Duikers, cables, and nets


Material Information

Duikers, cables, and nets a cultural ecology of hunting in a Central African forest
Physical Description:
xx, 416 leaves : ill. ; 29 cm.
Noss, Andrew J., 1964-
Publication Date:


Subjects / Keywords:
Hunting -- Africa   ( lcsh )
Hunting -- Environmental aspects   ( lcsh )
Hunting -- Economic aspects   ( lcsh )
Geography thesis, Ph. D
Dissertations, Academic -- Geography -- UF
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1995.
Includes bibliographical references (leaves 363-415).
Statement of Responsibility:
by Andrew J. Noss.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002058416
oclc - 34345367
notis - AKP6466
System ID:

This item is only available as the following downloads:

Full Text
/'r- '








Copyright 1995


Andrew J. Noss


This research was assisted by a grant from the Joint

Committee on African Studies of the Social Sciences Research

Council and the American Council of Learned Societies with

funds provided by the Ford, Mellon, and Rockefeller

Foundations. Additional funds were provided by two Grants-

in-Aid of Research from Sigma Xi, The Scientific Research

Society; and by the World Wildlife Fund under the United

States Department of Agriculture Agreement No. 93-G-155.

Many people in many places have contributed their

support to this project. Not all can be named here, nor can

words adequately express the debt that I owe. Most

importantly, I would like to thank my family for sustaining

me: James and Delores Samuels; Philip, Cecilia, Haldor,

Darcy, and Katie Noss; and especially my wife Lauren

Samuels. Their visits to Bayanga were especially uplifting.

I would like to thank my committee members for their

support and guidance, and for allowing me to wander off in

my own direction: my chair Ed Malecki, cochair Abe Goldman,

Nigel Smith, Tom Struhsaker, and Richard Bodmer.

Encouragement from Peter Schmidt, Hunt Davis, and the Center

for African Studies at the University of Florida helped me

decide to embark on this journey in the first place. While


I was writing the dissertation, Tom Struhsaker and Ken

Glander also provided invaluable access to facilities at the

Duke University Primate Center.

In addition to members of my committee, many people

have provided guidance by critically reviewing my research

plans and the dissertation: Serge Bahuchet, Russ Bernard,

Richard Carroll, Gustave Doungoube, Tamara Giles-Vernick,

John Hart, Barry Hewlett, Anna Kretsinger, Cecilia Noss,

Philip Noss, Kent Redford, Gery Ryan, Lauren Samuels, David

Wilkie, and Vivian Wilson.

In the Central African Republic, Mr. Modibo Walidou-

Chahir, Ministre de 1'Enseignement Superieur et de la

Recherche Scientifique generously granted me a research

permit, with the assistance of the Ministry's Chef de

Mission Mr. Remain Sopio. Ambassador Robert Gribbin,

Michael Keller, and Guillaumine Gazembeti at the U.S.

embassy went out of their way to help me obtain the

necessary documents.

Richard Carroll first described to me the Bayanga

region and WWF's Dzanga-Sangha project. He and other WWF

staff were continuously supportive and encouraging: in

Washington, Danyelle O'Hara; in Bangui, Doungoube Gustave,

Jean Luabeya, Delphine, Bernardin, Marcel, Frangois, Jean,

and Marc; in Bayanga, Allard Blom, Jean Marc Garreau, Ngaima

Honor&, Yangui Adolphe, Ngatoua Urbain, Victor Babon, and

Mopongo Auguste.

I would like to thank the people of Bayanga, especially

the hunters who taught me about themselves and the forest,

but who must remain nameless; my assistant Ngbongo Yves; my

helpers Blaise, Elbert, and Bertrand; and my neighbors Mama

Ngounja and her family who looked after me.

I would also like to thank the people of Mossapoula for

sharing part of their lives with me, and for teaching me

about their lives today, no longer just "people of the

forest". In particular I thank Makolea, self-proclaimed

"chef ti gbanda" (the chief of the net hunt), and my

assistant Matofi Fernand.

I also learned a great deal from the other researchers,

whose backgrounds, disciplines, and research topics provided

alternative, thought-provoking perspectives on actors and

events in the Bayanga region: Tamara Giles-Vernick, Michele

Goldsmith, David Harris, Anna Kretsinger, Jay Malcolm,

Justina Ray, Louis Sarno, and Andrea Turkalo. The Peace

Corps volunteers Anthony Aufdenkampe, Bonnie Dixon, John

Hanson, Yann Lussiez, Jo Shrout, Margo Stoddard, and Gillian

Grant did likewise. The hospitality at Dzanga, Koungana,

and Bai Hokou camps was particularly appreciated. Finally,

Jean-Marc and Pierrette Garreau and Allard Blom were also

extremely generous.










. iii

. xi

. xv

. xvii


. . . 9

Cultural Ecology . .
The Origins of Cultural Ecology
Carl Sauer . .
Julian Steward .
Cultural Ecology in Theory .
Human ecology .
Systems theory .
Ecological anthropology .
Cultural ecology .
Cultural Ecology in Practice
Subsistence economies .
Developing economies .
Cultural ecology themes .

Conservation and Development .
Biodiversity Conservation .
Socioeconomic Development .
Integration of Conservation and
Development .

Conservation, Development and Wildlife:
The Role of Hunting .
Hunting Methods . .
Pressure on Wildlife Resources .

. 9
. 10
. 10
. 11
. 13
. 13
. 14
. 15
. 16
. 19
. 20
. 22
. 23

. 25
. 26
. 28

. 30

S 38
. 42


Wildlife Management Practices .
The Subsistence Role of Hunting
The Economic Role of Hunting

Conclusion . .


The Ecosystem .
Ecosystem Characteristics .
Species Characteristics .
Atherurus africanus .
Cephalophus callipygus .
Cephalophus dorsalis .
Cephalophus monticola .
Population Characteristics

Population densities .
Age structure and sex ratios
Ecological interactions .
Estimated Population Densities in
the Bayanga Region .
Line transect surveys .
Net hunt encounters .

Line transect versus net hunt encounter
density estimates .

The Human System . .
Political History . .
Economic and Social History .

Concession system .
Coffee plantations .
Logging . .
Diamond mining . .
Integrated conservation and
development .
Safari hunting . .
Exploitation versus conservation
Population Centers in the Bayanga

Region . .
Bayanga . .
Mossapoula . .
Other settlements .

Ecological Impacts of Human Activities
Human Activities and Habitat Change
Implications of Hunting .
Changes in population structure
Changes in population density .
Sustainable harvest rates ..
Changes in the ecosystem .


. 53

. 56

. 57
. 58
. 64
. 66
. 67
. 67
. 68
. 69
. 69

. 73
. 78




I j 1 j

Relations Between the Ecosystem and the Human
System: The Role of Hunting 131
Elephants . ... .131
Hunting Legislation . 135
Bushmeat Markets . 139
Currency Devaluation . ... .143

Conclusion .

. 144


Methodology . .

Description . .
Snare Types . .
Snare Hunting Practices .
Snare Hunting Range . .

Impacts of Cable Snare Hunting on Wildli
Captures . .
Age Structure and Sex Ratios .
Wastage . .
Injured Escapes . .
Protected Species . .
Sustainability of Snare Harvest Rates

Economic Importance of Cable Snare Hunting
Employment . .
Income . .

Comparative Data .

Conclusion . .


Methodology . .

Description . .
Communal Net Hunting Practice .
Beliefs and Rituals .
Seasonality . .
Mossapoula Net Hunting Range


. 203


Impacts of Net Hunting on Wildlife .
Captures . .
Age Structure and Sex Ratios .
Vulnerability . .
Sustainability of Net Hunting Harvest

. 225
. 226
. 229
. 231
. 234



. 148


S. 160
. 161
. 167
. 168
. 170
. 174
S. 175

S. 182
. 183
. 187

. 193





Economic Importance of Net Hunting 240
Time Allocation to Net Hunting .. .241
Returns from Net Hunting . 252

Role of Women in Net Hunting . 262

Forest Hunting Camps . 265

Comparative Data . 272

Conclusion . . 281


Impacts on Wildlife . ... .286
Captures . . 287
Sustainability . ... .291

Economic Importance .. . 296
Employment and Time Allocation 297
Returns . . 298

Conclusion . . 302


Hunting and Cultural Ecology . 303
The Ecosystem . 304
The Human System . 305
Hunting and Change . 308

Biodiversity Conservation . 313

Socioeconomic Development . .. 316

Integrated Conservation and Development 320

Recommendations for Conservation and
Development, with Particular Reference
to the Dzanga-Sangha Project .. .326
Monitoring . . 327
Snare Hunting . 329
Net Hunting . . 332

Recommendations for Further Research 334












. 337

. 343

. 344

. 350

. 354

. 355

. 359

. 360

. 363

. 416


Table page

3.1 Game Species Activity, Weight, and Diet 66

3.2 Population Densities for Game Species at
Research Sites in Gabon, Zaire and
Cameroon . . 70

3.3 Age Structure for C. Monticola ... .71

3.4 Estimated Population Densities from Line
Transects . ... .77

3.5 Estimated Population Densities from Net Hunt
Encounters . . 79

3.6 Comparative Density Estimates: Line Transects
versus Net Hunt Encounters in a
Single Area . 88

3.7 Population Densities of Game Species in the
Bayanga Region and Elsewhere 89

3.8 Estimated Population Figures for Settlements
in the Bayanga Region ... .108

3.9 Bayanga Population Origins .. 109

3.10 Men's Employment in Bayanga .. 112

3.11 Mossapoula Censuses . .. 115

3.12 Impacts of Hunting on C. monticola:
Age Classes in Hunted versus Unhunted
Populations .. . 126

3.13 Sustainable Harvest Rates . .. 129

3.14 Head Tax Levied on Game Species .. 137

3.15 Observed Bushmeat Sales by Species 142

4.1 Observed Cable Snare Captures ... .162

Table pase

4.2 Reported Cable Snare Captures .. 165

4.3 Capture Rates for Fence Snares .. 166

4.4 Capture Rates for Foot and Neck Snares 166

4.5 Age Structure of Animals Captured in Snares .167

4.6 Sex Ratios of Animals Captured in Snares 168

4.7 Wastage from Cable Snares . .. 169

4.8 Injured Escapes from Cable Snares .. .171

4.9 Snare Injuries on Captured Animals ... .173

4.10 Annual Biomass Harvest by Snares and Guns 177

4.11 Bayanga Annual Snare Harvest Rates 178

4.12 Bayanga Annual Snare and Gun Harvest Rates 180

4.13 Sustainability Index by Species and
Hunting Method . .. 180

4.14 Bushmeat Market Prices in Bayanga .. 188

4.15 Weekly Returns from Snare Hunting .. 189

4.16 Annual Bayanga Bushmeat Production ... .191

4.17 Comparative Bushmeat Consumption .. .194

4.18 Comparative Snare Captures .. 195

4.19 Comparative Annual Snare Hunting Returns 196

4.20 Comparative Snare Hunting Yields 197

5.1 Descriptive Statistics for Mossapoula
Net Hunts . 207

5.2 Reasons Cited for not Net Hunting .. 218

5.3 Net Hunting Pressure within Mossapoula
Hunting Range . .. 225

5.4 Net Hunt Captures . .. 227

5.5 Net Hunt Captures by Method .. 228
























Net Hunt Captures by Women versus Men .

Descriptive Statistics for Forest Camp (N=8)
versus Day Hunts (N=73) .

Net Hunt Yields from Forest Camp Hunts (N=8)
versus Day Hunts (N=73) .

Comparative Net Hunt Statistics .

Comparative Net Hunt Captures .

Hunting Yields and Distance from Town .

S 264

S 270



S 274

S 275



Net Hunt Capture Rate .. ... 229

Age Structure of Net Hunt Captures .. .230

Sex Ratios of Net Hunt Captures .. 231

Vulnerability to Net Hunting by Species 232

Mossapoula Annual Net Hunt Harvest Rates 236

Sustainability Index by Species for
Net Hunt Harvests . .. 237

Annual Biomass Harvest by Mossapoula
Net Hunters . 238

Effects on Net Hunting Yields of
Releasing Young, Female, or
Pregnant/Lactating Animals .. 240

Mossapoula Individual Activity
Distribution . .. 244

Mossapoula Community Activity
Distribution . 245

Activity Distribution by Age Group 249

Net Hunt Returns . .. 255

Mossapoula Weekly Net Hunt Returns 257

Annual Net Hunting Returns .. 258

Annual Mossapoula Bushmeat Production 261












Species Captured by Snares versus Nets

Animals Captured by Snares versus Nets

Sex Ratios of Animals Captured by Snares
versus Nets . .

Harvest Rates by Snares and Nets .

Sustainability Index by Species and
Hunting Method . .

Age Structure of Snare and Net Captures

Weekly Returns per Hunter: Snares
versus Nets . .

Returns per Person/Day Hunting: Snares
versus Nets . .

Variation in Bushmeat Prices .

Annual Bushmeat Production by Snares and
Nets, and Bayanga Beef Sales ..


S 288

S. 289

. 290

. 293

. 293



S. 299

S. 300

S. 301


Figure page

2.1 Theory in Geography and Anthropology .. 10

2.2 A Systems Model of Human Ecology 18

3.1 The Central African Republic and the
Research Area . 57

3.2 Average Temperatures (oC): Bayanga,
1974-1984 . ... .59

3.3 Average Rainfall (mm): Bayanga, 1973-1984 .60

3.4 Vegetation of the Bayanga Region 61

3.5 Location of Line Transect and Net Hunt
Surveys . ... .75

3.6 Population Densities of Game Species: C.
monticola, A. africanus, C. dorsalis
and C. callipygus .. 83

3.7 C. monticola Population Densities from
Net Hunt Encounters . ... 84

3.8 A. africanus Population Densities from
Net Hunt Encounters .. 85

3.9 C. callipygus Population Densities from
Net Hunt Encounters . 86

3.10 C. dorsalis Population Densities from Net Hunt
Encounters . 87

3.11 Logging in the Bayanga Region . 98

3.12 Dzanga-Ndoki National Park and Dzanga-Sangha
Special Reserve . .. 101

3.13 Population Centers in the Bayanga Region .106























Bayanga Residential and Occupational
Mobility . .

Mossapoula Censuses: September, 1993 and
November, 1994 . .

Mossapoula Residential and Occupational
Mobility . .

Bayanga Snare Hunting Camps and
Snare Lines . .

Bayanga Snare Hunting Range .

Relative Distribution of Snare Captures and
Abundance for Primary Game Species .

Informant Occupation Time-Lines .

Net Hunt Leaders . .

Mossapoula Net Hunting Range .

Mossapoula Hunting Range Decline .

Species Vulnerability to Net Hunters .

Community Participation in Net Hunting

Individual Participation in Net Hunting

Time Allocation to Net Hunting .

Net Hunting Participation Rates by
Age Group . .

Time Allocation to Net Hunting by
Age Group . .

Annual Returns from Net Hunting
(kilograms) . .

Annual Returns from Net Hunting
(CFA francs) . .

Time Spent in Forest Camps .

Location of Mossapoula Forest Camps .

. 110


S 117

S 150

. 159

S 163

. 185

. 213

S 222

S 223

S 233

S 246


. 248

. 250

. 251

S 259

S 260

S 267

S 268


BSP Biodiversity Support Program
CAR Central African Republic
CFA Communaut6 Financiere Africaine
COICA Coordinadora de las Organizaciones Indigenas de la
Cuenca Amazonica
EIL Experiment in International Living
ICDP Integrated Conservation and Development Project
IUCN International Union for the Conservation of Nature
USAID United States Agency for International Development
WCED World Commission on Environment and Development
WCED World Commmission on Environment and Development
WCS World Conservation Strategy
WRI World Resources Institute
WWF World Wildlife Fund and World Wide Fund for Nature


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Andrew J. Noss

December 1995

Chairman: Dr. Edward J. Malecki
Major Department: Geography

This dissertation analyzes the relationships between

ecological and human systems, and between biodiversity

conservation and socioeconomic development, by focusing on

the role of hunting. The analysis addresses two research

questions. First, how does hunting affect wildlife

populations? Are harvests sustainable? Second, what is the

economic importance of hunting, relative to other economic

activities, for human populations?

The two primary hunting methods in the rainforests of

the south-western Central African Republic are cable snare

hunting by immigrant ethnic groups, and communal net hunting

by BaAka Pygmies. Data collection is based on participant

observation, activity records, interviews, market surveys,

and biological measurements of captured animals.


The primary game species for both hunting methods are

the brush-tailed porcupine Atherurus africanus and the

duikers Cephalophus monticola, C. callipygus, and C.

dorsalis. Net hunters encounter and capture few other

species. But virtually all mammalian species, as well as

several birds and reptiles, are captured by snares.

Net hunters probably overexploit C. monticola and C.

dorsalis, whereas snare hunters probably overexploit C.

callipygus. Both hunting methods are practiced in most

areas, in addition to gun hunting. Therefore current

hunting by all methods combined is probably excessive for

all game species. In addition to animals killed and

recovered by hunters, snares lose many animals to

decomposition and scavengers, and many others escape with


Although both hunting methods produce meat for

subsistence and for sale in local markets, snare hunting is

primarily market-oriented, while net hunting is primarily

subsistence-oriented. Both groups of hunters pursue mixed

subsistence and economic strategies, combining hunting with

other activities.

Human populations are increasing through population

growth and immigration, while economic expectations are

rising. At the same time, hunting ranges are declining as

other land use systems expand. Wildlife is an open access

resource belonging to the state. Local hunters do not


conserve or manage wildlife resources: they believe that

wildlife cannot be exterminated by snares or nets. Even if

wildlife is exterminated they will pursue other activities

or move elsewhere. Under these conditions, conservation and

development are incompatible.


This dissertation analyzes the relationships between

ecological and human systems by focusing on an activity

which links the two systems, namely wildlife exploitation.

Cultural ecology provides the theoretical framework for the

analysis. Accordingly, two parallel topics are defined.

The first topic relates to the ecosystem, in particular the

theory and practice of biodiversity conservation. Of

primary interest from a conservation perspective are the

impacts of hunting on wildlife and the sustainability of

wildlife harvesting. The second topic relates to the human

system, and the theory and practice of socioeconomic

development. Of primary interest from a development

perspective is the economic importance of wildlife

exploitation, as well as the relations between hunting and

other subsistence or economic activities. Each chapter

addresses both themes and explores the links between the


The research context is the Bayanga rainforest region

of the south-western Central African Republic (CAR). The

local population consists of several ethnic groups,

including both long-term local residents and recent


immigrants. Current economic activities in the region are a

continuation of historical resource exploitation and land

use patterns: logging, diamond mining, and coffee

plantations. In addition, people engage in cultivation,

fishing and hunting for economic and subsistence purposes.

The most recent regional development is the creation of a

national park to conserve a portion of CAR's remaining

forests and wildlife.

Research depends primarily on participant observation

of legal and illegal hunting activities. Additional data

are collected through activity records, interviews, market

surveys, and biological measurements of captured animals.

These methodologies generate two types of data: first,

ecological data regarding the wildlife resource base and the

impacts of exploitation on wildlife populations; and second,

economic data on the role of wildlife exploitation in the

local economy.

Chapter 2 locates the research in the context of

geographical and anthropological theory, specifically in

cultural ecology. In addition, the chapter relates the

research to the literature on biodiversity conservation and

socioeconomic development, and on efforts to integrate the

two. The section concludes by examining the relations

between hunting, on one hand, and both conservation and

development on the other.


Chapter 3 describes the ecosystem and the human system

in the research area. The ecosystem, like other tropical

rainforests, is complex. This section describes in

particular the resource base represented by the four primary

game species: the brush-tailed porcupine Atherurus

africanus, and the duikers Cephalophus callipygus, C.

dorsalis, and C. monticola. Of particular importance are

estimated population densities of these game species in the

Bayanga region, and predicted sustainable harvesting rates.

Components of the human system include political

authorities, economic activities, and population centers.

Change in each of these components over time in turn affects

patterns of wildlife exploitation. In order to illustrate

the influence of hunting on relations between the ecosystem

and the human system in the Bayanga region, the chapter

concludes by discussing the following cases and issues: the

ecological impacts of human activities, elephants, hunting

legislation, bushmeat markets, and currency devaluation.

Chapters 4 and 5 analyze the two primary local hunting

methods--cable snare hunting and communal net hunting

respectively--practiced by two distinct groups of hunters.

In each chapter, a description of the research methodology

and the hunting method precedes a discussion of the primary

research questions:

1) What are the impacts of the hunting method on
wildlife in terms of species distribution, sex ratios,


and age classes captured or injured? Is current
offtake by the method sustainable?

2) What is the economic importance of the hunting
method in terms of employment, time allocation, and
returns? What is the relationship between hunting and
other subsistence and economic activities?

Chapter 4 examines cable snare hunting by Bayanga

residents of several ethnic groups, based on participant

observation of snare hunting trips from September 1993 to

December 1994. Most snare hunting requires overnight trips

to check snare lines. Cable snares are illegal everywhere

in CAR because they are indiscriminate in capturing and

injuring virtually all mammals, as well as several species

of birds and reptiles. But many people depend on snares to

obtain food, as well as income from the sale of meat in

local markets. Cable for snares is much easier and cheaper

to obtain than the alternative--firearms. Snare hunting can

be a full-time occupation for weeks or months, though

Bayanga residents prefer formal employment.

Chapter 5 examines communal net hunting by the BaAka of

Mossapoula, based on participant observation of net hunts

from September 1993 to November 1994. Most net hunts are

day trips from Mossapoula, while in other cases groups hunt

from temporary camps established in the forest. Net hunting

is legal outside protected areas, and targets the most

abundant small/medium-sized terrestrial mammals--the

porcupine and duikers mentioned above. Net hunting also

provides food and income, but is one of several subsistence


and economic activities among which Mossapoula residents

alternate on a daily basis.

Chapter 6 compares the two hunting methods, in terms of

their implications for the ecosystem and the human system.

Chapter 7 concludes by reviewing the role of hunting within

the cultural ecology framework. The chapter also addresses

the implications of hunting for biodiversity conservation

and socioeconomic development across tropical forests, and

proposes recommendations for further research.

The following comments are general explanatory notes on

the text and the fieldwork.

BaAka: I use the general term "Aka" which unites the

Bayanga area Aka with those in the Lobaye area of CAR and

those in northern Congo, all of whom share a single

language. The Aka have been extensively studied by Serge

Bahuchet and others. Other terms in the literature for the

same people are "Biaka," "Babinga," and "BaMbenga." Louis

Sarno's (1993) more local term "BaBenzele" refers to the

Bayanga region BaAka. The BaAka are linguistically and

culturally distinct from the Baka of Cameroon and CAR west

of the Sangha river. When discussing the residents of

Mossapoula, I refer only to the BaAka residents.

Pygmy: The BaAka are one of several African forest

peoples, traditionally hunter-gatherers, known as "Pygmies."

The term refers to their small stature, and dates back to

Aristotle, Herodotus, Homer and Pliny (Demesse, 1978). When


referring to particular ethnic groups, I use the name the

group calls itself--Aka, Efe, Mbuti, etc. I have found no

alternative that refers to these groups collectively.

Bilo: I use the BaAka term "Bilo" to refer

collectively to all non-BaAka ethnic groups in the Bayanga

region (Kisliuk, 1991; Kretsinger, 1993). Other terms used

in the literature to refer to Pygmies' neighbors are not

accurate in the Bayanga context. The term "Bantu" is

incorrect, as several ethnic groups such as the Gbaya are

not Bantu-speaking, while Aka is a Bantu language (Bahuchet,

1985c; Bahuchet and Thomas, 1986; Greenberg, 1966). The

terms "villagers" and "farmers" no longer provide meaningful

distinctions where the BaAka too live in villages and

cultivate farms. When discussing Bayanga residents, I refer

only to Bilo residents.

Orthography: In the case of river and town names, I

use the French orthography for consistency with other maps

produced for CAR. In the case of personal names and words

in African languages, I use basic phonetic orthography.

Languages: I spoke Gbaya and fluent French before

beginning my research, and learned Sango during my

fieldwork. I did not learn the BaAka language, and this

constrained my data collection with those BaAka individuals

who do not speak Sango (in particular older women), although

my assistants served as interpreters. Because I worked with


Bilo as well as BaAka informants, Sango was the more useful

language to learn.

Assistants: Throughout my fieldwork I employed two

assistants: Ngbongo Yves from Bayanga, and Matofi Fernand

from Mossapoula. In addition, I hired two other Bayanga

residents to conduct surveys: Ndombe Blaise for a

neighborhood bushmeat market survey, and Bopokina Elbert for

a public market survey.

Informant compensation: The issue of compensation was

problematic throughout my fieldwork. In particular, I did

not want to provide material or financial rewards that would

encourage individuals to hunt more than they might

otherwise. Therefore, I did not provide daily rewards, but

several times during the course of my fieldwork purchased

clothes and shoes for informants to thank them for their

assistance. With respect to Bayanga snare hunting, my

informants were the 16 individual hunters that I

accompanied. With respect to Mossapoula net hunting, my

informants were all 246 adult and adolescent BaAka

residents, including those who rarely or never net hunted.

"The project": Throughout the dissertation I use this

term to refer to the integrated conservation and development

project described in Chapter 3. The project is administered

by the World Wildlife Fund, is headquartered in Bayanga, and

manages the Dzanga-Sangha Special Reserve and Dzanga-Ndoki

National Park.


"The logging company": I use this term to refer to the

sawmill located in Bayanga. Since 1972 three different

logging companies--Slovenia Bois, Sangha Bois, and Sylvicole

de Bayanga--have used this sawmill to exploit timber in the

Bayanga region (see Chapter 3).

CFA francs: The currency of the Central African

Republic is the Communaut6 Financiere Africaine (CFA) franc,

pegged at an exchange rate of 1 French franc = 50 CFA

francs, or $US1 290 CFA francs, until January 1994. At

that time the CFA franc was devalued to a new pegged

exchange rate of 1 French franc = 100 CFA francs, or $US1

580 CFA francs. See Chapter 3 for a discussion of this

currency devaluation.


This chapter begins by examining the theoretical

context for the research within the fields of geography and

anthropology, and particularly in the field of cultural

ecology. The chapter in turn relates the research to the

literature on biodiversity conservation, socioeconomic

development, and efforts to integrate the two. The

concluding section examines the relations between hunting on

one hand, and both conservation and development on the

other, in case studies across the tropics.

Cultural Ecology

This section reviews the development of the field of

cultural ecology, particularly through the work of Carl

Sauer and Julian Steward. Cultural ecology is further

described by examining other research in the field.

Cultural ecology integrates geography's people/

environment tradition with anthropology and ecology (Figure

2.1). The resulting theoretical framework considers the

systemic interdependency between society and nature. People

affect the environment and natural ecosystems by

appropriating natural resources and integrating them into


production processes--whether at the household, community,

or global levels. The natural environment in turn affects

social structures and culture as well as subsistence and

economic production systems.

Man/land or people/environment tradition Ecology

Systems theory


Cultural ecology
--cultural materialism <
--political ecology

Ecological anthropology

Cultural anthropology

Figure 2.1: Theory in Geography and Anthropology

Note: The following components of this figure are described
in greater detail below: human ecology, systems theory,
ecological anthropology, cultural materialism, and political

The Origins of Cultural Ecology

The fathers of cultural ecology are the geographer Carl

Sauer and the anthropologist Julian Steward.

Carl Sauer

Carl Sauer introduced the concept of culture to

geography in the United States, by defining human geography

as cultural history (Leighly, 1976). He saw landscape as a

cultural text (Cosgrove and Jackson, 1987): cultureue is


the agent, the natural area is the medium, and the cultural

landscape the result" (Sauer, 1925, p. 46). Thus, humans

transform natural landscapes into cultural landscapes

through the action of cultural forms such as population and

settlement patterns, land utilization systems, subsistence

systems, and communication systems (Parker, 1981; Sauer,

1941a; 1944; 1962a; 1963a; 1963b).

Sauer set the stage for the development of cultural

ecology. Most importantly, he described landscape as an

association of physical and cultural forms; a changing,

living scene where human activities unfold. He described

relationships between resources in the environment and

cultural forms. He emphasized the impacts of humans on the

environment, and the influences of the environment on human

activities (Sauer, 1925). For example, humans alter

habitats through settlement and agriculture, and they affect

particular species through domestication, dispersal, and

extinction (Sauer, 1941a; 1963b). At the same time, humans

adapt to the physical environment by organizing their lives

around the natural resources that their skills and values

identify (Gade, 1988; Sauer, 1941b; 1962a, 1962b).

Julian Steward

The anthropologist Julian Steward founded the field of

cultural ecology by integrating natural systems and cultural

systems into a single analytical framework. The framework's

three essential components are resources, technology, and


labor. Resources are determined by culture: only human

knowledge and technology make resources accessible (Murphy,

1977). Technology is a cultural trait, while the

application of labor is determined by cultural elements

including economic arrangements, social organization, and

demography (Orlove, 1980).

Based on these components, cultural ecology addresses

four principal topics: 1) the relations between the

environment and exploitation or production technologies; 2)

the behavior patterns resulting from appropriation and

production systems in a particular area using particular

technologies; 3) the effects of these behavior patterns on

other aspects of culture; and 4) the impacts on culture and

the physical environment of relations with other ethnic

groups and external institutions (Steward, 1955; 1968).

Cultures and social structures thus adapt not only to the

physical environment, but also to the social environment.

Steward defined a "culture core" as the central causal

mechanism influencing other aspects of culture. This

culture core included the cultural features most closely

connected to exploiting the environment, namely those

associated with subsistence systems and economic

arrangements (Moran, 1982). The environment only influences

the culture core, while more peripheral aspects of culture

and social organization including political and religious


patterns are subject to autonomous processes of cultural

history (Murphy and Steward, 1956; Steward, 1950; 1955).

Cultural Ecology in Theory

Cultural ecology developed in opposition to

environmental determinism, which emphasizes the empirical

physical environment as the primary factor that determines

human behavior and culture in general (Johnston, 1986;

Nugent, 1981). Instead, cultural ecology favors Franz Boas'

possibilism, in which the environment defines a set of

opportunities and limits among which people make choices to

attain goals (Brookfield, 1969; Moran, 1982). Cultural

ecology follows geography's people/environment or man/land

tradition, within the broader field of human ecology

(Ackerman, 1963; Kates et al., 1990; Stoddart, 1987).

Human ecology

Human ecology developed in the early 1900s among

sociologists who borrowed the ecosystem concept from

biologists to describe the social system as a human

ecosystem in which humans organize to maintain themselves in

a given environment. An ecosystem is a set of organisms

within a physical environment that has spatial dimensions.

Ecology is the study of the biological, physical, and human

components in an ecosystem, and of their interactions

through energy flows and nutrient cycles (Kormody, 1974;

Odum, 1975).


Human ecology is the process of forming viable

relations between population and environment. The emphasis

is on individuals and populations as biological entities

profoundly modified by culture. These entities affect and

are affected by the environment and each other. The

environment includes natural and human-modified biological

and physical elements as well as the cultures of peoples in

adjacent areas and beyond: the components impinge on one

another (Hawley, 1986). In the form of institutional

systems such as market and administrative forces, production

systems, and pressure groups, culture mediates between

humans and the biophysical environment (Barrows, 1923;

Carlstein, 1982; Smith and Reeves, 1989). Thus settlement

patterns and subsistence strategies reflect social behavior,

technology, resource opportunities and limitations, as well

as external and internal stimuli (Butzer, 1982; Helm, 1962;

Morrill, 1987; Porter, 1978). Some theorists describe human

ecology simply as human behavior: culture is therefore the

ecological equivalent of animal behavior (Bennett, 1990;

Rappaport, 1984; Vayda and Rappaport, 1968).

Systems theory

Systems theory, like ecology, provides a method for the

holistic analysis of human situations, rather than a theory

that permits rigorous problem selection (Brookfield, 1982;

Ellen, 1982). Generally speaking, a system is a framework

for describing and analyzing a complex array of unlike


variables that interact through connectivity, mutual

causality, and feedback (Butzer, 1990; Morren, 1986).

Systems analysis emphasizes the functions and interchanges

of component parts, through flows of energy, materials, and

information that can be quantified (Micklin, 1984; Odum,

1983; Rambo, 1982a; 1983; Sponsel, 1989). All interactions

are interdependent (Butzer, 1989). Systems theory enhances

the ecological framework of cultural ecology analysis

because it integrates environmental and human variables

(Stoddart, 1965).

However, it may be impossible to quantify certain

factors, particularly external factors that are significant

in an open system (Butzer, 1990; Ellen, 1982). For example,

socio-political and ideological factors must be included in

the analysis because the environment and technology are not

sufficient to explain contemporary ecosystem realities

(Parker, 1981; Orlove, 1980).

Ecological anthropology

Ecological anthropology and cultural ecology both

emphasize culture as a means of adapting to the environment

(Headland, 1986; Jochim, 1981; Parker, 1981), and both

examine the interactions of human populations with other

components of the local ecosystem and with external

influences (Helm, 1962; Moran, 1981a). Systems of

production are important links between population dynamics,


social organization, culture, and the environment (Orlove,


However, ecological anthropologists emphasize the

evolutionary nature of human adaptive strategies, including

sociocultural as well as physiological responses to problems

and stresses imposed by the local environment (Morren, 1986;

Orlove, 1980; Parker, 1981). Alternatives can be compared,

and overall success in human adaptability can therefore be

measured according to demographic, energetic, and

nutritional criteria (Moran, 1982).

Adaptive strategies include demographic shifts through

migration and changes in settlement patterns, socioeconomic

relations with other groups through trade links, and culture

traits like technology (Gross et al., 1979; Moran, 1982;

Morren, 1986; Tanaka, 1980). In addition, current adaptive

strategies are strongly influenced by regional level factors

including the goals of national and international actors

regarding production, conservation, and development at the

regional scale (Moran, 1981a).

Cultural ecology

Like human ecology, cultural ecology is defined

generally as the systemic relations between society or

social systems on the one hand, and nature or ecosystems on

the other (Butzer, 1989; Grossman, 1977; Moran, 1981a;

Turner, 1989). More specifically, cultural ecology refers

to the adaptation of human society to the local environment


through the flexible agent of culture (Bahuchet, 1985b;

Baksh, 1984; Butzer, 1990; Clarkson, 1968; Knight, 1992).

The local environment includes not only natural resources,

but also other human groups that influence adaptation

(Barth, 1969; Campbell, 1983; Netting, 1968; Sponsel, 1986).

Carl Sauer examined the results of human-environment

interactions visible in landscapes (Parker, 1981). Cultural

ecologists more recently have focused on the processes of

change. They argue that culture does not simply adapt to

environmental constraints or changing environmental

conditions: feedback processes between culture, technology

and nature alter the natural environment as well as the

human environment (Bargatsky, 1984; Bennett, 1976; Headland,

1986; Nowicki, 1985). People alter the habitat to make it a

more suitable place to live, while also making themselves

more suitable to live in that habitat (Denevan and Schwerin,

1978). Cultural ecology therefore describes the human use

of resources, and the human appropriation of nature

(Bargatsky, 1984).

Steward's cultural ecology has been criticized for

treating culture and environment as separate and opposed

entities, lacking the concept of dynamic mutually

interacting processes (Moran, 1990). In the 1960s cultural

ecologists refocused the field on the ecosystem as a unit of

analysis (Ellen, 1988). Nature and culture came to be seen

as two interlocking subsystems of a larger system (Figure


2.2) in which relations among components are circular

(Bennett and Chorley, 1978; Mikesell, 1978; Rambo, 1982).

Within the larger system, cultural ecology emphasizes the

predominant role of culture in mediating change and

governing feedback relationships (Sponsel, 1980).

Nevertheless, the ecosystem interacts with every component

of the culture system, and determines a set of freedoms and

constraints within which individuals exercise choice

(Parker, 1981; Sponsel, 1986).

Figure 2.2: A Systems Model of Human Ecology

Source: Rambo, 1983, p. 26. Reprinted with permission of
the author, A.T. Rambo.


The ecosystem concept has been criticized for

neglecting the critical role of external factors such as

larger economic and political systems, as well as the

historical background of contemporary environmental

circumstances. Defining ecosystem boundaries is also

problematic. Furthermore, ecosystem analyses tend to

emphasize long-term homeostasis and ignore changes taking

place in local ecosystems. Finally, they overlook diversity

and differentiation within local groups or populations, and

the options people have in adjusting to change (Lees and

Bates, 1990; Moran, 1990; Orlove, 1980).

Cultural Ecology in Practice

Cultural ecology has addressed a variety of topics in a

wide range of human societies. Given the critiques

described above, cultural ecologists have used a variety of

methods that identify particular subsystems or problems for

analysis, or that emphasize particular components of the

human/environment system.

Cultural materialism. Cultural materialism gives

primacy to the technological and economic components of the

cultural system, as they relate to modes of production and

reproduction (Abruzzi, 1980; Micklin, 1984; Sponsel, 1983).

These elements of social organization and culture are

practical adaptations permitting a particular population to

exploit its environment successfully (Harris, 1966; 1974;


1977). Like ecological anthropology, this field focuses on

population as a unit, and on biological concepts such as

carrying capacity, nutrient flows and energy cycling

(Orlove, 1980; Rappaport, 1984).

Political ecology. Political ecology emphasizes the

influence of political and social ideologies as well as

economic processes on the interactions of society and the

environment through the organization of labor, space, and

resources in the system of production. Exogenous political

and economic forces, as well as the internal structure of

society governing access to resources and modes of

production, condition local ecosystem-human system

interactions (Bassett, 1986; 1988; Bell and Roberts, 1991;

Hecht and Cockburn, 1989; Hyndman, 1994; Knight, 1992;

Pickles and Watts, 1992; Schmink and Wood, 1987; Watts,


Subsistence economies

Because of the practical difficulties involved in

analyzing all the components of ecosystem-human system

relationships, cultural ecologists have focused on

relatively small, isolated, and homogeneous systems in which

cultural institutions tend to be dominated by subsistence

activities (Fosberg, 1976; Headland, 1986; Mikesell, 1978;

Porter, 1978). Although relations with the physical

environment are particularly direct, impacts on the

environment are minimal (Bennett, 1976; Rambo, 1983).


Subsistence systems represent a means of regulating

energy flows between the human population and the ecosystem

(Nietschmann, 1984). Research in cultural ecology describes

how these societies exploit and alter their environment

through hunting, gathering, fishing, and agriculture

(Bahuchet, 1972; 1978; 1979a; 1979b; 1985b; Dornstreich,

1977; Marks, 1976; Nietschmann, 1971; 1973; Smole, 1962;

Vickers, 1979). Cultural ecology further examines cultural

adaptation to environmental constraints, for example to

argue that the lack of animal protein causes people to

establish complex food prohibitions to protect certain game

animals (Ross, 1978). Shifting agriculture also represents

an adaptation to environmental constraints such as poor soil

fertility, and abundant weeds and pests, as well as to

socioeconomic constraints such as scarce labor and

inefficient technology (Denevan, 1978). Finally, ecological

change can induce social change: the depletion of forest

resources results in more intensive exploitation, and the

dispersal of human settlements (Baksh, 1984; 1985).

Many subsistence societies in tropical forests adapt to

their environment by maintaining trading relationships with

neighboring societies. In its simplest manifestation, one

group exploits forest resources such as wild animals, while

the other cultivates food crops, and the meat-for-starch

exchange provides both groups with both products (Headland

1986; 1988; J. Peterson, 1977; 1978a; 1981; Terashima,


1986). Thus, dynamic relations with neighboring groups are

key factors influencing settlement patterns, mobility, and

even hunting techniques (Bahuchet, 1978; Endicott, 1979;

1984; Harako, 1981; Hoffman, 1988; Rambo, 1979a; 1982b).

Recent studies argue that the choice of foraging strategy is

an outcome of social and economic pressure by neighboring

groups (Eder, 1988b; Sellato, 1994).

Developing economies

In the case of more complex human-environment systems

with links to external political and economic forces,

researchers have focused on particular subsystems, in

particular agricultural subsystems (Brookfield, 1969;

Clarkson, 1968; Netting, 1968; 1986; Rambo, 1982a). The

systemic relationships incorporated in the analysis have

expanded to include not only the natural environment and

subsistence production, but also commodity production, the

social relations of production, cultural values, and

political and economic forces (Grossman, 1984). Others

emphasize the "human ecosystem" comprising four components

linked by systems of production: population dynamics,

social organization, culture, and environment (Headland,

1986). Finally, cultural ecologists have studied cultural

methods and institutions for regulating and managing natural

resources, such as fisheries, wildlife, forests, etc.

(Nietschmann, 1973; 1984).


Alternatively, one may begin with a particular human-

environment problem, and contextualize relations within

progressively wider or denser contexts (Vayda, 1983).

Similarly, Lees and Bates (1990) advocate beginning with a

specific environmental event and analyzing environmental

change and human behavior in response to this event. These

methods apply cultural ecology by examining the effects of

economic and social organization on the ecosystem, and the

impact of environmental change on socioeconomic systems.

But they do not depend on predetermined units of analysis or

specific system boundaries.

Cultural ecology themes

As described above, cultural ecologists analyze the

food base, activity patterns, group composition, resource

availability, relations with other groups, and change over

time in all these variables. Food production is considered

to be the primary human/environment nexus, and cultural

ecologists emphasize the coevolution of subsistence

strategies and certain features of the environment

(Bargatsky, 1984; Ellen, 1988; Knight, 1992; Mikesell,

1978). They identify environmental constraints and analyze

patterns of social adaptation to cope with change in

ecological and social systems (Butzer, 1990; Durham, 1976).

For example, in the context of Brazil's Transamazon

settlement scheme, which has significantly altered the

tropical forest, colonists must adapt to both ecological


constraints such as low soil productivity as well as

socioeconomic constraints such as limited access to

transportation and markets (Denevan, 1973; 1989; Smith,


In addition to local environmental and socioeconomic

constraints, institutions at the regional, national, and

international levels determine the nature of interactions

between people and the environment (Marks, 1989; Moran,

1979; 1993; Riddell, 1988). Thus the growth of extractive

industries and commercialization in Brazil has in many

places replaced Indian culture with caboclo culture, as

people opportunistically respond to new resources and new

environmental and socioeconomic constraints and

opportunities. They adapt through technological,

institutional and organizational change (Dove, 1993b; 1994;

Hardesty, 1986; Kates, 1987; Parker, 1981).

New economic systems thus offer new opportunities,

subject to new constraints such as limited land availability

and rising population density. Subsistence societies may

exploit new resources, adopt cash crops, and engage in non-

agricultural activities (Denevan and Schwerin, 1978;

Grossman, 1981; 1984; Nietschmann, 1972; 1973). Ecological

systems facilitate as well as limit human action, while

resource exploitation using particular technologies imposes

short-term and long-term effects on social and natural

systems (Smith, 1976a).


Cultural ecologists are also interested in traditional

resource use and management institutions as potential models

for development schemes in the Third World (Bennett and

Dahlberg, 1990; Butzer, 1989; Eden, 1978; Marks, 1984;

Nowicki, 1985; Sponsel, 1985; 1986). Other general topics

include the following: environmental and regional planning,

environmental constraints to development, the relations

between environmental potentials and economic opportunities,

the consequences of shifting from a subsistence to a market

economy, sedentarization, intensification of production,

land use and tenure, and tourism studies (Bennett, 1976;

Ellen, 1988; Hardesty, 1986; Knight, 1992; Mabogunje, 1984;

Mikesell, 1978; Murphy and Steward, 1956; Porter, 1978;

Rambo, 1982a). The following chapters address many of these

topics through a case study of wildlife exploitation,

linking hunting in the Bayanga region to wider ecosystem and

human system causes and effects.

Conservation and Development

The relationship between hunting and the ecosystem is

particularly important with respect to biodiversity

conservation: hunting directly reduces wildlife

populations, and excessive hunting may cause their

extinction. At the same time, the relationship between

hunting and the human system is particularly important with

respect to socioeconomic development: hunting provides

subsistence and economic resources to local populations.

Hunting therefore connects the two systems and reveals their

interdependencies. In order to sustain both systems,

biodiversity conservation and socioeconomic development must

be integrated, for example through the sustainable

exploitation of wildlife, and sustainable development for

the human population.

Biodiversity Conservation

The concept "biodiversity conservation" has many

meanings. Biodiversity refers to biological variability and

abundance at several scales: genes, species, communities or

ecosystems (McNeely, 1988; Reid and Miller, 1989; WRI et

al., 1992). Biodiversity conservation may therefore seek to

preserve particular ecosystem components (genes, species,

populations) as well as the patterns and processes that

complete ecosystems represent (Ehrlich and Wilson, 1991;

Noss, 1990; Reid and Miller, 1989).

In its strictest sense, conservation seeks to preserve

natural ecosystems free of human intervention, manipulation,

or management. Thus protected areas prohibit traditional

development activities such as agriculture, resource

extraction, and human settlement. Strict preservationists

emphasize the moral and aesthetic values of protecting

nature for its own sake, and the ecological services that


protected areas provide to society (Callicott, 1990; Lund,

1980; Nash, 1967).

However, conservationists began seeking economic

justifications for conservation efforts for two reasons.

First, conservation imposes economic costs on local people

in the form of land and resource alienation, together with

crop destruction and threats to personal safety from

wildlife (McNeely and Miller, 1984). Second, conservation

imposes economic costs on the nation in the form of

management expenses and lost opportunities to use the land

in more economically productive ways. At the local and

national levels, conservation competes with other forms of

land use, and the alternatives are usually evaluated on the

basis of short-term economic costs and benefits (Dixon and

Sherman, 1990; O'Riordan, 1988; Pullan, 1983b; Western,

1984a; 1984b). The economic incentives to exploit protected

area resources are strong, while protected area managers

lack the resources and the political commitment to prevent

destructive human activities.

The extreme "use it or lose it" view argues that unless

wildlife is economically valuable, it will be replaced by

more productive land use systems (Freese, 1994; Kock, 1995;

Luxmoore, 1989). Many conservationists are therefore

seeking viable methods for the sustainable use of wildlife,

methods that derive economic benefits without exterminating

species or degrading ecosystems.

Socioeconomic Development

"Development", like "conservation" is a term with many

meanings. Development implies the improvement or

fulfillment of potential (Daly, 1990b). Development can

have political or ethical implications, in terms of

achieving self-sufficiency and self-determination (Dasmann,

1984). But it is most often associated with improving

economic conditions according to criteria such as income,

efficiency, and productivity (Berger, 1992; Bodley, 1988;

Foresta, 1991). Thus the primary objective for many

development efforts has been to promote economic growth by

increasing the productivity of traditional economic and

subsistence activities, or by introducing new activities

(Burbridge, 1987). Development is achieved primarily

through technological and institutional innovation in food

production, resource extraction, industrial production, and

other economic sectors (Barbier, 1987).

Traditional development efforts therefore focus on

transforming natural ecosystems into more productive

artificial ecosystems managed by humans. In the case of

tropical forests the process has been extraction of

available resources through logging and mining, and

conversion to more productive land uses such as agriculture

or ranching. But these approaches have been severely

criticized from three perspectives:

1) Some bemoan economic growth without development for

the rural poor (Berger, 1992; Chambers, 1983; Independent

Commission, 1980; Lakshmanan, 1982; O'Riordan, 1989;

Redclift, 1984; 1987; Schumacher, 1973).

2) Others argue that generating short-term economic

benefits imposes severe long-term economic costs by

degrading the environment: tropical forests are

particularly fragile ecosystems (Denevan, 1989; Eden, 1978;

Nicholaides et al., 1985; N. Smith, 1976a; 1981).

3) Finally, proponents of "limits to growth" theories

argue that natural resources are limited, and the

environment's carrying capacity will soon be reached, given

current consumption patterns and population growth

(Brookfield, 1988; Daly, 1990a; 1990b; Fearnside, 1986;

Hardin, 1986).

"Sustainable development" was widely promoted in the

1980s as a means of addressing problems of poverty,

environmental degradation, and skewed resource distributions

that economic growth models have ignored (Barbier, 1987;

Fearnside, 1983; Lele, 1991; O'Riordan, 1989; Redclift,

1991; WCED, 1987). Most definitions of sustainable

development emphasize the maintenance of ecological and

economic viability through time. Development is

economically viable if economic benefits equal or exceed

economic costs. Development is ecologically viable if it

does not deplete biophysical resources or degrade ecosystems

(Brookfield, 1990; Goldman, 1995; Tisdell, 1988; Wilbanks,


Sustainable development promotes small-scale production

systems and appropriate technologies that minimize

environmental and sociocultural impacts (Browder, 1989;

Schumacher, 1973; Trainer, 1990). Geographers and

anthropologists in particular stress the importance of

adapting development activities to local economic, social,

and ecological conditions (Clay, 1988; Mabogunje, 1980;

Redclift, 1987; Richards, 1985; Watts, 1987; 1989). New

approaches to development therefore focus on the sustainable

use of natural resources and ecosystems, recognizing that

the conservation and management of natural resources is

essential to successful long-term socioeconomic development.

Integration of Conservation and Development

In turning to the sustainable exploitation of wildlife

and sustainable development respectively, development and

conservation practitioners have increasingly recognized the

mutual interdependence of their objectives (Lele, 1991).

Development depends on careful management of natural

resources, while conservation is not possible unless

development reduces pressure on natural resources.

Development can provide additional resources for

environmental protection, but environmental degradation

undermines development goals (McNeely, 1988; Whitaker, 1986;

World Bank, 1992).

Some researchers even argue that human cultural

diversity is a component of biodiversity, and that humans

are an integral part of nature (Bennett, 1975; Callicott,

1990). In particular, indigenous peoples are described as

"ecosystem people" who live within the ecosystem in a

sustainable manner (COICA, 1989; Colchester, 1992;

International Alliance, 1992; Klee, 1980; O'Connor, 1994;

Oldfield and Alcorn, 1991). Humans have helped to create

and maintain most tropical forest ecosystems; therefore to

exclude humans in the name of conservation will

fundamentally alter ecosystem processes and biodiversity

(Bahuchet, 1993; Bailey et al., 1990; Balee, 1989; Kormody,

1974; McNeely et al., 1990). Restricting people's access to

land and natural resources upon which they have historically

depended can also be disastrous for human societies.

Thus policymakers began to emphasize the role of parks

in regional and national development (MacKinnon et al.,

1986; McNeely and Miller, 1984; Pullan, 1983a), and the

potential for integrating production systems with the

environment (Brookfield, 1988). Wildlife, because of its

economic value, becomes a development resource (Abel and

Blaikie, 1986; CAR, 1992; Dalal-Clayton, 1988; Pullan,

1983b). But it will only remain a resource so long as

exploitation is not excessive.

In 1980, three of the world's largest conservation

agencies1 produced the World Conservation Strategy (WCS),

which defines three objectives for conservation: to

maintain essential ecological processes and life-support

systems; to maintain genetic diversity; and to ensure the

sustainable utilization of species and ecosystems which

support rural communities and major industries (IUCN, 1980).

The WCS recognizes that a major threat to conservation is

underdevelopment, and argues that conservation projects must

provide for economic growth to improve the economic status

and the quality of life for rural people (Erskine, 1985;

Halle, 1985). It also encourages development projects to

invent and apply methods that are ecologically sound, and

that conserve the living resources essential for human

survival and well-being (Allen, 1980; Talbot, 1984).

Efforts to integrate conservation and development have

adopted a variety of theoretical and practical models.

Conservationists have begun to promote the integration of

conservation efforts with development activities that

benefit local people and the national economy (Eltringham,

1984; Freese, 1986; Kundaeli, 1984). At the local level,

conservation projects recognize the legitimate rights of

rural people to utilize local natural resources (West and

Brechin, 1990). Furthermore, conservationists increasingly

1The International Union for the Conservation of Nature
(IUCN), in cooperation with the United Nations Environmental
Programme (UNEP) and the World Wide Fund for Nature (WWF).

seek the cooperation and support of local communities, as

well as their active participation in conservation

management (Barrow et al., 1993a; 1993b; Brown and Singer,

1990; Clay, 1985; Gibson and Marks, 1995; Kiss, 1990;

Western and Henry, 1979; Western and Wright, 1994). At the

national level, conservation projects promote the

sustainable exploitation of natural resources, through non-

consumptive (Caro, 1986; Lusigi, 1981; 1988; McIvor, 1989;

World Bank, 1990) as well as consumptive forms of wildlife

exploitation (Abel and Blaikie, 1986; Berry, 1986; Dalal-

Clayton, 1988; Lewis et al., 1990; Luxmoore, 1985; Metcalfe,


Integrated conservation and development projects

(ICDPs) combine core protection areas for tourism and

research with buffer zones or multiple use areas for

sustainable subsistence and economic activities (Wells et

al., 1992). The projects promote socioeconomic development

by providing health and education services, as well as by

introducing new economic opportunities and activities in

order to reduce local dependence on wildlife resources. At

the same time they promote biodiversity conservation by

protecting parks and restricting traditional economic

activities to ecologically sustainable levels and methods

(Brown and Singer, 1990; Du Toit, 1985; Ruitenbeek, 1992;

Sayer, 1987; Sharma, 1990; Western, 1984b; Zube and Busch,


In part ICDPs are a concession by international

conservation groups that police measures alone are not

sufficient to protect wildlife: in the long run local

poverty and population growth will overwhelm park

authorities' resources and commitment to conservation. At

the same time, ICDPs represent the hope that sustainable

exploitation of natural resources by local people is

possible, i.e., that careful management can generate short-

term benefits and long-term economic growth without

depleting the resource base. A proliferation of

publications argue that conservation and development are

compatible (BSP, 1993; Clad, 1984; Cleaver and Schreiber,

1994; IUCN, 1980; IUCN et al., 1991; Ledec and Goodland,

1988; McNeely and Miller, 1984; Oldfield and Alcorn, 1991;

WCED, 1987). However, evidence of success in the field is

scarce, and arguments to the contrary are numerous (Browder,

1990; 1992; Mares and Ojeda, 1984; Mordi, 1991; Redford,

1992; Redford and Sanderson, 1992; Robinson and Redford,

1994b; Struhsaker, 1990).

A comparative assessment of ICDPs across the tropics

finds that few provide viable alternatives to extensive

resource utilization practices, and they have difficulty

producing and distributing benefits from conservation. The

emphasis on local participation can lead ICDPs to respond

increasingly to short-term needs at the expense of long-term

issues. Although ICDPs encourage local participation and

empowerment, local people's needs and interests may include

"unacceptable" activities that overexploit or degrade the

environment (Brandon and Wells, 1992; Wells and Brandon,


Some critics argue that rural development components

are primarily a means to reduce tensions between parks and

local people, but do not attempt to provide sustainable

livelihood alternatives (Ghimere, 1991; Jeanrenaud, 1992).

According to others, ICDPs too often emphasize the promotion

of income generating activities and the provision of social

services, while neglecting resource management objectives

(Durbin and Ralambo, 1994). These problems can be addressed

in the implementation of specific projects.

However, more serious criticisms of ICDPs suggest

fundamental contradictions between conservation and

development objectives. Some development activities

directly undermine conservation objectives: roads improve

access for illegal woodcutters (Stocking and Perkin, 1992).

In addition, agricultural extension can make rural areas

more appealing, and may draw additional people near

protected areas, thus increasing the pressure on wildlife

resources (Oates, 1995; Southgate and Clark, 1993).

Immigration, wealth accumulation and population growth limit

the positive effects of development for local populations,

while even limited exploitation of wildlife resources

produces biotic impoverishment (Kremen et al., 1994).

Furthermore, in the long term the sustainable exploitation

of wildlife cannot support continuously rising human

populations and needs with a wildlife resource base that

fluctuates with environmental conditions and whose habitat

is declining (Barrett and Arcese, 1995).

The ICDP model assumes that by providing economic

benefits and alternative economic activities, local people

will alter their behavior with respect to wildlife.

Evidence suggests that this assumption is erroneous. Many

rural households have surplus labor and low opportunity

costs for their time. Therefore they can adopt new

activities promoted by the ICDP without abandoning those

activities that the ICDP seeks to eliminate, such as

poaching (Barrett and Arcese, 1995; Southgate and Clark,

1993). For most individuals, the benefits they receive from

the ICDP are less than the costs they incur. In addition,

ICDPs generally provide public services to communities as a

whole, and cannot exclude particular individuals who

continue to poach wildlife (Gibson and Marks, 1995).

Finally, the link between indigenous peoples and

biodiversity conservation is severely strained under market

pressures. Indigenous peoples have the same capacities,

needs, and desires to overexploit the environment as do

other peoples. Given more efficient technologies, growing

populations, and external markets for forest products,

resource exploitation by indigenous peoples no longer is

compatible with biodiversity conservation (Browder, 1992;

Gray, 1990; Marks, 1989; Plotkin and Famolare, 1992;

Redford, 1991; Redford and Sanderson, 1992; Robinson and

Redford, 1994b). These peoples have numerous economic

incentives to "mine" natural resources (Barbier et al.,

1994; R. Smith, 1981; Southgate and Clark, 1993).

Furthermore, development based on forest product

extraction,2 if it is successful in creating markets and

raising prices, inevitably fosters overexploitation as

external political and economic forces intercede (Dove,

1993a; Hanson, 1992; Jeanrenaud, 1992; Ryan, 1991). Others

argue that, because of political and social forces for

wealth accumulation, and the lack of understanding of

complex natural systems, resources are inevitably

overexploited, and sustainable use is impossible (Ludwig et

al., 1993). Thus wildlife is in a Catch-22 position: if it

has no economic value, alternative land uses will eliminate

wildlife habitat; yet if wildlife is valuable,

overexploitation is inevitable (Prescott-Allen, 1982).

In general, ICDPs require serious local and national

political commitment, conducive legislation, appropriate

institutions, low or stable population densities, widespread

use of traditional or appropriate technologies for resource

extraction, and effective protected area management--a set

of conditions unlikely to be found anywhere (Wells et al.,

2For example, extractive reserves in the Amazon.


1992). Notwithstanding these critiques, numerous efforts to

integrate development and conservation are underway across

the tropics, because of the local, national, and

international support for both biodiversity conservation and

socioeconomic development in tropical forests.

Conservation. Development and Wildlife:
The Role of Hunting

Hunting is an important subsistence and economic

activity for forest residents throughout the tropics. Most

forest inhabitants pursue a mixed subsistence and economic

strategy that combines hunting with cultivation, commercial

extraction of forest products, wage labor, etc.

Hunting Methods

Traditional3 hunting technologies are often very

closely adapted to particular ecological conditions: the

desired prey species, the availability of materials for

making weapons, the forest structure, the season, and even

the time of day (Bahuchet, 1988). Hunting methods also

depend on the particular historical and cultural context:

peoples in similar environments adopt different technologies

because of their contacts with other groups (Bahuchet,

1978). Finally, the economic and social environment

determines which technologies among those available are

3Many hunting methods that date back to pre-European
contact or pre-westernization times are still used today.


selected or commonly used: accessibility of technology,

skill and knowledge, human settlement or hunting group size,

subsistence versus commercial objectives, as well as

personal preference and prestige.4 As these cultural and

environmental conditions evolve over time, the role of

hunting in turn continually changes.

Some animals, such as tortoises, are captured by hand.

Clubs are used for killing creatures that could be captured

by hand, that are trapped in nets, or that are cornered by

dogs. Groups of hunters with spears, sometimes working with

dogs, hunt large game including elephants (Bahuchet, 1988;

Brosius, 1991; Harako, 1976; Sandbukt, 1988; Terashima,

1983; Vallois and Marquer, 1976).

Snares are widely used, and are extremely variable both

in terms of the prey captured and the particular techniques

employed. Pit traps, with or without sharpened and poisoned

stakes in the bottom, are used to capture large terrestrial

animals such as pigs and okapi. Gravity traps employ a

spear or log that falls when an animal triggers the trap.

4An important debate in the literature on Zaire Pygmies
discusses the reasons for the reliance on nets versus bows
and arrows by neighboring groups in the Ituri forest. Both
ecological and socio-economic explanations are proposed.
Ecological explanations focus on the structure of the forest
(Roscoe, 1990), the availability of plants used for making
nets (Bahuchet, 1989; Harako, 1976), or the density of
wildlife populations (Milton, 1985; Terashima, 1983).
Socio-economic explanations emphasize the importance of
commercial meat hunting and the value of women's labor
(Bailey and Aunger, 1989; Hayden, 1981; Wilkie and Curran,
1991), or population densities and relations with
neighboring farmers (Abruzzi, 1979; 1980).


Noose snares can be used for virtually any bird or animal,

and are placed on terrestrial or arboreal animal trails.

Alternatively, baits can be used to attract the prey. Nylon

and cable wire are widely used because they are stronger and

last longer than cord made of natural fibers (Schultz,

1986). Snares are often set around crops, both to protect

farms from wildlife depredations and to acquire meat from

animals attracted by the crops (Bahuchet and de Garine,

1990; Carpaneto and Germi, 1992; Orejuela, 1992). But

snares are also set out in the forest, from 20 to as many as

500 in a snare line that is regularly checked (Almquist,

1992; Hewlett, 1977; Infield, 1988; Wilson and Wilson,

1991a). In some cases hunters build long fences with gaps

in them, and place noose or gravity snares in the gaps

(Dwyer, 1990; Sandbukt, 1988).

A variety of projectile weapons are used for killing

terrestrial and arboreal animals of all sizes at a distance:

the bow and arrow, the crossbow, and the blowgun. Hunters

often apply poison to the arrows or darts. A single hunter

may search for game tracks and sounds, or hide in a blind

near a fruiting tree or an animal trail. Hunters also call

animals to bring them within range (Demesse, 1980; Harako,

1976; Harner, 1972; Stearman, 1989a). In Latin America

groups of hunters hunt together when a herd of white-lipped

peccaries has been sighted. The Efe and the Boyela in Zaire

conduct group bow hunts: a group of hunters forms a large


semi-circle, and a group of beaters drives game toward the

hunters (Sato, 1983; Terashima, 1983).

Finally, nets are a specialized hunting technology used

in African savannas and forests (Almquist, 1992; Arom and

Thomas, 1974; Berry et al., 1986; Demesse, 1980; Koch, 1968;

Newman, 1970; Sato, 1983; Sevy, 1972; Tanno, 1981;

Terashima, 1980), and also in India to hunt monkeys (Sinha,

1972; Williams, 1969). The Aka and Mbuti probably adopted

nets from neighboring farming peoples, but few

agriculturalists use nets today. Mbuti hunting nets measure

approximately 1.2 meters high and from 30 to 100 meters

long. They are normally used by a group of men and women

hunting together: several nets are joined together to form

a large semi-circle or a closed circle, and beaters drive

animals into the nets (Ichikawa, 1983). This is an

efficient technology for capturing small terrestrial mammals

weighing less than 20 kilograms, but larger animals are

usually able to break through or leap over the nets.

Hunting methods depend not only on ecological factors,

but also on the role of other economic and subsistence

activities. Farmers tend to hunt near their residences,

venturing out for day hunts or longer stays during periods

when less agricultural labor is required. More mobile

hunter-gatherers, like the Aka of CAR, the Mbuti of Zaire,

the Agta of the Philippines, or the Ache of Paraguay, often


hunt for extended periods from forest camps that can be

several days' walk from roads or permanent settlements.

Across the tropics, modernization means that

traditional cross-cultural diversity in hunting methods is

giving way to the shotgun as an all-purpose, all-season

hunting weapon. The shotgun has a greater range than

traditional weapons, it is often more accurate, and heavy

shot is not deflected by intervening vegetation between the

hunter and the target (Hames, 1979; Rambo, 1978).

The shotgun does have limitations, including the high

cost of guns and ammunition (Beckerman, 1980; Chin, 1984;

Laurent, 1992; Nietschmann, 1972; Paolisso and Sackett,

1985; Saffirio and Scaglion, 1982), reliability problems

with wet powder and misfiring (Yost and Kelley, 1983), and

the noise of firing which scares away other game (Alvard,

1993; Baksh, 1984; Rai, 1982; Ross, 1978). There are also

legal restrictions on owning guns: many countries require

firearm licenses; and Malaysia, Indonesia and the

Philippines have totally banned shotguns under martial law

(Endicott, 1979a; Griffin, 1981; Hoffman, 1986; Morren,


Pressure on Wildlife Resources

Indigenous forest hunter-gatherers adapted to their

environment by living in small, mobile, and dispersed

groups. As they depleted wildlife resources in one area,


they split into smaller groups or moved their camp elsewhere

(Baksh, 1985; Vickers, 1980). Forest farmers also generally

lived in small dispersed settlements. The pressure on

wildlife was geographically and temporally extensive, with

fallows in space and time allowing wildlife populations to

recover and maintain themselves despite human exploitation.

These fallows resulted from a variety of cultural practices:

hunting zone rotation (Almquist, 1992; Balee, 1989; Hames,

1980; 1986; Morren, 1986; Sponsel, 1980), buffer zones

between antagonistic groups (De Boer, 1981; Gross, 1975;

Sellato, 1994), hunting sanctuaries (Turnbull, 1983),

seasonal trekking (Moran, 1981b), and seasonal hunting

(Bahuchet, 1985b; Dwyer, 1982; Eder, 1988a). Traditional

forest hunting was probably sustainable because of low human

population density, simple technology, and subsistence-

oriented production (Redford, 1991; Schmink et al., 1992;

Uquillas, 1985; Vickers, 1991).5

These conditions rarely exist in tropical forests

today. Wildlife exploitation is becoming more intensive, as

both temporal and spatial fallows are decreasing: more

people are hunting continuously in smaller and smaller

areas. The human population in forest regions is growing,

with larger and more permanent population centers (Bahuchet,

5However, even subsistence-level production with simple
technologies may severely depress populations of certain
species, for example slowly-reproducing primates (Godoy and
Bawa, 1993; Peres, 1990; Redford and Robinson, 1985).


1985a; Bahuchet and Guillaume, 1979). A larger number of

people, including immigrants and urban dwellers, now depend

on the same wildlife resource base. Hunting is increasingly

market-oriented (Seeger, 1982).

Large-scale government colonization plans in Brazil and

Indonesia have encouraged landless people in high-density

areas to migrate to relatively unpopulated forests. But in

other countries similar population shifts are occurring

without formal government incentives (Hart and Hart, 1984;

Peterson, 1992a; 1992b; Wilkie, 1987a). Forests comprise an

important frontier that was not exploited more intensively

when there were easier and less marginal environments to

exploit. Other lands are often no longer available for poor

people who must farm to feed themselves. In addition,

valuable products exist in forests: timber, minerals such

as gold and diamonds, and animal products such as ivory,

skins, and bushmeat.6 Technologies and capital are now

available to exploit all these resources intensively.

Hunting is an important subsistence and economic

adaptation that allows forest peoples to survive. But the

available hunting area is declining as people migrate to

forest regions, and as exploitative activities convert

forest to other uses such as cattle ranches, farms,

plantations; or simply degrade wildlife habitat by

6"Bushmeat" is a term from west Africa that refers to
all meat from wild animals killed for food.


intensively extracting resources, for example through

logging and mining (Griffin, 1985; 1989; Headland, 1986;

Rodgers et al., 1991; Stearman, 1990). Furthermore, the

international movement to preserve tropical forests is

providing external funds for tropical countries to create

protected areas in forests, areas which often prohibit or

constrain hunting activities (Doungoub6, 1990; Nations and

Hinojosa, 1989). At the same time, shifts to new

technologies such as firearms and cable snares, combined

with production for external markets as well as subsistence

needs, all tend to increase the intensity and efficiency of

hunting activities. Although few researchers have collected

both socioeconomic and ecological data,' it is likely that

harvest rates are exceeding reproduction rates for many

exploited species.

Wildlife Management Practices

Forest peoples in most cases exhibit a fatalistic

attitude towards wildlife, exploiting wildlife resources as

if they are unlimited. Some hunters associate hunting

yields with mystical spirits: if yields decline it is

because the spirits have not been properly treated, not

because people have overhunted (Bulmer, 1982; Dwyer, 1982).

7The exception is a research team in Peru which has
collected abundance, productivity, and harvest rates for
particular game species in a particular location (Bodmer,
1994; Bodmer et al., 1988a; 1988b; 1994; 1995).


Much indigenous knowledge of wildlife, intimate though it

may be, focuses on harvesting as opposed to managing these

resources. Therefore, most forest peoples do little to

conserve harvests. The primary concern is for short-term

yields, not for long-term conservation of wildlife (Bulmer,

1982). Hunting is usually opportunistic: any animal of any

species encountered is pursued, and hunters do not

discriminate by sex or age of prey, capturing females and

young whenever they are available (Alvard, 1993; 1994; Hames

and Vickers, 1982; Redford and Robinson, 1985).

In the past, there were always other unoccupied areas

to which people could move when local resources were

depleted, and permanent environmental degradation did not

result from their activities (Johnson, 1989; Kaplan and

Kopische, 1992). In the modern context, perhaps they lack

economic alternatives and must emphasize short-term needs at

the expense of long-term sustainable resource management.

Alternatively, they may fail to anticipate future scarcity

resulting from overexploitation because resources have

always been abundant in the past.

Taboos and other cultural beliefs may serve to manage

or protect wildlife by reducing the motivation to hunt

particular species (Silberbauer, 1981; Smith, 1976b;

Sponsel, 1980), and by distributing hunting pressure across

a greater diversity of species (Arhem, 1976). However,

traditional belief systems are increasingly superseded by


western culture (Bennett, 1962; Dwyer, 1982; Hames, 1991).

Market pressures undermine taboos to the extent that the

market places a premium on previously unexploited species,

encouraging individuals to violate cultural norms for

material gain (Berlin and Berlin, 1983; Moran, 1981a). As

other game declines, pressure increases to hunt tabooed

species (Nietschmann, 1973; Peres, 1990). In some cases,

although people will not consume certain species of game,

they will kill that species when encountered because the

meat can always be sold (Johnson, 1989; Sponsel, 1980). In

other cases, previously "inedible" or tabooed species are

being reclassified as edible (Hames and Vickers, 1982; Speth

and Scott, 1989; Yost and Kelley, 1983).

When there are few visible signs of degradation, there

are also few incentives to conserve wildlife resources, and

formal institutions or controls on hunting activities do not

develop (Harms, 1987; Mulongo, 1989; Testart, 1977). Mobile

populations not tied to a specific resource base and defined

territory gain little from prudent resource use (Gadgil et

al., 1993). Alternatively, people may not perceive

declining wildlife as a risk at all, or as a risk that they

are able to address (Kottak and Costa, 1993). Often they

are unable to restrict access to hunting areas and wildlife

(Bahuchet, 1979b; Hames, 1986; Orejuela, 1992; Rodgers et

al., 1991). Rather than a "tragedy of the commons" there is

a tragedy of open access resources (Feeny et al., 1990;


Hardin, 1968; McCay and Acheson, 1987). Professional

hunters from elsewhere have little difficulty in entering

the area to pursue their short-term economic goals, while

other immigrants can also freely establish residences and

farms. Local people then lack the incentives to conserve

wildlife resources because they bear all the costs and are

able to capture few of the benefits in return.

The Subsistence Role of Hunting

The subsistence role of hunting for human populations

depends on the other subsistence activities possible in the

physical and cultural environment. Across the tropics,

wildlife meat or "bushmeat" is an important subsistence

resource for forest peoples. In west and central Africa,

60-80 percent of meat consumed by forest residents is

bushmeat (Ajayi, 1979; Asibey, 1974; 1987; Heymans and

Meurice, 1973; Martin, 1983; Prescott-Allen and Prescott-

Allen, 1982).

The role of wildlife meat in the diet varies

considerably over time, and among neighboring ethnic groups

and settlements. People who do not farm depend more on

wildlife for subsistence, both directly consuming the game

they capture, and indirectly to the extent that they sell

meat in exchange for other foods. Central African Pygmies

traditionally maintained trading relationships with farmers

of other ethnic groups (Bahuchet and Guillaume, 1979; 1982;

Bahuchet and Thomas, 1986; Berry et al., 1986; Delobeau,

1989; Terashima, 1986). Similar long-standing symbiotic

trading relationships occur in Brazil (Milton, 1984), and

across Asia (Gardner, 1969; Griffin, 1981; 1984; Hoffman,

1988; J. Peterson, 1978a; 1981; Rai, 1982; Rambo, 1979a;

1980; 1982b). Although meat is traded, production is for

subsistence at the local level: hunters provide the

subsistence requirements of meat for both groups, while

farmers provide the subsistence requirements of starch.

Cultivation methods and crops can affect hunting

intensity and frequency. Certain crops like manioc require

relatively little attention. Manioc farmers can therefore

spend extended periods away from their fields hunting in the

forest (Nietschmann, 1973). By dispersing fields throughout

the forest, forest peoples can effectively expand their

hunting range (Hart, 1979; Posey, 1985). On the other hand,

crops that require much more care, for example rice or

maize, reduce the time available for hunting, and often

increase the reliance on trapping (Behrens, 1984; Carpaneto

and Germi, 1992; Dwyer, 1985; Dwyer and Minnegal, 1991;

Sandbukt, 1988; Sato, 1983). Thus, increasing dependence on

agriculture may reduce the subsistence role of wildlife

(Bahuchet, 1985b; Johnson, 1977). But by providing greater

food security, which allows population density and

settlement size to increase, cultivation may increase

hunting pressure, particularly on large game species which


provide the highest returns (Speth and Scott, 1989; Werner

et al., 1979).

Some researchers argue that cultivation can increase

the abundance of game species, and therefore the potential

subsistence role of hunting. Traditional shifting

cultivation creates a mosaic of forest patches at different

stages of regrowth' (Bailey et al., 1990; Barnes et al.,

1991; Hart and Petrides, 1987; Ross, 1978; Sato, 1983;

Wilkie, 1987b; Wilkie and Finn, 1990), increases the

abundance of forest/forest edge ecotones (J. Peterson, 1977;

1981; 1982), and provides food sources for wildlife in both

current and abandoned farms (Linares, 1976; Posey, 1982;

1985). The vegetation diversity benefits many opportunistic

or colonizing wildlife species, or those which depend on two

or more habitat types, including important game species like

pigs and deer in Asia, and rodents in Africa and Latin

America (Dufour, 1990; Speth and Scott, 1989).

The Economic Role of Hunting

The boundary between subsistence and commercial

production is often indistinct: all meat that is not

consumed can be sold in local markets (Peres, 1990). For

many forest peoples, hunting has for generations or even

centuries had economic as well as subsistence objectives

(Bower, 1989; Healey, 1990; Rambo, 1979b; Sellato, 1994;

8Selective logging can perform a similar function.

Spielmann and Eder, 1994). For the past 50 years,

commercial hunting of crocodiles for skins has helped the

Sirion6 of Bolivia through annual periods of agricultural

scarcity, and allows them to obtain luxury goods (Stearman

and Redford, 1992). In fact, hunting is an integral part of

all forest-based activities, a "subsidy from nature" without

which other commercial activities like rubber tapping or

Brazil nut collecting would not be possible (Hecht et al.,

1988; Redford, 1992; 1993). Likewise, hunters supply meat

to gold miners and loggers (Alvard, 1995; Stoll, 1992).

As urban populations increase, rural populations become

ever more important as food producers. Urban areas,

especially in Africa, exert a particularly strong pressure

on hunters in forest areas to produce bushmeat (Anadu et

al., 1988; Colyn et al., 1987; Falconer, 1990; Steel, 1994).

This market provides a major income-earning opportunity for

forest peoples who have in many cases benefitted little from

development elsewhere in the country (Infield, 1988).

Furthermore, professional hunters are increasingly turning

their attention to forest areas as wildlife resources are

depleted in more accessible regions (Rodgers et al., 1991).

Urban dwellers in Ghana make weekend hunting trips to forest

areas to earn supplementary income (Asibey, 1980).

The bushmeat trade increases the intensity of hunting

by expanding the population base consuming the meat, and the

time spent hunting (Colyn et al., 1987; Hart, 1978).

Commercial meat traders, by provisioning hunters with food

and hunting equipment, can also promote the expansion of

hunting into more remote areas that were previously

inaccessible. Thus the distribution of Mbuti hunters in

Zaire no longer depends on the distribution of farming

villages to supply their carbohydrate needs (Hart, 1979;

Peacock, 1984). Trade also alters hunting methods by

diffusing new technologies9 from outside forest areas, and

by causing hunters to focus on more marketable prey species

(Hayden, 1981).

The long-term indirect effects of trade on hunting and

social change are difficult to predict (Godoy and Bawa,

1993). Trade may induce specialization by particular groups

in hunting: trade allows neighboring ethnic groups to

coexist by emphasizing narrower ecological niches, thus

increasing the potential human carrying capacity of the

environment and reducing the potential for conflict

(Bahuchet et al., 1990; Berry et al., 1986; Fox, 1969;

Headland, 1986; Milton, 1984; Peterson, 1978b; Turnbull,

1965). In Asia, the trade in wildlife products has affected

hunting practices for centuries: inducing relatively

sedentary agriculturalists to become mobile hunter-gatherers

in order to more effectively harvest forest products for the

market (Hoffman, 1986). Trade has encouraged the Miskito of

'For example, nets and crossbows in central Africa, and
shotguns everywhere.


Panama to emphasize meat production at the expense of

subsistence agriculture, to the extent that hunting

activities conflict with cultivation for time, space and

labor (Nietschmann, 1972; 1973). Alternatively, certain

individuals within groups become professional hunters (Ayres

et al., 1991; Carpaneto and Germi, 1992). The persistence

of these specialist individuals or groups through time

clearly depends on the persistence of wildlife as an

economically important resource. Intensive exploitation may

deplete wildlife to the point where it can no longer support

specialized hunters. As they turn to alternative economic

and subsistence activities, hunting will decline.


Based most importantly on the work of Julian Steward,

cultural ecology combines theories and methodologies from

geography, anthropology, and ecology. Cultural ecology

attempts to grapple with the extremely complicated problem

of human-environment interactions through a systems approach

that integrates ecological and socioeconomic systems. It

has provided a general framework for analysis of

interactions, rather than specific methodologies of its own.

Research has emphasized people-environment relationships in

developing countries. In this way, cultural ecology can

play an important role in guiding current efforts to

integrate conservation and development, two subjects which


are particularly problematic in tropical forests. In recent

years, natural scientists promoting ecosystem conservation

and social scientists advocating human system "development"

have begun to recognize the interdependencies of their two


This dissertation seeks to build upon and advance the

theory and practice of cultural ecology described above.

The field work continues in the geography and cultural

ecology traditions to integrate disciplines and research

subjects. Rather than analyzing the entire

people/environment system, the dissertation focuses on

hunting in order to investigate ecosystem/biodiversity

conservation and human system/socioeconomic development, and

attempts to integrate the two in a particular African case


Wildlife exploitation for subsistence and commercial

purposes can contribute to socioeconomic development. If

exploitation is sustainable, it can also support

biodiversity conservation. But sustainable exploitation

depends on appropriate management practices and the ecology

of the resource base.

People are currently harvesting wildlife, and will

continue to do so, throughout tropical forests because wild

animals provide both protein and income. The complete

prohibition of hunting is unrealistic and unenforceable.

Integrated conservation and development projects seek to


direct this exploitation toward more sustainable practices

and levels. Such a course of action requires accurate

information on the ecosystem, in the form of biological and

ecological data on local game species, combined with

information on the human system, in particular on hunting

methods, motivations, and harvest rates.

This dissertation provides both types of information

for the Bayanga region of the Central African Republic based

on 16 months of fieldwork, and complemented by data from

research elsewhere. The following chapter describes the

ecosystem and the human system of the Bayanga region.

Chapters 4 and 5 report the results of the study of cable

snare hunting and communal net hunting respectively.


The cultural ecology framework emphasizes the relations

between two systems: the ecosystem and the human system.

The research site is the Bayanga region of the Central

African Republic, the area south of Nola that is bounded by

the Congo border to the east, the Cameroon border to the

west (Figure 3.1). This chapter examines the components of

each system in the Bayanga region.

The regional ecosystem, like other tropical

rainforests, contains high species diversity of plants and

animals. This chapter focuses on the resource base

represented by the primary game species and implications for

their sustainable harvesting. Components of the human

system include political authorities, economic activities,

and population centers. Change in each of these components

over time in turn affects patterns of wildlife exploitation.

In order to illustrate the influence of hunting on relations

between the ecosystem and the human system in the Bayanga

region, the chapter concludes by discussing several cases

and issues: elephants, hunting legislation, bushmeat

markets, and currency devaluation.

Figure 3.1: The Central African Republic and the Research

Source: Carroll, 1986b.

The Ecosystem

The tropical rainforest ecosystem of the Bayanga region

is complex, and offers many resources that have long been

exploited by colonial and independent governments, foreign

companies, immigrants, and local residents. This section

describes general ecosystem characteristics, then focuses on

wildlife resources, which are exploited by local residents

for food and for markets. The principal game species are

described in detail, with particular attention given to


population densities, predictions of sustainable harvest

rates, and the impacts of harvesting.

Ecosystem Characteristics

The Bayanga region lies at the upper limit of the great

Congo basin tropical rainforest. Relief is relatively flat,

averaging 350 meters in altitude to the east of the Sangha

river, with hills rising to 700 meters to the west of the

Sangha. The entire region is drained by the Sangha river,

flowing south to the Oubangui and the Zaire rivers.

Seasonal temperature variations are slight: monthly

averages vary only 20 Celsius. But diurnal variations are

as high as 140 C (Figure 3.2). Average annual rainfall of

less than 1,400 millimeters is low for a rainforest

(Richards, 1952). A three-month dry season lasts from

December to February, though it does rain during this period

(Figure 3.3). Because of these relatively constant and

stable climatic conditions, most animals are permanent

residents that do not make long-distance seasonal

migrations, and reproduction continues throughout the year.

Maximum, minimum, and monthly averages

301- 1



Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Figure 3.2: Average Temperatures (OC): Bayanga, 1974-1984

Source: Carroll, 1986b.

Average annual rainfall I












0 -

May Jun Jul Aug Sep Oct Nov Dec

Jan Feb

Mar Apr

Figure 3.3: Average Rainfall (mm): Bayanga, 1973-1984

Source: Carroll, 1986b.

Forest types include evergreen, deciduous and

evergreen-deciduous transition forests. Closed forest

formations cover 94 percent of the 8,330 km2 area from Nola

south, including mixed forests and patches of

Gilbertiodendron dewevrei monodominant stands on well-

drained soils (Figure 3.4). Poorly-drained soils are

covered with almost pure stands of Guibourtia demeusii,

Gilbertiodendron dewevrei, or raffia palms (Raphia sp.).


1357 mm



SMarshy clearing (bi)
Mixed dosed forest
SDepleted/secondary forest
Gilbertiodendro dewevrel forest
International boundaries
Sangha river

Figure 3.4: Vegetation of the Bayanga Region

Source: Carroll, 1986b.


0 10 25
I Ieter

9sl E,


Patches of savanna occur to the northeast. Herbaceous

marshy forest clearings called "bai" are maintained, and

perhaps created, by heavy wildlife usage, particularly

elephants searching for minerals in the soil (Carroll,


Wildlife diversity, as in other tropical forests, is

high (see Appendix 1 for a species list). Wildlife

exploitation focuses primarily on mammals, although many

species of birds, reptiles, and insects are also collected.

The most important mammal species from a conversation

perspective are the forest elephant (Loxodonta africana

cyclotis) and the lowland gorilla (Gorilla gorilla gorilla)

which have been surveyed locally by Carroll (1986c; 1988)

and Fay (1989; 1991).1

In addition, a variety of ungulate species are present:

forest buffalo (Syncerus caffer nanus), bongo and sitatunga

antelope (Tragelaphus euryceros and T. spekei), seven

species of duikers (Cephalophus spp.) and two types of pigs

(Potamochoerus porcus and Hylochoerus meinertzhageni) (Fay

et al., 1990; Wilson and Wilson, 1989). Fifteen species of

primates occur, including chimpanzee (Pan troglodytes) and

several species of monkeys (Cercopithecus spp., Cercocebus

spp., Colobus guereza, and Procolobus badius). The

carnivores include leopard (Panthera pardus), golden cat

1Additional research is in progress: on elephants by
Andrea Turkalo, and on gorillas by Melissa Remis and Michele


(Profelis aurata), and several mongoose and genet species

(Viverridae) (Blom, 1993; Carroll, 1986b; 1986c).2

Most forest mammals that are primary consumers are

frugivorous, though the larger species also browse. Grazing

is limited to the several dozen marshy clearings and is

probably only important to buffalo. However, many other

species also frequent bais to obtain minerals such as

calcium. The Dzanga bai, 300 meters by 150 meters, is most

noteworthy for the spectacle of elephant agglomerations

exceeding 100 individuals.3 In addition to the elephants,

present virtually 24 hours a day since anti-poaching efforts

took effect in 1988, other species seen at Dzanga include

buffalo, bongo, sitatunga, black-fronted duiker, bush pig,

giant forest hog, as well as flocks of African grey parrots.

Gorillas visit some of the other bais.

Wildlife abundance and dispersal is related to the

nature of food resources. The diversity of trees and the

relatively weak seasonality ensures that some species are

fruiting at all times of the year, but the food resources

are patchily distributed. In order to exploit these

resources, animals which forage in groups move around

extensive areas. These animals include the diurnal primates

2Justina Ray has studied mongooses in the Bayanga

3Unlike savanna elephants, the forest subspecies forms
only small groups of 2-4 individuals (Merz, 1986; White et
al., 1993).


and the larger terrestrial mammals such as elephants,

buffalo, bongo, and bush pigs. The smaller terrestrial

mammals such as sitatunga and all duikers) are relatively

sedentary and solitary.

The four principal game species exploited by local

people, for subsistence and for the trade in bushmeat, are

the brush-tailed porcupine (Atherurus africanus) and three

duikers (Cephalophus callipygus, C. dorsalis, and C.

monticola). These species are described in detail below.

Species Characteristics

Appendix 2 provides biological information for the four

game species, based on measurements of captured animals in

the Bayanga region and on detailed studies by researchers

elsewhere in Africa. Information on duiker reproduction is

inconsistent, but the most accurate information is available

from Vivian Wilson's on-going breeding research at

Chipangali Wildlife Trust in Zimbabwe. Duikers bear only

one young at a time. They breed continuously throughout the

year, although birth rates are somewhat higher in particular

seasons of the year. Individuals of all species reach

sexual maturity (first oestrus) at 11 months, and after a

gestation period of 7 months produce their first offspring

at 18 months. Conception occurs again within days of

parturition. Therefore female duikers reproduce every 210

days, and are pregnant virtually continuously. Reproductive


rates in the wild may be somewhat lower, perhaps one young

per female per year. Longevity is probably 8-10 years.

Very little is known about A. africanus, but it is thought

to reach sexual maturity at two years of age, and to

reproduce one young at a time after a gestation period of

100-110 days (Rahm, 1962).

Individuals of all duiker species are generally

solitary, except for C. monticola which often forage in

pairs or even small groups including two parents and a young

animal (Dubost, 1980). A. africanus forages alone, but is

less territorial than the duikers, with several individuals

and families sharing a sleeping area (Emmons, 1983).

All species are principally frugivorous, though the

duikers browse as well, and A. africanus digs for roots and

tubers. In addition, duikers consume small amounts of

animal matter and insects in order to obtain necessary salts

(Kingdon, 1982). Duikers consume fruit in direct proportion

to fruit abundance: when fruit is not available, more

leaves are consumed (Feer, 1989a). In the rain forests of

Gabon and Zaire fruits comprise more than 70 percent of the

diet throughout the year (Table 3.1).4

4But in South African dry forests, where fruit is not a
reliable resource, C. monticola depends on leaves for 70
percent of its diet (Faurie and Perrin, 1993).


Table 3.1: Game Species Activity, Weight, and Diet
(diurnal/nocturnal, kgs. percent of fruit/leaves in diet)

Activ Weight Fruit Leaves Fruitc Fruitd

C. callipygus" D 20.1 82.7 16.3 86-89 88
C. dorsalisa N 21.7 72.9 26.6 86-89 75
C. monticolaa D 4.9 78.5 20.3 75
A. africanusb N 3.3

Activity: D=diurnal, N=nocturnal.
Weight: average for species, in kg.
Fruit: percent of fruit in diet (dry weight).
Leaves: percent of leaves in diet (dry weight).

Sources: From Gabon: aDubost, 1984; bEmmons, 1983; and
cFeer, 1989a. From Zaire: dHart, 1985.

Because ecological conditions in the Bayanga region are

relatively stable and predictable, ranges and diets likewise

do not vary much. Most duikers exhibit little weight change

in the dry season, when food resources are least abundant,

although the larger duikers expand their home ranges during

this season (Dubost, 1984). All four species are

territorial, and they probably remain on the same territory

throughout their lifetime. In South Africa, C. monticola

are so strongly territorial that they rarely leave their

home range even when beaters shouting and shaking leaves try

to drive them out: instead they double back through the

line of beaters or hide (Bowland, 1990). More detailed

information on the four game species is provided below.

Atherurus africanus

A. africanus is strictly nocturnal, hiding by day in

thick brush or in burrows abandoned by other animals. It

may use the same shelters for many years, with well-defined

paths to established feeding areas. It forages alone, but

often joins others in dens during the day. Studies in Gabon

find home ranges averaging 13.3 hectares. However, it is

possible that members of a clan share a territory (Emmons,

1983; Kingdon, 1982; Rahm, 1962).

Cephalophus callipygus

C. callipygus males maintain a home range of 27-49

hectares, while females' home ranges are 40-42 hectares.

Territories are larger in the dry season. Reproduction is

continuous throughout the year. C. callipygus prefers

primary forest with relatively thin undergrowth (Estes,

1991; Feer, 1989a; 1989b).

Cephalophus dorsalis

C. dorsalis is strictly nocturnal, and is much less

active than the diurnal C. callipygus, covering in its daily

foraging only one-fourth the distance that the latter covers

(Estes, 1991). It hides in tree hollows or very thick brush

by day, but at night forages in more open areas. Studies in

Gabon find male home ranges (79 hectares) overlapping two

female home ranges (41 hectares), and territories are larger

in the dry season (Dubost, 1983; Feer, 1989a; 1989b).

Research in Zaire reports territories averaging 15.2

hectares, but individuals also will make direct long-

distance moves to known food patches (Hart, 1985).5

5Reported differences in home range size may result
from different research methods, or from habitat

Cephalophus monticola

C. monticola occupies all types of forest where the

undergrowth is sufficiently clear (visibility greater than

20 meters), including primary and secondary forest resulting

from cultivation and logging (Fimbel, 1994). It forages in

these clear areas, but rests and hides from danger in denser

vegetation and brush (Estes, 1991; Hanekom and Wilson,

1991). It is exclusively diurnal, and is active for a

longer period in the dry season, perhaps because less food

is available.

Studies find male-female pairs sharing territories of

3.7-6.4 hectares in Zaire (Hart, 1985), and 2.5-4 hectares

in Gabon (Dubost, 1980). The home range varies little with

the seasons, remaining quite stable throughout the adult's

life. Each family member protects its territory from

outsiders of the same sex. Males may venture out of their

territories but only 100-300 meters to a particular fruiting

tree, and for a few hours at most. Male-female pairs forage

together or alone, but are sometimes accompanied by their

own or migrant young. The young migrate to find unoccupied

territories, and annual population turnover is from 7-10

percent (Dubost, 1980; 1983). Reproduction occurs

throughout the year, but is relatively more common in the

dry season (Mentis, 1972).

Population Characteristics

This section describes population characteristics of

the four game species, most importantly population

densities, age structure, and sex ratios. These

characteristics are critical in determining the effects of

hunting on game species, and especially in evaluating

whether hunting is sustainable.

Population densities

Densities are the population characteristic most often

measured in the field (Table 3.2). Density estimates may

vary because of differences in methodologies: track and

pellet counts, line transect surveys, and radio-telemetry.

Variations among research sites also result because of

differences in ecological interactions such as vegetation

types or competition with other frugivores, as well as

differences in patterns of human intervention: hunting

pressure and habitat disturbance.

Table 3.2: Population Densities for Game Species
at Research Sites in Gabon, Zaire and Cameroon
(individuals per square kilometer)

(1) (2) (3) (4) (51) (6)
A. africanus 77.7 12.0a 45. ib 30.0
C. monticola 70 30.4-53.0 62-78c 14.9 61-69
C. callipygus 10.7-15.5 0.6-6.7 3.2
C. dorsalis 6.3-8.7 2.5-6.8 1.5

Sources: For Gabon: (1) Feer, 1988; 1989b; 1993; (2) Lahm,
1993; (3) bEmmons, 1983; and CDubost, 1980. For Zaire: (4)
Hart, 1985; (5) Wilkie, 1987b. For Cameroon: (6) Payne,
Methods: I have included duiker densities only for sites in
Gabon and Zaire, because the same guild of duiker species
occurs in these sites as in the Bayanga region.
Feer, Dubost, Emmons and Hart used radio-telemetry to
measure home ranges. This is the most accurate method of
determining population densities. bEmmons provides only
average home ranges for A. africanus, which I have used to
calculate densities assuming seven individuals per adult
home range.
Hart also conducted censuses using groups of net
hunters, as well as pellet and track count surveys. Wilkie
conducted track and pellet count surveys as well.
Lahm and Payne conducted day and night line transects.
"Lahm recorded only encounter rates for A. africanus, which
I have used to calculate approximate densities according to
Payne's (1992) mean sighting distance for the species in
night transect surveys.

Age structure and sex ratios

Emmons (1983) reports that males comprise 52 percent of

subadult A. africanus, and 58 percent of adults. The only

age structure and sex ratio information on duikers in the

literature is for C. monticola in Gabon (Table 3.3).


Table 3.3: Age Structure for C. Monticola
(number and percent female/male by age classes)

In region (results from hunting)
Embryos Births MO M1 M2 Subad Adults Repr
Number 20 10 11 8 29 22 54 76
Female (%) 30 50 62 75 59 69 44 51
Male (%) 70 50 38 25 41 31 56 49

In study area
Embryos Births MO M1 M2 Subad Adults Repr
Number -- -- 5 13 28 23 79 102
Female (%) -- -- 60 68 54 67 49 53
Male (%) -- -- 40 32 46 33 51 47

Notes: MO = 0-4 months old, no molars visible.
M1 = 4-10 months old, 1st molar erupting, but not 2nd.
M2 = 10-20 months old, 1st and 2nd molars out, but not 3rd.
Subad = sub-adults, 20-28 months, complete dentition.
Adult = older than 28 months, tooth wear visible.6
Repr = Total reproducing individuals including adults and

Source: Dubost (1980), p. 247. Reprinted with permission
of the publisher, Blackwell Wissenschafts-Verlag GmbH,

Dubost assumes that all individuals in the subadult and

adult age classes are breeding. However, young individuals

without territories probably cannot breed successfully, so

the actual breeding population may be smaller than the

number of sexually mature individuals. The age and sex

structure data for C. monticola exhibit a strong deficit of

young males in the population--these are the individuals

most constrained by intraspecific competition for unoccupied

territories (Dubost, 1980). Young are forced by their

'These age categories do not correspond to those of
Vivian Wilson, who finds captive duikers reaching sexual
maturity, and presumably having complete dentition, within
one year. However, most mammals grow more quickly and
reproduce at an earlier age in captivity than in the wild.


parents to emigrate and find unoccupied areas to claim for

themselves. Their survival therefore depends on being able

to find a suitable territory. Territoriality is determined

by the availability of food resources as well as competition

for enemy-free space. Mortality results from young being

forced into more unsafe environments, for example outside

protected areas or near human settlements.

Ecological interactions

The net effects of other species on the game species

are complex interactions of predation, competition, and

facilitation. Natural predators of duikers and porcupines

in the Bayanga region include leopards, golden cats, eagles,

pythons and large vipers. Other natural forms of mortality

include old age, disease, starvation, and accidents.7

Natural mortality removes territorial adults and creates

gaps for young individuals without territories to colonize.

Mortality rates from natural causes for young duikers less

than two years of age are high: 30 percent for C.

monticola, 70 percent for C. callipygus, and 87 percent for

C. dorsalis (Dubost, 1980; Feer, 1988).

The game species also compete with other frugivores,

for example, monkeys, squirrels, birds, and elephants

(Gautier-Hion et al., 1980). One study suggests that duiker

7During my fieldwork net hunters found a dead sitatunga
mired in mud, and a hippopotamus died near Bayanga in the
same way. Several years previously a dead elephant was
found trapped between a tree and a river bank.


densities are lower where abundant elephants compete with

duikers for fruit (White, 1994a). But game species derive

benefits from other animals at the same time: arboreal

frugivores pick fruits and drop them to the ground, so

duikers often follow and feed below mixed monkey troops

(Fay, 1989; Feer, 1989a). An estimated 50 percent or more

of fruit available to terrestrial ungulates consists of

fruit or parts of fruit discarded by primates (Hart, 1985).

Furthermore, elephants benefit duikers by clearing trails

through the forest, and by maintaining bai clearings where

minerals are available (Carroll, 1986a; 1986c).

Estimated Population Densities in the Bayanga Region

Researchers have employed a variety of techniques to

estimate population densities of terrestrial wildlife in

African forests: day and night line transects, pellet and

track counts, and radio-telemetry (Dubost, 1980; Emmons,

1983; Feer, 1989b; Fimbel, 1994; Hanekom and Wilson, 1991;

Hart, 1985; Heymans and LeJoly, 1981; Lahm, 1993; McCoy,

1995; Nummelin, 1990; Payne, 1992; Struhsaker, n.d.; White,

1994a; Wilkie, 1987b). Radio-telemetry is the most accurate

method, yet also the most expensive and time-consuming. All

other methods are problematic, especially for counting small

cryptic animals such as duikers in thick vegetation. Census

methods are therefore more reliable for determining indices

of abundance rather than absolute density estimates.

Line transect surveys

I chose to conduct day-time line transect surveys for

several reasons. Because a conveniently-located set of

transects already existed, this method was the easiest and

conflicted least with other research activities. In

addition, pellet and track counts are unable to distinguish

between C. callipygus and C. dorsalis. Finally, night

transects can be dangerous where elephants, gorillas, and

gun hunters are numerous.

Between December 1993 and October 1994 I conducted a

total of 97.7 kilometers of surveys along a set of transects

established by the World Bank-funded Projet d'Amenagement de

Ressources Naturelles between Mossapoula and Bayanga (World

Bank, 1990). The transects are four parallel and straight

paths roughly two meters wide, each two kilometers in length,

and 500 meters apart. The perpendicular paths connecting

the ends of the four parallel trails--two 500 meter sections

at the east end, and a single 1,500 meter section at the

west end--were also censused. Figure 3.5 shows the location

of line transect surveys, as well as the net hunt censuses

described below. Two observers walked together at a pace of

1-1.5 kilometers per hour, beginning between 10 a.m. and

noon, and covering 5-6 kilometers per day. For each animal

seen or heard, I recorded the following information:

species, number, angle, perpendicular distance from


transect, animal to observer distance, forest type, and

time. Duikers often whistled in alarm as they flushed.

Figure 3.5:

Location of Line Transect and Net Hunt Surveys

The method makes the following assumptions: 1) all

animals on the transect line are seen; 2) each animal's

position is fixed before it moves away from the observer; 3)

no animal is counted twice; 4) distance measurements are

exact; and 5) sightings are independent events (Anderson et

al., 1979; Burnham et al., 1980; Eberhardt, 1978). All

assumptions were probably violated in practice. Duiker

flight responses depend on the vegetation. Where the


understory is relatively clear, duikers tend to move away

quietly and are not observed at all, or they flush further

away from the observer than their original position. Where

the understory is thicker, duikers more often freeze and

flush when the observer is very close, or remain hidden.

Duikers, and especially the nocturnal bay duiker C.

dorsalis, may hide in thick brush in the middle of the day

when most censuses were conducted. Angle and distance data

are estimates and were not measured exactly. Finally the

sample size for all species is too small for reliable

density estimates.8 Nevertheless, I present the

information because no other data are available on duiker

population densities for the Bayanga region (Table 3.4). No

A. africanus were encountered.

Population density is calculated using the following

formula (Whitesides et al., 1988):

D = (N 1000) / 2LW

The width of the census path W is two times the mean

sighting distance for each species, assuming that the same

number of animals beyond this distance were seen and

included as were missed within the sighting distance. Two

alternative sighting distances are used in the table above:

W1 is the perpendicular animal-to-path distance, and W2 is

the animal-to-observer distance. The differences in density

'Buckland et al. (1993) recommend a minimum sample size
of 60-80 individuals.


estimates according to the two sighting distances are

statistically significant for C. monticola (Wilcoxon

matched-pairs signed-ranks test, T, = 0, p < 0.01).

Table 3.4: Estimated Population Densities
from Line Transects (animals/km2)

N L E W1 SD (+) D1
C. monticola 33 97.7 0.34 9.03 7.57 18.70
C. callipygus 4 97.7 0.04 18.25 9.44 1.12
C. dorsalis 1 97.7 0.01 20.00 -- 0.26

N L E W2 SD (+) D2
C. monticola 33 97.7 0.34 16.11 9.36 10.48
C. callipygus 4 97.7 0.04 22.00 13.49 0.93
C. dorsalis 1 97.7 0.01 20.00 -- 0.26

Notes: N = number of individuals seen
L = length of transect (km)
E = encounter rate (individuals/km)
W1 = mean animal to path distance (m)
W2 = mean animal to observer distance (m)
D1 = population density (individuals/km2) based on animal to
path distance
D2 = population density (individuals/km2) based on animal to
observer distance
SD = standard deviation of W1 and W2 respectively.

Although the transects covered both mixed (8,325 meters

S80 percent) and monodominant Gilbertiodendron dewevrei

forest (2,175 meters 20 percent), only two observations (5

percent), both of C. monticola, were in monodominant forest.

The transect area was regularly visited by net hunters from

Mossapoula, as well as by residents of both Bayanga and

Mossapoula seeking other forest products. Therefore human

disturbance was high, resulting perhaps in low actual

densities. Transect surveys may also underestimate actual


densities at this site if animals are wary and avoid

detection by census takers.

Net hunt encounters

In addition to the line transect censuses, I have also

estimated population densities based on encounters during

net hunts from Mossapoula and from forest camps (Table 3.5).

Ichikawa (1978) also used encounters within the enclosed

circle of nets to estimate population densities for C.

monticola of 70-120/km2. Other researchers have censused

duiker populations using groups of observers with nets,

setting the nets randomly and counting only the animals

inside the enclosed area (Dubost, 1980; Hart, 1985; Koster

and Hart, 1988). However, because I was observing hunters,

the census path was not random: they do not hunt in

Gilbertiodendron dewevrei forests or in marsh/swamp

forests.9 And within the mixed forest where they hunt,

they encircle and search areas with patches of thick brush

or treefalls, where animals are more likely to hide.

9In general, Gilbertiodendron dewevrei forest has a
very open understory, with few hiding places. Marsh and
swamp forest is too thick and wet for net hunting.

Table 3.5: Estimated Population Densities
from Net Hunt Encounters (animals/km2)

Distance from

<5 kms (N=39 hunts)
C. monticola
A. africanus
C. dorsalis
C. callipygus


-1 n LrmsQ

C. monticola
A. africanus
C. dorsalis
C. callipygus










SD (+)

SD (+)

SD (+)

SD (+)

10-15 kms (N=6 hunts)

C. monticola
A. africanus
C. dorsalis
C. callipygus

C. monticola
A. africanus
C. dorsalis
C. callipygus



SD (+)

SD (+)

Notes: N = number of animals encountered (captures and
escapes) by the entire net hunting group: N1 = encounters
inside the net circles; N2 = total encounters during hunt.
D = population density (individuals/km2). D1 is based
on the area inside the net circles. D2 is based on the
total area searched by the net hunting group, assuming a
path width between one and two times the net circle
diameter, and a path length based on the net circle diameter
and a 100-200 meter distance between casts.

10-15 kms (N=6 hunts

(N=22 hunts)

The first density estimate (Dl) is based on the area

inside the circles enclosed by the nets. The area is

determined by the number of nets used per hunt (maximum 20),

an average net length of 18.27 meters (N=54), and the number

of times the nets are set. I assume that all circles during

a given hunt are the same area, but the area varies between

hunts according to the number of nets. The number of

animals "censused" includes all captures and escapes from

inside the circles. I have excluded occasions when an animal

was seen first, and the nets were then set to capture that


The second density estimate (D2) is based on the total

area searched by the net hunters. In practice, the width of

the search path varies considerably: at times the hunting

group is close together near the nets, at other times the

group is spread over a considerable area as individuals lag

behind or detour to gather other forest products. I

calculate the daily path width using two alternative

measurements: one and two times the diameter of the net

circle. The distance between casts also varies. I have

calculated the daily path length according to the circle

diameter, the number of times the nets are set, and two

alternative distances between sets: 100 and 200 meters.

The number of animals censused includes all animals seen

during the hunt. The result for each hunt is a range of

densities according to alternative path width and length

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd