The core problem in Nepali politics today is intra-party differences. There is a deepening rift between the Maoist dogmatists and pragmatists. The NC's right wing tendencies are re-emerging, drowning out centrist voices. It is a wonder that the MJF is a party at all, given the hatred among the top leaders.

But the most interesting ideological and leadership struggle is within the UML. The UML is thriving, despite being rejected by the people, because the NC and Maoists are fighting. The party is suffering from a severe existential crisis, and can't figure out where it has gone wrong.

Any UML leader begins his assessment of the Maoists by claiming that the rebels had to give up their 'ultra-left adventurism' and accept the Madan Bhandari line. According to this version, the 'people's war' was wrong. The Maoists could rise because of the mess created by the NC, army and then the king. The UML had little role through the 90s and was thus not to blame. This narrative lauds the party's 'balanced' policy during the peace process, where it adopted a policy of critical engagement with Maoists.

With this line, and a strong organisational base, the party thought it would be rewarded by the electorate. Post-April, there was disarray. Excuses were found, Maoist intimidation was held responsible and the leadership changed. Many in the UML still feel their line was correct but the tainted leadership was being punished for compromises during the 1990-2005 period.

Now voices arguing for greater proximity to the Maoists have asserted themselves. Bamdeb Gautam is the most active proponent of this line. Most of his cadre from the ML days went on to join the Maoists and his stated aim is to merge the UML and Maoists. Jhalanath Khanal may not go that far but believes in 'broader left unity'.

But those demanding greater distance from the Maoists have also gained strength. The conservative K P Oli school has been skeptical of the peace process and has viscerally opposed the Maoists. He draws his support from party workers who have often been thrashed by the Maoists and want to maintain a distinct identity.

If the UML has got the well deserved image of a flip-flop party, it is because of this internal struggle. And the man who has reconciled those interests, though personally inclined to the Oli view, is Madhav Nepal. He is still the decisive player in the party and will stage a comeback during the convention.

What has complicated the plot is that the line between the ideological battle and personal rivalry is blurred. Oli and Gautam may represent the two extremes, but they have come closer recently. Oli supported Gautam to head the UML team in government, while Khanal was rooting for Amrit Bohara. Oli now expects Gautam to reciprocate the support during the convention, when Khanal's leadership will be questioned. The result of that power struggle will determine the UML's approach to the Maoists, and thus the duration of this government.

Whether the UML can revive itself will depend on the ideological position it adopts, its effectiveness in government, the way the NC and Maoists evolve, the nature of polarisation in the next few years, and whether it can retain its grip on the lower levels of the bureaucracy.

The UML has to confront other challenges. It has dismally failed to deal with the Tarai's identity politics. The party structure is non-inclusive. Its leaders make prejudiced statements. It has lost its base among Madhesi intermediate castes and Dalits, and faces the danger of being reduced to an exclusively pahadi party.

It also needs to reassess its relationship with NGOs run by party sympathisers. These have been the UML's proxy patronage dispensation networks. But getting coopted into funding civil society has blunted the political edge of the cadre. UML workers and other sidekicks in 'civil society' are derisively dismissed as doing 'dollar ko kheti'. Nowhere do democracy, left politics, and civil society intersect as clearly as inside the UML. Its future decisions will have an impact in all three spheres.