PILGRIM, 2006 - A Cross Cultural Study Into LEK - Tese

A Cross-Cultural Study into Local Ecological Knowledge
S. E. Pilgrim
A thesis submitted for the degree of Ph.D.

Department of Biological Sciences
University of Essex
October 2006
Dedicated to my
Mother and Father
with love
Summary
For decades now, since the birth of industrialisation, human populations, particularly in the Western
world, have become less and less reliant upon and connected with their local land. With this
departure from the land has come a departure from traditional knowledge systems. Generations of
accumulated observations are being lost or replaced by modern knowledge systems. With
industrialisation now spreading to remote regions threatening traditional knowledge bases, this
investigation looks at inter- and intra-cultural variation in knowledge, particularly in terms of
economic development and resource dependence within communities in the UK, India and
Indonesia.
Significantly lower ecological knowledge levels were observed where economic development was
high and resource dependence low. As level of resource dependence of a community decreased, the
age at which ecological knowledge of individuals became saturated increased and rate of knowledge
acquisition slowed. Progressive loss in the younger generations was observed at the industrialised
sites. The teaching methods of formal education and the influence of television were found to be
contributing to this pattern of loss.
At the developing study sites, ecological knowledge was found to be gender-differentiated. This is a
product of societal roles and daily activities, unlike in industrialised areas where lifestyle choices
were the most important predictor of ecological knowledge. For instance, ecological knowledge was
higher in individuals that grew up in rural areas, lived in rural areas during adulthood and made
frequent visits to the countryside. Word-of-mouth and direct experience were found to be the most
effective modes of knowledge transfer across all sites. The revealed patterns of knowledge loss
contribute to our understanding of the future of ecological knowledge bases globally and action that
may be taken to prevent further decline in the light of economic development.
Acknowledgements
I would like to take this opportunity to thank my supervisors Professor Jules Pretty and Dr David
Smith for their constant support, encouragement, patience and direction. I would also like to
acknowledge Chris and Pete Hailes who have given me constant support during UK field work and
provided me with a second home to come back to after many long hours of interviewing. I am
grateful to John Devavaram, Erskine Arunothayam and A. Thangham for their assistance and
guidance in India and to Julian Clifton, Chris Majors and Le Mido for their encouragement, support
and help in Indonesia. Also thanks to Jo Peacock and Rachel Hine for their advice and support
particularly whilst writing up.
Several people have been a constant source of encouragement and strength throughout, as well as an
essential sounding board, in particular Michelle Hailes, Helen Smith, Leanne Cullen and Steve
Cooley. Also a huge thank you to Ben Morrison for years of love, kindness, support and faith in me.
Finally a big thank you to my family, namely both my nans, together with Ian and Denise for their
support and love, and most importantly my mother and father, who showed never-ending faith,
patience, support and love throughout my years of study.
This work was funded by a NERC studentship (NER/S/A/2002/10365).
List of Contents
Summary
Acknowledgements
List of Contents
List of Figures
List of Tables
List of Abbreviations
Page
3
4
5
10
14
15
Chapter 1. Introduction
17
1.1. Composition of LEK

1.2. Transfer of LEK
1.3. The functional values of LEK
1.3.1. Natural resource conservation
1.3.1.1. Self-management
1.3.1.1.1. The development of local management systems
1.3.1.1.2. The attributes of self-management systems
1.3.1.1.3. The adaptability of self-management systems
1.3.1.1.4. Self-management systems in the future
1.3.1.2. State-management
1.3.1.3. Co-management
1.3.2. Food stability and genetic diversity
1.3.3. Pharmaceutical industry and economic generation
1.3.4. The promise of future findings
1.3.5. Cultural capital
1.4. Factors leading to LEK loss
1.4.1. Urban migration and urbanisation
1.4.2. Modernisation of lifestyles and technology
1.4.3. Extinction of experience
1.4.4. The introduction of schooling and curriculum changes
1.4.5. Economic development and marketisation
1.4.6. Assimilation and integration
1.4.7. Language loss
1.5. General population distribution of LEK
1.5.1. Gender differences
1.5.2. Generational differences
1.6. Aims of thesis
17
19
20
20
21
22
22
24
25
27
28
29
30
31
32
34
34
35
36
36
37
38
39
40
41
43
44
Chapter 2. Methods
49
2.1. Introduction
2.2. Study sites
2.3. Ethnobotanical and ethnozoological interview development
2.3.1. Interview changes and methodological development
2.4. Sampling strategy
49
51
53
54
59
Chapter 3. India case study: Virudhunagar, Tamil Nadu
64
3.1. Introduction
3.1.1. Indias dependence upon the land
3.1.2. Indias biological diversity
3.1.3. The state of Tamil Nadu
3.1.4. Virudhunagar district
3.1.5. Tiruchuli block
64
64
68
69
72
73
3.2. Site specific methodology

3.2.1. Study site selection
3.2.2. Species selection
3.2.3. Sampling strategy in the field
3.2.4. Ethnobotanical interview structure
3.2.5. Statistical analysis
74
74
74
75
77
78
3.3. Results
3.3.1. The effect of location
3.3.2. The effect of wealth
3.3.3. The effect of gender
3.3.4. The effect of age
3.3.5. The effect of species utility
3.3.6. The effect of healer status
79
80
80
83
83
87
89
3.4. Discussion
91
91
92
93
94
96
97
Chapter 4. UK case study: Comparing rural Lincolnshire, urban Kent

and coastal Essex
98
4.1. Introduction
4.1.1. The Lincolnshire study site
4.1.2. The Kent study site
4.1.3. The Essex study site
98
99
102
105

4.2.4. Knowledge assessment
107
107
108
109
113
114
4.3. Results
4.3.1. The effect of site
4.3.4. The effect of income
4.3.5. The effect of occupation
4.3.6. The effect of childhood location
4.3.7. The effect of length of residence
4.3.8. The effect of frequency of countryside visits
4.3.9. The effect of visiting the countryside for different purposes
4.3.10. The relationship between attitude towards ecological knowledge
and knowledge level
4.3.11. The relationship between knowledge source and knowledge level
4.3.12. The effect of quality of experience
115
115
116
120
127
127
128
128
131
131
135
4.4. Discussion
4.4.10. The relationship between attitude towards ecological knowledge
and knowledge level
145
145
146
147
148
148
148
149
149
150
151
135
140
151
153
Chapter 5. Indonesia case study: Kaledupa, SE Sulawesi
155
5.1. Introduction
5.1.1. The Wakatobi Marine National Park
5.1.2. The island of Kaledupa
5.1.3. The Orang Pulo and the Orang Bajo
155
156
158
161

165
165
166
167
169
171
5.3. Results
5.3.1. The effect of ethnic group
5.3.5. The effect of primary occupation
172
173
181
184
184
184
5.4. Discussion
191
191
193
195
196
196
Chapter 6. Country comparisons
198
6.1. Introduction
198
6.2. Methodology
201
6.3. Results
6.3.4. The effect of experience and different sources of transfer
203
203
204
204
206
6.4. Discussion
209
209
210
211
212
Chapter 7. General discussion
214
7.1. Developing a standardised protocol for ecological knowledge assessment

within and between communities across different cultures
7.2. Understanding the effect of economic development on knowledge
distribution both within and between communities
7.3. Discovering the key factors that lead to high and low levels of ecological
knowledge based on quantitative methods and if these differ between
localities
7.4. Determining whether progressive knowledge loss is occurring within the
chosen communities and the reasons for its existence or absence
7.5. To understand patterns of knowledge acquisition with age at all sites and to
test the hypothetical acquisition curves shown in Fig 1.5.2
7.6. Detecting differences in familiarity with different species groupings and
reasons for such differences
7.7. General conclusions
7.7.1. Generic conclusions
7.7.2. Developing region conclusions
7.7.3. Industrialised region conclusions
214
216
219
220
223
225
227
227
228
228
References
233
Appendix I
Examples of species flashcards used in fieldwork
248
Appendix II
India case study: Wild flowering plant species list
250
Appendix III
UK case study: Wild species list
252
Appendix IV
Indonesia case study: Wild species list
255
Appendix V
India case study: Data recording sheet
257
Appendix VI
UK case study: Data recording sheet
259
Appendix VII
Indonesia case study: Fishers and farmers data recording sheets
264
Appendix VIII
Recalculating the HDI figure for the Indonesian study site
280
10
List of Figures
Page
Figure 1.5.1. Levels of ecological knowledge and understanding
Figure 1.5.2. Theoretical acquisition curves
46
48
Chapter 2. Methods
Figure 2.2.1. The 3 chosen study regions
53

Figure 3.1.1. Map of India showing the southern state of Tamil Nadu
Figure 3.3.1. Comparing levels of other use knowledge across 2
village locations
Figure 3.3.2. Comparing wealth rank across 2 village locations
Figure 3.3.3. Mean levels of community ecoliteracy and relationship with
community wealth rank
Figure 3.3.4. Mean levels of community knowledge variance and relationship
with community wealth rank
Figure 3.3.5. Levels of knowledge variance in 5 villages and relationship with
(A) mean levels of community ecoliteracy and (B) mean levels of
community use knowledge
Figure 3.3.6. Comparing the mean levels of agricultural use knowledge
between men and women
Figure 3.3.7. Comparing the mean number of uses known per recognised
species between adults and young people
Figure 3.3.8. Comparing differences in level of health knowledge between
adults and young people
Figure 3.3.9. Comparing differences in level of health knowledge across 2 age
groups in both women and men
Figure 3.3.10. Comparing the changes in mean ecoliteracy with age between
women and men showing the maximum age of knowledge
saturation for both
Figure 3.3.11. Levels of species familiarity with local villagers and relationship
with the number of uses a species holds
Chapter 4.
Figure 4.1.1.
Figure 4.1.2.
Figure 4.1.3.
Figure 4.1.4.
UK case study: Comparing rural Lincolnshire, urban Kent

and coastal Essex
Map of England showing the location of the study sites within the
3 selected counties
Map of Lincolnshire County indicating the location of the rural
study site
Map of Kent County indicating the location of the urban study
site
Map of Essex County indicating the location of the coastal study
site
70
81
81
82
82
84
85
85
86
86
88
90
99
102
104
106
11
Figure 4.3.1. Comparing UK site differences in (A) level of total ecoliteracy,

117
(B) use knowledge (total number of uses) and (C) use knowledge
(% species for which uses were identified)
Figure 4.3.2. Comparing UK site differences in level of terrestrial ecoliteracy 118
Figure 4.3.3. Comparing the mean % of species that respondents were able to 118
identify in the different species groups
Figure 4.3.4. Comparing rural gender differences in (A) level of mammal
119
ecoliteracy and (B) level of flower ecoliteracy
Figure 4.3.5. Comparing urban gender differences in (A) level of total
121
ecoliteracy, (B) level of flower ecoliteracy and (C) level of crop
ecoliteracy
Figure 4.3.6. Comparing coastal gender differences in level of flower
122
ecoliteracy
Figure 4.3.7. Comparing age group differences in total ecoliteracy at (A) the
123
rural site, (B) the urban site and (C) the coastal site
Figure 4.3.8. Comparing age group differences in (A) use knowledge (total
124
number of uses) and (B) use knowledge (% species for which uses
were identified)
Figure 4.3.9. Comparing age group differences in (A) level of bird ecoliteracy, 125
(B) level of flower ecoliteracy, (C) level of crop ecoliteracy and
(D)level of marine ecoliteracy
Figure 4.3.10. Comparing rural age group differences in level of mammal
126
ecoliteracy
Figure 4.3.11. Comparing differences in total ecoliteracy between different
129
occupation groups at (A) the rural site, (B) the urban site and
(C) the coastal site
Figure 4.3.12. Comparing how differences in childhood environment affect total 130
ecoliteracy
Figure 4.3.13. Comparing how differences in years lived at a site affect total
130
ecoliteracy
Figure 4.3.14. Comparing how differences in frequency of countryside visits
132
affect total ecoliteracy
133
affect (A) level of mammal ecoliteracy, (B) level of bird ecoliteracy,
(C) level of flower ecoliteracy and (D) level of crop ecoliteracy
Figure 4.3.16. Comparing how differences in reasons for visiting the countryside 134
affect total ecoliteracy
Figure 4.3.17. Comparing differences in reasons for making countryside visits
136
at (A) the rural site, (B) the urban site and (C) the coastal site and
how they differ between the 3 sites
Figure 4.3.18. Views towards knowledge importance ranked and relationship
137
with an individuals total ecoliteracy
Figure 4.3.19. Comparing the % knowledge contribution from (A) television,
139
(B) parents/relatives, (C) friends/locals, (D) occupation and (E)
hobbies between the 3 UK sites
Figure 4.3.20. Comparing the % knowledge contribution from (A)
141
parents/relatives and (B) occupation between men and women
12
Figure 4.3.21. Comparing the % knowledge contribution from (A) books, (B)
142
friends/locals, (C) hobbies, (D) television, (E) education and (F)
parents/relatives between the different age groups
Figure 4.3.22. Comparing quality of experience scores between the 3 UK sites
144
Figure 4.3.23. Comparing quality of experience scores between the different age 144
groups
Chapter 5. Indonesia case study: Kaledupa, SE Sulawesi
Figure 5.1.1. Map showing the location of the Wakatobi Marine National Park, 157
Indonesia
Figure 5.3.1. Differences in levels of use knowledge indicated by (A) total
174
number of uses listed and (B) number of uses known per identified
species between 2 ethnic groups
Figure 5.3.2. Differences in (A) economic use knowledge and (B) consumption 175
use knowledge between 2 ethnic groups
Figure 5.3.3. Differences in the use knowledge of children (aged 15-19 years) 176
between 2 ethnic groups
Figure 5.3.4. Levels of ecoliteracy of (A) terrestrial plants, (B) terrestrial
178
animals and (C) marine species between 2 ethnic groups
Figure 5.3.5. Comparisons of the mean changes in total ecoliteracy with age
179
between 2 ethnic groups showing the maximum age of knowledge
saturation for both groups
Figure 5.3.6. Levels of Bajo mens marine ecoliteracy across all 4 age groups 179
measured illustrating knowledge saturation in the 3 eldest groups
Figure 5.3.7. Levels of support for local traditions (ranked) across 2 ethnic
180
groups
Figure 5.3.8. Comparing Bajo and Kaledupan (A) mean levels of wealth with 182
maximum age of knowledge saturation and (B) mean use knowledge
Figure 5.3.9. Mean levels of community use knowledge in 6 villages and
183
relationship with mean primary household income
Figure 5.3.10. Mean levels of community use knowledge in 6 villages and
183
relationship with the % of resource dependents within a community
Figure 5.3.11. Levels of variance in community ecoliteracy in 6 villages and
185
relationship with mean primary household income
Figure 5.3.12. Levels of variance in community ecoliteracy in 6 villages and
185
relationship with mean levels of community use knowledge
Figure 5.3.13. Comparing the levels of total ecoliteracy between men and
186
women
Figure 5.3.14. Comparing the levels of Bajo use knowledge between men and
186
women
Figure 5.3.15. Comparing patterns of age group differentiation in (A) total use
187
knowledge and (B) overall ecoliteracy levels
Figure 5.3.16. Comparisons of the mean changes in ecoliteracy with age between 188
men and women showing the maximum age of knowledge saturation
Figure 5.3.17. Comparing fishers and farmers levels of ecoliteracy of (A)
190
terrestrial plants and (B) terrestrial animals
13
Chapter 6. Country comparisons

Figure 6.3.1. Comparing age differentiation in plant knowledge between the
UK, India and Indonesia
Figure 6.3.2. Comparing the mean changes in plant ecoliteracy with age
between the UK, India and Indonesia
Figure 6.3.3. Mean plant ecoliteracy in the UK, India and Indonesia and
relationship with human development index and per capita
Figure 6.3.4. Comparing mean levels of plant ecoliteracy with community
knowledge variance in the UK, India and Indonesia
Figure 1.5.2. Theoretical acquisition curves
Figure 7.6.1. Ecological knowledge distribution patterns at the 3 sites based
on economic development levels
205
205
207
207
224
230
14
List of Tables
Page
Chapter 1.
Table 1.1.1.
Introduction
Distinctions between LEK and the Western definition of science
18
Chapter 2.
Table 2.2.1.
Table 2.3.1.
Table 2.3.2.
Table 2.4.1.
Methods
Key characteristics of study site regions
Species selection criteria
Stages of methodological development at all 3 sites summarised
Code of best practices for LEK research
52
56
58
62
Chapter 3.
Table 3.1.1.
India case study: Virudhunagar, Tamil Nadu

Annual averages in rainfall over the state of Tamil Nadu, southern 70
India (1999-2003)
Key characteristics of villages selected for study
74
Summary of respondent group sizes and demographics from each 76
village
Table 3.2.1.
Table 3.2.2.
Chapter 4.
Table 4.1.1.
Table 4.2.1.
Table 4.2.2.
Table 4.2.3.
Table 4.2.4.
Table 4.2.5.
Table 4.2.6.
Table 4.3.1.
Table 4.3.2.
Table 4.3.3.
Table 4.3.4.
Chapter 5.
Table 5.2.1.
Table 5.2.2.
Table 5.3.1.
UK case study: Comparing rural Lincolnshire, urban Kent and

coastal Essex
Site characteristics of the 3 areas selected for study within the UK 98
Rural villages key characteristics and attributes
107
Urban wards key characteristics and attributes
108
Coastal towns key characteristics and attributes
108
Rural sampling strategy showing minimum target quotas for
111
all sub-samples
Urban sampling strategy showing minimum target quotas for
112
all sub-samples
Coastal sampling strategy showing minimum target quotas for
113
all sub-samples
Statistical results for age differences in ecoliteracy levels of the
127
different species groups at all 3 sites
Statistical results for the effect of countryside visits on ecoliteracy 131
levels of different species groups
Statistical results for the association between % source
138
contribution and ecological knowledge level
Statistical results for the association between quality of
143
environmental experience and knowledge of each species group
Indonesia case study: Kaledupa, SE Sulawesi
Demographics and characteristics of villages selected for study
Breakdown of site sampling strategy: Village sub-sample quotas
Levels of fishing practices exploited by 2 ethnic groups and
details of the practices
166
168
177
15
List of Abbreviations
ASHE
BC
df
CCSR
CES
CIRAN-MOST
COREMAP
DEFRA
DNA
EIA
EMDA
GDI
GDP
HDI
HPI
HYV
IIED
IUCN
km
LEK
LIPI
mm
NGO
NIC
ODPM
OECD
ONS
p pp
pa
PGR
Rp
Rs
SA
SDNP
SE
SEPASAL
SOC
SPEECH
SPSS
TCDC
TIDCO
UK
UN
Annual Survey of Hours and Earnings

Before Christ
Degrees of Freedom
Cathie Marsh Centre for Census and Survey Research
Centre for Ecological Sciences
Centre for International Research and Advisory Networks and
Management of Social Transformations Programme
Coral Reef Rehabilitation And Management Project
Department for Environment, Food and Rural Affairs
Deoxyribonucleic Acid
Energy Information Administration
East Midlands Development Agency
Gender-related Development index
Gross Domestic Product
Human Development Index
Human Poverty Index
High Yielding Variety
International Institute for Environment and Development
International Union for Conservation of Nature and Natural Resources
Kilometres
Local Ecological Knowledge
Lembaga Ilmu Pengetahuan Indonesia (Indonesian Institute of
Science)
Millimetres
Non-Governmental Organisation
National Informatics Centre
The Office of the Deputy Prime Minister
Organisation for Economic Co-operation and Development
Office of National Statistics
Purchasing Power Parity
Per Annum
Population Growth Rate
Rupiah
Rupees
Maximum Knowledge Saturation Age
Sustainable Development Networking Programme
Standard Error
Survey of Economic Plants for Arid and Semi-Arid Lands
Standard Occupation Classification
Society for Peoples Education and Change
Statistical Package for the Social Sciences
Technical Cooperation among Developing Countries
Tamil Nadu Industrial Development Corporation
United Kingdom
United Nations
16
UNDP
WMNP
WWF
United Nations Development Programme

Wakatobi Marine National Park
World Wide Fund for Nature
17
The last two decades have witnessed heightened interest in the knowledge base termed traditional,
indigenous and local ecological knowledge (Gadgil et al., 1993; Mauro & Hardison, 2000;
Turner et al., 2000; Byg & Balslev, 2001; Davis & Wagner, 2003; Folke, 2004; Kaihura, 2004) or
ecoliteracy; ecology being the study of the natural systems around us and literacy being the
intellectual frameworks that support this knowledge (Fisher, 2005; Haigh, 2005). With this has
evolved an understanding of its evolution, its composition and its distribution in traditional
communities. It is thought that this knowledge base is primarily an attribute of pre-scientific
societies, where knowledge accumulation is orally-transferred, slow-paced and geographicallyisolated, and yet highly detailed and accurate despite changing environmental conditions.
In the 5-7 million years we spent as hunter-gatherers, our knowledge base evolved with the
ecosystems within which it exists (Pretty, 2002). It has further developed as a result of historical
continuity of resource dependence and closeness of relations between a society and its environment,
understanding both environmental conditions as well as complex ecological interactions locally
(Gadgil et al., 1993; Davis & Wagner, 2003). This knowledge base has been termed Local
Ecological Knowledge (LEK) or ecoliteracy in this study, reflecting its place-based nature rooted in
the history, geography and culture of a site without being static, outdated and retrogressive, with all
that is traditional being its dynamic nature and mode of oral transfer (Ekpere, 2004).
1.1. Composition of LEK

LEK evolves through one generations detailed observations and experiences being orally
transferred to the next. The following generation combine this knowledge with their own
observations in the field including changes in environmental conditions. Consequently a stockpile of
experiences and observations based on close daily interactions of consecutive generations is formed
(Gadgil et al., 1993; Pierotti & Wildcat, 2000; Davis & Wagner, 2003). LEK as a knowledge base is,
therefore, considered place-based and integral to its holders as it contains their own experiences in
addition to that of their ancestors and evolves with them throughout their lives. This belief is
strongly held by many traditional communities including, for instance, the Inuit of the Canadian
Arctic (Wenzel, 1999).
18
A key feature of LEK is its non-static ever-evolving nature, responding to ecosystem changes when
and where they happen. Unlike Western science, which is recorded in texts making it static and
globally-applicable, LEK is a situated practice that goes unwritten, allowing it to adapt to the
inevitable evolution of local ecosystems upon which it is based (Long et al., 2003). Thus like true
scientists, indigenous people inherit knowledge, test it in their own environments and then compare
their own observations adapting the knowledge base accordingly (Alexiades, 1999; Pierotti &
Wildcat, 2000; Roth, 2004).
LEK has often been thought of as unscientific and therefore unreliable, despite its similarities to
Western science; evolved as a situated practice drawn from the observations of a series of actors
working in particular social, political and ecological contexts (Brodt, 2002; Davis & Wagner, 2003).
However, unlike scientific knowledge which is general, segmented and universally-relevant, LEK is
detailed, holistic and adapted to context (Roth, 2004) (Table 1.1.1). Although often termed
traditional knowledge, nothing about this knowledge base is traditional, with the exception of its
transfer routes. In actual fact, it is exceedingly current (Gilchrist et al., 2005). Broken down, LEK is
essentially a collection of rules-of-thumb derived from trial and error processes that form a
cumulative body of knowledge classifying and describing environmental objects, activities and
events in the locality as they happen (Gadgil et al., 1993)
Table 1.1.1. Distinctions between LEK and the Western definition of science
Local ecological knowledge
Tacit knowledge
Knowledge of experience
Simultaneous here and now knowledge
Practical knowledge base
Adapted from Rahman, 2004
Western scientific knowledge

Explicit knowledge
Knowledge of mind
Sequential there and then knowledge
Theoretical knowledge base
Knowledge content ranges from best practices for harvesting natural resources to understanding
resource occurrence, distribution and the environmental interactions affecting this. Specific
knowledge base composition of a society will depend upon what is important to a societys cultural
beliefs, to its survival and to that of its homelands (Davis & Wagner, 2003; King, 2003). Knowledge
content can be highly heterogeneous, both spatially and temporally, especially in highly stratified
19
societies like India where differential access to learning resources exists (Gadgil et al., 1993; Brodt,
2002).
The LEK base of many societies goes a long way towards forming their cultural traditions, belief
systems and worldviews (Wenzel, 1999). Consequently many traditional communities consider
human society as integral to nature rather than external to it, and value all living things as distant
relatives that deserve respect (Berkes et al., 2000; Pierotti & Wildcat, 2000; Salmon, 2000).
1.2. Transfer of LEK

Necessary to LEK survival in traditional communities is its continual transfer between subsequent
generations. Traditional knowledge transfer occurs via two routes; word-of-mouth and experience
through observation. The former is via story-telling practices between networks of friends and
relatives; the latter is through observing the behaviour and actions of friends and relatives. Despite
effective transfer methods, the tendency of LEK to go unwritten and outside of recorded history
often results in the disillusion that it is an inferior knowledge base (Gadgil et al., 1993; Scoones &
Thompson, 1994; Young, 2002).
Oral transmission commonly occurs vertically through family lines from parents and grandparents
telling narratives, including stories and songs, repeatedly, either formally at social occasions or
informally between family members late into the night (Alexiades, 1999; Turner et al., 2000; Singh
& Singh, 2005). The Bajo of Indonesia, for instance, hold ritual curing ceremonies and inter-village
feasts crucial to LEK transfer and inheritance (Rachman, 1993). Members of the Blackfoot tribe of
North America recall memories of elders stories lasting all night and story-tellers using mnemonic
devices from talking sticks to wampum belts to retell tales (Peat, 1996). In some communities intergroup connections including trading routes also contribute to existing LEK bases (Turner et al.,
2000). The second mode of LEK transfer in traditional communities is personal experience, acquired
through observation and practical implementation. In many traditional communities, this involves
children silently observing land management practices, participating in them and then applying the
practical skills themselves (Wenzel, 1999; Turner et al., 2000; Brodt, 2002; Harmon & Loh, 2004),
making it as much a skill of perception and practical demonstration as a concept (Sillitoe, 1998).
20
Unlike text book information, tacit knowledge learned through direct experience, such as riding a
bicycle, is unlikely to be forgotten (Rahman, 2004). Cree children living in the Canadian sub-arctic
inherit essential hunting skills this way, until one day they are considered ready to go out alone and
put what they have learned into practice (Peat, 1996). Healing is a prime example of elders acting as
role models and passing on their learned knowledge, however some claim to learn healing
techniques through other methods such as their dreams (Alexiades, 1999; Vandebroek et al., 2004).
Traditional practices derived from the transfer of local knowledge and self-enforced through
community custom include the protection of keystone, sacred or medicinal species, rotational
resource use and temporal harvesting restrictions (Feit, 1988; Turner et al., 2000; Drew, 2005;
Ghimire et al., 2005; Singh et al., 2006a). Such practices are an antecedent of farmers in-depth
understanding of local weather conditions, pest and plant relations, soil fertility and local species
classification (Singh & Singh, 2005). For instance, indigenous non-cultivators are believed to know
as many as 200 local plant categories, reaching 500 amongst cultivators, and approximately 390
animal categories (Krech, 2005).
1.3. The functional values of LEK

Historically, LEK was nothing less than a life-line for gatherers to harvest edible plants at the
correct time and for hunters to identify potential prey and find their way home. Consequently,
aboriginal cultures gave birth to generation after generation of excellent naturalists, or else they
perished (Pyle, 2003). Reliance upon LEK as a survival tool is still common today in developing
areas rich in biodiversity but deprived of modern facilities, such as healthcare and supermarkets. In
regions like this, LEK acts not only to sustain the local population through meeting their physical,
social, cultural and spiritual needs, but also to maintain biodiversity through traditional management
practices (Yupari et al., 2004).
1.3.1. Natural resource conservation

Highly diverse landscapes, the primary focus of conservation efforts, have remained occupied by
human populations throughout history, and will continue to be into the future. Human activities are
21
inevitably placing pressure upon the environment and subsequently changing its composition both
locally and globally (Hooper et al., 2005, Smith et al., 2006). Human activities aimed at maximising
productivity (domestication and monocropping) and globalising environmental control are creating
phase shifts both gradually (by global market demands or climate change) and more abruptly
(hurricanes or disease outbreaks), the latter of which are increasing in frequency (Folke et al., 2005).
Conservationists in the field have to make concise decisions on the realistic options available for the
management of resources within environmental systems, aside from the speculation of management
options widely discussed in theory (Alcorn, 1993). In a deviation from previous trains of thought,
policy makers, conservationists and academics alike, have recently come to consider the importance
of local knowledge systems in resource management, as a consequence of the poor outcome of
government efforts (Fazey et al., 2006).
1.3.1.1. Self-management
LEK-based management systems have entered the limelight as an alternative to the failures of
exclusion conservation and to the traditional top-down management approach (Davis & Wagner,
2003), as seen in rainfed areas during the Green Revolution for example. This aimed to maximise
yields using good water supply and high-quality chemical fertilisers. However, the lack of financial
and environmental (good soil and water) resources in most areas were not accounted for, and
productivity subsequently declined in the areas that needed it the most (CIRAN-MOST, 2003). It is
being realised increasingly that conservation initiatives that take into account peoples needs and
resources are more likely to be successful. Hence community resource management practices,
developed primarily for long-term survival, can often be successfully used to conserve local
ecosystems today (Parlee et al., 2005).
A community is perceived as a spatially-defined unit with a well-defined social structure (Agrawal
& Gibson, 1999). In the past, intrusive strategies and externally-planned regulations excluding
community stakeholders have overruled traditional knowledge bases and systems of management
often at a great cost to local communities with limited conservation benefits yielded. Shortcomings
have included government instability, monitoring and enforcement limitations and inaccessible,
unavailable or outdated science. Hence there has been a recent surge of resources into community-
22
management schemes from international agencies, including the World Bank, Worldwide Fund for
Nature and The Nature Conservancy (Agrawal & Gibson, 1999).
1.3.1.1.1. The development of local management systems

For ninth-tenths of human existence on earth, hunting and gathering pressures exerted on any
ecosystem have had to be constrained for the survival of the community and wildlife populations
through to the present day. Indigenous groups had no choice but to learn sustainable methods of
harvesting local resources without depleting them, based on their own detailed knowledge, in order
to survive (Ghimire & Pimbert, 1997). The Raramuri of Mexico, for instance, have sustained
themselves and their children for generations using 350 different local plant species as food and
medicines. Despite local resource extraction, they have succeeded in upholding the biodiversity
levels of the region for over 2000 years, through practices such as selective harvesting and
environmental modification (Salmon, 2000; LaRochelle & Berkes, 2003). Developing knowledge
bases that understand how to sustain biodiversity whilst harvesting natural resources reflects
indigenous peoples stake in their local environment for both their own survival and that of future
generations (Gadgil et al., 1993).
Sustainable management techniques are key to all subsistence-based sedentary societies, including
fishing, horticultural and agricultural-based communities, and contribute to animal husbandry,
forestry and agroforestry, pharmaceutical industries, marine science and fisheries (Gadgil et al.,
1993; Ruddle, 2000). Ecosystem knowledge, that ensures long-term resource sustainment, includes
detailed information on physical environments, species behaviour, aggregations, taxonomies,
specific habitats and resource abundance (Ruddle, 2000). Ethiopians have successfully hunted civets
for musk for generations by employing stringent local regulations based on their own intricate
knowledge, including releasing females and avoiding areas where previously caught females are
sighted (IIED, 1994).
1.3.1.1.2. The attributes of self-management systems

Self-management practices derive from a sense of ownership and responsibility (Phillips et al.,
1994) for the resource combined with an in-depth knowledge of the system, its components and
their interactions, reinforced by a set of socially-accepted informal rules. These rules comprise the
23
backbone of the community with respect to worldview and are termed social norms. Some have
even evolved more recently in a local response of resistance to government-enforced management
efforts (Feit, 1988).
The nature of locally-enforced, socially-accepted behaviour norms can strongly influence the
success of self-management (Agrawal & Gibson, 1999). For instance, where established norms act
to limit fishing to certain areas of a coral reef at certain times and protect areas of regeneration, such
as aggregation and spawning sites, community norms ensure management is self-sustaining.
However, where accepted norms promote excessive exploitation, for instance in Indonesia where
certain marginal tribes believe that using mined coral to build foundations makes the house stronger,
societal norms may act to the detriment of the environment. Strategies for the intrinsic alteration of
long-existing community norms are unknown (Rudd et al., 2003).
Social norms held by the islanders of Ahus Island, Papau New Guinea, include prohibiting spear and
net fishing and strictly limiting invertebrate harvests within six demarcated areas (tambu) of the reef
lagoon upon which they rely. Tambu fishing is only permitted for significant ceremonial occasions
that occur up to three times annually. At these times, tambu areas are opened very briefly (2-3
hours) for intensive exploitation to provide food for consumption at the ceremony. As a result, target
species extracted from the restricted areas are over 20% larger and fish biomass caught, 60% higher.
Such management practices are embedded in tradition rather than conservation-oriented, but
succeed in yielding the same results. Despite intense resource dependence for both sustenance and
economic incentives, compliance is successful through the perceived legitimacy of behavioural
norms in benefiting the whole of society, combined with moral influence from peers. Also,
exclusive ownership rights over resources in the direct vicinity induce a sense of community
responsibility reinforcing compliance (Cinner et al., 2005).
The treatment of resources, according to these norms, devolves into a local pattern of customary,
and thus expected, behaviour by which everyone conforms as thought to be in the best interests of
the community. No government enforcement is required, but instead local taboos exist, reducing
cost expenditure by government bodies on enforcement sanctions, monitoring and punishment of
rule-breakers, that arises with other forms of management (Rudd et al., 2003). This evolution of
24
traditional practices, that act to sustain the local ecosystem in its complexity, forms a selfmanagement system that is adaptive to phase shifts in the environment as they begin to emerge.
Costs of making collective decisions are reduced between individuals sharing a spatial unit and
communal norms, and although social stratification is inevitable in any human coercion, even in
terms of gender relations, power differentials are likely to be far less pronounced and far better
understood within an established community (Agrawal & Gibson, 1999). Thus LEK-based
management, as a form of social capital, empowers societies that are otherwise powerless by
enabling them to self-manage their own resources (Ruddle, 2000; Davis & Wagner, 2003).
1.3.1.1.3. The adaptability of self-management systems

Many self-management systems have proved successful at maintaining a resource over a long
historical period (Feit, 1988). However over time, any system of management would have to adapt
to the dynamic nature of an environmental system. Past observations of changes in ecosystem
composition and phase shifts enable traditional self-management systems to contain a delicate
feedback system that allows for subtle changes in the environment to be detected and accounted for
in the relevant adjustment of practices (Fazey et al., 2006). This is the key component by which a
management system is deemed successful over time and requires an intricate understanding of the
ecology of the system and its components.
Appreciating and adapting to the uncertainty of nature is key to agricultural systems globally. Entire
crop yields can be destroyed by a single pest or disease. The rural poor in particular rely heavily on
their ancestors knowledge that has sustained crop yields, even under adverse conditions, for
generations (Pyle, 2003; Fenta, 2004). Consequently, the rural poor are responsible for a great deal
of our traditional global seed variety by passing on seeds through family and community networks
(Soleri & Cleaveland, 1993).
Several decades ago in Bangladesh modern agricultural methods were introduced in an effort to
maximise rice production. A quarter of a century on, local soils have been found to be depleted of
organic matter, nitrogen, phosphorus, zinc, sulphur, boron and other essential elements, water bodies
are suffering from increased sedimentation and there has been widespread genetic erosion reducing
25
8500 cultivars of rice to only a few dozen today. All as a consequence of the abandonment of
traditional LEK-based management practices (Ahmed, 2004).
Through traditional biomonitoring techniques that involve observing shifts in resource abundance,
local managers can improvise upon pre-existing practices (often by self-imposed restrictions or
altered efforts) to accommodate environmental changes and, thus, successfully manage resources for
future generations. In addition to biotic changes, traditional management systems have had to adapt
to anthropogenic changes as other peoples enter the area. External threats to resources from
outsiders have become more frequent in recent decades, but throughout history resource users have
been forced to extend management practices into new areas to adapt to these impositions and will
need to continue to monitor external access in the future (Feit, 1988; Agrawal & Gibson, 1999).
Hence LEK-based self-management of ecosystems does not have a history of continual success in
the light of environmental change, but rather a history of disruption, adaptation and self-renewal
(Feit, 1988).
1.3.1.1.4. Self-management systems in the future

When the Maasai were forcibly relocated from their homelands in Kenya, their traditionalist form of
pastoralist self-management ceased, resulting in widespread scrubland invasion. As a result,
antelope pastureland declined and ecosystem collapse followed (Pretty, 1995). A similar situation
arose when the Maori tribes of India were relocated. Subsequently paspalam grasslands declined as
did wild bird populations (Pretty, 2002). The introduction of regulations forcing hunters to catch
large male caribou on Ellesmere Island, Canada, led to similar tragedy. The advice of the natives,
that these individuals ensure population stability by protecting the young and the pregnant and
finding food, was neglected by the authorities, hence population collapse followed (Ghimire &
Pimbert, 1997). Therefore, such in-depth knowledge has the potential to aid conservation efforts
globally in the future, in the designation and management of protected areas for instance, especially
in the tropics where systems are diverse and local resource dependence is high, yet scientific data
are limited (Ruddle, 2000).
Despite an assemblage of successful self-management systems, with survival of the community and
ecosystem acting as testimony of not only local knowledge but its effective application (Feit, 1988),
26
sustainability is not an automated response to self-management. Thus it may be possible to reestablish equitable human-environment partnerships that meet similar conservation goals today. As
already outlined, for a management system to be successful and self-enforcing it requires four
components; intricate knowledge and understanding of the ecosystem, ownership responsibility,
adaptive capacity and the social institutions capable of enforcement. Where one of these
components is lacking, self-management is likely to collapse.
It is important to acknowledge that not all indigenous societies have attained a harmonious balance
of utilisation and conservation through their knowledge-based systems (Sillitoe, 1998; Gilchrist et
al., 2005). Localised failures in management have led to species extinctions in different regions all
over the world (Gadgil et al., 1993; Murdoch & Clark, 1994). This is most likely to occur where
people are new to environments and only recently settled. Overexploitation is a product of local
peoples not yet able to understand, classify and name local species (Skutnabb-Kangas, 2004).
Conversely, there have been examples of the LEK of industrialised communities making a
significant contribution to local conservation. For instance, in Sweden expert site-specific
knowledge was used to successfully sustain watersheds and wetlands suffering from exploitation
pressures (Olsson & Folke, 2001; Olsson et al., 2004).
Increased consumption pressures from population expansion combined with technological
innovations and institutional alterations have the capacity to deplete traditional practices.
Additionally, marketisation acts to place cash values on common property systems. This has the
potential to create local intensification of resource extraction, the abandonment of traditional
resource collection methods and the knowledge bases that sustain them, and cause environmental
degradation. These pressures, combined with weak non-specific property rights, act to renegotiate
resource extraction incentives (Agrawal & Yadama, 1997; Agrawal & Gibson, 1999). Therefore, if
isolated communities become externally resource-reliant, their intrinsic concern and knowledge of
local resource bases may become diluted and devoid of purpose, causing management systems to
collapse (Smith et al., 2006).
27
1.3.1.2. State-management
State-management is a form of management deriving from the legal authority of the nation state.
When national interests lie in conserving a local resource, management of that resource is often
taken out of the hands of local users and transferred to the hands of professional policy makers and
scientists (Feit, 1988). State management exerts authority from a distance and overrides the benefits
of territoriality and personal interests held by settlements local to the resource. Instead of being
locally-derived, the knowledge system state-management relies upon evolves further afield in
Western scientific institutions. Thus it relies upon a different, some might say disconnected, set of
components to local systems of community management.
Although new institutions have in mind a sustainable management concept, by disregarding local
knowledge, environmentally and culturally, enforcement is jeopardized. Management practices
imposed in state systems are often based upon conceptual theory of sustainability, rather than
generations of observations. In truth, no outsider can hope to acquire a fraction of the botanical,
climatological and ecological knowledge and understanding of a community elder. Since effective
sampling of all of the worlds ecosystems is impossible, state management theory is based upon
results generated from periodic surveys. These tend to be isolated in a specific time and place,
incorporating few of the environmental parameters at work in an ecosystem.
The synthesis of local knowledge on site permits anticipation and response to environmental
alteration, unlike state theory whereby changes are only detected when they reach large-scale phase
shifts. State management responds to such changes by returning to desktop-modeling based on data
from similar ecosystems and relying upon assumptions to fill in gaps in the data. This means the
management body cutting itself off from local ecological understandings and cultural norms in a
situation where it can ill-afford to do so (Feit, 1988).
Self-management, therefore, has the capacity to provide more rapid responses to sudden and often
complex environmental changes. Government agencies, on the other hand, lack the ecosystem
experience of local resource users and focus upon quantitative predictions of potential outcomes
(Lundquist & Granek, 2005). Hence the knowledge available to these two different managers is
inherently different and culturally bound if government bodies stay on this isolated route (Drew,
28
2005). For as long as governments continue to view traditional folklore as an obstacle to resource
management rather than a tool, state management systems will most likely fail (Dove, 1988).
However, where local knowledge and understanding of an ecological system has been depleted, the
scientific theoretical basis of state-management may be the only option. As the predictability of a
system decreases, with loss of local knowledge systems, uncertainty as to the sustainable levels of
resource withdrawal increases and, thus, management tends more towards theoretical insurance
functions to prevent environmental degradation. Hence, management capacity inevitably shifts from
the realms of local control to state control as LEK is lost (Rudd et al., 2003).
1.3.1.3. Co-management
Recent decades have seen a trend in widespread protected area designation under state-management,
many employing exclusion theory. This entails excluding all local resource users from future access
to local ecosystems and undermining their ability to exploit at sustainable levels as they have been
doing for generations. Thus local costs of such designations have been considerable, whilst
conservation benefits have yet to be seen (Stocks et al., 2006). Therefore, co-management of
resource systems is likely to become a more prevalent option in the future where self-management is
unviable (Smith et al., 2006).
Local input remains essential to co-management to fill in state gaps in knowledge, to determine
realistic, culturally-viable objectives and to build stakeholder ties, whilst the government provides
supporting policy to legally enforce this relationship (Gilchrist et al., 2005). Successful comanagement takes into account social situations and economic constraints of the region (Hooper et
al., 2005). Upon recognizing their autonomy, authority and in-depth environmental understanding,
legal force is applied to reinforce pre-existing community practices in co-management systems,
adopting their self-enforcing, adaptive and sustainable techniques (Feit, 1988; Gilchrist et al., 2005;
Stocks et al., 2006).
Traditional communities worldwide that have established a delicate balance between resourcereliance and ecosystem maintenance, conserving ecosystem goods and services for use by humans,
have done so on the whole by managing the local environment in a mosaic of different successional
stages throughout history (Gadgil et al, 1993). By enhancing biodiversity through these ecocultural
29
landscapes, the network of intra-ecosystem connections increases, maximising ecological resilience.

Hence, if one species and its subsequent function are lost, it is more likely another species will exist
to replace that function in a complex, biodiverse ecosystem (Capra, 2000). LEK-based land
management techniques include rotational land and resource use, sacred groves, the protection of
keystone species or restraints on their collection periods, employing local experts as stewards to
monitor and advise on resource use and using fire to create patches suitable for prey species (Gadgil
et al., 1993; Khare, 1998; Berkes et al., 2000). All stem from an in-depth understanding of local
species and their complex interactions.
1.3.2. Food security and genetic diversity

We currently face a global extinction crisis, with 25 % of the worlds plant and animal species
expected to become extinct by 2050 (McNeely & Scherr, 2001). With biodiversity levels estimated
to be between 5 and 15 million species worldwide, most of which remain undiscovered (only 1.5
million species currently known), and current species extinction rates exceeding those previously
recorded by approximately 100 to 1000 times, species are currently being lost at a faster rate than
we can identify them (Pretty, 2002; King, 2003).
Current extinction rates make the conservation of LEK even more pressing. The more knowledge
humanity can conserve, the more likely we are to find sustainable solutions to global problems
where other solutions fail (King, 2003). The wild relatives of our staple foods for instance, and the
knowledge held on them, may provide vital crop alternatives and food security for deprived regions
in adverse farming conditions in the future (Anishetty, 2004). The wild plants found today in
traditionally-managed systems offer hidden harvest opportunities to remote, isolated communities
that hold in-depth knowledge about them. This information is particularly important in times when
environmental conditions are severe and food availability is limited, in periods of drought for
instance. LEK and its associated management practices are also responsible for the majority of
domesticated plants and animals worldwide and, thus, a vast portion of global genetic diversity
(Mauro & Hardison, 2000).
Where urbanisation is impinging upon previously marginalised lands, such as in the Andean
highlands, LEK loss is correlated with a decline in local land races and breeds (Young, 2002). It is
30
estimated that of the 6500 domestic animal breeds that exist globally today, most of which originate
from traditional communities, a third of them are threatened with extinction (Pretty, 2002).
Consequently genetic diversity is declining, threatening the wild relatives of todays staple crops
and global food security.
It is estimated that the crop varieties bred and developed by traditional farmers are worth US$15
billion to the global seed industry today, and if protected, could play a significant role in developing
sustainable food productions systems in the future (Tansey, 2004). When a fungal blight eliminated
the musk melon crop of California, germplasm from resistant varieties was imported from India
saving an industry worth millions (Sahai, 2004). This potential may be depleted in the future with
current extinction rates of both species and knowledge.
1.3.3. Pharmaceutical industry and economic generation

Another industry into which knowledge of local species uses is a major contributor is the
pharmaceutical industry. The current value of the world market in traditional medicines, including
herbal products and raw materials, is estimated at US$60 billion (Zhang, 2004). However this is
likely to constitute only a fraction of the overall potential, since only a small portion of tropical
rainforest species have been assessed for their medical value to date. Invaluable species historically
identified as a result of the ethnobotanical practices of indigenous peoples include quinine, curare
and ginseng (Ruiz Muller, 2004).
Interest in medicinal uses of plants was first documented in a publication by Rumphius in 1775
entitled Herbaria Ambionensis (Kamil, 2004). India, in particular, is a nation heavily reliant upon
traditional medical practices, with over 95% of its medicinal resources collected from the wild
(Shankar, 1998). The Vietnamese alone use over 3000 different plant species to meet their primary
healthcare needs (Le Quy, 2004). Even in the UK, the Department of Health announced in 1995 that
25% of all NHS general practice referrals are to the sector of complimentary medicine (Posey,
1999).
It is estimated that one-third of the global population lack access to modern healthcare facilities. In
India alone, traditional medicine constitutes an indispensable source of primary healthcare for 70%
31
of the population and in Africa, 80% of the population of member states rely upon traditional
healthcare treatments (Zhang, 2004). The success and necessity of traditional medicine has been
truly demonstrated in Pangani district of Tanzania where traditional healers are able to treat over
2000 opportunistic diseases that take hold of HIV and AIDS patients, enabling some victims to
survive up to five years longer (Gorjestani, 2004).
Therefore, traditional systems of healthcare have been of immense importance to communities, both
on large and small-scale. Many rural poor communities worldwide have developed their own
medicinal uses for local species on which they depend. For instance, in the Dolkha district of Nepal,
where modern healthcare is absent and poverty persists, locals rely upon 58 local species to provide
113 medicinal remedies to heal a wide variety of ailments (Shrestha & Dhillion, 2003). This
situation is symptomatic of many poor communities in developing nations.
At present, only 150 of the 30,000 traditional medicinal species are marketed worldwide (Kothari et
al., 1998). The tribal communities of India alone are believed to currently exploit over 7500 local
species for medicinal purposes (Sahai, 2004). In 1999, retail of alternative remedies reached $20
billion with only 1% of this profit deriving from African and Middle Eastern sales, reflecting growth
in the European market and subsequent market export expansion from developing nations (Mhame,
2004). Therefore, the global implications of expanding this market would not only benefit human
health worldwide, but would also improve economic security for rural and indigenous communities
in developing regions globally, providing that property rights are protected and benefits are
equitably distributed locally.
1.3.4. The promise of future findings

If so many ecosystem goods and services are currently being exploited to the benefit of human
populations worldwide, who is to say what potential opportunities will be uncovered in the future?
LEK offers the opportunity to save both time and money on research into these opportunistic uses
(Ruddle, 2000). However, in order for these intrinsic values to be discovered in the future, drastic
measures are required for the conservation of both species and knowledge diversity. In conserving
these pools of capital, we buffer ourselves against large-scale disasters in the future (Kaushik, 2004).
32
A 1.7 billion population increase is expected by 2020, with 80% of the increased food demand
anticipated to fall in developing regions. Through sustainable food production, medical care and
income generation via the manufacture of traditional products, LEK may aid the alleviation of
poverty and food insecurity in the future (McNeely & Scherr, 2001; Hamwey, 2004). Hence, LEK
has become a popular policy target in the international arena in recent years, including in Article 8
(j) of the Convention on Biological Diversity. This article emphasises the importance of indigenous
resources, including knowledge and innovations, and states that any exploitation of these resources
should result in the equitable sharing of benefits amongst the original holders. The agreement has
been signed by 180 countries and parties to date (Latiff & Zakri, 2004).
Arguments to conserve LEK often focus greatly on the material benefits it offers in terms of drug
discovery, raw materials for biotechnology and agricultural innovations (Mauro & Hardison, 2000).
However, in addition to enhancing knowledge and resource diversity, LEK also maximises global
cultural diversity, contributing to the formulation of community worldviews, social institutions,
customary practices, spiritual beliefs, differentiated-societal roles and overall identity (Singh et al.,
2006a).
1.3.5. Cultural capital

LEK represents the most fundamental component of cultural adaptation to the natural environment
(Oviedo et al., 2004). As a knowledge base it provides an element of social capital and power to
the otherwise powerless by giving them the tools to self-manage their own systems sustainably
(Ruddle, 2000; Davis & Wagner, 2003; Gorjestani, 2004; Singh et al., 2006a). Thus it acts as a
form of cultural capital that can enhance ethnic pride and is passed on through generations via
traditional stories, songs, rituals and practices that connect traditional communities with their
homelands and, subsequently, to their ancestors that previously lived off it (Gadgil et al., 1993;
Barrett, 1995; Pfeiffer, 2002).
It has also been suggested that LEK creates within its holders a unique affinity for their local
environment, enabling individuals to bond with and subsequently care about their non-human
neighbours (Nabhan & Trimble, 1994). In traditional societies, this affinity is forged from a young
age in childhood. For instance, the children of nomadic sea-faring Bajo tribes of Indonesia were
33
traditionally paddling their own boats by four years old and fishing by seven (Adi and Sudarman,
1994). Pyle raises concerns as to whether the aesthetics of the virtual can ever truly replace
experience of the actual in industrialised regions. He asks, Can images on the web ever replace the
creative mastery behind a spiders web? (2001, p.21). If they cannot, then LEK and the affinity it
generates with nature is likely to be lost in future generations, as is ecosystem diversity, as local
species are cared about less and less (Matsumoto, 2003).
Thus LEK as a knowledge base offers solutions to local, national and global environmental
challenges by providing information on the use of locally available resources (King, 2003; Kaushik,
2004; Hassan et al., 2005). Locally, ecoliteracy enables natural resources to be used for income and
food. The closeness with nature derived from ecoliteracy also meets the social, cultural and spiritual
needs of many traditional societies (Gadgil et al., 1993; Barrett, 1995; Pyle, 2001; Pfeiffer, 2002;
Yupari et al., 2004). Ecoliteracy also provides resource managers with information about sustainable
practices, levels of resource withdrawal and monitoring information (Wright, 1985; Gadgil et al.,
1993; IIED, 1994; Berkes et al., 2000; Ruddle, 2000; Turner et al., 2000; Long et al., 2003; Folke,
2004). In addition, ecoliteracy has the capacity to enhance support for local management efforts
(Lundquist & Granek, 2005), since people are unlikely to care for that which they do not know
(Pyle, 2001, 2002, 2003; Matsumoto, 2003).
Nationally, ecosystem goods that hold commercial value have the opportunity to boost the national
economy (Posey, 1999; Mhame, 2004) as well as improve regional food security (McNeely &
Scherr, 2001; Hamwey, 2004). Internationally, ecoliteracy has provided information on a huge
number of medicinal species for the pharmaceutical industry to date (Ghimire & Pimbert, 1997;
Shreshta & Dhillion, 2003; Gorjestani, 2004; Le Quy, 2004; Ruiz Muller, 2004; Zhang, 2004), and
the promise of future discoveries holds high hopes for the cure of some major diseases (Ruddle,
2000; UNDP, 2004).
34
1.4. Factors leading to LEK loss
1.4.1. Urban migration and urbanisation

As already mentioned, unlike other knowledge bases LEK is not static, instead it has the capacity to
increase, decrease and remain constant (Pacn, 2004; Pilgrim et al., 2006a). In recent years, the
documented loss of LEK from developing regions worldwide has been attributed to a number of
large-scale phenomena including urban migration. This is the process whereby people migrate from
resource-rich rural locations in pursuit of urban jobs and cosmopolitan lifestyles promised by the
city. This is especially prevalent in younger generations who perceive little or no value to inheriting
their ancestors LEK and, instead, are drawn to cultures of commercialism, consumerism and quickturnover labour-selling urban jobs portrayed by the media. The demands of modern states for cash
via taxes, school fees and healthcare accelerate this process (Barrett, 1995; Brodt, 2002; Kaihura,
2004).
Previous studies have shown that LEK is commonly lost with increased access to modern facilities,
such as healthcare and commerce. In the Ozark Mountains of Missouri for example, knowledge of
folk medicine is sustained only in those communities that are remote enough to remain isolated from
urban centres (Nolan & Robbins, 1999). In these communities, traditional practices and local
resources are relied upon on a daily basis, ensuring the transfer of LEK to younger generations.
Rapid and extensive environmental change as a result of urbanisation throughout the latter half of
the last century has depleted many areas of green space, including many habitats that children of
previous generations once played in and subsequently bonded with. Backwoods have been cleared
for property, corner stores for malls and roadways for bicycle tracks (Johnson, 2000), robbing
children of their natural outdoor playgrounds. Consequently, for each generation a new starting
point is formed from which they see environmental degradation occur throughout their lifetimes,
each more modernised and urbanised than the last. Thus, the full expanse of environmental
depletion and damage will never be fully appreciated by a single generation of observers. This
phenomenon is termed Environmental Generational Amnesia by Kahn (2002).
35
Throughout history, human populations have proven to be highly adaptable, adjusting continually to
their new physical environments and ecological conditions. One way of doing this is by replacing
knowledge of the old environment with an in-depth knowledge and understanding of the new
environment. This is causing ecological knowledge in industrialised societies, including plantlore, to
be replaced with knowledge more applicable to the new environment of economy and industry for
example (Ruddle, 2000). This is particularly prevalent where there has been a sudden severe
environmental change resulting in expansive species loss. Where this happens, plants and animals
that were once present and are no longer in the vicinity are no longer talked about (Barrett, 1995).
1.4.2. Modernisation of lifestyles and technology

Thus urban and suburban development is starving children and future generations of their right to
experience and interact with nature. Pyle (2002) perceives direct experience as key to bonding with
the local environment and creating future generations of naturalists. Lifestyle changes, due to
economic development and technological advances, are also responsible for reduced time spent in
direct contact with nature. Changes include the replacement of oral traditions, such as story-telling
by elders, with television watching, surfing the net and computer games, as the rural poor get
assimilated into a more dominant, technologically-minded society (Nabhan & Trimble, 1994;
Johnson, 2000; Turner et al., 2000). Other factors, such as safety concerns and over-scheduling, are
drawing children in particular in from the outdoors (Pyle, 2001, 2003). The average time a child
spends out of doors has greatly declined from 86 minutes per day to 42 minutes from 1981 to 1997
(Orr, 2002), although the direct impact of this decline on LEK has yet to be quantified.
Reflective of lifestyle changes are losses in family traditions. These include the nuclear family
replacing the extended family, reduced social networks, increased mobility and vehicular
transportation, and a decreased sense of place, all of which act to limit a childs spontaneous
encounters with nature and minimise time spent in the company of community elders (Kellert,
2002). Such lifestyle transformations greatly threaten LEK by jeopardising its traditional transfer
routes; word-of-mouth and direct experience (Nabhan & Trimble, 1994; Pyle, 2001; Orr, 2002).
36
1.4.3. Extinction of experience

Whilst young peoples opportunities to directly experience nature are declining, their opportunities
to experience nature indirectly or virtually are rapidly expanding through an increased number of
zoos, museums and botanical centres, TV documentaries, films and internet images (Kellert, 2002).
The growing availability of indirect experience in industrialised societies today, replacing direct
experience, is thought to be causing a cycle of human and ecosystem degradation Pyle (2002, 2003)
terms the Extinction of Experience. This is a hypothetical phenomenon by which the less time a
person spends in the natural environment, the less they know and recognise. Consequently, they act
less to protect those species with which they are unfamiliar. In turn, more species extinctions occur
unabated and less is the desire to experience the outdoors, leading to a hopeless downward spiral of
disaffection and degradation leading to disconnection.
Historically, children would participate in subsistence activities with community elders as soon as
physically able (Adi & Sudarman, 1994; Tansey, 2004). Independence from the land, urban
influence and the spread of formal education has resulted in raised expectations of younger
generations. This is illustrated in Rao et al.s study (2003) by the willingness of young people to
leave their homelands for little compensation and the chance of an urban life. Elders on the other
hand are more strongly tied to their roots and the homelands of their ancestors, staying resident on
them at all cost (Ghimire & Pimbert, 1997).
1.4.4. The introduction of schooling and curriculum changes

The introduction of formal education to developing regions has altered traditional lifestyles in two
ways; firstly, by removing children from their family for long periods during the day and, secondly,
by altering the content of their childhood teachings from that of their parents. Formally-educated
children, on the whole, no longer know or value the names and uses of their neighbouring species
learned through narratives of elders, instead, Western curricula teaches the big picture of ecology,
defining ecological functions, ecosystems, roles, habitats and interactions (Pyle, 2001; SkutnabbKangas, 2004). Consequently, many children can no longer identify the non-human neighbours on
their doorstep but are able to define the words ecology and ecosystem with little idea of their
own place in them.
37
In the past, natural history education involved more interaction with the environment, but in recent
years laboratories and hard statistics have replaced butterfly nets and botany walks (Pyle, 2001). The
passive learning techniques of formal education are gradually replacing traditional interactive
methods of teaching in developing communities globally, which consist of watching, listening and
doing. Indigenous children are gradually being taught to recount a list of facts instead of taking part
in story-telling and traditional practices. A community elder once accused a young man of writing
her stories down in order to forget them, implying we have no need to remember things once we
learn how to record them (Peat, 1996).
Western education puts heavy emphasis on scientific fact and textbook learning and, as a result, has
demeaned the importance of LEK and parental learning in many regions of the world (Nabhan &
Trimble, 1994; Kothari et al., 1998). Many Western scientists disregard and devalue LEK as
superstitious stories based on fiction rather than fact, particularly since it often goes unrecorded.
Consequently, generations of indigenous children are now being educated outside of their own
culture and, as a result of media influence through television and the internet, are gradually being
integrated more and more into other cultures. Where this is happening, children no longer value the
experiences and knowledge of their elder relatives and are yearning for the culture of the West (Peat,
1996). Thus the primary knowledge sources of environmental learning have shifted from traditional
family networks and direct interaction with the environment, to 84% of school children now gaining
their environmental knowledge from the media and formal education (Nabhan & Trimble, 1994).
1.4.5 Economic development and marketisation

On a larger inter-cultural scale, variation in knowledge level has been found to exist as a result of
differences in scale of economic development and, thus, local resource dependence (Pilgrim et al.,
2006b). For instance, once access to a modern health clinic became available to one Ese Eja village
based in the Amazon, medicinal resource dependence declined as did local knowledge of natural
pharmaceuticals (Alexiades, 1999). In India, landless agricultural labourers that were highly
resource dependent were found to hold more knowledge of their local environments than
landowners, herders and traders (Gadgil et al., 2000). Similarly, on Nicaraguas Atlantic coast, the
rural poor were far more aware of local medicinal plant species than their urban cousins (Barrett,
1995). Thus knowledge differentials exist within and between communities, reflecting cultural
38
norms, societal roles and changes in traditional resource dependence as a consequence of local
development.
This combination of urbanisation, loss of traditional livelihoods and formal schooling has led to
traditional communities, previously isolated and self-sustaining, having increased trade access and
access to modern facilities, such as health clinics. Modern healthcare, in particular, is creating a
deviation from traditional healing using inexpensive widely-available local species, to expensive
Western medicines and hospital stays. On the one hand, world trade has created market demand for
locally-derived resources and traditional products increasing the economy of deprived traditional
communities (Hamwey, 2004). On the other, with the globalisation of markets has come widespread
loss of locally resource-dependent, self-sustaining communities and the knowledge that they hold
(Gadgil et al., 2000).
Markets, monetarisation and commodification, therefore, impact peoples relationships with their
resources (Pilgrim et al., 2006c; Stocks et al., 2006). Traditional products are being replaced by
external goods (Byg & Balslev, 2001) and, consequently, local resource reliance is declining and
traditional knowledge is being replaced by knowledge of trade and commerce (Pilgrim et al.,
2006c). For instance, convenience goods are replacing local wild foods, and supermarket shopping
is replacing collection and preparation. No longer are the general public their own producers,
processors and preservers of food, many even lack the knowledge to recombine the ingredients of
convenience foods for nutritious meals (Anishetty, 2004; Tansey, 2004). As a result, the world has
suffered widespread decline in local resource use and control in exchange for increased corporate
interests (Gadgil et al., 2000).
1.4.6. Assimilation and integration

Departure from traditional lifestyles, and the knowledge systems underlying them, is seen by many
authorities as beneficial to native tribal groups beginning them on the path towards development.
However, the reality is that isolation from the land and loss of traditional practices can be the demise
of a once self-reliant community. For example, the Innu of Labrador were forced off their lands and
sedentarised in the late 1950s after 8000 years as nomadic hunters. Subsequent loss of livelihoods
resulted in loss of self-esteem, culture and identity in a society to which they do not belong. This has
39
had both direct and indirect effects on the mental and physical health of the Innu (Samson and
Pretty, 2005). Indirect physical consequences include loss of species that they once relied upon for
healthcare and food and deviation from traditional hunting techniques that required extreme physical
activity. Direct consequences, as a result of community demise, include heavy drinking, suicide,
solvent abuse and sexual abuse (Samson, 2003).
1.4.7. Language loss

One final factor contributing to LEK loss in indigenous and tribal communities is widespread
language loss. Languages have been described as the DNA of cultures because they encode and
transfer all of the inherited knowledge of a society. Thus when a language is lost, like a DNA
sequence, the unit of knowledge and human thought it represents is lost forever. With this comes a
loss of oral traditions and narratives that pass on customs, worldviews and survival skills, including
LEK, that are central to a culture, hence a diminished sense of identity often follows (King, 2003;
Samson, 2003; Skutnabb-Kangas, 2004).
It is anticipated that 50-90 % of the worlds 6800 languages will come under severe threat within the
next century. Consequently, the knowledge bases held within those languages will become
threatened. As many as 95% of the worlds languages had less than one million native users in 1999,
and 500 were spoken by less than 100 people. Like biodiversity, the worlds linguistic hotspots are
centred on developing regions in indigenous and minority cultures that are under the most pressure
to assimilate into dominant society and global markets (Maffi, 2002; King, 2003). If assimilation
occurs in the future and causes the widespread extinction of indigenous languages, then generations
of localised knowledge will most likely be lost with far reaching implications both locally and
globally.
In 1993, Kellert and Wilson suggested that if LEK continues to be diminished with each passing
generation, then future populations may not simply be disconnected with nature, but may suffer a
fear of the natural environment, termed Biophobia. Biophilia is the hypothesis that humans have a
genetic predisposition towards nature as a result of our ancestors dependence upon their local
environments. The biophilia hypothesis suggests that humans have consequently evolved to prefer
natural environments to built ones as a result of historical connections (Kellert & Wilson, 1993).
40
With the loss of LEK causing widespread disconnection from nature, the concern is that a form of
biophobia may override this genetic predisposition and that humans may evolve a fear of nature as it
becomes increasingly unfamiliar and unknown.
Thus despite natural history and environmental books being vast in number, an abundance of field
centres, increasing membership of environmental societies and ecotourism rapidly expanding for
both day trippers and holiday-makers, LEK is under threat and declining worldwide. Even in
England LEK was relied upon in the not so distant past. For instance, horsemen of the last century
could treat equine ailments using only their own in-depth knowledge of local plant species. This
included using deadly nightshade to cure colds and apples for fevers (Pretty, 2002). LEK is therefore
being lost via two routes as a result of widespread processes of modernisation, urbanisation and
commercialisation. These include lack of transfer between generations and hybridisation or
replacement with extra-local knowledge. More often than not these two scenarios mutually-reinforce
one another advancing the rate of LEK decline (Ruddle, 2000).
The recent Millennium Ecosystem Assessment (Hassan et al., 2005) has shown the considerable
economic value of the worlds marine and terrestrial ecosystem services. However, virtually all
large scale damage to the global environment is caused by economic activities (Nadeau, 2003). This
is ironic as in the long-term environmental degradation will result in severe economic losses both
locally and globally, and in many cases the economic value of natural resources is still largely
unrealised. Conserving biodiversity worldwide has been estimated to cost US$300 billion (Myers,
2002). However, some time and money could be spared if the experiences of traditional societies
were used to our advantage today (Veitayaki, 1997). Therefore, the value of LEK held by resourcereliant communities and the management solutions offered by retaining this knowledge base and its
corresponding practises are extensive.
1.5. General population distribution of LEK

A notable characteristic of LEK, known for a long period, is its intra-cultural heterogeneity
(Rappaport, 1979; Garro, 1986; Sillitoe, 1998). However, this often goes unconsidered in
ethnobotanical studies, with researchers either making the assumption that knowledge is evenly
41
distributed (Alexiades, 1999) or studies have been too small-scale to accurately determine levels of
variation (Tsuji, 1996).
1.5.1. Gender differences

A well-studied area of ethno-botanical distribution is gender differentiation. However,
generalisations in this field are often made based on results from geographically-localised research.
The majority of research in this area shows that although transferred vertically through family lines,
LEK can often be gender-isolated (Singh et al., 2006b). For instance, culinary traditions are
primarily passed between female relations in rural areas of France, Italy, Portugal and Spain today
(Howard-Borjas, 2002; Howard, 2004).
One explanation for this is barriers in knowledge transfer. Gender-differentiated LEK transfer was
demonstrated in a study by Browner (1991). In this study, 73% of women claimed to learn their
knowledge of traditional medicines from female relatives compared to 71% of men who acquired it
from their fathers or grandfathers. Similar gender-differentiated transfer in medicinal knowledge
was observed in communities in the Ozark Mountain region of Arkansas and Missouri. Here granny
women transferred plant medicinal knowledge to their daughters and granddaughters. Yarb doctors,
on the other hand, whose healing powers were dependent on supernatural beliefs, communicated
their knowledge to their sons and grandsons only (Nolan & Robbins, 1999).
Gendered knowledge is thus defined as that held by women or by men, but not by both (Howard,
2004). It has frequently been attributed to traditional labour divisions and cultural roles and
responsibilities that exist within traditional communities (Garro, 1986; Alexiades, 1999; Turner et
al., 2000; Pfeiffer, 2002). For example, amongst the Ese Eja of the Amazon it is mens work to hunt,
fish, gather brazil nuts, gather timber, and clear swiddens, whereas womens responsibilities include
childcare, planting, harvesting and gathering all other natural products required in daily life
(Alexiades, 1999).
Whether men or women constitute the primary LEK holders within a community has been widely
discussed. A number of studies have found men to be the most knowledgeable, whether identifying
mammals in Sumatra, Indonesia, edible fruit trees in Burkina Faso or holding agro-ecological
42
knowledge in Jharkland, India (Jewitt, 2000; Kristensen & Balslev, 2003; Nyhus et al., 2003). This
has been most frequently attributed to the extensive freedom men possess and the larger area
covered in their resource collection activities compared with women. Amongst the Tado people of
Flores Island, Indonesia, for instance, women were found to know more species closer to home,
whereas mens hunting activities meant that they knew more species further a field (Pfeiffer, 2002).
Nabhan & Trimble (1994) suggest that, in general, boys are the ones to roam free and explore new
landscapes throughout childhood, whereas girls are encouraged to stay close to home. This pattern
continues in adulthood amongst many communities in rural India for example. Here, men are able to
interact with farmers outside of their own communities by participating in a diversity of farming
activities, whereas women have restricted roles as manual labourers (Jewitt, 2000). However, this
view devalues womens role in maintaining local ecosystem health and their in-depth knowledge of
species interactions and uses in the local area (Singh et al., 2006b).
Womens knowledge of traditional medicine has often been portrayed as inferior, often being
termed old wives tales. However, domestic womens healing techniques treat between 70% and
90% of illness episodes encountered everyday worldwide. Despite this, the title of Shaman or
suchlike, symbolic of social status, is reserved for men (Howard-Borjas, 2002; Howard, 2004). A
Shaman is considered to be the primary healer within most traditional communities, able to treat a
number of serious ailments. This is the case even in regions like India where womens medical
knowledge has developed over 40 centuries and is able to cure most common ailments and
childhood diseases without external consultation (Warren, 1997). Consequently, a number of studies
have shown women to be the key knowledge holders (Mandany & Bugahoos, 1998; Alexiades,
1999; Singh et al., 2006b).
King (2003) describes women as central to maintaining seed variation and, therefore, to agri- and
genetic diversity. She views them as the primary custodians of a crop-focused knowledge base that
has evolved over 4000-5000 years. This connectedness has been attributed to womens community
roles that invest more time in the local environment as a whole (Warren, 1997). In Central American
Hispanic communities for instance, womens diverse array of localised responsibilities includes
collecting plants for medicines, food, fibres, utensils, cosmetics and ornamentals (Howard, 2004).
43
It has been suggested that mens knowledge focuses upon maximising economic gain from local
species and womens on the storage and preparation of local species to meet daily nutritional
requirements. Consequently, women are thought to be responsible for growing as much as 59-80%
of the worlds food (Warren, 1997). Womens close social networks embedded in local tradition and
lifestyle maximise their capacity for knowledge transfer (Howard-Borjas, 2002; Howard, 2004).
Social constructionist feminists perceive this link between women and their local environments as
having a historical and contextual basis. Thus the connection is thought to be socially-constructed as
a result of gendered roles throughout history and culture is thought to be responsible for nurturing
these roles (Plumwood, 1993; Mellor, 1997). Spiritual feminists, on the other hand, support the
feminine principle by which women are the primary knowers as a consequence of their innate
affinity towards nature. This elemental connection is thought to be the result of female biology
being controlled by nature and assumes that women, therefore, have an innate understanding of their
ecosystems (Shiva, 1997). This idealistic view eliminates the possibility that women could ever
exhibit signs of violence towards nature and generalises that women have no part in ecological
degradation, even in industrialised regions where populations have long been independent of their
local resources.
1.5.2. Generational differences

Widespread knowledge demise with altering landscapes and lifestyles is thought to be causing
progressive knowledge loss amongst the younger generations of many communities (Tsuji, 1996;
Turner et al., 2000; Douglas, 2002; Sargent & Browner, 2004; Birchenough, 2005). Community
elders have often described young people today as far less knowledgeable and interested in their
local environments than they were as children (Nabhan & Trimble, 1994; Barrett, 1995; Crate,
2002). In Madagascar for example, the young people of traditional communities whose ancestors
heavily relied on locally-made palm products for survival, take no interest in learning the skills
needed to manufacture these products today, now that alternative imports are available (Byg &
Balslev, 2001). Also, in Tado communities, Indonesia, young women are unable to maintain the
culinary traditions and recall the recipes of their predecessors (Pfeiffer, 2002).
44
One of the first attempts to quantify generational knowledge decline was made across India as part
of the Peoples Biodiversity Registers Programme, an attempt to document isolated pockets of
Indian LEK from different regions. Data from Kaihad village, Himachal Pradesh, revealed drastic
intra-community knowledge loss in younger generations, with adults over 50 years able to identify
70% of local flowering plants and knowing uses for only 5%, adults between the ages of 30 and 50
able to identify 25% of local flowering plants but naming uses for only 1%, and young people
unable to identify any of their local plant species (Gadgil et al., 2000).
There are a growing number of studies into ecological literacy or knowledge within countries. Very
few, though, have been cross-cultural, and most have lacked the necessary quantitative data across
whole communities to describe the phenomenon of loss. Previous research has generally targeted
isolated areas and qualitatively documented local knowledge (Ghimire et al., 2005). Many studies
have focused solely on the identification of experts and few have taken the widely-utilised
knowledge and practices of laypersons into account, despite their contribution to local resource
management (Davis & Wagner, 2003). Limited sample sizes are another shortcoming of earlier
research (Alexiades, 1999; Martin, 2004). In essence, previous studies have been qualitative, smallscale, culturally-specific and geographically-localised (Sillitoe, 1998; Fazey et al., 2006). This study
is unique in that it aims to combat this paucity in previous research by being quantitative, crosscultural and geographically-widespread.
1.6. Aims of thesis

The importance of LEK as a knowledge base has long been understood as a critical resource for
indigenous peoples and the rural poor. Despite this, there have been no in-depth studies attempting
to quantify and analyse the phenomenon of loss; which sectors of the community are most affected,
what modern society attributes are affecting local knowledge bases and how rapidly this loss is
occurring within and between communities. Furthermore, key sources of knowledge transmission
essential to the preservation of LEK amongst different communities have yet to be determined
through quantitative methods.
45
LEK distribution in relation to development has been widely discussed in theory but, to date a hole
in the research exists looking at intra- and inter-community knowledge levels in different regions of
the world in relation to their economic status and human development levels. Also, a number of
studies have looked at factors affecting LEK in isolation such as age or gender, but none to date
have carried out an overall assessment evaluating all of the impacting factors and their importance.
Therefore, at present there is a lack of knowledge as to the future of LEK bases in communities
worldwide and the direct factors most attributable to its current demise.
In summary, loss of LEK is of global importance threatening the long-term security of ecological
systems and human societies alike. Many attempts have been made to qualitatively document LEK
for future safe-keeping. However, few studies have tried to locate the holes that LEK is leaking from
using quantitative methods and find solutions to prevent future leakage.
Indigenous people living in and off a particular ecosystem for generations develop an intimate
relationship with it and, consequently, encode its every characteristic linguistically. As well as
naming all of its components, they formulate worldviews based on distinguishing between proper
and improper thoughts about the environment (Krech, 2005). Four levels of ecological knowledge
have been recognised (Berkes, 1999): i) the names of biotic (e.g. plants, animals) and abiotic (e.g.
soils, water, weather) components of ecosystems; ii) the functions and uses of each component; iii)
the land and resource management systems and the social institutions that govern them; and iv) the
worldviews and cosmologies that guide the ethics of people in the system (Fig 1.5.1).
46
Worldviews and
cosmologies
Management
systems
Uses and functions
Names
Figure 1.5.1. Levels of ecological knowledge and understanding

This research focused on quantifying the first two levels of ecological literacy in communities (the
names of biotic components of ecosystems and the functions and uses of each component) in the
three countries of India, Indonesia and the UK.
This thesis aimed to determine the following
1. A standardised protocol for ecological knowledge assessment within and between
communities across different cultures.
2. The effect of economic development on knowledge distribution both within and between
communities.
According the literature, economic development leading to modernisation of lifestyles should act
to damage local knowledge bases (Alexiades, 1999; Gadgil et al., 2000; Stocks et al., 2006), a
key hypothesis of this study. This study aimed to prove this decline quantitatively and see if the
rate of decline is the same across all sites.
47
3. Key factors that lead to high and low levels of ecological knowledge based on quantitative
methods and if these differ between localities.
Modern education is a key factor in the literature believed to be reducing LEK levels globally
(Nabhan & Trimble, 1994; Kothari et al., 1998; Pyle, 2001; Skutnabb-Kangas, 2004). Therefore,
a hypothesis of this study was that local knowledge becomes threatened, particularly in younger
generations, where modern education systems exist. This thesis also aimed to detect any other
factors responsible for LEK decline as well as factors that increase local knowledge levels.
4. Whether progressive knowledge loss is occurring within the chosen communities and the
reasons for its existence or absence.
Knowledge loss with age within communities has been described time and time again in the
literature (Nabhan & Trimble, 1994; Barrett, 1995; Gadgil et al., 2000; Byg & Balslev, 2001;
Crate, 2002; Pfeiffer, 2002). Therefore, a hypothesis of this study was that knowledge levels of
elder generations will exceed that of younger generations where local resource dependence no
longer exists.
5. The patterns of knowledge acquisition with age at all sites and to test the hypothetical
acquisition curves shown in Fig 1.5.2.
48
A
Knowledge level
B
C
Age
Figure 1.5.2. Theoretical acquisition curves. Curve A represents a highly resource dependent society
with rapid acquisition and saturation. Curve B represents a less resource dependent society with less
rapid acquisition. Here knowledge saturation occurs later in life and at a lower level than community
A. Both curve A and curve B show a decline in knowledge after the plateau, indicative of memory
loss with age. Curve C represents an industrialised community with no resource dependence where
knowledge level never reaches that of community elders and knowledge levels are on the whole
low.
6. Differences in familiarity with different species groupings and reasons for such differences.
49
Chapter 2. Methods
2.1. Introduction
LEK loss has been brought to light globally and put onto the worldwide stage through policies such
as the Convention on Biological Diversity. Its decline has been attributed to a variety of large-scale
processes including modernisation (Peat, 1996), urbanisation (Nabhan & Trimble, 1994), urban
migration and sedentarisation (Rao et al., 2003; Samson, 2003), biophobia-promoting disconnection
(Kellert & Wilson, 1993) and extinction of experience (Pyle, 2003). Despite such well-defined
theory, we still lack the empirical data to describe the complex phenomenon that is LEK loss both
within and between different cultural communities, locally and globally. It is only through detailed
case-study comparisons that we will be able to fully understand this process in relation to
development and its impacting factors (Zent, 2001). Thus the direct forces acting upon LEK, their
origin (be it local, regional or further afield) and mode of interaction are little understood, despite
widespread theory (Tansey, 2004).
Past LEK studies have tended to focus on compiling a comprehensive inventory of indigenous
species names and uses within a biodiverse and characteristic ecosystem (Byg & Balslev, 2001).
Expert elders have been targeted and the longest compilations of species names are deemed the most
successful research projects. Despite awareness of intra-cultural variation, community knowledge
has been assumed to be uniform throughout the vast majority of LEK research (Howard, 2004).
Few studies take into account laypersons LEK and the dissemination of this knowledge within
society and, as a result, very few studies to date have focused on measuring the mean LEK levels of
communities (Gadgil et al., 2000). This information is particularly relevant where resources are
locally-managed, as it is not only community experts that use and manage species within these
systems. In particular, no work to date has looked at community knowledge variance (the gap
between the most and least knowledgeable).
Levels of knowledge variance may be key to the success of community management schemes if
increased variance in knowledge creates variance in local views of nature leading to a non-uniform
response to management locally. It seems reasonable to hypothesise that as economic development
50
increases and resource dependence decreases, mean knowledge level is likely to decrease and LEK
variance increase within a community (as laypersons become less reliant on local knowledge to
meet their daily needs). This may potentially lead to a loss of collective action by the community as
a whole. If this is the case, then mean knowledge levels and their variance would be more central to
predicting the management capacity of a community than the maximum knowledge levels measured
when focusing on experts. Hence, it is essential to develop a standard replicable methodology to use
in future ecological knowledge research that withstands rigorous public inspection whilst
representing the true breadth, depth and differentiation that exists (Davis & Wagner, 2003).
Developing a standardised protocol that assesses decline quantitatively and comparatively with
precedence would also allow the integration and interpretation of knowledge from a variety of
locations and cultures (Gilchrist et al., 2005). Without such uniform methods, results are difficult to
replicate and compare (Phillips et al., 1994). Future methodology, therefore, needs to be
accomplished in documenting knowledge variance and at identifying the relevant social factors that
cause differences (Zent, 2001). Thus a priority of this thesis is to develop a methodology that meets
all of the above criteria and successfully combines socially-extracted data with scientific analytical
techniques for potential policy formulation and conservation suggestions (Phillips et al., 1994).
Zent (2001) succeeded in developing a methodology that meets some of these requirements, in
particular, quantifying LEK loss. However, the method of marking off an area containing an array of
local species which respondents are expected to identify lacks large-scale viability. Firstly, it isolates
plant knowledge as the only component of LEK and, secondly, it eliminates extreme ecosystems
where biodiverse species-rich patches may not exist but still resources are vital, for instance in areas
of prevalent drought. Also, it requires travel to and from the marked site by respondents, reducing
the chances of attaining a high number of willing participants and, therefore, a large representative
sample.
Thus the field of LEK research could greatly benefit from establishing a standardised methodology
that quantifies LEK changes and provides an accurate indication of intra-community knowledge
distribution, without interviewing every single community member on the details of every local
species (Martin, 2004). The methodology used in this study was continually developed and refined
51
to attain a code of best practices in LEK research that is globally-applicable. This chapter aims to
provide an overview of the basic methodological approach used at all sites and the rationale behind
it, site-specific adaptations will be discussed accordingly in subsequent chapters. Methods described
here aimed at determining LEK level and content, knowledge variance, key transfer sources, and the
most and least knowledgeable groups within and between different cultures and regions. As
previously stated in Chapter 1, the focus of this LEK assessment will be upon names and functional
uses, rather than management practices, stories and worldviews.
2.2. Study sites

Study site selection aimed at targeting a wide range of communities, environments and pressures
permitting cross-comparisons to be made and generalisations on LEK distribution to be drawn. This
also addresses the assumption made by many researchers about LEK research that it is small-scale,
culturally-specific, and geographically localised (Sillitoe, 1998, p.233). LEK content is thought to
be determined by three key factors; the physical environment, the cultural setting and the
educational system. By selecting study sites from different regions geographically, culturally,
ecologically and economically, this methodology was aimed at determining if any impacting factors
exist at all sites, despite vast site differences, that are reducing LEK globally.
A set of selection criteria for the study sites were assembled. These were as follows
1. The selection of sites had to incorporate a range of environments, pressures, cultures and
levels of economic development.
2. Work in developing regions had to be carried out in collaboration with established local
organisations who have already built up reciprocal relations of trust in order to provide
access into sites and local acceptance. However, influence by local NGOs on knowledge
systems had to be minimal.
3. Sites had to be accessible and representative samples within them attainable.
4. Levels of resource reliance (in terms of meeting daily needs and livelihood dependency) and
resource protection had to be varied between sites.
5. Sites had to be segregated into villages or wards enabling samples to be taken from a variety
of communities for intra-site comparisons.
52
6. Sites had to contain natural resource bases locally that are common property and open-access
to all.
Based on these criteria, three regions for study were selected; an area of Tamil Nadu in southern
India, the Wakatobi Marine National Park southeast of Sulawesi, Indonesia, and three very
different localities in the UK (Table 2.2.1).
Table 2.2.1. Key characteristics of study site regions
Index characteristic
HDI score
HDI rank
GDP per capita (p pp US$)
HPI score
HPI rank
Population (millions)
Annual PGR
Life expectancy
Established NGO (nongovernmental organisation)
Degree of urbanisation
Biodiversity level
Resource reliance
Local resource protection
HDI
India
0.60
(0.66 Tamil Nadu
statea)
Medium
2892
31.3
58
1,070.8
1.4
63.3
SPEECH
Indonesia
0.70
(0.64 southeast
Sulawesib)
Medium
3361
17.8
41
217.4
1.1
66.8
Operation Wallacea
Low
High
High
None
Low
High
High
National Park
UK
0.94
High
27,147
14.8
15
59.3
0.3
78.4
N/A
High
Low
Low
Local, national and
international
designations
Human Development Index (2003)

HDI are figures calculated from life expectancy, literacy rate, gross enrolment and GDP, however, at
the remote study sites in India and Indonesia these figures are likely to be lower than the national
figures given here. a2001 regional figure from Tamil Nadu Human Development Report (2003).
b
2002 regional figure from Indonesia Human Development Report (2004).
GDP Gross Domestic Product (p pp US$) (National figures) (2003)
HPI
Human Poverty Index (UK figures calculated based on HPI-2 index used for high income OECD
countries) (National figures) (2003). The higher the HPI score and rank, the more deprived a country
(maximum rank 177).
PGR Population Growth Rate (National figures) (2003-2015 estimates)
Source: UNDP (2003)
53
UK
India
Indonesia
Figure 2.2.1. The 3 chosen study regions (Worldpress, 2006)
2.3. Ethnobotanical and ethnozoological interview development

Semi-structured ethnobotanical and ethnozoological interviews were the key tool used for measuring
LEK levels (species names and uses), assessing content and quantifying differences in distribution
within and between sites (Kala, 2005). Interviews were designed to differentiate between the most
and least knowledgeable community members and identify the key characteristics of the two groups
(Browner, 1991). Interviews consisted of closed questions which elicited fixed answers and yielded
quantitative results. This simplified data aggregation and analysis and eliminated ambiguous
answers, interviewer bias and extraneous contextual noise (Zent, 2001).
Quantitative methodology is useful in the study of LEK to provide comparative, measurable data
(Davis, 2003). During interviews, interviewees were shown a selection of photos of common local
species and knowledge was assessed on three levels
1. Species recognition (Do you recognise this species?)
2. Species identification (What is the name of this species?)
54
3. Species use knowledge (Can you use this species and, if so, what for?)
Since study sites included some industrialised non-resource dependent communities and focused on
laypersons knowledge, primary species names were accepted as correct, for instance vetch (primary
name) instead of common vetch (primary name with modifier). Where names given were unknown,
local experts were consulted to ensure that no alternative vernacular names were marked as
incorrect.
Ethnobotanical species flashcards have been used in various studies aimed at determining LEK
differences (Balmford et al., 2002; Nyhus et al., 2003). Despite this, flashcard identification is still
an incomplete measure of overall ecological knowledge since it only uses a sample of common local
species. However, it is a highly effective tool for quantifying LEK levels for comparison between
and within groups, the purpose of this study (Nyhus et al., 2003). Essentially, flashcard
identification acted to validate and measure the abstract concept that is LEK within the time and
resource constraints of this work. For instance, taking each participant to a biodiverse patch of the
environment or collecting good quality specimens to show all interviewees is time consuming and,
therefore, unviable where high sample quotas are required or local species are protected. Examples
of flashcards used in interviews are given in Appendix I.
Despite its uniqueness, LEK should be scrutinised just as other forms of knowledge are questioned
and validated (Huntington, 2000; Gilchrist et al., 2005). Hence, local experts were used at all sites to
validate that the species used in interviews were common to the site and that uses listed by
respondents were correct.
2.3.1. Interview changes and methodological development

Initially, ethnobotanical flashcards alone were used in interviews to test respondents species
knowledge (Balmford et al., 2002; Nyhus et al., 2003). Photographs of wild flowering plants
(including trees) classified as common to the study site were identified and collated. Flowering
plants were concentrated on initially for the following reasons
1. They are primary producers at the bottom of the food chain.
2. They are found at all study sites.
55
3. They have uses and values that may be transferred orally, through stories for example.
4. Many are threatened in different areas due to habitat destruction.
5. As wild relatives of todays staple crops, they are essential to sustaining the agri-gene pool
and agricultural systems as a whole.
6. They are the basis of many livelihoods historically and continue to be in many resourcedependent societies today.
7. They act as hidden harvest, particularly in resource-deprived, environmentally-stressed
regions, and have an economic value.
8. They are often managed as common property resources accessible to all.
9. Conserving them acts to conserve local biodiversity.
Demographic data collected at the first study site in southern India included age, gender and village
of residence. Age as a variable when examining knowledge acquisition is questionable since
increased age is naturally associated with knowledge accumulation in any cultural context (Zent,
2001). Ideally to compensate for this, LEK studies should be carried out with people from the same
age cohort at different times in their lives, for instance by returning to the site every 5-10 years,
however, this was not a viable option for the time constraints on this research. Therefore, in this
study, respondents from different age cohorts were interviewed and compared to give an indication
of LEK acquisition with age.
Following analysis and interpretation of the India data, additions were made to the interview
structure, including recording respondents occupation, years resided at the site and environment of
childhood. Species categories used in interviews were expanded from wild flowering plants alone to
incorporate crops, mammals and birds. This generated a greater volume of data for analysis,
especially at the UK study site where species uses were rarely known.
In the past, LEK research has generally focused on assessing the knowledge of just one species
grouping from a single trophic level, most frequently plants (Hooper et al., 2005). This study is
unique by comparing knowledge of different species groups from different trophic levels. Crops,
mammals and birds were chosen over other species groupings for the reasons listed in Table 2.3.1.
These groups were the most visual, have been known to the public for generations and are easily
56
classifiable in comparison with other species groups (e.g. insects). In all groups, the species most
common to the area with an ecology that permits survival in a variety of local habitats were selected
(full species listings are included in Appendices 1-3).
Table 2.3.1. Species selection criteria
Species
grouping
Mammals
Birds
Crops
Wild
flowering
plants
Group characteristics
Highly characteristic group thus much public interest and media

coverage
Fewer species in this group than others
Much research focused on this group
Effects of habitat change obvious on this group
Common at all sites
Characteristic grouping
Less media coverage although common as a hobby
High variety locally
Crucial as food source
Important economic status
Important culturally, especially in rural areas
Historically farming important as job source and community basis
Diversity and local knowledge believed to decline with urbanization
Current or historic cultural significance
Current or historical community dependence
Primary producers and food of keystone species
Common and varied at all sites
Conservation crucial to agri-gene pool
By examining the knowledge of a range of different species groupings (flowering plants, crops,
mammals and birds) this methodology made a more comprehensive assessment of local knowledge
than previous studies in the field. This enabled analysis to see if the knowledge of certain species
groups is more successfully transferred than others, for instance if knowledge of local plants is more
successfully transferred than knowledge of mammals in a gatherer community. This gives an
indication of the importance of different species groups to the different communities studied and is
unique to this study (Gilchrist et al., 2005).
In addition to expanding the species identification exercise, structured questions were incorporated
looking at alternative factors that may act to influence LEK. At the UK study sites, these included
57
how frequently a respondent visited the countryside, for what purpose when they did, their
household income and their attitude towards the importance of LEK. Respondents were also asked
the source of their own local knowledge in order to establish the origin of each individuals LEK
and the most effective sources of transfer (Alexiades, 1999).
Following pilot interviews, fixed responses were constructed based on respondents answers to the
added questions. This facilitated data analysis, reduced the ambiguity of questions, reduced
interview time and avoided forcing peoples knowledge into etic categories constructed by the
interviewer. For instance, frequency of time spent in countryside was divided into daily, weekly,
monthly, rarely and never. Where respondents answered other or none of the above this was
recorded on site (Houtkoop-Steenstra, 2000).
Species uses were divided into categories for the purposes of data handling and manipulation. Use
categories included food (humans), construction, farming (both agriculture and livestock), health
(human medicinal) and economic (for sale). These were based on pilot interview responses from
each site combined with answers given in similar studies exploring the uses of local resources
(Phillips et al., 1994; Pfeiffer, 2002). Species use categories varied slightly between sites based on
local markets and livelihoods. All uses listed outside of these categories were recorded on site to
again, avoid forcing respondent information into etic categories.
Where economic income could not be defined by occupation or annual salary, such as in remote
regions of Indonesia where a diversity of income sources is frequently pursued, migration is
frequent and work is seasonally-dependent, alternative economic indicators had to be used. Sitespecific ranking systems were constructed based upon land ownership, social status, house building
materials and ownership of consumer goods.
At the final study site in Indonesia, Likert-scale questions were incorporated into the interview
structure. This allowed the views, perceptions and behaviours of the two ethnic groups on site to be
compared. It also permitted an assessment of how these factors relate to LEK in a resource
dependent society (Laine et al., 2004).
58
At all stages of method development, interviews were piloted and restructured where necessary prior
to use on the general public. This ensured that questions were clear and being interpreted correctly
by respondents to attain the desired responses (Houtkoop-Steenstra, 2000). Photograph quality was
assessed by carrying out 25 pilot interviews inviting comments on image quality. Photographic
flashcards included up to three images of each species, for instance including the leaves and flower
of a plant species. Where pilot respondents commented negatively on the quality of the photographs,
images were either replaced or removed.
All respondents were given the same introductory information informing them of the basic interview
structure without revealing the detailed aims of the study. Wherever possible interviews were
carried out in private to prevent third parties from influencing respondents answers (Holstein &
Gubrium, 1995; Houtkoop-Steenstra, 2000). However, where spectators were unavoidable they were
asked not to contribute at any stage, and if they did, results were discarded. Methodological
development at the three sites is summarised in Table 2.3.2.
Table 2.3.2. Stages of methodological development at all 3 sites summarised
Site 1:India
Collate local species
images
Validate local species
collection with local
expert(s)
Site 2:UK
Construct new expanded
interviews increasing
demographic data
collection, expanding
species identification and
adding in structured
questions
Pilot interviews
Site 3:Indonesia
Construct Likert-scale
questions
Compile economic
indicators
Collate local species
images
Collate local species images

Final alterations
Run interviews

expert(s)

expert(s)
Pilot interviews
Pilot interviews
Final alterations
Construct fixed answers to
structured questions and
final alterations
Run interviews
Run interviews
59
2.4. Sampling strategy

Too little discussion of the methodology used in LEK research has taken place in the past, especially
in terms of informant selection. Davis and Wagner (2003) found that of the 65 studies they
reviewed, only 22 had some form of detailed documentation describing the methods employed.
Most studies investigating LEK, medicinal ethnobotanical knowledge in particular, focus upon
identifying experts as key primary informants, disregarding the knowledge of laypersons that often
hold different epistemologies and utilise LEK in their daily lives (Browner, 1991; Alexiades, 1999;
Nolan & Robbins, 1999; Davis & Wagner, 2003; Howard, 2004; Vandebroek et al., 2004). This
threatens to omit data and misrepresent knowledge distribution within a society, especially
considering LEKs well-known heterogeneity (Howard-Borjas, 2002; Vandebroek et al., 2004).
Focusing on expert knowledge is unsuitable for LEK studies aimed at investigating community
distribution, content, variation and modes of transfer. By looking at the knowledge of male and
female laypersons of all ages within a community, whose ethnobotanical knowledge is tested
empirically often on a daily basis, this methodology identified a gradient from the most to the least
knowledgeable groups. This is impossible by focusing on expert knowledge alone (Barrett, 1995;
Howard-Borjas, 2002).
Limited sample sizes are another shortcoming of many earlier LEK studies. Largely limiting the
number of informants often leads to a biased sample which overlooks differences in vertical and
horizontal knowledge distribution that have been found even in the smallest of communities
(Alexiades, 1999; Martin, 2004). Informants should be selected from all social strata and ethnic
groups found at a site in order to fully represent the full diversity of knowledge distribution. Hence
all voices need to be heard, in particular those of the silent majority (such as women and the rural
poor) who, in the case of local knowledge, may be the richest in society. All too often the
ethnobotanical knowledge of very few is taken to represent the LEK base of an entire culture or
region (Howard-Borjas, 2002).
Many earlier studies were biased towards men (Phillips et al., 1994), some even omitted women
completely from their research considering their knowledge to be inferior, overlooking any gender-
60
differentiation that exists (Alexiades, 1999). Where this is the case, a hole in the research exists.
This omission is particularly prominent in pharmaceutical research whereby researchers commonly
seek out the local healer or Shaman, roles typically filled by men in traditional communities,
overlooking womens daily contributions to healing practises in the home. Awareness is only just
being raised of the vital contribution that women have made to primary healthcare treatments
worldwide (Howard, 2004).
Although sampling strategies used in this study varied between sites as a result of differing
population sizes and accessibility, common goals were maintained based on past shortcomings.
These included to recognise the factors responsible for creating a gradient in knowledge within a
community, to examine the knowledge content of even the most isolated and deprived groups and to
attain a large representative sample at all sites that successfully illustrates knowledge distribution
patterns both within and between communities.
Stratified cluster sampling was used in this study. Cluster sampling was used to select a random
sample of villages/towns/wards within the chosen locality. Then stratified sampling took place
within those villages/towns/wards, ensuring that all sub-populations were represented. Where
possible, sub-populations were sampled at a ratio that reflected their proportions within the actual
population. For instance, the proportion of men and women interviewed at a site was representative
of the proportions with which they existed within the actual population. Where detailed census data
for a region were unavailable, sub-populations were sampled at equal proportions to ensure that each
group was fairly represented.
Thus designated quota sizes were pre-determined and respondents were identified within them.
Where possible haphazard sampling techniques were used, as a means of identifying respondents at
random whilst purposely filling designated quotas. One technique of haphazard sampling frequently
employed in LEK research is chain referrals or snowball sampling (Huntington, 2000; Brodt, 2002;
Davis & Wagner, 2003). Snowball sampling is particularly useful in quota sampling. It entails
respondents being asked to provide several names of other local residents, within the required age
ranges, who may be willing to participate. Names from this list were then contacted at random. The
minimum age for sampling was 15 years. This means that interview structure did not need to be
61
simplified for younger respondents and ethical approval and parental consent was not needed prior
to interviewing. Site-specific sampling methods will be described in the relevant chapters.
Large sample sizes are more likely to give an accurate result close to the value of the true population
(Gilchrist et al., 2005). A number of rules of thumb as to sample sizes suitable for statistical analysis
were adhered to in this study. These recommend that sample sizes in complex correlational research
should reach 30 or more (this applies to individual sub-samples), in multi-variate research the
overall sample size should exceed the number of variables being considered by 10-fold and sample
sizes of 10%, where viable, are recommended (Roscoe, 1975; Hill, 1998). Reasons for variations in
sample size between sites will be discussed within the site-specific chapters. A recommended code
of best practices for LEK research developed in this study is given in Table 2.4.1.
62
Table 2.4.1. Code of best practices for LEK research

Method formulation
Study site selection
Corresponding protocol
Check that the area is safe
Ideally a workable field base should exist to provide entry into local
communities
Species selection
Formulate a list of species commonly found at the study site
Validate the presence of these species with local experts
Image selection
Ensure that species photographs include background elements to
illustrate size and include all species components e.g. a leaf for
plants
Pilot all photographs to check visual quality
Formulating interview
Know the culture e.g. asking a persons level of education may
questions
offend in some parts of the world
Ensure that demographics to be collected are viable in the actual
population e.g. income is often variable and ages unknown in some
remote communities
Pilot interviews to ensure any ambiguity is eliminated
Pilot interviews in actual population to gain an indication of answers
to questions and, where possible, make questions closed with predesignated categories based on these answers
Familiarising interviewer
In unfamiliar cultures work with a long-established and respected
with respondents
local organisation such as an NGO
Spend time learning cultural taboos and accepted behaviour
Where necessary, wear local dress
Go through permission channels to enter villages and arrange forms
of payment for respondents time in advance where necessary
Sample selection
Ensure that all sub-populations are fairly represented
Avoid focusing on experts and select respondents by the most
random means possible e.g. haphazard where quotas need filling
Ensure sample sizes are large enough to be statistically viable
Minimum samples recommended for statistical testing were collected at all sites and SPSS 12.0
statistical package was used for constructing databases for the handling, analysis and manipulation
of data (Kinnear & Gray, 1999). Individual statistical tests will be discussed in the methodology of
each site-specific chapter. Both parametric and non-parametric statistical tests were employed.
Parametric were used where data collected were normally distributed, whereas non-parametric
statistical tests were used to provide a more accurate and conservative result for heterogeneous
populations where data were non-normally distributed. More often than not data collected in this
study were found to be non-normally distributed and hence non-parametric tests were used.
On the whole Mann Whitney-U and Kruskall-Wallis statistical tests were used, and their parametric
equivalents (t-test and analysis of variance), since the aim of this study was to compare mean
63
ecoliteracy levels between groups of respondents to identify the most and least knowledgeable
within populations. Regression analysis was not used in this study since it can only be used if the
nature and direction of a relationship are known prior to testing. Finally, Spearmans rank and its
parametric equivalent (Pearsons product moment) were used to assess the strength and direction of
relationships between variables where there is likely to be an association rather than a causal
relationship (i.e. a dependent and independent factor).
64
3.1. Introduction
India, Asias second largest nation, ranks seventh largest country in the world stretching over an
area of 3,287,263 square kilometres, covering over 2% of the worlds land mass. It reaches from 8
4 to 37 6N latitude and from 68 7 to 97 25E longitude with a coastline stretching 7516 kilometres
(Gadgil, 1993; SDNP, 2002). As the second most populous nation due to a population explosion
from 300 million in 1947 to over one billion today, combined with unequal income distribution,
extreme pressure is placed on environmental resources and infrastructure, hence food production is
critical to this extensive nation (EIA, 2001).
3.1.1. Indias dependence upon the land

As much as 86.8% of Indias vast land area is under agriculture, forestry, pasture or another form of
biomass production. Consequently, there is virtually no high quality land left uncultivated
nationwide, although the population this finite land mass has to support is rapidly expanding (India
State of Environment Report, 2001). Despite widespread land dependence, over 70% of the country
is afflicted with low and unpredictable rainfall preventing intensive cultivation and dependable
yields even in the main cropping season (Nyrop, 1975). Therefore India, despite climatic constraints
and vast political reform under both British colonisation and independent rule, has remained preeminently an agricultural country. Even those pursuing industrial occupations often retain their
agricultural skills and practices as a subsidiary form of income (Gadgil, 1972).
The rural poor, dependent primarily on the limited income of agricultural labour and own food
production capacity, constitute the single largest poverty-stricken group (Farrington et al., 2003).
Nevertheless, agriculture is Indias primary economy in terms of employment accounting for over
60% of livelihoods, but only generating 26% of national GDP (TIDCO, 2002a). Hence poverty and
malnutrition is a widespread problem in rural India. Between 1996 and 1998, 207 million nationals,
including a disproportionate number of scheduled castes and tribes, were unable to access enough
food or purchasing power to meet the minimum human health nutritional requirements and, as a
65
result, over half of the populations children are falling underweight, girls in particular (Ludden,
1999; Farrington et al., 2003).
Indias climate dictates two seasons of farming. The Karif season takes place between July and
September (summer monsoon period). Karif crops include rice and constitute two-thirds of the
annual harvest. From October to March (winter monsoon) is the Rabi farming season, which
includes crops such as wheat (Nyrop, 1975). Despite its climatic constraints, India continues to be
an important exporter of many major crops including sugarcane, groundnut, tea, rice, pulses,
tobacco, cotton and castor seed (Nyrop, 1975).
A structure of social order has persisted in India for over 3000 years termed the Caste System. This
is a form of social hierarchy whereby Hindu nationals are born into a caste and subsequently
constrained to it for life. A persons caste determines where they live, who they marry, and their
livelihood pursuits. It first evolved when darker skinned Aryan tribes migrated from the plains of
central Asia between 1500 and 500BC. Still today the lowest social groups can be depicted by their
darker skin tones.
Up to four social castes exist in mainstream society and below these in the order of ranking lie those
groups thought to be unclassifiable and below the ranking system, termed the Untouchables.
Members of this designation are considered indigenous outcastes and traditionally lack all societal
privileges, restricted to the dirtiest of jobs. They are now frequently referred to as Dalit
communities, a term coined by the society members themselves meaning suppressed (Glaeser,
1995).
Not only are Dalit communities suppressed of their rights to follow chosen occupations and mix
with the remainder of Indian society but, due to exceedingly low literacy through denied access to
education, they are deprived of any social standing in local decision making. Therefore, like many
minority groups, their accumulated knowledge of the land passes through generations orally
unabated (Gadgil, 1972; Nyrop, 1975). Farming decisions such as combining water, crops, soils,
manure and labour are crucial to the self-sustainment of under-privileged groups and have to be
made accurately and efficiently to ensure survival. Age-old methods of agricultural prediction relied
66
on for centuries by these groups include insect patterns, animal behaviour, cloud shapes and even
the shapes and sounds of thunder and lightning, derived from generations of detailed observations
(Ludden, 1999).
Thus LEK can be key to both the prediction of and survival through extensive famine periods,
usually caused by the failure of the southwest monsoon in this region but also occasionally by
adverse winter weather conditions and insect blights. Key to survival through such periods is
knowledge of hidden harvest opportunities, including edible wild plants and non-poisonous
palatable animals. When and where food scarcity prevails, locals are forced into alternative forms of
subsistence including grass seeds, green vegetables and even local leaves and flowers that can be
mixed in high proportions with any small amounts of high-priced grain that families are in
possession of. This knowledge has been relied upon and transferred between Indian hunter-gatherer
societies as far back as 6000BC (Bangalore CES, 2003).
Knowledge of the neem tree (Azadiracta indica) is a prime example of Indian ecological knowledge
that has been shared with the world community for benefits to be reaped by all. For a long-time in
India the neem tree has been both revered and respected for its multitude of invaluable uses. A key
asset to farmers is the ability of different neem extracts in keeping almost 200 species of insect at
bay, these include the diamond leaf moth and floral thrips. Chemical tests carried out long after rural
farmers discoveries revealed bioactive ingredients within the extracts that act to kill the larval stage
of the target insect, prevent it from feeding or inhibit the female from depositing eggs depending on
how it is applied. However, concerns have been raised as to whether these uses and their application
methods will be forgotten in traditional communities with the introduction of synthesised chemicals.
Other applications include in medicines, toiletries, cosmetics, livestock protection and public health
(UNDP TCDC, 2003a).
Due to 70% of Indias vast population inhabiting rural areas (EIA, 2001), a number of policies have
been introduced attempting to improve agricultural livelihoods and living conditions for the rural
poor, the most significant of which was the Green Revolution that aimed at keeping food production
growth ahead of population growth in developing regions (Lipton, 2004). Using investments from
the Rockefeller and Ford Foundations, the Green Revolution began with the transfer of genetic
67
material from Japan and Mexico. Scientists then experimented with these new crop varieties
alongside local farmers in Mexico and produced a dwarf variety of wheat, one of the first High
Yielding Varieties (HYVs) developed. This new seed variety, alongside HYVs of rice, maize,
sorghum and others, was key to improving food production systems across Asia and Latin America.
Therefore combining Western science with a farmer first participatory approach yielded the Green
Revolution with mixed outcomes (Biggs & Smith, 1998; Poole & Penrose Buckley, 2006).
The Green Revolution succeeded at increasing crop yields for growing populations even on poor
quality farmland and thus providing food security, improving diets through increased calorie and
protein intake per person due to reduced food prices, increasing labour and wage rates for the rural
poor in particular and reducing poverty in rural areas (Hazell & Ramasamy, 1991; Rosegrant &
Hazell, 2001; Hazell, 2003; Lipton, 2004, 2005). Consequently the Green Revolution has doubled
food production and greatly reduced poverty across Asia over the last three decades as one of the
only agricultural movements where landless labourers and small-scale farmers have gained as much
as large-scale farmers (Asian Development Bank, 2001). As well as increasing the agricultural
economy, the Green Revolution also boosted the non-farm economy of Asia and the availability of
non-farm jobs. This was primarily due to increased marketing and processing demands from the
farm economy combined with increased consumer demands due to increased wage rates (Hazell and
Ramasamy, 1991). Therefore, the Green Revolution succeeded in improving food security and
economic conditions in Asia and Latin America, however it did not come without costs.
Despite reducing economic disparities between large and small-scale farmers, the Green Revolution
failed to reduce disparities in income between farmers working on rainfed marginal lands compared
to those with irrigation resources, as HYVs did not yield large quantities where water resources
were limited. Hence it yielded little benefits for drought-stricken regions such as Sub-Saharan
Africa which still lacks food security (Hazell & Ramasamy, 1991; Asian development Bank, 2001;
Hazell, 2003; Lipton, 2004). In addition, the chemical fertilisers used to maximize HYV production
polluted local water sources and water tables declined, micronutrients were depleted from soils,
monocultures reduced local biodiversity and there was no stabilization of food production in the
event of drought (Hazell & Ramasamy, 1991; Hazell, 2003; Lipton, 2004). Thus in 1970, 23 million
rural households across India remained landless, and despite vast labour contributions, only 10% of
68
all agricultural income was accrued by these households (Nyrop, 1975). Despite this, it is likely that
poverty and famine would be much higher today if it were not for the innovations of the Green
Revolution (Hazell, 2003).
Pressures on agriculture and economic development have led to rapid industrialisation all over India,
placing further strains on the land through activities of deforestation, land degradation and soil
erosion. As a result urban migration has increased both seasonally and permanently with people
seeking the higher wages of the textile and leather industries to name a few (TIDCO, 2002a). With
industrialisation has come the arrival of Western allopathic medicine, although this tends to be
expensive and concentrated in urban areas. The popularity of this modern medicine in urbanised
areas remains as a post-colonial hangover which encouraged people to disparage traditional systems
of medicine using local species and opt for modern and costly pharmaceuticals (UNDP TCDC,
2003b). Hence, today, detailed medicinal knowledge based on cultural traditions and tried and tested
methods is concentrated amongst the rural poor, in particular the socially-deprived groups such as
Dalit communities and women. In these areas modern medicine is inaccessible financially and
physically. The potential value of this traditional medicinal knowledge has only recently been
realised on a global scale. Hence recent policy introductions at the national level encourage
medicinal knowledge protection and co-development with modern techniques (UNDP TCDC,
2003b).
3.1.2. Indias biological diversity

In the face of widespread cultivation, India has retained its biological diversity ranging from
landscapes of tropical evergreen forests and mangroves to dry deciduous forests and scrub desert
(Gadgil, 1993). This range provides habitats for 7.6% of the worlds mammal species, 12.6% of the
worlds birds (55 of which are endemic to the country) and 6.0% of flowering plant species (33% of
which are endemic constituting 15,000 species).
Despite high landscape diversity and designated protected areas (including national parks and
sanctuaries), 3000-4000 of Indias flowering plants are said to be under some degree of threat
(SDNP, 2002), hence India became party to the international Convention of Biological Diversity in
June 1992. This aimed at providing a framework for the sustainable management and conservation
69
of natural resources. Derived from this, an Act of Parliament was passed termed the Biological
Diversity Bill in 2000 aimed at protecting national resources and ensuring equitable benefit sharing
of their goods and services. This in turn instigated the construction of the National Biodiversity
Authority of India given the task of conserving national biological diversity and monitoring
external, environmentally-impacting activities (Biological Diversity Bill, 2000; SDNP, 2002).
3.1.3. The state of Tamil Nadu

India, for its vast size and high population density, has been divided up into political regions
comprising of 7 unions and 24 states (Bangalore CES, 2003), allowing for a multi-level system of
governance at the levels of union, state, district, block and to some degree village-level. For the
purpose of this study a number of villages within the block of Tiruchuli were chosen, within the
district of Virudhunagar that is situated within Indias most southern state, Tamil Nadu. Within
traditional rural areas the Panchayati raj system of local self-governance still exists, promoting
grass-roots decision making capacity and self-development in theory. In practice, however, the
power of this tiered system is most concentrated amongst state government officials and electors.
Thus the village Panchayat exists as an ancient, but in most cases powerless, institution (Nyrop,
1975).
Tamil Nadu is Indias fourth largest state, extending over 130,058 square kilometres with a
population of 62.1 million according to the 2001 census (Tamil Nadu Human Development Report,
2003; Government of Tamil Nadu, 2004). Bordered by Karnataka and Andhra Pradesh in the north,
the eastern Ghats and Kerala in west, the Bay of Bengal to the east and the Indian ocean to the
south, it is encompassed by 1000 kilometres of fragile coastline (Tamil Nadu Human Development
Report, 2003; India Travelogue, 2004) (Fig 3.1.1).
The state is divided into five major physiographic regions; the Kurinji or mountainous region, the
Mullai or forested region, the Palai or arid zone, the Marudham or fertile plains and the Neidhal or
coastal zone (India Travelogue, 2004). The four largest river basins in the state include the
Pennaiaru, the Vaigai, the Tamaraparani and the Cauvery, the latter of which and its tributaries
provide Tamil Nadu with its major source of water and power (TIDCO, 2002b; India Travelogue,
2004). The official state language is Tamil and the primary religion Hinduism (India Travelogue,
70
2004). The 2001 census revealed that 44% of the population resides in urban areas compared with
56% in rural, and three-quarters of the latter pursue primary occupations in agricultural production
(Government of Tamil Nadu, 2004; India Travelogue, 2004).
Tamil Nadu
Figure 3.1.1. Map of India showing the southern state of Tamil Nadu (Multimap, 2006,
1:40,000,000)
With such a vast rural population it is no surprise that 44% of the land area of Tamil Nadu,
excluding forested zones, is under cultivation (TIDCO, 2002b). The problems faced by farmers of
this region include fragmented land holdings (averaging at 0.93ha (Government of Tamil Nadu
Policy Notes, 2005)), lack of investment capital, urbanisation pressure, poor infrastructure and no
standardised route of information dissemination (TIDCO, 2002c). In addition to these institutional
problems, the tropical climate of Tamil Nadu suffers high temperatures (exceeding 40oC in the
summer months), poor soil conditions (especially in the plains where infertile degraded soil
prevails) and frequent drought caused by monsoon failure depleting ground water levels and
reservoir/tank storage (Table 3.1.1) (Rao, 1994; Ludden, 1999; Government of Tamil Nadu Policy
Notes, 2005; Tamil Nadu Agriculture Department, 2005).
Table 3.1.1. Annual averages in rainfall over the state of Tamil Nadu, southern India (1999-2003)
Year
Normal based on past averages
1999
2000
2001
2002
2003
Source: Tamil Nadu Agriculture Department (2005)
Rainfall (mm)
967.2
837.3
847.5
774.7
748.4
867.8
71
Tamil Nadu has six distinct agro-climatic conditions, enabling it to produce every type of fruit and
vegetable, and despite restricted rainfall, paddies are frequent (Ludden, 1999). Annual state food
grain production exceeds 10 million tonnes, of which rice constitutes approximately 80%. Other
cash crops of the state include sugarcane, tobacco, chilies, coconuts, coffee and cotton. Also
important as food crops are oilseeds, pulses, mangoes, bananas and spices (TIDCO, 2002b, 2002c;
India Travelogue, 2004).
Despite farming being the mainstay for most household economies within Tamil Nadu, agriculture
does not comprise the majority of the states economic income. Through establishing itself as the
second most industrialised state behind Maharashtra, Tamil Nadu has a well-developed and
diversified economy with a state domestic product of US$19.6 billion giving it the third highest
national state economy (TIDCO, 2002c; Tamil Nadu Human Development Report, 2003).
Agriculture and related activities only contribute 19.7% of this figure, manufacturing constitutes
32.7% and services 47.6% (TIDCO, 2002c). Primary industries include sugar, alcohol-based and
textiles production, namely cotton ginning, spinning and weaving (TIDCO, 2002b; India
Travelogue, 2004). Rural inhabitants often seek jobs in these industries during seasonal periods of
urban migration when drought reduces available workload on the land (Gadgil, 1972).
Along with state industrialisation came improvements to education. Literacy rates increased from
63% to 73% in the decade between 1991 and 2001. This figure is representative of 83% of men and
65% of women, although this is heavily concentrated amongst the higher castes in urban areas with
rural enrolment remaining very low. As a consequence of illiteracy afflicting two-thirds of the aged
population, extensive school construction has ensured that most children today have access to a
primary school within one kilometre. As a result, Tamil Nadu is now ranked third in education level
amongst all the states (Nyrop, 1975; Tamil Nadu Human Development Report, 2003; Government
of Tamil Nadu, 2004) and has an average human development index of 0.66, exceeding the national
figure of 0.60 (Tamil Nadu Human Development Report, 2003; UNDP, 2003).
Despite advances in development, Tamil Nadu still suffers from traditional caste differentiation and
rural deprivation. As much as two-thirds of the state population derives from backward classes and
72
denotified communities, mostly residing in one of the 64,000 nucleated villages (India Travelogue,
2004; Government of Tamil Nadu Policy Notes, 2005). Despite being one of the most industrialised
states, Tamil Nadu fails to achieve equitability in its wealth, with 13 million people in 2000 (21% of
the state population) living below the poverty line, 8 million from rural areas alone. Poverty and
malnutrition are particularly prevalent in the lowest castes and tribes of the state, amongst which
33% of households fall below the poverty line, illustrating sharp imbalances between rural and
urban areas and high and low castes (Tamil Nadu Human Development Report, 2003). Women in
particular suffer due to unequal access to food, healthcare and education. Their traditional domestic
roles are considered to be of little importance. However, rural women in actual fact make up the
majority of the agricultural labour force and work the longest hours for the lowest wages (Bryson,
1996; Tamil Nadu Human Development Report, 2003).
3.1.4. Virudhunagar district

States of India are divided into districts which constitute the principal subdivision politically
(Nyrop, 1975). Virudhunagar is a southern district of Tamil Nadu state known for its acute droughts,
erratic monsoons and tropical climate as a result of its location in the rain shadow of the Western
Ghats (Muller, 1982; Nathan, 1995; Devavaram et al., 1999). The dry climate sees little variation
between summer and winter temperatures, exceeding 40oC in April and May and rarely falling
below 20oC from October through to December (India Travelogue, 2004).
Virudhunagar district is landlocked on all sides. It contains no perennial rivers, just those fed by the
annual monsoons. The southwest monsoon falls between June and August and provides scanty
rainfall, however the bulk of the rain falls in the northeast monsoon between October and December.
Despite two monsoon periods annually, the rainfall across the district is still exceedingly limited
(NIC, 2003). Hence the climate of the region is defined as semi-arid tropical monsoon type,
characterised by high mean temperatures and low humidity (NIC, 2003).
Virudhunagar district had a population of 1,751,548 in 1991 according to census data. Up to 37% of
the land area is cultivated and provides sustenance to over half of the working population.
Cultivated crops include black gram, green gram, small millets, maize, groundnut, gingili, chili,
sugarcane and cotton, an important economic crop of the region. In addition to agriculture,
73
Virudhunagar produces over 70% of Indias matches and fireworks (for its dry climate and historical
exploitation of child labour) and has a widespread distribution of cottage industries amongst the
rural poor including matchbox making (NIC, 2003).
Madurai is the closest urban centre for Virudhunagar residents and constitutes the second largest
city in the state of Tamil Nadu covering 22 square kilometres (Emerging Planet India, 2003).
Virudhunagar, in 2001, had a literacy rate of 74% (increased by almost 10% over a decade) with
85% of men and 64% of women now being classified as literate (NIC, 2003; Government of Tamil
Nadu, 2004). Thus although human development is improving in the region (HDI = 0.66), like
gender development (Gender-related Development Index (GDI) =0.65; equal gender development
GDI =1), it still has a long way to go (Tamil Nadu Human Development Report, 2003).
3.1.5. Tiruchuli block

Virudhunagar district comprises two land types, the productive biodiverse slopes of the Western
Ghats and the black soil plains. The latter includes Tiruchuli block, characterised by black cotton
soil known locally as karisal and dominated by 40 rural Panchayat villages (NIC, 2003; Tamil Nadu
Maps, 2004). The NGO SPEECH (Society for Peoples Education and Economic Change) has been
working in this region to improve living conditions, socioeconomic stability and societal imbalances
in rural poor communities since 1986 (Glaeser, 1995).
Villages are key to traditional Indian social structure, defined as the smallest self-sustaining unit
both socially and economically, only trading externally for non-locally derived salt and spices
historically (Gadgil, 1972). With SPEECHs support, local villages in Tiruchuli block have
established water harvesting schemes, money lending initiatives, farmers working groups and
womens sanghas (Devavaram et al., 1999). Hence SPEECH staff have forged good reciprocal
relations of trust with local villagers over long periods.
74

Village communities in Tiruchuli block were chosen for study based on having had long-term
relationships of dependency with the land and its resources, in terms of both livelihoods and
sustenance, and having differing degrees of economic development and external influence (Table
3.2.1). In addition, they had sustained long-term relations with the local NGO, SPEECH, and were
situated varying distances from urban centres. In total, four rural villages were selected, Kottam,
Udayanampatti, Kallunmadam and Kariakalkulum. All rural villages were situated varying distances
from the small town of Tiruchuli (1 hour drive from Madurai) which lacks expansive development
but has an established market to which external products are transported. Also Sungureampatti was
chosen as a suburban village lying on the outskirts of Madurai.
Table 3.2.1. Key characteristics of villages selected for study
Village
Kottam
Udayanampatti
Ethnic composition
Dalit
Dalit and non-Dalit
Land ownership
Little ownership
Little ownership by
non-Dalit
Kallunmadam
Dalit and non-Dalit
Little ownership by
non-Dalit
Kariakalkulum
Non-Dalit
Land ownership
Sungureampatti
Dalit
Landless
Source: Erskine, personal communication, 14 July 2003
Location
Rural
Rural
Rural
Rural
Suburban

Wild flowering plant species found in the state of Tamil Nadu were collated from Kew Botanical
Gardens SEPASAL database (Survey of Economic Plants for Arid and Semi-Arid Lands) totaling
494 species (Royal Botanic Gardens Kew, 2003). Species whose soil preferences, habitat
requirements, rainfall or running water requirements, length of growing season or typical altitudes of
growth were outside of the study area conditions were then eliminated. Only those species with
habitat requirements and tolerances that fitted the ecology of the study site, including drought
resistance and high temperature tolerance, were retained. Added to this listing were a number of
species observed on a preliminary field visit in January 2003. The full listing included 71 wild
plants.
75
Internet search engines were used to find high quality images of the flowering plants chosen. Closeup photographs of the fruit and leaf of each species were used in conjunction with a whole plant
image. Thus in some cases a single species card held up to three different images of a plant in order
to clearly represent all of its components.
The final compilation of species images was validated and verified at the field site with a local
expert, Professor J. Jebaraj, a professor of local botany from American College, Madurai. He was
asked to identify local names and uses and to confirm that each species could be found growing
locally at the study sites. From this a database of local names was constructed. Where species were
described as not growing locally, the species in question was eliminated. The final listing included
48 wild plant species listed in Appendix II.

From each of the five villages, both men and women were sampled from two age groups, under and
including 20 years of age (defined as young people) and over 20 years (defined as adults) (20 years
being the typical age for getting married and starting a family in this region (J. Devavaram, personal
communication, 15 July 2003)). Respondents were divided into just two age groups upon
recommendation from members of SPEECH due to very few adults knowing their age or the year
they were born and few villagers surviving to an elderly age. Sampling was opportunistic.
Respondents were identified by a member of SPEECH who went ahead to the village and rounded
up any men or women from a specified age group available at the site, not just those known to be
local experts or healers.
A total of 192 respondents were interviewed. At this early stage of the research, respondents were
interviewed in groups of varying sizes. This was used to assess the optimal group size for accurately
assessing the ethnobotanical knowledge of a large population. A total of 25 groups were interviewed
with respondent numbers ranging from 1 to 29 people (Table 3.2.2).
76
Table 3.2.2. Summary of respondent group sizes and demographics from each village
Village
Kottam
Kottam
Kottam
Kottam
Kottam
Udayanampatti
Udayanampatti
Udayanampatti
Udayanampatti
Kallunmadam
Kallunmadam
Kallunmadam
Kallunmadam
Kariakalkulum
Kariakalkulum
Kariakalkulum
Kariakalkulum
Kariakalkulum
Kariakalkulum
Kariakalkulum
Sungureampatti
Sungureampatti
Sungureampatti
Sungureampatti
Tiruchuli womens
federation
Group
number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Group
size
7
28
3
9
13
2
12
6
8
10
24
6
10
1
1
1
1
1
1
1
5
5
3
5
29
Gender
Age
Location
F
F
F
M
M
F
F
M
M
F
F
M
M
F
F
F
M
M
M
M
F
F
M
M
F
Adult
Adult
Young
Adult
Young
Young
Adult
Young
Adult
Young
Adult
Young
Adult
Adult
Adult
Young
Adult
Adult
Adult
Young
Young
Adult
Young
Adult
Adult
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Suburban
Suburban
Suburban
Suburban
Rural
After observing group interviews and analysing the results, it was decided that individual interviews,
although not as time efficient, were a more accurate way of assessing ethnobotanical knowledge, a
key methodological development. Interviewing respondents on an individual basis eliminated bias
caused by passive participation of quieter group members and over-dominance by others inevitable
in group situations. Also, individual respondents could be interviewed within the privacy of their
own home away from other villagers keen to intervene and demonstrate their own knowledge. Thus,
based upon these preliminary findings, individual interviews were held at future sites.
A preliminary site visit was made in January 2003 to make connections with the NGO SPEECH,
visit villages, get an idea of the ecology of the region and the feasibility of the study and learn about
cultural norms and taboos. Interviews then took place during a two month field visit throughout July
and August 2003. One primary translator from the NGO was used enabling them to become highly
77
skilled at the interview process and understand the importance of not influencing respondents
answers (Twyman et al., 1999). Using a single translator ensured consistency in translation. On the
rare occasion that another translator had to be taken into the field they were briefed on the aims of
the study and interview process by both the interviewer and the primary translator.
3.2.4. Ethnobotanical interview structure

Ethnobotanical interviews lasted 1-2 hours and were piloted on site with local members of SPEECH
staff. Wherever possible, a quiet accessible area of each village was identified for interviewing to
minimise interruptions. A standard introduction was given at the start of each interview introducing
the interviewer and stating that the study aims were to investigate plants of the surrounding area, not
to assess local knowledge. The age group, gender and village of residence of each group were
recorded. Respondents were shown the species flashcards one by one. Where group sizes were large,
it was often necessary to walk the photograph around the perimeter of the group until everyone was
satisfied that they had clearly seen the image.
After viewing the species image, the respondents were asked to raise their hand if they recognised
the species and knew its name. The number of raised hands were counted and taken to represent the
proportion of the group that could identify the species. It is possible that this method may create
some level of bias if more people raised their hands than actually knew the species. This may be a
consequence of peer pressure stemmed from knowledge assessment in a public arena. However it is
an efficient method of assessing ecological knowledge levels in a group situation, groups were
informed that the purpose of the study was just to gain information about local species and the plant
identification results yielded were not significantly higher than those yielded at the Indonesian study
site where different methods were used. Therefore, we can assume that this technique did not largely
bias the results collected.
The group members that recognized the species were then asked to name the species shown to them
to ensure that species identification was correct and in agreement with the pre-recorded names given
by Professor J. Jebaraj. Where local names differed, they were recorded on site. Local healers and
SPEECH members were later consulted to determine if the name given was a misidentification or
simply an alternative vernacular name.
78
Following identification, the respondents with raised hands were asked what the species in question
could be used for. All uses identified were recorded. This was repeated for all of the 48 plant
species. A copy of the data recording sheet used for interviews is included in Appendix V. Uses
were later recoded into categories including health, food, agriculture, building and other. These were
based on the most frequent uses listed and, therefore, considered to be the most important to wellbeing and livelihoods.
In addition, three unstructured interviews were carried out with J. Jebamalaidass, a healer and trainer
of traditional medicinal practices, a male healer and a female healer from Kariakalkulum. Questions
asked were open and probed the respondents views on changes in ecological knowledge, modes of
transfer and local ecosystem changes.

A database was constructed for the India site results using SPSS 12.0. This enabled the data to be
manipulated, inconsistencies detected and results statistically analysed. All data were tested for
normality (Kolmogorov-Smirnov test) and variance (calculated by coefficients of variance). To
account for group size variation influencing the data, the % of each group able to recognise a species
(indicated by raised hands) was calculated and used in analysis.
This was calculated firstly by working out the maximum score that would be achieved if all group
members were able to recognise 100% of the species shown to them. Then an actual score was
calculated based upon the % of species that were correctly recognised and the proportion of group
members able to recognise each one. For instance, if all group members in a group of 25
respondents were able to recognise all 48 species, 100% recognition would score 1200 (25
respondents x 48 plant species). If only 10 group members recognise 20 species, they score only 200
and thus have a naming ability of only 17% (200 as a % of 1200 total species) as a group. However
if 5 respondents were able to recognise 10 species and the other 20 were able to recognise 40, the
group would have a naming ability of 71% ([5 respondents x 10 species] + [20 respondents x 40
species]=850; 850 as a % of 1200=71%). This was used to measure the ecoliteracy level of each
respondent group.
79
Where data were found to be normally distributed, parametric tests were employed and means
compared. Non-parametric tests were used to analyse the remainder (health, food and construction
use knowledge) focusing upon medians and interquartile ranges. Comparisons in knowledge level
between age groups and villages were made using one-way analysis of variance. Gender
differentiation in ethnobotanical knowledge was analysed using t-tests, as were differences between
rural and suburban sites.
Since income was highly seasonal and depended upon annual harvested yields, an assessment of
village wealth was made through a ranking exercise. Villages were scored based on the caste group
that villagers belonged to (Dalit, non-Dalit and mixed) and land ownership (some, little, none).
Villages scored between 2 and 6, 2 being indicative of the lowest community wealth and 6 the
highest. A relationship between knowledge level and knowledge variance with wealth rank was
tested for using Pearsons product moment.
Comparisons in knowledge content were carried out by looking at the number of uses identified in
the different use categories by different groups. Comparisons in knowledge of agricultural uses and
other uses between age groups and villages were made using one-way analysis of variance.
Kruskall-Wallis was used for health, food and construction uses. Gender differentiation in
knowledge of agricultural and other uses was tested for using t-test. Mann Whitney-U was used for
health, food and construction uses. Use knowledge was measured as total number of uses listed and
number of uses listed per identified species. The effect of healer status on level of health knowledge
was also tested using Mann Whitney-U. Finally, an association between species familiarity amongst
locals and its usefulness was tested for using Pearsons product moment.
3.3. Results
Despite methodological developments derived from experiences at the India study site (including
altering respondent numbers and interview content), some interesting results were nevertheless
yielded that contribute to the current understanding of LEK. The data looked at the most and least
80
knowledgeable groups based on age, gender, village of residence and wealth rank. Differences in
knowledge content were also examined.
Ethnobotanical knowledge was measured on three levels; (i) ecological literacy (in terms of local
species recognition), (ii) use knowledge (number of local species uses identified) and (iii) type of
use knowledge, such as food and health (number of species uses known in the different use
categories). Differences in knowledge level (i) is indicative of primary knowledge differentiation
(species familiarity), differences in level (ii) is indicative of secondary knowledge differentiation
(species use knowledge) and differences in level (iii) is indicative of specialist knowledge groups
(knowledge content). On the whole, primary ecological knowledge (level (i)) appeared to be
homogeneous, but heterogeneity, and thus barriers in transfer were revealed upon analysis of
knowledge levels (ii) and (iii). Barriers in the transfer of use knowledge were found both within and
between communities.

The effect of physical location was examined by comparing ecoliteracy levels and use knowledge
between communities from two different environments, rural and suburban. Total ecoliteracy and
plant use knowledge on the whole was not significantly different between villages from the two
sites. Across all five villages, respondents identified on average 46% of local plant species shown to
them. Respondents listed on average 29 uses for the 48 plant species, with a mean of 1.5 uses per
species identified. Suburban village knowledge of other plant uses was significantly higher than
rural knowledge (t=-2.575, p<0.05, n=192). However, variation around the mean was found to be
much larger at the suburban site (Fig 3.3.1).

The effect of wealth was investigated by ranking key indicators of wealth from all five villages and
comparing the overall score to mean community ecological knowledge. Mean community wealth
rank was significantly lower in the suburban village than the rural villages (U=0.000, p<0.01,
n=192) (Fig 3.3.2). Community ecoliteracy was found to be inversely related to mean community
wealth (r=-0.830, p<0.001, n=163), although each village showed considerable variation around the
mean (Fig 3.3.3). Community knowledge variance (difference between the most and least
81
Other use knowledge

(number of other uses listed)
12
10
8
]
4
]
0
Rural
Sub-urban
Location of village
Figure 3.3.1. Comparing levels of other use knowledge

across 2 village locations (mean + 2SE, n=192)
Wealth rank
2
Rural
Suburban
Location of village
Figure 3.3.2. Comparing wealth rank across 2 village

locations (median + interquartile range, n=192)
Ecoliteracy (% total plants identified)
82
65
55
45
35
]
25
1
Wealth rank
Figure 3.3.3. Mean levels of community ecoliteracy and

relationship with community wealth rank (means + 2SE,
n=163)
Community knowledge variance

(coefficient of variance %)
30
25
20
]
15
10
1
Wealth rank
Figure 3.3.4. Mean levels of community knowledge

variance and relationship with community wealth rank
(means + 2SE, n=163)
83
knowledgeable village members) on the other hand was positively associated with community
wealth rank (r=0.825, p<0.001, n=163) (Fig 3.3.4).
Community knowledge variance was found to be inversely related to both community ecoliteracy
(r=-0.628, p<0.01, n=163) and community use knowledge (r=-0.575, p<0.01, n=163) (Fig 3.3.5).
Thus the least wealthy village of Sungureampatti had low knowledge variance and high levels of
ecological knowledge. In comparison Kariakalkulum, ranked the wealthiest village, revealed large
differences in knowledge level between expert and non-expert villagers (variance), but lower mean
knowledge levels overall.

Gender differences in ecological knowledge were assessed by comparing the ecoliteracy level and
plant use knowledge of men and women of all ages. Men and women did not show significant
differences in their levels of ecoliteracy and species use knowledge. Women knew equal numbers of
food, health, construction and other plant uses as men. However, mens knowledge of farming uses
was significantly higher than womens (t=-3.149, p<0.01, n=192), despite high variation in the
knowledge of both (Fig 3.3.6).

Age differences in ecological knowledge were examined by comparing ecoliteracy and plant use
knowledge between young people (20 years and under) and adults (over 20 years). The mean
number of uses per recognised species was found to be significantly higher in adults than in children
(t=-2.569, p<0.05, n=192), despite high variation around the mean (Fig 3.3.7). However, ecoliteracy
did not significantly differ between adults and young people.
Adult knowledge of health uses significantly exceeded young peoples (U=16.500, p<0.01, n=192),
although adult health knowledge showed higher variation (Fig 3.3.8). No such age differences were
revealed in knowledge of food, construction, farming and other uses for plant species shown.
A better picture of the distribution of rural Indian health knowledge was revealed by examining the
knowledge of young people and adults of each gender separately (Fig 3.3.9). A significant
84

30
A
]
Kariakalkulum
25
20
Kottam
]
Kallumadam]
15
Sungureampatti
Udayanampatti
10
30
35
40
45
50
55
Community mean ecoliteracy

(mean % local species identified)

30
B
]
Kariakalkulum
25
20
Kottam
]
]
15
Sungureampatti
Kallunmadam
]
Udayanampatti
10
20
25
30
35
40
Community mean use knowledge

(mean number of uses listed)
Figure 3.3.5. Levels of knowledge variance in 5 villages

and relationship with (A) mean levels of community
ecoliteracy and (B) mean levels of community use
knowledge (n= 163)
85
Agricultural use knowledge

(number of farming uses listed)
14
12
10
2
women
men
Gender
Figure 3.3.6. Comparing the mean levels of agricultural

use knowledge between men and women (mean + 2SE,
Uses per species (number of uses

listed per identified species)
n=192)
2.0
1.8
1.6
1.4
]
1.2
1.0
<20 years
>20 years
Age group
Figure 3.3.7. Comparing the mean number of uses known

per recognised species between adults and young people
(mean + 2SE, n=192)
86
Health use knowledge

(number of health uses listed)
30
20
10
0
<20 years
>20 years
Age group
Figure 3.3.8. Comparing differences in level of health

knowledge between adults and young people (median +
interquartile range, n=192)
(number of health uses listed)
Health use knowledge
30
Age group
<20 years
>20 years
A
20
10
0
women
men
Gender
Figure 3.3.9. Comparing differences in level of health

knowledge across 2 age groups in both women and men
(median + interquartile range, n=192)
87
difference in knowledge of local plant health uses was revealed between the four groups (H=13.555,
p<0.01, n=192), with womens knowledge exceeding mens and girls knowledge exceeding boys.
Overall adult women knew the most health uses but revealed large variation around the median,
whereas young men knew the least.
Although a species was denoted as medicinal if it had one or more health uses, some species could
treat up to 11 different illnesses. These included diarrhoea, jaundice, diphtheria, bronchitis and
conjunctivitis. However, on interviewing a healer from Kariakalkulum, he said that although he
treats over 1000 people in his own and nearby villages and uses over 100 plant species in healing, he
still does not use as many as his predecessor did, in this case his father-in-law.
Despite high variation in ecoliteracy, men and women revealed distinct patterns in knowledge
acquisition with age (Fig 3.3.10). Mens knowledge was found to be saturated by the age of 20 years
and stay relatively constant thereon, declining gradually in later life. Women, on the other hand,
continued to acquire ecological knowledge until the maximum age of 50 before loss occurred with
old age. Knowledge acquisition rate of women exceeded that of men between the ages of 20 and 50,
suggesting that men gain ecological knowledge at a younger age and plateau earlier. Women, on the
other hand, appeared to begin acquisition of ecological knowledge at a later age, but then
accumulate it far more rapidly than men and up until a later age. Thus the mean age of knowledge
saturation for women was later than for men.

The relationship between the number of uses listed for a plant species and the proportion of the
population that recognised that particular species was examined to test if the usefulness of a species
had any bearings on its familiarity amongst locals. Plant familiarity was found to be positively
associated with its usefulness (total number of uses it holds) (Rs=0.720, p<0.001, n=48), with
species that had three or more uses in general being recognised by 40% of respondents or over (Fig
3.3.11).
88
Ecoliteracy (% total plants identified)
80
70
60
50
women
40
men
30
20
15-20yrs
21-50yrs
50+yrs
Age group
Figure 3.3.10. Comparing the changes in mean ecoliteracy

with age between women and men showing the maximum
age of knowledge saturation for both (mean + 2SE,
n=192). Symbol (*) indicates maximum age of knowledge
saturation
89

The effect of healer status on a persons depth of health knowledge was examined by comparing the
number of health uses listed for local plants by healers with those listed by non-healers. No
significant difference was found in level of health knowledge between the two groups.
Local species familiarity (% of

respondents that can identify each species)
90
100
A
A
A
A
80
60
A
A
A
A
A
40
20
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
0
0
Species usefulness (total

number of uses of each species)
Figure 3.3.11. Levels of species familiarity with local

villagers and relationship with the number of uses a
species holds (n=48)
91
3.4. Discussion

The result that ecoliteracy and total use knowledge did not vary between rural and suburban villages
was evidence against the relationship between ecological knowledge declining with reduced
distance and isolation from urban centres (Nolan & Robbins, 1999). Instead, it demonstrated the
effect of continued resource dependence (sustained at both locations) on ecological knowledge
acquisition. This is not evidence for the rejection of the hypothesis that local urbanisation causes a
loss of ecological knowledge, but that continued resource dependence is more important and can
override this effect and thus is a more powerful predictor of LEK level.
Rural communities relied upon the land and its resources due to having few alternative options,
whereas the suburban community of Sungureampatti had alternatives but could not afford to pursue
them since household income was low and often dependent upon the earnings of widowed women.
Hence local resources were harvested to subsidise diets and income wherever possible. Therefore,
the rural villages lacked purchasing access whereas Sungureampatti lacked purchasing power. For
this reason, villages in both locations retained dependence upon the land and its resources and thus
held high levels of ecological knowledge.
Despite similar ecological knowledge levels, suburban villagers knowledge of other plant uses was
significantly higher than rural villagers. The category of other uses included ornamental, religious,
cosmetic and sporting (such as using local plant products as the bat, ball and wicket in a game of
cricket). This demonstrated the importance of horizontal knowledge transfer in densely populated
urban/suburban regions. Sungureampatti was closely surrounded by many suburban villages and
urban residencies from which villagers could acquire additional knowledge of unique species uses
(on top of essential survival knowledge known by all) from knowledge bases that have developed
independently from their own. Such knowledge is unlikely to be shared with more remote, isolated,
rural communities. High suburban knowledge variation, therefore, was most likely indicative of the
differing opportunities of villagers to experience and interact with outsiders and share their
knowledge bases.
92

Indicative of reduced household income due to women being the sole wage earners in many
households, mean community wealth was significantly lower in the suburban village than the rural
villages. Large variation in rural wealth demonstrated intercommunity differences, particularly
between non-Dalit land owners and landless Dalit communities.
An inverse relationship was revealed between community wealth rank and ecoliteracy, thus the
wealthiest non-Dalit village of Kariakalkulum had the lowest levels of ecological knowledge. This is
evidence again for the importance of resource dependence upon ecological knowledge acquisition
(Gadgil et al., 2000; Pierotti & Wildcat, 2000; Pretty, 2002), supporting the hypothesis that
increased knowledge is held by communities with higher levels of resource dependence. Where
purchasing power was low through low income (reflected by wealth rank), communities were forced
to subsidise diets and income using local natural resources and, thus, were more likely to hold indepth knowledge of these resources as a tool for daily survival. Communities with market
purchasing power, on the other hand, can support themselves independent of local resources and,
therefore, have less need to acquire in-depth knowledge of local plant species and their uses. Large
variation around mean knowledge levels was most likely indicative of different households within a
community having differing levels of income and thus varied resource dependence levels.
Community knowledge variance being positively associated with wealth rank and inversely related
to knowledge level (both ecoliteracy and use knowledge) was evidence for the importance of
equitable knowledge transfer where high levels of resource dependence are prevalent. In
Kariakalkulum, where resource dependence was low and purchasing power high, it is not essential
for every person to have high levels of resource knowledge. Instead, certain individuals were
experts, most likely as a result of personal interest. On the other hand, in Sungureampatti, where
natural resources supplement lifestyles, it was most likely imperative that every community member
hold a high level of knowledge for use on a daily basis to exploit local resources in order to sustain
both themselves and their families.
Heightened variance in communities with low ecoliteracy levels demonstrated the importance of
sampling at random and including laypersons. Davis and Wagner (2003) stressed the importance of
93
identifying experts to focus upon in interviews when examining ecological knowledge. However,
these results show that experts exist even within less knowledgeable communities and, considering
the levels of LEK variance revealed, can give a misleading indication of community knowledge
levels if interviewed alone.

No differences were found between the knowledge levels of men and women. This is indicative of
the frequent ecosystem interaction of both groups despite differentiated societal roles. Thus this
result is not evidence against the importance of societal roles on knowledge level, but instead it is
likely that gender-differentiated roles in this region do not lead to differential levels of experience.
For instance, women combined responsibilities as agricultural labourers with looking after children
in the home, whereas men shared agricultural responsibilities with women in addition to being the
primary pastoralists. Although this increased mens time in contact with nature, in monotonous
landscapes where both men and women interact with local species on a daily basis, both were found
to hold high levels of LEK.
Although overall ecological knowledge levels were not gender-differentiated, mens pastoral
responsibilities were supported by significantly higher levels of farming use knowledge (including
livestock uses) than women. This is evidence for the importance of societal roles and responsibilities
in creating local knowledge experts resulting from increased experience of certain activities. Also,
the opportunities for oral transfer between individuals with shared roles (pastoralists in this case)
have the potential to further boost specialist knowledge. High variation in both men and womens
knowledge of agricultural plant uses was most likely indicative of livelihood differences, between
those who were just pastoralists, those that were crop farmers and those that pursued both farming
activities. Similar patterns of expertise were not revealed in knowledge of food or construction uses
of local plant species, indicating that both men and women were fully aware of these species uses,
most likely through personal experience (women cooked the meals, but men and women ate them;
men built the houses, but both women and men lived in them).
94

Unlike total ecoliteracy, the mean number of uses per recognised species significantly differed
between the two age groups, with adults able to name significantly more uses per recognised species
than children. This suggests that ecoliteracy (species naming ability), the primary level of local
knowledge, was being successfully passed on to future generations at a young age. Use knowledge,
on the other hand, was either passed on in its entirety above the age of 20 years or was being
progressively lost between generations (Douglas, 2002; Birchenough, 2005). The former is more
likely in resource dependent communities where primary ecological knowledge is passed on through
oral transfer in childhood and secondary level use knowledge is learned later in life through a
combination of oral transfer and personal experience.
However, the villages of the two locations have undergone substantial social and ecological changes
in the past decade that may inhibit secondary knowledge transfer. Widespread drought has caused
the loss of a number of local species as has the invasion of dry scrubby Prosopis plants as a
consequence of low rainfall (S. Rajeswari, personal communication, 21 July 2003). In addition,
formal education has been introduced within the last decade (J. Devavaram, personal
communication, 15 July 2003), relocating childhood experience from the fields into the classroom.
Formal schooling is known to heighten childrens cultural awareness and consequently heighten
aspirations, often resulting in future generations departing from the traditional lifestyles of their
parents to follow dreams of cosmopolitan lifestyles associated with urban living (Peat, 1996;
Ghimire & Pimbert, 1997; Rao et al., 2003).
These two factors combined means that the use knowledge of younger generations may never reach
that of their parents. Ecoliteracy may survive as the most primary knowledge level indicative of a
recent history of resource dependence, but as young people gain independence from their traditional
lifestyles, species use knowledge is likely to be progressively lost. This is a consequence of reduced
oral transfer and experience as a result of a decline in local resource dependence and the
introduction of formal schooling.
Of the three healer interviews carried out, two admitted to knowing substantially less than their
predecessors and all three said that they have not passed on their in-depth knowledge to their own
95
children. Reasons given for this included urban migration and a lack of interest shown by younger
generations. High variation in the mean number of uses listed per recognised species by both adults
and children was most likely indicative of differing levels of resource dependence of different
households and communities.
Knowledge of health (mostly medicinal) uses of local plant species was significantly higher amongst
adults than young people, most likely due to adults having greater experience in dealing with illness
(Garro, 1986). This is indicative of the importance of roles and responsibilities in gaining access to
health knowledge. Most couples in this region marry between the ages of 18 and 20 and soon after
start a family (J. Devavaram, personal communication, 15 July 2003). Local knowledge of
medicinal species is a necessity to parenting in a region lacking faith in and access to modern
medicine. Hence health knowledge is orally transferred (as a person is unlikely to be experimental
with the healthcare of their child) in bulk to young adults, particularly women, as they become new
parents and their dependence upon local plants to cure common childhood ailments emerges.
Overall, elder womens medicinal knowledge was found to exceed mens, similar to the results of a
study by Singh et al. (2006b). Adult womens in-depth health knowledge (Fig 9) demonstrated the
importance of their primary household roles of raising children and caring for family members to
knowledge acquisition (Garro, 1986; Bryson, 1996; Alexiades, 1999; Turner et al., 2000; Pfeiffer,
2002; Westermann et al., 2005). This supports the hypothesis that societal roles play a key part in
knowledge content (Warren, 1997).
Adult womens medicinal knowledge was greater than adult mens, just as young womens
knowledge was greater than young mens. This suggested that one of two knowledge transfer
processes is taking place. Either elder women, the primary holders of health knowledge, only pass
small quantities of this knowledge on to other societal groups and retain the majority of the
knowledge themselves or a horizontal transfer barrier exists. That is elder women transfer their
health knowledge to young women and elder men to young men, but no health knowledge transfer
occurs between the sexes. Therefore, mens health knowledge will never reach the level of womens
and individual health knowledge levels will only peak in later life, due to dependence upon oral
transfer.
96
High health knowledge variation amongst adult women was most likely indicative of differing levels
of resource dependence between different communities and households. For instance, some poorer
households are likely to be more reliant upon grandmother treatments than others that can afford
hospital fees of up to Rs25, 000 for admission alone (J. Jebamalaidass, personal communication, 26
July 2006). Grandmother treatments are the use of local plant species in medicines to treat common
ailments and childhood diseases, knowledge of which are held primarily by elder women. Also,
females with healers as husbands or fathers may have had enhanced health knowledge levels
through combining knowledge of specialist healer treatments with knowledge of traditional
grandmother remedies.
The different pattern of LEK acquisition with age between men and women was further evidence for
the importance of oral health knowledge transfer to women later in life. Men most likely gain orallytransferred knowledge combined with experiential knowledge early in life (many boys are
pastoralists of the family herd from a young age). Hence they acquire most of their knowledge at a
young age and gain little more through daily interactions due to the monotonous landscape of the
region.
Women, on the other hand, help with family and household chores from a young age, but lack the
experience-boosted knowledge that young men acquire. However, as women become independent of
their families through marriage and childbirth, many gain environmental experience through helping
their husbands in the field. This is in addition to acquiring orally-transferred health knowledge
through relationships with other women (Westermann et al., 2005). This explains why womens
knowledge was found to saturate at a later age than mens (maximum of 50 years for women
compared with 20 years for men). High variation in the ecoliteracy levels of both men and women
was most likely reflective of differences in resource dependence between different households and
communities based on income.

The positive relationship between the number of locals able to recognise a species and the number
of uses that particular species held was indicative of the importance of resource dependence upon
97
the ecological knowledge of this region. The more useful a species was to local residents in daily
life, the more frequently that particular plant was most likely talked about and knowledge of it
transferred. Thus a key feature of resource dependent communities is likely to be increased
familiarity with the most harvested local species (Gilchrist et al., 2005). This is further evidence in
support of the key role resource dependence plays in ecological knowledge transfer.

No differences detected between healers and non-healers (laypersons) health knowledge was
indicative of continued local reliance upon traditional grandmother remedies (J. Jebamalaidass,
personal communication, 26 July 2006). The respected status of healers in this region was derived
from their ability to treat ailments that cannot be treated within the home by women. Therefore,
healer knowledge was in-depth and focused upon treating serious disorders and ailments. However,
laypersons (usually women) were found to hold an equally important, but different, form of health
knowledge focused upon treating common ailments, especially those encountered in childhood.
This investigation has yielded quantifiable evidence for the hypothesised importance of resource
dependence and societal roles on ecological knowledge within and between communities and
revealed that informant characteristics affect the distribution of knowledge, that is some people
know more through sharing more and living in certain situations (Garro, 1986). To investigate
distribution patterns further and their relationship with current and historical societal roles,
knowledge distribution in non-resource dependent communities with a history of resource reliance
will be examined. Therefore, three developed communities of the UK will be investigated in the
next chapter to see if past patterns of resource reliance and gender-differentiated roles continue to
affect knowledge levels today or if other factors come into play where natural resource dependence
is no more.
98
Chapter 4. UK Case Study: Comparing rural Lincolnshire, urban Kent and coastal Essex
4.1. Introduction
The UK was the most developed of the three study sites, with expansive urban centres drawing in
populations seeking cosmopolitan jobs, city pay and lifestyles. Although no area of the UK has gone
unaffected by urban development, certain areas have been affected more than others. Some remote
rural villages have remained somewhat connected with the land far more than their urban
counterparts, despite a common rural history. Increased accessibility in the UK permitted intracountry comparisons to be made based on differing levels of urban development.
A total of three study sites were chosen in the UK with high accessibility, different landscapes and
ecology, differing levels of urban development and disconnection from the land and smaller
residential divisions (for stratified sampling). Rural villages of the East Midlands were compared
with urban wards of the southeast to examine the impacts of urbanisation and economic
development on ecological knowledge and disconnection from the land. Coastal towns of East
Anglia were also studied as an intermediary and to compare local marine knowledge with terrestrial
(Table 4.1.1) (Fig 4.1.1).
Table 4.1.1. Site characteristics of the 3 areas selected for study within the UK
County
Lincolnshire
Landscape
Rural
Population (in 2001)a
646645
Area (in square km)
6000b
Population density (per
108
square km)
GDP per head pa (in
10,751e
1998)
Characteristic industry
Agriculture
a
figures from ONS (2001)
b
figures from AtUK (2005)
c
figures from GenUKI Kent (2004)
d
figures from Essex County Council (2005)
e
figures from EMDA (2004)
f
figures from Kent County Council Bulletins (2004a)
Kent
Urban
1,329,718
3085c
431
Essex
Coastal
1,310,922
3670d
357
11,621f
11,640d
Urban
Maritime
99
Lincolnshire
Essex
Kent
Figure 4.1.1. Map of England showing the location of the study sites within the 3 selected counties
(Multimap, 2006, 1:4,000,000)
4.1.1. The Lincolnshire study site

The East Midlands covers 12% of Englands total land area and holds 7% of the total UK
population. Within it are a vast number of cities, modern towns, market towns and much smaller
settlements. As many as 36% of the regions population (1.5 million people) inhabit rural areas. This
figure exceeds the national average (EMDA, 2004). Lincolnshire is a county within this region that
covers almost 6000 square kilometres (AtUK, 2005). It is Englands second largest county and is
bordered by Yorkshire to the north from which it is separated by the Humber, Northamptonshire,
Cambridgeshire and Norfolk to the south, Nottinghamshire, Leicestershire and Rutland to the west,
and the North Sea coastline to the east (Thirsk, 1957; GenUKI Lincolnshire, 2004).
Despite its large land mass, Lincolnshire is thinly populated with a mean density of 108 people per
square kilometre of land (compared with a mean value of 383 people per square kilometre for
100
England as a whole (ONS, 2005a)) and a total population of 646,645 (ONS Lincolnshire, 2001;
EMDA, 2004; AtUK, 2005). Reflective of the East Midlands as a whole, Lincolnshires population
of elder residents slightly exceeds the national average whereas its population of younger people
falls slightly below (EMDA, 2004).
Lincoln city is the county town of Lincolnshire and enjoys a drier sunnier climate than most of the
UK, due to its eastern location just south of the countrys midpoint (EMDA, 2004; AtUK, 2005). Its
climate, combined with low land relief and coastal border create a diverse array of landscapes within
the county, combining Wolds and heath uplands, with fens, marshes and sand dunes (Thirsk, 1957;
EMDA, 2004). Despite this diversity, as much as 87% of Lincolnshires vast land area remains
under agricultural production (EMDA, 2004).
Traditional crops of the county include oats, wheat, clover, beans, turnips and barley. Traditional
practices employed in the region, many of which are being re-introduced today, include crop
rotations, leaving set-aside and using natural fertilisers (Young, 1970). Lincolnshire has been
renowned for its high quality livestock and the success of its selective breeding programmes over
the generations. It is not rare for farmers to know the entire family tree of their herd (Young, 1970).
Some local livestock farmers have even developed their own local veterinary practices including the
self-diagnosis of animal diseases, such as the rot, and treatments for them (for instance by using
common ground-sel to cure the greasy heels of a horse) (Young, 1970).
Traditionally, the county was divided into three subdivisions; the hills and woodlands of Kesteven in
the southwest, the fens of Holland to the south and, the area that this study focuses on, the northern
county division of Lindsey dominated by open farmland (AtUK, 2005). The agricultural revolution
saw extensive woodland clearance and drainage of fens. Forests became tracts of grazed grassland,
wetlands were developed into dry cropping lands and an occupational shift emerged from geese
rearers, fishers and wildfowl catchers to ploughmen and farm labourers (Beastall, 1978; EMDA,
2004). Increased flooding due to a reduction in land level and enclosures amalgamating lands
(reducing the number of smallholdings) created social problems within the county, further
increasing the already wide gap between the wealthy and poor (Beastall, 1978).
101
Even today Lincolnshire remains heavily dependent upon its farmland, with two-thirds of the land
area continuing to be tilled (EMDA, 2004; AtUK, 2005). Thus arable farming is predominant and
highly intensive across the county and pasture lands have fallen over the last 50 years (EMDA,
2004). Despite this, employment in agriculture in the region has declined by approximately 25% in
the decade between 1981 and 1991 (EMDA, 2004) and the workforce of the county (approximately
200,000) now falls mostly into the service sector (129,000 people) and manufacturing (44,000
people), with only a proportion of the remainder surviving in the food sector (EMDA, 2004).
In 1998, the GDP per head across the county averaged at 10,751 pa. This fell below both regional
(11,848 pa) and national (12,548 pa) averages (EMDA, 2004) and is reflected by the low level of
rural earnings which fall below equivalent work in urban areas, as well as behind regional and
national averages for rural employment (EMDA, 2004). The result is a large population of
commuters; 57% of the population of rural districts of Lincolnshire now travelling outside of their
home district for work (7% higher than the regions average) (EMDA, 2004).
Changes in traditional agricultural practices have resulted in species extinctions within the county.
Changes such as a decline in untilled land, increased field sizes, widespread chemical application
and a loss of grasslands have caused some frequently sighted species to be lost including the
lapwing and the Green woodpecker (EMDA, 2004). Hedgerows, verges and ditches have been lost
at a greater rate in Lincolnshire than across the rest of the UK (EMDA, 2004), hence designations
where species diversity still prevails are important to maintaining the natural landscapes of the
county.
This study focused on four rural villages (Kexby, Upton, Willingham and Marton) near the town of
Gainsborough, that is situated in the district of West Lindsey and characterised by monotonous
landscapes of flat arable fields (Fig 4.1.2). Farming and its associated trades dominate the area and,
traditionally, farmland tenancies were passed between generations of family members, as was indepth local knowledge (Rose, 2000; AtUK, 2005). The chosen villages, once traditional farming
settlements, remain surrounded by farmland although much of the traditional woodland has gone.
That which remains is used by local farmers to keep pheasants (Rose, 2000).
102
Rural study site
Figure 4.1.2. Map of Lincolnshire County indicating the location of the rural study site (Multimap,
2006, 1:200,000)
Modern development is both limited and relatively recent within the selected rural villages. Large
farms, machinery, new housing and even electricity have only been introduced within the last 100
years. During this time there has been a loss in smallholder farms, hedgerows and traditional farm
workers combined with the closure of village schools and chapels. Consequently, economic
development and improved transport have altered rural village life from what it once was (Rose,
2000).
4.1.2. The Kent study site

A well-developed urban area of the southeast was selected for comparison with rural Lincolnshire.
The southeast is also traditionally rural in its roots, but has developed at a far greater rate than other
areas due to its location so close to the capital. As London has developed, the once rural towns of
the southeast have simultaneously developed in infrastructure, transport, employment status and
income. As a result dense populations prevail, with over a quarter of the UKs population
103
(constituting one-third of Englands population as a whole) resident in the southeast (DEFRA,

2004). Of this population, up to 70% inhabit the 18 principal urban towns and cities, thus housing
pressure is intense (Kent County Council, 2004).
Historically Kent is a maritime rural county, situated at the southeast extremity of the UK. Thus the
east and the south of the county are bordered by sea, the remainder of the south is bordered by
Sussex, the west border lies on the boundary of Surrey and to the north Kent is outlined by the
Thames (GenUKI Kent, 2004). Kent held a total population of 1,329,718 in 2001, with a mean
density of 431 inhabitants per square kilometre (ONS Kent, 2001; KENTnet, 2005). Covering an
area of 3,085 square kilometres, Kent is Englands ninth largest county incorporating a variety of
landscapes from coastline and marshland to countryside and urban landscapes (GenUKI Kent,
2004).
Southeast Englands physical diversity owes to its topography and climate. Kent was traditionally a
farming county, long-described as the Garden of England and known for its horticulture, hop
production and paper-making (supported by vast areas of ancient woodland) (GenUKI Kent, 2004;
KENTnet, 2005). Despite intensive industrialisation, 61% of the land area of the southeast remains
under agricultural production today (falling 19% short of the national average). Another 15% of the
land area remains as woodland and 23% of the landscape is dominated by infrastructure. This means
that the southeast is the most developed area of the UK outside of London, reflected by only 1.64%
of the countys workforce currently in the agriculture, hunting and forestry industry (Kent County
Council, 2003; DEFRA, 2004).
Kent has a complex transport network in both road and rail with particularly strong connections to
London (DEFRA, 2004). Reflecting high levels of development and transport links with London,
the economic performance of the southeast is closely linked with the capital. Approximately 18% of
the citys jobs are filled by commuters from the east and southeast (comprising of over 500,000
jobs) (DEFRA, 2004). Despite this, the average household earnings of Kent are among the lowest in
the southeast as a result of continued economic differentiation between rural and urban areas
(DEFRA, 2004; Kent County Council Bulletins, 2004b). The GDP of Kent per head in 1998 was
104
estimated at 11,621 pa compared with the national average of 12, 548 (Kent County Council
Bulletins, 2004a).
Kents countryside has been under increasing pressure over the past 50 years from agricultural
changes and development demands including road building and infrastructure development (Kent
County Council, 2004). Despite this, Kent has managed to retain a great deal of its rural landscape
to date through protective legislation. This study focuses on an area situated in the north of the
county, closely connected with London and heavily urbanised. Remaining tracts of land previously
farmed or used for pasture are protected by the Green Belt. Four urban wards were chosen; Bromley
Common and Keston, Crystal Palace, Hayes and Coney Hall, and West Wickham. All have easy
access to London, residential landscapes, urbanised industries, high population densities and good
transport networks (Fig 4.1.3).
Urban study site
Figure 4.1.3. Map of Kent County indicating the location of the urban study site (Multimap, 2006,
1:2,000,000)
105
4.1.3. The Essex study site

Essex is a county of East Anglia bordered by Cambridge and Suffolk to the north separated from by
the River Stour, Kent to the south separated by the Thames, Middlesex and Hertfordshire to the west
and the North Sea to the east. It is ranked as Englands eleventh largest county covering 3670 square
kilometres. Essex is ranked as the sixth most populous county in England with a population of
1,310,922 in 2001. This equates to a population density of 357 residents per square kilometre,
although this varies greatly between locations (ONS Essex, 2001; Answers Corporation, 2005;
Essex County Council, 2005).
North of the border of outer London and outside of the county town of Chelmsford, Essex
residential settlements comprise chiefly of small towns, villages and hamlets. Much of the land area
(up to 73%) is rural as a consequence of human cultivation over centuries (Answers Corporation,
2005). In 2001, two-thirds of this land was still tilled for crops, although livestock grazing has
declined (Answers Corporation, 2005). Coastal towns, on the other hand, focus more on maritime
industries, including oyster fisheries that have been of great importance to coastal communities of
this area over a long period (Answers Corporation, 2005). The tourist industry of Essex is also
heavily concentrated around the coastal towns of the region and worth an estimated 392 million in
2001 (Essex County Council, 2005).
Essex had an estimated GDP of 15,056 million in 1998, amounting to 11,640 GDP per head. This
figure is greatly affected by the high frequency of out-commuting into London. In 1991, as many as
25% of workers (170,000 people) commuted outside of the county for work, the majority headed for
London (Essex County Council, 2005). Due to its proximity to the capital, commuting is a feasible
and appealing option to those wanting to combine the countryside of Essex with London wages and
job opportunities.
Like Kent, the London border of Essex is protected from further urban development by the Green
Belt. However, the most important habitat type within the county in terms of wildlife conservation is
its coastal zone, stretching to a length of over 300 miles and comprising the largest habitat type
within the county. It supports a rich variety of wildlife and is of national and international
106
importance as bird feeding grounds, supporting over 182,000 birds in winter including the little tern
ranked as having European importance (European Marine Sites, 2005).
In addition to having ecological and landscape value, the coastal estuaries of Essex have held social
and amenity value to local residents throughout history (European Marine Sites, 2005). For the
purposes of this study, three towns along the coastline of Essex were selected; West Mersea, Frinton
and Brightlingsea, all of which sustain working marinas (Fig 4.1.4). The towns are all estuarine or
coastal in location and habitat, have a long-working history of maritime activities including social
networks based upon maritime pursuits and still gain a portion of their income from such activities
today.
Coastal study site
Figure 4.1.4. Map of Essex County indicating the location of the coastal study site (Multimap, 2006,
1:1,000,000)
107

As described, four villages were identified for study in a rural region of Lincolnshire. All villages
were primarily rural with a proportion of villagers retaining land-based livelihoods. The major land
use of the area was arable farming and local histories were centred on family farming. All villages
chosen were situated a minimum of five miles from a major town. Selected villages had a maximum
population size of around 500 and maintained networks including Young Farmers groups (Table
4.2.1).
Table 4.2.1. Rural villages key characteristics and attributes
Village
Kexby
Marton
Upton
Willingham
325
508
386
501
Distance from town
5.0
5.2
5.7
6.0
(miles)b
Agricultural-based
Young Farmers
Young Farmers
Young Farmers
Young Farmers
networks
(Corringham)
(Corringham)
(Corringham)
(Corringham)
Schools
None
None
None
None
a
figures based on 1991 estimates from Travellers World (2004) adjusted by 4% according to county
population changes over the decade (ONS Lincolnshire, 2001)
b
figures from AA Route Finder (2005)
The four wards of urban Kent, bordering London, were selected for study based on a number of
criteria including having built-up residential landscapes with a minimum population size of 10,000
(defined by The Office of the Deputy Prime Minister (ODPM) as the definition of an urban area)
(ONS, 2005b). All were situated within the borders of outer London. All wards had good transport
links and, therefore, easy commuter access to the capital (travel time within one hour) and were
situated on or near the green belt (Table 4.2.2). Therefore, remnants of rural land persisted at the
sites in the absence of the livelihoods that created them.
108
Table 4.2.2. Urban wards key characteristics and attributes

Ward
Distance from central
London (miles)b
Travel time via road to
central London
(minutes)b
Public transport to
London
Schools
a
Bromley
Common and
Keston
14,150
11.2
Crystal Palace
Hayes and
Coney Hall
West Wickham
11,250
7.6
14,900
14.5
14,300
13.2
46
33
44
59
Train
Train and bus
Train
Train
Primary and
secondary
Primary and
secondary
Primary and
secondary
Primary
figures from CCSR (2004)

figures from AA Route Finder (2005)
In East Anglia, the three coastal towns selected for study were chosen for their positions on the
north Essex coastline. All towns required an intermediary level of development (exceeding that of
rural Lincolnshire but falling short of outer London development levels) and had to sustain some
level of income from traditional maritime industries for selection. All three also retained some social
groups based on their maritime roots and activities (Table 4.2.3).
Table 4.2.3. Coastal towns key characteristics and attributes
Town
Brightlingsea
Frinton
West Mersea
Population (in
8146a
23,056b
6600c
2001)
Traditional
Fisheries (including
Coastal resort
Oyster fishery
maritime
oyster)
Sea fishing
Shipbuilding
industries
Shipbuilding
Sail-makers
Maritime
Sailing
Yacht
Sailing
clubs
Yacht
Yacht
Schools
Primary and secondary
Primary and secondary
Primary
a
figures from Tendring District Council (2004)
b
figures from Newspaper Society database (2004) adjusted to incorporate 0-14 year olds based upon census
age group proportions for the county (ONS Essex, 2001)
c
figures from Town Guides (2004)

Local species lists for the UK study sites included plants and animals classified as common or
abundant across the whole of the UK. In addition to wild flowering plants, mammals, birds, crop
plants and coastal species (exclusively for use at the coastal sites) were included. The 18 most
109
common mammals and 20 most common bird species found across the UK according to DEFRA
statistics (2004) were used. A full list of wild flowers identified as common or abundant all over the
UK was compiled initially using Roses wild flower key (1981). From this, species found primarily
in rural habitats were eliminated due to the reduced likelihood of them being found at the urban site,
as were species with highly-specific soil requirements, such as those needing moist or acidic
conditions. Using these criteria 16 species were finally chosen based on their abundance and affinity
to most habitat types. In addition to wild plants, a selection of 13 commonly grown crop plants was
compiled using DEFRAs National Crop Listing (2001).
Images of coastal plants and animals were included to assess if local knowledge of economicallyimportant marine species exceeded that of terrestrial species in traditional maritime communities. A
selection of common coastal species was compiled based on the British Marine Life Study (2004).
This list was verified by Dr David Smith (lecturer in marine biology at the University of Essex) who
added any extra species known to be economically important along the East Anglian coastline. The
final list included 19 coastal species. Full UK species lists including Latin and vernacular names are
given in Appendix III.
Internet search engines were used to collate high quality images of the 86 different species. Only
images of highest quality were used in the identification survey. These were verified by 25 peers not
included in the study. Species cards were created labeling each species by number. Where
necessary, up to three images were used to represent all the different components of a species, for
instance showing the fruit of a plant and/or the leaf.

Calculation of sample sizes varied at the three sites due to the exceedingly large population sizes at
the urban and coastal sites in comparison with the rural site. However, at all sites target quotas were
calculated based upon gaining statistically-viable, representative samples of both men and women
from four different age groups.
Age groups were chosen based on attempting to minimise within group knowledge variance. The
four age groups selected were 15 to 19 years, 20 to 29 years, 30 to 49 years, and 50 years and over.
110
These age ranges represent different life stages, each one with a potentially different source of
knowledge acquisition. Thus it is possible for a person to hold a similar depth of ecological
knowledge at the upper end of an age group as they did at the lower end. For instance, 15-19 year
olds are representative of the life stage between finishing school and starting a career or pursuing
higher education, thus ecological knowledge acquired is primarily from schooling and childhood
experience.
The 20 to 29 year age group is representative of a different life stage in which people typically settle
down with partners, thus ecological knowledge during this period will be influenced by the interests
of his/her partner and outdoor pursuits. From 30 to 49 years, work and raising a family are usually a
priority. Experience is key to knowledge acquisition during this time of life. Finally, individuals
aged 50 and above are usually key to knowledge sharing, often with minimal acquisition. This is not
to say that people ever stop learning, but that a theoretical plateau is reached whereby the majority
of experience-based knowledge has already been learned by this age.
Due to the small population sizes of the rural villages, it was possible to interview 10% of the total
populations (Roscoe, 1975). Sample quotas were calculated firstly by attaining total population
estimates for each of the four villages for 1991 (Travellers World, 2004). These figures were
amended (by a reduction of 4%) to account for demographic changes across the whole of
Lincolnshire over the decade. Then, based on Lincolnshire census data, the proportion of men and
women that fell into the chosen age groups could be calculated for Lincolnshire county as a whole
(ONS Lincolnshire, 2001). These figures could then be applied to the total population size of each of
the four villages in order to calculate the expected number of men and women in each age group in
each village at the time of study.
From this, it was possible to calculate the number of respondents needed for interview from each
subpopulation to meet the target quota of 10%, according to Roscoes rules of thumb (1975). Where
the sub-sample quota fell well below 30 individuals, 20% of the population was interviewed
(applied to age groups 15-19 and 20-29) (Roscoe, 1975). A total of 178 interviews were carried out
at the rural site, exceeding 10% of the total population aged 15 and above (Table 4.2.4).
111
Table 4.2.4. Rural sampling strategy showing minimum target quotas for all sub-samples
Age group
Gender
Population
15 - 19
M
F
20 - 29
M
F
30 - 49
M
F
50 +
M
F
Kexby
Actual
Sampled
10
2
9
2
17
4
16
4
34
3
45
5
59
6
76
8
Marton
Actual
Sampled
16
4
14
2
26
6
26
6
69
7
70
7
92
9
118
12
Upton
Actual
Sampled
12
2
11
2
20
4
19
4
52
5
54
5
70
7
90
9
Willingham
Actual
Sampled
15
4
14
2
26
6
25
6
68
7
69
7
90
9
117
12
Minimum sampled
12
20
20
22
24
31
41
The population sizes of the selected urban wards were too large to sample 10% with time and
resources constraints (1 interviewer) (ONS Kent, 2001). Therefore, sample sizes were calculated
based on attaining a minimum of 30 respondents from each sub-population ensuring viability for
statistical testing (Roscoe, 1975). To ensure representative sampling, age groups and genders were
sampled at ratios that reflected their proportions in the actual population. For instance, if 8% of the
population aged 15 and upwards comprised women aged 20 to 29, then 8% of the total sampling
quota would comprise women in this age group.
Firstly, the proportion of the countys population represented by each age group (aged 15 and
above) was calculated (ONS Kent, 2001). The smallest sub-population (15 to 19 year olds) was
represented by the minimum number viable for statistical sampling (30 respondents). The other subpopulation sampling quotas were calculated as ratios of this figure according to the proportions of
the total population that they constitute. For instance, if there are twice as many 20 to 29 year olds as
there are 15 to 19 year olds then the ratio is 2:1, and the sampling quota for 20 to 29 year olds
calculated to be 60. This method resulted in an extremely large total sampling quota impossible to
fill with time constraints. Therefore, the three groups with sample sizes greatly exceeding 30 were
halved (ensuring that all sub-samples still contained a minimum of 30 respondents). This gave a
minimum total sample size of 241. However, in total, 287 interviews were carried out at the urban
site.
112
Once the total target number of respondents for sampling had been calculated, the number to be
sampled within each of the four wards was designated. This was calculated based on their relative
population size in proportion to one another (CCSR, 2004). For instance, the largest numbers of
interviews were carried out within the largest ward. Once ward sampling quotas had been
designated, sub-population sizes to be sampled within each ward were calculated based upon actual
sub-population proportions for the county (ONS Kent, 2001). For instance, if 35% of the countys
population (over 15 years) comprised people aged 30 to 49 years, then 35% of each wards sampling
quota was made up of this age group. Finally, gender ratios were calculated within each age group.
For example, if the countys gender ratio for a particular age group was 51% women to 49% men,
then the sampling quota for that age group in each ward was divided according to this ratio.
Therefore, if 23 people in a ward were to be interviewed within that age group, 12 would be women
and 11 men (Table 4.4.5).
Table 4.2.5. Urban sampling strategy showing minimum target quotas for all sub-samples
Age group
Gender
Population
15 - 19
M
F
20 - 29
M
F
30 - 49
M
F
50 +
M
Bromley
Common and
Keston
Actual
Sampled
440
4
424
4
764
4
785
4
1947
11
2015
11
2314
11
2752
13
Crystal Palace
Actual
Sampled
350
3
337
3
607
3
624
4
1548
8
1602
9
1839
9
2188
11
Hayes and
Coney Hall
Actual
Sampled
463
4
447
4
805
4
827
5
2050
11
2122
12
2436
12
2898
15
West
Wickham
Actual
Sampled
445
4
429
4
772
4
794
4
1968
11
2036
11
2338
11
2781
13
15
15
15
17
41
43
43
52
Minimum sampled
The same sampling strategy employed at the urban site was used at the coastal study site, where
actual town populations were again too large to sample 10%. Based on these calculations, the
minimum number of interviews to be carried out at the coastal site was 227. However, in total 246
respondents were interviewed (Table 4.2.6).
113
Table 4.2.6. Coastal sampling strategy showing minimum target quotas for all sub-samples
Age group
Gender
Population
15 - 19
M
F
20 - 29
M
F
30 - 49
M
F
50 +
M
Brightlingsea
Actual
Sampled
244
3
226
3
468
3
455
3
1153
8
1179
8
1329
9
1563
10
Frinton
Actual
Sampled
692
10
641
9
1326
10
1287
9
3265
23
3336
23
3760
26
4424
31
West Mersea
Actual
Sampled
198
3
183
2
379
3
368
2
935
6
955
7
1076
7
1267
9
16
14
16
14
37
38
42
50
Minimum sampled
Haphazard sampling methods including door-knocking and chain referrals or snowball sampling
were used (Huntington, 2000; Brodt, 2002; Davis & Wagner, 2003). Haphazard sampling is a useful
way of identifying respondents at random to fill pre-designated quotas. Interviewees were identified
by randomized door knocking, visiting public places (including high streets, schools and public
libraries) and attending meetings of local clubs (such as sports clubs, Womens Institutes and local
branches of the Royal British Legion). Interviews were carried every day of the week at all times of
day to ensure that no groups (e.g. full-time workers) were eliminated. UK research was carried out
between April 2004 and December 2005. All respondents were interviewed on an individual basis.

A standard introduction was given to all respondents introducing the interviewer and the supporting
institution. A brief explanation of what was expected of the respondent was given and the aims of
the study were described as researching local plants and animals of the area and their local names.
Emphasising that the study was an assessment of knowledge was avoided for fear of eliminating
respondents less confident in their own ecological knowledge.
Initially, a respondents personal demographics, including age, gender, occupation, village/ward of
residence, length of residence and environment of childhood (rural, coastal, urban or traveled) were
recorded. The species identification task followed. Species recognition (whether or not the
respondent recognised the species), in addition to species identification (ability to name the species),
was recorded. Also, the respondent was asked to name any known uses. This ensured that the most
114
basic levels of knowledge were detected (recognition) and could be compared with more in-depth
knowledge bases (uses known) in a country heavily urbanised and independent of its natural
resources.
Vernacular, genus and family names were all accepted as correct, for instance a kidney vetch could
be described as vetch, since this study aims to assess community knowledge held by the general
public, rather than expert knowledge. Where unusual vernacular names were given, books and the
internet were used as reference points to assess accuracy (Mabey, 1996). The respondents were
shown one species card at a time and data were recorded on a data recording sheet included in
Appendix VI.
Closed questions followed assessing each respondents frequency of countryside visits, reasons for
visits, income level, knowledge source (respondents were also asked to estimate the % contribution
from each source) and view of the importance of environmental knowledge (respondents were asked
which of five statements they most agree with). Interviews lasted between 15 and 60 minutes and
were piloted at both rural and urban study sites prior to testing. Piloting enabled the rephrasing of
ambiguous questions and closed question categories to be verified and amended where necessary to
avoid forcing the respondents answers into etic categories.

The data were compiled in a database constructed using SPSS version 12.0, enabling detection of
inconsistencies, database manipulation and statistical testing. All data were tested for level of
variance (calculating coefficients of variance) and normality (Kolmogorov-Smirnov test). No
variables were found to produce a normal distribution and, therefore, non-parametric statistical tests
were employed.
The knowledge levels (% identification, total uses listed and % of species for which uses were
identified) and variations in knowledge (coefficients of variation) of respondents from different
sites, age groups, income brackets, occupations and childhood environments were compared using
Kruskall-Wallis. Since few respondents were willing to give an indication of their household
income, income levels were calculated based upon scoring job titles according to the Standard
115
Occupation Classification for the UK (SOC, 2000) and then using the Annual Survey of Hours and
Earnings (ASHE, 2005) to estimate average earnings.
Kruskall-Wallis was also used to test the effects of frequency of countryside visits, reasons for visits
and number of years lived at a site on knowledge level. Differences in ecoliteracy, variations in
knowledge (coefficients of variation) and knowledge source contribution between men and women
were tested using Mann Whitney-U. However, differences in knowledge sources between
respondents from different sites and age groups were tested using Kruskall Wallis.
Spearmans rank was used to test for associations between (i) source contribution and knowledge
level, (ii) knowledge level and view of ecoliteracy importance and (iii) knowledge source
contribution and view of ecoliteracy importance. Associations between knowledge level and quality
of local environmental experience were also tested for using Spearmans rank. Quality of local
experience was calculated based on a scoring system between one (lowest quality of environmental
experience) and 20 (highest quality of environmental experience). Respondents could score between
one and four points for each variable assessed; how frequently they visited the countryside, their
reason for visiting, where they live now, how long they have lived there for and their childhood
environment. Therefore, people that have lived in a rural area for a long period, visit the countryside
frequently for purposes that involve being out of doors and that grew up in a rural area were ranked
as having the highest quality of local environmental experience. However, people living in urban
areas, having grown up in a town or city, who only pass through the countryside on the rare
occasion had the lowest quality of experience (Ryan, 2005).
4.3. Results

The effect of site on ecological knowledge was examined by comparing mean ecoliteracy levels
between rural, urban and coastal residents. Total ecoliteracy was found to be significantly different
between the three sites (H=46.136, df=2, p<0.001, n=711) with rural inhabitants being the most
knowledgeable (naming 51% of local species), coastal the least knowledgeable (naming 42% of
local species) and urban residents with intermediate levels of knowledge (naming 44% of local
116
species). A similar pattern was attained when considering species use knowledge. Significant
differences between the total number of uses identified (H=103.764, df=2, p<0.001, n=711) and the
proportion of local species for which uses were identified (H=124.052, df=2, p<0.001, n=711) were
found between the three sites (Fig 4.3.1).
Upon further analysis, it appears that paucity in the marine species knowledge of coastal residents
lowered their mean knowledge levels. If considering knowledge of local terrestrial species alone
(excluding the marine species identification data of coastal residents), rural residents remain the
most knowledgeable, coastal residents have intermediate levels of knowledge and urban have the
least (H=33.067, df=2, p<0.001, n=711) (Fig 4.3.2). A significant difference was found between the
mean level of terrestrial knowledge (naming 48% of local species) and marine species knowledge
(naming 24% of local species) of coastal respondents (U=8159.000, p<0.001, n=246).
Mammals were found to be the most identifiable species group with over 77% of the species
correctly identified on average. Birds and crops were the next most identifiable groups with an
average of 43% and 41% being identified respectively. Flowers and marine species were the least
correctly identified with only 25% and 24% identified respectively. Therefore, by including marine
identification in overall ecoliteracy scores, the mean knowledge level of coastal residents was
reduced (Fig 4.3.3).
Respondents from the three different sites significantly differed in their frequency of countryside
visits (H=120.815, df=2, p<0.001, n=711). It was found that rural residents and coastal residents
visit the countryside more frequently (on average weekly) than urban-dwellers who visited on
average monthly. No significant differences in knowledge variation between the three sites were
observed.

Gender differences in ecological knowledge at the three sites were examined by comparing the
ecoliteracy of men and women from all age groups. At all sites some gender differences in
knowledge were found. At the rural study site, despite overall ecoliteracy levels of men and women
not being found to be significantly different, knowledge content was found to vary. Men were able
117
Ecoliteracy (% total species identified)
100
A
80
A
A
60
40
20
A
0
rural
urban
coastal
Site
Use knowledge (number of uses listed)
40
S
30
A
A
A
A
20
A
A
A
10
rural
urban
coastal
Site
Use knowledge (% species
for which uses were identified)
40
C
S
S
30
A
A
20
10
0
rural
urban
coastal
Site
Figure 4.3.1. Comparing UK site differences in (A) level

of total ecoliteracy, (B) use knowledge (total number of
uses) and (C) use knowledge (% species for which uses
were identified) (median + interquartile range, n=711).
Symbol (o) indicates outliers, symbol (*) indicates
extreme outliers
118
Terrestrial ecoliteracy (% terrestrial

species identified)
100
A
A
A
80
A
A
60
40
20
A
0
rural
urban
coastal
Site
Figure 4.3.2. Comparing UK site differences in level of

terrestrial ecoliteracy (median + interquartile range,
% Identification (mean % of species that

respondents were able to identify in each group)
n=711). Symbol (o) indicates outliers
100
80
60
40
20
mammals
birds
flowers
crops
marine
Species group
Figure 4.3.3. Comparing the mean % of species that

respondents were able to identify in the different species
groups (mean + 2SE, n=711)
Mammal ecoliteracy (% mammals identified)
119
100
A
80
A
60
S
S
S
S
40
20
0
women
men
Flower ecoliteracy (% flowers identified)
Gender
100
A
B
A
80
A
A
A
A
60
40
20
0
women
men
Gender
Figure 4.3.4. Comparing rural gender differences in (A)

level of mammal ecoliteracy and (B) level of flower
ecoliteracy (median + interquartile range, n=178). Symbol
(o) indicates outliers, symbol (*) indicates extreme outliers
120
to identify significantly more mammal species (U=3222.000, p<0.05, n=178) and women more
flowering plants (U=3146.500, p<0.05, n=178) (Fig 4.3.4).
At the urban site, women were found to have higher levels of overall ecoliteracy than men
(U=8555.000, p<0.05, n=287). This result was due to their ability to identify on average more local
flowering plants (U=7176.000, p<0.001, n=287) and crop species (U=8520.000, p<0.05, n=287)
than men (Fig 4.3.5). At the coastal site, although overall ecoliteracy was found not to vary between
men and women, women were able to identify significantly more local flowering plants than men
(U=5738.500, p<0.01, n=246) (Fig 4.3.6). No significant gender differences were found in
knowledge of local species uses, in variation of knowledge levels and in frequency of countryside
visits.

Differences in ecoliteracy between people of different ages were looked at by comparing the mean
knowledge levels of respondents from the four different age groups studied. At all three sites, mean
ecoliteracy levels were found to significantly increase with age (rural; H=24.183, df=3, p<0.001,
n=178, urban; H=52.963, df=3, p<0.001, n=287, coastal; H=87.326, df=3, p<0.001, n=246) (Fig
4.3.7).
A similar pattern was found in knowledge of species uses. Both total number of uses identified
(H=50.300, df=3, p<0.001, n=711) and proportion of species for which uses were identified
(H=46.245, df=3, p<0.001, n=711) significantly increased with age (Fig 4.3.8). This pattern of
increased ecoliteracy with age was demonstrated at all sites in knowledge of local birds, flowering
plants, crops and marine species (only at the coastal site) (Fig 4.3.9). However, knowledge of
mammal species was only found to show a slight incline with age at the rural study site and none at
the urban or coastal (Fig 4.3.10) (Table 4.3.1).
121
100
A
80
60
40
20
A
0
women
Gender
men
100
B
80
A
60
A
A
40
20
0
women
Gender
men
Crop ecoliteracy (% crops identified)
100
C
80
60
40
20
A
A
0
women
Gender
men
Figure 4.3.5. Comparing urban gender differences in (A)

level of total ecoliteracy, (B) level of flower ecoliteracy
and (C) level of crop ecoliteracy (median + interquartile
range, n=287). Symbol (o) indicates outliers, symbol (*)
indicates extreme outliers
122
100
80
60
A
A
A
40
20
A
women
men
Gender
Figure 4.3.6. Comparing coastal gender differences in

level of flower ecoliteracy (median + interquartile range,
n=246). Symbol (o) indicates outliers, symbol (*) indicates
extreme outliers
123
100
A
80
60
40
20
0
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
100
B
80
60
40
A
A
20
A
15-19yrs
20-29yrs
30-49yrs
0
50+yrs
Age group
100
C
80
60
A
40
20
A
0
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
Figure 4.3.7. Comparing age group differences in total

ecoliteracy at (A) the rural site, (B) the urban site and (C)
the coastal site (median + interquartile range, n=711).
extreme outliers
124
40
S
30
A
A
20
A
A
10
0
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
40
Use knowledge (% species

for which uses were identified)
B
A
30
20
10
0
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
Figure 4.3.8. Comparing age group differences in (A) use

knowledge (total number of uses) and (B) use knowledge
(% species for which uses were identified) (median +
interquartile range, n=711). Symbol (o) indicates outliers,
symbol (*) indicates extreme outliers
125
100.
A
A
80
60
40
20
0
15-19yrs
20-29yrs
30-49yrs
Bird ecoliteracy (% birds identified)
100
80
60
S
40
A
20
50+yrs
A
15-19yrs
80
60
40
20
0
15-19yrs
20-29yrs
A
30-49yrs
Age group
20-29yrs
30-49yrs
50+yrs
Age group
50+yrs
Marine ecoliteracy (% marine species identified)
Age group
100
100
A
A
A
80
A
A
30-49yrs
50+yrs
A
60
40
20
0
15-19yrs
20-29yrs
Age group
Figure 4.3.9. Comparing age group differences in (A) level of bird ecoliteracy,
(B) level of flower ecoliteracy, (C) level of crop ecoliteracy and (D) level of
marine ecoliteracy (median + interquartile range, n=711). Symbol (o) indicates
outliers, symbol (*) indicates extreme outliers
Mammal ecoliteracy (% mammals identified)
126
100
80
60
A
A
A
40
20
0
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
Figure 4.3.10. Comparing rural age group differences in

level of mammal ecoliteracy (median + interquartile range,
n=711). Symbol (o) indicates outliers, symbol (*) indicates
extreme outliers
127
Table 4.3.1. Statistical results for age differences in ecoliteracy levels of the different species groups
at all 3 sites
Mammal
ecoliteracy
Rural
H statistic
10.807
df
3
Significance
p<0.05*
n value
178
Urban
No significant
H statistic
difference
df
Significance
n value
Coastal
No significant
H statistic
difference
df
Significance
n value
df
Degrees of Freedom
Kruskall-Wallis statistical test used
Bird
ecoliteracy
Flower
ecoliteracy
Crop
ecoliteracy
46.160
3
p<0.001***
178
44.142
3
p<0.001***
178
24.183
3
p<0.001***
178
55.088
3
p<0.001***
287
45.964
3
p<0.001***
287
90.622
3
p<0.001***
287
98.746
3
p<0.001***
246
71.613
3
p<0.001***
246
57.212
3
p<0.001***
246
Marine
ecoliteracy
55.058
3
p<0.001***
246
Age was also found to affect the frequency with which a person visits the countryside, with
significantly more visits being made by elder people than younger (H=20.245, df=3, p<0.001,
n=711). On average, people over the age of 30 visit the countryside weekly compared with people
29 years and below who only visit monthly. No significant difference in knowledge variation
between the age groups was recorded.

The effect of income on knowledge level was looked at by comparing the ecoliteracy of groups of
respondents with differing levels of household income. Ecoliteracy was found not to significantly
vary with income, as was knowledge of species uses. In addition, no significant differences were
found in knowledge variation, frequency of countryside visits and type of environmental experience
between respondents from different income groups.

The effect of differing occupations on knowledge was tested by comparing the ecoliteracy levels of
respondents from different job sectors (excluding retired respondents). Occupation was found to
128
have a significant effect upon ecoliteracy levels at all three sites, although the differences in
knowledge caused by differing occupations at the rural site were less pronounced (rural; H=25.611,
df=5, p<0.001, n=143, urban; H=37.747, df=5, p<0.001, n=221, coastal; H=61.823, df=5, p<0.001,
n=182) (Fig 4.3.11). At all sites, respondents working in the environmental and education sectors
were found to have the highest levels of ecoliteracy, unlike students who were found to have the
lowest knowledge levels at all sites. No significant differences in knowledge variation were found
between people from different occupation groupings.

When considering the effect of childhood location on ecoliteracy, the data from all three sites were
combined to test for a relationship regardless of current area of residence. The ecoliteracy levels of
respondents who grew up in four different environments (rural, urban, coastal and traveled) were
compared. Childhood environment was found to have a significant effect upon ecoliteracy levels
(H=51.621, df=3, p<0.001, n=711). Respondents who spent their childhoods in rural areas or
traveled between sites were found to know more than respondents who grew up in coastal or urban
areas (Fig 4.3.12).
Childhood location was also found to significantly affect the frequency of visits a person makes to
the countryside (H=82.459, df=3, p<0.001, H=711). People who traveled during their childhood or
lived in rural or coastal areas, on the whole, visited the countryside more frequently (on average
once a week) than urban residents (on average once a month). However, there were no significant
differences in knowledge variation between the groups of respondents from different childhood
locations.

The effect of length of residence in an area on ecoliteracy was tested by combining the data from all
sites to test if, regardless of site, the longer a person inhabits a locality, the more knowledge they
accumulate of their local species. The ecoliteracy levels of groups of people that have lived at the
study sites for different lengths of time were compared. Length of residence was found to have a
significant effect upon overall ecoliteracy levels (H=96.888, df=7, p<0.001, n=711). A pattern was
detected whereby knowledge declined slightly between 0-4 and 5-9 years living at a site, after this
129
100
A
80
60
40
20
A
0
environment/farming
healthcare
student
education
miscellaneous
housewife/unemployed
Occupation
100
B
80
60
40
20
0
environment/farming
healthcare
student
education
miscellaneous
Occupation
100
C
80
A
A
60
A
A
40
20
A
0
environment/farming
healthcare
student
education
miscellaneous
Occupation
Figure 4.3.11. Comparing differences in total ecoliteracy between different

occupation groups at (A) the rural site, (B) the urban site and (C) the coastal
site (median + interquartile range, n=546). Retired respondents were
removed from analysis. Miscellaneous includes office work, construction,
retail and administration work. Symbol (o) indicates outliers
130
100
80
A
A
A
urban
coastal
60
40
20
0
rural
traveled
Environment of childhood
Figure 4.3.12. Comparing how differences in childhood

environment affect total ecoliteracy (median +
interquartile range, n=711). Symbol (o) indicates outliers
100
80
A
A
A
A
0-4
5-9
60
40
20
0
10-19
20-29
30-39
40-49
50-59
60+
Years lived at site
Figure 4.3.13. Comparing how differences in years lived

at a site affect total ecoliteracy (median + interquartile
range, n=711). Symbol (o) indicates outliers
131
ecoliteracy gradually increased until 40-59 years of residence (Fig 4.3.13). At this point ecoliteracy
began to decline again. No significant difference in knowledge variation and frequency of
countryside visits was detected between the groups.

The effect of countryside visits was assessed by combining the data from all three sites to test if,
regardless of site, increased countryside visits act to heighten a persons knowledge of their local
environment. Ecoliteracy was found to be significantly associated with frequency of visits
(H=79.465, df=4, p<0.001, n=711) and to increase with increased frequency of countryside visits
(Fig 4.3.14). This pattern of heightened knowledge with increased countryside visits was detected in
all species groupings with the exception of marine species that did not significantly differ (Table
4.3.2) (Fig 4.3.15).
Table 4.3.2. Statistical results for the effect of countryside visits on ecoliteracy levels of different
species groups
Mammal
Bird
ecoliteracy
ecoliteracy
73.029
46.355
H statistic
4
4
df
p<0.001***
p<0.001***
Significance
n value
711
711
df
Degrees of Freedom
Kruskall-Wallis statistical test used
Flower
ecoliteracy
51.635
4
p<0.001***
711
Crop
ecoliteracy
96.122
4
p<0.001***
711
Marine
ecoliteracy
No
significant
difference

The effect of a persons primary reason for visiting the countryside on ecoliteracy was tested by
comparing the knowledge of groups of people that visit the countryside for different purposes. An
individuals primary reason for visiting the countryside was found to significantly affect their level
of knowledge (H=51.332, df=8, p<0.001, n=639). For instance, people visiting the countryside
primarily for occupation and hobbies knew significantly more than people who simply passed
through the countryside on their way to somewhere else (Fig 4.3.16).
Knowledge variation caused by peoples differing reasons for visiting the countryside was greatest
at the coastal site (H=38.601, df=8, p<0.001, n=240), intermediate at the urban site (H=16.902,
132
100
80
A
A
A
A
A
60
40
20
A
A
0
never
rarely
monthly
weekly
daily
Frequency of countryside visits
Figure 4.3.14. Comparing how differences in frequency of

countryside visits affect total ecoliteracy (median +
100
100
A
80
60
A
A
A
A
S
S
40
S
S
20
Bird ecoliteracy (% birds identified)
Mammals ecoliteracy (% mammals identified)
133
0
never
rarely
monthly
weekly
A
A
60
40
20
daily
never
rarely
monthly
weekly
daily
100
100
S
80
60
A
A
40
20
80
80
A
A
60
40
20
0
never
rarely
monthly
weekly
daily
A
A
never
rarely
monthly
weekly
daily

affect (A) level of mammal ecoliteracy, (B) level of bird ecoliteracy, (C) level of
flower ecoliteracy and (D) level of crop ecoliteracy (median + interquartile range,
n=711). Symbol (o) indicates outliers, symbol (*) indicates extreme outliers
134
100
A
80
A
A
A
A
A
A
60
40
20
A
A
0
day out/holiday
hobby
relax/leisure
dog walking
occupation
passing through
visit friends/relatives
walking/exercise
other; e.g. fresh air, school trips
Reason for countryside visits
Figure 4.3.16. Comparing how differences in reasons for

visiting the countryside affect total ecoliteracy (median +
135
df=8, p<0.05, n=239) and was not significant at the rural site. Walking and exercise was found to be
the primary reason for visiting the countryside in the UK. Residents from urban and coastal areas are
more likely to visit the countryside to relax than rural-dwellers and rural residents are more likely to
visit for the purposes of occupation and dog walking. Urban residents are the most likely to visit the
countryside for holidays and day trips (Fig 4.3.17).
4.3.10. The relationship between attitude towards ecological knowledge and knowledge level
The relationship between attitude towards the importance of environmental knowledge and level of
ecoliteracy was tested. The data from all three sites were combined to test if a person who values
ecoliteracy more is more likely to hold higher levels of knowledge regardless of site. Ecoliteracy
was found to be significantly associated with attitude towards environmental knowledge (Rs=0.351,
p<0.001, n=708) and to increase with increased view of its importance (Fig 4.3.18).

The relationship between knowledge source and knowledge level was examined by testing for an
association between % contribution from a source and overall knowledge level. Parents/relatives,
friends/locals and hobbies were all found to increase a persons ecoliteracy by boosting their
knowledge of mammals, birds, flowers, crops and marine species (Table 4.3.3). Hobbies as a
knowledge source had the strongest positive association with overall knowledge level.
Occupation was found to be positively associated with ecoliteracy by increasing knowledge of
mammals, birds, flowers and crops, although not marine species where occupation had no effect.
Books were only found to be positively associated with knowledge of birds, flowers and marine
species, but to have no affect on mammals and crops.
Television and education, on the other hand, were found to have no overall association with
mammals, and to have a negative effect on respondents knowledge of birds, flowers, crops and
marine species. Therefore, overall ecoliteracy level was negatively associated with % contribution
from television and education. Hence, people who rely most upon the television and schooling for
ecological knowledge are likely to be less knowledgeable about their local environment.
136
% Respondents that listed reason
50
A
40
30
20
10
0
day out/holiday
hobby
relax/leisure
dog walking
occupation
passing through
walking/exercise
Reason for visits to countryside
50
B
40
30
20
10
0
day out/holiday
hobby
relax/leisure
dog walking
occupation
passing through
walking/exercise
50
C
40
30
20
10
0
day out/holiday
hobby
relax/leisure
dog walking
occupation
passing through
walking/exercise
Figure 4.3.17. Comparing differences in reasons for

making countryside visits at (A) the rural site, (B) the
urban site and (C) the coastal site and how they differ
between the 3 sites (n=639)
100.0
100
% of total species named
137
A
A
80 80.0
60 60.0
40
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
40.0
A
A
A
A
A
A
A
A
20 20.0
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
0 0.0
importance of knowledge ranked
Importance of knowledge
Figure 4.3.18. Views towards knowledge importance

ranked and relationship with an individuals total
ecoliteracy (n=708). Score 5 indicative of respondents that
view LEK as the most important and score 1 the least
138
Table 4.3.3. Statistical results for the association between % source contribution and ecological
knowledge level
Source
Rs statistic
Books
0.132
Television
-0.218
Parents
0.108
Friends
0.119
Hobbies
0.282
Occupation
0.148
Education
-0.147
Spearmans Rank statistical test used
n value
711
711
711
711
711
711
711
Significance
p<0.001
p<0.001
p<0.01
p<0.01
p<0.001
p<0.001
p<0.001
Association
Positive
Negative
Positive
Positive
Positive
Positive
Negative
A similar pattern was observed when testing for a correlation between knowledge source
contribution and view of the importance of ecoliteracy. Books (Rs=0.125, p<0.01, n=711),
friends/locals (Rs=0.099, p<0.01, n=711), hobbies (Rs=0.213, p<0.001, n=711) and occupation
(Rs=0.093, p<0.05, n=711) are all weakly, but positively, associated with view of ecological
knowledge importance. The higher the % contribution from these sources to a persons ecological
knowledge, the higher they view ecoliteracy. Hobbies revealed the strongest association.
No such association was found between parents/relatives and importance rank. Television (Rs=0.186, p<0.001, n=711) and schooling (Rs=-0.075, p<0.05, n=711), however, were found to be
weakly, but negatively, associated with importance rank. That is the higher the % contribution from
these sources, the lower a persons view of the importance of environmental knowledge.
Patterns in knowledge acquisition from different sources were tested for. No differences in the
contributions made by books and schooling were detected between the three study sites. However,
television as a knowledge source was found to contribute more to the knowledge of urban and
coastal residents than rural (H=9.358, df=2. p<0.01, n=711). Parents/relatives (H=11.534, df=2,
p<0.01, n=711), friends/locals (H=15.963, df=2, p<0.001, n=711) and occupation (H=14.700, df=2,
p<0.01, n=711) all contributed to a significantly greater extent at the rural site. Hobbies as a source
of ecological knowledge acquisition contributed significantly more at the coastal site and least at the
rural site (H=9.341, df=2, p<0.01, n=711) (Fig 4.3.19).
139
A
A
80
A
60
40
20
0
rural
urban
100
coastal
% Knowledge contribution from parents/relatives
% Knowledge contribution from television
100
B
80
60
40
20
0
rural
urban
C
80
60
A
40
S
S
S
S
S
S
S
S
20
S
rural
urban
coastal
% Knowledge contribution from hobbies
80
D
S
60
S
S
40
S
S
20
S
S
S
S
S
S
S
S
S
S
rural
urban
S
S
S
S
S
S
S
S
S
S
80
S
A
A
60
S
coastal
Site
Site
40
100
100
coastal
Site
% Knowledge contribution from occupation
% Knowledge contribution from friends/locals
Site
100
Figure 4.3.19. Comparing the %

knowledge contribution from (A)
television, (B) parents/relatives, (C)
friends/locals, (D) occupation and (E)
hobbies between the 3 UK sites (median
A
A
+ interquartile range, n=711). Symbol

(o) indicates outliers, symbol (*)
20
indicates extreme outliers

0
rural
urban
Site
coastal
140
Books, television, friends/locals, hobbies and education were all found to contribute knowledge to
equal extents between men and women. Parents/relatives, however, were found to be significantly
more important to women as a knowledge source (U=56179.500, p<0.01, n=711) and occupation
contributed significantly more to men (U=56089.000, p<0.001, n=711) (Fig 4.3.20).
Books (H=32.298, df=3, p<0.001, n=711), friends/locals (H=8.667, df=3, p<0.05, n=711) and
hobbies (H=33.236, df=3, p<0.001, n=711) were all found to contribute significantly more to elder
peoples knowledge bases. The contribution from television (H=43.160, df=3, p<0.001, n=711) and
schooling (H=15.037, df=3, p<0.01, n=711), however, was found to significantly decline with age.
Therefore, young people were the most reliant on these knowledge sources. Parents/relatives were
found to contribute significantly less to 50 year olds and upwards than to the three younger age
groups (H=10.891, df=3, p<0.05, n=711) (Fig 4.3.21). Occupation was the only knowledge source
found not to significantly differ in its contribution to the knowledge of different age groups.
Knowledge source contribution was found not to differ between groups of people with different
levels of household income.
The effect of quality of experience on ecological knowledge was examined by comparing
ecoliteracy levels between groups of people that attained different quality of experience scores based
on their frequency of visits, area of residence, childhood environment, years lived in an area and
activities pursued within the countryside. Ecoliteracy was found to be positively associated with the
quality of a persons environmental experience (Rs=0.378, p<0.001, n=711). Positive correlations
between knowledge and quality of experience were found for all species categories (Table 4.3.4).
% Knowledge contribution from

parents/relatives
141
100
A
80
60
40
20
0
women
men
% Knowledge contribution from occupation
Gender
100
B
S
80
S
S
60
40
20
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
women
men
Gender
Figure 4.3.20. Comparing the % knowledge contribution

from (A) parents/relatives and (B) occupation between
men and women (median + interquartile range, n=711).
extreme outliers
% Knowledge contribution from books
A
A
80
A
A
60
40
20
0
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
S
% Knowledge contribution from hobbies
100
S
S
80
S
60
S
40
S
S
S
S
20
S
S
S
S
S
S
S
S
S
S
S
15-19yrs
20-29yrs
30-49yrs
% Knowledge contribution from television
100
% Knowledge contribution from friends/locals
142
100
B
80
60
S
40
S
S
A
A
20-29yrs
30-49yrs
50+yrs
S
S
S
S
S
0
15-19yrs
Age group
A
100
D
A
80
A
A
60
40
20
50+yrs
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
A
100
A
A
80
S
20
A
A
60
40
20
A
A
% Knowledge contribution from

parents/relatives
% Knowledge contribution from education
A
A
Age group
100
80
60
40
20
0
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
Figure 4.3.21. Comparing the % knowledge contribution from (A) books, (B)
friends/locals, (C) hobbies, (D) television, (E) education and (F) parents/relatives between
the different age groups (median + interquartile range, n=711). Symbol (o) indicates
outliers, symbol (*) indicates extreme outliers
143
Table 4.3.4. Statistical results for the association between quality of environmental
experience and knowledge of each species group
Mammal
Bird
ecoliteracy
ecoliteracy
Rs statistic
0.322
0.360
Significance
p<0.001***
p<0.001***
n value
711
711
Spearmans Rank statistical test used
Flower
ecoliteracy
0.313
p<0.001***
711
Crop
ecoliteracy
0.416
p<0.001***
711
Marine
ecoliteracy
0.296
p<0.001***
711
Significant site differences in quality of environmental experience were revealed (H=405.156, df=2,
p<0.001, n=711). Rural residents were found to have the highest quality of environmental
experience, followed by coastal residents and then by urban residents (Fig 4.3.22). Quality of
environmental experience was also found to significantly increase with age (H=29.291, df=3,
p<0.001, n=711) (Fig 4.3.23). No such differences were revealed between men and women. The
quality of an individuals environmental experience was also found to be positively correlated with
their view towards the importance of ecoliteracy (Rs=0.297, p<0.001, n=711). Therefore, people
with a higher quality of environmental experience viewed knowledge of the environment as more
important.
144
Experience score
20
15
10
0
rural
urban
coastal
Site
Figure 4.3.22. Comparing quality of experience scores

between the 3 UK sites (median + interquartile range,
Experience score
20
15
10
0
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
Figure 4.3.23. Comparing quality of experience scores

between the different age groups (median + interquartile
range, n=711)
145
4.4. Discussion

Rural ecoliteracy, in terms of both local species identification and knowledge of uses, exceeding that
of urban and coastal areas is indicative of the importance of frequent experience to environmental
awareness as previously suggested by Nabhan & Trimble (1994) and Pyle (2003). However, just
living in the countryside is not necessarily confirmation of increased countryside experience, hence
this investigation assessed frequency of interaction. The finding that rural respondents make visits to
the countryside more frequently (weekly) than urban residents (monthly) indicates that experience is
most likely a key predicting factor of knowledge level.
Coastal respondents, however, also claimed to visit the countryside on a weekly basis on average,
but held lower knowledge levels of ecoliteracy than rural respondents (but more than urban based on
terrestrial knowledge). The enhanced awareness of rural respondents is most likely a product of
daily wildlife encounters on their doorstep (passive interaction) combined with outdoor visits (active
interaction). For instance, when leaving the house or looking out the window, rural residents have
the opportunity to observe nature rather than having to actively seek it out.
It could be argued that coastal respondents have marine species on their doorstep and yet their
knowledge of them compared to their knowledge of terrestrial species is greatly reduced. This could
be explained by knowledge of marine species being less successfully transferred than that of
terrestrial species. Perhaps sources of information on maritime species are less easily available than
terrestrial sources, for instance fewer books may be available and fewer elders knowledgeable on
marine systems. Also, marine species are less frequently sighted than many terrestrial species. For
instance, when walking through fields you are more likely to see a rabbit or fox than you are to see
an edible crab or starfish when walking on the beach.
Despite years of dependence on wild plants for sustenance and survival (Mabey, 1996; Pretty,
2002), wild flower knowledge was found to be very low among members of the UK public
interviewed. Poor knowledge of flowering plants can probably be explained by our lack of reliance
upon them, unlike crop species, combined with their less characteristic appearance, unlike animals.
146
Enhanced knowledge of mammals is most likely a result of their iconic status in our culture (Pretty,
2006). For example, they are often featured as main characters in books and used as symbols for
major wildlife organisations (such as the panda for the World Wide Fund for Nature (WWF, 2006)).
High levels of variation in knowledge were revealed across all of the UK sites. This is most likely an
outcome of departure from resource dependence in industrialised regions enabling both experts and
non-experts to co-exist equally well.

Many earlier studies have hypothesised that women may have a connection and deeper
understanding of nature than men (Shiva, 1997; Warren, 1997). Plumwood (1993) and Mellor
(1997) attributed this to womens historical roles in the identification, collection and application of
wild plants. Historically, women would collect wild flowers to use in cooking and medicines
(particularly for use in childcare). In support of this idea that ancestral roles have influenced
womens knowledge content today, women in this study were found to have increased knowledge of
flowering plants at all sites. At the urban site in particular, where opportunities to experience the
environment were reduced, women knew more crop species too and, therefore, were more
ecoliterate than men.
However, at the rural site men knew more mammal species than women and, as a result, overall
ecoliteracy levels between men and women did not significantly differ. This was most likely due to
increased visits to the countryside and an increased number of outdoor jobs available to men in rural
areas, such as farming. No difference in overall ecoliteracy between men and women was also
observed at the coastal site where frequency of visits to the countryside was also high (despite
increased wild flower knowledge of women). This indicates that more frequent experience of nature
appears to reduce gender differentiation in knowledge. This, in turn, suggests that as communities
interact more frequently with nature, a symptom of resource dependence, gender differentiation in
knowledge levels subsides. Therefore, making it unlikely that womens increased ecoliteracy is a
symptom of historical roles (when interaction was most frequent) and may have a deeper, perhaps
innate, explanation (Shiva, 1997). More research in this area would have to be carried out to confirm
or disprove this.
147

At all three sites, clear patterns of increased knowledge, in terms of both species identification and
knowledge of uses, were revealed with age. This is indicative of one of two scenarios. Either a
progressive loss of ecoliteracy is taking place between generations (Douglas, 2002; Birchenough,
2005) or ecological knowledge is attained continually throughout life in non-resource dependent
communities not reaching the level of parents and grandparents until late in life. To decipher
between these two scenarios, a study needs to be carried out on a group of individuals measuring
their ecoliteracy at different ages throughout their lives. Unfortunately, the time available to this
investigation did not permit this.
However, this study did reveal that a large number of young people now depend on television and
schooling for environmental education. This is in contrast to their elders who relied more upon
friends and hobbies (word-of-mouth and direct experience). Combining this information with the
negative association revealed between knowledge level and % contribution from television and
education (indirect virtual experiences), it is highly likely that the ecoliteracy of todays younger
generations will never reach that of their elders and, therefore, progressive loss is occurring.
Additional evidence for this is the decreased outdoor experience of younger generations. This was
demonstrated by people over the age of 30 being more likely to visit the countryside weekly than
people below this age who visit it on average monthly.
All species groups revealed this pattern of progressive knowledge loss with the exception of
mammals, which only revealed a slight generational loss at the rural site. This again highlights the
iconic status of animals, in particular mammals, in childhoods today (Pretty, 2006). Childrens
books, in particular, often anthropomorphize mammals for use as main characters, for instance
badger, rat and mole in Wind in the Willows (Kenneth Grahame (Wikipedia, 2006a)) or characters
in Beatrix Potter books (Wikipedia, 2006b). Childrens toys are also often based on animals.
Therefore, children become confident at recognising a wide variety of mammals from a young age.
This emphasises the effect that forming bonds and becoming familiar with wildlife from a young
age can have (Nabhan & Trimble, 1994).
148

Ecoliteracy, attitude towards the importance of environmental knowledge and frequency and type of
countryside experience did not vary with income level. This re-emphasises that experience is key to
ecoliteracy level and only factors that affect level of environmental experience have the ability to
subsequently affect ecological knowledge.

At all sites the highest levels of ecoliteracy were held by those working in the environmental and
education job sectors. Students were found to be the least knowledgeable. The environmental sector
includes jobs such as farmer, park ranger and site manager, all of which enhance time spent
experiencing nature and therefore ecoliteracy.
High ecoliteracy within the education sector is indicative of a career that relies upon the acquisition
and transfer of knowledge for success. However, despite high levels of ecoliteracy held by teachers,
the results revealed that this knowledge is not being successfully passed on to students. This
suggests that the failure to transmit ecoliteracy is not due to a paucity in the knowledge of the
teaching staff but is a result of structured curricula eliminating local knowledge and continual
emphasis on meeting academic targets.
As mentioned, students were found to have the lowest levels of ecological knowledge. This reveals
that educated young people that have gone through years of formal education have little awareness
of that which is on their doorstep. It clearly shows how formal education is failing to create the next
generation of naturalists in the UK (Orr, 1996; Peat, 1996; Pyle, 2001; Skutnabb-Kangas, 2004;
Haigh, 2005) and, instead, is producing reams of intelligent and educated individuals with an
ignorance of their local surroundings. This means that we may no longer have the staff qualified to
teach local knowledge in the future, even if curricula are altered and nature walks and butterfly nets
are reintroduced (Pyle, 2001).

Indicating the importance of experience from a young age, individuals that enjoyed rural childhoods
were found to have higher levels of ecoliteracy than individuals from urban and coastal childhoods.
149
Individuals that traveled between sites in childhood also held levels of knowledge similar to that of
rural inhabitants. This indicates that moving between sites from a young age acts to increase adult
ecoliteracy, arguably by enhancing a young persons experience of different environments and
therefore their ecological awareness from childhood.
It is likely that an affinity is forged in childhood between the environment and individuals from rural
areas, or those that traveled, which lasts for a lifetime and motivates them to encounter and learn
about local wildlife in the future (Nabhan & Trimble, 1994; Pyle, 2002; Ryan, 2005). This
emphasises the importance of maximising experience of natural environments during childhood,
perhaps through frequent day visits, to creating ecoliterate adults. The finding that childhoods spent
travelling or in rural locations led to more frequent visits being made to the countryside in adulthood
is evidence in support of this.

A clearly defined pattern was found to describe the relationship between time spent at a site and
ecological knowledge. In general, knowledge level was found to be higher when a person has only
lived at a site for 0-4 years than for 5-9 years. From this period onwards knowledge level was found
to gradually increase until between 50 and 60+ years of residence when it starts to decline.
This pattern most likely represents an initial interest in moving to a new area, often with new
habitats and different wildlife. This initial interest then wanes and ecoliteracy gradually increases
through accumulated experiences of the environment over time. (This study found that people who
have lived in an area for 50 or 60 years were no more likely to make countryside visits than people
who have lived there for five years, therefore accumulation of observations over time is key here.)
The final dip in knowledge is most likely representative of memory fallibility with age (Krech,
2005).

Supporting the finding that direct experience is highly important to the acquisition of ecological
knowledge, it was revealed that the more an individual visits the countryside, the higher level of
ecoliteracy they were found to have. This effect may be cyclical. For instance, frequent experience
150
from a young age may enhance ecoliteracy and ecoliteracy, in turn, may promote a desire to
frequently encounter the environment. Therefore, just as the extinction of experience can create a
downward spiral of disconnection (Pyle, 2002, 2003), perhaps enhancement of experience can create
an upward spiral of affection and reconnection. In all species groupings the difference in ecoliteracy
levels between those who visit the countryside daily and weekly were found to vary very little. This
suggests that policies to promote day visits to the countryside on average once a week may be
sufficient to significantly increase ecoliteracy amongst the UK population.
Crop ecoliteracy was found to be higher than expected in the group of individuals that never visit the
environment. This is most likely due to a number of plants in this category being eaten as vegetables
enabling recognition despite limited experience of them in the wild. Marine plants and animals was
the only species group for which increased visits did not enhance local knowledge. Specific
activities and hobbies may be needed to enhance marine species experience and, therefore,
ecoliteracy including fishing, curio-collection and rock pooling, rather than just walking and
observing.

This study revealed that, in addition to the frequency of an individuals countryside visits affecting
their level of knowledge, personal reasons for visiting the countryside also have an effect.
Individuals who simply pass through the countryside (in a car for example) encounter a lot less than
those walking out of doors experiencing all the sights, sounds and smells of the natural environment,
and subsequently know a lot less.
The most knowledgeable groups were found to be those that visit the environment for the purposes
of pursuing hobbies and occupations. These groups all frequently encounter the outdoors in spite of
bad weather conditions all year round, in comparison with groups that visit solely for the purpose of
holidaying or visiting friends. In these groups, visits can be infrequent and time outdoors limited.
Environmental occupations and hobbies, fishing or farming for instance, also both require a basic
knowledge of the environment to participate in them. Therefore, the basic skill required combined
with frequent outdoor experience most likely makes individuals in these categories the most
knowledgeable.
151
In 2004, some 1.5 billion day visits were made to the countryside in the U.K. (The Countryside
Agency, 2004). When comparing the three sites, it was clear that the most frequent reason for
members of the UK public to visit the countryside was to go walking or pursue some other form of
exercise. Rural residents visiting the countryside for different reasons did not have significantly
different levels of ecoliteracy. It is likely that any differences are buffered by passive interaction
(encountering the surrounding countryside in their daily lives). Urban residents are the most likely to
visit for the purpose of holidays. Urban and coastal residents were more likely to take a trip to the
countryside just to relax than rural residents living in the countryside. This suggests that the
countryside provides a welcome leisurely break to many pursuing urban lifestyles.
4.4.10. The relationship between attitude towards ecological knowledge and knowledge level
High ecoliteracy was found to be correlated with a stronger view of the importance of ecological
knowledge. This is most likely a cyclical process. Viewing environmental knowledge as important is
more likely to encourage a person to gain this knowledge However, once an individual learns more
about their environment they may value local ecosystems more and, therefore, want to learn more
about them. This cyclical process is likely to derive from an initial affinity to learn about and bond
with nature, most likely derived from frequent outdoor encounters early in life.

Clear differences were found in the contribution of different sources to the ecoliteracy of the UK
public. The result that hobbies, parents/relatives, friends/locals, books and occupation all have the
capacity to increase a persons ecoliteracy illustrates the importance of word-of-mouth sources, and
to some extent books, in the transfer of ecoliteracy. However, ecoliteracy was negatively associated
with % acquisition from television and school. This illustrates that the more a person relies on
television and formal education to provide them with their local knowledge (therefore relying less
on informal contacts and personal activities), the less ecological knowledge a person is likely to
hold. This demonstrates the failure of indirect images (on television and in textbooks) and virtual
images (on the internet) in transferring ecoliteracy and acts as further proof that the value of
frequent, direct experience and word-of-mouth transfer to ecoliteracy acquisition cannot be equaled
(Kellert, 2002; Pyle, 2002, 2003).
152
The same patterns of association were revealed when considering the effect of knowledge source on
an individuals view of the importance of ecoliteracy. This supports the idea that direct experience
does not only contribute to ecological knowledge but also has the capacity to create a deeper bond,
an affinity with nature, by which an individual values knowing about their plant and animal
neighbours (Nabhan & Trimble, 1994). In a non-resource dependent society this may be the only
means by which to do this.
Most likely due to reduced access to green space and knowledgeable people, urban and coastal
residents were found to rely significantly more on television and formal education for ecological
knowledge acquisition. Rural residents, however, relied more upon direct word-of-mouth (through
fellow workers, parents and friends). Environmental hobbies were found to be more frequently
pursued by coastal and urban residents seeking relaxation and, therefore, were found to be a more
important source of acquisition at these sites.
Men were found to be more likely to gain their knowledge of the environment from jobs and women
from parents and relatives. Therefore, men are more likely to exchange LEK with work colleagues
than women are, whereas women are more likely to discuss issues such as the environment with
other family members. This may be a continuation of traditional societal roles where LEK transfer
was gender-differentiated and women gained their knowledge primarily from discussions with other
women on how to use local wild plants to fulfill their household roles, such as cooking and childcare
(Plumwood, 1993; Mellor, 1997).
Evidence that young peoples ecoliteracy is unlikely ever to reach the levels of their elders is given
by the finding that young people are more likely to rely upon television and classroom learning for
environmental knowledge than their elder relatives, as has already been discussed. Today, the
emphasis is strongly on young people to complete their education and attain qualifications, whereas
in the past the emphasis was on young people to find work (often on the land) and to look after their
families (Tansey, 2004). Consequently, elder respondents in this study relied mostly on
friends/locals, hobbies and books for acquisition of their own ecoliteracy, contrary to the situation
today.
153
Interestingly, parents/relatives were found to contribute less to the knowledge of over 50 year olds
than to the other three age groups. This may reflect natural memory loss having forgotten what they
learned in their childhoods and remembering what they learned in their adult years. Also the war
years meant that many families were separated and many children did not have elder relatives
around them during childhood.

Those with the highest quality of environmental experience in terms of frequent interaction,
interaction from a young age, interaction over a long period and a motivation to be outdoors, had
higher levels of ecoliteracy. This result is evidence in support of direct experience being of the
utmost importance to ecoliteracy, not only in terms of frequency but in terms of quality (Ryan,
2005).
Rural residents were found to have the highest quality of experience compared with coastal and
urban residents. Urban residents had the lowest quality of environmental experience most likely
resulting from living in built-up areas with reduced green spaces. Increased quality of experience
was found to positively correlate with an individuals view of the importance of ecoliteracy. Again
enforcing the idea that experience creates a deeper affinity with nature, not only creating a deeper
level of knowledge but also a higher intrinsic value for knowing about nature (Kellert & Wilson,
1993; Nabhan & Trimble, 1994; Kellert, 2002; Pyle, 2003; Karjalainen & Habeck, 2004).
This study has given evidence for the importance of word-of-mouth transfer and direct experience in
an individuals acquisition and view of ecoliteracy in non-resource dependent societies. In addition,
it has highlighted the failure of formal education and television in creating a new generation of
environmentally-aware naturalists and, as a result, we are seeing a progressive loss of ecoliteracy
over time. If this is the case that the knowledge of younger generations will never meet that of their
elders and this progresses generation after generation, will we soon be in a place where
industrialised regions have no naturalists left to teach future generations the plants and animals on
their doorstep? Will no-one notice when these plants and animals disappear? Is this a worldwide
phenomenon or just concentrated in developing regions? The following chapter examines
154
generational knowledge differences to see if ecoliteracy decline is occurring within resourcedependent communities of remote Indonesia where economic development is gradually encroaching.
It also looks to see if direct experience and word-of-mouth transfer are just as important to
knowledge acquisition within these communities where resource dependence is still prevalent.
155
Chapter 5. Indonesia: Kaledupa, SE Sulawesi
5.1. Introduction
Indonesia constitutes the worlds largest archipelago state and comprises over 17,500 islands of
which just over a third are inhabited (Elliott et al., 2001). It is considered a global biodiversity
hotspot with culturally- and ecologically-important tropical forests, mangrove systems, seagrass
meadows and 51,020 square kilometres of coral reef (17% of the worlds total) fringing many parts
of its extensive coastline (Burns, 2002; Pedju et al., 2004; May, 2005). Threatening this expansive
and valuable ecosystem is economic development and population growth. As many as 206 million
people were resident in Indonesia in 2001 and this total continues to increase at an estimated 1.5%
annually (Elliott et al., 2001; Burns, 2002).
As much as 60% of Indonesias population lives along its extensive coastline relying heavily on
marine resources for protein (Elliott et al., 2001). Local fisheries are responsible for meeting these
needs, 90% of which are artisanal (May, 2005). Simultaneous pressure is exerted from the recent
commercialisation and commoditisation of south-east Asias natural resources, particularly geared at
the export of mari-products to Japan and China (Robinson & Paeni, 1998). This external pressure
was derailed by the establishment of Indonesian law stating that non-Indonesian companies are only
allowed to exploit the national waters for the purpose of scientific research, but international
industries side-stepped this by providing the low wage Indonesian labour force with technologically
advanced fishing gear and boats (Elliott et al., 2001).
Hence coastal waters experience pressure from overexploitation due to ever-expanding consumption
requirements, in addition to pollution, destructive fishing practices, coral mining and boat damage,
all a symptom of the ever-expanding coastal population relying upon the goods and services of local
marine systems (Elliott et al., 2001). Sustainable development of this deprived archipelago and
state-enforced management over the vast coastline are not a real possibility with weak institutional
capacity and funding. Thus self-management of fisheries, by the communities dependent upon them,
is a key option for a region so constrained by poverty and population expansion, and has recently
156
become a viable one under government reform and decentralisation of powers (Elliott et al., 2001;
Burns, 2002; May, 2005).
Despite deprivation, considerable advances have improved the national human development levels
of the region, including a decline in infant mortality, a rise in adult literacy, increased gender
equality, in terms of schooling and labour, and a rise in per capita income (UNDP Human
Development Reports, 2005). However, it is unlikely that such national improvements have reached
the marginal tribes of the region. The state support decentralisation, democratic governance and
building upon widespread capital, believing in the long term this will provide the key to human
development and free competitive markets. However, advances in human development have done
little to alleviate deprivation for the most resource dependent communities. Many of these
communities remain on or below the poverty line and hence are unable to access improved facilities
that incite a cost, including schools and doctors surgeries (UNDP Human Development Reports,
2005).
5.1.1. The Wakatobi Marine National Park

Sulawesi is the fourth largest island of the Indonesian archipelago. It is situated with Borneo to the
west and the Moluccas to the east and covers a land area of 159,000 square kilometres. The
southeast peninsula province of Sulawesi incorporates a large number of islands and lies between 36o latitude and 120o 45-124o 06 longitude (Tourism of Indonesia, 2005). The remoteness of this
archipelago region has resulted in high diversity and endemism both on land and in the surrounding
waters. This region is situated on the theoretical division, termed the Wallace Line, that demarcates
the transvergence from Australian flora and fauna to that of Asia, thus it is considered of immense
importance in terms of biogeography and evolutionary biology (Smith, 2004).
At the centre of this biodiverse Wallacea region is the Tukan Besi archipelago, designated a
protected area in 1996 (Fig 5.1.1). The 13,900 square kilometres of the Wakatobi Marine National
Park (WMNP) incorporate all of the islands, atolls and reef systems of the archipelago and
constitutes the second largest Marine National Park in Indonesia (Pedju et al., 2004; May, 2005).
157
The WMNP has five main zones, three of which permit local fishing activities (Elliott et al., 2001).
Zonation and regulation enforcement aims to protect the invaluable ecosystems whilst ensuring local
resource requirements of the 88,000 people living within the parks boundaries are met long into the
future. The park employs over 45 rangers to police the park for external and/or destructive fishers
(COREMAP, 2002). External fishers exploiting the Wakatobi , particularly in the calm periods from
April to May and October to November, come from South Sulawesi, Kendari, Maluku, Irian Jaya
and even as far as the Philippines, Singapore and Hong Kong (Elliott et al., 2001).
WMNP
Figure 5.1.1. Map showing the location of the Wakatobi Marine National Park, Indonesia
(Multimap, 2006, 1:4,000,000)
With recent modernisation and economic development spreading from South Sulawesi, including
the introduction of motorised boats over the last 15 years with the import of cheaper engines from
China, resource control of the Wakatobi waters is becoming ever more stringent (Robinson & Paeni,
1998; Operation Wallacea, 2004). Those most impacted are local, low-income families who heavily
rely upon marine resources for subsistence and household income. Goods provided include fuel,
construction materials and foodstuffs (Operation Wallacea, 2004).
158
Natural resource dependence derives from the remoteness of the regions settlements limiting
income-generating capacity. The focus of the economy is on subsistence fishing and farming. Since
manufacturing potential is limited on the remote islands, any consumer goods must be imported
(Smith, 2004). Tourism has gradually crept into the Wakatobi providing fishers with alternative
incomes, although capacity is limited when considering infrastructure, food demand and waste
disposal potential (Elliott et al., 2001).
5.1.2. The island of Kaledupa

The WMNP consists of four main islands, the largest of which is Wangi Wangi home to a
population of 76,000 (Elliott et al., 2001). The second most populated island is Kaledupa with a
rapidly expanding population of 16,600. Economic development capacity is limited on Kaledupa by
food production, fresh water supply, power, transport and communications. Of the 16,000 plus
people spread across 19 villages, only 1516 people across 8 villages have access to electricity (May,
2004).
Farming, mostly of cassava and maize, and fishing the 78 kilometres of fringing reef are the main
livelihoods of Kaledupas inhabitants with a limited number of occupations available in trading,
construction and the civil service (May, 2004; Operation Wallacea, 2004). Kaledupan healthcare
remains primarily traditional in its resources with 31 Shamans, who hold extensive orally-acquired
knowledge of local medicinal plant uses, to only one Western doctor and no hospital facilities.
However, a shift in favour of Western medical practices with development is emerging. Education
facilities are also limited with 22 primary schools, four junior schools and only one secondary
school. Most children remain deprived of education above primary level (Burns, 2002; Operation
Wallacea, 2004).
With limited capacity to trade (only three markets and eight co-operatives are established on
Kaledupa), few alternative income sources and expanding population sizes, most families exercise
their rights to fish to some extent, whether it be land farmers to supplement their income or agar
farmers to supplement their food supply. Thus sustainable methods employed by artisanal fisheries
are critical to the maintenance of this essential marine resource.
159
Traditionally, local daily catch fed households with any remainder being shared among friends and
family. Today, local markets situated on Kaledupa (at Buranga, Ambeua and Langge) have acted to
commercialise fishing efforts to a degree, however improved transport and globalisation have had a
far greater effect. They have intensified demand from the larger islands of Wanci and Tomea, and
even as far a field as Singapore and Hong Kong for the live food and ornamental trade (May, 2004;
Pedju et al., 2004). Hence, this region has already undergone some degree of shift from subsistence
to commercial fishing over the past 15 years (May & Coles, 2004). Although the Chinese have been
recorded trading with Indonesian fishermen in trepang harvest as far back as the 15th century (Adi
& Sudarman, 1994), the concern today is that with growing commercial pressures (channeled
through middlemen and collector boats) being exerted upon artisanal fisheries, fishing efforts may
be further intensified beyond sustainable levels. This is likely to happen as cash-poor fishers begin
to see the monetary value of maximising their catch for export, above and beyond subsistence levels
(Elliott et al., 2001).
Fishing is a year-round activity off Kaledupa, and typically fishers go on expeditions for two weeks
at a time and then return home for several days rest. Local fishing methods employed in the waters
surrounding Kaledupa include 12 types of hand line, 12 different nets, 6 traps, gleaning and spear
fishing (May, 2004). Artisanal fisheries worldwide are responsible for generating sustenance and
income for low-income coastal communities and Kaledupa is no exception (Smith, 2004). Local
fishers claim destructive fishing to be damaging fish stocks off Kaledupa, but do not perceive overfishing and the shift from low efficiency subsistence practices to modern commercial techniques
(including the introduction of motorised boats) as a threat (May , 2004; Smith, 2004). External
fishers are generally blamed for destructive fishing and stock decline, through techniques including
fertiliser bombs and potassium cyanide poison (Elliott et al., 2001; Pedju et al., 2004).
In addition to over fishing, coral mining is a major threat to the reef systems of the region and their
resources (Pedju et al., 2004). Coral mining was legal until 1992, although the use of dead coral as a
free resource in the foundations of house building and road construction was not widely used until
after the 1960s when demand rocketed. Believed to make houses stronger many local women, in
particular indigenous women, blind to the consequences, continue to mine the coral illegally (Burns,
2002; Operation Wallacea, 2004).
160
Due to generations of local fishers perfecting catch techniques throughout the seasons, the marine
knowledge of communities on and around Kaledupa has become highly specialised and detailed.
They can provide current habitat information, in addition to describing detailed changes in stock
abundance and diversity over a longer-term period (Rudd et al., 2003; Smith, 2004). This local
resource knowledge may hold the key to predicting the impacts of resource withdrawal and what
levels can be permitted in that particular social and natural environment, essential to local
community management.
The low social standing of fisher families gives them limited opportunities to opt out of fishing and
pursue alternative careers, thus knowledge bases are transferred to the next generation uninterrupted
(Smith, 2004). The fact that local fish stocks have fluctuated over time but still meet the nutritional
needs of families in the region for generations is testimony to the success of traditional management
in sustaining local resources. This may not be a conscious effort at conservation, but still does not
deplete the effectiveness of fishing using a wide variety of customary techniques to target the local
variety of marine resources, meeting conservation targets as a by-product (Operation Wallacea,
2004; Smith, 2004).
This rationale has been applied off the east coast of Kaledupa where a stake-holder area has been
designated for community-management. Since national enforcement of park regulations is
financially and institutionally unviable across all areas of the park, self-management is more likely
to succeed in meeting local needs both environmentally and culturally (Elliott et al., 2001). This is
supported by the Wakatobis recent separation from the government district of Buton, giving the
newly formed government a mandate to manage its natural resources independently and locally
(Pedju et al., 2004). However, local communities express a utilitarian, functional view of marine
resources that may be indicative of a local desire for economic development and lifestyle
improvement. Thus a sudden financial injection into this region, permitting socio-economic
development via the purchase of modern gear, may act to destabilise traditional management
systems and further increase resource withdrawal. This in turn may act to deplete the resource base
that has sustained local populations up until this point (Polunin, 1984; Akimichi & Supriadi, 1996;
Smith, 2004).
161
5.1.3. The Orang Pulo and the Orang Bajo

In 2001, Kaledupa was home to 49.6% of the Wakatobis full-time fishers, although this included
only 5.7% of the island population. Instead, the great majority were derived from an ethnic group
who inhabit the reef flats surrounding the island and are under represented in the region, termed the
Orang Bajo (Bajo people, the South Sulawesi spelling of Bajau) (May, 2004; May & Coles, 2004;
Pedju et al., 2004). Thus, in addition to its biological diversity, the Wallacea region is also a hotspot
for cultural diversity (Smith, 2004). The Bajo only constitute 13% of Kaledupas population, but
make up the majority of fisher families, relying on marine resources as their primary protein source.
The islanders, however, termed the Orang Pulo (island people), have more varied diets including
chicken and goats as well as fish (May, 2004, 2005).
Unlike the Bajo that have maintained their traditional maritime roots, the Pulo (more frequently
referred to today as Kaledupans) now pursue livelihood options dominated by farming and
administration (May, 2005). Both groups are socially segregated with different traditions, languages
(Bahasa Kaledupa spoken by the Pulo and Bahasa Sama spoken by the Bajo) and lifestyles
(Operation Wallacea, 2004; May, 2005). The Pulo islanders are descendents of the nearby Butonese
with a combined history on the land (as agriculturalists) and at sea (as traders and pirates), that today
practice fishing and farming. The Bajo, on the other hand, are originally boat-dwelling, marine
nomads who traveled throughout the Malay Archipelago. Their cultural beliefs and worldviews
revolve around the sea and its products as a result of a long-term, historical dependence on marine
resources (Akimichi & Supriadi, 1996; May, 2005). Thus, despite living in each others presence for
generations, these two ethnic groups have evolved separately and maintained their own traditions
through limited interaction with one another.
Both ethnic groups suffered severe political pressure to assimilate during the 1950s-1960s causing
substantial deviation from traditional lifestyles. The Bajo were forced into sedentarisation on the
intertidal zone of the reef flats, constructing stilt houses out of locally-available materials. The Pulo
were made to sell the majority of their boats and adopt a new centralised government. Thus both lost
a portion of their maritime histories and were exposed to commercial pressures for the first time
(Elliott et al., 2001; May, 2005). Increasing external influence continued with mobility and media
162
access, simultaneously traditional customs and livelihoods waned, as younger generations had their
material expectations raised and began to seek jobs external to the area (Operation Wallacea, 2004).
Pulo villages are socially structured in their existence and elect a democratic head, termed the
Kepala Desa, who acts to both formally lead and informally advise the villagers. This role is usually
filled by a knowledgeable, well-respected elder man. Overseeing the activities of the villages within
the sub-district is the Camat, the head of the sub-district government. The role of the Camat is to
work alongside the Kepala Desas, the police, the army and the navy to regulate and control activities
such as coastal fishing (Elliott et al., 2001). The average income on the island is estimated at 20,000
Rupiah per person per day (GB1.25 / US$2.18) (Smith, 2004).
A great deal of Kaledupas land area has been converted to agriculture, although concerns exist that
once the agricultural capacity of the island has been reached, more people will turn to marine
resources to subsidise their lifestyles (Robinson & Paeni, 1998). Seaweed farming has become a
supplementary income source to many in recent years as a low effort and economically viable
option. However, this inevitably increases competition for space with fishers in this intensively
exploited region, causing fishers to venture further a field than was previously necessary. The value
of seaweed lies in a key ingredient, carageenan, which can be extracted and used within the food and
cosmetic industries. Despite high start-up costs, seaweed farming requires no licence on Kaledupa
and generates income all year round with little further input, even during the calm season when
increased catch creates a fall in the market price of fish (Smith, 2004).
Due to the traditional nomadic lifestyles of the Bajo, minimal research has been carried out on them
to date. That which has been conducted provides us with some level of understanding of their
community infrastructure and societal roles, mostly in relation to employment and economic change
(Smith, 2004). However, their knowledge to date is an untapped resource of generations of
observations, as a result of living with and relying on the marine environment.
Government initiatives attempted to incorporate the Bajo into what they perceived to be mainstream
Indonesia in the 1950s and1960s in an initiative to secure the peripheries of this vast archipelago.
Despite government efforts, the Bajo have remained isolated and, thus, sustain a great deal of their
163
ancestors cultural values and traditions. Minimal changes have been imported into the Bajo culture
from mainstream society and, as a result, they fit the UN working definition of an indigenous
population remaining isolated and marginal (Smith, 2004). Through sustaining their culture and
peripheral location in remote Indonesia, the Bajo have maintained their dependence upon marine
systems for food, lifestyles and traditional customs. Therefore, WMNP zoning schemes have the
most detrimental affect on the lives of the Bajo, as key stakeholders of the region (Elliott et al.,
2001; Smith, 2004).
The Bajo culture believes all individuals to be free and equal, thus instead of one designated leader,
several knowledgeable respected elders assume leadership responsibilities in Bajo communities.
Traditionally, the Bajo have had few economic opportunities and those that have been made
available are generally at a lower wage than the islanders receive, however economic activities
within the villages are gradually increasing. These include selling homemade products and farming
agar, as well as exploiting fish markets using more modern, high-efficiency fishing techniques, such
as gill nets. This is a consequence of increased exposure to externals and growing market forces
(Smith, 2004).
Traditionally the Bajo, similar to many indigenous peoples, fulfill roles dictated by gender, whereby
women pursue more domestic nurturing roles within the home incorporating low-key fishing
activities such as gleaning and net making, and men fulfill roles of provision, of both food and,
nowadays, income through more extensive fishing pursuits (Pedju et al., 2004; Smith, 2004). In Bajo
villages, three or more generations often share a stilt hut or at least live within close proximity.
Social capital and knowledge transfer is maximised through the establishment of traditional social
support networks strengthened by partaking in daily roles and activities together such as cooking
and eating.
Although the region is primarily Muslim, the Bajo have retained their animistic beliefs. Such beliefs
incorporate rituals that make offerings to sea spirits during rites of passage and to request a good
catch for example (Burns, 2002; Smith, 2004). Traditionally, the Bajo perceive daily fish catch to be
pre-determined by the spirits, contrary to any long-term understanding of marine systems and
sustainable fishing techniques they have acquired (Smith, 2004). Bajo spiritual beliefs are reflected
164
in traditional medical diagnoses and treatments and in the maintenance of Shamanistic practices. A
Bajo person is believed to get ill if they have disrespected the sea and its components or forgotten
their ancestors that have passed away whose spirits remain kindred with the waves (Nimmo, 1990;
Adi & Sudarman, 1994). In comparison with traditional techniques, Western medicine is a costly
alternative and the Bajo believe it only capable of curing ailments inflicted by God, not those by
spirits (Smith, 2004).
Bajo children rarely have access to the same educational opportunities as the Pulo. Those that do
attend school usually only do so until elementary level (Smith, 2004). Their traditional nomadic
lifestyles, whereby children would be paddling their own boats by the age of four and fishing at
seven years old, saw little use for and a cultural distance from formal education. Instead, education
consisted of the informal oral transfer of marine knowledge between generations combined with
personal experience (Adi & Sudarman, 1994).
So strong are Bajo ties to the sea that not only are their livelihoods and beliefs based around the
oceans, but also are their stories and names. It has been suggested that they are so in-tune with the
marine environment that they can navigate their way home using only the stars, locate fish
aggregations feeling the vibrations of their paddle and tell the day of the month based on the
seaweed budding cycle (Adi & Sudarman, 1994). Hence it is possible that no other indigenous tribe
have retained such a strong link with the marine environment throughout their history.
As a nomadic tribe, the Bajo employed traditional practices to fish areas before moving site, thus
rotating their resource exploitation. Today, however, the Bajo exploit only three types of fishing
ground; Palilibu situated close to home, Pongka involving a fishing trip of several days and Sakai,
the furthest fishing ground, often involving a trip lasting months at a time and a man taking his
entire family along. Due to extensive male absence from the home, women have long filled the roles
of household heads, with little choice but to budget the money, balance the diet and bring up the
children (Smith, 2004).
As a consequence of the introduction of modern techniques, markets and technologies, combined
with sedentary lifestyles and in-depth knowledge attuned to locating resources, pressure on local
165
marine systems of the region has intensified. This threatens to alter local incentives, social taboos
and accepted practices for resource collection. Although many Bajo men retain lifestyles of
migration pursuing seasonal work opportunities, local fishers have nevertheless seen some
invertebrate populations around the coast of Kaledupa decline, with less than 10% of current catches
comprising of mature specimens (COREMAP, 2002). Therefore, it is likely that local incentives to
fish will shift from subsistence exploitation to financially-driven overexploitation, threatening local
ecosystems, if economic development continues unabated in this region (Tomascik et al., 1997).
Operation Wallacea is a self-funded, research driven conservation organisation that was established
in the Wakatobi in 1995. Its environmental and cultural research programmes focus on and around
the islands of Hoga, Kaledupa and Lintea within the co-managed, resource-rich marine environment
in the northeast of the WMNP (Smith, 2004). Through its seasonal influx of students and
researchers, mostly between June and September, Operation Wallacea provides locals with an
alternative form of income seasonally. It contributes to the local economy directly by hiring local
staff and landlords, and indirectly by increasing local food demand (Operation Wallacea, 2004;
Smith, 2004). Operation Wallacea was the base for this research as an organisation with pre-existing
relations of trust with local communities. Operation Wallacea staff also arranged interview payment
schemes ensuring that partaking communities were fairly and equitably compensated for their time
invested in research.

Villages were chosen for study based on having maintained some level of resource dependence upon
local species for daily consumption and/or income generation. Both ethnic groups were represented
by the villages chosen as were a range of dependence/development levels. All villages selected had
pre-existing, workable relationships of trust with Operation Wallacea over an extended period
(excluding Lahoa, the third and final Bajo village of the island, which had not previously been
worked with).
166
A total of three Kaledupan villages were studied, Buranga Ollo and Kasuwari, and compared with
the only three Bajo villages surrounding the island, Sama Bahari (Sampela), Lahoa and Mantigola.
The latter two are classified for political purposes as sub-villages of Tanomeha and Horuo
respectively, but stand separately on the reef flat, retain cultural isolation and are independent
communities in their own right. Kaledupan villages were chosen from a selection which Operation
Wallacea had pre-agreed financial arrangements with, for their support in research. Buranga, Ollo
and Kasuwari were selected for their accessibility, their mixed population of both fishers and
farmers and for their economic differentiation from each other.
All Kaledupan respondents in this study had access to schooling and the opportunity to pursue
careers in administration. However, Buranga residents, as a result of being situated close to the
island capital of Ambeua, had more job opportunities than Kasuwari residents, who were situated
the furthest away. Opportunities to pursue careers in administration were, on the whole, not
available to Bajo villagers and schooling was far less accessible. Lahoa residents had no access to
formal education. Sampela had an established primary school although it was not formally in use
due to poor attendance by both students and teachers alike. Mantigola, on the other hand, had an
established and functional primary school and many students here pursue secondary education in the
nearby sub-village of Umala on land (Table 5.2.1).
Table 5.2.1. Demographics and characteristics of villages selected for study
Village
Total
Total
Total
Total
Primary
populati mena
womena
househ household
ona
oldsa
incomeb
Buranga
998
469
509
316
6.1
Ollo
1265
609
656
324
5.4
Kasuwari
1019
496
523
288
3.0
Sampela
1106
552
554
251
3.4
Mantigola
1131
555
576
217
2.8
Lahoa
135
67
68
30
1.3
a
from the Chamat Kaledupa (2005)
b
mean values from Pilgrim et al. (2006b) (income in millions Rp)
% Natural
resource
dependentsb
53
69
62
80
73
75
Ethnic
group
Location
Pulo
Pulo
Pulo
Bajo
Bajo
Bajo
Coastal
Coastal
Coastal
Reef
Reef
Reef

Species categories used in the identification exercise included marine species, terrestrial animals,
birds and terrestrial plants (this was not divided into wild and crop plants as many wild species are
167
harvested). Experts well-acquainted with the Wallacea region were asked to advise on the most
common local species sighted on the island and in its surrounding waters. Dr Smith (Director of
Marine Research for Operation Wallacea), Dr Morrison (lecturer in plant biology and supervisor on
a number of field trips to the region), Dr David Kelly (a post-doctoral researcher familiar with the
study site) and Dr Julian Clifton (Director of Social Research for Operation Wallacea) were all
consulted on species flashcard compilation. The final species collection included 24 marine species,
11 terrestrial animals, 14 birds, and 25 flowering plant species.
Internet search engines were used to find the highest quality photographs of each of the 74 species.
These photographs were then used to make flashcards where each species was labeled only by
number. Again, where necessary, two or more images were used for one species to represent all of
its components, such as the fruits and leaves of a plant. Prior to testing, four experts were consulted
to verify the presence of all species at the study site, and along with local community elders, helped
compile a full list of local species names in Bahasa Kaledupa (spoken exclusively on Kaledupa),
Bahasa Indonesia (the national language) and Bajo (spoken exclusively by the Bajo). Only two
species were identified as being absent from the region and therefore eliminated. The final
compilation of 72 species is included in Appendix IV.

Sampling quotas were calculated with the aim of attaining a minimum of 30 per sub-sample making
data viable for statistical testing (Roscoe, 1975). Sampling 10% of the population was not viable
since total population sizes of the villages were unknown prior to interviewing and no data existed
on the population proportions represented by different age groups. Therefore, using minimum
sample sizes ensured statistical viability whilst representing all sub-populations. Also, limiting
within-village sample sizes permitted horizontal expansion of the study across a greater number of
villages, allowing for inter-community comparisons in knowledge to be made.
Men and women were sampled evenly across the four age groups. The same age groups were
employed as were used at the UK sites (15-19 years, 20-29 years, 30-49 years and 50 years and
over). This enabled intra-community patterns (such as progressive loss) to be compared between the
UK and Indonesia. On average, 32 interviews per village took place, across six villages, resulting in
168
a final sample size of 192 (Table 5.2.2). This meets Schoemakers (2004) specification rule that the
total sample size for any form of social sampling should exceed 100.
Due to the small size of Lahoa (30 households), combined with the need to sample respondents from
independent households, the quota size for sampling from this village was halved to 16 respondents
and the sample sizes for the other two Bajo villages were increased to 40 to compensate. To assess
the impact of primary occupation, farmers and fishers were sampled at equal ratios within the
Kaledupan villages where these two primary occupations persist (all Bajo are fishers, very few have
access to land). Interviews were carried out individually during a two month field visit between July
and September 2005.
Sampling was purposive. Upon approaching a household a particular individual was requested
specifying age, gender and occupation where relevant. Every second household within a village was
approached until sampling quotas were filled. If a person of that particular specification did not live
there or was unwilling to be interviewed then the next selected household was approached.
Interviewing respondents from the same household was avoided to prevent intra-household transfer
concealing differences in community knowledge levels. Interviews were carried out at all times of
day, every day of the week, to ensure that no working groups were eliminated. Where possible,
interviews were carried out within participants houses to avoid external input from third parties. In
some cases where it was proving difficult to find the final respondents to fill village quotas,
haphazard sampling using the snowball technique was employed.
Table 5.2.2. Breakdown of site sampling strategy: Village sub-sample quotas
Age
Gender
group
15-19 Men
15-19 Women
20-29 Men
20-29 Women
30-49 Men
30-49 Women
50+
Men
50+
Women
Village totals
Ethnic totals
Site total
Buranga
fish farm
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
32
Ollo
fish farm
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
32
96
Kasuwari
fish farm
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
32
Sampela
all fish
5
5
5
5
5
5
5
5
40
192
Mantigola
all fish
5
5
5
5
5
5
5
5
40
96
Lahoa
all fish
2
2
2
2
2
2
2
2
16
169
One primary translator was designated to the study, who was fully briefed as to the aims of the
research and the importance of not influencing the respondents answers or giving them hints. Using
just one translator ensured consistency of translation (Twyman et al., 1999). The translator spoke
Bahasa Indonesia and Bahasa Kaledupa. In the Kaledupan villages, a local guide (usually a wellrespected villager) was used upon request of the Kepala Desas from whom permission was sought
prior to interviewing. Being accompanied by a familiar, well-respected local ensured that a
relationship of trust and openness was created. Local guides also provided information on village
residents, cultural taboos and traditional customs. In the Bajo villages, two different guides were
employed for use across the three villages. Bajo guides facilitated translation between Bahasa
Indonesia and Bahasa Sama (Bajo). Both guides were highly competent in Bahasa Indonesia as well
as Bajo and had the aims of the research and their roles as translators outlined to them prior to
interviewing.

As at the other sites, species identification was used as a measurement strategy for LEK, enabling
knowledge levels to be compared with socioeconomic and demographic data collected. Structured
interviews were constructed and reviewed by Dr Clifton (Director of Social Research for Operation
Wallacea). In addition to measuring LEK, they collected quantitative data on awareness, attitudes
and perceptions, changes in resource levels, local behaviour and economic wealth. Interviews were
divided into four sections including case demographics, local knowledge assessment, structured
questions on livelihoods and primary knowledge source, and Likert-scale questions (Laine et al.,
2004). Experienced translators knowledgeable about local norms were asked to advise on interview
content. If any questions could possibly cause offence to local villagers (such as enquiry into level
of schooling) or were too complicated, they were either removed or re-written. Interviews were
piloted prior to testing.
The material wealth of each respondent was scored based upon the floor, roof and wall materials of
their house and ownership of consumer goods. If this could not be done by eye, respondents were
then asked these details. Village of residence, gender, age and primary occupation of each
participant was recorded. The respondent was asked if they had any alternative sources of income
170
and if they had lived in the village all of their life. The species identification exercise followed. For
each card the respondent was asked if they recognised the species and, if so, to name the species and
what, if anything, it could be used for.
Following the species identification task, respondents were asked a series of closed questions. These
included who taught them about the local plants and animals, whether they share this knowledge
with other locals, what methods they use to fish/what crops they grow and what they consider to be
the most important threats to the sea/land. Respondents were then asked if they could distinguish
between a bomb-caught fish and a fish caught by a non-destructive technique and, if so, whether
they would purchase the fish caught by destructive methods. This question looked at local attitudes
towards destructive fishing. Finally, respondents were asked whether they had been visited by any
other organisations and Bajo respondents were asked the frequency of their visits to land.
For each of the above questions, a list of categorical responses was compiled from the pilot
interviews. This saved time when recording answers in the field (instead of recording full answers a
box could simply be ticked). Where answers were given that were outside of these categories, they
were recorded separately to avoid forcing respondents answers into etic categories. Closed
questions ensured that translation was straightforward and questions were not open to interpretation,
especially where answers were being translated twice before being recorded. The data recording
sheets used for fishers and farmers during interviewing can be found in Appendix VII.
The concluding section of the interview included 26 Likert-scale questions. The respondents were
asked whether they strongly agreed (sangat setuju), agreed (setuju), were indifferent (ragu),
disagreed (tidak setuju) or strongly disagreed (sangat tidak setuju) with each statement read to them.
On the whole, this concept was well understood by respondents. The Likert-scale questions were
randomly ordered and assessed a variety of the respondents views and behaviours. These included
environmental understanding and value, view of knowledge importance, pro-environmental
behaviour, desire for future continuity, self-efficacy, marine park understanding and view, utilitarian
value of the environment, view of environmental deterioration and support for traditional practices.
Interviews lasted between 20 and 55 minutes.
171
Additional market surveys were carried out with six local sellers, evaluating the market values of all
saleable species used within the knowledge assessment. Values for marine species were recorded
per kilogram, values for livestock per animal and values of marketable crops per fruit. Therefore for
marketable species, the impact of local economic value on species familiarity was assessed.

All data were entered into a database using SPSS program 12.0. Once data entry was complete,
editing programs could be used, frequency distributions constructed, data exploration carried out to
ensure the data were free of inconsistencies and statistical tests run. All data sets were tested for
normality (using Kolmogorov-Smirnov test) and variance (by calculating coefficients of variance).
The majority of data were not normally distributed warranting non-parametric statistical testing,
with the exception of plant use knowledge.
The ecoliteracy of each respondent was measured by calculating the % of local species shown to
them that they were able to identify. Local knowledge levels of each species category were also
calculated, including marine species knowledge, terrestrial plants and terrestrial animals. Use
knowledge was calculated as the total number of uses listed and mean number of uses given per
identified species. As well as this basal analysis, knowledge data were further analysed to see if
certain groups of respondents held specialised knowledge of particular species groups and their uses.
Comparisons between individuals of different genders, ethnic groups and primary occupations were
made using Mann Whitney-U (parametric t-test for plants uses). Comparisons between individuals
in different age groups, from different villages, relying on different knowledge sources and with
different wealth ranks were made using Kruskall-Wallis (parametric one-way analysis of variance
for plants uses).
Wealth rank was calculated through a scoring system whereby the lowest score was 1 for the
cheapest material and 3 for the most expensive. For instance, for having walls constructed from
palm a respondent would score 1, compared with 2 for wooden walls and 3 for concrete. Extra
points were added for the ownership of consumer goods including a television and stereo. Scores
172
were then totaled to give a figure ranging from 4 to 11 indicating the level of wealth of the
respondent, termed the wealth rank.
Associations between knowledge and Likert-scale findings were tested for using Spearmans rank,
as were associations between different Likert-scale factors including environmental understanding
and value, view of knowledge importance, pro-environmental behaviour, desire for future
continuity, self-efficacy, marine park understanding and view, utilitarian value of the environment,
view of environmental deterioration and support for tradition. Any cultural differentiation in Likertscale scores was tested for using Mann Whitney-U. Spearmans rank was used to test for
associations between number of alternative income sources and LEK level, and number of fishing
methods employed/crops grown and LEK level.
To compare gender, ethnic and occupational differences in knowledge of consumption, income,
ornamental, health, farming, religious, construction and other uses knowledge, Mann Whitney-U
was used. These use categories were derived from respondents most frequent answers. Age group
differences in knowledge of these uses were tested for using Kruskall-Wallis. Finally, an association
between market value and species familiarity was tested for using Spearmans rank.
5.3. Results
The results of this section were aimed at testing the importance of ecological knowledge to resource
dependent societies and the impact of economic development on community knowledge bases from
this region. The importance of gender-differentiated roles to experience and knowledge levels was
looked at. Also, distribution of community knowledge with age was examined as was the impact of
primary occupation upon level and content of knowledge.
Two different levels of ecological knowledge were measured; (i) ecological literacy (in terms of
identification ability indicating species familiarity) and (ii) use knowledge (number of local resource
uses listed indicative of resource reliance). Knowledge level (i) is the primary level of resource
familiarity and level (ii) is the secondary level. To possess level (ii), the person must already hold
level (i), but this does not hold true in reverse. Differences in ecological literacy and use knowledge
173
were determined between different groups within communities (intra-community), as well as

between different communities (inter-community). Both intra- and inter-community knowledge
appeared to be heterogeneous.
5.3.1 The effect of ethnic group

The effect of ethnic divisions upon ecological knowledge was examined by comparing ecoliteracy
and use knowledge between Kaledupan and Bajo respondents. Total ecological literacy of Bajo and
Kaledupan respondents was not significantly different, both were able to name over 73% of their
local species. However, Bajo use knowledge (number of uses listed) was significantly higher than
Kaledupan despite high knowledge variation in both groups (U=2384.000, p<0.001, n=192), as was
the number of uses listed per identified species (U=2384.500, p<0.001, n=192) (Fig 5.3.1).
Of the uses listed, the number of economic uses listed by the Bajo for local resources was highly
variable, but still significantly exceeded the number of economic uses identified by Kaledupan
respondents (21 and 4 uses respectively) (U=746.500, p<0.001, n=192). However, Kaledupans knew
significantly more consumption uses than Bajo respondents (37 and 31 uses respectively)
(U=1761.500, p<0.001, n=192) (Fig 5.3.2). Despite a lack of formal schooling, the use knowledge
of Bajo children (aged 15-19) significantly exceeded the knowledge of formally-educated
Kaledupan children (U=108.000, p<0.001, n=48) (Fig 5.3.3).
Although no ethnic differences in total ecoliteracy were detected, the knowledge levels of different
species groupings differed between ethnic groups. Significantly higher levels of knowledge of local
terrestrial species (both plants and animals) were detected amongst Kaledupan respondents, able to
174
90
A
80
A
70
60
50
40
30
20
Bajo
Kaledupan
Ethnic group
B
listed per identified species)
Use knowledge (number of uses
2.0
1.5
1.0
0.5
Bajo
Kaledupan
Ethnic group
Figure 5.3.1. Differences in levels of use knowledge

indicated by (A) total number of uses listed and (B)
number of uses known per identified species between 2
ethnic groups (median + interquartile range, n=192).
Symbol (o) indicates outliers
175
Economic use knowledge (number

of economic uses listed)
50
A
40
30
S
S
S
S
S
20
A
A
A
A
A
A
10
0
Bajo
Kaledupan
Consumption use knowledge

(number of consumption uses listed)
Ethnic group
50
B
40
30
20
A
10
0
Bajo
Kaledupan
Ethnic group
Figure 5.3.2. Differences in (A) economic use knowledge

and (B) consumption use knowledge between 2 ethnic
groups (median + interquartile range, n=192). Symbol (o)
indicates outliers, symbol (*) indicates extreme outliers
176
90
80
70
60
50
40
30
20
Bajo
Kaledupan
Ethnic group
Figure 5.3.3. Differences in the use knowledge of children

(aged 15-19 years) between 2 ethnic groups (median +
177
name 10% more plants and 6% more animals than the Bajo (plants; U=2494.000, p<0.001, n=192,
animals; U=3369.500, p<0.01, n=192), although the Bajo showed larger variation in terrestrial
knowledge. Conversely, Bajo respondents revealed significantly higher levels of marine knowledge
than Kaledupan respondents in that they were able to name 7% more species (U=1305.000, p<0.001,
n=192), although Kaledupans showed larger variation in marine knowledge (Fig 5.3.4).
Amongst the Bajo, self-efficacy was positively associated with pro-environmental behaviour
(Rs=0.279, p<0.001, n=192), which was found to significantly increase with increased level of
marine ecoliteracy (Rs=0.253, p<0.05, n=96). Bajo knowledge accumulation was found to occur
earlier in life (pre-15 years) than Kaledupan acquisition. Hence Bajo saturation occurs at an earlier
age (maximum 15 years) than Kaledupan (maximum 30 years) (Fig 5.3.5). Many Bajo men reached
total marine knowledge saturation, able to recognise and identify 100% of local marine species
shown to them. This was unseen in other groups (Fig 5.3.6).
Support for local traditional practices was found to significantly increase with increased knowledge
of marine species uses (Rs=0.396, p<0.001, n=192). Hence Bajo fishers exhibit significantly higher
support for traditional practices than Kaledupans (U=1802.500, p<0.001, n=192) (Fig 5.3.7).
However, support was found to decline in Bajo respondents with increased levels of terrestrial
knowledge (Rs=-0.293, p<0.01, n=96). Both ethnic groups, particularly the Kaledupans, showed
considerable variation in their support for local traditions. In terms of fishing methods, the Bajo
were found to maintain traditional low intensity fishing practices, such as spear fishing, at a level
that far exceeds that of Kaledupan fishers (Table 5.3.1).
Table 5.3.1. Levels of fishing practices exploited by two ethnic groups and details of the practices
Fishing
method
Spear
Bubu
Fish fence
% Bajo that
employ method
60.4
1.0
0.0
% Kaledupans that
employ method
16.6
20.4
12.6
Details
Traditional, low intensity, highly selective
Modern, high intensity, non-selective
Modern, high intensity, non-selective
As well as support for local traditions, pro-environmental behaviour and marine park understanding
revealed a significant positive association with LEK level (Rs=0.156, p<0.05, n=192; Rs=0.245,
178
Plant ecoliteracy (% plants identified)
100
90
80
70
60
50
A
40
A
A
30
20
Animal ecoliteracy (% animals identified)
Bajo
Kaledupan
100
B
90
80
70
60
50
40
30
20
Bajo
Marine ecoliteracy (% marine species identified)
Ethnic group
Ethnic group
Kaledupan
100
90
80
A
70
60
A
A
S
S
50
A
40
A
A
30
20
Bajo
Ethnic group
Kaledupan
Figure 5.3.4. Levels of ecoliteracy of (A) terrestrial plants,

(B) terrestrial animals and (C) marine species between 2
ethnic groups (median + interquartile range, n=192).
Symbol (o) indicates outliers, symbol (*) indicates extreme
outliers
179
80
75
A
B
B
B
70
Kaledupan
Bajo
65
15-19yrs
30-49yrs
20-29yrs
50+yrs
Age group
Figure 5.3.5. Comparisons of the mean changes in total

ecoliteracy with age between 2 ethnic groups showing the
maximum age of knowledge saturation for both groups
(mean + 2SE, n=192). Symbol (*) indicates maximum age
Marine ecoliteracy (% marine species

identified)
of ecological knowledge saturation
100
80
60
40
20
0
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
Figure 5.3.6. Levels of Bajo mens marine ecoliteracy

across all 4 age groups measured illustrating knowledge
saturation in the 3 eldest groups (median + interquartile
range, n=48). Symbol (*) indicates extreme outliers
180
Support for local traditions
1
Bajo
Kaledupan
Ethnic group
Figure 5.3.7. Levels of support for local traditions

(ranked) across 2 ethnic groups (median + interquartile
range, n=192). Score 5 indicative of respondents with the
highest support for traditional practices and score 1 the
least
181
p<0.01, n=192 respectively). However, desire for future continuity was found to be inversely related
to marine species knowledge (Rs=-0.191, p<0.01, n=192).
There was no significant difference detected in knowledge source contribution between the two
ethnic groups, with over 85% of both respondent populations claiming that their primary knowledge
source was themselves. However, 7% of Bajo respondents claimed that family are important to
knowledge acquisition compared with only 2% of Kaledupans, and 7% of Kaledupans claimed that
school is important compared with only 1% of Bajo.

The effect of income on knowledge was investigated by comparing (i) intra-community knowledge
levels of respondents from different wealth ranks and (ii) inter-community knowledge levels of
villages from different average income levels. There were significantly higher levels of wealth
amongst Kaledupan communities (U=2449.000, p<0.001, n=192). Wealth rank was found to be
inversely related to use knowledge (Rs=-0.157, p<0.05, n=192) and delay the age by which
knowledge saturation occurs (Rs=0.413, p<0.001, n=192). Therefore, wealthier individuals (for
instance from Kaledupan communities) on the whole have lower levels of species use knowledge
and reach ecological knowledge saturation later in life (Fig 5.3.8). Despite local economic
dependence upon natural resources, no association was detected between species familiarity (% of
people able to identify a particular species) and its market value, for both marine and terrestrial
groups.
The six chosen villages showed no significant difference in overall ecological literacy level, but
differed significantly in use knowledge (H=67.780, p<0.001, n=192). There was found to be an
inverse relationship between use knowledge and mean primary household income of a village (Rs=0.211, p<0.01, n=192), with the exception of Sampela which far exceeded expected knowledge
levels for its mean level of income (Fig 5.3.9). Mean use knowledge was also found to significantly
increase with increased levels of community resource dependence (Rs=0.505, p<0.001, n=192) (Fig
5.3.10).
182
15
A
13
Wealth rank
Max SA=20
Max SA=30
11
3
Bajo
Kaledupan
Ethnic group
90
A
80
A
70
60
50
40
30
20
Bajo
Kaledupan
Ethnic group
Figure 5.3.8. Comparing Bajo and Kaledupan (A) mean

levels of wealth with maximum age of knowledge
saturation and (B) mean use knowledge (median +
interquartile range, n=192). Symbol (o) indicates outliers.
Max SA indicates maximum age of knowledge saturation
Mean community use knowledge

183
70
]
Sampela
Lahoa
60
Kasuwari
]
]
Mantigola
Ollo
]
50
Buranga
]
40
1000
2000
3000
4000
5000
6000
Mean household income (1000Rp)
Figure 5.3.9. Mean levels of community use knowledge in

6 villages and relationship with mean primary household
income (mean + 2SE, n=192). Mean household income is

based solely on the income of the primary earner
70
]
60
]
]
50
40
50
60
70
80
90
% Resource dependents in community
Figure 5.3.10. Mean levels of community use knowledge

in 6 villages and relationship with the % of resource
dependents within a community (mean + 2SE, n=192)
184
The variance in ecoliteracy between the most and least knowledgeable groups within a community
(measured by coefficient of variance) increased significantly with mean primary household income
(Rs=0.882, p<0.001, n=192) (Fig 5.3.11). Community knowledge variance was also found to be
inversely correlated with mean community use knowledge. Thus poor but ecoliterate communities
like Lahoa have lower knowledge variance between individuals, but where ecoliteracy is lower such
as in the wealthier village of Buranga, the difference in knowledge level between experts and nonexperts is far greater (Rs=-0.438, p<0.001, n=192) (Fig 5.3.12).

Gender differences in ecological knowledge were examined by comparing the ecoliteracy and use
knowledge of men and women from all age groups. Although both men and women showed high
variation in knowledge, men had significantly higher levels of ecological literacy than women
(U=2601.000, p<0.001, n=192) (Fig 5.3.13). This was true for local knowledge of animals
(U=3393.500, p<0.01, n=192), birds (U=1915.500, p<0.001, n=192) and marine species
(U=3008.500, p<0.001, n=192), but not for plants. However, whilst Bajo mens knowledge of local
uses was significantly higher than womens (U=883.500, p<0.05, n=192) despite high variation in
both genders (Fig 5.3.14), Kaledupan use knowledge did not significantly differ between the sexes.

Age-based patterns of ecological knowledge distribution were looked at by comparing the
ecoliteracy and use knowledge of respondents from four different age groups. Despite high variation
across all ages, use knowledge was found to significantly differ between the four age groups
(H=11.972, p<0.01, n=192), unlike total ecoliteracy which did not vary with age (Fig 5.3.15).
Despite this, both measures of ecological knowledge showed a similar trend in knowledge
accumulation with age, peaking between the ages of 20 and 29 and declining in the later years of
life. Ecoliteracy accumulation with age was found to be more rapid in men than women and
maximum age of knowledge saturation was lower for men (Fig 5.3.16).

There was no significant difference in total ecological literacy or use knowledge between fishers and
farmers. However, when examining knowledge content, farmers knowledge of terrestrial plants and
185

21
A
19
17
A
15
13
11
5
1000
3000
2000
5000
4000
6000
Mean household income (1000Rp)
Figure 5.3.11. Levels of variance in community

ecoliteracy in 6 villages and relationship with mean
primary household income (n=192). Mean household
income is solely based on the income of the primary
earner

A
20
A
16
A
12
A
8
A
A
45
50
55
60
65

(mean number of uses)
Figure 5.3.12. Levels of variance in community

ecoliteracy in 6 villages and relationship with mean levels
of community use knowledge (n=192)
186
100
90
80
70
60
A
A
A
50
40
women
men
Gender
Figure 5.3.13. Comparing the levels of total ecoliteracy

between men and women (median + interquartile range,
90
80
70
60
50
40
30
20
women
men
Gender
Figure 5.3.14. Comparing the levels of Bajo use

knowledge between men and women (median +
187
90
80
70
60
50
40
30
20
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
100
B
90
80
70
60
50
40
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
Figure 5.3.15. Comparing patterns of age group

differentiation in (A) total use knowledge and (B) overall
ecoliteracy levels (median + interquartile range, n=192).
Symbol (o) indicates outliers
Ecoliteracy (% total species named)
188
80
75
men
70
A
A
women
65
15-19yrs
20-29yrs
30-49yrs
50+yrs
Age group
Figure 5.3.16. Comparisons of the mean changes in

ecoliteracy with age between men and women showing
the maximum age of knowledge saturation (mean + 2SE,
n=192). Symbol (*) indicates maximum age of ecological
knowledge saturation
189
animals was significantly higher than fishers (plants; U=870.000, p<0.05, n=96, animals;
U=884.500, p<0.05, n=96) (Fig 5.3.17). No such difference was found between fishers and
farmers marine ecoliteracy. However, fishers marine ecoliteracy level was found to significantly
increase with the number of fishing methods that they use (Rs=0.245, p<0.01, n=144).
190
100
A
90
80
70
60
50
40
30
Fisher
Farmer
Animal ecoliteracy (% animals identified)
Primary occupation
100
90
80
A
70
A
60
A
50
40
30
Fisher
Farmer
Primary occupation
Figure 5.3.17. Comparing fishers and farmers levels of

ecoliteracy of (A) terrestrial plants and (B) terrestrial
animals (median + interquartile range, n=192). Symbol (o)
indicates outliers
191
5.4. Discussion

The pattern of Bajo use knowledge exceeding that of Kaledupan communities, even in uneducated
children, was evidence for the importance of secondary ecological knowledge in communities with
continued resource dependence (Gadgil et al., 1993; Pretty, 2002; Davis & Wagner, 2003). It also
indicates that formal schooling most likely acts to hinder local knowledge transfer between
generations as previously suggested by Kothari et al. (1998) and Pyle (2001). Despite this, there was
no difference in overall ecoliteracy level (primary knowledge) between the two groups. This is most
likely indicative of historical resource dependence. For instance, Kaledupans are able to identify the
local ecosystem components they once relied upon, but they can no longer recall the full range of
uses they were once employed for. This most likely represents the early stages of ecological
knowledge loss.
Intense natural resource dependence for income purposes by the Bajo explains the wide range of
economic uses they were able to list, as they have few income alternatives (Smith, 2004). The Bajo
rely solely on marine resources for protein, but will not eat any marine species that can be sold for
profit. Instead, they sell the tasty catch and consume the less palatable species that lack market
value. Kaledupans, on the other hand, listed less economic uses and, instead, a diverse array of
species for consumption. This reflects their increased wealth, enabling them to consume both the
more and less palatable species, as well both marine and terrestrial sources of protein.
Differences detected in ecoliteracy content between the two ethnic groups demonstrated the
importance of ecosystem interaction to knowledge acquisition. Differences in ecosystem interaction
between the two groups are most likely derived from different social norms and daily activities
(May, 2005). For instance, the social norms and daily subsistence patterns of the Bajo dictate their
ties to the sea and all of its products (Akimichi & Supriadi, 1996; May, 2005), enhancing their
maritime knowledge. Only non-traditionalists amongst the Bajo frequent the land and hold any landbased knowledge (traditionally nomadic Bajo tribes only set foot on land to bury dead relatives.)
Similarly, heightened terrestrial knowledge by Kaledupans stems from their land-based lifestyles
192
and agricultural histories (May, 2005), although many retain some level of maritime knowledge
through their own fishing pursuits.
Self-efficacy is indicative of a self-belief in the impact of a persons own behaviour (Meinhold &
Malkus, 2005). Increased self-efficacy with increased maritime knowledge amongst the Bajo most
likely reflects their increased understanding of the impact that human behaviour can have on the
environment as a result of intense ecosystem dependence over a long period. For instance, a person
with in-depth knowledge and extensive experience of local reef systems has most likely observed
changes in the system under human exploitation and, thus, comprehends their own capacity to elicit
the same sorts of changes. Kaledupans lack the same level of understanding and connectedness that
results from continued resource dependence by the Bajo (Adi & Sudarman, 1994), and thus do not
express this same relationship.
The pattern of pro-environmental behaviour being positively correlated with LEK level is most
likely a consequence of LEK creating a deeper affinity with nature which, in turn, arouses a desire to
act positively towards local wildlife. A similar positive relationship was revealed between marine
park understanding and LEK level. This can most likely be explained by frequent marine park users
(i.e. the most resource dependent) acquiring the highest levels of marine resource knowledge as a
result of daily ecosystem experience and, at the same time, relying on an in-depth understanding of
marine park zones and regulations in order to access these resources.
An inverse relationship between desire for future continuity and marine knowledge is most likely a
consequence of resource dependent traditionalists, knowledgeable about their marine species,
having a strong desire for change towards more economically-secure and stable lifestyles. The most
resource dependent groups that utilise local knowledge on a daily basis (primarily Bajo fishers) also
revealed strong support for local traditions, in particular Shamanistic medical practices. The high
cost of modern facilities, such as healthcare and schooling, in comparison with cost free local
practices most likely causes this level of support (Jodha, 1986).
However, those that have already departed from traditional fishing lifestyles, and subsequently
traditional knowledge bases, are likely to have increased food security and improved quality of life
193
as a result of raised income levels, and as a result, they desire this level of economic stability and
development to continue into the future. For instance, Kaledupans deviation from traditional
lifestyles and resource dependence is reflected by their reduced support for local traditional
practices. However, differing levels of income exist within Kaledupan communities, most likely
causing the high variation observed in support for local traditions. Maintenance of traditional low
intensity fishing practices is further evidence for Bajo poverty and subsequent resource dependence.
Despite differences in resource dependence and knowledge levels, members of both ethnic groups
claim themselves to be their own primary sources of ecological knowledge acquisition. This
highlights the importance of personal observation and experience to ecological knowledge gain in
this region, most likely derived from a history of intense natural resource dependence, even amongst
the Kaledupans whose dependence has waned with recent economic development (Wenzel, 1999;
Turner et al., 2000; Brodt, 2002; Harmon & Loh, 2004; Rahman, 2004).
A pattern of early knowledge acquisition and, thus, early saturation by the Bajo was evident. This is
indicative of intense resource dependence by the Bajo, whereby having high levels of knowledge
from a young age could mean the difference between survival and struggle (Adi and Sudarman,
1994; Pretty, 2002). In such remote, resource dependent communities, knowledge is often acquired
at an early age (pre-15 years) (Smith, 2004). This suggests that it is gained through both experience
and oral transfer in the home during childhood prior to extensive field experience (Alexiades, 1999;
Turner et al., 2000; Singh & Singh, 2005). Total marine knowledge saturation of Bajo men, the
primary resource providers (Pedju et al., 2004; Smith, 2004), demonstrates the importance of regular
interaction and experience, through societal roles, to the acquisition of ecological knowledge
(Nabhan & Trimble, 1994; Kellert, 2002; Pyle, 2002, 2003).

Indicative of reduced resource dependence and increased external purchasing power, Kaledupan
communities had, on the whole, a higher wealth ranking than the Bajo. The wealth ranking used was
indicative of material wealth and comfort rather than monetary wealth assessed through household
income. The direct correlation between wealth rank and age of knowledge saturation, and inverse
association with use knowledge, clearly demonstrates the relationship between wealth, resource
194
dependence and resource knowledge. Where overall wealth is low, as in Bajo villages, community
members are forced to acquire high levels of LEK at a young age in order to supplement both food
and income (Adi & Sudarman, 1994). Where this is not the case and daily nutritional needs can be
met by purchasing goods at a local market, communities can afford to acquire knowledge and build
up observations gradually over time. Therefore, in the wealthier Kaledupan communities resource
use knowledge is generally lower and acquired later in life. It seems likely that all species of market
value are relied upon (either bought or sold) at the study site, not just those of high economic value,
since market price is not indicative of local species familiarity.
Overall, a high level of ecoliteracy (species identification ability) was revealed across all villages,
indicative of a history of resource dependence in all communities (May, 2005). Therefore, the ability
to recall species names has persisted as a remnant of a once rich ecological knowledge base, even in
more economically developed communities. Hence high ecoliteracy levels can prevail in
communities with differing levels of current resource dependence, providing they share a not too
distant history of resource reliance.
Resource use knowledge, on the other hand, appears only to persist where natural resources are still
heavily relied upon for their goods and services. This is clearly demonstrated by use knowledge
being inversely related to mean primary household income, but directly related to the proportion of
resource dependents within a community. The village of Sampela had higher knowledge levels than
expected for its level of income (Fig 9). This may be symptomatic of a recent deviation from natural
resource reliance in the light of recent economic development that has not yet yielded sufficient time
for loss of local use knowledge, or may simply be caused by survey discrepancies if villagers have
exaggerated their income levels.
Community knowledge variance showed a pattern of incline with increased primary household
income and decline with increased use knowledge of a village. This further supports the
hypothesised importance of ecological knowledge transfer between all community members in
resource dependent villages. In Buranga where income is high and, therefore, dependence upon
local resources and knowledge is low, the gap in knowledge between the most and least
knowledgeable can afford to be large. This is because locals are no longer reliant upon this
195
knowledge for survival and, instead, rely on market purchases to meet their daily nutritional
requirements. Whereas the villagers of Lahoa, with the lowest income levels and highest
dependence upon local resources, cannot afford to have low levels of species use knowledge.
Instead, all villagers have high levels of knowledge on how to use local species to meet the daily
resource needs of themselves and their families. Therefore, it appears that resource dependence, a
product of local income, not only dictates ecological knowledge level, but also the levels of
knowledge variance within a community.

Mens local knowledge was found to exceed womens across most species groups (birds, animals
and marine species). This demonstrates the importance of gender roles in knowledge acquisition
(Garro, 1986; Bryson, 1996; Alexiades, 1999; Turner et al., 2000; Pfeiffer, 2002; Westermann et al.,
2005). It appears that mens primary roles as the resource providers, through hunting and resource
collection (Smith, 2004), provide them with increased opportunities to experience nature and,
therefore, enhance their knowledge of local ecosystems. This is compared with womens primary
roles based in the home raising children and caring for family members (Pedju et al., 2004). Mens
travel further a field also permits increased interactions with residents from other villages, and even
other islands, increasing opportunities for external knowledge transfer and acquisition. However,
mens increased experience of nature did not significantly increase their knowledge of wild plants,
most likely due to the uncharismatic nature of plants compared with birds or other animals,
combined with womens reliance upon them in the home for cooking and medicinal treatments.
Such gender differentiation, however, was only exhibited by Bajo men and women, where
traditional lifestyles and gender-differentiated roles still persist (Pedju et al., 2004; Smith, 2004).
The use knowledge of Kaledupan men and women did not significantly differ. This is further
evidence that use knowledge is a product of societal roles that no longer exist in many Kaledupan
communities. Thus knowledge levels within these communities no longer reflect the genderdifferentiated roles of resource collection they may well have done in the past. Therefore, when
investigating gender differentiation in ecoliteracy, traditional cultural roles must be taken into
account if these patterns of knowledge are to be fully understood.
196
Unlike ecoliteracy, use knowledge significantly varied with age. However, both measures of
knowledge showed a similar pattern of acquisition with age; peaking between the ages of 20 and 29,
and then slowly declining in later years (Fig 15). This pattern is most likely caused by the most
active age group (20-29 year olds) travelling the furthest to collect resources and experimenting with
more exploitation methods, subsequently acquiring the most diverse ecosystem experience and
knowledge. Prior to this age, LEK acquisition is gradually increasing. After this age, resource
collection continues, but is likely to stay more local (due to family responsibilities) and specialised,
as the younger, fitter villagers continue to travel the furthest and experiment with more modern
techniques. In addition, knowledge starts to be forgotten later in life, symptomatic of natural
memory loss with age (Krech, 2005), resulting in further ecological knowledge decline.
The Indonesian communities studied demonstrate this pattern well, although acquisition with age
was found to be more rapid in men than women. This is again indicative of womens reduced
opportunities to experience local wildlife due to their household responsibilities (Pedju et al., 2004;
Smith, 2004), unlike men who generally have daily encounters with a wide range of species
enabling them to build up knowledge of local ecosystems more rapidly.

Although no overall difference in the ecological knowledge levels of fishers and farmers was found
to exist, knowledge content was found to differ. The enhanced terrestrial knowledge of farmers is
clearly indicative of their increased experience of the land and its components compared to fishers.
There was no evidence for enhanced maritime knowledge in fishers, most likely due to many
farmers of Kaledupa embarking upon fishing in their free time to supplement family protein
requirements (May, 2005). Therefore farmers gain experience, and subsequent knowledge, of both
ecological systems.
Increased marine knowledge in individuals that used a wide variety of fishing techniques
demonstrates the importance of diversified experience in knowledge acquisition. Employing a range
of fishing practices leads to diversity of catch and fishing ground exposure, enhancing local
observations and, therefore, knowledge. This provides further evidence that community roles,
197
responsible for frequency and type of environmental interaction, are clearly linked to an individuals
own knowledge level as well as community knowledge distribution.
This investigation provided quantifiable evidence that level of resource dependence (a product of
wealth) is directly related to an individuals knowledge level and is a key predictor of LEK
distribution within and between communities. In addition, societal roles in resource collection were
found to be crucial to explaining intra-community knowledge differences. However, to be sure that
these results reflect the true relationship between resource dependence and ecological knowledge
globally, cross-country comparisons need to be carried out. The following chapter compares the
differences observed in ecoliteracy between India, Indonesia and the UK to test if levels of resource
dependence are the key predictor of inter-community knowledge differences worldwide. It also
looks to see if factors causing intra-community differences shift with level of economic
development.
198
Chapter 6. Country Comparisons
6.1. Introduction
For several decades ecological knowledge, its decline and management solutions, have been
receiving global attention. However, despite its widespread importance, studies of ecological
knowledge on the whole have been small-scale, culturally-secluded and geographically-isolated
(Tsuji, 1996; Furman, 1998; Mandany & Bugahoos, 1998; Turner et al., 2000; Byg & Balslev, 2001;
Olsson & Folke, 2001; Crate, 2002; Kristensen & Balslev, 2003; LaRochelle & Berkes, 2003; Negi,
2003; Nyhus et al., 2003; Shrestha & Dhillion, 2003; Hunter & Rinner, 2004; Ladio & Lozada,
2004; Talay et al., 2004). Many statements derived from these isolated studies form the basis of our
understanding of ecoliteracy today (Brodt, 2002). However, it is dangerous to over-generalise from
the results attained from such small-scale research since there is no evidence that the conclusions
drawn reflect the situation in other communities close by, let alone nationally or even globally
(Teune, 1990). Therefore, geographically-isolated studies have the potential to be isolated in their
findings also.
To date, only a few ecoliteracy studies have explored the knowledge of different regions and dared
to make comparisons (Gadgil et al., 2000; Ruddle, 2000; Eisler et al., 2003; Vandebroek et al.,
2004). The researchers of these studies realised the place-based nature of ecological knowledge and,
based on this, made comparisons between patterns of knowledge distribution in different physical
and cultural environments. Outcomes revealed a number of factors that have the capacity to greatly
affect knowledge levels, even between neighbouring communities, including access to modern
medicine and market imports (Nolan & Robbins, 1999; Gadgil et al., 2000; Byg & Balslev, 2001).
Such research has taught us of the importance of placing local knowledge systems in their
geographic, cultural, political and socio-economic contexts.
Cross-cultural studies to date have focused upon the differences in knowledge levels between sites
and, consequently, have taught us a great deal about the heterogeneity of ecological knowledge
(Garro, 1986; Sillitoe, 1998; Gadgil et al., 2000). However, they have failed to look at the
similarities in knowledge that exist across geographical and cultural boundaries (Sillitoe, 1998;
199
Ghimire et al., 2005; Fazey et al., 2006). Detecting similarities in ecological knowledge could be
key to combating global patterns of loss. For instance, few studies have addressed the process of
knowledge acquisition and the success of different modes of transfer (Alexiades, 1999). By
describing the most effective and ineffective modes of knowledge transmission across cultures, the
potential exists to not only describe widespread patterns of knowledge loss but to explain them. This
may in turn inform policy decisions with regards to the protection of local knowledge systems into
the future.
According to Davis & Wagner (2001), ecological knowledge research should represent the full
breadth, depth and comparability of knowledge systems whilst ensuring the outcomes are
sufficiently justified to stand up to rigorous public inspection, however only the latter has been
focused upon in most previous studies. Even cross-cultural studies, to date, have not addressed the
full extent of differentiation between knowledge systems, since industrialised and developing
knowledge bases have yet to be compared. This is important to understand global patterns of
knowledge distribution in the light of the continual spread of economic development into developing
regions.
Earlier work has dealt with the two regions independently, focusing on either industrialised (Nabhan
& Trimble, 1994; Pyle, 2001) or developing locations (Gadgil et al., 2000; Vandebroek et al., 2004),
avoiding comparing the two contrasting regions despite their co-existence. Adeola (1998), however,
did make cross-national comparisons between 24 developed and less-developed countries, but
avoided the issue of knowledge differentiation focusing on concern for environmental issues,
awareness of environmental problems and pro-environmental activities. This type of extensive study
with a focus on ecological knowledge would teach us a great deal about the effect of development
upon knowledge systems.
Some may say that comparing the knowledge systems of resource-dependent societies and
industrialised, disconnected regions is futile, with the latter knowing far less than their dependent
counterparts. However, comparing patterns of community distribution and effective sources of
transfer may reveal global patterns in knowledge acquisition that transcend economic development
and geographical and social boundaries. For instance, similar knowledge transfer processes may
200
exist in the face of different political, economic and cultural situations resulting from similar
histories of resource dependence, which may be key to maintaining or boosting community LEK
levels in the future. Comparing distribution patterns may also give an indication as to the future of
traditional knowledge systems in the light of impending development.
In general, indigenous knowledge has been the focus of previous work (Gadgil et al., 1993; Mauro
& Hardison, 2000; Pfeiffer, 2002). This is a consequence of extensive writing on the value of
indigenous knowledge and practices, particularly to self-management schemes (Feit, 1988; Ghimire
& Pimbert, 1997; Rudd et al., 2003; Cinner et al., 2005). Knowledge in industrialised regions,
however, is also crucial to local management practices, not only to provide information on local
ecosystem dynamics, but also to ensure public support of co- and state-management schemes into
the future (Lundquist & Granek, 2005).
By making cross-comparisons between developing and industrialised regions, this study aims to lay
to rest the assumption that all ecological knowledge research is small-scale and culturally- and
geographically-isolated (Sillitoe, 1998). All knowledge systems will be put in the context of
geographic location, culture, economic status and current worldviews and, taking precautions not to
over-generalise, will be compared as the entirely different systems that they are (Teune, 1990).
Cross-cultural comparisons will be made to determine whether the progressive decline of ecological
knowledge is a global phenomenon or isolated to industrialised areas and if different factors
controlling knowledge levels exist where levels of resource dependence differ.
By comparing three sites with completely different economic, cultural and political histories, this
chapter aims not only to compare knowledge levels but also to compare patterns of community
distribution. For instance, is it possible that elder women are wild flower experts globally regardless
of geographic location or culture? Also comparing such different sites enables the detection of any
key factors that are responsible for the loss or transfer of ecological knowledge globally. Therefore
by comparing the knowledge of respondents from the UK, India and Indonesia, this study aims to
Determine if any patterns in inter-community knowledge levels exist that relate to levels of
economic development.
201
Determine if any similar patterns in community knowledge distribution exist, between different
genders, age or wealth groups for example.
Determine if any similarities exist in the transfer of ecological knowledge or if different sources
are more effective in different regions of the world.
Understand any changes in knowledge distribution patterns that occur with industrialisation, not
only in terms of content but also transfer.
Determine if any key factors exist that, through policy-promotion globally, have the potential to
cease or reverse the process of ecological knowledge decline.
6.2. Methodology
The methods used for data collection at all sites have already been described in previous chapters,
therefore this section just focuses on the rationale behind country comparisons made and the data
analysis carried out. As a result of haphazard sampling methods, semi-structured interviews with
species flashcards and % identification ability being used in the assessment of ecological knowledge
at every site, data from all sites could be used for comparative purposes. Due to only plant
ecoliteracy being measured at the India study site, and plant identification being strongly correlated
with overall ecoliteracy at the UK and Indonesia study sites (UK; Rs=0.762, p<0.001, n=711,
Indonesia; Rs=0.579, p<0.001, n=192), plant identification scores were the focus of comparisons
made in this chapter, taken to be reliable indicators of overall ecoliteracy levels.
SPSS 12.0 statistical package was used for database construction and manipulation, recoding of data
and to carry out statistical tests (Kinnear & Gray, 1999). Non-parametric tests were used due to
working with non-normally distributed data. As well as comparing overall knowledge levels, the
results of this chapter aim to compare the differences and similarities in patterns of community
distribution and acquisition between the UK, India and Indonesia.
Differences in patterns of knowledge distribution with age between the three locations were tested
for. Firstly, the data for each site were split into adults and young people. Over 20 years of age was
taken to be indicative of adults with livelihoods and sometimes families. The age group of 20 years
and under was taken to be indicative of young people that have gained their knowledge to date
202
primarily from family and schooling. The mean difference in knowledge level between adults and
young people within each community was then calculated (11 UK communities, 6 Indonesian
communities and 5 Indian communities). The knowledge differences within each community from
India, Indonesia and the UK were all then compared and Kruskall-Wallis statistical test was used to
see if they were significantly different.
Knowledge acquisition curves were constructed for each country by calculating the mean
knowledge level of 20-29 year old respondents, 30-39 year olds, 40-49 year olds and 50-70 year old
respondents. This gave an indication of the mean knowledge levels of the different age groups from
all three sites at the time of study. The construction of knowledge curves also revealed any plateaus
in knowledge gain. Where a plateau was found, the knowledge of the age group in question was said
to be saturated since it had reached the knowledge level of community elders.
Different patterns of gender knowledge distribution between locations were tested for firstly by
calculating the mean difference in men and womens knowledge levels within the UK, India and
Indonesia communities studied. Then significant differences in gender knowledge distribution
between the different communities were tested for using Kruskall-Wallis. This tested to see if
economic development had a differential effect on the knowledge levels of men and women.
The effect of development upon ecological knowledge was examined by testing for an association
between mean ecological knowledge and the GDP and HDI of the sites. GDP figures and HDI
scores were attained from the UNDP Human Development Reports (2003). However, due to vast
regional differences in development levels, the state HDI for Tamil Nadu was used. Also, the HDI
and GDP figures for Indonesia were amended to account for the high poverty of the region, the high
proportion of indigenous Bajo resident at the study site and the poor education facilities. By
adjusting the life expectancy to account for mean Bajo life expectancy, reducing the adult literacy
figure and the GDP (to account for the illiterate, low earning Bajo population), the HDI for the
region was recalculated to be 0.57 instead of the national figure of 0.70 or regional figure of 0.64
(see Appendix VIII for details of recalculation). Spearmans Rank was used to test for an association
between GDP/HDI and ecological knowledge level.
203
Community knowledge variance was compared between the three sites. By calculating the
coefficient of variance for the knowledge of respondents from each community, a value for the %
variance between the most and least knowledgeable individuals from each village/town/ward was
attained. Mean levels of variance for the UK, India and Indonesia were then calculated, enabling
mean knowledge levels and community variance to be tested for an association using Spearmans
Rank. Data and observations were also taken from the results of previous chapters to make
quantitative and qualitative comparisons of impacting factors and key sources of transfer.
6.3. Results
The results of this section were aimed at comparing mean LEK levels and different knowledge
distribution patterns observed at the study sites of the three countries. The results from each country
were collated and compared to explore if any patterns in knowledge distribution exist that relate to
economic development levels and if any are sustained throughout the three sites despite
development differences. Knowledge of local plant species (% plants identified) was used as an
indicator of overall ecoliteracy.
Initially, age and gender differences at each site were compared to see if any patterns were sustained
throughout all sites or whether age and gender-differentiated knowledge patterns altered with levels
of development. Knowledge acquisition with age was also compared between the three countries in
terms of rate of acquisition and average age of saturation.
The effects of GDP, as an indicator of wealth, and HDI (a product of adult literacy rate, gross
enrolment of schools, GDP and life expectancy) on ecoliteracy and knowledge variance were tested.
Finally, observations made at the various sites in terms of the most effective sources of transfer and
the importance of experience and word-of-mouth to ecoliteracy levels were examined, as were the
negative effects of television and modern education systems.

The effect of age on ecological knowledge was examined firstly by calculating and comparing the
mean knowledge levels of respondents under the age of 20 with those aged 20 and over at all three
204
sites. The differences recorded between the two age groups within the different countries were then
compared. The differences in knowledge between the two age groups significantly differed between
the three sites (H=14.478, df=2, p<0.01, n=1095). The only country in which knowledge level
significantly differed between the two age groups was the UK (U=16920.500, p<0.001, n=711),
with over 20 year olds knowing on average 13% more species than under 20 year olds. Within the
India and Indonesia study sites, differences in knowledge levels between the two age groups were
non-significant with elder people knowing only 2% more plant species at the Indian study sites and
young people knowing only 5% more species, on average, at the Indonesian study site (Fig 6.3.1).
In the UK, therefore, elder people were able to recognise more species than younger. This meant
that knowledge level was found not to peak until the age of 70 on average. In India, however,
knowledge levels are higher on average and acquisition more rapid, with knowledge saturation
occurring by the age of 50. Finally, Indonesia revealed the highest levels of ecological knowledge
and the fastest rate of knowledge acquisition, with knowledge levels on average being saturated by
the age of 30 (Fig 6.3.2).

The effect of development of gender-differentiation of ecological knowledge was tested by
comparing the mean differences in knowledge between men and women in the UK, India and
Indonesia. Patterns in knowledge differentiation between men and women were found not to
significantly differ between the three sites. At all three sites, womens plant ecoliteracy slightly
exceeded that of mens. Despite this gender difference in knowledge only being significant at the
UK study site (U=54888.000, p<0.01, n=711) and not at the other sites, level of economic
development was found to have no significant effect on this pattern

To examine the effect of economic and human development on ecoliteracy, the mean knowledge
levels of respondents from the three sites were tested to see if there was an association with the GDP
and HDI figures. The association between level of development (economic and human) and
ecological knowledge was found to be significant and negative (Rs=-0.471, p<0.001, n=1095). (The
statistical result attained was the same when testing for an association between both GDP and
Difference in plant ecoliteracy between the old

and young (difference in % plants identified)
205
20
10
-10
UK
India
Indonesia
Country
Figure 6.3.1. Comparing age differentiation in plant

knowledge between the UK, India and Indonesia (median
+ interquartile range, n=1095). Negative differences were
yielded where younger generations knowledge exceeded
that of elder generations
100
80
C
60
Indonesia
India
B
A
40
A
UK
A
20
0
20
30
40
50
60
70
Age
Figure 6.3.2. Comparing the mean changes in plant

ecoliteracy with age between the UK, India and Indonesia
(mean + 2SE, n=1095). (*) Indicates age of knowledge
saturation at the India and Indonesia study sites
206
knowledge and HDI and knowledge since HDI is a product of GDP.) Therefore, this study found
that as the economic status of a community increases, and consequently human development level
increases, community knowledge of the local environment decreases (Fig 6.3.3).
Knowledge variance and ecoliteracy level were also tested for an association and were found to be
negatively associated (Rs=-0.471, p<0.001, n=1095) (Fig 6.3.4). Therefore, as mean community
knowledge level increases, the difference in knowledge between the most and least knowledgeable
community members was found to significantly decrease. This in turn correlates with development.
As the economic development of a community increases, mean ecological knowledge level
decreases and the variation between the most and least knowledgeable becomes greater.
Within countries, differences in local wealth were only found to have a significant effect upon
ecoliteracy in India and Indonesia (India r=-0.830, p<0.001, n=163; Indonesia Rs=-0.157, p<0.05,
n=192), but not in the UK. In both India and Indonesia wealth rank was found to be indirectly
related to ecoliteracy. Therefore, it appears that the wealthier and less resource dependent a region
becomes, the lower ecological knowledge levels fall and the less effect household income has on
local knowledge levels (as in the UK). However, in less wealthy regions where resource dependence
and local knowledge is still high, small differences in the wealth of households can notably affect
knowledge levels.

In both India and Indonesia the importance of societal roles in resource collection and use were
responsible for community knowledge differentiation. Roles that involve more frequent interaction
with nature, dependence upon local species and coercing with others that hold similar knowledge,
for instance for motherhood or farming, enhance a persons LEK levels. Therefore, in some cases,
societal roles make the overall ecosystem knowledge of certain groups in society higher, for instance
where men are the primary hunter-gatherers and encounter wildlife daily. In other cases, however,
individuals may simply develop an area of expertise, like elder women in the medicinal uses of
plants. Either way what appears to dictate a persons knowledge level and content is their experience
of nature.
Mean plant ecoliteracy (% plants identified)
207
70
Indonesia GDP $2521
60
50
]
India GDP $2892
40
30
UK GDP $27,147
20
0.60
0.70
0.80
0.90
Human Development Index
Figure 6.3.3. Mean plant ecoliteracy in the UK, India and

Indonesia and relationship with human development index
and per capita GDP (+ 2SE, n=1095). GDP US$ estimates
Community knowledge variance (coefficient

of variance %)
derived from purchasing power parity (p pp)
UK
50
40
30
W
India
20
30
40
50
Indonesia
60
70
Figure 6.3.4. Comparing mean levels of plant ecoliteracy

with community knowledge variance in the UK, India and
Indonesia (n=1095)
208
In the UK, experience was also found to be a crucial predictor of ecological knowledge, however it
was gained not through societal roles, but through lifestyle decisions. Experience was higher in
those people that have lived in rural areas for a long period and make frequent visits to the
countryside for reasons than involve being out of doors. Ecological knowledge was found to be
highest in these groups of people that maximise their quality and frequency of environmental
experience. Therefore, key at all three sites was the overwhelming importance of experience to LEK
levels.
Also evident at all three sites was the importance of oral communication to ecoliteracy transfer. The
continuation of societal roles, and the skills needed to fulfill them, at both the India and Indonesia
study sites indicates that at least a portion of ecological knowledge is transferred orally between
people of shared roles. For instance, Indonesian women knew that collecting coral for use in house
foundations was believed to make a house stronger. Without oral transfer, the women may have
learned how to mine the coral for foundations, but not the reasons for doing it. In the UK, the most
successful sources of knowledge acquisition included friends/locals, parents/relatives, hobbies and
occupation (the latter two most likely involve books as well as conversing with others that have
similar interests). This again emphasises the importance of word-of-mouth knowledge transfer to
sustaining high levels of ecological knowledge. However, what was also highlighted at all sites was
the failure of television and modern education systems in the transfer of ecological knowledge.
In the UK, the damaging effect of television and schooling upon ecological knowledge transfer was
revealed by the negative association detected between % contributions from these sources and
knowledge level (television Rs=-0.218, p<0.001, n=711; education Rs=-0.147, p<0.001, n=711).
Therefore, the more an individual relied upon television or formal education to acquire their
ecological knowledge, the lower their mean knowledge levels. This pattern was also indirectly
observed at the Indonesia study site by the negative relationship revealed between primary
household income and knowledge level. In communities more economically developed, televisions
were more common and children were more likely to attend school. Therefore, these two factors are
likely to contribute to the lower ecoliteracy levels found in wealthier households of this region.
209
6.4. Discussion

Indicative of progressive knowledge decline in industrialised regions, UK communities
demonstrated significantly higher levels of age differentiation in ecological knowledge than the
Indian and Indonesian communities studied. Therefore, UK elders were able to identify significantly
more local plant species than younger generations unlike Indian and Indonesian elders. This could
be indicative of one of two scenarios; a pattern of progressive loss could be occurring (Douglas,
2002; Birchenough, 2005) or in the UK ecological knowledge gain could continue throughout life,
until becoming saturated (to the levels of their ancestors) later in life. As already discussed in
Chapter 4, the former scenario is the most likely due to a shift in knowledge sources away from
experience and oral transfer and towards television and formal education (Nabhan & Trimble, 1994;
Kellert, 2002; Pyle, 2003). Therefore, it seems that the UK is undergoing progressive loss of
ecological knowledge, unlike resource dependent communities of India and Indonesia that are
retaining knowledge levels between generations, indicative of the importance of local knowledge to
resource dependence lifestyles.
If progressive loss is occurring in industrialised regions as thought, it seems fair to assume that,
although the knowledge of UK respondents was found to peak at the age of 70, this is not in fact
indicative of a saturation point and a plateau (matching ancestral knowledge levels) is not reached.
Instead, it simply indicates that LEK levels are higher in elder generations as a consequence of more
years to experience and learn about nature. Thus knowledge saturation (whereby a generation has
acquired all of the knowledge held by the previous generation) is unlikely ever to occur in regions
like the UK where LEK is no longer successfully transferred between parents/grandparents and
children and is consequently being lost with each passing generation.
In India and Indonesia however, knowledge saturation was found to occur by the age of 50 and 30
respectively, demonstrating the importance of acquiring knowledge at a younger age in resource
dependent communities. Therefore, where local plant species are important to daily activities such
as construction, eating and medicinal treatments, ecological knowledge is passed on to younger
generations at an early age (Adi & Sudarman, 1994). The pattern of Indonesian knowledge levels
210
and acquisition rate exceeding that of Indian knowledge levels and rate is most likely a result of
increased resource dependence at the Indonesian study site. Therefore in reference to the knowledge
acquisition curves hypothesised in Figure 1.5.2 (Chapter 1), it appears that Indonesia reflects
hypothetical acquisition curve A (highly resource dependent society; rapid acquisition and
saturation), India represents curve B (less resource dependent society; less rapid acquisition and
saturation), and the UK curve C (industrialised society; slow acquisition and no saturation). Thus the
findings from these three sites provide evidence in support of the knowledge acquisition curves
hypothesised in Chapter 1.

On the whole, gender differentiation in knowledge did not differ significantly between the three
sites. Therefore, economic development does not cause ecological knowledge gain by men and
women to significantly differ. Although the only significant difference in men and womens plant
knowledge was detected at the UK study site (men were able to identify 32% of local plant species
compared to women who could identify 35% on average), more in-depth analysis revealed that
slight gender differences did exist at the other two sites (Indian men identified 45% and women
47%, Indonesian men identified 68% and women 69%). Therefore, it appears that women know
slightly more about their local plants than men at all three sites, however the social roles filled by
men at the resource dependent sites minimise this differentiation.
This is evidence in support of previous studies claiming that the women of any community know
more about their local plant species than the men (Plumwood, 1993; Mellor, 1997; Shiva, 1997;
Warren, 1997). However, both Plumwood (1993) and Mellor (1997) claim that women have a
heightened connection with local plants that is indicative of their heightened knowledge throughout
history resulting from their traditional roles in society. Comparison between sites shows that, in fact,
knowledge levels are less differentiated where resource dependence is still high and traditional
societal roles still prevalent. Therefore, societal roles do not appear to be the cause of gender
differentiation giving strength to Shivas (1997) claims that women are innately more deeply
connected with their environment than are men.
211

A clear pattern of reduced knowledge with increased levels of development was demonstrated by
this study. As GDP, and therefore human development, of a region increased, ecological knowledge
was found to decrease. This indicates that as communities become wealthier and economies become
less self-sustaining and dependent on local resources, so knowledge of those resources is no longer
an essential survival tool (Byg & Balslev, 2001). As a consequence, ecological knowledge
transmission is most likely reduced as knowledge of local markets and global economies becomes
more important and talked about (Gadgil et al., 2000; Stocks et al., 2006). Therefore, a downward
spiral is entered into whereby resource dependence is removed, knowledge transmission is reduced
and substitution for other forms of knowledge inevitable, resulting in a cycle of progressive loss
with each generation. Therefore, doubts must be realised as to the future of ecological knowledge
survival in the face of modern patterns of economic development.
The result that community knowledge variance is inversely related to knowledge level and,
therefore, directly related to development is indicative of ecological knowledge independence with
loss of resource dependence. As communities become wealthier and less resource dependent, they
also become less dependent on the knowledge that facilitates local resource use (Gadgil et al., 2000).
Therefore, individuals with very little knowledge of their local surroundings can survive just as well
as experts in industrialised societies, creating high variation in knowledge levels. This is contrary to
resource dependent regions where ecological knowledge is not an asset but a survival tool (Pretty,
2002). In these communities, holders of the most knowledge have increased survival chances.
Therefore, variance is low since all community members need a minimum LEK level for daily
survival.
At the India and Indonesia study sites, where resource dependence is still high, community
knowledge levels were found to decline with increased community wealth. No such pattern was
found at the UK study sites since communities are no longer dependent on natural resources in this
part of the world and, therefore, knowledge of them is not a product of survival but one of personal
interest unrelated to income. However, where resource dependence is still high and a predictor of
daily activities, income has a significant impact upon knowledge levels. Wealthier households with
enhanced purchasing power can trade in local markets and are, therefore, less dependent on the local
212
environment to meet all of their daily needs. However poorer households, that have little or no
purchasing power, are forced to collect any resources they require for daily life from their
surroundings. Therefore their increased reliance due to reduced wealth leads to an increased
knowledge of local resources.

At all sites, be it through societal roles and daily tasks or personal lifestyle choices including area of
residence and countryside visits, what appears key to ecological knowledge is level of
environmental experience. This is evidence in support of Pyles (2002, 2003) extinction of
experience which hypothesises that the less a person spends time in contact with nature, the less they
know and recognise. Consequently, the less they notice environmental changes and the less they act
to protect the unfamiliar. As the environment becomes less diverse with increased species
extinctions and people become more disconnected with nature, the less their desire to experience the
outdoors leading to a hopeless downward spiral of disconnection and degradation.
The results of this study support this hypothesis and emphasise the importance of direct engagement
with nature, whatever the purpose for interaction, collecting food or walking the dog for instance
(Karjalainen & Habeck, 2004). Therefore, communities can afford to develop whilst retaining their
ecological knowledge providing they sustain high levels of interaction and, ideally, as many
traditional uses for local species as possible. In addition to the importance of direct experience, the
importance of oral transfer in the transmission of ecological knowledge was also revealed at all
sites, as found in a number of previous studies (Alexiades, 1999; Turner et al., 2000; Singh & Singh,
2005). Therefore, regardless of development level, traditional sources of ecological knowledge
transfer that have been relied upon throughout history, oral transfer and direct experience, appear to
be the most effective.
This conclusion is further supported by the failure of television and formal education in the transfer
of ecological knowledge revealed by this study. Despite textbooks and virtual images (over the
internet) being the key tools of education today, they fail to be as effective in ecological knowledge
transfer as traditional informal methods (Peat, 1996; Kellert, 2002; Pyle, 2002). Therefore,
213
pixellated screens and pictures in textbooks are unable to replace the effectiveness of butterfly nets
and botany walks in teaching children about their local environment (Pyle, 2001).
Economic development is responsible for heightened dependence on television screens and
textbooks for education, making traditional teachings of parents and grandparents that sustained
human populations for generations, devoid (Nabhan & Trimble, 1994; Peat, 1996). Modern formal
education systems have devalued informal teachings and experience with emphasis on exams and
curricula. In addition, the introduction of the virtual (computer games) is luring children away from
experiencing the actual (playing out of doors) to such an extent that childrens time spent out of
doors is reported to have reduced from 86 minutes per day in 1981 to 42 minutes in 1997 (Orr,
2002).
In conclusion, economic development is altering modes of environmental education whilst drawing
children in from the outdoors. These factors combine to create rapid extinction of experience and
knowledge that may be irreversible in the future as we create generation after generation of educated
individuals containing fewer naturalists than ever before. Therefore, curricula-reform and policies
encouraging maximum environmental experience appear to be key to sustaining ecological
knowledge into the future (Orr, 1996; Johnson, 2000; Matsumoto, 2003; Haigh, 2005). However,
reform needs to be rapid if there is any hope of creating knowledgeable individuals taught by
todays declining pool of naturalists.
214

Prior to this study there was little understanding of how ecoliterate people acquired their knowledge
and the rate at which community knowledge declined as wealth increased. However, despite being
small-scale, geographically localised and culturally-isolated, earlier research did reveal the
heterogeneity of ecological knowledge within and between communities (Garro, 1986; Sillitoe,
1998; Gadgil et al., 2000; Ghimire et al., 2005; Fazey et al., 2006), which has led to the current
understanding that ecoliteracy is both site-specific and heavily dependent upon local societal roles.
Hence conclusions drawn to date have mostly been restricted to the sites of study. Therefore, the
main aim of this thesis was to broaden the current understanding of ecological literacy by
developing a standardised protocol for assessing LEK, both primary (naming) and secondary (uses).
This protocol was intended for use cross-culturally in order to fully understand the effect of
development on ecological knowledge levels and transfer. In doing so, various other questions were
answered about the patterns of knowledge distribution and the success of different transfer methods.
These will be discussed in turn.
7.1. Developing a standardised protocol for ecological knowledge assessment within and
between communities across different cultures
Previously no standardised methodology existed for the cross-cultural assessment of ecological
knowledge and past studies have differed in their approach making the results incomparable and
even unreliable (Tsuji, 1996; Davis & Wagner, 2003; Vandebroek et al., 2004). This study
developed a novel standardised protocol for assessing ecological knowledge for comparative
purposes. The aim was not to develop a methodology that assessed total community knowledge, but
instead to assess differences in mean knowledge levels and content both within and between
communities. The protocol identified the most and least knowledgeable groups, their characteristics
and primary knowledge sources. It was developed to be culturally-generic and employable at
different sites across the world.
Combining structured interviews with photographic flashcards was an effective method for
assessing ecological knowledge at entirely different locations (Balmford et al., 2002; Nyhus et al.,
215
2003). By assessing both primary (species naming ability) and secondary (species uses) knowledge
levels of non-resource dependent communities could be compared with resource dependent ones.
Conclusions and hypotheses drawn from previous small-scale studies were tested for their
widespread applicability and a new understanding of ecological knowledge patterns and transfer was
developed based on analysis of the results.
Perhaps most importantly this methodology generates quantitative data about the distribution of
ecological knowledge within communities for cross cultural-comparison. Most previous studies,
with a few geographically-isolated exceptions (Gadgil et al., 2000), have been primarily qualitative
and focused on documenting existing knowledge bases and uses for a selection of local species (Byg
& Balslev, 2001). Although potentially valuable in the future to identify opportunistic uses, such
studies fail to understand widespread patterns of knowledge loss or causes through changes in
transfer methods, required to combat future ecoliteracy loss. Both of these were successfully
identified in this study. Therefore, in addition to recording specialised species uses by local
communities, this methodology successfully identified differences in knowledge between the most
and least knowledgeable groups, key characteristics of these groups, what factors threaten to reduce
ecoliteracy and what can raise community knowledge levels. This provides future researchers with
an understanding of modes of knowledge transfer, their success rates, and ways to reverse future
decline.
This methodology also challenged the suggestion made by Davis and Wagner (2003) that ecological
knowledge research should focus upon the identification of experts. The methodology used here
clearly reveals high levels of knowledge variation within communities and differences in content
between experts and laypersons. Therefore, experts may be able to list a great number of uses for
local species, but this does not necessarily reflect the true knowledge level and content of the
community as a whole.
Hence experts should not be the sole focus of future LEK studies, as has been previously suggested,
for two reasons. Firstly, knowledge variation increases as economic development increases,
widening the gap between expert and layperson knowledge. Secondly, minority groups often
ignored in research, such as women, often have knowledge levels equal to that of experts, although
216
the content differs. For instance, in the case of women this may result from their primary role as
child carers curing common childhood ailments in comparison with healers who focus on the
treatment of more serious afflictions (Garro, 1986; Alexiades, 1999; Turner et al., 2000; Pfeiffer,
2002).
Although assessing local knowledge using just a selection of the local species prevents an
assessment of total ecological knowledge, it yielded results that allowed inter- and intra-community
comparisons and identified groups that were particularly deprived of knowledge. Restricting
interview length in this way permitted large sample sizes and cross-cultural comparisons over a
wide geographic spread.
One limitation of this study is that it assessed progressive knowledge loss using groups of
respondents of different ages. To ensure that progressive loss was definitely taking place and that
knowledge was not simply acquired throughout life, the ideal would be to measure the knowledge of
a single cohort at different ages (every 5-10 years) throughout their lifetimes. However, there was
not sufficient time available to this study to support this in-depth sampling and the method used still
yielded interesting results about generational differences in knowledge and transfer methods. In
conclusion, combining structured interviews with photographic flashcards is an effective method of
assessing distribution patterns of ecological knowledge both within and between communities
(including communities from different cultures and geographic locations) and examining the
attributes that make certain groups more knowledgeable than others.
7.2. Understanding the effect of economic development on knowledge distribution both within
and between communities
To date, a great deal of speculation has centred on the effect of development on ecological
knowledge. However, few studies have investigated and quantified the direct relationship between
community wealth/income and knowledge level and what factors, driven by increased wealth, are
responsible for ecoliteracy demise. This research is the first to have considered this relationship
cross-culturally in regions with differing levels of economic development.
217
This study clearly revealed a strong inverse relationship between ecoliteracy and wealth, not only
within and between localised communities, but also between communities internationally. This
confirms the research hypothesis made in Chapter 1 that economic development leading to
modernisation of lifestyles acts to damage local knowledge bases. It also provides quantitative
evidence for earlier theories describing the negative effects of industrialisation upon knowledge
levels (Johnson, 2000; Turner et al., 2000; Kaihura, 2004) and the direct relationship that exists
between resource dependence and environmental knowledge both locally and globally (Pretty,
2002). Hence communities worldwide face a period of ecoliteracy extinction in the light of growing
resource independence and market globalisation. This was observed even at the local level as
remote, resource dependent communities, with recently increased income (from trading local goods
with outsiders and increased market access), have begun to lose their secondary knowledge (local
uses).
Therefore the aim of this thesis to achieve an understanding of the effect that development has upon
ecological knowledge has been met on two levels, both locally and globally. Unlike previous
studies, these conclusions are supported by quantitative data (Ghimire et al., 2005). Recorded trends
reveal a distinct pattern of decline, from an initial loss in secondary knowledge of local species uses,
to a subsequent loss in knowledge of local species names (primary knowledge) over time with
continued economic development and detachment. This thesis also found a direct relationship
between community wealth and variation in knowledge levels revealing a larger gap in knowledge
between experts and laypersons in industrialised communities than resource dependent ones. This
emphasises the importance of looking at laypersons knowledge, in addition to experts, in future
LEK studies. This is the first time that knowledge variance has ever been quantified and a reason
given for its differing levels.
With modern patterns of development, these results have implications for biodiversity conservation
and resource management schemes in communities worldwide. For example, loss of ecoliteracy
from the general public reduces their stake in the environment and, subsequently, threatens public
support for local conservation efforts in industrialised regions (Pyle, 2001, 2002, 2003; Lundquist &
Granek, 2005). This in turn permits species extinctions to continue unnoticed and unabated by local
residents. However, this investigation detected a positive relationship between self-efficacy and pro-
218
environmental behaviour with knowledge level. This implies that increased LEK levels increase an
individuals self-belief in the capacity of their own environmental actions. A valuable extension to
this study would be to determine the strength of this relationship and to assess if a similar
relationship exists between level of knowledge and willingness to participate in conservation
schemes.
In developing regions, a decline in ecoliteracy also threatens biodiversity but via a more direct route.
That is as local knowledge declines so local management practices, including collection methods
that have sustained the diversity of a locality for generations, are lost (Feit, 1988; IIED, 1994;
Ruddle, 2000; Cinner et al., 2005). As these traditional management techniques are abandoned and
eventually forgotten by younger generations, local diversity is likely to decline (studies have shown
that traditional management practices usually act to sustain higher levels of biodiversity by
maintaining habitats in a mosaic of different successional stages (Gadgil et al, 1993)) and species
composition alter.
Ecoliteracy decline also undermines a communitys capacity for self-management (Phillips et al.,
1994). For instance, as the in-depth knowledge base that has been used to guide local management
decisions for generations is lost, a shift occurs from reliance upon actual local observations to
reliance upon state theory and predictive modeling. Such schemes as locally-managed Lake Racken
in Sweden (Olsson & Folke, 2001) and the waters around Ahus Island, Papua New Guinea (Cinner
et al., 2005) are less likely to occur as local knowledge and connectedness decline. Even comanagement schemes could be threatened by a future decline in ecoliteracy as local support derives
from a local awareness of the environment, How can we care about that which we do not know?
(Pyle, 2001).
Therefore, by maintaining and/or maximising the ecological knowledge base of communities, using
methods and targeting groups identified by this study, two potential outcomes exist; (i) local support
for and participation in conservation activities will most likely be heightened in industrialised
communities and (ii) traditional management practices, and the diversity they create, will be
sustained in developing regions. However, ecoliteracy decline must be addressed soon if the future
of resource management is to be anything other than state-controlled.
219
7.3. Discovering the key factors that lead to high and low levels of ecological knowledge based
on quantitative methods and if these differ between localities
Many studies have described some level of ecoliteracy decline and have attributed it to recent
introductions, for instance of a modern health clinic or imported market goods (Barrett, 1995;
Alexiades, 1999; Brodt, 2002; Kaihura, 2004; Stocks et al., 2006). However, although these may
explain the indirect factors driving knowledge loss, they do not explain the intermediate factors
directly causing dissemination failure. For example, introducing a modern healthcare clinic does not
automatically cause people to forget their knowledge of local plants and animals, so what changes
take place in ecological knowledge transfer as a result of this introduction that result in such
decline?
This study looked directly at the changes in transmission methods caused by development that lead
to knowledge loss. Since modern patterns of economic development are unlikely to be reversed, it is
crucial to identify successful dissemination methods and if these differ between sites. Therefore, this
thesis is unique in that it identified the sources from which ecological knowledge is transferred, how
key sources alter with development and which are the most and least successful, in the hope of
offering solutions to combat decline in the face of widespread industrialisation in the future.
The results revealed that traditional methods of ecoliteracy transmission, word-of-mouth transfer
and direct experience, remain the most successful at all study sites across the world regardless of
resource dependence and development level. As has been widely discussed in theory but never
previously quantified, the failure of modern education in local knowledge dissemination has been
shown in this study to be a major cause of knowledge decline at all study sites (Nabhan & Trimble,
1994; Kothari et al., 1998). This is due to a shift in knowledge source reliance from daily experience
with elder relatives to the learning tools of formal education (text books and the internet) and
supports the hypothesis made in Chapter 1 (that local knowledge becomes threatened where modern
education systems exist). This also supports the writings of Nabhan and Trimble (1994), Kellert
(2002) and Pyle (2002). The failure of television in transferring ecological knowledge (again an
issue that has been widely discussed in the literature but never quantitatively proven (Kellert, 2002))
220
was also identified in this study. This again illustrates the invaluable role of direct experience and
oral transfer in creating LEK experts.
This suggests that in the light of development, curricula need rapid re-evaluation and revision to
encompass more field visits so that children can learn about the outdoors through direct experience
and through interactions with friends and family members rather than just via textbooks and
television (Nabhan & Trimble, 1994; Pyle, 2001; Orr, 2002). Policies are needed that encourage
weekly visits to the countryside in industrialised regions and help maintain traditional management
practices and resource use in developing regions, if we hope to produce ecoliterate generations in
the future.
Raising community knowledge levels through the promotion of traditional methods of transfer
offers conservation benefits by producing naturalists amongst younger generations and human
health benefits associated with reconnecting people with their environments. Studies have shown
such benefits to be both physical and mental (Pretty et al., 2005). This is particularly important
considering that almost half of the worlds population live in fast-paced urban environments today
(Sustainable Development International, 2004). Therefore, the results of this study show that
experience and oral transfer are key to sustaining high ecoliteracy levels into the future, and school
curricula and television sets need to be re-evaluated for their place in environmental education.
Returning to traditional methods of transfer could potentially offer some solutions to local resource
management, biodiversity conservation and human health challenges.
7.4. Determining whether progressive knowledge loss is occurring within the chosen
communities and the reasons for its existence or absence
For any hope of ecoliteracy preservation and environmental reconnection in the future, and policies
in support of these actions, it is crucial to understand where loss is occurring from and which groups
are most affected, if we are to plug the holes from which knowledge is leaking. Key questions
include: Is progressive loss occurring everywhere or is it isolated within certain communities? Is it
evenly distributed within communities or is loss greater in some groups than others? Where is the
progressive loss of ecological knowledge most rapid?
221
Finding the answers to these questions is essential if we are to identify target groups for education
and reconnection programmes, particularly if we hope to produce future generations of naturalists
(Pyle, 2002). As a consequence of the tendency of past ecological knowledge studies to be
geographically-localised and culturally-isolated (Sillitoe, 1998; Fazey et al., 2006), no studies to
date have attempted to give an overview of global patterns of knowledge decline. Some previous
studies, however, have been successful in identifying pockets within communities where loss is
occurring (Gadgil et al., 2000; Birchenough, 2005), but what they have failed to define is the criteria
that define these groups and whether these criteria are site-specific or applicable to all communities
worldwide.
Since a worldwide reversal of knowledge loss is inconceivable, identifying key groups to direct
efforts at raises the chances of success. This entails identifying the regions where it is occurring
most rapidly and whether certain groups within communities are more susceptible to becoming
disconnected and ecologically illiterate than others. Being able to target efforts is a criterion
essential to many public policies worldwide. Therefore, being able to identify target communities,
and then automatically knowing within those communities which group(s) is most likely to be
lacking in knowledge is key to policy success. Thus target groups must be identified on two levels.
Firstly, communities either entering or already in a period of knowledge loss must be identified, and
then the population groups within them that are likely to be the least informed due to poor or nonexistent transfer methods must be isolated.
This thesis successfully met the aims described above, illustrating that progressive knowledge loss is
a symptom of economic development and departure from resource dependence as hypothesised in
Chapter 1 (knowledge levels of elder generations will exceed that of younger generations where
local resource dependence no longer exists) (Douglas, 2002; Birchenough, 2005). Demonstrating
this, progressive loss in younger generations was found to be most marked in UK non-resource
dependent communities. Here, knowledge loss is rapid and both primary and secondary forms of
ecological knowledge are in decline. Progressive loss of secondary knowledge is occurring but at
reduced rates at the India study site that has undergone a degree of economic development in recent
years.
222
The only site that did not reveal any progressive knowledge loss with age was Indonesia where
young people were still meeting the knowledge levels of their predecessors and a high level of
resource dependence continues. However, with the introduction of schooling and the expansion of
local markets, this may not be the case for much longer. Therefore an inter-community scale of loss
exists, from no loss in resource dependent societies to the most rapid loss in industrialised regions,
such that all communities with limited or no resource dependence are likely to have entered a period
of progressive LEK decline.
In addition to addressing the period of loss currently faced in economically-developed or developing
communities, local governments and policy-makers should also consider the repercussions of future
development on the local knowledge of traditional societies before injecting a sum of money or
increasing transport and market connections to a previously isolated site (Nolan & Robbins, 1999).
That is not to say that development should be restricted, but that precautions should be taken to
promote the continuation of traditional management practices and resource uses in order to preserve
them locally, and the knowledge that facilitates them, in the face of impending development.
Intra-community patterns of knowledge loss show that young people that live in urban areas, that
come from wealthier families (in developing regions), that have undergone formal schooling and
that experience the outdoors infrequently, generally, have a paucity in their ecological knowledge
and are unlikely to ever reach the knowledge levels of their parents and grandparents. Therefore,
when entering a community undergoing progressive knowledge loss, education programme efforts
should focus upon these groups of people.
However, an alternative to the introduction of educational programmes would be the revision of
school curricula as already discussed (Orr, 1996; Johnson, 2000; Matsumoto, 2003; Haigh, 2005),
especially since the target groups are young urban people in formal education. Therefore, increasing
field visits and addressing methods of ecological teaching, in urban schools in particular, may
contribute to combating further decline and even reverse the trend.
223
Such actions are essential if we hope to produce future generations of naturalists with a view
towards continuing these teachings in the future (Orr, 1996; Pyle, 2002). If current rates of loss are
permitted to continue unabated, then one day in the future we may face the situation where we are
not in a position to save ecological knowledge. Instead, we may have to recreate the knowledge
bases that have sustained us for generations and taken hundreds of years to gather.
This thesis succeeded in identifying the communities undergoing the most rapid rates of progressive
knowledge loss and identified the target groups on whom knowledge preservation efforts should be
focused. One factor that may encourage policy-makers to invest in ecoliteracy preservation is if
human health can be shown to be suffering as a consequence of progressive knowledge decline and
its coinciding environmental disconnection (Nabhan & Trimble, 1994; Pyle, 2003). This study
revealed that the least knowledgeable groups are less likely to visit the countryside and experience
nature. The benefits of being out of doors and exercising in the environment have recently been
shown in a study by Pretty et al. (2005). However, if in addition to this, future research could
demonstrate reduced health in these groups as a direct result of environmental disconnection, this
may provide a key tool in persuading policy-makers of the importance of acting to protect
ecoliteracy now rather than later.
7.5. To understand patterns of knowledge acquisition with age at all sites and to test the
hypothetical acquisition curves shown in Fig 1.5.2
The hypothesised knowledge acquisition curves made in this study, based on what is already known
about the relationship between resource dependence and LEK (Pretty, 2002), make it unique when
compared with previous studies. Looking at knowledge level differences with age in the different
communities enabled the process of acquisition in the different environments to be examined. This
thesis provided evidence in support of the hypothesised acquisition curves (Fig 1.5.2.),
demonstrating the relationship between economic development and resource dependence with
knowledge acquisition, level and saturation rate.
224
A
Knowledge level
B
C
Age
Figure 1.5.2. Theoretical acquisition curves. Curve A represents a highly resource dependent society
with rapid acquisition and saturation. Curve B represents a less resource dependent society with less
rapid acquisition. Here knowledge saturation occurs later in life and at a lower level than community
A. Both curve A and curve B show a decline in knowledge after the plateau, indicative of memory
loss with age. Curve C represents an industrialised community with no resource dependence where
knowledge level never reaches that of community elders and knowledge levels are on the whole
low.
Therefore, the findings of this study revealed that economic development greatly affects community
knowledge acquisition. As predicted, acquisition is slower and more gradual in industrialised
societies (such as the UK) where resources are no longer essential to survival. However, in
developing communities (such as India and Indonesia) where resource collection and utilisation is
still a part of daily life, knowledge acquisition is far more rapid and knowledge levels far higher. In
these communities, knowledge levels plateau off when the knowledge of younger generations
reaches that of their parents and grandparents, and a slight drop in knowledge occurs with old-age.
Thus it appears that a scale of community knowledge acquisition exists, gradually becoming less
rapid as economic development of a community increases.
225
Hence knowledge conservation measures are required in developing regions to promote traditional
practices and resource uses in order to sustain current knowledge levels and acquisition rates.
Industrialised regions, on the other hand, need to be targeted by policies attempting to reintroduce
LEK acquisition from a young age, either through the introduction of education programmes,
increased countryside visits or through altered curricula (Orr, 1996; Johnson, 2000; Matsumoto,
2003; Haigh, 2005). Such policies are essential for developing communities to be able to selfmanage their own resources in years to come and to ensure public support of future co-management
schemes in industrialised regions (Lundquist & Granek, 2005).
7.6. Detecting differences in familiarity with different species groupings and reasons for such
differences
It is important to understand whether populations know more about some species groupings than
others in order to understand whether previous studies using just one species group (usually plants)
to test the overall knowledge levels of a community are reliable. It would also be useful to know if
certain species groupings are more well-known than others then why they are. For instance, if some
species are more well-known due to their usefulness, characteristic appearance or abundance, then
can these attributes be used to promote knowledge levels of other species (threatened locally for
example) raising them to an equal level of familiarity?
No research in the past has purposively used a variety of different species groups to assess local
knowledge levels and quantitatively compared public knowledge levels of the different groups
(Gadgil et al., 2000; Alexiades, 1999; Byg & Balslev, 2001; Nyhus et al., 2003). In general, past
studies have focused on assessing local knowledge of just one species grouping, in order to make a
judgment of overall community knowledge. This study is unique in its quantitative assessment of the
knowledge held on several species groupings and how accurate plant identification is as an indicator
of overall LEK levels. By looking at the species groups that different populations know the most
about and attempting to explain these differences, this study takes an original in-depth look at
ecological knowledge content.
226
The results of this study found that knowledge content of communities and factors affecting content
varied according to level of resource dependence. The most resource dependent communities
studied, in remote Indonesia, showed a high level of knowledge of all species within the ecosystems
they most relied upon. For instance, the Bajo held an expanse of knowledge about all marine species
in their local waters. In India, however, reduced resource dependence meant that not all species
within locally-exploited ecosystems were well-known and there was a clear relationship between
knowledge of local plants and their usefulness to the daily lives of locals. Therefore, the usefulness
of a species or ecosystem is clearly the biggest predictor of knowledge content in developing
communities, since this decides the environments and species that locals most frequently interact
with.
One way to maintain this bond between traditional communities and their local ecosystems is to
promote the traditional use of local resources, and therefore regular ecosystem interaction, as a
matter of cultural pride even in the face of economic development. However, with modern patterns
of development encroaching, causing a decline in natural resource dependence and a shift towards
formal systems of education, this pattern of species knowledge based on dependence and daily
ecosystem interactions may be lost, as has occurred in UK communities.
Where resource dependence no longer exists and daily requirements are met through transported
goods and imports such as in the UK, it is the iconic status of species that is the key predictor of
public awareness of them (Pretty, 2006). For instance, in the UK, the publics knowledge of
mammals far exceeds that of other species groups despite being the least common, most infrequently
sighted and of very little use in our daily lives (even historically when compared to wild flowers
(Mabey, 1996)). This information may be highly valuable in efforts to enhance local knowledge
levels of other species groupings.
If childrens books and characters could focus on other species in addition to mammals, such as
trees, flowers or birds, or perhaps a combination, then children may develop an affinity with and
basic knowledge of a wider selection of species from a young age. Therefore, by enhancing the
iconic status of other species groups, childrens books and characters may be used as a tool in
ecoliteracy enhancement in future generations. Piloting this idea, evaluating its worth and ensuring
227
that enhanced knowledge of mammals cannot be attributed elsewhere may be a valuable extension
to this study.
If promoting the iconic status of any species from a young age is an effective way of raising public
awareness of that species in younger generations, then conservationists may be able to use this idea
to raise awareness of plants and animals threatened with extinction that are little known about.
Raising public awareness through increasing the iconic status of a species may act to heighten local
support and conservation efforts to save the species in question, illustrated in the World Wide Fund
for Natures Save the Panda campaign (WWF, 2006).
In addition to the potential benefits offered in the field of biodiversity conservation, these results
offer valuable insight to future researchers in the field of LEK. For instance, this study found that
assessing the plant knowledge of an individual gives a good indication of their overall LEK level
(with 76% accuracy in the UK). However, using mammal knowledge as an indicator of total LEK
level is less reliable (only 66% accurate in the UK). The iconic status of mammals was the cause of
this bias, as already discussed, with even the least knowledgeable groups able to identify the
majority of mammals in industrialised regions. These findings suggest that previous studies focusing
on mammals, in regions where media sources such as television and books are available, lack a
degree of accuracy and should not be used in future comparative work. However, using wild flower
knowledge as an indicator of community LEK level appears to be an effective tool, yielding good
results with minimum effort, although for true intra-community comparisons to be made, knowledge
of a variety of local species groups should be assessed.
7.7. General conclusions
7.7.1. Generic conclusions

1. The quantitative knowledge assessment procedures used during this investigation are
reproducible across different cultures and geographic locations.
2. The highest community knowledge levels are attained when economic development is at its
lowest and resource dependence at its highest.
228
3. The effects of community wealth vary between locations having no effect upon local
knowledge levels in industrialised communities, but showing an indirect relationship with
knowledge in resource dependent communities.
4. Variance in community knowledge between the most and least knowledgeable increases as
communities become more developed.
5. Knowledge acquisition rate increases and age of saturation decreases with increased level of
resource dependence.
6. A scale of progressive loss is occurring. The most rapid loss (primary and secondary
knowledge) is occurring in industrialised communities, a slower rate of loss (secondary
knowledge) is occurring at developing sites with low levels of economic development, and
no progressive loss is occurring at the most remote resource dependent sites.
7. Individuals from developing communities know the most about species and ecosystems that
are of the most use to them in their daily lives, whereas individuals from industrialised
communities know the most about mammals due to their iconic status.
8. The current structure and curricula of formal education are having a negative effect on young
peoples ecological knowledge levels and their view towards the importance of ecoliteracy,
as is television.
9. Traditional sources of knowledge transfer, including word-of-mouth and direct experience,
are the most effective methods of ecoliteracy transfer across all regions.
7.7.2. Developing region conclusions

1. Gender-differentiated knowledge in resource dependent communities is a product of social
roles and daily livelihood activities.
2. Laypersons often hold the same amount of knowledge as experts (healers), but content may
differ.
7.7.3. Industrialised region conclusions

1. Womens heightened wild flower knowledge is unlikely to be a product of historical societal
roles as previously hypothesised.
2. People who live in rural areas are more knowledgeable about their local environments.
229
3. Occupations have a significant effect upon knowledge levels, with teachers and people
working in the environmental sector knowing the most.
4. Childhoods spent in rural locations or travelling between different areas result in more
frequent visits to the countryside and higher ecoliteracy levels in adulthood.
5. Increased length of residence in an area increases an individuals local knowledge.
6. More frequent visits to the countryside lead to increased knowledge levels and the view that
ecoliteracy is of significant importance to society.
7. Where communities are no longer resource dependent, the most frequent reasons for them to
make countryside visits is for the purpose of walking or some other form of exercise.
8. Knowledge levels are higher in individuals that visit the countryside frequently for purposes
including their occupation or outdoor hobbies. However, individuals that visit the
countryside only to pass through it on the way to somewhere else, alongside non-visitors, are
the least knowledgeable.
9. Knowledge levels were found to be positively correlated with attitude towards the
importance of ecological knowledge.
The progressive loss of ecological knowledge we are now facing is the result of a shift in knowledge
transfer methods with economic development, from word-of-mouth and direct experience to reliance
upon the media and formal education (Nabhan & Trimble, 1994; Peat, 1996; Kellert, 2002; Orr,
2002) (Fig 7.6.1). Therefore, Pyles (2002, 2003) extinction of experience is most likely a key factor
contributing to knowledge loss, mutually-reinforced by a reduction in oral transfer from friends and
family (Peat, 1996).
230
High resource
dependence
Low economic
development
Intermediate resource
dependence
Intermediate economic
development
No resource
dependence
High economic
development
Indonesia
India
UK
High LEK
Low variance
No progressive loss
Intermediate LEK
Intermediate variance
Slow progressive loss
of secondary
knowledge
Low LEK
High variance
Rapid progressive
loss of primary and
secondary knowledge
Figure 7.6.1. Ecological knowledge distribution patterns at the 3 sites based on economic
development levels
Essentially, two levels of knowledge loss have been detected within this study; (i) progressive loss
with age within communities and (ii) regional loss (across many communities) with economic
encroachment. The former is most severe within industrialised communities, whereas the latter is of
current concern in developing regions starting to undergo economic development. Thus the
implications of both levels on industrialised and developing regions are different but equally
important.
In industrialised regions, further knowledge loss promises to create an even greater disconnection
between humans and nature in the future. This is likely to further reduce the amount of time future
generations spend out of doors, as we have recently seen in children (Orr, 2002). Less time out of
doors and more time spent immobile in front of the television or PC is likely to create a vast array of
231
health problems for industrialised countries to cope with in the future. These include physical
problems such as obesity and obesity-related diseases, including coronary heart disease,
hypertension and Type II diabetes (Pretty et al., 2005). In addition, are the mental problems that
further disconnection from the environment and reduced time out of doors can cause, including
depression and anxiety. With urban development and fast-paced city lifestyles attracting more and
more people, reconnecting people with nature and the outdoors may be key to reducing stress levels
and related disease, such as hypertension, in urban populations of the future (Pretty et al., 2005).
Therefore, introducing children to their local environment from a young age, and teaching them
about the species within it, can create adult generations with an affinity for nature that make more
outdoor visits throughout life, as shown in this study. This can have individual health benefits as
well as national benefits, reducing NHS expenditure for example, if pursued large-scale. In addition,
teaching people about their local species is the most likely means of making people care for them
(Pyle, 2001). Thus by educating children about their local plants and animals, future generations will
hopefully start to care about the species on their doorstep and notice when those species that they
care about decline.
The other strand of loss is across regions, in sites where economic development and marketisation of
natural resources is gradually encroaching. Here, it is the in-depth knowledge of traditional
communities that is most at threat from altered resource collection incentives, a shift towards
modern knowledge of economy and industry and a loss of traditional lifestyles. Losing knowledge
of local species uses, such as consumptive and pharmaceutical, has implications on a number of
scales. Firstly to local communities, who lose the ability to exploit local species for foods and
medicines in times of harsh environmental or strained economic conditions, as well as losing the
local income streams these species can provide. Secondly, nationally, the country loses a potential
source of food security and, again, an income source. Finally, internationally, human populations in
the future will have lost the knowledge of which wild species potentially offer food security to areas
affected by poverty, in addition to losing future opportunistic uses that have yet to be discovered, in
the cure of some major diseases for example.
232
Local knowledge loss in developing regions also threatens traditional systems of management that
have maintained ecosystems and resource bases throughout history. Traditional systems of
management, and the knowledge that they are based on, have anticipated environmental changes,
observed phases shifts and sustained human populations for generations. In addition to threatening
local, often biodiverse, ecosystems, loss of traditional knowledge and practices is also likely to
threaten the future of similar ecosystems worldwide. For example, this may happen as knowledge of
sustainable management practices is lost that could have been transferred to similar ecosystems
under threat in other parts of the world. As well as the practical losses caused by the decline of
ecological knowledge in developing regions, cultural losses are also likely. For instance, departure
from traditional resource management systems and local species uses threatens a break in cultural
continuity for many indigenous or tribal groups which, subsequently, reduces global cultural
diversity (King, 2003).
Prior to this investigation, there was a lack of information on the widespread distribution of
ecological knowledge and patterns of decline. Results obtained during this thesis have contributed to
our understanding of knowledge distribution patterns globally and have also added information to
the field on where LEK loss is occurring, the causes of loss and techniques that may be used to
combat further decline.
233
References
AA Route Finder (2005) Route Finder. Available from AA website http://www.theaa.com/
travelwatch/planner_main.jsp.
Adeola, F.O. (1998) Cross-national environmentalism differentials: Empirical evidence from core
and noncore nations. Society and Natural Resources, 11 (4), pp.339-364.
Adi, C.M. & Sudarman, P. eds. (1994) Bajau, 2nd edition. Jakarta, Yayasan Sejati.
Agrawal, A. & Gibson, C.C. (1999) Enchantment and disenchantment: The role of community in
natural resource conservation. World Development, 27, pp.629-649.
Agrawal, A. & Yadama, G.N. (1997) How do local institutions mediate market and population
pressures on resources? Forest panchayats in Kumaon, India. Development and Change, 28,
pp.435-465.
Ahmed, F.U. (2004) The protection of traditional knowledge in Bangledesh. In: Twarog, S. & P.
Kapoor eds. Protecting and promoting traditional knowledge: Systems, national experiences
and international dimensions. Geneva, United Nations, pp.185-192.
Akimichi, T. & Supriadi, D.A. (1996) Marine resource use in the Bajo of North Sulawesi and
Maluku, Indonesia. Senri Ethnological Studies, 42, pp.105-119.
Alcorn, J.B. (1993) Indigenous peoples and conservation. Conservation Biology, 7, pp.424-426.
Alexiades, M.N. (1999) Ethnobotany of the Ese Eja: Plants, health and change in an Amazonian
society. Ph.D. thesis, City University of New York.
Anishetty, M. (2004) Conservation and utilization of plant genetic resources for food and
agriculture: Strengthening local capacity for food security. In: Twarog, S. & Kapoor, P. eds.
Protecting and promoting traditional knowledge: Systems, national experiences and
international dimensions. Geneva, United Nations, pp.33-40.
Answers Corporation (2005) Essex County. Available from Answers.com website
http://www.answers.com/topic/essex-2.
ASHE (2005) Annual Survey of Hours and Earnings. Available from ONS (Office of National
Statistics) website http://www.statistics.gov.uk/downloads/theme
_labour/ASHE_2005/Table14_7a.xls.
Asian Development Bank (2001) Agricultural biotechnology, poverty reduction and food security. A
working paper. Available from http://www.adb.org/Cocuments/Books/Agri_Biotech/default.
asp.
AtUK (2005) Lincolnshire. Available from AtUK website http://www.atuk.co.uk/england/
lincolnshire.htm.
Balmford, A., Clegg, L., Coulson, T. & Taylor, J. (2002) Why conservationists should heed
Pokmon. Science, 295, pp.2367.
Bangalore CES (2003) Geography and major biomes. Available from Bangalore CES (Centre for
Ecological Sciences) website http://ces.iisc.ernet.in/hpg/cesmg/indiabio.html.
Barrett, B. (1995). Herbal knowledge on Nicaraguas Atlantic coast: Consensus within diversity.
Journal of Community Health, 20, pp.403-421.
Beastall, T.W. (1978) The agricultural revolution in Lincolnshire. In: Barley, M. ed. History of
Lincolnshire. Lincoln, The History of Lincolnshire Committee.
Berkes, F. (1999) Sacred Ecology. Philadelphia, Taylor and Francis.
Berkes, F., Colding, J. & Folke, C. (2000) Rediscovery of traditional ecological knowledge as
adaptive management. Ecological Applications, 10, pp.1251-1262.
234
Biggs, S. & Smith, G. (1998) Beyond methodologies: Coalition-building for participatory

technology development. World Development, 26, pp.239-248.
Biological Diversity Bill (2000) India biological diversity bill. Available from NIC (National
Informatics Centre) website http://envfor.nic.in/legis/others/biobill.html.
Birchenough, A.C. (2005) The use of local knowledge to map the distribution of endangered birds
on Pacific Islands. In: Evans, S.M. & Foster-Smith, J. eds. Proceedings of the 3rd
International ENSUS Conference: Promoting Knowledge and Understanding of the
Environment, 13 April, 2005, University of Newcastle upon Tyne, UK. University of
Newcastle upon Tyne, School of Marine Science and Technology, pp. 37-45.
Brodt, S. (2002) Learning about tree management in rural central India: A local-global continuum.
Human Organisation, 61, pp.58-67.
Browner, C.H. (1991) Gender politics in the distribution of therapeutic herbal knowledge. Medical
Anthropology Quarterly, 5, pp.99-132.
Bryson, L. (1996) Revaluing the household economy. Womens Studies International Forum, 19,
pp.207-219.
Burns, T.G. (2002) The driving forces and environmental effects of a shift from reef to pelagic
fishing practices in Sampela, southeast Sulawesi. Available from Operation Wallacea
website http://www.opwall.com/Indonesia%20marine%20reports/2002%20
research%20reports%20-%20Sampela%20fisheries.htm.
Byg, A. & Balslev, H. (2001) Diversity and use of palms in Zahamena, eastern Madagascar.
Biodiversity and Conservation, 10, pp.951-970.
Capra, F. (2000) Ecoliteracy: A systems approach to education. In: Crabtree, M. Ecoliteracy:
Mapping the terrain. California, Berkeley, Learning in the Real World, pp.27-35.
CCSR (2004) Population of Bromley. Available from CCSR (Cathie Marsh Centre for Census and
Survey Research) website http://www.ccsr.ac.uk/research/migseg/ bromley.doc.
Cesar H.S.J., Burke, L. & Pet-Soede, L. (2003) The economics of worldwide coral reef degradation.
Cesar Environmental Economics Consulting (CEEC), World Wildlife Fund/ICRAN,
Netherlands, 24pp.
Chamat Kaledupa (2005) Pemerintah Kabupaten Wakatobi: Lapchan Leadaan Penduduk Kecamatan
Kaledupa. Unpublished.
Cinner, J.E., Marnane, M. J. &. McClanahan, T.R. (2005) Conservation and community benefits
from traditional coral reef management at Ahus Island, Papau New Guinea. Conservation
Biology, 19, pp.1714-1723.
CIRAN-MOST (Centre for International Research and Advisory Networks and Management of
Social Transformations Programme) (2003) Register of best practices on indigenous
knowledge. Available from UNESCO website http://www.unesco. org/most/bpikreg.htm.
COREMAP (2002) Management objectives and performance criteria. Unpublished.
Crate, S.A. (2002) Viliui Sakha oral history: The key to contemporary household survival. Arctic
Anthropology, 39 (1-2), pp.134-154.
Cultural Survival (2006) Promoting the rights, voices and visions of indigenous. Available from
Cultural Survival website http://209.200.101.189/home.cfm.
Davis, A. (2003) Thoughts on approaches to designing and conducting ecological knowledge social
research. Human Ecology, 31 (3), pp.463-489.
Davis, A. & Wagner, J.R. (2003). Who knows? On the importance of identifying experts when
researching local ecological knowledge. Human Ecology, 31, pp.463-489.
235
DEFRA (2004) Topography, geology and main features. Available from DEFRA (Department for
Environment, Food and Rural Affairs) website http://www.defra.gov.uk/erdp/docs
/sechapter/section11/topography.htm.
DEFRA (2005) Environmental statistics. Available from DEFRA(Department for Environment,
Food and Rural Affairs) website http://www.defra.gov.U.K./environment/statistics
/index.htm.
DEFRAs National Crop Listing (2001) Plant varieties and seeds. Available from DEFRA
(Department for Environment, Food and Rural Affairs) website http://www.defra.gov.uk
planth/pvs/nlguide.htm.
Devavaram, J., Arunothayam, E., Prasad, R., Pretty, J. (1999) Watershed and community
development in Tamil Nadu, India. In: Hinchcliffe, F., Thompson, J., Pretty, J., Guijit, I. &
Shah, P. eds. Fertile ground: The impacts of participatory watershed development. London,
IT Publications.
Dixon, S.P., Birchenough, A.C., Evans, S.M. & Quigley, M.P. (2005) Childrens knowledge of
birds: How can it be improved and can it be used to conserve wildlife? Transactions of the
Natural History Society, Northumbria, 64, pp.121-134.
Douglas, G. (2002) Environmental knowledge in the people of Pacific Islands. In: Proceedings of
the 2nd International ENSUS Conference: Engaging the public in science, 16-18 December,
2002, University of Newcastle upon Tyne, UK. University of Newcastle upon Tyne, School
of Marine Science and Technology.
Dove, M.R. (1988) The real and imagined role of culture in development: Case studies from
Indonesia. Honolulu, University of Hawaii Press.
Drew, J.A. (2005) Use of traditional ecological knowledge in marine conservation. Conservation
Biology, 19, pp.1286-1293.
EIA (2001) India: Environmental Issues. Available from EIA (Energy Information Administration)
website http://www.eia.doe.gov/emeu/cabs/indiaenv.html.
Eisler, A.D., Eisler, H. & Yoshida, M. (2003) Perception of human ecology: Cross-cultural and
gender comparisons. Journal of Environmental Psychology, 23 (1), pp.89-101.
Ekpere, J.A. (2004) Sui generis legislation and protection of community rights in Africa. In:
Twarog, S. & Kapoor, P. eds. Protecting and promoting traditional knowledge: Systems,
national experiences and international dimensions. Geneva, United Nations, pp.235-240.
Elliott, G., Wiltshire, B., Manan, A. & Wismer, S. (2001) Community participation in marine
protected area management: Wakatobi National Park, Sulawesi, Indonesia. Coastal
Management, 29 (4), pp.295-316.
EMDA (2004) Lincolnshire. Available from EMDA (East Midlands Development Agency) website
http://www.emda.org.uk/about/county.asp?countyV=lincoln.
Emerging Planet India (2003) Madurai general information. Available from Emerging Plant website
http://www.madurai.com/general.htm.
Essex County Council (2005) Essex binaries. Available from Essex County Council website
http://www.essexcc.gov.uk/vip8/ecc/ECCWebsite/content/binaries/documents/et02v5nt.pdf.
European Marine Sites (2005) Why has the Essex Estuaries area been designated a candidate
European marine site? Available from European Marine Sites website http://www.
essexestuaries.org.uk/european/estuaries/euro_marine_site/why_ems.htm.
Evans, S.M., Dixon, S. & Heslop, J. (2005) Environmental knowledge: How good is it and where
does it come from? In: Evans, S.M. & Foster-Smith, J. eds. Proceedings of the 3rd
International ENSUS Conference: Promoting Knowledge and Understanding of the
236
Environment, 13 April, 2005, University of Newcastle upon Tyne, UK. University of

Newcastle upon Tyne, School of Marine Science and Technology, pp.1-15.
Farrington, J., Saxena, N.C., Barton, T. & Nayak, R. (2003) Post offices, pensions and computers:
New opportunities for combining growth and social protection in weakly integrated rural
areas? London, The Overseas Development Institute.
Fazey, I., Proust, K., Nedwell, B., Johnson, B. & Fazey, J.A. (2006) Eliciting the implicit knowledge
and perceptions of on-ground conservation managers of the Macquarie Marshes. Ecology
and Society, 11 (1), Available from Ecology and Society website
http://www.ecologyandsociety.org/vol11/iss1/art25/.
Feit, H.A. (1988) Self-management and state-management: Forms of knowing and managing
northern wildlife. In: Freeman, M.M.R. & Carbyn, L.N. eds. Traditional knowledge and
renewable resource management in northern regions. Edmonton, Boreal Institute for
Northern Studies and International Union for the Conservation of Nature and Natural
Resources, pp.72-91.
Fenta, T. (2004) Using farmers traditional knowledge to conserve and protect biodiversity: The
Ethiopian experience. In: Twarog, S. & Kapoor, P. eds. Protecting and promoting traditional
knowledge: Systems, national experiences and international dimensions. Geneva, United
Nations, pp.25-28.
Fisher, F. (2005) Ecoliteracy and metaresponsibility: Steps to and ecology of mind. Systemic
Practice and Action Research, 18 (2), pp.133-149.
Folke, C. (2004) Traditional knowledge in social-ecological systems. Ecology and Society, 9 (7),
Available from Ecology and Society website http://www.ecologyandsociety
.org/vol9/iss3/art7/.
Folke, C., Hahn, T., Olsson, P. &. Norberg, J. (2005) Adaptive governance of social-ecological
systems. Annual Review of Environment and Resources, 30, pp.441-473.
Forestry Commission (2006) Forest school project could point way for evaluating forest schools in
Wales. Available from Forestry Commission website http://www.forestresearch.gov.uk/
newsreel.nsf/HomepageWebPressReleases/0F50DCEE6631469D80257107004D7826.
Frisk, J., Kaiser, F.G. & Wilson, M. (2004) Environmental knowledge and conservation behaviour:
exploring prevalence and structure in a representative sample. Personality and Individual
Differences, 37, pp.1597-1613.
Furman, A. (1998) A note on environmental concern in a developing country: Results from an
Istanbul survey. Environment and Behaviour, 30 (4), pp.520-534.
Gadgil, D.R. (1972) The industrial evolution of India in recent times: 1860-1939. 5th ed. London,
Oxford University Press.
Gadgil, M. (1993) Biodiversity and Indias degraded lands. Ambio, 22, pp.167-172.
Gadgil, M., Berkes, F. & Folke, C. (1993) Indigenous knowledge for biodiversity conservation.
Ambio, 22, pp.151-156.
Gadgil, M., Seshagiri Rao, P.R., Utkarsh, G., Pramod, P., Chhatre, A. & members of the Peoples
Biodiversity Initiative (2000) New meanings for old knowledge: The peoples biodiversity
registers program. Ecological Applications, 10, pp.1307-1317.
Garro, L.C. (1986) Intracultural variation in folk medical knowledge: A comparison between curers
and non-curers. American Anthropologist, 88, pp.351-370.
GenUKI Kent (2004) Kent. Available from GenUKI website http://users.ox.as.uk/~malcolm
/genuki/big/eng/KEN/.
GenUKI Lincolnshire (2004) Lincolnshire. Available from GenUKI website http://www. genuki.org
.uk/big/eng/LIN/.
237
Ghimire, K.B. & Pimbert, M.P. (1997) Social change and conservation: Environmental politics and
impacts of National Parks and protected areas. London, Earthscan.
Ghimire, S., McKey, D. & Aumeeruddy-Thomas, Y. (2005) Heterogeneity in ethnoecological
knowledge and management of medicinal plants in the Himalayas of Nepal: Implications for
conservation. Ecology and Society, 9 (6), Available from Ecology and Society website
http:www.ecologyandsociety.org/vol9/iss3/art6/.
Gilchrist, G., Mallory, M. & Merkel, F. (2005) Can local ecological knowledge contribute to
wildlife management? Case studies of migratory knowledge. Ecology and Society, 10 (20),
Available from Ecology and Society website http://www.ecologyandsociety
Glaeser, B. (1995) Housing, sustainable development and the rural poor: A study of Tamil Nadu.
London, Sage Publications.
Godoy, R., Reyes-Garcia, V., Byron, E., Leonard, W.R. & Vadez, V. (2005) The effects of market
economies on the well-being of indigenous peoples and on their use of renewable natural
resources. Annual Review of Anthropology, 34, pp.121-138.
Gorjestani, N. (2004) Indigenous knowledge for development: Opportunities and challenges. In:
Government of Tamil Nadu (2004) Population Dynamics. Available from Government of Tamil
Nadu website http://www.tn.gov.in/dear/ch17.pdf.
Government of Tamil Nadu Policy Notes (2005) Backward classes, most backward classes and
minorities welfare department: Policy note 2004-2005. Available from Government of Tamil
Nadu website http://www.tn.gov.in/policynotes/bcmbc2004-05-1.htm.
Haigh, M. (2005) Greening the university curriculum: Appraising an international movement.
Journal of Geography in Higher Education, 29, pp.31-48.
Hamwey, R. (2004) Traditional knowledge and the environment: Statement by the United Nations
Environment Programme. In: Twarog, S. & Kapoor, P. eds. Protecting and promoting
traditional knowledge: Systems, national experiences and international dimensions. Geneva,
United Nations, pp.345-346.
Hardin, G. (1968) The Tragedy of the Commons. Science, 162, pp.1243-1248.
Hassan, R., Scholes, R. & Ash, N. eds. (2005) Current state and trends, Vol. 1: Ecosystems and
well-being. Millennium Ecosystem Assessment. London, Island Press.
Hazell, P.B.R. (2003) Green Revolution: Curse of blessing? In: Mokyr, J. ed. The Oxford
encyclopedia of economic history. Oxford, Oxford University Press.
Hazell, P.B.R. & Ramasamy, C. (1991) The Green Revolution reconsidered: The impact of highyielding rice varieties in south India. Baltimore, The John Hopkins University Press.
Hill, R. (1998) What sample size is enough in internet survey research? Interpersonal Computing
and Technology, 6, Available from Interpersonal Computing and Technology Journal
website http://jan.ucc.nau.edu /~ipct-j/1998/n3-4/hill.html#martin.
Holstein, J.A. & Gubrium, J.F. (1995) Qualitative research methods: The active interview. London,
Sage Publications.
Hooper, D.U., Chapin, F.S., Ewel, J.J., Hector, A., Inchausti, P., Lavorel, S., Lawton, J.H., Lodge,
D.M., Loreau, M., Naeem, S., Schmid, B., Setl, H., Symstad, A.J., Vandermeer, J. &
Wardle, D.A. (2005) Effects of biodiversity on ecosystem functioning: A consensus of
current knowledge. Ecological monographs, 75, pp.3-35.
Houtkoop-Steenstra, H. (2000) Interaction and the standardised survey interview: The living
questionnaire. Cambridge, Cambridge University Press.
238
Howard, P. (2004) Gender bias in ethnobotany: Propositions and evidence of a distorted science and
promises of a brighter future. In: Proceedings of the International Society of Ethnobiology
9th International Congress and Society for Economic Botany 45th Annual Meeting: Panel 12
Gender issues in ethnobotanical and ethnobiological research and related development
practice, 13th-17th June, 2004, University of Kent, Canterbury. University of Kent,
Department of Anthropology.
Howard-Borjas, P.L. (2002) Gender relations in local plant genetic resource management and
conservation. In: Doella, H.W. & DaSilva, E. eds. Encyclopedia for Life Support Systems:
Biotechnology. Cambridge, EOLSS Publishers.
Hunter, L.M. & Rinner, L. (2004) The association between environmental perspective and
knowledge and concern with species diversity. Society and Natural Resources, 17, pp.517532.
Huntington, H.P. (2000) Using traditional ecological knowledge in science: Methods and
applications. Ecological Applications, 10, pp.1270-1274.
IIED (International Institute for Environment and Development) (1994) Whose Eden? An overview
of community approaches to wildlife management. London, IIED.
India EIA (2001) India: Environmental issues. Available from India EIA (Energy Information
Administration) website http://www.eia.doe.gov/emeu /cabs/indiaenv.html.
India State of Environment Report (2001) India State of Environment Report: Priority Issues.
Available from NIC (National Informatics Centre) website http://envfor.nic.in
/soer/2001/soer.html.
India Travelogue (2004) Tamil Nadu. Available from India Travelogue website
http://indiatravelogue.com/dest/tam/geog.html.
Indonesia Human Development Report (2001) Human development challenges in a democratic,
decentralised Indonesia. Available from UNDP website www.hdr.undp.org/reports/detail
_reports.cfm?view=144.
Jewitt, S. (2000) Unequal knowledges in Jharkland, India: De-romanticizing womens
agroecological expertise. Development and Change, 31, pp.961-985.
JNCC (2005) Essex estuaries. Available from JNCC (Joint Nature Conservation Committee)
website http://jncc.gov.uk/ProtectedSites/SACselection/sac.asp? EUCode=UK0013690.
Jodha, N.S. (1986) Common property resources and rural poor in dry regions of India. Economic
and Political Weekly, 11, pp.1169-1182.
Johnson, J.M. (2000) Design for learning: Values, qualities and processes of enriching school
landscapes. Washington, DC, American Society of Landscape Architects.
Kahn, P.H. (2002) Childrens affiliations with nature: Structure, development, and the problem of
environmental generational amnesia. In: Kahn, P.H. & Kellert, S.R. eds. Children and
nature: Psychological, sociocultural, and evolutionary investigations. Massachusetts, The
MIT Press, pp.93-116.
Kahn, P.H. & Kellert, S.R. eds. (2002) Children and nature: Psychological, sociocultural, and
evolutionary investigations. Massachusetts, The MIT Press.
Kaihura, F.B.S. (2004) Indigenous knowledge and rural livelihood improvement. PLEC News and
Views, New Series 4 March, pp.2-6.
Kaiser, F.G., Wlfing, S. & Fuhrer, U. (1999) Environmental attitude and ecological behaviour.
Journal of Environmental Psychology, 19, pp.1-19.
Kala, C.P. (2005) Indigenous uses, population density, and conservation of threatened medicinal
plants in protected areas of the Indian Himalayas. Conservation Biology, 19, pp.368-378.
239
Kamil, S. (2004) The protection of traditional knowledge in Indonesia. In: Twarog, S. & Kapoor, P.
eds. Protecting and promoting traditional knowledge: Systems, national experiences and
Karjalainen, T.P. & Habeck, J.O. (2004) When The Environment comes to visit: Local
environmental knowledge in the far north of Russia. Environmental Values, 13, pp.167-186.
Kaushik, A. (2004) Protecting traditional knowledge, innovations and practices: The Indian
experience. In: Twarog, S. & Kapoor, P. eds. Protecting and promoting traditional
knowledge: Systems, national experiences and international dimensions. Geneva, United
Nations, pp.85-90.
Kellert, S.R. (2002) Experiencing nature: Affective, cognitive, and evaluative development in
children. In: Kahn, P.H. & Kellert, S.R. eds. Children and nature: Psychological,
sociocultural, and evolutionary investigations. Massachusetts, The MIT Press, pp.117-152.
Kellert, S.R. & Wilson E.O. eds. (1993) The biophilia hypothesis. Washington, DC, Island Press.
Kent County Council (2003) Employment, occupations and workforce. Available from Kent County
Council Land Use and Transport Policy Unit website http://66.249.93.104/
search?q=cache:BcDKjqXxOX4J:www.kent.gov.uk/NR/rdonlyres/1440E578-0074-4E54AB41-C6B6BEC8C7B/1075/cen403employment.pdf+kent+county+council
+Employment,+occupations+and+workforce.&hl=en&gl=uk&ct=clnk&cd=1.
Kent County Council (2004) Local transport plan for Kent 2001/02 to 2005/06. Available from
KCC (Kent County Council) website http://www.kent.gov.uk/sp/planning/LTP/environment
&heritage.html.
Kent County Council Bulletins (2004a) GDP estimates. Available from KCC (Kent County
Council) website http://www.kent.gov.uk/sp/lutp/spig/Bulletins/Economy%20
and%20Labour%20Market/ECN302_GDP%20Estimates.pdf.
Kent County Council Bulletins (2004b) Earnings in Kent 2000. Available from KCC (Kent County
Council) website http://www.kent.gov.uk/sp/lutp/spig/Bulletins/Economy%20
and%20Labour%20Market/Ecn101.pdf.
KENTnet (2005) A brief history of Kent the Garden of England. Available from KENTnet
website http://www.kentnet.co.uk/kentindex/kenthist.htm.
Khare, A. (1998) Community-based conservation in India. In: Kothari, A., Pathak, N., Anuradha,
R.V. & Taneja, B. eds. Communities and conservation: Natural resource management in
south and central Asia. London, Sage Publications.
King, L. ed. (2003) Sharing a world of difference: The earths linguistic, cultural, and biological
diversity. Paris, UNESCO-Terralingua-World Wide Fund for Nature.
Kinnear, P.R. & Gray, C.D. (1999) SPSS for windows made simple. Sussex, Psychology Press.
Kothari, A., Pathak, N., Anuradha, R.V. & Taneja, B. eds. (1998) Communities and conservation:
Natural resource management in south and central Asia. London, Sage Publications.
Krech, S. (2005) Reflections on conservation, sustainability, and environmentalism in indigenous
north America. American Anthropologist, 107, pp.78-86.
Kristensen, M. & Balslev, H. (2003) Perceptions, use and availability of woody plants among the
Gourounsi in Burkino Faso. Biodiversity and Conservation, 12, pp.1715-1739.
Ladio, A.H. & Lozada, M. (2004) Patterns of use and knowledge of wild edible plants in distinct
ecological environments: A case study of a Mapuche community from northwestern
Patagonia. Biodiversity and Conservation, 13, pp.1153-1173.
Laine, A., Lavonen, J. & Meisalo, V. eds. (2004). Current research on mathematics and science
education. Research Report 253. Helsinki, University of Helsinki, Department of Applied
Sciences of Education.
240
LaRochelle, S. & Berkes, F. (2003) Traditional ecological knowledge and practice for edible wild
plants: Biodiversity use by the Raramuri in the Sierra Tarahumara, Mexico. International
Journal of Sustainable Development and World Ecology, 10, pp.361-375.
Latiff, A. & Zakri, A.H. (2004) Biodiversity and traditional knowledge: The Malaysian experience.
In: Twarog, S. & Kapoor, P. eds. Protecting and promoting traditional knowledge: Systems,
Le Quy, A. (2004) The use and commercialization of genetic resources and traditional knowledge in
Vietnam: The case of crop and medicinal plants. In: Twarog, S. & Kapoor, P. eds. Protecting
and promoting traditional knowledge: Systems, national experiences and international
dimensions. Geneva, United Nations, pp.7-14.
Lipton, M. (2004) Crop science, poverty and the family farm in a globalizing world. In: Proceedings
of the 4th International Crop Science Congress: New directions for a diverse planet, 26
September - 1 October, 2004, Brisbane, Australia.
Lipton, M. (2005) The family farm in a globalizing world: The role of crop science in alleviating
poverty. 2020 Discussion Paper 40. Washington DC, International Food Policy Research
Institute.
Loh, J. & Harmon, D. (2005) A global index of biocultural diversity. Ecological Indicators, 5 (3),
pp.231-241.
Long, J., Tecle, A. & Burnette, B. (2003) Cultural foundations for ecological restoration on the
White Mountain Apache reservation. Ecology and Society, 8 (1), Available from Ecology
and Society website http://www.ecologyandsociety.org/vol8/iss1/art4 /main.html.
Ludden, D. (1999) The new Cambridge history of India IV: An agrarian history of south Asia.
Cambridge, Cambridge University Press.
Lundquist, C.J. & Granek, E.F. (2005) Strategies for successful marine conservation: Integrating
socioeconomic, political, and scientific factors. Conservation Biology, 19, pp.1771-1778.
Mabey, R. (1996) Flora Britannica. London, Sinclair-Stevenson.
MacDonald, E. (2002) Assessing the human development status of three Kaledupa villages.
Available from Operation Wallacea website http://www.opwall.com/Library/
Indonesia/marine%20socioeconomic.shtml.
Maffi, L. (2002) Endangered languages, endangered knowledge. UNESCO. Oxford, Blackwell
Publishers.
Mandany, I.M. & Bugahoos, K.A. (1998) An assessment of 6th, 9th, and 12th grade students
environmental knowledge in Bahrain. Environment International, 24 (3), pp.325-330.
Martin, G.J. (2004) Ethnobotany: A methods manual. London, Earthscan Publications.
Matsumoto, K. (2003) The nature journal as a tool for learning. Available from New Horizons for
Learning website http://www.newhorizons.org/strategies/environmental/ matsumoto.htm.
Mauro, F. & Hardison, P.D. (2000) Traditional knowledge of indigenous and local communities:
International debate and policy initiatives. Ecological Applications, 10, pp.1263-1269.
May, D. (2004) Reef fishing activity in the Kaledupa stakeholder area. Available from Operation
Wallacea website http://www.opwall.com/2004%20Kaledupa%20reef%20fishing.htm.
May, D. (2005) Folk taxonomy of reef fish and the value of participatory monitoring in the
Wakatobi National Park, SE Sulawesi, Indonesia. Unpublished.
May, D. & Coles, T. (2004) Proposed strategy for the recovery of the Kaledupan coastal fisheries.
Available from Operation Wallacea website
http://www.opwall.com/Library/Indonesia/fisheries%20and%20farming.shtml
McNeely, J.A. & Scherr, S.J. (2001) Common ground, common future: How ecoagriculture can
help feed the world and save wild biodiversity. Washington, DC, IUCN and Future Harvest.
241
Meinhold, J.L. & Malkus, A.J. (2005) Adolescent environmental behaviours: Can knowledge,
attitudes, and self-efficacy make a difference? Environment and Behaviour, 37, pp.511-532.
Mellor, M. (1997) Feminism and Ecology. Cambridge, Polity Press.
Mhame, P.P. (2004) The role of traditional knowledge in the national economy: Traditional
medicine in Tanzania. In: Twarog, S. & Kapoor, P. eds. Protecting and promoting
Mller, M.J. (1982) Selected climatic data for a global set of standard stations for vegetation
science. London, Dr. W. Junk Publishers.
Multimap (2006) Online maps to everywhere. Available from Multimap website http://www.
multimap.com/.
Murdoch, J. & Clark, J. (1994) Sustainable knowledge. Geoforum, 25, pp.115-132.
Myers, N. (2002) A convincing call for conservation. Science, 295, pp.447-448.
Nabhan, G.P. (2002) Coming home to eat: The pleasures and politics of local foods. New York,
W.W. Norton and Company.
Nabhan, G.P. & St. Antoine, S. (1993) The loss of floral and faunal story: The extinction of
experience. In: Kellert, S.R. & Wilson, E.O. eds. The biophilia hypothesis. Washington, DC,
Island Press, pp.229-250.
Nabhan, G.P. & Trimble, S. (1994) The geography of childhood: Why children need wild places?
Boston, Beacon Press.
Nadeau R. L. (2003) The wealth of nature: how mainstream economics has failed the environment.
New York, Columbia University Press.
Nathan, K.K. (1995) Droughts in Tamil Nadu: A Qualitative and Quantitative Appraisal.
Available from Drought Network News website http://drought.unl.edu/pubs/dnn/
dnnarchive.htm.
Negi, C.S. (2003) Role of traditional knowledge and beliefs in conservation: Case studies from
Central Himalaya, India. Man in India, 83, pp.371-391.
Newspaper Society (2004) Frinton-On-Sea: Demographic profile of the area. Available from
Newspaper Society website http://www.nsdatabase.co.uk/locationdetail.cfm?
locationid=623.
NIC (2003) Virudhunagar district. Available from NIC (National Informatics Centre) website
http://www.virudhunagar.tn.nic.in/.
Nimmo, H.A. (1990) Religious rituals of the Tawi-Tawi Bajau. Philippine Studies, 38, pp.166-98.
Nolan, J.M. &. Robbins, M.C. (1999) Cultural conservation of medicinal plant use in the Ozarks.
Human Organisation, 58, pp.67-72.
Nyhus, P.J., Sumianto & Tilson, R. (2003) Wildlife knowledge among migrants in southern
Sumatra, Indonesia: Implications for conservation. Environmental Conservation, 30, pp.192199.
Nyrop, R.F. (1975) Area handbook for India, 3rd ed. Washington, DC, American university.
Olsson, P. & Folke, C. (2001) Local ecological knowledge and institutional dynamics for ecosystem
management: A study of Lake Racken watershed, Sweden. Ecosystems, 4, pp.85-104.
Olsson, P., Folke, C. & Hahn, T. (2004) Social-ecological transformation for ecosystem
management: The development of adaptive co-management of a wetland landscape in
southern Sweden. Ecology and Society, 9 (2), Available from http://www.ecologyandsociety
ONS (2001) Census 2001. Available from ONS (Office of National Statistics) website
http://www.statistics.gov.uk/census2001/access_results.asp.
242
ONS (2005a) Population: UK population grows to 59.6 million. Available from ONS (Office of
National Statistics) website http://www.statistics.gov.uk/cci/nugget.asp?id=760.
ONS (2005b) About National Statistics and ONS. Available from ONS (Office of National
Statistics) website http://www.statistics.gov.uk/geography/nrudp.asp#rural.
ONS Essex (2001) Census 2001: Essex County. Available from ONS (Office of National Statistics)
website http://www.statistics.gov.uk/census2001/pop2001/print_v /essex_print.asp.
ONS Kent (2001) Census 2001: Kent County. Available from ONS (Office of National Statistics)
website http://www.statistics.gov.uk/census2001/pyramids/printV/29.asp.
ONS Lincolnshire (2001) Census 2001: Lincolnshire. Available from ONS (Office of National
Statistics) website http://www.statistics.gov.uk/census2001/pyramids/printV /32.asp.
Operation Wallacea (2004) Facts about the Wakatobi. Available from Operation Wallacea website
http://www.opwall.com/Library/index.shtml.
Orientalia (2003) Enact law to protect traditional management. Available from Orienatlia Articles
website www.orientalia.org/article575.html.
Orr, D.W. (1996) Educating for the environment: higher education's challenge of the next century.
The Journal of Environmental Education, 27, pp.7-10.
Orr, D.W. (2002) Political economy and the ecology of childhood. In: Kahn, P.H. & Kellert, S.R.
eds. Children and nature: Psychological, sociocultural, and evolutionary investigations.
Massachusetts, The MIT Press, pp.279-304.
Oviedo, G., Gonzales, A. & Maffi, L. (2004) The importance of traditional ecological knowledge
and ways to protect it. In: Twarog, S. & Kapoor, P. eds. Protecting and promoting
Pacn, A.M. (2004) The Peruvian proposal for protecting traditional knowledge. In: Twarog, S. &
Kapoor, P. eds. Protecting and promoting traditional knowledge: Systems, national
experiences and international dimensions. Geneva, United Nations, pp.175-180.
Parlee, B., Manseau, M. & . K Dene First Nation (2005) Using traditional knowledge to adapt to
ecological change: Densoin monitoring of Caribou movements. Arctic, 58 (1), pp.26-37.
Peat, F.D. (1996) Blackfoot physics: A journey into the native American universe. London, Fourth
Estate.
Pedju, M., Santiaji, V., Hanan, S. & Fudge, J. (2004) Report on a baseline survey in Wakatobi
Marine National Park to assess resource status, use and perception March-October 2004.
SE Sulawesi, Joint Program Office. Available from Operation Wallacea website http://tncseacmpa.org/downloads/Wakatobi%20Baseline%20 Survey%20Report%20V1.pdf.
Pfeiffer, J. (2002) Gendered interpretations in eastern Indonesia: Ethnoecology in the transition
zone. In: Lansdowne, H., Neilson, W. & Dearden, P. eds. Communities in Southeast Asia:
Challenges and responses. Victoria, Centre for Asia Pacific Initiatives, pp.43-63.
Phillips, O.L., Gentry, A.H., Reynel, C., Wilkin, P. & Glvez-Durand B.C. (1994) Quantitative
ethnobotany and Amazonian conservation. Conservation Biology, 8 (1), pp.225-248.
Pierotti, R. & Wildcat, D. (2000) Traditional ecological knowledge: The third alternative
(commentary). Ecological Applications, 10, pp.1333-1340.
Pilgrim, S.E., Smith, D. & Pretty, J. (2006a) A cross-regional quantitative assessment of the factors
affecting ecoliteracy: Policy and practice implications. Ecological Applications, in review.
Pilgrim, S.E., Cullen, L., Smith, D.J. & Pretty, J. (2006b) Ecological Literacy is Lost in Wealthier
Communities and Countries. Conservation Biology, in review.
243
Pilgrim, S.E., Smith, D. & Pretty, J. (2006c) Hidden Harvest or hidden revenue? The effect of
economic development pressures on local resource use in a remote region of southeast
Sulawesi, Indonesia. Indian Journal of Traditional Knowledge, in press.
Plumwood, V. (1993) Feminism and the mastery of nature. London, Routledge.
Polunin, N.V.C. (1984) Do traditional marine reserves conserve? A view of Indonesian and New
Guinean evidence. Senri Ethnological Studies, 17, pp.267-283.
Poole, N.D. & Penrose Buckley, C. (2006). Innovation challenges, constraints and opportunities for
the rural poor. Background paper for the International Fund for Agricultural Development,
Rome. Available from http://www.ifad.org/events/gc/29/panel/e/poole.pdf.
Posey, D.A. ed. (1999) Cultural and spiritual value of biodiversity: A complimentary contribution to
the global biodiversity assessment. London, Intermediate Technology Publications.
Pretty, J. (1995) Regenerating Agriculture: Policies and practice for sustainability and self-reliance.
London, Earthscan.
Pretty, J. (2002) Agri-Culture: Reconnecting people, land and nature. London, Earthscan.
Pretty J. (2004) How nature contributes to mental and physical health. Spirituality and Health
International, 5, pp.68-78.
Pretty, J. (2006) The earth only endures. London, Earthscan, in press.
Pretty, J., Guijt, I., Thompson, J. & Scoones, I. (1995) A Trainers' Guide to Participatory Learning
and Action. IIED Participatory Methodology Series 1. London, IIED, p.250.
Pretty, J., Peacock, J., Sellens, M. & Griffin, M. (2005) The mental and physical health outcomes of
green exercise. International Journal of Environmental Health Research, 15, pp.319-337.
Pretty, J. & Smith, D.J. (2004) Social capital in biodiversity conservation and management.
Conservation Biology, 18, pp.631-638.
Pyle, R.M. (2001) The rise and fall of natural history: How a science grew that eclipsed direct
experience. Orion, 20, pp.17-23.
Pyle, R.M. (2002) Eden in a vacant lot: Special places, species, and kids in the neighbourhood of
life. In: Kahn, P.H. & Kellert, S.R. eds. Children and nature: Psychological, sociocultural,
and evolutionary investigations. Massachusetts, The MIT Press, pp.305-328.
Pyle, R.M. (2003) Nature matrix: Reconnecting people and nature. Oryx, 37, pp.206-214.
Rachman, A.M.A. (1993) Tribal information capture and environmental knowledge. In: Proceedings
of Lembaga Ilmu Pengetahuan Indonesia (LIPI) (Indonesian Institute of Science)
International Seminar on Bajau Communities, November 22-25, 1993, Jakarta Indonesia.
Jakarta, LIPI.
Rahman, A. (2004). Development of an integrated traditional and scientific knowledge base: A
mechanism for accessing and documenting traditional knowledge for benefit sharing,
sustainable socio-economic development and poverty alleviation. In: Twarog, S. & Kapoor,
P. eds. Protecting and promoting traditional knowledge: Systems, national experiences and
Rao, R.R. (1994) Biodiversity in India: Floristic aspects. India, Bishen Singh Mahendra Pal Singh,
Dehra Dun.
Rao, K.S., Nautiyal, S., Maikhuri, R.K. & Saxena, K.G. (2003) Local peoples knowledge, aptitude
and perceptions of planning and management issues in Nanda Devi Biosphere Reserve,
India. Environmental Management, 31 (2), pp.168-181.
Rappaport, R.A. (1979) Cognized models. In: Rappaport, R.A. ed. Ecology Meaning and Religion.
California, North Atlantic Books, pp. 97-144.
Robbins, L.C. (1935) An essay on the nature and significance of economic science. London,
Macmillan.
244
Robinson, K. & Paeni, M. (1998) Living through histories: Culture, history and social life in south
Sulawesi. Canberra, Australian National University, Department of Anthropology.
Roscoe, J.T. (1975) Fundamental research statistics for the behavioural sciences, 2nd ed. New York,
Holt Rinehart & Winston.
Rose, D. ed. (2000) A tale of two villages. Unpublished.
Rose, F. (1981) The wild flower key: British Isles - N.W. Europe. London, Penguin Books.
Rosegrant, M.W. & Hazell, P.B.R. (2001) Transforming the rural Asian economy: The unfinished
revolution. Hong Kong, Oxford university Press.
Roth, R. (2004) Spatial organisation of environmental knowledge: Conservation conflicts in the
inhabited forest of northern Thailand. Ecology and Society, 9 (5), Available from Ecology
and Society website http://www.ecologyandsociety.org/vol9/iss3/art5/.
Royal Botanic Gardens Kew (2003) SEPASAL database. Available from Kew Gardens website
http://www.rbgkew.org.U.K./ceb/sepasal/internet/.
Rudd, M.A., Tupper, M.H., Folmer, H. & van Kooten G.C. (2003) Policy analysis for tropical
marine reserves: Challenges and directions. Fish and Fisheries, 4, pp.65-85.
Ruddle, K. (2000) Systems of knowledge: Dialogue, relationships and process. Environment,
Development and Sustainability, 2, pp.277-304.
Ruiz Muller, M. (2004) Regulating bioprospecting and protecting indigenous peoples knowledge in
the Andean community: Decision 391 and its overall impacts in the region. In: Twarog, S. &
Ryan, R. L. (2005) Exploring the effects of environmental experience on attachment to urban natural
areas. Environment and Behaviour, 37 (1), pp.3-42.
Sahai, S. (2004) Commercialisation of traditional knowledge and benefit sharing. In: Twarog, S. &
Salmon, E. (2000) Kincentric ecology: Indigenous perceptions of the human-nature relationship.
Ecological Applications, 10 (5), pp.1327-1332.
Samson, C. (2002) A Way of Life that Does Not Exist: Canada and the Extinguishment of the Innu.
Newfoundland, St. Johns, ISER Press.
Samson, C. & Pretty, J. (2005) Environmental and heath benefits of hunting lifestyles and diets for
the Innu of Labrador. Food Policy, in press.
Sargent, C.F. & Browner, C.H. (2004) Gender politics and the distribution of knowledge about
plants for reproductive health in Jamaica. In: Proceedings of the International Society of
Ethnobiology 9th International Congress and Society for Economic Botany 45th Annual
meeting, June 14, 2004, Canterbury UK.
Scoones, I., Melnyk, M. & Pretty, J. (1992) The hidden harvest: Wild foods and agricultural
systems: A literature review and annotated biography. London, IIED, The Sustainable
Agriculture Programme.
Scoones, I. & Thompson, J. ed. (1994) Beyond farmer first: Rural peoples knowledge, agricultural
research and extension practice. London Intermediate Technology, Publications.
SDNP (2002) Geography and major biomes. Available from SDNP (Sustainable Development
Networking Programme) website http://sdnp.delhi.nic.in/resources /biodiv/biodivframe.html.
Shankar, D. (1998) Conserving a community resource: Medicinal plants. In: Kothari, A., Pathak, N.,
Anuradha, R.V. & Taneja, B. eds. Communities and conservation: Natural resource
management in south and central Asia. London, Sage Publications.
245
Shiva, V. (1997) Staying alive: Women, ecology and development. London, Zed Books.
Schoemaker, J. (2004) Social science sampling. Invited lecture at the WWF and TNC workshop
session: Social science research, December 2004, Bali.
Shrestha, P.M. & Dhillion, S.S. (2003) Medicinal plant diversity and use in the highlands of
Dolakha district, Nepal. Journal of Ethnopharmacology, 86, pp.81-96.
Sillitoe, P. (1998) The development of indigenous knowledge: A new applied anthropology. Current
Anthropology, 39, pp.223-252.
Singh, R.K. & Singh, P.K. (2005) Fertility management dynamics of soil: Exploration of farmers
hidden wisdom. Asian Agri-History, 9, pp.291-303.
Singh, R.K., Singh, D. & Sureja, A.K. (2006a) Community knowledge and conservation of
indigenous biodiversity: Exploration of hidden wisdom of Monpa tribe. The Journal for
Transdisciplinary Research in Southern Africa, 2 (1), pp. 73-102.
Singh, R.K., Sureja, A.K. & Singh, D. (2006b) Amta and Amti (Hibiscus sabdariffa L.): Cultural
and agricultural dynamics of agrobiodiversity conservation. Indian Journal of Traditional
Knowledge, 5 (1), pp.151-157.
Skutnabb-Kangas, T. (2004) On biolinguistic diversity: linking language, culture and (traditional)
ecological knowledge. Invited plenary lecture at the interdisciplinary seminar: At the limits
of language. Madrid, Universidad Autnoma de Madrid and Cosmocaixa, Department of
Biology and Department of Linguistics and Philosophy.
Smith, D. (2004) Interim marine field report July-August 2004. Available from Operation Wallacea
website http://www.opwall.com/Library/Indonesia/marine%20management%20and%20
overall.shtml.
Smith D.J., Pilgrim, S.E. & Cullen, L. (2006) Coral reefs and people. In: Pretty, J., Morrison, J.,
Ball, A., Benton, T., Guivant, J., Lee, D., Orr, D., Pfeffer, M. & Ward, H. eds. Sage
Handbook on Environment and Society. London, Sage Publications, in press.
SOC (2000) Standard Occupation Classification 2000. Available from ONS (Office of National
statistics) website http://www.statistics.gov.uk/methods_quality /ns_sec/soc2000.asp
Soleri, D. & Cleaveland, D. (1993) Seeds and strength for Hopis and Zunis. Seedling, 10 (4), pp.1318.
Stocks, A., McMahan, B. & Taber, P. (2006) Beyond the map: Indigenous and colonist impacts and
territorial defence in Nicaraguas BOSAWAS Biosphere Reserve. Working paper. Idaho,
Idaho State University, Department of Anthropology.
Sustainable Development International (2004) World Urban Forum. Available from Sustainable
Development International website http://www.sustdev.org/index.php?
option=com_events&task=view_detail&agid=6&year=2004&month=09&day=13&Itemid=1
Talay, I., Gunduz, S. & Akpinar, N. (2004) On the status of environmental education and awareness
of undergraduate students at Ankara University, Turkey. International Journal of
Environment and Pollution, 21 (3), pp.293-308.
Tamil Nadu Agriculture Department (2005) Policy Note 2004-2005: Agriculture. Available from
Government of Tamil Nadu website www.tn.gov.in/policynotes/agri2004-05-2.htm.
Tamil Nadu Human Development Report (2003) Synopsis of Tamil Nadu HDR. Available from
UNDP website www.undp.org.in/hdrc/shdr/TN/.
Tamil Nadu Human Development Report (2005) Status of human development in Tamil Nadu.
Available from Government of Tamil Nadu website http://www.undp.org.in/
hdrc/shdr/TN/Chp2.pdf.
246
Tansey, G. (2004) A food system overview. In: Twarog, S. & Kapoor, P. eds. Protecting and
promoting traditional knowledge: Systems, national experiences and international
dimensions. Geneva, United Nations, pp.41-58.
Tendring District Council (2004) Brightlingsea. Available from Tendring District Council website
http://www.tendringdc.gov.uk/localplan/written/cpt11.htm.
Teune, H. (1990) Comparing countries: Lessons learned. In: yen, E. ed. Comparative methodology
: Theory and practice in international social research. London, Sage Publications.
The Countryside Agency (2004) The Countryside Agency publications. Available from The
Countryside Agency website http://www.countryside.gov/Publications/articles/
Publication_tcm2-19382.asp#.
Thirsk, J. (1957) English peasant farming: The agrarian history of Lincolnshire from Tudor to
recent times. London, Routledge & Kegan Paul.
TIDCO (2002a) Tamil Nadu policies: Contract/corporate sector farming policy. Available from
TIDCO (Tamil Nadu Industrial Development Corporation) website http://www.
tidco.com/tn_policies/tn_policies/contract_ farming.asp.
TIDCO (2002b) Focus Tamil Nadu: Agriculture profile. Available from TIDCO (Tamil Nadu
Industrial Development Corporation) website http://www.tidco.com/tn_policies/focus
_tamilnadu/agriculture_ profile.asp.
TIDCO (2002c) Focus Tamil Nadu: Economic profile. Available from TIDCO (Tamil Nadu
Industrial Development Corporation) website http://www.tidco.com/tn_policies/
focus_tamilnadu/economic_profile_of_tamilnadu1.asp.
Tomascik, T., Mah, A.J., Nontji, A. & Moosa, M.K. (1997) The ecology of the Indonesia Seas. The
ecology of Indonesia series, Vol. 8. Oxford, Oxford University Press.
Tourism of Indonesia (2005) South-east Sulawesi. Available from Tourism of Indonesia website
http://www.indonesia-tourism.com/south-east-sulawesi/history.html
Town Guides (2004) West Mersea. Available from Town Guides website http://www.townguides.org/tgo/essex/west_mersea/introduction-welcome_to.asp.
Travellers World (2004) England. Available from Travellers World website http://www.travellersworld.info/countries/c1.html.
Tsuji, L.J.S. (1996) Loss of Cree traditional ecological knowledge in the western James Bay region
of Northern Ontario, Canada: A case study of the sharp-tailed grouse, Tympanuchus
phasianellus phasianellus. The Canadian Journal of Native Studies, 16 (2), pp.283-292.
Turner, N.J., Ignace, M.B. & Ignace, R. (2000) Traditional ecological knowledge and wisdom of
aboriginal peoples in British Columbia. Ecological Applications, 10, pp.1275-1287.
Twyman, C., Morrison, J. & Sporton, D. (1999) The final fifth: Autobiography, reflexivity and
interpretation in cross-cultural research. Area, 31 (4), pp.313-325.
Tyrrell, M. (2005) The social reproduction and growth through practice of marine knowledge and
skill. Unpublished.
UNDP (2003) UN Human Development Index Report. Available from UNDP website
http://hdr.undp.org/reports/global/2003/.
UNDP (2004) Model Laws for the protection of biodiversity knowledge in developing countries.
Available from UNDP website www.undp.org/tcdc/bestprac/social/cases/03biodivesrity.html.
UNDP Human Development Reports (2005) Human development challenges in a democratic,
decentralised Indonesia. Available from UNDP website
http://www.hdr.undp.org/report/detail_reports.cfm?view=144.
247
UNDP TCDC (2003a) Preparation of neem biopesticides at farm level. Available from UNDP
TCDC (Technical Cooperation among Developing Countries) website
http://www.undp.org/tcdc/bestprac/social/cases/13-neem.htm.
UNDP TCDC (2003b) The promotion of indigenous systems of medicine in India. Available from
UNDP TCDC (Technical Cooperation among Developing Countries) website
http://www.undp.org/tcdc/bestprac/social/cases/01-medicine.htm.
Vandebroek, I., Van Damme, P., Van Puyvelde, L., Arrazola, S. & De Kimpe, N. (2004) A
comparison of traditional healers medicinal plant knowledge in the Bolivian Andes and
Amazon. Social Science and Medicine, 59, pp.837-849.
Veitayaki, J. (1997) Traditional marine resource management practices used in the Pacific islands:
An agenda for change. Ocean and Coastal Management, 37, pp.123-136.
Warren, K. ed. (1997) Ecofeminism: Women, culture, nature. Bloomington, Indiana University
Press.
Wenzel, G.W. (1999) Traditional ecological knowledge and Inuit: Reflections on TEK research and
ethics. Arctic, 52, pp.113-124.
Westermann, O., Ashby, J. & Pretty, J. (2005) Gender and social capital: The importance of gender
differences for the maturity and effectiveness of natural resource management groups. World
Development, 33 (11), pp.1783-1799.
Wikipedia (2006a). Wind in the Willows. Available from Wikipedia website
http://en.wikipedia.org/wiki/The_Wind_in_the_Willows
Wikipedia (2006b). Beatrix Potter. Available from Wikipedia website
http://en.wikipedia.org/wiki/Beatrix_Potter
Worldpress (2006) World maps and country profiles. Available from Worldpress website
http//:www.worldpress.org/map.cfm.
Wright, A. (1985) Marine resource use in Papua New Guinea: Can traditional concepts and
contemporary development be integrated? In: Ruddle, K. & Johannes, R. eds. The traditional
knowledge and management of coastal systems in Asia and the Pacific. Indonesia, Jakarta
Pusat, UNESCO, pp.79-100.
WWF (2006) WWF-UK. Available from WWF (World Wide Fund for Nature) website
http://www.wwf.org.uk/core/index.asp
Young, A. (1970) General view of the agriculture of Lincolnshire: A reprint of the work drawn up
for the consideration of the Board of Agriculture and Internal Improvement 1813. Devon,
David & Charles.
Young, K.R. (2002) Minding the children: Knowledge transfer and the future of sustainable
agriculture. Conservation Biology, 16, pp.855-856.
Yupari, A., Jaramillo, L., Lojenga, R.K., Briceo, S. & Snchez, R. (2004) UNCTADs BIOTRADE
Initiative: Some considerations on access, benefit sharing and traditional knowledge. In:
Zent, S. (2001) Acculturation and ethnobotanical knowledge loss among the Piaroa of Venezuala:
Demonstration of a quantitative method for the empirical study of traditional ecological
knowledge change. In: Maffi, L. ed. On biocultural biodiversity: Linking language,
knowledge, and the environment. London, Smithsonian Institution Press, pp.190-211.
Zhang, X. (2004) Traditional medicine: Its importance and protection. In: Twarog, S. & Kapoor, P.
eds. Protecting and promoting traditional knowledge: Systems, national experiences and
248
Appendix I
Examples of Species Flashcards used in Fieldwork
Species 18
Species 28
249
Species 41
Species 28
250
Appendix II
India case study: Wild flowering plant species list

Species no.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
Latin name
Common name
Portulaca oleracea
Tamarix aphylla
Grewia flavescens
Grewia villosa
Azadirachta indica
Dalbergia sissoo
Vigna aconitifolia
Bauhinia rufescens
Caesalpinia coriaria
Senna siamea
Tamarindus indica
Acacia baileyana
Acacia farnesiana
Acacia leucophloea
Acacia longifolia
Acacia melanoxylon
Acacia tortilis
Albizia lebbeck
Albizia saman
Pithecellobium dulce
Prosopis juliflora
Citrullus colocynthis
Citrullus lanatus
Coccinia grandis
Eucalyptus crebra
Calotropis procera
Salvadora persica
Datura metel
Amaranthus graecizans
Withania somnifera
Euphorbia tirucalli
Aloe vera
Bothriochloa insculpta
Bothriochloa pertusa
Brachiaria ramosa
Cenchrus biflorus
Cenchrus ciliaris
Cenchrus pennisetiformis
Cenchrus setigerus
Common purslane
Athel tamarisk
Rough-leaved raisin
Common spike thorn
Neem tree
Indian rosewood
Marechal
Kharroub
Divi-divi
Kassod tree
Tamarind
Cootamundra wattle
Sweet acacia
White-barked acacia
Sydney golden wattle
Blackwood
Umbrella thorn
Woman's tongue
Rain tree
Manila tamarind
Mesquite
Wild gourd
Watermelon
Ivy gourd
Narrow-leaved ironbark
Milkweed
Toothbrush tree
Downy thorn-apple
Spreading pigweed
Indian ginseng
Pencil tree
Aloe
Creeping bluegrass
Indian bluegrass
Browntop millet
Indian sandbur
Buffel grass
Slender buffel grass
Birdwood grass
251
40.
41.
42.
43.
44.
45.
46.
47.
48.
Cymbopogon martinii
Cynodon dactylon
Dactyloctenium aegyptium
Dichanthium annulatum
Pennisetum pedicellatum
Saccharum spontaneum
Setaria verticillata
Barleria prionitis
Mangifera indica
Palmarosa
Bermuda grass
Egyptian grass
Kleberg blue-stem
Kyasuma grass
Wild sugarcane
Bristly foxtail
Porcupine flower
Mango
252
Appendix III
UK case study: Wild species list

Species no.
Latin name
Common name
Mammals
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Sorex araneus
Oryctolagus cuniculus
Talpa europaea
Sylvaemus sylvaticus
Clethrionomys glareolus
Rattus rattus
Sciurus carolinensis
Pipistrellus pipistrellus
Erinaceus europaeus
Felis catus
Lepus capensis
Mustela nivalis
Vulpes vulpes
Meles meles
Capreolus capreolus
Mustela vison
Lutra lutra
Halichoerus grypus
Common shrew
Rabbit
Mole
Wood mouse
Bank vole
Common rat
Grey squirrel
Pipistrelle bat
Hedgehog
Feral cat
Brown hare
Weasel
Red fox
Badger
Roe deer
Mink
Otter
Grey seal
Birds
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Troglodytes troglodytes
Fringilla coelebs
Turdus philomelos
Erithacus rubecula
Passer domesticus
Parus caeruleus
Columba palumbus
Phylloscopus trochilus
Alauda arvensis
Prunella modularis
Anthus pratensis
Parus major
Phasianus colchicus
Emberiga citrinella
Sturnus vulgaris
Turdus philomelos
Corvus frugilogus
Corvus corone
Wren
Chaffinch
Blackbird
Robin
House sparrow
Blue tit
Wood pigeon
Willow warbler
Skylark
Dunnock/Hedge sparrow
Meadow pipit
Great tit
Pheasant
Yellow hammer
Starling
Song thrush
Rook
Carrion crow
253
37
38
Sylvia communis
Phylloscopus collybita
White throat
Chiffchaff
Flowering plants
39
Ranunculus acris
40
Ranunculus ficaria
41
Viola riviniana
42
Stellaria holostea
43
Vicia sativa
44
Lotus corniculatus
45
Potentilla erecta
46
Convolvulus arvensis
47
Veronica officinalis
48
Glechoma hederacea
49
Prunella vulgaris
50
Plantago lanceolata
51
Bellis perennis
52
Silybum marianum
53
Hieracium pilosella
54
Hyacinthoides nonscripta
Meadow buttercup
Lesser celandine
Common dog violet
Greater stitchwort
Common vetch
Common birds foot trefoil
Tormentil/Cinquefoil
Field bindweed
Heath speedwell
Ground ivy
Self-heal
Ribwort plantain
Daisy
Milk thistle
Mouse-ear hawkweed
English bluebell
Crops
55
56
57
58
59
60
61
62
63
64
65
66
67
Beta vulgaris
Brassica napus
Avena sativa
Hordeum vulgare
Secale cereale
Triticosecale rimpaui
Triticum aestivum
Zea mays
Solanum tuberosum
Phaseolus vulgaris
Pisum satium
Spinacea oleracea
Vicia faba
Sugar beet
Oil seed rape
Oats
Barley
Rye
Triticale
Wheat
Maize
Potato
French bean
Pea
Spinach
Broad bean
Marine
68
69
70
71
72
73
74
75
76
77
Crassostrea angulata
Crassostrea gigas
Asterias rubens
Nucella lapillus
Mytilus edulis
Patella vulgata
Littorina littorea
Gibbula umbilicalis
Cerastoderma edule
Carcinus maenas
Common oyster
Portuguese oyster
Common starfish
Dog whelk
Mussel
Common limpet
Edible periwinkle
Topshell
Edible cockle
European shore crab
254
78
79
80
81
82
83
84
85
86
Cancer pagurus
Nereis diversicolor
Arenicola marina
Fucus vesiculosus
Ascophyllum nodosum
Salicornia europaea
Atriplex portulacoides
Limonium vulgare
Eryngium maritimum
Edible crab
Ragworm
Lugworm
Bladder wrack
Knotted/Egg wrack
Glasswort/Samphire
Sea purslane
Common sea lavender
Sea holly
255
Appendix IV
Indonesia case study: Wild species list

Species no.
Latin name
Common name
Marine
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Bohadschia
Holothuria scabra
Chelonia midas
Triaenodon obesus
Thunnus albacares
Dugong dugon
Protoreaster nodosus
Diadema
Cheilinus undulatus
Tridacna gigas
Octopus cyanea
Panulirus ornatus
Crocodylus porosus
Balistapus undulatus
Chaetodontidae
Pomacentridae
Lethrinus genivittatus
Parupeneus barberinus
Scarus psittacus
Siganus argenteus
Cromileptes altivelis
Lutjanus campechanus
Laticauda colubrina
Taeniura lymma
Sea cucumber (Bohadschia spp.)

Sea cucumber (Holothuria spp.)
Green turtle
White tip reef shark
Tuna
Dugong
Chocolate chip star
Sea urchin
Napoleon wrasse
Giant clam
Octopus
Ornate rock lobster
Crocodile
Orange-lined triggerfish
Butterflyfish
Damselfish
Threadfin Emperor
Dash-dot goatfish
Common parrotfish
Streamlined spinefoot rabbitfish
Grouper
Snapper
Banded sea krate
Blue-spotted ray
Animals
25
26
27
28
29
30
31
32
33
34
35
Rattus rattus
Mus musculus
Cynopterus brachyotis
Varanus salvator
Python reticulatus
Hemidactylus frenatus
Cervus elaphus
Capra hircus
Bos taurus
Felis catus
Canis lupus
Rat
Mouse
Fruit bat
Monitor lizard
Snake
Gecko
Deer
Goat
Cow
Cat
Dog
256
Birds
36
37
38
39
40
41
42
43
44
45
46
47
48
49
Gallus domesticus
Cacatua sulphurea
Megapodius reinwardt
Todirhamphus chloris
Merops ornatus
Tanygnathus megalorynchos
Chalcophaps indica
Haliaeetus leucogaster
Egretta sacra
Fregata minor
Oriolus chinensis
Saxicola caprata
Passer montanus
Cinnyris jugularis
Flowering plants
50
Ipomoea pes-caprae
51
Barringtonia asiatica
52
Casuarina equisetifolia
53
Rhizophora stylosa
54
Mangifera indica
55
Tilia europaea
56
Carica papaya
57
Artocarpus heterophyllus
58
Musa acuminata
59
Tamarindus indica
60
Moringa ovalifolia
61
Dioscorea
62
Hevea brasiliensis
63
Coffea arabica
64
Theobroma cacao
65
Cocos nucifera
66
Rutaceae
67
Syzygium aromaticum
68
Zea mays
69
Oryza sativa
70
Gossypium hirsutum
71
Capsicum annum
72
Manihot esculentum
73
Piper nigrum
74
Anacardium occidentale
Chicken
Yellow-crested cockatoo
Orange-footed scrubfowl
Collared kingfisher
Rainbow bee-eater
Great-billed parrot
Emerald dove
White-bellied sea eagle
Pacific reef egret
Great frigatebird
Black-naped oriole
Pied chat
Tree sparrow
Olive-backed sunbird
Goatfoot/Beach morning glory
Box Fruit
She-oak
Mangrove tree
Mango
Common lime
Papaya
Jackfruit
Banana
Tamarind
Murungai
Yam
Rubber
Coffee
Cocoa
Coconut
Citrus lemon
Clove
Maize
Rice
Cotton
Chili
Cassava
Black pepper
Cashew
257
Appendix V
India case study: Data recording sheet

Village:
Name:
Number in interview group:
Age group:
<20 yrs
Gender:
M
Spp Species name
no.
1 Portulaca oleracea
2 Tamarix aphylla
3 Grewia flavescens
4 Grewia villosa
5 Azadirachta indica
6 Dalbergia sissoo
7 Vigna aconitifolia
8 Bauhinia rufescens
9 Caesalpinia coriaria
10 Senna siamea
11 Tamarindus indica
12 Acacia baileyana
13 Acacia farnesiana
14 Acacia leucophloea
15 Acacia longifolia
16 Acacia melanoxylon
17 Acacia tortilis
18 Albizia lebbeck
19 Albezia saman
20 Pithecellobium dulce
21 Prosopis juliflora
22 Citrullus colocynthis
23 Citrullus lanatus
24 Coccinia grandis
25 Eucalyptus crebra
26 Calotropis procera
27 Salvadora persica
28 Datura metel
29 Amaranthus
graecizans
30 Withania somnifera
31 Euphorbia tirucalli
>20 yrs
F
No. that
No. that
No. that
recognise know name know uses
species
Uses
258
32 Aloe vera
33 Bothriochloa
insculpta
34 Bothriochloa pertusa
35 Brachiaria ramosa
36 Cenchrus biflorus
37 Cenchrus ciliaris
38 Cenchrus
pennisetiformis
39 Cenchrus setigerus
40 Cymbopogon
martinii
41 Cynodon dactylon
42 Dactyloctenium
aegyptium
43 Dichanthium
annulatum
44 Pennisetum
pedicellatum
45 Saccharum
spontaneum
46 Setaria verticillata
47 Barleria prionitis
48 Mangifera indica
259
Appendix VI
UK case study: Data recording sheet

Age:
Male/Female:
Occupation:
Village/Town live:
Years lived in area:
Childhood: rural/urban/coastal (circle)
Extra (e.g. member of any environmental groups such as The National Trust, keen gardener or
studied this area):
Spp
no
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Recognise
(tick or
cross)
Name
Use
Extra
260
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
261
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
262
How often do you spend time in the countryside? Please tick accordingly
Daily
Weekly
Monthly
Rarely
Never
If so, for what reason do you visit the countryside?
Approximately how much was your annual household income for 2003?
< 15 000
15 001 - 20 000
20 001 - 25 000
25 001 - 35 000
35 001 - 50 000
50 001 - 75 000
>75 000
Dont know
Where did you get your own knowledge of the local environment from and how much did you get
from each source? Please rate accordingly e.g. 30% books 70% television
Books/magazines
Television
Parents/relatives
Friends/locals
Hobbies (e.g. fishing, gardening, bird watching)
Occupation (e.g. farming)
School/further education
Other (please state)
To what extent do you think it is important to maintain knowledge of local plant and animal species?
Please tick accordingly
Extremely important Local ecological knowledge is vital to the maintenance of local wildlife
263
and its uses as a life source for future generations.

Quite important Local knowledge should be maintained to protect local wildlife, but other
factors take higher priority.
Indifferent If people have an interest in wildlife then they should learn about it, but it is not a
problem if no-one is interested.
Quite unimportant A decline in local ecological knowledge is a shame, but not a problem.
Extremely unimportant A loss in ecological knowledge is expected since it is irrelevant in
todays society.
The data collected from these surveys will not be given to any external parties and will only be used
in aggregate for data collation and analysis for the purpose of this project
Contact Details
If you do not mind being contacted again regarding this research then please complete this section.
Your details will not be passed on to any other parties.
Full Name:
Address:
Telephone:
E-mail:
264
Appendix VII
Indonesia case study: Fishers data recording sheet
Date
Time
Translator
Village
Respondent
Bajo
Non-Bajo
Equipment:
Electricity
Generator
Satellite dish
Radio
Telephone
Outhouse
Wood
Brick
Concrete
No walls
Other
Walls:
Bamboo
Roof:
Coconut
Metal
Asbestos
Bamboo
Wood
Concrete
Floor:
Language:
Bajo
Bahasa Kaledupa
Bahasa Indonesia
1. Your gender?
Male
Female
Other
265
2. What is your age?

15-19 years
20-29 years
30-49 years
>50 years
3. How do you make most of your money and do you have any other ways of making money?
4. Have you always lived here?
Yes
No
5. If no, how long have you lived here and where did you live before?
Spp
no
Recognise
(tick or
cross)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Name
Use
266
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
267
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
6. Who taught you about the plants and animals here?

Me
Parents
Other family
members
Friends
7. Do you tell other people about the plants and animals here?
Yes
No
Other
268
8. How do you fish?

Hand line
Hand line
(reef wall)
(reef flat)
Refuge/pot
trap
Bubu trap
Gill net
Seine net
Other net
Trawl
Fish fence
Hand/
Poison/
bomb
Octopus
fishing
Other - state
spear gun
Reef
gleaning
(surabi)
(suku)
9. What are the most important threats to plants and animals here?
Too many people
Fishermen from other places
Cyclones, storms, tsunamis
Cutting down mangroves
Tourism
Traditional fishing
techniques
No-one watching over fishing
Gleaning
Climate change (local term)
Coral mining
Destructive fishing (give e.g.)
Waste and dirty water
Too many boats
Catching animals and fish

than we are not allowed to
(give e.g.)
Too many fishermen
Other (state)
10. Can you tell the difference between a long line and bomb-caught fish?
Yes
No
11. If yes, if at market there was fish caught by long line and fish caught by bomb fishing, which
would you buy?
Long line?
Bomb?
Either?
12. Have you been visited by any other organizations like Operation Wallacea?
Yes
No
269
14. The plants and animals of the sea and land are
special and should be left alone
15. I believe there is much I can do to look after
the plants and animals here
16. Plants and animals are here mostly for our use
17. The plants and animals here are not important
to me at all
18. The plants and animals here are important to
me as a way of making money
19. The plants and animals here are important as
food
20. The plants and animals here are important for
medicines
building houses
the village as a whole and who we are
my religion
24. The marine park is good for protecting the
Strongly
disagree
Disagree
Indifferent
Agree
Strongly
agree
13. If yes, which organizations?
270
plants and animals for my children in the future

25. The marine park is not good for me and my
family now
26. In the future I want my children to use the
plants and animals here as we do
27. I think it is important to tell others what I
know about the plants and animals here
28. I do not think that what I know about the
plants and animals here will be important in 50
years time
29. Even if it damages other plants and animals, I
believe in fishing in a way that catches the most
fish in the shortest time
30. I understand why the marine park is there and
what it does
31. I know most of the marine park rules
32. I would be willing to help keep watch over the
reef to stop destructive fishing [give examples of
destructive fishing]
33. I put any young fish that I catch back.

34. If other people keep taking all the plants and
animals they want to use then one day we will
run out
35. I use every fish I catch, there is never any
271
waste
36. The number of fish in the sea here has gone
down
37. There are less different types of fish than there
were in the past
38. I do things to help protect the plants and
animals here
39. I trust traditional healers (local term) rather
than modern doctors to cure mine and my
familys ailments
272
Indonesia case study: Farmers data recording sheet

Date
Time
Translator
Village
Respondent
Equipment:
Electricity
Generator
Satellite dish
Radio
Telephone
Outhouse
Wood
Brick
Concrete
No walls
Other
Walls:
Bamboo
Roof:
Coconut
Metal
Asbestos
Bamboo
Wood
Concrete
Floor:
Language:
Bajo
Bahasa Kaledupa
Bahasa Indonesia
Other
1. Your gender?
Male
2. What is your age?
15-19 years
Female
20-29 years
30-49 years
>50 years
3. How do you make most of your money and do you have any other ways of making money?
273
4. Have you always lived here?

Yes
No
5. If no, how long have you lived here and where did you live before?
Spp
no
Recognise
(tick or
cross)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Name
Use
274
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
275
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
6. Who taught you about the plants and animals here?

Me
Parents
Other family
members
Friends
Other
7. Do you tell other people about the plants and animals here?
Yes
No
8. What crops do you farm?

Rice
Cocoa
Coffee
Chili
Coconuts
Cotton
Bananas
Other
276
9. What are the most important threats to plants and animals here?
Too many people
Farmers from other places
Cyclones, storms, tsunamis
Cutting down mangroves
Tourism
Traditional farming
techniques
The use of chemicals on crops

and land
Cutting down forest
Climate change (local term)
Coral mining
Growing larger areas of just

one crop
Waste and dirty water
People growing a smaller

variety of crops
Poor soil quality
Drought
Fishing techniques
Flooding
Other (state)
10. Can you tell the difference between a long line and bomb-caught fish?
Yes
No
11. If yes, if at market there was fish caught by long line and fish caught by bomb fishing, which
would you buy?
Long line?
Bomb?
Either?
12. Have you been visited by any other organizations like Operation Wallacea?
Yes
No
14. The plants and animals of the sea and land

are special and should be left alone
15. I believe there is much I can do to look after
Strongly
disagree
Disagree
Indifferent
Agree
Strongly
agree
13. If yes, which organizations?
277
the plants and animals here

16. Plants and animals are here mostly for our
use
17. The plants and animals here are not
important to me at all
me as a way of making money
19. The plants and animals here are important as
food
medicines
building houses
the village as a whole and who we are
my religion
24. The marine park is good for protecting the
plants and animals for my children in the future
25. The marine park is not good for me and my
family now
26. In the future I want my children to use the
plants and animals here as we do
278
27. I think it is important to tell others what I

know about the plants and animals here
28. I do not think that what I know about the
plants and animals here will be important in 50
years time
29. Even if it damages other plants and animals, I
believe in farming in a way that produces the
largest crop yield in the shortest time
30. I understand why the marine park is there
and what it does
31. I know most of the marine park rules

32. I would be willing to learn about and teach
others of sustainable farming techniques
33. I have to use chemicals to increase the
amount of food I grow
34. If other people keep taking all the plants and
animals they want to use then one day we will
run out
35. I use all the food I grow, there is never any
waste
36. The number of wild plants here has gone
down
37. There are less different types of wild plants
than there were in the past
279
38. I do things to help protect the plants and

animals here
39. I trust traditional healers (local term) rather
than modern doctors to cure mine and my
familys ailments
280
Appendix VIII
Recalculating the HDI figure for the Indonesian study site
Life expectancy
The mean age of household heads (eldest person in a household) was 10 years younger for Bajo than
Kaledupans therefore the mean life expectancy can be reduced by 5 years to account for 50% of
interviewees deriving from Bajo villages.
Original life expectancy (National figure) (UNDP, 2003) = 65.8 years
Amended life expectancy = 61.8 years
Adult literacy
The mean % of Kaledupan and Bajo combined that have acquired some form of primary level
schooling is approximately 50%. On average only 1 in 10 Bajo can read and write (C. Majors,
personal communication, 14th August 2005).
Original adult literacy (National figure) (UNDP, 2003) = 87.9%
Amended adult literacy = 50.0%
Combined gross enrolment
Since there is no estimation of the % of Bajo children that attend primary, secondary and tertiary
schooling, national figures were used. This should not greatly affect the estimate since enrolment is
only weighted at one-ninth of the whole calculation.
Gross enrolment (National figure) (UNDP, 2003) = 66
GDP
Average primary household income for Bajo was found to be on average 50% less than Kaledupan
levels therefore GDP was reduced by 25% to account for 50% of interviewees deriving from Bajo
villages.
Original GDP (National figure) (UNDP, 2003) = 3361 p pp US$
Amended GDP = 2521 p pp US$
HDI
HDI was recalculated using these amended figures.
Original HDI (National figure) (UNDP, 2003) = 0.70 (0.64 for SE Sulawesi region omitting Bajo)
Amended HDI = 0.57
This is likely to be a highly conservative estimate but gives a better indication of the human
development levels of the region in the light of the high indigenous population living there.

PILGRIM, 2006 - A Cross Cultural Study Into LEK - Tese

Uploaded by

Copyright:

Available Formats

PILGRIM, 2006 - A Cross Cultural Study Into LEK - Tese

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PILGRIM, 2006 - A Cross Cultural Study Into LEK - Tese

Uploaded by

Copyright:

Available Formats

A Cross-Cultural Study into Local Ecological Knowledge

A thesis submitted for the degree of Ph.D.

1.1. Composition of LEK

Chapter 3. India case study: Virudhunagar, Tamil Nadu

3.2. Site specific methodology

Chapter 4. UK case study: Comparing rural Lincolnshire, urban Kent

4.2. Site specific methodology

Chapter 5. Indonesia case study: Kaledupa, SE Sulawesi

5.2. Site specific methodology

Chapter 6. Country comparisons

Chapter 7. General discussion

7.1. Developing a standardised protocol for ecological knowledge assessment

Chapter 3. India case study: Virudhunagar, Tamil Nadu

UK case study: Comparing rural Lincolnshire, urban Kent

Figure 4.3.1. Comparing UK site differences in (A) level of total ecoliteracy,

Chapter 6. Country comparisons

India case study: Virudhunagar, Tamil Nadu

UK case study: Comparing rural Lincolnshire, urban Kent and

Annual Survey of Hours and Earnings

United Nations Development Programme

1.1. Composition of LEK

Western scientific knowledge

1.2. Transfer of LEK

1.3. The functional values of LEK

1.3.1. Natural resource conservation

1.3.1.1.1. The development of local management systems

1.3.1.1.2. The attributes of self-management systems

1.3.1.1.3. The adaptability of self-management systems

1.3.1.1.4. Self-management systems in the future

landscapes, the network of intra-ecosystem connections increases, maximising ecological resilience.

1.3.2. Food security and genetic diversity

1.3.3. Pharmaceutical industry and economic generation

1.3.4. The promise of future findings

1.3.5. Cultural capital

1.4. Factors leading to LEK loss

1.4.1. Urban migration and urbanisation

1.4.2. Modernisation of lifestyles and technology

1.4.3. Extinction of experience

1.4.4. The introduction of schooling and curriculum changes

1.4.5 Economic development and marketisation

1.4.6. Assimilation and integration

1.4.7. Language loss

1.5. General population distribution of LEK

1.5.1. Gender differences

1.5.2. Generational differences

1.6. Aims of thesis

Figure 1.5.1. Levels of ecological knowledge and understanding

2.2. Study sites

Human Development Index (2003)

Figure 2.2.1. The 3 chosen study regions (Worldpress, 2006)

2.3. Ethnobotanical and ethnozoological interview development

2.3.1. Interview changes and methodological development

Highly characteristic group thus much public interest and media

Collate local species images

Validate local species

Validate local species

2.4. Sampling strategy

Table 2.4.1. Code of best practices for LEK research