Wednesday, 7 July 2010

The Afghan War isn't winnable - read Alex de Waal on why

The media continue to echo the government's assumption that the war in Afghanistan is winnable, and that it is possible to 'set a timetable' for withdrawal over a period of years.
If you want to understand why this is all wrong, then I suggest you read this remarkable and brilliant recent article:
Dollarised

Alex de Waal writes about the political marketplace

State-building isn't working, and it isn't for lack of trying. The European and American countries that go by the name 'the international community' have poured expertise, money and troops into Afghanistan, the Democratic Republic of Congo and Sudan, to name only the biggest and most challenging countries. But the more effort that is expended, the more troublesome these countries seem to become. It is not clear how 30,000 more American troops, or special corruption courts, will succeed in Afghanistan where the effort of the last seven years has failed. In Congo and Sudan, peacekeeping missions and elections, heralded as the way to lift these countries out of prolonged crisis, now effectively deepen their entrapment.

International state-builders begin with a blueprint of what a modern country ought to look like, and how it ought to be run: Afghanistan needs to become more like Austria, and Sudan more like Sweden. Many of their citizens agree, and leave for developed countries if they get the chance. The economists and political scientists who advise international institutions argue that no country should follow its own unique rules, and human rights advocates insist there should be no second-best solutions for countries just because they are poor and war-torn. Too often, the same coterie of international civil servants decamps from contracts in Kosovo to East Timor to Liberia, bringing with them the same working culture and the same formula for state-building.

State-builders ignore vernacular politics, to the detriment of the countries they leave at the end of their contracts. From within the UN compound or behind the embassy walls, forces such as kinship and patron-client networks are readily denigrated as 'tribalism' or 'corruption'. But as Hamid Karzai's travails illustrate, without these well-tested mechanisms no ruler can be expected to run a turbulent country like Afghanistan. Real politics in countries like Afghanistan, Congo and Sudan operate much like village politics or even family politics, on the basis of personal affinity, loyalty and reward. The same principles and practices are found at all levels: the astute village chief has the skills he needs to be a functional head of state, while the journalist for a provincial newspaper can rival the professor of politics for insight. Western policymakers call such countries 'fragile states'. Their formal state structures are not strong enough to resolve political disputes or manage national budgets, which makes them problematic interlocutors for Western governments and international institutions. The World Bank defines a fragile state as one that cannot efficiently handle foreign assistance. In some cases, civil war has devastated the social fabric, creating what we might call fragile societies as well. Often it was precisely the strength of its social fabric which allowed a country to withstand foreign invasions and colonial occupation with its social order intact; today, that strong social fabric renders state institutions incapable of effective government.

Patronage is the circulatory system of real politics and in the modern era the heartbeat of the system is possession of the symbols of sovereignty, which allows a ruler to allocate aid funds or mineral revenues. Patronage can be inefficient and corrupt, and can contribute to political and economic crisis, even war, but patron-client systems may also function as a repository of trust and security. In places where formal state institutions do not provide stability and services, patronage mechanisms can dispense resources, sometimes in a way that is recognisably fair. No Afghan, Congolese or Sudanese would readily forfeit actually existing patronage systems for the uncertain promise of formal institutions.

How has Omar al-Bashir managed to remain in power in Sudan for 20 years, despite several civil wars, economic crisis and international ostracism? Unusually for a military officer who launched a coup, he is not an autocrat so much as chairman of the board, and an unruly board it is, consisting of Islamist ideologues, party bosses and security chiefs, each of whom has his own fiefdom and funds. It's rule by cabal not charisma, and it is worth trying to understand how he managed it. I am told, by those close to Bashir, that he himself is puzzled.

The activities of advocacy groups such as Human Rights Watch and the Save Darfur Coalition has been a mixed blessing in Sudan. These organisations generate a lot of information and interest, but their rush to condemn means that almost no one takes the trouble to study the political skills of Sudan's rulers. At the Darfur peace talks, I had the opportunity to observe the head of the Sudanese delegation, Dr Majzoub al-Khalifa, at close quarters. He had an impressively thick skin, and could withstand any insult that was hurled at him. He was also skilled at what is known in the vernacular as 'Jellaba politics', after the class of riverine traders who historically dominated commerce in the Sudanese peripheries and beyond. We might call it 'retail politics', or more precisely, 'retail patronage politics'. It is the ability to weigh up the price, in money, of a particular individual's loyalty and make him an offer (it is a very gender-specific exercise); it is also about reading the market so as to know the likelihood that the price will rise or fall in the future.

The Darfur rebels complained about Majzoub's 'Jellaba politics', but they too were active participants in the political market. During the last days of the peace talks in April and May 2006, while the formal negotiations were conducted over the text of the Darfur Peace Agreement, informal but more important bargaining was going on over the price of a deal. The key issues that determined whether the chairman of the Sudan Liberation Movement, Abdel Wahid al-Nur, would sign the document were not contained in the 87 pages of detailed provisions: what mattered was the amount of money in the compensation fund, which would be under his personal control (and thus available for patronage purposes). He was offered $30 million but demanded a minimum of $100 million. He also demanded a personal pay-off, reportedly of $5 million, as a signing fee. The negotiations over the text were conducted in formal sessions, in English; the bargaining over the price was done behind closed doors, in Arabic. A year and a half later, another negotiation was conducted between the leader of one of the largest Arab militias in Darfur and the Sudanese government. This had the virtue of being uncomplicated because there was no international mediator and no need to pretend that the talks concerned anything other than the price itself. On agreeing to the offer, Mohamed Hamdan Hamiti commented that he expected Khartoum to deliver on 40 per cent of the deal, adding that that would be enough for 18 months. The idea of Darfur as a political marketplace, an auction of loyalties, was not greeted with dismay: the Sudanese have absolutely no difficulty with the politics of the souk.

Another term in the political vernacular is tajil, 'delay', from which is derived the particular skill of 'tajility', strategic delay or the art of procrastinating until one's counterpart is exhausted or removed. This is the prevailing approach to formal agreements, which there is great reluctance to sign because of the ever changing conditions of the marketplace. Deals on money and resources can be struck quickly because they are temporary, but more permanent arrangements are a cause for concern. Now that every politician has a mobile phone and every field commander a satellite phone, price bargaining that used to take months can now be over in a matter of hours.

In principle, the tools determining the price at which a bargain will be struck can include electoral votes, the allegiance of parliamentarians, editorial columns in newspapers, public petitions and protests, work stoppages and the like; they could also include supernatural or spiritual sanctions. In many countries with weak states, among them Senegal and Tanzania, violence can be held in check by strong patronage systems. Elsewhere, however, violence is a bargaining tool in its own right and has a tendency to supplant other methods. It may be a means of acquiring assets, of affirming (masculine) identities, or of managing group boundaries and sustaining group cohesion. It also serves as a means of communication with the other party. The insurgent uses targeted violence against the government to get himself noticed and command a good price for his allegiance. He may also want to seize assets in order to become a bigger player. The government uses violence against the insurgent, or his community, in order to weaken his asset base and drive down the price. Elites don't attack one another directly and are remarkably civil to each other. Often they live in the same urban neighbourhoods, while the fighting takes place in faraway villages. Reciprocal violence against their human and material asset bases is merely a part of the bargaining process, though occasionally one side or the other breaks the rules and there is an episode, usually brief, of all-out war, after which low-level criminal-political violence resumes.

Bashir owes his longevity to his astute monitoring of the Sudanese marketplace, both the inner circle of his quarrelsome cabal and the turbulent, impetuous provincial elites, and to his readiness to ensure that just enough is paid. He looks set to survive, but whether Sudan can do so is another question: Bashir may yet preside over the country's transition from a unified sovereign territory to a geographically fragmented political marketplace, in which sovereignty is a commodity to be bartered.

Political scientists and policymakers like models, and the political marketplace is amenable to being modelled. The 'marketplace', here, is more than a metaphor, given that prices can be tracked along with supply and demand, and speculative bubbles identified. But like the market in protection run by mafia gangs and drug lords, the political marketplace has perverse features, including the possibility that competition will drive prices up rather than down. Imagine a country run by negotiation between a ruler and a set of provincial elites. Each elite member controls a unit: a tribe, a militia, an insurgent group, a commercial operation, or something similar. The ruler controls most of the resources, by virtue of sovereign rents, including foreign aid. The provincial elites can't overthrow the ruler and the ruler can't remove the provincial elites.

The provincial elites want the highest price from the ruler for their allegiance. They covet positions in government as well as resources, including trading licenses, local taxation powers and straight cash. In a system where violence is not normally a bargaining tool, provincial elites can use elections, demonstrations, boycotts and strikes to bring the ruler to the table. He will want to pay the lowest price for their allegiance and can threaten to withhold resources, letting them wither in the political wilderness. In a variant of this game, he can sponsor rival elites and even license them to undermine those who demand too high a price. In states where violence is an option, it can be used both by the ruler and the elites. The crucial need is to call the price of loyalty and get it right, and then to determine how long it will be before the price is up for renegotiation. There are other questions too, about the nature of the ruler (he may be no more than today's compromise among competing factions) and the extent to which a provincial elite has control of its constituency. But let's assume that the ruler and the provincial elites are relatively well defined and stable.

Countries that follow this model share three characteristics. First, as states they lack the capacity to manage political conflict within their territories. In Africa, the DRC and Sudan conform to the model in this respect but Ethiopia does not; in Asia, Afghanistan and Pakistan fit, but not Iran. Second, these countries are large, which allows political entities smaller than the state to enjoy a quasi-autonomous existence. This may also be possible in small states, but proximity and tight cultural affinity make for different forms of political bargaining. This framework cannot be applied to a small country like Burundi without major modifications and not to Rwanda at all. There is an important spatial dimension: the model would apply to Sudan, the DRC, Afghanistan, Chad, Somalia, Nigeria – all countries with dispersed populations.

Third, while the ruler requires the allegiance of key provincial elites to stay in power, he cannot rely on their resources – he has to have his own. If the day to day tactical challenge for a ruler is how to keep down the price of the elites' loyalty, the strategic challenge is how to sustain a patronage system that is affordable and stable. Fortunately for a ruler, he has sovereign rents, especially foreign aid and mineral revenues. Unfortunately for a state-builder, the forces of globalisation have made patronage less affordable and political markets more volatile. There has been price inflation in loyalty, and rulers nowadays are struggling to pay. In good years a ruler may be able to meet the demands of the elites. An economic crunch will not reduce those demands, merely his ability to pay. One response is to try to plunder natural resources (or license members of the elite to do so), divert aid, run up debts and sell off state assets at fire-sale prices to the elites. Another is to reduce the ambit of patronage, by shedding some of the beneficiaries, even if this generates grievance and conflict. A third alternative is to license one provincial elite to pillage its neighbour's assets – the cheapest form of controlling a borderland and one that can ignite intractable ethnic conflicts. A classic case is the counterinsurgency in Darfur.

A patronage system can also face a crisis as a result of increased demand. If the price of allegiance is driven up as new buyers enter the marketplace – neighbouring states, criminal cartels, international agencies – a ruler can face the same dilemma. Sustainable patronage systems are inclusive: either all the political elites are included at all times, or there is a predictable rotation, leaving those who are out of favour to focus on finding a way back under the existing system. Rulers such as Zaire's Mobutu Sese Seko, who were experts at the patronage system, rotated rival elites. It was a cheaper way of maintaining the system and had the advantage of preventing those in favour from building a durable patronage system of their own. In the African political vernacular, democracy usually means 'fair shares' or 'everyone at the table', i.e. the dining table. Constituency-based parliamentary systems are the preferred model because the rationale for choosing a representative is to ensure that he or she can secure state resources for a district. The 'winner takes all' system of executive power goes against this.

To manage a sustainable patronage system well is to limit the use of violence for bargaining purposes – which means that the security institutions of the state must not become too powerful, because that would risk a coup d'état. So the strategy is to multiply and divide: establish a range of security services, each as a specialised patronage network, thereby diluting the threat to the ruler from his own security officers. At the same time, it's important to try to keep the rules peacefully: one of the underacknowledged achievements of Mobutu and his political opponents was, for almost 30 years, to minimise the role of violence in Zairean political bargaining.

The independence generation in Africa had its share of civil wars and violent coups, but they tended to be confined within a single country, and were amenable to resolution either by force or negotiation. The patterns of violent conflict were distinct from those of today. Quite probably, there was more violence and destruction in Biafra, Angola, Burundi, Chad, the Ethiopian revolution, the guerrilla wars in Eritrea and Tigray, the Somali civil wars of 1988-91, southern Sudan in the late 1980s and early 1990s, than there is at present. But the characteristic pattern of intractable lawlessness was absent. We did not see the emergence of regional, overlapping conflicts, such as Liberia-Sierra Leone-Guinea, or the Great Lakes, or Sudan-Chad-Central African Republic, until more recently.

This shift in the nature of conflict is linked to the monetisation of patronage: the way cash payments, and especially payments in convertible currencies, have become an ever larger component of patronage. It is a general trend rather than a universal one (there are some interesting exceptions), and the reasons it has occurred are not difficult to understand. With economic liberalisation and the growth of informal and international criminal economies, and above all with economic globalisation, convertible currency drives out all other currencies, monetary and non-monetary, in which loyalty can be bought and sold. Material reward is at the centre of any patronage system, but in the last generation patronage has become not only monetised but 'dollarised'. Political markets, no longer confined to a single country, are now joining up across borders and becoming global. Symbolic rewards such as titles and ribbons are valued less – cash is what counts.

Rulers who have a legitimacy deficit, either because they have recently seized power and need to broaden their support, or because their political capital has run down, dispense cash. The ruler of a neighbouring country who wants to rent allegiance across the border will use cash. In Afghanistan the CIA pays off warlords in dollars. Drug cartels move into West Africa, finding embattled rulers ready to rent their sovereignty for cash. When several contiguous countries are faced with the same phenomenon, as most of sub-Saharan Africa is, an integrated, deregulated market will follow. This makes the weaker states into supplicants vis-à-vis their cash-rich neighbours, and sovereignty itself becomes tradable. For example, the position of the Central African Republic, on the peripheries of the Nile Valley political marketplace (centred on Khartoum), the trans-Saharan marketplace (patron: Libya, with N'djamena in an intermediate position) and the Congo Basin/ Great Lakes marketplace (with rival buyers in Kinshasa, Kigali and to a lesser extent Kampala), makes it a subordinate actor. The former president of CAR, Ange-Félix Patassé, was in fact the junior partner in a coalition with the Congolese rebel leader Jean-Pierre Bemba, because Bemba had more money.

In any marketplace, flows of information drive the bargaining. Until the telecoms revolution, and especially the advent of the Thuraya satellite phone, negotiations had to be conducted face to face and in sequence. Before this, by dint of controlling the media, two-way radios and the telephone network, the ruler possessed two huge advantages. He knew more than everyone else and he knew that once a bargain had been struck, it would be difficult for a provincial leader to go elsewhere and renegotiate a better one. Today deals can be negotiated and renegotiated round the clock – and even during the course of a single battle.

International engagements in weak and fragile states are conceived with development, humanitarian assistance, peacemaking, peacekeeping and institution-building in mind. These interventions are rarely designed or implemented with regard to the political marketplace. Often, international policymakers and diplomats ignore the ubiquity and significance of patronage systems: they pretend that patronage doesn't exist, or they recognise it but think of it as an aberration that can be safely ignored, or an abuse that should not be recognised. Some policies are clearly intended to dismantle patronage systems or supplant them. Yet on occasion, when the interests of major states are at stake, international players deliberately use the marketplace. During the Cold War, the Western powers paid for the loyalty of rulers such as Siad Barre, Daniel arap Moi, Jaafar Nimeiri and Mobutu Sese Seko, not to mention governments in Pakistan, Thailand and South Korea, and insurgents in Afghanistan and Laos. In 2001, large cash payments were made to Afghan power brokers to rent their allegiance for as long as it took to overthrow the Taliban.

All forms of international intervention influence the functioning of the marketplace. They skew the price of loyalty, by providing a buyer or seller with more resources or more options, or by directly renting the allegiance of certain actors, and they accelerate the dollarisation and regionalisation of the market. Interventions such as sanctions can help demonetise the system, or can drive it towards criminality. As yet we don't have the analytical framework or the monitoring data properly to assess how policies affect the marketplace, but we can make some informed guesses.

In a country such as Nigeria, the government regularly launches anti-corruption drives. Given that all Nigerian politicians are part of patronage mechanisms and therefore 'corrupt' to a greater or lesser degree, the question arises, who will be targeted in a war on corruption? The answer is, those who have demanded too high a price and have fallen from the ruler's favour. An anti-corruption campaign, ostensibly aimed at 'institution-building', becomes an instrument for manipulating patronage and, by extension, an exercise in corruption itself. Within a political marketplace, the only way a member of an elite can respond cogently to an anti-corruption effort is to offer a pay-off to the political master of the anti-corruption tsar. Abdicating from the patronage system is a short cut to political oblivion.

A political agreement in the marketplace should be seen less as a legal contract, more as a rental transaction. Neither party has much confidence that a contract can be enforced by law. Any deal signed with a ruler who has external support remains good only for as long as that support continues at existing levels. If it is withdrawn or scaled back, rival elites will want to renegotiate. In Congo, therefore, we would expect the departure of the UN force, MONUC, to be followed by new rebellions by leaders of provincial elites demanding a better deal. In Afghanistan, if US forces are expanded and the Karzai government negotiates a deal at the moment of its greatest strength, it will last only as long as those particular market conditions persist. When American troops begin to leave, bargaining will resume, most likely with violence on either side. In both cases, the presence of international forces and international assistance, of unknown duration, introduces uncertainty into the political marketplace, which makes a solution more difficult, not less.

Elsewhere, international engagement is either intended to weaken a government, or has the effect of bidding up the price of the opposition's allegiance. This is the case for Darfur, where international ostracism of Khartoum and the readiness of the international community to give a platform to poorly organised rebel groups with little political or military capability has over-inflated the price which rebel leaders believe they can charge. During the peace talks, the rebels figured that the Save Darfur campaign would allow them to demand a higher price than anything on offer from Khartoum, while the Sudanese government negotiators anticipated that US commitment to the draft agreement would render the activists irrelevant, so that they could pay a lower price for putting an end to the rebellion. The government feared a nationwide inflation of the price of loyalty, with copycat demands from other provincial elites for a similar deal – just as the Darfurians' demands had been increased by the precedent of the southern Sudanese peace deal. The rebels continue to hold out, expecting that intense international interest will translate into a high price, while the government continues to regard this as a speculative bubble. Should the international community withdraw from Darfur, the government will be proved right. It will be tempted to renegotiate the deal to its advantage, probably by using force. The faithful implementation of any agreement reached today depends on international engagement continuing at much the same level.

In a conflict in an institutionalised state, whose adversary will also, typically, be institutionalised, peace negotiations can take the classical form. It would have been possible in principle to negotiate a workable peace agreement between the president of Sri Lanka and the leader of the Tamil Tigers: had the two reached an agreement in good faith, it could have held on the basis of the political decisions they made. It was possible for the force commander of the UN mission in Eritrea and Ethiopia to deal solely with the chiefs of staff of the two armies, knowing that the orders handed down by them would be respected. But in Sudan, Congo or Afghanistan, the ruler is only as strong as today's bargains with members of the political elite, who retain a great deal of autonomy. The deal will be voided when circumstances change, and to the frustration of international peacemakers and peacekeepers, a commitment at the top does not translate into action throughout the system. This is even true of something as elementary as a ceasefire. The president may announce a ceasefire, but the generals in the office of the chief of staff will need to meet to agree what this means in practice and to negotiate with the provincial commanders, warlords, police chiefs and militia commanders as to how they should interpret it.

The temptation is either to accuse the leader of perfidy or to recognise that there is a structural weakness in the system, requiring a bottom-up approach. The peacemaker may draw the implication that the conflict is compounded of different local conflicts, each of them in turn compounded of micro-conflicts, and that peace is made by dealing with each one at its own level. This is how international mediators, including a growing number of conflict-resolution NGOs, become involved in local peacemaking. Where there is international concern, more senior diplomats become engaged too. Special representatives of the secretary general or even UN Security Council ambassadors may nowadays try to micromanage district-level disputes or even village conflicts. The difficulty with this approach is that every intervention is a distortion of the market. The more local the engagement, the greater the relative weight of the international actor involved, and hence the greater the distortion. An international mediator will never be as knowledgeable, skilled or patient as a local actor and will invariably be drawn towards one side or the other. Given that every agreement reached is good only for that particular set of conditions, these interventions are inherently unsustainable.

The end result is that peace operations become entrapped. They cannot resolve conflicts and the more they try to do so, the greater the part they play in the dynamics of the marketplace itself, meaning they can't now withdraw. Many peacemakers become disillusioned: they see missions without end. Cynicism spreads from the fringes to the centre of these operations, with more and more people asking, 'What are we doing here?' In turn, smaller-contributing countries, who may have sent soldiers out of a sense of solidarity or principle, are likely to keep them there less because of their original political motives than because of the rewards of remaining loyal to the superpower (allegiance-pricing is at work among the intervening states as well). The guiding purpose in Afghanistan will soon become salvaging Nato, and in Sudan, saving the UN's face – if it has not already done so.

The spread of norms of 'good governance' – which requires adherence to international standards of government and involves supplying aid for elections and NGOs, conferences, training workshops and educational opportunities – has contributed to a worldwide spread of democratisation and civil society. The growth in conflict resolution has helped reduce the numbers of conflicts and people killed. But it also reduces the particularities of national political cultures and enables the monetisation of political markets. It takes a state with considerable self-confidence and financial autonomy to resist the pressures to play this game. Those that hold out against the intrusions are often unsavoury – North Korea and Burma, for instance – but there are two sides to international intervention and peacekeeping operations, as there are to globalisation itself.

The likelihood of dollarised political marketplaces becoming capable, legitimate, institutionalised states is receding. In retrospect, there was a brief 30-year window after the Second World War, when conditions were propitious for building functioning states where none existed before. More recently, the forces of globalisation have made it far harder, perhaps impossible, for would-be states to join the club. The dynamics of the dollarised political marketplace, which are more powerful than international state-building practices, are entrenching a pattern of deregulated governance in global borderlands. The recent trend towards representative politics, pronounced in African countries without strong state institutions, will be hard to sustain under these pressures, and is likely to be reversed. The prospects for building democratic states in central Asia are fading too. Instead, the logic of the political marketplace points towards continuing low-level conflicts, part criminal and part political, in which violence serves as a principal bargaining tool in an auction room of allegiances.

Places as diverse as Afghanistan, Nigeria, East Timor and Somalia may appear remote from the circuits of international capital yet they are no less globalised than Singapore or the Netherlands: it's just that their global connections are the kind we prefer not to acknowledge. The rich world's demand for recreational drugs, combined with its politicians' insistence that supplying this market is a criminal activity, funnels billions of dollars into unregulated political marketplaces. The global war on terror disburses further billions through poorly monitored loyalty payments in the most volatile parts of the world. Western governments spend significant sums trying to establish functioning sovereignty amid the factional politics of 'fragile' states, and civil, inclusive and affordable patronage systems are being swept away. For the unfortunate populations of these countries, this is a loss just as devastating as the weakening of state institutions, and one less recognised.

RR's final comment: If the illiterates who rule us were to read this, then they might understand why they either need to commit to staying in Afghanistan indefinitely, until they are forced out - or do the sensible thing, and stop propping up this corrupt government, and get out now...

0 Comments:

Post a Comment

Subscribe to Post Comments [Atom]

Links to this post:

Create a Link

<< Home

1. 2. 3. Rupert's Read: The Afghan War isn't winnable - read Alex de Waal on why 4. 12. 15. 16. 17. 18. 19. 20. 23. 24.

25. 26. The Afghan War isn't winnable - read Alex de Waal on why 27. 28.

29.
The media continue to echo the government's assumption that the war in Afghanistan is winnable, and that it is possible to 'set a timetable' for withdrawal over a period of years.
If you want to understand why this is all wrong, then I suggest you read this remarkable and brilliant recent article:
http://www.lrb.co.uk/v32/n12/alex-de-waal/dollarised
Dollarised

Alex de Waal writes about the political marketplace

State-building isn't working, and it isn't for lack of trying. The European and American countries that go by the name 'the international community' have poured expertise, money and troops into Afghanistan, the Democratic Republic of Congo and Sudan, to name only the biggest and most challenging countries. But the more effort that is expended, the more troublesome these countries seem to become. It is not clear how 30,000 more American troops, or special corruption courts, will succeed in Afghanistan where the effort of the last seven years has failed. In Congo and Sudan, peacekeeping missions and elections, heralded as the way to lift these countries out of prolonged crisis, now effectively deepen their entrapment.

International state-builders begin with a blueprint of what a modern country ought to look like, and how it ought to be run: Afghanistan needs to become more like Austria, and Sudan more like Sweden. Many of their citizens agree, and leave for developed countries if they get the chance. The economists and political scientists who advise international institutions argue that no country should follow its own unique rules, and human rights advocates insist there should be no second-best solutions for countries just because they are poor and war-torn. Too often, the same coterie of international civil servants decamps from contracts in Kosovo to East Timor to Liberia, bringing with them the same working culture and the same formula for state-building.

State-builders ignore vernacular politics, to the detriment of the countries they leave at the end of their contracts. From within the UN compound or behind the embassy walls, forces such as kinship and patron-client networks are readily denigrated as 'tribalism' or 'corruption'. But as Hamid Karzai's travails illustrate, without these well-tested mechanisms no ruler can be expected to run a turbulent country like Afghanistan. Real politics in countries like Afghanistan, Congo and Sudan operate much like village politics or even family politics, on the basis of personal affinity, loyalty and reward. The same principles and practices are found at all levels: the astute village chief has the skills he needs to be a functional head of state, while the journalist for a provincial newspaper can rival the professor of politics for insight. Western policymakers call such countries 'fragile states'. Their formal state structures are not strong enough to resolve political disputes or manage national budgets, which makes them problematic interlocutors for Western governments and international institutions. The World Bank defines a fragile state as one that cannot efficiently handle foreign assistance. In some cases, civil war has devastated the social fabric, creating what we might call fragile societies as well. Often it was precisely the strength of its social fabric which allowed a country to withstand foreign invasions and colonial occupation with its social order intact; today, that strong social fabric renders state institutions incapable of effective government.

Patronage is the circulatory system of real politics and in the modern era the heartbeat of the system is possession of the symbols of sovereignty, which allows a ruler to allocate aid funds or mineral revenues. Patronage can be inefficient and corrupt, and can contribute to political and economic crisis, even war, but patron-client systems may also function as a repository of trust and security. In places where formal state institutions do not provide stability and services, patronage mechanisms can dispense resources, sometimes in a way that is recognisably fair. No Afghan, Congolese or Sudanese would readily forfeit actually existing patronage systems for the uncertain promise of formal institutions.

How has Omar al-Bashir managed to remain in power in Sudan for 20 years, despite several civil wars, economic crisis and international ostracism? Unusually for a military officer who launched a coup, he is not an autocrat so much as chairman of the board, and an unruly board it is, consisting of Islamist ideologues, party bosses and security chiefs, each of whom has his own fiefdom and funds. It's rule by cabal not charisma, and it is worth trying to understand how he managed it. I am told, by those close to Bashir, that he himself is puzzled.

The activities of advocacy groups such as Human Rights Watch and the Save Darfur Coalition has been a mixed blessing in Sudan. These organisations generate a lot of information and interest, but their rush to condemn means that almost no one takes the trouble to study the political skills of Sudan's rulers. At the Darfur peace talks, I had the opportunity to observe the head of the Sudanese delegation, Dr Majzoub al-Khalifa, at close quarters. He had an impressively thick skin, and could withstand any insult that was hurled at him. He was also skilled at what is known in the vernacular as 'Jellaba politics', after the class of riverine traders who historically dominated commerce in the Sudanese peripheries and beyond. We might call it 'retail politics', or more precisely, 'retail patronage politics'. It is the ability to weigh up the price, in money, of a particular individual's loyalty and make him an offer (it is a very gender-specific exercise); it is also about reading the market so as to know the likelihood that the price will rise or fall in the future.

The Darfur rebels complained about Majzoub's 'Jellaba politics', but they too were active participants in the political market. During the last days of the peace talks in April and May 2006, while the formal negotiations were conducted over the text of the Darfur Peace Agreement, informal but more important bargaining was going on over the price of a deal. The key issues that determined whether the chairman of the Sudan Liberation Movement, Abdel Wahid al-Nur, would sign the document were not contained in the 87 pages of detailed provisions: what mattered was the amount of money in the compensation fund, which would be under his personal control (and thus available for patronage purposes). He was offered $30 million but demanded a minimum of $100 million. He also demanded a personal pay-off, reportedly of $5 million, as a signing fee. The negotiations over the text were conducted in formal sessions, in English; the bargaining over the price was done behind closed doors, in Arabic. A year and a half later, another negotiation was conducted between the leader of one of the largest Arab militias in Darfur and the Sudanese government. This had the virtue of being uncomplicated because there was no international mediator and no need to pretend that the talks concerned anything other than the price itself. On agreeing to the offer, Mohamed Hamdan Hamiti commented that he expected Khartoum to deliver on 40 per cent of the deal, adding that that would be enough for 18 months. The idea of Darfur as a political marketplace, an auction of loyalties, was not greeted with dismay: the Sudanese have absolutely no difficulty with the politics of the souk.

Another term in the political vernacular is tajil, 'delay', from which is derived the particular skill of 'tajility', strategic delay or the art of procrastinating until one's counterpart is exhausted or removed. This is the prevailing approach to formal agreements, which there is great reluctance to sign because of the ever changing conditions of the marketplace. Deals on money and resources can be struck quickly because they are temporary, but more permanent arrangements are a cause for concern. Now that every politician has a mobile phone and every field commander a satellite phone, price bargaining that used to take months can now be over in a matter of hours.

In principle, the tools determining the price at which a bargain will be struck can include electoral votes, the allegiance of parliamentarians, editorial columns in newspapers, public petitions and protests, work stoppages and the like; they could also include supernatural or spiritual sanctions. In many countries with weak states, among them Senegal and Tanzania, violence can be held in check by strong patronage systems. Elsewhere, however, violence is a bargaining tool in its own right and has a tendency to supplant other methods. It may be a means of acquiring assets, of affirming (masculine) identities, or of managing group boundaries and sustaining group cohesion. It also serves as a means of communication with the other party. The insurgent uses targeted violence against the government to get himself noticed and command a good price for his allegiance. He may also want to seize assets in order to become a bigger player. The government uses violence against the insurgent, or his community, in order to weaken his asset base and drive down the price. Elites don't attack one another directly and are remarkably civil to each other. Often they live in the same urban neighbourhoods, while the fighting takes place in faraway villages. Reciprocal violence against their human and material asset bases is merely a part of the bargaining process, though occasionally one side or the other breaks the rules and there is an episode, usually brief, of all-out war, after which low-level criminal-political violence resumes.

Bashir owes his longevity to his astute monitoring of the Sudanese marketplace, both the inner circle of his quarrelsome cabal and the turbulent, impetuous provincial elites, and to his readiness to ensure that just enough is paid. He looks set to survive, but whether Sudan can do so is another question: Bashir may yet preside over the country's transition from a unified sovereign territory to a geographically fragmented political marketplace, in which sovereignty is a commodity to be bartered.

Political scientists and policymakers like models, and the political marketplace is amenable to being modelled. The 'marketplace', here, is more than a metaphor, given that prices can be tracked along with supply and demand, and speculative bubbles identified. But like the market in protection run by mafia gangs and drug lords, the political marketplace has perverse features, including the possibility that competition will drive prices up rather than down. Imagine a country run by negotiation between a ruler and a set of provincial elites. Each elite member controls a unit: a tribe, a militia, an insurgent group, a commercial operation, or something similar. The ruler controls most of the resources, by virtue of sovereign rents, including foreign aid. The provincial elites can't overthrow the ruler and the ruler can't remove the provincial elites.

The provincial elites want the highest price from the ruler for their allegiance. They covet positions in government as well as resources, including trading licenses, local taxation powers and straight cash. In a system where violence is not normally a bargaining tool, provincial elites can use elections, demonstrations, boycotts and strikes to bring the ruler to the table. He will want to pay the lowest price for their allegiance and can threaten to withhold resources, letting them wither in the political wilderness. In a variant of this game, he can sponsor rival elites and even license them to undermine those who demand too high a price. In states where violence is an option, it can be used both by the ruler and the elites. The crucial need is to call the price of loyalty and get it right, and then to determine how long it will be before the price is up for renegotiation. There are other questions too, about the nature of the ruler (he may be no more than today's compromise among competing factions) and the extent to which a provincial elite has control of its constituency. But let's assume that the ruler and the provincial elites are relatively well defined and stable.

Countries that follow this model share three characteristics. First, as states they lack the capacity to manage political conflict within their territories. In Africa, the DRC and Sudan conform to the model in this respect but Ethiopia does not; in Asia, Afghanistan and Pakistan fit, but not Iran. Second, these countries are large, which allows political entities smaller than the state to enjoy a quasi-autonomous existence. This may also be possible in small states, but proximity and tight cultural affinity make for different forms of political bargaining. This framework cannot be applied to a small country like Burundi without major modifications and not to Rwanda at all. There is an important spatial dimension: the model would apply to Sudan, the DRC, Afghanistan, Chad, Somalia, Nigeria – all countries with dispersed populations.

Third, while the ruler requires the allegiance of key provincial elites to stay in power, he cannot rely on their resources – he has to have his own. If the day to day tactical challenge for a ruler is how to keep down the price of the elites' loyalty, the strategic challenge is how to sustain a patronage system that is affordable and stable. Fortunately for a ruler, he has sovereign rents, especially foreign aid and mineral revenues. Unfortunately for a state-builder, the forces of globalisation have made patronage less affordable and political markets more volatile. There has been price inflation in loyalty, and rulers nowadays are struggling to pay. In good years a ruler may be able to meet the demands of the elites. An economic crunch will not reduce those demands, merely his ability to pay. One response is to try to plunder natural resources (or license members of the elite to do so), divert aid, run up debts and sell off state assets at fire-sale prices to the elites. Another is to reduce the ambit of patronage, by shedding some of the beneficiaries, even if this generates grievance and conflict. A third alternative is to license one provincial elite to pillage its neighbour's assets – the cheapest form of controlling a borderland and one that can ignite intractable ethnic conflicts. A classic case is the counterinsurgency in Darfur.

A patronage system can also face a crisis as a result of increased demand. If the price of allegiance is driven up as new buyers enter the marketplace – neighbouring states, criminal cartels, international agencies – a ruler can face the same dilemma. Sustainable patronage systems are inclusive: either all the political elites are included at all times, or there is a predictable rotation, leaving those who are out of favour to focus on finding a way back under the existing system. Rulers such as Zaire's Mobutu Sese Seko, who were experts at the patronage system, rotated rival elites. It was a cheaper way of maintaining the system and had the advantage of preventing those in favour from building a durable patronage system of their own. In the African political vernacular, democracy usually means 'fair shares' or 'everyone at the table', i.e. the dining table. Constituency-based parliamentary systems are the preferred model because the rationale for choosing a representative is to ensure that he or she can secure state resources for a district. The 'winner takes all' system of executive power goes against this.

To manage a sustainable patronage system well is to limit the use of violence for bargaining purposes – which means that the security institutions of the state must not become too powerful, because that would risk a coup d'état. So the strategy is to multiply and divide: establish a range of security services, each as a specialised patronage network, thereby diluting the threat to the ruler from his own security officers. At the same time, it's important to try to keep the rules peacefully: one of the underacknowledged achievements of Mobutu and his political opponents was, for almost 30 years, to minimise the role of violence in Zairean political bargaining.

The independence generation in Africa had its share of civil wars and violent coups, but they tended to be confined within a single country, and were amenable to resolution either by force or negotiation. The patterns of violent conflict were distinct from those of today. Quite probably, there was more violence and destruction in Biafra, Angola, Burundi, Chad, the Ethiopian revolution, the guerrilla wars in Eritrea and Tigray, the Somali civil wars of 1988-91, southern Sudan in the late 1980s and early 1990s, than there is at present. But the characteristic pattern of intractable lawlessness was absent. We did not see the emergence of regional, overlapping conflicts, such as Liberia-Sierra Leone-Guinea, or the Great Lakes, or Sudan-Chad-Central African Republic, until more recently.

This shift in the nature of conflict is linked to the monetisation of patronage: the way cash payments, and especially payments in convertible currencies, have become an ever larger component of patronage. It is a general trend rather than a universal one (there are some interesting exceptions), and the reasons it has occurred are not difficult to understand. With economic liberalisation and the growth of informal and international criminal economies, and above all with economic globalisation, convertible currency drives out all other currencies, monetary and non-monetary, in which loyalty can be bought and sold. Material reward is at the centre of any patronage system, but in the last generation patronage has become not only monetised but 'dollarised'. Political markets, no longer confined to a single country, are now joining up across borders and becoming global. Symbolic rewards such as titles and ribbons are valued less – cash is what counts.

Rulers who have a legitimacy deficit, either because they have recently seized power and need to broaden their support, or because their political capital has run down, dispense cash. The ruler of a neighbouring country who wants to rent allegiance across the border will use cash. In Afghanistan the CIA pays off warlords in dollars. Drug cartels move into West Africa, finding embattled rulers ready to rent their sovereignty for cash. When several contiguous countries are faced with the same phenomenon, as most of sub-Saharan Africa is, an integrated, deregulated market will follow. This makes the weaker states into supplicants vis-à-vis their cash-rich neighbours, and sovereignty itself becomes tradable. For example, the position of the Central African Republic, on the peripheries of the Nile Valley political marketplace (centred on Khartoum), the trans-Saharan marketplace (patron: Libya, with N'djamena in an intermediate position) and the Congo Basin/ Great Lakes marketplace (with rival buyers in Kinshasa, Kigali and to a lesser extent Kampala), makes it a subordinate actor. The former president of CAR, Ange-Félix Patassé, was in fact the junior partner in a coalition with the Congolese rebel leader Jean-Pierre Bemba, because Bemba had more money.

In any marketplace, flows of information drive the bargaining. Until the telecoms revolution, and especially the advent of the Thuraya satellite phone, negotiations had to be conducted face to face and in sequence. Before this, by dint of controlling the media, two-way radios and the telephone network, the ruler possessed two huge advantages. He knew more than everyone else and he knew that once a bargain had been struck, it would be difficult for a provincial leader to go elsewhere and renegotiate a better one. Today deals can be negotiated and renegotiated round the clock – and even during the course of a single battle.

International engagements in weak and fragile states are conceived with development, humanitarian assistance, peacemaking, peacekeeping and institution-building in mind. These interventions are rarely designed or implemented with regard to the political marketplace. Often, international policymakers and diplomats ignore the ubiquity and significance of patronage systems: they pretend that patronage doesn't exist, or they recognise it but think of it as an aberration that can be safely ignored, or an abuse that should not be recognised. Some policies are clearly intended to dismantle patronage systems or supplant them. Yet on occasion, when the interests of major states are at stake, international players deliberately use the marketplace. During the Cold War, the Western powers paid for the loyalty of rulers such as Siad Barre, Daniel arap Moi, Jaafar Nimeiri and Mobutu Sese Seko, not to mention governments in Pakistan, Thailand and South Korea, and insurgents in Afghanistan and Laos. In 2001, large cash payments were made to Afghan power brokers to rent their allegiance for as long as it took to overthrow the Taliban.

All forms of international intervention influence the functioning of the marketplace. They skew the price of loyalty, by providing a buyer or seller with more resources or more options, or by directly renting the allegiance of certain actors, and they accelerate the dollarisation and regionalisation of the market. Interventions such as sanctions can help demonetise the system, or can drive it towards criminality. As yet we don't have the analytical framework or the monitoring data properly to assess how policies affect the marketplace, but we can make some informed guesses.

In a country such as Nigeria, the government regularly launches anti-corruption drives. Given that all Nigerian politicians are part of patronage mechanisms and therefore 'corrupt' to a greater or lesser degree, the question arises, who will be targeted in a war on corruption? The answer is, those who have demanded too high a price and have fallen from the ruler's favour. An anti-corruption campaign, ostensibly aimed at 'institution-building', becomes an instrument for manipulating patronage and, by extension, an exercise in corruption itself. Within a political marketplace, the only way a member of an elite can respond cogently to an anti-corruption effort is to offer a pay-off to the political master of the anti-corruption tsar. Abdicating from the patronage system is a short cut to political oblivion.

A political agreement in the marketplace should be seen less as a legal contract, more as a rental transaction. Neither party has much confidence that a contract can be enforced by law. Any deal signed with a ruler who has external support remains good only for as long as that support continues at existing levels. If it is withdrawn or scaled back, rival elites will want to renegotiate. In Congo, therefore, we would expect the departure of the UN force, MONUC, to be followed by new rebellions by leaders of provincial elites demanding a better deal. In Afghanistan, if US forces are expanded and the Karzai government negotiates a deal at the moment of its greatest strength, it will last only as long as those particular market conditions persist. When American troops begin to leave, bargaining will resume, most likely with violence on either side. In both cases, the presence of international forces and international assistance, of unknown duration, introduces uncertainty into the political marketplace, which makes a solution more difficult, not less.

Elsewhere, international engagement is either intended to weaken a government, or has the effect of bidding up the price of the opposition's allegiance. This is the case for Darfur, where international ostracism of Khartoum and the readiness of the international community to give a platform to poorly organised rebel groups with little political or military capability has over-inflated the price which rebel leaders believe they can charge. During the peace talks, the rebels figured that the Save Darfur campaign would allow them to demand a higher price than anything on offer from Khartoum, while the Sudanese government negotiators anticipated that US commitment to the draft agreement would render the activists irrelevant, so that they could pay a lower price for putting an end to the rebellion. The government feared a nationwide inflation of the price of loyalty, with copycat demands from other provincial elites for a similar deal – just as the Darfurians' demands had been increased by the precedent of the southern Sudanese peace deal. The rebels continue to hold out, expecting that intense international interest will translate into a high price, while the government continues to regard this as a speculative bubble. Should the international community withdraw from Darfur, the government will be proved right. It will be tempted to renegotiate the deal to its advantage, probably by using force. The faithful implementation of any agreement reached today depends on international engagement continuing at much the same level.

In a conflict in an institutionalised state, whose adversary will also, typically, be institutionalised, peace negotiations can take the classical form. It would have been possible in principle to negotiate a workable peace agreement between the president of Sri Lanka and the leader of the Tamil Tigers: had the two reached an agreement in good faith, it could have held on the basis of the political decisions they made. It was possible for the force commander of the UN mission in Eritrea and Ethiopia to deal solely with the chiefs of staff of the two armies, knowing that the orders handed down by them would be respected. But in Sudan, Congo or Afghanistan, the ruler is only as strong as today's bargains with members of the political elite, who retain a great deal of autonomy. The deal will be voided when circumstances change, and to the frustration of international peacemakers and peacekeepers, a commitment at the top does not translate into action throughout the system. This is even true of something as elementary as a ceasefire. The president may announce a ceasefire, but the generals in the office of the chief of staff will need to meet to agree what this means in practice and to negotiate with the provincial commanders, warlords, police chiefs and militia commanders as to how they should interpret it.

The temptation is either to accuse the leader of perfidy or to recognise that there is a structural weakness in the system, requiring a bottom-up approach. The peacemaker may draw the implication that the conflict is compounded of different local conflicts, each of them in turn compounded of micro-conflicts, and that peace is made by dealing with each one at its own level. This is how international mediators, including a growing number of conflict-resolution NGOs, become involved in local peacemaking. Where there is international concern, more senior diplomats become engaged too. Special representatives of the secretary general or even UN Security Council ambassadors may nowadays try to micromanage district-level disputes or even village conflicts. The difficulty with this approach is that every intervention is a distortion of the market. The more local the engagement, the greater the relative weight of the international actor involved, and hence the greater the distortion. An international mediator will never be as knowledgeable, skilled or patient as a local actor and will invariably be drawn towards one side or the other. Given that every agreement reached is good only for that particular set of conditions, these interventions are inherently unsustainable.

The end result is that peace operations become entrapped. They cannot resolve conflicts and the more they try to do so, the greater the part they play in the dynamics of the marketplace itself, meaning they can't now withdraw. Many peacemakers become disillusioned: they see missions without end. Cynicism spreads from the fringes to the centre of these operations, with more and more people asking, 'What are we doing here?' In turn, smaller-contributing countries, who may have sent soldiers out of a sense of solidarity or principle, are likely to keep them there less because of their original political motives than because of the rewards of remaining loyal to the superpower (allegiance-pricing is at work among the intervening states as well). The guiding purpose in Afghanistan will soon become salvaging Nato, and in Sudan, saving the UN's face – if it has not already done so.

The spread of norms of 'good governance' – which requires adherence to international standards of government and involves supplying aid for elections and NGOs, conferences, training workshops and educational opportunities – has contributed to a worldwide spread of democratisation and civil society. The growth in conflict resolution has helped reduce the numbers of conflicts and people killed. But it also reduces the particularities of national political cultures and enables the monetisation of political markets. It takes a state with considerable self-confidence and financial autonomy to resist the pressures to play this game. Those that hold out against the intrusions are often unsavoury – North Korea and Burma, for instance – but there are two sides to international intervention and peacekeeping operations, as there are to globalisation itself.

The likelihood of dollarised political marketplaces becoming capable, legitimate, institutionalised states is receding. In retrospect, there was a brief 30-year window after the Second World War, when conditions were propitious for building functioning states where none existed before. More recently, the forces of globalisation have made it far harder, perhaps impossible, for would-be states to join the club. The dynamics of the dollarised political marketplace, which are more powerful than international state-building practices, are entrenching a pattern of deregulated governance in global borderlands. The recent trend towards representative politics, pronounced in African countries without strong state institutions, will be hard to sustain under these pressures, and is likely to be reversed. The prospects for building democratic states in central Asia are fading too. Instead, the logic of the political marketplace points towards continuing low-level conflicts, part criminal and part political, in which violence serves as a principal bargaining tool in an auction room of allegiances.

Places as diverse as Afghanistan, Nigeria, East Timor and Somalia may appear remote from the circuits of international capital yet they are no less globalised than Singapore or the Netherlands: it's just that their global connections are the kind we prefer not to acknowledge. The rich world's demand for recreational drugs, combined with its politicians' insistence that supplying this market is a criminal activity, funnels billions of dollars into unregulated political marketplaces. The global war on terror disburses further billions through poorly monitored loyalty payments in the most volatile parts of the world. Western governments spend significant sums trying to establish functioning sovereignty amid the factional politics of 'fragile' states, and civil, inclusive and affordable patronage systems are being swept away. For the unfortunate populations of these countries, this is a loss just as devastating as the weakening of state institutions, and one less recognised.

RR's final comment: If the illiterates who rule us were to read this, then they might understand why they either need to commit to staying in Afghanistan indefinitely, until they are forced out - or do the sensible thing, and stop propping up this corrupt government, and get out now...
30. 31. 32.