home       biography       publications       speeches       organisations       images       @contact

Preventing Deadly Conflict: The Role And Responsibility Of Governments And NGOs"

Public Lecture by Hon Gareth Evans AO QC, President, International Crisis Group, for the Centre for Study of Human Rights, London School of Economics, 2 February 2001

The Cost of Conflict

Everyone agrees that it's a lot less costly to prevent conflict than to react to it after the event. But very few policy makers ever seem to take the force of this observation seriously enough to do something about it. Maybe they would if they stopped now and again to do some calculations.

I tried an exercise of this kind in the early 90s, to fairly startling effect. I tasked the Australian Foreign Ministry to put together an estimate, from all available information, public and private, of what it cost the allies to fight the 1991 Gulf War - just the allies and just the dollar cost, not even beginning to take into account the massive human and physical destruction costs on the Iraqi side, or reconstruction costs in Kuwait.

Also, given that Iraq's invasion of Kuwait the previous August had represented a rather spectacular failure of early warning and preventive diplomacy by regional countries, the major powers and the UN system, I asked them at the same time to make a guesstimate of what it might have cost to have had in place at the time some more orderly and focused diplomatic machinery for conflict prevention - not on the naïve assumption that any such machinery would certainly have made a difference, but on the basis that something like it just might have.

The results were instructive. The cost of establishing under UN auspices six preventive diplomacy units in six regional centres around the world, with total professional staff of 100 and plenty of funds for travel, would have been at the time around $21 million a year. The cost, by contrast, of waging the six-week Gulf War was for the US alone $61 billion, and for the allied coalition as a whole, $70 billion. (It was also instructive to note that at the time we did this exercise, in 1993, the budget for all UN peace keeping operations combined was $3.7 billion, and the regular budget for the whole UN was just over $1 billion.)

The cost of policy failure of course is not just dollars but lives, and untold human misery. The Swedish peace research institute, SIPRI, estimates at 1.4 million the number of battle deaths just in armed conflicts between states in the ten years since the end of the Cold War.1 But most of the conflicts waged around the world in the last decade have been within rather than between states, and no credible aggregate estimate has been made by anyone of the total deaths there. We simply know that the numbers have beenstaggering, with over 800,000 slaughtered in the 1994 Rwandan genocide alone, and another 145 000 in Bosnia, including 7000 at Srebrenica in the worst mass murder in Europe since the Nazi Holocaust.

Add to those killed the number of rape victims, up to 50, 000 in Bosnia and up to 500,000 in Rwanda. And add again the number of people displaced or made refugees in conflicts around the world just in the last ten years: 18.5 million, according to the UNHCR's latest figures. Every single one of those lives lost, or lives wrecked, represents a failure of conflict prevention.

The Problem of Intervention

When prevention does fail, and deadly conflict erupts, there are very few policy options left to the international community. Let me spend a little time - before we come back to prevention - exploring what they are.

One very traditional option - employed periodically through the Cold War years , for example in Cyprus, and as recently as the Ethiopia-Eritrea war last year - is for diplomatic peace making efforts to be initiated, and if the combatants are susceptible to rational argument (or maybe less subtle forms of diplomatic persuasion), and can be brought to agree at least on a ceasefire, that can be followed by the deployment of a peace keeping mission to monitor, supervise and verify the peace thus agreed.

The trouble is that in the world of the 1990s and beyond - with most of the conflicts internal in character, with less well defined belligerents, less capacity for coercive but non-military measures like sanctions to have an impact, less capacity to negotiate anything, and even less capacity to make a negotiated agreement stick - that traditional line of approach has not, most of the time, seemed very relevant. As Brian Urquhart has recently put it:

So long as they scent possible victory, regional warlords, embattled ethnic leaders, or politically ambitious thugs are seldom susceptible to legal arguments, international pressures or humanitarian appeals.

Although the end of the Cold War actually made some long-term civil conflicts easier to settle - notably in Cambodia - it also meant the end - most obviously in the Balkans - of the often artificial cohesion that the superpowers had forced, for better or worse, upon the member countries of the blocs that owed them allegiance, and major new violence erupted accordingly. This coincided with the spread of ever more graphic instant international television coverage, forcing viewers in major countries to confront images of human suffering in far away places that previous generations had been able to ignore: the so-called CNN factor.

These things together created both the occasion and the demand for governments, and in particular the UN, to do something. And that something - it seemed it had to be, when prevention failed, and other options were not obvious - was military intervention.

And so, during the 1990s, we have seen the emergence (or some would say re-emergence, because interventionist doctrines in one form or another have been around for a long time) of the doctrine of the right of humanitarian intervention. The core idea is that if human rights - and in particular the most basic of them all, the right to live - are being violated, or are about to be violated, on a massive scale within a country, then there is a moral and legal right for others to forcibly intervene in that country, without its consent and irrespective of its claims to sovereignty, in order to stop that violation occurring.

And military intervention, for humanitarian or human rights protecting motives, is exactly what has happened - with or without Security Council endorsement - throughout the 1990s, over and again - in Somalia for example, in Northern Iraq, Liberia, Bosnia, Haiti, Rwanda , Sierra Leone and in Kosovo. (East Timor is a marginal case, because Indonesia's consent, although unquestionably the result of heavy international pressure, was essentially genuine - and the Australian-led intervention would not have occurred without it.)

The trouble is that - as a result of the way in which that right has been claimed and actually exercised on too many occasions throughout the 1990s - the initial consensus about the permissibility of military intervention in extreme cases, which seemed to be developing not only among lesser mortals but in the Security Council itself, seems to have evaporated. It has given way to endless, and increasingly bitter, disputation as to whether there is such a right, and if so when and how it should be exercised.

The basic reason for the shift in attitude, which was gathering momentum before Kosovo, has been clear enough: so many of these interventions have been too little too late, misconceived, poorly resourced, poorly executed - or all of the above. The generally acknowledged success of the military response to Iraq in the 1991 Gulf War (a straightforward case of UN-Charter authorised peace enforcement in response to cross-border aggression), and of more or less traditional peace keeping operations such as those in Cambodia, Namibia and Mozambique in the early 1990s, gave way to the debacle of the international intervention in Somalia in 1993; the pathetically inadequate response to genocide in Rwanda in 1994; and the utter inability of the UN presence to prevent the murderous 'ethnic cleansing' in Srebrenica in 1995.

NATO's intervention in Kosovo in 1999 brought matters to an intensely controversial head. Security Council members were divided; the legal justification for military action without new Security Council authority was asserted but largely unargued; the moral or humanitarian justification for the action, on the face of it much stronger, was clouded by allegations that the intervention triggered more carnage than it averted; and the means by which the NATO allies waged the war continue, justly or not, to be much criticised.

The result of all this is that the so-called right to humanitarian intervention, and the sea of questions that swirl round it, is probably the single most controversial unresolved foreign policy issue of the 1990s. Questions like whether it exists legally; when it arises morally; whether Security Council authority is necessary or sufficient; in what circumstances it should be exercised; how it should be exercised; and what are the political conditions that might justify and enable its exercise, all seem as far away as ever from being resolved as we enter the new century.

The central question was starkly posed by UN Secretary-General Kofi Annan in his 1999 speech to the General Assembly4; and put again in his 2000 Millennium Report5 in these terms:

…if humanitarian intervention is, indeed, an unacceptable assault on sovereignty, how should we respond to a Rwanda, to a Srebrenica - to gross and systematic violations of human rights that affect every precept of our common humanity?

He partly answered his own question by going on to say:

But surely no legal principle - not even sovereignty - can ever shield crimes against humanity…Armed intervention must always remain the option of last resort, but in the face of mass murder it is an option that cannot be relinquished.

The trouble is that the Secretary General didn’t quite bring his audience with him. The debate that followed his initial speech to the General Assembly, and which has continued in these terms ever since, saw fervent supporters of intervention, and anxious defenders of sovereignty, dig themselves into two deep trenches from which they have not yet emerged. The supporters of intervention focused single-mindedly on the humanitarian or human rights imperative; the defenders of state sovereignty, while not indifferent to claims about the need to redress human suffering, worried what pig-in-a-poke powers they would be giving their former colonial masters to abuse this time round.

It isn't my primary purpose today to talk about this intervention versus state sovereignty issue as such -but it does bear upon my main theme of conflict prevention in a way that I shall explain. Because the issues are interconnected, I want to say a little about an ambitious exercise has recently been set in train, in which I personally am much involved, to try and find a way for the protagonists in this debate to dig their way out of their trenches.

The Canadian Government and a number of major foundations announced at the General Assembly last September the establishment of a new International Commission on Intervention and State Sovereignty (ICISS), co-chaired by Mohamed Sahnoun of Algeria and me.6 We were asked to wrestle with the whole range of issues rolled up in this debate - legal, moral, operational and political - and to bring back a report within a year that would help the Secretary General and everyone else find some new common ground.

That will not be an easy task: in fact a Latin American colleague described it to me yesterday, in rather a surreal turn of phrase, as like trying to milk an ant with boxing gloves! But I think there are at least four good reasons why the new Commission may be able to add value to what has become a rather sterile and set piece debate.

First, because it is the most representative and consultative exercise yet attempted in this area. The Commission is evenly divided between North and South in its composition. From the South there is Mohamed Sahnoun, Fidel Ramos, Cyril Ramaphosa, Eduardo Stein and Ramesh Thakur; from the North, in addition to me, Lee Hamilton, Gisele Cote-Harper, Michael Ignatieff and Klaus Naumann; with, in addition, Vladimir Lukin from Russia, and the former head of the ICRC Cornelio Sommaruga, whom our Chinese friends might describe as a Northener with Southern characteristics. The Commission will meet in Asia and Africa as well as North America and Europe, and hold roundtables and other consultations in Latin America, the Middle East, Russia and possibly China as well. Our ears are open.

Second, because the exercise will be comprehensive in scope - addressing not just the legal and moral dilemmas which have been at the heart of most of the academic and policy debate about coercive intervention so far, but operational and political issues as well. It will try and take into account and build upon all the best thinking and writing that has been done on the issue around the world so far, with our Report planned to have two accompanying volumes, one of newly commissioned research, and the other an anthology and annotated bibliography of previous writing.

Third, because the whole exercise will have a sharply practical political focus. None of us want to see the report disappearing from sight soon after its release, having no other life than in LSE classrooms… To avoid that fate, it has to become part of serious inter-government debate, and that needs a significant government to pick up the ball and run with it. That is the role that Canada plans to play, and it has already been active in starting to build a coalition of the interested to carry the debate forward, hopefully with the Secreary General himself playing a major part in the enterprise.

Fourth, we hope the new Commission's report will add value by being innovative - bringing some genuinely new ways of thinking about the issue into the debate. We have already agreed to drop some of the terminological baggage the debate has been carrying, by using the neutral expression "intervention" rather than the loaded one "humanitarian intervention" , which latter has tended to enrage both humanitarian relief organisations, who don't like the H-word being used to describe what may be a military intervention, and also many of those on the pro-sovereignty side of the argument, who think the H- word is a 'hurrah' one which stacks the cards in favour of intervention in a given case before the argument has even started.

The Commission's most important single contribution, however, may be to reconceptualise the core concept of the right to intervene. We have in mind - and are now testing this in all our consulations - trying to turn the way in which we all think about the issue upside down, so that we talk not about rights but responsibilities, and not about intervention but protection - with "the right to intervene" becoming rather "the responsibility to protect". This immediately puts the focus where, arguably, it always ought to be - not on the countries throwing their weight around, for better or worse, but on the victims of conflict, who need the assistance of others if they are to be protected from suffering.

One of the virtues, perhaps, of looking at the issue this way - and now I return to the main theme of this lecture - is that it brings the need for conflict prevention to centre stage. The "responsibility to protect" obviously embraces, with equal clarity and force, not only the responsibility to respond effectively if suffering is occurring or threatened but also the prior responsibility to prevent suffering . (Perhaps it embraces a third element as well, the responsibility to follow through, mend and reconstruct after suffering has occurred - but that's a discussion for another day.)

We have in the past tended to think of prevention and intervention as being quite separate and distinct concepts, connected only by chronological sequence, with intervention being triggered as an option after prevention failed (if indeed it was ever tried). What talk about the responsibility to protect might conceivably do is establish not just a chronological but a substantive connection - the main normative consequence being that governments and intergovernmental organisations would need to take much more seriously, and treat as a core obligation rather than as a marginal afterthought, the responsibility to act to prevent deadly conflict.

And that, for all the reasons I stated at the outset, would be a Very Good Thing. Coercive intervention after the event is always going to be controversial, always going to be hard to mobilise, and as likely as not to be too little too late anyway. Prevention, with conflict as with so much else, beats the hell out of cure. But all that said - how do we make prevention actually work?

Making Prevention Work

For effective conflict prevention, three essential conditions have to be met. There has to be knowledge of the fragility of the situation, and the risk of impending conflict - so called "early warning". There has to be a set of appropriate and policy measures available that are capable of making a difference - the so called "preventive toolbox". And there has to be the willingness to apply those measures - the issue of "political will".

Early Warning. So far as early warning is concerned, there is no doubt that governments and intergovernmental organisations need all the help that they can get. As budgets for foreign offices around the world tend to go on shrinking - in my view, and I say this not only as a member of the foreign ministers' club, bizarrely and indefensibly - there seems to be, even in large services like America's and Britain's, less capacity than ever to track and monitor fragile situations and crises and conflicts in the making. There is a role, as a result, to be usefully played by non-government players including the media and, especially, NGOs - a theme to which I will return before closing.

All that said, it is easy to exaggerate the extent to which lack of early warning is a serious problem. My own suspicion - and the recent pathbreaking reports on the Rwanda tragedy certainly support this - is that far too often it is an excuse rather than an explanation, and that the problem is not lack of warning but lack of timely response. Most of the time key governments, the UN and certainly the major regional organisations, know perfectly well that there is a problem looming: they just hope that it will go away; or even if they know that it won't, they don't get around to doing anything about it because it is an iron law of both politics and the bureacracy that the urgent always drives out the important.

The Prevention Toolbox. If governments and intergovernmental organisations are minded to do something about conflict prevention, what should they do? What are the measures available in the preventive toolbox? To explore this issue in any useful detail would involve looking at much more than we have time for now - the different causes of conflict - greed, grievance or something else - to which those tools may be responsive; what kinds of measures may be appropriate when; and how those measures might most usefully be operationalised. All I can really do now is hint at the breadth and complexity of the options available by summarising under two headings - structural measures and direct measures - the most familiar preventive options available.7

Structural Measures

Political. Democratic institution and capacity building; constitutional power sharing and redistribution arrangements; confidence building measures between different community groups; funding and training support for democratic opposition groups.

Economic. Development assistance and cooperation to address inequitable distribution of resources; promotion of economic growth and opportunity; aid made conditional on structural reform.

Legal. Building effective and non-corrupt judicial and police systems; formal protection of minority rights; establishment of local human rights commissions.

Military. Reform, professionalisation and civilian control of the military; encourage adherence to arms control and disarmament regimes.

Direct Measures

Political/Diplomatic. Fact finding missions; dialogue and mediation; international appeals - from subtle and encouraging to 'naming and shaming'; threat of political sanctions - diplomatic isolation, suspension of organisation membership; targeted personal sanctions - visas, bank accounts ; non-official 'second track' dialogue and problem-solving workshops.

Economic. Threat of trade and financial sanctions; withdrawal of investment; threat of withdrawal of IMF or World Bank support; aid curtailment.

Legal. Mediation; arbitration; adjudication; threat of application of international criminal process.

Military. Preventive deployment; threat of use of military force.

This list still only scratches the surface of discussing the range of options available - not only at the stage of preventing the initial outbreak of conflict, but also at the later stages of preventing its escalation and preventing its recurrence, bearing in mind that in these later stages of prevention, many of the techniques available are indistinguishable from those involved in conflict resolution.

The question on which I do feel it is worth spending a little more time on here - both because it is so important and because it is so often sped over - is the third condition for effective conflict prevention, the mobilisation of political will.

Mobilising Political Will

The difficulty with most discussions of political will is that we spend more time lamenting its absence than analysing what it means. We tend to talk about it as a single missing ingredient - the gelatine without which the dish won’t set.

The trouble with this metaphor or any other way of thinking about “political will” as a single, simple factor in the equation is that it understates the sheer complexity of what is involved. To mobilize political will doesn’t mean just finding that elusive packet of gelatine, but rather working your way through a whole cupboard-full of further ingredients. What then are those ingredients?

Institutional Measures. In the first place, some of them are institutional. There is a great deal to be said, in any organisational setting, to have some organisation - i.e. in this context, some institutionalisation and routinisation of prevention, someone or some group within the system whose responsibility it is to think about prevention, and devise and recommend up the decision-making food chain appropriate policy responses. Another way of putting it is in terms of deliberately trying to build, through more focused arrangements of this kind, a "culture of prevention" .There have been some encouraging recent developments of this kind in a number of foreign affairs and development bureaucracies, including in Britain.

More substantial organisational innovations are possible, for example the possible regional preventive diplomacy centres I spoke about at the outset. Probably the most successful example of the approach of creating a new organisation on the principle that there is thereby a better chance of its functions being employed, has been the OSCE's appointment of a High Commissioner on National Minorities. Max Van der Stoel's full time application over many years to the task of devising and encouraging preventive measures to stop ethnic communities tearing at each other's throats in many parts of Europe has been hugely more effective than a series of case-by-case diplomatic excursions is likely to have been.

If you are going to get things happening in political and governmental systems, overcoming the inbuilt inertia which seems to inflict them all, the other crucial institutional element is some decision-forcing mechanism - something to bring the issue to the top of the decision-maker's in-tray. In the case of conflict prevention, there is nothing as readily implementable as for example the agenda forcing associated with the annual budget process, though regularly timetabled report and review sessions can work with the occasional idiosyncratic minister who happens to possess a disciplined cast of mind.

The best attention getter and decision forcer, apart from a real crisis itself, is likely to be media attention - and it is always a prime objective of those outside government, including NGOs, to concentrate decision making minds in this way.

The Right Arguments for the Right People. Institutional innovations will only take far in mobilising the political will to get something actually done about prevention. From my own experience both in government and beating on the doors of government, you just have to recognise that there are key individuals at or near the top of the food chains, and you have to at the end of the day find good arguments that will appeal to them. The well-equipped political- will- mobiliser needs to be equipped, as I see it, with five different kinds of argument in favour of preventive action.

Moral Argument. However base and self-interested their actual motives may be, governments always like to be seen - - both internally and internationally - - as acting from higher motives. Preventing human suffering, and all the catastrophic loss and misery associated with deadly conflict, can act as both an inspiring and a legitimising motive in almost any political context. Getting this moral motive to bite means, however, being able to convey a sense of urgency and reality about the threat to human life involved in a particular situation - - always difficult when there’s no blood or amputees for CNN to film.

Financial Argument. The best financial argument for preventive action is that it is likely to be cheaper, by many orders of magnitude, than responding after the event - whether through military action, humanitarian relief assistance, post-conflict reconstruction, or all three.

Part of the credibility of the financial argument depends on being able to make the case that if the situation does erupt, then there will be no choice but to react in other ways, including ultimately militarily. That is always an argument that has had credibility in the Balkan.. Elsewhere, where the countries that can make a difference have a less well developed sense of responsibility - - as with a crisis almost anywhere in Africa - the financial cost saving argument for prevention is a little less strong. But there will always be pressure on the major developed countries to contribute to organisations like UNHCR as they try to clean up the mess, so this argument generally has at least some resonance.

National Interest Argument. Preventive action can often serve a country’s national interests as very narrowly and traditionally defined in security and economic terms. Avoiding the disintegration of a neighbour, with the refugee outflows and general regional security destabilization associated with it can be a compelling motive in many contexts. National economic interests can often be equally well served - by keeping resource supply lines, trade routes and markets undisrupted. Whatever may have been the case in the past, these days peace is generally regarded as much better for business than war.

There is another dimension of the national interest which is highly relevant here. It is what I have been arguing for some years is every country's national interest in being, and being seen to be, a good international citizen. There is a lot of direct reciprocal benefit to be gained in an interdependent, globalised world where nobody can solve all their own problems: my assistance for you today in solving your drugs and terrorism problem might reasonably lead you to be more willing to help solve my environmental problem tomorrow. The interest in being seen to be a good international citizen is simply the reputational benefit that a country can win for itself, over time, by being regularly willing to pitch into international tasks for motives that appear to be relatively self-less.

Domestic Politics Argument. Making an argument that will address domestic political concerns is a subtler business than just calculating what the majority reaction will be. Governments - even those directly dependent for their support on the ballot box - often do things without knowing what is the majority view, and even when they know that the majority sentiment might be against the proposed action. What matters more is that they have arguments that will appeal to, or at least not alienate, their own political support base; and that they have arguments that they can use to deflate, or at least defend against, the attacks of their political opponents.

A good deal of the political debate, when it comes to preventive measures, is really shadow play, in which it’s enough that the government be seen to be acting competently and credibly. As often as not, there won’t be a majority reaction to have to take into account, and if there is one, it will be muted. This is because few issues, when it comes to prevention, are really likely to have strong political salience, in the sense of directly winning or losing a government enough votes to be relevant to its survival.

The kind of arguments a government will find most useful in dealing with its supporters, and its opponents, will largely depend on what kind of government it is. Governments of a left-of-centre persuasion, if there are any these days, will by and large find more comfort with their support base in moral arguments, and in national interest arguments of the good international citizenship variety, while governments of the right have traditionally drawn more comfort from economic and security interest arguments.

Manageability Argument. There is one final kind of argument that it is very helpful to be able to make in advocating any kind of preventive strategy to any kind or colour of government in the world - and that is that the commitment is somehow finite, and won’t be the forerunner for more and more demands: in other words, that it is limited, and manageable.

It’s not always easy to persuasively make this argument. It has to be acknowledged that prevention won’t always be successful - and making an effort at prevention might be seen, reasonably, as implying a commitment to full-scale reactive response if prevention fails, although there will always be separate criteria to be applied, and separate arguments to be had, on that score. Certainly in the case of deterrent prevention by use of military threats, credibility can only be maintained if there is an absolute willingness to follow it through.

One of the biggest fears that all governments have is the fear of precedent: that having been drawn into a preventive response in this situation, what’s to stop the country being drawn into a never-ending and ever-expanding set of such responses, in a world where conflict, and the threat of conflict, seem to be forever growing.

The best possible argument to be able to make in response to this fear is that which is suggested by the work of Ted Robert Gurr and his team at the University of Maryland - viz. that notwithstanding appearances, conflict not only between states but within states is declining, and there is every reason to believe that this trend will continue. In their just-published survey, Peace and Conflict 2001,8 which is formidably detailed and meticulous, they conclude that while serious armed violence obviously persists in a number of parts of the world "the turbulence that accompanied the end of the Cold War was largely contained by the end of 2000", and that the "number and magnitude of armed conflicts within and among states have lessened since the early 1990s by nearly half". This is extremely comforting for governments - and extremely useful as a result for those advocating higher government action profiles in the conflict potential areas that do remain.

Generally, I don’t think there is any reason for particular despair on the subject of political will. It is always going to be difficult to mobilise for the preventive causes we care about, but it’s not impossible. But there is a lesson here especially for NGOs - that we have to work much harder at advocacy than we have been, and adopt a much more extensive repertoire of arguments, and much more carefully targeted arguments, than we have been in the habit of doing.

This brings me at last to say something - as promised in the circulated title of this lecture - about the role and responsibility of NGOs in the context of conflict prevention.

The Role of NGOs - and the International Crisis Group

This is not the occasion to be waxing either lyrical or indignant about the growing place of non-governmental organisations in international affairs.9 It's sufficient to say that the last time someone tried to count, for the Global Governance report in 1995, there were 29 000 of us operating internationally, and a lot more have been created since; that NGOs operate in as many different areas as life itself and defy easy classification that we are no longer as easy to ignore as we used to be; and that by and large governments and intergovernmental organisations are seeing increasing virtue and utility in drawing upon the skills and experience that a great many NGOs can offer.

In the peace, security and human rights band of the spectrum, in which my own organisation , the International Crisis Group, is located, if one leaves aside the aid and humanitarian relief organisations (whose work does quite often overlap what the others of us are doing), the most active, prominent and useful NGOs generally concentrate on one or other of three quite distinct kinds of activity, viz, thinking, talking or doing.

The 'thinking' NGOs are by and large the research institutes and think tanks, whose product is books and papers, and who engage in data gathering, idea generating, network building, paper publishing and conference organising. Their rationale tends to be contributing to the ideas pool and general debate, though some are more sharply focused.

The 'talking' or advocacy NGOs - for example Asia Watch and Amnesty - engage in research and analysis, but their primary emphasis is on spotlighting governmental abuses and engaging in tom-tom beating advocacy accordingly.

The 'doing' NGOs - and Search for Common Ground, International Alert and the Community of San Egidio are examples - tend to focus on field operations which bring people together, build confidence, and mediate disputes; they also tend to be much involved in improving governance through training and general capacity building programs.

My own organisation is something of a unique hybrid, but I think a good example of what NGOs can and should be doing to help the cause of conflict prevention. ICG concentrates essentially on conflict prevention and containment rather than on conflict resolution or peace building more generally; and, while being very much field based, engages in research, analysis and advocacy - ie. thinking and talking - rather than operationally doing things.

The group was formed in 1995 by a group of prominent internationalists and foreign policy specialists who wanted to create a wholly independent new organisation which might be capable of injecting - through high quality analysis, policy prescription and advocacy - a little more substance and responsiveness into the international conflict prevention effort.

ICG now has a team of 55 worldwide, with field operations in Africa, Asia and the Balkans and advocacy offices in Brussels (the headquarters),Washington, New York and Paris;and a budget of $6million with funding coming 40 per cent from sixteen governments, 44 per cent from major foundations and 16 per cent from individuals. Last year we produced and distributed in printed form and electronically, to many thousands of recipients, 49 reports and briefing papers on the prevention or containment of conflict in eighteen countries; and had in the last twelve months seven million hits on and half a million visitors to our website, www.crisisweb.org .

Measuring the impact and effectiveness of an organisation like ours is extremely difficult. You can measure a whole lot of easily measurable things - which will give you a picture of the level of output, of activity, of media exposure, of website access, of the number of supportive letters and comments from Foreign Ministers, or of the increase or decrease in funding support which all this is generating.

You can measure the extent to which your policy recommendations bear fruit, in the sense of being implemented by someone, somewhere at some time, wholly or in part, but that exercise has its limitations. We make hundreds of recommendations in the course of a year, ranging from the very narrowly focused - eg. local government administration in Mitrovica or Mostar - and others much larger, eg. a security guarantee for Montenegro or a resetting of peace priorities in Central Africa. Many of them are implemented, but there is always the difficulty of establishing a cause and effect relationship when they are; and when they are not, there are often other good explanations than that they were wrong-headed, for example that they were overtaken by events, or were seen as right but not yet timely, or affordable.

About the only personal contribution I think I can completely confidently claim to have made to the overthrow of Slobodan Milosevic last year, despite all my innumerable face to face meetings and speeches and print and electronic media excursions on the Balkans, was acknowledged in this letter I received in November after addressing an Institute for International Mediation and Conflict Resolution (IIMCR) seminar in the Hague a few weeks earlier:

One of our female students from Belgrade was particularly energised by your talk. As a leading member of Otpor, she served as the go-between in getting the gentleman to drive his construction crane into the parliament building.

There's no doubt that you need a certain masochistic streak to get involved in the conflict prevention and containment business. Because the blood isn't yet running in the streets the media don't find it nearly as fascinating as conflict resolution, and the attention of decision makers is hard to grab. The most frustrating thing of all is that when a government or an intergovernmental body, urged on by NGOs like mine, does actually put together a conflict prevention or containment strategy which triumphantly succeeds, so that instead of the feared violence nothing at all happens, then you can be absolutely certain that nobody will notice. And of course for a politician, to perform good works without anyone actually noticing is like having a tooth pulled without anaesthetic.

The frustrations notwithstanding, this is a deeply satisfying business to be in. Nothing is worse to contemplate - against the background of all the horror that has been wrought this last century, and all the avoidable horror of the last ten years - than the thought of the pain and terror and misery that lies ahead for so many men, women and children if we fail yet again to prevent what is preventable, and deadly conflict again breaks out. To play a part, however small, in making that horror just a little less likely, is to be as richly rewarded as one could ever be.