P International Journal of Social Forestry (IJSF), 2009, 2(1):43-66

ISSN 1979-2611, www.ijsf.org

© Copyright 2009 CSF.

THE INSTITUTIONAL SPHERE OF COFFEE FOREST

MANAGEMENT IN ETHIOPIA: LOCAL LEVEL
FINDINGS FROM KOMA FOREST, KAFFA ZONE

Till Stellmacher1 and Peter P. Mollinga2

Abstract
Ethiopia’s coffee forests have witnessed high rates of deforestation during the last
decades. Main reasons identified are intensified forest resource utilisation and
expansion of smallholder agriculture. These are major drivers, however, as the
processes and impacts are mediated and promoted by institutional arrangements
through which intensification and expansion unroll. This paper interprets
institutional arrangements primarily as rules and regulations. It provides an
understanding of the particular (informal) forest resource use rights of smallholders
in Kaffa Zone, South-western Ethiopia. Given the path dependent character of land
tenure and property rights institutions, this research takes a historical perspective. In
the case study area, the coffee forests are historically divided into use right plots
individually held by local peasants. The nationalisation of all land in Ethiopia in 1975
was the major institutional turning point in which responsibility for forest use and
management was by decree shifted from local peasants towards newly established
centralised state entities. These bodies neither had experience, expertise nor financial
resources to tackle the challenge of forest governance. In practice, state control did not
‘reach’ the forest areas, and rather created a muddled and frequently changing
institutional framework that turned out to have no, or only minimal actual impact.
Consequently, use and management of forest resources in large parts of Ethiopia
remained defined by traditional use rights.

Keywords: coffee forest, institutional analysis, Ethiopia.

Introduction
Ethiopia is the cradle of worldwide Arabica coffee. The montane
rainforests in South-Western Ethiopia are the only place in the world where

1 Center for Development Research, Department of Political and Cultural Change, Walter-Flex-
Str. 3, 53113 Bonn, Germany, Tel: 0049-(0)228-734907, Fax: 0049-(0)-228-731972, e-mail:
t.stellmacher@uni-bonn.de
2 Center for Development Research, Department of Political and Cultural Change, Walter-Flex-
Str. 3, 53113 Bonn, Germany, Tel: 0049-(0)228-734918, Fax: 0049-(0)-228-731972, e-mail:
pmollinga@uni-bonn.de
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

coffee naturally grows in its original habitat. This coffee forest ecosystem
presents a biodiversity hotspot of worldwide importance. An estimated 30%
of the total Ethiopian coffee production originates from forest and semi-forest
coffee cultivation systems, contributing about 20% of the whole export
earnings of the country (Abebaw & Virchow 2003). In these cultivation
systems, coffee grows naturally under the full coverage of a primary
rainforest canopy and is utilized as a non-timber forest product with very low
labour and cash input (Teketay 1999).
Forest and semi-forest coffee is the most important cash crop in Kaffa
Zone; the coffee producers, however, live in extreme poverty. While their
livelihood traditionally depends on low-yielding subsistence agriculture and
the utilization of forest coffee for income, coffee producer prices are, however,
considerably low and highly fluctuating.
At the same time, the coffee forests of Kaffa Zone witness alarming
deforestation, at annual rates of up to 9% (Boum 2002). This is mainly due to
the expansion of smallholder agriculture and over-utilization of timber and
non-timber forest products driven by poverty. This development does not
only promote change of local climate, land degradation, erosion and scarcity
of forest products—all aggravating the poverty cycle—but also leads to the
irreversible loss of forest biodiversity and the coffee gene pool.
Empirical field research providing the background of this paper has been
conducted in Kaffa Zone—right in the geographical center of the Ethiopian
coffee forests in South-Western Ethiopia. Methodologically, a combination of
structured and semi-structured interviews, focus group discussions, and
visual communication techniques such as mental forest mapping and timeline
mapping was used.

Theoretical Background
According to the “Institutional Analysis and Development” framework
(Kiser & Ostrom 1982, Ostrom & Gardener et al. 1994, Ostrom & Burger et al.
1999, Koontz 2003, Ostrom 2004), decisions to deplete or destroy a forest are
produced in a social sphere where individuals interact over processes and
impacts, called the ‘action arena’. The decision making processes within the
action arena are influenced by three groups of exogenous variables, namely a)
the attributes of the natural resources (forest in our case), b) the attributes of
the community (the forest users), and c) the institutions.3

3 The interaction between humans and nature, people and natural resources, is mediated by

technology and institutions. This paper focuses on the institutional arrangement of forest
resource use. Technological mediation is absent both in the IAD framework and in the paper.
The technologies for the production of forest coffee in Ethiopia are extensive technologies with
low labor and capital input generating low output. It is therefore assumed that that institutional

44
 The institutional Sphere of Coffee Forest Management (Stellmacher & Mollinga)

In the past, scientists concerned about forests and forest loss identified the
attributes of the forest user communities, with their specific historical,
cultural, social and economic background, to be the main determinants for
forest depletion and destruction in the so-called developing countries.
Research focused primarily on demographic (population growth) and
economic (increased market pressure) factors as the primary threats for the
concerned forests.
This supported the argument that endangered forests need to be
prevented from the (over-) utilisation by an ever-growing population with an
ever-growing propensity to extract forest resources for economic gain. This, in
turn, offers justification for exclusionary forest conservation approaches,
which, in Africa, are often associated with nationalisation of forestland.
Nevertheless, after many unsuccessful attempts it became evident that ‘no go
conservation’ cannot be an appropriate instrument to actually halt forest
degradation and loss (Poteete & Ostrom 2002).
Since the early 1990s, an increasing number of studies revealed that
variables other than population pressure might play a role in depletion and
loss of forests in developing countries. Arun Agrawal, among others,
indicated that despite rapid population growth and increasing market
pressure, some local communities are able to manage forest resources in a
sustainable way, while others—in a similar situation—overuse and destroy
them. There is strong evidence that institutions are the decisive ‘missing
factor’ in this regard as these direct demographic, social and economic
developments within a given environment by providing incentives and
disincentives that shape human decision making and action in regard to forest
resource use, management and conservation (Agrawal 1995, Agrawal &
Yadama 1997, Bodin & Tengö et al. 2005). Consequently, understanding the
institutional arrangements through and within which forest resource users
operate is essential when thinking about sustainable forest management and
conservation.
Studying institutions has a long and established tradition among scholars
interested in social science, starting from Aristotle’s times. Ever since, a great
number of approaches from various disciplines were developed, resulting in
different theoretical frameworks with no common research programme,
methodology, and understanding what is actually an institution (Immergut
1998). A prevalent much quoted definition in the social - including economic -
sciences is the one of Douglas North, who understands institutions to be
“humanly devised constraints that shape human interaction [and] structure

mediation is the more important element for understanding the process and impact of forest
resource depletion.

45
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

incentives in human exchange, whether political, social, or economic [as well as]
define and limit the set of choices of individuals" (North 1990).
Beyond shaping human—human interactions, institutions can have a
considerable role in shaping human—nature relationships. In this sense,
institutions represent the critical role in the relations between humans and
forests as they can determine and regulate human access to forest resources,
and “bring order into disorder” (Nørgaard 1996). Institutional arrangements
provide a structure that says which person has which particular right to use
which forest resource to what extent, denote manners in which way goods
and services are to be used, as well as conservation measures that are to be
accomplished. As such, institutions are not neutral. They may contradict
individuals’ attempts at gaining short term benefits and can exclude or
include individuals or groups from access to resources (Carswell & de Haan et
al. 1999). Hence, the analysis of institutions needs to address matters of
institutional plurality, imbalances among actors, and conflicts arising out of
these, especially when resources are central to appropriators’ livelihoods - as
forest resources often are in rural areas of the developing world (Watson
2001). Accordingly, the concerned institutions demand explicit enforcement,
which entails that beyond their pure existence, institutions need to be
complemented by implementation and enforcement instruments that are able
to address the consequences of contravention (GTZ 2004).

The Changing Land Property Rights and Regimes in Ethiopia

Property rights are understood as “an enforceable authority to undertake
particular actions in specific domains” (Ostrom 1990), or—more applied to the
research topic concerned—as a bundle of institutions at the constitutional
level that can entitle actors with authorized access to resources and to the
benefits derived from them as well as to the potential to deny these benefits to
others (Bromley 1986). Land is an important production factor in the agrarian
society of rural Ethiopia. Land rights have an impact on forest resource use,
management and conservation, as well as on the livelihoods of forest users.
Whether forest land is the property of the state, an individual, a group of
individuals, or a combination of the above and whether tenure and property
rights are actually executed, these are the critical concerns.
The most important land property rights system in Ethiopia in terms of
geographical dispersion and formal recognition is the one imposed by the
state. Below therefore, we discuss the state-initiated land policies as well land
administration as its executive structure. State-initiated land policies are the
“formal decisions, laws and programmes” (Keeley & Scoones 2000) designed
for the different administrative levels. Land tenure and property rights
arrangements are ‘path dependent institutions’, in which history determines
not only the present situation but also future developments and prospects for

46
 The institutional Sphere of Coffee Forest Management (Stellmacher & Mollinga)

change. Policies evolve and change over time as a “result of the interplay
between context specific circumstances and the changing effectiveness of
different networks of actors in the policy debate” (Keeley & Scoones 2000). In
our analysis, considerable emphasis is thus given to the past, commencing
with the time of Haile Selassi I.
In the Emperor’s time before 1974, land tenure and property right
arrangements in Ethiopia were some of the most complex in the world. They
included private, communal, serfdom, state and church land property right
arrangements, among others (EEA/EEPRI 2002). Kaffa Zone, from the end of
the 19th century onwards, had been largely dominated by a landlord—tenant
system (neftegna—gebber in Amharic) with feudal characteristics in the form of
the gult system. After the Ethiopian revolution in 1974, the new military
government, popularly known as the derg4, proclaimed the abolition of the
political system and the award of all ‘land to the tiller’, i.e., assignment of land
tenure rights to the peasants. The subsequent land reform of 1975 aimed at
putting this proclamation into practice and was probably “one of the most
radical land reform[s] ever attempted in Africa” (Pausewang 1990) laying the
foundation of a land tenure system based on the socialist model. All land
tenure and right systems in Ethiopia—including the gult—were abolished,
landlords were dispossessed and land holdings, whether farmland, grazing
land or forests, were nationalised and came under direct possession of the
state (Stellmacher 2007a).
The administration of land was vested nationwide into a newly
established “Ministry of Land Reform and Administration”. At the local level,
the enactment of the “Proclamation to Provide for the Nationalization of
Rural Land No. 71/1975” led to formation of Peasant Associations, the lowest
administrative units in which a number of villages and hamlets (gots in
Amharic) were grouped together (McCarthy 2001). Peasant Associations were
given full control over the distribution of land and, starting from 1975, they
gave usufruct land rights to the peasants living in their administration as well
as to Agricultural Producers’ Cooperatives. The condition for obtaining land
use rights was the peasants’ permanent physical residence in the Peasant
Association and their capability and willingness to farm themselves and to
meet a number of administrative dues and obligations (EEA/EEPRI 2002,
Pankhurst 2002).
In the post-revolutionary years, the weaknesses of the nationalization
approach became evident. Most notably, since Peasant Associations
administrations frequently withdrew and reallocated land tenure rights from
peasants—often on short notice and without compensation. That uncertainty

4Derg (Amharic), also referred to as dergue, stands for committee or council used as a short form
for “Armed Forces Coordinating Committee”.

47
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

reduced the incentives to invest in land maintenance and improvement

measures, aggravated the problem of soil mining, and led to short term
oriented measures on the allotments (Marena Project n.d.). In addition, land
could not be used as a collateral to secure credit, resulting in low
creditworthiness and only marginal financial investment by the peasants
(EEA/EEPRI 2002).
After a violent regime change in 1991, the new transitional government,
dominated by the “Ethiopian People’s Revolutionary Democratic Front”
(EPRDF), took power. Notwithstanding the fact that economic liberalisation
was emphasised as one of the key transformation objectives, the “Constitution
of the Federal Democratic Republic of Ethiopia” approved in 1995 and valid
to the present day states: “Land is a common property of the Nations, Nationalities
and Peoples of Ethiopia and shall not be subject to sale or to other means of exchange”
(Article 40). In addition, the 1995 constitution states that legal rules and
regulations should be enacted for the utilisation and conservation of land and
other natural resources (Article 51), and that the governments of the newly
established regional states are designated to have more responsibility in
administering land and other natural resources under the federal laws (Article
52). Two years later, the new “Federal Rural Land Administration
Proclamation, No. 89/1997” was enacted, in force until the present day. By
this proclamation, the power of land administration, i.e., the “assignment of
holding rights and the execution of distribution of holdings” (FDRE 1997), was
shifted from the national bodies to the regional states. The system of Peasant
Associations as the lowest administrative level has remained without much
modification until today, except that the name was changed into kebele
association.
In summary, the revolutionary nationalisation of all land holdings in 1975
was the major institutional turning point in which land and forest
responsibilities were shifted from local to central institutions and authorities.
However, it soon became clear that the change of an institutional setting and
the transfer of rights and duties from one institutional system to another does
not necessarily bring a change for the better, that approving new rules and
regulations does not guarantee their enforcement in reality, and that
proclaiming new administrative bodies does not mean that they are actually
effective in the achievement of their objectives. In practice, the new state-
initiated bodies neither had the necessary organisational structures nor the
experience, the expertise or the financial and infrastructural resources to
tackle the cross-cutting and nation-wide challenge of governing the Ethiopian
forests.
The “National Forest Priority Area” approach, commenced in 1980, can be
historically portrayed as an ambitious attempt to tackle the problem of loss
and degradation of the primary forests in Ethiopia. However, it did not work

48
 The institutional Sphere of Coffee Forest Management (Stellmacher & Mollinga)

out in practice. This failure can be mainly traced back to the fact that attempts
were made to administer and conserve forests in a top-down manner, hence
by means of a centralised body most of the time under the “Ministry of
Agriculture and Rural Development”, and integrated into the overall political
structure. Similar to nationalisation and top-down forest policies in other East
African countries such as Tanzania and Kenya, also the ruling decision
makers in Ethiopia overestimated the state’s institutional capability,
effectiveness and efficiency for implementation, enforcement and monitoring
of their forest concerned policies. Concomitantly, pre-revolutionary land use
rights persisted, and customary forest use, management and conservation
practices were neglected. The state efforts at environmental protection in
general and primary forests in particular, in force since 1975, did not bring the
designated results under any Ethiopian government. State control did not
‘reach’ the forest areas. Instead, it created a muddled and ever changing
institutional framework that turned out to have no or minimal impact in
terms of the stated objectives. We subsequently discuss the underlying
institutional determinants that actually work in practice when it comes to use,
management and conservation of the concerned coffee forests.

Understanding Institutions from the LocalEexperience: The Case of Komba

Village
Komba village is located in Gimbo District of Kaffa Zone. It is an
administrative unit of the Southern Nations and Nationalities Peoples
Regional State (SNNPRS). Kaffa Zone is known as the birthplace of Arabica
coffee. According to an Ethiopian legend, the etymological origin of coffee
and its miscellaneous variants in different tongues can be traced back to
‘Kaffa’, its place of origin.
One third of Kaffa Zone is covered by montane moist forests. The terrain
is dominated by a dissected tableland with flat to moderately undulating
terrain on areas above 1,500 meter above sea level (asl), with a maximum
altitude of 3,350 meter asl (Ersado 2001). The area enjoys some of the highest
rainfall in Ethiopia (Mayne & Tola 2002).
Livelihoods are mainly based on subsistence, rain-fed farming and
horticulture (maize and enset as the major staple foods) and the utilisation of
forest products. Coffee from forest, semi-forest and garden production
systems as well as honey are the main cash crops. Plantation production plays
a marginal role in socio-economic and ecological regards (Stellmacher 2007b).
Komba village was chosen as a case study for three reasons: a) it is
situated within Koma forest, one of the largest coffee forest areas of Kaffa
Zone and facing high rates of deforestation, b) the utilisation of forest coffee
plays a crucial role in the income generation of the local population, and c) the
village with around 40 households is relatively small allowing in-depth

49
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

research. The local level case study approach justified the combination of
qualitative and quantitative instruments of empirical social research. In total,
interviews with 160 key informants (elders, kebele and iddir chairmen, NGO
workers) and ten focus group discussions with farmers were conducted.
Supplementary socio-economic data was obtained by means of a total
household census in Komba village. Data obtained was cross-checked and
triangulated. Main field research was conducted between 2003 and 2006.
Subsequent field visits to Komba village showed that the empirical evidence
provided is still relevant today as the major institutional, socio-economic and
environmental conditions and constraints remain unchanged.
Findings show that despite the creation of a new institutional structure by
the state, the people-nature interrelation in parts of rural Ethiopia continue to
be influenced by community-initiated institutions that originate from within
the local society, have historical legitimacy, are adapted to local realities, and
are reproduced over generations. Local institutions, however, are not
unchanging in form and meaning, nor equitable in access and impact. There
are a great range of institutions, with different roles and functions, and
operating in different ways with a diverse set of rules and responsibilities.
The fact that community-initiated institutions in rural Africa persisted
during decades of influence and interference from ‘the outside’, e.g. European
imperialists or the centralised national state, is not new as such. It is, however,
not thoroughly understood local institutional arrangements impact on the
use, management and conservation of the Ethiopian coffee forests. In Komba
village in Koma forest of South-western Ethiopia, two sets of institutional
arrangements are of particular importance, namely:
a) the institutional arrangements for collective coffee forest management
(including harvest of forest coffee berries), which have taken the
organisational form of neighbourhood working groups;
b) the legislative and juridical structures that provide the operational set
of institutions determine land and forest resource use rights.
These are discussed in the following sub-sections.

Collective Forest Management: daddo and dabbo

In many African societies, one of the benefits of being member of a
community is the access it provides to the labour of other community
members on a non-cash basis (Tache & Irwin 2003). In the coffee forests, a
wide range of activities are carried out by neighbourhood working groups, be
it the relatively light work of forest coffee harvesting, the production and
putting up of beehives which requires more knowledge and skills or—at the
top of the ‘hard labour’ scale—timber-logging and construction of tukuls, the
thatched round huts omnipresent in South-western rural Ethiopia. For
collective labour, two types of working groups are relevant in Komba village,

50
 The institutional Sphere of Coffee Forest Management (Stellmacher & Mollinga)

daddo and dabbo. Both provide short term work teams beyond kinship ties on
the basis of equal labour exchange in times of high labour demand. In both
working groups, the households being assisted are expected to provide
sufficient food (maize and kocho5) and a beverage (coffee and tella6), for the
workforce involved, hence the arrangements provide for a certain ‘food for
work’ aspect. A daddo working group consists of about three to five people
working together half-day. Dabbo is “much bigger than daddo” (23/11/2004:
Alemayehu Ketto, secretary of Komba iddir) in terms of workforce involved,
expenditure of time and action radius7 (Stellmacher 2007b). All households in
Komba village are engaged in daddo and dabbo at any time of the year. Labour
is exclusively carried out by adult men, whereas women are busy preparing
food and beverage for the workers. If a household requires a larger numbers
of workers at a specific time, the respective head of the household can ask at
an iddir meeting (see the discussion below) for work assistance.
Beyond guaranteeing a high level of reciprocity and social cohesion, these
working group arrangements provide a substantial level of social control
against individual (over)use of forest resources. The working systems are not
spontaneously evolving, but are arranged by and respond to a higher level
village administration, the iddir.

The ‘Formal Tradition’ of iddir

Iddir is an Ethiopian phenomenon. It can be found all over the country
and even among Ethiopian communities abroad, across all social classes,
ethnicities, and religions. Nevertheless, although often thought to be, iddir is
not an ancient institution. The first iddir developed as late as during the Italian
occupation in the 1930s (Abbot & Hailu et al. 2000, Pankhurst 2002). There is
no clear and consistent definition of what iddir actually is. Literature describes
it as a financial institution informally organised by local communities in
which members regularly contribute to a common pool with a view to
mutually support each other (Aredo 1993). There are, in fact, several types of
iddir that vary in terms of organisational arrangements, aims, rules and
regulations, as well as social and economic membership. Some iddir are, for
example, potentially open to all members of a community, others are
restricted to a certain group of people (women iddir, youth iddir etc.).

5 A staple food made from enset, a plantain-like perennial crop endemic to Ethiopia, often
refereed to as the ‘false banana’.
6 A self-brewed light beer.
7 The people in Komba village differentiate between forest areas relatively close to their

homesteads, called kubboo in the local language Kaffichio, which are used and managed
individually on a daily basis (e.g. by women for firewood collection), and more far-off forest
areas, named kuddoo, that are entered only in groups, which is also due to security reasons like
the protection from wild animals.

51
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

In Komba village, there is only one type of iddir, which is multifunctional.

Following the stringent ethnical differentiation in Komba village, however,
Komba iddir is in fact subdivided into one iddir for Kaffa and one for Mandjah
people (subsequently termed Komba Kaffa iddir and Komba Mandjah iddir).
Both are governed by an iddir committee, composed of five committee
members each, namely chairman, treasurer, secretary and two additional
members with advisory functions.
The Komba Kaffa iddir and Komba Mandjah iddir are sub-iddir of the
higher level Agama iddir. Agama iddir, in turn, is one out of five umbrella iddir
in Yeyebitto kebele (district). Membership in one iddir implies being a member
of the higher level one, thus: “if you are a member of Komba sub-iddir, you are also
a member of Agama iddir” (7/10/2004: Hailemariam Gebre, former chairman of
Komba Kaffa iddir).
All households of Komba village are members of either the Komba Kaffa
or the Komba Mandjah iddir. Households are represented as members by the
head of the household.8 They meet for general iddir assemblies once a month
on the communal grazing ground located right at the centre of Komba village.
These meetings serve different purposes. They are occasions at which iddir
committees are nominated, elected and dismissed by the iddir members, by
means of voting by a show of hands. There is no legislative period, and only
“when a committee member makes bad things, he will be substituted by another
person” (7/10/2004: Hailemariam Gebre, former chairman of Komba Kaffa
iddir). The cited chairman of the Komba Kaffa iddir was in charge from 1991 to
2004. We have no information on the reasons behind the termination of his
tenure.
General iddir assemblies are not only being held for the appointment of
internal positions but serve the function of ‘plenary sessions’ in the broader
sense. A regular point on the agenda is the payment of the monthly
membership contribution to the iddir treasurer. Both Komba Kaffa and Komba
Mandajah iddir membership fees amount to 0.25 birr per household (about
0.025 EUR) per month, payable in cash. Admission of new members is
another duty undertaken at the general iddir meetings. In case of newly
founded households with kinship linkages from within the iddir, the case
seems rather simple: “when a son [offspring of a household which is iddir
member] gets his own house he directly becomes an iddir member” (30/11/2004:
Komba Kaffa iddir group discussion). For settlers coming from ‘outside’ a
stricter procedure to become iddir members is applied in which applicants
have to present a letter of recommendation from their former iddir.
Thereupon, the iddir committee will decide on the new admission “when he

8 This includes female headed households. In Komba village, there is no pure women iddir as

found in other communities in Ethiopia.

52
 The institutional Sphere of Coffee Forest Management (Stellmacher & Mollinga)

does not have the letter, we will refuse him” (7/10/2004: Hailemariam Gebre,
former chairman of Komba Kaffa iddir). When a larger number of settlers
relocate in the area they can ask for the incorporation of their own sub-iddir
under the existing umbrella one. This happened to be the case with
newcomers from Kambata (see discussion below).
Beyond being a multi-purpose insurance system, the Komba iddir
associations hold and execute significant functions as administrative and
juridical bodies. This includes not only a definite member roster, but a written
statute, to which all household members are bound. For the management of
the nearby forest, the Komba Kaffa iddir statute9 contains two relevant articles
(see box below).

Rules and Regulations of the Komba iddir Concerning Forest Management

1) If any iddir member from the village stands against the rule and regulation of
the forest management, for example, when ordered to keep the forest from any
attacks, first, he will be punished for birr 50, second for birr 75 and lastly he
will be formally accused and referred to the kebele social court.

2) If any iddir member is absent from any [working group] call concerning the
forest management and development, he will be punished first for birr 5,
second for birr 10 and lastly he will be rejected from any share of the forest
product.
Signed by the iddir committee

(translation from Amharic)

This extract of the Komba iddir rules and regulations exemplifies the
formality and distinctiveness of community-initiated institutions—otherwise
often referred to as being ‘informal’—towards management of forest
resources by the population concerned. The iddir is the most significant
community-initiated institution concerning forest use, management and
conservation in the research area. The iddir provides a clear and long standing
organisational structure at village level, with definite membership and—fairly
unknown from literature—written by-laws concerned with the forest
management activities of its members.
Beyond iddir, there is one other community-initiated decision making
structure with decisive impact on the forest resource users’ decisions and
activities in the research area, which is the group of the elderly, locally known
as shimagile.10

9 The Komba Mandjah iddir statute could not be obtained.

10 Shimagile (Amharic) literally means “old man”.

53
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

The Elderly Who do “not Fear Others”

By tradition, village societies in Ethiopia are gerontocratic. Age is
accorded a high socio-political value and particularly older men possessing
greater economical resources enjoy a high social status (Vaughan and
Tronvoll 2003). This is expressed in the respect paid towards the elderly and
the influence of this group that extends beyond households and extended
kinship networks to local communities. Nevertheless, although being a
shimagile often coincides with ‘being older’, their reputation comes less from
their seniority, but more from being a ‘wise’ and ‘rich man’. This tends to
coincide with belonging to an influential family clan11 and a status of an
‘unassailable’ who “does not fear others any more” (4/5/2004: Mesfin Tekle,
Farm Africa/Kaffa Zone Rural Development Desk). Elders are not elected in a
democratic procedure, and they do not represent the interests of the village
community as a whole, rather those of certain peer groups. Women, the
younger generations and persons from less influential family clans are
basically excluded. In Komba village, a total number of eight elders were
identified, four each from the Kaffa and Mandjah ethnicity, and all male.
Formal education, age and religious affiliation were found to be varying
considerably.
The shimagile can be defined to be the lowest legislative and juridical body
beyond mere kinship relations in Komba village. This village internal social
structure holds considerable authority in decision making and enforcement of
traditional norms and rules, and manages the affairs of the villager to a
certain extent. In the use, management and conservation of the coffee forest,
the elders hold a teaching and advising function, in the sense that they pass
on local knowledge to the local resource users and monitor its compliance.
This is indicated in the following two statements of inhabitants of Komba
village: “The shimagile give advice how to manage the trees, not to cut them at a very
early stage” (25/11/2004: Yerango Ambo Gedawo), “the shimagile said we did
not have to disturb the trees. For cardamom they told us to clear every unwanted
material from the plant. They set restrictions to cutting wanza trees [important for
lumber, equipment and furniture making]. They did not pronounce penalties, but
they give us warnings not to do again” (26/11/2004: Abeto Mamo). The primary
role of the elders is to encourage compliance with village internal rules and
regulations, also concerning the management of coffee forest resources, and to
preside as a group over disputes or conflicts that go beyond the scope of
nuclear families and remain outside state courts. They are the “respected people
to negotiate quarrel” (30/11/2004: Komba iddir group discussion). The function
of the elders is embedded in a specific code of conduct in which the iddir can
assign juridical cases to the ‘elders’, as specified by an iddir chairman: “If two

11 Clan is understood as a number of households with social ties bound together by a tight

kinship network.

54
 The institutional Sphere of Coffee Forest Management (Stellmacher & Mollinga)

people have a quarrel, I [the iddir chairman] try myself to solve the problem, if not
possible, I call the iddir committee. The committee tries to solve the problem, but if
they fail then they give the problem to the elders. […] They report to the iddir
chairman what their decision is […]” (7/10/2004: Hailemariam Gebre). This
‘protocol of respect’ was delineated in a group discussion with Komba Kaffa
iddir members as follows: “Firstly the iddir is informed about a problem, then the
elderly come and talk to the iddir head, than the elderly take the individuals in dispute
to other places and then they report to the iddir” (30/11/2004: Komba iddir group
discussion).
In sum, the elderly constitute, together with the iddir, the village-level
administrative and juridical basis to administer, manage and enforce the
community forest property rights.

“The Forest Belongs to Us”: Community-Initiated Forest Use Rights

Empirical research from Komba village and the Bale mountains coffee
forest area provides evidence that beyond the ‘official’ land legislation
initiated by the Ethiopian state, long established arrangements of definite
forest use rights do exist in the coffee forest areas (Zewdie 2003a, Zewdie
2003b). Their roots are definitely found in history. During Imperial times, the
neftegna in Bonga town (see map) owned all land in Bonga forest. His district
officer, the chiquashum, managed his belongings in the area between Weshi
River and Wushwush, including Koma Forest. Within Koma Forest, the
chiquashum in turn entrusted local representatives (gacheukurro in Kaffichio
language) as local ‘village officers’, who assigned use rights for agricultural
and forest land to the smallholders, the gebber, from whom he, in return,
obtained tax payment.
Population of today’s Komba village is composed of former gebber (or
their descendants) from six hamlets (Bushasha, Cangatarra, Gokesha, Kabah,
Kama and Kidah), who held land use rights in Koma forest in the Emperors
time. The former chairman of Komba Kaffa iddir narrates the historical context
as follows: “First, the forest belonged to the chiquashum, but he distributed it
permanently to the farmers. But I got my land from my father. The gacheukurro once
gave it to my father and my father gave it to me. […] The chiquashum gave the order
to give 10 percent of our harvest to him. This was for all crops. When we collected
coffee from the forest, he got 10 percent, also from the honey. […] I had to pay 5 kg
honey per year for the forest use to the chiquashum (for my 70-80 ha). In Haile
Selassi’s time, we used the honey to pay the chiquashum, now we have the honey for
ourselves” (12/11/2004: Hailemariam Gebre).
After the revolution in 1974, the institutional set up fundamentally
changed. Neftegna, chiquashum, and gacheukurro were dispossessed and lost
their privileges and had to begin as ‘normal peasants’. The Yeyebitto Peasant
Association was established embracing a dozen villages with some hundred

55
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

households in total (cp. section 2. above). Also former gebber of the concerned
six hamlets were provided with some acreage of agricultural land for farming.
In the words of the former chairman of Komba iddir: “The derg gave land to all
people that they could live equally. The derg gave some land to me. I got six ha of farm
land only, but there was no reform on forest land” (12/11/2004: Hailemariam
Gebre). This narrative provides an insight into how the regime change of 1974
and the following reform of land use and property rights was perceived and
experienced as a gebber and forest resource appropriator at ‘local level’. The
peasants’ own words illustrate that the dergs’ land reform concerned the re-
distribution of land that was meant for agricultural purpose, but did not
include land of the coffee forests in this area. De jure, the Yeyebitto Peasant
Association took over responsibility for Koma forest, but the state-initiated
institutions concerning its use, management and conservation were not
enforced. Consequently, from the peasants’ perception the de facto property
rights of the coffee forest did not change. “Starting from Haile Selassi time the
forest belonged to us. […] This did not change during the derg” (12/11/2004:
Hailemariam Gebre).
Accordingly, from the mid-1970s onwards, resilient community-based
and newly established state-initiated forest tenure systems drifted apart and
created a mode of legal pluralism. The former gebber continued to execute
‘their’ forest land rights rather than within the scope of the newly established
Peasant Associations or higher level state bodies.
The institutional arrangement was consolidated with use rights
transferred from generation to generation, mainly by patrilineal inheritance,
in the way that “when a father divides his forest, the first son gets the most, the
second and third etc. less. The boys who are engaged in school or in governmental jobs
often get no forest” (5/11/2004: Abetu Mamo). We found no empirical evidence
in the case study area that these use rights are—or have been in the past--
subject to disposal.
But what about the effectiveness of these institutions? If they are not
legitimately recognised by the state but initiated by the communities
themselves, what happens in the case of malfeasance, misfeasance or
nonfeasance, for example, if community members do not appreciate the forest
use rights and pick coffee cherries in plots that ‘belong’ to others.
The effectiveness of institutions is very much linked to the actual
availability of sanction tools which can be applied in case of rule breaking.
For the management of forest resources in Komba village, the community
case law system foresees three kinds of pressure and disciplinary media to
exert its functions, namely: exclusion, fining, and directing the case to the
kebele court, known as fered shengo, for arbitration. The most radical sentence
that the iddir can impose on its members is a membership revocation, hence
the permanent exclusion of a household from the iddir system. In

56
 The institutional Sphere of Coffee Forest Management (Stellmacher & Mollinga)

reminiscence of the fact that all households of Komba village are iddir
members in the way that “without iddir you lack social acceptance” (7/10/2004:
Hailemariam Gebre, former chairman of the Komba Kaffa iddir), it is clear that
execution of this sentence would be tantamount to stigma and social
ostracism, and hence it becomes a strong pressuring instrument. However,
this sanction is ultima ratio and has not been applied in recent history of the
Komba Kaffa iddir. A more frequently applied sanction for behaviour that is
judged to be objectionable is fining. The amount of fines increases with
repeated indictment and was reported to peak at 75 birr. As a third possible
juridical reaction, the cases can be referred from the community-initiated to
the state-initiated law system. This comes about when the former cannot
reach consensus to resolve the case, or on the request of the accused. This
provides evidence that state-, and community-initiated institutional systems
are not necessarily fully independent from each other, but may co-operate at a
certain point. The handover of cases between the community-initiated to the
state-initiated legal systems in rural Ethiopia has also been described by Fule
and Tadesse (1996), although in a somewhat different institutional setting and
course of action.12
The perpetuation of this institutional system has implications that are
positive as well as negative from environmental and social points of view. On
the one hand, the transmission of forest land rights by inheritance implies a
certain commitment to the ‘forefathers’ and strengthens the emotional bond of
the forest ‘owners’ with ‘their’ forest plots. On the other hand, this system is
self-contained with little potential to react to changes from outside, be it
ecological, socio-economic or demographic. The existence and the distribution
pattern of the individually-held forest plots in the coffee forest are part of the
special original inhabitants’ common knowledge and so far in no way
‘formally’ recorded or documented. Most importantly for the management of
resources in Koma forest, the system does not allow the involvement of
people from ‘outside’, as new settlers. No one out of 54 forest plots identified
in Koma forest was reported to be obtained by a new settler household. The
issue of new settlers living right next to Koma forest is briefly described in the
next section.

“I am not from This Area”: The New Settlers and the Community-Initiated
Institutions
Even before the invasion of the Amharic Empire at the end of the 19th
century, the cultural and ethnic composition of the population living together
in the then Kaffa Kingdom was a complex patchwork. Since the violent

12 In their case study village located in the Rift Valley area of central Ethiopia, legal cases are at

first brought forward to the kebele which then hands them over to the elderly (Fule & Tadesse
1996).

57
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

integration of the area into the Ethiopian state, diversity increased greatly due
to big long-distance population movements from the other Ethiopian regions,
often strongly facilitated or even forced by the ruling government. The state’s
involvement has already begun in the Emperors time, but became more
considerable during the derg, with the implementation of nation-wide
resettlement programmes with hundreds of thousands of households being
relocated. As a result of this policy, the present, Kaffa Zone is one of the most
multi-ethnic parts of the country. Komba village and its surroundings are no
exception, with two ethnicities living in the village alone, and communities of
Kambata, Oromo and Amhara ethnicities at only a few minutes walking
distance.
One rationale behind the derg’s resettlement programmes was to allow a
more intensive use of land. Forest land had been identified to be
‘underutilised’ meaning low population density, seemingly low agricultural
output, and hence low economic value—at least according to the official
records of the “Ministry of Agriculture and Rural Development”. In that view,
Kaffa Zone in general and the then Yeyebitto Peasant Association in particular
provided an ideal area of destination, with large tracts of land covered by
relatively undepleted primary forests with comparatively few people around
and inside. In search for adequate resettlement areas and agricultural land for
new settlers from Kambata region13, the then Peasant Association
administration chose an area south of the national road, until then a part of
Koma forest, which was almost totally clear-cut. The ethnic, cultural and
religious background of Kambata people is considerably different from that of
Kaffa and Mandjah people. For example, the Kambata in Yeyebitto kebele
speak the Cushitic Kambata language and are without exception affiliated to
Catholic Christianity, while the locals speak Kaffa language and follow
predominately Orthodox Christianity. Environmentally, the region of the
Kambata origin can be described as savannah lowland, without any tracts of
primary forest.
With the successive arrival of Kambata settlers, the size of the population
living adjacent to Koma forest increased, involving a growing demand for
land for dwellings, infrastructure and grazing, as well as forest resources like
firewood and timber. Beyond population increase, the new settlers brought
along a different institutional and knowledge background, not adjusted to the
specific ecological and management conditions of the coffee forest
environment.

13 Kambata (also known as Kambatta or Kembata) is located in Central Ethiopia and is

administratively part of Sidama zone in the most eastern tip of the “Southern Nations and
Peoples’ Regional State” (SNNPRS). The region is characterised as one of the most densely
populated and impoverished regions of Ethiopia.

58
 The institutional Sphere of Coffee Forest Management (Stellmacher & Mollinga)

As already mentioned above, the new settlers’ households could not

obtain traditional forest use rights in Koma forest. Nevertheless, they are
knowledgeable of the existence of the forest land use rights held by the local
people. There is evidence that new settlers are aware of their own ‘secondary’
status by acknowledging the old settlers ‘primary’ use rights at least
regarding the use of forest coffee. This is reflected inter alia in the answer of a
new settler from the Amhara dominated community Bita Chega who moved
from Wollo region in Northern Ethiopia to some four kilometres north-west
of Komba village in the year 1985 in the course of the dergs’ resettlement
programmes. To the question whether people from Bita Chega village do
utilise forest coffee from the nearby Komba forest he responded: “We do not
have own coffee in the forest. The forest is occupied by other people” (26/6/2003:
Seyoum Tefera). These words match the statement of an inhabitant from
Washi village, located about four kilometres to the west of Komba village:
“Around Mulah (Bita Selam) and Agama village there is coffee in the forest. Only the
people in these villages get the coffee in the forest” (25/6/2003: Tadesse Lola).
The diverse historical and ethnic background of the new settlers raises the
question not only of whether they are acquainted with the fact that Koma
forest is fragmented into forest plots individually ‘owned’ by the original
population, but also of whether they accept the rules and regulations that are
attached to this forest ‘ownership’, and act accordingly. In order words, do
new settlers in a rights system created and maintained by the original
community, in which they themselves do not have any chance to acquire
rights over a forest plot or resource, actually respect this system?
The issue of “disobedience” with respect to the community-initiated forest
related institutions is a delicate one, and took much effort and time to
investigate. A major question is what happens in the case of non-compliance
by people that are not living in the sphere of jurisdiction of the village iddir or
the elderly? Do ‘rule violators’ need to fear any penalty when they use the
forest resources according to their own understanding? How are the
institutionalised authorities involved in the resolution of conflicts between the
different groups of forest resource appropriators? To elaborate on these
issues, two different ‘truths’, the one of new settlers and the one of the
original people are presented in the form of interviewees’ statements.
First we discuss the perception and experience of new settlers. The
following statement was issued by a peasant who settled in Komba village on
his own instigation in the year 2001, thus not within the scope of the dergs’
resettlement programmes. Although he is of Kaffa ethnicity and—compared
to the Kambata people—ethnically a ‘native’, he (respectively, his household)
does not posses use rights in Koma forest, which begins some twenty meters
behind his tukul house. When he was initially asked whether he ever collects
coffee from the forest he responded: “I do not go into the forest to collect coffee

59
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

because I am not from this area” (14/9/2003: #32). After many months of field
research when more confidence had been established between the researcher
and the interview partners, the question concerning the utilisation of forest
resources from the nearby forest (not only coffee) was answered more
specifically, partly contradicting the previous statement: “I get firewood from
everywhere within Koma forest, also the coffee and spices. I do not have to ask for
permission. Also for house construction materials I simply go into the forest. The
owner does not complain. Also the forest owner steals coffee somewhere else. Even the
hanging of beehives is sometimes possible without permission, otherwise the owner is
very serious. The serious people kick the thieves and take all the products from the
thieves. But the forest is not controllable, that is why I do not want to have my own
forest” (28/11/2004: #32). Other settlers living adjacent to Koma forest
commented in a similar way: “Anybody can pick coffee, it is impossible for the
owner to prevent stealing. But the owner does not try to prevent. Sometimes they go
into the forest, and if they see a thief they take the products from him and give him a
warning. It is possible to use both firewood and house construction materials. But
timber for selling was forbidden. Timber thieves would have been brought by the
owner to the kebele and accused. […] There are demarcations, but I do not know
where the plots are” (28/11/2004: Mekuria Wolde).
Nevertheless, the issue of validity and effective reach and enforcement of
the community-initiated forest use rights, and the ‘stealing’ of forest resources
respectively, look different from an original settlers’ point of view, hence from
the perspective of the forest plot ‘owners’. A peasant of the Mandjah ethnicity
who grew up in Gokesha got told the ‘stealing narrative’ from a rather
contrasting angle: “Others are not allowed to take coffee, to take firewood, to hang
beehives, spices, fruits, medicinal plants. Four days a week (Tuesday, Thursday,
Saturday, Sunday), I keep my forest from thieves by patrolling. When I meet the thief,
I will catch him. He begs me not to bring somewhere, then I allow him to go, but he
has to give [the forest product] what he gathered. After the third time I will bring
him to the kebele. The kebele will punish him. They put him into prison for two to
three days and give him advice not to do it again. I did not allow other people to take
materials, only house construction materials, or for ploughing material, I give
permission to any people of my kebele. If people from different kebeles were stealing, I
bring them to my kebele, and they will write a letter to the other kebele”
(26/11/2004: Abeto Mamo).
These statements illustrate that with a larger heterogeneity and
fragmentation of forest resource users, the strict adherence to the existing
community-initiated institutional framework is relaxed. This is basically due
to the fact that the people that are “not from this area” are less involved in,
bound to or knowledgeable about the community-initiated institutional
setting regarding forest resource use, management and conservation than the
original inhabitants “from this area”. The new settlers are aware of the
existence of certain rules and regulations initiated by the original

60
 The institutional Sphere of Coffee Forest Management (Stellmacher & Mollinga)

communities concerning use of the forest resources, but perceive them to be

not binding personally. This situation is conflictive as households of all
communities, despite their historic or ethnic background, depend on
resources from Koma forest for their very livelihoods.

“The Kebele Knows”: Between Local ‘Tradition’ and State Policies

The kebele administration provides the state-initiated institutional
framework through which governmental policies ought to be implemented
‘on the ground’ (see section 2). For use and management of forest resources,
the kebele is vested with the authority to enforce state-devised operational
institutions, hence rules and regulations meant to guide activities of forest
resource appropriators in its area of responsibility. But how does the kebele
execute this mandate in practice? How does it deal with the local level legal
pluralism that developed in the coffee forest areas? To elaborate on these
questions, interviews were conducted with the kebele chairman and vice
chairman. The first issue concerns the kebele chairman’s perception whether
the kebele administration is in practice capable to regulate forest use and
management activities in its area of authority and to prevent Koma forest
from undesirable human interference such as wood cutting: “The kebele cannot
catch the people, they hide themselves. […] From 1986-89 [1992/93-1995/96
Gregorian calendar], I was vice kebele chairman, from 1994 [2000/01 Gregor. cal.]
up to now main chairman. […] We never caught anybody, because the forest is very
vast. When we wanted to catch them, they throw everything down and run away”
(29/11/2004: Mengisto Mamo, kebele chairman). According to the kebele
chairman, ever since the early 1990s, not a single person has been accused of
disobedience of forest conservation state laws in Yeyebitto kebele. This
statement reflects how the responsibility of the state exceeds its practical
capability to control it at local level. The kebele administration is de jure
responsible for maintenance of state ‘law and order’ but has no de facto
implementation capability to enforce the state forest resource conservation
policies.
But how do the kebele representatives perceive the community-initiated
forest use rights within its own area of responsibility, and how does the kebele
administration deal with this issue? Is there - as is evident through the words
of the above cited Mandjah peasant - juridical interaction or even cooperation
between the two institutional ‘worlds’? When the chairman of the Yeyebitto
kebele was asked whether he is informed about the subdivision of Komba
forest into use right plots he stated: “The kebele knows the traditional forest
division, but the traditional users do not pay forest use taxes.” (29/11/2004:
Mengisto Mamo). The Yeyebitto kebele vice-chairman also stated that “even if
the [forest] land officially belonged to the state, it unofficially belongs to the people.
The kebele knows and accepts it” (24/11/2004: Abebe Abafogi). The threshold

61
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

between toleration and collaboration in the way that community-initiated

forest use rights can institute legal proceedings at the kebele court, the fered
shengo, could not be fully ascertained in the interviews conducted. The kebele
chairman expressed the circumstances in the statement: “The traditional
demarcation is not documented, and there are no written documents. But the fered
shengo acknowledges the traditional rights, if the forest owner comes with tangible
evidence” (29/11/2004: Mengisto Mamo). This, however, seems to be more the
theoretical case than a practical reality: “Since I am the leader of the kebele nobody
came to me with traditional forest issues. […] People say ‘my honey, my coffee etc. is
stolen’, but they do not bring the thieves” (29/11/2004: Mengisto Mamo). These
statements make clear that the lowest state representative in the case study
area is a) knowledgeable about the existence of community-initiated
institutions, b) does not disapprove or take action against them, and c)
acknowledges them in the way that malfeasance against the community-
initiated institutions can principally be prosecuted in the legal court of justice.
The individuals working in the kebele administration have themselves
grown up in this area, have manifold kinship and peer relationships on site;
moreover, their own households’ livelihood also critically depends on the
extraction of resources from Koma forest. To put it into the provocative words
of an NGO representative working in Bonga area: “The use of forest products is
allowed by the kebele because also kebele leaders use them” (4/5/2004: Mesfin
Tekle).
When combining the forest resource appropriators’ statements with the
kebele representatives’ ones, it becomes evident that there is no stringent
division between community-initiated and state-initiated institutional
‘worlds’ with regard to their cognisance of forest resource use and
management activities, and that the institutional systems complement each
other. The kebele acts as the “the linkage between informality and government”
(4/5/2004: Mesfin Tekle) and leaves the basic responsibility for forest use and
management issues with the community-initiated units, elders and iddir, that
get their authorisation primarily from historically developed social standards
and norms. Forest-related rule and regulation compliance is ensured by
mechanisms of social sanctioning which are powerful in a relatively cohesive
society with common norms and values. Similar findings are presented by
Alula Pankhurst who did research on irrigation in South Wello in Central
Ethiopia. He writes that “disputes over irrigation tend to be solved by elders, with
the threat of sanctions by burial associations [being iddir] in recent years. Only if the
matter cannot be resolved informally, […] the case be taken to government
institutions” (Pankhurst 2002).

62
 The institutional Sphere of Coffee Forest Management (Stellmacher & Mollinga)

Conclusions
In this article, the local level institutional sphere of forest resource use
and management is illustrated on the basis of data and insights gathered by
means of empirical research in the coffee forests of Kaffa Zone, South-western
Ethiopia. Evidence shows that there are historically-evolved and community-
initiated institutional structures that regulate the use, management and
conservation of forest resources. By combining concerted executive and
juridical functions at village level, the institutions partly fill in for governance
failure of the Ethiopian state. However, the research findings also indicate
that the community-initiated institutional system rests on social sanctioning
and consensual conventions limited to groups of original settlers only, rather
than on tangible and universally accepted agreements among all people living
in a particular forest area and using its forest resources. This critically limits
the practical assertiveness of the community-initiated institutional sphere,
and resulted in a local institutional vacuum in which both state-initiated and
community-initiated forest use, management and conservation rules and
regulations are violated by local people.
What can be concluded when we reminisce about the ongoing and rapid
depletion and loss of the last montane moist forests of Ethiopia? The article
shows a situation of legal pluralism in which local people have no
contractually warranted and effectively enforceable property use rights for
the forests they live in and on which resources their livelihoods critically
depend. Neither the state nor the community-initiated bodies alone are in a
position to govern the forests and their ecological, socio-economic and
cultural functions and services. This unclear and uncertain situation offers
both, original people and new settlers, little means and incentives to apply
future oriented sustainable use and management practices, and hence
promotes the depletion and loss of the Ethiopian montane coffee forests.
In the last few years, environmentally and socially concerned
development organisations recognized the importance of a more sustainable
use and conservation of the last Ethiopian primary forests. Consequently, an
increasing number of externally-funded ‘Forest Management’ projects were
established, opting for different approaches. However, they all tend to aim at
institutional change by the creation of forest users groups or societies with
determined rules and regulations, hence yet another institutional ‘world’.
Instead, already existing structures should be ‘taken on board’ and bolstered.
An expedient approach is to provide local level community-initiated bodies,
particularly the iddir, with better capacities to act and decide on forest use,
management and conservation. Formal recognition and legal backup of these
institutions by the state would be a necessary first step into this direction.

63
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

References
Abbot, P., Hailu, A. et al. 2000. “Community Organisation for Natural
Resource Management: the Potential for Community-based Wetlands
Management Planning.” Ethiopian Wetlands Research Programme: 34.
Abebaw, D. and Virchow, D. 2003. “The Microeconomics of Household
Collection of Wild Coffee in Ethiopia: Some Policy Implications for In-situ
Conservation of Coffea arabica Genetic Diversity.” Workshop Economy
Analysis of Policies for Biodiversity Conservation, Venice, BIOECON.
Agrawal, A. 1995. "Population Pressure = Forest Degradation: An
Oversimplistic Equation?" Unasylva 181(46): 50-58.
Agrawal, A. and Yadama, N. 1997. "How do local institutions mediate
markets and population pressure on resources? Forest panchayast in
Kumaon, India." Development and Change 28: 435-465.
Aredo, D. 1993. “The Informal and Semi-Formal Financial Sectors in Ethiopia:
A Study of the Iqqub, Iddir, and Savings and Credit Co-operatives.”
African Economic Research Consortium. Nairobi. 21.
Bodin, Ö., Tengö, M. et al. 2005. The value of small size: loss of forest patches and
threshold effects on ecosystem services in southern Madagascar. Stockholm:
Stockholm University: 38.
Boum, D. 2002. “Farming in Tsetse controlled Areas of Eastern Africa.
Ethiopia National Component.” Addis Ababa., Farming systems and
Natural Resource Management. Project 7. ACP. ET 086. .
Bromley, D. W. 1986. “Closing Comments at the Conference on Common
Property Resource Management.” Conference on Common Property
Resource Management, National Research Council. Washington, D.C.:
National Academy Press.
Carswell, G., de Haan, A. et al. 1999. Sustainable Livelihoods in Southern Ethiopia.
Brighton: Institute of Development Studies.
EEA/EEPRI. 2002. Land Tenure and Agricultural Development in Ethiopia. Addis
Ababa, Ethiopian Economic Association/Ethiopian Economic Policy
Research Institute.
Ersado, M. 2001. “Inventory of Woody Species in Bonga Forest.” Technical
Report No. 1. Addis Ababa, Institute of Biodiversity Conservation and
Research.
FDRE. 1997. “Federal Rural Land Administration Proclamation No. 89/1997.”
Federal Democratic Republic of Ethiopia. Addis Ababa.
Fule and Tadesse. 1996. Turufe Kechema. Ethiopian Village Studies. P. Bevan and
A. Pankhurst. Addis Ababa, Oxford, Department of Sociology, Addis
Ababa University, Center for the Study of African Economies, Oxford.
GTZ. 2004. Natural Resources and Governance: Incentives for Sustainable Resource
Use. Eschborn, Environment and Infrastructure Division, Gesellschaft für
Technische Zusammenarbeit: 64.

64
 The institutional Sphere of Coffee Forest Management (Stellmacher & Mollinga)

Immergut, E. M. 1998. “The Theoretical Core of the New Institutionalism.”

Politics and Society 26(1): 5-34.
Keeley, J., and Scoones, I. 2000. “Knowledge, Power and Politics: the
Environmental Policy Making in Ethiopia.” The Journal of Modern African
Studies 28(1).
Kiser, L. L. and Ostrom, E. 1982. “Three Worlds of Action: A Metatheoretical
Synthesis of Institutional Approaches.” Ostrom, E. Strategies of Political
Inquiry. Berverly Hills: Cage, pp. 179-222.
Koontz, T. M. 2003. An Introduction to the Institutional Analysis and Development
(IAD) Framework for Forest Management Research. First Nations and
Sustainable Forestry: Institutional Conditions for Success. Vancouver:
University of British Columbia.
Marena Project (n.d.). Characteristics of traditional forest management. Marena
Briefings (Ethiopia). Brighton, Forum for Social Studies, Centro de
Experimentação Florestal, University of Sussex. ET 10.
Mayne, R. and Tola, A. 2002. “Crisis in the Birthplace of Coffee.” Oxfam
International Research Paper: 1-12.
McCarthy, S. 2001. “The History of Agricultural Cooperative Development in
Ethiopia.” Cooperative Business Today. Addis Ababa, Voca-Ethiopia. 1: 5.
Nørgaard, A. S. 1996. “Resdiscovering Reasonable Rationality in Institutional
Analysis.” European Journal of Political Research 29: 31-57.
North, D. 1990. Institutions, Institutional Change and Economic Performance.
Cambridge: Cambridge University Press.
Ostrom, E. 1990. Governing the Commons: The Evolution of Institutions for
Collective Action. New York: Cambridge University Press.
Ostrom, E. 2004. “Applying the IAD Framework.” Workshop in Political
Theory and Policy Analysis, Indiana University.
Ostrom, E., Burger, J., et al. (1999). “Revisiting the Commons: Local Lesson,
Global Challenges.” Science 284: 278-282.
Ostrom, E., Gardener, R. et al. 1994. Rules, Games, and Common-Pool Resources.
Ann Arbor: The University of Michigan Press.
Pankhurst, A. 2002. “The Influence of the State and Market on Local Level
Management of Natural Resources: Case Studies of Forests, Irrigation and
Pasture Sites in South Wello, Ethiopia.” BASIS CRSP Management Entity.
Department of Agricultural and Applied Economics, University of
Wisconsin-Madison.
Pausewang, S. 1990. The Peasant Perspective. Ethiopia: Rural Development
Options. S. Pausewang, F. Cheru, S. Brüne and E. Chole. London and New
Jersey: Zed Books Ltd., pp. 213-226.
Poteete, A. and Ostrom, E. 2002. An Institutional Approach to the Study of Forest
Resources. IFRI, Indiana University.

65
International Journal of Social Forestry, Volume 2, Number 1, June 2009: 43-66

Stellmacher, T. 2007a. “The Historical Development of Local Forest

Governance in Ethiopia. From the Imperial Times to the Military Regime
of the Derg.” Afrika Spektrum 3(42).
Stellmacher, T. 2007b. Governing the Ethiopian Coffee Forests: A Local Level
Institutional Analysis in Kaffa and Bale Mountains. Aachen: Shaker
Publishing.
Tache, B. and Irwin, B. 2003. “Traditional Institutions, Multiple Stakeholders
and Modern Perspectives in Common Property. Accompanying change
within Borana pastoral systems.”Securing the Commons 4.
Teketay, D. 1999. “History, Botany and Ecological Requirements of Coffee.”
WALIA, Journal of the Ethiopian Wildlife and Natural History Society 20: 28-
50.
Vaughan, S. and Tronvoll, K. 2003. “The Culture of Power in Contemporary
Ethiopian Political Life.” Sida studies 10: 184.
Watson, E. 2001. Inter Institutional Alliances and Conflicts in Natural Resource
Management. Preliminary Research Findings from Borana Oromia Region,
Ethiopia. Cambridge, Marena Research Project.
Zewdie, Y. 2003a. “Forest Access: Policy and Reality in Kafa, Ethiopia.” Leisa
Magazine 19(3): 21-22.
Zewdie, Y. 2003b. “Not by Maize Alone: Forest Access and Rural Livelihoods
in Southwest Ethiopia.” ETFRN News 39/40.

66