Getting to ‘not yet’: Colombia’s Peace Agreement Referendum

This blog post, by Prof Christine Bell, PSRP Programme Director, addresses the result of the peace agreement referendum that took place in Colombia, last Sunday, 2 October 2016. This is the third of a series of blog posts related to the Colombian Peace Process.

A referendum on the Colombian Peace Agreement between the Government and the FARC, appears to have failed to pass on Sunday 2 October 2016.

According to reports the no vote led by 50.2% to 49.8%, a difference of fewer than 54 000 votes out of almost 13 million cast, with only 38% of the electorate casting a vote, although there are suggestions that a recount is in order.

Nonetheless, this is a devastating blow to the Colombian peace process, which – wisely or not – had conditioned further progress on a successful referendum, a result also required by a ruling of the Constitutional Court.

Are referenda on peace agreements normal?

How often, are peace agreements conditioned on a referendum?

As far as peace agreements go, there is no ‘normal’.  Each one has its own rhythm.  However, referenda on peace agreements or to bring in ‘peace agreement constitutions’ have been used in other peace processes, notably: Cyprus, Democratic Republic of Congo, Guatemala, Indonesia/East Timor, Iraq, Northern Ireland, Somalia (see further data base on the issue).  The Northern Ireland Agreement was probably most similar to the Colombian one (although the referendum was there a cross-border one, relating to the territorial nature of the conflict).  Historically, an important and similar referendum on the transitional constitution in Spain post Franco, both enabled the transition but created enduring controversy with Basque and Catalan separatists.

Referenda on complex political arrangements that involve packages rather than clear ‘yes’ or ‘no’ propositions are notoriously high risk (see recent ‘Brexit’ vote in UK).  In the Northern Irish peace process it was a successful strategy – but only just.  While 71.1% voted for the agreement this included only a small majority of the Unionist/Protestant community, without whose consent the agreement would have been undermined.

In other contexts, failed referenda have been fairly devastating and finished otherwise workable peace processes with developed agreements reached between the main parties to the conflict.  Referenda on constitutional reform pursuant to the peace agreement failed to carry the reforms in Guatemala; a referenda ended the peace process when a final accord was not carried in Cyprus; and a ‘Serb only’ referenda failure in Bosnia delaying settlement of that conflict by some years.

Are referenda a good idea in peace processes?

Why then do people think referenda are a good idea for adopting peace agreements?  Peace agreements are rarely just about ceasefires.  Like the Colombian one, they often have constitutional dimensions and implications. Yet these agreements are agreed primarily between those who have been at the heart of the conflict including – as in this case – armed actors.  The need for some sort of broader public legitimacy can therefore be very high, and a referenda can seem like a useful tool for achieving it.

The clear risk is that with so many provisions and issues dealt with, people who are opposed to just one dimension of the peace agreement can end up voting against it.  In Northern Ireland the argument that people should not ‘cherry-pick’ the agreement was a strong campaign message for those who sought a ‘yes’ vote and was largely accepted by the voters.

The Colombian peace agreement at nearly 300 pages long (297), is the longest peace agreement produced in any intra-state conflict.  Within those 297 pages is the longest and most complex provision on victims’ rights and accountability (68 pages).  It is likely that most of those who voted for the agreement had not read all of it, or even its portions on dealing with the past. Contrary to Colombian press reports it did not provide a simple amnesty for FARC.  In fact the detail of the provision on the past reflects the complexity of the  issue and the nuanced way it was dealt with to ensure both accountability and capacity to end the conflict.  The provision’s complexity was due to the various factors needing addressed.  These included the need to work within the shadow the International Criminal Court, who would not have permitted a blanket amnesty; a need to reckon with the complexity of past amnesty processes agreed with other armed groups with which this process had to be squared; and the complex business of reconciling multiple victims’ needs, all within some sort of clearly set out and workable modalities.

How exactly it would have all worked out, could not have been judged until the implementation stage.  However, this type of detail goes missing in public discussion on ‘the agreement: yes or no’, particularly where debate is rushed. In fact the detail itself may have created opposition as legalistic detail in Colombia often obscures what will actually transpire in implementation and is distrusted – a distrust that has a basis.

Was the referenda handled as well as it could have been?

The Government had a difficult call to make – whether to go to a vote fast and try to secure a referenda quickly before any opposition to the agreement had time to mobilise and enthusiasm was at its height; or whether to move more slowly and build a campaign which could address doubts.

Opting for the former now looks like a miscalculation, but the benefit of hindsight is a wonderful thing.  In fact, the peace process itself had timetables for implementation built in as necessary to the parties reaching agreement which had set the pace.  And more time is not always a good thing where cementing of a peace agreement is concerned.  Sometimes a long campaign fuels discontent which moves the public away from, rather than towards peace.  Delaying a referendum can open up a window in which ‘spoilers’ to a peace agreement are incentivised to undertake a spectacularly violent acts, in an attempt to undermine public trust (think the real IRA’s Omagh bomb in Northern Ireland, a few months after the referendum, which killed 28 people in an attempt to undermine the peace process).  These concerns must have been uppermost in everyone’s minds.

Where to from here?

A large factor in the narrow ‘no’ majority was undoubtedly a Colombian weariness and scepticism over peace agreements.  The Colombian peace processes, depending on how one counts, has had 117 jointly signed agreements between armed actors and the government over the last 25 years: 68 pre-negotiation agreements outlining principles and modalities for negotiations aimed at enabling talks; 38 providing for specific substantive commitments aiming to achieve peace; and 9 providing for implementation or renewal of previous commitments.  These agreements have arisen over what can be regarded as 5 distinct phases of the peace process (see further Bell, O’Rourke & Matzner 2015).  Against such a context perhaps campaigners for ‘yes’ needed to explain better, not just this peace deal but why, when so many agreements with one group have failed to end the conflict as a whole, this peace agreement was going to fare any better.

The best answer was that this 25 year ‘slow drip’ of a peace process was indeed inching towards a form of completion with this agreement. This stage of the process had been innovative and participative, and the agreement was more holistic than any of the agreements which came before in addressing the many layers of conflict and violent crime, and in squaring the need to move from conflict with a measure of accountability.

Ironically, it has been a feature of past failed Colombian peace processes that agreements were not signed, or signed and not robustly implemented, because parties to the conflict and society at large felt that time was on their side, and having squeezed to the max, they could step back and wait so as to get a better deal next time around.  The costly repeated lesson has been that after each failure the conflict became more difficult to solve, not easier, with more layers, more armed groups, more geopolitical factors and more egregious violence than before.

With the only plan ‘b’ of those who opposed this deal (notably Uribe, who had two terms as President in which to bring peace), that of ‘renegotiating a better deal’, it would seem that this lesson has not been learned.

The biggest majority in Colombia yesterday were those who did not vote.  That must have felt like a vote for apathy: a signal that people were a bit fed-up with politicians and armed groups and wanted to signal ‘a plague on both your houses’.

However, those people now have to live with the fact that staying away involved exercising a choice.  To stay away was to exercise a choice to end this process in this form, and send those politicians and armed groups back to the same drawing board that people are frustrated they are still at, and from which they had emerged with an honourable compromise which had a real chance.

About the Author: Professor Christine Bell is Assistant Principal (Global Justice), Co-Director, Global Justice Academy, and Professor of Constitutional Law, School of Law, University of Edinburgh. Her research interests lie in the interface between constitutional and international law, gender and conflict, and legal theory, with a particular interest in peace processes and their agreements. She is the Principal Director of the Political Settlements Research Programme.

Photo: A small rally in favour of the Colombian peace process, by Peter Angritt – Creative Commons