It is easy to be skeptical about the latest news on the Afghan-negotiation front: a “preliminary agreement” for the Taliban to open an office in Qatar. This is, after all, merely a procedural step that will give the Taliban a fixed address and identifiable interlocutor for negotiating purposes. It says nothing about substantive issues involving the future political structure of Afghanistan. Nor does it point to any Taliban intentions or bottom lines. Some of what the Taliban said when making the announcement about the new office suggested their overriding interest is in getting NATO forces out of Afghanistan.

Modest though this step is, however, it is just the kind of basic procedural matter that has often been one of the biggest hurdles to overcome in peace negotiations. The recent experience in Afghanistan itself shows how big it can be. There was the episode more than a year ago of dealing with a supposed senior Taliban leader who instead turned out to be an imposter. Then last September there was the death of former president and head of the Afghan peace council Burhanuddin Rabbani, who met an ostensible Taliban peace negotiator who instead was a suicide bomber with an explosive hidden in his turban.

Such setbacks, or the dangers associated with them, are not unique to conflict in Afghanistan. When Charles de Gaulle was trying in 1960 to negotiate an end to the Algerian war that had been underway for six years, he granted an interview to three rebel chiefs who said they were interested in making a separate peace. In doing so, de Gaulle was assuming personal risk comparable to the risk that Rabbani unluckily took on last year. The Algerian rebels were not subjected to a body search because that would have destroyed the climate of trust and confidence that the meeting was intended to create (although a security man armed with a submachine gun was concealed behind a curtain in the presidential office). In negotiations to end the Vietnam War, it took a year just to agree on the shape of the conference table (although the negotiations that really mattered would be conducted secretly in parallel by Le Duc Tho and Henry Kissinger).

Several reasons make peace negotiations difficult to get under way. One is a fear of showing weakness. That not only makes belligerents reluctant to make the first move; it also encourages them to take an especially hard substantive line at the outset. Some of that may be reflected in the remarks this week by the Taliban spokesman, who seemed to be interested only in hastening a NATO withdrawal. Another reason is the distrust that accumulates during the war. Overcoming that distrust was what the risk-taking by de Gaulle and Rabbani was all about. Yet another hurdle in insurgencies is that it is impossible to separate questions of the status and legitimacy of interlocutors in a peace negotiation from substantive issues to be negotiated. The reason it took so long to agree on the shape of the furniture in the Vietnam talks was that the status of the Viet Cong and whether it had an existence separate from North Vietnam was an issue both of standing to negotiate and of how the war ought to be settled. Similarly, the status of the Taliban as negotiators cannot be separated from questions of what role they should be permitted to play in a new Afghan political order.

Negotiations on Afghanistan will be at least as difficult as those that settled some of those earlier conflicts. But negotiations are necessary and inevitable. No one is going to win this war. Imparting a modicum of stability to Afghanistan will require many bargains to be struck, many of which will involve the Taliban.

As preliminary talks turn into fuller negotiations, there will be other hurdles to overcome. One is suggested by some of the commentary about the opening of the office in Qatar—to the effect that this might help to minimize Pakistani influence. Trying to constrain Pakistan's role in this whole process would be a mistake. The Pakistanis are perfectly positioned to be spoilers. As Lyndon Johnson once remarked about retaining J. Edgar Hoover in his administration, it is better to have him inside the tent [urinating] out than to have him outside the tent doing the opposite. And once you involve the Pakistanis you necessarily have to involve the Indians and other regional players, whose involvement is necessary anyway regardless of how much more complicated this makes the whole process.