The 36^th round of peace talks between the Colombian government and the Colombian Revolutionary Armed Forces (FARC) opened last week in Havana, Cuba. The optimism from the last two rounds, when negotiators agreed on measures to deescalate the conflict, appears to be giving way to renewed skepticism about the ability of the two parties to deliver on their promises.

A Gallup poll released on April 29 revealed that the popularity of Colombian President Juan Manuel Santos, who has staked his presidency on achieving a peace deal, dropped from 43% to 29% in the past two months. A second poll over the weekend by Ipsos-Napoleón Franco confirmed the results.

Attitudes toward the peace talks showed a similar disenchantment. Those backing the process dropped from 69% to 52% in the Gallup poll, while the number of respondents who believe the FARC should be militarily defeated jumped from 25% to 42%. The 72% support Santos enjoyed for starting the talks two months ago has now dropped to 57%. These trends are problematic for the peace process, given Santos has promised any accord reached in Havana will be ratified by the Colombian population.

The polling results showed a parallel increase in support for former president Alvaro Uribe, who has been a staunch opponent of the peace process. Uribe’s support climbed 12 points to 59% in the Gallup poll and 16 points to 57% in the Ipsos-Franco poll. With congressional elections coming up in October, campaign season is well underway and the results may be a bellwether of things to come.

Peace’s poor showing in the polls reflects a political moment characterized by scandals and a strike in the judicial system, a badly managed teachers’ strike, economic worries, and growing perceptions of insecurity more broadly, particularly in urban areas. Perhaps most significant, however, is the national outrage and indignation over a FARC attack on a military encampment two weeks ago in the southwestern department of Cauca, which killed 11 soldiers and injured some two dozen more. At least one guerrilla was also killed, and the military presence in the area was heightened.

The FARC and the government agree the deaths are a tragedy, but both have avoided taking responsibility and demanded the other do so. Santos condemned what he called a ”vile” and “deliberate” attack that violated the unilateral ceasefire upheld by the FARC since last December. The president renewed the bombings of FARC camps suspended following the largely successful rounds of talks earlier this year. He called on the country’s armed forces to “unleash all the offensive actions necessary to protect the civilian population and… our troops.” Humberto de la Calle, the government’s lead negotiator, pronounced “the path of confidence and hope that had been built has been damaged.”

FARC leader Pastor Alape highlighted the “incoherence of the government to be ordering military operations against a guerrilla force that has declared a truce,” and insisted that the talks “should not be broken for any reason.” The FARC called on the international community and the Broad Front for Peace, a coalition of Colombia’s leftist parties, to investigate the case. The Broad Front’s final report called for further investigation both by the FARC and by an independent commission established by the negotiating parties.

Meanwhile, FARC peace delegation commander Pablo Catatumbo called the Cauca incident “regrettable,” and reiterated it was “a reaction of guerrilla units to the siege and harassment of the armed forces.” He underscored that the ceasefire does not exclude “defensive actions.”

It remains unclear whether the attack in Cauca was a defensive action as suggested by FARC leaders, a renegade attack by local FARC contingents, or an effort on the part of FARC to position itself for the debates in Havana, which relate to questions of truth, justice, and reparations.

What is clear is that the parties—albeit the FARC under protest—are experiencing the consequences of conducting a peace process without benefit of a bilateral ceasefire. Of 34 ceasefire agreements signed with armed groups around the world in 2014, only three, including the indefinite one declared by the FARC in December last year, have been unilateral. Both types of ceasefires present serious challenges in terms of verification, but a unilateral one undoubtedly incurs higher political costs for those who undertake it.

The existence of a declared ceasefire raises the expectations of the population that it will be respected and that rigorous monitoring and verification protocols will be enacted. When it is perceived to have been violated, trust in the process can be expected to plunge. Trust can, however, be recovered if the incident is well managed and used to move the process forward.

When a similar crisis erupted last November over the FARC detention of a high-ranking general in FARC-dominated territory in the Chocó department, the government suspended the talks and called on the international community to assist in finding a solution. The crisis was quickly resolved and the process surged ahead. At the time, everyone agreed it would be important to set up a mechanism for managing future crises, but this does not appear to have happened. It is not too late to still do so.

Despite the best intentions of those at the negotiating table, it is important to remember that each side will still have its share of individuals and groups looking to spoil the process. I have previously argued that the best approach is to try to co-opt or isolate these actors.

Implications for Peace

That the parties are back at the negotiating table after the recent violence is no small achievement. In past decades, similar attacks were sufficient to overturn peace talks in La Uribe, Caracas, Tlaxcala, and Caguán. This time everyone decried the bloodshed, but few called for a halt to the talks, though increasing numbers are seeking conditions and a deadline for reaching agreement.

Nonetheless, Bogota faces an ongoing credibility gap regarding the talks in Havana. The number of Gallup respondents who don’t believe the negotiators will reach a peace deal jumped from 44% to 56% in the past two months. The percentage of skeptics was even higher, at 69%, in the Ipsos-Franco poll. This is partly based on a mistaken perception that the talks are not going anywhere, when in fact all parties acknowledge consistent progress, and believe a final agreement is achievable.

The Colombian process compares favorably with peace processes elsewhere. It has seen only two minor interruptions that were quickly resolved, and the parties have stayed at the table to work toward a final agreement. Negotiators have met on a regular basis in Havana for 36 rounds, for just over two years, and appear to have built confidence that has resulted in three provisional agreements on agrarian development, political participation, and illicit crops and drugs, as well as an unprecedented joint demining project.

Winning the battle for public opinion will require a concerted effort to not only educate the public about what is being negotiated, but to make them stakeholders in the process. The engagement of victims, women’s groups, and LGBTI representatives in the peace talks in the past year was a positive innovation and provides a model for how to do this. Strategic engagement of members of other sectors such as indigenous, Afro-Colombians, youth, labor, press, military, religious, and police could help create and empower efforts to defend the peace process at home.

There is every indication a peace deal will still be reached. Implementation will be the greater challenge. The war has created a reservoir of distrust that is quickly tapped and not easily transformed. For the peace process to succeed, commitments must be made and honored. As government negotiator de la Calle said, “When hope is broken, it is time for faith…. Dialogue… is the instrument that can put an end to the war in the least painful, least prolonged, and above all, the most firm and enduring way.” Reciprocal, bold, and generous gestures are needed to keep peace on track.