Are Afghan negotiations hopeless? In the wake of last month’s assassination of Burhanuddin Rabbani, the head of the country’s High Peace Council, the mood in both Afghanistan and the United States is pessimistic, to say the least. But negotiations are still possible, and understanding why that’s the case, as well as the difficulties of succeeding, requires understanding the history of similar negotiations, quieting Afghan suspicions, and abandoning myths that cloud public discussion. Saying negotiations are possible is a long way from saying that they will necessarily succeed. But that does not mean they are not a sensible option after more than a quarter century of conflict.

The first step towards successful negotiations is the quelling of a number of popular myths about how they are supposed to proceed:

Myth: Negotiations are an alternative to fighting. This pernicious idea has no historical foundation. During the American Revolution, the end of the Algerian revolt against the French, and civil wars in Namibia, El Salvador, and Angola, fighting continued during negotiations and sometimes spiked as one side tried to break the will of the other. This is natural; the negotiations succeeded only when all the parties recognized that fighting could not achieve their maximum goals. The late Israeli Prime Minister Yitzhak Rabin understood this well when he observed that in negotiating with the Palestinians in the midst of terrorism, “We will continue the process as if there is no terror. And we will fight the terror as if there is no process.”

In Afghanistan, our frequent calls for negotiations may signal weakness. The result will be a toughening of Taliban terms, and the need to fight longer to prove we are serious. However, the idea that we should refuse negotiations until we have a stronger position on the battlefield is equally mistaken. Such a posture would cut us off from exploring what terms might be acceptable. It is also a political mistake because without achieving clear battlefield success we will begin negotiation while looking weak and, again, cause a stiffening of insurgent terms. We need to be open to talks while simultaneously not appear to be dependent on them.

Myth: Negotiations provide a rapid solution. If there is anything that is clear from the history of successful negotiations in such situations, it is that they take years to succeed. A long period of posturing often proceeds serious negotiations. This is the stage we seem to be at: talking about talking but not yet really engaged in serious proposals and counter-proposals.

Possible myth: The Taliban are not serious about negotiations. The recent assassination of former Afghan president Rabbani is now cited to suggest that negotiations are impossible. Here the issue is less one of myth than of jumping too quickly to one conclusion when several different ones are possible. The assassination does not fit normal Taliban patterns in which spectacular actions are closely coordinated with a media campaign. In this case the Taliban spokesman has waffled, first claiming the attack, then denying it, then saying it is being “investigated;” all signs of confusion and lack of ownership. The attack is politically embarrassing for the Taliban in the immediate aftermath of a public statement from leader Mullah Omar suggesting, without much specificity, new flexibility toward negotiations. Nor is there any gain for the Taliban in the killing; they can refuse negotiations or stall without perpetrating an assassination that may well harden the will of others to resist them.

There are several other spoilers who would have had reasons for the murder. Al Qaeda is a prime suspect. Western leaders have set a clear condition for a negotiated settlement—the Taliban must break with Al Qaeda. However, there is no proof of Al Qaeda responsibility and other candidates are possible. One is the Haqqani network. Their appetite for negotiations is unclear. Pakistan’s intelligence service, the ISI, could also be sending a message that negotiations can only proceed with their good graces. Even Afghan domestic elements that fear negotiations could be responsible. What the assassination does suggest is that there are divisions within the insurgency that will add to the difficulties of negotiations.

Having dispatched of these popular myths, what should we do? Here are three of the most important things the U.S. must pursue:

First, we should continue negotiations while ridding ourselves of illusions that they offer fast or easy alternatives to war. In fact, the greatest contribution the United States could make toward the possibility of successful negotiations would be to make clear that we will persevere in fighting, and then support, over the long term, an Afghan army tough enough not to lose. The current absence of clarity on just this point is a continuing reason for the Taliban to hold back, and to see whether our will may break.

Second, the U.S. can negotiate but only Afghans can maintain a peace. It is imperative that we and the Afghan government negotiate as one body. Only in this way can we keep the talks focused on real issues rather than Taliban efforts to exploit the divisions between ourselves and Karzai. Close political/military coordination is essential so that, for example, we do not kill people we are talking to.

Third, there is enormous suspicion by many Afghan minorities, civil society groups, and political liberals of negotiations led only by President Karzai and warlord supporters. This disparate group fears that, faced with an imminent and premature NATO withdrawal, Karzai will cut a deal to hold office that essentially brings the Taliban back into control. Afghans have seen too many agreements that were broken. Rather than submit to such an agreement or even one that looks like it could bring the Taliban back to power, many opponents have told me they would restart a civil war. Only the U.S. can give the necessary assurances that key civil and political rights will not be sacrificed. We can help preserve space for the growth of civil, gender, and political rights, but only if we are clear about our long term intention to remain involved; this clarity is currently absent, especially in light of our announced date of turnover of the security lead to Afghans in 2014.

History suggests that the path to a negotiated end to an insurgency is long and difficult, that it requires fighting while talking, and that consistency of purpose is essential. There are no guarantees. Negotiations take all sides to agree. Given divisions in the insurgency and the interests of Pakistan, agreement may be elusive. But it is likewise too early to decide that negotiations are futile. We need to stop agonizing and keep trying, even as we keep fighting. After all, what have we to lose by doing so?

Ronald Neumann is the President of the American Academy of Diplomacy but expresses only his own opinion. He served as ambassador to Algeria, Bahrain, and Afghanistan (2005-07) and is the author of The Other War: Winning and Losing in Afghanistan.