PDF Available

Gender Politics in Semi-Democracies: Comparing gender quota implementation across political parties in Tanzania1

Elin Bjarnegård Uppsala University

Pär Zetterberg Uppsala University

Paper to be presented at the International Political Science Association (IPSA) World Congress Montréal, Quebec July 19-24 2014 Panel: Gender Politics and Regime Type

1

We want to thank Nina Carlsson and Katarina Galic for excellent research assistance.

1

Abstract The characteristics of hybrid regimes that are neither fully autocratic nor full-fledged democracies, but stalled somewhere in between, are increasingly in focus. This paper investigates potential gendered consequences of the particularities of these states. We already know that semi-democratic countries tend to have fewer women in politics than either autocracies or democracies (Bjarnegård & Melander 2011, Fallon et al. 2012), but we know less about why this is the case. It has mainly been explained by characteristics of democratic and autocratic regimes, such as the fact that it is easier to implement legal gender quotas in strong states. Because political parties are the main gatekeepers to women’s representation, the impact of different regime types on their recruitment practices should be in focus. In dominant party states, however, political parties operate under very different circumstances that also impact on their recruitment practices. We investigate how three political parties in semi-democratic Tanzania – dominant government party CCM and opposition parties Chadema and CUF - implement gender quotas. We find that whereas the resource-strong and bureaucratized CCM has a strong and influential women’s wing and well-established practices for quota implementation, the opposition parties are operating in a less formal manner letting informal institutions and personal connections guide their selection.

2

Introduction The political system of a country impacts on the opportunity structures and behavior of all actors operating within that system. This paper focuses specifically on how political parties and their propensity to be more inclusive with regards to women are impacted not only by the level of democraticness of a country, but also by the type of regime they operate within. Research has demonstrated that semi-democracies, at the middle of a quantitative democracy scale, are the least inclusive in terms of the representation of women in parliament (Bjarnegård and Melander, 2011, Fallon et al., 2012). At the same time, we know that in most regimes, democratic or not, political parties are the main gatekeepers to women’s political representation (Norris and Lovenduski, 1995, Norris, 1996). Considering that the regimes found at the middle of the democracy scale have very different autocratic pasts and evolve into very different types of environments for political parties, there is a need to take a closer look at what semi-democraticness entails for political parties and women. This paper zooms in on a type of semi-democracy in which we are likely to find large differences between political parties: the dominant party state of Tanzania. In dominant party states, the governing party often has a long experience of being the government and there is generally a resource gap between the dominant party and the opposition parties (e.g. Levitsky and Way, 2010b). One of the differences this may lead to is that opposition parties have fewer resources to put on organizational development, including establishing formal practices for selecting candidates. It has been suggested that women generally fare better in such parties that are rule-bound and bureaucratized than in less organized and more informal parties (e.g. Czudnowski, 1976, Guadagnini, 1993, Lovenduski and Norris, 1993, Norris, 1996, Bruhn, 2003, Caul Kittilson, 2006, Escobar-Lemmon and Taylor-Robinson, 2008). We study three different political parties in Tanzania: the dominant party CCM and the two largest opposition 3

parties, Chadema and CUF. In particular, we analyze the way in which they implement the special seats that are reserved for women. For these special seats, parties have to come up with a list consisting only of women candidates. We want to investigate whether there are discernible qualitative differences in how women fare also in candidate selection procedures where women do not compete against men. We find that women do, indeed, have more influence over the candidate selection procedure for the special seats in the dominant, and more bureaucratized, CCM than in the opposition parties Chadema and CUF: The women’s wing of CCM has worked its way to status within the party that is both formally and informally strong, and the candidate selection procedure has been delegated to this wing. Its dominant status also makes it easier to ensure representation from across the country and thereby using a decentralized selection system. It is more difficult to decentralize the selection procedure in the other two parties, who can only count on a few candidates each. In Chadema, resources have started to come in from abroad, organizational development is underway and a women’s wing was established recently. The establishment of procedures for selecting women for the special seats was, however, stalled by internal problems with corruption. CUF is both the poorest and the least bureaucratized party of the three, and it is also the party in which women have the least say about who gets nominated for the special seats. Because the special seats are distributed according to the overall election results, the dominant CCM also got the majority of the seats (67 out of 102 seats), while the electorally smaller parties Chadema and CCM did not get as many women representatives (25 and 10, respectively)2. The paper proceeds as follows: we start by outlining theories about gendered effects of democratization in general and the status of women in politics in semi-democracies in 2

For a table of election results, see appendix.

4

particular. We move on to building our theoretical argument about why we should expect countries in the middle of the democracy scale to work very differently for parties as well as for women. We elaborate on case selection and describe the background of the dominant party state of Tanzania. An analysis of the selection procedures for the special seats for each of the three parties follows, as well as a discussion about the implications of our results.

Gendered Democratization Our starting point is the empirical phenomenon that countries that are neither fully autocratic, nor fully democratic, perform the worst when it comes to the political representation of women. Male parliamentary dominance is most pronounced in these semi-democratic states, and when states move towards the middle of the democracy scale, either democratizing or reversing a democratic development, the representation of women goes down. Although this phenomenon has only recently been started to gain attention (Paxton et al., 2010, Bjarnegård and Melander, 2011, Fallon, et al., 2012), explanations have been put forward for why different regimes perform differently when it comes to the gendered composition of their parliaments. Autocracies and one-party-states are found at the lower end of the democracy scale and often have a relatively high proportion of women in parliament. These states are often strong, and it is possible for them to implement electoral gender quotas and to appoint women to political positions. Because many of these countries are communist countries, it is often in line with their ideology to boost gender equality in legislature. Autocratic one-party states thus fare relatively well when it comes to the numerical representation of women. When it comes to the actual power of the women elected, however, it is more questionable. They are elected into bodies that have little or no formal influence over policy-making, and they are not empowered by the support from their constituents, but by their proven loyalty to the governing party 5

(Fallon, et al., 2012, Bjarnegård and Melander, 2013). When these countries democratize, women suffer from the fact that they had merely achieved formal power, while they were left out of the real power centers and from the informal networks that continue to matter after the regime changes (Bjarnegård and Melander, 2013). Often, although there is a regime change the people in power remain the same because they are elected under the new system. This change in regime type, but not in regime personnel, also means that new pathways to power do not immediately present themselves to women in the new setting (Fallon, 2008). Women being side-stepped by informal channels is also the case even when they were active the liberation movement, because they were excluded from the formation of political parties that takes place after transition (Friedman, 2000, Jaquette, 2001, Waylen, 2007, Fallon, 2008). In some cases, primarily in post-communist settings, the exclusion of women from the informal spheres of power and the ‘forced emancipation’ causing women to have double burdens seem to have disillusioned large numbers of women to the extent that they left politics altogether at the time of liberalization (Jaquette and Wolchik, 1999, Matland and Montgomery, 2003). At the other end of the scale, the benefits of liberal democracy do not start to kick in until a state has reached a very high democratic level. Over time, however, stable liberal democracies offer increased opportunities for women’s political participation. Fallon et al demonstrate how women’s representation increases if democracy scores continuously improve. With each new democratic election women gain the knowledge they need in order to navigate the new political system (Fallon, et al., 2012). This slow learning process may be particularly important to women and other previously excluded groups (Jaquette and Wolchik, 1999, Tripp, 2001). With time, the powerful gatekeepers also have time to assess the politically inclined women, and women have time to mobilize. With a stable political system in place, stability and predictability cease to become the main concerns of political parties, and 6

international norms, transfer of power, and the extent to which representatives mirror society may become parts of the equation (Fallon, et al., 2012, Bjarnegård, 2013). With a system of accountability, the possibilities for women representatives to actually reach positions of power also increase. The complexities of the relationship between women and democratization have been highlighted by many. The type of autocratic regime, the role that women play in the liberation movement, the demands by international actors, the experience women gain over time – all may impact on how well women fare (Yoon, 2001, Lindberg, 2004, Tremblay, 2007, Waylen, 2007, Viterna et al., 2008, Paxton, et al., 2010, Krook, 2013). What takes place in between the autocratic stage and the fully democratic state can not merely be interpreted as a temporary transition, however. It is fairly straightforward to understand the mechanisms behind the curvilinearity of representation if and when states make a more or less smooth and steady transfer from autocracy to democracy. The analyses demonstrating the curvilinear relationship do not only build on such transitions, however. It has been increasingly acknowledged that many states do not go from one end to the other, but are stalled and stuck in between dictatorship and democracy (see e.g. Widmalm and Oskarsson, 2010 on 'promethocracy'). These stable semi-democracies merit some more attention. What are the characteristics that lead them to recruit fewer women? What are the possibilities of women to attain real influence in a semi-democratic state?

Varieties of semi-democraticness “Transitions from non-democratic regimes to polities that are at least nominally democratic have brought some significant changes in citizenship, participation, and policymaking, as well as disillusionment that changes have not been as far-reaching as had been hoped” (Waylen, 2007, 199). 7

In this quote from the concluding chapter in Waylen’s book ‘Engendering Transitions’, she notes that while many authoritarian states start upon a process of democratization, they become merely “nominally democratic”. There are many states in today’s world that are just that: nominally democratic. They are neither full-fledged democracies, nor dictatorships. Despite introducing multiparty elections, they have not democratized (Levitsky and Way, 2010a). Democratic institutions are usually in place and elections may be held with regular intervals. Several political parties are allowed to contest these elections, women are allowed to stand for election, and suffrage is universal. Yet, many states cannot be considered democratic. It may be because they do not respect civil liberties and human rights, or because elections are flawed. It may be because political violence is common or because the military has a lot of power. It may be because some political parties have more resources than others and because media is not objective in its reports. These regimes may be relatively stable, or they may be going back and forth between different regime types, causing an unstable and unpredictable political sphere. In other words, there are many varieties of semidemocraticness, and therefore these states also go by many names. Collier and Levitsky has simply called them democracies with adjectives (Collier and Levitsky, 1997), while O’Donnell famously refers to delegative democracies (O'Donnell, 1994). Diamond does not call them democracies at all, but rather focuses on the inherent mix between democratic and authoritarian elements in coining the overarching concept hybrid regimes (Diamond, 2002). Sometimes we see definitions that consider them primarily authoritarian with elements of democraticness instead of the other way around, such as competitive authoritarianism (Levitsky and Way, 2010a) or electoral authoritarianism (Schedler, 2006). They are given entirely new names, such as prometheocracy (Widmalm and

8

Oskarsson, 2010) or they are referred to as lacking of qualities such as democratic consolidation (Schedler, 1998, 2001). These semi-democratic polities have received more and more attention. Initially, they were regarded as democratizing but transitory states, being on a temporary trajectory from autocracy to democracy. They were treated as a residual category or as partial forms of democracy (Levitsky and Way, 2002). When research started zooming in on them, particularly by measuring democracy as a continuous variable with the possibility of curvilinear effects, some shared characteristics of these semi-democratic states were found. Apart from the starting-point of this paper – that semi-democracies as a group tend to include fewer women in their parliaments (Bjarnegård and Melander, 2011, Fallon, et al., 2012) – other studies have shown that semi-democracies have lower administrative capacity than either autocracies, where control is exerted from above, or than democracies, where control and accountability come from below. The analyses of such findings did, however, often keep to the view of the transitory status of these states, demonstrating that whereas strong autocratic institutions lose control and new actors enter the state when a country democratizes, new democratic institutions and requirements are not yet fully in place (Bäck and Hadenius, 2008). The mix between democratic rule and more autocratic methods in these countries make them inherently susceptible to instability, because the democratic institutions provide the opposition with at least some real opportunities for challenging the government (Levitsky and Way, 2002). Politicians in semi-democracies are therefore particularly prone to rely on informal networks such as patron-client relationships because the democratic institutions are not seen as reliable (Keefer, 2007, Keefer and Vlaicu, 2008, Bjarnegård, 2013).

9

As the years after the third wave of democratization passed, it became more and more evident that the stage these countries are in is, if not permanent, at least not merely transitory and temporary. When democratization research started looking into why democratization processes were stalled, it also embarked on a more detailed categorization of these regimes and the differences between them (Geddes, 1999). It is not just a matter of quantitative difference in degree of competitiveness, but a matter of qualitative differences in kind (Hadenius and Teorell, 2007). There is a large diversity of regimes in this ‘gray zone’, and research has started to unpack and categorize the different types (Carothers, 2002). With hybrid regimes as an over-arching name for all states that mix authoritarian and democratic features, it nevertheless becomes clear that these ingredients can be mixed and combined in different ways. As Levitsky and Way put it: “different mixes of authoritarian and democratic features have distinct historical roots, and they may have different implications for economic performance, human rights, and the prospects for democracy”(Levitsky and Way, 2002, 52). For instance, the dynamics of politics in Thailand are different from the dynamics of politics in Tanzania. Thailand is plagued by coups administered by what should be independent, democratic state institutions, such as the military and judiciary. The elected government is thus overthrown by institutions that are loyal to the elite establishment rather than to election results (Bjarnegård and Melander, 2014). In Tanzania, on the other hand, the elected government controls the state institutions to the extent that the opportunities of the opposition are severely restricted (Levitsky and Way, 2010b). Whereas in the Thailand, we see an oscillation between military dictatorship on the one hand and relatively democratic reform periods on the other hand, in Tanzania, we rather see a standstill. Although they are both within the group hybrid regimes, they have very different historical legacies that in turn create very different political challenges and dynamics for political parties and for the opportunities for women in politics. Earlier research by Bjarnegård (2013) has started to unveil the 10

gendered consequences of the oscillating, unstable Thailand where political actors strive to ascertain a stability and predictability they cannot find in the political system. Instead they invest in informal clientelist networks that can not be legally overthrown, simply because they do not exist as a formal or legal entity. In the search for stability politicians also tend to recruit new predictable and loyal politicians, who also fit the bill when it comes to being able to administer clientelist services. In order to be thus selected, you need a homosocial capital, i.e. resources and mutual trust, that Thai politicians are more likely to find among other men. This search for informal stability cuts across all parties, and is therefore a hurdle to women across the board. The other scenario, exemplified by Tanzania, poses very different challenges for parties and women. By closely studying Tanzania, this paper will delve deeper into the gendered dynamics of competitive authoritarian states, or dominant party states, where, as Levitsky and Way describe it, the playing field is uneven (Levitsky and Way, 2010b). One way for the incumbent regime to undermine democratic competition with non-democratic measures is to provide unequal access to state institutions, resources, and the media. Instead of using repression or electoral fraud as the primary means to maintain power, such a resource gap stalls democratization in countries like Botswana, Georgia, Kyrgyzstan, Malaysia, Malawi, Mozambique, Senegal, Tanzania and Venezuela. Keeping the playing field skewed means that autocratic incumbents can stay in power without resorting to repression. When the opposition is systematically denied the state resources that the incumbent government controls, it becomes difficult, though not impossible, for opposition candidates to win the election. In line with Levitsky and Way, we define an uneven playing field as “one in which incumbent abuse of the state generates such disparities in access to resources, media, or state institutions that opposition parties’ ability to organize and compete for national office is seriously impaired” 11

(Levitsky and Way, 2010b, 57). The concept is distinguished from more common incumbent advantage by differentiating between the many places where patronage and corruption affect the quality of democracy in many places, from the places where it undermines democracy itself (Levitsky and Way, 2010b). In a synthesis on twenty years of democratization literature, Geddes puts forward the idea that differences in democratization experience are rooted in the characteristics of the authoritarian regime, in other words in different regime legacies. She shows that whereas military regimes are among the most fragile, one-party states tend to be very long-lived (Geddes, 1999). Hadenius and Teorell, while building on Geddes, make an important addition: they distinguish between one-party regimes where opposition is not allowed, and dominant party regimes, where an opposition is allowed to contest elections (Hadenius and Teorell, 2007). This is particularly important because a dominant party regime often follows when multipartyism is introduced by a one-party regime. In sub-Saharan Africa in particular, oneparty regimes have tended to be transformed into dominant party systems, where opposition is allowed, but where the playing field is skewed in favor of the government party (Morse, 2014). Baker looks at the autocratic African leaders who were in power when the democratization movement started and finds that seven years later, half of them still held the highest office, having transformed themselves into democratic leaders (Baker, 1998). If this is true for the individual leaders, it is also true for the political parties. Boogaards finds that although a majority of African states have enjoyed multiparty elections, it is also the case that a majority of African states have seen no change of government (Bogaards, 2004). What these dominant party systems have in common is that basic democratic institutions are in place and there is some space open for contestation – but little real prospects for power reversal (Bratton and van de Walle, 1997, Carothers, 2002, Bogaards, 2004). Analysts seem more and more 12

resigned to the fact that these systems are likely to be long-lived and that we need to analyze them as new types of regimes in their own right (Herbst, 2001). Greene has developed a resource-argument that accounts for the dominant party equilibrium, both on the part of the dominant party as well as on the part of the opposition. The lack of possibilities for the opposition to seriously challenge the government party can be attributed to two things: first, to the resource gap between the governing party and opposition parties, and second, to the fact that the government party has found ways in which to increase the costs of participation in the opposition. Huge resource gaps are created when the dominant party is able and willing to tap state resources in order to finance large-scale canvassing and distribution of clientelist services and goods in order to make sure that voters are favorably inclined. At the same time, there are considerable costs involved for any person deciding to join the opposition. The most important cost is merely the reverse of the resource gap: by going to an opposition party instead of to the government party, your political career will likely be damaged, you will not get access to large campaign funds that can also help you get your message out, and you will be left out from old boys’ networks and potentially important business contacts kick-backs. Only the very ideologically inclined politicians will make that choice, career politicians are likely to go with the governing party. An additional cost can be the threat of repression and physical violence. Although repression can not be at the same systematic scale as in many autocracies, physical intimidation and election violence is common in many dominant party systems, and opposition politicians seem to be particularly targeted (Greene, 2007). Conditions favorable to democratization in these states, and thus conditions that are largely absent in stable dominant party states, are divisions within the incumbent regime and a unification of the opposition forces (Teorell, 2010).

13

Uneven playing fields for parties - and for women? It is thus clear that political parties in dominant party states operate under very different circumstances and offers very different types of opportunities for potential candidates. The resource gap between parties is likely to come with consequences related to internal party organization, where more resources facilitate bureaucratization and a branching out of different tasks. The costs of being a candidate for the ruling party are also much smaller than they are for an opposition party, both in terms of access to campaign funding and a welldeveloped partisan infrastructure as well in terms of not having to live under the constant threat of intimidation or physical violence from the ruling party. The proposition that semi-democracies are the least inclusive regimes when it comes to women in the legislatures thus needs to be nuanced in the context of a dominant party state where the very semi-democraticness lies in the fact that different parties operate under different circumstances and thus provide different opportunities for candidates. We thus need to gauge the different types of parties in a dominant party state from a gender perspective. First, we find it highly likely that the dominant party has a more developed party organization than its competitors. Dominant parties often have a long history of being the governing party, thus having developed a party infrastructure that is in many ways merged with the state structure. In a dominant party, the drafting of elaborate laws and regulations that specify how the state as well as the party is governed has developed over time in a dominant party. It has been suggested that women generally fare better in such parties that are rule-bound and bureaucratized and that parties who follow rules that are “detailed, explicit, standardized, implemented by party officials, and authorized in party documents” (Norris, 1996, 202) when they select candidates tend to end up with more female candidates than parties that select candidates in a more informal manner. In more bureaucratized parties, it is easier for potential candidates who are not from the inner circle to understand what it formally takes to 14

get in (e.g. Czudnowski, 1976, Guadagnini, 1993, Lovenduski and Norris, 1993, Norris, 1996, Bruhn, 2003, Caul Kittilson, 2006, Escobar-Lemmon and Taylor-Robinson, 2008). Parties that are less organized and more informal in character often have few formal rules that are not strictly implemented. Patronage based personal relationships trump written rules, which also means that “procedures may vary from one selection to another” (Norris, 1996, 203). Dominant parties, having an autocratic past, are also apt at developing large partisan infrastructures in order to control every corner of the country. The party exerts control by a combination of cooptation and coercion. It is important for dominant party states to create attractive career paths for a large number of loyal party members (Fjelde, 2010), including women. Moreover, many former one-party states have a socialist background, and the parties carry with them an ideological commitment to equality between the sexes. This is one reason why we see gender quotas have been used to ensure a relatively high representation of women in many authoritarian communist states (Bjarnegård and Melander, 2013). The question in these countries, where parliaments are generally not very powerful, is whether quotas are merely artificial and cosmetic rather than a road to real political power for women (cf. Dahlerup, 2007). The answer may come when one-partyism is abandoned, and legislatures are given more power. If quota women are then given career opportunities and leeway to work independently, and if quotas, in addition, serve as a stepping stone to the open seats, they may in fact work to transform structural inequalities between men and women and create what Darhour and Dahlerup call sustainable representation (Darhour and Dahlerup, 2013). We also propose that opposition parties in a former one party state do not have access to state infrastructure and resources, and they have not had the opportunity to create a party structure in which party loyalists can climb. They have no clear perks to offer candidates who decide to run for their party. Nor do they have the same experience with quotas as the government party 15

in a former one party state often has. Thus, they generally rely on personal relationships and trust to a much greater extent, in general as well as in the implementation of electoral gender quotas. In addition, opposition parties have much fewer ‘safe seats’ than the governing party. Research has also shown that women are generally seen as weaker candidates than male candidates. This is why majoritarian election systems, when the party has to focus on one candidate rather than on putting together balanced lists, tend to favor male candidates who are often perceived to be stronger and more likely winners in a ‘winner-takes-it-all’ race (Matland, 1998). A study of Thailand shows that within a majoritarian system, there were fewer female candidates in ‘close race’ constituencies, than in ‘safe’ constituencies, where the party was secure and could field ‘anyone’ and win anyway (Bjarnegård, 2013). Opposition parties in dominant party states may therefore not be able to ‘afford’ the risk of fielding a female candidate. In reserved seats quota systems, however, a certain number of seats are reserved for women and distributed after the election has taken place, and women thus do not compete with male candidates. In such a system women run the risk of becoming outsiders in the own party, because they are not involved in the actual campaign (Bjarnegård and Zetterberg, 2011). On the other hand, women in such systems do not challenge male positions and parties can safely select women without having to bypass or get rid of male candidates in the process. The internal power balance of the party is usually not threatened by the introduction of a reserved seats quota. Tripp and Kang show that reserved seats do not only constitute the quota type that is most favorable in terms of outcomes of the number of female legislators but that this quota type is also more common in nondemocratic – authoritarian and semiauthoritarian – settings. In such countries, Tripp and Kang claim, quotas can be symbolic in character or strategically 16

set up to create new patronage networks or to boost regime legitimacy internationally. It remains to be demonstrated, they argue, “whether female parliamentarians can have a significant impact in these authoritarian and semiauthoritarian regimes” (Tripp and Kang, 2008, 358). The present paper thus sets out to investigate a reserved seats quota in a dominant party regime setting – special seats in Tanzania - in order to investigate whether differences in party resources and opportunities have gendered consequences even in this type of quota where the implementation does not affect the actual number of women elected.

The dominant party state of Tanzania Tanzania is clearly a dominant party state, where the dominant party is Chama Cha Mapinduzi (CCM). CCM has stayed in power since multipartyism was introduced in 1992 and it has maintained control of at least two-thirds of the legislature, the Bunge ever since the first election in 1995. The party sometimes boasts that they do not need to cheat, because they are so certain of winning anyway (Hoffman and Robinson, 2009). Morse attributes CCM’s success mainly to the structure of the party, which was established when the country was a one-party state led by CCM’s predecessor Tanganyika African National Union (TANU) . Tanzania’s first president Julius Nyere developed an African version of socialism partly based on communal development. As a consequence, much attention was paid to institutionalizing the party not only at the national, but also at regional and local levels. Party branches were established in every ward and village throughout Tanzania and within those units party cells were established for every ten homes. Party branches were also established in large institutions and workplaces. A decentralization policy gave a lot of power to regional and district party offices and a lot of attention was paid to institutionalizing nominations for party office (Morse, 2014, 659).

17

The TANU nationalist movement emphasized gender equality in line with its socialist ideology. TANU founder Nyere said that “women suffered from inequalities which had nothing to do with their contribution to the family welfare. […] This is certainly inconsistent with our socialist conception of the equality of all human beings. […] If we want our country to make full and quick progress now, it is essential that our women live on terms of full equality with their fellow citizens who are men” (Nyere 1968, 109 as cited in Yoon, 2008). The TANU party attracted many women members to the point that it sometimes had more women than men as card-carrying members. One of its strategies for mass mobilization was investing into formal institutions, where the Union of Tanzanian Women was one important institution created (Yoon, 2008). Throughout multipartyism, CCM has managed to keep most of its candidates and defection has been surprisingly limited. Importantly, when Nyere started to advocate a democratic transition, it was because important external donors demanded it, not because there was an increasingly vociferous and strong internal opposition that needed to be satisfied. CCM leaders calculated that if they themselves introduced democracy before the opposition grew strong, they would be able to shape the new system in their favor (Hoffman and Robinson, 2009). They succeeded in this and the strong ties between party and state that are characteristic of a one-party state, failed to be broken by the introduction of multipartyism (Levitsky and Way, 2010b). It is also evident that CCM uses the new system and their existing country-wide infrastructure and its access to state funds efficiently. Its election campaigns are sophisticated and very expensive, with huge rallies with speeches, entertainment, and distribution of CCM t-shirts, hats, and posters. Hoffman and Lindsey claim that in the 2005 election campaign, the present president Jakaya Kikwete attended ninehundred rallies and spoke to 70,000 people a day (Hoffman and Robinson, 2009, 126). 18

The two main opposition parties, Chadema and the Civic United Front (CUF), have improved their organizational capacity lately, but have not been able to attain the same nationwide presence or voter loyalty. Part of the explanation for this is probably that these parties are not as broad. The Chadema party was founded by business elites and the CUF has most of its support from Muslims on the islands of Pemba and Zanzibar. But the parties are also not as institutionalized across the country as is CCM. The institutions of the opposition parties are not clearly written down, nor is there membership registry or any functioning communication or report system between branches and the central level. Both parties have, however, put a lot of effort into establishing party branches throughout Tanzania, but defection to CCM is common at the local level. Some opposition party officials even admit that their party branches are “manned by opportunistic individuals waiting to be bribed by CCM”. (Morse, 2014, 669). There are also differences between the opposition parties, however. Chadema has an internal organization for altering leadership positions, for instance, whereas CUF is much more personalistic, led, since its creation, by Maalim Seif and Ibrahim Lipumba. Chadema also launched “Operation Sangara” in 2005, emphasizing local leadership training as well as recruitment at local levels (Morse, 2014). Not everything in the electoral landscape of Tanzanian politics can be attributed to differences in internal structure of the political parties, however. CCM does use methods to coerce and control the opposition, although many claim it is strategic rather than systematic. By regulating civil society and the media and by sometimes resorting to physical violence, they are able to frighten the opposition if they do not manage to boost their support. Importantly, CCM controls the police and security services, and can therefore usually get away even with arresting opponents or committing physical acts of violence in the name of politics. One of 19

many examples include how CUF-leader Lipumba, during the 2005-campaign allegedly received death threats via cell phone text messages and was beaten and robbed during his campaign (Hoffman and Robinson, 2009).

Analyzing quota implementation in three Tanzanian political parties Tanzania has a legal electoral gender quota, called a special seat system. The special seat system has been in place since 1985, i.e. even before multipartyism was introduced, but the way in which the seats are distributed has been reformed and the number of seats reserved for women has continuously increased. It is a form of reserved seats, where seats are allocated to women according lists put together by the parties and to the proportion of votes the respective party received in the election. In other words, the law applies equally to all parties, but the parties who fare well in the election also get the additional bonus of putting the largest number of special seat women in the parliament. The special seats now make up 102 out of 357 and thus constitute a real power factor (Yoon, 2008). It is not only the election of special seats representatives that differ from the election of the other MPs, the special seat MPs also represent regions that correspond to several ordinary constituencies. This, in combination with the fact that political parties nominate and select the women, contributes to special-seat women having a different form of accountability to constituents than their colleagues who represent ordinary constituencies (Yoon, 2008). Although the special seat system was created to give women the experience they would need in order to move on to the ordinary constituency competition in the first-past-the-post system, very few women actually do move on to open seats. The special seats also seem to be the main road for women into the Tanzanian parliament: without the special seats there would be less than 10 percent women in the parliament (Yoon, 2013).

20

Only parties who get more than five percent of all valid votes can propose names of specialseat candidates to the National Electoral Commission. In the latest election three parties met this threshold: CCM, Chadema, and CUF, and we therefore study the way in which these three parties implemented the special seats quota (Yoon, 2013). We make use of internal party regulations and interview material gathered during field work in Tanzania in November 2013. We analyzed party regulations to assess the precision of the parties’ own internal rules. We interviewed party gatekeepers to get an impression of the extent to which the internal rules were actually followed and in order to get a deeper understanding of where decisions were taken and which priorities and choices that was important in the selection of the special seats. We also interviewed female candidates for special seats as well as women’s wings in order to determine the extent to which they were actively involved in the selection via a sort of functional decentralization (c.f. Hazan, 2002, Zetterberg, 2009) and to assess the power women have in the selection and implementation process itself and what this may imply for possibilities of subsequent political influence. We conducted a little over 20 interviews. Research assistants helped us to book interviews and also conducted some of the interviews, but most of them were conducted by the authors. Interviews were semi-structured, recorded, and subsequently transcribed and analyzed using the software Atlas.ti. In short, we assess whether the candidate selection process for the special seats is -

Bureaucratized and transparent, i.e. specified in party documents and carried out accordingly

-

territorially decentralized within the party structure

-

functionally decentralized to a women’s wing or similar body

21

Results The analysis is structured as follows: We assess the above criteria for one political party at a time: CCM, Chadema and CUF. The first step is to look at available party documents to see whether candidate selection for the special seats is formally regulated and specified. The second step is to compare the formal rules with the accounts given to us during interviews. We then look at the content of the actual candidate selection process to determine (1) whether it is territorially decentralized so that party branches or local party offices at different stages have a clear role to play and (2) whether it is also functionally decentralized to a women’s wing. We end by explicitly comparing the three cases and discussing the implications of the analysis. CCM We find that CCM by far has the most bureaucratized procedure for selecting candidates for the special seats, and that these candidates also have some power within the party. Thus, due to the dominant status of CCM, they also have some power to influence Tanzanian politics. Most of the few women who have been able to use the special seats as a stepping stone for ordinary constituency seats are also from the CCM. The whole process by which women are selected for the special seats is clearly outlined in the CCM party documents, in particular in the regulations of the CCM women’s wing, the Union of Tanzanian Women (UWT). The process is fully functionally decentralized to the UWT, who is in charge of the entire process up until the final rubberstamp decision made by the party executive. The process is also territorially decentralized and involves every level of the country and party. The interviews about the de facto process reflect the written process in detail. In other words, the CCM procedure for selecting special seats candidates seem to be highly bureaucratized. 22

The procedure starts at the district level. All aspirants have to collect election forms from their district, and submit their applications to the district offices of the UWT. The district UWT then forwards applications to the Regional Secretary. The Regional Congress (consisting of ward and district and regional leaders) of the UWT votes to come up with a shortlist of five names, who are then forwarded to the National Congress, usually ranked according to the number of votes they received at the Regional Congress. Candidates are scrutinized at the central national level of the UWT and if they pass the scrutiny the top two are usually selected for each region, while the remaining three tend to go to a national basket to be ranked by the UWT national congress. Sometimes the ranking of the candidates changes at this stage. If information that is not favorable to the party or the candidate emerges during the scrutiny, the Regional Office of the UWT does not have to abide by the vote count from the regional congress, but they can change the internal order of the candidates on the list. The CCM National Executive Committee makes the final decision on the list, and sends the list to the Election Commission (Chama Cha Mapinduzi Women Wing). The UWT is thus in charge of almost the entire process, and the UWT is also guaranteed representation in the National Executive Committee3. The nomination for the special seats within CCM is competitive at each level, and in order to become a candidate, you should be a long standing, loyal and active member of CCM and UWT (Yoon, 2008, see also Chama Cha Mapinduzi Constitution). The party needs to come up with one single ranked list to send to the Election Commission, but for a party like CCM, who is certain of getting a large proportion of the vote, this is not a major problem. The UWT National Congress uses a type of lottery in deciding which regions should come first on the list, and they are confident that the votes will

3

Interview at National Headquarters UWT, November 2014. [More specific information to be added in later version.]

23

be enough to ensure that they have regional representation of special seat Members of Parliament throughout the country4. Interestingly, the influence and status of the UWT within CCM is not something they have brought with them from the one-party regime, but rather something they have achieved after multipartyism was introduced. Killian describes early special seat recruitment as being entirely in the hands of the national executive of CCM and where opinions and even favored candidates of the UWT were sometimes entirely bypassed (Killian, 1996). During a reform of the CCM Constitution, the nomination for the special seats was entirely delegated to the UWT and as a party women’s wing in Tanzania, the UWT now has a unique power position within the party and when it comes to determining the special seats candidates. The type of background that is favorable for winning the candidacy at the Regional Congress is documented experience of activities within the UWT. The UWT also looks upon special seat candidates as their extended arm into the parliament. They are expected to work with women’s issues in accordance with UWT’s policies and with a focus on the particular region they represent5.

Chadema Chadema is a party that is undergoing change of its internal organization, and this is also visible in the nomination of the special seats. They have received considerable funds and assistance from a sister-party in Denmark for their internal organization, and are working with

4

Interview at National Headquarters UWT, November 2014. [More specific information to be added in later version.] 5

Interview at National Headquarters UWT, November 2014. [More specific information to be added in later version.]

24

capacity building and with decentralizing the party organization, including functional decentralization to the women’s wing. The Chadema women’s wing, Bawacha, was not established until in 2006 and thus, up until recently, the women’s wing was not involved in the nomination of special seat candidates6. The recently updated formal rules have, however, not yet materialized as fully established practices in the party. The formal, written procedure is quite similar to the one used by CCM. Eligible candidates are not only members of Chadema, but they should have contributed to party building and have made outstanding contributions to the party. Because the women’s wing of Chadema, Bawacha, is not as established as its counterpart in CCM, membership of the party is emphasized, rather than membership of the women’s wing. The recent changes did, however, formally delegate the procedure to Bawacha. It states that aspirants fill out forms and a district general meeting of Bawacha is held to elect candidates from among the pool of aspirants. The top three names are highlighted, but all names are taken to the Bawacha headquarters for scrutiny. The National Executive Committee of the Bawacha calls all qualified candidates from the district level and have an election, where the members of the Bawacha National General Meeting are eligible to vote. The candidates are simply listed according to the number of votes they receive in this meeting (Criteria and Procedure of Chadema Special Seat MPs). Because it has, in practice, been party leaders that nominate candidates for the special seats the procedure has been less transparent than the one in CCM and susceptible to corruption (Yoon, 2008). In 2010, the women’s wing was supposed to be responsible for the selection of special seat candidates, in accordance with the new guidelines. But two days before the final

6

Interview with Hon. Suzan Lyimo, chairperson of Bawacha (MP), November 2014. [More specific information to be added in later version.]

25

selection, there was evidence of corruption in the process, and the Central Committee of the party had to override the women’s wing and hire a consultant from the University of Dar Es Salaam who came up with graded criteria that were applied at the last minute, in order to come up with a list at all7. Although they seemed to have followed the new procedures up until the last stage, many things went wrong. According to Susan Lyimo, the Chairperson of the Bawacha, there was a lot of corruption in the process and some CCM-loyalists fielded their wives and relatives as fake candidates and Chadema failed to prevent this from happening. She attributes the failure to discover this until at the very last stage to the party’s status of an opposition party in a dominant party state. Opposition parties do not have the same presence at the lowest levels. At the large final meeting, where the vote was to take place, there were rumors that there were members who were not supposed to vote. It eventually surfaced that a group had gone to the districts and brought 50 women who were told to vote for certain candidates to the meeting in Dar Es Salaam. When they found the list of names they were supposed to vote for, Bawacha handed the process over to the party again.8. The Secretary General of Chadema also says that they do not have any particular expectations of the special seat MPs, other than that they are expected to contribute to all national programmes9. It is difficult to have any particular expectations, because the special seats MPs in effect have a larger area to cover than an ordinary MP, but they also do not get the

7

Interview with Secretary General of Chadema, Dr. Slaa. November 2014. [More specific information to be added in later version.] 8

Interview with Hon. Suzan Lyimo, chairperson of Bawacha (MP), November 2014. [More specific information to be added in later version.] 9

Interview with Secretary General of Chadema, Dr. Slaa. November 2014. [More specific information to be added in later version.]

26

constituency development fund that other MPs get. There are no funds for working on women’s issues. In addition, Chadema can not, like CCM, expect to have special seat MPs everywhere. They therefore try to change the discourse and call themselves national MPs, instead of regional, and they try to give every special seat MP a zone to represent and to visit. They have not yet seen to what extent this system will work, however10. The special seats are also very numerically important to Chadema, as they represent a majority of their seats in the legislature, the Bunge 11. Party loyalty among the few parliamentarians that this opposition party has is therefore likely to be highly valued, which might make it more difficult for the women’s wing to gain full control of the candidate selection procedure.

CUF The selection procedures of the Civic United Front (CUF) were the least bureaucratized, transparent, and by far the most centralized, both in terms of territorial centralization and functional centralization. The Executive Committee and the General Assembly of the party has large leeway in selecting constituency candidates, although they collect names from the local level. When it comes to the special seats, the wording of the internal rules is quite vague: “At any time where there is a need for special seat for women; for the parliament or a house of representative; the general assembly will set the special rules on obtaining MPs and the respective representatives” (Article 92(5) of the Constitution of the Civic United Front). CUF does have a women’s wing but it is clearly not very involved in selecting the special seats. They are discussing ways in which the women’s wing could be more involved in the 10

Interview with Hon. Suzan Lyimo, chairperson of Bawacha (MP) and with Hon. Naomi Kaihula (MP), November 2014. [More specific information to be added in later version.] 11

See table of election results in the appendix.

27

selection process, but no new rules have yet been implemented. The Executive Secretary of the women’s wing, Nuru Bafadhil, describes the process as going straight from districts’ suggestion of names to the National Executive Committee of the party. The women’s wing can give comments and recommendations, but the ranking of the names is entirely in the hands of the National Executive Committee, and the criteria they use are largely unknown to outsiders12 . What is evident is that party loyalty matters, rather than documented work for women’s issues. The Executive Secretary even claims that working with women’s issue is not at all important as a candidate for a special seat, because once in the parliament, you will work for and with everyone, not just for and with women. In particular because the CUF has such few women, they see themselves as working for all of Tanzania, not for a particular region13. The party often seems to select special seat candidates from the party’s strongholds, primarily the islands of Zanzibar and Pemba14. A woman’s wing member at a branch in Dar Es Salaam did not know how the special seats women were selected, or who they were: “We don’t know the names of the women on the special seats. They just come to the ward and then they leave. [….] They are working in parliament with these issues”15. Discussion Comparing the three parties in Tanzania, and the manner in which they select candidates for special seats, we can confirm our expectation that the dominant party CCM does not only 12

Interview with Nuru Awadh Bafadhil (CUF), Executive Secretary, Women’s Wing and with a local CUF women’s wing November 2014. [More specific information to be added in later version.] 13

Interview with Nuru Awadh Bafadhil (CUF), Executive Secretary, Women’s Wing. November 2014. [More specific information to be added in later version.] 14

Interview with a local CUF women’s wing November 2014. [More specific information to be added in later version.] 15

Interview with Muzney, a district member of the CUF women’s wing in Kinondoni District, Dar Es Salaam. November 2014. [More specific information to be added in later version.]

28

have more resources but that they also have used these resources to develop the party, both vertically, by decentralizing it territorially, and functionally, by delegating power to the women’s wings. In terms of gender politics, it is thus not only the level of democracy that matters, but also what type of regime and what type of party you stand for that matters. In the dominant party state of Tanzania, conditions for numerical as well as effective representation of women politician are clearly much larger in CCM than in Chadema and CUF. Numerically, CCM receives a majority of the 102 special seats, and the special seat system thus works to boost CCM’s dominant presence in the legislature even more. Beyond numbers, the CCM women’s wing has a clear mandate to select the most suitable candidates. Knowing that they will get a large majority of the votes, they can also ensure that each region in Tanzania gets its own women’s representative. Being a parliamentarian in Tanzania is also a lot easier if you have the ruling party behind you. In other words, the CCM representatives are also in a better position to actually enact gender-friendly legislation and to rise in ranks and reach even more influential positions within the party and government. The opposition parties get much fewer special seat parliamentarians and because the selection procedures are unclear, it is more difficult to know who they represent. Party loyalty, rather than work for women’s issues, seems to be paramount. In a dominant party system, any additional parliamentary candidate loyal to the opposition is important enough for the party leadership to want to maintain control. Clearly, prospects of enacting legal change or getting more influence are few for opposition candidates.

Conclusion This paper points to the necessity of analyzing the different characteristics of such hybrid regimes that are neither fully autocratic nor full-fledged democracies, but stalled somewhere in between. This paper investigates potential gendered consequences of the particularities of one type of hybrid regime: the dominant party state. The analysis nuances the idea about why 29

countries at the middle of the democracy scale tend to have fewer women in parliament than either autocracies or democracies and demonstrates that in certain types of state, the type of party matters most for the opportunities of women politicians. In Tanzania, as in other dominant party states, political parties operate under very different circumstances. The dominant party CCM has more resources, a more bureaucratized organization, and more decentralized selection procedures than either of the two opposition parties. We demonstrate that this also impacts on their recruitment practices. Whereas the resource-strong and bureaucratized CCM puts the entire candidate selection for the special seats in charge of a strong and influential women’s wing that is decentralized to every local level, the opposition parties are operating in a less formal and transparent manner letting personal connections and party loyalty guide their selection more than documented activities related to women’s issues. The findings have several potential consequences that need to be further investigated. First, we need to better understand to what extent these findings alter our understanding that semidemocraticness in itself is detrimental for women’s participation. Second, the vulnerability not only for opposition parties in dominant party states, but also for gender politics to develop in such states, should be acknowledged and further investigated.

30

References Baker, Bruce. (1998) The Class of 1990: How Have the Autocratic Leaders of Sub-Saharan Africa Fared under Democratization? . Third World Quarterly 19:115-27. Bjarnegård, Elin. (2013) Gender, Informal Institutions and Political Recruitment: Explaining Male Dominance in Parliamentary Representation. Houndmills, Basingstoke: Palgrave Macmillan. Bjarnegård, Elin, and Erik Melander. (2011) Disentangling Gender, Peace and Democratization: The Negative Effects of Militarized Masculinity. Journal of Gender Studies 20:139-54. ———. (2013) Revisiting Representation: Communism, Women in Politics, and the Decline of Armed Conflict in East Asia. International Interactions 39:558-74. ———. (2014) Thailand's Missing Democrats: Reds, Yellows and the Silent Majority. Foreign Affairs http://www.foreignaffairs.com/articles/141454/elin-bjarnegard-anderik-melander/thailands-missing-democrats. Bjarnegård, Elin, and Pär Zetterberg. (2011) Removing Quotas, Maintaining Representation: Overcoming Gender Inequalities in Political Party Recruitment. Representation 47:187-99. Bogaards, Matthijs. (2004) Counting Parties and Identifying Dominant Party Systems in Africa. European Journal of Political Research 43:173-97. Bratton, Michael, and Nicolas van de Walle. (1997) Democratic Experiments in Africa: Regime Transitions in Comparative Perspective. Cambridge: Cambridge University Press. Bruhn, Kathleen. (2003) Whores and Lesbians: Political Activism, Party Strategies, and Gender Quotas in Mexico. Electoral Studies 22:101-19. 31

Bäck, Hanna, and Axel Hadenius. (2008) Democracy and State Capacity: Exploring a JShaped Relationship. Governance: An International Journal of Policy, Administration, and Institutions 21:1-24. Carothers, Thomas. (2002) The End of the Transition Paradigm. Journal of Democracy 13:521. Caul Kittilson, Miki. (2006) Challenging Parties, Changing Parliaments: Women and Elected Office in Contemporary Western Europe. Columbus: The Ohio State University Press. Collier, David, and Steven Levitsky. (1997) Democracy with Adjectives: Conceptual Innovation in Comparative Research. World Politics 49:430-51. Czudnowski, Moshe M. (1976) Political Recruitment. In Handbook of Political Science: Micropolitical Theory, edited by Fred I. Greenstein and Nelson W. Polsby, pp. 155242. Reading, MA: Addison Wesley. Dahlerup, Drude. (2007) Electoral Gender Quotas: Between Equality of Opportunity and Equality of Result. Representation 43:73-92. Darhour, Hanane, and Drude Dahlerup. (2013) Sustainable Representation of Women through Gender Quotas: A Decade's Experience in Morocco. Women's Studies International Forum 41:132-42. Diamond, Larry. (2002) Thinking About Hybrid Regimes. Journal of Democracy 13:21-35. Escobar-Lemmon, Maria, and Michelle M. Taylor-Robinson. (2008) How Do Candidate Recruitment and Selection Processes Affect Representation of Women? In Pathways to Power: Political Recruitment and Candidate Selection in Latin America, edited by Peter Siavelis and Scott Morgenstern. University Park: Pennsylvania State University Press. Fallon, Kathleen M. (2008) Democracy and the Rise of Women's Movements in Sub-Saharan Afriva. Baltimore, MD: The Johns Hopkins University Press. 32

Fallon, Kathleen M., Liam Swiss, and Jocelyn Viterna. (2012) Resolving the Democracy Paradox: Democratization and Women's Legislative Representation in Developing Nations, 1975 to 2009. American Sociological Review 77:380-408. Fjelde, Hanne. (2010) Generals, Dictators, and Kings: Authoritarian Regimes and Civil Conflict, 1973-2004. Conflict Management and Peace Science 27:195. Friedman, Elisabeth J. (2000) Unfinished Transitions: Women and the Gendered Development of Democracy in Venezuela. University Park, PA: Pennsylvania State University Press. Geddes, Barbara. (1999) What Do We Know About Democratization after Twenty Years? Annual Review of Political Science:115-44. Greene, Kenneth F. (2007) Why Dominant Parties Lose: Mexico's Democratization in Comparative Perspective. Cambridge: Cambridge University Press. Guadagnini, Maria. (1993) A "Partiocrazia" without Women: The Case of the Italian Party System. In Gender and Party Politics, edited by Joni Lovenduski and Pippa Norris. London: Sage. Hadenius, Axel, and Jan Teorell. (2007) Pathways from Authoritarianism. Journal of Democracy 18:143-57. Hazan, Reuven Y. (2002) Candidate Selection. In Comparing Democracies 2: New Challenges in the Study of Elections and Voting., edited by Lawrence LeDuc, Richard G. Niemi and Pippa Norris. London and Thousand Oaks: Sage. Herbst, Jeffrey. (2001) Political Liberalization in Africa after Ten Years. Comparative Politics 33:357-75. Hoffman, Barak, and Lindsay Robinson. (2009) Tanzania's Missing Opposition. Journal of Democracy 20:123-36.

33

Jaquette, Jane. (2001) Women and Democracy: Regional Differences and Contrasting Views. Journal of Democracy 12:111-25. Jaquette, Jane S., and Sharon L. Wolchik. (1999) Women and Democratization in Latin America and Central and Eastern Europe. Baltimore and London: The John Hopkins University Press. Keefer, Philip. (2007) Clientelism, Credibility, and the Policy Choices of Young Democracies. American Journal of Political Science 51:804-21. Keefer, Philip, and Razvan Vlaicu. (2008) Democracy, Credibility, and Clientelism. Journal of Law, Economics, and Organization 24:371-406. Killian, Bernadeta. (1996) A Policy of Parliementary 'Special Seats' for Women in Tanzania: Its Effectiveness. Ufahamu: A Journal of African Studies 24:21-31. Krook, Mona Lena. (2013) Gender Quotas and Democracy: Insights from Africa and Beyond. Women's Study International Forum 41:160-63. Levitsky, Steven, and Lucan A. Way. (2010a) Competitive Authoritarianism: Hybrid Regimes after the Cold War. Cambridge: Cambridge University Press. ———. (2002) The Rise of Competitive Authoritarianism. Journal of Democracy 13:51-65. ———. (2010b) Why Democracy Needs a Level Playing Field. Journal of Democracy 21:5768. Lindberg, Staffan I. (2004) Women's Empowerment and Democratization. Studies in Comparative International Development 39:28-53. Lovenduski, Joni, and Pippa Norris. (1993) Gender and Party Politics. London: SAGE. Matland, Richard. (1998) Women's Legislative Representation in National Legislatures: A Comparison of Democracies in Developed and Developing Countries. Legislative Studies Quarterly 23.

34

Matland, Richard E., and Kathleen A. Montgomery. (2003) Women's Access to Political Power in Post Communist Europe. New York: Oxford University Press. Morse, Yonatan L. (2014) Party Matters: The Institutional Origins of Competitive Hegemony in Tanzania. Democratization 21:655-77. Norris, Pippa. (1996) Legislative Recruitment. In Comparing Democracies: Elections and Voting in Global Perspective, edited by Lawrence LeDuc, Richard G Niemi and Pippa Norris, pp. 184-215. Thousand Oaks, London, New Delhi: Sage Publications. Norris, Pippa, and Joni Lovenduski. (1995) Political Recruitment. Gender, Race and Class in British Parliament. Cambridge: Cambridge University Press. O'Donnell, Guillermo. (1994) Delegative Democracy. Journal of Democracy 5:55-69. Paxton, Pamela, Melanie M. Hughes, and Matthew A. Painter. (2010) Growth in Women's Political Representation: A Longitudinal Exploration of Democracy, Electoral System and Gender Quotas. European Journal of Political Research 49:25-52. Schedler, Andreas. (2006) Electoral Authoritarianism: The Dynamics of Unfree Competition. Boulder, CO: Lynne Rienner Publishers. ———. (2001) Measuring Democratic Consolidation. Studies in Comparative International Development 36:66-92. ———. (1998) What Is Democratic Consolidation? Journal of Democracy 9:91-107. Teorell, Jan. (2010) Determinants of Democratization: Explaining Regime Change in the World, 1972-2006. Cambridge: Cambridge University Press. Tremblay, Manon. (2007) Democracy, Representation, and Women. Democratization 14:53353. Tripp, Aili Mari. (2001) The Politics of Autonomy and Cooptation in Africa: The Case of the Ugandan Women's Movement. The Journal of modern African studies 39:101-28.

35

Tripp, Aili Mari, and Alice Kang. (2008) The Global Impact of Quotas: On the Fast Track to Increased Female Legislative Representation. Comparative Political Studies 41:33861. Waylen, Georgina. (2007) Engendering Transitions: Women's Mobilization, Institutions, and Gender Outcomes. Oxford: Oxford University Press. Widmalm, Sten, and Sven Oskarsson. (2010) Prometokrati: Mellan Diktatur Och Demokrati [Prometheocracy: Between Dictatorship and Democracy]. Lund: Studentlitteratur. Viterna, Jocelyn, Kathleen M. Fallon, and Jason Beckfield. (2008) How Development Matters. International Journal of Comparative Sociology 49:455-77. Yoon, Mi Yung. (2001) Democratization and Women's Legislative Representation in SubSaharan Africa. Democratization 8:169-90. ———. (2013) Special Seats for Women in Parliament and Democratization: The Case of Tanzania. Women's Study International Forum 41:143-49. ———. (2008) Special Seats for Women in the National Legislature: The Case of Tanzania. Africa Today 55:61-86. Zetterberg, Pär. (2009) Engineering Equality? Assessing the Multiple Impacts of Electoral Gender Quotas. In Department of Government: Uppsala University, Sweden.

36

Appendix. Election Results Tanzania, 2010 Directly

Special

Appointed/

elected

seats

Zanzibar

259

186

67

6

48

23

25

0

Civic United Front (CUF)

36

24

10

2

NCCR-Magueuzi

4

4

0

0

United Democratic Party

1

1

0

0

Tanzania Labour Party (TLP)

1

0

0

1

Others

1

0

0

0

Party

Revolutionary Party of

Total

Tanzania (CCM) Chadema (Party for Democracy and Development)

Men: 224 Women 126 Source: IPU

37