Front Cover
 Title Page
 About the authors
 Table of Contents
 Author's acknowledgments
 Description of enset and syste...
 Ecology and environment
 Uses of enset
 Agronomy and production manage...
 Harvesting and processing
 Gender issues
 Ensent farming systems: Three case...
 Food security and sustainabili...
 Future prospects
 Back Cover

Group Title: The "tree against hunger" : enset-based agricultural system in Ethiopia
Title: The "tree against hunger"
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00053942/00001
 Material Information
Title: The "tree against hunger" enset-based agricultural system in Ethiopia
Physical Description: v, 56 p. : ill., maps ; 28 cm.
Language: English
Creator: Brandt, Steven A
Conference: Seminar on Enset in the Food Security of the Horn, (1997
Publisher: American Association for the Advancement of Science
Place of Publication: Washington D.C
Publication Date: c1997
Subject: Ensete -- Congresses -- Ethiopia   ( lcsh )
Ensete -- Utilization -- Congresses -- Ethiopia   ( lcsh )
Genre: bibliography   ( marcgt )
conference publication   ( marcgt )
non-fiction   ( marcgt )
Bibliography: Includes bibliographical references (p. 53-56).
Statement of Responsibility: Steven A. Brandt ... et al..
Funding: Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
 Record Information
Bibliographic ID: UF00053942
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 003093113
oclc - 41892827
lccn - 99210714

Table of Contents
    Front Cover
    Title Page
        Page i
    About the authors
        Page ii
    Table of Contents
        Page iii
    Author's acknowledgments
        Page iv
        Page v
        Page 1
        Page 2
    Description of enset and systems
        Page 3
        Page 4
    Ecology and environment
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
    Uses of enset
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
    Agronomy and production management
        Page 19
        Page 20
        Page 21
        Page 22
    Harvesting and processing
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
    Gender issues
        Page 29
        Page 30
    Ensent farming systems: Three case studies
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
    Food security and sustainability
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
    Future prospects
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
    Back Cover
Full Text



. ._i..- wa






Enset-Based Agricultural Systems in Ethiopia

Steven A. Brandt, Anita Spring, Clifton Hiebsch,
J. Terrence McCabe, Endale Tabogie, Mulugeta Diro, Gizachew Wolde-Michael,
Gebre Yntiso, Masayoshi Shigeta, and Shiferaw Tesfaye
American Association for the Advancement of Science
Awassa Agricultural Research Center
Kyoto University Center for African Area Studies
University of Florida



Steven A. Brandt (Department of Anthropology), Anita Spring (Department of
Anthropology), and Clifton Hiebsch (Department of Agronomy) are faculty members at the
University of Florida. Gebre Yntiso is a graduate student in the Department of Anthropology.

Endale Tabogie, Mulugeta Diro, Gizachew Wolde-Michael, and Shiferaw Tesfaye
are researchers at the Awassa Agricultural Research Center in Awassa, Ethiopia.

J. Terrence McCabe is a faculty member at the Department of Anthropology and the Institute
for Behavioral Science, University of Colorado.

Masayoshi Shigeta is based at the Center for African Area Studies, Kyoto University, Japan.

1997 American Association for the Advancement of Science



Foreword ........


. . . . . . . . . ..... iv

. . . . . . . . . . V

Introduction .......................

Description of Enset and Systems ......

Ecology and Environment ............

History ...........................

Uses of Enset ......................

Agronomy and Production Management

Harvesting and Processing ...........

Livestock .........................

Gender Issues .....................

... ...... .. ...... ....1

................ .... .3




.................... 19


................... .25


Enset Farming Systems: Three Case Studies .................. 31

Food Security and Sustainability ...........................41

Future Prospects ....................................... 49

References ............................................53



Authors' Acknowledgments

Many people have contributed directly and indirectly to this publication, which has its origins m
the "Seminar on Enset in the Food Security of the Horn," held April 17, 1997 at the American
Association for the Advancement Of Science (AAAS) in Washington, D.C. We are particularly grateful
to Ato Lemma Mitiku, director, and Ato Adebecho Wachiso, former director, of the Southern
Nations, Nationalities, and Peoples Regional Government (SNNPRG) Bureau of Agriculture; Ato
Kelsa Kena, manager, and Ato Elias Urage, former manager, of the SNNPRG Agricultural Research
Center for their enthusiastic support of enset research. We would also like to thank Deutche
Gesellschaft fur Technische Zummenarbeit (GTZ) for its continued financial support of the Enset
Needs and Assessment Project (ENAP), and in particular Dr. Thomas Labahn, director of
GTZ/Ethiopia and W/t Rahel Teferi of GTZ/Ethiopia's project support services office. We also thank
the University of Florida Center for African Studies and Office of Research, Technology and Graduate
Education for financial support.
Thanks are also due to Ato Bekalu Molla and W/t Asnaketch Tensaye, consultants during Phase 1
of ENAP, and Dr. Zinabu Gebre-Mariam, W/t Yewelsew Abeba, and Ato Admasu Tsegaye of Awassa
College of Agriculture, Ato Assefa Amaldegen, Ato Ousman Surur, W/t Seble Shimles, Ato Tariku
Menjeye, Ato Tesfaye Habte and Ato Tesfaye Tadesse of BAR, Ato Esetu Anito and Bizuayehu Haile of
ARC, and W/o Meselech Melke of the SNNPRG Bureau of Culture and Information for their help in
completing Phase 2 of ENAP We also express our thanks to Ms. Melanie Brandt for map illustrations,
and to Sung A. Lee of the University of Colorado Department of Anthropology and Ethiopia's
Institute of Agricultural Research for advice.
We would also like to thank the AAAS Sub-Saharan Africa Program, and its director, Dr. Peter R.
Schmidt, for initiating and organizing the Seminar, as well as coordinating all aspects of this publica-
tion. Our thanks also go to John Schoneboom, who has provided invaluable editorial assistance as
part of the AAAS responsibilities.
Finally, we wish to thank the farmers and their families, as well as the administrative officials in
our study areas, for their willingness to share with us knowledge of this unique plant.


Africa faces over the next decade an ever-increasing
need to achieve sustainability in agricultural pro-
duction. In the Horn of Africa one of the primary
obstacles to sustainability has been the threat of famine. Yet,
as this important booklet shows, there are parts of Ethiopia
that survived famine during the 1980s because of the uti-
lization of enset as part of the subsistence system. Also
known as the false banana, enset is very likely the most
unstudied domesticated crop in Africa. It helps to feed
approximately ten million people in Ethiopia and Eritrea (in
restricted pockets). Given that it figures so importantly in
the diet of contemporary Ethiopians and that it has acted as
a famine buffer, why has it been so neglected? The answers
are complex. They are partly related to cultural perceptions,
politics, and history This short monograph unwraps much
of the mystery surrounding enset. It explores its history, not-
ing that enset was once much more widespread in Ethiopia.
It also explores its food characteristics and the different
agro-economic conditions under which it is grown as an
important part of the diet.
Because the development agendas of Western aid agen-
cies still focus on cereal grains, particularly maize, enset
continues to be ignored. Even though the Ethiopian govern-
ment has recently elevated enset to the status of a national
crop, it is not clear that this move will propel critical
research that is needed to realize its agricultural potential in
other regions of Ethiopia. Nor does it ensure that enset
remains an integral part of the subsistence systems where it
is already being cultivated successfully
Over the last seven years, I have closely followed the
progress of research on enset. For the most part this
research, in addition to work undertaken by Ethiopians, has
been initiated and carried out by colleagues in the
Department of Anthropology at the University of Florida.
While initially anthropological in focus, the research has
since expanded to include agronomy, soil science, econom-
ics, history, and other ancillary sciences. It has also incorpo-
rated collaborating scientists from the University of Addis
Ababa, Kyoto University, the University of Colorado, and the
Awassa Agricultural Research Center in Ethiopia. In this
sense it is an excellent model of interdisciplinary collabora-
tion and interaction.
In the Sub-Saharan Africa Program of the American

Association for the Advancement of Science (AAAS), we felt
there was a need to disseminate the results of preliminary
research conducted by the enset research team. Their pre-
liminary research results have important implications for
future agricultural development in Ethiopia, as well as for
the sustainability of existing enset systems. Also, it is clear
that policymakers and agricultural specialists do not under-
stand the plant, its economic potential, its cultural limita-
tions, its famine-buffer potential, and its threatened sustain-
To overcome these deficiencies in general knowledge, we
organized a full-day symposium in April 1997 under AAAS
sponsorship at its Washington headquarters. Experts from
institutions that invest in agricultural development in Africa
and pertinent Ethiopian scientists were invited to participate
in the symposium. Representatives from JICA, USAID, the
World Bank, and Catholic Relief Services provided critical
reactions to the papers presented by Brandt, Shigeta, Yntiso,
McCabe, and Hiebsch. The discourse was lively and point-
ed. Discussion focused on a number of key issues, such as
the cultural stigmas attached to enset food products, the
dangers of assuming wholesale transferability of enset agri-
culture to other regions in Ethiopia, and problems being dri-
ven by increasing populations and shrinking farmland in
some enset producing areas.
We at AAAS are delighted with this product from the
symposium. It opens the door for further exploration into
one of Africa's unknown food resources in a format that is
clear and easy to understand. It is set up in such a way that
key questions about enset are answered concisely and with-
out technical mystification. We also refer the reader to the
AAAS Sub-Saharan Africa Program web site
(http://www.aaas.org/international/ssa/ssa.htm), where this
publication will be found with each question hypertext-
linked to its answer.

Peter Schmidt
Sub-Saharan Africa Program



ince Ethiopia's tragic drought and famine-prone
decades of the 1970s and 1980s, researchers and poli-
cymakers have been particularly concerned with find-
ing long-term, sustainable solutions to Ethiopia's food secu-
rity needs. The majority of extension, development, and
research on Ethiopian agriculture has focused upon the
cereal-based systems of the highlands of northern, central,
and eastern Ethiopia, and to a lesser extent upon the shift-
ing cultivation economies of subtropical and lowland west-
ern Ethiopia. There has been considerably less research on
Ethiopia's other major agricultural complex, the enset agri-
cultural system of the highlands of southern Ethiopia.

Enset (Ensete ventricosum) is the main crop of a sustain-
able indigenous African system that ensures food security in
a country that is food deficient. Enset is related to and
resembles the banana plant (Plate 1, page 12) and is pro-
duced primarily for the large quantity of carbohydrate-rich
food found in a false stem (pseudostem) and an under-
ground bulb (corm). More than 20 percent of Ethiopia's
population (more than 10 million people the precise
number of enset users is unknown), concentrated in the
highlands of southern Ethiopia (Figure 1.1), depend upon
enset for human food, fiber, animal forage, construction
materials, and medicines.

1.1 o n u c
shwn th ditibto of doetiae S SO .



.Smlen- -
I 'Lalibela
LaOe' ._*



I ',



-- -



General Area of Enset Cultivation r-'

0 100 200 300 400 500 600 Kilometeis










Enset agriculture has received surprisingly little exten-
sion, development, or research attention, perhaps because:
1) the majority of enset farmers live in one of the least
developed regions of Ethiopia, making access and logistics
difficult; 2) the system is unique when compared to cereal
farming; 3) production processes are complex; and 4) there
is the perception that it is eminently successful, sustainable,
and trouble-free.
Since the 1950s Ethiopian and international scientists
have carried out enset research, but much of this work was
undertaken by isolated researchers and was often focused by
discipline on specific topics and poorly funded. In the early
1990s multidisciplinary, multinational teams of agronomists
and social scientists conducted pilot studies and held dis-
cussions with several Ethiopian institutions and individuals
in order to determine: 1) whether a more detailed under-
standing of enset agriculture could contribute to Ethiopia's
present and future food security needs, and, if so, how; 2)
what the current status of enset extension and research was;
and 3) the potential for collaborative investigations.
The general conclusion was that an integrated and com-
prehensive study of the biological, agricultural, ecological,
social, and economic components that make up enset-based
agricultural systems was greatly needed if Ethiopia was to:
1) increase production and distribution of enset products,
not only within rural southern Ethiopia, but for urban mar-
kets; 2) transfer enset-based agricultural systems, or parts
thereof, to other, non-enset growing regions of highland
Ethiopia; and 3) determine if the sustainability of enset agri-
culture was under threat in the short or long term.
In order to initiate the development of such integrated
and comprehensive multidisciplinary projects, the
International Workshop on Enset was held in December
1993 in Addis Ababa, under the auspices of the Ethiopian
Institute of Agricultural Research and the University of
Florida. With over 60 participants and 32 papers presented,
the purpose of the workshop was to: 1) bring together for
the first time Ethiopian and other researchers from interna-
tional, national, and nongovernmental organizations
involved or interested in enset agriculture; 2) determine the

current state of knowledge on enset; 3) increase the
Ethiopian and international public's awareness of the impor-
tance of enset-based agriculture in Ethiopia; 4) identify
future avenues of enset investigation; and 5) devise a long-
term, interdisciplinary plan for extension, development, and
research on enset-based farming systems (Abate et al, 1996).
Current research efforts in Ethiopia are largely relegated to
the Southern Nations, Nationalities, and Peoples Regional
Government (SNNPRG), where the Awassa Research Center
and Areka Research Station are conducting various enset stud-
ies with a team of eight to ten scientists. Several researchers at
Awassa College of Agriculture and Addis Ababa University
also have enset-related projects. However, no specific techni-
cal packages are currently being promoted to farmers. In
1995 the Enset Needs Assessment Project was initiated as a
direct outgrowth of the International Workshop. Funded by
Deutche Gesellschaft fur Technische Zummenarbeit (GTZ),
Germany, and coordinated by the SNNPRG Bureau of
Agriculture in collaboration with Ethiopian and foreign insti-
tutions, the goal of the project is to provide baseline data for
extension, development, research, and policy agendas for
national and international institutes, individual researchers.
national policymakers, and donor agencies. The first phase.
comprising a literature review, rapid rural appraisal, and
informal surveys of three major ethnic groups that use enset
as a staple or co-staple, has been completed. Phase Two, the
design and collection of household, yield, market, and pro-
cessing questionnaires, and the collection of additional data
from other enset-based ethnic groups, is in the process of
analysis by many of the authors of this publication.
International donors have in general been reluctant to
commit funds for enset research. However, in July 1997 the
Ethiopian Ministry of Agriculture designated enset as a
national commodity, which may pave the way for changes in
research and extension programs. As these programs are for-
mulated, more complete information on enset systems will
be required. It is in this spirit that this booklet has been put
together. The objectives here are to: 1) bring together and
put into focus what we know and what still needs to be
researched in order to document a sustainable system; 2)
map out future research agendas for national and interna-
tional scholars; and 3) provide information to government
policymakers and donors for potential interventions to assist
enset producers. This publication has been prepared in the
form of questions and answers, to provide an accessible
approach to the subject, and to elucidate what we know and
what needs to be known for future work and interventions.



I What does enset look like?
Enset looks like a large, thick, single-stemmed banana
plant (Plate 2, page 12). Both enset and banana have an
underground corm, a bundle of leaf sheaths that form the
pseudostem, and large leaves (Figure 2.1). Enset, however,
is usually larger than banana, with the largest plants up to
10 meters tall and with a pseudostem up to one meter in
diameter. The leaves are more erect than those of a banana
plant, have the shape of a lance head, and may be five
meters long and nearly one meter wide. Banana plants nor-
mally form suckers or clusters of plants at the base, but
enset does not.
The stem has three parts. The upper-most portion is the
pseudostem, which is made of a system of tightly clasping
leaf bases or leaf sheaths. The pseudostem may be two to
three meters tall and contains an edible pulp and quality
fiber. The underground corm is really an enlarged lower por

tion of the stem. It may be up to 0.7 meters in length and in
diameter. A short section of stem near the soil line, between
the pseudostem and corm, is the true botanical stem. Leaves
and the single flower head initiate from the true stem at its
center, grow up through the middle of the pseudostem, and
emerge at the whorl in the middle of the leaf bases. Enset has
a fibrous rooting system that grows out from the corm.
At maturity, a single flower head emerges, which forms
multiple flowers, fruit, and seeds. The entire head, which may
be nearly one meter in length, hangs downward from a stalk
in the center of the plant. Many of the small, banana-like
fruits (enset is sometimes called false banana) on each flower
head produce several irregularly shaped black seeds, each
about one centimeter across. Most wild and a few cultivated
plants are produced from seed, and have more than one par-
ent. Most domesticated plants, however, are propagated from
suckers, and are clones of their one parent. Most plants are
harvested before or at early stages of flower formation.

2.1 Parts of te

I What is the botanical classification of
SI enset and how is it distributed?
Sr Enset belongs to the order Scitamineae, the family
Musaceae, and the genus Ensete. Banana is in the same family
Sas enset, but in the genus Musa. Although further research
still needs to be done on the taxonomy and distribution of
enset species, current data reveal two wild enset species dis-
li- tribute over much of Asia, and four wild species in sub-
Saharan Africa and Madagascar (Baker and Simmonds,
1953; Simmonds, 1958). Ensete ventricosum, the only known
INFLORESCENCE wild species in Ethiopia, is concentrated in the southern
highlands, but also grows in the central and northern high-
lands around Lake Tana, the Simien Mountains, and as far
\F SHEATH north as Adigrat and into southern Eritrea (Simoons, 1960
and 1965; and observation by the authors).
In spite of the extensive distribution of wild enset, it is only
S in Ethiopia that the plant has been domesticated. Wild enset
propagates naturally by seed, and is restricted in Ethiopia to
elevations of approximately 1,200 to 1,600 meters above sea
level. However, farmers almost always propagate domesticated
"' enset vegetatively, and recognize more than 50 different vari-


f,! -

. a I I


eties, clones, or landraces (Alemu and Sandford, 1996;
Shigeta, 1991; Zippel, 1995). Domesticated enset (also classi-
fied taxonomically as Ensete ventricosun) is planted at eleva-
tions ranging from 1,100 to more than 3,000 meters, indicat-
ing the extent to which its natural distribution has been
expanded artificially through domestication. Vernacular names
for domesticated enset include enset (Amhara), asat (Gurage),
weise (Kambata), and wassa (Sidama), among others.

I What are the enset-based
systems of Ethiopia?
There are four major agricultural systems in Ethiopia: pas-
toralism, shifting cultivation, grain-based cultivation, and enset-
based cultivation (Westphal, 1975). Within the enset agricultur-
al system, four major enset sub-systems can be recognized,
based upon environmental, agronomic, and cultural criteria, as
well as the extent to which people depend upon the plant as a
staple crop (Westphal, 1975; observation of authors).
One such sub-system is where enset is the staple food and
main crop. Such groups as the Sidama and Gurage grow enset
(Figure 2.2) in dense plantations, and are highly dependent

K;TT~iii *.i .T iiTai0]-0n -C

upon cattle to produce manure for fertilizing enset fields
(Plate 3, page 12). The main enset product is kocho, a fermented
bread-like food that is consumed locally as well as exported to
urban markets. Population densities in these communities are
commonly 200 to more than 400 persons per square kilometer.
Another enset sub-system uses enset as a co-staple with cere-
als and tuber crops. The Gamo, Hadiya, Wolayta, and Ari,
among other groups of SNNPRG, depend upon enset as a co-
staple in this manner (Plate 4, page 12). Within an ethnic group
such as the Hadiya, there may be differences between house-
holds, with wealthier or higher resource households using cere-
als more than enset, and lower resource households being
entirely dependent upon enset (Spring et al, 1996). Cattle are
important for manure to fertilize enset fields, while oxen are
used to plow cereal fields. Both kocho and amicho (boiled enset
corm) are eaten. Population density among these groups is high,
sometimes with more than 200 people per square kilometer.
A third enset sub-system relies upon cereals as the most
important crops, with enset and root crops of secondary impor-
tance. Such groups as the Oromo farmers of southwestern
Ethiopia exemplify this system, where both the hoe and plow
are used to grow cereals. Enset is grown largely for security rea-

Addis Ababa


. 5' -


.. -- -: -

11 IT
'5- L. .. .. S
^ -,,^ _.- =. .:
." 'i " ,; i" :t _. -- .-L

- ,^ .' .. -

I a '-"
\ EF J

* Areas of Enset Cultivation \

0 100 200 Kilometers --""-"
____ _____ _________^ ^ L u k n ___ -*_____________

sons (i.e., if cereal crops fail) and eaten
in the form of kocho and amicho.
Livestock are important for transport
and plowing, but far less so for pro-
ducing manure as enset fertilizer
The fourth enset sub-system is
where root crops are of prime
dietary importance, cereals are of
secondary importance, and enset is
of minor importance. Groups such
as the Sheko in southwestern
Ethiopia practice hoe-based shifting
cultivation, in which yams and taro
are the most important crops. while
enset, cereals, and cattle-herding are
of minor importance. Traditionally)
enset is processed for eating simply
by cutting the corm into pieces and
cooking over hot stones. Population
densities are low in these groups.
and settlements are small and dis-



I Where is enset grown and what is its
range of environmental adaptation?
Domesticated enset is planted at altitudes ranging from
1,200 to 3,100 meters. However, it grows best at elevations
between 2,000 and 2,750 meters. Most enset-growing areas
receive annual rainfall of about 1,100 to 1,500 milllimeters,
the majority of which falls between March and September.
The average temperature of enset growing areas is between
10 and 21 degrees centigrade, and the relative humidity is
63 to 80 percent.
Detailed studies on the effects of environmental con-
straints such as temperature and water availability have not
been conducted on enset. Therefore, comments about the
range of adaptation and the effects of environment on such
characteristics as plant growth, time to maturation, yield,
and pest management are from preliminary observations.
Enset is not tolerant to freezing. Frost damage on upper
leaves is commonly observed above 2,800 meters above sea
level, and serious stunting is seen above 3,000 meters. For a
certain range below 1,500 meters, the constraint to enset
plant growth probably is more related to available water
than to high temperatures. In most areas of Ethiopia below
1,500 meters, the total rainfall and the length of the rainy
season decrease, and the potential water use by plants
increases because of the greater evaporative demand. Most
enset plantings below 1,500 meters have supplemental irri-
gation or are small enough in size that household waste
water may be applied.
There has been much speculation about the drought tol-
erance of enset. Farmer interviews suggest that those popu-
lations dependent upon enset have never suffered from
famine, even during Ethiopia's tragic drought and famine
prone decades of the 1970s and 1980s. Several authors
(e.g., Bayush, 1991; Shigeta, 1990) have noted that enset
tolerates short season droughts that have seriously damaged
annual crops, especially cereals. During the dry spell, only
the edge of older leaves and the outer leafsheath are visibly
affected, and the plant resumes normal growth after the
onset of the rainy season.
Characterization of enset drought tolerance is a vital
issue in clarifying the role of enset in Ethiopian food securi-

ty Observed drought tolerance and its attributes must be
carefully interpreted. Will enset grow and produce success-
fully where the average annual rainfall is less than 1,100
millimeters, or where the dry season has an average length
greater than where it is currently grown? Research data are
needed to answer these questions. It is hypothesized here
that, once enset is established, it can tolerate occasional
years of unusually low total rainfall or a short rainy season.
During that stress year, enset plants may gain little addi-
tional weight, but they can survive and provide an all-
important food source (which can also be stored for
months and years) when there is failure of crops that pro-
duce an annual harvest. In environments where enset is
adapted, enset can serve a vital role similar to that of live-
stock, i.e., providing food "on the hoof' for famine years.
However, enset will fail in environments of consistently low
rainfall or short rainy seasons.
Enset is not affected by occasional heavy rainfall. This
resilience is attributed to the plant's stiff leaves, which resist
large rain drops. In fact, one of the main attributes of enset
is that it protects the soil from erosive rainfall. The main
danger of heavy rains to enset is that roots and the corm do
not tolerate water-logging for long periods. For that reason,
enset is usually grown in soils that do not have high water
tables and are well drained.
Enset grows well in most soil types, if they are sufficient-
ly fertile and well drained of water. Cattle manure is used as
the main organic fertilizer. Manure increases water holding
so that soil water endures longer into the dry season, and
reduces the negative effects of the high clay content of verti-
sols. The ideal soils in enset growing areas are moderately
acidic to alkaline (pH 5.6 to 7.3) and contain two to three
percent organic matter.



I How does the production of
enset affect the environment?
Observations in areas that have been planted with enset
for many years suggest that native soils have been altered
positively by the long-term application of manure.
Compared to native soils that have not been similarly treat-
ed, these modified soils are likely to be more fertile and
have better physical characteristics, such as water holding
capacity. Enset's perennial canopy of leaves and the abun-
dant accumulation of litter also reduce soil erosion and
organic matter decomposition to a minimum. Because enset
production improves soils, particularly with adequate
manure, many enset fields have been in continuous produc-
tion for decades, if not centuries. A current fear is that sig-
nificant increases in human population and decreases in ani-
mals and manure may cause reductions in crop yields and
soil fertility, thereby reducing the long-term sustainability of
the enset system. Increased use of fertilizer may not com-

pensate for the manure loss because of the multiple roles
that manure plays in improving soils biologically, chemically;
and physically.
Enset affects the physical environment around houses
where it is most commonly grown. Enset serves in the same
role as trees, providing people, other plants, and animals
with protection from wind and sun. Having a field that par-
tially encompasses the homestead is considered aesthetically
desirable by enset-based societies; enset beautifies the
Ethiopian landscape by its thick, dark green foliage (Plate 1).
Enset is also likely to affect the macro-environment of an
area in a positive manner. It has been commonly observed that
species like enset, with deep roots and leaf canopies of long
duration, improve the hydrological dynamics of an area. as can
easily be measured at the watershed level. As the proportion of
these species increases with respect to annual species, water
infiltration increases and surface runoff decreases, resulting in
more water in the soil and aquifers. The result is increased
water availability and greater volume and duration of discharge
to springs, decreasing the effective length of the dry season.



I What are the origins of enset agriculture?
Given the restricted geographic distribution of domesti-
cated enset and the degrees of complexity and variability in
contemporary enset agricultural systems, agronomists and
biogeographers have long considered the Ethiopian high-
lands to be the primary center of origin for enset agriculture
(Harlan, 1969 and 1992; Sauer, 1952; Vavilov, 1951).
Anthropologists, archaeologists, historians, and other schol-
ars have also developed theories that argue for the domesti-
cation of enset in Ethiopia as early as 10,000 years ago.
Stiehler (1948), one of the first scholars to consider enset
origins, believed that the indigenous hunter/gatherers of
southern Ethiopia were the first to cultivate enset. He also
proposed that enset agriculture was later introduced to the
northern Ethiopian highlands by Cushitic-speaking peoples,
only to be replaced by such crops as wheat, barley, and t'eff
following the migration of Semitic-speaking groups into
northern Ethiopia. In a similar vein, Murdock (1959) sug-
gested that sometime in prehistory "Sidamo tribes" (i.e.,
Omotic and eastern Cushitic-speaking groups) of southwest-
ern Ethiopia independently brought enset under domestica-
tion. Later, central Cushitic-speaking peoples of northern
Ethiopia (i.e., the Agaw) also began to grow enset and a wide
range of other crops, and were quick to incorporate wheat,
barley, cattle, goats, and sheep into their economy once these
domesticates were introduced into Ethiopia from Dynastic
Egypt. Soon thereafter, cattle became important to enset
farmers as a source of manure for fertilizing their fields.
Another theory proposes that Nilo-Saharan speaking
farmers were forced out of the lowlands of eastern Sudan and
western Ethiopia some 4,000 to 5,000 years ago because of
the increasingly drier climates of the mid-Holocene (Clark,
1967 and 1976). Migrating east to the Ethiopian highlands,
they introduced farming to the indigenous hunter/gatherers
of highland Ethiopia and Eritrea, who began cultivating enset
and other indigenous Ethiopian domesticates on their own.
Drawing largely upon historical-linguistic data, Ehret (1979)
proposed another theory that argues for a much earlier date
for the beginnings of enset food production, perhaps as early
as 10,000 years ago. He suggested that Omotic-speaking
peoples, responding to a food crisis at the end of the

Pleistocene, first increased their consumption of wild enset
and then eventually domesticated the plant. Sometime
between 8,000 and 5,000 years ago, cattle, sheep, and goats
were introduced into Ethiopia from the Sudan and were
rapidly incorporated into existing enset systems.
Recognizing the need to explain how enset agriculture
evolved into the diverse systems that exist today, S. Brandt
(1984 and 1996; and Brandt and Fattovich, 1990) devel-
oped a model that expands upon previous theories by argu-
ing that the arid conditions of Ethiopia during the height of
the Last Glacial some 18,000 to 10,000 years ago (Gasse et
al, 1980) resulted in major changes in the environment and
in the abundance and predictability of critical resources. The
highlands of southern Ethiopia became an environmental
refuge where "complex" hunter/gatherer systems emerged,
which used certain wild animals and plants, including enset,
as dependable, stress-relieving food resources. Between
10,000 and 5,000 years ago, enset was fully domesticated
and a system of shifting cultivation emerged.
By the mid-Holocene (4,000 to 5,000 years ago), the intro-
duction into Ethiopia of foreign domesticates such as cattle,
sheep, and goats (Brandt and Carder, 1987), as well as wheat
and barley, resulted in the establishment of more intensive
forms of agricultural production in the highlands. These
forms included the use of the plow, irrigation, and terracing,
as well as the greater utilization of manure as a means to
maintain the fertility of enset without having to practice shift-
ing cultivation. Increasing population densities may have
forced some societies to develop additional methods of inten-
sification, including techniques to postpone consumption and
prevent surplus crop spoilage (e.g., the fermentation and stor-
age of enset in deep earthen pits). Over the last 3,000 years
new socioeconomic and political alliances resulted in the
establishment of chiefdoms and states in highland Ethiopia,
dependent to various degrees upon enset food production.



I What is the historical evidence for
enset agriculture in northern Ethiopia?
Today the vast majority of enset farmers live in southern
Ethiopia. However, historical evidence suggests that enset
may have once played a much more important role in the
agricultural practices of central and northern Ethiopia. The
earliest recorded evidence of enset agriculture in northern
Ethiopia is from the "Royal Chronicles"-medieval manu-
scripts written by priests in the liturgical Ethio-Semitic lan-
guage of Geez. There is a single passage dating to 1590
mentioning Oromo peasants growing enset for food south of
the Blue Nile River (Pankhurst, 1996).
European travelers of the 1600s and 1700s provide infor-
mation on enset agriculture. In the early 17th century,
Manuel de Almeida, a Portuguese Jesuit traveling through
northern Ethiopia in the area south of Lake Tana and north
of the Blue Nile, noted that enset was "the sustenance of
most of the people ... The tree itself is eaten, either sliced
and boiled, or crumbled and ground into meal which they
put into pits in the ground where it keeps for many years
..." (Almeida, 1954). In 1640 Jeronimo Lobo, another
Portuguese priest traveling in the region, described enset as
"a tree peculiar to this country" which "when cooked ...
resembles the flesh of the turnips, so that they have come to
call this plant 'tree of the poor' even though wealthy people
avail themselves of it as a delicacy, or 'tree against hunger,'
since anyone who has one of these trees is not in fear of
hunger" (Lockhart, 1984).
The 18th century Scottish traveler James Bruce described
enset as grown in "large, thick plantations" south of Lake
Tana, "exposed for sale" in local markets and as "food in
great quantity" growing in "great perfection at Gondar."
Furthermore, he stated that it was "the general opinion" that
enset was "naturally produced in every part of Abyssinia,
provided there is heat and moisture" (Bruce, 1790).
Although R. Pankhurst (1996) has questioned the accuracy
of some of Bruce's descriptions of enset, there is little doubt
that enset was a significant crop in the Lake Tana region at
that time.
However, by the 1840s, enset had apparently all but dis-
appeared as a food source in the north. Charles Beke, a
British traveler, provided a detailed description of farming in
the Lake Tana area as a region dedicated to cereal produc-
tion and consumption with little enset (Beke, 1844).

The reasons) for the rapid demise of enset in northern
Ethiopia remains unknown and unstudied. Possibilities
include disease and drought. It is also possible that the dra-
matic socio-political events that took place in northern
Ethiopia between the mid-1700s and mid-1800s played a
critical role in the rapid reduction of its production. In
1769, following the collapse of the once-unified Kingdom of
the Solomonic Dynasty, northern Ethiopia entered the peri-
od known as the "Era of Princes" or Zemane Masafent.
During this turbulent period, northern Ethiopia was
racked by socio-political and economic insecurity and unrest,
brought on by the rapid rise and fall of petty kingdoms,
increasingly more dependent upon tax and tribute from their
desperate peasantry. Kaplan (1992) states that "for the popu-
lation in general and the peasants in particular, the Zemane
Masafent was a period of severe hardship. In the best of
times the lot of the peasants and in particular those who
labored as tenant farmers was not a happy one. For them the
endless military conflicts of the Zemane Masafent aggravated
an already difficult situation. The soldiers of the different
regional armies lived off the land, ravaging both enemy terri-
tories and those of their masters. Insecurity poverty and
depopulation were characteristic of this period.-
The consequences of this difficult period for enset farm-
ing could have been two-fold. First, peasants may have been
unable to devote the minimum two to three years necessary
to re-establish enset farms and regain associated livestock
during unending periods of insecurity, destruction, displace-
ment, and depopulation. In this situation, it would have
been much simpler to plant and harvest annual cereals.
Second, faced with rising debts from standing armies and
other war-related activities, landlords and nobility may have
directly and indirectly placed considerable pressure upon
the peasants to emphasize more prestigious, surplus-produc-
ing, and income-generating crops such as cereals, rather
than subsistence "peasant food" like enset. A somewhat
analogous situation occurred during the 1600s in the Kaffa
kingdom of southern Ethiopia. Here, the desire of the royal
court and elites for the "prestige- foods of t'eff and other
cereals spurred them to demand cereals as tribute. since
cereals "were better for tax collectors since they could be
stored, divided and moved" (Orent, 1979). Whatever the
causes, by the end of the 19th century when King Menelik
conquered surrounding regions to create the modern map of
Ethiopia, enset food production in the north was practically



Unfortunately, archaeological and historical research into
the origins and evolution of enset agriculture is just begin-
ning, so specific data are lacking (Brandt, 1984 and 1996;
Clark, 1988; McCann, 1994; Pankhurst, 1996; Phillipson,
1993). Therefore, the various theories scholars have con-
structed are untested and will remain so until long-term
archaeological and historical research is initiated.

I What role has enset played in the
agricultural policies of Ethiopia's
recent and current governments?
During Haile Selassies's reign from the 1920s to 1974,
and in particular after World War II, Ethiopia's Ministry of
Agriculture launched major initiatives to increase food pro-
duction. Among these initiatives was the establishment of
Ethiopia's first agricultural university, funded and staffed in
large part by the United States. Haile Selassie's government
gave explicit instructions to focus upon cereal agriculture
and other income-generating crops such as coffee; enset was
virtually ignored.
Following the Ethiopian Revolution of 1974, the commu-
nist-inspired military dictatorship established small research
programs and experimental stations for enset, but provided
little in the way of operating funds or staff. It also tried to
establish Soviet-type collective farms in enset growing
regions, with the usual abysmal results. After the fall of
Mengistu in 1991, the current Ethiopian government has
shown more interest in enset and recognized its importance
to the people of the south. In 1997 the government declared
enset a "national crop" worthy of significant increases in
research and development funding.





I What are the food uses of enset?
The major foods obtained from enset are kocho, bulla and
amicho. Kocho is the bulk of the fermented starch obtained
from the mixture of the decorticated (scraped) leafsheaths
and grated corm (underground stem base). Kocho can be
stored for long periods of time without spoiling. The quality
of kocho depends on the age of the harvested enset plant, the
type of clone (variety), and the harvesting season. Moreover,
within one plant, the quality is influenced by the part of
leafsheath and corm processed. The preferred type is white
in color and is obtained from the innermost leafsheaths and
inner part of the corm, while the lowest grade is blackish
and is obtained from the outer leafsheath and corm.
Although many different dishes are prepared from kocho
(Plate 61, page 17), a pancake-like bread is the most com-
mon. Kocho prepared as a fermented enset bread has also
become extremely popular at restaurants that serve the
Ethiopian delicacy of kitfo (raw ground beef mixed with but-
ter and spices). The combination of kocho and kitfo is now
virtually required at restaurants. (Plate 5c, page 13).
Bulla is obtained by: 1) scraping the leafsheath, peduncle,
and grated corm into a pulp; 2) squeezing liquid containing
a starch from the pulp (Plate 6f, page 16); 3) allowing the
resultant starch to concentrate into a white powder; and 4)
rehydrating with water. It is considered the best quality
enset food and is obtained mainly from fully matured enset
plants. Bulla can be prepared as a pancake, porridge, or
Amicho is the boiled enset corm, usually of a younger
plant (Plate 5b, page 13). Enset plants may be uprooted for
preparing meals quickly if the amount of enset harvested is
insufficient, or for special occasions. The corm is boiled and

consumed in a manner similar to preparation methods for
other root and tuber crops. Certain clones are selected for
their amicho production.

I What are the non-food uses of enset?
Enset provides fiber as a byproduct of decorticating the leaf-
sheaths. Enset fiber has excellent structure, and its strength is
equivalent to the fiber of abaca, a world-class fiber crop. About
600 tons of enset fiber per year are sent to factories. In rural
areas the fiber is used to make sacks, bags, ropes, cordage,
mats, construction materials (such as tying materials that can be
used in place of nails), and sieves (Plates 5g and 5h, page 14).
Fresh enset leaves are used as bread and food wrappers,
serving plates, and pit liners to store kocho for fermentation
and future use (Plate 6g, page 16). During enset harvesting
enset leaves are used to line the ground where processing
and fermentation take place.
The dried petioles and midribs are used as fuel, and to make
mats and tying materials for house construction (Plate 5f, page
14). The dried leafsheaths are used as feed and wrapping mate-
rials. The pulp from the dried leafsheaths, petioles, and midribs
is used as cleaning rags and brushes, baby cushions/diapers,
and cooking pot stands. Dried leafsheaths are used as wrappers
for butter, kocho, and other items to transport to local markets
(Plate 5e, page 12). Enset leaves are an important cattle feed,
especially in the dry season when grasses are scarce. Leaves are
carried into the house for stall feeding of cattle during the
nighttime (Plate 5d, page 13).
Particular clones (or varieties) and parts of enset plants
are used medicinally for both humans and livestock to cure
bone fractures, broken bones, childbirth problems (i.e.,
assisting to discharge the placenta), diarrhea, and birth con-
trol (as an abortifacieant).




Plate 1 Plate 2

Plate 3 Plate 4

1. Enset plants surround the house providing shade, security, and a close-by food supply
2. Wide spacing between nearly mature enset plants in this woman farmer's field. The distance between
plants is being measured by Ethiopian scientists.
3. Enset as a staple crop in Sidama. Notice that enset is grown close to the home, while cattle graze outside
the enset area, which are sometimes fenced to keep them out.
4. Enset and annual cereal crops are mixed in this system of wealthier households in one area in Hadiya.
Notice the planting on slopes and the hedge fencing between crops to restrict livestock.



Plate 5a Plate 5b

'late 5c Plate 5d

5a. A woman selling kocho at a local market. Note that enset leaves are used like plates or mats on which to
display the product. Only woman market enset food products.
5b. Husband and wife selling amicho, the small, immature corm, at the market. The amicho will be boiled and
eaten like a potato.
5c. Chef in restaurant in Addis Ababa displaying a variety of cooked foods made of bulla (including mix-
tures with butter, cheese, meat, and dark greens) and kocho (rolled pancake bread).
5d. Boy feeding enset leaves to cattle.



Plate 5e

Plate 5f

Plate 5g

Plate 5h

5e. Enset leaves being sold in the market. They will be used as wrappers for fresh food. as well as for cooking
food on the griddle.
5f. Man using cordage made from dried enset petioles and midribs as construction material to build a house
5g. Men selling rope at a local market made from enset fiber (gleaned from processed pseudostems-
Most rope sellers and buyers are men.
5h. Woman at local market selling a basket sieve both made from enset fiber and used to sift kocho and bul!a
before cooking.


Plate 6a

Plate 6c

6a. Man uproots enset plant for processing.
6b. Woman has removed the outer leaf sheaths.
6c. Women pull apart the pseudostem.
6d. Woman decorticates pseudostem using a bamboo scraper.



Plate 6b

Plate 6d


Plate 6f

Plate 6g

Plate 6h

6e. Women pulverize corm in situ using serrated wooden tool.
6f. Squeezing starch (bulla) from decorticated and chopped enset.
6g. Woman putting decorticated and chopped enset in pit lined with enset leaves.
6h. Woman taking fermented kocho out of the pit. Note the two qualities of kocho. The darker.
poorer quality kocho is from the edges of the pit.


Plate 6e


Plate 6i

Plate 6j

Plate 61

6i. Woman chopping the fibrous enset using a wooden device to protect her hands; nonetheless, women
often have scars on their hands.
6j. "Pearling" enset to remove fibrous remnants and create a textured product before cooking.
6k. Shaping the kocho for additional chopping and fiber separation before cooking.
61. Cooking kocho into the pancake-like flat bread on an iron griddle.



Plate 8

Plate 7

Plate 9

7. Enset research trial at Areka Research Station. Note that thus far, there have not been any trials carried
out by farmers on their own lands.
8. Man incorporating manure and preparing the land for enset cultivation using an iron pointed hoe.
Notice the two stages of enset plants in the background, the smallest being one to two year old suckers.
and the larger ones being three to four year old transplants.
9. Lalibela, northern Ethiopia. Enset grows in the garden of a household. Enset is not processed for food.
but instead the leaves are used to wrap bread for baking.

Photo Credits
Assefa Amaldegen Plate 5d
Steven Brandt Plates 3, 5b, 5f, 9
Clifton Hiebsch Plates 5c, 6a, 6b, 6c, 6e, 6f, 6h, 6j, 8
Anita Spring Plates 1, 2, 4, 5a, 5e, 5g, 5h, 6d, 6g, 6i, 6k, 61, 7



I How is enset produced?
Compared to most crops, particularly annuals, the pro-
duction of enset involves many more steps. Suckers are usu-
ally produced from the two- to four-year-old corms (10 to
20 centimeters in diameter) and the true stem. (The mother
corm piece may be a whole corm or some portion of it.)
These mother corm pieces are obtained by harvesting
healthy plants, cutting off the pseudostems, removing the
roots, and cutting out the center or apical bud, from which
leaves and the flower stalk develop. Because of dominance
by this apical bud, lateral buds on the true stem do not usu-
ally develop; but once the apical bud is removed, these lat-
eral buds form suckers around the periphery of the mother
corm piece.
The mother corm pieces are usually planted in a nursery,
often with manure, where they can receive extra care.
Suckers are also formed from plants left in situ with the
pseudostem and apical bud removed. It is common for a
farmer to have 5 to 15 mother corm pieces each year.
Usually from 20 to 100 suckers form per corm piece. These
suckers are usually allowed to grow for one year before
transplanting, although they may be transplanted sooner or
even left for a second year if the farmer has excess planting
Suckers are transplanted using a hand hoe, usually to an
area that has been well prepared with added animal manure.
At and beyond this stage of sucker transplanting, there is
tremendous variation in management. Plants may be trans-
planted only once or up to four times, at ever wider spacing.
Not all plants within a farm or a field may receive the same
transplanting management. Some plants may be harvested at
a young age (two to three years) for amicho and some may
later be harvested for kocho. This variation in transplanting
and harvest management seems to be a function of ethnic
group, household needs, and available resources (such as
land, labor, capital, and other food crops in the system).
Elevation primarily affects the number of years that plants
are left at each stage, because cooler temperatures slow plant
growth. By contrast, manure speeds plant growth and time
to harvest.

A general objective of most enset transplant systems
seems to be to maintain a leaf canopy that covers the soil for
most of the year. Therefore, small plants are spaced close
together, e.g., half a meter apart, and/or are intercropped
with other species or larger enset plants. As increased plant
size and leaf canopy allows for efficient use of a wider spac-
ing, some or all plants in a field may be transplanted to
another field. For example, there are up to three additional
transplantings of all plants in the Gurage system, while in
the Sidama system some plants are thinned and moved to
another field. Whether taken from an area where all plants
are removed or from an area where there is selective thin-
ning, the plants may be: 1) transplanted to a uniform stand
of only removed plants; 2) incorporated with plants of simi-
lar size but different ages and previous management; 3)
planted in open spaces between taller (either uniformly
sized or variably sized) plants; or 4) intercropped (e.g., with
coffee or citrus trees).
Enset may be grown alone in uniform stands of similarly
sized plants, in mixed stands .of enset plants of different
sizes, ages, and clones, or in a combination of these at dif-
ferent stages. Enset may also be intercropped with other
species, in which case there is a tendency to intercrop
younger enset plants with annual crops (such as maize and
cabbage), and older enset plants with perennials (such as
coffee and citrus). In either intercrop type, farmers recognize
that the growth rate of enset is decreased. There are, howev-
er, no research data quantifying the effects of such cropping
strategies on the performance of enset or other crops in the
system, although some trials are underway at Areka
Research Station (Plate 7). Also, there is no extension infor-
mation available for a group of best management strategies.
The age of enset plants to be harvested may be uniform
or variable. For example, most plants harvested by the
Gurage are nearly mature (although poorer households may



have to harvest immature plants), while the Sheko tend to
harvest many young plants for amicho. The Gamunya har-
vest plants of varying ages. Within an ethnic group, duration
to harvest is affected by elevation (temperature), the age and
characteristics of particular clones, the intended uses for
food or cash, management (such as plant spacing and
manure rates), and the wealth level of the household
(addressed in more detail below under section on "Case
Studies"). Farmers tend to believe that it is better to harvest
plants at or near plant maturity, and that harvesting younger
plants indicates an inadequate food supply or poverty. The
importance of harvesting nearly mature plants is particularly
stressed by male farmers, while female farmers in some of
the same ethnic groups (e.g., North Omo) indicate they pre-
fer to harvest smaller plants for better taste and ease of fer-
mentation (Habte-Wold et al, 1996; Spring, 1996a; Tibebu
et al, 1996; see also section on "Gender Issues," below).
All research to date on the yield of enset has been with
nearly mature plants, and generally has been at spacings
designed to maximize yield per plant in the minimum dura-
tion. However, this is not the apparent objective of most
farmers' strategies. There are no data available to compare
yield or land use when harvesting many closely spaced,
younger, smaller plants versus fewer, widely spaced, older,
larger plants. Studies on combinations of these and inter-
cropping (either mixed sizes, clones of enset, or mixed
species) strategies also are nonexistent. There is no exten-
sion information available on the spacing, timing, intercrop-
ping, and harvesting of enset, while for other crop species
these facts are considered baseline extension information.
Ideally, farmers use cattle manure on enset (Plate 8). It is
common for enset to receive available manure before other
crops. In an ideal enset system, ruminant animals such as
cattle, sheep, and goats graze on large areas of grassland and
are then housed at night in corrals where manure is collect-
ed. This manure is then applied to enset and to a lesser
extent to other crops. For many communities and individual
households, however, manure is often scarce or nonexistent
because inadequate grazing land or lack of resources limit
animal numbers. In reality, increasing human population
densities and/or the disappearance of grazing lands in poor-
er households lead to declining animal numbers and
manure quantity per household, and in turn to decreasing
enset yields.

In a few households with sufficient capital, fertilizer is
beginning to be used on enset. This is particularly evident in
the Sidama region, where farmers are accustomed to putting
fertilizer on coffee, and have cash incomes from coffee sales.
In the absence or shortage of cattle manure, some have tried
inorganic fertilizer (diammonium phosphate) on enset. The
results are mixed; growth is greater, but food yields do not
increase correspondingly
As available land per capital becomes more limited, the
role of ruminant livestock for manure supply becomes one
of the greatest threats to the future of this highly successful.
sustainable, indigenous system. The potential for alternatives
has not yet been researched. There are no research data
available as a base to use for advising farmers on the rates of
manure and inorganic fertilizer that should be applied to
enset. Similarly, there is no information on the effect of soil
type, environment, age and size of enset plant, harvesting
management, intercropping, or any other variable on opti-
mum rates of manure or fertilizer. Improved pasture and
cut-and-carry systems to augment ruminant meat and milk
production, let alone manure production, have not been

I What are the diseases of enset?
Diseases are collectively the most severe biological prob-
lem facing enset. The damage that diseases can cause and
the lack of knowledge about or implementation of preventa-
tive strategies contribute to the severity of enset plant dis-
eases. Diseases are caused by several bacteria, nematodes.
fungi, and viruses. Bacterial wilt, caused by the bacteria
Xanthomonas campestris pv musacearum, is the most threat-
ening to the enset system. Bacterial wilt attacks plants at any
stage, including full maturity When bacterial wilt, or any
other cause, kills an enset plant late in its life cycle, it is a
particularly serious loss. The farmer has already invested
several years of land, labor, and resources into the plants
production. In some enset-growing areas, such situations
have caused farmers to abandon their enset farming and
replace it with annual crops.
Enset is attacked by numerous diseases in addition to
bacterial wilt. They include enset corm rot, enset sheath rot,
and enset dead heart leaf rot, caused by an unknown bacter-
ial pathogen and fungus, respectively, as well as root-knot,
lesions, nematodes, and virus diseases.
The most important factors responsible for spreading dis-
ease of bacterial wilt include disease-infected planting mate-



rial, contaminated farming and processing tools, and human
and animal vectors. The only research-recommended control
measures for diseases are cultural measures to prevent the
movement of the causal agent. For bacterial wilt, these mea-
sures include the use of healthy, disease-free suckers for
planting material; destruction and controlled movement of
diseased plants; cleaning of equipment that has come in
contact with diseased plant material; and rotation of crops.
The specific practices required for realization of these con-
trol measures can require sizable investment of additional
care and labor. If a farmer is not knowledgeable about the
cause of disease, is not convinced that the additional effort
will make a difference, or has insufficient labor, the control
practices are not likely to be adopted. While there is still
much to research about enset diseases, adoption of known
prevention could be part of an extension campaign.
However, the current lack of extension programs presents
the main limitation to disease management.
Bacterial wilt is easily spread by any object touching the
contaminated parts of the plant or processed enset (i.e.,
kocho). Contaminated cutting and processing tools, in par-
ticular, spread the disease. Cutting enset leaves for animal
feed and wrappers may spread the disease from one plant to
another. It is also postulated that mole rats, which burrow
underground, can cause contamination as they tunnel from
one plant to another. Similarly, snakes and insects going
between plants, as well as the presence of cattle walking
through fields, could also contribute to the spread of bacter-
ial wilt. All of these transmission agents need to be
In the case study areas (described in more detail below),
virtually all enset fields of Gurage households are infected
by bacterial wilt, while about half of the Hadiya households
have infected enset fields. By contrast, the majority of
Sidama households report no enset diseases on their farms,
although a few households report wilt in the older enset
plants. Farmers mention that bacterial wilt is more severe at
high altitudes, but more research is needed to confirm this.
The reason for these differences is probably related to farm-
ers' knowledge of the methods of disease spread and conta-
mination of tools.
Typical bacterial wilt symptoms in enset plants above two
years old is that the innermost leaf sheaths become yellow-
ish and droop. Usually only older plants are attacked; how-
ever, in one area studied the disease was observed attacking
younger enset plants (even the one-year-old suckers).

The control measures used by farmers are inadequate, and
seem to facilitate the distribution of the disease. In areas of
greatest infestation, farmers loan or borrow farming and pro-
cessing tools. Therefore both men and women engaged in
cultivating, processing, and cutting leaves spread the disease.
Some households control bacterial wilt by uprooting and
discarding infected enset plants as a cultural control mea-
sure; however, few households use sanitation methods for
their tools. Too frequently, infected enset plants are disposed
of near the enset farm, and, as such, they may be a potential
source of the disease inoculum and its re-spread. Some
farmers fallow the enset field and practice rotation with
annual crops. By contrast, in the area that is relatively free of
bacterial wilt, farmers practice control measures such as
uprooting the infected bacterial wilt enset plants and keep-
ing them away from the household, other enset plants, and
cattle. Farmers also try to keep healthy plants away from
contaminated farm and processing tools.
Some households decorticate and process infected enset
plants at their early stage of infection. Some women separate
the kocho from infected plants (bulla is never made from
infected plants) from the kocho of healthy plants by putting
them in separate pits. The wealthier households have the
choice of utilizing diseased plants or not. They may chose to
process this lower quality kocho separately and sell it. The
poor have less choice and may use it for home consump-
tion, purchase it, or receive it from more affluent households
as payment for their labor or craft product.
Farmers note that certain enset clones have relatively
high tolerance against bacterial wilt and that particular
clones revive after infection has occurred, while other clones
of similar age group are wiped out by the disease.
Many Gurage farmers report disease problems related to
kocho stored in pits, especially for an extended period of
time. A tooth-shaped fungal mycelial growth (species not
identified) on the upper surface of stored kocho in the pit is
reported. The disease starts around the wall of the pit, caus-
es a bad small, and the hocho becomes highly compacted
because of dehydration. Farmers report that this problem is
common in kocho stored for extended periods of time and is
easily transmitted from the nearby infected enset storage pits
or pit lining materials. Farmers try to control this problem
by regularly aerating the kocho and changing the pit.



I What mammals and insects attack enset?
Porcupine, mole rat, and wild pig attack enset plants in
the field. They usually damage the plant by feeding on the
corm and pseudostem. Among these pests, the mole rat
ranks number one. Since these animals are not microscopic
like the bacterial and viral diseases, farmers are knowledge-
able about them, and many employ effective management
practices. These practices include woven fences and ditches
around enset fields, to retard the movement of animals into
the field, and traps for catching them. Wealthier farmers use
steel traps to snare wild pigs and monkeys, and others traps
for porcupines and mole rats. Burning coals may be dumped
into the rat tunnels. Farmers also protect against porcupines
by digging pits around enset plants so that it is difficult for
the animals to get in or to climb up and get away. In some
areas, farmers organize themselves on a village basis to hunt
wild pests with sticks, spears, and machetes.
Insects have been considered a minor problem in enset
cultivation. However, over the last several growing seasons,
mealy bugs have been identified as a serious problem in cer-

tain regions. Mealy bugs are soft-bodied insects that feed on
the corm and roots. Enset plants infected by mealy bugs
show stunted growth; the damage appears more severe dur-
ing the dry season. Because they live underground, their
damage often goes unnoticed until serious loss has occurred.
As they are slow moving insects, mealy bugs are controlled
with methods similar to those used against diseases such as
bacterial wilt.

I What effect do weeds have on enset?
Weeds can cause greatly reduced plant growth while
enset plants are small, i.e., during the sucker stage and for
one or two years after the first transplant. The total land
area used for production during these stages is usually rela-
tively small compared to the farm size, and therefore weed-
ing can be accomplished with available labor. As the enset
plants become larger, the perennial leaf canopy and leaf lit-
ter on the soil surface prevent most weed growth. In annual
crop production, the labor available for weed control can be
a serious restriction to production. Although labor data are
not available, there is probably much less labor required for
weed control per ton of food in enset than on any annual
crop. The reduction in labor for weed control may be offset
in part by the additional labor required during harvesting
and processing of enset.



I How is enset harvested and processed?
Although enset is usually harvested just before flowering,
the preferred harvesting time is just when the plant flowers.
The time duration required to flower depends upon climatic
conditions, clone type, and management. Hence, the flower-
ing time varies from 3 to 15 years but is optimally around 6
or 7 years. Enset processing is carried out by women using
traditional tools (Plate 6), and the process is laborious, tire-
some, and unhygienic. The processing is done totally by
women in most ethnic groups; however, men occasionally
assist women, as among the Gamo (Plate 6a).
At harvest, leaves and older leaf sheaths are first removed
from the designated plants. The internal leaf sheaths (com-
monly up to two meters in length) are separated from the
pseudostem down to the true stem, which is about a 20
centimeter section between corm and pseudostem (Plate 6c).
Then the true stem is separated or stumped from the under-
ground corm. The concave side of the leaf sheath is peeled
and cut into pieces of about one meter length and split
lengthwise in order to shorten the leafsheath to a workable
size. Then the leafsheath is decorticated using a locally made
bamboo scraper while the leafsheath is held on an incline
(at 45 to 80 degrees from the ground) against a wooden
plank (Plate 6d). In some groups, women may sit on the
ground (often on enset leaves) and use one leg to hold the
leafsheaths in place, while in other areas they bind the
sheath to the board and stand to decorticate. The working
area used for decortication is covered with enset leaves.
There is variation in tools used (bamboo versus newly
adopted metal scrapers).

There is also variation in the way that the corm is grated
(Plate 6e). One practice is to uproot the corm and remove
any soil from its surface. Then the corm is grated separately
with a locally made wooden tool with a sharp serrated edge.
Another method is to grate the corm from the inside out
while still in situ in the ground.
After the completion of decorticating and grating, the
leafsheath pulp is spread on fresh enset leaves covering the
ground, after which the grated corm is spread on the
processed pulp. In some ethnic groups (e.g., Hadiya and
Sidama) a starter is added to aid in fermentation. This
starter consists of already fermented kocho to which various
spices and herbs are added. In other localities (e.g., Gurage),
fermenting agents are prepared from the inner portion of the
corm and then mixed with the decorticated pulp and grated
corm after some weeks. Turning, mixing, rinsing, and chop-
ping continue over a period of time until the mixture par-
tially ferments, when it is then referred to as kocho (Plate 6h-
61). The total time period for this fermentation to occur
ranges from 15 to 20 days. Then the fermented kocho is
stored in pits that are lined with enset leaves (Plate 6g and
6h). Pits vary in terms of size and depth, with some requir-
ing ladders. The kocho must be left in a storage pit for a
minimum of a month, but it can be stored for many months
and even for several years. Some women note that for long-
term storage, the kocho should be removed, the pit lining
changed, and then the kocho returned to the pit.



I Why are livestock important
in enset systems?
Regardless of elevation, ethnic group, or degree of depen-
dence on enset in dietary intake, it appears that livestock
play a critical role in maintaining soil fertility (and thus agri-
cultural sustainability). Livestock therefore play a critical
role in enset farming systems, as they provide: 1) manure for
important plant crops, including enset; 2) food, especially
milk and occasionally meat for the family; 3) traction for
plowing; and 4) a source of wealth that can be sold to pro-
vide cash in times of need. Additionally, in ethnic groups
that use equines for transportation and hauling, bundles of
enset are transported to local markets. Livestock are also
kept as an indicators of wealth and sources of prestige
among rural cultivators.
Manure is generally applied to crops grown in the vicini-
ty of the household, especially to those considered especially
important. Enset and coffee, where grown, are almost always
given priority in this regard. Enset plants may take twice as
long to mature without manure as they do with the applica-
tion of manure (McCabe and Lee, 1996). This view is fur-
ther supported by a recent survey, which shows that "ade-
quate farm manure is regarded as essential to successful
enset growing, and ... farmers with unproductive looking
enset plantations were those farmers who have the fewest
livestock" (Alemu and Sandford, 1991).
In systems where enset is mixed or secondary to cereals,
oxen are critical resources in the preparation and plowing of
fields for planting wheat, barley, t'eff, and other cereals.
Among farmers in the enset growing region, important live-
stock uses do not appear to include slaughtering to provide
meat for the household. However, the consumption of milk,
butter, and cottage cheese seems to make an important and
possibly critical contribution to an enset-based diet, which
in itself is very low in protein (Shank and Ertiro, 1996; Pijls
et al, 1994).

Livestock throughout the rural regions of Africa and
Ethiopia act as a store of wealth for a family or household,
and this is certainly true for enset cultivators. Households
depend on livestock as emergency resources for such things
as hospital fees, and they try to save livestock for large pur-
chases such as construction materials to build a new house.
The sale of milk and milk products can also make signifi-
cant contributions to household income. It appears that
until relatively recently, milk and milk products (such as
butter and cottage cheese) were reserved almost exclusively
for home consumption. However, during the past three
decades, the sale of these products has increased, and in
some cases has provided as much as 45 percent of annual
household income (Beneye, 1994).

I How are livestock managed
among enset farmers?
There are common themes in the management of live-
stock among the peoples who cultivate enset, although
regional and ethnic differences occur. Differences in manage-
ment practice may be found in the different ecological
zones: dega (highlands), weinadega (mid elevation), and kolla
(lowlands). There is also variation according to wealth cate-
gory, with the wealthier households possessing more live-
stock and requiring greater access to additional labor and
grazing lands.
The management of livestock involves both taking ani-
mals to pasture and bringing forage to livestock. Individuals
with one or two cattle will normally tether their animals in
the grassy area in the front or side of the homestead. Those
with more livestock will both tether their animals near the
house and take their animals to common grazing areas, if
they are available. In many villages, swampy or steep areas
are set aside for common grazing ("the commons"). Those
who are wealthy utilize the methods previously mentioned,
but also may take their livestock for periods of time to sec-
ond homesteads where the grazing resources are more abun-
dant. Usually these second homesteads are in the lowlands,
but in areas of uniformly high elevation they may be above
the elevation preferred for crop production. Access to the
common grazing areas is usually determined by farmers'



proximity to the commons. However, the use of the com-
mons appears to be changing, as more common grazing
land is being turned into land that is cultivated.
Stall feeding is the principal means by which livestock are
fed during the dry season throughout the enset growing areas,
and is a labor intensive activity Following harvest, crop
residues are given to the livestock, and among all enset grow-
ing groups, enset leaves form an important part of the dry
season livestock diet. Grass may be purchased, but it is more
often cut by women from homestead pastures or common
grazing areas. Cut enset leaves also contribute to livestock
diets in all areas where enset is grown, and they may be used
for as long as seven to eight months, or only for a couple of
months at the height of the dry season, depending on area
and ethnic group. Adebo (1992) reports that in one region
during the early and middle dry season, women cut grasses
from the village area, but when these resources were used up
they had to walk to the lowlands to cut forage for their live-
stock. This task entailed a seven to ten hour round trip.

I What are the major constraints
to livestock production?
Cows kept by rural farmers in the enset growing area
produce low quantities of milk. Estimates range from a low
of approximately 0.25 liters per day to a high of about two
liters per day during their seven month long lactation peri-
od. The amount of milk produced increases during the wet
season, as forage resources are more abundant and calves
tend to be dropped during this time. Low animal fertility is
also characteristic of the livestock kept by enset cultivators.
Mortality rates are high and enset cultivators typically buy
and sell livestock frequently.
Low productivity and high rates of mortality and turnover
strongly suggest that the livestock production system is
under significant stress. Current data indicate that the most
severe constraint is lack of adequate forage. A decrease in the
amount of land allocated for grazing per village, and the
transformation of some common grazing land to crop pro-
duction have contributed to this decline in forage resources.

I How has the system of livestock
management changed over the
last few decades?
Although it is typical for farmers throughout the world to
remember the past as a time of plenty in contrast to the
troubles of the present, the consistency of accounts in which
farmers kept far more livestock in the past than they do now
is striking. Preliminary research strongly suggests that there
has been a serious decline in the numbers of livestock held
by farmers on a household basis. What data are available
suggest that a typical household kept seven to eight head of
cattle, a number of small stock, and possibly a horse or two
during Haile Selassie's time, while now the average house-
hold keeps two to three cattle, and maybe two or three
sheep or goats.
This negative downward cycle is a result of increased
demands for cultivated land as a result of increasing popula-
tion pressure. Changes in the system of land tenure also con-
tribute to this trend. The Sidama material provides a good
example. Historically, during the feudalist period (1893-1935
and 1941-1974), Amhara lords left the management of the
land to the Sidama people, provided that sufficient taxes
were paid. Sidama elders then regulated and partitioned off
areas of land for grazing. This was also a time when several
families would band together and take turns spending up to
a year in the lowlands with their cattle. During this period, it
was typical for a village to allocate 30 percent of the land to
crop production and 50 percent for grazing.
During the recent socialist period (1974-1991). farmland
was divided and parceled out. In the Derg period, village
land was reallocated so that 50 percent of the land was for
crop production and 30 percent of the land for grazing.
Areas that once were forests or grazing lands became farms.
The only land not parceled out was swamp land (chaffa).
These periodically flooded areas are now left for grazing
(McCabe and Lee, 1996).
Although people describe the grass in swamplands as
tough and poor for grazing, the maintenance of the swamps
as grazing land is, as one man put it, "worth fighting for-
Unless a farmer owns land in other areas, the chaffas near
one's farm and one's own yard are the only sources of graz-
ing. For example, the two swamplands in one area of
Sidama were four and five hectares, respectively Twenty per-
cent of households in one region have second farms, mostly
in order to have access to additional swamplands and other
grazing areas (McCabe and Lee, 1996).



I Will the decrease in livestock numbers
and fertility threaten the sustainability
of the enset cultivation system?
All the above factors contribute to the progressive down-
ward spiral in the livestock production sector of the rural
economy There may be a decline in total livestock numbers
in general, but there is a definite decline for individual
households because of increasing population and limited
land. This decline will have an impact on manure produc-
tion and the availability of draught animals. It could also
have an impact on human nutrition. The cycle of increasing
impoverishment of the livestock component in this mixed
crop/livestock system is a serious cause for concern. The
multiple purposes of livestock cannot be replaced by fertiliz-
ers, and the sustainability of the enset cultivation system is a
result of the tight articulation of the crop and livestock pro-
duction systems.
For example, in the Sidama zone the reduction of com-
mon grazing lands has forced farmers to tether their animals
in their front yards. Here, most households retain only a few
cattle, while the number of donkeys and small stock has
been greatly reduced as compared to the past. With an
increasing population in an already densely populated area,
it is likely that the negative trend in livestock populations
will continue, with potentially severe impacts on enset pro-



I Why is it important to consider gender
roles and the contribution of women?
Gender roles (in terms of the division of labor for all
aspects of enset production and marketing) are of critical
importance (Woldetensaye et al, 1997; Spring et al, 1996).
Without women to process enset, there would be no food
produced and it would simply be an ornamental plant, as it
is in other parts of Africa and Asia. But women's work is
often relegated to lesser significance than men's. Both
researchers and farmers often believe that women are
involved "only" in processing and cooking of the enset, and
rank these tasks below cultivation tasks. Women, in fact, do
participate (in some areas and in some households) in pro-
duction activities (e.g., manuring and varietal selection), and
in households where there are no women knowledgeable
about enset clones and processing, enset is not eaten unless
others are paid to process and cook it.
Women in wealthy households became labor managers
by hiring poor women to process and poor men to cultivate.
Women in middle income and poor households exchange
labor for processing. Men are believed to be banned from
enset processing areas, but were observed helping among
the Gurage. Locally, women market small amounts of kocho,
bulla, and amicho to obtain money for household consum-
ables (e.g., kerosene and salt). They strategize as to the
amount of surplus kocho and bulla they can sell off and still
have enough for the household. Both sexes sell non-food
enset products (e.g., leaves, mats, rope, and other construc-
tion materials). Men keep cash from the sales of cash crops.
By contrast, little is known about the ownership or remu-
neration received from the bundles or "jumps" of enset sold

for the market in Addis Ababa. Do wealthier women have
surpluses to sell? Do wealthier men plant extra gardens and
hire labor to process the plants? Is there joint decisionmak-
ing and profit-sharing between the sexes on planting and
processing? Additional research on marketing, both locally
and in urban areas, is required.

I Are there gender issues in clonal
variation, in both selection and usage?
Habte-Wold et al (1996) argue that women farmers know
a great deal about the different varieties of enset, and that
"when men and women of the same household were inter-
viewed together, women tended to dominate discussion
about varieties, contrasting and comparing them and saying
what should be harvested at different ages. Men played a
greater role in discussing the cause of 'drop-out' other than
harvesting men stressed the desirability of harvesting
at maturity and the varieties which are normally harvest-
ed later, whereas women were more concerned with a bal-
ance of varieties which can be harvested at different ages."
In addition to the gender division of labor, there are gen-
der issues concerning varieties selected for planting and
time of harvesting. Both women and men farmers categorize
the varieties of enset into two categories, each with different
characteristics, and they distinguish each clone in terms of
its "maleness" or "femaleness" (Habte-Wold et al, 1996;
Alemu and Sandford, 1996; Spring, 1996a; Tibebu et al,
1993). This categorization has nothing to do with the bio-
logical or reproductive parts of the plant, but with a set of
qualities and characteristics related to desirability, time of
harvesting, fiber and food content, softness and hardness,
palatability, length of fermentation period, size, growth rates,
and resistance to disease and pests. The so-called "male
clones" mature later, and are harder but give a larger yield,
while the "female clones" mature earlier, are softer, less
fibrous, and more delicious. Men have a preference for the
"male" enset, because they say "there is less temptation for



the women to harvest the plant before maturity for the sake
of eating the delicious boiled corm," (amicho) as in the case
of "female" plants (Alemu and Sandford, 1991). However, in
some regions farmers do plant more "female" than "male"
plants. Whether or not there are gender-specific reasons for
these choices or if women manage to prevail in their own
preferences needs to be investigated.
As a result, farmers, depending on their own circum-
stances (in terms of location, land holding size, and wealth),

strategize and maintain different numbers of clones by -sex
and age, with a slight preference for the "female" plants.
Data from Habte-Wold et al (1996) show that farmers tend
to plant a ratio of 56:44 "female" to "male- plants. Absence
of men from the homestead to engage in off-farm work is
believed to restrict varietal diversification, but the reasons
for this, and why the wealthier have greater clonal diversifi-
cation, are still unclear and require further research. In
terms of such research, there is often a tendency for
researchers to focus on yield as the major criterion, while
other variables may be of greater concern and factored into
varietal selection by the farmers themselves.



I What are "enset farming systems"?
In contrast to agricultural systems that describe the pre-
dominant crop and livestock mixtures, the term "farming
system" is technically determined inductively based on a
configuration of agro-economical zones and cultural prac-
tices in relation to agricultural activities, farm enterprises
(e.g., crops, livestock, agroforestry), and off-farm/non-farm
enterprises (e.g., wage labor, crafts and trade skills, business
enterprises) (Spring, 1995a; 1995b). Within the enset sys-
tems, variations in production, distribution, the types of
farm and off-farm enterprises, and farmers' management
practices of enset cultivation can be described and analyzed
at the household and group level. Other variations occur in
planting (spacing and timing), fertilization manuringg and
mulching), indigenous disease and pest control, nursery and
transplanting techniques and timing, sucker propagation,
harvesting and processing techniques, labor patterns, and
marketing practices.
In order to study the enset farming systems and their
variation at the ethnic group and household level, a number
of surveys and studies haev been carried out. In terms of
investigations of the farming systems, diagnostic surveys
using rapid rural appraisals (RRAs) have been carried out by
several groups of researchers. In several regions FARM Africa
has assessed the diversity of farmers, farming systems, farm-
ers' constraints, and potential solutions and research activi-
ties. Several surveys are specifically on enset (Alemu and
Sandford, 1991 and 1996; Bull et al, 1995), while Zippel
and Alemu (1995) present a field guide to enset clones for
North Omo. Other publications on enset clones and their

gender characteristics are Alemu and Sandford (1996);
Habte-Wolde et al (1996); and Sandford and Kassa (1994);
also see Abate et al (1996). Informal surveys on enset have
been carried out by the Institute of Agricultural Research
(Raya et al, 1988; Degu and Workayehu, 1990; Shiferaw
Tesfaye and Bizauyehu Haile, 1995), and more recently by
the Bureau of Agriculture, SNNPRG (Spring et al, 1996).
The following case studies of the Gurage, Hadiya, and
Sidama zones provide some examples of the variation in
enset farming systems at the ethnic group and household
One peasant association (PA) in each zone was carefully
selected to include altitudinal variation (low, mid, and high),
accessibility, significant enset production, and cooperation of
leaders and farmers. A rapid rural appraisal and additional
studies of 60 households were then carried out in the three
zones. The Gurage and Sidama have an enset-dominated
system with variation caused by differential resource levels
among households. The Hadiya, by contrast, have two dif-
ferent systems: one in which enset is dominant and one in
which cereal crops are dominant and enset is secondary. The
major cash crops in the areas studied are coffee for the
Sidama, chat (a stimulant) for the Gurage, and cereals and
eucalyptus tree for the Hadiya.



I What was found in the
Gurage Case Study?
The Gurage identify themselves as "people of enset"
(Shack, 1966), and are one of the ethnic groups that depend
upon enset as their main staple. During group focus ses-
sions, community leaders identified four wealth or resource
categories (rich, middle, poor, and poorest of the poor)
based on the amount of livestock, enset plants, cash crops,
and houses owned (Figure 10.1). All Gurage in the PA stud-
ied have an enset-dominated farming system, although there
are differences between wealthier and poorer households in
terms of the types and amounts of cash crops and enset, and
the management of enset (Figures 10.2 and 10.3). Wealthy
households grow large quantities of chat and some coffee for

Figre 0.2 Gurag- e IYf P)
Farming System^^^^^^^^^

Wealth Wealth Indicators %

Degene 10 or more cattle, may give some for 15
(Rich) share-raising to poor people
Self-sufficient in clothes, school fees, food
May use contract land in addition to own
Harvests up to 100 enset plants/year
Owns 3 houses
Hires labor during planting, weeding, and
processing of enset
Sells up to 1000 birr of chat
Harvests up to 500 kg of coffee
Owns about 3000 eucalyptus trees
Gibtose About 5 cattle (no share-giving) 35
(Middle) Self sufficient in food, clothing, schooling
May harvest up to 50 enset plants/year
Owns two houses
Zega No cattle but share-raising them 35
(Poor) Sells his/her labor
Harvests up to 10 enset plants/year
Sells hay from own land as an alternative
source of income
Own one house
Gurmasa Livelihood depends on others for payment 15
(Poorest in cash or kind for labor & food
of poor) May share-raise cattle for milk and manure
Harvests less than 10 enset plants/year
Very small plots of land
Owns a very small house
Source: adapted from Spring et al. 1996.

1. Cropping System
land allocation

enset plantation size
2. Enset System
general management

3. Livestock
use of livestock

4. Manure

5. Processing

matured plants used

storage period

labor for processing

6. Non-Farm Activities
selling labor
7. Diet
contribution to diet
food availability

8. Sales/Income

cash, while poor households mostly produce craft items
(such as baskets and pottery) to earn income. Gurage farm-
ers, who have surplus enset and live near transportation cor-
ridors, send bundles of kocho and bulla to Addis Ababa for
sales to urbanites.
Houses are grouped close together, and cattle are the
main type of livestock and are grazed communally as well as
in side-yards. Non-food enset products such as mats, bas-


Figue 1.1 ealt Rakin by ommnit


enset > all other crops (chat for
cash; no cereals)
larger for wealthy; small for poor

wide spacing, especially in last
wealthy have many more than poor

mostly dairy cows
income source,manure.
milk and meat

wealthy have adequate. poor have
on enset plus other crops

wealthy process less frequently
than poor
wealthy process more plants
than poor
wealthy store for longer, poor
store for shorter
wealthy hire labor,
poor exchange labor
(FHH have labor constraints)
no starter; initially laid flat

off-farm activity .e.g.. crafts.

enset is main crop
high for all groups. period
unavailable is short

enset. chat
wealthy have good income
plus savings


Fiur 103G rg -frzy A ne
MaaeetPatcsFr otnu

Fiur 10. Clne by Wealt Cateory

1. Clones
2. Transplanting
Nursery time
3. Planting Method
structured row planting
4. Plant Density

5. Weeding/Cultivation
6. Manuring/Fertilization
application methods

season of application
7. Leaf Removal
number removed
8. Intercropping
9. Diseases and Pests
bacteria wilt
sanitary measures

less diversified more diversified

3 years

4 years

used by all households

wide used by all households
(4 m by 4 m)

less well


all households use ring and
fixed year round


many (severe)



serious in all households
little some
absent present


Rich Middle

1. Yeshirakinke 75 86
2. Ameratiye 100 57
3. Yekesiwe 100 86
4. Astara 100 43
5. Gurarge 100 57
6. Agadie 25 57
7. Badediet 25 14
8. Lemat 75 14
9. Kanchiwe 50 29
10. Kibinar 50 29
11. Siniwet 25
12. Separea 29
13. Yegendiye 25
14. Yireqiye 25 14
15. Nechwe 100 71
16. Ankefrye 25 14
17. Charkima 25
18. Oret 25 14
19. Gimbua 14
20. Deriye 25
21. Toshet 25
Source: adapted from Spring et al. 1996.



100 87
43 67
71 86
29 62
29 19
86 86

kets (made by women and girls), and construction materials
(made by men) are ubiquitous. Because of enset, food avail-
ability is high even for the poorest of the poor, although the
poor lack dietary diversity A major difference between
households is cattle; wealthier households have large herds,
while poor households have none and have to "share-raise"
calves by borrowing an animal and returning products, off-
spring, or the animal itself.
Unique to Gurage, compared with the other ethnic
groups, is planting enset in strict rows with wide spacing,
up to four meters apart in each direction (Figure 10.3).
Other differences are that cattle owners have more manure
to apply and do so frequently; some farmers fence their
properties; and more conscientious farmers weed more fre-

quently. The frequency of transplanting and weeding have
more to do with farming skills and labor availability than
with wealth. Considering clonal variation, the wealthy have
many more clones or varieties of enset plants than the mid-
dle and poor/very poor households (Figure 10.4).
In the study area, bacterial wilt is endemic regardless of
resource level, unlike the Hadiya and Sidama areas (see
below), and farmers are not knowledgeable about prevention
measures. Further, the sale of entire fields of chat, the major
cash crop, to Addis Ababa merchants may contribute to the
spread of the disease, since these merchants cut enset leaves
to wrap chat, perhaps using contaminated knives.
A starter is not used for fermentation (unlike in Hadiya
and Sidama), but enset is laid on the ground to begin the



process. Kocho is kept for long periods, with pits being
changed periodically. A fungus affects long stored kocho.
Figure 10.5 shows that rich households process many
mature plants less often and their storage period is longer
than poor households. The latter process small quantities
more frequently, use immature plants, and quickly consume
their stored kocho. Wealthier households hire labor for pro-
cessing, while the poor exchange labor. Unlike some other
groups, Gurage men help women in certain aspects of pro-
cessing, and it is not "taboo" for men to be present during
enset processing.
Female-headed households face labor constraints in
many aspects of enset production, and are more likely to be
poor. Both female-headed households and male-headed
poor households are often involved in off-farm activity (e.g.,
crafts, selling labor), and their diets are less varied.

I What was found in the
Hadiya Case Study?
The Hadiya (Figure 2.2) grow enset in a system with
cereals (wheat and barley). Wealthier farmers grow and con-
sume more cereals than enset, while poorer ones lease bits
of unused land and sell their labor for cereal production,
but eat mostly enset products themselves. As a result, there
are two systems, termed here Hadiya 1 and Hadiya 2. In the
former, cereals predominate over enset, while the latter is
completely enset-based. Wealth and livestock ownership are
highly correlated (Figure 10.6), and community leaders
divide their residents into wealth categories based on live-
stock ownership, size of the enset holdings, cash and cash
crops, and housing. Ironically, they do not mention land
holdings as a criterion. Land size is larger than in the other
areas studied; wealthy Hadiya have much more land, as well
as larger enset fields, than poor Hadiya households. Also,
cultivation of annual crops and eucalyptus trees seem to
contribute to soil erosion.

Fiur 10. Nube of Plnt Processedan
Legt of0 Strg by0 Welt Categoy
Yeeey PA .5-



Total Nov Aug
Plants to Jan.

<1 year 1to 2


years years

Mean 73 61 12 50% 25% 25%
Mean 35 30 5 75% 25%
Mean 20 17 3 100%
Source: adapted from Spring et al. 1996.

Figure 10.6WealthRankingbyCommunity
Leaders, Ana Lemo PA, HadiIa

Wealth Wealth Indicators %
(Rich) 1 horse and mule 8
1 donkey
Pair of oxen
2 to 3 cows
5 to 10 sheep
Minimum cash 2000 birr
3 timad enset plantation
About 10 flowered enset plants in the field
Sell cereal crops and pulses
Lends money on credit
2 to 4 houses, 1 large and the others small
May have more than 1 wife
Produces wheat, barley, beans, and pea by
contracting land with the poor
(Middle) 1 equine 50
*1 ox
1 to 2 cows
1 to 3 sheep
Minimum cash 1000 birr
2 timad enset plantation
1 large and 1 small house
4 to 5 flowered enset plants in field
Produces wheat and barley
(Poor) 1 sheep, cow, or calf 42
No oxen
No large enset plants in field
Usually leases land to the rich
Sells labor to the rich
Source: adapted from Spring et al. 1996.



Figure 10.7 Farming Sytems Types
Ana Lemo PA, Hadiya I

1. Cropping System
land allocation

enset plantation size
2. Enset System
general management
enset propagation

clone diversity
3. Livestock
dairy cows

use of livestock

4. Manure
5. Processing
# of plants
matured plants used
storage period
labor for processing

6. Non-Farm Activities

7. Diet
contribution to diet
food availability
period unavailable
8. Sales/Income



cereals > enset


own fields

enset > cereals

get suckers
as gifts

2 to 4 oxen no oxen
none, acquire
calves by loan

no equines
income source
milk and meat
draft power
enset and crops

low (2/yr.)
many (30-50/yr.)
more and longer
hire labor

manure, but less
milk, but less
little manure
mostly for enset

higher (3 to 5/yr.)
few (5-20/yr.)
family and
exchange labor

None or few involved in off-farm
activity e.g., crafts,
selling labor

cereals > enset

cereals > enset

higher + savings

enset > cereals

no enset
products sold
cereals sold
low and no savings

The diversity and number of livestock is greater than in
the other areas studied. Because of cereals, oxen for plough-
ing are essential. In addition to cattle, rich households have
oxen, as well as horses, donkeys, mules, and sheep. Poor
households have few animals and have to "share-raise" a
cow or calf loaned by the rich. They might rent/sharecrop
their small land parcels that are not planted in enset to rich-
er farmers for cereals production. The poor also sell crafts
and their labor for food and cash (Figure 10.7).
Figure 10.8 shows that Hadiya I farmers transplant more
frequently, and have wider spacing (some using row planti-
ng, as opposed to random planting, which is the norm), bet-
ter management, manuring, and disease control measures.
Bacterial wilt is a problem for many farmers, but the wealthy
have the choice of discarding, processing, selling, or giving
away diseased plants. Poorer households often are the recip-

FigurHOe510.f Uariy Ia (AnaLmoP 1A):SJystem'
iffesin M

1. Clone

2. Transplanting
(3 to 4 times)
3. Planting
4. Plant Density
5. Weeding/
6. Manuring/

7. Leaf Removal
8. Intercropping

9. Diseases
and Pests

more plants are
more clones
are available
(2 times)
some use row
use wide spacing
better managed

well manured
ring application
season specific
fewer removed

(bacteria wilt in 50%)
sanitary measures:
better awareness
fencing present
can uproot and
dispose of
bunds around plants

fewer plants are
limited number
of clones available
less frequent

random planting

use closer spacing
weakly managed

poorly manured
broadcast application
year round
more severe removal
some intercropping
because of land
(bacterial wilt in 50%)
poorer awareness

no fencing
does not uproot or
dispose of
no bunds



ients, which could contribute to the spread of the disease on
their farms. Animals pests also attack enset, but richer farm-
ers build bunds around the plants. The rich have more enset
clones (Figure 10.9) than the middle and poor wealth cate-
Women in wealthier households are enset farm managers;
they pay poor women to do their processing. A starter
(gamama) is used in processing enset. Figure 10.10 shows
that the rich and middle groups process less often (two to
three times per year) using large quantities of plants, while
the poorer households process small and more immature
plants more frequently. None of the poor store kocho over a
year, and these households are frequently short of food.

Figure 10.9 Cones by Weal h Categry, An

* F-ir 11 Num of PaSntsProcessed
and Le~ngth of StoragebyWealtih Category,
Yeferezye PA, 0 0 *.0r

Wealth Total Oct. Aug
Category Plants to Jan.

to June


<1 year 1 to 2 >2
years years

Mean 43 31 7 5 40% 60%
Mean 51 29 12 10 34% 66%
Mean 22 10 8 5 100%
Source: adapted from Spring et al. 1996.

Clone Name
1. Gimbo
2. Siskela
3. Shate
4. Gishira
5. Agade
6. Sapara
7. Oniya
8. Unjama
9. Disho
10. Kaseta
11. Astara
12. Sormanicho
13. Torora
14. Tebuta
15. Bedadeda
16. Zobra
17. Hayiwona
18. Woshamada
19. Gariya
20. Mesmesicho
21. Kembotra
22. Mariye
23. Merjia
24. Merza
25. Mandulk
26. Bosina
27. Ashamosa

Rich Middle Poor/ Mean






20 25




86 95
86 87
29 45
57 62
29 38
14 38
25 81
14 25
29 10
14 5
29 61





Source: adapted from Spring et al. 1996.



Figr 10 1W at akigb o mnt


Wealth Indicators

Duresa Land size more than 1 ha
(Rich) Land use very efficient
4 cows
2 sheep
1 donkey
Housing type large and well managed
Plants 800 to 1000 suckers per year
Has all stages of enset that is well
Harvests more than 4.8 qts = 480 kg
of unhulled coffee in addition to
amount for household use
Mererima Land size = 0.5 ha
(Middle) Land use efficient
2 cows
1 sheep
1 donkey
Housing type medium
Plants 400 suckers per year
Has all stages of enset that is
well structured
Harvests more than 3.6 qts = 360 kg of
unhulled coffee in addition to amount
for household use
Buticho Land size < =0.25
(Poor) land use inefficient
1 or no cow
House type small and poor quality
Plants 50 suckers
Does not have all stages of enset
Enset plantings are not well structured
Harvests only enough coffee for
household use

Source: adapted from Spring et al. 1996.

I What was found in the
Sidama Case Study?
Similar to the Gurage, the Sidama system is entirely enset
based, and there is some variation between households. Farm
size and livestock numbers in general are smaller than in the
other areas studied, but wealthier households have larger, more
diversified i-.1J l-l" and cash crops. Enset is planted randomly
and often intercropped with coffee, vine crops, and fruit trees.
There is no erosion because there are neither ox ploughing nor
annual crops, and the enset fields are well mulched and
manured. This area seems to be the most innovative in its adop-
tion of new technologies, as a result of farmers being organized
into coffee cooperatives and having steady incomes. Community
leaders estimate that 20 percent of households (compared to 15

1. Cropping System
Enset-dominated (diet, land allocation)
Coffee as cash crop (next to enset in land allocation)
Fruit trees, sugar cane, chat for supplementary cash
Maize, haricot bean, yam taro for supplementary food for rich
2. Enset System
Clones wider diversity in rich lower diversity in poo
Stage of enset all in rich HHs some in poor HHs
Management rich: better poor: worse cultivation



cultivation in rich

3. Livestock

4. Manure


milk, manure, sale
own cows, sheep, goats, calves
few or none, acquire from rich
by loan for manure

Type rich: inorganic poor: manure
and manure
5. Processing/Harvesting/Storage
Frequency 2 times/year, short time
/storage: long time
Labor hire or exchange labor household only
Pits many and wide few
6. Food Availability
Rich HHs wider range and longer period (Aug. to April)
Poor HHs narrower range and shorter period
(Aug. to Jan.)
7. Diet
Enset dependent for all HHs, but rich can supplement with cereals
8. Sales/Income
Rich high-coffee (dried), fruit trees, sugar cane,


9. Expenditui


livestock, chat
low for coffee (fresh), non-farm and off-farm
diverse foods throughout the year, clothes
purchase food and earn income for shorter
high for rich low for poor

percent for Gurage and 8 percent for Hadiya) are in the rich cat-
egory (Figure 10.11). The number of livestock is declining on a
per household basis as grazing lands became scarcer because of
increasing human population. Cows, the main livestock type,
are tethered in front of the house. Poor households are cattle-
deficient and might borrow a cow for manure and dairy prod-
ucts. Wealthier farmers are purchasing inorganic fertilizer (DAP)
to make up for cattle manure deficiencies. Although enset is still
the preferred food (Figure 10.12), households also supplement
their diets with cereals (usually maize) and other foods.


Figure 10 .1 E amigSsemi o
Badgallo PA, HHSidama^^^^


Figure 1 3 Mn S m F
aW* Coninuml n an VaUmmep~'r.iablesI,I.Boa:I.1ugillFI

1. Clones
number and types diver
Plantation size large
2. Transplanting
Frequency 1 anc
3. Planting Method
planting type all ha
4. Plant Density all ha
5. Weeding/Cultivation
management good
6. Manuring/Fertilization
application one s

7. Leaf Removal
8. Intercropping
9. Diseases and Pests
bacterial wilt
awareness of sanitary

and b

sified less diversified

12 times

Ive random, unstructured
ve close spacing


ide application
manure inorganic (DAP)



some crops all crops are

less serious for all

little some
well fenced not well fenced

F [Igure10.4gCopnes by Wealth Categoryof
Ho sehods B .oa S 0alg A S.dm (, 1,cI

Clone Name
1. Ado
2. Genticha
3. Midasho
4. Ewasho
5. Gullumo
6. Mindraro
7. Agana
8. Chacho
9. Medie
10. Birra
11. Derassie-Ado
12. Gossalo
13. Gedimie
14. Gena
15. Hekechie
16. Haho
17. Gerbo
18. Kitickha
19. Siriro
20. Arisho

Rich Middle






Poor/ Mean
100 100
100 1 00
67 82
17 12
17 12
17 12

33 31
17 19
17 19

S 12


Source: adapted from Spring et al. 1996.

Enset management practices are correlated with wealth
categories, farming skills, and availability of resources, and
there is variation in intercropping, fencing, and manuring
(Figure 10.13). This zone has the lowest incidence of bacter-
ial wilt, and farmers seem to know about its spread by cont-
aminated tools. There is little erosion, even on farms planted
on slopes. Unlike other enset growing areas, farmers pur-
chase enset suckers from highland farms rather than using
their own. The rich have more clones than the middle and
poor farmers (Figure 10.14).
A starter (gamacho) is used to ferment kocho, and many
farmers put the storage pits inside their homes to prevent
theft. Women are beginning to use iron scrapers instead of
bamboo ones for decortication and cloth squeezers for bulla.

I What conclusions can be drawn about
systems variation from the case studies?
Between ethnic groups where enset is a staple (Gurage
and Sidama) and a co-staple for some of the population (the
wealthier households among the Hadiya) there are differ-
ences in many aspects of enset cultivation clonall variation.
plant spacing, disease prevalence, and manuring) and pro-
cessing (tools and starter used, location, size, and disease of
pits). The gender division of labor varies between groups
and households, and there are differences in the mix of
farm, off-farm, and non-farm enterprises.
Most differences in enset systems have been attributed to
altitude. These case study data show that within an area.
variation can be found based on household resources and
farm enterprises, rather than only on altitudinal differences.
Wealthier households have resources to maintain self-suffi-
ciency, educate their children, vary their diets (changing to a
cereal-based diet, if they chose), hire labor for farming and
processing, and build many houses. They cope better with
enset and livestock diseases because of diversification of



clones and livestock. Poorer households lack clonal varia-
tion, are dependent on the richer ones for work and live-
stock to share-raise, and consume only enset-based diets
with few protein sources. They work for others, have little
cash or consumer goods, and are vulnerable to disease and
Land holding size is another key element determining the
amount of enset and other crops planted, as well as available
grazing land for livestock. The rich have the largest amount
of land and enset plantations per household, followed by
smaller amounts for the middle, and very small amounts
(often only enset fields) for the poor and very poor. Land size
strongly correlates with wealth, although community leaders
list livestock, enset plants, and cash as the indicators.
Wealthier households have greater clonal variation, as
well as having more mature and a larger number of enset
plants (future research will determine if clonal variation is
an advantage or something like a status symbol that the
wealthy can better afford). They have other income sources,
a more diverse diet, and an obligation to help the poor by
giving them livestock to "share-raise." They process larger
numbers of enset plants infrequently, and do not experience
famine, unlike poorer households, which have shortages

from time to time. Women in wealthier households reduce
the drudgery of enset processing by hiring labor. Poor
women have the double burden of working and processing
on their own farms and selling their labor for such tasks to
wealthier households.
There are no technical packages or extension advice on
enset production being promoted to farmers. But innovation
and intensification are occurring in small ways, particularly
among the Sidama, where increasing population growth
produces severe land shortages (even to the point of using
grazing lands for settlement and enset cultivation). Sidama
farmers with cash from coffee sales are purchasing inorganic
fertilizers, and women are changing the location of pits and
adopting an improved scraper. Gurage farmers are exploiting
increased interest in enset by urban dwellers, by sending
surpluses to Addis Ababa.



I How do enset-based farming systems
contribute to food security in Ethiopia?
Enset-based farming systems play an important role in
food security in Ethiopia. The exact role and value relative
to other farming systems cannot be addressed without
examining enset production and consumption in relation to
the concept of food security. Food security can be explained
in terms of: 1) adequate availability of food in line with pre-
sent population and demographic growth; 2) the nutritional
adequacy of food intake; 3) annual stability of the food sup-
ply; 4) access to food (through production or the market)
(Brandt, 1990; Webb and von Braun, 1994; and FAO,
1996); and 5) the sustainability of the food production
capacity over the long term. Each of these five features relat-
ing to food security is discussed briefly
Some of the most dense rural populations of Ethiopia are
located in regions practicing enset-based farming in the
southwestern highlands. Rahmato (1996) notes that among
the Wolayta, as landholding size declines, there is an
increase in the cultivation of enset. These observations indi-
cate that the human carrying capacity (i.e., the number of
people per unit of land area that can be adequately fed by
the food produced on the same land area) of enset and
enset-based farming systems is high and is likely greater
than other crops and cropping systems for the same agroe-
cology and inputs.

I How does the quantity and quality of
human food produced from enset in
enset-based systems affect potential
human carrying capacity as compared
to other systems?
Although enset-based farming systems seem to support
higher population densities than other farming systems, it is
difficult to compare these systems, because of a lack of
quantitative research data. The human carrying capacity of
enset-based agricultural systems is more difficult to quantify
than systems based on annual cereal crops for at least four
reasons: 1) enset yields are difficult to determine and have
not been quantified; 2) enset food products have a low,
inadequately-verified, protein content with an unknown
amino acid distribution; 3) enset's low protein content
necessitates that the protein contribution from associated
foods be more diligently considered; and 4) nutrient cycling
among enset fields and other fields are not yet evaluated.
Enset yields are difficult to measure and evaluate because:
1) plants are grown for multiple and variable numbers of
years; 2) the spacing of individual plants may be changed
several times; 3) enset may be grown in complex mixtures
with other species, as well as other enset clones and other
sized enset plants; 4) the weight gain of food in an enset
plantation for a year may not be the same as the amount
harvested by the farmers during that year; and 5) in addi-
tion to human food, there are many other enset products
obtained from each plant. Also, the huge volume harvested
from one plant and from an area, particularly in relationship
to cereals, contributes to the perception among both farmers
and scientists that the yield of enset is tremendous.
However, in reality, the content of water, energy, and pro-
tein, the area and time use by the plants, as well as other
aspects must be considered in order to interpret the actual
food yield from this huge volume. Box 11.1 and Figure 11.1
provide an example of the complexity of evaluating enset
yield. This example shows that the average annual yield of
34 farms was 5,000 kilograms of kocho per hectare, in addi-
tion to other products that were not measured, such as fiber
and animal feed.
Yield and human carrying capacity of enset and annual



crops under the same conditions have not been compared.
In such a comparison, at least two considerations must be
made. First, enset usually grows in regions with a long
growing season, commonly nine months. In the environ-
ments where it is possible to double crop and get two annu-
al crops, yield and human carrying capacity of enset should
be compared to a sequence of two crops. Second, compar-
isons of human carrying capacity should consider the abili-
ties of the systems to supply the nutrients, particularly ener-
gy and protein, required by humans.
The importance of considering the requirements and sup-
plies of both energy and protein in determining human car-
rying capacity are illustrated in Box 11.2 and Figures 11.2
and 11.3 with a comparison of two hypothetical cropping
systems, i.e., 1) enset and dry bean and 2) maize/sweet
potato and dry bean. To simplify the example, only crops
are included; but the relevance of consuming high-protein
animal food products with low-protein kocho is apparent.
From the comparison of enset and bean with maize/sweet
potato and bean (Box 11.1 and Figures 11.2 and 11.3), it

can be seen that comparing just the yields or energy content
of enset with maize/sweet potato is inadequate for determin-
ing the ability of the crop to support dense human popula-
tions. Even at the highest protein content, the enset, at
5,000 kilograms of kocho per hectare per year, and bean
support 15.5 adults per hectare, while the maize/sweet pota-
to, at 4,000 kilograms per hectare per year, and bean sup-
port 18.2 adults per hectare. Another concern is the amount
of bean required in the enset diet. The required bean con-
sumption of 53 to 71 kilograms per year is two to three
times greater than the amount provided by a typical African
diet, which is 10 to 26 kilograms per year of pulses plus
groundnuts (Aykroyd et al, 1982). Thus a diet with a large
proportion of enset may require the addition of a higher
protein source than bean, which is why high-protein animal
food products are so important in this system. A serious
concern in enset producing regions, is that as population
density or poverty increases, the opposite may be occur-
ring-consumption of kocho increases while consumption of
animal products decreases.

I 0 x 1 1 E es of ye c a wt r e to w t c t us of ae ae an tm

Using kocho weights, from a sample of five plants from each of 34 farms (17 near Emdibir, 10 in Kambata, and 7 in
Sidama) (Makiso, 1976), yield is calculated (Hiebsch, 1996) as dry weight per unit area per unit time [e.g., kg/(ha yr)].
based on the farmers' transplant management presented by Makiso (1976) (Figure 11.1). In this three transplant system,
suckers develop during the first year from the mother corms; the suckers are transplanted (1st transplant) to a 1.0 m X
0.5 m spacing for one year; then the plants that are still alive are transplanted (2nd transplant) to a 1.5 m X 1.5 m spac-
ing for two years; and then transplanted (3rd transplant) to a 2.5 m X 2.5 m spacing for four years; for a total of eight
years. Personal observations in similar locations indicate that the plant management in Figure 11.1 is stylized and sim-
plified, as there is variation among the locations, farms, and plants on a farm and some enset is intercropped.
The example in Figure 11.1 is designed to provide most of the food energy required by a family of five to six from
hocho. Although fermented kocho contains about 50 percent water, the calculated yield is for dry, i.e., waterless, kocho,
since water does not provide energy Each year, 80 eight-year-old plants, which utilized a total of 2,455 m2/yr as they
passed through the eight stage/years, are harvested from the 4th-year, 3rd-transplant plot of 500 m2. In order to harvest
80 plants each year, and continue on into the future, all eight stage/years must be present each year, thus requiring
2,455 m2 each year, i.e., 2,455 m2-X-1 yr = 2,455 m2 yr. Thus, the same area-X-time is required for the 80 harvested
plants during their eight years as for all stages/years in a given year. Therefore, yield is the dry weight of kocho produced
by the 80 plant divided by the total area-X-time required to produce them, i.e., 2,455 m2 yr. Based on these estimates of
water content, spacing, and timing, the average yields for the 34 farms was 5,000 kg dry kocho/(ha yr). Yields from high-
ly managed research plots (Bezuneh, 1984) range from 5,900 to 9,500 kg of dry kocho/(ha yr).



I How does the quantity and quality of
animal feed produced from enset in
enset-based systems affect potential
human carrying capacity as compared
to other systems?
The low-protein portion of an enset plant is eaten by
humans and the high-protein portion is either recycled to the
soil, used as a wrapping material, or fed to animals. Thus the
entire cycling of protein through other components of the sys-
tem, particularly animals, has a greater impact on human nutri-
tion and human carrying capacity than in cereal-based systems,
in which the high-protein portion is eaten by humans.
Human food from mature enset plants comes primarily
from the corm and an extracted pulp from pseudostem leaf
sheaths. Together the corm and leaf sheaths have 0.037 kilo-
grams of protein per kilogram of dry matter (Fekadu, 1996).
The remainder of the plant, which is mostly leaves, is about


26 percent of the plant, and contains 0.160 kilograms of
protein per kilogram of dry matter. Therefore, both the pro-
tein content and the total amount of protein is greater in the
portion not eaten by humans. The recycling of all these
products (non-human portions, human foods, and animal
manure) all have important consequences for the human
carrying capacity of the system. By comparison, the stems
and any remaining leaves of cereals and other tuber crops
that are left for animal feed are usually of low protein value
and in some cases are unacceptable as animal feeds.
Nitrogen, which is 16 percent of the protein, is often the
most limiting chemical element in a farming system. In the
enset system, the larger portion of the nitrogen does not
pass directly from the enset plant to humans. Rather, it is
cycled through animals. Therefore a quantitative under-
standing of the cycling of nitrogen in the enset system is dif-
ficult to measure, but important.

maaemn syte of rereene wit an anua havs of 80 plns tha 0i0s
sufficiet to suply the fod energ for a fve to si person ousehold

Year Area-X-time
Spacing for for each
for a No. of each plant stage/year t
Stage t stage total Plants ( (m X m) (m2 yr)
------ stage/year ---------
Sucker nursery 1 1 5 1.0 X 1.0 5
1st transplant 1 2 100 1.0 X 0.5 50
2nd transplant 1 3 89 1.5 X 1.5 200
2 4 89 1.5 X 1.5 200
3rd transplant 1 5 80 2.5 X 2.5 500
2 6 80 2.5 X 2.5 500
3 7 80 2.5 X 2.5 500
(harvested plants) 4 8 80 2.5 X 2.5 500

Total area-X-time 2455

SYield ( 30.6 kg fresh kocho 0.5 kg dry kocho \ 80 plants = 0.5 kg dry kocho
l plant 1.0 kg fresh kocho 2455 m2 yr m2 yr

Yield =( 0.5 kg dry kocho 10000 m2 5000 kg dry kocho
S m2 yr / ha / ha yr
SThe 80 plants harvested at the end of 8 years (4th year of 3rd transplant) went through all 8 stage/years;
all 8 stage/years are present each year so that 80 plants can be harvested per year into the future.
SAssumes that 5 mother corms in the sucker nursery produce 100 robust seedlings; and assumes
approximately 10% loss of plants at each transplanting.
Source: Makiso (1976) and Hiebsch (1996)



I How does enset contribute to the
stability of the annual food supply
and reduce food shortages, particularly
during drought years?
The presence of enset in the farming system contributes sig-
nificantly to the stability of the food supply by several mecha-
nisms. Enset can: 1) be stored for long periods; 2) be harvested
at any time during the year; 3) be harvested at any stage over a
several year period; and 4) survive stress years that reduce other
food sources. It could even be argued that since enset requires
from three to over ten years to mature, the frame of mind
required to produce enset contributes to a general prepare-for-

the-future mentality, which has other behavioral consequences.
As described in the section on processing, kocho is stored
in nearly-airtight "containers" (i.e., pits), in a fermented
state, which greatly retards loss. Farmers report that kocho
may be kept for several years in this way It is important to
note, however, that only the wealthier households may actu-
ally store kocho for more than one year (Spring et al, 1996;
and above); this product may be a status symbol analogous
to an aged wine for special occasions.
Mid-season food shortages can be alleviated because
enset can be harvested at any time during the year. If kocho
is going to be prepared, then the farmer must plan ahead by

Bo 1. 2 C n to oh c

The objective is to determine how many adults can be supported on one ha of land based on energy and pro-
tein requirements and production. The two systems are (1) enset and a double crop of two dry bean (Phaseolus
vulgaris) crops, and (2) a double crop of maize followed by sweet potato and a double crop of dry beans. The
enset used in this example is illustrated in Figure 11.1 and has a kocho yield of 5,000 kg/(ha yr). In support of an
untested hypothesis that enset has higher yields than a double crop, the maize and sweet potato are assumed to
yield 2,000 kg each of dry edible food/ha, for a total of 4,000 kg/(ha yr), or 1,000 kg/(ha yr) less than enset. The
diet from system (2) has equal portions of maize and sweet potato. The yield of the bean is assumed to be 800
kg/ha for each of two crops per year for a total of 1,600 kg/(ha yr).
The requirements for an adult human used in this example are (2,200 kcal/day) X (365 days/yr). or approxi-
mately 800,000 kcal/yr for energy (FAO/WHO, 1973) and (0.050 kg/day) X (365 days/yr), or approximately 18
kg/yr for protein (NRC, 1989). Since the energy contents of kocho, maize and sweet potato, and bean are similar,
the energy yield and the number of adults that could be supported based on energy are closely related to the food
yield, e.g., enset produces 19,000,000 kcal/ha and has enough energy to support 23.8 adults/ha (Figure 11-2).
Protein yield, however, produces quite a different picture. Reported kocho protein content varies by more than
three-fold from 0.012 to 0.037 kg of protein/kg of dry kocho (by comparison, cassava is 0.030 and maize is 0.108
kg of protein/kg of dry food). Although with most foods, protein content has been verified repeatedly, with kocho it
is not known whether the variation reported is due to actual protein content differences or to the laboratory tech-
niques. The amount of protein produced by the crops is quite variable, with enough for 3.3 adult/ha at the 0.012
kg of protein/kg value for kocho and 20.4 adults/ha for beans.
In either of the kocho-based or maize/sweet potato-based system, a proper quantity of bean can be combined to
provide the required energy and protein. These combinations for kocho at the three protein levels and for
maize/sweet potato are indicated under "weight of food" in Table 11.3. For example, at a protein content in kocho
of 0.029 kg/kg, 150 kg of kocho and 59 kg of bean will provide the required 800,000 kcal and 18 kg of protein-
The land areas required to produce those quantities of carbohydrate-rich food and bean are indicated under -land
required to feed 1 adult." For example, 0.030 ha of land is required to produce 150 kg of kocho and 0.037 ha to
produce 59 kg of bean, for a total of 0.067 ha/adult. Potential human carrying capacity (adult/ha) is the inverse of
ha/adult and ranges from 13.9 for the lowest protein value for kocho to 18.2 for maize/sweet potato. In order for
enset and bean to have the same human carrying capacity as maize/sweet potato and bean (i.e., 18.2 adults/ha).
the yields of kocho would need to be between 7,200 and 13,100 kg/(ha yr), as compared to 4,000 kg/(ha yr) for
maize and sweet potato combined.



F igure 1. N i c yield [ n suu ply
of dr koh fro 0 net maze swe poao an bean,0 0

Nutrient content '
Energy Protein hydrate Fat
kcal/kg ------ kg/kg of food --------

Nutrient yield :

Energy Protein
Xcal/ha kg/ha

Number of adults that
can be supplied with

Energy Protein
-------- adult/ha -----

rich food

* maize
* sweet potato
* average of maize
& sweet potato

Protein-rich food
dry bean

3,800 0.012 t 0.96
0.029 t
0.037 t
4,100 0.108 0.83
4,000 0.052 0.93




0.080 0.88 0.03

3,900 0.230 0.72 0.01

19.0 60

16.2 320

6.2 368

23.8 3.3



7.8 20.4

Reported protein contents of kocho are, in kg of protein/kg of dry food, 0.012 (ENI 1981), 0.029 (Pijls
1994), and 0.037 (Besrat 1979); other nutrient contents of kocho (ENI 1981); nutrient contents of other
crops (various sources).

Nutrient yield for energy and protein is based on dry food yields of
(1) 5,000 kg/(ha yr) for kocho,
(2) 4,000 kg/(ha yr) for a double crop of maize and sweet potato combined, and
(3) 1,600 kg/(ha yr) for two crops of dry bean that are double cropped.

S Nutrient yield is calculated by multiplying dry food yields by nutrient content, e.g.,
kg of food/(ha yr) X kcal/kg of food = kcal/(ha yr), thus
5,000 kg kocho/(ha yr) X 3,800 kcal/kg of kocho = 19,000,000 kcal/(ha yr).
Units for annual energy yield, indicated as Xcal/ha, are millions of kcal/ha.

Number of adults (adult/ha) that can be supplied with energy or protein is based on either:
(1) an energy requirement for one adult of (2,200 kcal/day) X (365 days/yr) for approximately
800,000 kcal/yr or
(2) a protein requirement for one adult of (0.050 kg/day) X (365 days/yr) for approximately 18
kg/yr. Since amino acid content and protein digestibility in kocho are not known, to prevent under-
estimating protein requirements, minimum protein requirements of 0.030 kg/(adult day) reported
by FAO/WHO (1973) are not used in this example.

Number of adults (adult/ha) that can be supplied with energy or protein is calculated by dividing
nutrient yield by the energy or protein requirements of an adult, e.g., based on energy from kocho,
[19,000,000 kcal/(ha yr)]/[800,000 kcal/(adult yr)] = 23.75 adult/ha.





Fiur 11. Weiht of (1 0.h an bean or (2 maz/we poat and bean

necessary ~ ~ to proid th adl rqie ntof8000 .0. ad1 k roen


"carbohydrate food"

"carbohydrate food"
bean (2 crops/yr)

"carbohydrate food"

"Carbohydrate Food" (kg protein/kg dry food)

maize &
kocho sweet potato

(0.012) (0.029) (0.037) (0.080)

Weight of Food (kg food/yr)W
138 150 156 184
71 59 53 14
209 209 209 198

Assumed Yields [kg/(ha yr)]
5000 5000 5000 4000
1600 1600 1600 1600

Land Area Required to Feed 1 Adult (ha)9
0.028 0.030 0.031 0.046
0.044 0.037 0.033 0.009
0.072 0.067 0.064 0.055

Potential Human Carrying Capacity (adult/ha)*





Yield of Kocho Required for a Potential Human
Carrying Capacity of 18.2 adults/ha [kg/(ha yr)]#
13,100 8400 7200

SRequirements: (2200 kcal/day)(365 day/yr) = 800,000 kcal/yr (et in );
(0.050 kg protein/day)(365 day/yr) = 18 kg protein/yr (pt in T)
' Weight of the carbohydrate food (c) and bean (b) solved by simultaneous equations:
et = ecc + ebb and Pt = pcc + pbb, where e is the energy (kcal) and p is the protein required in total (t)
and based on the contents in the "carbohydrate food" (c) and bean (b); the total (et or pt) is the sum
of energy or protein from c and b.
Land area (a): ac = c/yc, ab = b/yb, at = ac + ab, where y's are the assumed yields.
Human carrying capacity (HCC): HCC = l/at.
SYield (y) of kocho (k) required to have HHC equal to maize/sweet potato and bean (HHC2) i.e.,
18.2 adults/ha: yk = ekHCC2/(1.0 MCC2b/yb).



about one month as this is required by the fermentation
process. If the plants are harvested for amicho, they may be
used immediately. Because of the storage and harvest-timing
characteristics of enset, if a farmer has enough enset plants,
there is no "hunger period" as is common in cereal farming.
The last two mechanisms, which work together, are proba-
bly more important for adding stability to the year-to-year food
supply. Enset can be eaten at any stage of growth, over a sever-
al-year period after the corm reaches about 10 to 15 centime-
ters in diameter. Under good growing conditions, this condition
may occur during the first transplant stage. There is great varia-
tion among ethnic groups as to both the acceptability and the
practice of harvesting young enset plants. In years when other
foods are in short supply, usually caused by drought, more
enset plants may be harvested than was originally intended. For
example, in the system described in Figure 11.1, the harvest
may include the 80 plants in the fourth year of the third trans-
plant stage, as well as any number of younger plants. However,
the younger plants harvested will not be available to harvest in
the future. At the beginning of the next growing season, the
farmer will likely need to implement a strategy to recover. The
enset plants themselves may also contribute to the recovery,
since the remaining enset plants in the "prematurely" harvested
field may grow faster during the next season because of lower
plant density and reduced competition. Although these last two
mechanisms may provide great stability to the food supply in
an enset-based farming system, no research has been conducted
either on the effect of "early" harvest of enset on present and
future food supplies or on strategies implemented by farmers
that are facilitated by these enset characteristics.
In enset producing regions, no matter how small the land
holding, enset is grown. Even families referred to as "land-
less" have a house with enset around it. Enset serves the
multiple purposes described above, as well as providing a
dependable food source.

I How does enset contribute to the
sustainability of food production?
The ability to provide a long-term, sustainable food sup-
ply, with minimum off-farm input, is probably the most note-
worthy characteristic of enset, and is a primary motivation
for this publication and current interest in enset. An obvious
and principal contribution to sustainability is the minimal
soil erosion involved in enset's cultivation. Enset provides a
perennial leaf canopy over the soil and a heavy mulch cover
from leaf litter. Soil erosion is not seen in enset fields. This

situation is in stark contrast to fields of annual crops, partic-
ularly at the beginning of the rainy season when there is no
soil cover by the annual crop. The perennial leaf canopy also
may reduce maximum soil temperatures and, thereby,
decrease organic matter decomposition rates. There are no
research data to support these common visual observations
of reduced soil erosion.
A curious aspect related to soil erosion is that enset is
most commonly planted around the house, and the house is
usually on the most level location on a farm. If there is slope
variation on a farm, annual crops commonly occupy steeper
fields than enset. Therefore the soil erosion observed in
enset versus annual crops is not just related to the crop.
There is a need for research comparing enset and annuals on
sloping fields. If suspicions are confirmed, extension activi-
ties need to be implemented to encourage more production
of enset on erosive locations.
Leaching losses of plant nutrients, particularly nitrogen,
may be reduced by enset as compared to annual crops. This
should be possible because of the continuous soil occupa-
tion by roots. At the beginning of the rainy season and after
maturation, annual crops have little root proliferation and
little affect on nutrient leaching. For established enset, roots
already proliferate the soil profile at the beginning of the
rainy season. Also the large mass of the plant should serve
as a storage reserve, reducing the availability of the nutrients
in the soil for leaching.
The main negative feature of enset, its low protein con-
tent in the human-food portion, may contribute positively
to its sustainability Soils are depleted with continual
removal of crop products. This is common with off-farm
sales or with on-farm consumption without recycling of
waste products, including human excrement. Removing
low-protein (low-nitrogen) kocho from a site should have
less impact on the nitrogen status of the soils than removing
cereal crops, as long as the high-nitrogen portion is cycled
within the farm.
It is common to find enset fields that have been produc-
tive for decades. The mechanisms that allow this long-term
high productivity with minimum external inputs need a
great deal of future study with the objectives of improving
the enset system and transferring components of its success
to other systems.



I Who will benefit most from greater
knowledge of and improvements
in the enset systems?
The most directly and significantly affected stakeholders
to benefit from accelerated research and development activi-
ties related to enset systems are the subsistence farm family
and the local communities. Several characteristics of the
enset plant and systems described above, e.g., a large stored
food supply available when other foods are in short supply,
are particularly valuable for subsistence, low-resource farm-
ers living in a highly variable environment. However,
Ethiopian society as a whole also benefits because of the
preservation of natural resources through wise management
of the enset systems.

I What is the potential of enset products
playing a larger role in the diet of the
urban populace?
During the last two decades, two things have happened
to make enset food products significantly more popular
among the urban populace of Addis Ababa and surrounding
communities. First, the grain markets have experienced a
considerable increase in the price of cereals, while the price
of kocho has remained relatively constant. Urbanites shop-
ping in the markets of Addis Ababa, especially those from
enset-growing regions, are choosing to purchase enset prod-
ucts both for taste and to make their limited incomes go fur-
ther. Although considerably more research needs to be done
on the marketing and pricing of enset products, a cursory
survey of sellers at the main market, Mercato, revealed over
120 women sellers of kocho and bulla.
Second, there has been a breakdown in the cultural per-
ception of enset food products as "peasant food." As previ-
ously mentioned, kocho has become extremely popular at
restaurants and is almost "required" to be eaten together
with the Ethiopian delicacy of kitfo (raw, ground beef mixed

with butter and spices). Informal interviews and observa-
tions at a sample of Addis Ababa restaurants indicate those
establishments that specialize in often run out of high quali-
ty hocho due to poorly developed enset marketing and trans-
port systems. All of this suggests there could be considerable
opportunity for supplying and increasing the demand for
enset in urban markets.

I What is the potential of enset
cultivation being introduced or
re-established in regions outside
its main area of use?
In a study of agriculture in the former Illubabor region of
southwestern Ethiopia, the Sombo peoples, who traditional-
ly were cereal farmers dependent upon t'eff and maize,
experienced two starvation periods (Ishihara, 1993). The
second was in 1984-85 when the peasants migrated from
their villages in search of food, a considerable number of
them dying on the way Some traveled as far as Wolliso,
where they learned how to cultivate enset. Returning to
Sombo, they introduced enset agriculture, and kocho soon
became an important part of their diet. In 1992, when cere-
al crops were severely damaged by excessive rainfall, they
lost 50 to 90 percent of their cereal harvest and most of
their coffee beans to disease. However, they were able to
avoid famine because of their increased dependence upon
enset. This case suggests that fear of hunger and starvation
can be a powerful incentive to try to grow a new crop, even
a multiyear one such as enset.
The northern town of Lalibela, famous for its eleventh
century rock-hewn churches, is also the site where thou-
sands of people died as a result of the mid-1980s famine.
Some farmers in Lalibela grow a few enset plants near their
houses (Plate 9) in order to use the leaves to wrap bread for
baking. Like other northern Ethiopian farmers, those farm-
ers contacted had no knowledge of enset as food. Surprised
to learn that enset could be eaten, they expressed interest in
learning to cultivate and process enset for food as a means
of increasing food security.



However, before enset farming can be introduced to new
areas, a systematic survey throughout Ethiopia of the distri-
bution of wild and cultivated enset, as well as a study of the
history of enset use, should be undertaken. Trial farms
would need to be established where the mechanisms for
introducing planting materials, cultivation and processing
techniques, and cooking methods are provided. An adequate
number of livestock would also have to be available for the
production of manure and milk products. Furthermore,
remaining social stigmas about eating a fermented product
or a "peasant food" would have to be overcome.

I What extension and development work
could be implemented in the near future
to assist enset farmers?
Based on the research carried out so far, there are many
potential future interventions. These can be grouped into
the following unranked categories: A) extension information
to farmers concerning enset diseases; B) improvement and
mechanization of enset processing; C) improved livestock
breeds, pastures, and health and nutrition; D) increased and
improved production of protein-rich food crops; and E)
marketing assistance for enset products and improving
transportation and retailing networks.
A. Extension information to farmers con-
cerning enset diseases: Chief among the important
topics is an extension campaign to educate both
women and men farmers about the nature and spread
of bacterial wilt disease. Disease control requires an
integrated approach by the farmer and the farm com-
munity. Disease, particularly bacterial wilt, is not con-
trolled by outside inputs, but rather by a commitment
on the part of the farmer to follow proper sanitary
procedures, e.g., the use of clean cutting and process-
ing tools. This commitment only comes with knowl-
edge and its acceptance. This extension activity has
tremendous need and potential benefit-a benefit that
will have a spill-over effect into the understanding and
prevention of human and animal diseases.

B. Improvement and mechanization of enset
processing: The traditional methods of processing
may reduce the quality and quantity of enset food and
fiber. Research has been conducted in several institu-
tions (the Institute of Agricultural Research at Nazaret
and at Awassa, the Ministry of Agriculture, and Awassa

College of Agriculture) to develop improved process-
ing devices. Efforts have been made to modify: 1) the
decorticater that separates the leaf-sheath pulp from
the fiber; 2) the pulverizer that grates the corm into
fine pieces; 3) the kneader that squeezes out unwant-
ed water from fermented kocho; and 4) the shredder
that chops the fiber present in the fermented kocho.
Thus far such devises are primarily experimental and
have had little testing; farmer acceptance has not
occurred because of cost and inaccessibility (Metshen
and Abate, 1994). Adoption of these improved tools
should be pursued for their value in reducing labor
and increasing uniformity of products. There is also
potential for its dissemination as part of a cottage
industry development package.

C. Improving livestock breeds, animal
health/nutrition, and pastures, as well as using
enset leaves for enhanced feeds: All too often
researchers and extensionists ignore the importance of
livestock in maintaining the productivity (and with
respect to enset, the sustainability) of agricultural sys-
tems. McCorkle (1992) notes that within mixed pro-
duction systems, researchers have treated cultivation
and stock raising in virtual isolation and/or ignorance
of one another. Attention to animal nutrition and
health, improved pasture and forages. as well as
improved breeds and animal culling, would all have
positive effects on enset cultivation systems. Farmers
themselves ask development personnel and the cur-
rent researchers for improved veterinary services,
especially in regard to animal health. Since so much of
the enset system depends on cattle, assistance in
improving livestock breeds, training farmers how to
cull herds, and providing information, capital. and
planting materials for improved pastures and forages
are critical. Further, the role of enset leaves as a com-
ponent of silage and feed concentrates has not been
explored, but could have great potential to enhance
feed for a variety of livestock.

D. Increased and improved production of
protein-rich food crops: Haricot bean, lentil.
chickpea, and other seed legumes (pulses) supply pro-
tein-rich foods and are already important components
of the enset system. One of the main limitations of
enset food products is the low protein. There is



already much research data and extension information
on the production and utilization of these pulses. On-
farm trials are being conducted in the enset region on
some of these pulses. The productivity and nutritional
quality of the enset system could be improved through
a concerted extension effort to increase the produc-
tion, yield, and utilization of these pulses.

E. Marketing assistance and improving
transportation and retailing networks: Little is
known as to what could facilitate the marketing of
enset food and non-food products at local and urban
markets. There are many components of this challenge
that need exploration and intervention, including:
product supply and demand, transportation infra-
structure, supply of capital, market facilities, storage,
and packaging.

I What research agenda is necessary
to improve the understanding
of enset-based systems?
Since enset and enset-based systems have received little
study relative to many other crops and systems, the research
door is wide open. Much information considered to be base-
line for other crops has not been collected with respect to
enset. For example, almost no research has been conducted
on the effects on growth and yield of different clones; plant
density, spacing, and duration at a given spacing; transplant-
ing methods; manure and/or fertilizer amendments; propa-
gation techniques; and environmental conditions (i.e., tem-
perature, water, and sunlight).
While these baseline agronomic topics need to be
addressed, several larger issues related to food security and
natural resource preservation in Ethiopia must also receive
attention. They include: a) the effects on human nutrition as
population density and poverty reduce the amount of ani-
mal food products and cattle manure, and b) the sustainabil-
ity of enset systems (in terms of productivity and prevention
of soil degradation) as animal manure and other natural
resources become more scarce. Furthermore, the human
population carrying capacity of the various enset systems
under current practices needs study and estimation.
Interventions for both the maintenance of the indigenous
systems and the intensification of those systems to develop
new forms of production and processing require additional

Research trials are being carried out under station condi-
tions at Areka Research Station and Awassa Research Center
in southwestern Ethiopia (Plate 7). They relate to intercrop-
ping, fertilizer input levels, and measurement of yields. On-
farm studies, with farmers as trial cooperators, should be
implemented, as has been done with cereals, legumes, and
tubers (Franzel and Van Horten, 1992). These on-farm,
farmer-cooperator trials are particularly important because
of the complexity and diversity of enset management at the
household level, which, as previously mentioned, are not yet
fully understood by research scientists.

I What are some of the socioeconomic
and gender issues in need of further
Socioeconomic variables and agronomic management
practices are intimately related. One current puzzle is why
wealthier households have greater clonal variation, as well as
a greater number of clones in general. Is it because of their
larger land holding size, greater social networks for obtain-
ing more varieties, greater labor to plant and care for larger
enset holdings, larger incomes to purchase other varieties,
better all-around management practices, or the prerogative
of wealth? Research needs to be carried out on the reasons,
as well as on the consequences for household and surplus
Many of the previous sections noted that more research is
needed on a variety of topics. These include the following:

I Transplanting methods and harvest
management seem to be a function of
ethnic group, household needs, and
available resources (such as land, labor,
capital, and other food crops in the
system). But which are the critical
What is the importance of harvesting mature versus
immature plants? For example, North Omo male farmers
prefer to harvest mature plants, while women of the same
group prefer to harvest smaller plants for better taste and
ease of fermentation. The reasons for this distinction,
whether it is observed in other groups, and the conse-
quences for enset yields need careful analysis.
Since a major constraint on production is bacterial wilt,
socioeconomic and gender variables relating to disease



transmission agents (e.g., contamination by cutting and pro-
cessing tools, livestock, etc.) need to be examined.
The demographic composition of the household, the
amount of resources of the total household, as well as the
resources of individual members disaggregated by gender
need careful study. Access to and control over resources are
important to determine, as well as the overall carrying
capacity of a given area in terms of population density.
The division of labor by gender and the remuneration
received for selling surplus enset products need to be inves-
tigated. If men receive the profits from women's processing
labor, women's incentives are reduced. The effects of the
market on both men and women in terms of planting and
processing decisions need to be studied.
Some farmers plant more "female" than "male" plants.
This pattern needs to be investigated as to whether or not
there are gender-specific reasons for these choices, or if
women manage to prevail in their own preferences.

I What is the future of enset as a
sustainable agricultural system?
The present situation in the southern highlands of
Ethiopia provides a window into the workings of the issue
of sustainability in agricultural systems. Farm households in
some areas (e.g., Sidama, Gedeo, and Wolayta) are attempt-
ing to deal with the stress of limited land for cultivation and
grazing, as well as with rapidly growing populations. As
Rahmato (1996) warns for the Wolayta region:

The "triumph" of enset cultivation can now be seen
in the demographic mountain it has managed to throw
up, and the systematic crisis it has, unwittingly,
brought upon itself. Agricultural intensification, the
primary response of peasants to resource scarcity and
population pressure, has failed to arrest the intensifi-
cation of competition for resources and the accelera-
tion of demographic expansion. Moreover, while in
the past population growth may have stimulated
change and adaptability in the enset system, the
immense demographic pressure on the land today is
unlikely to induce technical progress, and may in fact
drive the system toward regression.

Research and development are needed to address sustain-
ability issues and the place of enset as a major contributor to
the food security of Ethiopia, or to search for alternative
agricultural systems. This research and development part-
nership needs to accelerate on all fronts that address the
biophysical and socioeconomic merits and limitations of the
various enset systems.




Abate, T., Hiebsch, C., Brandt, S., and Gebremariam, S.,
eds., 1996. Enset-Based Sustainable Agriculture in Ethiopia.
Addis Ababa: Institute of Agricultural Research.

Adebo, S., 1993. Dache Gofara Peasants Association: Report of
Diagnostic Survey. Redd Barna, Farm Africa, and Ethiopian
Ministry of Agriculture, North Omo. Addis Ababa: Farm Africa.

1992. Hanaze Peasants Survey. SOS Sahel,
Farm Africa, and Ethiopian Ministry of Agriculture, North
Omo Region (January 1992). Addis Ababa: FARM Africa.

Alemu, K., and Sandford, S., 1996. Enset Landraces. In
Abate et al, eds.

1991. Enset in North Omo Region. FRP
Technical Pamphlet No 1. Addis Ababa: FARM Africa.

Almeida, M., 1954. Some Records of Ethiopia, 1593-1646.
Translated and edited by C.E Buckingham and G.WB.
Huntingford. London: Hakluyut Society.

Aykroyd, WR., Doughty, J., and Walker, A., 1982. Legumes
in Human Nutrition. Rome: Food and Agriculture

Baker, R.E.D, Simmonds, N.W, 1953. The genus Ensete in
Africa. Kew Bulletin 3: 405-416.

Bayush, T., 1991. Community Management of Crop Genetic
Resources in the Enset Complex Farming System of Southern
Ethiopia: a Case Study from Sidama Region. Masters Thesis.
Agricultural University of Norway.

Beke, C., 1844. Abyssinia Being a Continuation of Routes
in That Country. Journal of the Royal Geographic Society 14:3.

Besrat, A., Mehansho, H., Bezuneh, T., 1979. Effect of vari-
etal difference and fermentation on protein quantity and
quality of ensete. Nutrition Report International 20(2):245-50.

Beyene, E, 1994. Present Situation and Future Aspects of Milk
Production, Milk Handling and Processing of Dairy Products in
Southern Ethiopia. Ph.D. Thesis: Agricultural University of

Bezuneh, T., Feleke, A., and Bauie, R., 1967. The
Cultivation of the Genus Ensete in Ethiopia. Soil and Crop
Science Society of Florida Proceedings 27:133-41.

Bezuneh, T., 1996. An overview on enset research and the
future technological needs for enhancing its production and
utilization. In Abate et al, eds.

1984. Evaluation of some enset ventricosum
clones by yield with emphasis on the effect of length of fer-
mentation on carbohydrate and calcium content. Tropical

Brandt, H., 1990. Food Security Aspects in Price and Market
Policies for Grain-Based Food Systems of Sub-Saharan
Africa. In Eshetu Chole, ed., Food Crisis in Africa: Policy and
Management Issues. New Delhi: Vikas Publishing House.

Brandt, S.A., 1984. New Perspectives on the Origins of
Food Production in Ethiopia.. In J.D. Clark, S.A. Brandt,
eds., From Hunters to Farmers: The Causes and Consequences
of Food Production in Africa. Berkeley: University of California

Brandt, S.A., 1996. A model for the origins and evolution of
enset food production. In Abate et al, eds.

Brandt S.A., and Carder, N., 1987. Pastoral rock art in the
horn of Africa: Making sense of udder chaos. World
Archaeology 19:194-213.

Brandt, S.A., and Fattovich, R., 1990. Late quaternary
archaeological research in the horn of Africa. In P
Robertshaw, ed., A History of African Archaeology. London:
Curry pp. 95-108.

Bruce, J., 1790. Travels to Discover the Source of the Nile, in
the Years 1768, 1769, 1770, 1771, 1772, and 1773. 5 vols.
Edinburgh: Edinburgh University Press.

Bull, M., and Konde, A., 1995. Report of Diagnostic Survey of
Bashilo Dalo Peasant Association in Arbegona Wereda. Addis
Ababa: FARM Africa.



Clark, J.D., 1988. A review of the archaeological evidence
for the origins of food production in Ethiopia. In Taddese
Beyene, ed., Proceedings of the Eighth International Conference
of Ethiopian Studies. Addis Ababa: Institute of Ethiopian

-, 1976. The domestication process in sub-
Saharan Africa with special reference to Ethiopia. In Origine
de I'Elevage et de la Domestication. Nice: UISPP

--, 1967. The problem of neolithic culture in sub-
Saharan Africa. In WW Bishop and J.D. Clark, eds.,
Background to Evolution in Africa. Chicago: University of
Chicago Press.

Degu, G., and Workayehu, T., 1990. Initial Results of
Informal Survey: Areka Area Mixed Farming Zone, Welayita
Awraja, Sidamo Region. Addis Ababa: Institute of Agricultural

Diro, M., and Tabogie, E., 1994. Enset (Enset ventricosum):
Surveying Base Cropping Systems in some Enset Growing
Regions of Ethiopia. Paper presented at the Second Ethiopia
Horticultural Workshop, 1 to 3 Dec., Addis Ababa, Ethiopia.

Ehret, C., 1979. On the antiquity of agriculture in Ethiopia.
Journal of African History 20:161-177.

ENI (Ethiopian Nutrition Institute), 1981. Expanded Food
Composition Table for Use in Ethiopia. Addis Ababa: ENI.
FAO/WHO (Food and Agriculture Organization/World
Health Organization), 1973. Energy and Protein Requirements.
Report of Joint Expert Committee. Rome: FAO.

Fekadu, D., 1996. Potential of Enset (Ensete Ventricosum) in
Ruminant Nutrition in Ethiopia, Part III. M.Sc. Thesis,
Swedish University of Agricultural Sciences.

Franzel, S., and van Houten, H., eds., 1992. Research with
Farmers: Lessons from Ethiopia. Wallingford: C.A.B.
International for Institute of Agricultural Research Ethiopia.

Gasse, E, Rognon, P., and Street, FA., 1980. Quaternary his-
tory of the Afar and Ethiopian Rift lakes. In M.A.J. Williams
and H. Faure, eds., The Sahara and the Nile. Rotterdam:

Gebre, Y., 1996. Economic and socio-cultural significance of
enset among the Ari of southwest Ethiopia. In Abate et al, eds.

1995. The Ari of Southwestern Ethiopia: An
Explanatory Study of Production Practices. Social
Anthropology Dissertation Series No. 2, Department of
Sociology, Anthropology, and Social Administration, Addis
Ababa University

Habte-Wold, T., Sandford, S., and Kindness, H., 1996. The
age and sex structure of enset plantations in north Omo. In
Abate et al, eds.

Harlan, J. R., 1993. The tropical African cereals. In T. Shaw,
P. Sinclair, B. Andah, and A. Okpoko, eds., The Archaeology
of Africa: Food, Metals and Towns. London: Routledge. pp-53-

1969. Ethiopia: a center of diversity Economic
Botany 23:309-314

Hiebsch, C., 1996. Yield of enset ventricosum: A concept. In
Abate et al, eds.

Ishihara, M., 1993. Coffee and food crop production in
Qadamasa, Illubabor region of southwest Ethiopia: A prelimi-
nary research report. Nilo-Ethiopian Studies Newsletter 1: 14-18.

Jensen, A., ed., 1959. Altvolker Sud-Athiopiens. 1. Stuttgart:
WK. Verlag.

Kaplan, S., 1992. The Beta Israel (Falasha) in Ethiopia. New
York: New York University Press.

Kindness, H., 1994. Household Cash Income Sources and
Income Generating Activities in Welaita, North Omo. Addis
Ababa: FARM Africa.

Konde, A., ed., 1993. Report of Diagnostic Survey of Debre
Medhanit Tabia in Dedebana Derga-Agen Wereda. Addis
Ababa: FARM Africa.

Konde, A., and Bull, M., 1995. Report of Diagnostic Survey of
Gallo Argessa Peasant Association in Awassa Zuria Wereda.
Addis Ababa: FARM Africa.

Lockhart, D., trans., 1984. The Intinerario of eronimo Lobo.
London: Hakluyt Society.



Makiso, T., 1980. Agronomic report for 1979. Pub. No. 63.
Wolaita Agricultural Development Unit.

McCabe, J.T., 1996. The Role of Livestock in the Enset
Cultivation System: A Review of the Literature. Agricultural
Research Center, Awassa, mimeo.

McCabe J.T., and Lee, S.A., 1996. The role of livestock in
the enset cultivation system. Paper presented at the Annual
Meetings of the American Anthropological Association, San

McCann, J.C., 1995. People of the Plow. Madison: The
University of Wisconsin Press.

McCorkle, C.M., ed., 1992. Plants, Animals, and People:
Agropastoral Systems Research. Boulder: Westview Press.

Methzun, T, and Abebe, Y., 1994. Study of Enset: Processing
and Development of Enset Processing Tools in the Southern
Regions of Ethiopia. NORAG.

Ministry of Agriculture, 1983/84. General Agricultural Survey,
Preliminary Report. Addis Ababa: Ministry of Agriculture.

Murdock, G.P, 1959. Africa: Its Peoples and Their Culture
History. New York: McGraw Hill.

NRC, 1989. Recommended Daily Allowances. National
Research Council, Subcommittee on the Tenth Edition of the
RDAs. Washington, DC: National Academy Press.

Pankhurst, R., 1996. Enset as seen in early Ethiopian litera-
ture: History and diffusion. In Abate et al, eds.

1986. The History of Famine and Epidemics in
Ethiopia prior to the Twentieth Century. Addis Ababa: Relief
and Rehabilitation Commission.

Phillipson, D.W, 1993. The antiquity of cultivation and
herding in Ethiopia. In Thurston Shaw, Paul Sinclair, Bassey
Andah, and Alex Okpoko, eds., The Archaeology of Africa:
Food, Metals and Towns. London: Routledge.

Pijls, L., Timmer, A., W/Gebriel, Z., and West, C., 1994.
Cultivation, preparation and consumption of Enset in
Ethiopia. Journal of Science, Food and Agriculture. July 16.

Rahmato, D., 1996. Resilience and vulnerability: enset agri-
culture in southern Ethiopia. In Abate et al, eds.

Sandford, J., and Kassa, H., 1996. The effect of gender on
resource contribution, decision making and influence: A
comparison between enset, teff and maize. In Abate et al,

Sandford, S., and Reece, A., 1992. Proceedings of the
Workshop on Farmers' Participatory Research. Addis Ababa:
FARM Africa.

Sauer, C.O., 1952. Agricultural Origins and Dispersals. New
York: American Geographical Society

Shack, W, 1966. The Gurage: A People of the Enset Culture.
Oxford: Oxford University Press.

Shank, R., and Ertiro, C., 1996. Enset Crop Assessment: A
Linear Model for Predicting Enset Yield and Assessment of kocho
Production. Addis Ababa: UN Emergencies Unit for Ethiopia
and World Food Program. Manuscript on file.

Shigeta, M., 1991. The Ethnobotanical Study of Ensete (Ensete
ventricosum) in Southwestern Ethiopia. Ph. D. dissertation,
Center for African Area Studies, Kyoto University.

-, 1990. Folk in situ conservation of ensete
(Ensete ventricosum [Welw.] E.E. Cheesman): Towards the
interpretation of indigenous agricultural science of the Ari,
southwestern Ethiopia. African Studies Monographs, Kyoto,

Simmonds, N.W, 1958. Ensete cultivation in the Southern
highlands of Ethiopia: A review. Tropical Agriculture 35:302-

Simoons, E, 1965. Some questions on the economic prehis-
tory of Ethiopia. Journal of African History 6:1-13.

-, 1960. Northwest Ethiopia: Peoples and Economy.
Madison: University of Wisconsin Press.

Spring, A., 1996a. Diagnostics, Design, and Interpretation
Using RRAs to Assess Enset Farming Systems in Southern
Region. Mimeo.



--, 1996b. Gender issues and farming systems
research and extension in enset agriculture in Ethiopia. In
Abate et al, eds.

-, 1995a. Agricultural Development and Gender
Issues in Malawi. Landham and London: University Press of

1995b. A Rapid Rural Appraisal of the Environmental
Contaminants in the Lower Black River Morass, Jamaica.
Miami: North-South Center.

Spring, A., Haile, B., Tesfaye, S., Abebe, Y., Amaldegen, A.,
w/Michael, G., Tabogie, E., Surur, O., Tsegaye, A., Shimles,
S., Habte, T., Menjeye, T., and Tadesse, T., 1996. Enset
Farming Systems in Southern Region, Ethiopia: Report on a
Rapid Rural Appraisal in Gurage, Hadiya, and Sidama Zones.
Addis Ababa: GTZ, mimeo.

Stiehler, W, 1948. Studien zur Landwirtschafts und
Siedlungsgeographie Athiopiens. Mit 3 Karten. Erdkunde

Tabogie, E., 1990. Enset in Ethiopian Agriculture. Institute of
Agricultural Research News, Areka Research Centre.

Woldetensaye, Asnaketch, 1997. The Ecology of Ensete ven-
tricosum in Ethipoia. Doctoral dissertation. Department of
Soil Sciences, Swedish University of Agricultural Sciences.
Acta Universitatis Agriclturae Sueciae, Agraria 78.

Terefe B., and Tibogie, E., 1989. A review of the available
research recommendations and future strategies on enset. In
Proceedings of 19th National Crop Improvements Conference,
April 1987. Addis Ababa: Institute of Agricultural Research-

Tesfaye, S., and Haile, B., 1995. Diagnostic Survey Report of
Enset Based Farming Systems of Maraka Woreda, North
Omo. Awassa: Institute of Agricultural Research. Mimeo.

Vavilov, I., 1926. Studies on the Origin of Cultivated Plants.
Leningrad: Institute of Applied Botanical Plant Breeding.

Webb, P, and von Braun, J., 1994. Famine and Food Security
in Ethiopia: Lessons for Africa. New York: John Wiley and

Westphal, E., 1975. Agricultural Systems in Ethiopia.
Wageningen: Centre for Agricultural Publishing and

Zippel, K., and Alemu, K., 1995. A Field Guide to Enset
Landraces of North Omo, Ethiopia. FRP Technical Pamphlet
No 9. Addis Ababa: FARM Africa.


Directorate for International Programs
1200 New York Avenue, NW
Washington, DC 20005

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs