The Taliban publicly stated on Saturday their commitment ‘to pay extra attention to not harming civilians’ during their upcoming spring offensive, just hours before deploying a 12-year old boy to kill himself and three other (civilians) in an attack on local government offices in the east of Afghanistan. The same day, as part of their UN-authorised campaign to employ ‘all necessary measures…to protect civilians’, NATO missiles struck a residential district in Tripoli, killing Col. Gaddafi’s youngest son and three of his grandchildren.

War, no matter on whose behalf it is fought or what its supposed justification, is never a particularly effective instrument for protecting civilians. Apart from the obvious difficulties of distinguishing civilians from combatants in the first place, and the fact that combatants will often deploy themselves near or even among civilian populations precisely for their own protection, there is also the troublesome fact that the ‘smartest’ bombs in the most advanced arsenals in the world are still not that smart. No matter who ‘wins’ a war, it is the civilians who invariably lose - they lose their families, their homes, their livelihoods, their communities, their dignity, and sometimes even their sanity.

There are still no reliable figures for the number of civilians killed in Libya since the NATO bombing began and of course no way of comparing that number with how many might have been killed had there been no such bombing. However if we really want to ensure the protection of civilians rather than regime change or some other more or less benign war objective, we should be examining more closely the tools and instruments available for protecting civilians and start moving away from the false dichotomy of doing nothing or going to war.

Mary Kaldor, in her article ‘Libya: war or humanitarian intervention?’, proposes as an alternative to the current NATO bombing a ground deployment of (military) peacekeepers and setting up of safe havens to protect civilians in Libya from further onslaught by Gaddafi forces. She foresees them needing a robust mandate and ongoing air support to ensure they can fight back if attacked and are not simply overrun as in Srebrenica or Rwanda, where peacekeeping troops signally failed to protect the civilians under their charge.

When peacekeeping troops on the ground have been successful at protecting civilians, however, it has generally been through a process of negotiating and bargaining with the various commanders and warlords who may be threatening a particular region or group of people. Rarely has brute force been effective in such circumstances, for the simple reason that engaging militarily with an armed group that is already fighting a war is unlikely to get that group to stop fighting. It merely forces armed groups to change tactics, exactly as they do when they are being attacked from the air. Of course the peacekeepers may be able to ‘win’ such a fight with the help of overwhelmingly superior force, but then they cease to be peacekeepers and become combatants in a conflict that is almost guaranteed to increase rather than decrease the number of civilian casualties.

Col. Bob Stewart, former commander of NATO troops in Bosnia, tells of a time when he was leading a column of tanks and was stopped at a Serb checkpoint that refused to let him through. With the ability to call in whatever extra support he needed from NATO, he had at his command at that time more firepower than any country has had since the dawn of civilisation, but if he used even a tiny fraction of that firepower to blast his way through the checkpoint, the repercussions – especially on civilians who were behind those lines – could have been huge and unpredictable. So instead he called the BBC and they began interviewing and questioning the Serbs at the checkpoint, who immediately opened the gates and let the NATO tanks through. In that case, it was publicity rather than firepower that got the gates open.

There is no dictator so ruthless or warlord so evil that they are not susceptible to the wide range of pressures and influences that can be brought to bear in these kinds of situations. Sometimes all it takes is the mere presence of somebody unexpectedly present. The presence of a journalist – even of a tourist – has been proven time and again to radically change the way Palestinians are treated at an Israeli checkpoint on the West Bank, for instance.

Sometimes raising an issue with the right person can get a change of policy or behaviour towards civilians in the area. Subtle diplomatic skills may be required to convince someone that it is in their interests to avoid damaging their own reputation and tarnishing their cause with attacks on, or abuse of, innocent civilians. And sometimes it requires a complex array of social and peer pressures from colleagues, political and moral pressures from their support base or from higher up the food chain, economic pressures from their paymasters or future clientele, diplomatic and legal pressures from the outside world and whatever else may be brought to bear on the situation.

Of course this kind of negotiation does not and cannot work in every case. When there are snipers shooting at everything that moves and taking orders from no one, as in the worst days of the siege of Sarajevo, nothing may protect civilians as effectively as a well placed bullet. However, in far more cases than the world has yet to realise, the key to protection of civilians is not a more robust use of force but actually the use of direct negotiation and quiet diplomacy. This is how civilians are being protected, right now, in war zones all over the world.

Take Mindanao in the Philippines, for instance, where 20 years of war has killed tens of thousands and displaced hundreds of thousands, mostly out of view of the world’s media. A fragile ceasefire is currently keeping the two sides from full-scale engagement with each other, but there are still skirmishes, hit and run attacks and unauthorised movements of men and equipment happening every day.

Part of the ceasefire agreement there involves the deployment of 86 unarmed ‘civilian protection monitors’ who work with local communities to identify violations of the ceasefire and try to prevent them from escalating into a full breakdown of the ceasefire. They do that by knowing the commanders on both sides by name and having their phone numbers on speed dial. When there is a problem, for instance an attack by one side on the other or a movement that creates panic in the local area, the civilian protection monitors are immediately on the phone with local commanders on both sides, with community leaders, and if necessary with people higher up the chain of command who can do something about it – and nine times out of ten this works to de-escalate the situation, prevent further violence and stop people fleeing their homes.

There is no magic to this. It is simple, straightforward dialogue and negotiation with the people who can control whether other people are safe or not. And it works. As soon as you bring guns, tanks and air support into the picture, you are talking about something quite else, which more often than not doesn’t work – except to ratchet up the hatred and the violence and get more civilians killed in the long run.

In the case of Mindanao, the civilian protection is carried out by a civil society organisation (Nonviolent Peaceforce) with no recourse to military force or even the threat of military force. In other cases, of course, one of the pressures that may convince a commander or warlord to back down may well be the implicit or even explicit threat of military force if all else fails. That is still more effective than actually using military force, and far preferable for all the reasons already given.

In the case of Libya, if the aim were purely and simply the protection of civilians, a first step would probably involve discussions with both pro-Gaddafi and rebel forces at a very local level regarding their understanding of and commitment to international humanitarian law and the normative rules of war like the Geneva Conventions. It would almost certainly involve a more detailed analysis of other factors that could influence their behaviour in particular circumstances, for instance Muslim prohibitions against killing fellow Muslims, kinship connections between the different parts of Libya, economic factors that could backfire on them, issues of duty, honour, pride that relate to the humane treatment of the enemy, protection of women and children, preservation of important cultural artefacts like mosques, etc.

Where these kinds of pressures and influences were not effective, the strategy would be to ratchet up the stakes involved, making sure that individual combatants and their commanders knew that the effects of their actions were being carefully watched and monitored, seeking out moderates and allies in both camps who had an understanding of the dangers and implications of ignoring such very basic rules of warfare as the distinction between combatants and non-combatants, using diplomatic and political pressures from external allies and sympathisers to press home the importance of playing by those rules...

Finally, a strategy for civilian protection in Libya would ultimately depend on a relationship with Gaddafi himself. While much has been made of Gaddafi’s threat on the eve of the UN Security Council resolution to ‘show no mercy’ to the rebels in Benghazi, he also called for a ceasefire within minutes of the resolution being passed and has a long history of yielding to political demands from the international community rather than invite further attacks or sanctions that might weaken his regime.

There is no knowing now whether he would have agreed to some kind of non-military arrangement for the protection of civilians if negotiations had been tried before the cruise missiles were launched. What we can know is that there are never only two options to choose between in such situations, and the sooner we explore and develop what some of those other options might be, the better able we will be to protect the civilians who need our help.