Some sweep to power. Many more crumble. Why? And which way will Zimbabwe's 2018 coalition go?
For the past two decades, the phenomenon of the opposition coalition has gained growing traction and interest across Africa.
In 2000, a group of opposition parties in Senegal joined forces as the Sopi (or "Change") alliance. Together, they defeated the incumbent president and ended 40 years of one-party dominance.
In 2002, Kenya's opposition repeated the trick. In the 1992 and 1997 elections, losing parties had cumulatively gained over 60% of the vote. But this time around, they grouped together as the National Rainbow Coalition (NARC). This united opposition swept to power, removing the party that had governed Kenya since 1963.
Since then, pre-electoral coalitions have changed governments again in Senegal, as well as in Liberia, Madagascar, Malawi, Mali, Mauritius, Nigeria and The Gambia.
When elections are held in 2018, Zimbabwe hopes to join this growing list.
Morgan Tsvangirai's Movement for Democratic Change-Tsvangirai (MDC-T) and Joice Mujuru's National People's Party (NPP) have agreed - in principle - to team up. A host of other opposition parties have also provisionally joined, including: Welshman Ncube's MDC, Dumiso Dabengwa's Zimbabwe African People's Union (ZAPU), Simba Makoni's Mavambo/Kusile/Dawn (MKD), Tendai Biti's People's Democratic Party (PDP), and Elton Mangoma's Renewal Democrats of Zimbabwe (RDZ).
This would be a broad and impressive coalition, bringing together many well-known faces and politicians who have electoral support outside of traditional opposition strongholds. But for every successful opposition alliance Africa has seen, there have been several more that have crumbled after early optimism or fallen flat at the ballot box.
[Zimbabwe's opposition coalition: Avengers Assemble or Suicide Squad?]
Why do coalitions sometimes become more than the sum of their parts and generate a huge surge of support? Why do they often fragment and collapse?
Fighting each other vs. fighting together
One crucial indicator of whether an opposition coalition will succeed is how polarised the political landscape is. This can determine the degree to which parties are able to join forces coherently and without undermining their own reputation and principles.
According to political scientist Nicolas Van de Walle, opposition coalitions only work when they appear capable of winning and thus prompt members of the ruling party to defect. These defectors not only bolster the ranks of the opposition, but can bring supporters with them and sway undecided voters.
Ahead of Nigeria's 2015 elections, for example, the All Progressives Congress was significantly strengthened by mass defections from the ruling People's Democratic Party (PDP). Similarly, in Zambia in 2016, dozens of defectors from the ruling Patriotic Front (PF) and Movement for Multi-party Democracy (MMD) drastically improved the electoral fortunes of the United Party for National Development (UPND).
[How Zambia's once insuperable MMD returned to power by disappearing]
However, this strategy is not straightforward. To begin with, it can be difficult to encourage members of the ruling party to cross the aisle. And when they do, it can be tough to persuade opposition supporters to vote for someone who was, until recently, part of the government.
The more deeply polarised the political landscape, the harder this is.
Uganda, for example, is at the other end of the spectrum to Nigeria or Zambia where defections are not particularly costly. In Uganda, the main opposition Forum for Democratic Change (FDC) has long defined itself in stark contrast to the ruling National Resistance Movement (NRM). It emphasises the persecution it has experienced at the hands of the ruling party, which it characterises as illegitimate and unjust.
This makes it hard for the FDC to encourage defections from the NRM, which it consistently attacks in no uncertain terms. Moreover, when figures within the ruling party do defect, it can be risky for the FDC to bring them into the fold without undermining its own image.
In 2016, the FDC faced a dilemma when the opposition alliance it was part of voted for the recently-expelled former Prime Minister Amama Mbabazi to be its flag-bearer. The FDC was confronted with the prospect of backing a former insider in the very government it had long denounced. Afraid of alienating its base and diluting its anti-regime brand, the FDC decided to leave the coalition.
When it comes to Zimbabwe, the environment looks similarly polarised, especially between the main opposition MDC-T and the ruling ZANU-PF. The MDC-T claims to be the democratic saviour to the ZANU-PF's illegitimate authoritarianism; ZANU-PF presents itself as the liberator hero to the MDC-T's foreign subservience.
But unlike the FDC in Uganda, the MDC-T seems to be - at least in principle - less averse to allying with the long-standing government insider, Joice Mujuru. Nevertheless, the fundamental irreconcilability between the images of the MDC-T and ZANU-PF brings a certain riskiness to this decision. What does it say about the vociferous opposition party that it now says it is prepared to stand alongside a former ZANU-PF stalwart and vice-president? How will its supporters react?
In Zimbabwe, however, there are added complications arising from the fact that the hostile political climate also stretches to relations between some opposition parties. The MDC-T, for example, has used polarising rhetoric not just to condemn the ruling party, but also to criticise the opposition groups that emerged from a split in 2005. Tsvangirai's faction branded this MDC breakaway as "sell outs" and "traitors".
This rhetoric made attempts at a rapprochement in 2008 and 2013 more difficult. It will also make joining forces trickier ahead of 2018, especially given that many opposition groups have splintered even further since then. The PDP, for example, is the result another split in the MDC-T from when Tendai Biti walked out in 2014. And the ZRD is the result of fissure in the PDP.
It can be difficult to build stable and effective structures when so many bridges have been burned.
Who will lead the coalition?
The main hurdle at which most opposition coalitions fall is in picking its leader. This contest is often keenly fought, particularly since the benefits of the presidency are so great in most African countries.
The decision of who should be the figurehead is least contentious when there are recent and reliable indicators of party strength, such as the results of parliamentary by-elections. With this data, it is more straightforward to work out which candidate has the most recognition and support.
However, this kind of information doesn't guarantee an easy process. In Zambia, for example, the opposition UPND won a series of unexpected by-elections victories between 2011 and 2016. Its candidate Hakainde Hichilema also garnered 46.7% of the vote in the 2015 presidential by-election, losing by just 27 000 votes.
Nevertheless in 2016, when the UPND tried to form a coalition with opposition leader Edith Nawakwi - who got 0.9% in 2015 - Nawakwi insisted that she should lead the alliance. She said that she had supported Hichilema in a 2006 coalition and that now it was his turn to support her. The parties went their separate ways.
In Uganda 2016, the choice of who should head up the coalition was also a source of disagreement and ended up breaking apart the alliance. In this instance, the uncertainty over the relative popularity of the two potential candidates made it harder to judge who would be the best-placed candidate.
The FDC's Kizza Besigye had the broadest national reach and most organised structures, but had not surpassed 37% in three previous presidential runs. Meanwhile, former PM Mbabazi was an unknown quantity as an opposition figure, but was well-known nationally and had insider knowledge about the ruling party's election strategies. When Mbabazi was chosen, the FDC refused to back him and left, leading to the breakdown of the coalition.
Zimbabwe's nascent coalition is now in a similar situation. Tsvangirai is a veteran opposition figure with a proven track record of mobilising supporters, while Mujuru is an untested but well-known former ruling party insider with support in ruling party strongholds and close contacts in the intelligence services and police. It is uncertain which figure would draw the most voters and which will prevail in the contest to lead the coalition.
In terms of measuring the MDC-T's support, the series of splits and a three-year electoral boycott make it difficult to judge. But the 2017 Afrobarometer survey suggests that the opposition has lost ground since the 2013 elections, when Tsvangirai got just 34% of the vote. According to the study, the opposition is trusted by just 32% of the population, compared to 65% who trust the president and 56% the ruling party.
This may give more ammunition to those who'd prefer to see Mujuru as the flag-bearer. But it remains to be seen if the MDC-T would accept this outcome, or make the same decision as the FDC in Uganda.
Keeping the lower ranks happy
However, it is not just the leader of the coalition that matters. Political parties are comprised of hundreds of functionaries with their own ambitions and goals, and alliances frequently collapse as a result of vested interests at lower party echelons.
Ahead of Zambia's 2011 elections, for example, a pact between the two largest opposition parties at the time - the UPND and the Patriotic Front (PF) - was apparently scuppered by PF Secretary-General Wynter Kabimba. Kabimba had his own presidential ambitions and knew that he would be pushed down the pecking order under a coalition.
A similar thing happened in Zimbabwe in 2013. In that situation, two of Tsvangirai's inner circle that reportedly opposed a coalition with the breakaway MDC due to fears of losing their own positions in the hierarchy.
These concerns also arise around parliamentary races. Opposition parties that typically compete for the same seats face much more internal resistance to coalitions than those with different, complimentary constituencies.
In Kenya, for example, coalitions are frequently formed between relatively geographically contained, ethnic-based parties. Because the parties within these groupings - such as the recently formed National Super Alliance - rarely compete for the same seats, coalitions in Kenya face relatively little resistance from the lower ranks.
By contrast, negotiations between the two MDC factions in Zimbabwe in 2007 ultimately failed, partly because the MDC-T insisted on contesting two seats held by the other party in the opposition's shared stronghold in Matabeleland. Both sides refused to back down.
Ahead of 2018, Zimbabwe's opposition groups will face these discussions once again. But it is possible that they will be easier this time around. Because of repeated fragmentation, many of the resulting parties looking to form a coalition are smaller and newer.
This may mean that they are less able to make strong demands. It may also mean that negotiations are more about bringing party leaders on board than appeasing each grouping's structures. Because of this, the talks may bypass complex internal party dynamics and side-step vested interests lower down the party chain.
Zimbabwe 2018: Can a coalition win?
While 45% of Zimbabweans polled by Afrobarometer expressed support for the idea of an opposition coalition, there are still many answered questions and tricky challenges facing the nascent coalition in the run up to 2018.
Can the animosity between different factions be put aside? Will opposition supporters accept the inclusion of Mujuru, a decades-long ZANU-PF insider?
How will the presidential candidate be picked, based on what calculations and agreements? And how will those less pleased by the choice react?
Will a coalition deal involve running joint candidates in each constituency? And if so, how will those asked to shelve their ambitions respond?
These are tricky questions. But in many ways, they are just the start. Even once these dilemmas are resolved, there is still the ultimate question of whether even a perfectly-coherent and functional opposition coalition has much chance of winning. Bringing together a range of opposition parties is the first step in defeating the ruling party, not the final blow.
On this front, the prospects for the opposition in Zimbabwe do not look particularly rosy.
Trust in the opposition is low. Old methods of party mobilisation using organised labour are no longer an option given skyrocketing unemployment and informal livelihoods. And the impact of new social movements - such as #ThisFlag and #Tajamuka - is likely limited given that they are predominantly urban-based.
Meanwhile, ZANU-PF has shifted into election mode, doling out urban land in an effort to shore up support and turning the screws on vocal opponents. The ruling party may be riven with internal factionalism, but it's unclear if the opposition can turn this to their advantage.
The MDC-T remains the most organised opposition party with the largest organisational reach. If it could make it work, a broad coalition would bolster its ranks and could give it further appeal. But there remain serious concerns in the opposition including poor strategic thinking, complacency, a tendency towards authoritarianism and internal fractionalisation.
Even if the 2018 vote is a straightforward contest between a ruling party and a truly united opposition, the election is still likely to be one of fairly poor choices.