Federico Battera.

Abstract

This paper presents an overview of the trajectory of the State and Democracy system in Somaliland (former Northwest region of Somalia). It investigates the historical process and its nature since the 1991 declaration of independence. It analyzes the character of the State construction, its political culture and its main constraints and needs in a comparative perspective with Somalia. Since 1991, Somaliland has been able to re-build a consensual system rooted in its own culture by combining imported institutions with informal notions of governance. Bad legacies of the past--warlordism, authoritarianism, predatory attitudes, etc.--and the introduction of multipartitism could reverse the development and consolidation of the democracy and of a democratic culture. However, such system will endure if it will maintain its main features: exibility, localism, and consensualism.

Introduction

Since the Barre's flee from Mogadishu in January 1991, Somalia became a paradigmatic case of State failure. After the collapse of the former regime, Somalia was the recipient of several of peacekeeping and state-making operations supported by the International Community. However none was successful. The UN intervention, "Operation Restore Hope" (1992-95), was not only a fiasco but has left Somalia in worse condition than before. The U.S. withdrew its support in 1994 and has since then become averse to sending U.S. troops for peacekeeping missions (OTTAWAY, 2002). Other attempts to enforce peace from the top supported by the International Community were mainly centered around the warlords--between 1993 and 1997 (from the first Addis Ababa Agreement to Cairo Agreement)--or opened to 'civil society', as the 'community-based' national initiative of Arta, which led to the unsuccessful making of a 'Transitional National Government' (TNG), or the more recent Eldoret/Nairobi (2002-04). Conferences that tried to convene civil society representatives and the warlords with doubtful results: they did not provide Somalia with stability and sometimes they prolonged divisions, factionalism and warlordism. As a result, after thirteen years of attempts to create a viable Somali state, Somalia has been unable to maintain a stable and legitimate government. This article will address the issue of state-building and will focus on Somalia and Somaliland (1). I will start my analysis defining state-building concept in divided societies. Divided societies are societies where conflict has led to fragmentation, and internal mechanisms for conflict resolution have failed (2).

Traditionally, anthropologists describe Somalia as a society with a segmentary lineage system. According to anthropology the segmentary system is usually related to a 'stateless' condition without a central governing administrative body (FORTES & EVANS-PRITCHARD, 1940). Though, modern divisions are certainly linked to and fostered by pre-modern segmentary conditions, contemporary societal divisions that affect the contemporary Somali society are more complex and more directly related to the struggle for power and to the character of the state in the recent history. Divided societies are more likely those which have lost or are not able to make use of their own mechanisms to curb and manage conflict among groups. This seems the contemporary condition of the large part of the Somali society, with the exception of Somaliland where the state-building process has been more successful. For this reason I have treated Somaliland and Somalia as separate cases.

In the following sections, I will describe the characteristics of the Somali and Somaliland state experience from the colonization to the present. As in other African cases, de-colonization in Somalia was accompanied by the promise of democracy. The independence processes, in both the former British Somaliland and the Italian Somalia, were largely 'democratic' in the sense that people began to participate in the political process through the two major political parities, the Somaliland National League and the Somali Youth league. The failure of the Somali state was also the failure of the promise of democracy and Somalia thereafter changed into an authoritarian state.

Authoritarianism, centralism and the predatory character have been considered as the bad legacies of the previous regimes. Authoritarianism and centralism are colonial legacies; the predatory character was a feature of the Socialist regime that endured after its collapse of 1991 under the new form of warlordism. Such predatory conditions survived in Somalia after the 1991 collapse, but did not persist in Somaliland, which strove to develop a sort of what I have labeled as either progressing democracy or a consensual democracy. In the fourth and fifth sections, I will include analysis of another Somali case--i.e. the Puntland State of Somalia--as an unsuccessful example of state-making process and pacification. The Puntland case, which started its process in 1998 under the best auspices and in better starting condition in respect to the Somaliland case, allows me to underline the variables that help to explain the Somaliland success.

Legacies of authoritarianism and predatory culture define part of the so-called Somali 'civil society'. I will call it in the sixth section the 'uncivic civil society'. The term 'civil society' has become a sort of magic word in contemporary studies on democratization in developing countries. This cliche enables the state in developing countries to gain access to the UNDP or WB managed aid. However, using a model that focuses on an idea of 'civil society' is not always the best way to assess democracy in Africa. I will, therefore, analyze 'civil society' critically. In fact, in the case of Somali society, predatory and authoritarian attitudes survive side-by-side with a culture of mediation and anti-authoritarian attitudes. Democracy does not mean only check and balances, division of powers, free press and elections etc.; democracy is nurtured by practice and culture: willingness to compromise and reconcile, pragmatism, civility of political discourse, when endure, strengthen democracy (DAS GUPTA, 1995: 263-321). I will consider those attitudes as the Somali 'social capital', what Tocqueville referred to as "habits of the heart and the mind" (HYDEN, 1997: 3) (3).

The existence of a political class devoted to democracy and not simply to the accumulation of power is also important. The contemporary political class in Somalia is made up of low profile politicians of the Second Republic (1969-1991) or by a military class devoted for the past two decades to ravaging and spoiling the country. This has been, so far, one of the major problems to overcome differences in the rest of Somalia.

Localism, which is strongly embedded in the Somali political culture, is also a crucial factor. If correctly addressed, localism does not necessarily lead to fragmentation but help to articulate the political demand of the periphery to the center. During the First Republic, neither the periphery nor the center prevailed. A sort of political exchange developed between the two. But the maintenance of such equilibrium was totally reversed during the Second Republic: the political exchange was altered and the center developed a policy of exploitation towards the periphery (SAMATAR, 1989). I will discuss this issue in the fifth section. I argue that localism has been so far a resource for Somaliland, but the rest of Somalia failed to address the issue.

This has been another major obstacle to national reconciliation. I will conclude by suggesting that by looking at the failed state-building process in Somalia we learn that it is better for the international community to assess the situation before starting economic reconstruction projects. Sometimes by supporting a divided country efforts to achieve a stable reconciliation may be hindered than helped.

State-Building in Divided Societies

The term 'state-building' developed in the 1960s to describe the process taking place in Western Europe (4). The major aim of the state-building was to fuse the disparate population elements into a congruent whole, forging new loyalties and identities at national level at the expense of the particularistic ones turning subjects to citizens. This process took several decades to complete. State-building in non-European countries began during the period of decolonization. Colonialism introduced Africans to formal state procedure, legal frameworks, and through decolonization, to democracy. Political scientists were aware of the risks of extending the concept of state-building to African countries because of the existence of ethnic divisions. Ethnicism and tribalism were discredited as dangers for the state-building process and the single-party system was justified as a transitory but essential instrument to achieve national integration and avoid chaos.

The state crisis in Africa occurring in the 1980s led to a growing interest in state and nation-building in the region. Scholars were interested in whether the concept of state-building would work in a fragmented context. In particular, the introduction of the neo-patrimonialism concept for describing political regimes in Sub-Saharan Africa (MEDARD, 1982) and in Arab/Islamic countries rouse doubts about the possibility to develop an autonomous 'civil society'. Analysis focusing on the nature of state and regimes centered on power relations left no room to assess political change. The new wave of democratization during the 1990s led scholars to be more optimistic. It is true that many countries in Africa were failing dramatically but others succeeded and held several elections. Again, the Ivorian crisis of 1999 drove attention about the fragility of the state-building process in Africa and the risks that a state had to face before being firmly established.

In 2001 after the collapse of the Taliban regime, state-building operations regained a new momentum. Because of the vacuum left by the Taliban movement, a new government had to be created by the international community. The international community was determined to convince the public that a national government could be reconstructed through aid and external supervision. Optimism prevailed both in the U.S. and the U.N. Since the beginning of the 1990s the UNDP has dedicated several reports to building a new model for addressing the problem of state crisis. The report has indicated that the essential methods to link democracy and crisis management are building institutions, creating a rule of law and de-mobilization (5).

However reality sometimes differs from theory. As far as the Afghanistan reality is concerned further reports moved from the starting optimism and experts warned about the risks of a new Somali case (6). Obstacles were above and beyond identified as: the absence of a strong national identity, factionalism, the lack of an effective state culture, the incapacity to absorb economic assistance and the conflicting interests of the neighbors' powers which in the case of Afghanistan were prompt to destabilize the Karzai's government.

These obstacles undermine the development of the preconditions needed for working state-building processes. The case of Afghanistan is noteworthy because it shares many characteristics with Somalia. Afghanistan, like Somalia is deeply divided and state-building efforts are often thwarted by the differences in interests between several factions. Therefore, minimum level of agreement on common interests is often impossible to achieve. A second similarity between the two countries is that both have been prey to government authoritarianism and warlordism. Warlordism reproduces on a micro-scale what previous governments exerted on a national dimension. Communities become hostage to warlords though they are simultaneously deeply involved into the same economical mechanisms that lead to warlordism and divisions. In such conditions, severing the factors that feed warlordism and finding alternative solutions for hostage communities is extremely difficult.

Mismanagement of public resources, authoritarianism and the criminalization of security apparatus are strongly interrelated factors (BAYART, ELLIS, & HIBOU, 1997). Finding way out from such culture is very difficult. Sometimes assistance from outside is crucial but the more lasting solutions can be found only stimulating the best qualities of local culture. Both Afghanistan (7) and Somalia offer examples of internally pacified communities. I will later illustrate the Somaliland case as a successful example of pacified community. Pacified communities are those that were secured to peace and administered through bottom-up participation. Informal mechanisms to curb conflict, restrain accumulation of power and power-sharing formulas are also important, as I will later expose (8). Somaliland offers a good example of this kind.

The Somali State Experience

Somalia existed as a state only starting from the era of colonialism. Before colonialism, Somalia was ruled by scattered tribal sheikdoms. Historians and anthropologists described the Somali society as stateless, characterized by a strong de-concentration of power (LEWIS, 1961). Before colonialism, failures of centralizing attempts were also due to the resistance of a strong segmentary attitude (9). Somalia experienced colonialism under both Italian and British rule. Both the administrations combined direct rule (through District Commissioners and Residenti) with a degree of autonomy in the peripheral areas10. Both recognized and compensated chiefs ('Aaqils and Capi-stipendiati) with regular salaries. In particular, both the 'Aaqils and Capi-stipendiati were in charge of keeping peace providing the link between the District Commissioner and the people of the district.

The independence movement brought together the two legal systems established under the British and Italian rule. Although the merger of the civil and common law traditions in one judicial system created some difficulty, significant progress has been made in the administration of justice.

The First Republic (1960-69) was characterized by a dominant party system--after the osmoses of SNL to the SYL and the co-optation of major representatives of the HDMS into the SYL--but a good degree of free political competition continued. In particular, major candidates retained their local power and were able to negotiate their position into the dominant party either prior to or after elections. Re-alignment of positions was justified by loyalty to ones constituency rather than loyalty to a political party. After independence, the SYL changed from an ideological party committed to the unity of the 'Greater Somalia', into a de-ideologize party made up mainly of a congregation of influential notables with strong local interests. Corruption flourished also as a means of exchange between the center and the periphery, since government policies scarcely touch the periphery.

The Second Republic (1969-91), which was placed in power as a result of a military coup, was legitimized by the desire to end corruption and a stronger commitment to development and renewal of the ideal of a Greater Somalia. A new judicial model inspired by socialism in Eastern European countries was also established. This deeply affected the organization of the state. The introduction of the National Security Courts and the enforcement of the Public Order law of 1964 replaced the principle of an independent judiciary as it was seen incompatible with the socialist theory. The National Security Service (NSS) frightened society.

This move towards a more authoritarian government began the predatory trajectory of the Somali State. Corruption emerged again as a result of the increase in foreign aid and military assistance from Italy and the Gulf countries. Somalia had become important to the international community because of its position in the US/USSR confrontation over the Horn of Africa. Corruption under an authoritarian government transformed the balance between the center and the periphery, since the center was empowered by extraordinary financial and military resources. The regime lost its legitimacy however, and developed a predatory stance against its own people. By the end of the Cold War, with a collapsing regime, military apparatus mainly recruited along clan basis started to rely on the territory to endure. As a result, warlordism became a powerful governing force.

From the Collapse of the Post-Colonial State to the Emergence of Warlordism

Somalia's experience as an independent state was negative to say the least. The Somali predatory culture has its origins partially from tradition but predation under the Barre's regime was something new from the past. Traditionally, predatory practices were one of the mean of redistribution between pastoral groups, which raided each other during dry seasons. This was also an instrument that ensured to control the accumulation of wealth. This strategy occurred when groups tried to impose their hegemony on other groups or on vital territory, violating previous rule and agreements. But predation was not the traditional method governing power relations among groups. Negotiation, agreements, reciprocity and intermarriages were the preferred methods of control.

Traditional rules also distinguished between political and military leadership. Usually, the former was more stable, and often inherited; the latter was contingent relying on particular situations. Examples of warlordism appeared during the 'civil war' which ravaged Somalia during the colonial expansion. Warriors expelled from their own clan (dabato) offered their military service to the colonial authorities, the Sultan of Hobiyo, or the theocratic movement of the 'Mad Mullah'. Sometimes they set up composite gangs of warriors (bur-cad), which spoiled the territory (KEENADIID, 1984: 147, SAMATAR, 1982: 160). This however was a violation of traditional rules--xeer (SAMATAR, 1982: 199).

Despite its connection to tradition, warlordism is more directly tied to modern state and the mechanisms that ruled power relations. The Barre's regime took advantage from extraordinary foreign aid during the 1980s. State sovereignty was necessary to receive the aid that was managed only from Mogadishu. During this period, Barre was able to combine repression with a policy of co-optation of key clan members. Aid became the crucial resource in the creation of new patrimonial ties. In the 1980s aid lessened after the USSR severed its ties with the Horn of Africa and the Somali regime took a more repressive stance. The regime's stability was dependent only on the army recruited on loyal clans. With the end of the external support, the army collapsed into different units, which exerted their authority on the territory. Commanders turned to be warlords with their private army despoiling both territory and population. Such policy began with the repression in Somaliland, when the troops were left alone without logistic support.

With the collapse of the regime, warlords set up their own territorial basis relying mainly on clan support. Though gaining autonomy from collapsing state nonetheless warlords remained connected with the external world through informal and illegal channels. Much like the mafia, warlords need to be involved in global markets and politics. The UN intervention, operation 'Restore Hope', was unable to sever this relationship between the warlords and the global markets.

In 2000, UN shifted from a local approach--the so-called 'building blocks' approach which have sustained local reconciliation between 1997 and 2000 (Somaliland in 1993 and 1996-97, Puntland in 1998)--to a new national reconciliation approach calling for an initiative that was summoned in Arta (Djibouti). Such initiative, eventually, led to the creation of a 'Transitional National Government' (TNG). Again such initiative was based on the illusion that national reconciliation was conceivable in a deeply divided society. Such initiative appealed to Somali 'civil society', NGOs, the Diaspora, the 'traditional' leaders, etc. but trying to conceal the interests of a composite society with those of the warlords (11). Inevitably, the failure of Arta led to a stall of creative initiative for Somalia and a certain disillusion over the future of Somalia grew (12).

The Somaliland Path to Democracy and the Puntland Sliding Back to Authoritarianism

After ten years of attempts at national reconciliation, Somalia remains divided, conflict persists, and the international community seems unconcerned. Many regions in the country are hostage to warlordism and factionalism, while others, after having achieved a good interlude of stability and institutionalization, are in a state of civil strife. I will illustrate in this section how Somaliland has become the most striking example of pacification in the area though receiving little help from abroad.

The Barre regime was driven out of Somaliland in the beginning of 1991. Reconciliation initiatives among different Somaliland clans helped to create the Burco Congress in May 1991, which solidified Somaliland's independence. The Burco Congress represents the beginning of the state-building process in Somaliland. The other important step aiding the creation of state institutions occurred during the conferences of Booroma in 1993 and Hargeyesa in 1996. The Booroma Conference (January--May 1993) addressed the adoption of a political system of governance, the establishment of a new central administration and the issue of creating a power-sharing formula for all Northwestern communities. The Booroma National Conference paved also the way for a new constitutional process, whose outstanding outcome was a National Charter (13).

The National Charter stressed the role of Guurti, (the Assembly of 'titled' elders) (14). The SNM (Somali National Movement) leadership, the movement that led the liberation struggle and the region to independence, had passed to the Guurti its mandate powers during the Conference. The Guurti became therefore very important and powerful, able to include all the communities and clans into a regional institutional framework. Reconciliation conferences (shir beeleedyo) were held throughout the region between 1989 and 1995 (FARAH, 1998: 18) and a new political system based upon Somali cultural values was started. Somaliland did not rely on the international community for support but looked to its own mechanisms of reconciliation. The practice formalizing the role of the titled elders was not the only method used to achieve reconciliation. The Presidency, which was selected by the Guurti during the Booroma Conference, played also an important role of mediation starting a practice of consultation of the most prominent paramount chiefs (15) on the most important and delicate issues. A good example was the ability of President Igal to freeze the issue of the belonging of Sool and Saanag--the two eastern regions whose belonging is contested by the neighbor Puntland State of Somalia--when this was at stake between the Garowe Conference of 1998 and the Puntland crisis of 2001 (16).

Local guurtis, sometimes represented by more prominent elders than the National Guurti, helped in sustaining the effort of the state to satisfy the requests of the periphery when this were not fully channeled and satisfied by 'official' representatives. In those years, local administrations still relied on the appointment of the President, as regional and district councils were not yet established. However, local guurtis and other informal pressure groups and institutions were able to offer alternative channels of representation.

A good part of the achieved stability has to be nonetheless attributed to other factors. First of all, the attitude of the SNM forces, mainly recruited among the Isaaq (the major clan), from retaliation against Somaliland clans who sided back to Barre during the long war that previously had affected Somaliland (PRUNIER, 1998: 226). In those years, SNM was able to not feed war against Barrism with Isaaqism (17). This unusual ability of SNM, if compared to other anti-Barre forces (most of them turned to become instrument of the warlords after the regime collapse), to rapidly turn to peace is due to the prevalence of the civil wing over the military wing. Since the Booroma Conference of 1993, SNM military wing made a step back living the power to the civil wing and the Guurti. Again, during the Hargeysa Conference (1996-97), which followed a new outbreak of intercommunal conflict in Somaliland, the civil wing prevailed over the military wing. As a result, SNM dismantled before multipartitism was fully achieved in 2001 (18).

Secondly, an important factor of stability has been the commitment to reconciliation and democracy of the new political class, which emerged from the dismantlement of SNM under the leadership of President Igal (19). The Somaliland political class grew around the presidential entourage but split in several factions to re-assemble around new parties which were formed at the eve of the local elections of 2002. But what is strikingly different from other Somali realities is that since the Hargeysa Conference the Somaliland political class has avoided recurring to violence to promote its own interests.

A different trajectory has developed in the neighbor Puntland State of Somalia. In 1998, a Constitutional Conference was held in Garowe. A transitional Charter was approved for a term of three years initiated by Col. Abdullahi Yuusuf who was elected. The 'titled elders' (Isimo) were also included in the process following the Somaliland model (BATTERA, 2000). Apparently, compared to Somaliland Puntland started with better 'ethnic' conditions. Most of its population belongs to the same clan, the Majeerteen/Harti. Secondly, most of the area had been unaffected by civil strife that accompanied the collapse of the Somali state. Almost right after the withdraw of the Barre's army, Puntland had enjoyed a long period of stability and peace. Only during 1992 it experienced a short but intense conflict over the control of the region, between the SSDF (Somali Salvation Democratic Front), and al-Ittihad al-Islami, the strongest Islamist Somali organization. The Conflict was won by the SSDF and the region became rapidly peaceful with the reinstatement of the 'traditional' leaders (Isimo).

Once the conflict was over and peace was restored, the split of the SSDF between the two most important military/political leaders, Gen. Mohamed Abshir Muuse, Chairman of the organization, and Col. Abdullahi Yuusuf gained attention, though the region was kept from intercommunal conflict, progressing toward the Garowe Conference of 1998. But the struggle for power that had characterized previous rivalries continued to work underneath until the outbreak of 2001, when after the three years of transition an institutional crisis erupted violently opposing the dismissed President/Prime Minister, Col. Abdullahi Yuusuf, to his major opponents, Jama 'Ali Jama. Avoided during more critical years, civil strife erupted during the crucial phase of transition towards state institutionalization. So, if we look to the Northeast experience and we compare it with that of Somaliland, what went wrong?

The government's task was only to set up a local administration in order to secure minimum services and to gain the benefit from foreign aid, but the government was unable to settle intercommunal rivalries into a working framework of power sharing (20). The Transitional Charter formalized the government, however without a shared practice of mediation, constitutional rearrangements proved to be useless law exercises (MATTEI, 2000). Unlike Somaliland, Puntland failed also because the civil leadership was subordinate to military power. A sort of blank cheque to rule without any balance was given during the Garowe Conference to Col. Abdullahi Yuusuf without any serious commitment of the 'civil society' to democracy (21).

Local Political Models, Democracy and the State: The Pillars of a Flexible System

Somaliland is a remarkable example of the supremacy of civil power in a region largely dominated by military power (Eritrea, Ethiopia, Sudan and most of the Somali entities). Recent experiences show us how hard is to turn warlordism--as in the Liberian case--into a legitimate civil power also when this seems to be the result of free elections. Common interests to stay together are essential but are not enough. Nor warlordism, though sometimes internationally legitimated--as in the Liberian model--is an acceptable mean to get access to power. The Somaliland construction, though very far from accepted standards of transparency and participatory and accountable budgeting, is even more remarkable in the sense that: first Somaliland was able to harmonize interests, and second Somaliland practiced a political formula, centered around the good traditional practice of shir (22) and negotiation, that proved to be able to freeze and control tensions and rivalries when these arose. Such formulas when not a simple option but when respected and widely recognized contribute to strengthen state legitimation and, in perspective, democracy. Legitimacy is, indeed, fundamental in order to give stability to state constructions. As Prof. Mattei (2001) affirms, "(..) Legitimacy is the most powerful of the informal institutional constraints. The authority of the leader when considered legitimate is internalised in the brains hearts and souls of the armed men making them willing to obey. (..)".

Somaliland reconciliation succeeded because of the reconciliation of opposing interests. Once the consociational practices developed and structured, Somaliland gained legitimacy. The bad example offered by the rest of Somalia did the rest.

At the national level, on the contrary, consociational system could hardly work, because, as Prof. Horowitz pointed out, consociational democracy is a practical solution to ethnic conflict in countries that have already reached a certain level of consensus (BARNES, 2001: 98) (23). This is the case of Somaliland, but not of Somalia (24). Obstacles are not only in the serious differences from North to South but also in the divisive character of Somali society after ten years of civil strife.

There is an inner weakness which affects deeply divided society, characterized by communities internally divided by a dramatic and dynamic struggle for leadership: a deficit of representation. Once an agreement between parties is achieved, parties tend to divide in smaller fractions, and fractions withdrew from the agreement. Something like that happened several times in the recent Somali record of national and regional agreements. On the contrary, this situation is unlikely to happen when parties are internally pacified communities.

Formal and informal practices of consociational kind offer in the case of Somaliland a way to representation. In the case of Somaliland, the number of representatives and the constituencies were negotiated before the 1996-97 Constitutional Conference of Hargeysa was held. Hargeysa formalized the composition of the Upper and the Low Houses, though alternative channels of representation continued to be guarantee (25).

Consociational political attitudes work effectively not only when a minimum level of consensus has been reached but also through the de-concentration of power. Somaliland is a Presidential government but the Presidency holds only formally extraordinary powers. Though the law on local councils encountered obstacles because it menaced the Presidential prerogative to appoint local governors, a de-facto de-concentration of power has been in force even before the local elections of 2002. The same association of peripheral regions--Sool and Sanaag for example--is continuously negotiated through the concession of a variable autonomy (26).

In the introduction, I have argued that Somaliland is moving on the path towards democratization. I defined Somaliland as a progressing democracy. Political science has been recently challenge by the great diversity of postauthoritarian regimes. The result has been a great conceptual innovation, but with the risks of a growing confusion (COLLIER & LEVITSKY, 1997). I do not intend to contribute to this conceptual confusion by introducing a new version of democracy. I will add only that classify recent African experience of democratization is always difficult, giving the differences in culture and the complexity of interaction between different stratifications. Classifications may be certainly useful to make comparisons between political regimes but sometimes it is risky to trap one particular regime into a closed category. I prefer to look to democracy as a process, which never end. In particular, in the underdeveloped world and in the countries, which have faced a deep state crisis or exit from a civil war, democratization is more likely an unstable and irregular movement forward and backward.

Based on the Dahl categorization of polyarchy (1971), Somaliland has achieved during the last ten years a good level of civil and political liberties. However while there are organized political parties, it is doubtful that Somaliland has reached an all-inclusive level of political participation in the selection of leaders and policies. The practice of shir is only theoretically an egalitarian instrument of participation. Shir is hierarchically organized. Patrimonial allegiances, which organized the political life during the First and the Second Republic, survived and ordered shir decisions as well as 'national' politics. Furthermore, female participation is practically excluded in the traditional shir practice, though women were allowed to vote in the last presidential election (2003) and few female representatives were selected in the former Low House.

The informal mechanisms that I described above although ensuring a good level of pacification on the other hand may produce distortions in the political realm. In Somaliland, the same informal factors that support constructive consociational practices nurture neo-patrimonialism. And neo-patrimonialism could be an eroding factor of democracy. It could upset the balance strengthening governments against oppositions. Because democracy also promotes political change, neo-patrimonialism limits the prospect of change.

For such reasons, some scholars have questioned the real possibility of political change in Africa after the transitions of the 1990s (BRATTON & VAN DE WALLE, 1994). In particular neo-patrimonial networks have proven able to survive formal regime change.

In Somaliland the newborn multipartitism has been centered on two parties, the Kulmiye and the Udub (27). In the last presidential election informal agreements contributed to frozen contestation of the electoral counting by Kulmiye, which lost the election by a few votes (28). Things seems to be resolved by the promise of the winning party--Udub--to take up the electoral expenses of the opposition. This could be a precedent for future informal arrangements of consociational nature, which were not envisaged during the transition from a consensual system to a multiparty regime.

Party discipline is not very strong as representatives are more loyal to their own constituency than to their own party. Co-optation is more likely to occur individually with movements from one party to the other. Giving these conditions, opposition parties could fail to take roots, and the multiparty system is likely to become a dominant-party system.

However, these factors are not enough to label Somaliland regime in a different way (semi-authoritarian, pseudodemocracy, etc.) (29). In the case of Somaliland it is difficult to assert that neo-patrimonial networks are all monopolized by the center. The center is incapable of exerting direct control of the economy. This allows powerful competitors to appear. While incumbent government strives to endure, it is forced to accommodate these competing powers.

A leadership loyal to the democratic system, and flexible, accommodating, and consensual leadership styles have notably contributed to the democratic development of Somaliland (30). Flexibility, consensualism, etc. are characters that are deeply embedded in the Somali political culture but failed to work in the remaining part of the country, which continued to be driven by factionalism and warlordism.

State's Legitimacy Decline and the Road from Zero

The collapse of the Somali state did not occur because of the Barre's flight from Mogadishu. It began earlier when the government declared war against its own people during the late 1970s. Somalia then experienced corruption and criminalization of the state (31). These conditions led to the erosion and the decline of the state's legitimacy, and to what Clapham (1996) described as the externalization of state legitimacy. Because of this trajectory state-building must now start from scratch. The start from scratch means to first find out the common interests, which the government must rely. Roland Marchal (1996) in his excellent work on the Somali business class has described the deep ties between newborn administrations in Somalia and the business community. Administrations start, when the local business community agrees to set aside their differences and to set up common institutions to deal with ordinary business situations. In the case of Somaliland, Puntland and the TNG administration in Mogadishu, businessmen played an important role in the process of creating the local administrations. The establishment of a strong business class could be therefore an important starting point in the state building process, however it does not correct the problem of the state's lack of legitimacy. Somalia experienced already business-oriented polities during pre-colonial times--what Marchal (1998) calls Abban governance--and something similar happened during the Renaissance in Italy with the development of Signorie. However, it must not be underestimated the nature of the conflict in the South, the economy of war that feed it keeping in mind that new 'Abbans' largely benefited from war economy (32).

The problem becomes how to translate common interest into stable and legitimate rules that will be respected by the people. Somaliland succeeded to create the roots of a consensual democracy whose pillars are respect of an unwritten code of conduct and localism. Because problems are likely to become intercommunal, the elders played an informal but important role in preventing conflicts, which could rose between competing business circles33. Elders ensured mediation at the bottom of the system and the Presidency ensured the success of the system at the top, regulating the access to state benefits.

According to the Booroma Charter, the Chamber of elders was endowed only with a role of mediation, but the Constitution that resulted from the Hargeysa peace process endowed the Chamber of elders with an extended law-making role. Hence, the political system evolved to a sort of bicameralism. During the no-party phase that lasted until the establishment of multipartitism in 2001, the Chamber of elders sided with the Presidency and a good part of the House of the Representatives sided with the opposition (34).

Both, the role of supervisor acquired by the President and the bicameralism contrasts the new provisions introduced by the amended constitution of 2001. First by the fact that Somaliland is a Presidential system, with a President now directly elected by the people chosen amongst several candidates. Before the introduction of the election of the President, the Chamber of elders supported him due to his role of supervisor. The introduction of multipartitism could reverse such a formula. Multipartitism could cause polarizations across the two Houses and hinder the system's ability to ensure mediation at the top. Nevertheless, recent experiences demonstrate that the system can balance itself. The consensual attitudes and informal rules that governed in the past, if respected, will take care of the system in the future.

Localism also contributes to the further workability of the system (35). Clans are the 'natural' constituencies. With the de facto independence of Somaliland, demands to re-draw the districts coherently to the clan territories have been recorded. As a consequence, new districts were established even before the approval of the law on Local Councils. In 2001, the House of Representatives passed the law and local council elections were held in 2002. Before the approval of the law, local administrations relied on the appointment by the Presidency. However, local guurtis and other informal pressure groups and institutions were able to address political requests and avoid violent escalation of the protests. Localism has become, therefore, a resource for Somaliland while has been one of the major obstacle to national reconciliation in Somalia (36). The main question to answer remains therefore why the same factors that performed positively in Somaliland failed to work in Somalia.

Social Capital & the Legacies of Authoritarianism: the 'Uncivic Civil Society' & the State

In Somaliland, both the consociational mechanisms that ensured representation at the upper levels--the House of Representatives and the Guurti--and the informal ones that governed the distribution of executive posts were shaped by the same principles that consensually balance the center with the periphery. The periphery perceives the distribution of executive posts as the shortest way to gain access to development. According to a consociational and consensual logic, the informal mechanisms that governed the access to the 'national' cake were also capable of freezing conflicts and had the merit of de-politicizing developmental issues. Somali political culture is certainly authoritarian. However it combines authoritarian attitudes with a propensity for debate, political game-playing and the interdependence of the actors. As in the Coulon (1995: 512) description of Senegalese political culture, Somali 'big men' (37) can topple when unable to provide their followers with the benefits they expect to receive. Though patronage relations are by definition unequal, this dependence of the patron from his followers is also a way to control power by the weakest link in the chain.

Two models could describe the Somali political culture: the reconciliation model and the mobilization model. I borrowed both from Apter (1965). The reconciliation model derives from a very strong segmentary attitude as described by Ernst Gellner (1990) in his work on the Berbers of Morocco. This attitude corresponds exactly to what Ioan Lewis (1961) defined as 'pastoral democracy' and relies on the principle of not giving one person too much power (38). When a single person holds too much power he loses his following and different counterpowers appears. In contrast, the mobilization model is based on the belief that power should be concentrated in one person. This model has characterized the Somali political attitude towards the outside world when Somalia was strong enough to expand rather than characterizing Somali political attitude in case of an external aggression (39). As far as the internal effects of the mobilization model, aggressiveness towards the outside world tends to correspond to strong authoritarianism.

Probably, the entire recent Somali State experience could be explained as a failure in finding widely acceptable solutions to the risks of the concentration of power. Somaliland is certainly not the case, but so far a successful experiment to find solutions for power concentration. Unfortunately, it has not been so efficaciously followed by other Somali cases to whom the second model still seems more suitable.

In a certain way, warlordism is still a de-ideologize sub-type of the mobilization model. The mobilization model encourages aggressiveness, authoritarianism combined with predatory attitudes. Apart from Somaliland and partially the TNG it characterizes most of the warlords' fiefdoms.

During the First and Second Republics, Somalia has incorporated both models into its political process. Neo-patrimonialism is a characteristic of both models but has lead to different results under each. Neo-patrimonialism is more likely a resource for incumbents than oppositions, since governments are in the best position to nourish patronages compared to the opposition. However, in the case of the reconciliation model, strategies are more adaptive and inclusive and neo-patrimonial strategies tend to reconcile governments with oppositions. On the contrary, in the case of mobilization model, neo-patrimonial strategies tend to lead to repression and exclusion. I look at neo-patrimonialism and its corollaries, (patronage, clientelism, and corruption) on one hand, and the inclination to mediate and to pursue compromises on the other under the same cultural lenses. In the case of de-concentration of power, as it seems the Somaliland case, public arena become the object of competition of different patrimonial networks with none of them able to prevail definitively. This encourages competitors to reconcile and find the best solution to share power instead of concentrating it.

Willingness to compromise and to encourage civil political discourse strengthens democracy (DAS GUPTA, 1995). Such inclinations are what I refer to as Somali social capital. Legacies of authoritarianism and predatory culture may survive, but the harm affecting Somali civil society can be reduced through the practice of reconciliation.

Civil society is considered by Toqueville as the strongest bulwark against the tyranny of the majority (HYDEN, 1997: 6). For liberals, civil society is considered as a separated entity from the state. Powerful civil society could become an important counterbalance of State and tendencies towards authoritarianism. Today the so-called 'civil society' has become a magic word enabling developing countries to get access to the benefit of aid programs. The number of NGOs present in a particular state has been used as an indicator of the development of the civil society. This equation is linked to bad reputation held by many Third World African countries where the state has been characterized as corrupt, criminal, and repressive. In the last ten years, also Somalia experienced a surge in the creation of NGOs as an attempt to take advantage of the influx of foreign aid. Observers perceive this trend as an expression of a free society (40). However, NGOs are not a fully autonomous phenomenon. On one hand, they are becoming a kind of interface with donors; on the other, their commitment to democracy is doubtful since the state/society relationship has been deeply characterized by neo-patrimonialism and authoritarianism (41). In such cases, NGOs may readily fall victim to the politics of corruption. Propensity to rely on powerful patrons is a characteristic also of many Somali NGOs. The number of NGOs has risen because they are seen as a shortcut to development. I call this propensity the 'uncivic civil society'. Such conditions are in opposition with the foundation of the idea of social capital.

Bad legacies, such as authoritarianism and neo-patrimonialism, make the difference between 'social capital' condition and 'uncivic civil society'. Somaliland is exactly in the middle. Consolidation of democracy requires changes in both state and civil society (42). The fact that modern state has been imported to Africa and is alien from African culture does not imply that African society has not been deeply shaped and affected by it. Consequently, changes may only be understood at the intersection between state and society.

Ten-Years of Attempts at National Reconciliation: Lessons from the Somali State-building Experience and its Implications on the Future

Ten-years of failed attempts at national reconciliation is evidence of a lack of international creative solutions for state crisis in divided Somalia. Foreign intervention could at best help to end civil conflict, but is not enough to secure a stable path towards reconciliation and development. External resources are still not enough. This brings us to the lessons to be learned from Somaliland's state-building experience. It is important to determine what solutions are transplantable to other countries involved in similar experiences. More importantly, we must recognize what Somalia can learn from Somaliland.

The situation in Somaliland proves that a country must use its own resources coupled with good will and a determination to solve the state crisis. These 'internal' factors are crucial to deal with crisis' causes. With the departure of Barre's army Somaliland started on its way toward peace and reconciliation. As in Somalia, Somaliland was at risk of becoming victim to warlordism but was able to avoid this outcome though the 1995 crisis in Burco. Finally, Somaliland discovered a formula for a successful state-building process by combining imported institutions with informal notions of governance rooted in Somali culture. Nevertheless, 'internal' factors are indispensable but could not be sufficient.

And now I come to the role of aid as it relates to the peace-building efforts and democracy. Since democracy is important as a legitimizing factor, aid could be crucial. Somaliland is certainly not a perfect democracy: though local actors have enough power to redress central policies neo-patrimonial temptations could hamper political change. Aid should be directed respecting the equilibrium between parties and avoiding to strengthen too much the central power.

Supporting Somaliland transition with aid may also help the rest of Somalia. Aid in the form of economic reconstruction packages, can be the carrot in an incentives' game where the stick could be threatening sides with withdrawal of benefits, human rights commissions, and only in the harder situation military intervention (43). In the case of Somalia, this incentives' game should be easily played now with warlordism has exhausted, and with warlords' resources reduced--mainly khat and weapons smuggling--if compared to other divided contexts (Angola, Sierra Leone, Colombia, etc.). In this framework supporting Somaliland--but not necessarily the Somaliland government (44)--should be part of the incentives' game. Somaliland offers to Somalia a way out from conflict and warlordism.

Regional inclination to conflict must also be addressed. Regional context has been in the last years more an obstacle towards reconciliation rather than a resource. Historically, both Ethiopia and Kenya have been threatened by a stronger and united Somalia. Both have tried in recent years to contain the threat by directly or indirectly sponsoring warlords. Because at present, the idea of a strong, united, and aggressive Somalia is only rhetorical I do not think that this would be a serious obstacle for national reconciliation if the international community would sincerely intervene as a guarantor for the regional stability.

A second important lesson is that successful local reconciliations are easier than national or regional reconciliation efforts (45). Also when local reconciliation processes are difficult, local agreements are more likely to be respected than national agreements. The number of actors and interests are reduced and their composition is easier.

These considerations lead us to another important point; state-building operations sponsored by the international community must start with and respect local experiences. When the UN abandoned the building-blocks approach--i.e. to sustain local reconciliation before supporting national reconciliation--it renounced to its best device against warlordism. Though also the more successful Somaliland process was not smooth and exempted from set backs and warlordism attitude of some leaders, the Arta process of 2000 started as nobody had considered that warlordism had taken root before. It started from the presumption that Somalia must be at every cost a united country, ignoring the nine-year Somaliland process and, though unstable, the Puntland process. As far as Puntland is concerned, the Arta process had a destabilizing effect. At that time the TNG worked openly to destabilize Somaliland, with no success, while successfully destabilized Puntland (46).

At last, the destabilization touched also the TNG and between 2000 and 2001 all the numerous divisions that ravaged Southern Somalia since the collapse of the State blew up again. New realignment between former enemies led to a new organization--the SRRC (Somali Reconstruction and Restoration Council)--tied up by the desire to reduce TNG to another faction. At the end of 2001, in fact, the TNG reduced its control over the territory only to portions of Mogadishu, some spots of the country and a large part of the Upper Shabelle region (47). Notwithstanding, the international community recovered once again the project of national reconciliation to be held in Eldoret/Nairobi with uncertain results.

As discovered last century by the colonial forces, segmentary societies are very difficult to rule. Experience teaches us that foreign intervention may contribute to factionalism (48). Furthermore, top-down approaches to national reconciliation are hardly sustainable (49). Moving from warlordism is difficult since warlords created their own economy. So far, although exploited and predated, a good part of Somalia survived partially thanks to the warlords' economy. Therefore, in Somalia the question remains; how should Somalis re-build state institutions?

How could they move from warlords' culture? Constitutional formulas may help but only when nurtured by a non-authoritarian, constructive, and democratic political culture (50). Above I referred to Tocqueville's idea of social capital as the "habits of the heart and the mind". Civic attitudes and ideas of democracy are embedded in Somali culture, however there is always a risk that the society may shift from a civic to an uncivic society.

The making of a state remains the unavoidable end of the process. The state creates the framework within which a society grows; yet the society can shape the state according to its own cultural values. The Somaliland experiment demonstrates this attempt to blend modern state institutions with its own cultural values (51).

The traditional idea of a state-building in Africa is changing. The African state should include variable citizenship that should be more in accordance with traditional African culture without strict boundaries, and marked by a plurality of institutions and sovereignties. Informal approaches to the state-building process are likely to be more effective and flexible than formal constitutional arrangement. Stability and consensus will exist only if they are the main goals of political leadership.

REFERENCES

ALLEN, C. (1997). Who Needs Civil Society? In Review of African Political Economy, Vol. 24, Issue 73.
APTER, D. E. (1965). The Politics of Modernization, Chicago: University of Chicago Press.
BARNES, S. H. (2001). The Contribution of Democracy to Rebuilding Postconflict Societies. In The American Journal of International Law, Vol. 95, Issue 86, pp. 86-101.
BATTERA, F. (1995). Dalla tribU allo Stato nella Somalia nord-orientale: il caso dei Sultanati di Hobiyo e Majeerteen, 1880-1930, unpubl. PhD thesis, Universita degli Studi di Siena.
BATTERA, F. (2000). Remarks on the 1998 Charter of Puntland State of Somalia. In V. Piergigli & I. Taddia (eds.), "International Conference on African Constitutions" (Proceedings, Bologna, November 26th-27th, 1998), Torino: Giappichelli, pp. 173-91.
BATTERA, F. (2003). State Collapse & the Restoration of Law in Somalia: between Sari'a and Customary Law. In "Cooperazione, sviluppo e rapporti con l'islam nel Corno d'Africa", Roma, 7 giugno 2002, IsIAO, pp. 127-51.
BAYART, J.-F., ELLIS, S. & HIBOU, B. (eds.) (1997). La criminalization de l'Etat en Afrique. Bruxelles: Complexe.
BECKMAN, B. (1993). The Liberation of Civil Society: Neoliberal Ideology and Political Theory. In Review of African Political Economy, Vol. 20, Issue 58.
BRATTON, M. & VAN DE WALLE, N. (1994). Neopatrimonial Regimes and Political Transitions in Africa. In World Politics, Vol. 46, no.4, pp. 453-89.
CLAPHAM, C. (1996). Africa and the International System. The Politics of State Survival, Cambridge: Cambridge University Press.
COLLIER D. & LEVITSKY S. (1997). Democracy with Adjectives: Conceptual Innovation in Comparative Research, in World Politics, 49, 3, pp. 430-51.
COULON, C. (1995). Senegal: The Development and Fragility of Semidemocracy. In L. Diamond, J. J. Linz & S. M. Lipset, Politics in Developing Countries: Comparing Experiences with Democracy, Boulder: Lynne Rienner, pp. 493-530.
DAHL, R. A. (1971). Polyarchy: Participation and Opposition. New Haven: Yale University Press.
DAS GUPTA, J. (1995). India: Democratic Becoming and Developmental Transition. In L. Diamond, J. J. Linz & S. M. Lipset, Politics in Developing Countries: Comparing Experiences with Democracy, Boulder: Lynne Rienner, pp. 263-321.
DEUTSCH, K. & FOLTZ W. (eds.) (1963). Nationbuilding, New York: Atherton.
DIAMOND, L., LINZ, J. J. & LIPSET, S. M. (eds.) (1995). Politics in Developing Countries: Comparing Experiences with Democracy, Boulder: Lynne Rienner.
DRYZEK, J. S. (2003). Deliberative Democracy in Divided Societies: Alternatives to Agonism and Analgesia. Social and Political Theory Program--Research School of Social Sciences, Canberra: Australian National University.
FARAH, A. Y. (1998). Political Actors in Somalia's Emerging De Facto Entities: Civil and Military Relations in Somaliland and Northeast Somalia, Nairobi: War-Torn Societies Project, Somali Program.
FORTES, M. & EVANS-PRITCHARD, E. E. (eds.) (1940). African Political Systems. London: Oxford University Press.
GELLNER, E. (1990). Tribalism and the State in the Middle East. In P. S. Khoury e J. Kostiner (eds.), Tribes and State Formation in the Middle East, Berkeley: University of California Press.
HYDEN, G. (1997). Civil Society, Social Capital, and Development: Dissection of a Complex Discourse. In Studies in Comparative International Development, Vol. 32, No. 1, pp. 3-30.
ICG (2002), Somalia: Countering Terrorism in a Failed State. ICG Africa Report, no. 45.
ICG (2003), Somaliland: Democratisation and its Discontents. ICG Africa Report, no. 66.
KEENADIID, Y. C. (1984). Ina Cabdille Xasan e la sua attivita letteraria, Napoli: Istituto Universitario Orientale.
LE ROY, E. (1996). L'odyssee de l'Etat. In Politique Africane, n. 61, pp. 5-17.
LEWIS, I. M. (1961). A Pastoral Democracy: A Study of Pastoralism and Politics among the Northern Somali of the Horn of Africa, London: Oxford University Press.
LEWIS, I. M. (1995) (ed.). A Study of Decentralised Political Structures for Somalia: A Menu of Options, Report prepared by consultants from LSE commissioned by the EU (EC Somalia Unit).
LIJPHART, A. (1969). Consociational Democracy. In World Politics, Vol. 21, pp. 207-25.
LINZ, J. (1978). The Breakdown of Democratic Regimes: Crisis, Breakdown and Reequilibration. Baltimore: John Hopkins University Press.
MARCHAL, R. (1996). The Post Civil War Somali Business Class. Paris: EHESS.
MARCHAL, R. (1998). A Few Provocative Remarks on Governance in Somalia. Nairobi: UNDOS.
MARCHAL, R. (1999). Des contresens possibles de la globalization. Privatization de l'Etat et bienfaisance au Soudan et au Somaliland. In Politique Africaine, no. 73, 1999, pp. 68-81.
MARCHAL, R. (2003). Islamic Political Dynamics in the Somali Civil War: Before and After September 11. In "Cooperazione, sviluppo e rapporti con l'islam nel Corno d'Africa", Roma, 7 giugno 2002, IsIAO, pp. 89-126.
MATTEI, U. (2000). Patterns of African Constitutions in the Making. In V. Piergigli & I. Taddia (eds.), "International Conference on African Constitutions" (Proceedings, Bologna, November 26-27, 1998), Torino: Giappichelli, pp. 145-71.
MATTEI, U. (2001). Foreign Inspired Courts as Agencies of Peace in Troubled Societies. A Plea for Realism and for Creativity, paper presented at the workshop on "The Restablishment of the Rule of Law: The Case of Somalia" (Rome, December 10-11, 2001).
MEDARD, J.-F. (1982). The Underdeveloped State in Tropical Africa: Political Clientelism or Neo-Patrimonialism. In C. Clapham (ed.), Private Patronage and Public Power: Political Clientelism in the Modern State. London: Francis Pinter.
MOHAMED, A. A. (2002). Intergroup Conflicts and Customary Mediation: Experiences from Sudan. In Africa Journal in Conflict Resolution, no. 1.
OSAMBA, J. (2001). Peace Building and Transformation from below: Indigenous Approaches to Conflict Resolution and Reconciliation among the Pastoral Societies in the Borderlands of Eastern Africa. In Africa Journal in Conflict Resolution, no. 1.
OTTAWAY, M. & LIEVEN, A. (2002), Rebuilding Afghanistan: Fantasy versus Reality, Carnegie Endowment for International Peace, Policy Brief 12.
OTTAWAY, M. (2002). Think Again: Nation Building, Carnegie Endowment Publications.
OTTAWAY, M. (2003). Democracy Challenged. The Rise of Semi-Authoritarianism.
PRUNIER, G. (1998). Somaliland Goes it Alone, in Current History, May.
PUTNAM, R. D. (1993). Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press.
SAHLINS, M. D. (1977). Poor Man, Rich Man, Big Man, Chief: Political types in Melanesia and Polynesia. In S. Schmidt, L. Guasti, C. Lande & J. Scott (eds.), Friends, Followers and Factions, University of California Press, pp. 220-32.
SAMATAR, S. S. (1982), Oral Poetry and Somali Nationalism: The Case of Sayyid Mahammad 'Abdille Hasan, Cambridge: Cambridge University Press.
TILLY, C. (ed.) (1975). The Formation of National States in Europe, Princeton: Princeton University Press.
UNDP (2002). Human Development Report 2002. Deepening Democracy in a Fragmented World. New York: Oxford University Press.
YOUNG, C. & TURNER, T. (1985). The Rise and Decline of the Zairian State. Madison: The University of Winsconsin Press.

NOTES

(1) Until 1991, Somaliland has been an integral part of the Somali Republic. In 1991 the declaration of independence started a process of state-building. Declaration of independence has never been internationally recognized.
(2) Literature usually focuses on the national dimension--Lebanese, North-Irish, etc.--of the so-called 'divided societies'. I prefer to focus to a sub-national dimension. More often divisions affect communities as well as societies, dividing parties between moderates vs. radicals, 'hawks' vs. 'doves', young vs. elders, etc. I will consider not only the Somali society as 'divided' but also its single traditional divisions--clan, federation of clan, etc.--which are also internally divided. Such divisions are often articulated through generational conflicts or manifest violations of consolidated customary norms.
(3) For a good example of study on the social capital see Putnam (1993).
(4) Among the many who treated the issue see Deutsch--K. Deutsch & W. Foltz (1963)--and Tilly (1975). For long state- and nation-building have been confused, but with the 90s distinctions have been made in order to explain cases of state failure. It is not my intention to enter into the debate. I will use both the terms though as far as the African experience is concerned I prefer to talk about state-building.
(5) See UNDP (2002), in particular chapters 3 and 4.
(6) See Ottaway and Lieven (2002).
(7) In Afghanistan, Massoud-controlled Tajik areas (Panjshir) were secured to peace and administered through bottom-up participation before the Taliban-Tajik confrontation. The Panjshir model--i.e. internally pacified community--sharply contrasted with other areas--Mazar-i-Sharif, Pashtun areas, etc.--still fragmented between 'clan' fiefdoms. The Panjshir model remember Somaliland.
(8) Recent, though still few considered, literature has been devoted to customary mediation and indigenous approaches to conflict resolutions. For this study I considered Osamba (2001) and Mohamed (2002).
(9) One was the theocratic experiment of the so-called 'Mad Mullah', the other the Sultanistic state of Yuusuf 'Ali Keenadiid (see BATTERA, 1995)
(10) Though, apparently, only in the British system the 'Aaqils retained formal judicial powers within the Subordinate courts, within the Italian system, which formally established only some Qadi courts with Shari'a judges in charge, the Capi-stipendiati played also an unofficial judicial role in the more peripheral areas supporting or complementing the Residenti.
(11) The 'building block' approach derived not only from the confidence that localism could be a resource for reconstruction if correctly addressed but had been originally conceived as a stick-carrot instrument to promote legitimate local authorities against illegitimate authorities.
(12) In January 2004 a new agreement was signed in Nairobi announcing the world that the war in the South was definitively over. The agreement will establish a new power-sharing formula between clans. The low public attention which received demonstrates the prudence and disillusion that accompanied any Somali initiative.
(13) The National Charter of 1993 was followed by a Constitution in 1997 at the end of the Hargeysa conference, which was amended in 2000 and definitively approved by referendum in 2001. Main differences are that the Hargeysa Constitution (1997) transformed the parliamentary system into a bicameral system and the 2001 Constitution approved the transformation from a no-party system into a multiparty system.
(14) The Somaliland bicameral system comprised the Guurti--the Upper House--and a House of Representatives. The representatives of both houses have been selected according to a consociational rule which divides posts according to a delicate balance respectful of clan weight and the right of minorities. The same unwritten norm ruled the division of ministries and others bureau posts.
(15) Some of them were not directly involved in the daily political life in the Hargeysa Guurti but preferred to live in their own constituency and periodically were summoned in Hargeysa to have consultation with the President.
(16) During 1999, the conflict between Somaliland and Puntland over Sool and Sanaag seemed to escalate into war between the two entities. I was in Hargeysa at that time and had the opportunity to interview the Garaad of Warsangeli who was there to discuss the issue with President Igal. From the Warsangeli point of view the issue of the belonging of Sanaag was a false problem since the Warsangeli were glad to send their representatives to both the Houses of Garowe and Hargeysa and the Warsangeli elders were sitting both in the Hargeysa Guurti and the Garowe Isimada.
(17) During the war, which ravaged Somaliland between 1982 and 1990, most of the Barre's troops were recruited among non-Isaaq peoples of the Northwest and the Ogaadeen refugees appealing to the Daarood solidarity (Darodism) against the Isaaq of SNM. Part of them nurtured the so-called Dabargoynta Isaaqa ("The Isaaq exterminators") troops (see Africa Watch Report (ed.), Somalia. A Government at War with its own People, New York, The Africa Watch Committee, 1990, pp.44, 46, 69, 87-9, 123-4).
(18) During 1999, when the debate over independence became particularly hot in Hargeysa, forces tried to recover the SNM as an opposition party to Igal, who was accused by the strongest partisans of independence to be too cold about this issue, but this attempt to benefit from the credit endorsed by the SNM during the liberation struggle failed and with multipartitism new parties were constituted, but no one which recall SNM.
(19) Igal had been an experienced politician. He had passed through all the experiences of contemporary Somalia and survived its most delicate phases. He experienced decolonization, the unification of the Republic, and the creation of democracy, the de facto independence of Somaliland, and the Barre regime. He represents a good example of the African tradition of talented mediators.
(20) Contrarily to Somaliland, Puntland did not aim to get independence but committed itself to preserve the unity of a Federal Somali republic.
(21) I was there during the preparation of the Charter in May 1998. My impression was that Puntland delegates were ready to concentrate most of the power in the President/Prime Minister. Most of them already knew the name of the future strongman some months before his 'election'. Worried by the experts who prepared the Charter about the risks to bestow too much power with one person the delegates were confident about the future and stressed the necessity to have a transitory strongman to make things work.
(22) Shir means assembly. It is during the shir that Somalis express the best of their political culture.
(23) The key feature of consociationalism is to set institutional devices (proportionality, grand coalition, veto power) and foster cooperative attitudes between groups in segmented (and divided) societies (LIJPHART, 1969).
(24) The consociational option is one of the most interesting alternatives amongst the most for national reconciliation proposed by Prof I. Lewis in the "Menu of Options" (1995). But consociational options might not be enough in highly politicized societies cause they may contribute to crystallize divisions (DRYZEK, 2003).
(25) The same 'elders' category, which apparently seems to be highly formalized--most of the 'elders' inherited the post from their father--and represents the key to access to the Upper House, is actually very dynamic: a great competition for representation, though limited by a certain xeer norms, opposes fractions. Representation is continuously modified and new representation is allowed according to new negotiations and agreements. Such phenomenon of 'reproduction of eldership' demonstrates the vitality of the 'traditional' models of representation in the contemporary world (BATTERA, 2003: 129).
(26) At the point to tolerate them to maintain good relations with the neighbor Puntland and send representatives also to the House of Representatives and the Isimada (the Chamber of elders of Puntland) of Garowe.
(27) The new Constitution of 2001 had legalized the formation of political associations, which would be eligible to compete for seats on local councils, but only the three associations to obtain the highest percentage of the vote could be registered as political parties and go forward to contest parliamentary and presidential elections. As a result a third party was admitted to compete for the presidential elections (Ucid), when it obtained 15.85% of total vote.
(28) For comments on the electoral results see ICG (2003).
(29) Ottaway (2003: 3) defines semi-authoritarian systems those that "(..) are not imperfect democracies struggling towards improvement and consolidation but regimes determined to maintain the appearance of democracy without exposing themselves to the political risks that free competition entails.". Ottaway stresses the theoretical autonomy of semi-authoritarian regimes from the democratic ones. Semi-authoritarian regimes are specific regimes with a proper nature. The definition of Diamond, Linz and Lipset (1995: 8) of pseudodemocracy is very similar to that of the Ottaway, though Diamond, Linz and Lipset give room to the possibility of a pseudodemocracy to evolve into a full democracy and Ottaway gives no room to this possibility with the exception of a radical regime change: semi-authoritarian regimes are intimately constructed to maintain their authoritarian nature.
(30) On the importance of these factors to the democratic development see Linz (1978: 27-38) and Diamond, Linz and Lipset (1995: 17).
(31) On the predatory trajectory of the African state see Young & Turner (1985), and Bayart, Ellis, & Hibou (1997).
(32) In the last ten years, warlordism has been a profitable economic activity. Warlords' style of government has been characterized by fiefdom-liking territorial control and predatory mechanisms over human and territorial resources. Business operated in monopoly condition in the warlords' controlled fiefdoms.
(33) Elders retain what Marchal (1999) called the 'traditional ethos'.
(34) During this period, the Lower House found difficulties to make laws, as I was able to ascertain. Lack of competences hampered the Lower House to work as expected. But the no-party system of that period worked to mature in the Lower House a conscience of its own responsibilities. Government made law thanks to its decree power while the Lower House learned to discuss rejecting or enacting them. On the other hand, the Upper House, which was totally unable to understand the meaning of most of the laws made by the Government, was more concerned about the conflicting effect of the parliamentary game. It spent most of its time finding solutions to maintain a balance between the Government and the Lower House. Such learning effect of parliamentary game is of extreme importance for a young democracy.
(35) By localism I mean the strong prevalence of local dimension over the regional/national dimension.
(36) In Somalia, the debate over federalism versus centralism is understood only as a means of allocating governmental posts or as a way of dividing power.
(37) On the 'big man' see Sahlins (1977).
(38) Prof. Lewis was optimist (with some reservations) about the First Republic before its collapse. I again share the same optimism for Somaliland.
(39) From Gurey to Barre passing through the Sayyid and Generale Graziani. Contrary to a common opinion, in case of an external aggression, Somalis are more inclined to divide and not to unite, as illustrated during the colonial conquest.
(40) Though others (BECKMAN, 1993; ALLEN, 1997) argue that NGOs rhetoric could serve the neoliberal strategy of containment and de-legitimation of the state in the Third World.
(41) Allen (1997: 6) concludes that civil society is not necessarily a sufficient condition for democratic struggle to occur.
(42) Such consideration derives from a different approach from that of Tocqueville and neoliberals. What Hyden (1997: 9-10) calls 'The Regime School' draws its inspiration from Locke. Its assumption is that state and society are not separated but linked.
(43) In this sense, external intervention could be helpful when clearly defined by a mission mandate (OTTAWAY, 2002).
(44) Ottaway and Lieven (2002: 6) warned about the risks in making the central government the chief channel for international aid--"(..) Aid itself would become a source of future conflict (..)"--proposing an alternative localbased delivering. However, risks of strengthening local warlords through aid must not be ignored.
(45) For complexity, Somaliland is an example of regional reconciliation with several communities involved in the process.
(46) The Puntland crisis of 2001 was partly due also to the TNG attraction towards part of Puntland leadership.
(47) The international image of TNG was also worsened by the fact that at the end of 2001 most of the analysts starting to indicate it as tied up to fundamentalist networks (ICG, 2002: 18, MARCHAL, 2003).
(48) Intervention should not only restore pacification when altered but also severing those mechanisms that nourish warlordism offering alternative incentives. This is certainly very difficult, if we think the role that poppy-plantation played in the economy of Afghan war, and other resources--diamonds, oil, drugs, weapons, etc.--in other trans frontier conflicts.
(49) Up to now, the Bonn agreement, that legitimized the new Afghan regime, succeeded to strengthen pacification at the center but failed to give stability and full legitimacy to the Karzai's government. The state-building process in Afghanistan is likely to take a long time and without the commitment of IC injection of large amount of resources peaceful conditions will not endure.
(50) One of the most important obstacles to national reconciliation in Nairobi has been the incapacity to find formulas on which parties were eager to agree. The point is not on consociational formulas, but on the lack of a set of disincentives.
(51) Something in accordance to what Le Roy (1996: 5) call the indigenization of the state in Africa. Indigenization may further legitimacy, the accountability and the performances of the African state.

Federico Battera, Dept. of Political Sciences, University of Trieste, fbattera@yahoo.com