LEAVING TWO THIRDS OUT OF DEVELOPMENT: FEMALE HEADED HOUSEHOLDS AND COMMON PROPERTY RESOURCES IN THE HIGHLANDS OF TIGRAY, ETHIOPIA FULL REPORT Prof. Patricia Howard Erin Smith, M.Sc. Department of Social Sciences Wageningen University and Research Center The Netherlands For the Livelihoods Support Programme (LSP) Sub-programme “Improving people’s access to natural resources” Food and Agriculture Organization of the United Nations (FAO) September 2006 ii Foreword This report contains the full results of a study that was commissioned by the Livelihoods Support Programme (LSP) of the Food and Agriculture Organization of the United Nations (FAO). The LSP, which is supported in part by the UK Department for International Development (DfID), is helping to improve the impact of FAO interventions at the country level through the effective application of Sustainable Livelihood (SL) approaches. The research team consisted of a lead scientist, a senior researcher, an assistant field researcher, and three survey enumerators, as well as an assistant who supported transcription of field notes. The lead scientist, who was responsible for overall research design, implementation and reporting, was Prof. Patricia Howard of the Department of Social Sciences at Wageningen University in the Netherlands. The senior researcher, who was responsible for supporting research design, leading and implementing field research and co-authoring the report, was Erin Smith (M.Sc), also of Wageningen University. The field research was carried out over a four-month period during which the research team made considerable demands on many people who generously and graciously gave their time: first and foremost, the residents of Adiarbaetu, then the colleagues associated with the FAO Project, and lastly all of the DAs and many other officials at the kushet, tabia, and woreda levels. There are far too many names to mention; we thank you and hope to return something of value to you. A summary of this report has been published on the FAO LSP website: http://www.fao.org/sd/pepub_en.htm The results presented in this report reflect the findings and opinions only of the authors, and not of the FAO. Address for Correspondence: Prof. Patricia Howard Dept. of Social Sciences Wageningen University and Research Centre Hollandseweg 1 6706 KN Wageningen The Netherlands Email: Patricia.Howard@wur.nl i Glossary of Terms and Abbreviations AE Area enclosure; officially-designated land areas that have been enclosed seasonally or year-round to conserve and regenerate grass, trees and other vegetation. Areas are guarded and violations are punished. Baito The democratically-elected administrative and political council at tabia and kushet levels. BoANRD Bureau of Agriculture and Natural Resource Development; develops and implements regional agricultural policy CPR Common Property Resources Derg Regime Led by Colonel Mengistu Haile Mariam, the military government that overthrew Emperor Haile Selassie in 1975, that was itself deposed in 1991 by the EPRDF DA Development Agent; employees of the woreda-level Agricultural Departments who provide extension and advice to farmers Emic An insider’s understanding of how the world works EPRDF Ethiopian People’s Revolutionary Democratic Front, the political party that runs the government together with the TPLF ERHS Ethiopian Rural Household Survey Etic An outsider’s understanding of how the world works FAO Food and Agriculture Organization of the United Nations FH Female headed households Haile Selassie Emperor of Ethiopia from 1930 until his overthrow by the Derg Regime in 1974 IFPRI International Food Policy Research Institute Kushet Sub-divisions of tabias, consisting of villages and their surrounding land and resource areas MH Male headed households Meher Long and heavy rainy season usually from June to September NGO Non-governmental organisation Peasant Association Created by the Derg Regime to popularly administer policies and implement reforms at the local level Rist system The traditional system of land allocation based on descent lines that predominated for centuries over much of the highlands ii Tabia Sub-divisions of woredas; the lowest administrative levels in the Tigray region with populations from 5000-7000 Woreda In Tigray, Zones are sub-divided into administrative districts called woredas (alt. spelling: weredas, waredas) REST Relief Society of Tigray TPLF Tigray People’s Liberation Front, the political party that waged civil war against the Derg Regime and that, together with the EPRDF, runs the government in Tigray WAT Women’s Association of Tigray, a mass organization headed by the TPLF that promotes women’s participation in development and empowerment WFP World Food Programme of the United Nations Zones Tigray is divided into four major administrative and political areas called Zones iii Table of Contents Page Foreword i Glossary of Terms and Abbreviations ii Chapter 1 Leaving Two Thirds Out of Development: A Point of Departure 1 Postulations The Objectives What the Literature Does and Doesn’t Say, with a Focus on Bias and Gaps Organization of the Report Nota Bene 1 4 5 8 11 Dynamics of Development. Natural Resource Access Regimes, Household Differentiation, and Degradation in Tigray: A Literature Review and Analysis 13 1.A 1.B 1.C 1.D Chapter 2 2.A 2.B 2.C 2.D 2.E Chapter 3 3.A 3.B 3.C 3.D An Overview of the Smallholder Farming System and Livelihoods in Tigray Land Access and Reform, and Inter-generational Land Insecurity Women’s Assets, Female Household Headship and Poverty 2.C.1 Marriage Arrangements and Property 2.C.2 Women, Marriage, and Land under the Reforms 2.C.3 Divorce and Death of the Spouse 2.C.4 Inheritance 2.C.5 Decision Making, Bargaining Power within Households, and the Gender Division of Labour 2.C.6 Female Household Headship and Poverty Land Degradation and Common Property Resources 2.D.1 Resource Degradation Dynamics 2.D.2 Land Tenure, Common Property, and Resource Degradation 2.D.3 Woodlots and Trees 2.D.4 Enclosed Grazing Lands 2.D.5 Other Botanical Resources and Livelihoods Benefit Streams, Equity and Livelihoods in CPR Management 14 16 19 20 21 22 24 25 25 27 28 29 32 35 37 40 Research Design and Methods, or How Not to Get Lost Dealing with CPR Complexities 43 The Conceptual Framework 3.A.1 Livelihoods 3.A.2 Botanical Resources and their Access 3.A.3 Gender Relations and Botanical Resource Access 3.A.4 Household Headship Research Components and the Levels of Analysis Research Methods and their Field Applications 3.C.1 Quantitative and Qualitative Community Level Data 3.C.2 Contemporary and Retrospective Community Resource Mapping 3.C.3 Landscape Niche Calendars/Resource Distribution Study 3.C.4 Botanical Resource Cultural Valuation Study 3.C.5 Community Resource Access Study 3.C.6 Women’s Resource Access Histories 3.C.7 Species Access Study Household Level Data Collection: the Household Survey 43 43 44 45 45 46 48 48 51 52 53 54 54 54 58 iv Page 3.E 3.F Chapter 4 4.A 4.B 4.C. 4.D 4.E 4.F 4.G Chapter 5 5.A 5.B 5.C 5.D 5.E 5.F 5.G 5.H 5.I 5.J v 3.D.1 Content 3.D.2 Data Analysis and Limitations Constraints and Opportunities to Learn: Improving Research to Improve Decision Making 3.E.1 Lost in Time: Scarcity of the Ultimate Resource 3.E.2 Lost in Space: Study Site Selection 3.E.3 Lost in Translation I: The Research Team and its Composition 3.E.4 Lost in Translation II: Cultural Bias Embedded in Language and Methods 3.E.5 Lost in the Jungle: Botanical Identification 3.E.6 Found at Last: the Advantages of Living in the Field Suffer the Little Children, or How Not to Get Lost 58 62 62 62 63 64 64 65 65 66 The Research Site and Its Population 69 Situating Adiarbaetu 4.A.1 Social, Political, and Bureaucratic Organization 4.A.2 Transport and Market Access Water Access and Infrastructure Demographic Data and Household Composition Considering Headship Education Infrastructure and Attainment by Sex, Age, and Household Headship Health and Health Care Infrastructure, including HIV/AIDS Livelihoods Written Large 4.F.1 Land and Agriculture 4.F.2 Homegardens 4.F.3 Off-farm Activities 4.F.4 Food-for-work and Compulsory Community Labour Household Structure, Marriage and Divorce 69 69 70 71 72 Household Headship, Gender, and Livelihoods 81 Private Land and Its Use 5.A.1 Private Land Access 5.A.2 Land Tenure and Decision Making 5.A.3 Sharecropping 5.A.4 Land Use and Crop Residue Use 5.A.5 Land Improvements: Soil and Water Conservation Crop Production and its Distribution Livestock Production and its Distribution 5.C.1 Distribution of Livestock between Households 5.C.2 Distribution of Livestock within Households Trees on Private Land Homegardens Off-farm Activities and Livelihoods 5.F.1 Off-farm Income 5.F.2 Days Worked and Income Earned Community Work (including Food-for-work) Transfer Payments, Credit and Savings Membership in Community Organisations The Gender of Livelihoods 72 73 75 75 76 77 77 77 81 81 83 85 86 87 88 90 90 92 93 95 97 97 99 100 102 104 105 Page Chapter 6 6.A 6.B 6.C 6.D 6.E 6.F 6.G 6.H Chapter 7 7.A 7.B 7.C 7.D Common Property Resources, Resource Access and Use, and their Relations with Household Headship and Livelihoods 109 Common Land Areas in Adiarbaetu Formal Access Rules: Enclosures and Species Restrictions 6.B.1 Enclosures 6.B.2 Species Restrictions Common Land Areas and their Uses 6.C.1 Grazing Lands and Grass 6.C.2 Unenclosed Hillsides 6.C.3 Woodlots and Forests 6.C.4 Other Community Enclosures 6.C.5 Riversides and Other Borderlands 6.C.6 Churchyards (Sacred Groves) 6.C.7 On-farm Resources Seasonality and ’Intensity’ of CPR Use Informal Common Property Resource Access and the Case of Awelie Women, Female Headed Households and Common Land Areas Botanical Resources: A Closer Look 6.G.1 Fuelwood 6.G.2 Wild Foods 6.G.3 Medicinals 6.G.4 Fodder 6.G.5 Construction Materials 6.G.6 Ploughing and Other Tools 6.G.7 Bee Forage 6.G.8 Religious and Other Cultural Uses Common Property, Botanical Resources, and Livelihoods 110 111 111 111 113 114 116 117 119 119 120 120 121 124 127 129 130 133 135 136 137 138 138 138 141 Leaving Two-thirds Out of Development: A Point of Arrival 147 How Female Household Headship is Entailed in Development Dynamics The Social and Economic Position of Female Headed Households Livelihoods and CPR Access: Are Current Management Regimes Sustainable? Entry Points for Local Action 147 149 150 155 References 159 Appendix Tables 165 vi List of Boxes, Tables, and Graphs Page Boxes 1 2 3 4 5 6 7 8 The Research Objectives Main Research Design Elements A Framework for Assessing Natural Resource Tenure Question Guide for the Community Resource Access Study Question Guide for Women’s Resource Access Histories Question Guide for the Species Access Study Coffee Ceremony The Importance of Etan 5 43 45 55 56 57 139 140 Tables 1 2 3 4 5 6 7 8 9 10A 11 12 13 14 15 16 17 18 19A 20A 21A 22A 23 24 25A 26 27A 28 29A 30A 31A 32A 33A 34A 35A 36A vii Chronology of Droughts and Famines in Tigray and its Surroundings during the Last Two Centuries Uses of Enclosed Woodlots, 100 Villages in 50 Tabias, Tigray, 1998-99 Harvest of Woodlot Products 1997 to 2000 Uses of Restricted Grazing Lands, 100 Villages in 50 Tabias, Tigray, 1998-99 Research Objectives, Level of Analysis and Research Methods Data Collection Methodologies and Information Generated Emic Wealth Categories Main Characteristics of the Case Study Species Content of the Household Survey – Concepts, Variables, Dimensions, Indicators Source and Tenure of Water for Domestic and Agricultural Use by Sex of the Household Head Dependency Ratios by Sex of Household Head Educational Level by Sex of Household Head Education Completed by Sex by Age Cohort: Surveyed Household Population Education Completed by Age and Sex of Household Head Principle Crops Grown in Adiarbaetu and Area Marriage and Assets at Marriage, by Sex of Household Head Disposition of Children and Household Assets upon Termination of Last Union, by Sex of Household Head Total Landholdings by Sex of Household Head Number of Household Members by Total Landholdings Quality of Household Land Allocated by the Baito by Sex of Household Head Quality of Total Household Land by Sex of Household Head Location of Agricultural Plots by Sex of Household Head Tenure of Land Plots by Sex of Household Head How Acquired Land and Where Acquired Land by Person Acquiring Land Intra-Household Decision Making by Who Acquired the Plot Reasons for Sharecropping Land In and Out by Sex of Household Head Division of Agricultural Labour in Female Headed Households Land Use by Sex of Household Head Use of Crop Residues Harvested by Sex of Household Head Total Household Landholdings by Amount of Crop Residues Harvested Soil and Water Conservation Measures on Plots by Person(s) Responsible Crops Planted During Meher, by Area Planted and Sex of Household Head Crop Combinations, Meher Season, by Sex of Household Head Total Meher Crop Area as a Percent of Total Landholdings by Sex of Household Head Total Quintals Consumed per Meher Crop and Average Per Household Producing that Crop, by Sex of Household Head Total Kilos of Own Crops Consumed By Household Dependency Ratios 2 32 34 36 47 49 50 58 59 165 72 73 74 74 76 78 79 82 166 166 166 167 83 84 167 85 168 86 168 168 169 169 170 170 171 171 Page 37A 38A 39A 40A 41 42 43A 44A 45A 46 47A 48 49 50 51A 52 53A 54A 55 56 57 58A 59A 60A 61 62 63 64 65 66 67A 68A 69A 70A 71A 72A 73A 74A 75 76A 77A 78A 79A 80 81A Amount of Male Household Labour 16-64 Years by Total Household Crop Consumption in Kilos Number of Oxen by Crop Consumption in Kilos Number of Oxen by Total Meher Crop Area Total Crop Harvest in Quintals and Share of Harvest for Tenants, FH Numbers of Livestock by Type and Sex of Household Head Distribution of TLU by Sex of Household Head Total Farmland by Total Number of Tropical Livestock Units Control Over Livestock within Male Headed Households Live Animal Purchases, Sales and Consumption in the Past 12 months Sales, Income and Consumption of Animal Products by Sex of Household Head Trees Growing on Household Land by Species Uses Made of Trees Growing on Household Land Total Number of Trees by Size of Landholding Number of Trees per Household by Sex of Household Head Rights to Trees by Sex of Household Head Number of Homegarden Crops per Homegarden Crops Produced in Homegardens by Number of Households Reporting the Crop Division of Labour in Homegardening by Relation with the Household Head, MH Primary and Secondary Occupations of Heads and Wives of Heads Activities to Generate Income in Cash or Kind by Sex Days Worked and Income Earned over the Past Four Months, by Sex of Worker and Sex of Household Head Earnings by Activity Wages per Day by Sex and by Activity Participation in Community Work Activities by Household Headship and Sex Days Worked and Compensation Received, Food-for-Work, by Household Headship and Sex Credit and Savings by Sex of Household Head Membership in Organisations by Household Headship and Sex Adiarbaetu Common Land Areas Permissible Uses of Area Enclosures Species Restrictions All Uses by All Common Land Areas Users of Restricted and Open Grazing Lands and Days of Use Species Used by Common Resource Area, All Species and All Areas Species Collected From Unenclosed Hillsides and Uses Species by Species Use for Enclosed Woodlots and Unenclosed Forests and Woodlots Species and Species Use for Other Community Enclosures Species Reported as Collected from Riversides and other Boundary Lands and Uses Seasonality of Use,* by Major Use Categories and by Common Land Areas Use of Awelie Reported in the Household Survey and Resource Areas where Collection was Reported Major Use Categories by Sex of Household Head Sex and Age of Common Resource Users by Major Use Categories Species Mentioned by Residents of Adiarbaetu, Local Names, Possible Scientific Names, and Uses Twenty-five Most Important Plant Species According to Use Categories, Men's Focus Groups and Women's Focus Groups Fuels Used and Fuelwood Sources by Sex of Household Head Species Used for Fuelwood by Number of Times Reported 172 172 173 173 91 92 174 175 176 93 176 94 94 95 177 96 177 178 98 98 99 178 179 180 101 103 104 112 112 113 181 182 183 185 186 187 187 188 127 189 190 191 195 130 197 viii Page 82 83 84A 85A 86A 87 88A 89A 90 Fuelwood Users Average Days of Use by Major Use Category and by Common Resource Area Wild Food Plant Consumption Frequencies All Wild Foods Reported by Local Plant Name All Species Reported as Used by Uses Users of Wild Food Species All Medicinal Plants Reported, by Local Name Species Ranking according to Number of Times Use was Reported Important Species According to the Cultural Valuation Study 131 132 198 199 200 135 202 203 142 Map of Tigray and the Study Area Crop Residue Harvest by Total Livestock Units Crop Consumption by Male Labour Availability and Oxen Ownership Area Cultivated by Oxen Ownership Average Days per Year of Use by Major Use Category Seasonality of Use by Major Use Category Seasonality of Resource Use by Common Land Areas, All Areas Seasonality of Resource Use by Common Land Areas, Excluding Unenclosed Hillsides Use of Common Land Areas by Sex of Household Head 10 87 89 90 122 122 123 Graphs 1 2 3 4 5 6 7 8 9 ix 123 128 Chapter 1 Leaving Two-thirds Out of Development: A Point of Departure 1.A Postulations Drought, famine and poverty are not new to the Ethiopian highlands, but drought appears to be increasing in frequency and impact, and hunger has become a seemingly permanent feature of the region (Table 1). For decades it has been known that famine and hunger are not inevitable consequences of drought, and the Ethiopian Government, national and international development agencies, and NGOs have been attempting to understand and address the factors that come together to generate the drought-hunger-poverty nexus. Another phenomenon, which is certainly not new but which is also likely to be much more common and consequential now than in the past, is the high incidence of female household headship and the striking correlation between extreme poverty and female headed households, which seems to be as intransigent as drought. Being a member of a female headed household in highland Ethiopia means having a 35% chance of being destitute, compared with only an 8% chance if one belongs to a male headed household. Female household heads are far more likely to be landless and, when they do have access to land, 70% must sharecrop it out (losing around half of the yield in the process), in part because they lack access to enough labour and livestock to farm it themselves. Female headship and poverty are certainly no strangers in other parts of the world. However, in the case of highland Ethiopia, this very close relationship requires explanation, particularly considering that formal equality between men and women in access to assets such as land and livestock has been part of the region’s customary tenure and kinship system for hundreds of years, and today is embedded in Ethiopia’s constitution, as well as in many other laws governing property rights. While currently poor female headed households are supported by food aid and food-for-work programmes, their specific needs are otherwise barely addressed, and they are largely socially and economically excluded. Development dynamics appear not just to marginalize, but to continually generate these extremely poor households, as though they were a structural feature of particular economic policies, like underemployment or inflation. Another of the most important and tenacious problems that is argued to either cause or seriously aggravate the drought-poverty-famine nexus in the plough-based cereal farming system of the highlands is severe natural resource degradation, particularly manifest in soil erosion, deforestation, and devegetation. Government and international efforts to understand processes contributing to this degradation and to reverse it have been massive and certainly predate the renowned 1984-85 drought and famine. Successes in reforestation efforts, in protecting other resources such as grazing lands, and in implementing soils and water conservation measures on private holdings, have been significant, and some of the degradation appears to have been reversed. However, it appears that little success has been achieved in regenerating people’s capacities to earn their livelihoods without foreign aid, widespread malnutrition, and misery, and also continued degradation of resources, especially on common lands that are not protected through enclosure. Natural resource degradation also appears to be a structural feature of development dynamics in the highlands. It is argued herein that resource degradation, poverty, and female household headship are all manifestations of a particular development dynamic that has prevailed in the highlands for at least the past 30 years, but whose roots are clearly historical and tenacious. Drought is one of the external shocks that have sent the once lush, if not prosperous, highlands into crisis. Particularly in Tigray Region, which is the focus of the current research, another is civil war. These alone go very far to explain the gravity of the situation in Tigray today, a bare 15 years after the 17-year-long civil war ended, and a bare three years after the last major drought, and they also help to explain the very substantial number of widows in the region, many of whom are household heads. But it is well known that there are yet other factors involved. Institutions in the highlands, particularly those that are meant to manage land and other natural resources, have undergone substantial change since the mid-1970s. In that process, rights to exploit and manage both private and communal land resources, and the trees and 1 Table 1 Chronology of Droughts and Famines in Tigray and its Surroundings During the Last Two Centuries PERIOD CAUSE AND SEVERITY 1812-16 Severity unrecorded 1831 Severity unrecorded 1835-38 Large human and livestock mortality, cholera epidemic, known as ‘the years of stagnation’ 1864-66 Large human mortality 1876-78 Large livestock mortality 1880 Much loss of livestock 1888-92 Great Rinderpest epidemic, 90% livestock loss, 30% human fatalities, highland Ethiopia agriculture nearly destroyed 1913-14 The price of grain increased 30-fold and there was great starvation 1943 War induced famine in Woyane rebellion 1949-51 Drought and locust plagues 1953 Severity unrecorded 1957 The year did not have more than 10 rainy days 1958-59 Epidemic and locusts, famine and 100,000 people died in both Wello and Tigray 1964-66 Worse than 1984 drought 1971-78 50% of livestock lost, 200,000 dead in Tigray, Wello and northern Shoa 1984-85 Sequential rain failure; much livestock dead; 8 million people affected and about 1 million people dead nationwide. About 200,000 people from Tigray crossed border into Sudan 1990-92 Rain failure, estimated 4 million people suffering food shortage nationwide 2002-03 1.83 million people in Tigray in need of food aid Source: Beyene (2003), Table 4-1. other flora and fauna that are associated with them, have been redefined time and time again.1 In such processes, of course, women’s rights were also affected. There were three possible outcomes for them: reforms ameliorated pre-existing inequalities, reproduced or compounded them, or generated new inequalities. Such institutional reforms also meant that all highland smallholders confronted great uncertainty in terms of their property rights to land, trees, and common land areas which, it is widely recognized, often leads to resource degradation, as well as to poverty. Still, land and institutional reforms do not in and of themselves explain the substantial inequality in access to assets that especially widowed, divorced, and single women confront: in principle, at least, women’s rights to property during all land redistribution processes were protected by law, although it is often argued that officials in charge of reallocation in fact discriminated against them. Rather, it is argued in this report that the processes that have resulted in the current unequal access to assets between male and female household heads have only barely been affected by land reforms. The reforms were intended to create equality in land holdings, but they have failed to resolve land hunger. Land hunger, and the current lack of economic viability of most smallholdings, are generated by the same processes that result in the over-production of women headed households. The particular forms of household formation and dissolution, and the customary relations of marriage, divorce, and inheritance, are a highland legacy that led to large-scale conversion of common lands into agricultural holdings and, when this was no longer viable, to the diminution and fragmentation of farms. Setting up a new household, divorcing one spouse in order to marry another with more land, and having children with more than one woman, all presented other means, especially for 1As this report is being written, the Government of Tigray is preparing another plan to devolve management of most of the hillside land in the area to local councils. See Chapter 2 and Nedessa et al (2005) for more details. 2 men, to gain access to additional land. To mention only a few indicators that hint at the nature of the problem: one study in Tigray found that the average marriage lasted only 7.5 years; larger-scale data from the region show that the average number of children that a Tigrinian woman will have in her lifetime is around 6.8; and two-thirds of all households in Tigray own less than a half of a hectare of land. As will be discussed in various sections of this report, these same household dynamics are also responsible for generating a large number of poor female headed households. Nor is it casual that women constitute the majority of the disadvantaged: this problematic is firmly embedded in gender relations. Common property resources have in a sense ‘fallen victim’ to the dynamics discussed above, since common land was in effect the only land available to assuage the land hunger generated by high rates of household formation and dissolution. However, common land is an essential constituent of livelihoods in an area where the plough-based farming system is dependent on feed resources from communal grazing lands; where house construction, fuel, agricultural and household implements are all mainly derived from trees and other plant species that are found in only miniscule quantities, if at all, on private holdings; where medicines as well as fruits, vegetables, and other essential nutrients are also generally not supplied by the cereal and legume-based production system; and where cultural associations with indigenous and wild botanicals have such strong religious foundations that these continue to thrive in local churchyards, which are so rich in species diversity and density that other communally-protected forested areas pale in comparison. The institutional reforms that have occurred over the past thirty years, and the massive campaigns to halt degradation and reforest the highlands, have had a major impact on those common lands that have remained. Laws governing the management of such areas have changed repeatedly, and have also often failed to define many user rights. Those areas that are protected (‘enclosures’) have generated considerable environmental benefits, but policies have largely failed to consider the population’s needs for botanical resources other than grass and timber. Enclosures have largely failed to contribute to the livelihoods of the very poor who, lacking livestock and cash, cannot use most of the resources that they do provide. Common land areas that are not protected have been largely left out of conservation policy: large tracts of land have become de facto open access, while most highland populations rely on these lands to meet these multiple material and cultural needs. It is only realistic to expect that continued degradation of this ‘third’ of all livelihood resources will result. The outcome of the interplay between these factors means that much of the highland farming population cannot survive from their holdings, while their rights to exploit and manage non-farm natural resources and common land areas have been severely limited by environmental conservation and regeneration efforts, on the one hand, and by degradation ensuing from de facto open access on the remaining land, on the other. Female headed households, which are the product of civil war, high divorce rates, unequal division of marital assets, and irresponsible paternity, lack the resources to benefit substantially from the limited resource extraction allowed in enclosed areas. On the other hand, in comparison with households that have greater assets, they are forced to rely more heavily on areas that are open access and thus, together with others who are landless and poor, women heads aggravate the degradation that is one of their greatest banes. It is hoped that this introduction highlights the importance of exploring the possible interrelationships between common property resource access, resource degradation, poverty, and household headship in the highland context as a series of postulates. Currently, although there has been much research on development dynamics in the highlands, little is actually known about several of the inter-connections posited above, although they may be vitally important both to human welfare and to environmental recovery in the region. Much of the large-scale statistical data that are needed to substantiate such interrelationships are lacking; nor could the research that generated the results contained in this report provide them. The interpretation of the statistical data that does exist is in any case impossible without continuous reference to ethnographic and historical research findings. Much more work needs to be done, and it is hoped that the contributions made in this report can help to orient such work by providing important hypotheses. The global objective of this research was to investigate the interrelationships between female headship and access to natural resources in a specific development context in highland Tigray, where the Food and Agriculture Organisation of the United Nations operates a programme oriented toward enhancing food security. It was soon found, however, that the connections between common property resource access and female household headship were much more complex than anyone first as- 3 sumed: there are a substantial number of ‘intervening variables’ in which to become entangled. However, it is this process of disentangling both the direct and indirect relationships that sheds considerable light not only on the nature of the development dynamics in the region, but as well on the need for new approaches to resolving resource degradation, enhancing conservation, and contributing to livelihood security in Tigray, by addressing the ‘missing two-thirds’ of development. 1.B The Objectives An ongoing FAO project (GCP/ETH/056/BEL “Improving nutrition and household food security in Northern Shoa and Southern Zone of Tigray, Ethiopia”) (FAO n.d.) aims to improve the food security of the most vulnerable groups at the household level in communities in Gera Keya and Lalomama woredas in Northern Shoa Zone, Amhara Region, and in Hintalo Wajirat and Enderta woredas in Southern Zone, Tigray Region. Participatory Rural Appraisal exercises carried out in the project region had shown that the availability of and access to natural resources are important determinants of a household’s food security and nutrition, where the poor and female headed households are particularly at risk. The FAO wished to investigate further the possible interrelations between access to natural resources and poverty among female headed households, in order to add important information and value to the Belgian-funded FAO project, as well as to provide normative outputs for FAO and the Livelihood Support Programme’s Access to Natural Resources sub-programme. In the Terms of Reference for this project, the FAO indicated that a major reason for the poverty of female headed households in the project area is related to the fact that the farming system is plough-based and, although the majority of female heads have farmland, women generally do not possess oxen. Since their land holdings are small and culturally women are not allowed to plough, they must sharecrop out their land and hence cannot sustain their families. Since the benefits derived from their own crop land are insufficient to meet their food and other basic needs, female headed households, they argued, depend upon access to trees and wild and domesticated or semidomesticated plants occurring outside of their own cropland in forests, grazing lands, and borderlands, and in homegardens, to provide them with fuelwood, fodder, famine and other foods, utensils, etc. They were also reported to resort to daily labour and other income generating activities, including sale of charcoal and fuelwood, chickens, eggs, handicrafts, local alcoholic drinks and hair braiding. However, FAO reported, given that labour and business opportunities in remote areas are scarce, the household food security situation of most female headed households is extremely precarious. Therefore, it argued, improving access to tree and plant resources for this group could potentially improve their livelihoods. It concluded that understanding their current access to these resources, as well as dynamics of change in this access over time, was imperative. Within this context, this research was intended to investigate women’s and men’s current access to tree and plant resources and to explore the linkages between access (including to land on which these resources occur), gender, and livelihoods among the poorest populations in a study site in Tigray Region. Specifically, the FAO sought to address the following questions: • • • • • What are the major constraints and opportunities to accessing tree and plant resources for men and women in poor households, and particularly for female headed households, in the selected study site? Which tree and plant resources are currently and have in the last generation made important contributions to poor women’s livelihoods, and how is their access to these resources changing and why? What are the current constraints in access to tree and plant resources in these areas that are related to gender? How does this affect poor women’s livelihoods, and particularly the livelihoods of female household heads? What are the entry-points to strengthen access to tree and plant resources, particularly for poor women and female headed households?” These were subsequently translated into specific objectives that guided the research design, literature review, and choice of methods that are elaborated in Chapter 3, which are presented in Box 1. 4 Box 1 Research Objectives 1. To determine livelihood dependencies on wild and indigenous botanical resources and their dynamics. 2. To determine dependencies on different types of common lands. 3. To determine the importance and diversity of homegardens. 4. To determine differential access to common land resources. 5. To capture determinants of poverty and differentiation of households. 6. To capture determinants of livelihood strategies, especially those related to botanical resources. 7. To capture contributions of plants and trees to food security. 8. To capture the significance of gender relations and household headship, and the importance of plant, tree, and common land resources to women and female headed households. 9. To provide recommendations regarding possible means to strengthen the access of the poor, and particularly of female headed households, to botanical resources. The title of this report thus reflects the initial concerns of FAO’s Livelihood Support Programme when it commissioned the study (the point of departure), but it also reflects the conclusions reached after months of literature review and field investigation, which are presented throughout this report and summarized in Chapter 7. It was found that women headed households have in fact been almost completely neglected in policy making and in development interventions other than in food aid. Common property resources (CPR) were also found to constitute a very important part of livelihood resources – at least a third – but these resources have also in fact been very largely neglected. It was also found that neglect of both almost certainly together generate further ‘externalities’, including the reproduction of poverty and resource degradation, and as well as possibly the alarming increase in HIV/AIDS in Tigray. If this and the other postulates that are presented above have indeed been left out of development, what can be the reasons for this? 1.C What the Literature Does and Doesn’t Say, with a Focus on Bias and Gaps As is discussed in Chapter 2, there has been extensive research carried out in Tigray and in nearby regions of highland Ethiopia that is concerned primarily with pinpointing the causes of poverty and malnutrition in the region. Such research, some of which is based upon large-scale surveys and panel data, usually has as primary goals tracking these phenomena over time and helping to better target aid to the region. Further, Tigray has been the subject of numerous research projects focusing on the dynamics of environmental degradation and conservation, particularly since the Regional Government has placed great emphasis on halting erosion and deforestation, and because such degradation is thought to have a strong relationship with poverty and malnutrition. This means that much is already known about the factors leading to high levels of poverty, malnutrition, food insecurity, and degradation. It is also clear from the aforementioned research that particular population groups are overrepresented among the poor and destitute, especially female headed and other households that lack access to a minimum of livelihood assets such as labour, land, and livestock. Some research also goes into much greater depth in an effort to understand how marital assets are divided, how this affects women’s bargaining power within marriage, and how it leads to much inequality in asset division upon divorce or death, which therefore has led to poverty among female headed households. Women, it is shown, are still largely dependent on male largess if they are to survive from agriculture. They are also reported to be those who least benefit from land reforms, from CPR use, and even from the FAO project interventions in the study area, in spite of the fact that they are specifically targeted (Mirjam Kuzee, FAO, Pers. Comm. 2004). Research on CPR that has been reviewed for this study has been particularly concerned with assessing the efficacy of the mass efforts to conserve soils and water on farms, and to regenerate grazing lands and afforest areas in enclosures. Some has begun to explore the relations between access regimes, livelihoods, and equity, and the most recent research has been concerned as well with possible alternatives to community-based management, or devolution. 5 In spite of the apparent plethora of research related to the phenomena under investigation in the current study, it is argued in this report that are still many lacunae and some biases evident in the conceptual frameworks that have informed such research. Some of these biases are also evident in development assistance that has been provided to the Region, and some are also apparent in policy making. Many of those lacunae and biases that have been evident in the past are by now wellrecognized in development research and practice, and are no longer so apparent in the literature on Ethiopia, although it could be argued that they persist in policy making: gender bias that neglects women and gender relations; conservation bias that neglects the importance of CPR for local livelihoods; and a timber bias in forestry policies that neglects the role of non-timber forest products (NTFPs). One of the biases that seems to largely persist and that is not commonly recognized is the bias of agro-centricity2 that neglects the importance of non-cultivated resources in general, as well as cultivated and managed areas outside of agricultural fields (such as homegardens, fallow fields, borderlands etc.), both for the viability of farming systems and for livelihoods.3 The literature review that was conducted both at the outset and again at the conclusion of this research found the following gaps in the published literature in relation related to Tigray4 (and, it could often be argued, in relation to the highlands more generally): 1. Especially the significance of indigenous and wild botanical resources for rural livelihoods in Ethiopia is poorly understood, even though Ethiopia is one of the world’s ‘hotspots’ of biological diversity and is particularly known for being a centre of origin of a significant number of food plants. While recently there has been much research relating to plant genetic resources in Ethiopia, particularly to teff (Eragrostis teff) and enset (Ensete ventricosum), but also to wheat (Triticum vulgare), sorghum (Sorghum vulgare), legumes, and oil crops, this has barely extended to traditional horticultural and wild food resources, which are only very recently the subject of research in Ethiopia in general, and which apparently have barely been researched in Tigray. Leading authorities on wild and indigenous plant resources in Ethiopia commented: “Traditional Ethiopian vegetables do not figure very prominently in modern crop research and conservation programmes. Traditional vegetable crops are marginalized in modern agriculture while the wild and weedy vegetable species receive no special attention” (Asfaw 1995: 1). This same neglect extends to wild edible plants more generally: "While plants used in traditional medicinal practices have, comparatively, been well documented, wild edible plants have received very little attention” (Asfaw and Mesfin 2001: 53). Particularly, the contribution of wild food plants as ‘famine’ food and their actual and potential contribution to nutrition more generally has recently become a concern to aid agencies.5 However, within Ethiopia in general, and within Tigray Region and in the FAO project area in particular, there is very little information regarding wild plant and indigenous horticultural and tree resources. While the FAO Baseline Survey Report on Nutrition and Household Food Security in Northern Shewa and Southern Tigray (FAO 2004a) collected some data on homegardens, these were apparently limited to exotic vegetables and, while it mentioned the cultivation of a wild cactus (Opuntia ficus-indica L. Mill.) used for food and fodder, it made no other reference to wild plant foods, although it indicated that large percentages of the population are unable to feed themselves for long periods during the year. While food aid currently provides much of the caloric intake of the population in the reference area, it has not done so traditionally, whereas the strong seasonality of crop production has always resulted in lean periods where not only ‘famine foods’, but many other food resources from the wild are used that also constitute part of the culinary traditions of the Tigray peoples. This bias with respect to Tigray was encountered repeatedly when the researchers involved in this project were deployed to the field: they were informed by officials that there are no such resources left in Tigray. 2For a discussion, see Howard 2003b and Klappa 2006. While, especially due to the contemporary concern with biodiversity conservation, there is a very substantial and growing body of literature dealing with traditional use and management of such resources and their importance for livelihoods, the bias that contributes to the neglect of these resources has been barely conceptualized or discussed. 3This bias is obviously much less evident in recent forestry literature, especially that dealing with NTFPs. However, even this literature often neglects to consider resource areas outside of forests that are cultivated or semicultivated. In Ethiopia, such areas have been relatively more investigated by ethnobotanists. 4For a discussion of the literature covered and the limitations of this literature review, see Chapter 2. 5See, e.g. Barnett 2001; Guinand and Lemessa 2000; Kenyatta and Henderson 2001; Kew Royal Botanical Gardens n.d. 6 2. The literature on CPR use in Tigray a) focuses on area enclosures, whereas the management and use of common lands that are not enclosed are generally not discussed, although the latter constitute the majority of all common land. Therefore, the use of the botanical diversity that is typically associated with these non-enclosed areas6 is not considered. Further, b) the use of enclosed areas is discussed mainly in terms of their importance for grazing and timber, with little reference to their many other uses. Other than in the case of a single study (Jagger et al 2003), the contributions of enclosed areas to livelihoods are therefore under-estimated. The significance of such lacunae is exemplified in reference to fuelwood: although the exploitation of woody species for fuelwood has been cited as one of the primary reasons for the rapid disappearance of local forests in Tigray, and most studies recognize that fuelwood is a vital subsistence resource that is now extremely scarce in the region, not a single study was found that focused on fuelwood use or availability in the region. There are also apparently no central policies in place that are oriented toward solving the fuelwood crisis. Research on CPR access and management also focuses nearly exclusively on enclosed areas 3. and neglects non-enclosed common lands. The assumptions about how these lands were managed before the reforms are contradictory and apparently not grounded in research, so there is very little knowledge about the changes that were provoked by reforms, and still less knowledge about how the remnants of such customary systems continue to influence CPR management today, although it is often acknowledged that much contemporary common land management is still at least partly based on pre-existing indigenous systems. Only one study was found that researched one customary common land tenure regime prior to the reforms, as well as the changes that were effectuated after the reforms, and the authors of this study stated that no other literature on customary tenure was found in relation to common property prior to 1975. Due to this dearth of information, with a few exceptions, CPR studies in Tigray and elsewhere also do not consider the decline in such customary management systems as a cause of resource degradation, nor how such remnants may now affect resource management, equity, etc. Also, as this study shows, violations are under-reported and informal access rules, some of which may be related to customary management regimes, may be more important than formal access rules in terms of actual resource use; yet such informal access rules are not yet acknowledged in the literature. 4. There is little data on the access that different population groups have to CPR resources or on the distribution of benefits derived from CPR use among different population groups (e.g., men and women, of different wealth strata, female and male headed households), even in reference to enclosed areas. Those data that are reported refer mainly to timber and grass. No studies have been found that attempt to carefully assess the contributions that these resources make to the livelihoods of different population groups. 5. Extensive research on rural gender relations in Ethiopia and some research in Tigray reveals little about women’s work outside of agriculture, other than in off-farm occupations. There are now several studies focused on agriculture that examine women’s use of improved varieties and their roles in managing indigenous staple crops,7 but no literature was found that investigated women’s involvement or gender differences in wild plant gathering and management, homegardening, agroforestry or forestry, medicinal and ethnoveterinary practices and related knowledge, or women’s access to common land and botanical resources, other than anecdotally. A few assessments of enclosures have discussed women’s participation in local councils that manage such areas, but apparently none has investigated this in-depth. 6. With a few exceptions, the implications of household formation and dissolution for resource management, poverty, and female household headship in rural areas are also barely researched. The dynamics leading to high levels of female household headship have been more generally described, but as yet little is known about how reforms and the current legal framework have created incentives 6For discussions of botanical diversity outside of agricultural fields and its uses see especially Asfaw 1995; Asfaw 2001; Asfaw and Mesfin 2001; Deffar 1998; Williams et al 2001; Vivero n.d.). 7See e.g. Sandford and Kassa 1994; Tsegaye 1997; Tiruneh 2001 ; Tiruneh et al 1999; Tiruneh et al 2001. 7 or disincentives for divorce, the trends in divorce and irresponsible paternity, the rates of formation/dissolution of female headed households, and their consequences for poverty and phenomena such as the spread of HIV/AIDs. Nothing has been found on the actual or possible interventions of local institutions (e.g., baito, PA) in issues such as the division of assets upon divorce or widowhood, although it is mentioned in passing in a few studies that today such intervention appears to be virtually non-existent. The biases and lacunae mentioned above are certainly not unique to Ethiopia, but are related to more pervasive cultural assumptions about women’s roles and about what are usually termed ‘minor’ land-based resources. Such bias has its greatest impact when it is women who are mainly managing ‘minor’ non-cultivated resources (Howard 2003a). At a global level, it is argued that the relative lack of attention to both is not a casual phenomenon: “. . . because most plant use, management and conservation occurs within the domestic realm, and because the principal values of plant genetic resources are localised and non-monetary, they are largely invisible to outsiders and are easily undervalued” (Ibid: 4). What this implies is that the contribution of women and of ‘minor’ botanical resources to livelihoods and food security in rural Ethiopia, and in Tigray in particular, are unclear, as is much of the domain of access to and use of wild and indigenous botanical resources. All of this notwithstanding, as can be seen in Chapter 2 there are numerous excellent publications that delve deeply into different aspects of the development dynamics in the region. Without this literature, there would have been no possibility to design this study or interpret the findings. All of it provides great insight into development dynamics in the Ethiopian highlands, and the most recent studies particularly point the way toward greater understanding and hence toward more effective policy making. The research presented here draws heavily on this literature and on the methods and instruments employed in previous studies, and the report begins by reviewing that which has been most important. Given the lacunae mentioned above, the research proposed for Tigray turned out to be highly ambitious. There are numerous methodological challenges to researching the relations between access to CPR, their use, and livelihoods. If this research is to be serious, it requires a multi-disciplinary approach and the concomitant combination of qualitative and qualitative research methods. FAO was interested in the research in part because it wished to learn about methodologies that can be used to investigate access to and use of CPR. The relationship between botanical resource access and livelihoods has not often been investigated in regions that present such extremes of resource degradation and of CPR management complexity. A range of methods had to be used, and certain new methods were developed and tested in this context. It is hoped that the current report will contribute to the ongoing search for means to empirically capture the phenomena under study. In the absence of previous research on the topic, the study was considered to be exploratory, seeking to identify variables and their possible inter-relations, rather than to test specific hypotheses statistically or determine absolute levels (e.g. the degree of contribution of wild or indigenous plants to food security). Nevertheless, in general the research did intend to achieve reliability, e.g. the findings would be the same if repeated with the same or similar subjects in the same or a similar context, with the provision that even small variations in agroecological conditions may be highly significant and create substantial contextual differences that will affect the findings. However, given the ambitious nature of this research, the resource constraints that were present, and its exploratory nature, several of the objectives could not be fully realized. Nor could the research even begin to overcome the lacunae mentioned above: it provides a case study dealing with several of the issues raised, but not all. These constraints and limitations of the study are discussed particularly in Chapter 3, but are also mentioned in the chapters that present the empirical results. 1.D Organization of the Report As was mentioned above, Chapter 2 presents a review of the most relevant literature and the discussion of development dynamics in the Ethiopian highlands as they have been posited to date. It begins with an overview of the farming system and associated livelihoods in Tigray (for a map of the region and its administrative boundaries, see Graph 1). Against this backdrop, the first component of the natural resource access regime - private land access - is discussed both in relation to the traditional rist system and to land reform, as well as to what has been identified as a permanent feature of both that is significantly related to resource degradation and poverty: inter-generational land insecurity. 8 Inter-generational land insecurity and social differentiation among households are in turn related to household formation and dissolution, including marriage, intra-household power relations, divorce and death, where gender relations are a defining factor. Together with political instability that gave rise to the civil war, these factors are posited to explain the high incidence of female household headship and poverty among female headed households. Then, the second major component of the natural resource access regime - CPR access - is discussed both in relation both to the dynamics of natural resource degradation (of forests, grazing lands and farmland) and to private and CPR tenure, particularly in relation to the government’s responses to degradation that have resulted in enclosures intended primarily to conserve and regenerate grass and trees. The government-instituted CPR regimes, although often locally managed, have largely displaced traditional local institutions. These regimes tend to ignore the longstanding relationship between local plant biodiversity and livelihoods. Within these relatively new CPR configurations, household dynamics, including female headship, are also implicated, but are also largely overlooked, which has consequences for the dynamics which recycle poverty, hunger, and resource degradation. Chapter 3 presents an overview of the research design, the conceptual framework and definitions employed, and the methods and instruments used, as well as critical reflections on their implementation in the field and ways in which implementation affected the research outcomes. FAO expressed particular interest in identifying and elaborating methodologies to assess the relations between natural resource access and livelihoods, particularly in relation to botanical resources. The research design attempted to research these interrelations in-depth by generating both qualitative and quantitative ethnobotanical and socio-economic data. The lessons learned, which are presented in this chapter, can help to formulate future research to better capture the relations between CPR use, access, livelihoods, and poverty in the region, which is crucial for improved policy formulation and development assistance. Chapter 4 focuses on the micro-research context, providing an overview of Adiarbaetu, the research site, and its population, which serves as a necessary point of reference and reflection for the remainder of the report given that statistical data was generated only with a very small, purposefully selected and therefore biased sample, and qualitative data was collected from a relatively small number of informants. However, researchers did live in the study site for three months, and were thus able to come into contact with and observe a large number of people and resource areas, and to develop considerable rapport with the population. For this chapter, grey literature was reviewed, and local officials were interviewed and asked to provide tabia or kushet level data, which was complemented by data from the Household Survey (n = 30). The chapter thus presents information on living conditions and welfare in the kushet, which in part are determined by the presence and organization of basic public services and infrastructure at both village and household levels. Then, the information available on demographic characteristics, the composition of households, and marriage and divorce practices are presented, since these are known to vary according to locality and are important dimensions of development dynamics in the study area. Social, political and bureaucratic organization is also obviously important in resource management and distribution, and these are briefly outlined. Finally, an overview is presented of livelihoods and some of the factors that determine livelihood strategies. Chapter 5 takes an in-depth look at the livelihoods and livelihood strategies of 30 households in the study area, including 15 female and 15 male headed households that were selected to represent different emic wealth strata. The survey that was used to collect the data presented was based on the Ethiopian Rural Household Survey (ERHS), which was modified in several ways, especially to generate information on CPR access and use. The chapter systematically compares and contrasts the results with respect to male and female headed households, and also critically examines the relations posited in the literature reviewed in Chapter 2 with respect to social differentiation, asset ownership, and poverty. It begins by examining access to and use of private land, and then relates these to crop and livestock production and consumption, respectively. The existence and importance of trees and homegardens, and their roles in livelihoods, are also discussed. The chapter then explores off-farm self-employment and wage employment, and their contributions to livelihoods, and then addresses other sources of income and subsistence goods such as community work (including food-for-work), transfer payments, and credit access as well as savings. Membership in community organisations, including those providing for labour exchange, credit and other types of social security are also analysed. 9 Graph 1 Map of Tigray and the Study Area Hintalo Wajirat Woreda 10 Chapter 6 presents and analyses the field data collected in Adiarbaetu in relation to access to common land areas, the uses that the local population makes of these areas and their botanical resources, and the ways in which use and access differ among individuals and households. Local perceptions about common property areas and enclosures, about the values associated with botanical resources (their emic importance), and the benefits that the community and various social groups within it derive from them, are fundamental to this analysis. The discussion focuses in particular on gender relations and female household headship, and on formal and informal rules of access to CPR and how these affect CPR use, and then teases out the relationships between such access rules and poverty dynamics affecting especially female headed households. The chapter begins with a discussion of the different resource areas and of the formal access rules that govern them, as well as the means of control used to discourage violations. Findings are then presented regarding use and access by type of common land area, including grazing lands (restricted and unrestricted) and grass, hillsides, woodlots and forests (restricted and unrestricted), riversides and other borderlands, churchyards, and private farmland. The use patterns, users, and access regimes within each are described and discussed. Then, the seasonality and intensity of resource use (of both areas and types of uses) are examined and related to factors such as climate, enclosures, and resource demand patterns. The discussion then turns to informal rules of access to resources that are related both to traditions and cultural customs, as well as to exceptions that were found to be systematically made when formal rules are implemented, particularly in relation to those species whose use is prohibited, but that are also very important to livelihoods. The issue of equity in resource access is then explored by examining female headed household’s access to benefits from area enclosures and open access areas, and the effects that these enclosures have for their livelihoods. The chapter then examines the major botanical resource use categories, and the species that are most important within and across these categories, using an emic perspective. Actual use patterns are discussed with reference to the species and the reasons that particular species are used for each of the major emic use categories: wild foods, medicinals, fodder, construction materials, ploughing and other tools, bee forage, and religious and other cultural uses. Species use patterns and species preferences diverge considerably, since people are forced to use plant resources that are often far from desirable in terms of fulfilling their material and cultural needs. This is subsequently related to the problems and possibilities associated with existing access regimes, dealing particularly with the tensions between meeting the needs of the local population and conservation. The chapter concludes with a summary and reflections on the relations between CPR access and livelihoods, especially referring to female headed households, and their implications for sustainable resource management. Chapter 7 returns to the postulations, or point of departure, presented in Chapter 1, relating these to the literature review and development dynamics discussed in Chapter 2. The first issue addressed is the nature of development dynamics in the region and the number and poverty of female headed households, where it is argued that the latter are generated in part as a means for men to control additional land. The second issue address is whether CPR resource access and use regimes in the region are sustainable, given the development dynamics that are evident. In addition, recommendations for immediate action are made on the basis of the study results in Adiarbaetu, the study village, and are primarily focused on means to improve the lives and livelihoods of female headed households, including their access to botanical resources. Nota Bene The reader should be aware that, the first time that glossary items appear in the text, they are presented in bold. Words in Tigrinia or in Amharic are presented in italic, unless they are found within a quotation or are so commonly used in research that this is impractical. Vernacular plant names in English are presented in regular font; those in Tigrinia or Amharic are presented in italics. All scientific plant names are presented in italics, in accordance with taxonomic conventions. In chapters 5 and 6, the number of tables is so great that many are relegated to an appendix located at the end of the report. Tables that are presented in the appendix are represented by the letter “A” (e.g., 10A). 11 12 Chapter 2 Dynamics of Development. Natural Resource Access Regimes, Household Differentiation, and Degradation in Tigray: A Literature Review and Analysis This chapter traces the relationships between natural resource access regimes, resource use and degradation, poverty, and household differentiation in the highlands and in Tigray as these have been presented and posited in research. Given the seemingly intransigent nature of the problems of hunger and poverty in the region, and the subsequent need for international food aid, as well as the severity of natural resource degradation and its relation to the drought-hunger-poverty nexus, there are what appears to be a plethora of studies focusing on these dynamics. However, as mentioned in Chapter 1, there is a relative dearth of literature dealing with household differentiation and CPR, particularly in relation to those areas that are not enclosed, but also in relation to the areas that are. The same biases that explain such lacunae are also responsible for the dearth of research on non-cultivated botanical resources and livelihoods. A thorough review of the literature about these development dynamics is not possible in the current context; however, we have attempted to locate and review those studies that most directly address the inter-relationships between rights to natural resources, resource use and degradation, household differentiation and poverty, including all dealing with female headed households (FH). The review does not take into account much grey literature or literature published in languages other than English. Further, much of the literature that is reviewed does not directly address Tigray, but rather presents results on other areas in the Ethiopian highlands that share the farming system, the heritage of the rist system, and the predominance of the Ethiopian Orthodox Church. There is still much research to be done with respect to Tigray itself and, within Tigray, as is discussed in Chapter 3, there is also much diversity. The review therefore traces the grosser outlines of the relevant development dynamics in the region, drawing both upon large-scale quantitative research and smaller-scale in-depth case studies, with the objective of contextualizing the study area and developing grounded hypotheses. Settlement in Ethiopia is purported to have been concentrated in the highlands because neither malaria nor tsetse flies were found at these elevations (Williams et al 2003). The region has been continuously settled by humans since at least 3000 BC, and the ox-plough farming system has been traced back to at least four to five centuries BC.15 The tenacity and persistence of the cultures and economic formation that explain this continuity are surprising given the internal and external shocks to which they have been subjected, particularly over the last half century. However, recent development dynamics have plunged the region into crisis. These dynamics are very complex and, it can be argued, are as yet not well understood. There are at least six factors that are considered to contribute to this complexity. First, agroecological conditions are very difficult, since the region is both mountainous and prone to highly variable rainfall and periodic drought, which partly explain these shocks. Second, ploughing is a permanent feature of the farming system and, since oxen pertain to individuals and households rather than to other social collectivities, oxen ownership has been identified as the defining factor in household differentiation and it is also a powerful symbol of personal wealth. However, current cattle and other livestock populations are considered to be too high to be sustained. Third, the tenure system that persisted until the 1970s was ancient and highly sophisticated, involving both private and communal property and the associated local traditional institutions (especially clan and church), whereas land reform and institutional change have since been imposed from above in a manner that both complements and contradicts these traditions, and that both resolves and creates tenure insecurities. Fourth, severe resource degradation, particularly soil erosion and devegetation, affect most of the highlands. Fifth, much of the highlands, and particularly Tigray, has recently been the site of prolonged and intense civil war. Sixth, the process of demographic change in the region is en- 15For a discussion and references, see Beyene 2003. 13 twined with the specific form of land tenure, household formation, composition and dissolution, which give rise to the fragmentation and diminution of farms, to the point where most are currently economically unviable. Households are equal insofar as they are mostly all poor, so that the major distinctions made among them today are between the ‘poor’ and the ‘destitute’. Developing policies that permit both sustainable resource management and sustainable livelihoods is indeed a very great challenge that is not likely to be achieved in the short term. 2.A An Overview of the Smallholder Farming System and Livelihoods in Tigray The population of Tigray in 2005 was estimated to be around 4.3 million, where 81.2% of the population resided in rural areas. Tigray covers 80,000 sq. miles and has an annual average human population growth rate of 3% (Gebremedhin et al 2000). Ethiopia also has the largest livestock population in Africa and the tenth largest livestock population in the world. About 70% of the cattle and sheep and 30% of the goats in Ethiopia are in the highlands above 1,500 metres, whereas all camels are in the lowlands (Mengitsu n.d.). In Tigray alone, according to the 1998 agricultural census, there were around three million cattle, nearly one million sheep, 1.5 million goats, a half a million equines, and around 130 thousand camels (Gebremedhin et al 2004). There are more than 36 livestock units per km2 and more than 40 per capita, which are among the highest ratios in Ethiopia (FAO 2004b). The region is relatively dry and, as indicated in Table 1, subject to frequent drought. Farming in the northern highlands is nearly entirely dependent upon rainfall, which ranges from 450-980 mm annually (Ibid) and is strongly seasonal and erratic. The main cropping season, called meher, is from mid-June to September, when rains are concentrated. Cultivation varies depending on soil type, drainage, exposure, and especially altitude, but the basic pattern is pervasive Hoben (1973). There are three altitude zones: qolla (lowlands, < 1300 m.a.s.l.), weyna dega (midlands, 1300 – 2200 m.a.s.l.), and dega (highlands, > 2200 m.a.s.l.). Lowland crops include maize (Zea mays), pearl millet (Genus penissetum) and sorghum (Sorghum spp.); midlands crops are wheat (Triticum spp.), barley (Hordeum vulgare) and teff (Eragrostis tef) and, in highland areas, barley and potatoes (Solanum tuberosum) dominate. Pulses (e.g., field peas [Pisum sativum] and lentils [Lens escuelenta]), oil seeds, vegetables, and spices are also produced across the highlands, but the higher altitudes have the lowest species diversity. Crop productivity is generally low due to factors such as soil degradation and erratic rainfall: half of the households surveyed by FAO in the project area produced 100 kg of grains or less per year in 1994/95 (FAO n.d.). External inputs such as fertilizers and seed have not been found to have a substantial impact on crop productivity, mainly because of these factors (Gebremedhin et al 2002). Depending on the crop and soil type, land must be ploughed two to six times per year to permit sufficient depth of cultivation and soil granulation. Grain is harvested using sickles, and oxen or horses are used for threshing. Oxen are required for ploughing due to the dry climate and heavy soils (Aune et al 2001); ploughs are made of wood. Cattle provide traction power for 95% of grain production and also provide milk, meat, dung (which maintains soil fertility and provides a source of fuel), income, and personal savings which serves as a hedge in periods of drought and famine (Gebremehdin et al 2000, FAO 2004b). Animals have traditionally been naturally grazed, but today grazing lands do not meet minimum nutritional requirements, particularly in the dry season. Natural grazing predominated in Tigray. The availability of natural grazing land has been reduced substantially since much common grazing land was converted to farmland, while the creation of enclosed areas for woodlots (see Section 2.D.3) has also limited the area available for open grazing (Gebremedhin et al 2004, Chisholm 2000), relegating grazing to marginal lands (e.g., waterlogged, on steep slopes) that are unsuitable for cultivation, so that productivity is very low (Amare 1999, Mengistu n.d.). Currently, livestock feed sources in Tigray include crop residues (45%), grazing lands (35%), browse (10%), and crop aftermath (8%) derived from 3.6 million ha of cultivated land and 3.2 million ha of grazing land. About 68% of crop residues are fed to oxen and 20% to milking cows and newly castrated bulls, while the rest is fed to other livestock when natural grazing resources are in short supply (Gebremedhin et al 2004). Beles (Prickly pear – Opuntia ficus indica) an exotic cactus species introduced into Ethiopia in the latter half of the 20th century, is increasingly used for ani- 14 mal feed (Ibid). Important browse species have been reduced since these are used for fuel and construction purposes (Mengitsu n.d.). Amare (1999) pointed out that, with limited access to cut fodder resources, sheep and goat breeding may still be possible since they are fed "merely by letting them graze freely, supplementing this with a small amount of hay. Goats are the least constrained since their feeding habits allow them to thrive on hillsides and lowland areas unusable for anything else" (Ibid: 76). However, the lack of grazing land obviously limits the number of sheep that can be maintained. During the cropping season, fields are off-limits to grazers, so livestock have to be herded and the lack of grazing area places a limit on sheep breeding. Sheep and other livestock must also compete with oxen for access to grazing land. The farming calendar has several labour peaks, and there is a relatively marked gender division of labour only insofar as ploughing and sowing are usually considered to be strictly male tasks. Ploughing initiates in March-April for the meher and may continue into May-July, depending upon the soils. It occurs again in September-October if pulses are planted (Amare 1999, Beyene 2003). "The fact that farming is plough-based gives great importance to male labor in this farming system, a fact that puts households which lack male labor at a serious disadvantage” (Amare 1999). Labour peaks between June and August when fields must be cultivated and weeded. Beyene (2003) reported that particularly weeding is onerous, heavy, and highly time demanding, and it is mainly women who weed. Since weeding occurs during the period when food shortages are greatest, such labour demands have serious negative health consequences. Harvesting begins in October and lasts through November-December, presenting another labour peak. Little labour-saving technology is employed, and labour demands are often great over short periods due to the cropping calendar. Food availability is also strongly seasonal. The months of greatest food shortage are typically from late May to September, but this varies according to land and rainfall distribution. Dietary intake is also affected by food culture: there are many religious prohibitions, especially among Orthodox Christians. While meat is preferred, there are many animals that cannot be eaten, such as pig, camel, and many types of wild game. The Ethiopian Orthodox Church observes 250 fasting days per year when no meat or dairy products can be consumed. The majority of the food consumed in the highlands (Selinus 1971) consists of stews and sauces eaten with injera, a large pancake-like bread made ideally from teff but, since it is often in scarce supply, barley and other grains are substituted. During non-fasting days, the diet consists primarily of beef, goat, or mutton cooked in berbere16 sauce, and few or no vegetables are eaten unless meat is unavailable. However, on fasting days, meals consist of legume stews and vegetable dishes consisting primarily of potato, chard or cabbage, and occasionally beets and carrots, and wild food plants are also often consumed. In general in the highlands, households have relied on own grain production to satisfy their requirements for staple food and only purchase grain when harvests are insufficient. Grain sales have been the most important source of cash income (Amare 1999). In areas where livestock are important, animal and animal product sales are also a source of cash: particularly of hides, skins, cheese, eggs, and butter, although their contributions to total household income are usually reported to be small. Other common sources of cash are sales of pulses, tela and other local alcoholic beverages made from grain and honey, and skilled work such as weaving, healing, carpentry, building, and ironwork, “which can be significant sources of income but only a handful of individuals possess these skills" (Amare 1999: 58). The FAO (n.d.) noted that poor infrastructure and great distances to markets militate heavily against the success of market-oriented enterprises. Wolde-Georgis (1996) also noted that, because farms in Tigray are so small, they are barely capable of producing marketable surpluses. His local surveys in two rural villages in Tigray showed that only about 21% of respondents engaged in some kind of non-agricultural employment, and only 18% had some skill with which they could generate income. New employment sources that have emerged are related to conservation programmes and Area Enclosures (AE) (e.g., soil and water conser16This is a traditional Ethiopian spice mixture consisting of primarily of chillies (Capsicum spp.) and up to 28 different spices, including fenugreek (Trigonella foenum-graecum), ginger (Zingiber officinale), black pepper (Piper nigrum), coriander (Coriandrum sativum), cinnamon (Cinnamomum zeylanicum), and cardamom (Elettaria cardamomum). Recipes are personalized depending on the spices used and their ratios. 15 vation, nurseries, AE guards), some of which are paid and some of which are voluntary or compulsory. Food-for-work and other community programmes oriented toward infrastructure development and maintenance and conservation have become major sources of sustenance for the rural population in Tigray, and these are discussed further in Section 4.F.4. 2.B Land Access and Reform, and Inter-generational Land Insecurity The history of land access and tenure in the northern highlands of Ethiopia (the centre of the old Christian kingdom) is highly complex, both due to the ancient and sophisticated traditional tenure system (see e.g. Amare 1999, Fullerton Joireman 1996, Hoben 1973), and to the repeated land reform and redistribution processes that have occurred since the late 1970s. As Beyene (2003) noted, there is much similarity in the principles of land tenure systems prior to 1975 and after the reforms: that is, while the State completely re-engineered rural institutions and land access after 1975, it did so on the basis of land access principles that were well institutionalized in the customary system. Prior to the State reforms, the major forms of land access in Tigray and in other parts of northern Ethiopia and Eritrea were the customary rist and gulti systems. Hoben provided the most definitive account to date of the rist system, which differs from Western concepts of land ownership in fundamental ways: Most of the distinctive characteristics of rist are related to the fact that a field held by rist rights is not thought of as an enduring unit with fixed boundaries and a permanent location. It is not an object . . It represents a share of a much larger tract of land held corporately by the descendents, through any combination of male and female ancestors, of a legendary figure who is believed to have first held the tract of land as his rist. The individual's rights in a particular field are always limited by the rights of other descendents of the first holder, and his ability to obtain or maintain possession of the field is subject to the control of a corporation composed of all those recognized descendents who hold land in the tract . . . rist refers to the right a person has to a share of land first held by any of his or her ancestors in any line of [cognative] descent (1973: 11-12). The rist system was rural and institutionalized from the ‘bottom-up’. In it, all male and female descendants of the individual ‘founders’ were entitled to land from both their mothers’ and fathers’ lineages, where individuals had usufruct rights to family land which could not be sold or given away outside of the descent group. Until land reforms were instituted, newly married couples generally inherited small amounts of land and livestock from their parents through marriage gifts, but these were insufficient to sustain a growing household. They obtained additional land by incorporating marginal and unoccupied land or common grazing land. Litigation was another means to gain land access. While access was relatively egalitarian and the incidence of landlessness was low, population increase and legal conflicts proved to be the bane of the rist system. In southern and eastern Tigray, where the study site is located, the inheritance-based system was locally and spontaneously transformed into a chiguraf-guoses or ‘village’ tenure system where residency, rather than kinship, became the defining criteria for land access, largely because extended urban families threatened to take over rural land. While this system was more egalitarian than the rist system, allowing all residents to access land regardless of occupation or capacity to farm and responding to demographic change through periodic land reallocation, it led to insecure property rights, encouraged fragmentation of landholdings as well as successive reduction over generations, and prohibited accumulation (Beyene 2003, citing Bruce 1976). Inequalities in land access were much more common and problematic in southern Ethiopia and it was this that compelled the Derg Regime, upon taking power, to implement a “Land to the Tiller” programme. The main features of the 1975 reform were: 1) all rural land was nationalized and anyone who would personally cultivate the land could obtain up to 10 ha; 2) only women with no other means of livelihood could hire labour to work this land; 3) land sales, exchanges, or transfers were prohibited; and 4) 16 tenants and labourers would own the land that they worked at that time. The implications for private land access in Tigray were not dramatic, but the traditional rist institutions were eliminated and land distribution was arranged by the State, which at the same time created new local institutions: Peasant Associations (PA) (which allocated land), agricultural collectives, and cooperatives were established and took over most functions formerly fulfilled by traditional institutions. In Tigray, the reforms couldn’t be implemented immediately because of armed resistance on the part of the Tigray People’s Liberation Front (TPLF) which waged a 17-year war against the Derg Regime. Rather, in Tigray, the TPLF and the government implemented their own versions of land reform, which barely differed, in areas that they respectively controlled. Since these areas often shifted, this led to high levels of tenure insecurity: reforms created by one party were undone by the other (Beyene Ibid). The TPLF established the baito (community or village council) rather than the PA, which implemented reforms at the local level. Beyene summarized the land tenure rules in Tigray over the past three decades: 1) A person needs to reside as a farmer in a tabia (village) to get land from the tabia where, in general, the minimum residence required was two years. 2) Property-sharing in the case of divorce or death, regardless of capacity to cultivate (e.g., due to age or sex). 3) Generally a minimum age qualification as an adult: 22 years for a male and 15 years for a female. Adults get equal shares (each adult is entitled to receive one-half gibri – the local land unit), and four underage children count as one adult. 4) Rights are vested in individual members of a household: particularly husband, wife, and adult sons/daughters had titles to the land allocated to the household. 5) Area and quality of land were considered during land distribution. In Tigray, arable lands are classified into three categories according to soil depth, which is closely related to soil fertility. Distance and location of plots were also considered in most villages. Individuals were allocated both near and far-away plots. The change in government in 1991, when the TPLF and the Ethiopian People’s Revolutionary Democratic Front (EPRDF) assumed power, didn’t alter the land reform, but in 1995 the reforms were incorporated into the constitution, and power and administration were decentralized to nine Federal States. Whereas the central government retained ownership of all land and prohibited land transfers and sales, the regional governments were empowered to establish land distribution and registration regimes. Land registration was introduced in Tigray in 1998. It covered most farmers and signalled the end to land redistribution: Beyene (2003) argued that the main reason for ending redistribution was that landholdings had become so small that they were no longer economically viable. People have therefore had rights to the same holdings for a relatively short period of time: until the redistributions ended, tenure was too insecure to permit them to make long-term investments. As of 2001 it could be shown that, across Ethiopia, the land reform tended to redistribute land amongst those who had land but not to those who were landless, a trend that may have been exaggerated by a loosening of restrictions on land leasing which meant that farmers with access to oxen could lease or sharecrop land from those without access, thus resulting in a growing strata of richer peasants, although inequalities and landlessness are less of a problem in Tigray than in other regions (Mersha and Githinji 2005). In the FAO project area in southern Tigray, the landless young now constitute 10 to 25% of the population. Landholdings have fragmented to the point where they can no longer provide the food and income needed to sustain an average family: most landholdings are less than one hectare and many are only a quarter of a hectare. For this reason, families cannot divide land among their children (FAO n.d.). Population pressure and AE made for conservation purposes (see Section 2.D) have virtually eliminated the possibility to allocate village land to new families or migrants, so that many young families are now de facto landless (FAO Ibid, Beyene 2003). Sharecropping and, to a lesser extent land rentals, provide the principle legal means for land-poor households and landed households that do not have access to oxen or to someone to plough their land (including the majority of FH) to use their land to produce grain. On the other hand, it provides house- 17 holds that have sufficient labour and oxen to access additional land. Hoben (1973) noted that sharecropping (sometimes referred to in the Ethiopian literature as ‘tenancy’) was prevalent under the rist system for similar reasons. He also noted that the main disadvantage of sharecropping was the lower rate of return for the plot owner. He described the typical arrangement: tenants had to supply half of the seed, both of the oxen, and all of the labour, in turn receiving from a half to three-quarters of the harvest, depending on the type of grain (Ibid: 49). Amare (1999) reported that, with the land reform, the incidence of sharecropping decreased. Over time, decreasing land availability and increasing competition over land to sharecrop has led to a reduction in the proportion of the crop that is given to the sharecropper and increasing demands on sharecroppers to provide additional payback to plot owners (Beyene 2003). This would potentially counteract the disincentives that sharecroppers have to supply labour or other inputs which results in lower yields in comparison with owner-operators that Pender and Fafchamps reported for Tigray (cited in Teshome 2003: 133). Beyene (2003), in fact, found in his study villages in Tigray that sharecroppers managed the land in the short-term in the same way as owner-operators, since they are also interested in maximizing yields. However, longer-term management of such plots differed. Plot owners tended to sharecrop out plots that were further from their homesteads, as well as those plots over which tenure was less secure; plots that were sharecropped out received less manure and fertiliser and fewer trees were planted on them. In Hintalo Wajirat Woreda, where the present study site is located, Teshome (2003) also reported that mougia (a monetary inducement) may be offered by the sharecropper to motivate the plot holder to lease out his/her plot; together, the mougia and the quality of the plot serve to determine the share of yields. He gave the reasons that farmers in two irrigation systems leased land out: being female and thus not being able to plough (50%), not having oxen (10-20%), not being able to purchase fertilizer (50-60%), shortage of household labour (30-40%), uncertain access to water (60-70%), and the fact that farming interferes with off-farm activities (10-20%). He reported that the share of the yield received by the plot holder was usually one-third, and that household grain consumption was a function of this yield, where the grain produced was consumed within four to seven months, leading to severe food shortages. With respect to the relation between land tenure and development dynamics in Tigray more generally, Beyene (2003) argued that the lack of economic viability of landholdings, and tenure insecurity created by inter-generational instability, are contributing factors in land degradation and economic stagnation. He relates this to what Hoben (1973) argued is a key feature of the traditional rist system: the inter-generational discontinuity of the farm and the farm family, where "neither the household, nor the homestead, nor the estate of rist land upon which they depend, are enduring units of social organization" (Ibid: 64). Beyene argued that the current land allocation system prioritizes equity in land access over economic viability of landholdings and creates perverse incentives for landholders. He found in two villages in Tigray that there is a correlation between size of land holdings distributed in the villages and size of household, but “the per capita land allocated consistently decreases as the size of the family rises” (Ibid: 141). This, he argues, has driven people to establish smaller households. More land is distributed to household heads in comparison with adult children living in the same household, so that it is advantageous for people to establish their own households as soon as possible. Aging households lose access to their children’s labour, and household resources are often divided to facilitate this process. Upon death of the parents, the land is further divided between offspring. The process of land division is perpetuated inter-generationally, leading to land fragmentation. Thus, “. . . strategic economic development and longterm decisions, such as consolidation and ‘conservation’ of the farm, are undermined through the practice of redistributing land and its subsequent successive fragmentation . . the process sustains social conflict, competition and struggle over access to land” (Beyene n.d.: 16). He summarized the dynamic as follows: The dimension along which land holding varies . . is the developmental stage of the household. Normally, young people . . face the problem of access to land. Older people . . are landholders. In between these age groups, there are households at different stages, with different composition and varied dynamics of member mobility. The household developmental cycle continuously reproduces differences in the type and level of risk households perceive at any point in time (Ibid: 17). 18 Beyene did not discuss the significance of FH in this process,17 but we argue herein that familial instability, together with discrimination against women in resource allocation, can only contribute to increasing landholding fragmentation, decreasing size of holdings, insecurity of tenure, and hence destitution and resource degradation. As we demonstrate in this report, for women in Tigray, this differentiation becomes acute and often forces them into destitution, due to (a) marital instability and the subsequent high divorce rates; (b) what have been until recently high mortality rates or male absence due to civil war; (c) higher female household dependency ratios; (d) unequal division of household assets (with women accessing fewer resources of lower quality); and (e) unequal opportunities to farm and to participate in civil life. High rates of female household headship in turn further promote fragmentation and decreases in average landholding size. Women household heads are more dependent, ipso facto, on common land resources, but many such resources are in fact of minimal use to them due to enclosure, which in general are not governed with their resource requirements in mind, and those that remain open are under increasing pressure precisely because they are relatively free access and increasing numbers of people, both female headed and young households, depend upon them for bare subsistence. 2.C Women’s Assets, Female Household Headship, and Poverty We argue in this report that there are three factors that influence women’s land access in the Ethiopian highlands in general and in Adiarbaetu in particular. The first is what Fafchamps and Quisumbing (2001) signal as the continuity between traditional gendered land access and land reforms, where both manifest formal equality between men and women, while in practice resource allocation has been and still is biased against women. Local traditions regarding gendered resource access and household dynamics are anchored in marriage and divorce practices that were associated with the rist system. Reforms have both created continuity with and provoked changes in marriage and divorce practices and their relations to land access. This means that current inequalities between the sexes in resource access can be attributed to a) persistent local traditions regarding asset divisions in marriage, at divorce and upon death that are biased against women; b) a ‘hangover’ from such divisions in the not-too-distant past, which have affected a large number of women who are still alive today; c) biases against women in post-reform land allocations by the baito or PA; or d) some combination of these factors. Although Fafchamps and Quisumbing have studied the contemporary importance of marital assets to women’s household bargaining power across rural Ethiopia, there appears to have been little effort to date to determine their dynamics in Tigray, where gender relations appear to be more egalitarian than in other parts of the country.18 Marital assets, marriage, divorce, death, and land allocation pre- and post-reform are therefore discussed in turn, with the objective of explaining, first, why it is that FH are so numerous and, second, why it is that FH have fewer assets than MH. Within this, FH poverty and numerical importance are, we hypothesize, a result of the second dynamic, which Beyene (2003) in part elaborated upon: household formation and dissolution processes. The third dynamic, which we discuss in detail throughout this report, is the particular form that CPR management has taken post-reform, which has largely deprived FH of access to botanical resources that are vital to their welfare and livelihoods, forcing them, as well as other land-poor and destitute households, to over-exploit open access resource areas. Both FH and the plant resources that are essential to them have been largely neglected in development efforts: this ‘third’ of the population, and this ‘third’ of the livelihood resource base, have either gone unrecognized, been culturally defined as useless, or simply been neglected. 17He did find that sex was not a determining factor in the amount of land held in two villages that he studied in Tigray, but it was a determining factor in the amount of land actually cultivated, and he noted that female headship could be associated with vulnerability and poverty. 18Fafchamps and Quisumbing (2001) concluded that, as one moves from north to south across rural Ethiopia, women's status and therefore possibly their bargaining power with respect to access to household assets and decision making declines. The relative degree of gender equality that can be found in the north, they and others argue, is related to the historical predominance of the Orthodox Church and of the rist system in the region. 19 2.C.1 Marriage Arrangements and Property First we discuss traditions under the rist system and then during the post-reform period and contemporarily, but it must be kept in mind that there is substantial continuity between all of these eras. The discussion is made difficult because of the differences between formal and de facto rights and practices and because of the fact that local tradition plays a strong role even in post-reform land allocations: community authorities today appear to be reluctant to become involved in familial or private matters, particularly in relation to what happens to assets upon marriage and divorce. Also, no coherent discussion was found of these practices in Tigray in particular, so we must draw upon findings from studies that explore these issues in other highland areas. Hoben (1973) described marriage practices among the Amhara where the rist system, the Ethiopian Orthodox Church and the plough-based farming system predominated. His work is often cited by those who discuss traditional property arrangements in Tigray, so that it is possible to assume that the traditions he described were (and quite likely still are) valid for the study area. He described the ideal composition of the farming household: The minimal labor force required by a complete farming household must include a man who is responsible for ploughing and most other agricultural tasks, and a woman who is responsible for the management of food preparation and most other routine activities at the homestead as well as the provision of firewood and water . . . It should also have a boy of 7-12 years old (a girl can be substituted) to herd livestock; a girl over seven . . . or a young divorcee to help the mistress of the house with her work . . and a young man to help with the arduous year-round task of ploughing (Ibid: 44). Households were monogamous and first marriage occurred early: boys married in their late teens or early twenties and girls married as early as twelve or thirteen, “ideally just before menarche.” Most marriages were arranged and the couple did not meet prior to marriage. Matches were made by parents or at times elders, who attempted to ensure that both parties had actual or potential rist land rights. Property arrangements preceded marriage: "The object here is to make certain that equal amounts of moveable property will be given to the bride and groom" where cattle were an important part of endowments. The first marriage resulted in the creation of a new household by pooling labour, cattle and other livestock, and any cash given to the bride and groom in equal value by their respective parents, and eventually by pooling rist rights to land, usually upon the death of the parents. Then, "two men are chosen to represent the interests of the bride and groom respectively in future marital disputes or divorce proceedings" and banns are read; the last phase is the wedding. Weddings were major feasts given simultaneously at the bride and groom’s homes (Ibid). Marriages within the community were not favoured. Exogamy forbid marriage between those who shared a common ancestor in any line in six or fewer ascending generations, so that the bride and groom had at least a large part of their rist rights through different descent lines. Virilocality also predominated: a woman was expected to leave her native village to join her husband’s household. The couple remained with the husband’s parents for the first few years and contributed to the parent’s household before establishing their own residence. The groom’s parents would usually provide a small amount of land for their son to permit him to establish his homestead: where community land was available, it could be cleared in proportion to the amount of labour that was available to the new household. Hoben further explained that, in the rist system, land was classified into three types: ‘father’s rist’, which is land traced through the father's descent line, ‘mother’s rist’, which is land traced through the mother's descent line, and ‘wife’s rist’ which is that traced through the wife’s descent line. “. . . a man can claim 'wife's rist' only after his wife has given him a child and can keep it only so long as he continues to support that child. In other words, a man does not have any rights to rist in virtue of his marriage to his wife but only as trustee or custodian for the children he has with her. For this reason, wife's rist is also referred to as 'children's rist'" (Ibid: 136). Men could thus acquire new rist rights through every woman 20 with whom he had children: even when children were born out of wedlock, this did not negate their rist rights and hence their father’s claim to their mother’s rist. There were and are, however, other types of arrangements regarding marriage and marital assets. Woldegiorgis (1999) pointed out that, in livestock rich areas, parents typically would provide their marrying children with two head of cattle and up to ten to fifteen sheep. When one of the spouses could provide fewer animals or other compensating resources, the other spouse would be granted a certain number of animals as personal property, but where the poorer spouse could "earn half the value of any offspring as compensation for his or her labor input into the care of the animals" so that, even in this case, over time the couple’s livestock assets could equalize. Another arrangement reveals the degree to which the type and amount of assets that spouses brought into marriages affected their position within the household, as well as conditions of property division upon divorce: A different conjugal arrangement is one in which one spouse doesn't have any of the necessary assets that are required for a proper semanya marriage [such as that described above], and comes in as a hired person. This person does not have rights to … property in the household and gets an annual wage that usually amounts to a complete set of clothes and two quintals of grain or one sheep, in addition to room and board. This is an arrangement that arises primarily from labor deficiencies that single men or women sustain - male draught labor in the case of females and female domestic labor in the case of men. Such types of relationships, initiated because of the need to fulfil the economic and consumption needs of both parties, are frequently terminated when one of them becomes dissatisfied . . . (Ibid: 43). Upon divorce, the hired person leaves only with any accumulated salary. "In fact, there are quite a few people, usually women, who earn their living as such wage earners, staying in different households for varying periods of time. Older widows or divorcees may often find this an option if they are not in a position to be supported by their children" (Ibid). Amare (1999) also found that, in Amhara, women who have brought no resources into a marriage are easily replaced by more propertied women, and that divorced women with property are much more likely to remarry. Men are less likely to divorce women who are well endowed, which also means that poorer women are likely to "remain destitute and single as a result," their only alternative being to marry equally poor men and thus remain in poverty (Ibid: 49). 2.C.2 Women, Marriage, and Land under the Reforms Currently, the Ethiopian constitution provides for equal rights for men and women to use, transfer, administer, and control land, and to equal treatment with respect to divorce and inheritance. Tadesse (2003) pointed out that, until 1987, marriage was a prerequisite for the acquisition of land rights: the Land Law promulgated that year abolished this precondition, so that land could be granted to anyone who had reached a minimum age. Since the Derg reforms were implemented, the principle source of land for married couples has been the baito or PA. The study carried out by IFPRI (Fafchamps and Quisumbing 2001), based upon a survey of 1500 households in 15 villages across rural Ethiopia in 1997, found that the PA directly allocated two-thirds of the land held by households, whereas dowry and bride price account for only a small proportion of land transfers. Some researchers argue that FH’s inferior land assets are therefore directly attributable to gender bias in land allocations under the reform. Woldegiorgis (1999), for example, argued that the land allocated to FH was usually of minimum size and of poorer quality in comparison to that allocated to MH. Amare (1999) found that the proscriptions on women’s use of oxen for ploughing is interpreted by many local authorities to mean that land will be cultivated by men, and therefore they register household land in the name of the husband. He also argued that bias against women in PA land allocation prevailed upon death of the husband, resulting in a disproportionately large reduction in household landholdings when a male head of household died, whereas no land was taken from the household when a wife died, because the husband was expected to marry soon after, and 21 a woman might be taken care of by her son. Widows who formed FH therefore often had insufficient landholdings. Tadesse (2003) argued that baito or PA land allocation was even more biased against married women than against FH – as household heads, these women are members of PAs whereas married women are not, and baito or PA members had priority in land allocation. Wives had been excluded from PA membership since only household heads were registered – it was estimated that, in 1990, only 12% of PA members were women. This increased to around 20-25% by the early 1990s due to male mortality and destabilization resulting from the civil war, drought and famine (Woldegiorgis 1999), but this enrolment still refers mainly to female household heads. Still, Woldegiorgis’s argument is only convincing insofar as it can be demonstrated that the registration of land in men’s names itself leads to inequality in intrahousehold decision making or in the division of resources upon divorce. Fafchamps and Quisumbing (2001) made some important observations in this regard. First, they found that single adult men or women living outside of marriage constituted 22% of the sample: these tended to be widows and divorced women. Further, it was found that a large proportion of all of those respondents living in monogamous marriages were previously married: 35% of the husbands and 22% of the wives. A third of the marriages had ended in the death of one of the spouses, whereas the remainder ended in divorce or separation. The highest proportion of single women was found in Tigray, which they posited might reflect the high male mortality associated with the civil war. These data demonstrate the degree to which divorce is prevalent, both among those who are married and among those who are single. The same study reported that, while the baito or PA must still formally support land transfers following marriage, divorce or death, in practice they do not usually intervene except when villagers directly solicit it. If it can be supposed that the PA or baito allocated land equally to men and women within households based upon the formula mentioned by Beyene (see page 5, no. 3), which some studies have indicated is the case even if the land was registered to male heads, legally it should be equally divided upon divorce or death. However, if upon divorce or death the baito or PA does not interfere, then it is quite likely that local customs that are derived in part from the rist system enter into force. What happens to women’s assets upon divorce or upon their spouse’s death therefore continues to be a very important determinant of their household’s welfare, and a major reason for them falling into destitution. In other words, if women receive land from the baito or PA, irrespective of whose name it is registered in, they may lose control over this land upon divorce if local customs so dictate, and the PA or baito would do nothing about this unless specifically requested, which local customs would also tend to dictate against. Marriage arrangements today appear to differ little from those that were described under the rist system. In Tigray, however, the land proclamation raised the age of marriage: men must be 22 and girls must be 16. While arranged marriages predominate, increasingly couples choose their own partners, but such marriages still need to be approved by parents, where bloodlines of the spouses are considered to determine a potential spouse’s character. The endowments with which the groom and bride enter marriage has supposedly become a secondary consideration since these differ less today than they did in the past (Amare 1999), given that most household land is no longer inherited but rather allocated by the baito or PA. 2.C.3 Divorce and Death of the Spouse Divorce in Tigray is frequent, as Fafchamps and Quisumbing (2001) also showed for other parts of rural Ethiopia. Unless a couple marries in a church (and few do, given the strict religious rules that must be followed in such cases), marriage is seen as a civil contract that can be readily dissolved if either partner so desires. Bauer’s study (1985) in Tigray in the early 1980s showed that the average length of a marriage was only 7.5 years. He found that one incentive for divorce for men was that having children by a number of women permitted them to claim more land. Amare (1999) reported that high divorce rates find many grounds: laziness, unwillingness to perform gender-defined duties, illness, barrenness, and economic decline of the household. He suggested that the divorce rate may have declined since reforms due in part to the ‘disastrous’ consequences of household land division. 22 Although legally each spouse is entitled to equal shares of a household’s resources upon divorce, the literature indicates that culturally this is indeed not the case: women generally give up all or most of their share of the resources (Amare 1999, Bauer 1985). Amare (1999) described the general process and outcome of divorce in Amhara. "Divorce settlements are carried out by elders chosen by the ex-spouses" and only rarely go to court. In principle, land is generally divided in proportion to the number of family members staying with each divorcee (which Fafchamps and Quisumbing 2001 also ratified), where typically young children remain with their mother, and older children decide for themselves with whom to live. In reality, however, men are favoured in the division of land and children since they are thought to be the main producers and best able to defend children’s interests. Commonly-held livestock and other major assets would also be divided equally but, if there was only one ox, this would usually be allotted to the man. Houses, which are difficult to divide, would generally be allotted to men, and women would receive compensation, in part because it was assumed that women would return to their native villages. Divisions of assets on the basis of the principles mentioned above are only part of the picture. Hoben, when referring to the rist system, found that "A well-to-do man often allocates the use of a field or two to a divorced wife if she has children by him and has not married again. She, in turn, gives the land out in tenancy” (1973: 139). This is close to what appears to be the case in Adiarbaetu today (Chapter 5), and it implies of course that women are not receiving equal land shares upon divorce, and that their rights are still dependent on the largesse of their ex-spouses (that is, they are ‘usufruct’ rights that depend on the maintenance of their ex-husband’s minor children). Woldegiorgis (1999) reported that, if land is registered in both names, it will be shared equally between husband and wife upon divorce. However, Bauer (1985), referring to the situation in Tigray in the early 1980s, reported that the amount of communal and individual property that each spouse received upon divorce depended on the type of marriage contract. While asset division upon divorce in practice appears to differ depending on local customs, nearly all researchers agree that, up until land redistribution ended, virilocality was a major factor that deprived women of land to which they were entitled upon divorce. Fafchamps and Quisumbing (2001) argued that, if women remained in their husband’s localities, they could retain land from their marriage, but virilocality discouraged women from doing this: women would often return to their native villages where their kin resided and subsequently remarry. Amare (1999) discussed this phenomenon in some depth. When women remarried, they had to give up control of the land they gained in marriage since policy prohibited the use of land in a locality different from one's place of residence. Rather, the PA would reallocate a woman’s share to her ex-husband since he was expected to remarry within the locality. Upon divorce, elders and local officials often allocated most of the children and land to the husband because men were thought to be the 'producers', whereas women were seen as outsiders. On the other hand, if a poor man married a wealthier woman and moved to her locality then this situation could be reversed. But the usual outcome was that women left their land behind and took their share of movable property with them to their native villages, and then attempted to remarry. It was also reported that, if after divorce women decide to live independently in their husbands’ village, several factors might also work against them. Partitioning the household's land would be likely to reduce their chances of remarriage: single men are reluctant to marry these widows because there is stigma attached to the act of “coming into a woman's house,” and also because “it is not considered proper to marry the ex-wife of a man who could be a neighbor or relative" (Amare Ibid). Further, the amount of land a divorced or widowed woman would bring into a new marriage would be insufficient to form the basis for a viable household, particularly considering local officials' bias in land allocation in favour of men. Women’s access to land was therefore still realized mostly through marriage and not through independent means. As already indicated, land reallocation ended in Tigray after 1991, which also ended land reallocation upon divorce and therefore any intervention in divorce on the part of the baito or PA. This has allowed women to leave their former households and localities and still retain control of their share of land even after remarriage. “Quite a few women arrange to have land from a previous marriage cultivated for them in this manner, as they maintain their residence in another locality" (Amare 1999: 48). Nevertheless, some social disapproval is directed at women who successfully retain land from their previous marriage 23 and then marry other men in the same locale, since remarriage does not give these women greater rights to land, whereas the men they marry are thought to have gained an unwarranted addition to their land resources. The other factors discussed above that are not related to virilocality are, however, probably still depriving women of their access to land (even baito or PA allocated land) upon divorce. While discussion of the empirical findings in this report are reserved for later chapters, it is interesting to note here what women focus group participants from Adiarbaetu said in relation to the high divorce rates that they reported, which emphasizes the degree to which customary marriage and divorce relations prevail. The first reason that they gave is poverty, since women who do not have dowries cannot marry formally (with a contract); if a woman cannot do so, “then she can live with someone who doesn’t respect her or leaves her easily for another woman, a better woman with more money and resources.” Husbands leave wives who have no assets for women who do. “There is a need for proper money for a proper marriage. The dowry is not only for him but for his relatives as well. The expected dowry is five cattle, five goats or sheep, and land.” Yet another reason given was that “There is a culture of having mistresses. Women want to be with men even if they are married. This leads to divorce as well.” 2.C.4 Inheritance Even though land that is inherited or given as dowry represents only a small fraction of households’ holdings today, it is still important given that no new land reallocations are taking place, and the land that is inherited potentially remains with a person after divorce or death of the spouse. As mentioned previously, inheritance in the rist system did not formally discriminate between male and female heirs, and as well current law stipulates that all children receive equal shares of parents’ assets upon death. According to Woldegiorgis (1999), the Civil Code states that, if a man dies intestate, his wife does not inherit from him at all: the children and then other relatives inherit and, if there are no other heirs, his holdings revert to the State. Women may remain on their half of the land if it was common property or as a guardian to their minor children. In practice, irrespective of the legal situation, the likelihood that women equally inherit or are otherwise bequeathed land before or after their parents’ death depends upon local traditions. Amare (1999) found in reference to the Amhara that inheritance was in practice biased toward sons, since families sought to maintain land within their descent groups and, if the land passed through daughters to son-in-laws, it would then be likely to pass to the son-in-law’s decent group. Further, when landholdings are small, as they generally are, land would often not be divided amongst all of the heirs, but be passed on to the male heirs in order to avoid landholding fragmentation. Woldegiorgis (1999) further reported that some girls received land equally, others in a 2:1 ratio with respect to their brothers, whereas other parents provided land to their sons only, and instructed them to care for their sisters. A girl is expected to marry and, if she is widowed, she is expected to remarry. Today land divisions are likely to remain biased against women as families strive to avoid fragmenting their holdings even further. What becomes, then, of men and women after they have divorced or been widowed? An important determinant of the outcome is gender differences in the capacity to manage an independent household. While women, even if they have children to care for, are thought to be able to manage without remarriage, men are not thought to be able to do so without domestic labour. Remarriage is an option for women and necessary and inevitable for men. Partial households of either sex can join other partial households of either sex. For those women who attempt to earn their livelihoods independently on the land that they receive from their ex-spouse or parents, their living standard is likely to decrease due to unfairness in the division of land, as a result of a reduced share of the household's wealth upon divorce, or because of deficiencies in their initial contributions to the household (Amare 1999). A man who does not find a partner to marry or another FH to join with may hire a woman to carry out domestic tasks, and female heads may hire a male to perform male agricultural tasks while they themselves manage field production. Women focus group respondents in the study area reported that many of the female household heads’ husbands died in the civil war. Some chose to remain single, committing themselves not to remarry after their husbands died, and “Even those women who have many children will often opt to 24 stay alone because they fear how their children will be treated if they remarry. Sometimes men are reluctant to marry women who have children because they feel they will be burdened.” 2.C.5 Decision Making, Bargaining Power within Households, and the Gender Division of Labour Amare (1999) noted that generally households in Amhara lack the economic and consumption units that are found in many other African households. However, the amount of assets that men and women bring into a marriage, together with the gender division of responsibilities, generally determine decision making and bargaining power that each sex has within households. Writing on the basis of his observations among the Amhara in the 1960s, Hoben (1973) noted that decision making authority regarding production depended upon the type of marriage contract. If it was arranged by dowry, only the husband had the right to manage production, which was the case with most first marriages. But if each partner contributed equal shares to the formation of the household through a marriage arrangement, they jointly managed household production and assets. Amare (1999) also found that decisions about common property were joint, whereas decisions about personal property were the prerogative of the individual owner. Nevertheless, Amare also noted that men and women have their own spheres of daily decision making. Women, he reported, are responsible for homegardening and small livestock rearing, whereas men are responsible for decisions about what and how much to plant in fields. Women manage the grain stock to ensure adequate seed supply for planting and, if there is not enough seed, they sell liquor to obtain it or get it from kin or friends. Men exclusively buy and sell large livestock, wood, charcoal, and larger quantities of grain, and only men travel to regional markets, whereas women sell eggs, vegetables, chickens, and home-made liquor at local markets. Amare found no indication that there is a separation in the use of income: both men’s and women’s income is likely to be spent on clothes, kitchen or agricultural tools, seed and food. However, it is the male household head that is “primarily responsible for finding the resources to pay taxes with, who engages in claims and negotiations over land, and to whom directions from higher authorities are passed on. If there are no adult men present, the adult woman takes over this role" (Ibid: 45). Even so, economic decisions such as sales and purchases are to be jointly decided and, if this is not the case, it can be grounds for divorce or separation. The situation is likely to have changed with the reforms, particularly since most land that couples now access has been allocated through the baito or PA, which diminished the proportion brought into marriages by other means. Fafchamps and Quisumbing (2001) reported that, across the 1500 rural households surveyed in Ethiopia, control over finances and productive assets was usually joint. Decisions on what to produce were found to be the purview of the household head. By extension, decisions to rent out land or to give it away, for instance to children, are also predominantly taken by household heads. Decision making about livestock was found to be more complex: most was held jointly, and individually held livestock nearly always belonged to the head. The right to sell livestock, even when jointly owned, and to keep the proceeds, predominantly fell to household heads. Control over expenditures was also found to be centralised in the hands of the household head. 2.C.6 Female Household Headship and Poverty The major constraints that FH are commonly reported to confront are diminished household assets, including land, livestock, and farming implements (particularly ox ploughs), inability to plough their own land, lack of livestock feed, and shortage of male labour. Because FH are often unable to plough their own land, they must sharecrop out and, in the process, relinquish a major share of the harvest. Some women find means to mobilize oxen and male ploughing labour through ties with kin or ex-spouses, through labour-sharing arrangements, or by trading dung or crop residues but, in the latter case, this means that they have less of these resources for themselves (Amare 1999). Woldegiorgis (1999) mentioned that such ploughing arrangements often result in bad timing and poor harvests, since men plough women’s fields only after they have ploughed their own. She also noted that certain policies stipulated that PA members should plough the land for women farmers, but that this was not implemented. Graz- 25 ing animals, and cutting and piling hay and grain stalks, are also considered to be male tasks, for which FH who have no older sons need assistance. FH may have to sell animals that they have retained from marriage due to the lack of feed and the need to cover such expenses. Not surprisingly, Amare noted, “most female divorcees do not have any animals" (1999: 67). His conclusion explains perhaps why so many women do not contest the unequal division of household assets upon divorce: "the dire economic conditions awaiting women who choose to remain single, therefore, convince many divorced women to abandon their share in land," and look for a man with land to marry (Ibid: 67-68). The most recent major study of the relationship between female household headship and poverty in the north-eastern highlands of Ethiopia is found in Sharp et al (2000), which was based on a stratified random sample of 2,127 households, 14% of which were found to be ‘destitute’. Of the total sample, 21% were FH. Of the MH, 8% were classified as destitute, in comparison with 35% of FH. They found that the most important factor determining social differentiation and thus destitution was not how much land households legally owned, but rather how much they cultivated themselves: “the typical destitute household gives up control of half their farmland. Their effective control over farmland falls from .55 to .27 hectares…." (Ibid: 89). More than 70% of those that sharecropped land out were FH. Over half of the destitute households were not farming at all, compared with six percent of the non-destitute households. What was common across nearly almost all destitute households (92%) was that they did not own oxen. "Since oxen are perceived as the property of men, it is not surprising that oxen ownership is concentrated in households dominated by men: two-thirds of MH own one or more oxen, but three-quarters of FH own no oxen at all" (Ibid: 90). MH owned 91% of all livestock, almost three times that which FH owned. Beyene (2003), however, argued that social differentiation is not so much related to oxen ownership, as to the availability of male labour for ploughing. Since oxen are used only for ploughing and threshing, they must be used within a very concentrated space of time during the year. The number of oxdays per year required by any one household is quite limited, therefore it would be highly inefficient for all households to own oxen. Oxen require feed year-round, constitute a large initial cash outlay for their purchase, and illness and death of oxen are continuous threats. In order to efficiently utilize oxen, they need to be hired out to other farmers, which requires intensive use of male labour. He further reported that, across Tigray, the structure of oxen ownership is strikingly similar: very few households have more than two oxen, less than a quarter own two, some forty percent have none, and the rest have one. He concluded that: The number of oxen sustained is determined more by the community level needs . . The system of exchange and the fact that individuals control some rural resources, however differentiated, means that there is a mutual interdependency among villagers. These exchange systems allow oxen holders to make their assets available to the community members and to utilize them. These exchange systems, to a great extent, are reciprocal in value (Ibid: 154). This means that many households that do not own oxen do not have access to the other resources (e.g., male labour) required to sustain and intensively use them: for these households, exchange with oxen owners is the best means to avoid the risks associated with oxen ownership. This implies that it is at least equally important to examine FH’s access to male labour. In fact, Sharp et al (2000) reported that other destitute and poor households share certain characteristics with many FH besides landlessness or near landlessness, sharecropping and lack of oxen. Although destitute households were found to be smaller in size, they have higher dependency ratios than non-destitute households. Household size may decrease as an effect of destitution or as part of the process of becoming destitute: divorce and separation are common consequences, as well as causes, of destitution. Destitute households are more likely to lose their older children, whereas better-off households are more likely to gain labour through adoption, absorbing adult relatives, and hiring workers. Destitution, they conclude, relates “primarily to differences in control (ownership access) over key productive assets, notably draught oxen and male labour" (Ibid: 77-78). 26 Another factor that was related to higher levels of destitution among FH was that they are more than twice as likely to include people who are disabled, which suggests that women with disabilities, or who have children or others with disabilities to care for, are less likely to get married or face being divorced or abandoned if they become disabled while married. Respondents reported that married women “who become weak” will be “thrown away by their husbands” (Ibid: 105). Further, elderly women who live alone appear to be an especially vulnerable group. Among the elderly destitute, FH are overrepresented (34%), while MH are under-represented (21%). The consequences are that FH are more likely to be forced to ration their family’s food supply (71% consumed less than three meals a day, compared to 59% of MH), suffer longer periods of food shortage (26% had no food stored compared with 39% of male heads), and are twice as likely to live in poor quality housing. Still, the authors concluded that the sex of the household head was not a defining variable in identifying absolute destitution because two out of three FH were not destitute. They admitted that their definition of ‘destitution’ was quite rigorous and, if it is assumed that the gender inequality found among the poorest population subgroup is reproduced across the entire population, “then the majority of FH are certainly worse off by most indicators than the majority of MH” (Ibid: 78). However, since MH outnumber FH four to one, ignoring poor MH would imply that many destitute households would be ignored, and many FH that were not destitute would benefit (inappropriately) from development aid. They also found that there are few differences between the livelihood activities of the destitute and those of the rest of the population: only one activity, carrying water for payment, was not practised by any non-destitute household. There are other activities that are described as low status or ‘poor people's work’, such as selling fuelwood, that are more performed by people who are destitute, but these are also sometimes done by people who are not. They found very few households that received any remittances or cash gifts, but more of the non-destitute households received something in comparison with the destitute, which suggests that the destitute are so marginalised that they are excluded from social support networks. Lacking income and assets, the poor often turn to their better-off relatives and neighbours for support during crisis, but most of the destitute households cannot do this, which increases their vulnerability. They are also less likely to be active in religious and ceremonial social institutions such as mahebers, senbetes, and idirs, since they don’t meet the criteria for equal participation: they cannot meet the cost and social expectations that come with membership (only 53% of destitute were found to be members of religious societies compared to 81% of non-destitute) (Ibid). They are not totally excluded, however: communities are vital sources of assistance, and the destitute often exchange resources (as described above for FH) in order to gain access, e.g., to ploughing services. But, in general, the destitute, Sharp et al conclude, “are looked down on by other community members and are excluded from many social activities and community based associations.” The discussions above have highlighted the reasons that households in northern Ethiopia in general, and in Tigray in particular, generally have equitable but inadequate access to land. They further elucidate the reasons that, in spite of formal equality between men and women in land access, FH are particularly likely to be poor and destitute. For both FH and MH that lack access to enough labour, land, and oxen, sharecropping and the use of CPR, as well as food-for-work and other community work schemes, are essential to survival. Below it is argued that FH are particularly dependent on CPR, and degradation and inequitable access therefore seriously compound their difficulties. 2.D Land Degradation and Common Property Resources Deforestation and erosion are major problems in Ethiopia as a whole, but they are especially grave in the northern highlands and in Tigray. It is estimated that half of Ethiopia’s farmland is severely degraded, where top soil has been completely depleted on about two million hectares. Because of this, even slight fluctuations in rainfall can cause great losses. It is further estimated that 50-60% of all rainfall is lost as runoff that carries away 2-3 billion tons of topsoil annually (Wolde-Georgis 1996). Esser et al (2002) indicated that the severity of soil erosion in Tigray results from the mountainous topography, torrential rainfall, and the low degree of vegetational cover. “Soil erosion is severe in all of Tigray and poses a major 27 threat to continued agricultural production . . Virtually all topsoil and, in some places parts of the subsoil, has been removed from sloping land leaving stones or bare rocks at the surface” (Ibid: 6). The loss of vegetational cover in Ethiopia and in Tigray has indeed been acute. It was estimated that, at the turn of century, 40% of Ethiopia was covered with forests, which was reduced to about 16% by the 1950s. Current estimates are that less than 4% is covered by natural high forests which, furthermore, are severely degraded (Wolde-Georgis 1996,, Williams et al 2003). It is estimated that “deforestation has been occurring at the rate of about 163,000 ha per year” (Williams Ibid: 14, citing Reusing 1998). In Tigray, as late as the 1950s, nearly half of the region’s land area was still covered in woodlands and forests, while less than 30 years later, in 1979, 9% remained (Wolde-Georgis 1996). According to Gebremedhin et al (2000), forests and woodlots currently cover about 1.6% of the land area. Nearly all construction materials are imported, fuelwood shortages are acute and, "Despite the fact that about 40% of the total land area is used for grazing, shortage of feed sources is the major livestock production problem" (Gebremedhin et al 2004: 2). 2.D.1 Resource Degradation Dynamics In spite of the fact that there are numerous studies on diverse factors that contribute to land degradation and deforestation in Ethiopia and in Tigray, there are few that provide integrated assessments of the driving forces and consequences (Feoli et al 2002). Deforestation has been a concern in Ethiopia for a long time. In the past, in northern Ethiopia, leaders progressively moved their capitals southward as localities ran out of water and fuelwood. This trend continued until the time of Emperor Menelik II in the latter part of the 19th century. Due to deforestation around Addis Ababa and elsewhere, Emperor Menelik II introduced measures to promote planting of eucalyptus as a cash crop. Across Ethiopia, the main cause of deforestation and forest degradation today is reported to be agricultural expansion, including growing livestock numbers. Land clearing for settlements and forest fires are also contributors (Williams et al 2003). In addition to land clearing for agriculture, other major causes of deforestation and forest degradation in Tigray included tree cutting for fuel, timber, and agricultural implements (Gebremedhin et al 2000). Population growth, which is estimated to be around 3% for Ethiopia as a whole, has also been cited as a major factor leading to deforestation and land degradation (FAO n.d.), but this growth alone explains little. Nedessa et al (2005), for instance, reported in their study of AE management across nine woredas in four regions of Ethiopia that if we examine the relationship between human population, livestock densities and land degradation, we see that Ahferom, the area with the highest land degradation (47%), has one of the lowest population densities, and a relatively low number of animals per unit area. Alaba, on the other hand, has three times the population density of Ahferom . . and three times the number of animals per land unit, but only 18% degradation. From this data it is difficult to conclude that degradation is directly caused by high population and animal density (Ibid: 15). Rather, livelihoods and institutional factors that determine how an increasing population uses the natural resource base are argued to be the critical intervening variables. The high level of dependence of highland populations on livestock, particularly cattle and oxen, couples human population growth with livestock population growth. As indicated in Section 2.A, livestock populations are very high. Researchers generally concur that they exceed grazing land carrying capacity (Amare 1999,, Benin and Pender 2002,, Gebremedhin et al 2004). Despite this general consensus, there has been little integral research on the contribution of livestock populations to degradation in Tigray or in the highlands. The site-specific research that has been done shows that heavy to very heavy grazing pressure significantly increases surface runoff and soil loss, and reduces infiltration (Mwendera and Saleem 1997), as well as significantly reducing ground vegetative cover and biomass yields, especially on steeper slopes, leading to the predominance of less palatable species (Mwendera et al 1997). 28 While grazing on agricultural land has decreased dramatically over the past few decades as land scarcity has increased, when present, it “contributes to soil compaction and the need for frequent tillage to prepare fields for crops, making practices such as reduced tillage less feasible" (Gebremedhin et al 2004: 276), thereby perpetuating the need for draft oxen. There are other critical relations between livestock and crop production at farm level that have changed drastically, reducing the viability of both. Crop residues were important to maintaining the structure and fertility of soils by providing organic matter, but these residues are now mainly used for animal feed and fuel (Gebremedhin et al 2000,, Williams et al 2003). Dung was used for fertilizer but, with fuelwood shortages (see below), it is increasingly used for fuel (Amare 1999). The use of chemical fertilizers has increased, permitting the cultivation of more marginal land, but most farmers still do not use fertilizer or do so only intermittently due to high costs, ignorance about its use, and its inappropriateness for some soils which increases the risk of crop loss (Ibid). All of these problems have given rise to attempts to find solutions particularly through permanent and seasonal enclosure of grazing and other land to permit grass regeneration. The question remains, however, about whether the management of such grazing lands resolves or contributes to the problems that are discussed above. The implications of such efforts are discussed below. 2.D.2 Land Tenure, Common Property, and Resource Degradation Considering the importance of common land areas in the region, there is surprisingly little literature discussing their tenure, history, or use prior to the State-led creation of village AE beginning around 1974. The growing body of literature dealing with natural resource management and common property areas has focused on these AE, particularly on woodlots and grazing land (see e.g. Benin and Pender 2002, Berhanu 2004, Chisholm 2000 and 2004, Gebremedhin et al 2000, Gebremedhin et al 2004, Keeley and Scoones 2000, Shitarek et al 2001, Shylendra 2002). It has been reported that, before the Derg Regime, much of the land in Tigray was effectively open access (Berhanu 2004); however this perception is almost certainly due to the fact that little research has been done on traditional CPR management regimes. For example, Chisholm (2000) noted that, in the villages he studied in Eastern Tigray, forests had been carefully managed by local communities for a long time, and there were well-established rist rights permitting households to use clearly delineated sections of the forest area. Such management systems generally survived turbulent political periods, the incursions of outsiders and civil war, up until 1975 when they were abolished under the Derg Regime: even so, these rights have endured to a certain extent. He argued that, for the villages that he studied, deforestation accelerated under the Derg Regime because of its lack of clarity concerning tenure rights particularly on unenclosed hillsides: ‘outsiders’ began to enter these areas and cut trees and, since communities couldn’t control this, they began to cut the trees themselves. The literature on grazing lands is also not very clear about customary tenure regimes: some argue that grazing lands were completely open access, whereas others report that grazing land has long been managed by community institutions. One study was found (Ashenafi and Leader-Williams 2005) that researched a customary CPR management regime and the changes brought about by the 1975 reforms. The study was done in Gera Keya Woreda of Amhara Regional State where the rist system had also prevailed. While the customary system they describe was, according to them, ‘unique’ to the area, this does not mean that similar systems were not pervasive across the highlands. The system, called Qero, entailed the enclosure of common land areas . . from any type of use by the community for consecutive periods as long as 3-5 years. The length of closure largely depended upon the growth and recovery of the Festuca grass and the need felt by the community for harvesting it . . the length of closure de- 29 pended on the success of the local crop harvest and on the frequency of drought in the Guassa area (Ibid: 547).19 The Qero system was managed by a headman (Abba Qera or Afero) who was responsible for protection of the area, and who was elected unanimously by all CPR users. Users groups were also organized into several parishes, where each had a headman esquire (Aleqa or Chiqa-shum) who answered to the Abba Qera. Every adult male had to patrol the common areas to find violators and ensure that they were reported. Subsequently, violators received a punishment appropriate to the type of violation. With the 1975 reforms, all land officially became the property of the State, so the Qero system, its bylaws and its enforcement mechanisms, ceased to exist. Local informants reported that the new AEs that were established were meant to be patrolled by militias but, due to corruption and inefficiency, they were barely protected. Although this changed in 1991, nearly two-thirds of current residents felt that the current management system is also ineffective and, although only about a tenth thought that there should be a return to the Qero system, half thought that resource management should be in the hands of the community and should have new bylaws and better enforcement (Ibid). Reforms have been seen to have negative effects on resource management both on individual holdings and in common land areas. Amare (1999) argued that the degree to which reforms created individual land tenure insecurity varied across Ethiopia and thus the implications for investments in land conservation and tree planting varied. In Tigray, Chisholm argued that resource degradation is principally due to ‘external shocks’, essentially drought and . . the impact of land reforms and other State legislation which reduced the rights of rural households over tenure of trees, or at least made such tenure rights uncertain. The effect of such policy changes was to convert the property rights regimes governing trees from private or CPR rights to de facto open access: households were then effectively induced to over-exploit these resources as a consequence of the perverse effects of State policy (2004: 8). Berhanu (2004, citing Yeraswork 1996), also wrote about the negative effect of reforms for indigenous management in Wollo and North Shoa, and attributed the severe environmental degradation that took place in these areas to State intervention that led to the collapse of customary institutions, leading to de facto open access. The Derg Regime did attempt to tackle the problem of deforestation and resource degradation. Soon after taking power and, with assistance from the World Food Programme (WFP), soil and water conservation activities were initiated on a large scale. In 1980, the WFP consolidated the small, fragmented projects into one support programme entitled “Rehabilitation of forest, grazing and agricultural lands (ETH-2488)” (Nedessa et al 2005). In Tigray, more than 80,000 ha of hillsides were enclosed to most use in order to regenerate indigenous plant species but, by the end of this period, most of the AE were either harvested or destroyed (Gebremedhin et al 2000). The reasons given for this are varied, and include inaccurate scientific and technical knowledge, a standardized approach that failed to consider diverse agroecological conditions, the lack of local participation, the perception that such measures were coercive, and the fact that they eliminated previously existing local natural resource management institutions, creating greater tenure insecurity (Berhanu 2004, Chisholm 2004, Gebremedhin et al 2004, Nedessa et al 2005). In 1991, the TPLF decided to devolve much decision making around conservation and reforestation to the local level. It created the respective institutions based on a local level participatory planning approach and, with the support of the peasantry, initiated "mass-based soil and water conservation efforts, including a considerable amount of voluntary collective labour, and . . . the setting aside of heavily 19The adaptability of such a local system with respect to local agroecological conditions is one of its potential hallmarks, since another recent study (Nedessa et al 2005) reported that current grazing land management is not responsive to fluctuations either in climate or in livestock numbers (see discussion below). 30 degraded areas for revegetation," also on a large scale (Chisholm 2004: 8, see also Nedessa et al 2005). In Tigray, unlike in some other parts of the country, the land that was enclosed was that which was most heavily degraded (Ibid). Up to 1998, over 260,000 ha of degraded communal lands had been rehabilitated in Tigray through the creation of soil and water conservation structures, revegetation through tree plantations, protection, or AE to facilitate natural regeneration of grass for grazing as well as woodlot production (Berhanu 2004). By law, AE are administered and managed at regional and woreda levels by the Agriculture Department, which is responsible for formulating policies, rules and regulations, developing strategies and guidelines for management and administration of natural resources including AE, and designing technical packages and interventions. Currently, the objectives of AE according to BoANRD are to: • • • • • • Halt and reverse land degradation Check the adverse effects of run-off Create natural resources highly demanded by livestock, human beings and the land Improve the micro-climate of respective places and thereby maintain environmental stability in the region Create habitat for wildlife Conserve diminishing biological resources, mainly forest trees, shrubs, herbs and grasses (Nedessa et al 2005, citing Forester 2000). However, there is no separate bureaucratic structure that is in charge of natural resources, and no staff within the Departments has specific responsibility for their management. Field staff are not trained in AE management and, “While they are technically qualified and highly motivated, they lack the competence to deal with the complex social issues involved” (Ibid: 26), which results in an emphasis on technical management and conservation priorities, rather than on local sustainable use. At community level, the baito is responsible for formulating bylaws, planning, implementing, and monitoring and evaluating natural resource management, with technical assistance from woreda specialists and Development Agents. There are also social courts appointed by the baito that are responsible for hearing cases and taking legal action against violators. In addition to AE, the Regional Government in Tigray has created restrictions on the use of specific tree species that are enforced by DAs who are assigned to each village for this purpose. Species restrictions apply to all common land areas, regardless of whether they are open or enclosed. Restrictions are divided into two categories: species that cannot be cut for any reason, and those that can be cut only with permission. Species that cannot be cut for any reason are those that are most endangered (see Table 66 for the list of restricted species). According to Nedessa et al (2005), the Tigray government has now drawn up a detailed plan to allow for the development and management of unenclosed hillsides based on the region’s land use policy, where communities will be authorized to design and implement their own regulations. Baitos will own such hillsides and the CPR therein will be communally owned and used; those hillsides that are not developed can be allocated to individual households as private land. Baitos will be able to select user groups and also to determine how much land will be allocated to whom. However, these researchers assert, the plan “does not sufficiently guide on the management and utilization of resources from the AEs. The guideline also rules out the use of natural trees and protects wildlife in area enclosures. The lack of clear policies creates uncertainty at several levels and results in conflicting practice, even in the same woreda” (Ibid: 25). They draw this rather negative conclusion from their study of several other communities across Ethiopia that have developed and implemented such plans with a great deal of autonomy, but these have at times compounded problems of inequality and resource degradation, rather than resolved them. The immediate positive environmental effects of the AEs and the species restrictions are nearly universally acknowledged both by researchers and local populations. Nevertheless, there is a steadily growing body of literature questioning the effectiveness and sustainability of these measures, particularly 31 of AE, and there is also much concern about the nature of local community institutions and their effectiveness in the management of CPR, as well as the incentives that farmers have to implement conservation measures on farm (Berhanu 2004, Chisholm 2000 and 2004, Gebremedhin et al 2000 and 2004, Jagger et al 2003, Nedessa et al 2005, Shitarek et al 2001, Shylendra 2002). 2.D.3 Woodlots and Trees In Tigray, one of the most widely implemented efforts to reforest and halt environmental degradation is the development of enclosed woodlots. Shortly after the Derg Regime fell in 1991, woodlots were again promoted in a more participatory manner, with the Tigray Regional Bureau of Agriculture and Natural Resource Development (BoANRD), NGOs such as the Relief Society of Tigray (REST) and World Vision, and local tabia administrations. Generally, in collaboration with a baito, these agencies propose the area for enclosure, but the final decision is made at a general meeting of community members.20 "Between 1992 and 1996, about 49 million seedlings are reported to have been planted in community woodlots" with a survival rate of around 40% (and as low as 10%) (Gebremedhin et al 2000: 6). People are hired to guard the woodlots against violations and are paid in kind or allowed to harvest grass. AE including woodlots have been established mainly to achieve ecological regeneration rather than to provide economic benefits to communities (Gebremedhin et al Ibid). It is not possible to enter into detail about all of the successes and difficulties encountered in relation to these woodlots, which are discussed in several reports (Berhanu 2004, Chisholm 2000 and 2004, Gebremedhin et al 2000, Jagger et al 2003, Shitarek et al 2001, Shylendra 2002). It is possible, however, to discuss some of their major characteristics and summarize conclusions about their effects. Gebremedhin et al (2000) carried out the largest scale research on woodlots across Tigray. They reported that they are present in around 90% of the tabias, with nine woodlots per tabia of an average of eight ha each, although there is much variation evident. Most woodlots are managed by the kushet baito, but a third (averaging 18 ha each) are managed at tabia level by the tabia baito. Woodlots are most commonly planted to eucalyptus. Their uses are presented in Table 2. It is important to note that most non-livestock related uses are prohibited, including any that involve cutting trees or shrubs, as well as collecting virtually any resource other Table 2 Uses of Enclosed Woodlots, 100 Villages in 50 than fruits and seeds, and grass cutting Tabias, Tigray, 1998-99 is also tightly regulated. Violations are usually punished by cash fines set by VILLAGE TABIA ALL the baito or by the local court. Common USE MANMANWOODviolations in 1998 were for grass cutting AGED AGED LOTS when this was prohibited, grazing aniGrazing 0.6 8.9 2.3 mals and cutting trees or branches, and Cut and remove grass 71.1 39.9 64.7 were more common on tabia-managed Collect fuelwood 4.4 0.0 3.5 woodlots. Fines were highest for cutting Collect dung 1.0 0.0 0.8 trees. Cut and remove trees or Both this study and another branches 0.0 0.0 0.0 done somewhat later on a sub-sample Cut and remove shrubs 0.0 0.0 0.0 of these same woodlots (Jagger et al Collect leaves 0.0 0.0 0.0 2003) reported that the main benefit to Collect bark 0.0 0.0 0.0 date appears to be improved environCollect roots 0.0 0.0 0.0 mental conditions in the woodlots Collect fruits or seeds 60.0 49.1 57.8 themselves and in their immediate viBee keeping 61.1 38.4 56.4 cinity: decreasing topsoil erosion in Source: Gebremedhin et al 2000, Table 2. Based on a survey of 50 woodlots and increased soil protection tabias (164 woodlots) at altitudes above 1500 m. 20For discussions of some of the problems associated with AE established by NGOs and other ‘outside’ organizations as well as their comparison with locally-established AE, see Nedessa et al (2005) and Gebremedhin et al (2000, 2002, and 2004). 32 in adjoining farmlands, higher levels of soil nutrients compared to non-protected areas, and increased timber and bee forage availability. On the other hand, fuelwood scarcity may have been aggravated by woodlot creation, which leads to an increase in the use of crop residues and dung for fuel, thus contributing to environmental degradation (Gebremedhin et al 2000). They also mentioned that, since these areas also cannot be used for grazing, grazing land has been reduced, leading to the problems mentioned above (see also Chisholm 2004). Since permitted uses are quite limited, the remaining benefits from village-managed woodlots are very small. The issue of limited benefits was addressed in more detail in Jagger et al’s study, where they examined the effects of various degrees of woodlot devolution for sustainable land management and biodiversity preservation, small-holder incomes, empowerment in terms of decision making ability, and equity in the distribution of benefits. Their findings are important for the present study and therefore are discussed here in some detail. The study was conducted in Tigray in 2000-2001, and covered a sample of 185 woodlots at community (tabia), village (kushet), sub-village, and household level. Altogether, 120 groups of community informants (each including a baito representative, a member of the agricultural cadre, a woodlot guard, a woman, and another community member) were surveyed. Household woodlots were primarily promoted by BoANRD (75%) or by the tabia baitos as part of a pilot project where most of the woodlot owners were landless. Decision making capacity (‘empowerment’) was highest for the household woodlots, since only about 50% of their owners had to seek permission to harvest poles, fuelwood, or tree fodder from their plantations; in comparison, people have to seek such permission in 70-90% of the common area woodlots. Regarding the stream of benefits from income generation, the study’s findings are presented in Table 3. As the authors report, “Quantities of NTFPs harvested from woodlots are generally very low and unlikely to significantly contribute to incomes” (Ibid: 19). While, as can be seen in this report (chapters 3 and 6), it is doubtful that the methods employed fully captured the use of the products that are listed in the table (especially the category NTFPs), it is nevertheless clear that the income benefit streams derived are indeed low. This is explained not by actual production or production potential of these woodlots, nor by actual or potential demand for their products, but rather by the rules that strictly limit or forbid their use. Benefits are also limited by the fact that community members must purchase some of the products, especially grass and poles. The study estimated that, while grass is the major economic benefit, its contributions to household income on household managed woodlots is only about 60 birr, or US $8.00 per year. The same study provides data on the labour input into woodlots and shows that these are very high, especially for community managed woodlots; labour inputs decrease with devolution on a per ha basis. Community woodlots rely upon non-voluntary uncompensated labour, voluntary uncompensated labour, and voluntary labour that is compensated usually through food-for-work. After accounting for labour costs, which comprise the vast majority of woodlot costs, they concluded: Average annual net benefits for all types of woodlots are negative, though for woodlots under 11 years of age, average annual net benefits improve with devolution, indicating greater potential for income diversification when smaller groups or households manage woodlots . . . The contrast of high costs and very low economic benefits in a region where demand for woodlot products is very high is troubling. The need to improve access to woodlot products, both for subsistence and for income generation, is essential (Ibid: 21). Household managed woodlots do have the potential to add substantially to income once these are mature. However, in comparison with community-managed woodlots, they are not as effective in terms of providing environmental benefits, particularly in terms of biodiversity, since smallholders with more control over woodlots are more likely to plant fast growing exotics and to be more focused on woodlots as a source of income” (Ibid: 27). At present, environmental sustainability and income generation appear to be competing management goals. Many authors have also reported that the quantity and distribution of future benefit streams from community woodlot AE are unclear and that weak management plans and such undefined benefit streams represent their major weaknesses. Shylendra (2002) argued that the degree to which the rehabili- 33 Table 3 Harvest of Woodlot Products 1997 to 2000a,b COLLECTIVELY MANAGED WOODLOTS HOUSEHOLD WOODLOT MANAGED Community Village Sub-village PRODUCT WOODLOTS Percent of woodlots that harvested product Woodlot 0-4 years N=19 N=32 N=4 N=66 Poles 0.0 (0.) 0.0 (0.0) 0.0 (0.0) 1.5 (1.5) Thatching grass 51.5 (15.9) 66.0 (10.5) 31.3 (28.9) 7.5 (3.3) Fodder grass 47.9 (16.4) 74.9 (9.3) 45.2 (29.6) 19.7 (4.9) Handicraft grass 0.0 (0.0) 4.0 (3.7) 0.0 (0.0) 3.0 (2.1) NTFPs 1.2 (2.1) 1.7 (1.3) 0.0 (0.0) 1.5 (1.5) Woodlot 5-10 years N=12 N=37 N=3 N/A Poles 11.8 (8.7) 10.3 (6.3) 50.5 (37.5) Thatching grass 59.4 (18.6) 64.7 (10.3) 50.5 (37.5) Fodder grass 61.8 (18.1) 68.6 (10.0) 25.2 (28.8) Handicraft grass 5.9 (6.1) 6.3 (5.6) 0.0 (0.0) NTFPs 0.0 (0.0) 4.6 (3.2) 0.0 (0.0) Woodlot 11-25 years N=2 N=6 N=4 Poles 100.0 (0.0) 17.8 (13.9) 0.0 (0.0 Thatching grass 100.0 (0.0) 31.3 (21.2) 63.6 (30.9) Fodder grass 0.0 (0.0) 33.4 (26.4) 0.0 (0.0) Handicraft grass 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) NTFPs 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) If harvested, average quantity of product harvested per hectare per yeard Woodlot 0-4 years N = 19 N = 32 N=4 N = 66 Poles (number) No obs No obs No obs No obs Thatching grassc 23.0 (20.2) 10.0 (2.4) 1.6 (.) 2.2 (1.2) Fodder grassc 7.4 (3.4) 7.3 (1.6) 0.2 (0.5) 13.0 (11.5) c Handicraft grass No obs 0.3 (0.2) No obs 2.0 (0.3) NTFPsc N/A** 4.2 (2.6) No obs N/A** Woodlot 5-10 years N=12 N=37 N=3 N/A Poles (number) 1.2 (0.7) 3.2 (1.6) 12.0 (3.0) c Thatching grass 13.8 (6.1) 10.0 (1.6) 3.7 (0.3) Fodder grassc 19.4 (4.2) 31.2 (10.5) 7.5 (.) Handicraft grassc 0.3 (.) 0.9 (0.1) No obs c NTFPs No obs 0.4 (0.4) No obs Woodlot 11-25 years N=2 N=6 N=4 N/A Poles (number) c N/A** 3.9 (1.4) No obs Thatching grassc 13.1 (5.0) 14.6 (9.2) 5.3 (1.0) c Fodder grass 12.7 (3.0) No obs No obs Handicraft grassc No obs No obs No obs NTFPsc No obs No obs No obs Source: Jagger et al 2003, Table 4. in parentheses indicate standard error. bMeans and standard errors are corrected for sampling stratification and weights. cMeasures: thatching grass – headloads; NTFPs – handful. dRespondents were unable to provide information on quantities of poles and NTFPs harvested. eValues are conditional upon respondents indicating that they had harvested woodlot products at some point between 1997 and 2000. N/A = Not applicable. aValues 34 tation of common property areas is able to contribute to livelihood security of the poor depends largely on how far the intervention is able to address the question of equity in participation and benefit sharing by the community – this issue is addressed again below and is a major topic in the current research. Other than woodlots, six areas can be identified where trees are found in Ethiopia and also in Tigray: 1) State forests; 2) community forest sites; 3) protected forest areas (e.g. hillside AE); 4) natural communal woodlots; 5) individual plantations around homes and farms; and 6) churchyards (sacred groves) (Keeley and Scoones 2000, Shitarek et al 2001). Churchyards are unofficial enclosures, The only areas where one can find indigenous tree species in any quantity, and which are regenerating, are the church compounds and, in some cases, protected surrounding land . . Churches are especially important as guardians of the three main indigenous highland species: Hagenia abyssinica, Juniperus procera, Olea africana. These trees have a religious value and have probably been saved from extinction as a result. Because of the traditional law and powers of the church, people respect these forests, which are managed by the church committee (Shitarek 2001: 2). Churches have become important pockets of conservation and biodiversity in areas otherwise severely affected by environmental degradation (Bongers et al 2006, Binggeli et al 2003, Wassie et al 2004). While there is increasing literature and research on the role of churchyards in preserving biodiversity within Ethiopia, none of the CPR literature in Tigray discusses the use of churchyards as resource areas. With regard to tree planting on private land, several authors have reported that this has been constrained due to tree tenure insecurity. Shitarek et al (2001) reported that, until land reallocation ended, there was fear that private trees would be confiscated without compensation, and uncertainty as to whether homesteads would be retained due to villagisation and land reallocation. Further constraints to tree planting were cited as shortage of land around the homestead and lack of attention on the part of development programmes to private planting activities in comparison with community programmes. There were also prohibitive taxes on sales of high quality indigenous species which have recently been modified to encourage such planting and sales, but much of the rural population is unaware of this change. Gebremedhin et al (2000) indicated that tree tenure in Tigray is ambiguous since, although landholders own trees growing on their land, they need permission from the baito to cut them. They also reported that regional laws prohibit planting of eucalyptus and cactus on cultivated land, and that there have been few initiatives to promote tree planting on farms. However, tree planting on private land is no longer unusual in Tigray, particularly since the land reallocations came to an end and the regional government is now promoting a pilot project for this purpose. The most common forms of agroforestry include homestead plantations, scattered trees on farmlands, and farm woodlots, whereas windbreaks are less common. Plantations, particularly of eucalyptus, contribute substantial amounts of income for farmers in several regions of Ethiopia, and there is evidence of a process of conversion of agricultural land to eucalyptus plantations in the highlands, where farmers have reported receiving higher economic benefits from growing eucalyptus than from crop production (Williams et al 2003): this process is also evident in the study area, as is discussed in Section 5.D. 2.D.4 Enclosed Grazing Lands Few studies were found that assess the effects of grazing land regulations, but one recent study in Tigray covered 50 tabias (Gebremedhin et al 2004). They reported that use regulations regarding grazing areas have been long-established in Tigray: more than half of such areas were established before 1966, and only 17% were created since 1991. Most regulated grazing lands (68%) are indigenous initiatives managed by village organizations or elders, which organize members, draft regulations, and enforce rules. The State does not interfere in these, although the BoANRD has promoted some of these organizations and may provide technical and material assistance. Grazing lands have guards, and most villagers contribute to their payment in cash or kind. 35 All of the villages surveyed in 1998-99 had unrestricted grazing areas, and nearly 90% had one or more restricted grazing areas. The average total amount of restricted grazing land area per village was 38 ha, whereas each grazing area was on average 10.5 ha. Each village managed on average four restricted grazing areas, but there was much variation. More than half of the restricted grazing areas were designated for oxen only, whereas the remainder could be used for all livestock. In 42% of the cases, restricted grazing lands were opened for grazing from September to December, 29% from January to May, and 13% from June to August. In all villages, only oxen were allowed to graze during the rainy season (June to August), which is the period when feed supplies are most abundant. In comparison with enclosed woodlots, the use of restricted grazing lands was more liberal. As reported in Table 4, besides grazing livestock, most villages permitted dung collection, two-thirds permitted fruit collection, and slightly fewer allowed bee keeping. More than half allowed fuelwood collection, and a minority permitted grass cutting for feed or construction, but cutting trees or shrubs was never allowed. Penalites for violations were widely used and mainly consisted of cash fines. The most frequent violations were for grazing animals when not allowed, and cutting grass for feed and construction without permission. Other violations included cutting roots, branches, leaves, or trees, and collecting fuelwood in areas where it was not allowed. The study concluded that, in all communities, grazing lands have regenerated significantly because of restricted use. The farmers who were surveyed did not perceive any problems as a result of the restrictions, and Table 4 Uses of Restricted Grazing reported that the main benefit was the ability to graze liveLands, 100 Villages in 50 Tabias, Tigray, stock when feed supplies were short. “On average, 42% of 1998-99 households in each village received benefit from grazing PERCENT OF animals. Other benefits to rural households included cutting ALLOWED USES RESTRICTED grass for feed and other purposes, collecting dung, and colGRAZING AREAS lecting fuel wood from dead trees" (Ibid: 282). Grass cutting 22 Another recent study assessed the environmental Fuelwood collec- 53 and economic implications of grazing land AE across sevtion eral communities in Tigray and across Ethiopia. It raises Dung collection 90 other important issues that challenge the current models Fruit collection 66 upon which AE policies are based. Nedessa et al (2005) Cutting trees 0 point out that, when policies were initially formulated, it Bee keeping 60 was assumed that environmental objectives would be best Source: Gebremedhin et al 2004, Table 2. served by entirely excluding humans and livestock from AE for long periods. Studies showed that this policy was effective in increasing biomass as well as the number of species, which convinced both policy makers and community residents of their efficacy. However, the authors argue, there are still important questions about the assumptions that have been used about human-livestock relationships in grazing land management in Ethiopia, where policies are based on conventional range management models developed for temperate areas. These assume that areas have a constant livestock carrying capacity and that grazing land degradation is simply a matter of exceeding this capacity. However, in Ethiopia, especially climatic variability largely determines vegetation amounts and cover, so grazing systems must adapt through herd movements, different mixes of livestock types, and herd size fluctuations. Given these conditions, and also the differences in the degree of land degradation that are found across Ethiopia, the scale of enclosures becomes very important: “ . . in extensive systems where livestock grazing involves movements over large areas, restricting animals from portions of these areas will have consequences for the grazing system as a whole . . . It may be that the AE is improving biomass production in the area, but has resulted in increased grazing pressure on areas outside” (Ibid: 6). Another problem is that, in Tigray, many AE designated for grazing allow only oxen to be grazed, which can have an important negative impact on biodiversity, since grazing only one type of livestock leads to the domination of only certain species: heavy cattle grazing without goats, for example, can result in bush encroachment, which further reduces grass. Further, they noted that regulations about the use of woody species have led to an increase in woody vegetation where thinning or pruning is prohibited, thereby depressing the productivity of grass. 36 2.D.5 Other Botanical Resources and Livelihoods Nedessa et al (Ibid) cited Asefa et al (2003), who measured species diversity in various types of AE in Hawzen Woreda in northeast Tigray. Their study reported that species that were previously lost or reduced are indeed increasing in AE, and both herbaceous and woody species richness and diversity were found to be greater than in AE than in open areas. However, bush density had not increased. For herbaceous species, while richness, diversity, and density increased for the first three years after enclosure, species richness declined thereafter and became less rich than what was found in open areas. Bush species richness and diversity increased for the first six years and also declined thereafter, although bush density was not related to the age of AE. Tree species richness and diversity increased for the first three years, reduced in the fifth year, and increased again thereafter; tree density also progressively increased with the age of AE (Ibid). Therefore, it is indeed possible that single-species livestock grazing and tree planting may explain why the richness and diversity of other species declines. This draws attention to the availability of and access to botanical resources other than grass and timber that are essential for local livelihoods. Research on CPR in Tigray has concentrated on enclosed woodlands and grazing lands, and on grass and timber; within AE, little attention has been paid to other botanical resources. Other CPR areas, such as churchyards, open hillsides, roadsides and borderlands, and even private farmland that are accessed by community members for certain uses, have also received very little attention, although it is in these areas that most botanical resources can be found. Further, the use of botanical resources for purposes other than feeding animals or providing timber has received very little attention in Tigray, although studies in other areas of Ethiopia are increasing, particularly with respect to regions that are perceived to have substantial wild and indigenous vegetation. Tigray is not one of these regions. The dearth of research in turn appears to reflect the lack of attention that these areas and plant resources have received from development and government agencies in general. Since so little information was found about non-enclosed common land areas in Tigray and about non-grass and non-timber resources in general, the information presented below relates to Ethiopia in general, and is organized according to use. Fuelwood It is estimated that 90% of Ethiopia’s energy needs are provided by biomass, primarily fuelwood (Williams et al 2003, Vivero n.d.). Fuelwood is obviously a major concern to the population across most of deforested Ethiopia, although it is not one that has been addressed by AE. Williams et al (2003) reported that the most important direct contribution of forests to human diets is fuelwood and charcoal used for cooking. It was estimated that at least half of all fuel consumption is for cooking injera, the national staple food. Although it is illegal to manufacture charcoal in Ethiopia, it is still widely available. Fuelwood and charcoal are important to livelihoods not only because of the contributions made to the diet: Williams et al’s study on forest products showed that they are also the forest resources that are most commonly sold, not only during periods of economic hardship, but also as a routine economic activity. Fuelwood exploitation has been cited as one of the primary reasons for the rapid disappearance of local forests in Tigray (Gebremedhin et al 2000), however, not a single study was found during this review that focuses on fuelwood use or availability in the region. FAO (n.d.) reported that most households require an average of .3 ha of forested land to meet their fuelwood needs. In its project area, fuelwood is mainly collected from ‘wasteland’ [non-enclosed common lands], and trees are cut without replacement, which occurs since, “there is no clear management of wasteland areas as compared to communal forests” (FAO Ibid). This assertion, however, appears to be partly untrue: as was discussed above, species restrictions are enforced in Tigray and seriously affect fuelwood use. Nevertheless, as was made clear in previous sections, fuelwood usually cannot be collected from enclosed woodlots and, although it is permitted on restricted grazing land, its extraction is reported to be low (see below). As indicated earlier, therefore, a large share of fuel energy needs is satisfied by burning dried cow dung and, to a lesser extent, crop residues, with the subsequent negative effects on soils and crop productivity. 37 Certain positive trends in fuelwood availability have been reported. Williams et al (2003) wrote that, in certain highland areas, it has increased because households have planted eucalyptus trees, so that burning of animal dung has declined. Also, apparently since the government has suspended fertilizer subsidies, farmers are leaving more dung on their fields and thus the market for fuelwood is increasing. However, in Tigray, FAO (n.d.) noted that, although the sale of fuelwood is an important source of income for the poor, sales are decreasing and, due to fuelwood scarcity, households are spending a large portion of their time gathering brushwood and cow dung. This particularly affects women who appear to be the main fuelwood collectors in the region. Nevertheless, little effort has been expended to date to overcome the fuelwood crisis that is so evident in the highlands. In fact, ironically, fuelwood scarcity may have been aggravated by woodlot creation. Gebremedhin et al reported: . . . for example, 13 of the 100 sample villages reported that fuelwood had declined in rank as a source of fuel for cooking since 1991 . . and in all of these cases, shortage of fuelwood was cited as the reason for the change. In 11 of these cases, the rank of dung as a fuel source had increased and, in several burning of crop residues had increased in importance . .Thus, even though restrictions on using woodlots are leading to improved conditions of the woodlots, they may be contributing to declining soil fertility in the near term as dung and crop residues are increasingly used for fuel (2000: 17-18). Other reports on the impact of AE (Shitarek et al 2001, Shylendra 2002) also indicated that households have not been able to gain any significant benefits from AE in relation to their basic biomass needs. Wild Foods Until very recently, almost no research had been done in relation to traditional wild food plant use in Ethiopia, much less on the importance of such botanical resources for household nutrition and livelihoods (Asfaw 1995, Guinand and Lemessa 2000, Kew Royal Botanical Gardens n.d.). Very little of the research that has been done has been carried out in Tigray. Until recently, wild food use was considered to be strongly associated with famine, where the greater the food shortage, the higher the consumption of so-called ‘famine foods’ (Barnett 2001, Asfaw and Tadesse 2001).21 However, Asfaw and Tadesse (Ibid) estimated that Ethiopia is the site of some 12000 edible species, eight percent of which are currently used for food in Ethiopia. Of this eight percent, 25% are food crops whereas 75% are wild food plants. Local ethnobotanists have so far identified 203 wild food species. The type, amount, and use differ according to region, where especially altitude conditions the amount and species available: of the 203 wild food species known, 144 are found below 1500 mm, 148 between 1500-2400 m, and only 81 above 2400 m.a.s.l. (Ibid). Wild food consumption is also much influenced by seasonality. Contrary to what was thought in the past, it is now estimated that only 15% of the identified wild food plant species are used for food in time of famine and food shortage, whereas 85% form part of the regular diet of those who consume them. However, very little is known about the contributions that such plants make to nutrition, since research on this topic has only begun recently (Ibid, Guinand and Lemessa 2000) and there is no nutritional information available for most of the species consumed (Kew Royal Botanical Gardens n.d.). Nor does it appear that data has ever been collected on wild food consumption as part of any nutrition study in Ethiopia. Many studies have shown that wild foods contribute very substantial amounts of both calories and micronutrients to the human diet across rural areas of the developing world (for reviews, see Grivetti and Ogle 2000, Johns and Staphit 2004) and, in areas where food shortages are great, their nutritional contributions are certain to be even greater. In Ethiopia, despite the fact that wild foods seem to be most consumed by poorer households and ethnic minorities, there is no 21For a discussion of the definitions used for ‘wild’ food plants and ‘famine’ food, see Guinand and Lemessa 2000. A database of wild food plants that is maintained by the UN Emergencies Unit for Ethiopia can be found at: http://www.africa.upenn.edu/faminefood/fieldguide.htm 38 evidence of social stigma attached to their consumption as reported by previous research (Barnett 2001). In the study area, as reported in Chapter 6, it appears that wild food plants are consumed by all. While some have suggested that wild food plant consumption increases as food security increases, it is very likely that it has decreased in the highlands due to drought, resource degradation, continuing pressure on open access common land areas, and the creation of AE, where the collection of wild foods other than fruit or seeds is prohibited. Decreasing consumption of these plants has certainly negatively affected overall household nutrition, particularly among those populations that were already facing food shortage and nutritional deficiencies. As Asfaw and Tadesse argued, “full-scale ethnobotanical studies, implementing appropriate conservation measures and domesticating and integrating promising taxa into existing crops are favoured for the purpose of diversification of food sources, ensuring food security, and maintenance of biodiversity and environmental integrity” (2001: 47). Patterns of wild food plant use in the study area are presented in Chapter 6. Medicinal Plants Compared with wild foods, there is a substantial amount of literature concerning the use of medicinal plants in Ethiopia, the majority of which are considered to be wild species. According to Deffar (1998), some 600 plant species are use as medicine in Ethiopia, or somewhat in excess of 10 percent of Ethiopia’s vascular flora but, given Ethiopian floristic diversity, the number is likely to be much higher (FAO, cited in Fassil 2002). The same report indicated that over 85 percent of the rural population uses these plants as the primary source of healthcare for humans and animals. Fassil (Ibid) added that “continued reliance on traditional medicines is partly due to economic circumstances, which place modern health facilities, services and pharmaceuticals out of the reach of the majority of the population. However, in many cases, it is also attributable to the widespread belief in the effectiveness of many traditional therapies” (Ibid: 38). Deffar (1998) reported that traditional home-based medical care is generally the responsibility of women, however Fassil (2002) found in a study in the north-western highlands of Ethiopia (Bahir Dar Zuria Woreda) that middle-aged women and men had a greater breadth of knowledge of plant location, identification, collection, preparation, and administration in comparison with other groups. Women were more knowledgeable about preparation of treatments, and men knew more about medicinals used for cattle, as well as more about plants procured from the wild and in nearby forests, whereas women knew more plants procured around homesteads. The data presented in Chapter 6 from the study area in Adiarbaetu, however, shows that women are responsible for both traditional lay and specialist healthcare, although most adults in the community have some knowledge of medicinal plants. Medicinal plants are also often reported to be important sources of revenue for traders, who are not necessarily healers. In a few studies that report on marketing of medicinal plants, in major markets in Addis Ababa and in Eastern Ethiopia, most traders are women (see e.g. Teshome-Bahiru 2005, Letchamo and Stork 1991). Not all medicinal plants are collected from forests or common land areas: many are cultivated in homegardens or are found in agricultural fields (Asfaw 2001), and women often transplant medicinal plants found in the wild into their homegardens (Tsegaye 1997). Given the dearth of homegardens in the study area (see below), it is likely that the majority of medicinal plants that are collected are found in common land areas, so that the degradation and disappearance of tree and plant resources occurring across much of Tigray must also have a negative effect on medicinal plant availability. There are many other uses made of botanical resources found on common lands in the study area that are discussed in-depth in Chapter 6. It is useful here to briefly review the importance of homegardens since, although they occur on private land, across much of Ethiopia they are generally repositories of much plant biodiversity and of indigenous and wild plant species used for food, medicine, fuel, pesticides, fodder, and all other purposes that plants fulfil. Homegardens Asfaw (2001) summarized the research available to date on homegardens across Ethiopia. The largest proportion of the useful plants found in Ethiopia are cultivated in homegardens, where 172 crop species 39 distributed in 121 genera and 50 plant families have been recorded. About 90% of homegarden produce is used for own consumption. Homegardens contain many rare species as well as experimental crops: "In a countrywide survey, 52% of a total of 172 crops species found in home gardens were categorized as typical garden species, 28% were seen to be common both in home gardens and fields while 20% were typical crops that are occasionally found in home gardens in the study area or have special varieties grown in home gardens" (Ibid: 132). Three-quarters of the crops found in homegardens were used for food. Most reports indicate either that women are largely responsible for homegardens (Amare 1999, Woldegiorgis 1999) or that they are managed by men and women together (Asfaw 2001, Pankhurst 1992). In the past, repeated land reallocations and villagisation are reported to have created much tenure insecurity with respect to homesteads where homegardens are generally located, and meant that homegardens had to be repeatedly re-established. Under villagisation, households also often lost access to land on which to establish homegardens, since homestead plots were smaller and agricultural plots were further away (Amare 1999). While these processes also occurred in Tigray, apparently over much of the arid highlands, homegardening is severely constrained by land shortage as well as by the lack of water, implements, and seed, although, as is discussed in Chapter 5, there is now apparently an increase in the number of homegardens in the study area. 2.E Benefit Streams, Equity, and Livelihoods in CPR Management Two overarching problems have been identified by researchers concerned with issues of CPR management in Tigray: the first is the possibility to halt degradation and sustainably regenerate vegetative resources, and the second is the need to generate livelihood resources that can help to alleviate poverty, food insecurity and continued degradation in the area. Recent studies concur that these two problems are inseparable. Chisholm, for example, argued: The issue of whether . . CPR . . are managed sustainably is inextricably linked with the issue of whether households which use such CPRs can sustain their livelihoods as a result. Any comprehensive assessment of the effectiveness of community-based management of CPRs needs to focus on these two dimensions of sustainability: the sustainability of the natural resource itself, and the sustainability of the livelihoods of those households using the resource (2004: 3). Resources in the area have been depleted due to drought and overuse, and they have been overused because of changes in institutional and individual access rights. With the exception of restrictions on certain woody species, most common land across the highlands is de facto open access, since rights regimes around them are lacking or ill-defined. On the other hand, the restrictions in place in AE, while increasing the availability of certain resources, generate greater pressure on these open areas. Further, the economic benefit streams that flow from AE are both poorly defined and unequally distributed within and between communities. Berhanu (2004) argued that poor awareness of actual physical boundaries of resource areas, unclear legal ownership, and long-term benefit insecurity contribute to illegal use. Perceptions of benefits, in particular economic benefits, are crucial, giving people incentives to participate in conservation and improvement of CPR. Therefore, “the manner in which communities allocate the benefits among themselves can affect collective effort of the group negatively or positively depending on the degree of ‘fairness’ of the allocation” (Ibid: 34). Chisholm also pointed out that, unless the returns or the expected livelihood benefits are commensurate with the costs incurred, it will be difficult to ensure participation in environmental rehabilitation. There is consensus in the relevant literature that benefit streams from AE to date are largely environmental. Economic benefits for most people are as yet minimal but may become substantial in future, particularly as timber stands in woodlots reach maturity (see above). Even so, it is rarely clear how these future benefits will be divided, and there appear to be few mechanisms in place or anticipated to ensure popular participation when these decisions are finally made (Chisholm 2000 and 2004, Gebremedhin et al 2000, Jagger et al 2003, Shitarek et al 2001, Shylendra 2002). 40 Chisholm reported that the results of a study that he carried out with 303 households in Tigray indicated that current economic benefits and costs of AE are distributed unequally. For example, with respect to regulations that restrict access to grazing land, he argued that larger landowners may suffer least, since they "have readier access to [grass] substitutes: for example, they can meet a higher share of feed requirements from the hizati [restricted grazing areas] and from crop residues, and can supply a greater proportion of their fuel needs from animal dung," so these respondents didn’t object to such AE (2000: 17). He found that those households with more productive assets, especially labour power and oxen, were more likely to want to preserve the status quo or even tighten restrictions on grazing land. Those who had no oxen, on the other hand, were indifferent to grazing land restrictions, or complained that they had no access to these areas for grass cutting. Restrictions on forest use, on the other hand, were opposed by a majority of the respondents in his study. It has been shown above that there are serious restrictions on access to very important resources, particularly to those used for fuelwood but also for many other uses, if this involves anything other than removing fruit or leaves, whereas there appears to be a dearth of initiatives to increase such access in future. It has also been reported that some of the products from AE are sold, but the poor are not able to purchase them. The reliance of the population on these resources, particularly the poor who cannot produce their own trees nor purchase products that are produced in AE, is unlikely to decrease in future. The collection of culturally and materially essential CPR is therefore now largely confined to non-enclosed areas which are effectively open access. Given this resource pressure, these areas nearly certainly continue to degrade, and continued degradation can only result in even greater pressure and ever greater strains on livelihoods as well as on AE. More generally, inequities are compounded by the fact that, in most AE, all community residents are expected to make equal contributions to their establishment and maintenance either in labour or cash, for which they are only partially compensated through food-for-work programmes. Nedessa et al (2005) found that the richer households benefit triply: from the resources (grass and timber) in AE, from foodfor-work, and because there are fewer households with whom benefits have to be shared since the poor cannot use the most valuable AE resources (Nedessa et al 2005). They also found inequity in resource distribution in some areas because local bylaws exclude non-livestock owners from using AE at all. “The more land is enclosed (an increasing trend at present . . ), the more the poor will lose if discriminatory . . . [by]laws persist” (Ibid: 23). Even if households without livestock do obtain livestock in future, they may be denied rights to use AE, which amounts to de facto privatisation of these areas. Given the inequalities in the distribution of assets among MF and FH and the resulting high levels of destitution and poverty among the latter, it is appropriate to question to what extent women in general, and FH in particular, rely on CPR, and to what extent they benefit from the current access regimes. In relation to its project area, FAO argued with respect to FH, young landless households, and MH without oxen that “The benefit they derive from their own or communal land is minimal and steadily decreasing due to the continuous and far-reaching environmental degradation" (n.d.: 10). To date there have been few studies that specifically address gender relations in CPR access and use, but a few do, although only superficially. Shylendra (2002) studied the impacts of AE on livelihoods in Woeri Leke Woreda, Tigray. It was found that the participation and extent of involvement of women in decision making around CPR management and use was very limited, although a household survey showed that both men and women equally felt the need for women's involvement. The AE is managed by the baito and therefore women’s participation is determined by their overall participation in the baito system. In the baitos investigated, only 10 of the 90 members were women, despite the significant proportion of FH found in the area (58% in one village and 42% in the other). Given women’s limited representation, any special concerns that they may have about CPR have a limited chance of being addressed. Nevertheless, one of the baitos that was studied had not totally neglected women's needs: special provisions had been made to allow women from very poor households to collect fuelwood for sale from AE. Nedessa et al (2005) also investigated gendered participation and benefits from AE across several study sites in Ethiopia, including a few in Tigray. They noted that households have only one representative in the baito and, for MH, this is generally a male. The female participants in the baito are usually all 41 female household heads and, because they represent such as high percentage of households in Tigray, their representation is proportionately greater. However, when women do participate, they are usually passive “for cultural and religious reasons” (Ibid: 27). They also found that . . . women’s participation in developmental works such as construction of soil and water conservation structures, nursery and afforestation activities is fairly high: around 50% in Tigray and 30-40% in other regions. Women are getting incentives through participating . . . however, in some cases they are systematically excluded from the benefits . . Women are responsible for the collection of fuelwood, livestock feed, and water from long distances where these materials are scarce. AEs generate these materials. Women could collect these products from the AEs instead of walking long distances, saving them time and labor (Ibid). The reasons that they cited for inequity in access to the benefits of AE are particularly applicable to FH and may go far to explain why, in spite of formal equality in access to benefits, FH are likely to systematically receive fewer benefits than most MH, even though they may rely more on AE for secondary products. A series of studies have indicated that FH relative to MH may have less access to common land resources because of (a) low effective participation in institutions governing common land resource use; (b) lower availability of labour with which to make contributions to efforts to establish or maintain common land resources; and (c) lesser need for grazing land because of lower access to labour for herding and lower livestock numbers. The present study therefore examines in depth how current CPR access regimes have affected women in general and FH in particular; how the quality and availability of the resources that women have traditionally used have changed over the reference period (pre-1975 to present); and how this has affected these women’s livelihoods, and livelihood opportunities and strategies. 42 Chapter 3 Research Design and Methods, or How Not to Get Lost Dealing with CPR Complexities This chapter presents the research design and methods as originally envisioned, and reflects critically on field implementation and research outcomes. First, the initial conceptual framework for the research29 is outlined, including livelihoods, botanical resource access regimes, gender relations, and household headship. Then, the research components, levels of analysis, and methods are summarized. Expected versus actual results, as well as implications for reliability and representativeness of the data presented in the report, are also drawn out. The chapter concludes with critical observations about the constraints and opportunities encountered, and the utility and reformulation of the methods and the research design to facilitate future research, which is essential in the region given that the type of data that was generated on CPR and livelihoods is virtually absent. 3.A The Conceptual Framework The conceptual framework, based in part upon the literature review, sought to define and inter-relate the main concepts guiding the generation of field data, which are presented in Box 2. Box 2 Main Research Design Elements Main concepts • Livelihoods • Gender relations • Poverty • Female headed households • Access to botanical resources • Botanical resource contributions to livelihoods 3.A.1 Main relations between concepts • Changing dynamics of botanical resource access • Current gender-related constraints to botanical resource access • Contributions of botanical resources to livelihoods • Changing dynamics of contributions of botanical resources to livelihoods • Gender and other relations affecting differentiation between poor households and between male and female headed households, with respect to livelihoods and to botanical resource access and contributions to livelihoods Livelihoods Livelihoods was a key concept in the research and Ellis (2000) provided the general theoretical framework, although the way that this concept was operationalised and quantified was strongly influenced by the availability of the Ethiopian Rural Household Survey (ERHS) (Section 3.D). This survey was used to collect quantitative data at household level, after adapting it in various ways to capture botanical resource use and access, and their contributions to livelihoods. In Ellis’s conceptualization, livelihoods are comprised of the assets (natural, physical, human, financial, and social capital), the activities, and the access to these (mediated by institutions and social relations) that together determine the living gained by individuals or households. In rural areas particularly, livelihood diversification (e.g. the use of a range of assets and a range of activities) often contributes to a successful 29A full description of methods, guides, and survey questionnaires can be found in Howard, Patricia and Erin Smith. 2005. The relationships between tree and plant access and livelihoods in the Southern Tigray Region of Ethiopia: A focus on poor and female headed households. Research proposal and instruments. Research Project supported through a Letter of Agreement with the FAO Livelihoods Support Programme. Wageningen University and Research Center, Wageningen, the Netherlands, February. 43 livelihood strategy due to its capacity to ameliorate the effects of risk and seasonality, unpredictable events, and adverse trends. The ERHS attempts to measure these dimensions. While livelihoods vary both temporally (seasonally or from year-to-year) as well as in relation to shocks, adverse trends, and risks, these phenomena could only be assessed qualitatively in this research. 3.A.2 Botanical Resources and their Access For the purpose of this study, botanical resources were defined as domesticated and wild, purposefully planted and natural, exotic and indigenous plants that are cultivated and/or harvested and used by the population in the study area. The focus, however, was on indigenous and wild botanical resources rather than on traditional staple crops. It was assumed that knowledge, access to, and use of these resources are highly gendered and are also changing rapidly for numerous reasons, including land use change, commoditisation, decreasing labour availability, changing perceptions regarding the cultural value of these resources, resource degradation, etc. In the highlands, the issue of access to botanical resources must obviously be framed within the larger context of CPR management discussed in Chapter 2, which includes formal property regimes and especially AE. However, such formal property regimes are only part of the picture. The conceptual framework supposed that there is also “a multiplicity of pre-existing indigenous or traditional common property arrangements governing a variety of resources within the same landscapes” (Chisholm 2000) or what others have referred to as ‘natural resource tenure’. The research design combined and modified the frameworks utilized by Howard and Nabanoga (in press) and Kundhlande and Luckert (1998). It was meant to permit the complexity of social conditions that regulate access to and use of natural resources to be captured and, further, to detect incremental changes over time. It (a) acknowledges the existence of different sets of rules relating to different natural resources (e.g. to trees, water, land, wild plants) in different landscapes (e.g. enclosed and open woodlands, grazing lands, hillsides, agricultural fields, homesteads), and as well (b) is intended to analyse how the sets of rules correspond to social structures, that is, how rights and duties are distributed among groups within the community in accordance with their characteristics, e.g. age, sex, wealth, occupation, household size, kinship, etc. Natural resource tenures are defined herein as variations in property rights. Following Kundhlande and Luckert, property rights definitions have two components: “1) a good or service, and 2) social conditions that restrict or promote its use.” There are thus "as many kinds of property right structures as there are different combinations of social conditions which may be placed on the use of a good, or the provision of a service" (1998: 5). As was discussed in Chapter 2, the social conditions of land access in the highlands are complex, and even so they indicate little about the tenure or social rules regarding resources on the land, e.g. trees and other botanicals. ‘Tree tenure’ has long been recognized, but all natural resources are potentially subject to social conditions of use that vary according to the resource. Kundhlande and Luckert’s framework is presented in Box 3. A major problem for Western scholars is to understand or characterize emic concepts underpinning such social relations, as well as the mechanisms through which such rights are defined and sanctioned, since religious, ritualistic, and normative beliefs and traditions often constitute the principle basis of legitimacy for such social relations. Traditional and indigenous peoples may or may not have concepts of property rights per se, but apparently all do recognize what Ellen (1993a, 1998) refers to as ‘morals’ that are invoked to regulate access to natural resources found in various landscapes, although these may or may not be formalized in customary or formal legal codes. Spiritual or social ‘morals’ represent foundations of, and mechanisms to reproduce, intra-community and intrahousehold property relations, of which gender relations are an intrinsic feature. The hypothesis that was pursued in the research was that social rules of access to botanical resources (or ‘tenure’) would reflect the following complexity (Howard and Nabanoga, in press): Part Z of species A in landscape X can be used by person Y if the use is for B and Y abides by rule M, during season C Patterns were expected to emerge regarding so-called ‘bundles of rights’ for different groups of users. However, as is discussed below, there was insufficient time to complete all of the detailed qualitative research that was meant to capture these complex dynamics. 44 Box 3 A Framework for Assessing Natural Resource Tenure (from Kundhlande and Luckert 1998) Comprehensiveness Exclusiveness Use designation Duration Allotment type Transferability Operational requirements Operational control Security 3.A.3 The number and types of resource attributes over which households have control. A given type of tenure does not always confer the right to use all resources in an allotment The ability of a tenure holder to prevent other parties from freely enjoying the benefits of a resource Limits activities that tenure holders can carry out on a given type of tenure The period over which property rights can be exercised Whether the tenure specifies an area over which a household possesses rights, or whether the tenure specifies the volume of natural resources which may be collected by the household The ability of a resource owner to sell, bequeath, or gift an asset (including the benefits generated by the asset) to others Where resource users must adhere to certain regulations during the use of an asset, for resource management and in their harvesting practices Mechanisms for enforcing the conditions of tenure: e.g. penalties The tenure holder's perceptions of the dynamics of property rights, the ability to capture benefits generated by the resource in the future, and whether changes in tenure will occur Gender Relations and Botanical Resource Access It was hypothesized that the distribution of powers and obligations that are manifest in traditional ‘natural resource tenure’ regimes would reflect social structure, which determines the importance of particular botanical resources to particular people within it. It was also assumed that universal differences in systems of rights and duties by sex would be reflected in informal rights systems relating to botanical resources. The simplest explanation for this is that there is a gender division of labour, which means that rights are associated with men’s and women’s material needs and their distinct obligations to provide goods and services, as well as with their knowledge and abilities (human capital) that are required to carry out distinct tasks. According to this line of reasoning, access to botanical resources is gendered to the extent that a gender division of labour exists with respect to the use, management, and conservation of plants. It was also presumed that the gender division of labour is related to religious and other values and belief systems, and to concepts of masculinity and femininity and norms about behaviour that are appropriate for each sex. These value and belief systems and social norms prescribe the type of activities and responsibilities that are appropriate for men and women, as well as with whom men and women of different social positions can appropriately interact: e.g., they prescribe particular types of gender relations. Very importantly for plant and other environmental knowledge, these beliefs and norms extend to men’s and women’s relations to different physical spaces and environments (Howard 2003b). Not only do men and women use different spaces, they use the same spaces differently. If men and women access different spaces and environments, it stands to reason that their access rights also differ since rights represent a means to legitimize and reproduce such relations. 3.A.4 Household Headship The reasons for concern with female household heads in this research have already been laid out; however the definition of household headship is still problematic. ‘Household headship’ is a category of social status that can only be defined relatively and culturally. In most patriarchal cultures it is nearly by definition a male position. However, female household headship is also generally recog- 45 nized in cases where an adult male is completely absent. Headship generally refers to a decision making position of oversight, management, and authority, and hence it also often is related to age and kinship. Depending upon the position accorded to age in a culture, it may be the senior male in a household who has the most authority and management decision making; where age is seen to be a limiting factor, the eldest male’s authority may give way to a younger male’s authority. One’s relation to the household head also often has a significant effect on women’s status in a household: the wife of the head may have privileges over other women, e.g. particularly over daughters-in-law; it will also have an effect on the standing of other men if some are direct kin and others indirect kin of the head. Culturally-recognized headship does not mean that the head has absolute decision making power or authority. In fact, decision making is divided among household members in part according to domains of responsibility, but it also may depend upon command over resources: e.g., having higher income may lead to greater say over household assets. The research herein continuously sought to reveal such gendered decision making domains within households, particularly with respect to natural resources. 3.B Research Components and the Levels of Analysis Table 5 presents an overview of the research objectives and respective research methods that were used to generate data at four levels of inquiry: community-wide, household level, individual level, and (plant) species level. The community was defined as the population residing in a village (kushet) and its land area which is administered by a single village council (baito). Community-level research was based on interviews and discussions with key informants, expert informants, and focus groups of both sexes, different ages and wealth categories, and was intended to reveal patterns of human endeavour, land and resource use, and dependencies and dynamics of environmental, economic, and cultural change in the immediate context within which households and individuals operate and strategize. The community is in turn influenced by exogenous contexts and processes over which it has little influence or control (e.g. climatic, political, economic); it is at the level of the community where such contextual factors were researched to determine their influences on community-level dynamics. The household was a second key unit of analysis, where households are considered as differentiated units of production and consumption that together constitute a community, making it more or less heterogeneous. A household in this research was defined as a group of people who normally share a common cooking pot, or who otherwise co-reside in a homestead. Most household members are also kin. Kin generally divide and allocate a series of assets as well as their labour among themselves, and their individual livelihood strategies may be contingent upon or otherwise complement each other. The assets of the household, including its holdings, are typically numerous and may be held in various types of tenancy, and the household is also linked to the larger community through a wider farming/livelihood system that depends to a greater or lesser extent upon CPR (Bruce 1989). Variations among households in terms of their livelihood strategies and dependencies upon various types of natural resources are likely to be attributable to household headship, demographic differences, livelihood assets, and cultural factors. Data generated at the household level can also be used to understand intra-household dynamics, gender (including headship), and age differences, and the status of specific individuals depending upon e.g. their kinship relations or social status (e.g. as herbalists, crafts persons, income earners). These relations are, to a large extent, based upon more mesoand macro-level social relations and norms of behaviour (e.g. religious, cultural, political), but they may also reflect local context and idiosyncrasies of individuals and specific population sub-groups. Data about individuals were collected through one instrument aimed at household level (e.g. the Household Survey) and one aimed specifically at the individual (the Species Access Study). Data about individuals were confined to male heads and their wives, as well as female heads. Much individual data was collected through the Household Survey by interviewing these individuals separately. Further, such data were collected in the Species Access Study in reference to those people (male or female) who were significantly involved with such species. Women heads of households and married women were also interviewed qualitatively for their resource access histories. 46 Table 5 Research Objectives, Level of Analysis, and Research Methods COMMUNITY HOUSEHOLD INDIVIDUAL & SPECIES OBJECTIVES LEVEL LEVEL LEVEL Determine livelihood Botanical Resource Household Survey Species Access Study; dependencies on Valuation Study; Women’s Resource wild and indigenous Landscape Niche Access Histories botanical resources Calendar/Resource and their dynamics Distribution Study Determine dependVillage Survey; Household Survey Species Access Study; encies on different Community ReWomen’s Resource types of common source Mapping; Access Histories land and resources Landscape Niche Calendar/ Resource Distribution Study Determine imporHousehold Survey Species Access Study; tance and diversity Women’s Resource of homegardens Access Histories Determine differenCommunity ReHousehold Survey Species Access Study; tial access to comsource Mapping; Women’s Resource mon land resources Landscape Niche Access Histories Calendar/ Resource Distribution Study; Community Resource Access Study Capture determiVillage Survey Household Survey nants of poverty and differentiation of households Capture determiBotanical Resource Household Survey Species Access Study; nants of livelihood Valuation Study Women’s Resource strategies, especially Access Histories those related to plants and trees Capture contribuBotanical Resource Household Survey tions of plants and Valuation Study; trees to food security Landscape Niche Calendar/Resource Distribution Study Capture significance Village Survey; Household Survey Species Access Study; of gender relations Landscape Niche Women’s Resource and household head- Calendar; CommuAccess Histories ship; importance of nity Resource Access plant, tree and comStudy mon land resources to women and FH Species level data were also collected. Ethnobotanical studies are often focused on data generated at the species level, where it is the characteristics of the species that are investigated in relation to environmental, cultural, and economic dynamics. The rationale is quite evident for doing so in relation to species of great economic and/or cultural value in particular contexts, but it is not so evident when combined with community, household, and individual level data on livelihood systems. The species level data nevertheless allow for linkages to be established, e.g. between the uses that individuals make of species, the gender division of labour within households and household livelihood strategies, the use of landscape spaces, formal and informal rules of access, and use dynamics. Such 47 species level data can reveal heterogeneity where otherwise homogeneity would be suspected: e.g. a particular landscape may be under private tenure and generally off-limits to community members except with reference to a particular plant species, which is considered to be ‘open access’ to whoever wishes to collect it; or particular tree species may be considered to be owned by the person who planted it, even when found in a communal area. 3.C Research Methods and their Field Applications While the ability to generalize within or beyond the study population would have been desirable, this was not possible to achieve due to resource limitations. Quantitative data were collected at household level, but the sample size was too small and heterogeneous to permit valid statistical analysis; therefore all data from the survey are presented descriptively. It was decided to pursue depth of penetration into the research topics and their interrelations and sacrifice breadth of coverage. The in-depth understanding of behavioural and cultural determinants of the phenomena being researched also requires the use of qualitative methods. The research commenced with a literature review carried out at Wageningen University in the Netherlands, and resulted in the development of literature database in Microsoft Access and in the collection of pre-existing data (as well as survey questionnaires used to generate this data) in the study area, including ethnobotanical data.30 During the review, several major abstract databases were used.31 The annotations were entered into NVivo (Gibbs 2002) and coded; all qualitative data collected during the research were also entered into NVivo, coded, and analysed. The field research components were originally constituted at village, household, individual, and species levels, respectively. The methods that we intended to employ in each component were presented as a series of steps within each component; the components were also presented more or less sequentially, where the data collected in the previous steps could serve as precursors to the data collected in subsequent steps, either for sampling or for subsequent detailed questioning. Table 6 presents information summarizing the methods used. Since the recommendation domain for this study was the FAO project site in Southern Zone, Tigray, it was considered to be important that a minimal amount of data on these sites be available to permit a degree of comparison and generalization, which would have been permitted through the Village Survey. This survey would further allow two study sites to be purposively selected to represent particular phenomena of interest. It was assumed that differences in gendered access to common land and indigenous and native botanical resources would vary according to the general dependencies of the village populations on these resources, which would in turn be a function of the availability of such resources, and the availability of alternative livelihood resources such as wage employment and markets. The higher the dependency of the population on CPR and on indigenous botanical resources, it was hypothesized, the more likely that at least informal access regimes would recognize and regulate such access. It was also hypothesized that women, including FH, would have less access and fewer benefits. 3.C.1 Quantitative and Qualitative Community Level Data It was anticipated that two study sites would allow comparison of populations’ CPR dependencies. It was proposed that ten out of the twenty kushets in which the project was operating would be surveyed, which would permit selection of the two study villages and allow certain types of generalizations to be made at woreda level. In the end, due to time constraints, it was only possible to carry out the research in one village and, instead of selecting the village from a sample of 10 villages to which the Village Survey (see below) had been applied, a village was recommended to the researchers by the woreda official. 30See Asegom and Yohannes (1996), Asfaw (1995, 2001), Asfaw and Tadesse (2001), Azene et al (1993), Guinand and Lemessa (2000), Jansen (1981), McBurney et al (2004), Selinus (1971); see also “Websites consulted for ethnobotanical data” in the References section of this report. 31This includes: PROQUEST (from University Microfilms International), CAB Abstracts, Agricola, TREECD Forestry, Econlit, Tropag and Rural, and JSTOR, as well as Google Scholar and more general internet searches. 48 Table 6 Data Collection Methodologies and Information Generated METHOD Village Survey INFORMANTS Key informants Community Resource Mapping Male & female focus groups Retrospective Community Resource Mapping Botanical Resource Cultural Valuation Study Retrospective Resource Valuation Study Landscape Niche Calendars/Resource Distribution Study Elderly focus group Community Resource Access Study Key informants; Male & female focus groups Women’s Resource Access Histories Female focus group with 10 participants over age 50 Key informants – MH & FH, other men and women Species Access Study Household Survey Male & female focus groups Elderly focus group Male & female focus groups Male & female heads of households and spouses of male heads INFORMATION GENERATED Village economic, social, and environmental history; economy, ecology, agricultural patterns and calendar Primary landscape niches, land use, spatial distribution of land-based resources, user groups by landscape niche, differentiated by sex Same as above, pre-1975 Species and uses, cultural ranking of importance of each species (economic, cultural, material), species preferences Historically important species no longer available or used, retrospective ranking of importance, landscape areas and reason(s) for changes Species use per landscape niches per season; other landscape niche uses per season, by sex of users, by destination of use (consumption, sale, both); availability and exploitation of resources across landscape niches Formal resource tenure, access & use rules, sanctions, decision making mechanisms & participation; informal access rules for different resources (e.g. food, medicinal, grazing), landscape niches and user groups (e.g. children, elderly, women, healers) History of women’s access to land (common & private) & changes with land reform, villagisation, AE, drought and famine, and reasons; changing dependencies on plant resources and reasons for these changes Uses, parts used, destination (i.e. consumption, sale), and the landscape niche where the plant or plant part is collected; planting and access rights and the people and/or organizations responsible for implementing and enforcing access regulations; exceptions to access rules; history and dynamics of species’ use Household assets, employment, agriculture, livestock & homegarden production, income from work and production (kind, cash), division of finances, division of labour, marital history, formation and dissolution, intra-household resource access and tenure, use of communal resources, species use and dependencies, food sources and seasonality calendar The Village Survey that was conducted complemented the data collected from the FAO baseline survey (FAO 2004a) which focused primarily on nutrition. It was adapted from the village survey carried out by the Institute of Development Studies (IDS – UK) and Save the Children in Amhara (Sharp et al 2000), and collected data on: village location (distances to nearest services, markets for different products, etc.), local employment for men and for women, livelihood activities of men and women, land and water resources, communal land resources and land use as well as access regimes, crops and agricultural calendars, wild plant use, and agricultural extension services. As was indicated earlier, however, the Village Survey could only be implemented in one village, and only one village could be included in the research. 49 Several other methods were employed to generate data at community level, all of which were oriented toward identifying and understanding dynamics of resource use relating to common lands. Community level data were generated either using key informants such as baito or woreda officials, extension agents (DAs), or user association (e.g. grazing land, woodlot) officials when a formal or official perspective was required, or by using three focus groups when community perceptions were of primary interest. Focus groups were used when it was desirable to: a) capture several perspectives about the same topic, b) obtain people’s shared understandings of everyday life, or c) capture data generated by the interactions between the participants. Given that each data collection exercise, with the exception of Community Resource Mapping, built upon the results of the previous data collection exercise, the same focus groups were used for all related methods with the exception of the Women’s Access Histories.32 Participants for all focus groups were chosen with the help of village officials who are very familiar with the population in the study area. They were asked to provide the names of eight men and eight women of mixed economic status and ages and, for women, a mix of female household heads and wives. The elder focus group consisted of the three eldest members of the men’s and women’s focus groups, all of whom were over age 50. Emic wealth categories were generated with village officials to support the selection of a range of participants from different Table 7 Emic Wealth Categories wealth strata. The exercise resulted in the categories shown in Table 7. WEALTH EMIC DEFINITION The community-level data CATEGORY collection exercises each would Very Poor Don’t own a house (rent), landless, food secure provide a key set of data on reonly one month per year, one set of clothing source availability, distribution, purchased per year, fully dependent on fooduse and access, and the dynamics for-work activities thereof over time. The information Poor Own a house, have land, do not produce more would be provided at the level of than two quintals of grain (lasts about three landscapes as well as at the very months), daily paid labour is required for detailed species level. Each exercise survival added a ‘layer’ to the data, moving Medium Own house, own one-two oxen, have at least .5 from general and more abstract hectares of land, food secure for at least five (resource mapping), to increasingly months specific and concrete, although, Better off33 2+ oxen or 2 oxen and a cow, more than .5 ha toward the end of the research, the of land, food secure for 9+ months, good overexercises would be carried out that all income required greatest abstraction and capacity for synthesis from the participants themselves, e.g. their own assessments of communal tenure relations: the Community Resource Access Study, Women’s Resource Access Histories, and the Species Access Study. Each exercise would enhance participants’ understanding of community resource dynamics and better prepare them to discuss the more abstract concepts entailed in informal access rules and behaviour. The sequencing of exercises would also tend to reduce informant resistance to discussing particularly sensitive issues, such as uses made of areas where access is formally restricted. While it was not anticipated in the research design, field researchers spoke with children as often as possible to get an idea of their knowledge of plants, as well as access rules that might be specific to them. This was often achieved in everyday activities and conversations while researchers lived in the kushet. One children’s focus group was also arranged at the local school, and consisted of 12 children from eight to fourteen years of age. Discussion was unstructured in order to put the children at ease and to gather as much information as possible. Issues discussed included knowledge of plants 32In ethnobotanical research it is common that informants are selected based upon their recognized botanical expertise. We chose not to do this because we wanted to learn about the importance of botanical resources for the entire population. In any case, in Adiarbaetu it was found that, with the exception of medicinal plant knowledge, there are no particular ‘experts’ on the plants used. Villagers agreed that everyone uses all species. 33This category only applies to MH. Women leaders insisted there were no FH that would classify as ‘better off.’ They struggled with identifying any FH that fell into the medium wealth category. Most FH fall under the very poor and poor categories. 50 in general, knowledge of wild foods, access rules specific to children, and locations and times of wild food collection.34 3.C.2 Contemporary and Retrospective Community Resource Mapping Participatory Community Resource Mapping was important both to capture information about resource use and because a map of the kushet and its resource areas was unavailable. The outcome of the mapping exercises influenced the discussions and results of other focus group exercises, so it is important to discuss the problems encountered. Mapping was meant to capture villagers’ contemporary understanding of land use, and the spatial distribution of natural resource areas at a community scale. In addition, it was used to provide a comprehensive list of uses of landscape niches and to identify the main groups that use those niches. The reference period for the exercise was the past year. Focus groups were used to capture shared understandings (including definitions/taxonomies) of land uses and resources, and of the significance of different spaces to different groups of people. Villagers were asked to plot the location of areas where they obtain important products or where they use resources. This led to the identification of different landscape niches and resources such as agricultural land, grazing land, watering holes, woodlots, roadsides, specific tree species, etc. Once the participants had plotted the resources to their satisfaction, they were asked, for each landscape niche, what uses are made of the space (specific uses and/or species used), as well as which groups make use of the space (e.g., men, women, poor people, cattle owners, etc.). Symbols indicated the various uses and groups, and were included on the map. A third focus group consisting of elderly participants was then convoked to construct a single community resource map prior to 1975, the year of the great land reform. There was no official map of the kushet that could be used during community mapping exercises to provide a general outline of the kushet and help locate significant features as spatial reference points. One of the DAs therefore drew a map that was then introduced to the focus groups. Women preferred to draw their map in the dirt, and men preferred to draw on paper. Both maps were transcribed onto separate sheets of paper for future reference, and larger, more detailed maps were then created by researchers from these sketches, which were then reconfirmed for accuracy with the focus groups at the subsequent session. Elders were unable to create a retrospective community resource map: they could talk a great deal about the resources and resource areas in the past, but had difficulties putting this into a spatial context, explaining that, in the past, the amount of plant resources was so great that they were difficult to map. Plants were found and used in every location and there were few demarcations or restrictions on areas as there are today. Plant resource availability had changed so drastically over the past thirty years that it was impractical to try to map past resource areas and resources used.35 Although a map was not generated, the session was very valuable for gaining an understanding of the degree of change that has been experienced in the area in relation to plant resources. Mapping in general was a difficult exercise for most of the participants, especially for those who had never seen maps. Many differences were found between maps created by the men’s and women’s focus groups, as was expected. However, when analyzing the results, it became apparent that many were attributable more to how groups visualise their resources areas in relation to the map rather than to how they use or value these areas. For example, while on the map it appeared as if men and women used different woodlots, in reality the same woodlot is used, but the woodlot was located differently on the map. Since the maps were used as reference points throughout the research, inconsistencies and problems with this method in turn had an influence on many of the other methods, es- 34Children were found to be highly knowledgeable of both local and exotic plant species, and could speak about fruits and vegetables not found in the area and which perhaps they had never seen. They were able to list many wild foods, some of which were not mentioned by adults. They had much knowledge of access rules and restrictions, repeating many of the regulations that were reported during adult focus groups. Although they are very aware of the rules, they also stated they were able to avoid being caught or punished for violations because of their age. 35Participants at one point stated, “If you need to put it on a map then put plants everywhere, large trees everywhere. There wasn’t a place in the kushet that we didn’t use.” 51 pecially because it was often difficult to determine the specific landscape areas to which informants were referring. 3.C.3 Landscape Niche Calendars/Resource Distribution Study Since botanical resource availability is highly seasonal and is linked to specific landscape niches, the Landscape Niche Calendar method was developed for this study to provide data on botanical resources that were out of season at the time of the fieldwork, and therefore fill a critical data gap. Further, it was intended that the exercise would reveal perceptions about the distribution of botanical resources across landscape niches (existence of the resources), and the distribution of use across the niches (exploitation of the resources), which could then be compared to highlight differences between existence and exploitation, leading therefore to discussions regarding access, gender, etc. which were expected to explain such differences. It was also supposed that, when such calendars are used together with data from the Household Survey, the exercise could reveal how private and communal resource exploitation activities compliment each other within local livelihood and food security strategies. The reference period for the exercise was the past year. Focus groups were used in order to capture shared understandings of the phenomena under study, and of the significance of different resources, uses, landscape niches, and temporal periods for different groups of people, particularly for men and women. As a first step, a Landscape Niche Calendar was drawn up for each of the landscape niches that were identified during the community resource mapping exercises. Men’s and women’s focus groups were then asked which uses were made of the landscape niches; subsequently, the months during which the niche was used for this purpose were plotted. Then it was determined whether men, women, or both were the primary users during each period plotted. Subsequently, participants were asked to list the botanical species that were used within the landscape niche. As a second step, the focus groups were asked to use a pebble ranking method to show the relative degree of exploitation of different landscape niches with respect to the different uses and species mentioned, which would give the relative importance of each landscape niche. As a third step, the pebble ranking method was used to show the relative incidence (availability) of the different species across the landscape niches. Since there were insufficient resources available to permit a study to be carried out of the actual incidence of the important species across landscape niches (e.g. density counts per ha determined by sample plots), the perception of incidence reported by community members had to be used, although the data would not be reliable as an indicator of actual incidence, but rather would only indicate people’s shared perceptions of incidence. In implementation, the process of creating the landscape calendars was fairly straightforward and easily understood. However, since there were many resource areas used and many resources used in each area, the sessions became too long and participants became bored, which affected the quality of the data toward the end of the sessions (men were especially prone to this). Because the number of species used was quite high, respondents would often stop after having listed a certain number and then report only general uses. This was particularly true of medicinal species. Women in particular reported that, depending on one’s knowledge, any plant can be used for medicine, so they tended to indicate areas in which they collect medicinal plants rather than listing individual species. This problem, of having to recall and provide details about many species, has probably also biased some of the results of the Household Survey in relation to CPR and species use, leading to underreporting. The second objective, to assess resource distribution and compare this with exploitation, was not straightforward and, in the end, did not work. Since the number of uses reported for each landscape niche was high, the discussion of distribution and exploitation had to be completed in two sessions. During the first session, participants were asked to rank the various landscape areas according to the amount that each resource is used, whereas during the second session they had to rank according to the incidence of the resource (e.g. where the species grow the most). While participants had no problem ranking landscape niches according to actual use, they did have difficulties ranking them 52 according to availability of the species: they often seemed to be confused about the distinction.36 When the discussion turned to comparing availability versus use (e.g., areas where a species is used the most but is not very abundant and vice versa), participants insisted that the differences noted were due to errors on their part, and wanted to rank again in order to eliminate the differences. After this occurred several times, the effort was abandoned and more general questions were asked such as: Are there areas where X grows abundantly but where you do not use it? Are there areas where X is collected but it is difficult to find or is not very abundant? Why? A possible solution in future would be to rank on use and incidence simultaneously for each species, rather than to implement the exercise in separate sessions. 3.C.4 Botanical Resource Cultural Valuation Study The Botanical Resource Cultural Valuation Study (after Sheil et al 2002, Lykke et al 2004) was used to determine which botanical resources are the most important to the community, why they are important, and how this has changed over the past 30-40 years. Focus group participants were asked to free list the ‘most important’ botanical species that villagers use from common land areas, without initially considering how the term ‘important’ might be defined, but rather leaving this up to the participants’ intuition. People tend to think of those botanical resources that are most salient to them first, so the free-listing, together with any discussion that occurs about which species are ‘important’, provides at least an initial cultural consensus (Bousfield and Barclay 1950). While participants could easily have listed many more species, they were asked to limit the list to the 25 most important to facilitate the ranking exercise that followed, since ranking a larger number of species for several characteristics creates difficulties. However, the men’s focus group insisted on including one more species because, they said, it was important. Once a list of salient species was obtained, the next step was to begin to understand why these resources in particular are ‘important’ to villagers, where ‘importance’ implies that there are specific cultural and material values associated with the resources. While it would have been possible to simply assume that certain value categories, such as ‘food’, ‘medicinal use’, or ‘construction’ exist universally, this imposes an ‘outsiders’ (etic) perspective on value, and may over-emphasize material, while missing cultural, values. Although etic value classifications were understood by villagers, they were not necessarily true representations of how importance is understood or assigned within the village. Therefore, in this step, botanical resources were classified and ranked according to the villagers’ understanding. For each species listed, participants were asked their reasons for importance which generated a long list of uses, which could be considered to be emic value categories.37 The researchers, together with the focus group participants, then generated a ‘taxonomy’ of values using local terms and seeking consensus around these, where each category was exclusive and encompassed the scope of reasons for importance given by the focus group participants: that is, performing what anthropologists term ‘cultural domain analysis’ (Atran 1994, Boster and Weller 1990, Freeman et al 1987). An example of a distinction that women made when discussing the emic category ‘medicinals’ was that they refined it into three categories: general medicine, children’s medicine, and women’s medicine. Men included value categories such as ploughing tools and walking staffs. Participants were then asked to distribute pebbles according to the relative importance of each species within each category. This exercise was not possible to implement as anticipated. All focus group participants were illiterate. The only way to accomplish this was to have samples or photographs of the species listed that would permit them to identify the species during ranking. However, it was not possible to acquire samples or take photographs of all of the species due to time limitations, seasonality, and the location of many species. The method had to be adapted: instead of using the entire list of species, participants were asked to name the five most important species for each value 36This is one of the drawbacks of using translators: it was not possible for the lead researcher to understand how the concepts were translated for participants. It is therefore possible that many problems may have begun with translations. 37The discussion focused on the most important species and why they were important to participants. As a result, while many uses were identified for each species, this did not result in an exhaustive list: there are many other local use values that were not mentioned during these sessions, principally because focus group participants do not represent the entire community, and also because not all uses are considered to be very important. 53 category, and then to rank them. Five was the maximum that researchers felt could be retained in memory during ranking. Once ranking was completed, participants were also asked why they had ranked them in this manner. Since only the five most important species for each category were used, the relative differences in importance revealed in the ranking exercise was far less than would have been the case had the original free-lists been used. Despite limitations, the rest of the data obtained were highly pertinent. The final exercise asked the elderly focus group participants about resource dynamics over the past 30-40 years. They were requested to name any species of trees or plants that used to be important to villagers but that are either no longer used, or are far less important. For each new species named, they were then asked which value category(s) pertain to the species, from which landscape areas it is/was found, and the reason(s) for the change in use or importance. This provided an idea of dynamics affecting species and uses (rather than landscapes), and of the factors influencing change. Such information was useful in and of itself, and also provided a platform for further discussion around resource access, use, and availability. 3.C.5 Community Resource Access Study The Community Resource Access Study was focused on formal and informal or customary norms and rules regarding who can use what resource for what purpose and under what conditions. Formal access rules may or may not determine actual access practices, depending upon (a) the community’s respect for the authority of the institutions that set up the rules and the effectiveness of sanctions against violations; (b) the clarity or vagueness of the rules with respect to particular landscape niches, uses, or user groups; (c) the absence of rules regarding particular landscape niches, uses, or user groups; and (d) the existence of alternative, informal or customary rules or behavioural norms. The question guide used in this study is presented in Box 4. In order to explore these dimensions, the Community Resource Access Study was carried out at two levels. Formal rules, sanctions, and violations were discussed with community officials and other key informants who belong to the institutions that are responsible for creating/enforcing such rules. The informal or customary norms and rules regarding access were investigated through the focus groups. Interviews with key informants and officials regarding formal access rules followed Bruce (1989), whose schedule was adapted according to the findings of various studies on common resource management in the study region. Officials were interviewed, but they were very difficult to access due to their own time constraints and it was not possible to interview them opportunely: some of the important officials could not be interviewed at all. Only a limited amount of information about formal rules could be obtained at the outset; therefore, when discussing informal rules with focus groups, researchers lacked some of the important background information, and could not always readily recognize and understand differences between formal and informal rules, and sanctions, and how they interact or counteract each other. Nevertheless, it was possible to gain a good general understanding of the formal and some informal rules governing access to botanical resources from focus group discussions. C.3.6 Women’s Resource Access Histories Women’s Resource Access Histories were based on interviews with both female household heads and wives from different wealth groups, all of whom were over age 40, to investigate how changes in access rules have affected women specifically over the past thirty years. Participant selection occurred again with the assistance of the village officials. Three of the women also participated in the other women’s focus group, since it was thought that they could facilitate the discussion and put the other women at ease. In addition, four women from more rural areas of the kushet were selected to provide potentially contrasting histories. Discussions were held individually in two sessions in a semistructured format (Box 5). 3.C.7 Species Access Study The final qualitative method that was developed by the lead author and applied successfully in another research setting in sub-Saharan Africa (Howard and Nabanoga, in press) was the Species Access 54 Box 4 Question Guide for the Community Resource Access Study A. Formal Access (e.g. asked of Peasant Association head, head of grazing land committee, head of woodlot committee, etc.): Who is allowed access to (landscape area)? Anytime? What can they take? As much as they want? Does it matter if it’s for own consumption or for sale? Are all rules enforced? If not, why? Who enforces these rules? Why them? Regarding the organization: Who is a member? How do they get to be a member? Does everyone in this organization create the rules? Enforce the rules? If not, which members do? Why? What is the consequence if someone is found breaking these rules? Are there exceptions? Do these rules apply to children? Do they apply to gathering of medicinals? Do they apply to the very poor? In the past year, have many broken these rules? If so, what consequences did they face? Are there disputes over these rules? If so, by whom? Reasons for disputes? (Example: for woodlots) If trees cannot be cut or destroyed, can other products be taken (i.e. grass, fungi, sap, debris/twigs, leaves, other plant parts)? If so, by whom? Are they allowed to take as much as they want? Are they allowed to sell these products? Are there exceptions to these rules (children, medicinals, certain seasons, etc.)? B. Formal and informal access: Focus group questions: If a couple wants to build a house, where would they get the materials (poles/grasses)? How far away is that? Can they destroy the tree/plant, or only gather what has fallen? Do they need to ask anyone’s permission? If so, whom? Why that person? Can they take it at any time? As much as they want? Can anyone take materials from there? Any species? Can people from outside the village (perhaps name specific village nearby) also take materials from there? If not, why? How are they stopped? Who stops them? Why this person(s)? Grazing: Could the couple also graze their oxen there? Any other animals? Do they need to get permission? If so, from whom? Why that person? Can anyone graze their oxen there? If not, why? Where else can they take their oxen to graze? Medicinal Plants: If someone is ill, where do they get medicinal plants? How far away is that? Can they destroy the tree/plant or only gather what has fallen? Can they take any part of plant? Can they take as much as they want? Do they need to ask permission? If so, from whom? Why? Can they take it any time? As much as they want? Can anyone take it? Can they sell it? If not, why? Who stops them? How do they stop them? Why this person? Are there exceptions to these rules? What happens if the rules are broken? Wild Foods: If someone wanted to gather some (X wild food), where would they get it? How far away is that? Can they take as much as they want? Do they need to ask permission? If so, from whom? Why? Can they take it anytime? If not, why? Who stops them? How do they stop them? Why this person? Can anyone take X from there? Can they sell it? What if there is a famine/severe food shortage? 55 Box 5 Question Guide for Women’s Resource Access Histories • • • • What resources do you primarily use from communal land? From your own land? From other ‘private’ land? In your opinion, how much do these (resources from communal/own/private areas) contribute to your overall household welfare? To your own personal welfare? What landscape areas do you utilize in the village (in order of importance)? Why? For each resource area: • Are there any rules about access to this area? • Are these rules always followed? Are there exceptions to these rules? • What are the main resources you get from this area? For home consumption or sale? • Are you able to take as much as you want? Any plant/tree or plant/tree part? • Have access rules to (each area) or to (each resource) changed in the past 30 years or since your mother/grandmother was using them? If so, how and why? Tell me when the first change occurred and how it affected you/your mother/grandmother. The next change? Etc., prompting for known changes as necessary. • What about the quality or amount of (each resource)? Has it changed? If so, how and why? • Are there areas that you use more now than in the past? If so, why? Are there areas that you (your mother/grandmother) used more in the past than now? If so, why? How has this affected you (your mother/grandmother)? • In general, how did the land reform change your access/use of resources, or your access relative to your mother’s? Relative to your spouse’s? • In general, how did villagisation change your access/use of resources, or your access relative to your mother’s? • In general, how did/do AE change your access/use of resources, or your access relative to your mother’s? Relative to your spouse’s? • In general, how has land degradation affected your access/use of resources? Relative to your spouse’s? • At the time of these changes, were women included/consulted about the decisions and how it would affect them? Were men consulted? If there are differences between men and women, why were there differences, and how has it affected men and how has it affected women? • Do you feel women are included/consulted now about decisions regarding resource access? Are men consulted? If there are differences between men and women, why are there differences, and how has it affected men and how has it affected women? • Have you ever spoken out about the effect of these changes? If not, why? • Does the lack of /less male labour in your household affect your access to resources? If so, how? Do you have other ways of getting access to these resources? Please tell me about them. Study. The botanical resource access ‘formula’ discussed in Section 3.A.2 provided the conceptual framework for this exercise. It is difficult to achieve a detailed understanding of the complexity of access rights at the village landscape level, since these rights vary by species as well as by landscapes. Therefore, individual species users and particular species were investigated with a focus on determining gender differences. While initial criteria for selecting case study species were developed, these were adapted after the community level studies were finalized. For each species, it was planned that 10-15 individuals would be interviewed, including five female heads and five male heads and their wives, where the selection would be done on the basis of substantial involvement of the individual with the species, and a preference for selecting people for whom Household Survey data had been collected. However, if patterns in use and access were readily apparent after interviewing a few individuals of both sexes, then fewer individuals would be selected, and additional species would be studied. Initially, two species were chosen. The question guide (Box 6) was designed to collect species-disaggregated information on the uses, parts used, destination of the products (i.e. on-site consumption, home consumption or sale), and the landscape niche where the plant or plant part is 56 Box 6 Question Guide for the Species Access Study What parts of X do you use? For what purposes? For each use, is it for on-site consumption, home consumption, or sale? Where do you get X (landscape niche) for each use? Is X self-sown or planted (per landscape niche)? If planted, by whom, why (for each landscape niche)? Who cannot plant it and why (for each landscape niche)? If needed, who gives permission for planting (per landscape niche)? What are the penalties for any violation? If necessary, ask about land tenure for the landscape niches where the plant/tree or product is collected. Who can gather each part of X (by landscape niche)? Is permission needed (by landscape niche, by user categories)? If so, from whom? Is it limited by season? Are there any exceptions (children, elders, specialists, etc.) (per landscape niche)? Is there a limit to what can be gathered for home consumption by any of these persons? For sale? For on-site consumption? If so, what is the limit? Why is there a limit? Who sets the limit? Who enforces the limits? What are the penalties for violations of the above? Does any of the above change depending on use (e.g. medicinal use)? Additional questions that can be added, depending upon the type of resource (e.g. tree, woody plants and shrubs), include: Can you hang things on the tree or stand things against it or use it for anything else e.g. when curing hides or to hang beehives? Would you have to get permission to do this? If so, from whom? Can you cut any part of the living tree for timber, poles, or fuelwood in X space? If so, do you have to get permission to do this? If so, from whom? Is there a limit to the amount that you can cut? If so, what is the limit and who makes that limit? Is it alright to sell what you have cut? Can you cut down the tree? Do you have to get permission to do this? If so, from whom? Can you harvest tree products such as fruit, nuts, pods, bark, etc. in X space? If so, do you have to get permission to do this? If so, from whom? Do you actually harvest any products? Which? Is there a limit to how much you can you take of X? Who makes that limit? Is it alright to sell X? Is there a limit to what you can sell? Who establishes this limit? Can you use things that have fallen from the tree and are on the ground under the tree, e.g. twigs, leaves, or fruits in x space? If so, do you have to get permission to do this? If so, from whom? Do you actually gather anything that has fallen? What do you gather? Is there a limit as to how much you can take of x? Who sets that limit? Is it alright to sell x? Can you destroy the tree e.g. by uprooting or chopping down individual trees or by clearing a patch of forest? Can you lend the use of the tree to someone else? Can you lease the tree? Can you sell the tree together with the land? Can you sell the tree apart from the land? collected, all of which could be cross-tabulated with information about the species users. Additionally, information would be collected on planting and access rights, and on the people and/or organizations responsible for implementing and enforcing any access regulations. The questions considered exceptions that might exist for any access rule (e.g. for children or specialists), or depending on use 57 (e.g. for medicine). It was also assumed that, at the time of these interviews, researchers would have a good understanding of the land tenure and uses made of the landscape niches where these species are collected. Due to the fact that the research period was shorter than planned, and that the species access study was to be completed toward the end of the research period, there was insufficient time to collect all of the data that was anticipated. This represented a substantial loss, since many of most important access distinctions are found at species level. Any future research should therefore ensure that more time is taken to collect data at species level. Although time was limited, researchers collected as much species level data as possible. Two species were selected based on their importance as well as the contrasts listed in Table 8. Table 8 Main Characteristics of the Case Study Species AWELIE (OLEA EUROPEA SSP. CUSPIDATA) Highly valued All parts used Prohibited use Women and men have different uses Uses: fuelwood, ceremonial, medicine, toothbrushes, construction, tools Scarce HAHOTE (RUMEX NERVOSUS) Abundant use All parts used Not restricted Uses are common to all Uses: fuelwood, fencing, wild food, medicine Abundant Key informants were chosen based on sex and residence (central versus rural), and for their knowledge and use of awelie. Only four women from more rural areas could be interviewed. Two more women were found at the end of the research period who illegally collect and sell restricted fuelwood species, including awelie. Another key informant was a man known to use awelie illegally. Since all people interviewed use hahote, this species was discussed with all of these informants. While the official rules regarding access to these species are fairly straightforward, there were details about actual use that only became clear during these discussions. In fact, researchers gained the best understanding of use and access when they discussed individual species rather than general uses of plants and resource areas. It was also found that many of the methods dealing with more general resource areas, such as the maps and the landscape calendars, were not very effective in capturing the use of borderlands, riversides, and roadsides. However, species level exercises did elicit much more of this information. Similarly, very rich information on access, plant use, and livelihoods was collected during the short period dedicated to the Species Access Study. 3.D Household Level Data Collection: the Household Survey The Household Survey was the instrument used to generate data at household level, which also generated data on the subsets of individuals within households that are the special focus of this study: female and male household heads and their wives. It was used to capture data on all of the important variables under study, and permit the inter-relations between the variables to be explored at household level as well as, depending upon the variable and indicators in question, at the level of individuals. The latter permitted comparison not only between types of households (MH versus FH), but as well between groups of individuals according to factors such as sex, age, and education level. 3.D.1 Content The Household Survey captured the variables presented in Table 9. It was most auspicious that several high quality household surveys had been carried out in the vicinity of the research site in the recent past, and the respective questionnaires were carefully analysed for their utility for the present study. The process of questionnaire design was greatly facilitated by employing survey instruments that pursue similar objectives and that have been adapted, tested, and successfully used to generate data in similar cultural, economic, and agroecological settings. To the degree that the 58 CONCEPT Livelihoods Table 9 Content of the Household Survey – Concepts, Variables, Dimensions, Indicators* VARIABLES DIMENSIONS INDICATORS Assets Natural Fuelwood & water access Agricultural land & tenure (|}) Agricultural land quality Tree planting & tenure (|}) Use of fallow land and tenure (|}) SECTIONS II.B,C V.A & E V.A V.B & F V.B V.B. V.B. V.G V.J VI.A VI.B VI.C Physical Buildings and tenure (|}) Crops stored Land improvements (|}) II.A II.D V.B. Human Education level (|}) I. Financial Credit (|}) Savings (|}) II.E II.F IV.B Water conservation (|}) Soil conservation (|}) Livestock feed sources Use of homegardens Use of common land areas (|}) Use of plants and trees (|}) Remittances, gifts, food aid, other transfers) (|}) Social Membership in organisations (|}), including credit and savings organisations (|}) Activities/ diversification Employment (including self) & income Off-farm labour (|}) Self-employment (|}) Community work Community work (|}) Food-for-work (|}) Earnings (cash, in-kind) (|}) Unpaid labour for own-consumption Type and division of labour for HH reproduction and production (|}) II.G II.E, F IV.A IV.A IV.C IV.B, C IV.A, B, C VIII.A *|} indicates that data is sex-disaggregated 59 CONCEPT Livelihood (con’t.) Table 9 Content of the Household Survey – Concepts, Variables, Dimensions, Indicators* (cont.) VARIABLES DIMENSIONS INDICATORS Activities/ Agroforestry production & income Land use Crop production belg; meher Diversification (con’t.) Agricultural sales & income Sharecropping Tree inventory; tenure (|}) Destination of tree products Source of planting material for crops and trees (|}) HH labour supply & division Livestock production & income Land use Inventory and ownership (|}) Products & destination (|}) Feed sources Live animal sales & purchases Income from sales (|}) Income from hiring out oxen/ploughing Fodder sources Ethnoveterinary supplies HH labour supply & division Homegardening Food security 60 Crop production and livestock products consumption Food supply Plant diversity - consumption Land use Production (|}) Destination of production As source of goods for HH consumption Sources of planting material As source of foodstuffs HH labour supply & division Crops for own consumption Meat consumption from own slaughter Sources per season per food type Food frequency chart – plant food SECTIONS V.A. V.C & D V.C & D V.E V.F V.F VI.E VIII.A V.A. V.G V.G, H, I V.G V.G V.H V.H V.I VII.A, B, C, F VII.F VIII.A V.A. V.J V.J VI.D & F VI.E VII VIII.A V.C & D V.G VI.D VII questions used are identical, it is also possible to compare results with data generated by the other surveys. The Household Survey design was derived in large part from the Ethiopian Rural Household Survey co-implemented by the International Food Policy Research Institute (IFPRI), Addis Ababa University, and Oxford University in the UK, to provide a longitudinal data set which began with seven peasant associations in 1989, expanding to 15 that were covered in 1994, 1995, 1997, and 2004. Two kushets nearby Hintalo Waijerat woreda are covered in the ERHS sample: Geblen, and Harrasaw (Gebre Egziabher and Tegegne 1996, Tewodros and Derbew 1996). The 1997 round emphasized the ownership of assets and permitted the in-depth analysis of gender relations and marriage in this context, as reported in Fafchamps and Quisumbing (2001). The data generated by this and the previous survey rounds are available online and, therefore, the results can be readily compared, although there was insufficient time to include such a comparison in this report. Such a comparison would be very useful insofar as it could help to determine the degree to which the bias introduced by the sampling procedure, as well as by the small number of households, affected the results. The ERHS, however, does not include questions regarding access to common land resources, homegardens, or other renewable resources such as plants and trees. Also, the crop codes that were used in that survey had to be supplemented by the limited ethnobotanical data available that was compiled and cross-checked.38 Other questionnaires that were consulted include those by Sharp et al 2000, Chisholm,39 and Cabal (2000). Further, data provided by Teshome (2003) and Selinus (1971) proved to be valuable for certain terms and concepts that are important in the study area. The sample of households for the selected village ideally would have been randomly selected to permit more statistical analysis, but at the same time the sample had to permit the different wealth categories to be represented as well as female and male headed households. Also, there were insufficient resources to permit more than 30 households to be surveyed, and the survey itself was as yet exploratory (it could only be pre-tested in the field with three households). Thus, purposeful selection was the only reasonable strategy. The sample of 30 households therefore included 15 FH and 15 MH, equally representing each of the wealth categories although, as indicated earlier, there were no FH in the kushet that fit into the higher wealth category. Therefore, it was necessary to seek other indicators to discern between the upper third and the middle third groups among FH, which was done using the emic categories presented in Table 6. Besides ensuring that MH and FH were correctly identified, it was also desirable to ensure that at least a minimum percentage of the households selected were utilising renewable natural resources outside of agriculture. Therefore, when initially contacting the selected households to determine their willingness to participate in the survey, a Qualifying Survey was administered. The Qualifying Survey contained questions oriented toward determining household headship, wealth category, use of CPR and, finally, willingness to participate and availability of household heads and spouses. A list of 50 possible households was selected with the help of village officials, who were asked to ensure that MH and FH of various economic strata and ages were represented. The 25 FH were identified by leaders of the women’s association. Qualifying surveys were then conducted with these households. From these surveys, thirty households (15 MH and 15 FH) were chosen to take part in the Household Survey based again on the same criteria, assuring that there was as equal representation as possible of the different variables. In general, the qualifying survey served to reinforce information given by village officials with respect to MH. However, they proved to be important with respect to the selection of FH: at times, FH were eliminated if they had males over the age of sixteen in residence. Those FH with adult males were reinterviewed to determine that they were not de facto male headed and, if they were not, they were surveyed. 38The initial list of crops came from the ERHS codes. Crops have been added, as have scientific names and some vernacular names in Tigrinya and Amharic, using the ethnobotanical sources mentioned in the references, particularly Jansen (1981). 39Via personal communications, Nick Chisholm kindly provided a copy of the questionnaire he used to generate data for Chisholm 2000 and 2004. 61 To reduce the possibility for error and increase efficiency of implementation, national enumerators who had experience collecting data for the ERHS were employed to administer the survey. These enumerators were nevertheless trained and supervised by the field researchers.40 The survey was very long and required three repeat visits to complete. Several sections had to be administered to the head of household or the person in charge of agricultural production, several had to be administered to the female head or wife, and some had to be administered to both, which also implied difficulties with scheduling as well as repeat visits. The survey work was completed in one month. 3.D.2 Data Analysis and Limitations The Household Survey data was analysed using SPSS and Excel using descriptive statistics (mainly crosstabs and frequency distributions) since statistical tests on such a sample would be meaningless. The purposive sample certainly also limited the generalisability of the results since they introduce bias. The problem posed for data analysis and interpretation, then, is to consider whether the types of biases that were introduced into the sample are likely to have a significant effect on the results, which depends on the specific topic. For example, there is not much likelihood that the type of sampling that was used to select FH would have an effect on the type of marriage contract, but it may have an effect on participation in foodfor-work. It is therefore very important to triangulate the study results; that is, to combine the results of the quantitative and qualitative research to indicate areas where data are too inconsistent or contradictory to permit conclusions to be drawn. It is also important to bear in mind that the results obtained are only indicative, and that larger scale, more representative research must be conducted. Not all of the survey data collected could be analysed, again due to lack of resources, and not all of the data that were analyzed could be fully exploited. The survey was very lengthy and, even though not all variables were entered into SPSS, 1866 were created. Some of the sections in the survey were ineffective in capturing the information required (especially in relation to assets in marriage and divorce, and with respect to the attempt to capture sources of food, construction materials, seed, etc.), but more time spent in training and supervision could overcome some of these difficulties. The factors that introduce bias into the survey results, as well as into the results of the other methods employed and into the field research in general, are discussed in the next sections. Factors that are believed to have facilitated the field research, and improved or otherwise enriched the results vis a vis other studies that have been conducted in the highlands, are also highlighted. 3.E Constraints and Opportunities to Learn: Improving Research to Improve Decision Making Aside from the constraints and opportunities mentioned above that are specific to the research methods employed, there are others of a more general nature that affected research results. Below, the main general constraints encountered are discussed. 3.E.1 Lost in Time: Scarcity of the Ultimate Resource The research depended in part on the collection of quantitative and qualitative data from government and non-governmental officials, as well as intensive use of villagers’ time. Especially at the woreda level, officials have hectic schedules and overbooked agendas. Many work seven days a week and 12 or more hours per day, not including emergencies. Researchers must adapt their own schedules to fit those of the officials. This is also true in relation to villagers: all are giving up time in order to participate that would 40It was rather surprising to learn that the enumerators were not at all accustomed to being supervised once in the field, or to having their work checked. One of the three enumerators had very poor performance, a second had fair performance, and one was sufficiently conscientious to be qualified to supervise the others after a period of time. Initial mistakes were numerous, including on those sections of the survey that they had presumably previously administered in the ERHS. 62 otherwise very possibly be used in daily livelihood activities. While compensation for their time41 helps and shows respect, even paid time has opportunity costs. Therefore, it is essential to allow enough time to avoid over-burdening informants with schedules that interfere with their most important activities. In general, given the short research period and the large number of CPR utilized, the methods employed turned out to be too numerous and too time-consuming. This often became an issue, not only for the researchers, but as well for the participants. 3.E.2 Lost in Space: Study Site Selection For reasons mentioned earlier, the study site was chosen by the head of the BSF project in Hintalo Wajirat woreda. His first choice, Adiarbaetu, was accepted after confirming that it met the main selection criteria: the correct percentage of FH and the existence of a significant amount of botanical resources. As was discussed in Chapter 2, agroecological conditions in Tigray are diverse, and this diversity extends to plant life. As a result, there is a variation in plant use throughout the region due to differences in altitude, climate, and resource degradation, as well as tradition. There are many benefits to choosing a resource site randomly, without giving particular attention to plant use, especially when considering dynamics, since environments and access regimes have changed. However, since the time available for field research was limited and plant use was a focal point of the research, it was necessary to choose a site where plants were known to be available and used. Although in many ways it would have been ideal to conduct the village surveys in order to choose a study site, in the end the researchers benefited from immense local knowledge of the project head who chose a site that was appropriate. However, the choice of this particular study site (or any other single site) did have implications for the research findings. For example, according to residents, due to its altitude, the use of wild foods in Adiarbaetu is far less than in other areas of Tigray, especially those at lower altitudes. The size of the kushet was also a limiting factor. With approximately 950 hectares of mountainous terrain, it can take over two hours to walk from one side of the kushet to the other. Given the distance and lack of time in the field, it was impossible to include households or individuals representing the entire kushet in the Household Survey sample or in the focus groups. Distances also limited researchers’ ability to explore the entire kushet and to become familiar with, and physically examine, all of the resource areas used and discussed. Researchers attempted to compensate whenever possible. While it was originally intended that the research would focus on resource areas within the kushet, it also became clear that, due to resource depletion, villages use many areas outside of the kushet to obtain plant resources. This increased the difficulties posed if researchers are to explore all resource areas used. The Household Survey collected data on areas used within the kushet only, whereas focus group participants were purposefully asked to include all resource areas used. The area is very diverse, and living conditions are equally so. The part of the kushet located closest to the tabia centre is nearly urban (a ‘town’). Homes are located close together; water is available to some in their backyards, whereas for others it is never more than a five or ten minute walk away; there are small ‘shops’ and ‘restaurants’ run from people’s homes; and government offices, churches, schools, and a health centre are all within easy walking distance. There is even one ‘restaurant’ with a generator and satellite dish showing TV programmes, football games, and nightly movies. Those living in the more rural areas of the kushet do not have easy access to any of these resources: many live hours from the market, school, and even water sources. Because of these differences, it is likely that these more rural residents rely upon and use resources differently, including plant resources. Researchers attempted to 41All participants were compensated for their time. Households were paid a flat rate for participating in the survey, and focus group members were compensated according to the hours invested. All participants were repeatedly informed that the money given to them was for their time and not for the information provided. There is much controversy regarding such payments. However, it was decided that reimbursement for time not only shows respect, but also brings immediate assistance to many of the villagers who are in great need. At the end of the fieldwork, one of the women focus group participants invited the lead field researcher to have coffee, and showed her an ox that she had purchased from the sale of her spice crop and the money she had received from her participation in the research. 63 represent these differences as much as possible, and to capture information from other parts of the kushet. A day trip was made to the farthest area of the kushet to meet with some households and become familiar with more of the area, and researchers explored as many resource areas as possible.42 Regardless, the research was concentrated in the area surrounding the central, more ‘urban’ part of the kushet. Therefore, results are not necessarily representative of the entire kushet. 3.E.3 Lost in Translation I: The Research Team and its Composition When dealing with research that contains so many layers and methods, a properly composed research team is essential. The team consisted of a head researcher at Wageningen University; a non-Ethiopian woman ethnobotanist who supported the head researcher in Wageningen and led the field research in Tigray; a female Ethiopian gender expert who translated and supported focus group discussions; and two male and one female national enumerators. As discussed in Section 3.E.5, it is also essential to include a botanist in a research project of this nature. Also, since translation was required, adding another national research assistant to the team would have greatly facilitated data collection. One researcher and one translator are insufficient to permit facilitation, data recording, observation of group dynamics, and translation. Inadequate staffing led to gaps in the information that were not noticed at the time because there was too much to do at once, which results in loss of data. If a translator is used (in Ethiopia this may be the case even when all researchers are national, since many don’t speak the local language), then it is necessary to have at least one other assistant to help record and facilitate. Also, because of the need to use male and female focus groups, it was not possible to separate female household heads and wives. Whenever possible, questions were asked regarding FH during focus groups discussions, but replies came from all of the women. There were insufficient researchers to permit one to direct the discussions, one to translate, and one to annotate responses to indicate which woman responded in what way. Therefore, most information gathered about the differences, if any, between these women, came primarily from the Household Survey. For similar reasons, it is ideal to have two surveyors per household, one to ask the questions and interact with informants, and another to record the information. 3.E.4 Lost in Translation II: Cultural Bias Embedded in Language and Methods Many methods that appear to be simple and straightforward in one cultural context are completely foreign in another, leading to invalid or incomplete results. For example, mapping proved to be problematic in Adiarbaetu. While it might have worked better had there been an official map available and greater awareness on the part of the researchers of the geographical and agroecological layout of the area, they began the fieldwork with the assumption that, even if mapping was new to participants, it could be easily learned. In fact, it was completely foreign to the residents to visualize the environment in two dimensions from a bird’s eye view. Another very important factor in research related to CPR is that there are differences between etic and emic ways of understanding the environment and CPR, including different ways of perceiving and naming landscape niches and botanical resources, and of understanding their importance (valuation).43 Time did not allow for the exploration of emic environmental and botanical categories, which was important to this type of research. For example, it was only well into the field research period that researchers became aware of the confusions regarding the use of the terms ‘forest’ and ‘woodlot’. ‘Unenclosed hillsides’ was also used as a distinct land use category in the literature reviewed, and thus in the question- 42An official had to accompany the researchers on such journeys, even though one of the researchers was an Ethiopian woman. This person was usually the DA for Natural Resources. However, he was extremely busy and, along with researchers’ busy schedule, it was nearly impossible to find time to walk through these areas, especially those more than an hour away. 43Local taxonomies for environmental classification are the subject of much scholarly interest and debate, and specific methodologies have been developed to explore these. They are also considered to be very important for environmental and ethnobotanical research. See e.g. Berlin (1992) and Ellen (1993b). 64 naires and guides, but, in such a mountainous area, people considered that all landscape areas were hillsides. Using it as a separate category caused much confusion. Proper understanding of emic categories and, in this case, environmental knowledge, is essential to collecting and properly interpreting the data. 3.E.5 Lost in the Jungle: Botanical Identification Even in an area where botanical resources are severely depleted, dozens or even hundreds of species are used. Due in part to this depletion, many of the species that informants use are not found throughout the kushet, and some are only found beyond its borders. Further, many species were not in season at the time of the research, and time restrictions did not permit researchers to collect or photograph all of the species that were in season. It was therefore very difficult for researchers to know precisely which species were being referred to in discussions and in the survey data. While researchers did become very familiar with many of the local names and the plants themselves, voucher specimen collection was needed in order to scientifically identify the species and relate them to their vernacular names. This gap in information poses a serious problem in ethnobotanical research and for understanding CPR access, since many of the access rules are based on life forms, species, and plant parts (i.e. trees versus herbaceous plants, leaves versus branches). Numerous plants mentioned in the surveys or discussions have not been scientifically identified, and even the life form is unknown. Voucher specimens are essential for scientific identification, which must be conducted by a plant taxonomist at a reliable research institute or herbarium. Further, while a literature review was conducted to assemble information on Tigrinya plant names and their scientific equivalents, it appears that most ethnobotanical research in Ethiopia has focused on other regions, and vernacular names are most commonly in Amharic. Many species have also not yet been taxonomically identified in Ethiopia, and it can be presumed that Tigray is under-represented in herbarium collections. Since it was not possible to collect voucher specimens during field research, hundreds of photographs of plants were taken and vernacular names were recorded, where assistance was provided post facto by the National Herbarium at Addis Ababa University. However, until voucher specimens can be collected, all of the species identifications made in this research must be considered as unverified. It is therefore essential to include a botanist in any research team that proposes to investigate plant resource use in such circumstances. Given the difficulties in locating the species and the problem of seasonality, this must constitute a longer-term effort involving local institutions. 3.E.6 Found at Last: the Advantages of Living in the Field In-depth research demands that field researchers live in the communities being studied, at least during the field research period. This, however, is quite unusual in the Ethiopian context, and particularly in the highlands, apparently because conditions in these areas are so difficult, and because educated urban residents do not wish to be subjected to them.44 Further, residents of Adiarbaetu noted on several occasions that they have never had more than a few hours of contact with any foreign project official or researcher: most such visitors did not leave their vehicles. It is also very likely that much research done to date on CPR regimes has been done by researchers who come into villages only long enough to carry out their surveys, spending the nights and weekends in Mekele or other urban areas. However, it was the fact that the researchers lived in Adiarbaetu for three months that led to some of the most important research findings and that made the research outcomes particularly unique and important in the highland context. Researchers won the trust of the community. Simply the willingness to live in the community earned them respect and acceptance, which greatly facilitated openness of dialogue with nearly everyone with whom the researchers came into contact. Further, by living in the kushet, researchers were able to observe many of the dynamics discussed in the literature, and to gain greater insights into how community members perceive that development programmes are or are not assisting them, while checking these perceptions through observation. Since trust was built, many offi44This assertion is based upon several discussions with researchers and their supervisors at Wageningen University and with scientists and development workers in Ethiopia. 65 cials told researchers what they thought was and was not working with projects in the area, as well as how they felt their input and knowledge were being received. Overall, villagers and officials alike stated that ‘outsiders’, including fellow Ethiopians involved in the project management, usually treat them with very little respect. The researchers were told many times that they were the first outsiders to talk to the residents themselves. Most outsiders, if they visit the kushet at all, only deal directly with select officials. They also admitted that they, at times, tell such project officials what they think they want to hear rather than what they know to be true. Thus, development agencies still lack enough awareness about how their programmes fit into the reality of village life, which is a main reason for programme ineffectiveness. Most programmes expend most effort attempting to instruct, and the least effort attempting to learn, from communities. A substantial amount of the qualitative information presented in this report was gained during informal discussions with villagers during daily life. Living in the village presented innumerable opportunities for interaction and dialogue, as well as participant observation. It is highly doubtful that female heads of household would have revealed to researchers the fact that they earn their livelihoods from prostitution, or that landless men would have openly discussed their illegal CPR use, if researchers had not lived in the village and gained their trust. 3.F Suffer the Little Children, or How Not to Get Lost Research on CPR access and livelihoods, and particularly on botanical resources in these configurations, is absolutely essential, since this and other recent research has demonstrated that such resources are vital and yet neglected in current policies. Several recommendations for further research can be made based upon the research presented herein. First, the experience with the research methods indicates that several of those used were very valuable and are probably necessary if the dynamics around CPR and livelihoods are to be understood, and particularly if landless and FH are to be supported. Based upon the results of this research, the Household Survey CPR components should be reformulated and administered to a much larger sample across a much larger region. Before administering the survey, however, it is important to carry out cultural cognitive research, particularly on land use and botanical taxonomies, and as well to carry out mapping exercises at local level that permit precise identification of resource areas that can then be coded in the survey and correlated with the data. The community mapping exercise is essential, but should be done with participants while walking or driving through the area in question, rather than in a closed room. Only through physical reconnaissance is it possible to accurately identify resource areas and coordinate the data collected spatially. The Household Survey should be complemented through a sub-sample where complete listings are made of plant species and resource areas used, since the listings done during the survey were made ‘on the fly’, and qualitative research as well as some of the other data collected in the survey (e.g., food frequencies) demonstrated that more species are used than were initially listed. Further, the food frequencies section of the survey is inadequate to determine intake of wild foods, which should be more rigorously quantified for nutritional assessment. Similarly, more in-depth quantitative research is needed to adequately capture the livelihood value of plants and other resources collected from common land areas: e.g., quantities collected, consumed, and sold. The importance of preliminary research cannot be overstressed. There is great need to assemble information on local plant taxonomies, to collect voucher specimens, and to scientifically identify the vegetative resources that are being exploited. Research on harvesting pressure is crucial for those resources that are most exploited, particularly those used for fuel. Preliminary research must also test methods for cultural bias or appropriateness, and appropriate length of the exercises. Particularly with illiterate populations, there is a need to identify means by which participants can deal with a multitude of species and landscapes through photographs, drawings, dried plant parts, etc. Children are an important part of such research not only because they use CPR differently than adults, but also because their knowledge and use patterns reflect change: it is highly likely that children in the highlands have far less knowledge than their parents or grandparents, but the knowledge that they 66 do have may provide an accurate reflection of current resource use and availability, and of the threats to biodiversity and to livelihoods that result from knowledge erosion and changes in CPR access. Finally, this research has demonstrated that it is very important to combine community level data with individual level data, but that greater priority must be given to the latter because it provides greater detail and specificity. Focus group discussions are useful for gaining insights into how communities perceive resource dynamics and for a first approximation regarding access and use. However, it is individual level data generated through instruments such as the Household Survey and the Species Access Study that provide greater specificity and certainty about the phenomena under investigation. 67 68 Chapter 4 The Research Site and its Population This chapter presents an overview of the micro-research context, particularly focusing on the FAO project area in Southern Tigray and on Adiarbaetu, the research site, and its population. The data and analysis presented here provide a point of reference and reflection for the empirical findings presented in the remainder of the report, given that statistical data was generated only with a very small, purposefully selected sample, and qualitative data could only be collected from a relatively small number of key informants. Sources of information for this chapter include scholarly and grey literature, particularly the FAO Project Document (FAO n.d.) and baseline nutrition study (FAO 2004a), as well as the Village Survey that was conducted with local officials (see Section 3.C.1), who were asked to provide tabia or kushet level data. Where data are lacking, Household Survey data are presented. Social, political, and bureaucratic organization is obviously crucial in asset distribution, and is therefore briefly outlined. The administration and political organization of CPR in Tigray is discussed in Chapter 2, whereas the specifics of their functioning in the research area are discussed in Chapter 6. Further, living conditions and welfare are discussed, which are in part determined by the presence and organization of basic public services and infrastructure at both village and household levels (especially water, health and education). There is a very brief discussion meant primarily to contextualize livelihoods (including agriculture and livestock production, homegardens, off-farm income generation, and food-for-work), since few data were available at kushet level. These topics are covered indepth at the regional level in Chapter 2, and at household level in Chapter 5, whereas CPR use and access are discussed in chapters 2 and 6. The information available from the Household Survey on demographic characteristics and household composition (including marriage and divorce practices) are presented in this chapter since, as was discussed in Chapter 2, these are important dimensions of the development dynamics in the study area that vary considerably according to locality; therefore they present another important part of the context in which household livelihoods and resource use in Adiarbaetu are played out. 4.A Situating Adiarbaetu Adiarbaetu kushet is located in Bahre Tsaba tabia within Hintalo Wajirat woreda in the south-eastern highlands (Southern Zone) of Tigray, approximately 64 kilometres from Mekele, the regional capitol (see Map, Graph 1). The estimated total human population of the kushet in 2004 was 3075;61 comprised of 526 households, 374 of which are MH and 152 (28.9%) of which are FH. There are 48 MH and 58 FH that officials identified as landless, for 12.8% and 38.2% of these households, respectively. The kushet is centrally located within the tabia, with tabia offices just outside its border. The kushet itself encompasses a mountainous area of approximately 950 hectares. Due to its terrain and size, those portions of the kushet located near the tabia centre have greater access to facilities and infrastructure. Residents located in the outer, more rural areas of the kushet, are more isolated and need to walk long distances to benefit from infrastructure such as water points, markets, and schools. Land quality, access to water, plant resources, healthcare, educational, religious and legal facilities, markets and income generating opportunities, are all influenced by and differ greatly depending on location within the kushet. As a result, location has a substantial effect on livelihoods. However, as discussed in Chapter 3, it was not possible to adequately represent more rural areas of the kushet. 4.A.1 Social, Political, and Bureaucratic Organization Regions are governed by an autonomous Regional Council which has economic and judicial powers independent of the Federal Government. Tigray is divided into four Zones and 35 districts, or woredas, which have an average population of 100,000 (FAO n.d.). Woredas in turn are constituted of sev- 61Since there is no official census record for the kushet, this figure comes from village officials and was last updated in 2004. Officials reported that the last unofficial census was conducted in 2003 by FAO. Therefore, this number is an approximation. 69 eral tabias, or sub-districts. The tabia is the lowest administrative unit and generally has a population of 5,000-7,000. Each tabia has its own baito which is popularly elected every four years. Tabias are often comprised of several (5-6) kushets and their surroundings, and kushets have their own baitos. The regional government has given substantial powers to the baitos: all socio-economic functions within tabias and kushets are co-ordinated and executed by them (Berhanu 2004). The baito consists of an executive committee of six persons, where the office holders are responsible for activities relating to economic development, social services, security, and propaganda. There is also a social court (Mahebrawe) that acts as the independent body for justice. The Ethiopian People’s Revolutionary Democratic Front (EPRDF) politically manages the government in conjunction with the Tigray People’s Liberation Front (TPLF), which operates throughout Tigray down to the kushet level. Many of their ex-combatants have administrative roles in the Tigray apparatus, as do the chairpersons of the Women’s Association of Tigray (WAT), and other mass organizations (Teshome 2003). WAT, which was formed after the civil war ended in 1991 out of the TPLF’s women’s fighters association and local women’s associations (Meehan 2004), is reported to have 400,000 members, and its mission is to empower women to participate in development. It engages in research, capacity building, advocacy and networking, and economic diversification and skills training efforts. WAT is active in Adiarbaetu, and village women pay dues, attend meetings, and elect leaders who then represent them vis a vis village officials. However, WAT leaders reported very low participation, especially in meetings. There are superstructures at woreda level that develop and/or implement policies and provide services and goods to the tabias. Of particular interest in this research is the structure of administration and decision making relating to natural resource use. Aside from the Regional BoANRD (see Chapter 2), the Hintalo Wajirat Woreda Agriculture Department coordinates overall agricultural development and has more than 60 agricultural workers involved in extension, soil and water conservation, forestry, irrigation, home economics, and animal and plant disease control. Some 70% of these are Development Agents (DA) who “have day-to-day contact with farmers and are responsible for the implementation of agricultural extension at farm level” (Teshome 2003: 46). Many international and national organizations are permanently present in the region and, as FAO pointed out, “The devolution of authority means that assistance to the people of Ethiopia can also be decentralised" (n.d.: 19-20). An important NGO that has been operating in Tigray since the height of the civil war in 1978 providing food and rehabilitation assistance, is the Relief Society of Tigray (REST); other important national and international NGOs, such as the Tigray Development Association (TDA) and World Vision, also actively support rural development. Nevertheless, FAO (Ibid.) also reported that, apart from the region’s Integrated Food Security Programme (supported by the EU and Irish-Aid) and the Dedebit micro-credit programme. Hintalo Wajirat receives limited development assistance. The Ethiopian Orthodox Church undertakes a rural integrated development project in parts of the woreda. The FAO Project itself is implemented through existing Food Security Offices at regional, zonal, and woreda levels. 4.A.2 Transport and Market Access Until 2005, Adiarbaetu was relatively isolated from more urban centres within the woreda and certainly from the city of Mekele. A seasonal dirt road officially opened in March connecting the kushet by means of a daily bus to the town of Adigudam and to Mekele. While village officials classify this road as ‘all-weather’, researchers considered it to be seasonal. Although research was not carried out during the rainy season, some very muddy conditions were experienced when rain fell, and vehicles were stranded. It is certainly not as all-weather as paved roads found in other parts of the region. Before this, the only means of travel was by foot. Few people use animals for personal transport: only one person was observed on horseback during the entire research period, and this was not within the kushet. Although people may own donkeys and even camels, these are used almost exclusively as pack animals. Bus fare is too high for most to afford, so they still travel long distances by foot, including to Adigudam (28 kilometres away) for the weekly market. However, with this improvement in infrastructure and easier access to the rest of the woreda and to Mekele, many people expressed hopes of finding new work opportunities, better access to more markets, and overall improvement in livelihoods. 70 4.B Water Access and Infrastructure The climatic situation and the frequency of drought have been described for Tigray in chapters 1 and 2. Adiarbaetu presents no exception. The dry season lasts from late September to May, and the rainy season from late May to September. While the 1984-85 drought is the most famous, Adiarbaetu experienced the drought of 2003, and residents considered 2004 as a drought year as well. Villagers repeatedly mentioned lack of water and drought as the primary causes not only food shortages, but also of resource degradation.62 Water access is therefore extremely important. There are several pumps and wells scattered throughout the kushet. In the more central areas, water is even pumped to some homes, to a single tap usually located behind the house. However, very few residents have this type of access. Most ‘urban’ residents obtain water from central wells and water taps, or purchase water from those with private taps. Those living in the more remote areas have to travel farther to retrieve water for themselves and for livestock. Those with land near rivers, even seasonal ones, naturally have better access to water. Also, parts of the kushet are located in valleys where the water table is higher, allowing for more vegetative growth for longer periods. Those with homes in this area are able to dig wells and support more productive homegardens if their land and other assets also permit.63 However, in other parts of the kushet, particularly those at higher and rockier locations, water is a major problem. Residents of these areas depend on communal wells, rivers, and rainfall. There are many who depend on rainfall alone. Table 10A shows the source and tenure of water for the 29 households that provided information, disaggregated by sex of the household head. For domestic use, 90% relied upon public or communal access, whereas only 10% had private sources. MH relied considerably more on communal sources (33%) in comparison with FH (6.7%). Of the private sources, only MH had wells. Of the public sources, rivers and streams provided the majority of households with domestic water (43%), with a higher percentage of FH dependent on these (60%) in comparison with MH (27%), where the latter also had greater access to public and communal artesian sources (40%) and hand pumps (13%). Thus, it would appear from the survey that FH are disadvantaged in terms of access to technology and possibly more secure (e.g. artesian and underground) domestic water sources. Since water availability for agriculture and livestock production is so important to food security in the area, projects (including the FAO project) have attempted to address shortages by introducing water conservation measures such as ponds, dams, and irrigation schemes into the more central areas of the kushet, while some of the farmers who have land near a water source, including along riversides, are developing their own irrigation methods (see also Section 5.A.5). Nevertheless, the Village Survey indicated that, across the entire kushet, approximately 400 hectares are rain fed versus just 18 hectares that are irrigated. Researchers observed one irrigation pump operating in a section of agricultural land near the centre of the tabia. It was reported that only 20 farmers who belong to a farm cooperative benefit from this pump, of which one is a woman. Collection ponds have also been introduced. These are dug by hand, lined with plastic, and covered with rocks to collect rainwater. Many residents were enthusiastic about these additions, and said that they have made a difference although, as is reported in Section 5.A.5, many of the ponds constructed appear to be disfunctional due to a lack of plastic sheeting. The Household Survey results regarding sources of water for agricultural use presented in Table 10A show that 13% of the households had no access at all (or didn’t require it since they didn’t have land). Of those households with access, only one MH and one FH had private wells (whereas only two households had access to a public or communal well), and there was no other private source of water for agricultural use. The majority of such water is provided by public sources, where streams (47%) and hand pumps (13%) are the most important. Interestingly, more FH had access to hand 62Interestingly, the national members of the research team often encouraged informants to also consider the implications of misuse of natural resources and overpopulation, perhaps because outside agencies consider these to be the main culprits. This is not to say that informants did not raise these issues on their own. However, local perceptions were overwhelmingly that, while drought was not the only cause, it was the primary cause of food shortage and resource depletion. 63In fact, when researchers visited one village resident, he was found in his backyard digging a pit latrine. He was nearly knee-deep in mud, having hit water after digging less than a meter down. 71 pumps, whereas MH were more dependent on rivers and streams (reversing the relation that was found for domestic water sources, so that FH would appear to be more water secure): five of the 12 FH who use water for agriculture use either wells or hand pumps, in comparison to only three of the 14 MH.64 Overall it appears that, aside from rainfall, about half of the surveyed households are entirely dependent on rivers and streams for both domestic and agricultural use, and this population lives relatively close to the centre of the kushet. 4.C. Demographic Data and Household Composition Considering Headship Hintalo Wajirat woreda’s population is estimated to be over 130,000, where 18% are children under age five and 21% are women of childbearing age. Life expectancy at birth is only 48.5 years, and total fertility is 6.8 children per woman (FAO n.d.). Equivalent data were not available for the kushet. The Household Survey was not based on a representative sample of the population. However, it does permit comparison of the selected households based upon male or female headship. The 15 surveyed FH households had a total of 43 members, 24 (56%) of whom were under age 18; the average number of household members was 2.87. In nearly three-quarters of these households, there was only one adult present. In contrast, MH had a total population of 80, with a slightly lower percentage of Table 11 Dependency Ratios by Sex of members who were under age 18 (52%), but the Household Head (n=30) average number of household members was MH FH substantially higher, at 5.3. In nearly half of the DEPENDENCY No. Cum. No. Cum. households, there were three or more adults RATIO % % present; in none were there fewer than two. The 0.0 1 6.7 1 6.7 overall dependency ratio for all surveyed .2 - .5 4 33.3 3 26.7 households was .97. Table 11 shows that FH .51 – 1.0 4 60.0 4 53.3 have on average higher dependency ratios, 1.1 – 1.5 4 86.7 1 60.0 meaning that the greatest burdens fall upon 2.0 2 100.0 3 80.0 female heads who, as will be seen in Chapter 5, 3.0 0 1 93.3 on average have far less capital and income. 4.0 0 1 100.0 4.D Education Infrastructure and Attainment by Sex, Age, and Household Headship There is one school in a neighbouring kushet that provides education through grade eight. However, the nearest high school is located in Adigudam, 28 km away, and those wishing to attend must either relocate to the village, usually without their families, or make the long journey by foot. Because of this, it is difficult for children to attend high school. While the grade school is within easy walking distance from some areas of the kushet, particularly the more central areas, residents in other parts must walk up to two hours each way to attend. While there is no fee for school attendance, materials must be purchased and many families find this too costly. Although the importance and benefits of education are well known to residents, the need for children’s labour at home and in the fields is more immediate, and may eclipse the extended benefits of education. Children contribute significantly to household tasks and therefore overall livelihoods (see chapters 5 and 6). They can be seen herding smaller livestock such as sheep and goats, as well as collecting dung, fuelwood, and wild foods. Young girls often help in household tasks. Children are also often the primary source of childcare for their younger siblings. It is not unusual to see children as young as five with a sibling on their backs. As a result, although education is important, hardship means that it is not always the first priority for households, which may in part explain the very high illiteracy rates found in the area. However, data presented below about children’s education in poor FH indicate that poverty and the need for children’s labour alone cannot explain the lack of school attendance. 64Given that all other indicators in this study show that FH are more deprived in terms of access to assets in comparison to MH, it would seem that this must be due to a special initiative that, quite unusually, has focused especially on them. 72 Table 12 presents the completed educational level of the surveyed household heads. Women heads’ educational disadvantage is very clear: 80% have no schooling, compared with only 27% of male heads. When all females versus Table 12 Educational Level by Sex of Household Head all males in the sample are considered (n=30) (Table 13), it is also evident that females have a strong educational disFEMALE advantage among all age groups. Even EDUCATIONAL MALE HEADS HEADS among the 5-9 year-olds, more than LEVEL No. % No. % half of all girls have not attended None 4 26.7 12 80.0 school, in comparison with 40% of Some nursery school 1 6.7 1 6.7 boys and, if the 10-19 age group is Primary 1-3 0 1 6.7 added, more than 67% of the girls Primary 4-5 5 33.3 0 have never attended school, compared Primary 6th completed 2 13.3 0 with 31% of the boys. The older the Secondary 7-8 2 13.3 1 6.7 female, the higher the percentage that Secondary 9-11 1 6.7 0 has never attended school, a trend University 0 0 which is not noticeable among males. Totals 15 100.0 15 100.1 In total, nearly half of the children aged 5-9 do not attend school. On the other hand, 47% of 10-19 year-old males and 37.5% of 10-19 year-old females have completed some level of secondary education; one male even attended university. Girls’ opportunities, at least among some households, are improving vis a vis their mothers, of whom only one attended secondary school. How do children of FH fare in comparison with those belonging to MH? Table 14 shows a phenomenon which is quite striking: 45% of the people pertaining to MH have never attended school, in comparison with only 12% in FH. This might be explained by the difference in age composition of the households. Since FH have a higher dependency ratio, it is clear that they have a greater proportion of younger children, and educational rates for all children have been increasing over time. However, it is also quite evident that primary school attendance in FH is higher both at the lower level of primary (28% versus 22%), and especially at the upper (40% compared with 11%); in secondary school, attendance is equivalent, at 20%. Especially 5-9 year-old children from FH are more than twice as likely to attend primary school in comparison with children from MH. As will be seen in Chapter 5, most FH are very disadvantaged compared to most MH, suggesting that the higher educational rates among FH are due to a greater propensity to send children to school.65 4.E Health and Health Care Infrastructure, including HIV/AIDS In rural Tigray, communicable diseases such as tuberculosis, meningitis, malaria, diarrhoea, and HIV/AIDs, and non-communicable health problems such as malnutrition, are the leading causes of morbidity and mortality.66 REST reported that the estimated infant and child mortality rates for Tigray are between 100 and 150 per 1,000 live births, where rural areas have higher rates. It also indicated that the maternal mortality rate for Tigray is estimated at 1000/100,000 live births, which is higher than the national average of 700/100,000. Further, in Ethiopia in general, it is estimated that one out of every 14 adults could be infected with HIV/AIDs. Tigray is one of the country’s worst affected regions: By 1993, commercial sex workers (CSWs) in the two biggest towns of the region (Mekelle & Adigrat) were exhibiting HIV seroprevalence rate[s] as high as 70%, and a 1999 study in [the] Eastern zone of Tigray among women attending antenatal care 65A possible explanation could be that FH have less need for children’s labour since they have fewer livestock and less land, but children from 5-9 years old are not usually much engaged in agricultural tasks: older children are mainly engaged. FH would possibly have even greater need for the labour of younger children to fetch water, fuelwood, and other CPR, and to care for younger siblings. However, it is not possible to determine why these rates are so different without more research. 66Besides these severe health problems, other disorders that are prevalent in Adiarbaetu include parasitic and respiratory infections, constipation and haemorrhoids. 73 AGE GROUP % 5-9 10-19 20-29 30-49 50+ Total 4 1 1 2 1 9 40.0 5.9 25.0 22.2 25.0 20.5 5-9 10-19 20-29 30-49 50+ Total 6 2 8 10 6 32 54.5 12.5 72.7 76.9 100.0 56.1 AGE 74 NONE NONE % 5-9 10-19 20-29 30-49 50+ Total 7 2 6 4 1 20 58.3 11.8 66.7 80.0 100.0 45.5 5-9 10-19 20-29 30-49 50+ Total 1 0 0 0 2 3 14.3 0.0 0.0 0.0 100.0 12.0 Table 13 Education Completed by Sex by Age Cohort (n=101) NURSERY/ % PRIMARY 4-6 % SECONDARY PRIMARY 1-3 AND ABOVE Males Age Five and Older 4 40.0 2 20.0 0 2 11.8 6 35.3 8 0 0.0 0 0.0 3 1 11.1 5 55.6 1 0 0.0 2 50.0 1 7 15.9 15 34.1 13 Females Age Five and Older 5 45.5 0.0 4 25.0 4 25.0 6 2 18.2 0.0 1 2 15.4 0.0 1 0.0 0.0 13 22.8 4 7.0 8 % TOTAL % 0.0 47.1 75.0 11.1 25.0 29.5 10 17 4 9 4 44 22.7 38.6 9.1 20.5 9.1 100.0 0.0 37.5 9.1 7.7 0.0 14.0 11 16 11 13 6 57 19.3 28.1 19.3 22.8 10.5 100.0 Table 14 Education Completed by Age and Sex of Household Head (n=101) NURSERY/PRIMARY 1-3 % PRIMARY 4-6 % SECONDARY + MH 4 33.3 1 8.3 0 4 23.5 4 23.5 7 1 11.1 0 0.0 2 1 20.0 0 0.0 0 0 0.0 0 0.0 0 10 22.7 5 11.4 9 FH 5 71.4 1 14.3 0 2 13.3 9 60.0 4 0 0.0 0 0.0 1 0 0.0 0 0.0 0 0 0.0 0 0.0 0 7 28.0 10 40.0 5 % TOTAL % 0.0 41.2 22.2 0.0 0.0 20.5 12 17 9 5 1 44 27.3 38.6 20.5 11.4 2.3 100.0 0.0 26.7 100.0 0.0 0.0 20.0 7 15 1 0 2 25 28.0 60.0 4.0 0.0 8.0 100.0 showed an HIV seroprevalence rate of 9.4% (town 12.45%, rural 6.4%). As data from seven urban (hospital based) blood bank centers suggested, prevalence raised from 5-7% in 1995/1997 to 12.5% (with peak prevalence of 20% in some centers) in 1999. Moreover, hospital based data shows [that] there were 3106 (56.7% of suspected cases) new HIV infections between 1995 & 1999.67 The incidence of HIV/AIDs in the region and its relation with sex workers is particularly of concern given that the Household Survey determined that two of the 15 female household heads (13.3%) reported that they had been sex workers, and it is certainly possible, given the poverty levels among FH, that this livelihood activity was under-reported. Sanitation is also a major issue. While there is increasing education on the importance of pit latrines, the majority of the households still do not have them. Due to the limited amount of land available to households, latrines are often still dug close to water sources, leading to additional health problems.68 Among the 30 surveyed households, a total of 14, or less than half, had latrines. MH were nearly four times more likely to have a latrine (73%) in comparison with FH (only 20%), where one of these female heads obtained her latrine when she rented a house where a latrine was present. This is probably because, as is reported in Chapter 5, FH have less access to land and to the labour with which to construct them. The rest of the latrines reported were constructed by the current dwellers, usually by husbands and wives working together, although a few male heads and two of the female heads reported building them alone. A new health centre for the tabia officially opened in April of 2005. It is located within a short walking distance from central part of the kushet, making it again about a two to three hour walk from the more distant areas. It offers medical care, family planning, vaccinations, and HIV/AIDS testing, and has a pharmacy. Currently there is no fee for services except for more specialized prescriptions. There are also traditional birth attendants within the kushet. The majority of health problems are still treated with traditional medicine, either within the household or through specialized traditional healers.69 While people are beginning to use the new health centre, residents seem to have little knowledge of what is offered, and hold misconceptions about costs. 4.F Livelihoods Written Large 4.F.1 Land and Agriculture The woreda project head reported that 98% of woreda residents derive their livelihoods from agriculture and livestock production, most of which is subsistence-oriented. Little statistical data was available on these activities at the kushet level, so most of the discussion of land and agricultural production in this report is based on Household Survey data, which are presented and analyzed in Chapter 5. The most important crops grown in the kushet are presented in Table 15. All food-for-work activities are paid in wheat. There is a weekly market within the tabia with many vegetables on sale including chillies (Capsicum spp.), tomatoes (Lycopersicon esculentum), potatoes (Solanum tuberosa), cabbage (Brassica oleracea capitata), and Swiss chard (Beta vulgaris var. cicla). Focus group participants reported that, due to their increasing cash value, more people are beginning to cultivate vegetables for sale in addition to grain and legume crops. Some are also grown in homegardens (see below). 67Relief Society of Tigray (REST), Health Division, October 10, 2000. A strategy for HIV/AIDS related interventions. Online at: http://www.u-fondet.no/sw209.asp (accessed 8 June 2006). 68All information presented here aside from that cited and the Household Survey results derives from observations while living in the kushet as well as conversations with health and sanitation officials. While it is possible that the digging of latrines close to water supplies is due to improper education, it is more likely that there simply is no alternative since few people have enough land to keep them at a safe distance. 69The role of traditional plant medicinals in livelihoods is discussed in more detail in Chapter 6. 75 Table 15 Principle Crops Grown in Adiarbaetu and Area (ha) TOTAL AREA CROPS GROWN (ha) Barley (Hordeum vulgare) 115.0 Wheat (Triticum vulgare) 86.0 Teff (Eragrostis tef) 40.3 Horse Bean (Vicia faba) 25.5 Sorghum (Sorghum spp.) 25.0 Peas (Pisum sativum) 15.0 Chick Peas (Cicer arietinum) 15.0 Flax (Linum ussitatissimum) 10.0 Maize (Zea mays) 10.0 Others 9.0 Within the kushet, 12% of the land is found at 0-1500m, 11% at 1500-2500m, and 77% is above 2500m. Cultivated land that is rain fed totals 350.75 ha, where land quality is average and the degree of erosion is characterized by officials as ‘medium’. There are 61 ha of grazing land, 116 ha of woodlots, and 400.25 ha of land in other uses. River gullies encroach on farmland during the rainy season, leading many farmers to lose most or all of their land. Measures to mitigate this include construction of soil and stone bunds, although these are often washed away. Chapter 5 presents data from the Household Survey on land tenure, access, and agricultural production, as well as on land quality and soil conservation measures. The woreda and the study village confront the same dynamics with respect to livestock that were discussed in Chapter 2. FAO (2004a) reported that over 90% of the households in Hintalo Wajirat do not possess individual grazing land, and that they are therefore divesting themselves of small livestock such as sheep and goats. Credit programmes have been set up by various organizations to permit households to purchase livestock, including oxen. However, not all people can participate. Particularly female heads complained that they are often not eligible for this type of credit, since it requires collateral in the form of household assets. Fodder depletion is a serious issue, and loss of livestock due to disease and malnutrition occurs particularly during drought years. Grazing land enclosures have been created in the kushet and these and their use and administration are discussed in Chapter 6. Livestock production in the surveyed households is discussed in Chapter 5. 4.F.2 Homegardens As was discussed in Chapter 2, contrary to other parts of Ethiopia where high plant species diversity and intra-specific diversity are found in homegardens, the number and extent of homegardens in the study area are very limited. The FAO project assumed that nutrition could be improved by increasing vegetable consumption: it reported that many households already grew hamli talien (Amaranthus graecizans?; Brassica spp.?) (a wild green), as well as a few other vegetables such as onions and tomatoes, which would provide a basis for increasing homegarden production. It sought specifically to increase production and consumption of oil as a carrier for fat-soluble vitamin A. It also sought to promote homegardening as an income generating activity, particularly as an alternative for those households with very little or no land, and mentioned specifically FH and those without oxen, since homegardens don’t need to be ploughed (FAO n.d.). It anticipated that, as a result of the project, homegardening would be practised by at least 75% of households without oxen, both for own consumption and for sale. It also sought to increase small-scale production of indigenous fruit trees and vegetables selected by the communities. The results of the nutrition survey carried out by FAO (2004a) showed that about 75% of the households surveyed in the project area had no gardening tools, while the remaining households had only a hoe. It noted that only a few households grew vegetables on plots ranging from eight to 13 m2. Production ranged from 13.5 kg per year for kale, 15.5 for garlic, 22.0 for onion and 58.0 for potato; fruits of any kind were non-existent. The report did note, however, that beles (Prickly pear; Ficus opuntia-indica) are cultivated around many homesteads in areas of 36 to 40 m2, where the mean production reached 150270 kg per household (the fruit are eaten and the pads are fed to livestock). It is reported that there are currently only 106 homegardens within the kushet, which means that less than a third of all households have them. However, many of these are very small and cannot produce more than a few vegetables. Villagers cite water shortage and lack of land as the primary reasons for the absence of a homegarden, but other factors cited include poor soil, livestock destruction, and the inability 76 to purchase seed. It also appears that homegardens were not a traditional form of land use in this area. The data on homegardens from the Household Survey are presented in Chapter 5. 4.F.3 Off-farm Activities According to the Village Survey, livelihood activities in Adiarbaetu are relatively sex-segregated. Men’s livelihood activities include staple crop production, the production of cash crops such as spices and vegetables, trade in livestock and petty goods, collection and sale of stone, wage labour performed on a daily basis, and food-for-work activities. For women, the most common livelihood activities include tela production and sale, sale of other beverages such as tea and coffee, petty trade, hair braiding, sharecropping out land, wage labour on a daily basis, and food-for-work activities. Both men and women may also earn income from the sale of products from trees grown on their own property, as well as from homegardens. However, only men migrate for work. Locally, men are able to find employment in surrounding kushets in construction and carpentry where they earn about 30 birr (about US $3.30) per day. If this work is unavailable, they may also travel farther. Often they travel to Raya, which is approximately three to four hours by car from Adiarbaetu, where they can find abundant employment as day labourers on farms. However, considering the travel costs, net earnings are significantly lower, at approximately 10 birr per day. This type of employment is not without risk. Officials reported that delays in payment or failure to pay for work completed are common. The risk of injury is high in construction work in particular. For many people, the time spent away from their own farms means that their production suffers. 4.F.4 Food-for-work and Compulsory Community Labour Food shortages are a serious problem throughout Ethiopia, but particularly in drought-prone Tigray. Food shortages are reflected in the high prevalence of wasting and stunting among children. According to REST, in 1998, 18.2% of the children under age five were wasted, 61.7% were stunted, and 48% were underweight in Hintalo Wajirat (reported in FAO n.d.). Within Adiarbaetu, an annual food shortage period of two months or less means that it is a ‘good year’; in drought years, food shortages are experienced for six months or more. Many individual households experience far longer periods of food shortage per year, as is discussed in Chapter 5. Food-for-work has therefore become ‘a way of life’ in Tigray and in the study area. In Hintalo Wajirat Woreda, over fifty percent of households were reported to have participated in food-for-work (Teshome 2003). A person received three kilograms of wheat for participating in public works such as dam construction or soil and water conservation. In addition, food aid recipients are obliged to clean their kushets, construct public servants’ houses, or maintain rural roads for five days per month. In addition to food-for-work, community members must contribute twenty days of unpaid labour per year for public works. Soil and water conservation activities, including stone terraces, soil bunds, check dams, flood diversions, trenches, and maintenance works, constitute many of such activities, as do contributions to the development and maintenance of woodlots (Teshome 2003, Jagger et al 2003). Some of this work is carried out through the World Food Programme, some through REST, and some through the Ethiopian Social Rehabilitation and Development Fund (Teshome 2003). 4.G Household Structure, Marriage, and Divorce In the study area, of the total population captured in the Household Survey that was 17 years and older (a total of 61 people), 27.9% were single or never married, 47.5% were married, 1.6% were separated, and 8.2% were widowed. In addition to having a higher percentage of single people than that found in the IFPRI study (Fafchamps et al 2001, see also Chapter 2), most singles were younger (in the 17 to 21 yearold age group), and the youngest married person was 19 years old. Of the surveyed household heads, Table 16 shows that all male heads are currently involved in a union, and only one male head had previously been married. On the other hand, four of the female heads reported that they were single, and three of these had never married. Of the single women, one had two 77 Table 16 Marriage and Assets at Marriage, by Sex of Household Head MH FH TOTAL MARITAL INDICATORS No. % No. % No. % Marital status of head (current) (n=30) Single, never married 0 0.0 3 20.0 4 13.3 Married 14 93.3 0 0.0 13 43.3 Separated 1 7.7 0 0.0 1 3.3 Divorced 0 0 9 60.0 9 30.0 Widowed 0 0 2 13.3 2 6.7 Missing 0 0 1 6.7 1 3.3 Number of times married previously (n=30) Never 14 93.3 3 20.0 17 56.7 Once 1 7.7 7 46.7 8 26.7 Twice 0 0 4 26.7 4 13.3 Three times 0 0 1 6.7 1 3.3 How long first union lasted 1-2 years 0 0.0 2 16.7 2 15.4 4-10 years 0 0.0 6 50.0 6 46.2 11-20 years 1 100.0 3 25.0 4 30.8 More than 20 years 0 0.0 1 8.3 1 7.7 Totals 1 100.0 12 100.0 13 100.1 Who chose last spouse Self 3 21.4 6 46.2 9 33.3 Own parents 11 78.5 5 38.5 16 59.3 Own elders 0 0.0 1 7.7 1 3.7 Another person 0 0.0 1 7.7 1 3.7 Totals 14 99.9 13 100.1 27 100.0 If spoke to last spouse before marriage Yes 0 0.0 4 30.8 4 14.8 No 14 100.0 9 69.2 23 85.2 Asset status before marriage Farmed independently 2 14.3 6 46.2 8 29.6 Had own land 2 14.3 3 23.1 5 18.5 Had own house 1 7.1 2 15.4 3 11.1 Had own livestock 1 7.1 0 0.0 1 3.7 Totals 6 42.8 11 84.6 17 63.0 Asset status upon or after marriage Received land from parents 1 7.1 1 7.7 2 7.4 Received livestock from parents 3 21.4 5 38.5 8 29.6 Brought own land into last marriage 2 14.3 3 23.1 5 18.5 Received land when she/he married you 8 57.1 3 23.1 11 40.1 children, two lived alone, and one was 17 years old and lived with her young nephew and 70 year-old mother. Thirteen female household heads reported having a first union that ended. Female heads more frequently reported that they themselves chose their previous spouse, but in this case these women were more likely to have been married more than once (30% of the cases), and to have farmed independently prior to marriage (50% of the cases). However, Fafchamps et al (2001) found in the 1500 surveyed households that involvement in the choice of a spouse was not more common among previously married individuals than among people who were never previously married. In the Household Survey, women heads were more likely than male heads to have farmed independently before marriage, and slightly more 78 likely to have had their own land and house, which probably reflects the differences in marital histories. Nevertheless, the percentage of heads having own assets prior to marriage in general was low (less than 20% for land, 11% for a house, and less than 4% for livestock). This was also the case with the percentage of heads who received land from their parents upon or after marriage, which was equal for men and women heads, but also very low, although heads were more likely to have received livestock, particularly the women. This reflects the more general situation found by Fafchamps and Quisumbing (Ibid) that dowry, bride price, and other ritual gifts account for only a small percentage of land transfers (see Section 2.C), but it does not reflect their conclusion that land and livestock that are inherited after marriage come primarily from the husband's family, or that daughters barely inherit anything from their parents. Interestingly, more female heads reported that they brought land into their marriage in comparison with male heads, although more men reported receiving land when they married their spouses. The one man gave ‘disrespect’ as the reason for his divorce whereas, of the female heads, all divorced women said that the reason was ‘incompatibility’. One woman said that her union ended because her husband had entered the army, and five stated that their last union ended due to death. Nearly 50% of the female heads have been married once, slightly more than a quarter had been married twice, and one woman had been married three times. As was reported in Section 2.C.2, focus group participants in Adiarbaetu gave poverty as a principle reason for high divorce rates. Women who marry without a dowry cannot marry formally (with a contract), and their husbands easily abandon them for wealthier women. Keeping a concubine or having children by other women is another common phenomenon that leads to divorce. The war has contributed to the high number of widows. Women whose husbands died in the war receive benefits from the government of approximately 50 birr per month. What happens to assets upon divorce or widowhood is obviously crucial to female heads. First, of those heads that had divorced, the one previously divorced man indicated that his wife remarried in the same place of residence. Of the women divorcees, two reported that their husbands remained single and still live in the same place, whereas three said that their ex-spouses moved away and remarried; and one didn’t know where her former husband was. Table 17 Disposition of Children and Household Assets upon Termination of Last Union, by Sex of Household Head (n = 30) MALE FEMALE HOUSEHOLD ASSET AND HEADS HEADS TOTAL WHO OBTAINED No. % No. % No. % Custody of young children Not applicable Myself Custody of older children Not applicable Myself Land Not applicable Myself Husband’s livestock Not applicable Myself Wife’s livestock Not applicable Myself Cash savings Not applicable Myself 15 0 100.0 0.0 7 8 46.7 53.3 22 8 73.3 26.7 15 0 100.0 0.0 13 2 86.7 13.3 28 2 93.3 6.7 14 1 93.3 7.7 12 2 85.7 14.3 26 3 89.7 10.3 15 0 100.0 0.0 13 1 92.9 7.1 28 1 96.6 3.4 14 1 93.3 7.7 14 0 100.0 0.0 28 1 96.6 3.4 15 0 100.0 0.0 13 1 92.9 7.1 28 1 96.6 3.4 79 Upon divorce or death of their spouse, Table 17 shows that women generally reported that they retained custody of all children, but very few reported that they retained any of the household’s land or livestock. Eight of the female heads said that they retained custody of children (both younger and older), for 100% of the cases where there were children involved. The one divorced man indicated that he retained the household’s land and his wife’s livestock, whereas only two of the women reported receiving any household land when their unions ended (17% of those women whose first union ended). No woman reported that she obtained either the house or the trees belonging to the household, and only one woman reported that she kept her husband’s livestock. One woman said that she kept the household’s cash savings. However, the data relating to how female heads acquired their land show that the situation is more complex than this, and that it is likely that more women heads have access to land through their previous marriage(s). In Table 24 it can be seen that a total of 53.1% of the plots that women heads possess are owned together with either their (ex)’spouse’ or with an ‘original owner’ who would be either a spouse or a parent. Further, Table 23 shows that 24% of the plots that women households owned were obtained from ‘another’ source, which is also possibly an (ex)spouse, or a father of their children with whom they were never married. What this would imply is that, even though women are divorced or otherwise separated from their previous spouses, they ‘own’ land together. This would reflect the traditional practices that were described by Hoben when discussing the rist system: he noted that, upon divorce, "A well-to-do man often allocates the use of a field or two to a divorced wife if she has children by him and has not married again. She, in turn, gives the land out in tenancy” (1973: 139). Or, it may imply that women heads do not see this land as belonging exclusively to themselves, but rather recognize that their ex-husbands retain some type of claim to the land, e.g. in the case of her death or if the children return to his household. Alternatively or additionally, it may reflect the contemporary de jure context: an informal arrangement where spouses who are de facto living apart have not subjected their holdings to formal separation since this might imply that their landholdings would be subject to reallocation by the baito. This discussion reappears in Chapter 5 and in the conclusions in Chapter 7. 80 Chapter 5 Household Headship, Gender, and Livelihoods This chapter examines the livelihoods and livelihood strategies of 30 households in Adiarbaetu, where 15 female and 15 male headed households were selected to represent various wealth strata, as discussed in Chapter 3. In distinction to Chapter 6, the present chapter deals with private rather than common resources, and with the range of assets and activities that households engage in to gain their living. Livelihood strategies of different types of households are related to resource access dynamics in the study area. Livelihood diversification strategies used to procure food and other life essentials are compared between MH and FH, with the objective of critically examining the relations posited in Chapter 2 with respect to social differentiation and poverty. Intra-household relations within MH are also explored to the degree possible, in order to better understand the differences and similarities between women in FH and those in MH, and to address some of the issues relating to bargaining power that were raised in the literature. While the cash contributions that are made by CPR resources to livelihoods through value-added activities and trade are discussed in this chapter, all other CPR contributions to household livelihoods are covered in Chapter 6. This chapter begins with a discussion about access to and use of private land and other productive assets, and then relates these to crop and livestock production and consumption. The existence and importance of trees and homegardens, and their roles in livelihood diversification are also analyzed. The discussion then turns to off-farm employment and its contributions to livelihoods, as well as other sources of income and subsistence goods such as community work (including food-forwork), transfer payments, credit, and savings. Finally, the chapter presents information about participation in community organisations in order to examine the degree to which women and FH can avail themselves of social resources. The Household Survey that was used to collect the data was based on the ERHS (see Section 3.D) which was modified to generate data on the livelihood contributions of CPR as well as of onfarm botanical resources such as trees and homegarden species. As was also discussed in Chapter 3, sampling was purposive rather than random. Still other factors bias the results, including the fact that most of the households sampled reside in the more ‘urban’ part of the kushet. While the survey captured a good deal of information about assets and control and decision making related to them, it did not succeed well in capturing total income, total food consumption, and the contribution of social networks to livelihoods through gifts and other forms of exchange. The reliance on quantitative methods for capturing livelihood dynamics was great, so that further qualitative research is needed to complement the data presented. Heads of households were the respondents for the sections in the survey relating to land access and land tenure, agriculture, livestock, and trees. Respondents for the questions relating to homegardens were either the female head or the wife of the head. Women respondents also answered the questions about non-agricultural activities that their household members engage in to generate income in cash or in kind. While a central objective of the research was to determine why FH in the study area are so disadvantaged in comparison to MH, at the same time the discussion must address the factors that contribute to the poverty of many MH. No one in the study area is wealthy: even the households with the most assets are usually food insecure for a month or more per year, and would be considered by those living in more favourable contexts in Ethiopia to be poor. Especially in the case of severe drought or other calamity, these households can easily lose their assets and even their lives. However, it was not possible to analyze all of the data collected to determine the contribution of the development dynamics discussed in this report to MH poverty and differentiation; resources were insufficient and the sample size too small. 5.A Private Land and Its Use 5.A.1 Private Land Access As discussed in chapters 1 and 3, land scarcity and landholding fragmentation in the study area mean that agricultural production cannot sustain an average household, and families can no longer provide 81 a substantial amount of land for their offspring. Landholdings among the surveyed households were indeed very small, ranging from less than an eighth of a hectare to one hectare, where the majority of households had between one-eighth and half of a hectare. Two of the households were landless, one of which was male headed. Within households, landholdings were usually dispersed across two or more plots that varied in size, quality, and distance from the homestead: the 28 households with land held a total of 95 plots, for an average of 3.5 plots per household. Of the total number of plots, 64 pertained to MH, whereas FH had 31, or less than half as many. The total land area was 76.5 timads (9.5 ha), for an average of around one-third of a hectare per household. All FH together held slightly less than 21 timads, for an average of 1.4 timads (.175 ha) per household, whereas all MH together held 55.5 timads, for an average of 3.7 timads (.46 ha), or a ratio of around 3:1. Table 18 shows that more than two-thirds of MH held more than two timads, and that the most land to which any household had access was eight (one ha). In contrast, 80% of FH had two timads or less, and the maximum was 3.5 timads. The surveyed FH are, as the literature reports for FH in Tigray and other highland regions, seriously land deprived. As was discussed in Chapter 2, Table 18 Total Landholdings by Sex of Household Head the baitos used various criteria when allocating land, one of which was the SIZE* OF MH FH Total number of adults and children in a HOLDING % No. % No. % No. household. Table 19A shows that there Landless 1 6.7 1 6.7 2 6.7 is some relation between size of .25 - < 1 Timad 2 13.3 4 26.7 6 20.0 household and total landholdings in 1 to 2 Timads 2 13.3 7 46.7 9 30.0 the survey data, seen particularly in 2.1 to 5 Timads 6 40.0 3 20.0 9 30.0 the lower and upper ends of the spec> 5 Timads 4 26.7 0 0.0 4 13.3 trum (1-2 and 7+ household members), Total 15 100.0 15 100.0 30 100.0 but this doesn’t appear to be very *One timad = 1/8 ha strong, particularly for households with 3-6 members (the majority). This may be attributed in part to the over-representation of FH in the sample: among these households, those with 3-6 members had proportionately less land in comparison with MH in the same strata. Considering that FH also generally had higher dependency ratios (one-third had ratios of 2.0 or above, compared with only 20% of MH (Table 11)), which Sharp et al (2000) showed are positively correlated with destitution (Section 2.C.6), it can be inferred on this basis alone that these FH are at greater risk of poverty and food insecurity. Aside from the amount of land to which households had access, land quality was also considered in baito land allocation. The survey captured data on soil quality and slope as reported by farmers using local terms. Of the plots allocated to MH by the baito (Table 20A), 56% were of good (lem) soil quality, whereas only 24% of the plots that FH received were of comparable quality. About twothirds that FH held were of medium soil quality (lem-teuf), compared with a quarter of the baitoallocated plots that MH held; a slightly lower proportion of plots held by FH (14%) were of poor quality in comparison with MH (17%). From Table 21A, which considers all land to which households had access, it can be seen that FH in general had access to poorer quality soils in comparison with their male counterparts: of the plots to which MH had access, just over half were considered to have ‘good’ soil quality, compared with less than 40% of those to which FH had access. At the other extreme, only 12% of MH had plots with poor soil quality, compared with nearly 20% of the FH plots. This indicates that, when FH access land by means other than through the baito, this land also tends to have poorer soil quality in comparison with the land that MH access by other means. With regard to the slope of the land that was allocated by the baito, 90% of the plots that MH held were flat (medda), compared to 81% of the plots that FH held. MH also had a larger proportion of plots of moderate incline in comparison with FH (10% versus 5%), and it was only FH that reported having plots on steep slopes (6% of their plots). When plots accessed by all means are examined, the situation remains unchanged, other than the fact that FH tended to have a more equal proportion of land on moderate slopes. From informants’ reports, there appears to be little relationship between land quality and slope, so that poor quality soils are found on flat land, and good quality soils are found on land of moderate incline, but steeply inclined land generally has soil of only medium quality. In summary, FH tended to have less land access with lower soil quality in comparison with MH, and baito-allocated land found on steep slopes was only held by FH. 82 A final consideration in the allocation of baito land was the distance of the plot from the homestead. While plots that are closer to the homestead may be more desirable, soil quality considerations may mean that it is not rational for households to have all plots close by, and distribution of the plots across different agroecological niches was also considered to mitigate risk. Table 22A shows that a third of plots were within the homestead, and thus would include the house and any homegarden, whereas approximately another third were within a 5-10 minute walk. FH tend to have their plots within or very close to the homestead, or further than a 30 minute walk (84% of all plots), in comparison with MH (64% of the total). The higher percentage of FH’s plots that were at or closer to home reflects the fact that they had fewer total plots of land, and a larger proportion of this land was used for their homesteads. The type of data presented above are not available across the kushet, but there is no reason to think that the sample was biased in terms of representing types of households, other than the fact that these were relatively urban. The data presented either tends to indicate that baito land allocation was biased against FH or, alternatively, the land that FH reported as having been allocated to them by the baito was actually land allocated to the household during their marriages and, upon divorce or death of the husband, the household’s baito land was unequally divided, where women were given the worst of the households’ land. This latter interpretation tends to be supported by the data presented below on land tenure. 5.A.2 Land Tenure and Decision Making Land tenure in the survey referred both to who currently ‘owns’ each plot of land (that is, who has freehold tenure), and to how the land was acquired. Table 23 shows that nearly two-thirds of the plots to which households had access were owned by the head, whereas about 17% were rented or sharecropped in. MH were much more likely to be the sole owners of their plots in comparison with FH, where half of the FH owned plots together with the ‘original owners’ (presumably mainly exspouses). Only MH rented or sharecropped (‘leased’) land in, although the ownership category ‘head and tenant’, which only FH declared, refers to the fact that these women sharecrop out land. As is the case in other Table 23 Tenure of Land Plots by Sex of Household Head parts of Tigray (Beyene 2003), the main means of land access among MH FH TOTAL the surveyed households were TENURE OF PLOTS No. % No. % No. % freehold acquired from the local Head owns 47 74.6 13 40.6 60 63.2 baito, followed by sharecropping, Head & spouse own 0 0.0 2 6.2 2 2.1 and then rentals (Table 24). More Head & original 0 0.0 15 46.9 15 15.8 than two-thirds of all plots were owner allocated by the baito, another Head & tenant 0 0.0 2 6.2 2 2.1 23% were sharecropped or rented Tenant 16 25.4 0 0.0 16 16.8 in, whereas only a small fraction Totals 63 100.0 32 99.9 95 100.0 was inherited: land purchases are not legally permitted, but heads reported that 4.3% of the plots were purchased.79 In all cases where it was reported (29 out of 32 plots, or 90%), FH said that they acquired the land themselves, but this does not preclude that they actually obtained it through divorce, as is surmised below.80 Only seven percent of the total plots pertaining to MH were acquired by the wife, who was either allocated the land by the baito or purchased or inherited it. In 13% of the cases, both the head and wife acquired the plot and, in all of these instances, the land was allocated by the baito. Male heads alone acquired the majority of all plots (57%), generally having been allocated the land through the baito or having acquired it through sharecropping, principally from non-relatives. FH, on the other hand, reported that a much larger proportion of their land was allocated by the baito; however, inheritance is still an important source of land for FH, which is in contrast to what Fafchamps and Quisumbing (2001) found in the ERHS data, that women rarely received land from their own lineage. What the data show altogether is that women in MH 79Nedessa et al (2005) report that poor peasants across different regions of Ethiopia are informally selling land, and that informal land markets are prevalent even though they are officially illegal. 80As noted in Section 4.G, the survey was not very effective in capturing the division of assets upon divorce. 83 Table 24 How Acquired Land and Where Acquired Land by Person Acquiring Land BOTH MALE WIFE OF WIFE AND FEMALE TOTALS TENURE INDICATOR HEAD HEAD HEAD HEAD No % No % No % No % No % How acquired Allocated 31 58.5 2 50.0 7 100.0 21 72.4 61 65.6 Purchased 2 3.8 1 25.0 0 0.0 1 3.4 4 4.3 Inher1 1.9 1 25.0 0 0.0 5 17.2 7 7.5 ited/parents’gift 17 32.1 0 0.0 0 0.0 0 0.0 17 18.3 Sharecropped in* 2 3.8 0 0.0 0 0.0 2 6.9 4 4.3 53 100.1 4 100.0 7 100.0 29 Rented in 99.9 93 100.0 Totals Where acquired Baito 33 62.2 2 50.0 7 100.0 21 72.4 63 67.7 Husband’s par3 5.7 1 25.0 0 0.0 4 4.3 ents 0 0.0 1 25.0 0 0.0 1 1.1 Wife’s parents 15 28.3 0 0.0 0 0.0 1 3.4 16 17.2 Non-relative 2 3.8 0 0.0 0 0.0 7 24.1 9 9.6 Other 53 100.0 4 100.0 7 100.0 29 99.9 93 100.0 *Most likely under-reported (see FN 3, below). barely received land from the baito (only 5% of the plots): rather, baito land was registered either in the name of the couple (17.5% of the plots) or, most commonly, in the name of the male head (77.5% of the plots). This would tend to support the conclusion that divorced FH who reported having received land from the baito were actually reporting that land which they received from their exhusbands which was allocated to the latter through the baito, which would also explain the high percentage of plots that FH reported were co-owned. Since the baito does not generally interfere in land allocation upon divorce, land that FH access probably continues to be registered in the names of their ex-husbands. Thus, achieving greater equality in land access for FH would imply that baitos should become involved in divorce settlements, and should not register land exclusively in the name of the male head. The literature on gender and land access in northern Ethiopia that was reviewed in Chapter 2 indicated that decision making within households regarding major production resources tended to be egalitarian, particularly when such assets were held in common, or when women brought own assets into the household. It could be hypothesized that the importance of baito-allocated land in households’ asset portfolio would tend to reinforce egalitarianism in decision making since, even if land was registered in the head’s name, legally the land pertains to individuals within the household, and divorce is supposed to result in equal division of assets, which is also culturally sanctioned (see Section 2.C.2). Male respondents were queried about who can make decisions regarding what to grow on agricultural land, as well as on the disposition of land (giving it away or sharecropping it out), and the responses are presented in Table 25A. In MH, irrespective of who acquired the plot, men reported that they and their wives both make the decisions regarding what to cultivate, with the exception of a male who decides alone what is to be cultivated on a plot acquired by his wife. Female heads make this decision alone with the exception of two women, who reported that they and their tenants (sharecroppers) together decide what to cultivate. Once again, in relation to who can decide to give away a plot of land, men reported that they can do so only together with their spouses regardless of who acquired the land, with a single exception. The picture changes only slightly when it comes to deciding who can rent or sharecrop out a plot of land: in this case, two men reported that their spouses alone can decide this, even when it was the male head who acquired the land in the first place. There is no way of knowing why this would be the case. Thus, decision making regarding the use and deposition of agricultural land appears to be shared between husband and wife, and few informants perceive that outsiders (e.g. sharecroppers, landlords) have much influence. 84 5.A.3 Sharecropping Slightly less than half of the MH sharecropped land in, and Table 26 shows that all did so in order to gain access to more land to cultivate. All had at least 3.5 timads of land in total (presumably including the sharecropped land81) and three had more than five. A total of eight FH, or slightly more than half, sharecropped land out for various reasons, most of which revolved around oxen: either lack of access to oxen, prohibitions against women ploughing, or lack of access to dung (fertiliser) which could be provided by oxen, together with the economic incapacity to purchase fertiliser. Several women gave as an additional reason the fact that cultivation interferes with other non-agricultural activities. These results replicate fairly well what Teshome (2003) found for other villages in Hintalo Wajirat (see Section I.D) Nevertheless, most of these women still participate in cultivation and post-harvest activities when their land is sharecropped out, as can be seen in Table 27A. The division of agricultural labour among FH provides a strong indicator of the existence of cultural and economic restrictions Table 26 Reasons for Sharecropping Land In and Out by that women confront, and hence of Sex of Household Head* their dependence on male labour. Of those who had land that was cultiMH FH vated in the year prior to the survey, REASONS GIVEN No. % No. % only one ploughed it, and this same For sharecropping land in woman carried out all other agriculTo have more land to culti17 100.0 0 0.0 tural tasks alone. Two-thirds devate pended upon male sharecroppers, For sharecropping land out whereas the remaining third used Women can’t plough 6 22.2 either hired or ‘other’ male nonDoesn’t have oxen 0 0.0 10 37.0 household members who are landed Can’t afford fertilizer 0 0.0 5 18.6 to plough, where the last category Interferes with off-farm 0 0.0 5 18.6 most likely refers to ex-husbands activities 0 0.0 1 3.7 and fathers, and represents only 16% Plot is too distant of the cases where women relied on Totals 17 100.0 27 100.1 male labour. Planting was done by *Percentage of reasons given; multiple responses accepted. the same men who did the ploughing. Seven of the women heads, however, indicated that they participated together with men in cultivation (weeding, watering, pruning, etc.), and three indicated that they performed these tasks alone; only three reported that the sharecropper alone was responsible, indicating that, even when land is sharecropped out, women still perform certain agricultural tasks on these fields.82 In those households that sharecropped land out, sharecroppers exclusively performed all other tasks, including threshing (which requires the use of oxen), with one exception which was bird scaring, which a female head did herself. Aside from threshing, women heads were much more involved in post-harvest tasks: they predominate in all except for transporting crops from the field. Women reported that they alone sell crops, and most reported that they alone also select and store seed, although these women did not sell any crops and, in the literature, it is reported that both sharecropper and landowner have to provide the seed. Nevertheless, these patterns indicate that, while FH were much more dependent on non-household male labour for agricultural production, they also performed agricultural work even when they sharecropped land out. Regarding the quality of the land sharecropped in and out, MH and FH reported similar figures: a little under a quarter of the plots were considered as good quality (woferam), about two-thirds were considered to be average (mekakelgn), whereas only about a tenth were considered to be poor (keche) (for terminology see Teshome 2003). No data was collected on the distance of these plots from 81As is discussed below (Section 5.B), some men reported that they had access to less land than the amount that they reported was under cultivation. It appears that land that was sharecropped in was under-reported in the section of the survey that recorded land access, where informants were asked to provide data about “the plots that your household has used for agriculture, livestock, gardening or to grow trees during the last 12 months”. 82They may be more likely to do so when the sharecropper is an ex-spouse or father of their children. 85 the homestead nor on whether the owners use of fertiliser, so it is not possible to determine, as Beyene (2004) did, whether these plots were managed as well as other plots in the short or long-term (see Section 2.B). Also, the survey questions regarding the inputs that plot owners provided to sharecroppers and vice versa did not result in valid data.83 When looking at the relationship between the head and the person with whom they had entered into sharecropping arrangements, three-quarters of MH sharecropped in land belonging to nonhousehold members who were not kin (76%), and a quarter sharecropped in land belonging to nonhousehold members who were kin, probably their ex-spouses. Two FH reported that they sharecropped out land to kin and non kin-related household members, but this appears to be a misunderstanding of the concept of household used in the survey: both women own their plots together with ‘original owners’ and had no adult males in their households, which makes it highly likely that they sharecropped out to their former spouses. The majority (57%) entered into relations with nonhousehold members who were also not kin, and 29% sharecropped out to non-household members who were kin. Thus, FH were more likely to enter into relations with people who were either (ex-) household members or kin (43%, compared with 23% of MH). Several female heads provided additional information during the survey that illuminates the diversity of their land use strategies even in such a small sample, and demonstrates that these depend on factors such as kinship relations, exchange of resources, and own inventiveness and initiative. One woman who sharecropped land out reported that the sharecropper was another woman, which indicates that FH that do not have severe resource restrictions also seek to increase their agricultural production capacity. Another woman reported that she sharecropped out one of her plots because it is located three hours away in the lowlands. Regarding those women who did not sharecrop land out, one reported that she didn’t have to because her father provided her with the necessary services: she in turn provided him with her crop residues to feed his oxen. Yet another woman reported that she purchased her house and land (.25 timads) and that, instead of sharecropping, she exchanged crop residues with the person who ploughed her land, and hired labourers to do part of the agricultural work. One woman opted to plant eucalyptus on one of her plots and to sharecrop out the others. 5.A.4 Land Use and Crop Residue Use Because FH and MH have different land endowments, and because of the constraints that FH face in accessing male labour and other farming-related assets, it can be expected that they will use land differently. Table 28 shows that nearly Table 28 Land Use by Sex of Household Head 80% of all plots were cultivated. The second largest use category is MH FH TOTAL permanent crops which, as is disLAND USE % No. % No. % No. cussed in Section 5.D, mainly conCultivated 50 78.1 24 75.0 74 77.1 sist of Eucalyptus spp. A larger proGrazing 1 1.6 0 0.0 1 1.0 portion of FH had plots in permaFallow 1 1.6 0 0.0 1 1.0 nent crops, but two indicated that Permanent 8 12.5 6 18.8 14 14.6 this is land that they rent from a Homegarden 4 6.2 2 6.2 6 6.3 landlord, and they were not permitOther 64 100.0 32 100.0 96 100.0 ted to use the trees growing there. If these two are not counted, then the proportion of MH and FH with permanent crops is the same. Only one MH had any grazing land, and only one had any land in fallow, which confirms the processes that were described for Tigray as a whole (Section 2.A) where, due to land scarcity, grazing land has been converted to agricultural production and crop land is no longer left in fallow. As was also discussed in Chapter 2, the use of crop residues for fodder and fuel has a deleterious effect on soils, and such uses increase as grazing resources decline. MH were far more likely than FH to use crop residue: they were harvested from 48% of MH plots, versus 28.5% of FH plots, which probably reflects the fact that several of the FH sharecropped out plots and the sharecroppers retained the residue. MH harvested a total of 181 shekims of crop residue (87% of the all residue), 83Apparently some respondents understood the question to be “which inputs does the sharecropper provide” rather than “which inputs does the plot owner provide to the sharecropper.” In spite of various attempts to correct this, confusion persisted, leading to inconsistent data. 86 compared with FH which harvested only 27, or 13%. Slightly less than half of the crop residue harvested came from wheat, and another third came from barley, followed by about 15% from teff and less than 2% from peas. Table 29A shows that, while all of the crop residues that MH harvested were used for animal feed, FH used residues either for feed or for ‘other uses’ (that is, other than fuel or thatch, which were the other alternatives) that were not specified. It is likely that these residues were exchanged for ploughing services, as reported above, where they were subsequently used as feed. Table 30A shows the relation between total household land area and amount of crop residue harvested in the past 12 months. Of the ten households that owned land that was not planted entirely to trees and that reported harvesting no crop residues, all were FH. Clearly, the largest landholders harvested the most crop residues. Graph 2 suggests that there is also a positive relationship between the amount of crop residue harvested and the Total Livestock Units (TLU)84 owned, although it can also be seen that even households with no livestock harvested crop residues. As was discussed in chapters 2 and 4, crop residues represent the single most important source of feed in Tigray, and it also serves as a medium of exchange for households without oxen to obtain ploughing services. Graph 2 Crop Residue Harvest by Total Livestock Units (n=20) 35 Crop residue harvest (Shikems) 30 25 20 TLUs Crop residues 15 10 5 0 1 3 5 7 9 11 13 15 17 Tropical Livestock Units 5.A.5 Land Improvements: Soil and Water Conservation In the surveyed households, soil conservation measures had been implemented on 70 of the 95 plots. Table 31A shows that, of the plots held by MH, such measures were implemented on 54 (84% of the total), where the vast majority consisted of bunds (whether they are stone or soil bunds is unknown, and it can be assumed that these are level bunds, given that most plots were cultivated on flat to moderately sloping land). In contrast, FH implemented such measures on 56% of their plots, where the vast majority again consisted of bunds. This lower percentage is probably due to the fact that much FH land is occupied by homesteads. Water conservation measures were implemented less frequently, on 68% of all plots, and consisted mainly of ponds. MH again surpassed FH in implementing these measures: ponds had been constructed on three-quarters of MH plots, whereas FH implemented them again on 56%. Residents stated that, while they knew that ponds could be beneficial, they had not received the plastic lining that is supposed protect against absorption, and therefore water is quickly absorbed into the soil. A 84Total Livestock Units were estimated as follows: 1 Cow, ox or bull = 1 TLU; calves & heifers = .5 TLU; sheep & goats = .15 TLU; donkey = .065 TLU, and poultry = 0.005 TLU. 87 great deal is also lost to evaporation. A few women heads, including one who was a household survey respondent, reported that they had begun to dig ponds, but were unable to complete the work due to lack of time and the strenuousness of the labour. Throughout most of the kushet, soil is dry, hard, and rocky, making any digging very difficult. These women complained that they were unable to complete the work on their own and were offered no assistance. Only one person reported implementing any tree conservation measures, and this consisted of windbreaks. Table 31A also shows that, in MH, both soil and water conservation measures were implemented mainly by husbands and wives together (in 74% of the cases for soils and in 85% of the cases for water). No FH reported implementing such measures together with anyone else. There appears therefore to be a predominance of women involved in on-farm conservation efforts. In general it appears that the level of implementation of soil and water conservation measures on individual farms in the study households is quite high in comparison with that which is reported for other regions of Tigray (Esser et al 2003), which may be a result in part of the FAO project. 5.B Crop Production and its Distribution Most of the surveyed households were landowners, and most landowners produced crops, although a few (see below) only had tree plantations. All households were only able to produce crops during the rainy season (meher) in the year prior to the survey. The 28 households with land planted a total of 69 timads to crops, for an average of slightly less than 2.5 timads per household. A maximum of four crops were planted by any household. Table 32A presents the crops planted, the number of plots planted to each crop, and the area planted per crop. More than three-quarters of FH planted 1.5 timads or less, in comparison with only a fifth of the MH. FH and MH also clearly had different cropping patterns: FH planted 81% of their crop area to wheat, compared with MH which planted 38%, followed in area by barley (15.7% and 28.6%, respectively). Teff, the most preferred staple crop in the highlands, requires high labour inputs and has relatively low yields. It is a luxury crop for FH: only one planted her very small land area to it, as her single crop, whereas 22% of the total area pertaining to MH was planted to white or red teff. Table 33A shows that, whereas around 58% of all MH planted three to four crops (a combination of teff, barley, wheat, and sometimes peas), two-thirds of FH planted only wheat, and only 14% planted both wheat and barley. Two factors combine to explain the differences in cropping patterns between FH and MH. First, the proportion of total land area that FH planted to crops differs. Table 34A shows that a greater percentage of FH cropped a lower percentage of their total land area: 36% produced crops on 75% or less of their total holdings, compared with only 7% of MH. This is due in part to the fact that FH had less land, and therefore had to use a greater proportion of it for their homesteads. Second, as discussed above (see FN 41), a few MH apparently under-reported the total land that they cropped, in comparison to the land which they reported actually harvesting, so that the gap between MH and FH may be less than what appears in the table. The above-described differences in areas cropped and crop combinations are also reflected in differences in consumption of the harvested crops. First, it is important to note that, in keeping with the pattern reported for most of Tigray, only two households sold any part of their crop: one was the woman who had the most land of all FH (3.5 timads), who earned 100 birr (US$ 11), and one was a MH with less land, who earned 60 birr (US$ 6.60). All remaining production was either destined for households’ own consumption or for sharecroppers. A total of 64% of the MH consumed more than 1.5 qq, compared with only one FH who also had the most land; more than 90% of the FH with land consumed 1.5 qq or less, compared with 36% of the MH. More than half of the FH in fact consumed 50 kilos or less, compared with none of the MH. Table 35A compares consumption of the different crops by MH and FH, where the differences in average consumption per household that produced the crops are striking: MH’s consumption of wheat and teff was more than two and a half times that of average FH’s consumption, and barley consumption was nearly four times as high. The dependency ratio provides a partial means to assess the degree to which household crop consumption is distributed according to adult labour force availability. Table 36A presents the data, where it notable that those with the highest crop consumption levels (above three qq) had dependency ratios of 1.0 or lower, that is, where there is a positive balance between available workforce and dependents. On the other hand, of those households with a dependency ratio above 2, that is, where 88 the number of minor children is substantially greater than the number of working age adults, nearly 60% consumed 100 qq or less of their own crops. The amount of male labour available to the household (in this case, males aged between 10 and 64 years) and the number of oxen owned by the household may together explain the distribution of crop consumption between households. Tables 37A and 38A present the data, which is visible in Graph 3. In this graph it can be seen that the relationship that would be predicted by Beyene’s thesis (Section 2.B), that is, that it is greater access to male labour rather than oxen ownership that mainly determines a household’s wealth level, does not necessarily hold with respect to household crop consumption, but neither does there appear to be a strict relation between number of oxen and own crop consumption: this relation appears to hold only at the highest and lowest levels of consumption, and in these ‘extremes’, it does not appear that either oxen ownership or male labour access is more important. There appears to be a positive relationship between the amount of male labour and oxen ownership and own crop consumption, but the incidence of sharecropping among FH means that neither are absolutely necessary; the ability to hire oxen and male labour are mitigating factors. For example, the total lack of oxen did not prohibit landowners from consuming own crops, however all of those households that consumed the least (50 kg) owned no oxen, as did 67% of those who consumed only 75-100 kgs. On the other hand, owning oxen does not necessarily mean that crop consumption is very high, seen in the fact that 33% of those households that had two oxen consumed between 75-300 kg of their own crops. Graph 3 Crop Consumption by Male Labour Availability and Oxen Ownership Crops (Quintals), Males (10-64), No. Oxen 12 10 8 Males 10-64 6 Crop consumption No. Oxen 4 2 0 Table 39A and Graph 4, on the other hand, present the relationship between ox ownership and cropped area. This may more clearly reflect the relationship between oxen and overall household status insofar as the consumption data exclude the amounts that oxen owners produce but must share with others when they sharecrop land in, that is, around 50%. In this table, it can be seen that there is also the expected positive relationship between oxen ownership and cropped area. Of those who own two oxen, all cultivated more than two timads, and 71% cultivated more than four. Of those who own no oxen, none cultivated more than two timads, but most (69%) cultivated between one and two. But a larger percentage (60%) of those who cultivated 2.1 to 4 timads owned only one ox, while some who owned one ox cultivated between one and two timads, showing that oxen ownership may or may not represent a considerable advantage in terms of cultivated area. To conclude the analysis of differentiation in crop production and consumption, it is important to return to the situation prevailing among FH, where own crop consumption levels were so 89 Graph 4 Area Cultivated by Oxen Ownership (n = 25) 9 8 No. Households 7 6 No oxen 5 One ox 4 Two oxen 3 2 1 0 .25 - < 1 timad 1 - 2 timads 2.1 - 4 timads > 4 timads much lower than in MH. As indicated above, the explanation lies partially in the lower amount of land that FH held and in the lower proportion of total landholdings that were used for crop production, but Table 40A shows a further disadvantage that FH confronted due to their lower overall access to male labour and oxen that are discussed elsewhere in this report. This table shows that 53% of FH gave up 50% of their crop to sharecroppers. Of the seven FH that did not sharecrop out, three consumed only half of a quintal versus five that did sharecrop out. The remaining FH that didn’t sharecrop out consumed on average more than those that did sharecrop out. The sharecropping relationship is obviously necessary for many FH, but it is certainly not in general advantageous.85 5.C Livestock Production and its Distribution As discussed in Chapter 2 and above, livestock ownership, and particularly oxen ownership, has been considered to be a primary demarcation of social and economic status in the Ethiopian highlands, although there are questions regarding whether it is the lack of oxen per se, or rather the lack of access to male labour, that represents the primary constraint for FH and other poor households. Apart from oxen ownership, livestock provide milk, meat, eggs, and other products for own consumption, as well as income and a financial buffer in times of stress. Livestock ownership is not without its risks, and livestock morbidity and mortality are reported to be high. The lack of feed is the greatest constraint. 5.C.1 Distribution of Livestock Between Households Table 41 shows that, of larger livestock, oxen and donkeys were by far the most prevalent (27% and 21%, respectively) although, considering all other types of cattle together (calves, bulls, heifers, cows), these represented 31%. Small ruminants represented only 21%, where goats were more prevalent than sheep. No one owned horses or camels. Of MH, 20% owned no oxen, a third owned one, and 47% owned two. Average oxen ownership among MH was 1.2, whereas across the 30 households it was only .66; if these two figures are extrapolated to the entire population of the kushet and it is assumed that FH represent 30% of the total, then on average there would be .86 oxen per household, which is lower than the average that Beyene (2003) found across Tigray, of just under one. He also found a different distribution of oxen 85The one female head that had an ox had no adult male labour, but she did have two sons aged between 10 and 15 years. She most likely therefore exchanged use of her ox to secure ploughing labour, and a second ox for ploughing. She was the FH who owned the most land, and she did not sharecrop land out. 90 NO. OF HOUSHOLDS WITH: Calves Bulls Oxen Heifers Cows Sheep Goats Donkeys Totals Chickens Table 41 Numbers of Livestock by Type and by Sex of Household Head (n=30) MH FH MH None One Head Two or None One Head Two or Total Head More More No. % No. % No. % No. % No. % No. % No. % 10 66.7 4 26.7 1 6.7 13 86.7 2 13.3 0 0.0 6 75.0 9 60.0 6 40.0 0 0.0 15 100.0 0 0.0 0 0.0 6 100.0 3 20.0 5 33.3 7 46.7 14 93.3 1 6.7 0 0.0 19 95.0 13 86.7 2 13.3 0 0.0 15 100.0 0 0.0 0 0.0 2 100.0 8 53.3 6 40.0 1 6.7 14 93.3 1 6.7 0 0.0 8 88.9 12 80.0 1 6.7 2 13.3 15 100.0 0 0.0 0 0.0 6 100.0 12 80.0 0 0.0 3 20.0 15 100.0 0 0.0 0 0.0 9 100.0 7 46.7 4 26.7 4 26.7 15 100.0 0 0.0 0 0.0 15 100.0 74 58.3 37 29.1 16 12.6 15 100.0 4 3.5 0 0.0 71 94.7 6 40.0 1 6.7 8 53.3 14 93.3 0 0.0 1 6.7 52 81.3 FH Total No. 2 0 1 0 1 0 0 0 4 12 % 25.0 0.0 5.0 0.0 11.1 0.0 0.0 0.0 41.1 18.8 91 ownership than that reported in Table 41: between 37 and 47% owned one ox, whereas 17-25% of households owned two or more oxen. This may mean that the present study’s sample was biased toward wealthier MH, which would then affect the difference in assets between MH and FH, but this is not possible to determine without comparable kushet level data. The differences in livestock ownership between MH and FH were very strong indeed: not considering chickens, MH owned a total of 71 head versus FH’s four. FH owned, in total, two calves, one ox, and one cow. MH even owned four times as many chickens: only one FH had chickens as opposed to nine MH. Table 42 compares the distribution of Tropical Livestock Units (TLU), a standard measure whereby different species of different average Table 42 Distribution of TLU by Sex of Household size can be compared and described in relation Head to a common unit. Here it can be seen that twothirds of all FH had less than .5 TLU, compared MH FH TOTAL to none of the MH. Nearly half of the MH had 1 TLU No. % No. % No. % to 2 TLU, whereas slightly more than a quarter .01 to .5 0 0.0 2 66.7 2 11.1 had 5 TLU or more. Feed requirements are re1 to 2 7 46.7 0 0.0 7 38.9 lated to TLU, so that only half of the MH would 2.1 to 5 4 26.7 1 33.3 5 27.8 have had quite significant needs for livestock >5 4 26.7 0 0.0 4 22.2 feed, whereas only one FH had such a need. Totals 15 100.0 3 100.0 18 100.0 Examining the distribution of TLU by total farmland (Table 43A), it is apparent that there is not a strict relation between farm size and TLU. For those with >5 timads, two had only 1-2 TLU, one had 2.1 to 5 TLU, and one had more than 5 TLU. For those with .25 - < 1 timads, two had 1-2 TLU. This is likely to reflect the fact that farm size doesn’t determine the number of livestock that one can maintain simply because animals are not grazed on farmland, although, with larger farm size, more crop residues can be produced and more income is available with which to purchase and maintain livestock. 5.C.2 Distribution of Livestock within Households As a means to determine who controls livestock within households, the head was asked who can sell or give away the animals, who can keep the offspring, who can keep proceeds from the sales of animals, and who sells the animal products. Table 44A shows only data for MH since, in FH, there were few animals, and women heads reported that they alone were responsible. What is interesting about this data is the degree to which male heads reported that decisions about larger livestock are joint, even in the case of oxen, in relation to who can sell or give away larger livestock and who can keep any offspring. Only in about a third of the households did men report that selling or giving away larger livestock is their own prerogative, and in the rest it is a joint decision: wives had the sole prerogative in relation to chickens, and then in five of the eight households that had them. Regarding who keeps offspring, it was reported that all are joint property, again with the exception of chickens where in this case only two wives and one head were seen to be the sole owners of any chicks produced. However, when sales of live animals are considered, the picture changes significantly: in the case of oxen, calves, heifers, cows, goats and donkeys, two-thirds to 100% of all male heads said that they would be the sole recipients of the proceeds. Only in the case of bulls and sheep would the proceeds be shared in half or more of the households; in the case of chickens, women keep the proceeds themselves. The data on ‘who sells products’ is not very revealing, since little information was provided by the respondents, and the data that was provided didn’t always correspond to the data presented in Table 45A regarding sales (e.g., some respondents reported sales but didn’t indicate who sells the products). Nevertheless, sales of cow and chicken products were more often indicated to be the purview of wives. As long as larger livestock belong to the household, they are seen as joint property, but when these animals are sold and income is involved, the benefit streams go to the male head who apparently considers himself to be largely responsible for these animals. Women’s predominance in chicken raising is likewise recognized: men don’t seem to be so eager to insist that they make decisions together with their wives when it comes to small fowl. 92 Table 45A shows that oxen and goats represented the majority of livestock purchases and sales over the previous 12 months, where oxen represented nearly 50% of sales and goats 60% of purchases. One MH alone sold five of the nine oxen (probably as a trader). Sales of cows and sheep were the next most important market transactions. Goats were obviously purchased for consumption, representing 83% of all larger livestock consumed. Given the sales and consumption of sheep, it would appear that households are indeed divesting themselves of these animals as the FAO reported for the project area more generally (see Section 4.F.1). FH were responsible for a tiny fraction (7%) of all larger livestock purchases: only one purchased an ox. However, given the very low number of livestock owned by FH, they were still responsible for 22% of the larger livestock sales (one calf and four cows). The greatest amount of commercial activity in live animals was with respect to chickens: although few were purchased, a considerable number were sold and a single FH was responsible for 23% of the sales. Chickens were also the most consumed. However, only one FH consumed chickens in the 12 months prior to the survey, compared with six of the MH. The inequality between MH and FH in consumption of live animals is also striking: MH were responsible Table 46 Sales, Income and Consumption of Animal Products by for all consumption of both Sex of Household Head (n = 30) goats and sheep. Of animal product sales MH FH CONSUMED and consumption, the data preLIVESTOCK No. Total No. Total No. % HHs sented in Table 46 show that PRODUCT HHs Income HHs Income eggs represented 11% of the Milk/cream 1 20.0 5 16.7 total income derived from liveCheese/yoghurt 1 526.0 1 3.3 stock products, where six MH Hide/skins 5 103.5 and one FH sold eggs. One FH Meat/offal 0 1 66.0 10 33.3 sold chicken meat. Milk, cream, Dung cakes 0 0 cheese, and yoghurt were obEggs 5 98.0 1 20.0 8 33.3 tained from cows and sheep; Wool 0 0 only MH sold these products, Hiring out oxen 2 100.0 and one alone earned 500 birr Totals 14 847.50 2 86.0 24 (59% of all income earned by all households) from cheese and yoghurt sales. Hides and skins, and hiring out oxen, represented 27% of total sales, and these were exclusively sold by males in MH. Regarding animal products consumed, one FH consumed cow’s milk and cream and one consumed chicken meat, while one consumed eggs. This can be compared with MH, where five consumed cow’s milk and cream, one consumed sheep milk and cream, six consumed goat meat, two consumed chicken, and seven consumed eggs. These figures ratify FH’s disadvantages with respect to livestock ownership and the consumption and income benefits that are derived, but they also indicate that even relatively few of the MH earned much income from livestock products. Meat and eggs predominated over dairy products as consumption items, where meat came from small ruminants, rather than cattle. 5.D Trees on Private Land As was discussed in Chapter 2, several researchers have argued that tree planting on private land is and has been discouraged by land and tree tenure insecurity. If this is indeed the case, it does not appear to be discouraging tree planting on private land in Adiarbaetu, at least recently. The incentives to plant are clear. Elders reported that they used to procure tree products from the mountains but, since these have disappeared, “We now have to plant and take them from our backyards.” Others said that, from the time that enclosures and restrictions began to be implemented, people began planting trees (particularly eucalyptus) in their backyards, which they said they were encouraged to do. One of the women focus group participants provided concrete examples: 93 We planted this kokola tree [Euphorbia abyssinica?] used for fuelwood and construction] long ago, when I was a child. It’s also easy to transplant. We also grow it as a living fence around our land. I sell some of my kokola for transplants. Others do this as well, if they are growing it in abundance. People are also planting awelie [olive, Olea europea ssp. Cuspidata] but it is slow growing. Because our soil is no longer favourable, it takes a very long time to grow. Another woman said that she would like to have as much land “as I can imagine” to grow trees, pointing out that the previous woman cited has a lot of land, and can grow as many eucalyptus trees as she likes and cut them when she needs. “But, for me, I do not have this amount of land and thus I cannot get the benefits.” Households do indeed appear to be planting as many trees and other useful perennials as possible on whatever land they do own: “The beles and trees in the kushet that line the roads and other areas are privately owned. We all have them in our backyard. We use them as living fences.” Trees are also widely recognized as important for land conservation: as one man said, “Every tree can also protect.” The household survey captured the number and species of trees located on the land to which households had access, their uses, and rights of individuals within households to the trees. Table 47A shows that the vast majority (>80%) of all households had trees, although nearly a quarter had five or fewer. Slightly more than 20% had 26 or more trees, with three households having what may be termed ‘plantations’, that is, in Table 48 Uses Made of Trees on excess of 50 trees. In fact, two MH had planted all of their land to Household Land (n = 23) trees. Eucalyptus spp. clearly predominate in such plantations, USE No. % particularly a red gum variety: only households with smaller Fuelwood 11 51.2 stands planted significant portions of a blue gum variety. Other Construction 16 69.6 species include chihidi (cypress, Juniperus procera?, Cupressus Medicine 9 39.1 spp.?) (one household, which planted two, as well as four red Fodder 1 4.3 gums), and achachea (Acacia spp.), tebele (unknown), hahote (Rumex Other materials 4 17.4 nervosus?), and kokola (Euphorbia abyssinica?), pertaining to anBoundaries or other household, where all trees were self-sown rather than 1 4.3 markers planted. Table 48 shows that the most common use of these trees was for construction and fuel, followed by medicine. The category ‘other materials’ most likely refers to fencing and tool making (particularly ploughs). In only one case was it reported that eucalyptus trees provided animal feed. Table 49 gives data on the total number of trees by landholding strata, where it is evident that there is no clear positive relationship between the two variables. For example, one household with less than one timad planted all of its land to trees, as did one household with less than two timads, which Table 49 Total Number of Trees by Size had a total of 500 trees. Cumulatively, two-thirds of of Landholding all trees were planted on farms of less than two tiNO. mads. Indeed, more limited land access may proTOTAL TREES CUMULATIVE vide an incentive for tree planting, given that this LANDHOLDING No. % No. % may be the only way to earn sufficient income from None 5 .5 5 .5 such limited land resources. Both households are .1 to .99 timad 42 4.6 47 5.1 constituted by young couples that participate sub1 to 2 timads 556 60.8 603 65.9 stantially in food-for-work activities and that had 2.1 to 5 timads 86 9.4 689 75.3 two head of cattle (but no oxen); one of the male 5.1+ timads 225 24.6 914 99.9 heads trades in cloth, but otherwise there were no other sources of income, so it is not clear how these households subsist while waiting for their plantations to mature. Table 50 shows that MH were far more likely both to own trees and to have a higher average number of trees in comparison with FH. A third of the FH had no trees, and nearly 90% had fewer than six trees, compared to only 20% of the MH found in these two categories. Only two FH had more than six trees, and in no case did a FH have more than 25, compared to eight MH with more than 25 trees each. 94 Table 50 Number of Trees per Household by Sex of Household Head (n = 30) MH FH TOTAL TOTAL TREES No. % No. % No. % None 2 13.3 5 33.3 7 23.3 1-5 1 6.7 8 53.3 9 30.0 6-10 2 13.3 1 6.7 3 10.0 11-25 2 13.3 1 6.7 3 10.0 26-50 5 33.3 0 0.0 5 16.7 51-100 1 6.7 0 0.0 1 3.3 101-500 2 13.3 0 0.0 2 6.7 Totals 15 100.0 15 100.0 30 100.0 Moreover, some FH reported that they did not own the trees that grew in their homesteads since they rented the land: they could not use these trees other than for medicinal purposes. The one woman head who had 11 eucalyptus trees planted them about 15 years ago, and she uses them for fuel, construction, and medicine. No MH did not own the trees found in their homesteads. Table 51A provides data on tree tenure by sex of the household head. In the case where a woman was the head of a household that had trees (10 cases), where the trees did not belong to a landlord (eight cases), the woman head had all rights pertaining to the trees, which may be shared e.g. with another senior female (one case). In cases where a man was the head, in all but one the rights to destroy or sell trees reside with both he and his spouse, and in all cases but three (77%), both spouses also had the right to cut the trees or their branches. In eight cases (62%), only wives of heads had the right to take products without cutting: the question was probably interpreted to mean ‘who takes products without cutting the trees,’ rather than ‘who has exclusive rights to take,’ since it is obvious that tree ownership is perceived to be joint. This joint ownership may be related to the fact that a large number of respondents (seven, or 30%) reported that their trees were only recently planted, and most do not yet yield any products: as is the case with livestock, it may be that men consider their wives to be joint owners, but will keep the proceeds from sales themselves once these become available. Nevertheless, it appears that the concept of joint land ownership extends to trees growing on that land. This is reinforced by the fact that, in all MH where it was reported (nine), it was indicated that both husband and wife together decided to plant the trees. This is extraordinary with respect to the Sub-Saharan African context, where tree planting is often related to land ownership and is a male privilege. 5.E Homegardens Focus group participants said that, prior to the FAO project (around five years ago), those who had land near riverbanks began to grow vegetables for the market. However, over the past two years (due to the project), the number of homegardens has increased, and more people have become aware of their advantages as a source of income. They said that men’s involvement has increased considerably since irrigation pumps and important marketable crops have been introduced (e.g., cumin, potato), and now some of the new homegardens are ploughed by oxen instead of prepared by hand. The planting area is increasing, and men, women, and children all perform labour. Some families sell produce in Mekele. Elders were asked about the history of homegardening in the community. Homegardens, they assured, were not prevalent because the area was too dry, but when households had better access to water and, during the rainy season, women managed gardens in their backyards. Vegetables, particularly cabbage, lettuce, carrots, fenugreek, and sasug (Ocimum basilicum?) were grown solely for home consumption because the plots were small, although if there was any surplus, this might have been sold. In discussions with women, some said that they used to have small homegardens where they cultivated primarily spices and perhaps a few vegetables, and the cultivation and management of these gardens was mainly a female task. However, as homegardens have increased in size and number, gender 95 roles have begun to change, with men taking a large and, in many cases, primary role in their management. This shift is reported to be due to the increasing value of vegetables as a cash crop. Researchers said that they observed men working in homegardens far more often than women. However, the data reported in the Household Survey do not support these impressions.86 Only 13 of the 30 households surveyed (43%) had homegardens of any size. All but one of these pertained to MH (92%). Seven FH said that they do not have a homegarden because they don’t have sufficient access to land, while another reported that she had only a few crops due to a lack of space, and yet another reported that she will destroy her two trees to create the space to plant vegetables. Several female heads indicated their desire to have a homegarden: one said that, if she had a garden, she would grow fenugreek, hamli talien, spia (unknown) and eucalyptus, and would sell the surplus. Two other FH said that their soil is too poor to permit gardening (they tried to plant but the crops failed), and two also reported that a lack of water prohibits homegardening. One woman head poured out her story to the enumerators, and it is worth recounting here. Fifteen years ago, Teresa (not her real name) planted eleven eucalyptus trees to provide her household with fuel, construction materials, and medicine. Food-for-work told her to dig a well, but she couldn’t finish it because the work was too hard. They told her that she couldn’t get the seed she wanted for her homegarden because she didn’t finish the well. In any case, she doesn’t have a homegarden since her soil is too poor, and she doesn’t have enough water even for her household. She had planted gesho in her yard but the livestock ate it; she sold the livestock to send her children to school. She is growing ba-erir which she transplants from the wild, using it as medicine against the evil eye and as a fumigant. In comparison, only one male head reported that he doesn’t have sufficient land for a garden, although he also reported that he had planted 35 eucalyptus trees. Some MH did not have access to household land for gardening, but did have access to a cooperative garden where onion, cumin, garlic, potatoes, tomatoes, and carrots were produced for sale on a three ha plot, and another rented land for his wife’s homegarden. One MH said that he and his wife earned 1500 birr (US $165) from selling homegarden produce last year, although this year there was as yet no production, since the rainy season hadn’t started. Nevertheless, all of the crops that were reported as harvested from homegardens in the 12 months prior to the survey were destined for own consumption, and only food uses were reported. Data on the amounts grown were not collected, however one family reported that, during the last rainy season, five kg of green pepper, four kg of onions, and five kg of tomato were produced for own consumption. It is very probable that such crops can only be produced during the rainy season due to water shortage. Table 52 Number of Homegarden Crops per Table 52 shows that there were only a few speHomegarden (n = 13) cies grown in homegardens in the survey households: NUMBER OF CROPS No. % the maximum for any one garden was eight. Two 1 crop 2 15.4 households grew only one species whereas three re2-3 crops 3 23.1 ported growing six or more: the bulk (62%) produced 4-5 crops 5 38.5 two to five species, reflecting poor species diversity in 6 or more crops 3 23.1 comparison with other regions of Ethiopia (Asfaw Total 13 100.1 2001). However, no physical surveys of homegardens were carried out, which may have yielded higher species diversity since respondents may have underreported native, wild, or spontaneously growing species. The species that were grown are reported in Table 53A. Onions, cabbage, and garlic were the most common and lettuce, tomatoes, potatoes, carrots, cumin, green pepper, and beetroots were less common. All were exotics with the exception of gesho (buckthorn, Rhamnus prinoides L’Herit) (Getahun 1976). 86It is important to note that there was confusion about what constitutes a ‘homegarden’. Some (especially male) respondents referred to land outside of their homestead that was planted to vegetables. Enumerators were repeatedly told that ‘homegardens’ refers to gardens within the homestead where crops are produced, but it is clear that other ‘gardens’ were included in the survey data. In any case, there is no strict etic definition of homegardens and, if there was, it couldn’t readily be imposed on the respondents. 96 Table 54A presents the gender and age division of labour in homegardening as it was reported in the survey. Results are for MH only, since only one FH had a garden where she performed all asks together with her son. Essentially, women or girls participated in all homegarden tasks, although men were more involved in seed procurement and may do this alone, whereas wives and daughters were more likely to cultivate and harvest alone. The assertions that homegardens are predominantly a male domain and that this is due to increasing commodification does not hold among the surveyed households, since not only were homegarden tasks generally shared, no household sold any homegarden produce in the 12 months prior to the survey. 5.F Off-farm Activities and Livelihoods Although subsistence agriculture provides the basis for the livelihoods for the majority of Adiarbaetu’s residents, as can be seen from the foregoing discussion, most households generated little cash this way. Other income generating activities are necessary to supplement such production and, for the landless and many FH, these activities, rather than agriculture, provide their main source of sustenance. Food-forwork activities are also crucial and are discussed both here and in a later section. 5.F.1 Off-farm Income The female head or wife was asked to give the primary and secondary occupations of all adult household members. Table 55 shows that nearly two-thirds of male heads had own-account farming as their primary occupation while for another 20% it was a secondary occupation; for another 13%, paid farm labour was the primary occupation. Nearly half of male heads had either trade or food-for-work as secondary occupations, while another third had other secondary occupations, and no secondary occupation was reported for 20%. In sharp contrast, only 13% of all female heads reported farming as a primary occupation, and none reported it as a secondary occupation; nor did any report paid farm labour. Since FH cannot rely on crop production to sustain them through the year, non-farm income generation activities are vital to their financial and food security. Women heads were therefore more diversely occupied: aside from the four that reported either own-account farming or housewife as primary occupations, 40% had unique primary occupations, where 33% reported that they sold tela. One was disabled, whereas one was looking for work. Nevertheless, one woman informally reported that she works as a prostitute, and another reported that she did so until three years ago. Regarding secondary occupations, grain milling was reported by a third of all women, and 20% reported food-for-work. In contrast to women heads, wives reported exclusively that their primary occupation was either housewife or ‘domestic servant’, where the latter may be taken quite literally to mean paid servants within their own households, depending upon their marital contracts (See Section 2.C.1), or it may mean simply that this is how they see their function. Wives reported a more diverse set of secondary occupations: 43% declared themselves to be a farmer or family farm worker (obviously in a supporting role to their husbands), followed by food-for-work (29%), and selling tela (14%). Respondents reported that only heads and/or their spouses engaged in off-farm activities to generate income in cash or kind: children and other household members did not do so. All but one household reported such activities, and that which reported no activity was headed by a disabled woman. Table 56 shows that women predominated in such activities: of the 51 reported, women performed 73%, however the sample was biased since it included 50% FH. In general, the gender division of such labour was what was more or less expected: women principally earned income by using their domestic skills and capital, where homegarden production, food and drink preparation and sales, domestic work, and prostitution constituted 16 of the 37 activities (43%), while food-for-work constituted 46%, altogether totalling 89%. Trade represented 8% of women’s off-farm income generating activities, whereas it represented 36% of all such male activities. For men, food-for-work was as important a source of income in cash or kind as it was for women, particularly if it is considered that nursery work can also be considered as a type of food-for-work, since it is paid in kind with food. 97 Table 55 Primary and Secondary Occupations of Heads and Wives of Heads (n = 30) PRIMARY SECONDARY SEX, RELATION TO HEAD AND OCCUPATION OCCUPATION OCCUPATION No. % No. % Male head (n = 15) Farmer or family farm worker 9 60.0 3 20.0 Farm worker (for pay) 2 13.3 0 0.0 Party official/administrator/clerical 1 6.7 0 0.0 Trader 2 13.3 4 26.7 Food-for-work 1 6.7 3 20.0 Grain milling 0 0.0 1 6.7 None 0 0.0 3 20.0 Female head (n = 15) Farmer or family farm worker 2 13.3 0 0.0 Housewife 2 13.3 2 13.3 Domestic servant 1 6.7 0 0.0 Trader 1 6.7 0 0.0 Religious worker 1 6.7 1 6.7 Grain milling 1 6.7 5 33.3 Selling tela 5 33.3 1 6.7 Temporarily disabled/unable to work 1 6.7 0 0.0 Weaver 0 0.0 1 6.7 Food-for-work 0 0.0 3 20.0 Looking for work 1 6.7 1 6.7 Wife of head (n = 14) Housewife 8 57.1 0 0.0 Domestic servant 6 42.9 0 0.0 Farmer or family farm worker 0 0.0 6 42.9 Weaver 0 0.0 1 7.1 Trader 0 0.0 1 7.1 Selling tela 0 0.0 2 14.3 Food-for-work 0 0.0 4 28.6 Table 56 Activities to Generate Income in Cash or Kind by Sex FEMALE MALE ACTIVITY No. % No % Homegarden produce sale 1 2.7 0 0.0 Trade in grains & pulses 1 2.7 3 21.4 Transport by pack animal 0 0.0 1 7.1 Trade in livestock 1 2.7 2 14.3 Collection and trade of spices and aromatics 1 2.7 0 0.0 Drinks prep. and sales (mainly tela) 29.7 0 0.0 11 Food or tea preparation and sales 1 2.7 0 0.0 Basket making 1 2.7 0 0.0 Prostitution 2 5.4 0 0.0 Domestic work 1 2.7 0 0.0 Nursery work 0 0.0 2 14.3 Food-for-work 17 45.9 6 42.9 Totals 37 100.0 14 100.0 98 Table 57 Days Worked and Income Earned over the Past Four Months, by Sex of Worker and Sex of Household Head PER WORKER PER HOUSEHOLD DAYS WORKED AND Male Female MH FH INCOME EARNED, BY No. Cum. No. Cum. No. Cum. No. Cum. CATEGORY % % % % Total days worked 1-10 14.3 8.3 2 15.4 3 1 7.1 1 11-20 85.7 41.6 5 53.9 15 2 21.4 4 90.5 83.3 21-40 3 77.0 1 4 50.0 5 41-100 100.0 100.0 1 84.7 2 5 85.7 2 100-160 0.0 2 100.1 0 2 100.0 0 Totals 100.0 13 21 14 12 Total income earned 1 – 50 birr 0.0 3 20.0 0 1 6.7 3 30.0 51 – 300 birr 70.0 9 80.0 7 8 60.0 6 90.0 320 – 700 birr 100.0 3 100.0 3 5 93.3 1 100.0 701 – 1020 birr 0 0 1 100.0 0 Totals 15 10 15 10 5.F.2 Days Worked and Income Earned Data collected on days worked and income earned cannot be considered to be very reliable: respondents often have difficulty recalling precise amounts over long periods (in this case, four months). The data presented here should therefore only be considered as indicative of trends. While a larger number of women worked off-farm to earn cash income in comparison with men (21 versus 13, Table 57), there was a substantially lower number of men in the surveyed households who were of working age (21 men versus 37 women aged 15-64). The participation rate of these men in off-farm activities was higher than women’s (62% versus 57%, respectively). The fact that FH represented 50% of the total households sampled resulted in higher female participation compared with the general population: in fact, wives generated cash income in only 60% of the MH, whereas the female head generated cash in 80% of the cases. With a few exceptions, Table 58A shows that reported income earned was generally low. “There is no minimum wage legislation in Ethiopia but a wage of 120 Birr (US $15) per month is currently acceptable as minimum rate payable for unskilled workers.”87 Total reported income earned was divided by total reported days worked over the four-month period, resulting in an average wage per day: however, respondents were not asked how many hours were worked per day, so it is possible that the data reported below reflect part-time work. Of the residents sampled in Adiarbaetu who reported non-farm activities that generated cash income, only five (about 18%) earned 20 birr per day or more. Nearly half (46%) earned the equivalent of 15% or less of the minimum daily wage, of which 77% were women. Of all working women, 72% reported earning 10 birr per day or less (that is, 30% or less of the ‘minimum wage’ for unskilled workers), compared with only half of the male workers. Table 59A further shows that women constituted three-quarters of those who earned five birr or less per day of reported work, whereas they constitute 40% of those who earned 20-60 birr per day. There was no activity that consistently generated higher levels of income: although trade in livestock and foodfor-work have that potential, these activities generated only small amounts of income for some of those people who engaged in them. Interestingly, preparation and sales of ‘other drinks’ (mainly tela) can pro- 87Addis Ababa Chamber of Commerce com/aboutethioprofdetail.asp?CID=3). (accessed 21 Nov. 2005) (http://www.addischamber. 99 vide a substantial amount of income: while two women report earning less than five birr/day and most earned 6-10 birr/day, some reported earning above 20 birr/day, although this may actually be related to the commercial sex trade.88 In interviews, women heads stressed that the type of income generation activity is very important. Since many have small children, they need work that allows them to stay at home as much as possible, which they are able to do when they produce tela, the primary income generation activity of female heads. However, this creates great competition between women, and tela production far exceeds demand. Women have attempted to deal with this by rotating the days when they sell so that everyone can earn some income. Despite this, and because of excess production, tela sales alone cannot provide any woman with decent income. Tela production is also labour intensive. While the residues are used to feed livestock and can be sold or traded, the amount earned is not proportionate to the time spent in production. The use of wild plants in income generation appeared in at least two activities (collection and trade of spices and aromatics, and basket making), representing 5.4% of all of women’s activities (two out of 30 households). However, such activities appear to be under-reported: respondents volunteered more information about their income generating activities in regard to the wild plants that they collect in another part of the survey that sought data about food sources (see Section 6.G.2). One female head, who is deaf, reported that she retails kofkaf (unknown) and ba-erir once per week, year-round, for five birr per week, and she sells medaf talien (a seed from Argemone mexicana? which is made into an oil used to smooth injera oven tops) for two birr per week during the dry season. The female head who reported basket making as an income generating activity said that she specializes in this, working 80 days in the past four months to generate 400 birr in total (for only five birr per day average, but she probably works part-time). One woman reported that she trades in spices and aromatics, which is probably based on wild plants, but this generates only a small amount of income (less than five birr/day). Another woman reported that she specializes in an aromatic plant, ba-erir, which she transplants from the wild. It is also important to note that none of those households surveyed reported sales of fuelwood as an income generating activity, even though this is widely reported in Ethiopia as a means by which the poor earn a living. It is not surprising that fuelwood collection does not provide a source of income in Adiarbaetu, considering the reported shortages of fuelwood and the amount of time required for its collection (see Section 6.G.1). For FH, tela production was by far the most prominent income generating activity. Tela is a fermented drink made from grains such as wheat, barley, sorghum and gesho. Its prominence is not due to its income generating potential, since most women reported earning little more than a few birr per week. However, the rest products from tela are sold as animal feed or traded for ploughing services. Further, it is one of the few income generating activities that can be done entirely from home, which is crucial for FH who have small children. As was indicated above, prostitution was also a source of income for FH and is often connected tela sales. While the research did interview a few women who openly discussed these activities, their prevalence is probably far greater than the research revealed. As one woman head, who is also a leader of WAT, said: We know that (and the government says it too) women should not have relations with men [prostitution] because of AIDS. But we don’t have any income and they don’t give us any alternatives, so what are we to do? If they don’t want us to do this, they also need to give us an alternative that will work for us to support ourselves. 5.G Community Work (including Food-for-work) The Ethiopian government sponsors food-for-work in vulnerable drought and famine-prone woredas, and several national and international agencies also sponsor community work programmes that are remunerated mainly in food. The wage rate within government food aid programmes has been fixed at 3 kg a day, but at times it is paid at only 2.5 kg/day, which represent modest amounts. “A norm of 3 kg of 88The higher incomes reported from tela sales may actually be from prostitution. Tela sales and prostitution are linked: when selling tela from their homes, women also provide sexual services to their clients. 100 food and 120 g of oil per day is considered adequate to provide 1,800 calories to a family of six” (Quisumbing and Yohannes 2004: 9) but, in some locations (such as Adiarbaetu), oil is not provided. While work can be remunerated in cash, most recipients prefer payment in kind. In Adiarbaetu, only wheat was reported as in kind remuneration, although some respondents reported payment in cash (see above), which was not reported in the survey section dealing with community work. Twenty out of the 30 households surveyed participated in community work programmes for which they were paid over the twelve months prior to the survey. The total number of days invested was 2,087, for an average per household of 104 days. This is quite substantial when it is considered that, over a twelve month period, there are approximately 300 working days, so that food-for-work provided the equivalent of .33 fte employment per household. However, this amount varied substantially between households, from a minimum of six days (two households), or .02 fte, to a maximum of 270 days (one household), or .9 fte. Table 60A presents the data on participation in community work activities disaggregated by sex of the household head and sex of the person who performed the work. More than 80 percent of the activities (19 households) were performed for the official government food-for-work programme. The remaining were performed under the sponsorship of the woreda government (two households), another government agency, and Redd Barna (Save the Children Norway) (one household each). Men worked almost exclusively for the official food-for-work programme, whereas four women heads worked for other programmes. Nearly half consisted of digging wells, river diversion, and small dam construction, and more than a third were related to road building and maintenance, or soil conservation. Two households participated in irrigation activities and three in tree planting and nurseries. Women participated less in tree planting and nursery work, and more in road building and maintenance or irrigation activities. Women heads did not participate in soil conservation activities, whereas participation was equal between MH and FH in digging wells, river diversion and small dam construction. Differences between MH and FH in reasons for participating in community work were apparently strong. FH reported much more frequently that their primary reason was household food shortage (over half of the cases), whereas in MH, the primary reason was that they required any type of employment (50%, versus FH’s 27%), or temporary employment (33% versus FH’s 9%). What MH did with what they earned, however, differed little from what FH did: 75% of MH consumed what they earned, compared with 82% of FH, indicating that the need for cash was not greater than the need for food. The differences in reasons for participating thus more likely reflected perceptions about male versus female responsibilities (e.g. males are to be gainfully employed, women’s are to ensure food supply). Comparing the sexes, women in general (39%) were more likely to report food shortages as the reason for participation in comparison with men (10%). Only one man and one woman reported the need to make investments as the primary reason, and only five households reported selling part or all of their earnings. Table 61 shows that the total number of days worked in the previous 12 months by people in MH was 1,523, compared to a total Table 61 Days Worked in Food-for-work, of 567 for FH (a ratio of By Household Headship and Sex nearly 3:1), and an average of 101.5 days for MH versus 40.5 MH FH MALE FEMALE for FH. When disaggregating DAYS WORKERS WORKERS by sex, it is apparent that a WORKED No. % No. % No. Cum. No. Cum. larger proportion of women % % worked 30 days or less (about 1-10 1 8.3 2 18.2 1 10.0 4 17.4 a third) compared to men, 11-20 1 8.3 2 18.2 0 10.0 3 30.4 only 10% of whom worked 30 21-30 1 8.3 0 0.0 0 10.0 1 34.7 days or less. Comparing men 31-50 2 16.7 2 18.2 3 40.0 4 52.1 and women within MH, 51-100 3 25.0 4 36.4 4 80.0 7 82.5 women performed 50% of 101-150 4 33.3 1 9.1 2 100.0 4 100.0 total days in food-for-work Totals 12 100.0 11 100.1 10 23 99.9 and averaged 69.8 days, in comparison with men’s aver- 101 age of 75.5 days. Women in FH worked an average of 51.5 days. It can therefore be concluded that the greater labour power available in MH not only permitted substantially greater time investment in foodfor-work overall, but as well greater investment on the part of women. Men earned an average of 268 kg, versus women’s average of 213 kg. It was not possible to compare MH and FH since, in nine of the FH, earnings data were missing. In focus group discussions, women reported that food-for-work activities are essential, and are way they can avoid severe food shortage. But they argued that these activities are not geared towards women, and are especially difficult for female heads. Most entail very heavy work, such as carrying stones to build walls or dams, or digging wells in hard, rocky soil. These are difficult for even those who are young and healthy, but are impossible for older, ill, or disabled women. They are also dangerous for pregnant women. Food-for-work is said not to be an option for women with young children, unless they have childcare. Those living in the more rural parts of the kushet may have to travel long distances to participate. For FH, this means that they have even less time to care for children and manage other essential household tasks. It is often difficult for them to arrive on time, but after repeated tardiness, they are told that they may no longer participate. 5.H Transfer Payments, Credit and Savings Only two respondents reported receiving any transfer payments in the form of gifts, remittances, etc., and both were FH. One received a total of 40 birr as a remittance, and the other received 50 birr per month as a pension granted by the government to war widows. FH did not report receiving anything that could be considered as alimony or child support. Participation in credit and savings schemes is presented in Table 62 by sex of the household head. Data are not presented by sex of the individual household member who participate, since only one woman from a MH received any credit or had any savings: it was nearly always the head of household who either saved or received credit. Dedebit Credit and Savings Institution (DECSI) was created by REST, and specifically targets Tigray and women, although 60% of its credit recipients are reported to be male (Teshome 2003). The DECSI’s mission is to improve the well-being of people who do not receive services from formal sector banks. It provides (a) general loans for income generating activities in agriculture, trade, handicrafts, and services (referred to in the table as Dedebit/REST); (b) Agricultural Input Loans provided to agricultural extension programme beneficiaries to purchase inputs, and (c) Agricultural Package Loans (referred to in the table as Package) to achieve food security and alleviate poverty at household level through activities such as livestock production and provision of agricultural equipment. Of the 29 households that provided data, nearly two-thirds had a loan at the time of the survey: 50% of FH and 73% of MH. Women received the vast majority of their loans from Dedebit’s general loan programme, whereas two-thirds of men received Agricultural Package Loans: only one FH was the recipient of a Package loan. MH and FH also differed considerably with respect to the purpose of the loan: livestock purchases constituted a third of all MH loans, whereas trade and the purchase of a generator (presumably to operate water pumps) represented 25% each; loans for consumption constituted only 11%. FH’s loans, on the other hand, were primarily for trade (mainly tela), only one had a loan to purchase an ox, and the rest were to meet consumption needs (e.g., food). In focus group discussions and interviews with women heads, they reported that FH can obtain credit for livestock fattening. However, since fodder is so difficult to come by, in a drought year they would be unable to feed livestock, and would run the risk of losing them while still owing the debt. They said that these programmes have high interest rates and short repayment periods (as little as two years), so there is little room for manoeuvre in a bad year. Several women said that they are simply afraid to borrow large sums of money: they do not understand how credit programmes work and how they would repay loans. In one instance only had a woman from a MH taken out a loan, and this was to purchase food and pay school fees. The average amount of the loan also differed between FH and MH, with FH receiving on average 53% of what men received, which is not surprising given the purpose of the loans. Regarding repayment, FH and MH were on a par, where an average of 21-23% of the value of the loans had been repaid. Clearly, although FH were able to obtain credit to meet emergency needs and for petty trade, they 102 Table 62 Credit and Savings by Sex of Household Head (n = 29*) FH MH TOTAL INDICATOR No. % No. % No. % Credit Source Package 1 14.3 7 63.6 8 44.4 Dedebit/REST 6 85.7 4 36.3 10 55.5 Total 7 100.0 11 100.0 18 100.0 Purpose Livestock 1 11.1 4 33.3 5 23.8 Trade 6 66.7 3 25.0 9 42.9 Generator 0 0.0 3 25.0 3 14.3 Consumption 2 22.2 2 16.7 4 19.0 Amount Total 9600 28100 37700 Avg. 1371 2555 2094 Paid back Total 2225 6050 8275 Avg. 318 550 460 Avg./debtor 556 1513 1034 % paid 23.18 21.53 21.95 Savings Has savings 64.3 5 33.3 14 48.3 9 Amount** Total 2100 2300 4400 Avg. 300 460 314 Contributions 35.7 2 13.3 7 24.1 5 Equb 4 80.0 2 100.0 6 90.0 REST 1 20.0 0 0.0 1 10.0 Period Weekly 4 80.0 1 5 Monthly 0 0.0 1 1 Yearly*** 1 20.0 1 2 Amount/week 45 60 105 Avg./week 11.25 30 21 *Data on one FH was missing. **Total amount missing in two cases ***The amount saved by the female head was missing and one man also saves 50 per year. ****In one case the person with the savings in a MH was the wife. either did not seek or were unable to receive loans that would enhance their agri-cultural production capacity. In MH, it was men who received credit in the household’s name, unless the loan was for consumption. Savings presents quite a different picture. Of the 14 households that had any savings, more than two-thirds were FH, which might be surprising considering that they had far fewer assets and less income. However, the average amount of savings was lower in FH than in MH, which is certainly unsurprising, although it was still on average nearly two-thirds of the MH average. FH also made up more than 70% of those who contributed regularly to a savings programme, either an equb or REST (Dedebit). 103 An equb is the equivalent of a rotating credit and savings scheme that operates at local level. They are voluntary and depend on contributions that are usually made weekly. Periodically, drawings determine who will receive the funds that have been accumulated, and drawings continue until all members receive the funds. FH are apparently more motivated to build up a savings, which may be tied to their need to accumulate funds to pay school fees or purchase raw material for tela production; also, since they barely own livestock, they aren’t able to use them as savings, and they have less income and probably fewer social connections to support them when major financial outlays must be made or in case of emergencies. 5.I Membership in Community Organisations It has been argued (see Section 2.C.6) that destitute FH tend to be socially excluded, and participation in community-level organisations presents one indicator. Data on membership was collected in the Household Survey. Table 63 presents this data by household headship and by sex of household members who belong to such organisations. All of the persons who were reported to be members were either heads or their wives and, on average, each person who was a member of any organisation belonged to 2.2 organisations. There are several types of community oganisations in the region. One is the equb, discussed above, and a similar organisation is the IDDIR, which is a funeral association that provides a type of insurance in case of the death of a family member, providing food, cash, or labour to meet funeral expenses, which requires that a membership fee be paid. Membership in a church organisation typically means being a member of a mehaber, which are socio-religious associations that eat together monthly on the day of a Saint that the particular mehaber has chosen to honour, and that serve as social support mechanisms (and often provide credit) in times of hardship. Festive work parties are either debo or wonfel, where debo implies providing food and drink for a large number of people who provide labour for agricultural tasks on a timely basis, and wonfel involve reciprocal labour among smaller groups. The baito was discussed in Chapter 2, and membership in it permits representation in baito decision making. Grazing and irrigation committees provide oversight and decision making for grazing land and irrigation management. There is also a woodlot committee, but none of the surveyed households participate in it. Table 63 Membership in Organisations by Sex of Household Head and Sex of Members (n = 29*) GRAZ. IRRIG. ORGANISATION/ PA COM COM INDICATOR EQUB IDDIR CHURCH FESTIVE Total members 6 2 34 20 20 6 4 % 6.5 2.2 37.0 21.7 21.7 6.5 4.3 FH members 4 1 11 4 8 0 0 Within % 14.3 3.6 39.3 14.3 28.6 0.0 0.0 MH members 2 1 23 16 12 6 4 Within % 3.1 1.6 35.9 25.0 18.8 9.4 6.3 FH members as % 66.7 50.0 32.4 20.0 40.0 0.0 0.0 of Total MH members as % 33.3 50.0 67.6 80.0 60.0 100.0 100.0 of Total Women members 4 1 23 10 13 2 2 Within % 7.3 1.8 41.8 18.2 23.6 3.6 3.6 Men members 2 1 11 10 7 4 2 Within % 5.4 2.7 29.7 27.0 18.9 10.8 5.4 Women as % of 66.7 50.0 67.6 50.0 65.0 33.3 50.0 total Men as % of total 33.3 50.0 32.4 50.0 35.0 66.7 50.0 *Data on one FH were missing. 104 TOTAL 92 100.0 28 100.0 64 100.0 30.4 69.6 55 100.0 37 100.0 59.8 40.2 Table 63 shows, first of all, that only two people were members of an IDDIR, whereas the highest membership (37%) was in a church organisation, probably a mehaber. There were an equal number of female and male heads who were members of a mehaber, but many wives were also members. Festive work organisations represented 21% of total memberships, where male heads were far more represented than female heads (2.5:1), although the number of women and men participating in total was equal, since several wives were also members. This implies that FH who have greatest problems with labour shortage obviously have difficulty overcoming such problems through labour exchange. Equally important in terms of membership numbers is the PA or baito council. Female heads outnumbered male heads as members of the PA although, since some wives were also members, in total MH out-represented FH by 1.5:1. Within MH, five wives and seven heads were PA members, but no wife was the sole representative of her household. Overall in this biased sample, women outnumbered men as PA members. Membership in a grazing or irrigation committee was generally low. FH were not represented at all, whereas in two households both the head and his wife were represented (in one household, both the head and his wife were also on the grazing committee as well as on the irrigation committee). Again, women were not on such committees as sole representatives of their households. What these data suggest is that FH are members of community organisations and are more likely to be a member of the PA in comparison with wives, but they are barely represented in the committees that manage natural resources. Wives were more represented in church organisations, but were less represented in village governance than their husbands, and those who were members represented their households together with their husbands. Both of these situations are likely to lead to women’s voices being subsumed to men’s in governing councils, particularly given that the literature reports that, when women are members, they are generally passive. Nevertheless, it is not possible on the basis of the data presented to argue that FH are socially excluded, but it is clear that, while FH depend more on common land resources, they are not nearly as involved as they should be in their governance. Agricultural cooperatives could be an important source of support for FH who have land but lack the means to farm it themselves. Focus group discussants and officials reported that the number of agricultural cooperatives is increasing in the area, some of which have up to 20 ha of land. Women said that they do not have access to these cooperatives, which should be considered as a principle organizational constraint. A study carried out on the support programmes available to FH in Tigray reported that the Regional Cooperatives Bureau, which promotes agricultural cooperatives through the baito, does not prohibit women from participating but, “in practice, the majority of the cooperatives established to date are men only or predominantly” (Meehan 2003: 7). Informants in the kushet said that women sometimes rent their land to men who participate in cooperatives to grow high value crops, such as spices. According to a kushet official, 20 ha of cumin89 can bring in as much as 20,000 birr per year. Women’s exclusion from these cooperatives is said to be due to the fact that ploughing and field crop production are seen as male domains. Women are reportedly beginning to see the benefits of cooperatives and are showing interest in them, and in future may insist on taking part themselves.90 This should be an important option for FH, who could pool their land as well as their labour, making use of cooperative access to equipment and livestock, as well as male labour. 5.J The Gender of Livelihoods Ardiabaetu is in many ways a typical Tigrinya kushet, largely mirroring the dynamics, conditions, and cultures that have been described for the highlands in general. Land fragmentation and diminution have 89This is a vernacular name for more than one species. Two plants designed with this name that were observed by researchers appeared to be cumin (English - Cuminum cyminum) and dill (Anethum graveolens). It is likely there are still other species known by this name. 90In speaking to one of the men’s focus groups about women’s exclusion from these cooperatives, one participant who is a member told of a woman whose land he used to rent. She said that next year she will not rent her land since she wants to take part in the cooperative. She will help with all agricultural tasks, besides ploughing. 105 reduced average landholdings to a third of a hectare. Private grazing land is exceptional, as is leaving land in fallow, and upon marriage or death couples receive only a very small amount of land, if any, from their parents. While livestock numbers per household are not very high, particularly oxen are very important, and crop residues are only used for animal feed, while dung is often used for fuel, both of which have deleterious effects on the soil and on crop yields. Households may, however, be divesting themselves of small ruminants. There are substantial differences between households in the kushet which are based upon access to assets such as land, livestock, and male labour. FH had on average far less land than MH, and can be considered to be seriously land deprived. The quality (qua soils and slope) of FH’s land also tends to be lower. Given lower land access, a greater proportion of FH’s land is occupied by their homesteads. This probably explains why FH implemented water and soils conservation measures less frequently than MH, although female heads also reported that they couldn’t complete all of the work that was required since it was too heavy. The data tends to support the conclusion that the land that FH said was allocated to them by the baito was actually land allocated to the household during their former marriages and, upon divorce, the household’s baito land was unequally divided, where the land that women obtained was the worst. Much of FH’s land probably continues to be registered in the names of their ex-husbands. More than half of FH sharecrop their land out, mostly for reasons that revolve around oxen, and it appears that many FH sharecrop land out to their ex-spouses. While FH are much more dependent on non-household male labour for agricultural production, the majority nevertheless perform agricultural work, particularly in associated post-harvest activities. The benefits of agricultural production are therefore on average much lower for FH than for MH. Crop residues were harvested by half of the MH and by only 28% of the FH, since FH probably give up much residue to sharecroppers. Those that did harvest residues often exchanged them for ploughing services. Crop residues are important sources of feed for livestock that MH maintain. FH also planted on average a lower percentage of their land to crops, which is also apparently due to the fact that more of their land is occupied by their homesteads. They were unable to plant more than one or two crops, in comparison with MH, where half planted more than two. Between giving up 50% of the crops to sharecroppers and planting fewer crops on less land, more than 90% of FH consumed less than 1.5 quintals, compared with only about a third of MH. FH were particularly disadvantaged because of their lower access to male labour and oxen, which appear to be the factors most consistently associated with crop consumption levels. FH had higher dependency ratios on average, so it is clear that they are at greater risk of poverty and food insecurity. Given their limited access to resources, FH also clearly are able to maintain only a very small fraction of the livestock that MH can maintain. Livestock consumption was correspondingly highly unequal: only MH consumed large livestock and only one FH consumed chicken, compared with six MH. Such inequalities were also evident with respect to livestock products such as milk and eggs. Further, lack of land access and secure tenure also limit FH’s ownership of trees and homegardens, and the benefits to be derived from such ownership. There is no clear positive relationship between tree and land ownership, insofar as two-thirds of all trees were planted by households with less than two timads. However, MH were far more likely to own trees and had a higher average number of trees: 90% of FH had fewer than six compared with only 20% of MH. While most tree plantations are still young and do not yet yield many benefits, they will certainly bring major income streams in future, but FH will be deprived of such income. More than 90% of all homegardens likewise pertained to MH, and reasons that FH are unable to maintain homegardens include lack of access to land and water and poor quality of soils. Homegardens contributed to food consumption and at least one also contributed to household income. Unsurprisingly, only 13% of FH reported farming as their primary occupation, and none reported it as their secondary occupation. FH earn most of their living from tela production and sales, prostitution, grain milling, and food-for-work: 80% of all female heads reported income generation activities of this ilk. Aside from food-for-work, these occupations can be carried out entirely at home, which is a necessity for single mothers. Without viable alternatives, women become commercial sex workers and it is likely that the incidence of this activity was under-reported. Eleven (73%) of the FH participated in community 106 work programmes for which they were paid in food, compared to nine MH, but MH worked on average three times as many days. The greater labour power of MH not only permitted substantially time investment in food-for-work on the part of male heads, but as well as greater investment on the part of wives. Nor did FH report receiving in kind or in cash contributions that would appear to be alimony or child support. It is very clear, however, that FH strategize to survive and to improve their livelihoods in ways that make maximum use of the resources that they do have at their disposal. As was pointed out in Chapter 4, they attempt to invest in the future by sending their children to school, and here it was seen that they are more likely to save, and are equally likely as MH to take out loans and repay them. They seek to avoid sharecropping out land by exchanging crop residues and by hiring labour, and to best utilize the land that they don’t sharecrop out, e.g. by planting trees. Being largely unable to access formal labour markets and lacking formal education, they mobilize their domestic skills to earn income. They procure as much food as they are able through food-for-work. But, as indicated above and in other sections of this report, their constraints are numerous and nearly overwhelming, and their situation and that of their families is very precarious. The position of FH has its roots in the position of women more generally in highland culture, which is reflected not only in inequalities of resource divisions upon divorce and widowhood, but as well in marriage itself. Wives barely acquired land from the baito in their own names: rather, it was either registered in the name of the couple or, more often, in the name of the male head (77% of the plots). Production decisions within the household nevertheless appear to be relatively egalitarian, since it is perceived that major productive assets are held in common: what to cultivate, and how to use and dispose of agricultural land, are usually joint decisions, and both soil and water conservation measures are implemented jointly. However, when referring to livestock other than chickens, although men often reported that they are jointly owned, when they are sold the income pertains to the male head. While trees are also considered to be joint property, it is possible and even likely that, when these begin to yield substantial income, men will also control it. While the survey data showed that homegardens are jointly managed, focus group discussions indicated that men take control when they begin to yield substantial amounts of income. Wives consider themselves to be housewives or ‘domestic servants’, and about half reported that farming was their secondary occupation, which reflects women’s cultural status within the household as servants and farm helpers. While 60% of wives did something to generate income off-farm, women in general earned less than five birr per day. Wives participated as often as their husbands in food-for-work activities, but they invested somewhat fewer days in total. Women in MH barely received credit, which appears to be largely a male prerogative, and it was also nearly always the male head that had any savings. Wives were more often members of church organizations than their husbands, but only a third of all wives were members of the PA, and they were always members together with their husbands, just as they were only members of grazing and irrigation committees if their husbands were members. Women in MH are better off than their FH counterparts in terms of overall access to assets and to the benefit streams that derive from these, but their access obviously depends on their husbands, and marriage is a fragile institution indeed in Tigray. 107 108 Chapter 6 Common Property Resources, Resource Access and Use, and their Relations with Household Headship and Livelihoods This chapter presents and analyses the data collected in Adiarbaetu in relation to access to common land, the uses that the population makes of the areas and their botanical resources, and the ways in which use and access differ among individuals and households. Local perceptions about common land areas and enclosures, about the importance of botanical resources and the benefits derived from them, are fundamental to this analysis. The discussion focuses in particular on gender relations and female household headship, and on formal and informal rules of access to CPR and how these affect CPR use. It teases out the relationships between such access rules and poverty dynamics affecting both FH and the landless. The analysis presented in this chapter is based on a combination of sources, including interviews with village officials, focus group discussions, Women’s Access Histories, the Species Access Study, and the Household Survey. The Household Survey contained a module on CPR use where informants were women heads of households or the wives of male household heads, who were asked to report on behalf of their entire households. Many methodological problems were encountered that were mentioned in Chapter 3, but that are discussed in this chapter in greater detail not only to qualify research results, but as well to present lessons learned for future research around CPR. In the CPR module, women respondents were asked to report all uses (e.g., grazing cattle, collecting fuelwood, collecting medicinals, etc.) made of all common lands during the 12 months preceding the survey.103 A pre-defined list of common land areas was used to prompt respondents for answers. The women were also asked the sex and age of the person or persons who made that use, and their relation to the household head, to list the calendar months when the use was made in that common land area, to estimate the number of days that each use was made in each month, and to list the species used and their uses. When the number of uses is reported, this refers only to the number of times that households reported making a particular use of a particular common land area. It does not refer to the intensity of use: the proxy for this is the number of days that households make a specific use in a specific area. It must be noted that the use of women respondents for this module apparently biased the results in the same way that the use of male respondents would have created bias: women tended to report uses that were most significant to them, while leaving out some that were less familiar, particularly male uses such as hanging beehives and livestock grazing. The biases that may have been introduced have in part been compensated through the men’s focus group discussions, and the possible effects of such bias on the survey results are discussed throughout the chapter.104 Further, another module in the Household Survey asked open questions regarding perceptions of changes in household resource use over time, reasons for these changes, opinions about common land areas that are over-exploited and under-exploited, and reasons for such phenomena. This data is qualitative, and its analysis is incorporated into the text. There are several other methodological problems that have to do with the etic concepts that are derived mainly from the AE literature on Tigray, and attempting to use them with local informants. These problems have affected the results in ways that were unforeseen, but the ‘errors’ themselves provide points of departure, not only for interpreting the data, but as well for dealing with complexity in future CPR research in the area. 103Each use that was reported is referred to herein as ‘uses’; each time a species was associated with a particular use, this is counted as a ‘species use’; and each time a common land area was associated with a use, this was counted as a ‘common land area use’. 104Such bias could be avoided if both men and women were to separately report household uses, but this would generate double-reporting. Alternatively, all household members could report on their own uses, but again this would be likely to generate a certain amount of double-reporting, and would also take too much time. Another option would be to interview men and women together, but in this case women might be less likely to report. Different alternatives should be attempted and validated to determine who should be interviewed. 109 6.A Common Land Areas in Adiarbaetu The literature review led the research team to assume that there are very distinct common land areas in Tigray that can be readily distinguished in the field: nowhere was it mentioned that such distinctions are in practice problematic. The distinctions found in the literature include those related to tenure (‘private’ versus ‘common’), to regulations applicable to common areas (‘open’, ‘enclosed’, or ‘restricted’), and then to a mix of land use and land form types, including ‘woodlots’, ‘grazing land’, ‘hillsides’, ‘riversides and borderlands’, etc. However, the fieldwork in Adiarbaetu demonstrates that these distinctions generally represent etic rather than emic perspectives, and in practice are imprecise and often impractical. While local populations do make distinctions regarding common land areas, these are not as numerous or as static as what is implied in the literature, most of which relies primarily on quantitative methods and hence on rigidly defined categories. It was found that local tenure distinctions more or less coincide with those that are portrayed in the literature, especially those common lands that are subject to seasonal or permanent enclosure, privately held land where access is generally subject to permission of the owners, and all other areas which are considered to be open for use and where only species restrictions (see below) apply. However, as will be seen, those that are only seasonally enclosed may be perceived as ‘open’, especially by people whose uses of these areas are not directly affected by seasonal restrictions. For example, with respect to grazing land, many villagers appear to classify that which is only seasonally closed as ‘open’ or ‘unrestricted’, and only consider areas that are closed year-round as ‘restricted’, whereas the etic perspective is that grazing land that is closed both seasonally and year-round is ‘restricted’ or ‘enclosed’. This also led to much confusion and difficulty when interpreting the data. In addition, it was found that local informal rules of access and the basis for them were pervasive in the research area, and yet these are not acknowledged in the CPR literature on Tigray. Such informal rules are generally expressed as ‘exceptions’ to formal rules that may or may not be sanctioned by officials. There are principles related to equity, age, and need that underlie these informal rules and explain their application to particular species, spaces, and individuals. Illegal use that is related to such informal rules or ‘exceptions’ is most often not punished since it is either not reported by villagers and guards or overlooked by officials when reported. Villagers much less readily understood etic classifications of areas by land use and land form, such as woodlots, forests, hillsides, and grazing lands. Given that the area is very mountainous, a land form such as a hillside has many ‘land uses’, including areas with scattered trees (‘open woodlots’), degraded forests, and places where grazing, fodder, food and medicinal plant collection occur simultaneously or in different seasons. In the literature, hillsides are portrayed as though they are separate from ‘restricted grazing land’, although restricted grazing often occurs on hillsides. Locally, hillsides are considered as a land form where distinctions are made only according to sections that are permanently or seasonally enclosed. Similarly, people do not differentiate easily between forests and woodlots. They do use the term ‘grazing land’, but the use of such areas may be multiple, and may differ according to season. In addition, such distinctions lack consensus and are also clearly blurred depending upon perceptions, own uses, and factors such as seasonality. Since resource use occurs in many common land areas, terms such as ‘grazing land’ seem not to be useful if the intention is to establish exclusive land use categories or strict boundaries around common land areas.105 The etic distinctions mentioned above guided the formulation of the Household Survey CPR modules, which were necessarily developed and administered before the results of the focus group exercises were available. The survey results therefore are as questionable and ‘blurry’ as the definitions underlying them. However, there are certain use patterns that can be distinguished that show that villagers have some culturally significant concepts about land use and land forms that permitted them at least to respond to the etic classifications imposed upon them. 105As discussed in Section 3.E.4, in future, research must be done at local level using cultural consensus analysis (from cognitive anthropology, see references in Chapter 3) to determine land form and land use categories that are meaningful to local residents. Ellen (1998), and Howard and Nabanoga (in press), noted that land areas (e.g. forests) and resources are generally not perceived by local populations to be physically bounded, but rather constitute mosaics and networks of resources that are related to use, species availability, access rules, and types of users, which is in part what methods such as the Species Access Study are meant to capture. 110 6.B Formal Access Rules: Enclosures and Species Restrictions 6.B.1 Enclosures According to local officials, Adiarbaetu has eight communal resource enclosures (AE), three of which are enclosed year-round and five of which are seasonally enclosed (Table 64). However, there are many common land areas that do not appear in the table, because they are open year-round or because they are used by the tabia as a whole, rather than exclusively by the kushet. These latter areas, especially those that are ‘open access’, are those that are most used by villagers, as seen below. Most AE were created primarily to allow for grass regeneration for grazing, although grass is also collected from these areas for coffee ceremonies, thatch and basket making, and, to a limited extent, for construction. Permanent AE, such as the Arara forest area, were created for environmental protection and regeneration. As can be seen in tables 64 and 65, most of the tabia AE are associated with grazing, and only livestock access is limited. Humans are free to enter even during the enclosed period to use other resources, although they are not allowed to collect grass. All areas that are not enclosed are considered as free access year-round, both for humans and for livestock although, as is discussed below, there are other restrictions that prohibit the use of specific species no matter where they occur. Paid and volunteer guards enforce AE. The importance of AE, especially for grass growth, is such that villagers often report violations. Volunteer guards are paid in grass, and paid guards receive payment from the money collected for fines. The community elects all guards. Formal access rules and punishment of violations are discussed in Section 6.C for each respective AE type. Most seasonal AE are managed at the tabia level, and most permanent AE are managed by outside organisations, such as REST. As was indicated in Chapter 2, the local community is consulted before new areas are enclosed, and are given the opportunity to raise concerns and vote for or against the enclosure. Villagers and local officials in Adiarbaetu both reported instances when an area was proposed for enclosure and was left open after the majority of the community opposed. In focus group discussions, both men and women said that women were invited to, and participated in, such meetings. In general, the community supports the AE, although they recognize that these make their access to certain resources more difficult, which is what studies in other communities across Tigray have also reported (see Section 2.D). The environmental benefits of AE are clearly and fairly quickly visible, and this is what is highly appreciated. As one woman said, “If there were no enclosures, the environment would increasingly deteriorate, and this will ultimately affect us more.” However, as is discussed throughout this chapter, not all feelings are positive. 6.B.2 Species Restrictions In addition to the AE, restrictions that apply only to woody species have been created at regional level, and are also effective in Adiarbaetu. These restrictions are only mentioned in Nedessa et al (2005) and not in the other literature reviewed, possibly because most CPR research has focused on AE, where in any case the use of woody species is already tightly regulated. According to tabia officials, species restrictions began to be enforced approximately ten years ago, and have become more strict and numerous over time. Species restrictions apply to all common land areas, regardless of whether they are open or enclosed, but they do not apply to trees on private property. Restrictions are divided into two categories: species that cannot be cut for any reason, and those that can be cut only with permission (see Table 66). Species that cannot be cut for any reason are those considered to be most endangered. However, many exceptions are made in the enforcement of these restrictions. Also, just as was reported in Section 2.D, residents are often uncertain about which tree and plant resources can be used and which cannot. In general, species restrictions are enforced through checkpoints set up at various locations through out the kushet. There are local DAs who are assigned to each kushet for this purpose. However, as is the case with AE, the community at large enforces many of the regulations: villagers report violators. Violations can result in heavy fines or imprisonment. The severity and frequency of punishment depends on many factors, including age of the offender, the species cut, the amount taken, the state of the species taken (living versus dead), and the number of prior offences. First time offenders and children are often let off with a warning after their names are taken down. However, repeat offenders can face fines from 50–150 birr. Imprisonment is required for those 111 COMMON LAND AREA Mutmak Mihiradha’a Gerib Ede Kisadmahla Table 64 Area Enclosures in Adiarbaetu TYPE OF AREA ENCLOSED DATE OF AREA (HA) ENCLOSUR E Grazing, 15 All Year 2004 hillside, forest Grazing 39 Seasonal Before Derg Grazing 36 Seasonal Before Derg Hillside/ 63 Seasonal, grazing, All year 2003 woodlots (woodlots) Gomata Amharetto Woodlot Hillside 5 105 All year All year Arara Hillside forests, woodlots 61 Seasonal, All year (woodlots) Baharatd Hillside, forests 105 Seasonal 2005 Before Derg Before Derg, 2005 (woodlots) Before Derg MANAGED BY WFP None Tabia Seasonal Tabia Cut & carry Tabia, REST (woodlots) Tabia Tabia Seasonal, cut & carry (woodlots) Tabia, WFP (woodlots) Tabia GRAZING None Cut &carry Seasonal, cut & carry (woodlots) Seasonal Table 65 Permissible Uses* of Area Enclosures SEASONAL ENCLOSURES COMMON PERMANENT Open LAND AREA ENCLOSURES Season Enclosed Season Grazing Land Medicinals Grass Medicinals Aromatics Medicinals Aromatics Wild Food Aromatics Wild Food Fuelwood Wild Food Fuelwood Fuelwood Woodlots Medicinals N/A N/A Aromatics Wild Food Bee Hives Forests Medicinals Grass Medicinals Aromatics Medicinals Aromatics Aromatics Wild Food Wild Food Bee Hives Wild Food Bee Hives Bee Hives Hillsides Medicinals Grass Medicinals Aromatics Medicinals Aromatics Wild Food Aromatics Wild Food Wild Food Fuelwood Churchyards Medicine N/A N/A Wild Food *The species restrictions listed in Table 66 are also enforced across all of these areas. 112 Table 66 Species Restrictions CAN NOT BE CUT CUT WITH PERMISSION ONLY Chihidi (Juniperus procera, Cupressus spp.?) Cha’a (Acacia sp.) Awelie (Olea europea ssp. cuspidata) Thehetses (Dodonea angustifolia) Tata-aelo (Rhus natalensis (or Rhus spp.?) Sheila-en (Cadia purpurea) Shemfa (Ficus sur) Mengolhas (Dovylais abyssinica) Kunkura (unknown) Daro (Ficus vasta) Source: Village officials (in particular the Natural Resource DA) who are unable to pay the fines, and for those who commit multiple offences.106 The length of imprisonment is also dependent on the number of offences, the amount of fine owed, and other factors mentioned above. Cutting or destroying live trees is punished more severely than cutting or collecting dead or fallen branches, where the former incurs fines of 500 birr or imprisonment. Enforcement is also species dependent and is stricter for those pertaining to the ‘no cut’ list in comparison with those that require permission to cut. Even within the ‘no cut’ list, some species are reported to be more strictly protected than others. Likewise, permission is more likely to be granted for some species on the ‘permission required’ list than for others on the same list. Officials take cultural importance, level of depletion, and rate of regeneration into account when enforcing restrictions or extending permission: it can be presumed that all of these criteria represent more than just legally sanctioned locallevel self-determination. However, as discussed below, the existence of certain ‘morals’ or informal rules of behaviour, and the recognition of the importance of certain species to livelihoods, give rise to a ‘need for exceptions’ that is accepted by both villagers and officials. Beyond the species restrictions, which obviously affect the collection of fuelwood and construction materials, and the seasonal restrictions on grass collection, all other species can be used freely as long as land is not privately owned. Medicinal uses of any species present one of the important de facto (and hence informal) exceptions to all formal rules. Any plant, or any part of any plant, can be taken any time from any location if its use is medicinal. According to local informants, even restricted species can be used for medicine without risk of punishment. While officials generally agreed, they insisted that not all species can be used: the most restricted species, such as awelie, cannot be used even for medicinal purposes. Nevertheless, the use of restricted species for medicinal purposes in practice occurs freely: apparently guards and villagers do not report it. Community members’ attitudes toward species restrictions are ambiguous to say the least, and residents are also often disgruntled about them. Unlike the case with AE, they do not have a say in the creation of species restrictions, which are decided by the regional government and enforced by the DAs. While many residents are aware of and can see the environmental benefits of such restrictions, they also emphasize the negative effects that these have for their livelihoods and well-being. When discussing restrictions on species used for fencing material, for example, one woman stated: “I’m not able to fence my property properly and animals come and destroy my garden. Hyenas kill my livestock…there are more and more restrictions and I am now afraid to take anything.” Below it will be seen that the number of complaints about restrictions has at least in one case forced officials to negotiate a partial ‘solution’. 6.C Common Land Areas and their Uses According to both the focus group discussions and the Household Survey, all community members (men and women, children and the elderly, MH and FH, destitute and better off) use every common land area. However, especially enclosure status, the gender division of labour, and household head- 106The procedure is generally as follows: for a first time offence - confiscation and a warning; for a second offence – a fine; for a third or further offence - imprisonment. For children, only repeat offenders are fined. Technically, their parents can be imprisoned if their children are caught more than twice, but it was said that this has never occurred in the kushet. 113 ship affect the areas used, the frequency of use, the uses made, and the plant species used. Below, these patterns are described and discussed with respect to each type of common land area. 6.C.1 Grazing Lands and Grass In Adiarbaetu, just as in the rest of Tigray, grazing lands are used to feed livestock both through direct grazing and through grass cutting. Grass also provides thatch for roofing, basket making, and for the culturally important coffee ceremony (see Box 7). As a result, grass is one of the most important botanical resources in Adiarbaetu. According to one resident, in the past Adiarbaetu was almost lush with grass: “The entire village used to be covered with grasses and the homes were all made with thatch.” The elderly focus group discussion concurred, and explained why this is no longer the case: This kushet was full of grass up to our thighs and we had many marshy areas. It was so thick in places we would often struggle to walk through it. Now these grasslands have changed into settlement areas, such as where the main town is located. Even the place where you [the researchers] are living was a marshland. We used this area for livestock grazing and the open area for entertainment. Measures to counteract deforestation and degradation, such as the enclosure of land for woodlots, have also had an impact by limiting the amount of area available for open grazing. Five of the eight AE in Adiarbaetu were created largely in an effort to manage grazing land. AE are enforced during the rainy season, from July to October, solely to allow grass to regenerate. After October, these areas are once again open for grazing. Outside of these restricted grazing lands, there are a few other predominately grassy areas that are used primarily for grazing (designated as ‘open grazing land’ in the Household Survey), but most common land areas are used for grazing. Therefore, the term ‘grazing land’ also encompasses other common land areas such as ‘hillsides’. Even some of the permanent AE are used to collect grass through the ‘cut and carry’ system107: livestock are not permitted to graze, but grass can be taken from these areas at designated times. According to officials, due to the depletion of grazing land and the high livestock population, grazing and illegal grass cutting are the most frequent violations in the kushet, and occur particularly in areas that are seasonally enclosed. Violators are fined approximately 10 birr, although the amount can depend on the number of past violations and, in the case of ‘cut and carry’, the amount of grass taken, and can be as high as 50 birr. Grasses used for basket making and thatch require longer growth periods. Therefore, they are increasingly difficult to find in open areas and are now primarily harvested from AE. Certain AE are open on specific days before being opened to grazing (generally in September and October) to permit residents to cut specific grass species used for basket making and thatch. Generally, the opening is only for one day for a particular species, although occasionally it can be extended to a second day. Since the grass species used for these purposes mature at different times, harvest days are not simultaneous, resulting in a maximum of four days in total per year for harvesting. Difficulties were encountered when interpreting the data from the Household Survey especially in relation to the use of grazing land for grazing, since this was barely reported, even though all of the MH reported owning livestock other than chickens.108 This could have in been due to the fact that women were the respondents for the CPR module and they may have under-reported some of male adults’ use of CPR. However, it is also related to the fact that only two types of grazing land (restricted and unrestricted) were included as options in the CPR module. In this module, grazing was only reported a total of five times: three times on restricted and twice on unrestricted grazing land 107Cut and carry is a general term used for any grass that is collected rather than grazed, including grass that is sold or auctioned off from woodlots. 108Summarizing the results of the Livestock Module with regard to the existence and distribution of livestock, Table 42 reports the percentage of FH and MH who have livestock according to the total number of TLU, where only one FH had any livestock that required grazing or fodder. There is therefore no reason to suspect that FH heads under-reported the use of grazing lands for livestock feed. Of the MH, one-third (five households) had 1.5 TLU or less, meaning that grazing requirements for these households were fairly low. Only six households had more than three TLU, and therefore had quite substantial feed requirements. 114 (see Table 67A). Had men been the respondents, it is possible that they would have used the category ‘other community enclosures’ to report instances of grazing; however, women did not. In the Livestock Module, several more response options were available: restricted grazing land, unrestricted grazing land, other community land, and private land. By far the most commonly reported source was ‘other community land’ (28 instances). In contrast, ‘unrestricted grazing land’ was reported only twice. ‘Restricted’ grazing land use was reported five times, all of which was for grazing oxen. The latter two figures nearly replicate the reports on grazing in the CPR module. During the community mapping exercises, participants listed only one grazing land area as open year-round, and they reported using this area only when restricted grazing areas were enclosed, which can explain why so little use was reported of ‘unrestricted grazing areas’ for grazing purposes. Therefore, the category that was missing in the CPR module was ‘seasonally enclosed grazing lands’, which might have captured the use of ‘other grazing lands’ that was reported in the Livestock Module. This problem biases the results of the CPR module throughout this report with regard to the use of grazing land for grazing. Table 67A has been adjusted to incorporate the reports from the Livestock Module by adding 28 instances of grazing under ‘other community enclosures’, but this is the only place where this adjustment has been made since other important indicators (e.g., person using, amount of days used, seasonality) were not available from the Livestock Module. Most of the other reported uses of grazing land should not be much affected by this partial, albeit important, omission. Restricted Grazing Land Table 68A shows that only 40% of the surveyed households reported used restricted grazing land in the 12 months prior to the survey, eight (or two-thirds) of which were FH. Only five households reported that they used the areas for grazing, one more than the number reported in the Livestock Module. The vast majority of other uses were also grass-related: grass collection for coffee ceremonies, for basket making, and for fodder which, together with grazing, constituted 87% of all uses. The predominance of grass in restricted grazing land use is largely explained by the fact that these areas are meant to produce this resource and little else grows there. They were also occasionally used to collect a few medicinal and wild food species, such as echote talien (Agremone mexicana?). These areas are not used for construction or fuelwood since they contain few woody species. In spite of the differences in use for grazing, female users outnumbered male users (59% versus 41%), and all but one user was an adult. With respect to the days of use, one household reported using restricted grazing land the equivalent of every other day of the year, whereas another reported using it only one day per year; the average of all days reported divided by all households was 58 days per year. Unrestricted Grazing Land Table 68A shows that a greater percentage of surveyed households (slightly more than half) reported using open grazing land. FH represented only 37% of the households using these areas, and female users outnumbered male by 2.5:1. Adults outnumbered children by 5:1. It is clear that unrestricted grazing land is used quite differently than restricted grazing land, and it is used far more frequently, which is not surprising given that the latter are open year-round and that many more hectares are involved. Grass is collected for coffee ceremonies, but it is not collected for thatch at all, because it is cut before it can grow to sufficient heights. Together, grass-related uses constituted only 42% of total uses.109 In comparison with restricted grazing lands, unrestricted grazing lands are a source especially of medicinals (nearly 30% of total uses), as well as of a limited amount of wild food species and of fuelwood, which in part is a function of the fact that more households use open grazing lands at all. FH have equal access to both open and restricted grazing land, and they use open grazing areas a great deal more than MH: 71% of FH versus 29% of MH, respectively. Most FH, which have no livestock, use grazing land to collect grass for coffee ceremonies and thatch for baskets. Many also used open grazing land to collect medicinal species, fuelwood, and wild food. 109Fodder collection is not included in this percentage since most households reported that they collect beles and not grass for fodder from open grazing lands, which is not the case with restricted grazing lands. 115 6.C.2 Unenclosed Hillsides Unenclosed hillsides are by far the most frequently used common land area in Adiarbaetu: in the household survey, 29 out of 30 households reported using them. Little is known about the history of hillside tenure, and what little literature does mention is contradictory on this point. On the one hand, some state that hillsides have a history of relatively open access and, on the other, some report that, at least since the time of Haile Selassie up until 1974, hillsides were part of rist land and governed by its rules (Chisholm 2000). Since that time, however, they have been open access (see Section 2.D.2) and they are now largely stripped of trees. Since 1991, management of hillsides occurs at the tabia level. Unlike the case of enclosed woodlots and grazing lands, where use is largely restricted to particular kushets, hillsides are used by all kushets within a tabia without distinction (Chisholm Ibid, Pers. Com. Natural Resource DA). Open hillside areas are now only controlled by species restrictions which limit, but do not halt, the use of such species. As was mentioned earlier, open hillsides are also monitored via checkpoints to enforce species restrictions. However, there are only a limited number of checkpoints to enforce these restrictions over a large area, and violations are frequent. The most important use of hillsides is for fuelwood collection but, to a lesser extent, trees are also cut for construction and tools. In focus group discussions it became clear that men use these areas primarily for grazing, to obtain construction materials and, to a lesser extent, to collect wild foods, grass for thatching and ceremonial uses, and to hang beehives. Women focus group participants reported using them to collect fuelwood, aromatics for etan (Box 8), wild foods, medicinal plants, dung, and grass for coffee ceremonies. According to the Household Survey (Table 68A), the 29 households that used open hillsides reported a total of 61 uses, for an average of just over two uses per household. This table shows that hillsides were used most to collect wild plant foods (43%) and fuelwood (33%), which together constituted three-quarters of all uses: out of all common land areas, it is principally from open hillsides that both of these important resources are collected. Medicinals constituted another 10% of uses, and collection of grass for coffee ceremonies and aromatics were minor uses, at about 5% each. Table 69A lists the species that households reported collecting and the number of times that each species was reported, by common land area. Out of a total of 41 species that were reported, 27 (two-thirds) were collected from open hillside areas, which was the highest number of species reported for any area. Use of these species was mentioned 136 times, which represents 41% of the total (310) reported uses for all areas and all species. Table 70A shows the uses of the species that were collected from open hillsides. Collection of wild foods (30%) and fuelwood (45%) accounted for threequarters of the uses, and medicinals for another 10%: collection of aromatics and grass for coffee ceremonies constituted minor uses. Beles, which has now naturalized on hillsides, was the most frequently used species and was collected principally for use for human food, and to a much lesser extent for fodder. It was followed by hahote and tabub (Becium grandiflorum?), used principally for fuelwood, which are among the few woody species whose use is not restricted; as well, households often reported collecting kirshem, or miscellaneous small branches and twigs. Beles, hahote, and tabub alone represented just over half of all unenclosed hillside uses. The only other species that represented more than 5% of reported uses were awelie and hama shiro (Caralluma sprengeri?), where awelie was used mainly used for fuelwood and hama shiro as a wild food. Nine species were collected for medicinal use, seven for aromatic use, and grass was collected only for coffee ceremonies, whereas species used for construction were barely reported. Because hillsides cover such large areas, species density also varies greatly from one part of the kushet to the next. Focus group participants and others reported that certain hillsides are known and used for particular species. For example, during the spring holiday season, women use ar-aro (Kalanchoe marmorata?) to make a special steamed bread. While it is relatively abundant in more rural parts of the kushet, it is difficult to find in the hillsides surrounding the centre. In general, this contrast is true of other resources as well: residents in the more central, densely populated area of the kushet have a more difficult time finding many resources that are more abundant in more rural areas and, as a result, they often must walk far to collect them. Table 68A gives information about the users of CPR in hillside areas, and about intensity of use as measured in days of use per year. Women outnumbered men as users by 3:1, and adults outnumbered children by 4:1. Those who most often collected plant species from these areas were household heads, then wives, and then sons and daughters. FH and MH were equally represented 116 and used them more or less equally. The number of days of use reported in total was around 3,000, by far the highest for any type of common land area, for an average of 28 days per year per household, where much variation was reported: a maximum of 360 days and a minimum of one. 6.C.3 Woodlots and Forests Although elders reported that the kushet used to be “covered with forest”, there is relatively very little natural forest left today. Although the etic distinction between forests and woodlots as land use types appears to be clear, where forests supposedly naturally occur and woodlots are purposefully planted, at times villagers were confused about the terms and used them interchangeably. Part of this ambiguity stems from the fact that there are certain areas that once were woodlots and nurseries but that are no longer managed and now resemble forested areas. Ambiguity could also arise from the size of the areas in question. Woodlots and forested areas can range from a few to hundreds of ha (Berhanu 2004). Also, some hillside areas are now being reforested by organizations such as REST. Oddly, at other times informants seemed to make a clear distinction between the two, stating that woodlots were created and managed by the government and other organizations, whereas forests were natural and managed by the community. Therefore, depending on the informants’ understanding of these terms, results of the survey and other methodologies used must have varied. Adiarbaetu has two types of woodlots and forested areas: those that are enclosed year-round and those that are enclosed only seasonally. The Household Survey used two categories: ‘enclosed woodlots’ and ‘unenclosed woodlots and forest areas’, since it was designed on the basis of the literature review, which did not mention such seasonal woodlot enclosures. Many survey respondents indicated that they use ‘unenclosed’ woodlots and forest areas. What is interesting is the fact that uses that were reported for these areas were very clearly seasonal, whereas those reported for unenclosed woodlots manifested no seasonality at all (see graphs 7 and 8), which would suggest uniformity and accuracy of interpretation of the etic categories among household respondents. Enclosed Woodlots In Adiarbaetu, organizations such as REST manage some of the woodlots and reforestation areas, while others are managed locally by the tabia administration. Woodlots are generally enclosed yearround and are only opened at particular times that are determined by the tabia administration for timber and grass sales. While they are open a few days per year for grass sales, informants reported that often woodlots are only open a few days every four years for timber sales, depending on tree growth rates, which is what was often reported for woodlots in Tigray (see Section 2.D.3). Woodlots consist of either single species (Eucalyptus spp.) or mixed species plantations. The wood sold is most commonly used for construction and ploughing tools, whereas, as mentioned before, grass is sold for fodder, thatch, and baskets. In focus groups discussions, the majority of the respondents reported that they only use enclosed woodlots for timber for construction, and that they fulfil their fuelwood needs elsewhere. Enclosed woodlots are used by a limited population, at limited times, and for limited uses. While this is in general the case, particularly in relation to periodic sales of woodlot products, it was also learned that enclosed woodlots are frequently used to glean other products for which no payment is required. In focus group discussions, men reported that they use enclosed woodlots to hang beehives, while women reported using them to collect medicinal species. It was thought that, since these areas are enclosed, they have a higher diversity of both woody and herbaceous species; however, other studies in Tigray show that this is not necessarily the case, and that species richness, diversity, and density depend on the age of the enclosure as well as on management practices (see Section 2.D). The Household Survey results that are reported in Table 67A show that collection of medicinal species and grass are the most commonly reported uses (28.6% each) of enclosed woodlots, followed by collection of fuelwood and construction materials (14.3% each), and finally by collection of aromatics (7.1%). Table 68A reports that 30% of the households surveyed used enclosed woodlots for any purpose over the preceding 12 months, of which 80% were FH. Collectors were mainly female (69%), and also principally adults (75%), and it was the head of household who most collected (56%), in this case followed by offspring (25%) and then wives of male heads (13%). 117 FH seem to have no exceptional problems with access: rather, they appear to rely more on enclosed woodlots than MH. Women, including FH, use these areas primarily for non-wood products such as medicinals and grass for coffee ceremonies. Herbaceous and some woody species can be collected by anyone year-round. However, FH do seem to be limited with respect to access to woody species. In focus group discussions, women reported that, although they are often the primary caretakers of woodlots (planting and watering the trees as food-for-work activities), the majority do not benefit from timber or fodder grass, since they cannot afford to purchase the wood and do not have livestock which would require grass (if and when that is sold). As one female head put it, “No one prevents us from using the woodlots. The problem is that we do not have the money to buy anything that is sold. We just take care of the species and, when they have matured, they are sold cheaply, but we do not have enough money to buy them.” This reflects precisely what Nedessa et al (2005) reported (Section 2.E) when they discussed inequity in the distribution of benefits of AE. The ten households that reported days of use for enclosed woodlots reported a total of 340 days of use per year, for an average of 34 days per household, although the variance was great: a maximum of 150 and a minimum of one day per year of use. Since enclosed woodlots are only open a few times a year or over several years, it is not surprising that there is much variability in their use and, compared with other ‘open’ or seasonally-enclosed areas, they are used less frequently by fewer households. Unenclosed Forests and Woodlots What are referred to here as unenclosed forested and woodlots areas are in reality seasonally closed to livestock grazing to allow for grass regeneration. When they are open, grazing is allowed and grass may be collected using the cut and carry system. The use of trees growing in these areas is forbidden. Since unenclosed forests and woodlots are often the only areas outside of enclosed woodlots that contain large trees, men often hang their beehives in them. In focus group discussions, men reported only using unenclosed forests and woodlots to purchase grass and to hang beehives. Women reported using these areas more frequently to collect grass, fuelwood (only non-restricted species such as tabub and hahote), and medicinals. These uses were further collaborated by the Household Survey where, as Table 67A shows, the collection of medicinals alone accounted for 50% of the total reported uses, followed by fuelwood collection, at 25% of the uses. Grass collection for coffee ceremonies was more frequently reported in enclosed woodlots, which is probably due to the fact that unenclosed forests and woodlots are either seasonally enclosed to prohibit grass use or are subject to grazing. In fact, this was the only difference in use between enclosed woodlots and unenclosed forest and woodlot areas that appears to be significant, aside from the seasonality of their use, as mentioned above. Table 68A shows that, compared with enclosed woodlots, slightly more households (12 out of the 30 surveyed, or 40%), reported using unenclosed forests or woodlots for any purpose. Of those using unenclosed areas, somewhat more than half (58%) were FH, and nearly half of these FH also used enclosed woodlots. Collectors in seasonally enclosed forests and wooded areas were even more predominantly female (86%); the same percentage were adults. Again, half of the users were heads of household, followed by wives (28%), and then offspring (14%). The only instance where an adult male was reported to have used these areas was to collect materials to make tools. Grass was collected only for coffee ceremonies and aromatics were also collected, but to a minor extent (8% each of total reported uses). Unenclosed forests and woodlots were, however, used more often than enclosed woodlots, with a reported 535 days of total use and an average of 59 days per household. Again, great variance was reported: a maximum of 360 and a minimum of five days per year. Species Collected from Woodlots and Forested Areas Table 71A reports the species that households collected and the number of times that each was reported, by common land area. Of a total of 41 species used, 27 (two-thirds) were collected from enclosed woodlots or unenclosed forest and woodlot areas, where the number of species used from each area was about the same. Use of these species was mentioned a total of 49 times, representing 16% of the total use reports across all areas. The number of times these species were reported was also almost equal between the two types of areas. Of these, four were collected in unenclosed areas that were not 118 collected in enclosed areas, and six were collected only in enclosed areas, whereas the remainder (eight species) were collected in both areas. The use category with the highest number of species collected (nine) was medicinals, followed by fuelwood (seven). Eucalyptus, and then atush (Achyranthes aspera?), were the species most used for medicinals, whereas tabub and hahote were those most used for fuelwood, although two households reported using awelie, whose use is formally totally prohibited. It appears that there is little difference in terms of the number of useful species between enclosed and unenclosed areas, although there is some difference in the type of species. 6.C.4 Other Community Enclosures In the Household Survey CPR module, a catch-all category was included to permit respondents to report use of enclosed land areas that were not captured by the terms ‘enclosed woodlots’ and ‘restricted grazing lands’. This was due both to the fact that some of the literature reviewed initially was ambiguous about the types of AE that were created, and because of the concern that respondents might not share common concepts regarding the terms used. As was indicated above, this category apparently was not used to report grazing in seasonal enclosures, but in Table 67A the reports on grazing use in seasonal enclosures from the Livestock Module were added. Respondents could also have been referring to churchyards, where there was no special designation in the survey for them, however fuelwood collection is prohibited in churchyards but was commonly reported for ‘other community enclosures’. Many of the respondents in fact reported that they use ‘other community enclosures’, and the data reported has its own particular pattern both of uses and users. Without additional research, however, it is impossible to determine to which common land areas these data refer, and the principal interest in them here is to ensure complete recording of CPR use. Table 67A gives the uses reported, where nearly two-thirds were for grazing, followed by fuelwood collection (18%), and then medicinals and aromatics (4.5% each), where wild food plants and ‘others’ were only reported once. Excluding the use of these areas for grazing, which was exclusively a MH use, Table 68A shows that 90% of the other uses made of this category of AE were FH. Still excluding grazing, three-quarters of the users (Table 68A) were female, but even excluding grazing, in this case more males were reported to be users in comparison with any of the other AEs. Heads again represented more than half of the users, but daughters, sons, and other relatives used them considerably, and wives were, of course, barely represented. Three-quarters of the users were adults, but in this case a considerable number of the users were children, mainly from FH. The species that were collected are reported in Table 69A. Of the total of 41 species reported, nearly half (20) were collected from these areas, which was the second highest of all common land areas after open hillsides. Use was mentioned 59 times, representing 19% of the total species reports. Table 72A shows that, again, hahote and tabub were the species most commonly used, but the use of awelie was also frequent (five reports), followed by thehetses (Dodonea angustifolia?). Most use of these species was for fuelwood (64%), followed by medicinals (17%) and aromatics (12%). 6.C.5 Riversides and Other Borderlands While rarely at the forefront of research or institutional evaluations of resources, riversides, roadsides and other borderland areas may be important sources of botanicals. In Adiarbaetu, this proves to be especially true for FH. Some riversides in the area are enclosed to allow for the regeneration of reeds and other species used for housing and fencing and cannot be used for any other purpose: most are fenced off. Some residents also have farmland or homegardens near to or alongside rivers. While the exact rules of access to these areas are unknown and may very possibly be undefined, residents appear to believe that the riversides adjoining these farms and homegardens can be used only by the landholders. Many riverside areas, however, are understood to be open access, and are used for multiple purposes. Often, such areas are closer than other common lands used for the same resources (such as hillsides), and focus group participants said that they are used to supplement other CPR use. Riversides and borderlands provide small, yet accessible, pockets of important botanical resources. Riversides and other borderland areas are also subject to great seasonality of plant growth (see Graph 8). As a result, many of these areas are used predominately during the rainy season, when plant growth is at its peak. However, throughout the year, riversides are areas of lushness and moisture that are generally difficult to find in this arid region, so that many species that were once abun- 119 dant are now only found growing along them. As well, some species that are found only during the rainy season throughout the rest of the kushet can be found growing year-round along some riversides. As a result, men and women from both MH and FH considered that these areas are underexploited, and about a third said that they would like to use them more for homegardens or would like to see the community plant more trees there. The Household Survey (Table 67A) indicates that riversides and borderlands were used to collect medicinals (one-third of the uses), wild foods (also one-third) and grass for coffee ceremonies (25%). The only other use reported was fodder collection. While grazing was not reported, researchers observed it occurring along riversides, roadsides, and pathways. Many roads are lined with beles, which provides wild food and fodder, and ereh talien (Agave spp.), which is sometimes used for fibre. In the survey, riverside and borderland use was only reported by FH, and survey results also showed that it was these women who were most concerned about the depletion of resources in these areas. Since they do not appear to be used by the population as a whole, these ‘pocket resource areas’ are largely ignored by the tabia administration. While some women even suggested that these areas be officially enclosed, others recognized the limits of this as a solution, saying “You can’t enclose everything. In the end, drought is also a problem, not just people”. Table 68A shows that nine out of thirty households (30%) reported using riversides and boundary areas, all of which were FH. In focus groups discussions, however, men said they use riversides primarily for grazing. In the surveyed households, females outnumbered male collectors by 7:1, and heads represented more than two-thirds of all collectors, whereas their children made up the other third. The total number of days of use reported was 171, for an average of 19 per household, but variation was great with a maximum of 90 days and a minimum of one day reported. Species used in riversides and other borderlands are presented in Table 69A. Of a total of 41 species, only seven (17%) were collected from these areas, which was as low as that reported for restricted grazing land. Use of these species was mentioned a total of only 15 times, representing scarcely 5% of the total species reports. Beles, hama shiro and grass were most frequently used; the use of eucalyptus was reported twice, and all other species were reported as used only once. Table 73A shows that wild foods accounted for the highest percentage of use (48%), followed by medicinals (27%). There is certainly a disjuncture between the supposed species richness and diversity of some of these areas (particularly riversides) and the use that is made of them, both in terms of the number of species used and of the frequency of use. This, plus the facts that only FH reported using these areas and that several survey respondents felt that they were under-exploited, indicates that the fuzziness of access rules, or the de facto privatisation of these common land areas, may mean that the only users who are tolerated or dare to use them are very poor FH. 6.C.6 Churchyards (Sacred Groves) In Adiarbaetu, there is a clear visual demarcation between churchyards and most other common land areas, since the difference in vegetation density is so great. While the Household Survey did not include the category ‘churchyards’ in the CPR module and hence collected no data on their use, qualitative research revealed a surprising amount of use, particularly among women. It is forbidden on religious grounds for the community to take fuelwood, grass, or construction materials from churchyards; no trees may be cut and no grazing is allowed. However, medicinal species and wild foods may be collected. During focus group discussions, women indicated that they often use these areas, especially to collect medicinal species. Interestingly, while men are clearly aware of them and the species available therein, they did not report using churchyards at all. Residents in general appear to have much higher regard for the Ethiopian Orthodox Church in comparison to other institutions. This respect, in combination with religious beliefs, prevents violations: the very idea of violating these restrictions was incredulous to villagers. Cultural and religious restrictions are more authoritative than legal restrictions, which results in crucial pockets of conservation with high richness, density, and diversity of indigenous species (see Section 2.D.3). 6.C.7 On-farm Resources While crops produced on private farmland are the property of the landholder or sharecropper, several other botanical resources found on such land are considered by residents to be common property. 120 For example, children are often seen in the fields during ploughing time searching for a small root called kuenti (Cyperus bulbosus?), which is roasted and eaten. While children mostly collect them from their own families’ fields if they have them, they are also able to use others’ fields. Similarly, a wild edible green known as hamli talien,110 which is eaten during the rainy season, can also be taken from private farmland. An invasive species locally known as echote talien is found in abundance on farmland. The seeds (called medafe talien) are collected and ground into an oil which keeps the grill-like stove top that is used to make injera well seasoned, preventing the dough from sticking to it. These species are also found on common land, but they can be harvested from farmlands without asking permission. Many people surround their homes and line paths leading to their homes with living fences consisting primarily of, but not limited to, baharzef, kokola (Euphorbia abyssinica?), beles and ere talien (Agave spp.?). These provide them with valuable resources for construction, food, fodder, and medicine, as well as privacy, and protection from livestock and wild animals. All of these resources are the property of the owners, and generally are used solely by their households. If theft occurs, it is up to the owners to prosecute those responsible. However, violations are reportedly rare, and do not seem to be a problem or a source of conflict within the kushet. Children and others sometimes take beles fruit and eat them on-the-spot, but an owner is more likely to prosecute someone who takes the pads rather than the fruit, since the former are fed to livestock. As indicated before, medicinals may be taken by anyone from anywhere, even from private homesteads. 6.D Seasonality and ‘Intensity’ of CPR Use Seasonality is a very important factor in CPR management, in access rules and in determining the contribution of CPR to livelihoods. Efforts were made to capture seasonality through the qualitative Landscape Calendar and Resource Distribution Study carried out with focus groups, and through the Household Survey CPR module. In the latter, respondents were asked to report the number of days that a particular use was made of a particular common land area according to calendar months. The number of days for which collecting was reported represents a very rough estimate and may be overestimated, since (a) respondents’ recall is generally imprecise; (b) an entire day is not required to collect or use a particular resource, and (c) multiple resources are often collected at the same time, but days of use were requested for each use that was made of a single common land area. For the latter, obvious double reporting was eliminated when data were cleaned. It is still possible that some double-reporting was not detected, although there is no reason to think that this would occur more often for some uses or common land areas than for others (e.g., bias would not necessarily be introduced). What leads to under-estimation, however, is the fact that many botanicals are collected only sporadically (e.g., ‘on religious holidays’), which resulted in a relatively high number of missing cases. This data therefore provides only a very rough proxy for intensity of use. It is, however, a better indicator of seasonality. The seasonal use patterns are noteworthy, conform more or less to what would be predicted, and are presented in Table 74A and in graphs 5 through 8. The number of households and average annual number of days per household per major use category that were found through the Household Survey are found in Table 68A. Seasonality of use should depend on CPR availability and labour availability, given that the resources that people seek in community areas are generally scarce, and labour demand can be assumed to always be higher than actual availability on a seasonal basis. Looking at seasonality of use by major use category (Table 74A and Graph 6), little variation is apparent in those uses that are restricted (construction materials that rely on timber, grazing, and fodder collection), with the exception of aromatics, where most uses are of leaves and small branches from tree species (awelie, eucalyptus) which are available year-round. However, the use of aromatics still exhibits some seasonality, being lowest from February to May. This is most likely explained by religious restrictions on use for etan from February to May, which is a fasting period. Clearly, livestock must be fed year-round, and either grazing or collecting fodder are essential for this purpose. The slight decrease in grazing during May through September is probably explained by AE: restricted grazing areas are enclosed during much of this period to permit grass 110This is a common term in the area for different wild greens. In the literature it is usually reported as Brassica rapa, but the plants that were observed and photographed were not Brassica spp. Another possibility is Amaranthus graecizans, which in the Famine Field Guide is also called hamle tilian in Amargna. 121 Graph 5 Average Days per Year of Use, by Major Use Categories Construction Basket thatch Aromatics Cattle grazing Fodder Coffee ceremony Wild foods Medicinals Fuelwood 0 20 40 60 80 100 120 140 160 180 200 Graph 6 Seasonal patterns of collection/use per major use category 35 Reports of any use 30 Aromatics Fodder 25 Grazing cattle Coffee ceremony 20 Construction Fuelwood 15 Medicinals Thatch for baskets 10 Wild foods 5 0 Jan 122 Feb Mar Apr May June July Aug Sept Oct Nov Dec Graph 7 Seasonality of Resource Use by Common Land Areas, All Areas 80 70 Days of use 60 Roadsides & borderland Open grazing land 50 Restricted grazing land 40 Enclosed woodlots Open forests/wooded areas 30 Other community enclosures Unenclosed hillsides 20 10 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Graph 8 Seasonality of Resource Use by Common Land Areas, Excluding Unenclosed Hillsides 16 14 12 Days of use Roadsides & borderland 10 Open grazing land Restricted grazing land 8 Enclosed woodlots Open forests/wooded areas 6 Other community enclosures 4 2 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 123 regeneration. Fuelwood collection also must occur throughout the year, but there is lower resource availability during the summer period (June-August): it is wetter and woody species are greener, making useable fuelwood less accessible. Conversely, the collection of wild food and medicinal species shows a marked seasonal variation, with peaks occurring during these same months, since it is during this period when such species are most available: none of the wild foods are available yearround. Since the coffee ceremony is an integral part of Ethiopian culture and is performed on a daily basis year-round, collection of fresh grass for the ceremony is also a year-round practice, but this also peaks in June and again in September, with a slight trough in July-August. Its overall peak during the months of June through September is due to the increased availability of grass during the rainy season. Apparent differences within this period, such as in July and August, are most likely due to use during religious holidays, which are frequent during these months. Looking at the number of days of resource use for the different common land areas (Table 68A and Graph 7), it is clear that unenclosed hillsides had the greatest intensity of use which, in fact, is so much greater than in other common land areas, that it is necessary to eliminate this area and examine the remaining areas in a separate graph (Graph 8). Here it is easier to see that restricted grazing lands, forests and woodlots (both enclosed and unenclosed), and riversides and other borderlands were the least intensively used. Open grazing land and ‘other community enclosures’ fall between these areas and open hillsides in terms of intensity of use. Examining averages per household, it can once again be seen that unenclosed hillsides are used often by the vast majority of households (83%), whereas ‘other community enclosures’ were used the most intensively, but by minority (23%) of households. Grazing lands (both restricted and open) were the next most frequently used areas, by half of the households for an average of 80 days per year, followed by forests and woodlots (enclosed and unenclosed), used by a slightly higher percentage of households for an average of 53 days per year. Open and restricted grazing lands appear to be used equally. There are, however, significant differences between open forests/woodlots and enclosed community woodlots, where the latter are used by a slightly higher number of households, but much less frequently. A possible explanation for this difference is the use of woodlots for grass for both fodder and basket making, which occurs only during specific days of the year. Again, roadsides and other boundary lands are least used, by 20% of households an average of only 27 days per year. With regard to seasonality of use of common land areas, Graph 7 shows that unenclosed hillsides not only have the greatest amount of use overall, but as well the greatest use during the rainy season, reflecting the seasonality of wild foods and medicinal plant collection, as seen above. The use of riversides and other boundary lands also exhibits this seasonal pattern, again for similar reasons. Grazing lands are used somewhat more frequently during April and again in June, most likely due to the increased collection of medafe talien seeds, which occurs in April, and of other wild foods, which occurs in June. Unenclosed forests and woodlots show fairly continuous use, whereas enclosed woodlots show strong fluctuations: use is lowest in March and climbs fairly steadily until a peak in September, which possibly reflects the fact that these areas are opened briefly for grass sales. 6.E Informal Common Property Resource Access and the Case of Awelie While the formal rules regarding AE and particular species are understood and generally supported by the community as a whole, this does not mean that people always adhere to them. Violations occur and, depending on the resource and common land area, they are reported to occur quite frequently. But, before a particular use of CPR can be characterised as a violation, it has to be detected, perceived by the detector to be a violation, and reported. Once a violation has been reported, the next step is for the baito official or DA to act upon it or not. If it is acted upon, then it is up to the baito or local court to levy fines or even imprisonment as a punishment, where it still is possible that some violations go unpunished, or that punishment is less or more severe depending on various mitigating factors. Whether and how this entire process takes place depends not only on legal provisions, but as well on cultural norms, and a great deal of leeway for local decision-making is not only formally possible, but also clearly evident. Part of the explanation for this ‘leeway’ can be found in Section 2.D, which discusses monitoring systems in AE in general. However, the literature on AE in Ethiopia does not mention informal norms or rules of conduct regarding CPR access to which community member themselves adhere: rules that determine, for example, whether villagers (including AE guards) will identify a CPR use event as a formal violation or not, whether they will report it if they do perceive it to be 124 a violation, and whether officials will do anything about it. In fact, in Adiarbaetu there are at least two layers of rules regarding CPR access that are constantly interacting. Informal rules affect how formal rules are followed and implemented. In addition, in the absence of formal rules, for example around access to non-restricted species in open hillside areas, or around access to non-cultivated plants on private farmland, customary rules or ‘morals’ may still apply. Customs, religious proscriptions, social networks, and considerations relating to equity and social justice, as well as other factors, can create or influence these informal rules.111 At times, formal and informal rules might contradict and hence displace one another and, at others, they work together to form a network of access rules that govern the use of natural resources, as a form of legal pluralism. In general, such a network of access rules exists within Adiarbaetu. However, upon closer examination, access to CPR can be even more accurately characterised as a set of formal rules that are mediated by the culturally accepted notion of the ‘need for exceptions’, which is recognized even by the officials who are responsible for enforcing formal rules. The formal rules refer to AE and species restrictions. The ‘need for exceptions’ is more complex. This includes deeply culturally embedded notions about the importance of particular CPR for spiritual and physical wellbeing, and about the right or even obligation that everyone has to use certain CPR for these purposes; it also includes customary norms of access to resources such as wild food and medicinal species, even on private land, that may or may not have been embedded in the rist system but that are probably ancient and, as well, notions about rights that particular people (e.g., children, the elderly, the disabled, the poor) have inherently or as a result of their social status in the community. Such a ‘need for exceptions’ tacitly acknowledges that formal access rules discriminate against disadvantaged community members, and upholds the precept that everyone has a right to meet their subsistence needs, which has often been found to underlie customary resource tenure regimes (Howard and Nabanoga, in press). The interface between formal rules and the ‘need for exceptions’ is constantly mediated and played out in everyday life, and is most likely continuously changing. Ultimately, as Chisholm (2004) also found for his study areas in Tigray, residents and officials alike in Adiarbaetu are constantly struggling to negotiate a ‘fragile balance’ between the need to use and conserve CPR.. The existence of these ‘exceptions’ made it difficult at first to understand access issues in Adiarbaetu, since formal rules were often discussed with researchers without mention being made of these exceptions. For example, while grazing is forbidden in enclosed woodlots, researchers occasionally observed cattle grazing in them. When asked explicitly about this, officials said that the rule was ‘occasionally’ overlooked in the case of ‘younger cattle’. However, underlying these apparent contradictions, there is an acute understanding of the importance of CPR to livelihoods. While it was not possible in the short period of time available to learn about all of the exceptions (and hence informal access rules) that are at work in Adiarbaetu, it was possible to gain an understanding of some of their principles through cases.112 There is probably no other resource where the need for exceptions operates more transparently and illustratively than is the case with awelie (olive). Of all of the species on the ‘no use’ list, the prohibition against use of awelie is reportedly the most strictly enforced, since it is locally very valuable as well as endangered. According to researchers, O. europaea is a fleshy-fruited, secondary climax species of dry Afromontane forest and at present a major constituent of the rather species-poor forest relics in the [Tigray] region. Once established, the tree is drought resistant, but because of its multiple uses (e.g. durable timber, traditional ox ploughs, firewood and charcoal) both young and mature trees have been over-harvested dramatically in Ethiopia and as a result this valuable tree is now under threat of local extinction.113 111See Chapter 2 and Howard and Nabanoga (in press) for a discussion. research is needed with these exceptions, and their relation to the formal administration of violations, as a primary focus (see Chapter 7). 113M. Bart, “Dispersal and recruitment of Olea europaea ssp. cuspidata in degraded Afromontane savannah: a case study in the highlands of northern Ethiopia.” Forest, Nature and Landscape Research Division, Katholieke Universiteit Leuven. Online at http://www.kuleuven.ac.be/research/Researchdatabase/project/3E05/3E0509 17.htm. 112Further 125 In addition to the uses mentioned above, awelie is also culturally very important for medicinal and cultural uses such as etan (Box 8). Both officials and informants acknowledged that, officially, there must be no use of awelie for any purpose. In general, informants also acknowledged this rule. However, according to focus group discussants, there is ‘one exception’. Both men and women said that they are able to collect small amount of leaves and small branches for childbirth and post-partum medicinal use. Even for this purpose, they said, awelie cannot be taken freely: ‘permission is required’. Because of its importance, they continued, there is an understood exception to the rule, and guards and checkpoint officials allow small amounts to be taken. Researchers were therefore surprised to find that awelie wood products can be readily found throughout the kushet: they were observed in the markets and in piles in yards awaiting use, as well as on women’s backs (together with other restricted species) in large bundles of sticks used for fuel. When asked about fuelwood collection in particular, knowledgeable informants reported that, while it is illegal to collect awelie, use is allowed once one is within the borders of the town. Apparently, if residents are not caught collecting awelie but only using it, they are ‘safe’. This exception also explains the amount of awelie for sale at the market, the majority of which is illegally collected. When asked about this exception, officials confirmed it. It appears that this exception has arisen over time. When restrictions were first introduced, they applied to all wood taken from restricted species. Possession and use were considered to be violations and it was not necessary to be caught in the act of collecting the wood. However, residents complained because the use of awelie was too important, and insisted that some access to this species was necessary: without viable alternatives, it was not possible to stop all use. In the case of awelie, an agreement was reached between officials and residents that any wood that is already collected and has reached the boundaries of the central area can be used and sold without punishment. However, if caught with awelie wood beyond these borders, confiscation and fines would result. In similar acknowledgement to their importance to village life, restricted species that have been confiscated, including awelie, are auctioned off to villagers at certain times throughout the year. Therefore, because a species is on the restricted list, does not mean that it is not used at all. Use and collection are restricted, but are not forbidden. Community members have their own concepts of morality that affect their willingness to report violations. Researchers learned that one of the kushet’s residents is renowned for his knowledge and use of awelie. He often discussed rules affecting this species with researchers, and he grows several trees on his property in order to compensate for these restrictions. While he admitted that he had illegally collected and sold awelie in the past, he insisted that he no longer does so. However, according to officials and other residents, he is one of the most important sellers of illegally collected awelie in the kushet.114 While people say that reporting of violations against species restrictions is fairly common, he is not reported. When discussing this contradiction with another resident, she explained that he is an exception: He’s one of the poorest and he needs the income he gets from selling awelie. He has been selling for a long time. He doesn’t sell as much now as in the past. But, in the end, we benefit from this sale as well, since we have a source of awelie. But the same does not apply to others. If I see anyone else taking awelie that does not have permission, I will report them. I think others feel the same way. Although this man had not been given official permission to collect awelie (and formally cannot be given permission), community members and officials alike make an exception in his case. The results of the Household Survey (Table 75) add yet another layer to this complexity, demonstrating that residents commonly used awelie and were not hesitant to report this use to researchers: while they were told that the information that was collected about them would be confidential, their willingness to report this information must be based on something more than simply the trustworthiness of the researchers. Rather, it must be based on a de facto recognition that, while restrictions are required, the need for material and cultural well-being cannot be denied. 114When the lead researcher was discussing with one of the officials a conversation that she had had with this informant about access rules, he laughed in amusement, saying, “He breaks the rules all the time! He alone is responsible for the destruction of our awelie!” 126 The most frequent use of awelie was for fuelwood, followed by construction (together, over half of the uses), for which woody parts are required, but it is not known whether these uses implied cutting any tree parts. Use of awelie for medicinals, fumigation, and other aromatic purposes, which do involve cutting parts of the tree,115 constituted another 37% of the uses. When examining the areas which were the reported sources of awelie, it is apparent that some came from ‘open access’ areas, but most came from areas that are supposedly well-guarded against such use; enclosed woodlots and other community enclosures were the source 64% of the times that awelie was collected. If awelie is formally so heavily restricted and yet both formal and informal ‘exceptions’ are so prevalent, it is obviously very important to know which informal rules and de facto exploitation patterns apply to the other restricted species, as well as to AE, and what implications these have for CPR conservation as well as for livelihoods and for the poor. The case of awelie alone makes it clear that wild and indigenous botanical resources play a crucial role in Adiarbaetu: the de facto use of this restricted species provides needed income for poor households and allows others access to a limited amount of a locally important resource. Table 75 Use of Awelie Reported in the Household Survey and Common Land Areas where Collection was Reported USES No. % Fuelwood 13 30.2 Medicinal 9 20.9 Fodder 3 7.0 Fumigation 6 14.0 Construction 9 21.0 Aromatic 1 2.3 Boundaries 1 2.3 Other 1 2.3 Total 43 100.0 Common Areas Enclosed woodlots 5 22.7 Unenclosed forests/ woodlots 1 4.5 Other community enclosures 9 40.9 Unenclosed hillsides 7 31.8 Total 22 99.9 6.F Women, Female Headed Households, and Common Land Areas Research in Adiarbaetu demonstrates that the use of CPR is more extensive than previously reported in the AE literature: more areas are used by more people and for more purposes.116 According to the Household Survey, which compared an equal number of MH and FH’s use of CPR, while all areas were used by both types of household, FH made greater use of all areas, with the exception of unenclosed grazing land and unenclosed hillsides. Graph 9 and Table 76A shows that FH were the vast majority of users of enclosed woodlots (80%), other community enclosures (90%), restricted grazing land (67%), and roadsides and boundary areas (100%). They were also a slight majority among the users of unenclosed forests and woodlots (58%), and of unenclosed hillsides (52%), where the last category was used by 29 out of the 30 households surveyed. Open grazing lands were the only common land areas that were most used by MH, at 63% versus 38% of FH. While, as mentioned above, women were asked to report about their household’s use, they may have under-reported uses that only adult men in their households make of common land areas, especially hanging beehives and collecting wild foods to eat on-the-spot, although the latter was likely to occur outside of the kushet, and data were only collected on resources areas within the kushet.117 Therefore, the survey results might better approximate an unbiased comparison of women’s and children’s uses of CPR between FH and MH, although, as will be seen, adult male uses were reported by wives. Any missing data on adult male use would mean that MH use figures are lower than they would have been had adult males represented themselves, whereas FH barely contained adult males. However, even if the survey results are interpreted to be a more valid comparison be 115Besides leaves, small branches are also used, although the amounts taken for this purpose are quite small. the forestry literature and the ethnobotanical literature begin to encompass this scope. See, e.g., Williams, et al 2003, Vivero n.d., Asfaw 1995 and 2001, Asfaw and Tadesse 2001. Guinand and Demessa 2000. 117The use of unrestricted grazing lands was very possibly not under-reported since, as iscussed above, in other sections of the Household Survey where male household heads were the respondents, only five reported using restricted grazing land as a source of food for livestock. 116Only 127 Graph 9 Use of Common Resource Areas by Sex of Household Head MH FH 16 14 Number of uses 12 10 8 6 4 2 0 Enclosed woodlots Open forests and woodlands Other community enclosures Unenclosed hillsides Restricted grazing land Open grazing land Riversides & boundary lands tween women and children in FH and MH, it is still very clear from the data that FH use these areas substantially more often. When looking closer at how common land areas are used between household types, Table 76A shows that FH also predominate in particular use categories: in aromatics (75%), grass for coffee ceremonies (62%), medicinals (69%), wild foods (56%), and even fuelwood (57%). This predominance, however, is relative, and reflects the fact that FH use common land areas more than MH. The lower half of Table 76A shows within household distribution of reported uses, and there the distribution of uses among the categories is more similar between household types: a higher percentage of FH uses compared with MH uses are of medicinals (24% versus 15%), grass for coffee ceremonies (20% versus 16%), and aromatics (6.5% versus 3%), whereas uses that figured more prominently in MH were grazing (6% versus 1%), and construction and fodder (4% versus 1%), but these differences are not as great as those apparent in frequency of use. Table 77A presents the major use categories by sex and age of the user, without considering whether users are from MH or FH. Without considering age, females predominated in all use categories. Construction materials were only collected by females, and females predominated even in grazing and fodder collection, which may reflect reporting bias. It is therefore useful to focus on males. Males participated most strongly (30%+ of the users) in grazing and collecting grass for coffee ceremonies; somewhat (20-30%) in collecting fuelwood, medicinals, and wild foods, and little (<20%) in the collection of aromatics and fodder. Of those males who were reported to be users or collectors in common land areas, boys played only a minor role, where their participation was most common in fuelwood collection. Only one male child collected aromatics and only one grazed cattle. Women respondents did report adult male uses (most of which, again, were from MH), but these were fewer than their own uses in all use categories. It is therefore very difficult to say based upon the results of the Household Survey that there is a clear gender division of labour in CPR use in Adiarbaetu that would affect how FH use CPR. Qualitative research methods used in Adiarbaetu did, however, show that men have a more varied and extensive use of CPR than what is suggested by the Household Survey. In focus groups, in addition to primary uses of areas such as woodlots and grazing land for timber and grass, men reported using woodlots and forests to hang bee hives. All areas were seen as important for bee forage and hanging beehives was often one of the first uses that men mentioned for all common land areas. Men also reported that they collect medicinal species and wild foods as well (see below for a more detailed discussion of men’s and women’s plant use categories). Nevertheless, women and other members of MH do use CPR differently. FH’s greater use of CPR is generally explained by their lack of financial capacity to obtain these resources by other means. Many of the CPR that FH collect can be purchased at weekly markets. MH are more likely to 128 have the means to purchase CPR to supplement those they collect, particularly since collection costs a great deal of time due to resource scarcity (see below in relation to fuelwood). With less land overall than their MH counterparts, FH do not have as much access to grass from their own property, and therefore need to use common land to procure it. Since FH often make and sell tela and prepare and sell coffee, they would use grass for ceremonial purposes more often than MH, and also require more fuelwood for these purposes. Whereas in focus group discussions it was reported that wild foods are used by everyone, FH appear to make more use of wild food for longer periods throughout the year to supplement their very limited food sources. At such extreme poverty levels, FH have few options available to them other than to exploit CPR: those other options were discussed in Chapter 5, and some of them are comparatively very unattractive. While FH, and women in general, confront no formal obstacles to the use of restricted grazing land, and even appear to use them more than MH (although not for grazing), FH and women more generally are disadvantaged in comparison with MH and men. In focus groups it was reported that only women make baskets, although men sometimes harvest grass for their wives and, contrary to what was found in the survey, it is mainly men who harvest grass used for thatch. In women’s focus groups, particularly women household heads complained that grass stealing is prevalent. They explained that there were customary rules that protected women’s right to basket grass: areas where it grew were ‘for women only’, and men respected these rules; women harvested jointly and shared the grass equally. Now, one woman said: “The tabia controls it and the men will go and collect it at night. When the areas are open for cutting there is nothing left to collect. It is so frustrating it makes me want to cry.” Women in general, and female household heads in particular, are unable to go out at night unaccompanied by men for reasons of personal security, and they also need to be at home with small children. Since it is becoming increasingly difficult to find appropriate grass resources at all, grass and other materials for basket making are now found for sale at the weekly market. However, the poorest households, many of which are FH, cannot afford to purchase these materials, and are therefore dependent on AE. Despite the fact that FH appear to predominate in the use of common land areas, this does not mean that they necessarily receive greater benefits compared with their relatively richer male counterparts. In fact, FH use primary resources less than MH. For example, FH use common land areas less for grazing, construction materials, and fodder. The Household Survey shows that 80% of MH reported use of areas was for grazing, construction, and fodder. While FH are finding ways to use common lands, they are not benefiting from those primary resources (grass and timber) and enclosures that researchers and officials consider as the most important to livelihoods. FH’s apparent greater use of some of these common land areas is for ‘secondary’ resources, rather than for fodder grass or wood. Timber and grass are currently seen as high-value resources, and these are used and purchased mainly by men. Timber is used primarily for construction and ploughing tools; to make much use of it requires access to male labour. In focus group discussions, female heads reported that these resources are either too expensive, as is the case of timber from woodlots and seedlings from nurseries, or not beneficial, as is the case with grass for grazing, since they do not own livestock. 6.G Botanical Resources: A Closer Look The purpose of this section is to examine much more closely the importance of botanical resources to residents of Adiarbaetu, particularly for major livelihood uses such as fuel, food, medicine, construction, fodder, and others. Botanical resources also have vital functions beyond the material: they form an integral part of cultural identity. The Cultural Valuation Study represented an attempt not to only understand which botanical resources are important to local residents, but also why they are important. While outsiders can make assumptions about the importance of particular plant resources (e.g., timber, grass) to particular population groups, this can lead to bias, as is demonstrated throughout this report. Capturing and using emic perspectives on resource categories and their importance is critical both to quantifying CPR use and to managing CPR in ways that in fact respond to people’s needs, as well as to ensuring that incentives and rights are clear for local populations, so that they can sustainably manage these resources. Further, the importance of CPR varies according to socioeconomic and cultural characteristics of different groups within communities. In Adiarbaetu and in the highlands in general, gender divisions of labour and cultural obligations influence which botanical resources are seen to be important, the reasons for their importance, and their uses. 129 Given the degree of devegetation in Tigray, it would appear to be an unlikely place to find much use of non-cultivated plants. People continually expressed doubt that this research would uncover anything: a typical remark was, “There are no plants left in Tigray”. However, the research results proved otherwise: in Adiarbaetu alone, respondents mentioned what may be118 138 distinct species that they use with more or less frequency (Table 78A). Informants often said that “every plant is a medicinal”. If medicinal species had been specifically researched, the list would have expanded considerably. As the availability of resources has significantly decreased over the years, the amount, frequency and even use has in some instances changed, but elderly focus group discussions indicated that their importance has not. Gender roles and corresponding obligations have much influence on the species that are important to men and women, where some species are used more or less exclusively by either sex. However, of the species used by both, the uses that each makes of them are often different. During focus group discussions, men and women were initially asked to free-list what they considered to be the most important species and then to indicate why they are important (see Section 3.C.4 for details). Women created fifteen value categories, whereas men created seventeen (Table 79A). Below, the findings regarding the most important value categories are discussed. 6.G.1 Fuelwood In addition to grass, according to all indicators fuelwood constitutes the most important CPR use in the study area. As the primary source of biomass for fuel, it is essential to life in Adiarbaetu. In Table 80 it can be seen that all surveyed households used fuelwood as a cooking source, half used charcoal (which is formally illegal), more than 90% used dung (indicating the severity of fuelwood shortages in the area), and only one used kerosene. The use of dung has increased with the rapid decline in Table 80 Fuels Used and Fuelwood Sources by Sex of Household Head (n = 30) Fuel Sources MH % FH % Cooking fuel source (n = 15) (n=15) Fuelwood 15 100.0 15 100.0 Charcoal 9 60.0 7 46.7 Dung 15 100.0 13 86.7 Kerosene 1 6.7 1 6.7 How fuelwood is obtained Purchase only 0 0.0 2 13.3 Purchase and collect off-farm 11 73.3 6 40.0 Purchase and collect on- and off1 6.7 0 0.0 farm Collect off-farm only 3 20.0 7 46.7 Time walked to collect fuelwood (n = 15) (n=13) 30 to 60 minute walk 3 20.0 0 0.0 2 hour walk 5 33.3 0 0.0 3 hour walk 3 20.0 2 15.4 4 hour walk 2 13.3 3 23.1 5 hour walk 1 6.7 0 0.0 6 hour walk 1 6.7 8 61.5 Source: Household Survey 118The phrase ‘may be’ is used because, without botanical voucher collections and proper scientific identification of the plants mentioned (see Section 3.E.5), it is not possible to know whether the plants listed represent distinct species or different varieties of the same species, or the same species with different vernacular names. However, given that the research focused on more central areas of the kushet, that the population that reported species information was quite small, and that species used for medicinal purposes were always under-reported since researchers did not have time to list them all, the number of species that are used in Adiarbaetu is likely to be substantially higher. It can be assumed, however, that those species that are most important to a wider number of residents have been captured in these lists. 130 fuelwood sources and the increase in restrictions on most woody species, but otherwise fuelwood continues to be the primary source of fuel for most households. According to residents, much of the kushet was covered in thick forests as little as thirty years ago. As discussed in Section 2.D.5, FAO (n.d.) reported that households require an average of .3 ha of forested land to meet their fuel needs. With over 350 households in Adiarbaetu, a total of 105 ha, or around a ninth of the total area of the kushet, would be required to supply these needs sustainably. Many species traditionally used for fuelwood are slow growing and do not regenerate well. Awelie and other hardwood species, such as sarow (Acacia etbaica?) and chihidi (Cupressus spp.?, Juniperus procera?),119 are preferred: they burn long and hot, and generate less smoke. As was discussed above in the case of awelie, fuelwood is so essential to survival and in such high demand that collection of these restricted species still occurs frequently. Since women are the main collectors, they also run most risk of fines or imprisonment. However, as a result of deforestation and current restrictions on preferred species, fuelwood now mainly consists of smaller and softer wood such as hahote and tabub, which are small woody shrubs that regenerate quickly120 and are easy to cut Table 82 Fuelwood Users and carry. Together, these two constituted USER INDICATORS No. % nearly 50% of the species that surveyed Fuelwood collectors by sex and (n=38) households reported collecting in the past relation to head 12 months (Table 81A). In addition to these, Wives or female heads 23 60.5 miscellaneous small dried branches and Male heads 5 13.2 twigs (kirshem) are also used. While kirshem Daughters 7 18.4 must be used due to the shortage of larger Sons 3 7.9 pieces of wood, its collection is very time Division of fuelwood collection by (n=20) consuming, it does not burn hot, cannot be sex and age used to make charcoal, and produces a Only female head or wife 8 40.0 great deal of smoke, which is harmful to Only male head 0 0.0 the health of those who use it. Male head and wife 5 25.0 Table 82 shows that a little over Female head or wife and sons, 5 25.0 half of the households purchased as well as daughters collected fuelwood, but nearly a third did Male head and children 0 0.0 not purchase it at all, 70% of which were Only sons, daughters 2 10.0 FH. Nine out of ten households collect their fuelwood only off-farm, which corroboFuelwood collection areas (n=25) rates other survey findings that, while sevRestricted grazing lands 0 0.0 eral are developing tree plantations, these Other grazing lands 2 8.0 are currently too immature to yield fuelEnclosed community woodlots 2 8.0 wood. Nearly two-thirds of the MH are Unenclosed forest/woodlot areas 1 4.0 able to purchase as well as collect fuelOther community enclosures 5 20.0 wood, compared to only 40% of FH, which Unenclosed hillside slopes 15 60.0 reflects the economic differences between Riversides and other borderlands 0 0.0 the two types of households. The only two households that exclusively purchased fuelwood are FH, one of which is headed by a disabled woman. Although men will collect fuelwood, it is mainly a female task. Table 82 shows that, for 79% of all reports of fuelwood collection, females were the collectors (heads, wives, or daughters). In a 119There are various species, both native and introduced, known as chihidi. Introduced species, or at least those seen as such from villagers’ perspective, are often given the suffix talien (‘Italian’) since it is believed that the Italians introduced them. There are trees called chihidi and others called chihidi talien. Many species are distinguished the same way, such as ereh (Aloe spp.) and ereh talien (Agave spp.). It would be interesting to investigate the accuracy of emic distinctions between native and introduced species, as well as the criteria used for naming plants. For example, species known as unguleh and unguleh talien look very different and appear to have different uses. However, their fruits look similar, which is perhaps why they have similar names. 120There is a local saying about hahote that, when it is cut, it says, “You may cut me now, but I will have grown back again by the time you reach home”. 131 quarter of the cases, men collected it with their wives; in another quarter, women collected with their sons and/or daughters; in 10%, only sons or daughters collected, but in no instance did only men or men and children collect. Focus group informants said that men tend to collect larger and heavier pieces mainly as a secondary task while grazing livestock or travelling to markets, but this was not reflected in the survey data. Apparently, there has been a change in the gender division of labour in fuelwood collection. Both men and women reported that, in the past, men were primarily responsible for fuelwood procurement, since most of the species used produce hard and heavy woods where strength is needed both to cut and to carry. The relative abundance of such species also meant that they were not very difficult to find, and probably not much time was invested in procurement. However, since the use of such species is no longer permitted and they are also difficult to find, collection has become very labour-intensive and women are now the primary collectors. Although villagers never mentioned it, this shift may also have occurred because fuelwood is no longer an important source of income, and its use is now largely confined to the domestic realm. Due to the decline in fuelwood sources, particularly in areas closest to the centre which have the highest population densities, collection must be frequent and requires long distance travel by foot. Table 83 shows that the average number of collection days per household per year was 49.5, or about once per week. Two households reportedly spent more than 100 days per year (at least once every three days), four spent between 50 and 100 (once to twice a week), and a quarter reported 48 days (about once per week). Only five households reported collecting on considerably fewer days, and the Table 83 Average Days of Use by Major Use Category and by Common Land Area No. Average Variation Major Use Category House% Days/HH/ from holds Year Avg. Fuelwood Medicinals Wild foods Coffee ceremony Fodder Cattle grazing Aromatics Basket thatch Construction Totals Common Resource Area Restricted Grazing land Other Grazing Land Enclosed woodlots Unenclosed forests/woodlots Unenclosed hillsides Other comm. Enclosures Riversides and borderlands Totals 26 18 21 9 3 4 7 5 3 96 27.1 18.8 21.9 9.4 3.1 4.2 7.3 5.2 3.1 100.0 76 45 79 104 172 109 36 31 15 74.1 1.9 -29.1 4.9 29.9 97.9 34.9 -38.1 -43.1 -59.1 0.0 12 12 10 8 15.0 15.0 12.5 10.0 49.0 51.0 16.0 41.0 -12.0 -10.0 -45.0 -20.0 25 7 6 80 31.3 8.8 7.5 100.0 113.0 130.0 27.0 61.0 52.0 69.0 -34.0 0.0 Source: Household Survey. minimum reported was 24. The amount of time that it takes to collect fuelwood per collection event (Table 80) is more alarming: only 10% collected it within an hour’s walk from home, and nearly a third walked six hours or more. Only 7% of MH reported having to walk six hours or more, compared with 61% of FH. Economic differences certainly contribute to this inequality, since most MH are able to supplement collection with purchased fuelwood, allowing them to either reduce the average time spent collecting or to confine collection to areas that are nearer to home. Further, MH have greater access to other sources of fuel such as dung, which decreases their dependence on fuelwood, 132 whereas most FH do not own livestock. The amount of time spent in collection means less time available for other subsistence and income generating activities. Other factors contributing to the distance travelled include the amount of fuelwood needed for each household, which differs depending upon the number of household members and on the presence of any livelihood activities (such as tela production) that require substantial amounts of energy. It appears that fuelwood is no longer an important source of income for many of the poor, as it was once reported to be. People who have no other options can seek permission to collect dried pieces of restricted species on the ‘with permission only’ list for both consumption and sale. However, according to officials, only four people in the tabia have been granted such permission. This number is small for three reasons: first, permission is not readily granted, so people do not think they will receive it if they apply; second, the process of seeking permission is time-consuming and bureaucratic and, third, there is social stigma attached to this permission since it is only granted to the ‘poorest of the poor’. Informants said that few, if any, FH apply for permission, because collecting fuelwood for sale implies travelling long distances and carrying heavy loads. The wood collected is difficult to cut and heavy.121 Few residents, much less poor FH, have pack animals or sufficient male labour to assist them, and many cannot afford the time to travel the distances required. Since fear of rape is always present, women will usually not travel alone, and regular collection for sale would require that they travel together. However, due to dire economic need, there are women from FH who collect and sell fuelwood illegally. Of the two women who spoke to researchers about their illegal collection activities, both indicated that they had recently stopped selling fuelwood in favour of food-for-work activities, saying that the risk was too high and the distances too great. Both women were very aware of the restrictions and resulting fines, and both said that they had been caught many times. However, they were never fined or imprisoned since the size of the pieces that they were able to collect, and hence the total amount, was much less compared to what men take. Again it appears that officials recognize the ‘need for exceptions’. Much more research is needed into the dynamics and implications of fuelwood use in Tigray for livelihoods, resource conservation, poverty and destitution: it appears that no such study exists. Both the qualitative and quantitative research in Adiarbaetu showed that FH not only depend more on common land areas for fuelwood, but also spend a greater amount of their time in its procurement, which has obvious negative implications for their livelihoods and welfare. 6.G.2 Wild Food Plants Wild food was an important value category for both men and women (Table 79A) and, in free-listing exercises, women mentioned four species, whereas men mentioned ten. During the Cultural Valuation Study, where informants agreed on the twenty-five most important species, men listed nine wild food species, whereas women only mentioned one122 (hahote) which was first cited for uses other than wild food. This is not to say that women are not aware of wild foods or do not use them since, when asked specifically about them, they were able to list many. Women collect, purchase, and prepare wild foods to eat as part of their households’ regular meals (Table 84A). Men, on the other hand, more often travel long distances and use wild foods for sustenance during their travels. Thus, wild foods become an important food category for them, one that meets their dietary needs when little else can. Women do not travel as often or as far,123 and do not depend on wild foods in the same way. While men reported that they eat most of this wild food on-the-spot, they occasionally bring some home for their families.124 The wild foods that women mentioned tend to be found within the kushet or are 121Pieces sold in the market tend to be larger and of harder and heavier wood. Most of these species are no longer found within the kushet.. 122The other three wild foods that women mentioned when directly asked about them were kuenti, hama shiro, and hamli. 123While women do not confront cultural restrictions on travel, they generally only move within the kushet, always in the company of other people due to the high incidence of rape. Men often travel outside of the kushet for work and to attend markets in other tabias and in cities. As a result, men have greater access to different areas and elevations, as well as to the different species found in these locations. 124A very interesting study (Hildebrand 2003) on wild yam domestication in southern Ethiopia also found that this was the reason that men, rather than women, were most likely to have domesticated yams. 133 readily available at the market, and hence women usually procure them while performing other tasks, such as collecting or purchasing fuelwood and medicinals. Like men, children often eat wild food onthe-spot while herding or carrying out other tasks for their parents, and occasionally they bring them home. Despite the decrease in availability and hence use in the area, children still have much knowledge about wild foods. Research in Adiarbaetu showed that wild food consumption is not clearly associated with famine (e.g. times of extreme food shortage). Famine food was originally anticipated as category distinct from other wild foods. However, it became clear during the focus group discussions that none of the wild foods were used solely as famine foods. The only distinction that participants made was with regard to the amount eaten: during famine or extreme food shortage, wild foods make up a larger portion of their diets. The amount that people normally consume depends on personal preferences. For example, a debate ensued among respondents, where some said that they only eat hama shiro when needed, whereas others said that they eat it all the time because they enjoy it. It was also reported that all residents, regardless of wealth status, eat wild foods. The Household Survey included a Food Frequency Module (FFM) that attempted to capture the frequency with which different foods are typically served and eaten. Respondents were prompted for their consumption of a long list of foods that was derived from the ERHS crop list supplemented by other ethnobotanical and food studies carried out in the region (see Chapter 3, FN 16). Several of the foods on the list are wild foods. However, respondents were not prompted for wild plants that they consume that were not on the list, but it was found that many other wild botanicals are consumed in Adiarbaetu: Table 85A reports a total of 30 species that were mentioned by different respondents in different contexts. The results of the FFM for wild foods are presented in Table 84A. Here it can be seen that the most frequently consumed wild food is beles (fruit), which all households eat, although 14 (nearly half) reported consuming it ‘only rarely’. Most households (37%) reported that they consume it 20-50 times in the summer months when the fruit is available. While beles is essential to many households during this period of annual food shortage, it is also consumed in large amounts because residents enjoy the taste. A recent study also reported that Opuntia ficus-indica has much nutritional value as a rich natural source of edible polyunsaturated oil containing essential fatty acids (Ennouri et al 2006). The second most commonly consumed wild food in the FFM was mekmoko (Rumex abyssinicus?), the shoots and leaves of which are widely used as a vegetable. There is no information on the nutritional composition of the plant, but it is probably comparable to garden sorrel (Rumex acetosa L.).125 Wild greens, known as hamli talien (probably Amaranthus spp.), are also reported to be widely eaten during the rainy season, although in the survey only one household reported eating it once per week. The Ethiopia Famine Field Guide126 reports that the leaves of Amaranthus spp. are high in vitamin A and in protein (27.8%); the fresh leaves contain higher quantities of both calcium and phosphorus than cabbage. As the above examples indicate, in addition to being eaten regularly, wild foods have an important nutritional role in the regular diet as well as a critical role during periods of food shortage. There is very little that is known about the consumption or nutritional value of wild food plants in Ethiopia, but there is an increasing amount of research on this topic, such as that entailed in the African Wild Harvest’s Pilot Project (see Kew Royal Botanical Gardens n.d.). Several households reported in the CPR module that they collected wild food plants but didn’t report that they consumed them in the FFM. Table 86A lists all uses of all species that were reported in the CPR module, as well as total number of uses per use category. Wild food plants were reported 35 times, for around 10% of all reported uses of all species, which is lower than what was captured in the FFM. However, several of the species mentioned in the FFM were purchased at local markets rather than collected: only two of the species reported in the latter were common to the CPR data set (beles and hamli talien). Further, it is almost certain that these reports did not include consumption of wild foods on-the-spot: this type of consumption is notoriously under-reported, yet it appears to be frequent in Adiarbaetu. Table 87 shows that wild foods were collected by nearly two-thirds of all households, and that slightly over half of these households were FH. Females were again the main collectors (83%). Sons and daughters represented about 29% of the collectors, and the rest were heads or their wives. 125PROTA 126See 134 (Plant Resources of Tropical Africa) Online Database. See http://www.prota.org/ uk/About+Prota. Guinand and Lemessa, under websites in the references. Table 87 Users of Wild Food Species USER INDICATORS No. Wild food collectors by sex and (n=41) relation to head Males 7 Females 34 Heads 20 Wives 9 Daughters/Sons 12 Division of fuelwood collection by age Adult 34 Child 7 Proportion of HH using wild food species Total HH collecting wild foods as 19 % of surveyed HH Total FH collecting wild foods as 10 % of all HH collecting wild foods Table 83 shows that wild foods were collected with considerable frequency (an average of 79 days per year), with great variation evident: between 360 days and only one day. Especially the col17.1 lection of beles accounted for a substantial 82.9 number of days. It should be remembered 48.8 that beles is multi-purpose, since its leaves 22.0 also supply an important source of fodder 29.2 for livestock, and four households reported simultaneously collecting beles for fodder. While the vast majority of house82.9 holds in Adiarbaetu experience food 17.1 shortages every year, FH are often hardest hit, and local officials in the project area have reported that their household mem63.3 bers make up about 80% of all malnutrition cases in the area.127 52.6 However, environmental degradation, as well as drought, has led to a decrease in wild food consumption. Informants reported that many of the wild food species that were consumed in the past are now difficult to find. Species that were available year-round are now only found seasonally and in fewer numbers. Men reported that their use of wild foods during travel has changed significantly: “We used to use wild food more in the past than now. Now we are forced to eat injera and bread more than our wild food. Our eating habits have changed a lot because of the drought. Wild food is now scarce” Women similarly mentioned how wild foods used to play a much larger role in their diets: “We used to have a great deal of wild food, even in our mother’s time. Lots of aagam lahami, hama shiro, and aagam. We used to collect a lot of these plants when we were children. Our mothers used to have these as their main meal, particularly during etan. They wouldn’t eat anything else.” 6.G.3 % Medicinals Residents of Adiarbaetu principally rely on traditional medicine for their health care. Although very recently a modern health centre was established in the kushet (see Section 4.E), due to cultural preference as well as continued lack of facilities in the formal system, traditional medicine will most likely provide most health care for residents in the future as well, so that the availability of, and access to, medicinal plant species is of great importance (see Section 2.D.5). While residents are beginning to use the new health facilities, traditional medicine is the first line treatment for almost all health problems (FAO n.d.). Medicinal plant species are therefore crucial botanical resources for the residents of Adiarbaetu, and in the Household Survey they were consistently in the top three of all use categories for all common land areas (Table 67A). The gendered division of labour and knowledge with respect to CPR is most pronounced in relation to medicinal species. Women are both the lay health practitioners who treat the members of their households for common illnesses, as well as traditional health care practitioners to whom people turn when lay knowledge is insufficient. During free-listing exercises, men did not mention a single plant for medicinal use, whereas almost all of the species that women listed were used for medicinal purposes, and many were solely for medicinal use. Women used several categories to classify medicinal plants (women’s health, children’s health, and general use),128 whereas men used the general category ‘medicine’. When discussing species whose use was common to both men and women, women often mentioned medicinal uses of those species, whereas men did not. All of the species that women mentioned as ‘important’ to them that were not also mentioned by men were medicinals. Approxi127Reported in one of the PRA exercises carried out in the FAO Project area under auspices of the project. had more categories for medicinals but were asked to narrow them down to three for the Cultural Valuation Study. 128Women 135 mately a quarter of the species mentioned by women are used only for medicine. However, this does not mean that men do not have knowledge of medicinal species, and men can also be traditional healers. When asked directly, they were able to list many medicinal species and discuss their applications. Both men and women said that “everyone knows at least a little” about herbal medicine, but both men and women acknowledged that women know more, and are primarily responsible for health care. Some women are more knowledgeable than others and are known as specialists for certain ailments, such as eye infections, skin diseases, etc. Women’s knowledge and use of medicinal species means they use more common land areas year-round. Medicinal species can be taken from any location at any time. When collecting them from AE, permission is reportedly needed from the guards, but there is no limit to the amount that can be taken and, de facto, no species restrictions apply. Therefore, informants indicated, while men often do not use AE when they are enclosed, women use these them to collect medicinals. However, Table 67A shows that unenclosed areas are used much more for medicinal plant collection: only 23% of the use events were reported to have taken place in AEs. While churchyard enclosures were not included in the Household Survey, informants indicated that women also frequently use them to collect medicinal species. Table 86A shows that there were 75 reports of medicinal species use involving 16 species across different common land areas, which constitutes 20% of all reported species uses. Table 88A lists all medicinal plants named by residents during the entire research period, which total 35. All informants and even some officials said that there are restrictions on the practice of traditional medicine. According to Teshome-Bahiru (citing Teshome 2000), the Derg Regime, in an effort to keep ‘false healers’ (people with no traditional health knowledge who misrepresent themselves to clients as healers) from operating, instead persecuted all healers, and particularly spiritual healer. In rural areas, PAs were charged with taking draconian measures against them. “Furthermore, considerable numbers of healers were put behind bars for conducting false healing which was considered by Marxist cadres as vestiges of the old feudal regime” (Ibid: 45). After this government fell, he states, the successive government has tolerated herbalists and even given them a ‘warm welcome’, although the Ethiopian Orthodox Church still condemns the use of spiritual healers. This explains why there is a pervasive belief in Adiarbaetu that traditional medical practice is illegal. During focus group discussions, women said that they have been negatively affected by the ‘restrictions’: where they once earned an income as practitioners, they said they now do not, for fear of incurring heavy fines or imprisonment. Despite these complaints, traditional medicine was widely practiced, albeit in an atmosphere of fear. While medicinal plant collection is prevalent, the majority is for personal use. Medicinal species can be found in the weekly market, but the species sold are usually those that are not found within the kushet. Environmental degradation has also affected the diversity and populations of medicinal species within the area. When asked about the whether medicinal species could be sold as an income generating activity, women heads said that it would require to much time to travel the distances required to find the species needed in large enough quantities. It is also believed that restrictions on traditional medicine extend to the sale of medicinal species. 6.G.4 Fodder As discussed in Section 6.C.1, fodder of all kinds, and especially grass, is exceptionally important to livelihoods in Adiarbaetu. Grass is the preferred fodder for all livestock. However, as discussed in chapters 2 and 4, grass and grazing meet only a small proportion of all livestock feed requirements in Tigray and in the FAO project area. As a result, many other types of fodder are used to supplement grass, especially during the rainy season, when most grazing areas are enclosed. Supplemental fodders include crop residues (see Section 5.A.4), beles pads, and hatela, the remnants from tela production. During focus group discussions, men mentioned other species that provide fodder, including leaves from awelie, hahote, tehatses (Dodonea angustifolia?) and dander (Echinops spp., Carduus spp.?). Since the responsibility for feeding livestock primarily falls to men and boys, the knowledge of fodder species is also gender differentiated. Nearly 45% of the species mentioned by men during the free-listing exercise are used for fodder, whereas women mentioned none. When asked directly about fodder species, women were able to list some, such as beles and grass, and they also mentioned the use of hatela since it is they who produce tela and give or sell it to men or exchange it for ploughing services. Men’s knowledge is more diverse, apparent in the fact that they listed more species, as 136 well as species that are eaten only by certain animals, such as camels and donkeys. Such gendered knowledge also in part explains the lack of fodder collection reported in the Household Survey CPR module. Fodder collection was only reported by four households, and it is highly likely that use of fodder other than beles was under-reported. Another reason for under-reporting is that fodder species are often used on-the-spot rather than collected and, as mentioned earlier in reference to wild foods, it appears that respondents reported more resource extraction activities than resource use activities that occur wholly within common land areas. Since many households also grow beles on their property, there is less need to collect it from the wild. Grass that is collected from woodlots on a cut-and-carry basis was also probably not reported: it doesn’t appear that Household Survey respondents reported the use of resources for which they had to pay. The depletion of fodder resources is seen as a serious issue in the kushet. There is a constant struggle between the need to preserve such resources and the need to feed livestock. Alternatives and solutions are always being investigated. Heads of the tree nursery said that they are always experimenting with new species that are fast growing and can be used as fodder. Despite these efforts, many men and women reported having lost livestock to hunger and disease due to insufficient high quality fodder. Oxen are essential to agricultural production, and are expensive and not easy to replace. Many residents take out credit to purchase oxen that they must repay even if oxen die or cannot work. These uncertainties often prevent the poorest households from investing in livestock. Development agents present livestock fattening as an income generating option for the poor and particularly for FH, but several women heads said that this is not a viable option for them because of limited fodder resources and the subsequent high mortality that results. However, income is earned from the sale of fodder, including crop residues and beles pads. Women, particularly those from FH, justified their engagement in tela production, which is very time-consuming and where tela sales themselves do not bring in significant income, by the extra money they earn selling hatela for fodder. 6.G.5 Construction Materials While the use of stone is increasing, particularly for fencing, botanical species and mud still represent the primary materials used in most structures in the kushet. Unlike other areas in Tigray, it is rare to see any homes built from stone in the study area. The widespread use of wood for construction contributed to deforestation in the area. Species restrictions now curb this use, and AE can now provide construction timber. People are still allowed to cut some species for construction purposes if they obtain permission. But, as mentioned earlier, several factors limit this. Those wanting wood to construct extra rooms or other ‘unnecessary’ structures rarely succeed in obtaining permission to cut trees, especially those on the restricted list. As a result, Eucalyptus spp. now predominate as a source of timber, although it probably was little used for construction in the past. Some residents use trees from their own land for this purpose, but most must purchase it from enclosed woodlots. Its importance is based primarily on availability. Men mentioned many more native species, such as awelie and chenedoge (Otostegia integrifolia?), which they prefer for their strength and durability but, for reasons mentioned earlier, these species cannot be used for construction which, more than any other use, involves harvesting entire trees. In the Household Survey, the use of common land areas as sources of construction material other than grass for thatch was barely reported (only four ‘use events’ were recorded, Table 67A), although this certainly doesn’t mean that such use doesn’t occur on a larger scale, both because men’s uses were probably under-reported and because the sample size was so small. It is probable that few people included in their use reports instances when they purchased such materials from AE, either because sales have not occurred in the past year, or because respondents were unaware that these also constitute ‘uses’ of common land areas. Construction is primarily a male task. Regardless, women and men both mentioned species used for construction (six and ten species, respectively). While in theory FH have equal access to construction materials, women heads reported many times that they are unable to purchase construction materials from woodlots since they cannot afford them. There are other reasons that trees used for construction materials are not used equally by all members of the community. Many of the poorest, including FH, rent their homes, and either are not allowed to make improvements or, if they are allowed, may not have the incentive to do so if this involves investments, since such improvements could raise the value of the property and hence the likelihood of eviction. 137 6.G.6 Ploughing and Other Tools In a plough-based farming system, ploughs are essential to survival, not only because ploughing can’t occur without them, but also because the efficacy of ploughing has an effect on crop yields. Due to the division of labour and the cultural sanction against women ploughing, is not surprising only men mentioned species used to make ploughs (a total of eight). Changes in the use of botanical resources for making ploughs and other tools provide insights into how resource degradation and access restrictions directly affect men. Awelie is by far the most preferred wood for ploughshares and, as discussed above, it is on the ‘no use’ list. Other species mentioned, such as kan’cha (unknown) and roweh (Grewia bicolour?), are important for particular (often smaller) plough parts. However, since species restrictions and resource depletion restricted the use of awelie, men now have to rely on other species, which they say do not serve as well. Eucalyptus is the most commonly used, but it is lighter and not as durable. Because of the difference in weight, men said that ploughshares made from it do not plough well, since they cannot dig deep enough. Where ploughshares made from awelie might plough a field in only one round, those made with eucalyptus require two or three rounds to complete a field, and the blade must be repeatedly sharpened. Ploughshares made from eucalyptus also must be replaced more often, and most of the wood for this must be purchased. While women did not mention species used to make ploughing tools, this does not mean that they are unaware of them. In general, they mentioned species used for construction and ploughing as well as other tools for household use, such as mortars and pestles. Mortars and pestles are tools of particular importance to women since they are essential for cooking, especially to make berbere and to grind coffee. Smaller mortars are used to pound coffee, whereas larger mortars and pestles, often carved directly out of tree trunks in the yard, are made to grind berbere and other spices. Plants used for household tools include agee-era (Agave americana?), ashu-ena (Phoenix reclinata?), eucalyptus, daro (Ficus vasta?), shemfa (Ficus sur?), tabub, tambuk (Croton macrostachyus Hochst. ex Del.?) and thehetses (Table 78A). 6.G.7 Bee Forage Tigray is renowned in Ethiopia for its flavourful honey and, while most keep hives for their own consumption, bee keeping is also a relatively lucrative income generation activity: one kilogram sells for around 25 birr, and more can be earned for the purest quality or for those flavoured by specific bee forage species. Tedj is an often-consumed alcoholic beverage that is made from honey (a type of honey mead), which is both consumed and sold. As a result, bee forage is thought to be very important. Bee forage was an important plant value category for men that women didn’t mention, again because of the gender division of labour. While there are no sanctions that prevent women from keeping bees, it is traditionally seen as a male task because, men said, traditional hives are heavy and are hung from tall trees, so it is difficult for women to hang them and to extract honey. Men have excellent knowledge of the species that bees prefer and that make the best honey, where criteria are taste, clarity, and health benefits. Men hang beehives at home as well as in communal forests and woodlots. The decline in plant resources has also led to a decline in honey production and in quality, as well as price increases. Honey was once a staple for everyone in the region, but it is now a luxury for most people. While there are many species used by bees, the most important that men reported were girbia (Monthecium glandosum?), tabub, shua arni (Leucas abyssinica?), acachea (Acacia spp.?), and tahoge (Cynodon dactylon?). Men also mentioned one species that is used to make hives (agee-era). While men often reported using bee forage in focus group discussions, it was not reported in the Household Survey CPR module, which is probably due both to the gender bias mentioned earlier and because respondents tended not to report uses made of common land areas (other than grazing) that do not involve extraction. 6.G.8 Religious and Other Cultural Uses The use of plants for religious, ritualistic, and other cultural purposes is an integral part of everyday life in Adiarbaetu and in Ethiopia in general. Their strong associations with religious observances and piety, hospitality and neighbourliness, purity, cleanliness, and general well-being, means that their use is both culturally obliged and a strong indicator of identity and social status. Both men and 138 women are very aware of the importance of botanical species for these purposes and value them highly. Box sexes mentioned six categories of cultural and cosmological importance, where men mentioned sixteen (65% of the 25 species that men considered to be most important) and women mentioned twenty (80% of the most important). While the amount and type of species used for these purposes may have changed over time due to resource depletion and restrictions, other than in a few cases their importance and use appear not to have changed. The importance of emic categories becomes very clear when exploring cultural and cosmological uses. The categories delineated include ceremonial, etan, religious, evil eye, staff or walking sticks, and cosmetics. A few of these are presented below as illustrations. Ceremonial Uses Plants play a central role in coffee and other ceremonies, such as feasts and weddings, as decorations and aromatics. While coffee and wedding ceremonies are the most common, such species are used for any celebration, often covering the floor and framing doorways. Grass, toko berbere (Schinus molle?), baharzef, and chihidi are most commonly used although, if necessary, anything green will be used, especially if it is aromatic. Coffee ceremonies (Box 7) are an essential and culturally necessary part of daily life in Ethiopia and in Adiarbaetu, and grass collection for coffee ceremonies accounted for a high percentage of all use events reported. Grass or other green vegetation is used to cover the area of the floor where coffee is made. Aromatic species and resins are burned throughout the ceremony. Some of the species that are burned Box 7 The Coffee Ceremony are found locally, while others come from lower elevations and regions, Coffee (Coffea arabica), locally known as buna, is an and are either brought from these essential part of Ethiopian social and cultural life. areas by family or friends, or are While it is possible to get Italian style macchiatos in purchased at the market. Grass is larger cities, coffee is still primarily consumed in a the most preferred vegetation for traditional coffee ceremony. The ceremony is considcoffee ceremonies and, since the ered to be a symbol of friendship and respect. Taking ceremony is performed up to three place three times a day, it includes cleaning, roasting, times a day and only greens can be and pounding coffee, as well as brewing it in a tradiused, people expend a great deal of tional ceramic pot. The brewing area is decorated energy collecting it for this purpose. with aromatic grasses and other species, and incense Table 67A shows that grass is burned. Each person is given three servings of cofcollection for coffee ceremonies confee, each of which is slightly weaker than the one bestituted 14% of all uses made of diffore, and the third is considered as a blessing. A ferent common land areas, where snack, called kola (traditionally, roasted grain or legrestricted and open grazing land umes, nowadays perhaps popcorn), accompanies the were the source in more than half of coffee. The ceremony is traditionally preformed by the events, and unenclosed hillsides, women, however occasionally men (especially single and riversides and other boundary men) perform it. Coffee ceremonies are an important areas, were the sources in 22% of the social event within the family and the kushet: a time cases, respectively. Table 83 shows to discuss family and community affairs, politics, and that a total of nine households reto gossip. ported collecting grass for coffee ceremonies, and these households collected on average 104 days per year, indicating that the frequency of grass collection is indeed very high. Table 77A shows that it was mainly women (70%) who collected grass for coffee ceremonies, although men also frequently performed this task (30%); no children were involved. While it is increasingly difficult to find abundant grass, its ceremonial role in daily life is still strong. People also purchase grass and occasionally other greenery if they have the resources to do so or if it cannot be found, especially for important ceremonies, such as Saint’s Day celebrations, coffee ceremonies for guests, and religious holidays. As a result, grass sales can provide some additional income, particularly to households who have access to it on their property. 139 Fumigants and Etan One of the most important cultural uses of wild plants is for etan, which is both a ritual and the name for the group of plants that are used in it.129 Box 8 describes the ritual and its purposes. Etan is still important despite reported restrictions that have arisen that limit or prohibit its practice. Informants said that “it is now forbidden” for young women to perform etan at all, and therefore the traditional rite of passage described in Box 8 is apparently no longer practiced. However, older women’s practice and use of etan is “still allowed.” While some women said that it is even forbidden for married and older women to use etan, the prevalence and openness of its use leads one to question whether there really are formal restrictions, or whether this is related to restrictions on the use of awelie for etan, or on misunderstandings of proclamations such as those reported above for medicinal plants. No literature was found on the use of etan, but it was reported on the internet that its use is prohibited by the Ethiopian Orthodox Church because it is seen as ‘magical’ and hence pagan. Due to the perceived restrictions, fewer women now use etan, particularly younger women, and many older women expressed a fear that the tradition would be lost. Men said that, “The complexion of older women is better [they are more beautiful] because they had more rest and used etan as much as they wanted. The youngsters today are forced to toil. They are not allowed to use etan, so they are not as beautiful and healthy as before.” In the Household Survey, while collection of aromatics was reported, this was infrequent (five percent of all use events). However, due to the percepBox 8 The Importance of Etan tions about restrictions and limited availability of the species used, etan The etan ritual consists of fumigating the body with are also purchased from the weekly smoke from burning aromatic species. While its primary market. Therefore, overall the practice purpose is vaginal purification, women sit over the burnappears to be strong, at least among ing species enclosed in blankets, and as a result the entire the older generation. body is fumigated. Etan is used both for health and The species used for etan have beauty. Women and men said that fumigation makes reportedly changed. Where awelie was their skin “brighter” and keeps the body clean and smelland still is preferred, resource restricing nice. While all of this contributes to overall feelings of tions forbid its use unless it is grown good health and beauty, as a vaginal fumigant, etan’s on private property. In the past it was most important function is for reproductive health, and it used alone or with other species chois considered to be particularly important during pregsen for their aromatic and medicinal nancy and after childbirth. Traditionally, young girls beproperties, such as kulio (Euclea divigan etan at menses, around the age of twelve. At this norum?) and cher’ncha (Calpurnia time, as a right of passage, they were expected to fumiaurea?). However, now any aromatic gate themselves for six to nine months. A girl was not species is used, with the exception of allowed to leave the house during this period, and would Eucalyptus spp. and Schinus molle. One sit and fumigates herself twice a day. All of her female of the most popular is b’harer (unneighbours would join her and keep her company, “so known), which is also burned at cofshe is not bored”. Girls who did not follow this ritual fee ceremonies. It is an aromatic plant were considered to be dirty and an insult and embarfound at lower elevations that is sold rassment to their families. After this ritual period was almost exclusively by women in the completed, the ‘women’ who emerged could perform etan weekly market. It is either collected as desired, except during fasting periods. During fasting, from the wild or cultivated in homeit is prohibited to perform etan and to use cosmetics such gardens. While etan is the most prevaas sasula (Impatiens tinctoria), which is used to tattoo lent use of fumigants, they also have hands and feet in a manner similar to henna (Lawsonia other uses. Women mentioned some inermis). that are used in food and drink preparation, particularly to disinfect and flavour tela pots before brewing. Fumigants, such as sheila-en (Cadia purpurea?), are also used to disinfect milk containers and to preserve milk. Women also mentioned using sar’saro (Silene macroso- is also a common name in Amharic for frankincense (Boswellia spp.), but in this part of Tigray it is the general term used for fumigants and for the ritual described in Box 8. 129Etan 140 len?) as a fumigant to ward off snakes and spiders. However, women in Adiarbaetu do not sell it because the distance that one has to travel to collect it is too great. 6.H Common Property, Botanical Resources and Livelihoods Research in Adiarbaetu clearly demonstrates that botanical resources are a very important constituent of livelihoods, providing fuel, timber, food for humans and livestock, medicines, and a multitude of other resources that are considered to be essential to physical and spiritual well-being. Measures to regenerate and protect these resources have led to increases in some, but this has been accompanied by greater restrictions on their use. In such a context, it is crucially important to understand how these resources are used, what role they play in livelihoods, how conservation measures affect different groups of people, and how residents themselves strategize to maintain the ‘fragile balance’ between exploitation and conservation, since both are in their interests. The understanding of these relations is needed to create stronger links between poverty alleviation and resource management, so that both livelihoods and resource use become sustainable. An example from Adiarbaetu that illustrates this need is related to the introduction of new species that are intended to substitute for restricted indigenous species. If this substitution is to be successful, new species need to fulfil the material, cultural, and ecological functions of the indigenous species that they are meant to substitute. Considering only one such case in Adiarbaetu, while Eucalyptus spp. is now used as a source of wood for ploughs, it is far inferior to awelie and cannot fulfil the many other roles that O. europaea ssp. cuspidata plays both materially and culturally, such as for fuel and for etan. As a result, illegal use of this severely threatened tree species abounds, and informal access rules have emerged that represent a negotiated local strategy to permit both use and conservation of the species. However, it is not known to what extent the combination of such informal and formal access rules endanger or protect the species, nor for how long such an arrangement will persist, although these are very important questions. Botanical resources are both threatened and protected in part according to their importance to the populations that use and manage them. Policies in Tigray and in Ethiopia as a whole have been formulated on the basis of assumptions about the importance of certain species and certain resources to all rural people. The ‘importance’ of different botanical resources is, however, relative, subjective, multidimensional, variable over time and across social groups, difficult to quantify, and often locally determined. While the amount of use of a particular species is an indicator of its importance, factors such as the type of use (such as religious and cultural uses, for own consumption or for sale), and the sex and economic status of the user, also influence importance. As a result, it is essential to understand botanical resource use from both a quantitative and qualitative perspective. This research has attempted to determine importance using both types of indicators, and the results can be used to determine local priorities that help to achieve both a more effective and equitable balance between CPR conservation and exploitation. It has been shown in this chapter that grass and grazing lands, and timber and woodlots, are only part of the CPR resources that are vital to livelihoods, and yet apparently the remaining CPR have barely been considered by policy makers and researchers alike. Quantitatively, according to the Household Survey, the five most important botanical resources in Adiarbaetu are hahote, grass, tabub, beles, and baharzef and, although its use was ranked in sixth place, considering that it is a species on the ‘no use’ list, awelie (Table 89A). The importance of these species is not surprising since, together, they span all of the major emic use categories: fuelwood, fodder, food, construction materials, and ceremonial and cultural uses. Many of these species are multi-purpose and fulfil several of these needs simultaneously. While they are obviously very important species, the use of at least four of them is greater than that of other important species primarily because of their availability: villagers often indicated that they prefer to use other species, but cannot. Several of the most used species are exotics that were introduced because they are fast growing, drought resistant, and multipurpose, but not because they are ideal substitutes for those species that people have most valued and used. When assessing importance from a more qualitative perspective, many more ‘most important’ species come to light and, if these species are not used more, it is because scarcity or restrictions prohibits it. Two of these species can be found on the restricted species list (Table 66). It can be supposed that, if use of the other species in this table was reported by even one household in the CPR module, then the species is used by a far greater number of people across the kushet which consists of 350 families. 141 Still, the question of species’ importance in Adiarbaetu is not yet completely answered. Table 90 provides the consensus about the most important plants that was reached in the Cultural Valuation Study with separate men’s and women’s focus groups. The items in the table are listed in the same order in which the participants set them forth, so it also gives an indication of the salience of these species. Table 90 Important Species According to the Cultural Valuation Study MEN WOMEN Species Uses Species Uses Baharzef Awelie Etan, fuelwood, conConstruction, tools, fuelwood struction, medicine Awelie Construction, tools, bee Tambugh Medicine, abortifacient, forage, fuelwood, fodpestles der, etan, toothbrush Chenedoge Atush Construction, fuelMedicine – children, wood, charcoal, tools, yogurt preparation fencing Toko berbere Unguleh Pestles, bee forage, Medicine, evil eye construction, tools, Fuelwood Aagam Rambo-rambo Bee forage, wild food, Evil Eye, medicine fencing, firewood Moondaha Itsmeskel Wild food Medicine – skin and syphilis Tabub Tahatses Bee forage, brooms, Cosmetic, fuelwood, fuelwood, terracing construction, fencing, food preparation Hahote Sheila-en Fodder, wild food, Fencing, construction, Fencing fumigant –milk pots Dander Donkey/camel fodder, Hahote Fuelwood, etan, wild bee forage food Ar-aro Tabub Food (steamed bread) Fuelwood, bee forage, medicine This table illustrates the importance of emic categories of resource use when considering conservation priorities, the ‘need for exceptions’, and the possible means to strike balances between livelihoods and environmental objectives. While policy makers and other outsiders might consider that local plant medicinals, fumigants, and materials for making pestles are not essential resources, telling women in Adiarbaetu this is equivalent to telling the average northern European that aspirin and antibiotics are not essential, that apples and oranges are luxury items, that bathing is not important, and that they can very well do without gas and electricity, or go ahead and find them on their own. Throughout this chapter, it has been shown that gender relations nearly always significantly influence the importance and use of different botanical resources, so it is clear that conservation measures have specific implications for each sex. Further, as discussed in chapters 2 and 5, access to assets combines with household formation and dissolution dynamics that are strongly influenced by gender relations to create specific groups of individuals and households that can be structurally characterized as ‘underprivileged’, ‘marginalized’, or ‘excluded’, and where, in the case of the overall context of hardship in the highlands, these categories mean literally ‘destitute’. For these ‘poorest of the poor’, not only are CPR in general vital, but certain species in particular are vital. Survey results indicated that almost all botanical resources are used more by FH with the exception of grass and timber. The patterns of FH use of botanical resources illustrates the fact that use and importance of particular species are not only determined by preferences and availability. Three important principles can be discerned: 142 1. 2. 3. Not all users meet the preconditions for such use (that make use feasible or rational), where preconditions in the region include access to sufficient labour, land, livestock or cash. FH structurally lack most of the preconditions that make the use of grass and timber resources possible or rational. Because of these same conditions, poor and land-deprived households are forced to seek CPR substitutes for essential resources that better-off households are able to at least partially produce or purchase. Finally, FH and other ‘very poor’ or destitute households seek means to generate more value from the CPR and the labour to which they do have access. Principles two and three are illustrated by the fact that, because they lack access to other assets, many FH sell tela to earn income. This increases their demand for fuelwood and hence the importance to them of fuelwood species. While fuelwood species are important to all women, they are most important to FH not only because of principle 3, but also because FH are less likely to have substitutes, such as dung and crop residue, or the ability to purchase from the market (principle 2). In Tigray, access problems of the type addressed above are compounded by the existence of AE that are oriented toward producing or protecting resources that are mainly useless (unusable) to the landless and FH, and by restrictions on species that are very useful to them. This obviously has implications for existing species use patterns. For example, R. nervosus and B. gradiflorum are among the most frequently exploited species in the area, particularly for fuelwood. Their increased use is a direct result of degradation and of the restrictions that have been imposed on more preferred and important species, such as awelie and chenedoge. According to residents, in the past hahote was never used for fuelwood; people “hardly noticed it” other than as a source of wild food. Today it is used extensively, particularly for fuelwood, but this does not mean that its use is either sustainable (not subject to over-harvesting) or ideal (does not imply health problems from smoke inhalation or excessive energy expenditures for collection among women who are already malnourished). On the other hand, the preferred fuelwood species are not produced in AE: even if they were, the poor could not access them because they have no ability to pay. The case of awelie, which was one of the few species that researchers could explore in depth, is clearly illustrative of the problems that residents and officials alike confront when attempting to negotiate a balance between conservation and livelihood needs, and as well of the fact that the needs of the population cannot simply be written off or relegated to the courts. The importance of O. europaea ssp. cuspidate is due to the fact that it fulfils multiple livelihood and cultural needs, and it does so very well. While restrictions have prohibited its use and any violations are supposedly reported and fined, it was found that in fact it continues to be widely used: even in the Household Survey it appeared as the sixth most used species in the area. While some use is legal (when it is purchased in the market or at administrative auctions or is obtained from private property), there is obviously a very substantial amount of formally illegal use that goes unreported and unpunished. Despite very wide public awareness of the restrictions and of the threatened status of the species, and despite everyone’s consensus that the species must be protected and that violators should be reported, use is pervasive simply because the species is so important to the community. The tabia administration has been encouraging people to plant the species on private property, but this is a slow process, and few of the people who need it have land on which to plant it, or money with which to purchase it from those can. The Household Survey revealed no instance where households had planted awelie, although several reported that they have planted eucalyptus.130 Similar problems must exist with respect to all of the restricted species. They have reached their current rate of degradation because they are the most important and most used botanical resources, and many are also slow to regenerate. There is no simple response to the constant threat of local extinction. However, awareness of which species are important and why can help inform any conservation and reforestation efforts in the area regarding which species should be a priority, and that seek to allow these species to be exploited by those who most need them. While it is clear that botanical resources are very important to subsistence in Adiarbaetu, there is less indication that they contribute much cash income. Rather, their cash contribution appears 130For a discussion about why private landowners generally prefer to plant eucalyptus rather than indigenous species, see Jagger et al 2003. 143 to have declined over time as a result of resource degradation, enclosures, and species restrictions. However, in another kushet (Dedeba) that is located closer to Mekele, however, it is reported that the ease of market access has led to major problems in enforcing restrictions in AE: . . protection of the government owned forests situated in Enderta Woreda has been a problem for the government because the people living in the area harvest trees illegally and sell them in Mekele town. The government has over the years increased forest guards to protect the forest, but to no avail. Socio-economic conditions of the area are such that the people have less alternative opportunities and are thus compelled to sell wood from the government owned forest . . These factors indicate that it is important to see the AEs in relation to people’s broader livelihood strategies (Nedessa et al 2005:17) Adiarbaetu is certainly not better off than this neighbouring kushet, but until 2005 it did not have easy access to Mekele. Residents, however, remember that fuelwood and charcoal sales were once a good source of income, and it is probably only a question of time before the problems experienced in Enderta woreda emerge in Adiarbaetu. Many men and women repeatedly stated that the most important effect of the resource restrictions on their livelihoods are related to fuelwood. While restrictions certainly affect own consumption, they have had the most negative effects on those who sold fuelwood which, several sources have reported (Sharp et al 2000, FAO n.d., Williams et al 2001), has been the principle source of income for the landless. It was reported that, for some residents of Adiarbaetu, these sales provided supplemental income with which they could purchase coffee, sugar, and tea, but for others, such as the landless and FH, it provided their mainstay. In an effort to recognize this need, officials in Adiarbaetu do grant individuals who have no other options permission to collect and sell fuelwood; however, as discussed above, the number of people who have actually obtained permission is minute. The price of fuelwood has increased as sources have decreased, which is also stimulating landowners to plant eucalyptus, and those who sell it are able to earn more selling smaller amounts. Others, who before might have been able to pay, are now forced to collect. Also because of price increases and problems entailed in gaining permission for legal collection, illegal collection and sales occur. Many women can be seen selling fuelwood and charcoal in the weekly tabia market. However, all of these women are from MH, and all but one come from other kushets in the tabia.131 Most of these women said that their husbands collect the fuelwood for them. Therefore, it appears that few of the households that most need the income, such as FH, are benefiting from fuelwood sales although, as was reported above, two female heads reported they use to sell illegally, one of whom earned approximately five birr per week by selling mixed species, and the other of whom sold only awelie and chihidi, the most preferred species, and earned 18 birr per week. Nevertheless, an unanticipated effect of the restrictions on fuelwood collection and sales might be a rise in prostitution, which is an alarming prospect considering that it appears to already be prevalent among female heads and also considering that this is a major route for the transmission of HIV/AIDS in Tigray, which is spreading rapidly. Other species or plant products that are often sold as raw materials include beherere (unknown) and gesho which are used in the production of tela and tedj. However, beherere is found only at lower altitudes, so residents must generally purchase it. While gesho does sell for a relatively high price, it is cultivated. Only two women were observed selling it at markets. Land and water access limit the ability of the poor to cultivate this plant. There were very few botanical resources found in common land areas that were reported to be sold. Women reported that wild foods such as aagam lahami (Carissa spinarum) are sold but, since they are increasingly difficult to find, they are only occasionally available in the market. Beles is the most commonly sold wild fruit, and it is both collected in common land areas and produced on private land. Wild fruit sales are successful even when people can collect the fruits themselves because of the time and labour savings that this represents for the purchaser. Many older or disabled people are not able to collect, and therefore purchase wild foods in the markets. 131Since sale of fuelwood is illegal without permission, many women were of course hesitant to speak with researchers, fearing that they would be reported. More time would be needed to build rapport to fully understand the history and extent of their sales. The fact that the present research did not locate women from FH or women from Adiarbaetu who sell fuelwood does not mean that they do not exist. 144 In the Household Survey (see Section 5.F), the use of other wild species for income generation appeared in the categories ‘collection and trade of spices and aromatics’ and ‘basket making’, which represented about six percent of all of women’s activities, in addition to which it was found that one FH engaged in selling kofkaf (unknown), berere and medafe talien. The amount of cash generated was up to around five birr per day. In spite of the fact that little cash income is earned and that only a few residents sell wild botanical resources, these nevertheless constitute an important part of their livelihoods, as well as a major subsistence contribution not only to these households, but to nearly all households. In households struggling with drought and environmental degradation, reduced access to land, water, and common lands have diminished their ability to generate cash from wild and indigenous botanicals. Despite these difficulties, the people of Adiarbaetu tend to agree with and see the need for the present restrictions. This does not, however, mean that they do not also feel strongly about the loss of these essential resources, and it does not mean that they do not understand and exercise the ‘need for exceptions’. Further, they would like to be more involved in the decision making process. At the moment they feel that rules are imposed upon them. Their initial reactions to the species restrictions were unfavourable and, while they now acknowledge the need for them, most residents expressed their continued frustration at not having had real input. 145 146 Chapter 7 Leaving Two-thirds Out of Development: A Point of Arrival One of the key questions raised in this research was whether CPR resource access and use regimes are sustainable, given the development dynamics that are evident in the region. It is argued that the neglect of a wide range of botanical resources and of open access common land areas, and the neglect of the need to balance material and cultural demands of local populations with conservation, mean that current CPR management regimes are very probably unsustainable. A third of productive assets, or open access CPR, have been ‘left out’ of the development equation. A second key question that was raised is whether household formation and dissolution are important factors in development dynamics. It was suggested that they are, and that a key point of entry for understanding them is female household headship. It has not previously been asked whether FH, and their associated high levels of poverty, are simply remnants of historical events (civil war, drought and famine, unequal allocation of land by the baito) that will eventually disappear, or whether, to the contrary, they are continually reproduced by highland culture and economy. Here it is argued that, especially now that traditional tenure systems have broken down, land reallocation is no longer possible, and land purchase is illegal, access to productive assets, and hence interhousehold differentiation, is now essentially determined by gender relations, which are integral to the continuous efforts, especially of men, to control assets through marriage and divorce. Female household headship, it is argued, is not only an outcome of, but also a condition for, this pursuit. In the process, an increasing number of both female headed and young households are left with very limited private productive assets. These households and, for specific reasons particularly FH, must live by providing their limited assets to others, by gleaning subsistence resources from common land areas, participating in food-for-work, and turning their limited and fairly uniform set of skills and assets toward spheres of petty production, trade, and services that are, nearly by definition, oversupplied. Women are particularly disadvantaged because they face multiple cultural and material constraints that do not permit them to accede to a wider labour market and networks of social exchange. The general inattention to FH in particular, it can be postulated, is an outcome not only of the unwillingness of the State and other development agents to intervene in matters that are culturally sensitive and ingrained, but as well of the inability of the State to redress the problem, as it has so often done, through land reallocation. Those programmes that are oriented toward providing productive resources to households that lack them are also demonstrably targeted toward men. A third of the population, or FH, are thus ‘left out’ of the development equation. In addition to revisiting the postulations, recommendations for immediate action are made on the basis of the study results. Since the research was conducted on a small scale, the recommendation domain is likewise limited: the suggested actions are particularly applicable to Adiarbaetu and similar communities in Hintalo Wajirat woreda, and are primarily focused on means to improve the lives and livelihoods of FH, especially, but by no means exclusively, by improving their access to botanical resources. 7.A How Female Household Headship is Entailed in Development Dynamics Beyene (2003) argued that the rist system and land reallocations under reforms both led to intergenerational discontinuity in landholdings and in farm households. Land reallocations under the reforms created incentives to establish new households, since it was by this means that more land could be obtained, particularly since heads were allocated more land than their adult children. Such dynamics, while maintaining a degree of equity among households, nevertheless led to farm fragmentation and diminution, created disincentives to invest in land, and acted as a brake on accumulation and hence economic growth. Beyene, however, was only concerned with inter-generational instability, and did not broach the intra-generational instability of households and holdings that occurs with divorce: nevertheless, high divorce rates should have strongly contributed to the negative development dynamics that he discussed. Nor did he take into account means of gaining access to land other than as parental gifts upon marriage, inheritance upon parent’s death, and land allocation through the 147 State (baito), which researchers such as Amare (1999) did discuss. Most importantly, Amare and others showed that divorce and remarriage present another means to access land and other assets, particularly for men. The means by which men were and are able to gain access to additional land that were discussed in the literature and also encountered in Adiarbaetu are rooted in the rist system, and continue to be reproduced through cultural norms and expectations regarding gendered rights and obligations. Men divorce their wives when they are able to marry other women with more assets. Given that, upon divorce, marital assets are generally unequally divided, where men retain oxen, most or all of the land, often the house, and older (working age) children, it is rational for them to divorce, as long as they remarry. With remarriage, men stand to gain more than they stand to lose, as long as they are able to find a new wife who has at least as many assets as they are forced to relinquish to their exspouses. What was found in Adiarbaetu that was not discussed in the literature was another set of phenomena that may be equally important to explain why it continues to be in men’s interests to divorce and remarry. First, it was found that much of the land that FH held that they reported had been obtained from the baito was probably still registered in their ex-husband’s names, since land allocated to the household is usually registered in the name of male heads and the baito no longer intervenes in the event of divorce or inheritance. More than half of the plots that women heads held were owned together with their ‘original owner’, most likely their ex-spouse. Second, FH often sharecropped their land out to their ex-husbands, or otherwise traded their resources (e.g. crop residues) with them for ploughing services or other labour. Third, FH did not appear to receive any other contributions from their ex-husbands in the form of labour, income, or food. Thus, it appears that a common way for men to continue to access the land pertaining to their ex-wives, and to at least nominally provide their exwives and children with support, is to sharecrop their land in. This is land that men would have had to plough and work had they remained married so that, upon divorce, in effect all they are giving up is half of the harvest, which in any case would have been destined to feed their wives and children had they not been divorced.161 Since FH clearly do not receive enough food in this manner to provide even a substantial fraction of what they require to survive, this ‘contribution’ almost certainly costs men less than it would have had they remained married. Men are thus free to only partially support their ex-wives and children, and to remarry women who have assets to bring into a marriage, thus expanding their total assets. It is no longer possible for men to obtain land through the baito. Young men must obtain land to establish farm and homestead through their own and their wives’ parents, but the amount of land that can be obtained in this way for most is minimal, since there are so many heirs and so little land. The only other way to access sufficient landholdings, therefore, is through divorce, remarriage, and sharecropping. The lion’s share of land that is offered for sharecropping pertains to FH. Those men who are unable to access land in this way are thus likely to remain poor. In addition, while such cases were not encountered among the 30 surveyed households, it was reported in the literature that older children usually remain with the male head after divorce, while women receive custody of young children. Men would therefore retain access to much of the household’s labour force, while women retain the burden of having to care for children who can contribute only little, if anything, to household subsistence. The high dependency ratios of FH found in Adiarbaetu and in other studies in the highlands mean that either divorce occurs early in marriage and women keep custody of young children, or that divorce occurs later in marriage after adult chil161Further, men were able to access additional land by having children by more than one woman, insofar as having children by a woman, even out-of-wedlock, meant having claims to a woman’s rist rights. Thus, a man was rewarded as long as he was able to contribute to the support of minor children outside of his marriage. While this phenomenon should have disappeared since land is no longer obtained through rist, it may in fact have been perpetuated as a means for men to gain access to land to sharecrop in. It is reported that, given the cessation of baito land allocation, there is increasing competition over such land, leading even to the payment of inducements to landowners (Berhanu 2004). While the Household Survey unfortunately did not determine whether FH contained children born out-of-wedlock, it did show that several FH sharecropped out land to people who were reportedly ‘kin’, but who were possibly not ex-spouses or parents. If a woman has land to sharecrop out, one way for a man to establish a permanent sharecropping arrangement with her may be to have children with her. This is reinforced to the extent that both men and women appear to consider that sharecropping is a means for an exspouse to ‘support’ an ex-wife and her children, as was reported by Hoben (see Chapter 2), and as seems to be the case in Adiarbaetu. 148 dren have established their own households, or that adult children remain with their fathers’ households. While there is no data that would confirm that the first and last cases predominate, de facto FH bear a larger burden caring for minor children, so clearly men have rid themselves of such a disadvantage through divorce, while nevertheless maintaining ties (or ‘rights’) to their ex-wives’ land through these children. What may be a rational course of action for men nevertheless leads to diminishing landholdings and increasing land fragmentation overall. Landholdings are divided upon divorce. If, upon divorce, women receive a lower share of land than what they brought into their marriages, and if household baito land is also unequally divided, then, over time, divorced women have less land, if any, to bring into subsequent marriages. Further, even if women sharecrop their land out to their exhusbands, they still require land for a separate homestead. Some do not receive this land and are forced to rent their homesteads, but most do appear to have at least enough land for a homestead. Since there are more divorces than remarriages, the number of FH should be rising as a proportion of total households, although a number of FH are war widows and there is no longer civil war, so overall the proportion may decrease for some time. It was reported in the literature and also in Adiarbaetu that men are not inclined to marry women who have minor children because of the responsibility and economic burden for offspring that are not their own. Divorced women most likely bring few or no assets into a marriage. Overall, the amount of the land that men gain by remarrying should also be diminishing since women hold a decreasing proportion of all land, which would increasing men’s reliance on sharecropping land of their ex-wives. This could eventually lead to a disincentive to divorce, which would then decrease the proportion of FH. However, as Amare (1999) indicated, men are also less likely to marry women who have no property and, as landholdings diminish, parents are unlikely to bequeath or give their land to daughters. The proportion of young women who remain single therefore is also likely to increase. Quite possibly, many of these young women will end up maintaining children born out of wedlock, or engage in prostitution. There is obviously a good deal of speculation in the above, and there is little statistical evidence to support the analysis. But what the data from Adiarbaetu do suggest is that (a) it was very possibly not only bias in baito land allocation that led to lower FH holdings; (b) FH have clear relations with ex-spouses through sharecropping that are based upon unequal exchange; (c) FH have full responsibility for minor children and little, if any, access to adult male labour other than through such relations of unequal exchange; (d) men have little means to gain access to additional land other than by marriage, divorce, remarriage, and sharecropping; and (e) landholding fragmentation and diminution increase with marital instability. Women are in a sense ‘highjacked’ into such a disadvantaged position because of deeply held beliefs about the incompatibility of certain agricultural tasks, such as ploughing, with concepts of femininity, about men as farmers and hence as breadwinners, and, it might be added, about the social ‘uselessness’ of men without wives, and of wives without husbands. Divorce, on the other hand, is seen as a natural, and indeed even inevitable, outcome of women’s lack of bargaining power vis a vis men in the form of marriage assets, and a lack of competitive ability vis a vis other, wealthier women. As was often reported in the literature, while remarriage is an option for women, it is a necessity for men. However, the number and structural poverty of FH is determined not only by marriage, divorce, and access to agricultural assets, as is discussed below. 7.B The Social and Economic Position of FH Informants in Adiarbaetu often reported that female heads are seen as ‘useless’. Female heads constantly contend with this subtle, yet ever-present cultural assumption, which is held not only by men and women from MH, but as well by women heads themselves. As a result, the phenomena that give rise to their existence, their particular struggles and those of their dependents, are relatively invisible to them and to the community as a whole, making them for officials and other development agents the social equivalent of welfare recipients. This negative appraisal is strong and influences all of the actions taken to alleviate poverty and improve food security. Such a social valuation is likely to be a reflection of women’s general inferior status vis a vis men, which is compounded by female heads’ problematic status as single women or single mothers. However, it is also strongly related to the economic status of most FH, which have very limited productive assets, very limited access to labour markets, and few prospects of acquiring these outside of 149 (re)marriage. As has been shown, most landed FH households in Adiarbaetu live in part by providing their limited assets to others (e.g., land, crop residues). When they do produce on their own land, it is of lower quality and hence yield, a greater proportion is dedicated to the homestead and thus a lower proportion is planted to crops, fewer crops are planted, and less own production is consumed. They lack access to adult male labour with which to carry out many essential agricultural tasks. Almost none can sustain livestock (and thus they have no access to dung), or consume or sell livestock products. Almost none have homegardens, and very few have their own trees. Some, who have sufficient assets and who are able bodied, do farm themselves, and do find means to improve their holdings, e.g. through tree planting, soil and water conservation measures, and purchase of oxen. But these are very few. Female heads in Adiarbaetu appear to have little opportunity to sell their labour: unlike men, they are unable to travel long distances to find work and there is little paid work for them, other than food-for-work, within the kushet. The same constraints limit their abilities to access and market more valuable CPR, such as fuelwood, medicinals, and wild food plants. Thus, they mainly use their domestic skills and tools to produce exchange value (e.g., by producing tela), but in conditions of extreme competition not only with other female heads, but as well with women from MH. If a ‘profit’ (in this case, a wage higher than the implicit, for example, in food-for-work activities) were to be realized, then other women would quickly enter into production, since every woman in principle has all of the skills and assets that are required. ‘Profits’ would be eliminated, and implicit wages would be driven down to the point where they are now, far below the minimum. The only sector where female heads can probably use their capacities to generate a higher implicit wage is in prostitution, since not all women are able or willing to become commercial sex workers. The risks entailed in prostitution are, nevertheless, very high in terms of social ostracism, unwanted pregnancy, and disease, where HIV/AIDs not only presents the threat that such women will pay with their own and their children’s lives, but that they will increasingly become focal points of risk for their neighbours. Food-for-work should, therefore, represent the one means by which female heads can access subsistence goods without confronting stigma, inequity, and undue risk: it is intended that food-forwork programmes benefit those who are most in need. In fact, FH average fewer days of food-forwork than MH due to lower labour availability, and their participation also comes at higher costs in terms of physical effort, where they are expected to perform the same work that men perform, while consuming less than what women minimally require. The opportunity cost of their labour, and of the time spent travelling to and from food-for-work activities, is also higher than in comparable MH, given that most have minor children and lack access to childcare. The majority of all FH cannot produce, or generate sufficient income to purchase, the wide range of botanical resources that are required for daily life, including fuelwood, food (including those containing vitamins and other micro-elements, such as fruit and vegetables), medicine, aromatics, grass for basket making, thatching and coffee ceremonies, construction materials, and materials for making tools. They are unable to purchase the grass and timber that are sold from AE, and have no use for AE to graze livestock. Their only option is to glean those subsistence resources that they cannot survive without from common land. On the one hand, they glean from land where vegetation is sufficient but where they are not allowed to use the resources that they most urgently require, and on the other hand, they glean those resources that they most urgently require from land where vegetation is largely depleted. Across both of these areas, those CPR that are the most valuable to them are also the most restricted, and yet it is they who most often cannot desist from their use, so they also run the greatest risk of punishment. The particular form that CPR management has taken post-reform has largely deprived FH of access to botanical resources that are vital to their welfare and livelihoods, forcing them, as well as other households, to over-exploit open access CPR. Again, the position of FH is structural rather than incidental, and exploring this position reveals a great deal about the sustainability of, and threats to, current CPR management regimes. 7.C Livelihoods and CPR Access: Are Current Management Regimes Sustainable? As discussed in Chapter 2, a better understanding of the possibilities for the success and sustainability of the government- or community-promoted conservation measures depends on their relation with livelihoods. Likewise, a better understanding of livelihood processes that generate or perpetuate poverty also depends on an understanding of the contribution of CPR, and hence of the ways in which 150 resource degradation in these areas creates poverty and destitution. While poverty is not the primary cause of degradation in the area, it certainly exacerbates it, particularly when most CPR are now open access, and violations of access rules are apparently common. Most research, however, on such contributions in the highlands has focused predominately on two resource areas and two resources: grazing land and woodlots, and grass and timber, respectively. The contributions of these as well as other resource areas and CPR are as yet not adequately understood, and have been under-estimated (Williams et al 2001). Research in Adiarbaetu shows that a greater number of resource areas and CPR contribute to household livelihoods, and that it is the poorest, and especially FH, that most depend upon these resources. As discussed in some detail in Chapter 2, and as was ratified in the study of Adiarbaetu in southeastern Tigray, common land areas and CPR are essential to livelihoods, even in heavily degraded and drought-stricken regions. CPR access regimes have also changed repeatedly, with inevitable effects on livelihoods. Development dynamics have led to the fragmentation and diminution of individual farm holdings. Drought, the conversion of common lands for agriculture, and increasing livestock populations have led to the reduction and degradation of grazing lands. Traditional control over forest and tree resources also eroded: forests and woodlots were often converted to farmland, and deforestation of remaining areas proceeded apace. The development of a policy of enclosure to permit regeneration and environmental recovery has led to environmental enhancement, but has not been informed by a vision that promote linkages between environmental sustainability and livelihoods, or that recognizes that local populations are heavily dependent upon primary and secondary resources other than grass and timber. Although there are instances where particularly disadvantaged groups, such as the landless and FH, have been specifically permitted to benefit economically from enclosed areas, in general it appears that the economic benefits perceived from these enclosures have been either unequally distributed, in the case of restricted grazing lands, or minimal or negative, in the case of enclosed woodlots. Most CPR that local populations glean from common land areas remain invisible to policy makers and other officials who repeatedly informed members of the research team that plant resources are in such short supply in Tigray that they cannot constitute an important component of people’s livelihoods. Nevertheless, it is difficult to explain why, for example, the clearly recognized need for fuelwood and its acute shortage has been barely considered when developing policy around the use of enclosed areas. Is this possibly related to gender bias? Given that women are generally responsible for fuelwood collection and that they are far less represented in the local councils that are responsible for decision making about resource exploitation within enclosures, this is a very strong possibility. However, it can be argued that men’s needs for CPR are also barely fulfilled through enclosures, although the uses that AE do fulfill are much more closely related to their needs in comparison to women’s needs, or to the needs of those population groups that own few or no livestock and have little or no cash income. Also discussed in Chapter 2 was the fact that research relating to AE in Tigray is increasingly focused on the relation between conservation, livelihoods, and participation, including possibilities for greater devolution of decision making to local levels, including privatisation. These issues have emerged particularly since the benefit streams relating to AE do not appear to be supporting local livelihoods. While perceptions of environmental benefits clearly favour AE, the perception that residents have of economic benefits is also crucial in resource management, providing people with incentives to participate in conservation. Further, “the manner in which the community allocates the benefits among themselves can affect collective effort of the group negatively or positively depending on the degree of ‘fairness’ of the allocation” (Berhanu 2004: 34). To begin with, in most cases, it appears that communities had their own management regimes long before the State other outside organizations introduced AE. In Adiarbaetu, many residents reported that restrictions on grazing land and forested areas existed long before the Derg Regime. It was because there was already a local tradition of enclosures and resource management that was largely overturned during the Derg Regime, that the majority of residents have supported the AE introduced since 1991. Although little is known about traditional resource management regimes, these must have been effective, since CPR appeared to have been abundant enough to allow for the successive and repeated conversion of common lands into private holdings, at least until civil war and drought took their terrible tolls. The early reforms imposed a new style of top-down CPR manage- 151 ment which communities rejected, and which certainly also disorganized existing CPR regimes and made much community land de facto open access. There has been much more community participation and support for the current AE regime. Residents participate in the decision to designate areas as AE and have at times rejected the proposals made by higher officials to enclosure certain areas. Further, most seem to think that AE are an absolutely necessary part of the solution to the severe resource degradation that affects their communities. The short- and long-term environmental benefits are clear and are widely appreciated, and therefore AE are strongly supported. In the Household Survey, the main causes of degradation that respondents gave were drought, overuse of resources, and population increase, in that order. While some thought that AE contribute to their lack of resources and poverty, the majority said that they are an important part of the solution. In addition to the types of environmental benefits mentioned in Chapter 2, AE and species restrictions allow for the regeneration of important native species. Research in Adiarbaetu found that residents favour planting a variety of tree species, including many important indigenous species which are slow growing: while they acknowledged the need for fast growing species, they think that the majority of the trees planted should be indigenous. In the Household Survey, most of the respondents said that they would like to see an increase in tree planting outside of AE. Most also said that there should be more efforts to regenerate resources in other open access areas, such as hillsides, riversides and borderlands, but most stopped short of insisting that these areas should be enclosed. There is, however, greater ambiguity and complexity entailed in the perceptions and assessments of short and long-term economic or livelihood benefits from AE and species restrictions. The economic benefits that have been realized in the relatively short-term include grass for grazing and other uses such as basket making and, to a lesser extent, wood for construction. Woodlot enclosures have also led to the regeneration of other herbaceous and woody species to which residents have access, providing that species restrictions don’t also apply. But there have been few short-term economic benefits from AE with respect to a multitude of resources other than timber and grass, and access to some resources, such as fuelwood, has decreased, both since little can be extracted from AE, and because reliance and hence pressure on open access areas has necessarily increased with AE and species restrictions. In this respect, species restrictions have been difficult for the entire community, and there is much tension around them. Most believe that they are necessary and important. Their dissatisfaction is not related to the existence of restrictions, but rather to their scope. Although the long-term benefits are understood, they are not reconciled with immediate needs. As a result, much illegal use continues. In Adiarbaetu, it is clear that FH, particularly those without livestock, do not gain short-term economic benefits from timber and benefit only little from grass. However, they are the principle users of other ‘minor’ AE resources to which they would otherwise probably not have access. Their more intensive use of certain AE resources is both a reflection and cause of their poverty. Greater reliance on AE means, to them, more access to subsistence resources, and also more time and energy spent procuring subsistence resources that other people don’t have to procure by this means. FH’s lower reliance on timber and grass from AE is also both a reflection and cause of their poverty: they don’t meet the preconditions for their use and therefore cannot benefit. There is also a perception in Adiarbaetu of inequality in the use of common land areas. The poorest households, especially FH and the landless, often expressed their frustration at not being able to use common land areas more for income generation, for example through fuelwood collection and sales, which was prevalent in the past but nearly impossible today. Others suggested that some common land areas should be redistributed to the poorest, particularly the landless, for use as farmland or homegardens. Researchers have especially expressed concern about expectations for long-term economic benefits from AE and about how these benefits, if realized, will be distributed. The total value of timber stands in AE in Adiarbaetu appears to be quite high, but it is unclear whether such stands will be sold, or how the revenues from eventual sales will be distributed. In this sense, no one has clearly defined property rights. Community members appear to have few expectations that AE will directly contribute to their livelihoods in the long-term, probably because, as is the case with other communities across Ethiopia (see Chapter 2), they have no idea what types of long-term (or even short term) revenues might be generated, and these issues have not been open to discussion. For some, there is the hope that the increase in woody species will eventually permit them to sell fuelwood and charcoal. But most residents still have faith that they will realize long-term economic benefits. It appears 152 that, in general, people do not need to be assured that they will obtain economic benefits in the longterm, since environmental improvements and short-term benefits (e.g., access to grass) are enough to gain their support for such measures. But attitudes about both short- and long-term benefits appear to be fairly closely related to the economic status of those who express them. Initiatives to resolve botanical resource scarcity in the area have not been aimed at common land areas alone. As discussed in Chapter 5, private tree plantations have been encouraged particularly for fuelwood production for own consumption and sale. However, relatively few residents have thus far been able to plant trees, although those who have the land and economic means appear to do so. The same problems that affect enclosed woodlots also apply to private planting: it is often years before these trees can be used for fuelwood or timber. Seedling survival rate is also a concern. Since water is limited (compounded by livestock damage), many seedlings that are purchased do not survive. Other major issues that have been raised with regard to private tree planting as a possible solution for resource shortages relate to their possible detrimental effect on biodiversity (private landholders generally plant only trees that provide short-term returns, such as eucalyptus) and their implications for equity (Jagger et al 2004). Officials and residents of Adiarbaetu clearly are attempting to maintain a ‘fragile balance’ between resource conservation and exploitation, particularly through informal access rules that are best referred to as the ‘need for exceptions’, which also seem to be meant to redress some of the inequities in access to AE and other assets, such as land. As was shown, local informal access rules were pervasive in the research area, and yet rarely acknowledged in the CPR literature on Tigray, and they are based on principles related to equity, age, and need, including deeply culturally embedded notions about the importance of particular CPR to spiritual and physical well-being, and to the obligation that different groups have to use certain CPR for these purposes. It includes customary norms of access to resources such as wild food and medicinal species even on private land. There is tacit acknowledgement on the part of both officials and residents that formal rules discriminate against disadvantaged community members, and informal rules seem to uphold the precept that everyone has a right to meet their subsistence needs, which has often been found to underlie customary tenure regimes. Formal and informal rules appear to contradict one another, yet they actually work together as a form of legal pluralism. It is highly questionable, however, whether the existing situation actually provides anything other than a short-term solution to resource degradation, poverty, and the maintenance of the cultural integrity of local people. What can be asserted is that State intervention that is not complimentary to such local attempts to balance conservation and use only increase the difficulties confronted. Chisholm argued that external factors, such as the replacement of traditional management practices with State imposed management, has led to less direct community involvement, and therefore less concern for the management of resources (2004: 16). In Adiarbaetu, residents spoke proudly of the traditional community enclosures and clearly respect churchyards. From discussions, it appears that fewer violations occur in these areas than in those that have been established by the government and other entities. Community management as it plays out in Adiarbaetu is the only means to manage the multidimensionality of the botanical resources, the landscapes, and the community that depends upon them. While state involvement is important for establishing guidelines and resources for the management of CPR, these should clearly compliment traditional management practices and not usurp local attempts to maintain a balance between resource use and conservation. Community involvement, beyond the recognition and encouragement of traditional management practices, includes the consultation of all community members regarding AE and species restrictions. As research in Adiarbaetu illustrates, different households have different needs for and uses of common areas and botanical resources. Therefore, all user groups need to be included in decisions regarding the management of these areas, especially FH and the landless. While the entire community is allowed to vote on an enclosure before plans are finalised, there are few opportunities for community involvement prior to or after this event. Women heads reported frustration that their attempts to raise their concerns, particularly regarding fuelwood, have been largely ignored. Overall, these women support the AE and see them as necessary and beneficial, but they also realize that they, more than others, bear the costs, both because they provide much of the labour for their establishment and maintenance, and because they are often more dependant on those resources whose use has been restricted. 153 While research and policy making has focused on AE and on particular botanical resources, there is a much greater diversity of common land areas and of botanical resources that are affected both by policies and by degradation, and yet that are nearly entirely neglected in conservation schemes. All common land areas and all uses are entailed in and affected by resource degradation, and all must also be considered in resource management. For any resource management regime to be sustainable, it must be based on an understanding of the importance of CPR to local populations. With such understanding, access regimes can be established that create links between exploitation and conservation; without it, such links are nearly impossible to ensure. Blanket policy approaches to CPR, it might be asserted, lead to solutions that suffocate local communities. In summary, the results of the research presented in this report show that, in Adiarbaetu, CPR cannot be managed solely on an area basis. People seek specific botanical resources that are important to them no matter where they occur in the landscape. If restrictions on resource access are created in one area, then demand shifts to another. Species restrictions provide a partial response to this: it is only by restricting or preventing use of the most threatened species anywhere they can be found that their disappearance can be prevented. But, it is also clear from this research that formally declaring an activity as illegal does not prevent it from occurring. These are some of the most important issues that both the policies and the research on CPR use in Tigray have largely failed to address. If ‘open access’ areas and use of restricted species are not adequately dealt with, then these areas are likely to become deserts, with AE dotting the landscape like tiny oases. If the material resources that are produced in AE continue to largely fail to meet the most pressing needs of the population, their destiny can only be likewise to disappear. These neglected resources and common land areas, we argue, constitute the forgotten third of development in the region, and their neglect threatens to undermine the other two-thirds: farming and AE. The second set of issues that is crucial to address in CPR management in Adiarbaetu and all other villages in the highlands is that of destitution and social equity. Highland political culture has, for centuries, upheld principles of equitable distribution of key livelihood assets, and has both formally, and to a degree in practice, recognized women’s entitlements. However, many cultural and economic factors have recently combined to generate very high levels of poverty, especially among particular types of households: those headed by divorced or widowed women and those that are young. It can be demonstrated empirically that these households rely more on CPR and are the major users of the ‘minor’, if not ‘major’, CPR resources. There are no rules limiting FH access to resource areas or resources in particular. In fact, it was found that the conceptual approach to assessing rights to botanical resources presented in Chapter 2 was sufficient to establish patterns of access, but it did not offer explanations for the patterns encountered in Adiarbaetu. What was missing was a crucial substrate of information that refers to the preconditions for use of CPR, that is, the material conditions without which rights of access are unusable, which of course lies at the base of debates about social inequality. Equal rights are not equivalent to equal opportunities: equality of opportunity supposes the fulfilment of preconditions for access, such as assets, predisposition to seek opportunities, and lack of effective discrimination. Both FH and the plant resources that are essential to them have been largely neglected in development efforts: this ‘third’ of the population, and this ‘third’ of the livelihood resource base, have either gone unrecognized, been culturally defined as useless,162 or simply been neglected. FH are often forced into destitution due to higher dependency ratios, unequal division of household assets (with women accessing fewer resources of lower quality), and unequal opportunities to farm and to participate in civil life. FH are more dependent on common land resources, but many such resources are in fact of minimal use to them due to enclosure, which in general are not governed with their resource requirements in mind, and those that remain open are under increasing pressure precisely because they are relatively free access and increasing numbers of people, both female headed and young households, depend upon them for bare subsistence. The unintended but very real marginalization and neglect of these households, both in terms of access to private assets and of access to ‘major’ CPR benefits, can only result in an increasing environmental and human toll, as well as increasing inequality, in an area that already suffers from some of the worst human and environmental degradation on the planet. 162Open 154 access resources are often referred to in the literature as “wasteland”. 7.D Entry Points for Local Action Officials in Adiarbaetu admitted that, until recently, they “never paid much attention” to FH. It is only within the past two years, with initiation of the FAO project, that they have begun to pay more attention to these women and their difficulties. In the end, it is the neglect of their circumstances, alarming numbers, and the relations that generate these households, that ensures that they remain destitute. The success of any endeavour to reduce poverty, address food insecurity, and even to promote resource conservation and environmental rehabilitation, is dependent largely on understanding and addressing these issues, as well as adjusting interventions to the day-to-day realities of female heads. One of the primary opportunities for women to have a political voice within the community is through the WAT. Women pay dues, attend meetings, and elect leaders who then have representation among the village officials. However, WAT leaders reported that, although women pay dues, there is very low participation, especially in meetings. It is a relatively new phenomenon for women to speak out. Women leaders said that they constantly battled the perception, particularly when they first took up their positions, that they were ‘trouble makers’ and not ‘real women’. Overall, women are soft-spoken and reluctant to discuss their problems and ideas. Men’s organizations not only have more power within the village, but also are given more respect and support by outside government and non-governmental agencies. Women leaders said that male officials and leaders are often asked to attend meetings in Mekele and even Addis Ababa, with all travel expenses paid. They also receive financial support from the government. As yet, women leaders have not been asked to attend meetings outside of the tabia and do not receive money from the government. There has been increasing interest, especially in the past few years, in the situation and struggles of FH and the landless, since they are among the poorest in this region. However, the focus has tended to assume that they are separate categories. However, there are women who are not only female heads but are also landless. These women not only face all the struggles of FH, which have already been discussed, but also have no agricultural production and therefore completely dependent on other income earning activities, which are hard to come by. Therefore, while other female-headed households might face food shortages three to six months of the year, these women and their families face food shortages throughout the year. This in turn can affect their ability to be able to work at all, especially in strenuous food-for-work activities.163 With few alternatives, many turn to prostitution to be able to feed themselves and their families. While all families in Adiarbaetu are struggling with food insecurity and poverty, the landless, and particularly landless FH, are by far the most destitute and require yet more attention and support, not only from NGOs, but as well from international agencies and especially the Ethiopian government. Fuelwood should be the first priority of any programme oriented toward supporting the livelihoods of the poor, including those that are oriented toward improving food security and environmental conservation. Together with grass, fuelwood is by far the major use of plant resources taken from common land areas. With additional fuel sources, as regeneration occurs other plant resources would also become available, increasing diversity and possibly easing food shortages. Women, and especially female heads, spend much of their time collecting fuelwood, which entails opportunity costs as well as caloric expenditure. Women could spend this time on other activities. The species currently used for fuel produce a great deal of smoke, so that alternative fuel sources that produce less smoke would also result in improved health. Two potential sources of fuelwood are enclosed woodlots and unenclosed common land areas. Within enclosed woodlots, it appears that more species must be introduced that can fulfil multiple livelihood needs, including the need for fuel. Even so, if fuelwood produced in enclosures are sold, FH and the very poor will still not have access. Further, eucalyptus, which predominates in these plantations, does not seem to be a viable source of fuel: in spite of its availability, residents re163A twenty-three year-old landless female head rents a small home with most of the income that she earns from odd jobs. Food shortage is a constant problem and she often goes two days without food so that her small daughter has something to eat. As a result, she is weak and struggles to accomplish food-for-work activities. Often she is sent home. She sells what little grain she receives in order to buy other vegetables and food. She even has to borrow a small burner and charcoal in order to prepare her food. In spite of her dire situation, she was determined not to take up prostitution. 155 port that they barely use it for this purpose. A study that examines the possibilities and constraints of eucalyptus as a viable fuelwood source is imperative, since devolution of plantations to farm level (through private plantations) may not be a viable solution insofar as households appear to be planting only eucalyptus, and as well FH may lack the labour required as well as the ability to withstand the risks associated with high seedling losses. Alternatives to household-level devolution include the development of cooperative plantations on common land areas, where cooperatives could be managed by landless and FH, and which could provide not only production for own consumption, but as well for sales. Cooperative plantations would overcome some of the constraints that FH confront in terms of limited access to labour, and would pool risks. Other alternatives, such as fuel-efficient stoves, should certainly be considered, but they should be seen as complementary to, and not substitutions for, fuelwood production. To be effective, fuel-efficient stoves must be available to every household. Experiences with ponds in the area indicate that households often don’t receive all of the inputs that are required in order to make them completely functional. In any case, while they may reduce the need for fuelwood, they do not eliminate it, and households will continue to seek preferred sources of fuel such as awelie from open access areas. Environmental conservation would be enhanced both directly and indirectly. With an increased number of multiple-purpose, multi-species plantations, the positive environmental effects attributable to enclosed woodlots would be extended in space. Threats of extinction of preferred species on the restricted list would not only be reduced, but potentially reversed, if these species are purposefully planted and if use is regulated, e.g. through cooperatives. If sufficient fuelwood can be produced, dung can again be applied to soils; although it was not reported in the household survey that crop residues are used for fuel, their use for this purpose has been reported elsewhere, and increasing the supply of fuelwood would reduce this possibility. Further, multiple purpose plantations could also increase the production of tree fodder, and certain grass species could also be produced, albeit on a more limited scale, as they are in enclosed woodlots. A study that seriously considers the viability of multi-species, multiple purpose plantations that are designed with the participation of local residents after conducting research such as that presented here, is a necessary precursor. In spite of the fact that FH are recognized as the poorest and most food-insecure households in the region, and that they are singled out by some development programmes and projects for support, FH have no associations or institutions through which they can organise themselves, nor are they adequately represented in political bodies or decision making organisms. While some female heads belong to WAT, compared to women in MH, they have different problems and needs, and require an own association to permit collective political, economic and social action. An association would provide them not only with a collective voice vis a vis tabia and kushet officials, development organisations, NGOs, and WAT, but would also provide an opportunity for them to offer mutual support. As one of the women expressed it in a focus group discussion, female heads work constantly and rarely have time to discuss among themselves. Focus group sessions provided them with such an opportunity for the first time: “I have learned so much from them. It has been helpful to have the time to interact with other women and discuss the issues affecting us.” Awareness raising is a first step, and women heads need the opportunity to learn that their concerns and needs are similar and have similar determinants, which can be changed. Such an association could be an adjunct to the WAT. Better integration of FH and women in general within existing organisations is also imperative. While many FH belong to the PA or baito, this involvement is largely passive and the policies that are formulated, as is demonstrated throughout this study, obviously do not take their needs into consideration. When wives participate, it is highly likely that they simply add their own votes to their husbands’. Affirmative action is required, and this can only be achieved through political will and through an own organisation that agitates for such action. WAT leaders in the kushet said that it is a constant battle to get women involved; they see little to no benefit from the time invested in these associations. As long as their participation is ineffective, they will continue to be apathetic. Respect for their association, and political will to help them to overcome the constraints they confront when participating, must be forthcoming at all levels. The baito must once again become involved in the division of household assets upon divorce and death, but it must do so with a view toward ensuring that the welfare of female heads and their minor children are paramount. Upon death, the primary consideration of the baito should be the ability of the female head to support herself and her children. In divorce settlements, equality in the division of assets should not be the primary goal: rather, the primary goal should be to ensure the welfare 156 of children, and to meet the subsistence needs of all of the parties involved, which might entail an unequal division of assets in favour of female heads, or an equal division of assets coupled with ongoing alimony or child support through cash, food, or labour contributions (e.g. for ploughing), depending upon the specific situation of each household. It is only by ensuring that men’s responsibilities to their ex-spouses and children are met on a continual basis that divorce, childbearing out-of-wedlock, and irresponsible paternity can be deterred. If baitos are able to enforce AE and species restrictions through fines or imprisonment, and if they regulate access to food-for-work and administer communal work programmes, then they are certainly able to enforce child support and alimony contributions of divorced heads. Further, greater benefits from agricultural production for FH do not have to depend upon the elimination of cultural restrictions on ploughing or other agricultural work, which may still be a long time in coming. Nor does it necessarily depend upon increasing FH’s access to land or to other production assets in the short term. Despite not being able to plough, women perform most other agricultural work, and hold all of the knowledge and skills that are necessary to engage in own-account agriculture. As discussed in Chapter 5, agricultural cooperatives provide a means for female heads to use their land productivity without giving up half of their crops. In order to receive any kind of support, agricultural cooperatives should be obligated to have a minimum percentage of FH members, which should correspond to the percentage of FH in the kushet. The research on the use of botanical resources and of common land areas has shown that there are various activities and resources that might prove to be viable sources of income for FH and the landless, if certain conditions are met. Several of these bear further investigation, which would entail (a) marketing possibilities and margins; (b) viability of transplanting or cultivating wild species or, alternatively, studies of availability and harvesting pressure on wild species to determine to what extent higher levels of exploitation are possible; (c) production requirements in terms of land, labour, production space, inputs, equipment and capital and the facilities that different types of FH and poor households have for these; and (d) possibilities for cooperative production and marketing. Cooperative efforts could help to resolve constraints related to time, childcare needs, land, equipment and infrastructure access, as well as credit access. Among the primary complaints by women in regards to current business schemes is that they are not suitable to their conditions. Therefore, it is very important that they be included in the process from the beginning. Aside from the discussion above about fuelwood, which has the potential to be a viable income generating activity if plantations are established, below other examples of potential income generating activities are discussed. The research on the use of botanical resources and common lands showed that there are various activities and resources that might prove to be viable sources of income for FH and the landless, if certain conditions are met. Several of these bear further investigation, which would entail researching (a) marketing possibilities and margins; (b) viability of transplanting or cultivating wild species or, alternatively, studies of availability and harvesting pressure on wild species to determine to what extent higher levels of exploitation are possible; (c) production requirements in terms of land, labour, production space, inputs, equipment, and capital and the facilities that different types of FH and poor households have for these; and (d) possibilities for cooperative production and marketing. Cooperative efforts could help to resolve constraints related to labour, childcare needs, land, equipment, and infrastructure access, as well as credit access. The main complaints that female heads have with regard to current business schemes is that they are not suited to their specific conditions. Therefore, it is very important that these women be involved in the process from the beginning. Aside from fuelwood, which has the potential to be a viable income generating activity if plantations are established, other examples of potential income generating activities are discussed below. Due to the weight and height of beehives, men have been responsible for bee keeping. However, with the introduction of modern box hives, this would be a viable business for FH. Honey is quite valuable and could provide a substantial income, and production could be done through cooperatives or individually. However, it would prove to be difficult for those who do not have any place to hang the hives since, while it is possible to hang them in trees on common lands, men who do so said that they are constantly concerned about thieves, even though the structure and height of traditional hives makes this more difficult. Research is required to determine whether and how this activity could be made accessible and economically viable for FH and the landless. Gesho (dogwood - Rhamnus prinoides) is a small shrub or tree used in the production of tela and tedj, the favoured local alcoholic beverages. Demand for it is high, and it is said to have high 157 market value. However, it appears that very few women sell it. It is possible that its production, whether cooperative or individual, could provide good revenues for FH. Gesho can be easily grown from seed. It is reported to grow well in most soils, and that it is tough. It is, however, slow growing, and it also requires space. With the new road, more income could be obtained if women could find a way to access larger markets, such as those in Adigudam and Mekele. Several other types of wild plants frequently in the kushet and are often purchased at the local market, so they potentially offer income sources for FH, but only if they become available on a larger scale and closer to home, since the time currently spent in their collection is great. Some of these botanicals could be incorporated into multi-purpose woodlots, as are proposed above in relation to fuelwood. Food-for-work activities are often difficult for women to carry out and can be harmful to those who are undernourished, pregnant or ill. These programmes need to be re-evaluated to ensure that the work is suitable, and working hours should be revised to allow FH to accomplish other necessary tasks, such as caring for their children. Food-for-work could possibly be offered to women for child care services, if these women can be trained for this purpose. If possible, work should be more widely spread throughout the kushet. Improvements in sexual and reproductive health are essential, and must be promoted through education and awareness programmes, as well as by providing contraceptives and prophylactics free to the local population and specifically to FH. Although there is now a health centre offering HIV/AIDS tests and contraceptives, few women (and most likely fewer men) are aware of these services or of their benefits. While informational meetings on health and sanitation were observed, it appeared that women rarely attended. To reach women, it is important to work with the WAT, perhaps even having classes during some of the scheduled meetings. It is urgent that female heads engaged in prostitution be identified and targeted not only for HIV/AIDs testing and education, but as well to ensure that they have economic alternatives to prostitution. Such women generally do not wish to be commercial sex workers, but as long as they are unable to support themselves and their families in any other way, this will continue and they will therefore constitute focal points of risk. This is especially important given that men’s mobility will increase with the opening of the new access road to Mekele, which is one of the major sites of AIDs transmission in the region. Female heads who engage in prostitution are otherwise destitute; thus they should be the first to be targeted for the types of assistance discussed above. 158 References Amare, Yared. 1999. Household resources, strategies and food security in Ethiopia: a study of Amhara households in Wogda, northern Shewa. Monograph Series in Sociology & Anthropology. Addis Ababa: Dept. of Sociology and Social Administration and The Addis Ababa University Press. Asfaw, Zemede. 1995. Conservation and use of traditional vegetables in Ethiopia. In L. Guarino (ed.), Traditional African Vegetables. Proceedings of the IPGRI International Workshop on Genetic Resources of Traditional Vegetables in Africa: Conservation and Use 29-31 August 1995, ICRAF-HQ, Nairobi, Kenya. Promoting the conservation and use of underutilized and neglected crops 16. Online at http://www.ipgri.cgiar.org/publcations/HTMLPublications/500/begin.htm# Contents. Accessed 2-11-2004 Asfaw, Zemede. 2001. Home gardens in Ethiopia: some observations and generalizations. In Watson, J.W., Eyzaguirre, P.B. (eds.), Home gardens and in situ conservation of plant genetic resources in farming systems, Proceedings of the Second International Home Gardens Workshop, 17-19 July, Witzenhausen, Federal Republic of Germany. Rome: International Plant Genetic Resources Institute, pp. 125-139. Asfaw, Zemede and Mesfin Tadesse. 2001. Prospects for sustainable use and development of wild food plants in Ethiopia. Economic Botany 55(1):47-62. Asgedom, Yemane and Mekonnen Yohannes. 1996. Vegetation resources of central Tigray. In Öyhus, Arne Olav, and Gebremedhn Gebru (eds.), Rural exploratory studies in the central zone of Tigray, Northern Ethiopia: proceedings of a workshop. Addis Ababa: Noragric, pp. 35-44. Chapter 1 Ashenafi, Zelealem Tefera and N. Leader-Williams. 2005. Indigenous common property resource management in the Central Highlands of Ethiopia. Human Ecology, 33(4): 539-563. Chapter 2 Atran, Scott. 1994. Core domains versus scientific theories. In L. Hirschfeld & S. Gelman (eds.), Mapping the mind: Domain-specificity in cognition and culture. New York: Cambridge University Press. Aune, J.B., Bussa, M.T., Asfaw, F.G. and A.A. Ayele. 2001. The ox ploughing system in Ethiopia: can it be sustained? Outlook on Agriculture, 30(4):275-280. Azene, B.T., Brinie, A. and B. Tengnas. 1993. Useful trees and shrubs for Ethiopia. Identification, propagation and management for agricultural and pastoral communities. Regional Soil Conservation Unit, SIDA, Sweden. Barnett, Anna. 2001. Northern Ethiopia: what’s on the menu? Nutrition and Food Science 31(1):35-37. Bauer, Dan. 1985. Household and society in Ethiopia: an economic and social analysis of Tigray social principles and household organization. East Lansing, Michigan, Michigan State University African Studies Center Occasional Papers Series Northeast African Studies Committee, No. 6. Benin, Samuel and John Pender. 2002. Community management of grazing lands and impact on environmental degradation in the Ethiopian highlands. Presented at the Ninth Conference of the International Association for the Study of Common Property, Victoria Falls, Zimbabwe, June 17-21, 2002. Berhanu, Mulugeta. 2004. Institutions and collective action in community property resource management: the case of community woodlots in Hawzien, Tigray Region. Masters research paper, Institute of Social Studies, The Hague. Berlin, Brent. 1992. Ethnobiological classification. Principles of categorization of plants and animals in traditional societies. Princeton, NJ: Princeton University Press. Beyene, Atakilte. 2003. Soil conservation, land use and property rights in Northern Ethiopia. PhD Dissertation, Agraria 395. Swedish University of Agricultural Sciences, Dept. of Rural Development Studies, Uppsala. Beyene, Atakilte. n.d. Land tenure institutions and the smallholder-farming systems: challenges for agricultural development. Paper prepared for the 12th Annual Conference on the Ethiopian Economy, n.d. Online at: http://www.addischaber.com/ downloads/ docs1/14.doc (accessed 19-5-2006). Binggeli, P., Desalegn Desissa, Healey, J., Painton, M., Smith, J. and Zewge Teklehaimanot. 2003. Conservation of Ethiopian sacred groves. ETFRN News, 38, 37-38. Bongers, F., Wassie, A. Sterck, F. J., Bekele, T., and D. Teketay. 2006. Ecological restoration and church forests in northern Ethiopia. Journal of the Drylands 1(1): 35-44. 159 Chapter 3 Boster, J.S. and S. Weller. 1990. "Cognitive and contextual variation in hot-cold classification." American Anthropologist 92:171-179 Bousfield, W.A. and W.D. Barclay. 1950. The Relationship between order and frequency of occurrence of restricted associative responses. Journal of Experimental Psychology 40:643-647. Bruce, John, 1989, Community forestry rapid appraisal of tree and land tenure. Community Forestry Note 5. Rome: Food and Agriculture Organization of the United Nations (FAO). Cabal, C.R. 2000. Household food energy intake of semi-subsistent households in integrated croplivestock agricultural systems in the central highlands of Ethiopia. PhD Dissertation, University of Hawaii. Ann Arbor, Michigan: University Microfilms International. Chisholm, Nick. 2000. Maintaining a fragile balance: community management of renewable natural resources in Tigray, NE Ethiopia. Presented at Presented at: "Constituting the Commons: Crafting Sustainable Commons in the New Millenium", the Eighth Conference of the International Association for the Study of Common Property, Bloomington, Indiana, USA, May 31June 4. Chisholm, Nick. 2004. How should the commons be governed? Household preferences for CPR management regimes in Tigray, northern Ethiopia. Paper presented at the Tenth Biennial Conference of the International Association for the Study of Common Property, Oaxaca, Mexico, 1013 August 2004. Deffar, Girma. 1998. Non-wood forest products in Ethiopia. EC-FAO Partnership Programme Project GCP/INT/679/EC. Addis Ababa: European Commission Directorate-General VIII Development and the Food and Agriculture Organization of the United Nations (FAO), December. Online at: http://www.fao.org/docments/show_cdr.asp?url_file=/DOCREP/003/X6690E/ 6690E02.htm (accessed 5 June 2006). Ellen, Roy. 1993a. Rhetoric, practice and incentive in the face of the changing times: A case study in Nuaulu attitudes to conservation and deforestation. In K. Milton (Ed.), Environmentalism: The View from Anthropology. London and New York: Routledge. Ellen, Roy. 1993b. The cultural relations of classification. An analysis of Nuaulu animal categories from central Seram. Cambridge: Cambridge University Press. Ellen, Roy. 1998. Indigenous knowledge of the rainforest: Perception, extraction and conservation. In B. Maloney (Ed.), Destruction and Development of the Tropical Rainforest. Dordrecht: Kluwer, pp. 87-99. Ellis, Frank. 2000. Rural livelihoods and diversity in developing countries. Oxford: Oxford University Press. Ennouri, M., H. Fetoui, E. Bourret, N. Zeghal and H. Attia. 2006. Evaluation of some biological parameters of Opuntia ficus indica. 1. Influence of a seed oil supplemented diet on rats. Bioresour Technol. 97(12):1382-6. Esser, K.B., Vagen, T.G., Yibabe, T., and M. Haile. 2002. Soil conservation in Tigray. Noragric Report No. 5, Noragric. Agricultural University of Norway. FAO. n.d. FAO Government Co-operative Programme. Project of the Government of Ethiopia. Project Document GCP/ETH/060/BEL. Improving Nutrition and Household Food Security in Northern Shoa and Southern Zone of Tigray, Ethiopia. Rome: Food and Agriculture Organization of the United Nations. FAO. 2004a. Baseline survey report. Nutrition and household food security in Northern Shewa and Southern Tigray. Final Version. Project/GCP/ETH/060/BEL, Addis Ababa: FAO, March. FAO, 2004b. Livestock sector brief. Ethiopia. Livestock Information, Sector Analysis and Policy Branch (AGAL), May. Online at http://www.fao.org/ag/againfo/rsources/en/publications/sector_ briefs/ lsb_ETH.pdf (accessed 26 May 2006). Fafchamps, Marcel and Agnes R. Quisumbing. 2001. Control and ownership of assets within rural Ethiopian households. Food Consumption and Nutrition Division Discussion Paper No. 120. Washington, DC: International Food Policy Research Institute, August. Fassil, Hareya. 2002. Beyond Plants, professionals and parchments: The role of home-based traditional health knowledge and medicinal plant use in primary health care in Ethiopia. Ethnobotany Research and Applications 3(1): 37-50. Online at: http://www.ethnobotanyjournal.org/ vol3/ i1547-3465-03-037. pdf (accessed 5 June 2006). 160 Feoli, E., Veurich, L., and W. Zerihun. 2002. Evaluation of environmental degradation in northern Ethiopia using GIS to integrate vegetation, geomorphological, erosion and socio-economic factors. Agriculture, Ecosystems and Environment, 91(1/3): 313-325. Chapter 4 Freeman, Linton C., Romney, A. Kimball, & Sue C. Freeman. 1987. Cognitive structure and informant accuracy. American Anthropologist 89:310-325. Fullerton Joireman, Sandra. 1996. Contracting for land: lessons from litigation in a communal tenure area of Ethiopia. Canadian Journal of African Studies 30(2):214-232. Gebre Egziabher, Kiros and Solomon Tegegne, 1996, Geblen, Subhsaesie Woreda, Tigray. In Bevan, Philippa and Alula Pankhurst (eds), Ethiopian village studies. Dept. of Sociology, Addis Ababa University, Ethiopia and Centre for the Study of African Economies, Oxford, UK, June. Gebremedhin, Berhanu, Pender, John, and Tesfaye Girmay. 2000. Community natural resource management: the case of woodlots in Northern Ethiopia. EPTD Discussion Paper, No. 60. Washington, DC: International Food Policy Research Institute. Gebremedhin, Berhanu, Pender, John, Ehui, Simeon, and Mitiku Haile (eds.). 2002. Policies for sustainable land management in the highlands of Tigray, Northern Ethiopia. EPTD Workshop Summary No. 14. Washington, DC: International Food Policy Research Institute. Gebremedhin, Berhanu, Pender, John and Tesfaye Girmay. 2004. Collective action for grazing land management in crop-livestock mixed systems in the highlands of northern Ethiopia. Agricultural Systems 82(3):273-290. Gibbs, G. 2002. Qualitative data analysis: explorations with NVivo. London: Open University Press. Grivetti, L. E. and B. M. Ogle. 2000. Value of traditional foods in meeting macro and micronutrient needs: the wild plant connection. Nutrition Research Reviews, 13(1): 31-46. Guinand, Y.F. and D. Lemessa. 2000. Wild-food plants in Southern Ethiopia: Feflections on the role of ‘famine-foods’ at a time of drought. United Nations Development Programme. Emergencies Unit for Ethiopia. Online at: http://www.sas.upenn.edu/ African_Studies/eue_web/ famp0300.htm (accessed 2-11-2004). Hildebrand, Elisabeth. 2003. Motives and opportunities for domestication: an ethnoarcheological study in southwest Ethiopia. Journal of Anthropological Archeology, 22: 358-375. Hoben, Allen. 1973. Land tenure among the Amhara of Ethiopia: the dynamics of cognatic descent. Chicago, IL: University of Chicago Press. Howard, Patricia. 2003a. The major importance of ‘minor’ resources: women and plant Biodiversity, IIED Gatekeeper Series, Sussex, UK: International Institute for Environment and Development. Howard, Patricia, 2003b, Women and the plant world: An exploration. In P. Howard (ed.), Women and plants. Gender relations in biodiversity management and conservation. London & New York: Zed Press and Palgrave Macmillan, 2003, pp. 1-48. Howard, Patricia and Gorettie Nabanoga, In press. Transgressing boundaries: Gendered rights within forested landscapes in central Uganda. World Development. Jagger, Pamela, Pender, John, and Berhanu Gebremedhin. 2004. Woodlot devolution in Northern Ethiopia: Opportunities for empowerment, smallholder income diversification, and sustainable land management. Discussion Paper No. 107. Washington, DC: International Food Policy Research Institute. Jansen, P.C.M. 1981. Spices, condiments and medicinal plants in Ethiopia, their taxonomy and agricultural significance. PhD Dissertation, Landbouwhogeschool Wageningen. Wageningen, the Netherlands: PUDOC. Johns, Timothy and Bhuwon Sthapit. 2004. Biocultural diversity in the sustainability of developingcountry food systems. Food and Nutrition Bulletin, 25(2): 143-155. Keeley, James and Ian Scoones. 2000. Knowledge, power and politics: the environmental policymaking process in Ethiopia. Journal of Modern African Studies 28(1):89-120. Kenyatta, Catherine and Amiee Henderson (eds). 2001. The potential of indigenous wild foods. Workshop proceedings, 22-26 January 2001. USAID/OFDA and CRS Southern Sudan, April, pp. 31-45. Kew Royal Botanical Gardens. n.d. The collection, collation and verification of nutritional information on Ethiopian wild food plants. Online at http://www.rbgkew.org.uk/scihort/ecbot-harvpilot.html (accessed 5 June 2006). Klappa, Stephanie. 2006. Fallow farming: exploring subsistence in Krisa, far northwest Papua New Guinea, and beyond. PhD Dissertation, Dept. of Anthropology, University of Kent at Canterbury. 161 Kundhlande, G., Luckert, M.K. 1998. Towards an analytical framework for assessing property rights to natural resources: a case study in one economy. Unpublished manuscript. University of Alberta, Edmonton, Canada. Letchamo, Wudeneh and Hermann Storck. 1991. Spices, aromatic and medicinal plants trade in eastern Ethiopia. Acta Horticulturae 270: 247-253. Lykke, A.M., Kristensen, M.K. and S. Ganaba. 2004. Valuation of local use and dynamics of 56 woody species in the Sahel. Biodiversity and Conservation 13:1961-1990. McBurney, Rory P., Griffin, Caroline, Alison A., Paul, and David C. Greenberg. 2004. The nutritional composition of African wild food plants: from compilation to utilization. Journal of Food Composition and Analysis 17:277-289. Meehan, Fiona. 2004. Female headed households in Tigray, Ethiopia. A study review. DCG Report No. 35. Drylands Coordination Group c/o NORAGRIC, December. Mersha, Gebru and Mwangi wa Githinji. 2005. Untying the Gordian knot: The question of land reform in Ethiopia. ISS/UNDP Land, Poverty and Public Action Policy Paper No. 9. The Hague: Institute of Social Studies, 15 October. Mengistu, Alemayehu, n.d. FAO Country Pasture/Forage Resource Profiles Ethiopia. Online at http://www.fao.org/ag/AGP/AGPC/doc/Counprof/Ethiopia/Ethiopia.htm (accessed 26 May 006). Mwendera, E. J. and M. A. M. Saleem, 1997. Hydrologic response to cattle grazing in the Ethiopian highlands. Agriculture, Ecosystems and Enviroment, 64(1): 33-41. Mwendera, E. J., Saleem, M.A.M. and Z. Woldu. 1997. Vegetation response to cattle grazing in the Ethiopian highlands. Agriculture, Ecosystems and Environment, 64(1): 43-51. Nedessa, Bertru, Ali, Jawad, and Ingrid Nyborg. 2005. Exploring ecological and socio-economic issues for the improvement of Area Enclosure management. A case study from Ethiopia. Drylands Coordination Group Report No. 38, May 2005. Oslo: DCG, May. Online at: www.drylandsgroup.org/noop/file.php?id=535&d=1 (accessed 11 June 2006). Pankhurst, H. 1992. Gender, development and identity: an Ethiopian study. London (UK): Zed Books. Quisumbing, A. and Y. Yohannes. 2004. How fair is workfare? Gender, public works, and employment in rural Ethiopia. World Bank Policy Research Working Paper No. 3492. Available at SSRN: http://ssrn.com/abstract=650201 (accessed 2 August 2006). Sandford, Judith and Helen Kassa. 1994. The effect of gender on resource contribution, decision making and influence: a comparison between enset, teff and maize. In Sandford, Stephen (ed.), Gender, structure and landraces in peasant enset plantations in North Omo. FRP Technical Pamphlet No. 6. Addis Ababa, Ethiopia. Selinus, Ruth. 1971. The traditional foods of the Central Ethiopian highlands. Research Report No. 7, Uppsala, Scandinavian Institute of African Studies. Sharp, Kay, Deveraux, Stephen and Yared Amare. 2000. Destitution in Ethiopia’s Northeastern Highlands (Amhara National Regional State). Final Report. London & Addis Ababa: Institute of Development Studies and Save the Children, April. Sheil, D., Puri, R.K., Basuki, I., Heist, M. van, Syaefuddin, R., Sardjono, M.A., Samsoedin, I., Sidiyasa, K., Dhrisandini, Permana, E., Angi, E.M., Gatzweiler, F., Johnson, B., and A. Wijaya. 2002. Exploring biological diversity, environment and local people's perspectives in forest landscapes. Methods for a multidisciplinary landscape assessment. Jakarta, Indonesia: Center for International Forestry Research (CIFOR). Shitarek, T., Manaye, S., and B. Abebe. 2001. Strengthening user-rights over local resources in Wollo, Ethiopia. Issue Paper, Drylands Programme, International Institute for Environment and Development. No.103. Shylendra, H.S. 2002. Environmental rehabilitation and livelihood impact: emerging trends from Ethiopia and Gujarat. Economic and Political Weekly 3 August: 3286-3292. Tadesse, Z. 2003. Women and land rights in the Third World: the case of Ethiopia. In Wanyeki, Lynne Muthoni (ed.), Women and land in Africa: Culture, religion and realizing women's rights. London and New York: Zed Press, pp. 67-95. Teshome, Woldeab. 2003. Irrigation practices, state intervention and farmers' life-worlds in drought-prone Tigray, Ethiopia. PhD Dissertation, Wageningen University, the Netherlands. Teshome-Bahiru, Wondwosen. 2005. Impacts of urbanization on the traditional medicine of Ethiopia. Anthropologist 8(1): 43-52. 162 Tewodros, Assefa and Yared Derbew. 1996. Harresaw, Atsbi Woreda, Tigray. In Bevan, Philippa and Alula Pankhurst (eds), Ethiopian village studies. Dept. of Sociology, Addis Ababa University, Ethiopia and Centre for the Study of African Economies, Oxford, UK, June. Tiruneh, Addis. 2001. Socio-economic and gender issues in the selection, conservation and utilization of sorghum landraces in Ethiopia. CERTWID/IDR Addis Ababa University, December. Tiruneh, Addis, Mwangi, Wilfred, Verkuijl, Hugo and Teklu Tesfaye. 1999. Gender differentials in adoption of improved wheat varieties in Ada, Lume and Gimbichu woredas of the Central Highlands of Ethiopia. Mexico, DF: International Maize and Wheat Improvement Center (CIMMYT) and Ethiopian Agricultural Research Organization (EARO). Tiruneh, Addis, Tesfaye, Teklu, Mwangi, Wilfred, and Hugo Verkuijl. 2001. Gender differentials in agricultural production and decision-making among smallholders in Ada, Lume, and Gimbichu woredas of the central highlands of Ethiopia. Addis Ababa University, Center for Education, Research and Training on Women in Development, Addis Ababa, Ethiopia. Tsegaye, Bayush. 1997. The significance of biodiversity for sustaining agricultural production and the role of women in the traditional sector: the Ethiopian experience. Agriculture, Ecosystems and the Environment 62(2-3):215-227. Vivero Pol, Jose Luis. n.d. The livelihood value of forests in Ethiopia. Non-wood forest products for the food security of rural households. Paper to be presented to Natural Resource Perspectives ODI Network. Addis Ababa. Wassie, A., Teketay, D. and N. Powell. 2004). Ethiopian Orthodox Tewahido Church Forests: Their role in Restoring Forest Ecosystem in the degraded highlands of Northern Ethiopia. Proceedings of the 16 th annual Society for Ecological Restoration, Victoria, B.C.,Canada, 24-26 August 2004. Williams, Paula, Gebrehiwot, Alemayehu, Mariam, Ayele Gebre, and Berhanu Mengesha. 2003. The role of forestry in food security and income generation in Ethiopia: current situation and future potentials. EC/FAO Cooperative Programme to Support Food Security GCP/INT/758/EC/ ETH (9). Addis Ababa: FAO. Wolde-Georgis, Tsegay. 1996. Survival strategies and coping mechanisms of peasants to famine: the case of Tigray, Ethiopia. PhD Dissertation, University of Denver (Colorado). Ann Arbor, Michigan: University Microfilms International. Woldegiorgis, Original. 1999. Land tenure and gender. Paper presented at the National Workshop on Food Security through Sustainable Land Use: Policy and Implementation Issues on Institution, Land Tenure and Extension. January 14-15, 1999, Addis Ababa, Ethiopia. Organized by NOVIB Partners Forum on Sustainable Land Use, sponsored by SNV Ethiopia. Websites Consulted for Ethnobotanical Data: Laboratory for Forest, Nature and Landscape Research, Katholieke Universiteit Leuven. http://www.biw.kuleuven.be/lbh/lbnl/forecoman/eng/projbeschrijving.asp?n=4 Guinand, Y.F. and D. Lemessa. Ethiopia: Famine Food Field Guide. Online at http://www.africa.upenn.edu/faminefood/category3/cat3_Ficus_vasta.htm (accessed 10 June 2006). Chapter 5 Metadata African Organization - Information System (METAFRO). Prelude Medicinal Plants Database. http://www.metafro.be/prelude (accessed 10 June 2006). PROTA (Plant Resources of Tropical Africa) Online Database. See http://www. prota.org/ 163 164 Appendix Tables Table 10A Source and Tenure of Water for Domestic and Agricultural Use, by Sex of the Household Head SOURCE/TENURE MH % FH % TOT. Private domestic source Well 2 13.3 0 2 River/stream 0 1 6.7 1 Artesian 0 0 0 Hand pump 0 0 0 Other 0 0 0 Total 2 13.3 1 6.7 3 Public domestic source Well 0 2 13.3 2 River/stream 4 26.7 9 60.0 13 Artesian 2 13.3 1 6.7 3 Hand pump 2 13.3 0 2 Other 0 1 6.7 2 Total 8 53.3 13 86.7 21 Communal domestic source Well 0 0 0 River/stream 1 6.7 1 6.7 2 Artesian 4 26.7 0 4 Hand pump 0 0 0 Other 0 0 0 Total 5 33.3 1 6.7 6 Private agricultural source Well 1 6.7 1 6.7 2 River/stream 0 0 0 Artesian 0 0 0 Hand pump 0 0 0 Other 0 0 0 Total 1 6.7 1 6.7 2 Public agricultural source Well 0 1 6.7 1 River/stream 10 66.7 4 26.7 14 Artesian 0 0 0 Hand pump 1 6.7 3 20.0 4 Other 0 0 0 Total 11 73.3 8 53.3 19 Communal agricultural source Well 1 6.7 0 1 River/stream 0 0 0 Artesian 0 0 0 Hand pump 0 0 0 Other 0 0 0 Total 1 6.7 0 1 % 6.7 3.3 10.0 6.7 43.3 10.0 6.7 6.7 70.0 6.7 13.3 20.0 6.7 6.7 3.4 46.7 13.3 63.3 3.3 3.3 165 Table 19A Number of Household Members by Total Landholdings TOTAL LAND/ 1-2 3-4 5-6 7+ HOUSEHOLD No. % No. % No. % No. % MEMBERS FH Landless 1 20.0 .25 to < 1 Timad 2 40.0 1 12.5 1 50.0 1 to 2 Timads 2 40.0 5 62.5 2.1 to 5 Timads 2 25.0 1 50.0 > 5 Timads Total 5 100.0 8 100.0 2 100.0 MH Landless 1 14.3 .25 to < 1 Timad 1 100.0 1 14.3 1 to 2 Timads 1 33.3 1 14.3 2.1 to 5 Timads 1 33.3 4 57.1 1 25.0 > 5 Timads 1 33.3 3 75.0 Totals 1 100.0 3 99.9 7 100.0 4 100.0 Table 20A Quality of Household Land Allocated by the Baito by Sex of Household Head MH FH TOTAL QUALITY No. % No. % No. % Soil Quality Lem (good) 22 56.4 5 23.8 27 45.0 Lem-teuf (medium) 10 25.6 13 61.9 23 38.3 Teuf (poor) 7 17.9 3 14.3 10 16.7 Total 39 100.0 21 100.0 60 100.0 Slope Medda (flat) 35 89.7 17 81.0 52 86.7 Dagath-ama (moderate in4 10.3 1 4.8 4 6.7 cline) Geddel (steep) 0 0.0 3 14.3 4 6.7 Totals 39 100.0 21 100.1 60 100.1 Table 21A Quality of Total Household Land by Sex of Household Head (n=30) MH FH TOTAL QUALITY No. % No. % No. % Soil quality Lem (good) 33 51.5 12 38.7 45 47.4 Lem-teuf (medium) 23 36.0 13 42.0 36 37.9 Teuf (poor) 8 12.5 6 19.3 14 14.7 Totals 64 100.0 31 100.0 95 100.0 Slope Medda (flat) 57 89.1 25 80.6 82 86.3 Dagath-ama (moderate in7 10.9 3 9.7 10 10.5 cline) Geddel (steep) 0 0.0 3 9.7 3 3.2 Totals 64 100.0 31 100.0 95 100.0 166 Table 22A Location of Agricultural Plots by Sex of Household Head LOCATION WITH RESPECT MH FH TOTAL TO THE HOMESTEAD No. % No. % No. % Within homestead 21 32.8 12 37.5 33 34.4 Under 5 minute walk 9 14.1 8 25.0 17 17.7 5 to 10 minute walk 16 25.0 2 6.2 18 18.7 Above 10 to 30 minute walk 7 10.9 3 9.4 10 10.4 Further than 30 minute walk 11 17.2 7 21.9 18 18.8 Totals 64 100.0 32 100.0 96 100.0 Table 25A Intra-Household Decision Making by Who Acquired the Plot HEAD HEAD & HEAD & TOTAL WHO ACQUIRED SPOUSE TENANT PLOT No. % No. % No. % No. % Who Decides What to Grow Male head 0 0.0 10 76.9 0 0.0 10 37.0 Wife 1 8.3 2 15.4 0 0.0 3 11.1 Both head & wife 0 0.0 1 7.7 0 0.0 1 3.7 Female head 11 91.7 2 100.0 13 48.1 Totals 12 100.0 13 100.0 2 100.0 27 100.0 Who can Give Away Male head 0 0.0 9 69.2 1 100.0 10 37.0 Wife 0 0.0 3 23.1 0 0.0 3 11.1 Both head & wife 0 0.0 1 7.7 0 0.0 1 3.7 Female head 13 100.0 0 0.0 13 48.1 Totals 12 100.0 13 100.0 1 100.0 27 100.0 Who can Rent or Sharecrop Out Male head 0 0.0 7 63.6 2 100.0 9 34.6 Wife 0 0.0 3 27.2 0 0.0 3 11.5 Both head & wife 0 0.0 1 9.1 0 0.0 1 3.8 Female head 13 100.0 13 50.0 Totals 13 100.0 11 99.9 2 100.0 26 99.9 *One case was omitted where a male head reported that it is another landowner who can decide to rent or sharecrop out the plot that he acquired. 167 Table 27A Division of Agricultural Labour in Female Headed Households ONLY ONLY HEAD & HEAD & OTHER HEAD & HIRED HEAD SHARESHAREOTHER LANDED HIRED LANDED CROPPER CROPPER LANDED MALE LANDED MALE TASK MALE ALONE MALE ALONE Ploughing 1 8 2 0 2 Planting 1 8 2 2 Cultivation 3 3 4 1 2 Fertiliser/ pesticide 1 8 1 2 Bird scaring 4 7 1 Harvesting 1 8 1 1 1 Threshing 1 8 1 3 Drying, preparation 5 2 2 1 1 for storage* Storing 8 2 1 1 Transporting crops from 4 2 3 1 2 fields** Seed selection & Storage 11 1 Crop sales 12 *In one case, another female labourer and a hired landed male labourer performed the task. **In one case, the female head and her son performed the task. Table 29A Use of Crop Residues Harvested by Sex of Household Head MH FH TOTAL USE No. % No. % No. % Animal feed 31 100.0 2 33.3 33 89.2 Other 0 0.0 4 66.6 4 10.8 Totals 31 83.8 6 16.2 37 100.0 Table 30A Total Household Landholdings by Amount of Crop Residues Harvested (n= 27) TOTAL LAND AREA/ .1 TO .99 1 TO 2 2.1 TO 5 5.1+ TIMADS TIMAD TIMADS TIMADS AMOUNT OF CROP No. % No. % No. % No. % RESIDUES None 4 33.3 6 50.0 2 16.7 0 0.0 3-5 shekims 0 0.0 1 33.3 2 66.6 0 0.0 6-15 shekims 1 14.3 1 14.3 5 71.4 0 0.0 16-33 shekims 0 0.0 0 0.0 0 0.0 4 100.0 168 Table 31A Soil and Water Conservation Measures on Plots by Person(s) Responsible HEAD & RESPONSIBILITY/ MALE HEAD HEAD & WIFE OTHER HH FEMALE MEASURES MEMBER HEAD No. % No. % No. % No. % Soils Bunds (n = 70) 8 11.4 40 57.1 5 7.1 17 24.3 Terraces (n = 2) 1 50.0 0 0.0 0 0.0 1 50.0 Total Soils 9 40 5 18 Water Ponds (n = 65) 2 3.1 41 63.1 4 6.2 18 27.7 Dams (n = 1) 1 100.0 0 0.0 0 0.0 0 0.0 Total Water 3 41 4 18 Total measures (n=138) 12 81 9 36 26.1 Table 32A Crops Planted During Meher, by Area Planted and Sex of Household Head CROP AND MH FH TOTAL MH TOTAL FH TOTAL AREA No. No. Total Total Total % (TIMADS) Plots % Plots % Area % Area % Area Area White teff 10.0 19.0 0.25 1.6 10.3 15.0 .1 to .5 1 10.0 1 to 2 8 23.5 Red Teff 1.50 2.9 0.0 0.0 1.50 2.2 1 to 2 1 2.9 Wheat 20.0 38.1 12.9 81.1 32.9 48.1 .1 to .5 1 2.9 2 20.0 1 to 2 6 17.6 3 30.0 3 to 5 3 8.8 Barley 15.0 28.6 2.5 15.7 17.5 25.6 .1 to .5 1 10.0 1 to 2 7 20.6 2 20.0 3 to 5 2 5.9 Lentils 0.0 0.0 0.25 1.6 0.25 0.4 .1 to .5 1 10.0 Peas 4.0 7.6 0.0 0.0 4.0 5.8 1 to 2 4 11.8 Qumu 1.0 1.9 0.0 0.0 1.0 1.5 1 to 2 1 2.9 Other 1.0 1.9 0.0 0.0 1.0 1.5 1 to 2 1 2.9 Totals 34 100.0 10 100.0 52.5 100.0 15.9 100.0 68.4 100.0 169 Table 33A Crop Combinations, Meher Season, by Sex of Household Head (n=28) MH FH CROP COMBINATIONS No. % No. % Red teff, wheat, peas, qumu 1 8.3 0 0.0 Teff, barley, wheat & peas 2 16.7 0 0.0 Teff, barley & wheat 5 41.7 0 0.0 Teff, barley & peas 1 8.3 0 0.0 Teff, wheat 1 8.3 0 0.0 Barley, wheat 1 8.3 2 14.3 Only teff 0 0.0 1 7.1 Only barley 0 0.0 1 7.1 Only wheat 0 0.0 9 64.3 Only lentils 0 0.0 1 7.1 Only peas 1 8.3 0 0.0 Only qumu 1 8.3 0 0.0 Totals 12 100.0 14 100.0 Table 34A Total Meher Crop Area as a Percent of Total Landholdings by Sex of Household Head (n=28) MH FH TOTAL TOTAL MEHER CROP AREA AS A % OF No. % No. % No. % TOTAL LAND None 2 14.3 0 0.0 2 7.1 50% 0 0.0 3 21.4 3 10.7 51-75% 1 7.1 2 14.3 3 10.7 80-90% 3 21.4 4 28.6 7 25.0 91-100% 7 50.0 5 35.7 12 42.8 330% 1 7.1 0 0.0 1 3.6 Totals 14 99.9 14 100.0 28 99.9 170 Table 35A Total Quintals Consumed per Meher Crop and Average Per Household Producing that Crop, by Sex of Household Head NO. AVG/ TOTAL PRODUCING HOUSEHOLD MEHER CROP CONSUMED HOUSEHOLDS (QUINTALS) MH White Teff (Eragrostis teff) 8.05 6 1.34 Red Teff (Eragrostis teff) .50 1 .5 Barley (Hordeum vulgare) 5 2.55 12.75 Wheat (Triticum vulgare) 20.75 9 2.31 Peas (Pisum sativum) Missing 2 Lentils (Lens escuelenta) 0 0 Qumu (Sclerocarya bierra) 1 2.50 2.50 FH White Teff .5 1 .5 Red Teff 0 0 Barley 2.0 3 .67 Wheat 9.5 11 .86 Peas 0 0 Lentils .5 1 .5 Qumu 0 0 - Table 36A Total Kilos of Own Crops Consumed By Household Dependency Ratios (n=26) DEPENDENCY RATIO/TOTAL .2 TO .75 1.00 1.33 TO 1.5 2.0 TO 4.0 TOTALS CONSUMPTION No. % No. % No. % No. % No. % 50 75-100 150-300 305-500 500-1000 Totals 2 3 2 2 1 10 25.0 50.0 33.3 50.0 50.0 38.5 3 0 0 1 1 5 37.5 25.0 50.0 19.2 1 1 2 0 0 4 12.5 16.7 33.3 15.4 2 2 2 1 0 7 25.0 33.3 33.3 25.0 26.9 8 6 6 4 2 26 30.8 23.1 23.1 15.4 7.7 100.0 171 Table 37A Amount of Male Household Labour 16-64 Years by Total Household Crop Consumption in Kilos 0 1 2 3 TOTAL NO. ADULT MALES/ N % No % No % No % No % TOTAL CONSUMED o . . . . Distribution of Male Labour Ability by Consumption Strata 50 7 87.5 1 12.5 0 0.0 0 0.0 8 100.0 75-100 4 66.7 1 16.7 1 16.7 0 0.0 6 100.0 150-300 2 33.3 3 50.0 1 16.7 0 0.0 6 100.0 305-500 0 0.0 2 50.0 2 50.0 0 0.0 4 100.0 550-1000 0 0.0 0 0.0 1 50.0 1 50.0 2 100.0 Total 13 50.0 7 26.9 5 19.2 1 3.8 26 100.0 Distribution of Consumption Strata by Male Labour Availability 50 7 53.8 1 14.3 0 0.0 0 0.0 8 30.8 75-100 4 30.8 1 14.3 1 20.0 0 0.0 6 23.1 150-300 2 15.4 3 42.9 1 20.0 0 0.0 6 23.1 305-500 0 0.0 2 28.6 2 40.0 0 0.0 4 15.4 550-1000 0 0.0 0 0.0 1 20.0 1 100.0 2 7.7 Total 13 100.0 7 100.0 5 100.0 1 100.0 26 100.0 NUMBER OF OXEN/MEHER CROP AREA Table 38A Number of Oxen by Crop Consumption in Kilos NO OXEN ONE OX TWO OXEN % No. % No. % No. 50 kg 75-100 kg 150-300 kg 305-500 kg >500 kg Totals 8 4 1 0 0 13 50 kg 75-100 kg 150-300 kg 305-500 kg >500 kg Totals 8 4 1 0 0 13 172 TOTAL No. % Distribution of Oxen Ownership by Consumption Strata 100.0 0 0.0 0 0.0 8 66.7 1 16.7 1 16.7 6 16.7 4 66.7 1 16.7 6 0.0 1 25.0 3 75.0 4 0.0 0 0.0 2 100.0 2 50.0 6 23.1 7 26.9 26 Distribution of Consumption Strata by Oxen Ownership 61.5 0 0.0 0 0.0 8 30.8 1 16.7 1 20.0 6 7.7 4 66.7 1 20.0 6 0.0 1 16.7 3 60.0 4 0.0 0 0.0 2 40.0 2 100.0 6 100.0 5 100.0 24 100.0 100.0 100.0 100.0 100.0 100.0 33.3 25.0 25.0 16.7 8.3 100.0 NO. OXEN/ MEHER CROP AREA Table 39A Number of Oxen by Total Meher Crop Area NO OXEN ONE OX TWO OXEN No. % No. % No. % .25 - < 1 timad 1 - 2 timads 2.1 - 4 timads > 4 timads Totals 4 9 0 0 13 .25 - < 1 timad 1 - 2 timads 2.1 - 4 timads > 4 timads Totals 4 9 0 0 13 TOTAL No. % Distribution of Oxen Ownership by Crop Land Cultivated 100.0 0 0.0 0 0.0 4 81.8 2 18.2 0 0.0 11 0.0 3 60.0 2 40.0 5 0.0 1 16.7 5 83.3 6 50.0 6 23.1 7 26.9 26 Distribution of Crop Land Cultivated by Oxen Ownership 30.8 0 0.0 0 0.0 4 69.2 2 33.3 0 0.0 11 0.0 3 50.0 2 28.6 5 0.0 1 16.7 5 71.4 6 100.0 6 100.0 7 100.0 26 100.0 100.0 100.0 100.0 100.0 15.4 42.3 19.2 23.1 100.0 Table 40A Total Crop Harvest in Quintals of Female Headed Households and Share of Harvest for Tenants (n=14) TOTAL SHARE NONE .5 QUINTALS 1 QUINTAL FOR TENANTS/ No. % No. % No. % HARVEST .5 3 42.9 0 0.0 0 0.0 1.0 1 14.3 5 83.3 0 0.0 1.5 1 14.3 1 16.7 0 0.0 2.0 0 0.0 0 0.0 2 100.0 3.0 1 14.3 0 0.0 0 0.0 Totals 7 46.7 6 40.0 2 13.3 173 Table 43A Total Farmland by Total Number of Tropical Livestock Units (n=18) TLUS/ .01 TO .5 1 TO 2 TLUS 2.1 TO 5 > 5 TLUS TOTAL TLUS LANDHOLDING TLUS TLUS STRATA No. % No. % No. % No. % No. % Landless .25 - < 1 timad 1 - 2 timads 2.1 to 5 timads > 5 timads Totals 0 1 1 0 0 2 0.0 33.3 25.0 0.0 0.0 10.5 Landless .25 - < 1 timad 1 - 2 timads 2.1 to 5 timads > 5 timads Totals 0 1 1 0 0 2 0.0 50.0 50.0 0.0 0.0 100.0 174 Distribution of TLUs by Farmland Strata 1 100.0 0 0.0 0 0.0 2 66.7 0 0.0 0 0.0 3 75.0 0 0.0 0 0.0 1 14.3 4 57.1 2 28.6 2 50.0 1 25.0 1 25.0 9 47.4 5 26.3 3 15.8 Distribution of Farmland Strata by TLUs 1 11.1 0 0.0 0 0.0 2 22.2 0 0.0 0 0.0 3 33.3 0 0.0 0 0.0 1 11.1 4 80.0 2 66.7 2 22.2 1 20.0 1 33.3 9 100.0 5 100.0 3 100.0 1 3 4 7 4 19 100.0 100.0 100.0 100.0 100.0 100.0 1 3 4 7 4 19 5.3 15.8 21.1 36.8 21.1 100.0 Table 44A Control Over Livestock within Male Headed Households (n = 15) HEAD & SPOUSE WHO/TYPE OF HEAD SPOUSE ONLY TOTALS LIVESTOCK No. % No. % No. % No. % Who Can Sell or Give Away Calves 1 20.0 4 80.0 0 0.0 5 9.6 Bulls 2 33.3 4 66.6 0 0.0 6 11.5 Oxen 4 36.7 7 63.6 0 0.0 11 21.2 Heifers 0 0.0 2 100.0 0 0.0 2 3.8 Cows 2 33.3 4 66.6 0 0.0 6 11.5 Sheep 0 0.0 2 100.0 0 0.0 2 3.8 Goats 2 33.3 4 66.6 0 0.0 6 11.5 Donkeys 2 33.3 4 66.6 0 0.0 6 11.5 Chickens 0 0.0 3 37.5 5 62.5 8 15.4 Totals 13 25.0 34 65.4 5 9.6 52 100.0 Who Keeps Offspring Calves* Bulls 0 0.0 5 100.0 0 0.0 5 12.2 Oxen 0 0.0 9 100.0 0 0.0 9 22.0 Heifers 0 0.0 2 100.0 0 0.0 2 4.9 Cows 0 0.0 6 100.0 0 0.0 6 14.6 Sheep 0 0.0 1 100.0 0 0.0 1 2.4 Goats 0 0.0 5 100.0 0 0.0 5 12.2 Donkeys 0 0.0 5 100.0 0 0.0 5 12.2 Chickens 1 12.5 5 62.5 2 25.0 8 19.5 Totals 1 2.4 92.7 100.0 2 4.9 41 100.0 Who Keeps Money from Sales Calves 4 80.0 1 20.0 0 0.0 5 10.4 Bulls 0 0.0 5 100.0 0 0.0 5 10.4 Oxen 8 80.0 2 20.0 0 0.0 10 20.8 Heifers 2 100.0 0 0.0 0 0.0 2 4.2 Cows 4 66.6 2 33.3 0 0.0 6 12.5 Sheep 1 50.0 1 50.0 0 0.0 2 4.2 Goats 3 75.0 1 25.0 0 0.0 4 8.3 Donkeys 4 66.6 2 33.3 0 0.0 6 12.5 Chickens 0 0.0 0 0.0 8 100.0 8 16.7 Totals 26 54.2 14 29.2 8 16.7 48 100.0 Who Sells Products Calves Bulls Oxen 0 0.0 1 100.0 0 0.0 1 9.1 Heifers Cows 0 0.0 1 33.3 2 66.6 3 27.3 Sheep Goats 1 100.0 0 0.0 0 0.0 1 9.1 Donkeys 0 0.0 1 100.0 0 0.0 1 9.1 Chickens** 0 0.0 0 0.0 5 100.0 5 45.5 Totals 1 9.1 27.3 100.0 63.6 100.0 11 100.0 *Missing values; **One male household head reported that his son sells chicken products. 175 Table 45A Live Animal Purchases, Sales and Consumption in the Past 12 months (n = 30) TYPE OF BOUGHT SOLD CONSUMED LIVESTOCK No. % No. % No. % Calves 0 0 1 5.3 0 0.0 Bulls 0 0.0 0 0.0 0 0.0 Oxen* 3 20.0 9 47.4 0 0.0 Heifers 0 0.0 0 0.0 0 0.0 Cows 1 6.7 4 21.1 0 0.0 Sheep 1 6.7 5 26.3 4 17.4 Goats 9 60.0 0 0.0 19 82.6 Donkeys 1 6.7 0 0.0 0 0.0 Sub-totals 15 100.0 19 100.0 23 100.0 Chickens 8 34.8 44 69.8 23 50.0 Totals 23 17.4 63 47.7 46 34.8 Table 47A Trees Growing on Household Land by Species (n = 30) EUCALYPTUS EUCALYPTUS OTHER NO. OF TREES TOTAL BLUE RED SPECIES PER No. % No. % No. % No. % HOUSEHOLD farms farms farms farms farms farms farms No trees 7 17.9 1-5 trees 9 23.1 3 33.4 4 44.4 1 11.1 6-10 trees 3 7.7 0 0.0 3 100.0 1 33.4 11-25 trees 3 7.7 2 66.7 2 66.7 0 0.0 26-50 trees 5 12.8 1 20.0 5 100.0 0 0.0 51-100 trees 1 2.6 1 100.0 1 100.0 0 0.0 101+ trees 2 5.1 0 0.0 2 100.0 0 0.0 Totals 23 100.0 7 30.4 17 73.9 2 8.7 176 Table 51A Rights to Trees: to Destroy or Sell, to Cut, and to Take Products without Cutting, by Sex of Household Head (n = 23) RIGHTS MH FH Only head can destroy, cut and take products 0 7 Both spouses have all rights 5 0 Both spouses can destroy and cut, but only spouse 4 0 can take products without cutting Only head can destroy, both can cut, and only 1 0 spouse can take products without cutting Both can destroy, only head can cut, and only 1 0 spouse can take products without cutting Both can destroy, but only spouse can cut and take 2 0 products without cutting Head & another household member have all rights 0 1 (in the case reported, a grandmother) Landlord has all rights 0 2 Totals 13 10 Table 53A Crops Produced in Homegardens by Number of Households Reporting the Crop* (n = 13) CROP No. % Onions 9 69.2 Cabbage 8 61.5 Garlic 5 38.5 Lettuce 4 30.8 Tomatoes 4 30.8 Potatoes 4 30.8 Cumin 3 23.1 Carrots 3 23.1 Other crops 3 23.1 Green pepper 2 15.4 Buckthorn; local hops 2 15.4 (Rhamnus prinoides) Beetroots 2 15.4 *Does not include one report of ba-erir since its production was reported in another survey module. 177 Table 54A Division of Labour in Homegardening by Relation with the Household Head, MH Households (n=6) SEED SEED PROCURE- PLANT- CULTIVA- HARVESTING SELECTION, TASKS/ MENT ING TION STORAGE DIVISION OF LABOUR Head alone 2 0 0 0 2 Head & wife 2 6 3 3 1 Head & daughter 1 0 0 0 0 Head, wife & daughter 0 0 1 0 0 Wife alone 1 0 2 2 1 Wife & daughter 0 0 0 1 1 Totals 6 6 6 6 5 Table 58A Earnings by Activity EARNINGS IN BIRR ACTIVITY 1-50 51-300 301-700 No. % No. % No. % Sale of homegarden produce 1 12.5 Trade of grains and pulses 5 25.0 Trade of livestock 3 15.0 2 25.0 Spices/aromatics col./trade 1 25.0 Preparation/sales other drinks 1 25.0 8 40.0 3 37.5 Food or tea preparation/sales 1 25.0 Basket-making 1 5.0 Transport (pack animal) 2 10.0 Food-for-work 1 5.0 1 12.5 Nursery work 1 12.5 Other misc. activities 1 25.0 Totals 4 100.0 20 100.0 8 100 178 Total No. % 1 3.1 5 15.6 5 15.6 1 3.1 12 37.5 1 3.1 1 3.1 2 6.3 2 6.3 1 3.1 1 3.1 32 100.0 Table 59A Wages per Day by Sex and by Activity (n = 28) .01 TO 5 6-10 BIRR 11-19 BIRR SEX OF WORKER AND BIRR ACTIVITY No. % No. % No. % Sex Male (10) 3 23.1 2 40.0 2 40.0 Female (16) 10 77.0 3 60.0 3 60.0 Female heads (11) 5 38.5 3 23.1 2 15.4 Wives (5) 3 60.0 0 0.0 1 20.0 Total 13 100.0 5 100.0 5 100.0 Activity Sale of homegarden produce 1 100.0 0 0 (1) 1 33.3 0 3 66.7 Trade of grain and pulses (4) 2 50.0 0 0 Trade of livestock (4) 1 100.0 0 0 Spices/aromatics col./trade (1) 2 18.1 5 45.5 2 18.1 Prep./sales other drinks (11) 1 100.0 0 0 Food or tea prep./sales (1) 1 100.0 0 0 Basket-making (1) 1 100.0 0 0 Transport (pack animal) (1) 1 50.0 0 0 Food-for-work (2) 1 100.0 0 0 Nursery work (1) 1 100.0 0 0 Other miscellaneous (1) 13 46.5 5 17.9 5 17.9 Total 20-60 BIRR No. % 3 2 1 1 5 60.0 40.0 7.7 20.0 100.0 0 0 2 0 2 0 0 0 1 0 0 5 50.0 18.1 50.0 17.9 179 Table 60A Participation in Community Work Activities by Household Headship and Sex (household n = 23; worker no = 33) MH FH MALE FEMALE PARTICIPATION WORKERS WORKERS PARAMETER No. % No. % No. % No. % Programme or agency* Woreda government 0 0.0 2 28.6 0 0.0 2 8.7 Other government 0 0.0 1 14.3 0 0.0 1 4.3 Redd Barna 0 0.0 1 14.3 0 0.0 1 4.3 Food-for-work 12 100.0 7 63.6 9 90.0 19 82.6 Other 0 0.0 0 0.0 1 10.0 0 0.0 Type of work Road build3 25.0 2 18.2 1 10.0 5 21.7 ing/maintenance 3 25.0 0 0.0 2 20.0 4 17.4 Soil conservation activities 1 8.3 0 0.0 2 20.0 1 4.3 Tree planting/nurseries 5 41.7 6 50.0 5 50.0 10 43.5 Well-digging, river diversion 0 0.0 2 18.2 0 0.0 2 8.7 & small dam construction Irrigation activities 0 0.0 1 9.1 0 0.0 1 4.3 Solving water or other CPR disputes Reason for participating Needed any employment 6 50.0 3 27.3 5 50.0 8 34.8 Needed temporary employment 4 33.3 1 9.1 3 30.0 5 21.7 General food shortage 1 8.3 1 9.1 1 10.0 3 13.0 Household food shortage 0 0.0 6 54.5 0 0.0 6 26.1 For investment 1 8.3 0 0.0 1 10.0 1 4.3 How earnings were used Sold for cash 1 8.3 2 18.2 1 11.1 3 13.0 Consumed 9 75.0 9 81.8 6 66.7 18 78.3 Part consumed, part sold 2 16.7 0 0.0 2 22.2 2 8.7 *Additional possibilities included: User committee, Peasant Association, FAO/Belgian project, World Vision, REST, Irish, Donbesco and “other”. 180 MAJOR USE CATEGORIES TOTALS No. % Table 67A All Uses by All Common Resource Areas (n = 30) RESTRICTED OPEN UNENCLOSED ENCLOSED OTHER GRAZING GRAZING FORESTS/ WOODLOTS ENCLOSED AREAS AREAS WOODLOTS AREAS No. % No. Grazing Fodder Medicinals Fuelwood Basket Thatch Grass Coffee Wild foods Construction Aromatics Other Totals 33 4 30 35 5 26 34 4 7 4 182 18.1 2.2 16.5 19.2 2.7 14.3 18.7 2.2 3.8 2.2 100.0 3 0 1 0 4 6 1 0 0 0 15 20.0 0.0 6.7 0.0 26.7 40.0 6.7 0.0 0.0 0.0 100.0 2 2 7 2 0 8 2 0 0 1 24 Grazing Fodder Medicinals Fuelwood Basket Thatch Grass Coffee Wild foods Construction Aromatics Other Totals 33 4 30 35 5 26 34 4 7 4 182 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 3 0 1 0 4 6 1 0 0 0 15 9.0 0.0 3.3 0.0 80.0 23.1 2.9 0.0 0.0 0.0 9.7 2 2 7 2 0 8 2 0 0 1 24 % No. % No % No. . Distribution of Use Within Resource Areas 8.3 0 0.0 0 0.0 28 8.3 0 0.0 0 0.0 0 29.2 6 50.0 4 28.6 2 8.3 3 25.0 2 14.3 8 0.0 0 0.0 0 0.0 1 33.3 1 8.3 4 28.6 1 8.3 0 0.0 0 0.0 1 0.0 1 8.3 2 14.3 0 0.0 1 8.3 1 7.1 2 4.2 0 0.0 1 7.1 1 100.0 12 100.0 14 100.0 44 Distribution of Use Across Resource Areas 6.1 0 0.0 0 0.0 28 50.0 0 0.0 0 0.0 0 23.3 6 20.0 4 13.3 2 5.7 3 8.6 2 5.7 8 0.0 0 0.0 0 0.0 1 30.8 1 3.8 4 15.4 1 5.9 0 0.0 0 0.0 1 0.0 1 25.0 2 50.0 0 0.0 1 14.3 1 14.3 2 25.0 0 0.0 1 25.0 1 15.6 12 7.8 14 9.1 44 % UNENCLOSED HILLSIDES No. % RIVERSIDES AND OTHER BOUNDARY LAND No. % 63.6 0.0 4.5 18.8 2.3 2.3 2.3 0.0 4.5 2.3 100.0 0 1 6 20 0 3 26 1 3 1 61 0.0 1.6 9.8 32.8 0.0 4.9 42.6 1.6 4.9 1.6 100.0 0 1 4 0 0 3 4 0 0 0 12 0.0 8.3 33.3 0.0 0.0 25.0 33.3 0.0 0.0 0.0 100.0 84.8 0.0 6.7 22.9 20.0 3.8 2.9 0.0 28.6 25.0 10.4 0 1 6 20 0 3 26 1 3 1 61 0.0 25.0 20.0 57.1 0.0 11.5 76.5 25.0 42.9 25.0 39.6 0 1 4 0 0 3 4 0 0 0 12 0.0 25.0 13.3 0.0 0.0 11.5 11.8 0.0 0.0 0.0 7.8 181 USER INDICATORS Table 68A Users of Different Common Land Areas and Days of Use RESTRICTED OPEN UNENCLOSED ENCLOSED UNENCLOSED GRAZING GRAZING HILLSIDES FORESTS FORESTS AND LAND LAND AND WOODLOTS WOODLOTS No. User by sex and relation to head Total Males Females Days per year of use Avg. per household (n=10)* Maximum Minimum Proportions of households using Households using as % of all HH FH using areas as % of all user HH 182 29 12 17 587 58 % 100.0 41.4 58.6 180 1 No. % 31 9 22 1090 78 100.0 29.0 71.0 360 1 No. % 86 65 21 3004 28 100.0 75.6 24.4 360 1 No. 16 5 11 340 34 % 100.0 31.3 68.8 150 1 No. % 14 2 12 535 59 100.0 14.3 85.7 360 5 OTHER COMMUNITY ENCLOSURES No. 24 6 18 960 107 % 100.0 25.0 75.0 360 1 RIVERSIDES AND OTHER BOUNDARY LAND No. % 16 2 14 171 19 100.0 12.5 87.5 90 1 12 40.0 16 53.3 29 96.7 10 30.0 12 40.0 10 30.0 9 30.0 8 66.7 6 37.5 15 51.7 8 80.0 7 58.3 9 90.0 9 100.0 LOCAL NAME Aagam Achabiti Agol Andelie* Ar-aro Asagorha* Atami Atran Atush Awelie Baharzef Beles Biharere* Cherencha Chihidi Chiraro* Dander Echote talien Grass Hahote Hama shiro Hamli talien Haragrisha Icheo* Ichern* Iebeke* Kebekel* Kilialo* Merez OTHER COMMUNITY ENCLOSURES No. % 0 0.0 1 25.0 1 11.1 0 0.0 1 16.7 0 0.0 0 0.0 0 0.0 1 6.7 9 40.9 3 13.6 0 0.0 0 0.0 0 0.0 2 100.0 1 100.0 2 66.7 0 0.0 3 8.6 11 27.5 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 2 66.7 0 0.0 Table 69A Species Used by Common Resource Area, All Species and All Areas RIVERSIDES, ENCLOSED OPEN UNENCLOSED OPEN BOUNDARY WOODLOTS FORESTS/WOODED HILLSIDES GRAZING LAND AREAS No. % No % No % No % No % 0 0.0 0 0.0 0 0.0 1 100.0 0 0.0 0 0.0 0 0.0 0 0.0 3 75.0 0 0.0 1 11.1 1 11.1 1 11.1 2 22.2 2 22.2 0 0.0 2 100.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 4 66.7 1 16.7 0 0.0 1 100.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 100.0 0 0.0 0 0.0 0 0.0 1 100.0 0 0.0 0 0.0 1 6.7 1 6.7 2 13.3 5 33.3 4 26.7 0 0.0 5 22.7 1 4.5 7 31.8 0 0.0 2 9.1 2 9.1 7 31.8 4 18.2 2 9.1 3 9.1 0 0.0 0 0.0 27 81.8 2 6.1 0 0.0 0 0.0 0 0.0 1 100.0 0 0.0 0 0.0 0 0.0 1 100.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 33.3 0 0.0 0 0.0 0 0.0 1 33.3 1 33.3 4 11.4 4 11.4 1 2.9 3 8.6 10 28.6 0 0.0 1 2.5 3 7.5 22 55.0 3 7.5 3 25.0 0 0.0 0 0.0 8 66.7 1 8.3 1 25.0 0 0.0 0 0.0 3 75.0 0 0.0 0 0.0 0 0.0 0 0.0 2 40.0 2 40.0 0 0.0 0 0.0 1 100.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 100.0 0 0.0 0 0.0 0 0.0 0 0.0 1 100.0 0 0.0 0 0.0 0 0.0 1 100.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 33.3 0 0.0 0 0.0 1 100.0 0 0.0 0 0.0 0 0.0 RESTRICTED GRAZING No. 0 0 1 0 0 0 0 0 1 0 2 1 0 0 0 0 0 1 10 0 0 0 1 0 0 0 0 0 0 % 0.0 0.0 11.1 0.0 0.0 0.0 0.0 0.0 6.7 0.0 9.1 3.0 0.0 0.0 0.0 0.0 0.0 33.3 28.6 0.0 0.0 0.0 20.0 0.0 0.0 0.0 0.0 0.0 0.0 TOTALS No. 1 4 9 2 6 1 1 1 15 22 22 33 1 1 2 1 3 3 35 40 12 4 5 1 1 1 1 3 1 % 0.3 1.3 2.9 0.6 1.9 0.3 0.3 0.3 4.8 7.1 7.1 10.6 0.3 0.3 0.6 0.3 1.0 1.0 11.3 12.9 3.9 1.3 1.6 0.3 0.3 0.3 0.3 1.0 0.3 183 LOCAL NAME Merkatoos zbee Meshago Rambo-rambo Sarow Sar’saro Sheila-en Shewho* Shoha Shua arni Tabub Toko berbere Thehetses Totals OTHER COMMUNITY ENCLOSURES No. % 0 0.0 2 0 1 0 2 1 1 1 8 0 6 59 50.0 0.0 33.3 0.0 33.3 33.3 50.0 50.0 22.9 0.0 40.0 19.0 Table 69A Species Used by Common Resource Area, All Species and All Areas (con’t.) ENCLOSED RIVERSIDES OPEN UNENCLOSED OPEN BOUNDARY WOODLOTS FORESTS/WOODED HILLSIDES GRAZING LAND AREAS No. % No % No % No % No % 0 0.0 0 0.0 0 0.0 0 0.0 1 100.0 0 0 0 0 0 0 0 0 0 0 0 15 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.8 0 0 0 0 1 1 1 0 2 0 2 25 0.0 0.0 0.0 0.0 16.7 33.3 50.0 0.0 5.7 0.0 13.3 8.1 1 0 0 0 0 0 0 0 3 0 1 24 Source: Household Survey *No information that would help to identify the species in reference was available. 184 25.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.6 0.0 6.7 7.7 1 5 2 1 3 1 0 0 20 0 6 136 25.0 100.0 66.7 100.0 50.0 33.3 0.0 0.0 57.1 0.0 40.0 43.9 0 0 0 0 0 0 0 1 2 1 0 34 0.0 0.0 0.0 0.0 0.0 0.0 0.0 50.0 5.7 100.0 0.0 11.0 RESTRICTED GRAZING No. % TOTALS 0 0.0 No. 1 0 0 0 0 0 0 0 0 0 0 0 17 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 5.5 4 5 3 1 6 3 2 2 35 1 15 310 % 1.3 1.6 1.0 0.3 1.9 1.0 0.6 0.6 11.3 0.3 4.8 100.0 0.3 Table 70A Species Collected From Unenclosed Hillsides and Uses TOTAL MAJOR USES (TOTAL) LOCAL NAME Aroma Coffee ConFuelMediWild No. % -tic struction wood cinal food Achabiti 3 2.2 3 Aagam 1 0.7 1 Agol 2 1.5 2 Ar-aro 4 2.9 4 Atame 1 0.7 1 Atush 5 3.7 1 3 1 Awelie 7 5.1 1 5 1 Biharere 1 0.7 1 Baharzef 4 2.9 1 3 Beles 27 19.9 22 Grass 3 2.2 3 Hama shiro 8 5.9 8 Hamli talien 3 2.2 3 Haregrish 2 1.5 2 Hahote 22 16.2 1 20 1 Ichern 1 0.7 1 Iebeke 1 0.7 1 Kilialo 1 0.7 1 Medafe talien 1 0.7 1 Mesaguh 1 0.7 1 Rambo-rambo 5 3.7 1 3 1 Sar-saro 1 0.7 1 Seraw 2 1.5 1 1 Shewho 1 0.7 1 Shilean 3 2.2 3 Tebeb 20 14.7 1 19 Thetetses 6 4.4 5 1 Totals 136 100.0 7 3 2 61 17 41 Percentage 100.0 5.1 2.2 1.5 44.9 12.5 30.1 Source: Household Survey. Fodder Other 4 1 4 0.7 1 2.9 185 Table 71A Species by Species Use, for Enclosed Woodlots and Unenclosed Forests and Woodlots NO. USES ENCLOSED UNENCLOSED PER AREA & WOODLOTS FORESTS MAJOR USES (TOTAL) SPECIES WDLTS. Aroma Coffee ConsFuelMediOther Local Name No. % No. % -tic Cerem. truction wood cinal Agol 1 4.0 1 4.3 2 Andelie 2 8.0 0 0.0 1 Asagorha 1 4.0 0 0.0 1 Ateran 0 0.0 1 4.3 1 Atush 1 4.0 2 8.7 3 Awelie 5 20.0 1 4.3 1 1 2 1 Chrechah 0 0.0 1 4.3 1 Baharzef 2 8.0 7 30.4 1 8 Grass 4 16.0 1 4.3 5 Hahote 1 4.0 3 13.0 4 Icheo 0 0.0 1 4.3 1 Merez 1 4.0 0 0.0 1 Meshago 0 0.0 1 4.3 1 Shewo 1 4.0 0 0.0 1 Shilean 1 4.0 0 0.0 Shoha 1 4.0 0 0.0 Thehetses 2 8.0 1 4.3 1 1 1 Tabub 2 8.0 3 13.0 4 1 Totals 25 100.0 23 100.0 1 5 3 14 19 2 Source: Household Survey. 186 Total 2 1 1 1 3 5 1 9 5 4 1 1 1 1 0 0 3 5 44 Table 72A Species and Species Use, from Other Community Enclosures GRAND USES (TOTAL) TOTALS Coffee Basket Fuel- Medi Wild Local name No. % Aromatic Cerem. Thatch wood food cinal Achabiti 1 1.7 1 Agol 1 1.7 1 Ar-aro 1 1.7 1 Atush 1 1.7 1 Awelie 9 15.3 2 5 2 Baharzef 3 5.1 2 1 Chiparo 1 1.7 1 Cypress 2 3.4 2 Dander 2 3.4 2 Grass 3 5.1 1 1 Hahote 11 18.6 1 8 2 Kilialo 2 3.4 1 1 Mesaguh 2 3.4 2 Seraw 1 1.7 1 Shewho 1 1.7 1 Shoha 1 1.7 1 Shilean 2 3.4 1 1 Shiwakerni 1 1.7 1 Tabub 8 13.6 8 Thetetses 6 10.2 1 3 2 Totals 59 100.0 7 1 1 38 10 1 Percentages 100.0 11.9 1.7 1.7 64.4 16.9 1.7 Source: Household Survey. NO. USES/ SPECIES Table 73A Species Reported as Collected from Riversides and Other Boundary Lands and Uses TOTAL USES (TOTAL) LOCAL NAME Coffee Wild No. % Cerem. Medicinal food Agol 1 6.7 1 Atush 1 6.7 1 Baharzef 2 13.3 2 Beles 3 20.0 3 Grass 4 26.7 3 Hama shiro 3 20.0 3 Hamli talien 1 6.7 1 Total 15 100.0 3 4 7 Percentage 100.0 20.0 26.7 46.7 Source: Household Survey. Other 1 1 1.7 Fodder 1 1 6.7 187 Table 74A Seasonality of Use,* by Major Use Categories and by Common Resource Areas MAJOR USE CATEGORIES Fuelwood Medicinals Wild foods Coffee ceremony Fodder Grazing cattle Aromatics Thatch for baskets Construction All uses Common Resource Areas Grazing land Forest/woodlots Other comm. encl. Unenclosed hillsides Roadsides, borderlands All areas JAN 27 11 6 7 2 3 3 3 3 65 FEB 27 9 5 6 3 3 1 3 1 58 MAR 26 9 5 4 3 3 1 3 1 55 APR 27 9 7 6 3 3 1 3 1 60 MAY 27 9 11 5 3 2 1 2 2 62 JUN 23 14 23 11 3 2 4 0 0 80 JULY 22 16 24 9 1 2 4 0 0 78 AUG 22 16 24 9 1 1 4 0 0 77 SEPT 29 15 21 12 1 2 5 3 1 89 OCT 28 12 8 7 1 3 5 1 1 66 NOV 28 12 6 7 1 3 4 1 1 63 DEC 28 10 4 7 2 3 3 2 2 61 TOTAL DAYS 1984 1790 1650 906 516 436 253 157 45 7737 25.6 23.1 21.3 11.7 6.7 5.6 3.3 2.0 0.6 99.9 10 12 11 27 1 10 9 8 27 1 10 8 8 25 1 14 9 8 25 1 11 10 8 27 1 15 10 7 37 5 11 11 6 37 7 11 11 6 36 7 11 13 11 39 7 7 12 9 28 3 6 11 9 26 3 8 11 10 26 1 1111 1009 1315 3201 258 15.8 14.4 18.7 45.6 3.7 61 55 52 57 57 74 72 71 81 59 55 56 6894 98.2 Source: Household Surveys. *Total number of days that households reported using a particular resource area for a particular use in the previous 12 months. 188 % Table 76A Major Use Categories by Sex of Household Head (n = 30) MH FH TOTAL MAJOR USE CATEGORY No. % No. % No. % A. Distribution of uses between household headship categories Aromatics 2 25.0 6 75.0 8 100.0 Cattle grazing 4 80.0 1 20.0 5 100.0 Coffee 11 37.9 18 62.1 29 100.0 Construction 3 75.0 1 25.0 4 100.0 Fodder 3 75.0 1 25.0 4 100.0 Fuelwood 15 42.9 20 57.1 35 100.0 Medicinals 10 31.3 22 68.8 32 100.0 Other 3 75.0 1 25.0 4 100.0 Thatch 2 40.0 3 60.0 5 100.0 Wild food 15 44.1 19 55.9 34 100.0 Totals 68 42.5 92 57.5 160 100.0 B. Distribution of uses within household headship categories Aromatics 2.9 6.5 5.0 Cattle grazing 5.9 1.1 3.1 Coffee 16.2 19.6 18.1 Construction 4.4 1.1 2.5 Fodder 4.4 1.1 2.5 Fuelwood 22.1 21.7 21.9 Medicinals 14.7 23.9 20.0 Other 4.4 1.1 2.5 Thatch 2.9 3.3 3.1 Wild food 22.1 20.7 21.3 Totals 100.0 100.0 100.0 Source: Household Survey. 189 MAJOR USES Aromatics Coffee ceremony Fodder collection Construction Cattle grazing Fuelwood collection Medicinals Other products Thatch for baskets Wild food Totals 190 Table 77A Sex and Age of Common Resource Users, by Major Use Categories FEMALE TOTAL ALL MALE FEMALE MALE MALES ADULTS ADULTS CHILDREN CHILDREN No. % No. % No. % No. % No. % No. % 0 0.0 5 62.5 1 12.5 2 25.0 8 100.0 1 12.5 9 30.0 21 70.0 0 0.0 0 0.0 30 100.0 9 30.0 1 16.7 4 66.7 0 0.0 1 16.7 6 100.0 1 16.7 0 0.0 3 75.0 0 0.0 1 25.0 4 100.0 0 0.0 2 25.0 5 62.5 1 12.5 0 0.0 8 100.0 3 37.5 10 8 1 0 10 41 18.5 20.0 20.0 0.0 20.4 20.4 31 23 4 4 29 124 57.4 57.5 80.0 80.0 59.2 61.7 5 1 0 0 3 10 9.3 2.5 0.0 0.0 6.1 5.0 8 8 0 1 7 26 14.8 20.0 0.0 20.0 14.3 12.9 54 40 5 5 49 201 100.0 100.0 100.0 100.0 100.0 100.0 15 9 1 0 13 51 27.8 22.5 20.0 0.0 26.5 25.4 ALL FEMALES No. % 7 87.5 21 70.0 5 83.3 4 100.0 5 62.5 39 31 4 5 36 150 72.2 77.5 80.0 100.0 73.5 74.6 Table 79A Twenty-five Most Important Plant Species According to Use Categories, Men's Focus Groups and Women's Focus Groups Women Men Both TOOLS CONSTRUCBEE FAMINE WILD BEEHIVES GEN. FODDER CEREMONIAL TION FORAGE FOOD FOOD CONSERVATION Awelie Tabub Moondaha Hahote Tabub Agee-era Awelie Awelie Chihidi Tahatses Awelie Hahote Aagam Beles Chehendoge Hahote Baharzef Sheila-en Toko berbere Aagam Moondaha Kunkura Dander Senti (grass) Merez Aagam Beles Ar-aro Cha-a Shua arni Ashu-ena Chihidi Dander Shemfa Beles Beles Baharzef Shua arni Mengolhas Shemfa Girbia Chehendoge Beles Da-aro Mengolhas Tahatses Toko berbere Girbia Kanche-a Da-aro kanche-a Shua arni Tahatses Kunkura Kanche-a Kunkura Shemfa Sheila-en Kunkura Roweh Da-aro Kanche-a Senti (grass) Roweh Cha-a FUMIGANT/ FUEL PLOWING FUMIGANT TOOLS: MEDICINE MEDICINE DISINFECT WOOD FENCING TOOLS RELIGIOUS HEALTH/BEAUTY HOUSEHOLD GENERAL WOMEN'S Sheila-en Awelie Tahatses Baharzef Senti (grass) Hahote Tambugh Awelie Awelie Sar-saro Tahatses Sheila-en Toko berbere Ashu-ena Merez Tahatses Tambugh Itsmeskel Sheila-en Chihidi Kanche-a Kulio Tabub Tabub Shimeda Hahote Chehendoge Roweh Cherencha Shemfa Kebirowaled Tabub Aagam Haragrisha Da-aro Merez Merez Hahote Baharzef Ashu-ena Cherencha Kulio Beles Awelie Agee-era Ezenanchewa Cherencha Mengolhas Chihidi Chihidi Cha-a Haragrisha Baharzef Baharzef Chehendoge Shimeda Toko berbere Aserekuka Aagam Gulihan Shemfa Gelac Mengolhas Chehendoge Roweh Da-aro 195 Table 79A Twenty-five Most Important Plant Species According to Use Categories, Men's Focus Groups and Women's Focus Groups (con’t). MEDICINECHILDREN Atush Itsmeskel Mechalo EVIL EYE Unguleh Rambo-rambo Sar-saro Kulio FUMIGANTFOOD/DRINK PREP. Atush Sheila-en Chehendoge Awelie COSMETIC Tahatses Hahote Merez Kulio Chihidi (w/henna) ETHNOVETERINARY Cherencha Shimeda Gulihan Source: Cultural Valuation Study, Men’s Focus Groups and Women’s Focus Groups. 196 STAFF Awelie Kanche-a Roweh Table 81A Species Used for Fuelwood by Number of Times Reported NO. TMES LOCAL NAME USE % REPORTED Aagam 1 .8 Achabiti 4 3.2 Atami 2 1.6 Awelie 13 10.5 Baharzef 10 8.1 Beles 1 .8 Cherencha 1 .8 Chihidi 2 1.6 Dander 2 1.6 Hahote 29 23.4 Kobkub 1 .8 Kulio 2 1.6 Meshago 3 2.4 Sarow 3 2.4 Shiela-en 6 4.8 Shoha 2 1.6 Shua arni 1 .8 Tabub 29 23.4 Thehetses 11 8.9 Total 124 100.0 Source: Household Survey. 197 Table 84A Wild Food Plant Consumption Frequencies KUENTI AAGAM KUNTEE MEKMOKO HAMLI TALIEN FREQUENCY OF BELES GEBA HH CONSUMPTION Twice per day Once per day Twice per week Once per week 20 One to two 2 times/month In six “summer” months: 4-20 times 1 20-40 times 5 40-50 times 6 70-90 times 3 100 times plus 1 Rarely 14 4 13 Source: Household Survey, Food Frequency Module. 198 MULOO KOSHIM BIMHIGO 2 1 1 4 1 4 3 12 8 6 1 Table 85A All Wild Foods Reported, by Local Plant Name POSSIBLE ENGLISH TIGRINYA SCIENTIFIC NAME Aagam Carissa edulis1(Forssk.) Vahl Ar-aro Kalanchoe marmorata Atush Achyranthes aspera Awelie Olea europaea ssp. cusOlive pidata Beles Opuntia ficus-indica, Prickly Pear Opuntia vulgaris Bimhigo Buna Coffea arabica Coffee Chenedoge Otostegia integrifolia Daro Ficus vasta Echote talien Agremone mexicana Prickly Poppy Geba Zizyphus spina-christi Girbia Monthecium glandosum Hahote Rumex nervosus Hama shiro Caralluma sprengeri Hamli talien Amaranthus graecizans; Brassica spp. Kancha-a Koshim Kuenti Cyperus bulbosus Kunkura Kuntee Mengolhas Dovyalis abyssinica Rumex abyssinicus Spinach-rhubarb Mekmoko Moondaha Mougia Cynodon dactylon Muloo Shemfa Ficus sur Shoha Unknown Thehetses Dodonea angustifolia Toko Berbere Peruvian pepper Turingi Citrus medica (auranBitter Orange tium?) 199 Table 86A Part A: All Species Reported as Used by Uses (from the Household Survey) LOCAL FUELCOFFEE MEDIFOD- ORNA- BASKFUMIARONAME WOOD CEREM. CINE FOOD DER MENT ETS GANT MATIC Percentage 34.0 7.7 20.5 9.6 6.6 1.1 0.5 4.9 1.1 Total uses 124 28 75 35 24 4 2 18 4 Aagam 1 1 Achabiti 4 Agol 10 Ar-aro Asagorha 1 Atami 2 Atush 12 Awelie 13 9 3 6 1 Ba-erir 1 1 Baharzef 10 2 16 2 Beles 1 24 14 Cha-a Cherencha 1 1 Chihidi 2 Chiparo/ 1 chiraro Dander 2 Echote talien Grass 23 4 2 Hahote 29 3 1 3 Hama shiro 7 Hamli talien 3 Haregrisha 7 Kobkub 1 1 Kulio 2 1 1 Merez 1 Merkatoos 1 zbee Meshago 3 2 1 Rambo-rambo 1 Sarow 3 1 Sar’saro 1 1 Shiela-en 6 2 1 Shoha 2 4 4 Shua arni 1 Tabub 29 1 Thehetses 11 4 1 Thatching grass Toko berbere 1 1 1 1 Source: Household survey. 200 Table 86A Part B: Species Reported as Used by Uses (from the Household Survey) Food PloughLocal Name ConBoun- Environpreping Other Oil struction daries mental aration Tools tools Seeds Other Total Percentage 7.1 2.5 0.3 1.6 0.5 1.1 0.3 0.5 100.0 Total 26 9 1 6 2 4 1 2 365 Aagam 1 3 Achabiti 4 Agol 10 Ar-aro 6 6 Asagorha 1 Atami 2 Atush 12 Awelie 9 1 1 43 Ba-erir 2 Baharzef 1 1 2 43 9 Beles 39 Cha-a 1 1 Cherencha 1 3 Chihidi 1 3 Chiparo/chiraro 1 Dander 2 Echote talien 1 1 Grass 29 Hahote 1 40 3 Hama shiro 7 Hamli talien 3 Haregrisha 7 Kobkub 2 Kulio 4 Merez 1 Merkatoos zbee 1 Meshago 6 Rambo-rambo 1 Sarow 1 5 Sar’saro 2 Shiela-en 9 Shoha 10 Shua arni 1 Tabub 3 33 Thehetses 5 1 22 Thatching 1 1 grass Toko berbere 1 5 201 % 100.0 0.8 1.1 2.7 1.6 0.3 0.5 3.3 11.8 0.5 11.8 10.7 0.3 0.8 0.8 0.3 0.5 0.3 7.9 11.0 1.9 0.8 1.9 0.5 1.1 0.3 0.3 1.6 0.3 1.4 0.5 2.5 2.7 0.3 9.0 6.0 0.3 1.4 Table 88A All Medicinal Plants Reported, by Local Name ENGLISH TIGRINYA POSSIBLE SCIENTIFIC NAME Agol Unknown Andelie Asagorha Aserekuka Atush Achyranthes aspera Awelie Olea europaea ssp. cuspidata Olive Eucalyptus globulus, Eucalyptus Baharzef Eucalyptus camaldulensis Chenedoge Otostegia integrifolia Cherencha Calpurnia aurea Chihidi Juniperus procera, Cupressus spp Cypress Ezenanchewa Gelac Gelac Gulihan (Gulehane) Gulihe Hahote Rumex nervosus (?) Haragrisha(Harigressa) Zehneria scabra Itsmeskel Unknown Karoo-wared Unknown Kulio Euclea divinorum Merez Acokanthera schimperi Merkatoos zbee Bersana abyssinica Meshago Meriandra bengalensis Mitchano Unknown Moondaha More Rambo-rambo Sarow Acacia etbaica Sar’saro Shemeda Justica schimperiana Shoha Unknown Tabub Becium grandiflorum Tambugh Croton macrostachyus Hochst. ex Del. Thehetses Dodonea angustifolia Toko Berbere Schinus molle Peruvian pepper Unguleh Solanum sp. 202 Table 89A Species Ranking according to Number of Times Use was Reported LOCAL NAME No. % Total 306 100.00 Hahote 13.07 40 Grass 35 11.44 Tabub 11.44 35 Beles 33 10.78 Baharzef 22 7.19 Awelie 6.86 21 Atush 15 4.90 Thetetses 15 4.90 Hama shiro 12 3.92 Agol 9 2.94 Ar-aro 6 1.96 Haregrish 5 1.63 Rambo-rambo 5 1.63 Sheila-en 1.63 5 Achabiti 1.31 4 Hamli talien 1.31 4 Meshago 1.31 4 Dander 0.98 3 Echote talien 3 0.98 Kilialo 3 0.98 Sarow 3 0.98 Shewho 3 0.98 Chihidi 0.65 2 Shua arni 2 0.65 Aagam 1 0.33 Andelie 1 0.33 Asagorha 1 0.33 Atami 1 0.33 Ateran 1 0.33 Biharere 1 0.33 Cherencha 1 0.33 Chiparo 1 0.33 Icheo 1 0.33 Ichern 1 0.33 Iebeke 1 0.33 Kekebel 1 0.33 Merez 1 0.33 Merkatoos zbee 1 0.33 Sar-saro 1 0.33 Shoha 1 0.33 Toko berbere 1 0.33 Source: Household Survey 203
0
You can add this document to your study collection(s)
Sign in Available only to authorized usersYou can add this document to your saved list
Sign in Available only to authorized users(For complaints, use another form )