Executive Summary
This study describes four models of decentralised constitutional
government, focusing on aspects that may be appropriate for Somalis.
The four models are: (1) a confederation; (2) a federation; (3) a
decentralised unitary state with strong guarantees of local or
regional autonomy; and (4) a consociational, non-territorially based
form of decentralisation. It then looks at different elements of
government in modern democracies, such as the political economy, the
administration, the judiciary, defence and foreign relations, and
electoral designs, and outlines how each would be organised in the
different models. The report concludes with a menu of options, and
considers a final form, which would be the most decentralised:
functional cooperation, which could be applied to however many
states Somalis decide they want to establish, or even without the
creation of formal state machinery at all. The report does not
assume that the former Somali Republic can only be reconstructed as
a single state.
Introduction: The Uncentralised Somali Legacy
by Professor Ioan M. Lewis
Today, most Somalis under the age of 30 have no knowledge of the
democratic structures that were in place before Mohamed Siad Barre's
military rule began in 1969, nor of the colonial period. In
traditional Somali politics, all decision making is conducted
democratically (although formally excluding women), by segmentary
groups of kinsmen whose patterns of alliance and confrontation are
fluid and inherently unstable. Pastoralists are individualists who
meet in general assemblies, where all adult male family heads (or
elders) seek consensus, instead of taking decisions by majority
vote. Since so many males can be elders, leadership is often
difficult to maintain for very long. In this uncentralised, rather
than merely decentralised society, there is thus no tradition of a
centralised state nor any set political offices or ranked leaders.
The two most stable elements in Somali society are: 1) the ' dia
paying group': the contractually defined kinship group within clans
or subclans, which acts as a unit in paying and receiving damages
for injury and death; and 2) the clan: a grouping of several related
'dia paying groups'. Clans usually live within a roughly
defined territory, trade together, share and maintain wells and
other resources (camels bear a common clan brand). There are
approximately 100 clans in this country of some 5 million
inhabitants, though the size and power of each clan varies
considerably. Pastoral nomads belong to clans that are larger and
more widely spread out than clan members who live by cultivating and
in urban areas. All clans belong to a clan family - a group of
related clans which is the highest level of political solidarity
within the Somali nation or ethnic group.
Clan Culture
The six Somali clan-families are the Dir, Darod, Isaq, Hawiye,
Digil and Rahanweyn (also known as Digil-Mirifle). The Isaq live
primarily in the former British Somaliland; the Digil and Rahanweyn
are agro-pastoralists who live between the Juba and Shebelle Rivers,
where they produce grain and have their own distinct culture and
language (May or Maymay). They also adopt migrants
from other clans. The Darod and Hawiye clans are the most widely
distributed. There are also marginal Bantu groups along the rivers,
other minority urban communities (some of Arab origin), and
blacksmiths, leather-workers, and hunters who may seek to maintain
retain separate status from the Somali clan structure.
Somali men are traditionally either warriors or men of God.
Today, almost every male has an automatic rifle, and some have even
heavier weapons, which constitute a serious obstacle to the
formation of governmental structures. The country thus now consists
of clans and unstable clan-alliances with their militias, as well as
war lords with free-lance armed forces. There are also, in some
areas, small independent Islamic communities, often with their own
militias, providing educational and medical facilities and offering
support to orphan war victims. In some urban settings, particularly,
where other patterns of social control are failing, a fundamentalist
style of Islamic law is being applied strictly, controversial though
this is for many Somalis. There is in fact a deep-seated tension
between universalistic Islam and particularistic local customary
values and allegiances.
Over the past five years, there have been significant movements
of population as people have pushed into the south, where economic
and agricultural opportunities are greater. There has also been some
'clan-cleansing' during the same period, creating substantial
numbers of refugees and displaced persons. While pasturage is
traditionally not owned by specific groups, today clans are trying
to maintain control over some regions, especially where there are
water points and trading centres. In the absence of any effective
central government, clan elders are also consolidating their power
by making decisions in local clan councils, often in co-operation
with militias, who provide informal police functions.
The Colonial and Post-Colonial Experience
There was no Somali state before European colonisation, which
divided the nation into five parts (French, British, Ethiopian,
Italian and Kenyan). This division ultimately served to unite
Somalis, giving rise to Somali nationalism. After independence in
1960, which brought together the British and Italian Somalias,
formal local councils were set up and elections held. The Italian
legal system was largely adopted, while common features of Somali
customary law continued to be applied alongside the European codes
and shari'a law. The new state (the Somali Republic) was
divided into eight regions with 59 districts, administered by
regional and district governors.
The desire to unite all Somalis under one banner encouraged
national solidarity cutting across clan ties, but created conflicts
with neighbouring states, especially Kenya and Ethiopia. This unity
lasted until the late 1960s when Mohamed Haji Ibrahim Egal, the new
prime minister, made peace with the two neighbours. In what proved
to be the last civilian elections held in 1969, most voters and
candidates behaved according to their clan identifications, with
1,002 candidates contesting the 123 seats in the National Assembly.
Military Rule
In October 1969, General Mohamed Siad Barre took over in a
military coup. Siad Barre developed a highly centralised,
totalitarian state, and tried unsuccessfully to abolish clan
behaviour. He applied what he termed 'scientific socialism'.
Inspired by Marx and Lenin, Siad ruled through a revolutionary
council, which effectively favoured the Darod clans, and was adept
at manipulating clan rivalry in choosing ministers.
After the crippling 1975 drought the government adopted relief
measures which displaced many northern nomads to the south, where
they were settled in agricultural and fishing communities. Both
activities were despised by the nomads, and many preferred to
survive on food hand-outs while rebuilding their herds. Siad
supported the Ogadeni guerillas (his mother's clansmen) in their
secessionist struggle in Ethiopia, which by the late 1970s was
gathering momentum and led to all-out war in 1977/78. This proved
disastrous as Somalia had underestimated the strength of external
support for Ethiopia and lost the war. Defeat and the enormous
number of Ogadeni refugees who fled into Somalia gradually
destabilised Siad's regime. The former Soviet Union had abandoned
Somalia for the greater prize of Ethiopia, and Siad desperately
switched allegiance to the West to gain assistance. As western aid
gradually increased, the national economy collapsed, and Somalia
became dependent on this aid for survival, although the informal
economy thrived.
Current Patterns of Wider Alliance
Siad's peace agreement with Ethiopia in 1988, his arming of
friendly clans to fight his opponents, the liberation of the
'Republic of Somaliland' by the Isaq guerillas, of Majerteynia by
the SSDF, and his own overthrow in 1991 all helped to forge the
current Rambo-style political economy.
Clans are not the only political units on the ground today, and,
following wider economic and other strategic interests, are
themselves involved in shifting patterns of alliance. Outside
Mogadishu, there are currently at least three more ambitious
political groupings, each with its own specific strengths and
weaknesses: 1) 'Republic of Somaliland'/North-West, based on the
former Somaliland Protectorate, and including the Isaq clans and
their allies in the Gadabursi, Ise, Warsangeli and Dulbahante clans.
The main port is Berbera, the capital Hargeisa; 2) North-East
(Regions), based on the Majerteyn clan. The main port and capital is
Bosaso; 3) Digil and Rahanweyn (Mirifle) alliance of clans based in
the Inter-Riverine region of southern Somalia. The capital is Baidoa.
All three regimes claim local autonomy, but only Somaliland has
so far asserted full independence (without achieving international
recognition). This reflects the localisation of the Isaq and
associated clans. The clan connections and interests of the SSDF
Majerteyn in the North East, in contrast, extend far outside this
region into the southern part of Somalia (through the Harti
grouping). Here since June 1995 the demographically dominant Hawiye
have two 'governments' based in different parts of the old capital,
Mogadishu, and variously claiming 'national' status.
The Challenge
In conformity with their traditionally uncentralised political
culture, the majority of those Somalis who feel a need for modern
government (and not all do) advocate a lightly administered,
decentralised system. Since Somalis have so energetically 'rolled
back the frontiers of the state', it will be difficult to establish
decentralised and flexible political structures that can accommodate
the powerful inter-clan conflicts that pervade their politics. The
problems are compounded by the widespread distribution of modern
weapons and the practice of voting with the gun which has become so
deeply embedded. The challenge facing Somali political leaders will
be to persuade people that there are general common interests that
could be best served by wider forms of political structures than
exist today. They will then have to establish (perhaps along the
lines suggested in our Menu) loosely articulated forms of political
organisation capable of securing public approval.
1. Decentralisation Options for Democratic Reconstruction
Four decentralised models of government are outlined in this
report, drawn from the experience of other peoples whose history has
something in common with that of the Somalis. These are analysed
with reference to emerging institutions in the former Somalia. There
are three territorially-based models (a confederation, a federation,
and a decentralised unitary state), and one community-based model (consociational)
that can be combined with any of the territorially-based models.
Decentralised constitutional structures in other parts of the
world that have similarly developed through local initiatives should
be of interest to Somalis who wish to rebuild one or more states to
replace the discredited former republic. It is not yet clear how
many states should emerge from the old Somalia. There are three
possibilities: 1) to re-establish a united Somalia under either a
unitary or a federal constitution; or 2) to create a union or
confederation from two or more states; or 3) to form two or more
independent states with no constitutional links. Whatever is
decided, it need not be permanent. Arrangements can be built into a
new constitution that allow for re-unification or secession. Another
important issue is where sovereignty, i.e., supreme political
authority, should lie. There are three possibilities: 1) sovereignty
can rest with all adult Somali citizens; 2) it can also be vested in
the constituent territorial units of a Somali state or states; and
3) it can also rest with the clans.
A Confederation
A confederation is a union of separate but equal states linked by
international treaties. In confederations, sovereignty rests with
the states, whereas in federations it is shared between the centre
and the states (or provinces or regions). A confederation is
normally created for specific purposes, such as for common defence,
free trade, or to prepare for closer political union; and the centre
normally acts as a coordinating body only. For example, in many
federations, such as Switzerland, Germany, and the USA, a
confederation existed before the member states opted for a closer
form of co-operation.
In the European Union, which is still more of a confederation
than a federation, member states (15 now but there is no limit)
agreed to establish common institutions that are responsible for
certain areas of policy throughout the Union. Once these functions,
such as agricultural policy, are transferred to the centre, they
cannot be exercised independently at the state level. In order that
each state continues to have some control over these central
institutions, provisions in the treaties of union allow each state
to veto, or opt out of policies that the state feels damages its
interests.
In some departments, the EU has moved closer to a federation. The
European Court of Justice, for example, is considered the highest
court for EU members, and thus sovereignty has effectively been
transferred to the centre. Its underlying structure, however,
remains confederal because the transfer is voluntary, i.e., the
Court does not have any power to enforce decisions. European
experience here may have potential relevance for Somalis: one of the
motives inspiring European union was to prevent a repetition of the
two European wars that also engulfed the world in conflict earlier
this century. At some point in the future, and in conformity with
traditional political values, it would be possible for the different
Somali regions or states to create common institutions and policies.
Somalis could then work together in central agencies, with
representation from each state or region, protected by the sovereign
status of each region and by the right to veto, or opt out of
unwelcome proposals.
A Federal System
It is probably too early to contemplate any form of centralised
authority, but in theory, if enough trust developed between the
different Somali groups, they could create a federation, satisfying
those who want a more substantial central governing capability, and
those who want strong regional governments. In typical federations,
free and independent states come together, and transfer a limited
amount of sovereignty to the federal institutions, although others
have been created by decolonisation, or after a war, such as in
Germany, where the purpose was to limit the power of the central
authority. The centre and the regions (or provinces) share power,
and the centre is unable, on its own authority, to change the
constitution. There are three ways of organising federal-provincial
relations: 1) the provinces have most of the powers of government,
with the centre only exercising limited powers; 2) the centre has
most of the powers with the provinces having a limited amount; or 3)
the provinces and the centre share powers of government with
specific means of mediating disputes between them. It is also
possible to grant considerable autonomy - even the right to secede -
to some regions within any of these three.
It is important to repeat that forms of government are not
irreversible: a federation can become a confederation, and a
confederation can develop into a federation. There are, in addition,
many types of federation, and models exist that have never been
applied. Federations are designed to allow diverse peoples and
cultures to live together, with guarantees normally written into the
constitution to ensure that the centre does not seize all the power.
This is sometimes facilitated by removing the federal capital from a
major city and placing it in a small city in an unimportant region.
If Somalis wanted a federation, the capital could be moved from
Mogadishu.
Swiss experience has some similarities to recent Somali history.
Switzerland was forged after civil war, coups in various cantons,
attempts at constitution writing, and intervention by outside
powers. The federal plan was designed to bring together people torn
apart by religious and political differences, all equally suspicious
of central government. In the resulting Switzerland, the powers of
the centre are sharply defined and limited, whereas those of the
cantons are not (Option 1 listed above). The cantons may even have
their own constitutions, as long as they do not contradict the
federal constitution, and guarantee to continue to govern
democratically.
Switzerland also has a collective presidency with seven members
who take it in turn to serve as president; this assures that no one
person or group takes over. This method has been very successful -
there are no Swiss politicians famous outside Switzerland. There are
two other governmental institutions. The first, the Federal
Assembly, has two chambers, one with representation according to the
population in each canton; the other has two representatives per
canton. The second is the Federal Court, on which the four languages
of Switzerland are represented; and which mediates disputes between
the federal government and the cantons, and amongst the cantons
themselves. The Swiss are first citizens of their communes, then of
their cantons, and finally of Switzerland.
A Decentralised Unitary State with Guarantees of Regional or
Local Autonomy
Just as the differences between a confederation and a federation
can be slight, so too between a decentralised unitary state and a
federation. The main difference is that in a federal system, power
is shared between the centre and the regions whereas in a
decentralised system, the regions are subordinate to the centre. For
example, in Uganda or Papua New Guinea, although regions are allowed
considerable control over many of their affairs, the centre retains
the right to reduce this or to intervene if it wants to. In Uganda,
the decision to decentralise was partly designed to accommodate
strong regional pressures.
Decentralisation is a top-down process, which by delegating
power, can help to reduce the centre's control over the social,
economic and cultural life of its citizens. In contrast, the
formation of a federation, is a bottom-up process where local units
join together creating an additional layer of government above them.
This decentralised option might be more appropriate within the local
units of state or states that the Somalis eventually establish,
providing these reflect traditional structures of political
authority. Although any decision to pursue the confederal or federal
options would necessarily require widespread popular support, there
are two major aspects of decentralised unitary systems that could be
included in a future Somali constitution(s): 1) the constitution
could define the powers of local governments. These powers can
always subsequently be transferred to the centre if a local
government wishes; and 2) guarantees should be included to ensure
that larger units of government (such as provinces or the centre)
cannot abolish smaller local units.
Consociation: A Non-Territorial Option
Power sharing or consociational principles of government can
operate in confederations, federations and decentralised unitary
states, and were developed specifically for divided societies. South
Africa today is a model of a consociational democracy in its early
stages. Consociational democracies normally include four elements:
1) a grand coalition government, which includes representatives from
the major groups (or clans or clan-families) in society; 2) a quota
of seats or jobs or public moneys for each community according to
its population strength; 3) community responsibility for certain
matters, such as education; e.g., the distinctive interests of clans
and minorities could be respected wherever they lived or worked; and
4) minority veto power.
Under consociational arrangements, groups are separate but equal.
For these to work, people need to be convinced that it is better to
have a share of power than to risk the costs involved in trying to
seek full control. Three conditions need to be satisfied: 1)
competing communities must not try to integrate other groups or
establish their own separate sovereign state; 2) politicians must
strive to maintain these beliefs, economic and political stability,
and be mindful of the negative consequences of returning to a state
of war; and 3) the leaders of the communities must be able to act
independently so that they may be able to reach a compromise with
other groups without being accused of betrayal by their own
community. The Somali tradition of elders participating in
consensual decision-making at the local level demonstrates that such
principles could be appropriate here. When these conditions are not
met, consociations fail, as happened disastrously in Lebanon and
Cyprus.
The key feature of all four systems of decentralised government
is the ultimate location and distribution of power: in
confederations, power is above all held by the component states; in
federal systems, it is shared between the centre and the constituent
states; and in decentralised unitary states, it is located at the
centre though certain regions enjoy considerable autonomy.
2. The Political Economy of Decentralisation
Only the pastoral livestock industry managed to survive the
destructive economic effects of the Siad Barre regime, which
officially suppressed private enterprise, although thereby
stimulating an extensive informal and black market economy. In the
late 1980s, a structural adjustment programme was attempted but not
properly implemented, leaving Somalis with little experience of
successful economic policy. Once the state collapsed, the country
was left with many small traders and producers lacking public
services and regulation. There is little prospect in the future that
donors will grant the large sums of aid of the past, and foreign
investment can be expected to be slight for some time. Thus
political reconstruction, at whatever level, will need to operate
against a background of small-scale local economies.
The switch to private enterprise fits well with decentralisation,
which depends on the abilities of local groups to manage their
economies and resist pressure from the centre. If the state
transfers assets to the private sector, it lets success and failure
be determined by the market. Decentralisation may reduce conflict
and may minimise the economic benefits associated with political
office. Clan loyalties will play an important role in rebuilding the
private sector, but need not lead to conflict if success and failure
are determined by fair competition, i.e., providing the market is
regulated and policed. Some services, such as law and order, defence,
roads, health, education, etc., will still need to be provided
publicly, but there is no reason why some of these cannot be managed
by local authorities, with the assistance of donors and NGOs.
Modern states also require conditions that allow traders to
produce and exchange their goods through common property laws,
stable currency and a credit system. Many assume that local
initiatives are the only way to restore order, and that new local
authorities can combine to form a state while still retaining
significant autonomy. Regional Somali authorities, however, will
need access to personnel, financial resources and physical assets.
Inter-Regional Transfers
Inequality between regions will require attention if poorer
regions are to be integrated in all political formations. Market
economies often lead to an unequal distribution of income, which can
only be corrected through transferring income by both local and
central government. In confederations, such transfers are rare,
while in unitary states the centre controls taxes and grants and
thus can redistribute more easily. In federal systems, transfers
often create conflict. It is also difficult to establish criteria
that determine who gets what, and in which circumstances. It is
unlikely that Somalis will pay taxes to anyone or for anything
outside their immediate region, so such a redistributive system may
not be practical for some time. This means that local authorities
will have to strengthen their ability to raise funds to pay for
education, health, roads and water. Regional councils can also set
up committees to decide how resources will be distributed, as well
as to monitor spending. In the meantime, local authorities will have
to charge full cost fees for consumer services, which will eliminate
the immediate need to raise taxes and also give local people control
over the service that is provided. Foreign donors may also be able
to assist the poorer regions in restructuring. Communities can
continue to provide services on a voluntary basis, such as mending
roads, building schools, and digging wells.
Money
The collapse of the Somali Republic has removed controls over
trade and investment, and ended the creation of state-supported
credit and the growth of money supply. While the previous system
caused inflation, which was increased by foreign aid, the current
situation is deflationary since there is a shortage of money in
circulation.
Trade is currently conducted in US dollars, Saudi rials and
Somali shillings. If the local currency wears out or is rendered
worthless by unauthorised printing of new notes, trade will
continue, but the use of foreign currencies may expand if
remittances and payment for exports continue. In the medium and long
term, Somalis will have to decide whether they want to create
regional currencies or a central one. There is no reason why several
currencies cannot co-exist, although there would be financial costs
to their doing so.
Trade and Credit
Export taxes should be kept to a minimum in economies that depend
heavily on foreign trade. Most taxes today are collected at the four
major ports and a few larger airports, which give the groups in
control considerable leverage. It is obviously desirable that
agreement can reached on the distribution of these resources.
Attempts will also have to be made to discourage internal customs
barriers, which can stimulate conflict. Most federations try to
maintain free internal trade to avoid such disputes.
Currently, small producers and traders use their own assets and
family networks for credit, while international traders also use
foreign banks. Central banks may not be necessary, as experience has
shown, and private banks can often be more effective. Community
banking systems, such as those run by NGOs in Bangladesh, Bolivia
and South Africa, could also be tried.
Property Rights and Agriculture
A means of sorting out property claims, returning assets to
owners, and compensating those who cannot return to their property
will need to be established. The legal system, courts of law,
prisons, and police could be financed locally, possibly with donor
support. Other issues of concern include the destruction of
rangelands, illegal charcoal production, over-exploitation of
fishing resources, and toxic waste dumping.
To improve agriculture, the private sector needs better credit
facilities to buy fertilisers, crops, etc. If subsidies are to be
involved from foreign NGOs, they should be careful not to discourage
local initiatives.
Health and Water
Whether for people or livestock, health care requires many types
of services that are often expensive and specialised. The major
consideration now is access to clinics, drugs, and doctors or
healers. Many of these services can be provided privately on a
fee-for-service basis, supplemented by foreign NGOs, which would
work similarly in unitary or decentralised political settings.
Poorer consumers could be subsidised by communities or foreign
donors, although it is often difficult to establish standards for
deciding who needs these services the most. Wells and water rights
may continue to be operated by clans and individuals as before, with
some involvement of regional governments.
Education and Roads
As with health, similar problems arise for education. Private and
public systems can operate at the same time, paid for by local
parent-teacher associations and foreign donors. Local communities
may manage and regulate their own schools, irrespective of type of
government. Higher education will be more difficult to provide in
the short-term, though it too could be controlled by local
governments.
The most widely used roads should be maintained by the widest
possible groupings, while local roads and footpaths can be a local
responsibility, with some sub-contracting to private firms where
necessary.
Security
A major concern will continue to be the huge number of weapons in
circulation. New forms of community-based security could be
established which recognise the right to bear arms, but also subject
people to the authority of clans and local authorities. A citizen
army could always be mobilised if there were an imminent danger, but
meanwhile, regardless of the nature of the state, i.e., confederal,
federal or decentralised unitary, costs would be kept down and
control exercised by those with most authority at the local level.
3. Remaking Administrative Structures
The six basic requirements here are: 1) how to organise a stable
administration for even the most decentralised government; 2) how to
overcome the assumption that every position is a prize for which
rival groups compete, especially for those posts controlling public
funds; 3) how to create administrative structures that allow policy
to be implemented while also meeting the challenge set out in (2);
4) how to save funds by coordinating some services and functions of
government on a state-wide basis; 5) how to resist any temptation to
create a uniform administrative structure throughout the country
(regional structures should reflect the needs of the region); and 6)
how to finance and staff, over the long term, these public
administrative structures, bearing in mind that numbers of people
employed should be kept to a minimum.
Minimum Functions
There are seven minimum functions necessary for the effective
running of a state administration:
1) Police and maintenance of order. In a unitary system, local
governments can be responsible for police who work alongside
national police forces. Under confederal arrangements, police forces
would be organised by their own states, although there could be a
coordinating body at the confederal level. In a federal system,
local and national police could share certain duties. Consociation
means communal self-government and is easier to organise in the
judiciary than in the police - though fair representation of all
clans is possible in a national police force.
2) Defence. Issues to consider here include land border
protection, prevention of smuggling, protection of fisheries and
off-shore resources, and the response to internal disorder and
disasters. These would need to be covered in whatever form of
government is adopted.
3) Fiscal policy. The issuing of a national currency would
function at a central level in both unitary and federal states,
while taxation and public borrowing are normally shared by the
centre and regions/provinces. In a confederation, these
responsibilities would normally be handled by the member states.
4) Collection, management and allocation of central revenue.
Assuming there will be some funds gained from export and import
duties, the centre in a confederal model need not be involved in
this, whereas each state in a loose federation could control these
functions, although some authority would be necessary to secure
subsidies for poorer regions.
5) Judicial systems. These vary according to form of government
and are discussed in Chapter 5.
6) Public service management. This is necessary under all models.
7) External representation. This is also a general requirement,
and is discussed in Chapter 6.
Other Functions
In the medium term, other functions that would be essential for
the proper running of any administration include: posts and
communications, which need to be accessible throughout the
country in any type of system, although it is not imperative that
this be centrally controlled, and could even be privatised; major
public works, such as roads; and tertiary education.
In the short term, however, there are several concerns that need to
be addressed, though not necessarily at the central level. These
include: primary and secondary education, health, veterinary and
agricultural services. Already in many parts of Somalia and the
'Republic of Somaliland', NGOs are assisting in providing some of
these services, most of which are paid for on a fee-for-service
basis.
The scarcity of trained staff for any of these functions makes it
important to offer services on as wide a basis as possible. Some
government funding will be necessary to plan for the future,
although this often causes spiralling costs as demands increase for
ever more services.
Staffing
Typically those working in the public sector are recruited on
merit (by competitive exams, specified qualifications, and some
consideration for representation), and are normally hired for life.
Neither of these would seem to be appropriate for Somalis. Because
of the clan-based nature of Somali society, representation should be
the primary principle of recruitment and promotion. Difficulties
will arise, especially concerning how many clans or sub-clans or
regions should be given representation, and this is especially
problematic when such jobs are to be kept to a minimum in order to
ensure that a greater proportion of available funds is directed to
substantive service provision and not primarily to salaries. This
dilemma might be resolved by limiting the period of service,
rotating jobs, and, should sufficient trust be established, holding
a census to determine exact numbers.
4. The Judicial Branch of Government
This chapter does not directly discuss the situation that would
arise if the former Somalia were to split into several independent
states - although the same general considerations would hold and
they might wish to operate common judicial institutions. There are
six key areas to consider in decentralised systems: 1) the supreme
court (or the highest court in the state); 2) the coordination of
judicial systems; 3) multiple systems of laws; 4) Islamic law; 5)
administrative justice; and 6) human rights safeguards.
The Supreme Court
A constitution is a guarantee that the agreement leading to the
coming together of different peoples and groups will be observed. In
confederal and federal systems, and in some decentralised unitary
states, defending the constitution is normally the task of the
supreme court. As such, the powers of the supreme court are very
important for the proper functioning of the state.
There are several fundamental questions related to the supreme
court. Who appoints members of this court? There is no set formula.
They can be appointed by the president or prime minister, or they
can be elected by the legislature. The 1960 Constitution of the
Somali Republic allowed for the involvement of both, and it may be
appropriate to consider a similar option in the future, for however
many states finally emerge.
How many members should there be? A possibility would be to
follow the method of the European Court of Justice, which has 15
judges, one from each member state. The more the court reflects the
diversity of the state or states, the easier it will be to have its
decisions enforced.
How long should a judge serve, and how can judges be removed? The
advantage of long terms of office is that decisions can be made
independently of political pressures, but this may also create
friction between the court and the executive branch, and in turn
cause the executive to either remove judges or pack the court with
supporters. Short terms of office, on the other hand, often cause
political considerations to take precedence as the continuation of
employment is somewhat dependent on the approval of the executive.
These obstacles can be overcome, especially for state courts.
Judges may be elected for medium length terms, such as six to ten
years, as happens in Switzerland or in states of the US. An
independent body can also appoint and discipline the judiciary.
Mandatory retirement is another possibility. Just as in appointment,
removal from office can involve different branches of government,
especially in controversial cases, such as removal from the supreme
court.
The supreme court should be the final court of appeal for civil
and criminal matters. It has the responsibility to uphold the laws
of particular states or regions, while also ensuring that common
standards apply throughout the federation. Confederations would not
normally have a supreme court, although each member state would have
one. In any future Somali federation, by contrast, this court would
also have the responsibility developing unified federal law. To
resolve constitutional disagreements, a combination of arbitration
and referendum could be adapted to traditional Somali
dispute-solving procedures.
Multiple Judicial Systems
An initial problem in reconstructing a Somali state or states
will be finding enough qualified lawyers. In a confederation or a
federation, an important issue to resolve will be how to enforce
decisions made by courts in one state in another state, as well as
between the state and the central government in a federation, as
different laws develop in different areas of the country. One
solution would be for the states to agree to enforce each other's
judgements, irrespective of whether they were in line with the
public policy or legislation of the enforcing state.
In all matters, though, it is important that (con)federal laws
are enforceable throughout the (con)federation. This holds for
criminal as well as for civilian law: the existing Penal Code in
Somalia could be re-applied while courts are being established, and
used as a basis by the successor state or states on which to
construct their own criminal law.
Multiple Laws and Islamic Law
Federations and confederations may develop multiple systems of
laws, which often complicate their administration and
implementation. Somalis already have experience of multiple legal
systems since customary and Islamic law have existed alongside
English and Italian law. This previous Somali practice could
therefore provide a possible model for a future decentralised state
or states, and representative commissions could be established to
ensure that laws are harmonised and disputes resolved during this
process.
Until 1962, there were separate shari'a and non-shari'a
courts in Somalia, with different methods of combining and
ordering the various forms of law. Some similar formulation could be
applied if Somalis agreed to establish a federation or a
confederation.
Administrative Justice
The current international emphasis on 'good governance' has
implications for the judiciary. For example, the new constitutions
of Malawi, South Africa and Uganda all give citizens the right to
fair treatment by government agencies as well as the right of appeal
against government decisions. The question is where to locate such
mechanisms, at the state or central levels, or both. Most decisions
directly affecting citizens are probably best taken at the regional
level, as people generally prefer legal disputes to be dealt with
informally, using customary practices wherever possible.
Human Rights Implications
Some international laws are binding on states even if they do not
wish to be bound. Even so, international human rights standards will
need to be written into law, not only to protect the individual, but
also to reassure potential donors and foreign investors.
Many treaties already exist that try to protect minorities by
establishing political, education, language, and property rights, in
addition to the fundamental rights theoretically enjoyed by all
individuals, e.g., non-discrimination, freedom of conscience, and
religion. Since it is easier to frame laws protecting human rights
than to implement them, there is scope for a careful study of Somali
traditional concepts of rights, and their relation to
internationally recognised human rights.
International obligations are easier to legislate in unitary states
than in decentralised systems because the centre has authority over
the laws of the country. Yet at the same time, abuse of power is
also more likely in such states. To avoid this danger the problems
associated with federations and confederations need to be addressed
at the start. For example, in federal systems one component
state may not comply with international laws, or inconsistent laws
may prevent the application of human rights norms. It is possible to
overcome such problems by mutual coordination of laws or by creating
a common Charter of Rights, and a constitutional court to oversee
the process of coordination and implementation. The advantages of
federal systems are that laws can be adjusted according to local
needs. Federal systems also can protect minority rights, especially
in terms of language, religion, education, marriage and family
issues, where relevant. In confederal systems, each state
would have responsibility for ensuring that human rights obligations
are followed, although here too it would be desirable to create a
coordinating body, or even to allow a single organisation to frame
such laws.
5. Alternatives for Defence and Foreign Relations
Issues and problems related to foreign affairs will be similar
irrespective of how many states are created. As foreign relations
are relations between states, their conduct is generally guided by
central structures, though adjustments can be made to cater for the
needs of decentralised government. There are seven main concerns
related to foreign affairs and defence that need to be addressed: 1)
treaty-making; 2) foreign representation; 3) defence; 4)
citizenship; 5) foreign trade, customs and excise; 6) foreign aid;
and 7) currency.
Treaty Making
Treaties, whether they pertain to membership in international
organisations such as the UN, or bilateral relations between states,
are normally signed by some central authority, and then ratified by
the president (or head of state), parliament and/or constituent
states. All government systems, from a centralised, unitary state to
a confederation, operate in this manner, although some federal
states allow a degree of independence in treaty making. For example,
in Australia and Canada, the provinces may sign treaties that deal
with foreign trade matters, as long as such treaties do not conflict
with those signed by the central authority. Further, in situations
where the province or canton or state signs its own treaties, the
central authority may also retain the right to approve it. Conflicts
resulting from treaties signed by different authorities can be
resolved by an independently appointed group, comprised of members
from all segments of society.
Confederations allow for more decentralisation. The options for
treaty making in confederal systems are as follows: 1) the central
authority is granted full rights to conclude international treaties,
while constituent states retain control over domestic matters; 2)
member states have some independence in treaty-making as long as
they are not in conflict with the interests of the confederation;
and 3) member states can conclude treaties with foreign powers
independently, while at the same time sharing common services and
rules covering relations amongst themselves, as well as relations
with external states in particular areas. In very loose
confederations, such as the European Union, where members still have
not granted the central authority sole jurisdiction over all
treaty-making, even on issues which have been transferred, they
retain the right to veto legislation that conflicts with their
interests. For such a loose arrangement to work, there must exist a measure
of trust between the centre and the constituent parts. If
Somalis were eventually to agree to confederate along similar lines,
it would also be necessary for the component states to honour
pre-existing agreements signed by earlier heads of state of Somalia
(unless they were opposed to the principles of the new state or
states).
Foreign Representation
The right to conclude treaties generally also determines who
represents the new state or states in embassies abroad, at
multilateral institutions, and international conferences. In unitary
and federal states, the appointment of such representatives is
usually controlled by the head of state. In confederal systems, such
appointments will depend on how extensive the sharing of foreign
relations decisions is between member states. The range of options
for diplomatic representation may be illustrated by the European
Union (where member states each have their own foreign embassy
staff) and the United Arab Emirates (where such staff represent the
federation and come from all member states).
The key issues concerning foreign representation relate to recruitment
of staff and the cost involved.
The first difficulty is how to recruit personnel, whether on
merit or according to a system of community representation (based on
region or clan, for example), or on a combination of the two. In
theory, this dilemma should not arise in unitary or federal states,
but it would be a problem in confederal or decentralised unitary
states, and could be very serious in states with deep divisions
between groups, where no group trusts the others to fully represent
its interest, e.g., a Kurd in Iraq would be unlikely to trust a
non-Kurdish Iraqi to represent his or her interests in country-wide
institutions. In such a situation, the solution may be to rotate
offices between clans or regions, and establish safeguards so that
while in office, each representative would not be able to pursue his
or her group's interests at the expense of others.
As the new state or states will inevitably be short of funds, a
further difficulty arises concerning how to pay for such offices.
One solution would be to minimise the number of offices. For
example, there could be a delegation to the UN which might also
cover relations with the United States, a representative in Brussels
to handle the European Union as well as bilateral relations with
member states, a representative to work with neighbouring East
African states, one for the OAU, and a small group to cover regional
issues such as trade, refugees, drought and food security. The need
to deal with such issues will arise under any system of government
but in no case do they require a large diplomatic establishment.
Defence
Even in decentralised systems, it may make sense for the centre
to provide for common defence, in order to reduce the financial
burden that is inevitably involved. Indeed, it is to make such
savings that many federations and confederations are formed, as was
the case in the short-lived Senegambian Confederation, or the United
Arab Emirates. In most states, whether they be unitary or federal,
responsibility for declaring war and maintaining the defence of the
union rests with the centre.
In Africa, however, central control over defence has often proved
disastrous. In many countries, there should be no need for large
defence budgets, since they face no serious external threat. Yet
almost invariably the government devotes a substantial portion of
its expenditure to defence. The army often ends up usurping power,
and far from protecting society, threatens the civilian population.
The problem is even worse in states that have fallen apart due to
civil war, and where, as in the former Somalia, most adults are
armed. In these cases it is seldom possible to disarm the population
and/or abolish the armed forces without enlisting the support of an
outside power, which is likely to be unpopular and widely resisted.
Yet without a regional threat there is no need to reconstruct a
standing army. Once again, Swiss experience may offer some useful
lessons for Somalis: the Swiss have no standing army, but if an
emergency occurs, they can easily assemble a force from the
different regions of the country, particularly as all citizens have
to undergo a period of military service.
An additional safeguard to prevent internal conflicts between
clan militias or regions could take the form of a dispute settlement
group with representatives from all parts of the country and all
clans, which could be assembled in time of need. Such a group would
resemble the gathering of clan elders to settle inter-clan disputes.
If there is insufficient trust to set up such a group, at least
initially, an alternative would be to contract out this function to
an international or regional organisation, such as the OAU. This
would have the advantage that the donor community might be more
willing to reinvest in whatever state or states are established if
there are transparent mechanisms for preventing the abuse of power.
A final defence issue is the protection of the environment. In
the absence of a Somali government or governments, Somali waters are
already being overfished and used as a dumping ground for toxic
waste. In the short term, it might be possible to contract out
supervision of these activities to an outside agency to monitor and
safeguard Somali natural resources.
Citizenship
In most systems, issues related to citizenship, such as the
issuing of passports, naturalisation, emigration and immigration are
handled by a central authority. Some countries allow dual
citizenship, and in confederal systems, like the European Union,
citizens belong first to their state, and second to the union. This
could work for Somalis: they could have common citizenship, and also
belong to their own regions or states, where they would vote. Thus a
Somali could live in any part of the reconstituted state or states,
but perhaps only vote in his or her particular area. If this were to
be agreed, then each region could also issue its own passports, as
members of the European Union already do.
Foreign Trade
As Somalis have demonstrated, foreign trade can continue even in
chaotic political circumstances. But for trade to contribute to
economic recovery over the long term, it is important that it take
place in a stable and secure environment. In some federal systems,
the different provinces maintain independent trade relations with
foreign states, as do the states in most confederations. Providing
the rules, enshrined in the GATT, WTO and Lome conventions, are
applied, it would be possible to devolve responsibility for trade
policy to regional governments.
In most countries matters relating to customs and excise are
under the overall control of a central ministry. The main reason is
to prevent trade wars between the constituent states. For example,
without some regulation they might be tempted to compete with one
another by lowering import taxes. In unsettled conditions, this
would not promote healthy competition but make it more difficult to
maintain the peace, as it did in the Senegambian Confederation
before its collapse (the Gambia refused to give up its low tariffs
and liberal trade policy).
The general problems that come up with regard to trade relate to
the collection of taxes and redistribution of funds, normally from
wealthier regions to poorer ones, and how to convince those
controlling strategic assets, such as ports and airports, to give
them up. In the short term, it may be necessary to leave such assets
under their current 'managers'. Each region would have
responsibility for buying into whichever services (health,
veterinary, fishing, transport, etc.) they needed, raising the money
to buy the service by taxes within their region.
Foreign Aid
Decentralisation is probably necessary if development aid is to
revive. Donors increasingly attach political and economic conditions
to aid and will need to be assured that the local government(s) are
legitimate and stable, and that the money will be used wisely. From
this point of view decentralised governments would have an advantage
over unitary ones in that legitimate representatives from local
areas could manage (and petition for) projects, rather than having
them administered by a centrally located agency. Aid could be
managed in this fashion in confederal unions or federal states. Even
in decentralised unitary states responsibility for foreign aid could
be delegated to local authorities. Indeed it is possible that donors
might make such delegation a condition for the resumption of aid.
In addition, with regard to disaster relief, as opposed to
development assistance, attention will need to be paid to
re-establishing early warning systems so that future famines may be
prevented. International regional organisations could be enlisted in
the first instance, on the condition that they train Somalis to
staff their own offices in the long term.
Currency
Authorities controlling the money supply are normally centrally
located, though they may be independent of the government, as the
Federal Reserve is in the United States, or they may maintain
different values for domestic and foreign transactions, as the
Republic of South Africa did until recently. In some countries where
the local currency is very weak or has collapsed, a foreign currency
often circulates next to the local one. But there are no unitary
states or federations where rival currencies compete with each
other. This is because of the fear that the authorities in one state
or province might off-load their economic problems onto their
neighbours by depreciating their own currency to make exports
cheaper and imports more expensive. The situation in confederations
is more ambiguous (see 1 and 2 below).
There are four possible arrangements for currency that comply
with decentralised models. They are:
1) Each state or region could maintain its own currency and
exchange it at a rate agreed upon by all parties. In this instance,
it would be better to peg its value to a major international
currency or basket of currencies. Difficulties may arise, however,
if one region is wealthier than others.
2) A currency system modelled on the European ecu could
be established. This would take the previous idea one step further
by pegging each region's currency to a common one. This could work
as an interim solution while trust was being restored in the new
political structure. Eventually the common currency could replace
regional currencies altogether.
3) The pre-existing Somali shilling could continue to be used for
trading purposes along with a major international currency, such as
the dollar, which would gradually replace shillings as they wear out
or are devalued because of possible over-printing.
4) An independent monetary authority could be created for the
entire state or states. Initially, staffing for this organisation
could be contracted out to an international financial institution,
such as the IMF, which could also train Somali bankers.
6. Principles of Constitutional and Electoral Design
If Somalis decide on some form of decentralised democratic
system, they will need to design it in a way that will prevent a
return to authoritarian rule. It will need a separation of powers,
checks and balances, and some form of consociational or consensual,
co-operative government.
Parliamentary Democracy
Democracies can be organised in several ways. One is
parliamentary democracy, in which the legislature has most of the
power. In the United Kingdom, for example, the legislature (or
parliament) has significant control over the executive because the
authority of the prime minister and the cabinet ultimately rests on
the support of parliament. There are nine key features of this type
of system: 1) a concentration of executive power, normally in a
one-party government; 2) a mixing of executive and legislative
power; 3) a two-chamber parliament, with one chamber being less
powerful; 4) a mainly two-party system; 5) party competition
normally over left and right issues, such as economic policy; 6)
simple plurality rule which gives seats to the candidate with the
most votes; 7) unitary and centralised government, where local
governments can be abolished or curtailed by the centre; 8) an
unwritten constitution, with minimal judicial involvement in its
interpretation; and 9) representative democracy with little use of
referendums.
Parliamentary and majoritarian systems, which give power to the
elected majority, only work well in fairly homogenous societies that
are based on the concept of liberal individualism. In deeply divided
societies, such systems do not function well, and it is often the
case that a dominant community can take over and run the society for
its own purposes.
Consensual/Consociational Democracy
Another form of organising a democracy is the consensual or
consociational method, which is the opposite of the former system.
Consensual models try to maximise participation and representation
in government while also holding back attempts by a powerful group
to take control. This type of system may be more appropriate for a
divided society, and has something in common with traditional Somali
decision-making processes.
In consensual systems, there are eight distinctive features which
could work in a federal Somali state that was based on a multi-party
system, or in the independent states of a confederal system: 1)
executive power sharing that enjoys support from many communities,
including a multi-person rotating presidency with
representatives from each of the regions, veto powers, and separate
elections for the executive and parliament; 2) separation of
executive, legislative and judicial powers that allows for
co-operation between the branches of government, or perhaps in the
Somali case, a 'council of state' formed by elders and former
presidents and prime ministers who would mediate disputes and take
certain issues to the Supreme Court; 3) two federal chambers with
equal powers and minority and/or territorial representation; 4) a
multi-party system; 5) a multi-dimensional party system where
language and clan differences, for example, are included in party
platforms; 6) a proportional representation voting system in which
the elected chambers genuinely represent the population and
encourage some integration; 7) a type of decentralisation that
allows some communities or clans greater autonomy; and 8) a
constitution that gives minorities specified rights.
Consensual systems enable all those who are affected by political
decisions to take part in making those decisions. They also allow
distinct communities the opportunity to govern themselves, so long
as they respect the rights of others to do the same.
Electoral Options
There are numerous types of electoral arrangements that can be
adapted to a society's particular needs. In a federal system, each
state or region could equally choose separate systems so long as
they fall within overarching electoral rules. The same is true for a
confederation. It must be remembered, however, that no outcome is
certain in the different systems.
Proportional representation (PR) systems ensure a proportional
relationship between the votes for parties or candidates and the
seats they win. PR systems differ from non-proportional (NPR)
systems, which are often called majoritarian systems. There are
three ways of organising NPR systems: 1) plurality rule, where a
candidate is elected in a single person district and wins if he or
she has the most votes; 2) alternative vote, where a candidate only
wins if he or she has a majority of the votes; and 3) exhaustive
double-ballot procedure, where if no candidate wins in the first
ballot, there will be run-off elections between the top two
candidates.
The advantage of having a NPR system in the Somali context would
be that it would encourage parties to broaden their appeal to
include many clans. The disadvantages, however, are three: 1)
parliament could be filled with many small parties, which could in
turn make it easier for a strong party to take control; 2) the
'winner takes all' principle would not be acceptable to those that
lost; and 3) dominant coalitions of clans could try to damage weaker
clans or parties. PR systems, on the other hand, may be more likely
to produce a stable government, and would cope better with
population shifts. In a PR system, moreover, clans could be allowed
to form their own parties or coalitions if they so wished.
7. Conclusion: The Menu of Options
A concluding menu lists the advantages and disadvantages of the
three constitutional models considered in the report, i.e.,
confederal, federal and decentralised unitary state, and also of
consociational government. This analysis, which is already in
summary form, is not reproduced here.
Functional Co-operation
There is one final possibility to consider: co-operative
arrangements without an overarching political authority. Functional
services, in which most Somalis have an immediate interest, and on
which they might be prepared to co-operate include veterinary or
health services; fishery protection and licensing; educational
services such as curriculum development, the provision of textbooks,
teacher training, and the negotiation of scholarships for Somalis in
foreign universities; posts and telegraphs; weights and measures;
and currency. Thus if attempts to construct a confederation,
federation, or decentralised unitary state break down, or proceed at
a very slow rate, agencies could still be created to provide
specific services to meet certain needs. Indeed there is a
traditional Somali precedent in the concept of contractual agreement
(heer ).
There are two ways in which such functional agencies could be set
up, although they are not mutually exclusive. The first and
theoretically most desirable would be by agreement amongst those
Somalis from different clans, regions and/or 'states' who share a
particular functional interest. Such agencies would employ Somalis
and this might make external donors more willing to reinvest in
their future. The second would be to contract out such services to
an international agency, such as the World Bank. This would have
similar advantages as the previous one regarding employment and
foreign investment, and avoid the disadvantages that might spring
from continued inter-clan competition, even over the provision of
agreed welfare services. Such collaboration might foster wider
Somali sentiments of national solidarity.