Human Dimensions of The Conflicts Between People and Jaguars (Panthera Onca) in Brazil

HUMAN DIMENSIONS OF THE CONFLICTS BETWEEN
PEOPLE AND JAGUARS (PANTHERA ONCA)

IN BRAZIL
Photo by Arne Sucksdorff
A thesis submitted for the degree of
Doctor of Philosophy
Silvio Marchini
Wildlife Conservation Research Unit, Department of Zoology
Linacre College, University of Oxford
Hilary Term 2010

ABSTRACT
Human dimensions of the conflicts between people and jaguars (Panthera onca) in Brazil
Silvio Marchini Doctor of Philosophy

Linacre College Hilary 2010
The most urgent issue in jaguar conservation is the killing of jaguars by humans. In addition
to turning the killing of jaguars an illegal practice throughout most of the species range,
conservationists have approached jaguar killing within the framework of human-wildlife
conflicts, and concentrated on ways to decrease the economic damage caused by jaguars. We
hypothesized, however, that the ultimate motivation for persecution was not the loss of
livestock to jaguars (or attacks on humans – which are very rare), but rather the perceived
potential for these threats and for their impact on human livelihoods. We also hypothesized
that jaguar killing was not strictly retaliatory, and that the perceptions that determine the
human behaviour of killing jaguars were amenable to change through education and
communication. In the following chapters, we investigated people‟s perceptions of jaguar
impact on human livelihood in Amazonia and the Pantanal within the theoretical framework
of conservation psychology; the development of such perceptions in young people, their
influences on jaguar killing, and how to change them through education and communication.
Our study showed that perceptions behind the conflicts between people and jaguars in Brazil
have more to do with social and psychological factors than previously considered, and
therefore may be more amenable to change through education and communication than
generally believed. As predicted, we found perceptions of jaguar impacts on livestock to be
important elements in conflicts between people and jaguars, but they are not the only
predictors of the intention to kill jaguars. Fear, personal and social motivations, and internal
and external barriers to killing jaguars can also influence jaguar killing behaviour. The
relative importance of these factors in determining intention to kill varies with region and
affluence. Livestock loss is relevant, but it acts in combination with attitudes, knowledge and
perceptions of the economic situation to determine how people perceive the impact jaguars
have on their livelihoods. Attitudes and knowledge, in turn, are influenced by age, gender and
whether respondents live in urban or rural areas. We found that school-based communication
interventions can have a powerful effect on perceptions of jaguars among pupils, and that
pupils can successfully transfer that effect to their fathers. In our communication experiment,
fathers were influenced not only by the information explicitly conveyed in the content of
books and their children‟s homework, but also by the implicit message that a community
institution (e.g., the local school) - and therefore other community members - support jaguar
conservation more than they had realised. If jaguar killing in rural Brazil is to be prevented,
conservationists will need to find and support ways to make jaguar killing not only illegal and
unprofitable, but also socially and personally unacceptable.
ii
DEDICATION
This work is dedicated to
Manoel „Manolo‟ Pacheco (in memoriam),
for his friendship and unconditional support to my work,
and for taking me to the most stunning place on Earth: Amazonia.
iii
ACKNOWLEDGEMENTS
I sincerely thank my supervisor David Macdonald. His cordial advice during planning, field
work, analysis and writing of this thesis was invaluable. It has always been a great honour
working under his supervision. I also thank Sandra Baker for reviewing the chapter drafts and
Paul Johnson for providing statistical advice. Their help was most appreciated. I would like to
thank the other members of the WildCRU who provided support of some sort – intellectual,
logistical or moral – during these last five years: Christos Astaras, Alan de Barros, Dawn
Burnham, Adam Dutton, Merryl Gelling, Lauren Harrington, Jorgelina Marino, Thomas
Merckx, Tom Moorhouse, Greg Rasmussen, Philip Riordan, Diana Roberts, Claudio Sillero,
Lucy Tallents and Alex Zimmermann. Special thanks to Andrew Taber, for supporting my
coming to Oxford, and to Erika Cuellar for sharing with me the bittersweet adventure of being
a South American backcountry person in Oxford.
My studies at the University of Oxford would have not been possible without the generous
financial support of Kevin Duncan, or of the Robertson Big Cat Conservation Fellowship
received from the Wildlife Conservation Society (WCS), and grants from the Iris Darnton
Foundation and the People‟s Trust for Endangered Species. Field work in Amazonia and the
Pantanal was supported by the Whitley Fund for Nature, Anglo American, HSBC Brazil, O
Boticario Foundation, Fauna & Flora International, Chester Zoo, Cleveland Metroparks Zoo,
Woodland Parks Zoo and Floresta Amazonica Hotel. I am also grateful to the following
public schools in Alta Floresta: Mundo Novo, Ouro Verde, Aluízio Azevedo, Rui Barbosa,
Jaime Veríssimo de Campos, Guimarães Rosa, Benjamin Pádua, Geni Silverio, Marinês de Sá
Teixeira and Presbyterian.
I wish to thank Vitória da Riva. Dona Vitória brought me to Alta Floresta and ever since then
has been instrumental in turning the dream of the Escola da Amazônia into reality. Thanks to
all the people at the Cristalino Ecological Foundation (FEC): Márcia de Col Farias, Luiz
Cézar Dias Jorge and Thamiris Mendes for their help with administrative matters, and André
Araújo, Marie Petretto, Viviani Pereira and Queli Ferreira Silva for their assistance in the
field and data processing. I am particularly grateful to Claudio Vicenti and Tiago Henicka for
their assistance, companionship and for showing me the way – sometimes dusty and other
times muddy – to the properties and schools of rural Alta Floresta. I also thank Alexandre
Faria for facilitating my first contacts with ranchers in Amazonia and enthusiastically sharing
iv
with me his seemingly endless knowledge about the region and its wildlife. Special thanks to
Ricardo „Osmosis‟ Luciano, my co-author of People and Jaguars: a Guide for Coexistence.
His talent and passion for the project were much appreciated.
I am grateful to Rafael Hoogesteijn for translating the Guidelines for Coexistence into
Spanish and so enthusiastically supporting its printing; to Howard Quigley, Alan Rabinowitz
and Peter Crawshaw for writing the prefaces for the Spanish and Portuguese versions of the
book, Howard also translated it into English; to Edson Grandisoli, my partner in crime, for
keeping Escola da Amazônia alive while I was busy writing this and for taking that awesome
photo of the jaguar in the Pantanal; to José Márcio Ayres (in memoriam) for introducing me
to the world of Conservation and changing for the better the course of my career; and a
special thanks to Renato Aparecido Farias, my irmão in Alta Floresta, for being there 24/7,
always willing to help. His goodwill and support were simply vital to this project.
Very special thanks to my father Toninho Marchini, my mother Anna Maria and my sister
Maria Angela for their love and endless support. They are the only ones who know the long
and crooked path I took to get here. And very very special thanks to my wife Cris, whose
belief in my ability and confidence that one day I would finish, despite my groans to the
contrary, made this work possible. I am eternally grateful for her love, patience and
understanding.
Finally, I want to express my gratitude to the almost 600 adults and over 1100 children and
adolescents in Amazonia, Pantanal and São Paulo who gave up their time to share with me
their thoughts and feelings about the most fascinating representative of the Brazilian fauna:
the onça-pintada!
v
Contents
CONTENTS
ABSTRACT ii
DEDICATION iii
ACKNOWLEDGEMENTS iv
LIST OF TABLES ix
LIST OF FIGURES xi
CHAPTER 1 AN INTRODUCTION TO THE HUMAN DIMENSIONS OF THE 1

CONFLICTS BETWEEN PEOPLE AND JAGUARS IN BRAZIL
Introduction 1
Conflicts between people and jaguars 3
Conflicts between people and jaguars in Brazil 3
Jaguar impact on people. 4
Human impact on jaguars 6
Mitigating human-jaguar conflicts by changing jaguar behaviour 7
The human dimensions approach to understanding and predicting 9
conflict
Human dimensions of wildlife 9
Human dimensions of the conflicts between people and jaguars 9
The human dimensions approach to mitigating conflicts 12
Understanding, predicting and changing human behaviour 12
Mitigating human-jaguar conflicts by changing human behaviour 15
Study areas 16
Amazonia 18
Pantanal 20
São Paulo city 22
Structure of the thesis 22
CHAPTER 2 PERCEPTIONS OF JAGUARS AMONG CHILDREN AND 25

ADOLESCENTS IN BRAZIL: IMPLICATIONS FOR
CONSERVATION
Introduction 26
Review and hypothesiss 27
Methods 30
Results 36
Experience and knowledge 36
Prominence 39
General perceptions 44
Contents
Concern and attraction 47

Sources of information 48
Discussion and conclusions 50
CHAPTER 3 PERCEPTIONS BEHIND THE CONFLICTS BETWEEN 56

PEOPLE AND JAGUARS IN BRAZIL
Introduction 57
Methods 62
Photograph interpretation experiment 68
Results 71
Perceptions of jaguar impact on human livelihoods 71
The effect of region, land use and place of residence 77
Predicting perceptions of jaguar impact on human livelihood 79
Photograph interpretation experiment 84
CHAPTER 4 FROM PERCEPTIONS TO PERSECUTION: CONFLICTS 91

BETWEEN PEOPLE AND JAGUARS
Introduction 92
Linking attitudes to behaviour: the Theory of Planned Behaviour 93
Extending the Theory of Planned Behaviour 95
Modelling jaguar persecution 97
Methods 99
Results 105
Differences between Amazonia and Pantanal 105
Predicting intention to kill jaguars in Amazonia 106
Predicting intention to kill jaguars in the Pantanal 110
Relationship between intention and behaviour 114
CHAPTER 5 SCHOOL-BASED APPROACHES FOR IMPROVING 124

PERCEPTIONS OF JAGUAR AMONG CHILDREN AND
THEIR FATHERS ON THE AMAZON DEFORESTATION
FRONTIER
Introduction 125
Changing perceptions: information and elaboration 126
Disseminating perceptions: intergenerational learning and modelling 129
Methods 130
Results 137
Information and elaboration 138
Intergenerational learning 142
vii
Contents
CHAPTER 6 GENERAL SUMMARY AND RECOMMENDATIONS FOR 150

FUTURE RESEARCH
Findings 150
Implications 151
Future directions: zooming out for a broader perspective 155
Jaguars versus pumas 155
Conflicts between humans over jaguars 156
From ecological to multidisciplinary to interdisciplinary approaches 157
From ranch to region to range 158
After all, can people and jaguars coexist? 159
LITERATURE CITED 162
APPENDIX I THE JAGUAR IN BOOKS FOR CHILDREN AND 172

ADOLESCENTS IN BRAZIL
APPENDIX II QUESTIONNAIRES AND INTERVIEW SCHEDULE 173
APPENDIX III PHOTOGRAPHS OF DEAD LIVESTOCK 181
APPENDIX IV A GUIDE FOR COEXISTENCE: ACTIVITY BOOKS 183
APPENDIX V ELABORATION EXERCISE 191
APPENDIX VI PEOPLE AND JAGUARS: A GUIDE FOR COEXISTENCE 192
APPENDIX VII JAGUARS, LIVESTOCK AND PEOPLE IN BRAZIL: 204

REALITIES AND PERCEPTIONS BEHIND THE CONFLICT
APPENDIX VIII CONTEMPORARY VIEWS OF HUMAN-CARNIVORE 228

CONFLICTS IN WILD RANGELANDS
APPENDIX IX OTHER COMMUNICATION MATERIALS 252
viii
List of tables
LIST OF TABLES
Table Page
th
1.1 Reported number of people killed by large carnivores in the 20 Century 6
(Loe and Roskaft, 2004)
1.2 Study areas addressed in each data chapter. 17
2.1 Results of analysis of variance of knowledge of, concern about, and 38
attraction towards jaguars by students‟ gender, age1, place of residence
(rural/urban)2 and location3
2.2 Percentage of respondents whose answer to the question “Of all animals, 45
the jaguar is the most…?” revealed their prevalent perception of the jaguar
to be threatening, attractive or endangered.
2.3 Relative importance of different sources of information about jaguars: 48
importance scores and rank (between parenthesis) by location and age
class.
3.1 Summary statistics for each socio-economic variable and results of general 72
linear models, Tukey tests and χ2test comparing variables among sample
groups (p < 0.05; groups indicated by the same letter were not significantly
different).
3.2 Factor loadings and eigenvalues for the two factors extracted using 73
Exploratory Factor Analysis to describe perceptions of jaguar impact on
human livelihood. Variables selected to construct the final scales are shown
in bold.
3.2 Summary statistics for perceptions of jaguar impact on livestock and 75
human safety and results of general linear models and Tukey tests
comparing these variables among sample groups (p < 0.05; groups
indicated by the same letter were not significantly different).
3.4 Summary statistics for experiences, attitudes and knowledge about jaguars, 76
perception of change in jaguar abundance and perception of the economic
situation, and results of general linear models, Tukey tests and χ2test
comparing these variables among sample groups (p < 0.05; groups
3.5 General linear model predicting perception of jaguar impact on livestock in 80
Amazonia. Results shown in bold are significant.
3.6 General linear model predicting perception of jaguar impact on human 81
safety in Amazonia. Results shown in bold are significant.
3.7 General linear models of the effect of socio-economic variables on the 82
factors that significantly affected perceptions of the impact of jaguars on
human livelihoods in Amazonia. Results shown in bold are significant.
3.8 General linear model predicting perception of jaguar impact on livestock in 83
the Pantanal. Results shown in bold are significant.
3.9 General linear model predicting perception of jaguar impact on human 84
safety in the Pantanal. Results shown in bold are significant.
3.10 General linear models of the effect of socio-economic variables on the 85
factors that significantly affected perceptions of the impact of jaguars on
human livelihoods in the Pantanal. Results shown in bold are significant.
4.1 Comparison of variables between Amazonia and Pantanal; means, standard 106
deviations (SD) and GLM results.
ix
List of tables
4.2 Pearson correlations between variables and Cronbach‟s alpha coefficients 107
(shown on main diagonal) for respondents in Amazonia. Results shown in
bold are significant.* p  0.05; ** p ≤ 0.01. NB: A= Cronbach‟s alpha not
computed (single measurement).
4.3 Results of GLM predicting intention to kill jaguars in Amazonia. 108
4.4 Results of GLM examining the effect of perceived impact of jaguars on 109
livestock, and on human safety, and of property size, on the extended TPB
variables that influenced intention to kill jaguars in Amazonia.
4.5 Pearson correlations between variables and Cronbach‟s alpha coefficients 111
(shown on main diagonal) for respondents in the Pantanal. Results shown
in bold are significant.* p  0.05; ** p ≤ 0.01. NB: A= Cronbach‟s alpha
not computed (single measurement).
4.6 Results of GLM predicting intention to kill jaguars in the Pantanal. 112
4.7 Results of GLM examining the effect of perceived impact of jaguars on 113
livestock, and on human safety, and of property size, on the extended TPB
variables that influenced intention to kill jaguars in the Pantanal.
4.8 Proportion of Amazonia respondent who had and had not killed jaguars in 115
the previous five years, who did and did not intend to kill them in the
future.
4.9 Proportion of Pantanal landowners who had versus had not killed jaguars in 115
the previous five years, and who did versus did not intend to kill them in
the future.
5.1 Mean scores  standard deviations for, and paired t-tests between, pre-test, 139
post-test and delayed post-test on pupils‟ perceptions of jaguars.
5.2 Mean scores  standard deviation for, and paired t-tests between, pre-test 144
and post-test on fathers‟ perceptions of jaguars.
x
List of figures
LIST OF FIGURES
Figure Page
1.1 Historical and current jaguar distribution range 7
1.2 Map of Brazil showing (a) biomes, (b) expected areas of conflict between 16
people and jaguars over livestock based on jaguar occurrence and cattle
density (Silveira et al. 2008), and the three study areas: 1) southern
Amazonia; 2) northern Pantanal; and 3) São Paulo city.
1.3 Maps of the Brazilian Amazon showing the study area (red ellipses), (a) 18
deforested areas (grey), and (b) the „Arc of Deforestation‟ and ecoregions.
2.1 Percentage of respondents reporting a jaguar sighting (shown white), family 37
livestock depredation by jaguars (rural students only) (shown black), and
acquaintance with a victim of jaguar attack (shown grey).
2.2 Mean knowledge scores by age (● 10-11, ● 12-13, ● 14-15, ● 16-17, ● 18- 38
19) and location. Bars represent the 95% confidence interval for mean.
2.3 Cognitive prominence (black bars) (percentage of respondents who 40
mentioned various taxa first when asked “cite three animals that you know
to occur in the nearest forest”) and associated sighting rate (white bars)
(percentage of respondents who reported ever seeing the taxon in the wild)
in a) Amazonia (n=862), b) Pantanal (n=147) and c) Sao Paulo city (n=118).
2.4 Affective prominence (percentage of respondents in a) Amazonia (n=882), 41
b) Pantanal (n=147) and c) Sao Paulo city (n=118) who cited the taxon when
asked “What animal do you like most?” (white bars) and “What animal do
you dislike most?” (black bars). Sorting based on the summated citations for
like and dislike.
2.5 a) Mean cognitive prominence of jaguars by age class (● 10-11, ● 12-13, ● 43
14-15, ● 16-17, ● 18-19) and location, and b) mean negative affective
prominence (dislike) by location. Bars represent 95% confidence intervals of
the mean.
2.6 Most prevalent perception of jaguars (threatening, attractive, endangered or 45
other) by location and place of residence.
2.7 Mean standardized scores for a) concern, and b) attraction, by gender (○ 46
female, ■ male) and location. Bars represent the 95% confidence interval for
mean.
3.1 Perception of jaguar impact on livestock: difference between neighbouring 74
ranches and own ranch (positive values mean higher impact on neighbouring
ranches and zero means no difference) and property sizes in Amazonia and
Pantanal.
3.2 Relationship between perception of jaguar impact on livestock and attitudes 80
to jaguars at different levels of knowledge (scores 1 and 2: R2 = .460 and
scores 3 and 4: R2 = .198) among cattle ranchers in Amazonia. The
relationship was not significant for knowledge scores 5 to 7.
4.1 Diagrammatical representation of the Theory of Planned Behaviour. 95
4.2 Final model showing intention to kill jaguars in the Amazonia deforestation 110
frontier.
4.3 Final model showing intention to kill jaguars in the Pantanal. 114
5.1 Diagram showing assignment of pupils and fathers to experimental 132
treatments (and sample sizes). Dashed lines indicate that fathers were not
xi
List of figures
aware of the involvement of their children‟s school in the study.

5.2 Variation in pupils‟ perceptions of jaguars between pre-test, post-test and 140
delayed post-test in response to the following treatments: control (◊),
information (●), elaboration (▲), and information plus elaboration (■). *
denotes p < 0.05, ** denotes p <0.01.
5.3 Variation in attitudes towards jaguars and jaguar killing between pre-test 143
and post-test among pupils exposed to elaboration alone (▲ and Δ) and
elaboration and information combined (■ and □). Solid shapes denote
positive initial attitudes while hollow shapes denote neutral and negative
initial attitudes. * denotes p < 0.05, ** denotes p <0.01.
5.4 Variation in fathers‟ perceptions of jaguars between pre-test and post-test in 145
response to the following treatments: control (◊), book via conservation
organization (○), book via child‟s school (●), child‟s homework (▲), and
book via child‟s school plus child‟s homework (■). * denotes p < 0.05, **
denotes p <0.01
xii
Chapter 1 - Introduction
CHAPTER 1
AN INTRODUCTION TO THE HUMAN DIMENSIONS OF

CONFLICTS BETWEEN PEOPLE AND JAGUARS IN BRAZIL
To start with, I had to know something about the people, the country and the trees. And of the
three, the first was the most important.
Gifford Pinchot
Introduction
The jaguar (Panthera onca) is one of the most charismatic species of conservation concern in
Central and South America. From the ancient rituals of indigenous communities (Saunders
1998), to classic literature (Cunha 1954; Lobato 1933; Rosa 1968) and painting (e.g., Debret,
Florence, Rugendas, Vanderbuch), to tourism marketing images, and the current Brazilian
fifty-real bill, the jaguar has been - more than any other species in the continent - celebrated
as a symbol of power and beauty. Nonetheless, jaguars are not always welcome in rural areas
and their presence can be intolerable for many people. Ironically, this intolerance is rooted in
the similarity between jaguars and us, human beings. Jaguars and humans are approximately
the same size, we both eat meat, and, therefore, we contend for the same prey species (i.e.
native and domestic ungulates). The most common complaint about sharing the land with
jaguars is that they feed on what should be food exclusively for people: domestic cattle.
Another similarity between jaguars and humans is that both are formidable predators. Jaguars
are admired for their hunting skills, but feared for their ability to defend themselves or attack
humans, which very rarely happens. In reality, it is we, humans, who persecute and kill the
jaguars and, as a result, jaguars are disappearing.
1
The killing of jaguars by ranchers is indeed a major threat to the species (Nowell and Jackson
1996; Zeller 2007). Because jaguars are renowned for preying on livestock (Hoogesteijn
2000), one might assume that ranchers kill jaguars in an effort to reduce economic losses
associated with livestock predation. Conservationists have approached jaguar killing within
the usual framework of human-wildlife conflicts (Rabinowitz 2005), in which conflict is
defined as the situation when the behaviour of a (otherwise non-pest) wild animal species
poses a direct and recurring (real or perceived) threat to the livelihood and/or safety of a
person or a community and, in response, persecution of the species ensues (Macdonald et al.
2009). To date, research and conservation efforts have focused on the ecological (Azevedo
and Murray 2007; Cavalcanti 2008) and economic (Silveira et al. 2006) dimensions of the
conflict respectively. Little effort has been made to understand the link between jaguars
killing livestock and people killing jaguars. The killing of jaguars may not be strictly
retaliatory (i.e. in direct response to the negative impact caused by jaguars) and might be
motivated other than by economics. If we are to curb the killing of jaguars, first we have to
understand the underlying causes of this behaviour.
In this thesis, I use a broader approach than the usual human-wildlife conflict framework to
examine why people kill jaguars in Brazil. I address the human perceptions behind conflicts
among jaguars, livestock, and cattle ranchers. I adapt techniques from the social sciences to
examine how perceptions about jaguars develop in children and adolescents, and the factors
that influence ranchers‟ perceptions of jaguar impact on livestock and human safety. I
investigate how these perceptions translate into the killing of jaguars and how education and
communication can be used to improve perceptions of jaguars among farmers and cattle
ranchers, and consequently reduce jaguar killing. Finally, I discuss how information on the
ecological, economic, social and cultural dimensions of a human-carnivore conflict can be
2
integrated fruitfully into a strategy spanning from individuals to populations (of both jaguars
and humans), in an attempt to promote coexistence between jaguars, livestock and people.
Conflicts between people and jaguars
Around one third (36%) of the jaguar‟s global distribution overlaps protected areas (A.
Zimmermann, pers. comm.), but very few of these areas offer true protection for jaguars and
their prey. Indeed the edges of protected areas often become hotspots for human-wildlife
conflict (Woodroffe and Ginsberg 1998). Outside of protected areas, jaguars most often
occupy livestock ranches (Zeller 2007). Conflicts between ranchers and jaguars over
livestock are widespread and have been documented throughout the jaguar‟s range, (e.g.:
Belize: Rabinowitz 1986; Brazil: Crawshaw and Quigley 1991; Dalponte 2002; Conforti and
Azevedo 2003; Michalski et al. 2006; Azevedo and Murray 2007; Palmeira et al. 2008; Costa
Rica: Saenz and Carrillo 2002; Argentina: Schiaffino et al. 2002; Venezuela: Polisar et al.
2003; Scognamillo et al. 2002). Ecological, socioeconomic, cultural, and historical aspects of
the relationships between people and jaguars in Brazil render the country particularly
important for jaguar research and conservation.
Conflicts between people and jaguars in Brazil
Brazil covers 40% of the land area of Latin America. Jaguar abundance estimates are as
scarce for Brazil (Almeida 1986; Quigley and Crawshaw 1992; Soisalo and Cavalcanti 2006)
as for other parts of their range (cf Wallace et al. 2003, Maffei et al. 2004, Silver et al. 2004).
However Brazil encompasses the world‟s two largest jaguar population strongholds
(Sanderson et al. 2002) - the wetlands of the Pantanal (140,000 km2) and the rainforests of
Amazonia (3,400,000 km2) - and is thought to contain more than half of the world‟s c. 30,000
jaguars (Medellin 2009). The southern Pantanal of Brazil has the highest recorded density of
3
jaguars (estimates range from 6.7 to 11.7 individuals/100 km2) (Soisalo and Cavalcanti
2006). The Pantanal is also home to the largest jaguars on record, with males weighing an
average of 100 kg (females are typically 10-20% smaller) and the largest individuals reaching
158 kg (Hoogesteijn and Mondolfi 1992).
Brazil is also home to the world‟s largest commercial cattle herd (>200 million head) and is
the world leader in beef exports (Nepstad et al. 2006). Cattle ranching threatens jaguars
indirectly, insofar as it is the major driver for deforestation in Amazonia, this being the
primary cause of >66% of the habitat loss in the region (Nepstad et al. 2006). Between 1987
and 2006 an average of 18,000 km2 of prime jaguar habitat was lost in this region every year,
mostly from the Amazonian agricultural frontier (PRODES 2007). In the last two decades,
Brazil has lost larger areas of jaguar habitat than any other country.
For ecological and historical reasons, there is overlap between areas where beef production
flourishes and jaguars survive, namely the Pantanal and the agricultural frontier of southern
Amazonia (Thornton et al. 2002). Cattle ranchers have never rejoiced in this overlap, and
have a long tradition of killing jaguars (Hoogesteijn and Mondolfi 1992). In essence, jaguars
kill livestock, and ranchers kill jaguars in an attempt to prevent further losses. But overall,
what is the magnitude of the impact caused by jaguars on livestock and human livelihoods in
general?
Jaguar impact on people
Impact on livestock
Although livestock depredation on small ranches, stocked with small herds of cattle, can be
ruinous (Hoogesteijn et al. 1993), livestock losses to jaguars are generally small when
4
averaged over time and space. Average losses attributed to jaguars in Brazil range from 0.2 -
2.3% of livestock holdings in the Cerrado (Palmeira 2004), the Atlantic Forest (Conforti and
Azevedo 2003), southern Amazonia (Michalski et al. 2006), southern Pantanal (Azevedo and
Murray 2007), two ranches in northern Pantanal (Dalponte 2002), and a larger portion of the
northern Pantanal (Zimmermann et al. 2005). Sometimes, a major factor predisposing a
particular herd to jaguar predation is poor husbandry (Azevedo and Murray 2007; Michalski
et al. 2006). While the risk of predation is greater among cattle left unattended close to forest
cover (Azevedo and Murray 2007), documented losses of cattle to jaguar predation are
generally much fewer than those attributable to accident, snake bite, disease, parturition
problems, flood (Azevedo and Murray 2007) and even theft (R. Hoogesteijn, pers. comm.).
Furthermore, on some ranches, livestock depredation by puma (Puma concolor) may be more
common than predation by jaguars (Polisar et al. 2003), but it may be difficult for ranchers to
distinguish predation by jaguars and pumas.
Impact on personal safety
In contrast to the historical and contemporary record of unprovoked attacks on humans by
lions (Panthera leo) (Patterson 1907; Yamasaki et al. 1999; Saberwal et al. 1994), tigers
(Panthera tigris) (McDougal et al. 2001), leopards (Panthera pardus) (McDougal 1989),
pumas (Beier 1991; Deurbrouck and Miller 2001), wolves (Canis lupus) (Linnell et al. 2002;
McNay 2002), and bears (Ursus sp.) (Herrero and Higgins 1995) (see Table 1.1), attacks by
jaguars have been almost invariably associated with the cornering or injury of hunted jaguars
(Marchini and Luciano 2009a), or the defence of cubs or carcasses. In Brazil, the only
documented, unprovoked, fatal attack by a jaguar on a human occurred on 24 June 2008,
when a young fisherman was killed by a jaguar while sleeping in his tent on a bank of the
Paraguay River in the Pantanal (Marchini and Luciano 2009a). One speculation has been that
5
this unique incident was a result of jaguars in that area becoming habituated to people around
baits used to attract jaguars to be photographed by tourists.
Table 1.1. Reported number of people killed by large carnivores in the 20th Century (Loe and
Roskaft 2004)
Number of humans
Species
killed
Tiger (Panthera tigris) 12,599
Leopard (Panthera pardus) 840
Wolf (Canis lupus) 607
Lion (Panthera leo) 552
Brown bear (Ursus arctos) 313
Sloth bear (Melursus ursinus) 48
Black bear (Ursus americanus) 37
Puma (Puma concolor) 18
Human impact on jaguars
Irrespective of the magnitude of damage or threat posed by jaguars, they have been severely
persecuted; indiscriminate killing of jaguars is one of the most serious threats to their survival
across Latin America (Zeller 2007). Persecution is the major cause of jaguar mortality outside
of protected areas (Nowell and Jackson 1996) and, because they range widely, persecution
threatens jaguars within protected areas as well (Woodroffe and Ginsberg 1998). Different
methods and approaches are used to kill jaguars in Brazil. One such method involves using
dogs to find and pursue the jaguar; either the jaguar climbs a tree or turns at bay on the
ground, whereupon the hunters arrive and kill it. Alternatively, the hunters wait at night on a
rudimentary tree platform, set up a few metres above a half-eaten carcass or other dead or
live bait, and then shoot the jaguar on sight. Poison baits are sometimes used. Landowners
kill jaguars alone, or together with other landowners or a hired professional hunter.
6
Among the ranchers who face livestock predation problems, the most popular jaguar
„management‟ method remains the „triple S solution‟: shoot, shovel and shut up (although
shovelling is often omitted). Together with habitat loss, persecution has reduced jaguars to
46% of their historical range (Sanderson et al. 2002) (Figure 1.1). Up to around 1500, when
European colonization started, jaguars were widely distributed throughout Brazil, but now
they are extirpated from entire regions (Sanderson et al. 2002). Some jaguars remain in
fragments of the Atlantic forest and the Cerrado, but large jaguar populations are present only
in Amazonia and the Pantanal, where human population density has historically been low.
The jaguar is currently considered as near threatened in Brazil (Caso et al 2008).
Figure 1.1. Historical and current jaguar distribution range
Mitigating human-jaguar conflicts by changing jaguar behaviour
Jaguar impacts on people result ultimately from jaguars killing livestock (and potentially
killing people). Several measures have been recommended or implemented that aim at
7
changing – directly or indirectly – the behaviour of jaguars to prevent them killing livestock.
The use of barriers such as electric fences, guards, acoustic (explosives) and visual (lights at
night) repellents, translocation of problematic animals, and changes in husbandry practices
(e.g. keeping calves away from forest edges) are examples of such interventions (Hoogesteijn
2000). The effectiveness of these interventions has not been systematically evaluated and,
meanwhile, lethal control, although illegal, remains the main management practice adopted
by the ranchers in Brazil.
Preventative measures are aimed at decreasing the economic burden on ranchers who coexist
with jaguars, on the assumption this will reduce their motivation to kill them. However,
human persecution of jaguars may be related less to livestock predation than was previously
believed, and the economic justification for killing jaguars may be equally unclear. The
ultimate motivation for retaliatory persecution may not be the actual impact of jaguars on
human safety or livestock, but rather the cultural and social perceptions of the potential threat
that jaguars pose to them. In conflicts between people and carnivores, the perceived impacts
often exceed the actual evidence (Conover 2001; Chavez and Gese 2005, 2006; Chavez et al.
2005; Sillero-Zubiri et al. 2007). In addition, factors not related directly to jaguar impacts on
human livelihoods (e.g. perceived social status of jaguar hunters in the community and the
thrill of the chase) may also be involved in the persecution of jaguars, making it difficult to
evaluate the economic rationale for killing them. If we are to understand, predict and resolve
the conflicts between people and jaguars, we also need to consider the human side of the
conflict.
8
The human dimensions approach to understanding and predicting conflicts
Human dimensions of wildlife
Human Dimensions has emerged as a sub-discipline of wildlife management over the last two
decades (Decker and Purdy 1988). This area of interest developed among wildlife
professionals who were concerned with how people's values affect and are affected by
decisions about the management of wildlife populations, habitats, and people's use of these
resources (Purdy and Decker 1989). Typically, human dimensions research and management
have focused on values, beliefs, norms, attitudes and behaviours, as well as socioeconomic
and demographic characteristics of wildlife user-groups, with emphasis placed on
incorporating information on such factors into wildlife management decisions.
Human dimensions of the conflicts between people and jaguars
Jaguars are of outstanding significance to people in rural Brazil. Felines in general have had a
profound effect on human sensibilities since the beginning of time. Throughout history, as
described by Saunders (1994), jaguars have inspired fear, respect and emulation as the
embodiment of supernatural power. In the Americas the jaguar was invoked in art and
religion of Amerindians, from Amazonia and the Andes, to Meso-America and North
America. Depicted in gold, pottery and stone, and conjured up in shamanistic visions, they
were associated with sacrifice, cannibalism and war, and employed in the subtle symbolism
of metaphor as icons of power, aggressiveness, ambition and prestige. In Brazil, the 12,000-
year old rock painting of Serra da Capivara, depicting a jaguar, suggests that people and
jaguars have had a close relationship since ancient times. The jaguar is still a central figure of
the mythology of many peoples, including several Brazilian tribes (e.g. Matis), in which role
they are at the same time feared and admired.
9
The first Europeans in Brazil may have been profoundly impressed by the jaguar. After all,
Europeans were afraid of large cats such as lions and leopards (Miranda 2003). Here, the
powerful yaguar of the Tupis would be called „onça‟, a word of European origin. The word
„onça‟ comes from the Greek lugkos, which became lynx in Latin, and lonce in Provençal. In
the 15th Century the „l‟ would have been regarded as an article and therefore suppressed,
giving rise to once; in Spanish the term was already onza by 1495, and the same origin is
attributed to the Portuguese expression onça (Miranda 2003) (the English, on the other hand,
would keep the indigenous term jaguar).
The very first account of the relationship between people and onças in Brazil was left by the
French Jesuit Jean de Léry in 1557. Léry reports that “whenever it (the jaguar) can, it catches
an Indian, kills him, smashes and devours him”. The Portuguese priests José de Anchieta and
Pero Magalhães de Gândavo were the first authors to report predation problems caused by
jaguars in Brazil. In his „History of the Province of Santa Cruz’, published in 1575, Gândavo
provides detailed descriptions of a jaguar attack on cattle and on Indians.
Since then, jaguars have been mentioned frequently in a variety of forms, from explorers‟
accounts and classic literature, to arts and folklore. It seems unlikely that any other species
has been depicted more often in hunting situations than jaguars, with examples ranging from
the paintings of Johann Moritz Rugendas and Jean-Baptiste Debret in the 19th Century, to the
hunter‟s accounts of the 20th Century (Roosevelt 1914, Cunha 1918, Siemel 1953, Almeida
1976) to contemporary movies and soap-operas (e.g. “Amazônia” by Globo Network). In
children‟s literature the jaguar has been almost invariably depicted as the villain that is killed,
or otherwise dies, at the end of the story (e.g. Lobato 1933). In folklore, sayings involving the
word „onça‟ have a rather negative connotation such as amigo da onça (false friend). On the
10
other hand, jaguars are currently used as an icon by the Brazilian ecotourism industry, and are
depicted more often in marketing materials than any other mammal species in the Pantanal
and Amazonia (unpublished data). Jaguars are also used as mascots by several Brazilian zoos,
and are receiving growing attention from conservation-oriented popular media including the
internet, magazines and TV documentaries.
Despite the prominence of the jaguar in human life, studies on local perceptions of jaguars
had not been conducted in Brazil until very recently. The first such study was conducted by
Conforti and Azevedo (2003), who evaluated 72 livestock ranchers around Iguaçu National
Park in southern Brazil. They found that perceptions towards jaguars were not influenced by
the history of predation on the properties surveyed. Likewise, among the 50 landowners
interviewed in northern Pantanal by Zimmermann et al. (2005), attitudes towards jaguars
were more closely related to the age and relative wealth of respondents than to cattle losses,
with younger and wealthier ranchers holding more positive attitudes towards jaguars. In
southern Amazonia, over half of the 62 ranchers interviewed by Michalski et al. (2006) were
interested in implementing non-lethal methods to control predation levels on their ranches,
but about one third explicitly were not. Palmeira and Barrela (2007) evaluated 28
householders in a small, traditional, rural community in southeastern Brazil, where most
respondents held negative attitudes towards jaguars and wished the species could be
eliminated. In contrast, the majority of the 1,007 people interviewed by Santos et al. (2008) in
the five biomes of Brazil (i.e. Amazonia, Pantanal, Atlantic Forest, Cerrado and Caatinga)
were in favour of jaguar conservation. Altogether these studies revealed strong and
contradictory attitudes towards jaguars and, together with our previous results (Cavalcanti et
al. 2010), suggest that the perceived impact of jaguars on human livelihoods may often
exceed the evidence.
11
Except for the work by Santos et al. (2008), each of the foregoing studies was conducted at a
single site and evaluated only people who were directly involved in conflict with jaguars;
negative experience with jaguars – specifically, livestock predation – was addressed as the
predictor of conflict. Attitudes towards jaguars and / or jaguar management have been
assessed, but only Zimmermann et al. (2005) have examined the potential predictors of
attitudes, and nobody has reported on any relationship between attitudes and people‟s
tendency to kill jaguars. Knowledge about jaguars was assessed by Conforti and Azevedo
(2003), but the influence of knowledge on people‟s perceptions about jaguars has not been
examined.
The human dimensions approach to mitigating conflicts
As highlighted by Manfredo (2008), the common thread in the wide range of circumstances,
and species, that spawn human-wildlife conflicts is the fact that the course and resolution of
the conflict is ultimately determined by the thoughts and actions of humans. Because humans
are the constant in human-wildlife conflicts, approaches that aim to understand human
behaviour would best address this issue.
Understanding, predicting and changing human behaviour
Anthropology, sociology, psychology and their applications to communication, education and
marketing, form the basis of many theories used to understand and influence certain
conservation behaviours (Gardner and Stern 2002; Jacobson et al. 2006). While a significant
progress has recently been made in our understanding of the relationships between people
and nature in general, and the conflicts between people and large carnivores in particular,
from an anthropological perspective (Borgerhoff-Mulder and Coppolillo 2005; Dickman
2008; Hazzah et al. 2009), this study adopts a theoretical structure that is based on social and
12
cognitive psychology. The goals of social and cognitive psychology may have more in
common with those of conservation than is the case for branches of social science
preoccupied only with understanding human behaviour, in so far as both psychologists and
conservationists are interested in the practicalities of bringing about changes in human
behaviour. This study drawns upon the following theories: the Theory of Planned Behaviour
(Ajzen 1985), the Elaboration Likelihood Model of Persuasion (Petty and Cacioppo 1981),
and the Social Learning Theory (Bandura 1977). These theories are described briefly below.
Theory of Planned Behaviour
The Theory of Planned Behaviour (TPB) proposes that human behaviours are governed not
only by personal attitudes, but also by social pressures and perceived control over one‟s own
behaviour. According to the TPB, the most proximal determinant of a person‟s behaviour is
their intention to engage in that behaviour. In turn, behavioural intentions are influenced by
three main factors: attitudes towards the behaviour, subjective norms and perceived
behavioural control (PBC). Attitude scores reflect an individual‟s overall positive or negative
evaluation of performing the behaviour. Subjective norms represent an individual‟s
perception of whether people important to them would approve of them performing the
behaviour. Perceived behavioural control reflects the extent to which an individual perceives
the behaviour to be under their volitional control. Thus, according to TPB, people who have
positive attitudes towards killing jaguars, think that there is normative support for killing
jaguars, and perceive that they can easily kill jaguars (or pay someone else to do it for them),
should have strong intentions to kill jaguars. In addition, to the extent that PBC is a proxy for
the actual control (accurately accounts for both the internal factors [e.g. knowledge, skills,
courage] and external factors [e.g. legal barriers, money, equipment, help from others])
needed to perform the behaviour, it may also have a direct impact on behaviour.
13
Elaboration Likelihood Model of Persuasion
The Elaboration Likelihood Model of Persuasion (ELM) proposes that attitude change
depends upon the amount of thoughtful consideration (cognitive elaboration) that occurs in
response to a persuasive communication. According to the ELM, there are two types of
attitude change: 1) central route; and 2) peripheral route attitude change. These two routes of
attitude change differ in the amount of elaboration involved. Central route attitude change can
occur when there is significant cognitive elaboration (high elaboration likelihood), whereas
peripheral route attitude change can occur in the absence of cognitive elaboration (low
elaboration likelihood). Central route attitude change requires the motivation, and the ability,
to process cognitively a persuasive communication. For example, a person‟s attitudes are
more likely to change by the central route if they are presented with new information on a
subject they find stimulating, and this is done in a thought-provoking way. Attitudes are more
likely to change by the peripheral route if new information is presented on a subject they find
uninspiring, and without their active involvement. Central route attitude change is more
difficult to achieve deliberately than peripheral route change, but when it does occur, it lasts
longer, influences behaviour more, and is more resistant to counter-persuasion than
peripheral route change.
Social Learning Theory
People can learn from observing someone and following their lead, not just from verbal
interchange. The Social Learning Theory is the foundation for modelling, also called
observational learning. It recognizes that humans live in social environments, with other
people, and the relationships we have with these people are sources of learning and change.
14
The TPB, ELM, Social Learning and other theories of behavioural change [e.g. Motivational
Theory (Maslow 1954), Stages of Change (Prochaska et al. 1992), Diffusion of Innovation
(Rogers 1995)] have provided the theoretical framework for the four basic methods of
promoting conservation-oriented behaviours (Gardner and Stern 2002): 1) government laws,
regulations and incentives; 2) small group / community management; 3) education; and 4)
moral, religious and / or ethical appeals.
Mitigating human-jaguar conflicts by changing human behaviour
Efforts to deter jaguar killing among ranchers and farmers have focused largely on economic
incentives, e.g. monetary compensation for livestock loss (Silveira et al. 2006) and legal
prohibitions and sanctions, e.g. making jaguar killing illegal. Although compensation
programs have been implemented worldwide (Naughton-Treves et al. 2003; Swenson and
Andrén 2005), there is growing scepticism among conservationists about their effectiveness
and sustainability (Nyhus et al. 2003, 2005). As for legal approaches to deter jaguar killing,
in 1967, the Brazilian Wildlife Protection Act prohibited commerce in wildlife and products
derived from their capture, pursuit or destruction. The Convention on International Trade in
Endangered Species (CITES) of 1973 made it illegal to conduct international trade in jaguar
skins or parts for commercial gain. Commercial, sport, and recreational hunting are
prohibited in Brazil. In 1998 the killing of jaguars was rendered a crime under Brazil‟s
Environmental Crimes Act (Lei de Crimes Ambientais - 9.605/98). The CITES listing, in
combination with the Brazilian legislation and anti-fur campaigns, brought about a sharp
decline in the fur trade, helping to reduce the pressure on wild jaguar populations. However,
law enforcement has been relatively ineffective in remote areas of rural Brazil, such as the
Amazon agricultural frontier and the Pantanal. Jaguar persecution continues (Crawshaw
15
2002; Michalski et al. 2006) – now very rarely for (illegal) trade, but because of their
perceived threat to people and their livelihoods and also for social and personal reasons.
This study proposes a model to understand and predict the human behaviour of killing jaguars
and examines the potential of an integrative intervention - combining elements of education,
community management and incentives – to deter such behaviour.
The thesis
Study areas
Because this study examines the social factors in conflicts between people and jaguars, over
livestock, the work took place in two regions of Brazil that differ significantly in social
conditions, but with the common feature of having large populations of both livestock and
jaguars. These regions were the deforestation frontier in southern Amazonia, more
specifically in the districts of Alta Floresta and Novo Mundo, and the northern Pantanal, in
the districts of Cáceres and Poconé, both in the State of Mato Grosso (Figure 1.2.a). Indeed,
the two regions have been identified as hotspots of jaguar-rancher conflict (Silveira et al.
2008) (Figure 1.2.b). The deforestation frontier in southern Amazonia was colonised recently
- from about 35 years ago - by people from different parts of the country, including parts of
Brazil where jaguars were no longer present. In contrast, Northern Pantanal is home to a few
traditional families that have raised cattle in the region for generations. The perspective of
people who do not share the space with jaguars was also addressed. In order to investigate
perceptions regarding jaguars on a broad social spectrum, wealthier pupils of the largest
metropolitan area in the continent – São Paulo city – were evaluated to represent a social
sector that is both geographically and culturally distant from conflicts with jaguars (Figure
16
1.2). The Amazon deforestation frontier is in the subject of all data chapters in this thesis,
while the Pantanal is the subject of three, and São Paulo city, one (Table 1.2).
(a) (b)
Figure 1.2. Map of Brazil showing (a) biomes, (b) expected areas of conflict between people
and jaguars over livestock based on jaguar occurrence and cattle density (Silveira et al.
2008), and the three study areas: 1) southern Amazonia; 2) northern Pantanal; and 3) São
Paulo city.
Below is a brief description of the physical, ecological, historical, economic, social and
cultural aspects of the Amazon deforestation frontier, northern Pantanal and São Paulo city.
A summary of relationships between people and jaguars in each region – conflicts, research
and conservation – is also presented.
17
Table 1.2. Study areas addressed in each data chapter.
Thesis chapter/study area Southern Northern São

Amazonia Pantanal Paulo
city
Chapter 2. Perceptions of jaguars among children X X X
and adolescents in Brazil
Chapter 3. Perceptions behind the conflicts between X X
people and jaguars in Amazonia and
Pantanal
Chapter 4. Conflicts between people and jaguars: X X
from perceptions to persecution
Chapter 5. School-based approaches for improving X
perceptions of jaguars among children and
their fathers on the Amazonia
Amazonia
The Amazon Rainforest is a moist broadleaf forest in the Amazon Basin of South America.
The area, also known as Amazonia or the Amazon Basin, encompasses seven million square
kilometres (1.2 billion acres), though the forest itself occupies some 5.5 million square
kilometres, located within nine nations: Brazil (with 60 % of the rainforest), Colombia, Peru,
Venezuela, Ecuador, Bolivia, Guyana, Suriname and French Guiana. The Amazon contains
over half of the planet's remaining rainforests and comprises the largest and most species-rich
tract of tropical rainforest in the world. The Brazilian Amazon was subjected to massive rates
of deforestation during the 1970s and 1980s, following the implementation of several large,
central-government sponsored agricultural resettlement programmes, subsidised by generous
fiscal incentives for large-scale cattle ranching (Fearnside 1993; Nepstad 2006). In 2004, over
27 thousand square kilometres of forest were destroyed. Rates of deforestation have dropped
since then (PRODES 2007). In 2009, the Brazilian Amazon lost approximately 7 thousand
square kilometres of forest, most of this in southern Amazonia, within the so-called „Arc of
Deforestation‟ (Figure 1.3).
18
This study was conducted in Alta Floresta, a frontier town in Southern Amazonia (lat -
9.8779, lon -56.0975). Alta Floresta, founded as recently as 1976, has approximately 51
thousand inhabitants and hosts the fourth largest herd of domestic cattle in the State of Mato
Grosso (approximately 700,000 head) (Brazilian Institute of Geography and Statistics 2007)..
The majority of the ranchers of Alta Floresta came from the southern state of Paraná, where
they had little contact with wildlife and where jaguars were rare or absent.
Figure 1.3. Maps of the Brazilian Amazon showing the study area (ellipses), (a) deforested
areas (grey), and (b) the „Arc of Deforestation‟ and ecoregions.
The forests in Alta Floresta are extremely important from a conservation perspective.
Cristalino State Park (CSP), located in that area, has the greatest levels of biodiversity among
all of the protected areas in Brazil. It is home to 515 bird species, of which 50 are endemic to
the region; 70 mammal species; 43 reptile species; 29 amphibian species and 16 species of
commercial or sport fish. Because of the diversity of habitats in the park („terra firme‟ forest,
seasonally dry forest, flooded forest, swamps, rivers and rocky outcrops), plant diversity is
also exceptionally high. The region‟s exceptional biodiversity is a result of its position in the
transition zone between the Amazon forest and the „cerrado‟ savannas of central Brazil. CSP
19
also lies between two major tributaries of the Amazon River, which acts for some species as a
barrier to dispersal, thus explaining the concentration of endemic species in the area.
In 1999 at a workshop on Brazilian conservation priorities, the Alta Foresta area was placed
within the highest priority conservation category because of its high levels of endemism and
rich diversity of vegetation and habitats, as well as the great urgency with which the area
needs protection, given the sheer pressure from smallholders, logging interests and large
cattle ranchers (Capobianco et al. 2001). Specific urgent interventions identified for the Alta
Floresta area were anthropological studies, creation of protected areas, environmental
education and biological surveys addressing aquatic systems, reptiles, amphibians and
mammals (Capobianco et al. 2001).
Cattle depredation by jaguars is considered a severe problem in Alta Floresta (Michalski et al.
2006), and persecution and habitat loss are major threats to jaguars there (Michalski and
Peres 2005). According to a range of formal and anecdotal accounts of jaguar and puma
mortality obtained over the year 2003–2004, an estimate of 110–150 jaguars and pumas are
killed annually in Alta Floresta through direct persecution by professional hunters or ranch
staff, or by poisoned carcasses (Michalski et al. 2006).
Pantanal
The Pantanal is the world‟s largest wetland area, a flat landscape, with gently sloping and
meandering rivers. The region, whose name derives from the Portuguese word „pântano‟
(meaning „swamp‟ or „marsh‟ ), is situated mainly within the Brazilian states of Mato Grosso
and Mato Grosso do Sul. Small parts of the Pantanal are in Bolivia and Paraguay. In total, the
Pantanal covers about 150,000 square kilometres (58,000 sq mi). The Pantanal floods during
20
the wet season, when over 80% of the area may be submerged, and the world's richest
collection of aquatic plants is nurtured in the process.
The ecosystem of the Pantanal is under threat from human activities, including uncontrolled
recreational fishing, the hunting and smuggling of endangered species (caiman, parrots and
macaws), uncontrolled tourism, and deforestation and forest fires for agricultural use in
neighbouring areas (Swarts 2000). In recent rainy seasons, flooding has been extreme and
many cultivated areas surrounding the park have been inundated. Receding flood waters have
carried large amounts of pesticides into the rivers and lakes, killing a great number of fish.
Pressure for economic development (such as oil pipelines and shipping canals) is of particular
concern. A plan to dredge the Paraguay and Parana Rivers to allow ocean-going ships to
travel far inland could have serious consequences for the ecosystem by affecting the flooding
and drainage cycles (Swarts 2000).
This study was conducted in the Pantanal districts of Poconé (location of town of Poconé: lat
-16.2575, lon -56.6234) and Cáceres (lat -16.0588, lon -57.7034). Cattle ranching is the main
economic activity in Poconé and Cáceres and conflicts between cattle ranchers and jaguars
are a major issue (Dalponte 2002). Cáceres and Poconé have approximately 87,000 and
32,000 inhabitants respectively, and cattle ranching is their main economic activity (Brazilian
Institute of Geography and Statistics 2009). As with the Amazon frontier, northern Pantanal
rears large cattle herds (around 832,000 and 347,000 head in Cáceres and Poconé,
respectively) (Brazilian Institute of Geography and Statistics 2007) and livestock depredation
by jaguars is a major complaint among ranchers (Marchini 2003; Zimmermann et al. 2005).
21
The southern Pantanal has the greatest density of jaguars recorded in Brazil (6.7-11.7
individuals/100 km2; Soisalo and Cavalcanti 2006), but no estimates of jaguar density are
available for northern Pantanal or southern Amazonia where this study took place. Indeed
estimates of jaguar abundance are as scarce for Brazil as for other parts of their range (Maffei
et al. 2004; Silver et al. 2004; Wallace et al. 2003). Nonetheless, reports of livestock loss to
jaguars (Dalponte 2002; Michalski et al. 2006) and jaguar sightings (Paula et al. in press)
suggest that jaguars are also present in relatively large numbers in northern Pantanal and
southern Amazonia.
São Paulo city
We also evaluated children and adolescents from a private school in Sao Paulo city, the
largest metropolitan area in the continent, home to more than 19 million people. Despite its
size and regional influence, Sao Paulo is surrounded by large remnants of the Atlantic Rain
Forest (e.g., Serra do Mar State Park, with 350 thousand hectares), some of them still
inhabited by jaguars (Torres et al. 2008).
Structure of the thesis
This thesis includes four data chapters, which I describe briefly below, together with this
introduction and a concluding chapter. The data chapters are written in the format of
scientific papers and, for this reason, there is some repetition between chapters, e.g. sites are
described both in the introduction and in data chapters.
22
Chapter 2 – Perceptions of jaguars among children and adolescents in Brazil: implications
for conservation
The children and adolescents of Brazil are future decision-makers in a country that contains
more than half of the world‟s jaguars. Therefore, jaguar conservation would benefit from an
understanding of the perceptions of jaguars among young Brazilian people. To investigate the
variation and development of perceptions about jaguars, and to inform strategies for
conservation education and communication, I assess children and adolescents between ages
10 and 19 in urban and rural areas of both Amazonia and the Pantanal, and São Paulo city.
Chapter 3 – Perceptions behind the conflicts between people and jaguars in Amazonia and
Pantanal
I explore how perceptions of the impact of jaguars on livestock and on human safety vary
with region, place of residence, experience of jaguars (including reported livestock loss),
attitudes towards jaguars, knowledge of the species, and perceptions of changes in jaguar
abundance and the regional economic situation. To investigate further the role of negative
experiences versus attitudes and knowledge in determining perceptions of jaguar impact on
livestock, I conduct an experiment in which respondents are asked to interpret photographs of
dead cattle and assign the most likely cause of death.
Chapter 4 – From perceptions to persecution: conflicts between people and jaguars
I use the Theory of Planned Behaviour to examine the role of ranchers‟ perceptions, norms,
attitudes and intentions concerning jaguar killing, in determining their jaguar-killing
behaviour. I also investigate the influence of: 1) descriptive norm and social identity on
ranchers‟ intention to kill jaguars on their properties; and 2) the effect of perceptions of
jaguar impact on human livelihoods (livestock and human safety), and of property size, on
the variables that influence intention to kill.

23
Chapter 5 – School-based approaches for improving perceptions of jaguars among
children and their fathers on the Amazon deforestation frontier.
Based on the findings of Chapters 3 and 4 (landowners in Amazonia are more likely to kill
jaguars if they approve of the behaviour and believe that their neighbours commonly kill
jaguars, and these attitudes and social norms are affected by perceived impact of jaguars on
human livelihoods, and ultimately by attitudes towards, and knowledge of, the species) I
conduct a school-based experiment in the Amazon deforestation frontier to examine the
effects of information and elaboration on these psychological measures among pupils, and the
effects of information (communicated via printed materials including children‟s homework)
on these measures among pupil‟s fathers.
Chapter 6 – General summary and recommendations for future research.
I summarise the findings and discuss their implications for conservation. In addition, I
discuss future research that could draw upon my findings to expand our understanding of the
human dimensions of human-jaguar conflict. Specifically, I recommend comparative studies
involving jaguars and pumas, research on the conflicts between humans over jaguars,
interdisciplinary research, and a model to address human-jaguar conflicts at the levels: ranch,
region and range.
24
Chapter 2 – Perceptions among children and adolescents
CHAPTER 2
PERCEPTIONS OF JAGUARS AMONG CHILDREN AND

ADOLESCENTS IN BRAZIL: IMPLICATIONS FOR
CONSERVATION
It is time to look to the very roots of motivation and understand why, in what circumstances
and on which occasions, we cherish and protect life
E.O. Wilson
Abstract
The children and adolescents of Brazil are future decision-makers in a country that is home to
the majority of the world‟s c. 30,000 jaguars. Therefore, jaguar conservation would benefit
from an understanding of the perceptions of jaguars held by young people in Brazil. To
investigate variation in, and the development of, perceptions about jaguars, and to inform
strategies for conservation education and communication, we assessed children and
adolescents between 10 and 19 years of age in urban and rural areas of both Amazonia and
the Pantanal, and São Paulo city. Jaguars were more prominent in the perceptions of
youngsters than any other native mammal. Their prominence was not fully explained by
knowledge about the jaguar or direct experiences with the species, such as experiencing
family livestock depredation problems. Nonetheless, knowledge predicted selection of the
jaguar as the most liked species in rural Amazonia. The jaguar was predominantly perceived
as threatening, but also as beautiful and endangered. Adjectives attributed to jaguars were
mostly emotion-laden. The parents were the most important source of children‟s information
about jaguars in rural Amazonia and Pantanal. Prominence, perceptions and knowledge of
jaguars varied significantly with gender, age and location. Education and other opinion-
forming interventions that aim to increase tolerance towards jaguars among young people,
will be most effective if they capitalise on the prominence of the jaguar and the emotions it
elicits, as well as taking into account variations in prominence, perceptions and knowledge,
and involving the parents.
25
Introduction
Conflicts between people and wildlife often depend less on the real impact of wildlife on
humans, and more on people‟s perceptions of that impact and of the species involved
(Conover 2001). Such perceptions can affect human behaviour towards the species
concerned, for example, influencing whether a person decides to persecute or tolerate it, or to
resist or support a related conservation policy (Marker et al. 2003). In conflicts between
people and large carnivores, the perceived impact often exceeds the evidence for it (Sillero-
Zubiri et al. 2007), and people‟s opinions and behaviours are often divided (Scarce 2005). As
the largest terrestrial predator in the Neotropical lowlands, the jaguar (Panthera onca) is
simultaneously cherished as a symbol and reviled as a threat to livestock and human safety
(Cavalcanti et al. 2010). Consequently, while jaguar protection has been advocated by
representatives of the scientific, non-governmental and eco-tourist sectors, jaguar persecution
is perpetrated or supported by cattle ranchers, farmers and poachers (Marchini 2003).
Perceptions of jaguars among Brazilians are particularly important factors in conserving the
species because Brazil is home to the majority of the world‟s c. 30,000 jaguars (Medellin
2009), and it encompasses half of the species‟ current distribution (Torres et al. 2008),
including the two largest jaguar strongholds in the world (Sanderson et al. 2002), namely the
rainforests of Amazonia (3,400,000 km2) and the wetlands of the Pantanal (140,000 km2).
The future of the species arguably depends upon perceptions held by various interest groups
in the country, and in particular Brazil‟s young people who, as future decision-makers, hold
the key to the long-term planning required to save the jaguar. Nonetheless, little is known
about how Brazilians – children and adolescents in particular - perceive the jaguar. We
believe that a deeper knowledge of Brazilian young people‟s perceptions of jaguars, how they
are formed, and how and why they vary, could be pivotal for jaguar conservation. We sought
26
to understand these factors to better inform conservation, education, and communication
strategies.
Review and hypotheses
Large carnivores figure prominently in people‟s perceptions of animals and often elicit both
positive and negative feelings. Macdonald (2001) states that carnivores “are simultaneously
different things to different people (and sometimes also to the same people)”. For instance,
while lions and leopards in Tanzania are among the most prominent or important species to
the Maasai and Barabaig in the Rungwa-Ruaha region (Dickman 2005), and the favourite
game species in and around Selous Game Reserve (Leader-Williams and Hutton 2005), they
are also some of the least popular animals among rural populations adjacent to the Serengeti
National Park (Kaltenborn et al. 2006). Most people in the Brazilian countryside consider
jaguars either beautiful or dangerous, while 10% consider them to be both (Santos et al.
2008). Several studies have demonstrated that among children and adolescents large
carnivores provoke both attraction and concern (Andersone and Ozolins 2004; Skogen 2001).
Children in the United Kingdom and Spain like wolves but are afraid of them (WWF-UK
2000). Brazilian children in Amazonia, the Atlantic Rainforest and Pantanal are particularly
aware of the presence of jaguars in the biome (Santos et al. 2008). The jaguar features as the
main character in more books for children and adolescents than any other native species; in
some of these books the jaguar is the hero and in others the villain (Appendix I). We
hypothesized that jaguars feature more prominently than other native mammals in the
perceptions of Brazilian young people (Hypothesis 1) and that perceptions of jaguars among
young people are both positive and negative (Hypothesis 2).
27
Several characteristics can affect how an animal is perceived by people. Animals with large
body size, that are beautiful or attractive, similar to humans, that can be identified
individually, have social habits and economic value, that neither compete with nor cause
damage to humans, and that do not exhibit predatory behaviour are usually perceived more
positively than animals with opposite characteristics (Kellert 1996). The characteristic of
jaguars that has received most attention from conservationists is their potential to cause
economic damage to cattle ranchers through livestock predation (Hoogesteijn 2000).
However, Conforti and Azevedo (2003), found that perceptions towards jaguars among
landowners in southern Brazil were not influenced by the history of predation on the
properties surveyed. Likewise, among the landowners interviewed in northern Pantanal by
Zimmermann et al. (2005), attitudes towards jaguars were more closely related to
respondents‟ age and relative wealth than to cattle losses, with younger and wealthier
ranchers holding more positive attitudes towards them. Therefore, we hypothesized that
livestock loss has a relatively minor effect on young people‟s perceptions of jaguars
compared to other factors (Hypothesis 3).
Several studies have demonstrated that perceptions of animals differ with demographic and
socioeconomic status, including age, gender and place of residence (rural/urban) (Kellert
1996). For instance, Kellert and Westervelt (1983) demonstrated clear and predictable
development stages in children‟s perceptions of animals. Four distinct developmental stages
were identified: (a) under 6 years old, when animals are viewed in highly instrumental,
egocentric and exploitative ways, and children reveal the greatest fear and anxiety toward
nature; (b) age 6-9, when children become more aware of animals possessing interests and
feelings unrelated to themselves and recognize their right not to be harmed; (c) age 9-12,
when the child shows a dramatic increase in factual understanding and knowledge of animals;
28
and (d) age 13-17, when adolescents reveal a sharp increase in abstract, conceptual and
ethical reasoning about the natural world. Using Kellert‟s typology of attitudes (Kellert 1976;
1996), Eagles and Muffit (1990) and Bjerke et al. (1998) found similar age trends for some
types of attitude (e.g. „dominionistic‟ attitude scores [i.e. levels of interest in exercising
mastery and control over wildlife] decrease with increasing age), but contrasting results for
others (e.g. „naturalistic‟ and „ecologistic‟ attitude scores [i.e. levels of interest in
experiencing direct contact with wildlife and understanding the biological function of
organisms and their habitat, respectively] decreased with increasing age among Norwegians
but increased in North Americans). Røskaft et al. (2003) found that fear of brown bears and
wolves increased with age among young Norwegians. Despite the apparent contradictions in
findings, these studies all revealed that major transitions in perceptions about wildlife occur
during pre-adolescence and early adolescence. For this reason we selected 10 to 19 year olds
as subjects in this study and predicted that perceptions of jaguars would change significantly
over this age range (Hypothesis 4). Irrespective of age, women express significantly more
fear of large carnivores than do men (Bjerke et al. 1998; Røskaft et al. 2003), but also exhibit
greater concern for their conservation (Czech et al. 2001). Examples of differences in
perceptions of animals between rural and urban young people include in Norway stronger
dominionistic and moralistic attitudes in rural and urban areas respectively (Bjerke et al.
1998) and in Latvia greater support for lethal control of wolf and lynx among adult rural
inhabitants than urban dwellers (Andersone and Ozolins 2004). We hypothesized that gender
and place of residence also affect perceptions of jaguars among children and adolescents in
Brazil (Hypothesis 5 and 6, respectively).
Perceptions of animals have also been demonstrated to change with socioeconomic, cultural
and historical factors (Dickman 2008; Kellert 1996; Manfredo 2008). Northern Pantanal and
29
southern Amazonia both contain large populations of cattle and jaguars (see Study Areas
below), but the regions differ in many socioeconomic, cultural and historical respects.
Northern Pantanal is populated by traditional families that have raised cattle there for
generations and southern Amazonia is largely inhabited by recent immigrants from different
parts of Brazil, including areas where jaguars have long been extirpated and forests replaced
by pastures and crop fields (Cavalcanti et al. 2010). In São Paulo city (Southeastern Brazil),
people are not exposed to any threat that jaguars may pose to livestock or human safety.
Therefore, we predicted that perceptions among young people would vary across the study
area (Hypothesis 7).
While previous studies on perceptions of wild animals among young people have investigated
certain factors relevant to conservation, none has focused on a particular taxon and examined
simultaneously differences between biomes, socio-economic, cultural and age groups. This
study adopted a comprehensive approach, forming the first step in examining: 1) the relative
prominence of the jaguar among other species in the perceptions of children and adolescents;
2) the role of knowledge and experience in determining this prominence; 3) the variation in
prominence, knowledge and perceptions of jaguars in young people across ages, between
sexes, from rural and urban areas and in contrasting regions; and 4) the sources of
information that contribute to this knowledge and shapes those perceptions.
Methods
Study Areas and Participants
This study was conducted in Alta Floresta, on the deforestation frontier in southern Amazonia
and Cáceres, in the northern Pantanal. We also evaluated children and adolescents from a
private school in Sao Paulo city (Study Areas, Chapter 1)
30
All of the children and adolescents assessed for this study were school students. We
evaluated students in the 5th and 7th grades of elementary schools (expected ages 10 to 15)
and 1st and 3rd years of high schools (expected ages 15 to 18). In Amazonia and Pantanal the
schools participating in the study were public. In Alta Floresta, we worked with the only two
rural schools with both elementary and high school education (Mundo Novo and Ouro Verde
State Rural Schools) and two of the largest schools in the urban area (Vitoria Furlani da Riva
and Jaime Veríssimo de Campos State Schools). In Cáceres, we worked with one urban and
one rural school. The urban school (São Luiz State School) was selected at random and the
rural school (Laranjeiras Rural School) was the school closest to the floodplain and which
could be reached by car. All participating schools in Amazonia and the Pantanal had a single
classroom for each grade and all students in the selected grades were assessed. In Sao Paulo,
the school was selected at random from a short list of the most renowned private schools in
town. The selected school (Vera Cruz) had many classrooms for each grade. One classroom
of each selected grade was chosen at random and all the students in the room were evaluated.
Public rural schools of Amazonia and Pantanal and private schools in São Paulo city
represent two opposing extremes in the broad range of socio-economic and cultural contexts
in which Brazilian children and adolescents live. Most of the young people in Brazil lie
somewhere between these two extremes and therefore results from our survey may shed light
on the perceptions of other children and adolescents in the country.
Questionnaire Development and Data Collection
Self-completion questionnaires were used to collect data. Between March and November
2007 we conducted a pilot study using focus groups (Oppenheim 1998) in which groups of
10 to 15 elementary and high school students from the participating schools in Alta Floresta
31
discussed the positive and negative aspects of coexisting with jaguars. By listening to
students talking freely, we were able to identify salient beliefs, perceptions and peculiarities
of their parlance, which were then used in the design of the questionnaire. The questionnaire
was then pre-tested in one classroom at each of the four schools in Alta Floresta and revised
accordingly for use in the main study.
We visited the schools in Amazonia, Pantanal and São Paulo between September and
November 2008. SM introduced himself and his assistant as representatives of an educational
project from outside the school and gave the students a brief introduction. The introduction
was devoid of value statements and merely invited them to participate in a project involving
questions about their opinions about wildlife. SM stressed that there were no right or wrong
answers to the questions, that no grade would be assigned to the answers and therefore the
students could feel at ease to offer their views transparently. SM also emphasized the
importance of each student completing their questionnaires individually. Jaguars were not
mentioned in the introduction to avoid influencing students‟ answers to the open questions in
the first part of the questionnaire. Teachers were given the choice of staying or leaving the
room; a few stayed but did not interfere with the survey. Questionnaires and pencils were
distributed and questions were read out one at a time. We checked that students understood
the question and ensured that they were writing animals‟ names and numbering their
responses correctly. This approach allowed us to repeat questions and provide further
explanations when necessary, and help groups of students avoid confusion with their task.
The questionnaires were completed in approximately 30 minutes and returned directly to SM.
Questionnaires were divided into five sections: 1) socio-demographics (e.g. gender, age and
residence [urban/rural]); 2) prominence of jaguars relative to other species; 3) experience of
32
jaguars; 4) knowledge and perceptions of jaguars; and 5) sources of information about
jaguars (Appendix II).
Open questions were used to gauge the jaguar‟s prominence (i.e. how the species features in
people‟s hearts and minds). The question “Cite three animals that you know to occur in the
nearest forest” was used to assess the propensity of different species to come to mind in
response to an open question on knowledge (or belief) about occurrence (hereafter called
„cognitive prominence‟). Because the Portuguese names for jaguar and puma (Puma
concolor) are similar (onça-pintada and onça-parda, respectively), we asked the students to
make sure they had written the complete name of the animal they meant before moving on to
the next question. Cognitive prominence scores were calculated by assigning the numbers 3,
2 or 1 to respondents who mentioned the jaguar in the first, second or third options,
respectively, and 0 to those who did not mention the jaguar. The questions “What animal do
you like/dislike most?” were used to assess the propensity of different species to come to
mind in response to open questions on both positive and negative feelings towards animals
(hereafter called „affective prominence‟). Answers were coded as dichotomous variables,
using 1 for jaguar and 0 for other species.
We assessed three categories of experience with jaguars: livestock loss; acquaintance with
injured person; and sighting. The questions were: Has your family (i.e. household) ever lost
livestock to jaguars? When did that happen?; Do you personally know anyone who has been
injured by a jaguar? Who was it and how did that happen?; and Have you ever seen a live
jaguar in the wild? Where?”. Answers to the first part of each of these questions were in a
binary (yes/no) format. The secondary “when”, “who”, “how” and “where” questions were
included in an effort to validate the answer to the first part of the question.
33
For knowledge questions all answers were coded as binary variables (1 for a correct answer
and 0 for an incorrect answer or for “do not know”). Knowledge was measured using seven
questions so the resulting scale ranged from 0-7 (minimum to maximum knowledge).
Questions were posed as statements, three of which were correct and four incorrect. The
statements were: (i) The jaguar generally begins to consume its prey from the front while the
puma consumes the areas from the ribs backwards (correct); (ii) A jaguar‟s prey is usually
hidden and covered with leaves, while a puma‟s prey isn‟t (incorrect); (iii) The jaguar‟s prey
generally presents a bite mark at the base of the neck, while the puma‟s prey generally has a
bite on the throat (correct); (iv) The jaguar‟s footprint is longer than wide with thinner and
pointed toes, while the puma‟s footprint is slightly wider than long, with round toes
(incorrect); (v) A female jaguar produces on average 1 or 2 cubs every other year (correct);
(vi) Black jaguars are more dangerous to cattle than yellow jaguars (incorrect); (vii) Jaguars
kill more people every year in Brazil than do domestic dogs (incorrect).
Respondents‟ general perceptions of jaguars were assessed using the open question
“Complete the sentence: of all the animals, the jaguar is the most…”. Concern about the
impact of jaguars on human safety and property, and attraction to the species were assessed
through multiple questions including five statements regarding „concern for safety/property‟
(e.g. “If I had to walk in the forest, I would be afraid of encountering a jaguar”) and three
regarding „attraction‟ (e.g. “It would be fun if jaguars were seen more often”). Answers to
these questions were gathered using a 5-point Likert scale (i.e. strongly disagree, mildly
disagree, neither disagree nor agree, mildly agree, strongly agree).
The relative importance of different information sources about jaguars was assessed by
asking the respondents to complete the sentence “What I know about jaguars I have learnt
34
mostly from…”. Respondents could then tick one or more boxes in a list of ten options
(television, parent, teacher/school, book, nature/travel, zoo, internet, friend/neighbour,
magazine, other).
Data analysis
The independent variables were gender, age (coded according to the following age classes:
10-11, 12-13, 14-15, 16-17, and 18-19 years old), location (rural Amazonia, urban Amazonia,
rural Pantanal, urban Pantanal, and São Paulo city) and experience with jaguars (i.e. sighting
of a jaguar in the wild, household livestock depredation by a jaguar, and acquaintance with a
victim of personal attack by a jaguar).
Answers in either a binary (yes/no) or in 5-point scale formats to the questions of knowledge,
concern and attraction enabled us to construct summated rating scales and calculate scores for
each variable. The scales were adjusted to range from 0 (no knowledge, concern and
attraction) to 1 (maximum knowledge, concern and attraction). Cronbach‟s alpha was used to
estimate the internal consistency of the scales (Cronbach 2004). The high values of
Cronbach‟s alpha for the concern and attraction scales (Table 2.1) provide support for our
grouping of related questions. Cronbach‟s alpha for the knowledge score was at the lower
limit of acceptability (α = 0.63); this is most likely the result of the multi-dimensional nature
of the questions and might also derive from frequent guessing of the two-choice questions.
The construction of psychological scales and reliability analysis have been described in detail
elsewhere (DeVellis 2003; Vaske 2008).
Differences in cognitive and affective prominence among genders, age classes, and regions
were assessed using Pearson‟s chi-square test. Logistic regression was used to predict
35
cognitive and affective prominence in rural Amazonia and rural Pantanal based on experience
with jaguars and knowledge about the species. For this analysis, cognitive prominence was
coded as a dichotomous variable, scoring 1 when jaguar was mentioned (as the first, second
or third option), and 0 when it was not. We used log-linear analysis to examine relationships
between prevalent perceptions of jaguars and each of age class and region. Analysis of
variance was used to test for differences in knowledge, concern for safety/property and
attraction scores among genders, age classes, places of residence and locations; post-hoc
Tukey tests were used to assess differences between locations.
Results
In total 1,115 questionnaires were properly completed and analyzed, 886 of them in
Amazonia (567 urban and 319 rural), 151 in the Pantanal (82 urban and 63 rural) and 118 in
Sao Paulo. Responses were received from respondents who were: 10-19 years old in
Amazonia and the Pantanal; and 11-17 years old in São Paulo. Males and females were
approximately evenly represented in the final sample (53% males and 47% females). Age
classes were represented as follows: 10-11 years (21%), 12-13 years (26%), 14-15 years
(21%), 16-17 years (25%), 18-19 years (7%).
Experience and knowledge
As shown in Figure 2.1, only a minority of students reported seeing a jaguar in the wild. The
percentage varied significantly (χ2 = 74.9, df = 4, p = 0.001) among the five main survey
groups. Some of the sightings reported by Sao Paulo students had taken place in the Pantanal
during holiday trips. The majority of the respondents who reported a sighting told us that they
had seen the jaguar only once and that the sighting had been brief, typically a glimpse. The
percentage of students reporting household livestock predation by a jaguar was similar in
36
rural Amazonia and rural Pantanal (χ2 = 330.9, df = 1, p = 0.142). Reports of jaguar attacks
on people differed significantly (χ2 = 26.3, df = 4, p < 0.001) among the main survey groups
and were most frequent in rural and urban Pantanal, followed by rural and urban Amazonia.
Just 7% of Sao Paulo students reported personal acquaintance with a victim of jaguar attack
and the person attacked was either a ranch-hand on a property owned by the student‟s
household or someone met during a trip to the Amazon. None of the reported cases of attack
was fatal. Knowledge about jaguars did not differ significantly among age classes, but was
greater among male than female students (male: Mean = 0.30, SD = 0.01; female: Mean =
0.24, SD = 0.01), differed significantly among locations (Amazonia: M = 0.28, SD = 0.17;
Pantanal: Mean = 0.35, SD = 0.17; Sao Paulo: Mean = 0.19, SD = 0.18), and was marginally
significantly greater in rural than urban areas (rural: Mean = 0.34 , SD = 0.02, urban Mean =
0.29, SD = 0.09) (Figure 2.2 and Table 2.1.).
37
Figure 2.1. Percentage of respondents reporting a jaguar sighting (shown white), household livestock
depredation by jaguars (rural students only) (shown black), and acquaintance with a victim of jaguar
attack (shown grey) in different locations.
38
1.0
Mean Knowledge Score
0.0
Amazonia Pantanal São Paulo
Figure 2.2. Mean knowledge scores by age class (● 10-11, ● 12-13, ● 14-15, ● 16-17, ● 18-19) and
location. Bars represent the 95% confidence interval for mean.
Table 2.1. Results of analysis of variance of knowledge of, concern about, and attraction towards
jaguars by students‟ gender, age1, place of residence (rural/urban)2 and location3.
Cronbach‟s df MS F-value p-value
alpha
Knowledge 0.61
Gender 1 6.282 6.351 0.012
Age 4 0.268 0. 271 0.320
Rural/urban 1 6.657 6.789 0.054
Location 2 17.829 18.025 <0.001
Concern 0.77
Gender 1 138.062 18.997 <0.001
Age 4 9.340 1.285 0.274
Rural/urban 1 0.200 0.179 0.351
Location 2 77.532 10.668 <0.001
Attraction 0.84
Gender 1 0.011 0.002 0.963
Age 4 6.602 1.255 0.287
Rural/urban 1 3.008 3.134 0.097
Location 2 47.440 9.015 <0.001
1. Age classes: 10-11, 12-13, 14-15, 16-17, 18-19 years.
2. Only for Amazonia and the Pantanal.
3. Locations: Amazonia, Pantanal and São Paulo city.
Prominence
In support of Hypothesis 1, the jaguar was mentioned more often than any other species or
taxon as the first answer to the question “Cite three animals that you know to occur in the
nearest forest” (Amazonia 24%, Pantanal 37%, and Sao Paulo 28%), even despite its
39
elusiveness (it was seen by far the least often among the ten most prominent species in
Amazonia and Pantanal) (Figure 2.3). Capybara, monkey, snake and macaw also appear
among the ten most salient taxa in the three samples. However, the answers “monkey”,
“snake” and “macaw” do not refer to single species, but multi-species taxa (there are 83
species of primates, 330 species of snakes and 13 species of macaws in Brazil). The puma
was not among the 10 most prominent animals reported for any of the three locations.
The jaguar was also the most common answer to the question “What animal do you like
most?” (Amazonia 20%, Pantanal 28%, Sao Paulo 26%) (Figure 2.4). „The monkey‟ was the
second most liked animal in Amazonia and Sao Paulo, whereas in the Pantanal second place
was occupied by the symbol of the region, the jabiru stork. As for the question “What animal
do you dislike most?”, snake topped the ranking in each of the three locations, followed by
jaguar in Amazonia (9%) and Pantanal (20%). In Sao Paulo, the jaguar was the fourth most
disliked (3%), behind snake, spider and insects. These results support Hypothesis 2.
40
a) b) c)
Amazonia B Pantanal C São Paulo City
Jaguar Jaguar
Jaguar
Capybara Caiman
Snake
Monkey Capybara
Monkey
Paca Jabiru stork
Birds
Snake Snake Capybara
Tapir Tapir Golden lion tamarin
Armadillo Coati Toucan
Macaw Deer Macaw
Caiman Monkey Sloth
Peccary Macaw Marmoset
0% 100% 0% 100% 0% 25%
Percentage of respondents Percentage of respondents Percentage of respondents
Figure 2.3. Cognitive prominence (black bars) (percentage of respondents who mentioned various taxa first when asked “cite three animals that
you know to occur in the nearest forest”) and associated sighting rate (white bars) (percentage of respondents who reported ever seeing the taxon
in the wild) in a) Amazonia (n=862), b) Pantanal (n=147) and c) Sao Paulo city (n=118).
a) b) c)
Pantanal São Paulo City
Jaguar Jaguar
Snake Snake
Caiman Monkey
Jabiru stork Macaw
Macaw Spider
Anaconda Toucan
Monkey Insects
Deer Golden lion tamarin
Capybara Rat
Birds Frog
0% 30% 0% 30%
Percentage of respondents Percentage of respondents
Figure 2.4. Affective prominence (percentage of respondents in a) Amazonia (n=882), b) Pantanal (n=147) and c) Sao Paulo city (n=118) who
cited the taxon when asked “What animal do you like most?” (white bars) and “What animal do you dislike most?” (black bars). Sorting based on
the summated citations for like and dislike.
42
Neither cognitive nor affective (like and dislike) prominence of jaguars differed significantly
between student genders (p = 0.635, 0.095 and 0.945, respectively). However, cognitive
prominence differed among age classes (χ2 = 33.9, df = 12, p = 0.001) and locations (χ2 =
16.8, df = 6, p = 0.010), with greatest prominence at intermediate ages (Figure 2.5) and in
Pantanal, followed by Amazonia and then Sao Paulo (Means = 1.59, 1.20, 1.17 respectively).
Neither positive nor negative affective prominence differed significantly among age classes
(like: p = 0.220; dislike: p = 0.974). Differences among locations approached significance for
positive affective prominence (greatest in Pantanal 28%, then Sao Paulo 26% and Amazonia
20%, χ2 = 5.728, df = 2, p = 0.059), and differed significantly for negative affective
prominence (greatest for Pantanal 20%, then Amazonia 9% and Sao Paulo 3%, χ2 = 24.798,
df = 2, p < 0.001) (Figure 2.5). In support of Hypothesis 3, logistic regression did not reveal
any effect of experience, or knowledge, of jaguars on either positive or negative affective
prominence in rural Pantanal, while cognitive prominence was predicted by reports of
acquaintance with a jaguar attack victim [β = 1.548, SE = 0.580, df = 1, p = 0.008, Exp(β) =
4.704; Hosmer and Lemeshow: p = 0.20; R2 = 0.20 (Nagelkerke)] but not by knowledge of
the species or other measures of experience of the species. In rural Amazonia, positive
affective prominence was predicted by knowledge about jaguars [(β = 1.076, SE = 0.301, df =
1, p = 0.000, Exp(β) = 2.934; Hosmer and Lemeshow: p = 0.56; R2 = 0.33 (Nagelkerke)] with
students who were more knowledgeable about jaguars being more likely to choose the jaguar
as their favourite species. Experience with jaguars did not predict prominence in rural
Amazonia.
a) b)
Figure 2.5. a) Mean cognitive prominence of jaguars by age class (● 10-11, ● 12-13, ● 14-15, ● 16-
17, ● 18-19) and location, and b) mean negative affective prominence (dislike) by location. Bars
represent 95% confidence intervals of the mean.
General perceptions
The question “Of all the animals, the jaguar is the most…?” elicited a total of 80 adjectives,
among which „dangerous‟ and „beautiful‟ were the most frequent (23% and 19%,
respectively). The adjectives were grouped into four categories: „threatening‟ (e.g. dangerous,
ferocious, detrimental), „attractive‟ (e.g. beautiful, interesting, fascinating), „endangered‟ (e.g.
persecuted, hunted, extinct), and „other‟ (e.g. stinky, silly, unknown) (Table 2.2). Log-linear
analysis revealed that the relative frequency of the four categories differed with age class and
location (partial χ2 = 30.9, df = 4, p < 0.001; partial χ2 = 15.4, df = 2, p < 0.001), and there
was a significant interaction between age class and location (partial χ2 = 136.7, df = 8, p <
0.001). Jaguars were most frequently labelled as threatening, followed by attractive and then
endangered by all age classes in all locations except for 10-11 year olds (the youngest group)
and 14-15 year olds in Sao Paulo both of which most frequently considered them attractive.
Jaguars were described as threatening by 70% of respondents in the Pantanal, 55% in
Amazonia and 38% in São Paulo. The population most likely to perceive the jaguar primarily
44
as threatening lived in urban Pantanal (Figure 2.6). The population most likely to consider
them primarily attractive were those in São Paulo (36%), compared to 32% in Amazonia, and
23% in the Pantanal (Table 2.2). Respondents in the Pantanal were also least likely to
consider the jaguar primarily as endangered (4%) compared with 7% in Amazonia, and 9% in
São Paulo. In all three study areas, jaguars were most likely to be perceived primarily as
endangered by 16-17 year old students (16-18 in Sao Paulo): 12% in Amazonia, 8% in São
Paulo, and 4% in the Pantanal.
45
Table 2.2. Percentage of respondents whose answer to the question “Of all animals, the jaguar is the
most…?” revealed their prevalent perception of the jaguar to be threatening, attractive or endangered.
Location and age Jaguar perceived as

class Threatening Attractive Endangered Other
Amazonia
10-11 57 32 2 9
12-13 48 37 7 8
14-15 61 26 5 9
16-17 55 28 12 5
18-19 59 26 7 8
Total 55 32 7 8
Pantanal
10-11 82 18 0 0
12-13 72 28 0 0
14-15 81 16 3 0
16-17 67 25 4 4
18-19 80 20 0 0
Total 70 23 4 4
Sao Paulo city
10-11 39 44 6 11
12-13 41 34 7 17
14-15 33 37 7 22
16-18 50 38 8 4
Total 38 36 9 17
Figure 2.6. Most prevalent perception of jaguars (threatening, attractive, endangered or other) by
location and place of residence.
46
Concern for safety/property and attraction
As shown in Table 2.1, concern about jaguar attack on people and livestock differed
significantly between genders, being greater among female than male students (males: Mean
= 0.43, SD = 0.21; females: Mean = 0.53, SD = 0.18) and differed significantly among
locations, being greatest in Pantanal (Mean = 0.53, SD = 0.18), then Amazonia (Mean = 0.48,
SD = 0.21) and Sao Paulo (Mean = 0.42, SD = 0.15) (Figure 2.7). Attraction differed
significantly among locations, being greatest in São Paulo (Mean = 0.66, SD = 0.20) and
Amazonia (Mean = 0.64, SD = 0.20), followed by the Pantanal (Mean = 0.73, SD = 0.16) (F
= 9.015, df = 2, p < 0.001). These findings support hypotheses 5 and 7. We detected a
significant interaction between the impacts of location and age on attraction to jaguars (F =
5.347, df = 2, p = 0.005) (Figure 7b). Post hoc Tukey tests indicated that students in Sao
Paulo were significantly more attracted towards jaguars than were students in Amazonia and
Pantanal. There was no significant difference in attraction to jaguars between these two
locations. Contrary to hypotheses 4 and 6, knowledge, concern and attraction did not vary
with age and place of residence (rural/urban).
a) b)
1.0 1.0
Mean Attraction Score
0.0 0.0
Amazonia Pantanal Sao Paulo Amazonia Pantanal Sao Paulo
Figure 2.7. Mean standardized scores for a) concern for safety/property, and b) attraction, by gender
(○ female, ■ male) and location. Bars represent the 95% confidence interval for mean.
47
Sources of information
Television was the most important source of information about jaguars for urban children and
adolescents (except for 10-13 year olds in urban Pantanal), whereas the parents were the most
important information source for rural respondents (Table 2.3). School, teachers and books
were the next most important sources of information about jaguars in Amazonia and rural
Pantanal and the zoo was the second most important source of information in São Paulo. The
internet was more important to urban than rural respondents in Amazonia and Pantanal,
whereas friends and neighbours were more important to rural respondents.
48
Table 2.3. Relative importance of different sources of information about jaguars: frequency and rank (between parenthesis) by location and age
class.
Location and Source of information

age class Television Parent Teacher/ Book Nature/ Zoo Internet Friend/ Magazine Other
School Travel Neighbour
Rural Amazon
10-13 0.66 (2) 0.81 (1) 0.56 (3) 0.55 (4) 0.52 (5) 0.16 (9) 0.18 (8) 0.34 (6) 0.25 (7) 0.02 (10)
14-19 0.79 (2) 0.80 (1) 0.54 (4) 0.61 (3) 0.41 (6) 0.25 (8) 0.14 (9) 0.42 (5) 0.30 (7) 0.05 (10)
Urban Amazon
10-13 0.65 (1) 0.45 (5) 0.51 (2) 0.46 (3) 0.37 (6) 0.33 (7) 0.49 (4) 0.19 (9) 0.26 (8) 0.01 (10)
14-19 0.79 (1) 0.73 (2) 0.47 (5) 0.54 (3) 0.46 (6) 0.29 (9) 0.52 (4) 0.30 (8) 0.34 (7) 0.04 (10)
Rural Pantanal
10-13 0.57 (2) 0.86 (1) 0.50 (3) 0.50 (3) 0.36 (6) 0.36 (6) 0.07 (9) 0.43 (5) 0.14 (8) 0.00 (10)
14-19 0.65 (2) 0.82 (1) 0.59 (3) 0.29 (4) 0.18 (8) 0.24 (6) 0.12 (10) 0.29 (4) 0.24 (6) 0.18 (8)
Urban Pantanal
10-13 0.51 (3) 0.65 (1) 0.42 (5) 0.28 (6) 0.53 (2) 0.28 (6) 0.47 (4) 0.19 (8) 0.16 (9) 0.05 (10)
14-19 0.84 (1) 0.81 (2) 0.65 (4) 0.65 (4) 0.67 (3) 0.42 (8) 0.58 (6) 0.40 (9) 0.49 (7) 0.09 (10)
Sao Paulo
10-13 0.72 (1) 0.34 (6) 0.36 (5) 0.40 (4) 0.42 (3) 0.51 (2) 0.32 (7) 0.09 (9) 0.21 (8) 0.02 (10)
14-18 0.68 (1) 0.22 (7) 0.51 (2) 0.35 (5) 0.40 (4) 0.51 (2) 0.31 (6) 0.03 (10) 0.20 (8) 0.06 (9)
Discussion and conclusions
Cognitive and affective prominence of jaguars was exceptionally strong among the young
people of Brazil, regardless of gender, age or location. Our results suggest that jaguars have a
stronger presence in the hearts and minds of Brazil‟s children and adolescents than does any
other native mammal. Conservation educators should capitalise on this prominence to capture
the attention of a wider audience and improve their own effectiveness. Indeed, the
widespread prominence of the species renders jaguars particularly sensitive and valuable
barometers of perceptions of the natural world among the young people of Brazil.
The high profile of the jaguar was not fully explained by the experiences of the respondents
(and their families), or by their knowledge of the species. Jaguars were particularly prominent
in the thoughts of young people in the Pantanal, where jaguars are more abundant and more
frequently seen than elsewhere, they are considered to be a threat to livestock, and where the
first officially documented unprovoked fatal jaguar attack on a human in Brazil took place in
June 2008. However the children and adolescents of Sao Paulo city were also more
cognitively and affectively aware of jaguars than of other species. The fatal attack in Pantanal
was widely publicised by the local media and might account for the apparent relationship
between reported acquaintance with a victim of jaguar attack and the cognitive prominence of
jaguars there, since young people in the region were evaluated only three months after the
incident. The news of this attack may have had a particular impact on those who believed
they knew another attack victim. Indeed, jaguar attacks on people have a prominent place in
story-telling in rural Brazil and the attack reports provided by our respondents may have been
inflated by cognitive biases such as an availability cascade [i.e. a self-reinforcing process in
which a collective belief gains more plausibility through its increasing repetition in public
discourse (colloquially, "repeat something often enough and it will become fact”)] (Kuran
50
and Sunstein 1999) or the availability heuristic (i.e. answers depend on what is more
available in memory, which is biased toward vivid, unusual, or emotionally charged episodes
such as a jaguar attack ) (Tversky and Kahneman 1973). Nonetheless, negative experiences
with jaguars, such as livestock predation on their familiy‟s property or personal acquaintance
with a jaguar attack victim, did not have any effect on either the positive or negative affective
prominence (like or dislike) of the species. While jaguar conservation has concentrated on the
concrete experiences that people have with jaguars – notably livestock depredation - our
results emphasise the importance of other factors on empathy or hostility towards jaguars. If
persecution of jaguars is heavily influenced by perceptions and prejudices acquired at an
early age, and at least substantially unrelated to direct experiences, then the capacity for
education to play a major role in jaguar conservation is clear. Despite being the second
largest terrestrial carnivore in the country, causing damage to livestock (Hoogesteijn 2000),
attacking and killing people in Brazil and elsewhere (Quigley and Herrero 2005), and sharing
half its popular name with the jaguar, the puma was surprisingly unimportant in our sample.
The significance of the puma to children and adolescents in Brazil has most likely been
overshadowed by the exceptional prominence of its larger relative, the jaguar.
The jaguar was predominantly perceived as a predator that threatens human safety and
property. Such a perception may elicit feelings of fear and anger. Jaguars are also perceived
as beautiful, and aesthetic appreciation is another emotional response. Emotion-laden
adjectives were far more frequently attributed to the jaguar than were emotionally neutral
ones. This apparent association in young people between key perceptions of jaguars and
emotional responses is also likely to be found, to some extent, in adults. Although most
studies on the human dimensions of wildlife have adopted a cognitive approach to
understanding human behaviour (Manfredo 2008), our results corroborate the notion that
51
emotions may also play a role in determining attraction or hostility towards wildlife, and
large carnivores in particular. Therefore, our understanding of the conflicts between people
and jaguars, and the design of more effective strategies to solve or mitigate them, may be
improved by the understanding of the interrelationships of cognitive concepts such as values,
norms and attitudes with affective concepts such as emotions.
Surprisingly, knowledge about jaguars, concern about their threat to human safety and
property, and attraction to the species did not change significantly over the age range
represented by this study. This suggests that lasting perceptions and prejudices about jaguars
may be acquired before the age of 10, and that without intervention these will persist at least
into early adulthood. Further research should be targeted at children under the age of 10 to
examine the origin and development of their perceptions about jaguars; their education
should then be targeted accordingly. We contend that education and communication
interventions might also be used effectively to increase knowledge and improve perceptions
among older children and adolescents. This could be especially successful if variations in
knowledge and perceptions among genders and locations, and in cognitive prominence with
age, were taken into account. In rural Amazonia, for example, where young people are less
knowledgeable about jaguars than in the Pantanal and knowledge predicts the choice of the
jaguar as the most liked species, information-intensive campaigns may increase knowledge of
and therefore empathy towards the jaguar, especially if children are targeted at 12-13 years
old, the age-group in this study to which jaguars had greatest cognitive prominence.
Increasing knowledge might also contribute to jaguar conservation in rural Pantanal. Many
young people in rural Pantanal will grow up to make decisions about the management of
cattle and natural resources on their household‟s property but, despite being more
knowledgeable about jaguars than other groups in this study, their knowledge about jaguar
52
ecology and predation issues was nevertheless rudimentary. For example, only 17% of young
people from rural Pantanal could distinguish jaguar and puma footprints. Our observations
also suggest that educational interventions in the Pantanal would be more effective if they
addressed fear of jaguars, particularly among girls, and concern about the impact of jaguars
on livestock, and targeted 16 and 17 year olds, amongst whom jaguars had greatest cognitive
prominence and were most likely to be perceived as endangered.
If the future of the jaguar does depend on perceptions of the species among Brazilian children
and adolescents, then advocates for the species should be alert to the influence, identified
here, of television and schools as information sources. It is possible that respondents perceive
television and schools as sources of information in general, rather than specifically about
jaguars, but nevertheless they could serve as useful media for recommendations about
communication interventions. Television might have the greatest potential to influence
perceptions of jaguars among urban children and adolescents. Jaguars have been featured in
only a few documentaries shown on Brazilian television compared to lions, cheetahs, sharks
or crocodiles. Documentaries shown on public television typically emphasize the predatory
behaviour and elusive and mysterious character of the jaguar. Conservation organizations
should consider producing videos that address the importance of the jaguar, clarify its real
impact on human livelihoods, and explain how to minimise this impact. Videos could be
broadcast by local public television networks at relatively low cost in areas where human-
jaguar conflict is an issue.
School-based conservation education might be particularly effective in Brazil because
primary school enrolment rates are relatively high in the country, and a large proportion of
Brazilian children and adolescents could therefore be reached through their schools. Primary
53
school enrolment in Brazil (97%) is greater than in many other countries of conservation
concern, e.g. Ethiopia 45%, Namibia 73%, Bolivia 78%, Zimbabwe 82%, Botswana 85%
(Unesco 2007). In Mato Grosso State, for example, 97.6% of school-age children are enrolled
at a primary school (Brazilian Institute of Geography and Statistics 2007). On the other hand,
the quality of Brazilian schools - in particular the public schools - is considered unsatisfactory
(Unesco 2008), and this might offer another opportunity for conservation. School directors
and teachers generally recognise their human and financial limits to addressing conservation
issues even at local or community scales, and are often willing to cooperate with conservation
organizations. Conservationists could integrate jaguar conservation education into school
curricula by acting as a resource to the school. Providing professional development for
teachers and field experiences for both teachers and students might substantially increase
their willingness to cooperate. Providing resource materials on jaguar conservation might also
help, as biology textbooks convey very little information about jaguars. Zoos emerged as
another important source of information about jaguars among São Paulo students. Despite the
potential role of zoos in education for conservation (Tunnicliffe and Scheersoi 2009), it might
not be cost-effective over most of the species‟ Brazilian range to take school children to see a
jaguar in a zoo because of the large distances involved (Marchini and Luciano 2009).
However, in rural Pantanal and Amazonia where jaguar persecution occurs, the parents
appear to be the most important factor in shaping young peoples‟ perceptions of jaguars. In
these areas jaguars are persecuted by adults, whose influence (along with that of other adults)
over their children contributes to the perpetuation of persecution. Young people‟s views of
the jaguar in rural Brazil are locally constructed, in an informal, family setting, and often
passed orally between generations. Therefore, efforts to improve the accuracy of the
perceptions of rural children and adolescents regarding jaguars in Brazil will ultimately need
54
to be directed through their parents. However, given the low human population density in
rural areas of Brazil, the lack of free time available to many adults, and the resistance of some
landowners to outsiders, schools may also be effective conduits through which to reach
adults. In short, we suggest that students can help educate their parents. Reaching tens of
students in a classroom, or hundreds on the school‟s soccer pitch, might be more cost-
effective than visiting landowners one-by-one at home, providing that students can, in turn,
influence their parents. Children are known to influence their parents‟ purchasing decisions
(Flurry and Burns 2005; Roedder-John 1999) and environmental attitudes (Ballantyne et al.
1998; Duvall and Zint 2007). Adults also hold keen interest and strong feelings (both concern
and attraction) towards jaguars (Cavalcanti et al. 2010), which might enhance inter-
generational learning about jaguar-related issues. An important applied research question,
therefore, is whether, and how most effectively, children can influence their parents‟ attitudes
and behaviour regarding jaguars in rural Pantanal and Amazonia.
55
Chapter 3 – Perceptions behind the conflicts
CHAPTER 3
PERCEPTIONS BEHIND THE CONFLICTS BETWEEN
PEOPLE AND JAGUARS IN AMAZONIA AND PANTANAL
Opinion is ultimately determined by the feelings, and not by the intellect
Herbert Spencer
Abstract
In an investigation of the perceptions behind the conflicts between people and jaguars in
Amazonia and Pantanal, we explored how perceptions of the impact of jaguars on
livestock and on human safety vary with region, place of residence, experience of
jaguars (including reported livestock loss), attitudes towards jaguars, knowledge of the
species, and perceptions of changes in jaguar abundance and the regional economic
situation. Livestock loss and threat to human safety were not the only predictors of the
perceived conflict with jaguars. Livestock loss acted in combination with attitudes,
knowledge and perceptions of the economic situation to determine how people perceive
the impact jaguars have on their livelihoods. Attitudes and knowledge, in turn, were
influenced by age, gender and whether respondents lived in urban or rural areas. An
experiment in which respondents were shown photographs of dead livestock and asked
to ascribe the cause of death revealed an interaction between attitudes and knowledge:
of respondents whose knowledge of the species was low, those with negative attitudes
towards jaguars assigned a larger number of photographs to jaguar depredation. Our
evidence suggests that attitudes and knowledge can affect the conclusions a rancher
draws from finding the carcass of a cow, or even from noticing that a cow is missing.
The impact of attitudes and knowledge on perceptions was greatest for smaller
properties. Landowners with smaller holdings had more negative attitudes and poorer
knowledge, and the lack of difference in attitudes between the owners of smaller
holdings where cattle were or were not raised shows that conflict with jaguars on small
properties has less to do with livestock loss than is usually believed. The owners of
56
smaller holdings believed that depredation was more serious on neighbouring properties
than on their own, which suggests that perceptions of conflict with jaguars amongst this
category of landowner were shaped primarily by what is heard from other people, and
not by personal experience.
Introduction
Wherever people and jaguars live in close proximity, they generally come into conflict
(Loveridge et al. 2010). As the largest wild felid in the Western Hemisphere, and an
opportunistic predator, the jaguar poses a direct and recurrent threat to large livestock
and to human safety. The persecution of jaguars is assumed to be a direct response to
these threats. However, the ultimate motivation for persecution may not be the loss of
livestock to jaguars (or attacks on humans – which are very rare), but rather the
perceived potential for these threats and for their impact on human livelihoods
(Cavalcanti et al. 2010). Deep-rooted prejudices, hatred and fear towards jaguars may
affect those perceptions. Distorted perceptions could prove perilous to a threatened
species, exacerbating the risks posed by retribution killing, and rendering irrelevant
many ecologically-based conservation actions and mitigation measures.
Conflicts between ranchers and jaguars over livestock have been documented
throughout the jaguar‟s range, (e.g. Argentina: Schiaffino et al. 2002; Belize:
Rabinowitz 1986; Brazil: Cavalcanti 2008; Costa Rica: Saenz and Carrillo 2002;
Guatemala: Soto 2008; Venezuela: Polisar et al. 2003). Livestock depredation on small
ranches, stocked with small herds of cattle, can be ruinous (Hoogesteijn et al. 1993).
However, livestock losses to jaguar are generally small when averaged over time and
space. Average losses attributed to jaguars in Brazil range from 0.2 - 2.3% of livestock
holdings in the Cerrado (Palmeira 2004), the Atlantic Forest (Conforti and Azevedo
57
2003), southern Amazonia (Michalski et al. 2006), southern Pantanal (Azevedo and
Murray 2007), two ranches in northern Pantanal (Dalponte 2002), and a larger portion
of the northern Pantanal (Zimmermann et al. 2005). Sometimes, a major factor
predisposing a particular herd to jaguar predation is poor husbandry (Azevedo and
Murray 2007; Michalski et al. 2006). While the risk of predation is greater among cattle
left unattended close to forest cover (Azevedo and Murray 2007), documented losses of
cattle to jaguar predation are generally much fewer than those attributable to accident,
snake bite, disease, parturition problems, flood (Azevedo and Murray 2007) and even
theft (R. Hoogesteijn, pers. comm.). Furthermore, on some ranches, livestock
depredation by puma may be more common than depredation by jaguars (Polisar et al.
2003).
As for human fatalities and injuries, in contrast to the historical and contemporary
record of unprovoked attacks on humans by lions (Panthera leo) (Patterson 1907;
Yamasaki et al. 1999; Saberwal et al. 1994), tigers (Panthera tigris) (McDougal et al.
2001), leopards (Panthera pardus) (McDougal 1989), pumas (Puma concolor) (Beier
1991; Deurbrouck and Miller 2001), wolves (Canis lupus) (Linnell et al. 2002; McNay
2002), and bears (Ursus sp). (Herrero and Higgins 1995), attacks by jaguars have been
almost invariably associated with the cornering or injury of hunted jaguars, or the
defence of cubs or carcasses. In Brazil, the only documented, unprovoked, fatal attack
by a jaguar on a human occurred on 24 June 2008, when a young fisherman was killed
by a jaguar while sleeping in his tent on a bank of the Paraguay River in the Pantanal.
One speculation has been that this unique incident was a result of jaguars in that area
becoming habituated to people around baits used to attract jaguars to be photographed
by tourists.
58
Irrespective of the magnitude of damage or threat that they pose, jaguars have been
severely persecuted; killing them indiscriminately is one of the most serious threats to
their survival across Latin America (Zeller 2007). Persecution is the major cause of
jaguar mortality outside of protected areas (Nowell and Jackson 1996) and, due to their
wide-ranging movements, threatens jaguars within protected areas as well (Woodroffe
and Ginsberg 1998). Together with habitat loss, persecution has reduced jaguars to 46%
of their historical range (Sanderson et al. 2002). In Brazil, despite the prohibition of
jaguar hunting in 1967, persecution of jaguars continued and contributed to the species
becoming critically endangered in the Caatinga, endangered in the Cerrado and
vulnerable in the Atlantic Forest (Paula et al. in press), although country-wide their
status is near threatened (Caso et al 2008). Brazil contains more than half of the world‟s
c. 30,000 jaguars (Medellin 2009), and encompasses the two largest strongholds for the
species (Sanderson et al. 2002): the Amazonian rainforests and the wetlands of the
Pantanal. In both Amazonia and the Pantanal, however, jaguars occur mostly outside of
protected areas. Less than 8% of the Brazilian Amazon is formally under full protection
(Instituto Socioambiental 2009) and 95% of the Pantanal is privately owned (Seidl et al.
2001). Most encounters between people and jaguars take place on cattle ranches.
Therefore, the future of jaguars is likely to be tightly linked to the perceptions of
Brazilian ranchers and ranch managers. However, despite the rapid growth of academic
interest in perceptions of wildlife in general (Manfredo 2008) and carnivores in
particular (Baker and Macdonald 2004; Dickman 2008; Kalternborn and Bjerke 2002;
Røskaft et al. 2007; Treves and Karanth 2003), little is known about perceptions of
jaguars and the factors that influence them.
59
The first study on local perceptions of jaguars in Brazil was conducted by Conforti and
Azevedo (2003), who evaluated 72 livestock ranchers around Iguaçu National Park in
southern Brazil. They found that perceptions towards jaguars were not influenced by the
history of predation on the properties surveyed. Likewise, among the 50 landowners
interviewed in northern Pantanal by Zimmermann et al. (2005), attitudes towards
jaguars were more closely related to respondents‟ age and relative wealth than to cattle
losses, with younger and wealthier ranchers holding more positive attitudes towards
them. In southern Amazonia, over half of the 62 ranchers interviewed by Michalski et
al. (2006) were interested in implementing non-lethal methods to control predation
levels on their ranches, but about one third explicitly were not. Palmeira and Barrela
(2007) evaluated 28 householders in a small, traditional, rural community in
southeastern Brazil, where most respondents held negative attitudes towards jaguars and
wished the species would be eliminated. In contrast, the majority of the 1,007 people
interviewed by Santos et al. (2008) in the five biomes of Brazil (i.e. Amazonia,
Pantanal, Atlantic Forest, Cerrado and Caatinga) were in favour of jaguar conservation.
Altogether these studies revealed strong and contradictory attitudes to the species and,
along with our previous results (Cavalcanti et al. 2010), suggest that the perceived
impact of jaguars on human livelihoods may often exceed the evidence.
Except for the work by Santos et al. (2008), the foregoing studies were each conducted
at single sites and evaluated only people who were directly involved in conflict with
jaguars; negative experience with jaguars – specifically, livestock depredation – was
addressed as the predictor of conflict. Attitudes to jaguars and/or jaguar management
have been assessed, but only Zimmermann et al. (2005) have examined the potential
predictors of attitudes, and nobody has reported on any relationship between attitudes
60
and people‟s tendency to kill jaguars. Knowledge about jaguars was assessed by
Conforti and Azevedo (2003), but the influence of knowledge on people‟s perceptions
about jaguars has not been examined.
We assessed peoples‟ perceptions of the impact of jaguars on livelihoods in Amazonia
and the Pantanal, and explored relationships between these perceptions and socio-
economic variables such as age, gender, place of residence (urban/rural) and property
size, plus psychological variables such as experiences, attitudes and knowledge about
jaguars. We hypothesized that perceptions of jaguar impact on human livelihoods are
not explained solely by the loss of livestock to jaguars, or by attacks on humans, but
socio-economic and psychological factors as well. Given the socio-economic and
cultural differences between the Amazon deforestation frontier and the Pantanal (see
below), we expected cattle ranchers in the two study areas to differ in their perceptions
of the impact of jaguars on their livelihoods, irrespective of the damage posed by
jaguars (Hypothesis 1). We also expected that perceptions of the impact of jaguars on
livelihoods and attitudes to jaguars would not differ between cattle ranchers – who are
exposed to both livestock loss and personal damage by jaguars – and farmers - who are
exposed to personal damage but not livestock loss (Hypothesis 2). Furthermore, we
hypothesized that the perceived impact of the jaguar on human safety would not be
necessarily greater among rural residents – who are potentially exposed to the attack by
jaguars – than among urban residents – who are not (Hypothesis 3). We expected
negative experiences with jaguars (i.e. attack of jaguars on livestock and people) to
determine perceptions of jaguar impact on livestock and human safety (Hypothesis 4).
However, we hypothesized that attitudes to jaguars and knowledge about the species
would also influence these perceptions (Hypothesis 5). Age, gender and property size
61
were expected to indirectly affect perceptions of impact by their effect on attitudes and
knowledge (Hypothesis 6). To investigate further the role of negative experiences
versus attitudes and knowledge in determining perceptions of jaguar impact on
livestock, we conducted an experiment in which respondents were asked to interpret
photographs of dead cattle and assign the most likely cause of death. We hypothesized
that respondents with stronger negative attitudes to jaguars and less knowledge about
them would assign more photographs to jaguar predation (Hypothesis 7).
Methods
Study areas and participants
This study was conducted in Amazonia and Pantanal (Study Areas, Chapter 1). In
Amazonia we worked in the districts of Alta Floresta and Novo Mundo, on the frontier
of deforestation in the north of the state of Mato Grosso, southern Amazonia. Alta
Floresta was founded in 1976 and colonized by migrant farmers, mostly from southern
Brazil. Properties surveyed in Amazonia were located between 9°23'2.13"S and
9°49'30.86"S and 56°20'25.59"W and 55°25'25.36"W. In the Pantanal, we worked in
the neighbouring districts of Cáceres and Poconé, in the south of the state of Mato
Grosso. The Pantaneiro landowner is typically from long-established, land-owning
dynasties (Barros 1998), and cattle ranching is their main economic activity (Brazilian
Institute of Geography and Statistics 2007). As with the Amazon frontier, northern
Pantanal hosts large cattle herds (around 832,000 and 347,000 head in Cáceres and
Poconé, respectively; Brazilian Institute of Geography and Statistics, 2007) and
depredation attributed to jaguars is a major cause of complaint by the local ranchers
(Marchini 2003; Zimmermann et al. 2005). Ranches surveyed in the Pantanal were
62
located between 16°14'54.69"S and 17°26'58.17"S and 58°18'5.40"W and
56°22'28.04"W.
While sharing the characteristic of hosting large population of both cattle and jaguars,
northern Pantanal and southern Amazonia differ in many social and cultural aspects, as
the former is populated by traditional families that have raised cattle there for
generations and southern Amazonia is largely inhabited by recent immigrants who came
from different parts of Brazil (Cavalcanti et al. 2010).
In both study areas, rural and urban residents were surveyed. In rural areas, the sample
unit was the property, with either owner – wife or preferably husband - from each
surveyed (in a few large ranches, the ranch manager was interviewed when neither of
the owners were available). The Association of Rural Workers of Alta Floresta, Cáceres
and Poconé provided a list of members, with landowner‟s name and contact
information, from which we could randomly select ranches. However, the lists were not
comprehensive. The lack of property registers was particularly pronounced in the
settlements around Cristalino State Park in the municipality of Novo Mundo. Therefore,
a combination of techniques was added to ensure randomness and representativeness.
Google Earth imagery (Google Inc. 2009) was used to map all the roads within the
study areas (available road maps do not include minor roads). A systematic sampling
strategy was then used on each road, selecting every other property along the road. In
Amazonia, ranches were easily accessible by land and all interviews were carried out on
site, while in the Pantanal some properties near the Transpantaneira highway could be
visited but many others were inaccessible overland. Hence, many ranchers were
interviewed in their second residences in town. In Amazonia, properties where the main
63
economic activity was non-stock agriculture were also included in the survey. Urban
residents were interviewed in Alta Floresta and Poconé. On the most central streets of
each town, we selected every fifth home or shop until 50 owners in each study area had
been interviewed.
Our intended survey population was the Amazon deforestation frontier and the
Pantanal; our sample frame represented all major interest groups within this (i.e. urban
residents and cattle ranchers in both regions and farmers in Amazonia). Although we
used different sampling methods to reach these target groups, we achieved high
response rates from each group, and results from our survey can be generalized to other
parts of the Amazon deforestation frontier and Pantanal.
Data collection
The study was conducted between March and October 2007 (Amazonia) and between
February and May 2008 (Pantanal). Face-to-face interviews were used. Conflict with
jaguar can be a sensitive issue in rural areas, and some respondents may have held
negative attitudes towards individuals and institutions that advocate wildlife
conservation. Such attitudes can be stronger towards people and institutions perceived
as foreign to the region (Marchini 2003). Therefore, some measures were taken to avoid
response and interviewer biases. All interviews were conducted by SM, who is
Brazilian. SM was always accompanied by a male local field assistant during the
interviews. SM was as objective and neutral in appearance and behaviour as possible.
Survey visits were made using a vehicle marked with the logo of a locally renowned
educational project of Alta Floresta, of which SM is the coordinator. SM introduced
himself as a representative of the educational project, rather than of a conservation or
64
overseas organisation. He explained to respondents that the purpose of the survey was to
collect data on people‟s perceptions of wildlife, and that these would ultimately
contribute to the educational project. The word „jaguar‟ was not mentioned during the
introduction.
We conducted a pilot study using qualitative, semi-structured interviews (Oppenheim
1998) with 130 people (90 rural and urban residents in Amazonia, and 10 ranchers and
30 urban residents in the Pantanal). By listening to people talk freely, we were able to
identify salient beliefs, perceptions, and peculiarities of the local parlance, which were
then used in the design of the questionnaire, and in adjusting our language to the target
groups. During the piloting process, open-ended questions were replaced incrementally
by clear, quantitative questions that would produce data suitable for statistical analysis.
Ultimately we designed a structured interview (Oppenheim 1998) (Appendix II) to
examine the following: (1) perceptions of the impact of jaguars on human livelihood;
(2) experience of jaguars; (3) attitude towards jaguars; (4) knowledge about jaguars and
depredation; (5) socio-demographic variables; and (6) description of the property. Two
elements that emerged from the initial semi-structured interviews were also
incorporated: perceptions of (7) change in jaguar abundance and (8) the regional
economic situation. The variables are described in more detail below.
Perceptions of the impact of jaguars on human livelihood. We asked 8 questions, from
an initial pool of 15 questions, designed to measure perceptions of jaguar impact using a
6-point scoring system that reflected the magnitude of the impact (0 none to 5 very
high). Questions were designed to assess beliefs about past damage and perceptions of
risk of future damage at different geographical (ranch / neighbourhood / district) and
65
demographic (personal / household / community) scales. Questions about the impact of
jaguars on livestock were asked only to cattle ranchers. Responses were summated to
obtain the respondent‟s total score.
Experience of jaguars. Three categories of experience with jaguars were assessed:
livestock loss; acquaintance with injured person; and sighting. The questions were open-
ended: What percentage of your herd was lost to jaguars in the last 12 months? (cattle
ranchers only); How many people do you personally know that have been injured by a
jaguar?; How many times have you seen a live jaguar in the wild?
Attitude towards jaguars. We assessed respondents‟ attitudes towards jaguars,
according to (Ajzen and Fishbein 2000; Ajzen 2001)‟s definition of attitudes, “the
tendencies people have when they view an entity with some degree of favourability or
unfavourability”. Attitudes were measured using a combination of Likert scales (e.g.
strongly disagree, disagree, neither agree nor disagree, agree, strongly agree) and
semantic differential statements (where respondents are offered a choice of two polar
opposite responses e.g. good-bad) (Oppenheim 1998). Each question consisted of a
statement about jaguars, and two extreme response options describing the potential
range of attitudes to the statement, e.g. very sad – very happy. Respondents were asked
to select a point on a five-point scale between those options that reflected their view.
We coded each answer -2, -1, 0, +1, or +2. Some questions were phrased in favour of
jaguars, and others against, in an effort to balance any potential bias associated with
question direction. However responses that were favourable to jaguars were always
coded positively, and responses that were unfavourable were always coded negatively,
so that we could construct a summated rating scale to summarise an individual
respondent‟s overall attitude towards jaguars. We used Cronbach‟s alpha to improve the
66
internal coherence of the scale by discarding items to maximize the alpha value (Vaske,
2008). Five questions were used so the summated rating scale ranged from -10 to +10.
The questions were as follows: (1) You would like the jaguar population in the region
to: decrease a lot - increase a lot; (2) If all the jaguars disappeared forever from the
region, you would feel: very sad - very happy; (3) What you feel towards jaguars is
better described as: dislike a lot - like a lot; (4) The jaguar has its value, even if it does
not generate any income to you: strongly disagree - strongly agree; (5) If you had to
walk on your own in a forest where there are jaguars, you would feel scared: strongly
disagree - strongly agree.
Knowledge about jaguars and depredation. For knowledge questions all answers were
coded as dichotomous variables, using 1 for a correct answer and 0 for an incorrect
answer and “do not know”. Knowledge was measured by seven questions so that the
scale ranged from 0 (minimum knowledge) to 7 (maximum knowledge). Questions were
posed as statements, three of which were correct and four incorrect. The statements
were: (i) The jaguar generally begins to consume its prey from the front while the puma
consumes the areas from the ribs back (correct); (ii) A jaguar‟s prey is usually hidden
and covered with leaves, while a puma‟s prey isn‟t (incorrect); (iii) The jaguar‟s prey
generally presents a bite mark at the base of the neck, while the puma‟s prey generally
has a bite on the throat (correct); (iv) The jaguar‟s track is longer than wide with thinner
and pointed toes, while the puma‟s track is slightly wider than long, with round toes
(incorrect); (v) A female jaguar produces in average 1 or 2 cubs every other year
(correct); (vi) Black jaguars are more dangerous to cattle than yellow jaguars
(incorrect); (vii) Jaguars kill more people every year in Brazil than do domestic dogs
(incorrect).
67
Perception of change in jaguar abundance. A five-point semantic differential scale was
used to measure belief concerning changes in jaguar abundance: Jaguar abundance on
your property is currently: decreasing a lot (coded as -2) to increasing a lot (coded as 2).
“No opinion” was coded as 0. This question was asked to rural residents only.
Perception of the regional economic situation. Two questions on a five-point semantic
differential scale with answers coded from -2 to 2 were used to measure perception of
economic situation: the economic situation of (respondent‟s main economic activity) in
the region is currently: very bad (-2) to very good (2), and The regional economic
situation is currently: declining fast (-2) to improving fast (2). Responses were
summated to obtain the respondent‟s total score, which ranged from -4 (very bad and
declining fast) to 4 (very good and improving fast).
Socio-demographic and property variables. Data on age, gender and education, and on
property size and land use (rural residents only) were obtained through fill-in-the-blank
questions. Education was coded as follows: illiterate (0); incomplete elementary school
(1); complete elementary school (2); incomplete high school (3); complete high school
(4); and college (5).
Photograph interpretation experiment
In addition to the main interview survey, a stratified random sample of respondents (55
ranchers in Amazonia and 15 in the Pantanal) was asked to interpret a set of 10
photographs of dead cattle. Four of the photographs presented evidence of predation by
jaguar, three presented evidence of predation by puma, and three did not present
evidence of predation by any particular predator (Appendix III). Photographs measured
approximately 13 x 21 cm and were laminated in transparent plastic. They were handed
68
to the respondents one by one in a random order. Respondents were given time to
examine each photograph before being asked the following open-ended question: What
was the cause of the death of this animal? Answers were coded as dichotomous
variables, using 1 for any kind of reference to predation by jaguar and 0 for other causes
and “do not know”. Whether the answer was correct or not was not taken into account,
as the experiment was not intended to measure respondents‟ knowledge about
depredation but rather to explore the effects of experience, attitudes and knowledge on
their interpretation of the photographs.
Data analysis
Statistical analyses were run in SPSS 16. First we used Exploratory Factor Analysis
(EFA) with a varimax rotation, a simple data reduction technique, to reduce the
interview data on perceptions of jaguar impact (8 questions) among cattle ranchers to a
smaller set of variables that explained key themes (represented by factors) in the data.
The meaning of the factors is interpreted by inspecting the pattern of correlations
between the original input variables and the output factors (the factor loadings). We
used the results of the EFA to produce two new „perception of jaguar impact‟ scales and
further refined these for inclusion in analyses. The internal consistency of the two scales
was examined using Cronbach‟s alpha reliability coefficients. Cronbach‟s alpha
measures the extent to which item responses (i.e. answers to survey questions) correlate
with each other (Cronbach 2004). The statistic can range from 0.00 to 1.00. A
Cronbach‟s alpha of 0.90 implies that scale is 90% reliable. By convention, an alpha of
0.70 is considered an adequate scale in human dimensions research (Vaske 2008).
69
General linear models, post-hoc Tukey tests and χ2 tests were used to examine broad
differences in survey variables between sample group means (Table 3.1). Pearson
correlation coefficients were used to examine the correlations between predictor
variables and perceptions of jaguar impact on human livelihoods among cattle ranchers
in Amazonia and the Pantanal. Next, two sets of General Linear Models (GLM) were
performed to develop a predictive model of perception of jaguar impact on human
livelihoods for ranchers in the two study areas. The first GLM analysis examined the
prediction of perceptions of jaguar impact on livestock and human safety. Instead of an
automated stepwise procedure, a structured comparison of specific models defined a
priori based on different explanation of perceptions of impact was carried out.
Sequential sum of squares (type I models) were used so that each effect in an individual
model adjusted only for preceding effects in the model. The following models were
assessed: (Model 1) negative experiences with jaguars: loss of livestock to jaguars and
acquaintance with people who had been injured by a jaguar, (Model 2) significant
predictors of Model 1 plus psychological variables: attitude towards jaguars, knowledge
about jaguars and depredation, sighting of jaguars in the wild, perception of increase in
jaguar abundance and perception of economic situation, and (Model 3) significant
predictors of Model 2 plus socio-economic variables: age, gender, education and
property size. The second set of GLM examined the effect of the socio-economic
variables on the negative experiences and psychological variables that significantly
affected perceptions of jaguar impact on human livelihoods in the first GLM analysis.
General linear models were also used to examine the effect of knowledge, attitudes, and
reported livestock loss on the number of photographs attributed to jaguar depredation in
the photograph interpretation experiment.
70
Results
Characteristics of sample
We conducted a total of 463 interviews including 52 in urban Pantanal, 50 in urban
Amazonia, 48 in rural Pantanal and 313 in rural Amazonia. The rural Amazonia sample
included 92 properties where agriculture was the main economic activity and no cattle
were raised. Table 3.1 presents additional information about the sample characteristics.
Perceptions of jaguar impact on human livelihoods
Refinement of the scales
The exploratory factor analysis extracted two factors from the 8 jaguar impact
statements, explaining 66% of the variance. Table 3.2 displays the factor loadings and
eigenvalues associated with each factor. The variables most strongly correlated with
Factor 1 (factor loadings = 0.45 to 0.83) were four statements on livestock depredation .
Factor 2 was most strongly influenced by four statements relating to impact on human
safety (factor loadings = 0.58 to 0.92). As a result, perception of impact could now be
described using two scales representing „perceived impact on livestock‟ (identified by
Factor 1) and „perceived impact on human safety‟ (identified by Factor 2). Cronbach‟s
alpha reliability coefficients were used to examine the internal consistency of each
scale. The reliability coefficients for the items in each factor were 0.88 (perceived
impact on livestock scale) and 0.86 (perceived impact on human safety scale).
71
Table 3.1. Summary statistics for each socio-economic variable and results of general linear models, Tukey tests and χ2test comparing variables among sample
groups (p < 0.05; groups indicated by the same letter were not significantly different).
Group (n)
Urban Rural
Non-ranchers Ranchers
Amazon Pantanal Amazon Amazon Pantanal
Variable (unit or scale) (n=50) (n=52) (n=92) (n=220) (n=48) dffactor dferror F / χ2
Age Range 21 to 74 18 to 66 17 to 77 17 to 82 27 to 77 4 452 10.39
(years) Mean ± SD 44.0 ± 13.0a 37.5 ± 10.5a 43.4 ± 13.1a 42.4 ± 13.1a 54.8 ± 11.8b p < 0.001
Gender Male % 76.0 59.6 72.8 73.3 100.0 4 - 22.98

(male/female) p < 0.001
Education Range 1 to 5 1 to 4 0 to 3 0 to 5 0 to 5 4 458 87.85

(0 to 5) Mean ± SD 2.68 ± 0.88ac 3.06 ± 0.94a 0.96 ± 0.55b 1.26 ± 0.76b 2.29 ± 1.43c p < 0.001
Property size Range - - 2.4 to 144 2.4 to 21,600 750 to 27,000 2 358 153.77
(ha) Mean ± SD 33.9 22.9a 298.8 1,581.1a 5,524.0 5,342.4b p <0 .001
72
Table 3.2. Factor loadings and eigenvalues for the two factors extracted using Exploratory
Factor Analysis to describe perceptions of jaguar impact on human livelihood. Variables
selected to construct the final scales are shown in bold.
Factor loadings
Variable Impact on Impact on
livestocka human safetyb
1. Damage associated with depredation ever 0.83 -0.17
caused by jaguars to you.
2. Damage associated with depredation ever 0.79 -0.15
caused by jaguars to your neighbours.
3. Risk of any damage associated with 0.75 -0.09
depredation to you in the next 12 months.
4. Risk of any damage associated with 0.71 -0.25
depredation to your neighbours in the next 12
months.
5. Number of people ever hurt by a jaguar in -0.12 0.92
the neighbourhood.
6. Number of household members ever hurt by -0.12 0.87
a jaguar.
7. Risk of you being hurt by a jaguar in the -0.20 0.74
next 12 months.
8. Risk of a household member being hurt by a -0.13 0.67
jaguar in the next 12 months.
Eigenvalues 3.27 1.83
Percentage of total variance 36.2 29.8
a
Perceived impact on livestock: Cronbach‟s alpha = 0.88
b
Perceived impact on human safety: Cronbach‟s alpha = 0.86
Perceived impact on livestock and human safety
In support of Hypothesis 1, cattle ranchers in Amazonia and the Pantanal differed in
their perceptions of the impact of jaguars on their livelihoods. Perceived jaguar impact
on livestock was significantly greater among ranchers in the Pantanal than in Amazonia
(Table 3.3). Owners of smaller landholdings (by area) tended to perceive a higher
impact of jaguar predation on their neighbours‟ ranches than they did on their own
ranch. This discrepancy between perceived impacts on neighbours‟ versus own ranch
decreased with increasing property size (Pearson correlation; r = - 0.522, p < 0.001)
(Figure 3.1).
73
Figure 3.1. Perception of jaguar impact on livestock: difference between neighbouring ranches
and own ranch (positive values mean higher impact on neighbouring ranches and zero means no
difference) and property sizes in Amazonia and Pantanal.
Perceived jaguar impact on human safety differed significantly between ranchers in
Amazonia and the Pantanal, with the latter holding the lowest perception of impact
among the five groups (Table 3.3). In rural Amazonia, cattle ranchers and farmers did
not differ in their perception of jaguar impact on human safety, in accordance with
Hypothesis 2. Urban residents in Amazonia had the strongest perceptions of jaguar
impact upon human safety, supporting Hypothesis 3.
74
Table 3.3. Summary statistics for perceptions of jaguar impact on livestock and human safety and results of general linear models and
Tukey tests comparing these variables among sample groups (p < 0.05; groups indicated by the same letter were not significantly different).
Group (n)
Urban Rural
Variable (unit or scale) (n=50) (n=52) (n=92) (n=220) (n=48) dffactor dferror F α
Perception of Range - - - 0 to 15 2 to 14 1 267 142.79 0.88
jaguar impact Mean ± SD 2.52 ± 2.61 7.58 ± 2.89 p < 0.001
on livestock
(0 to 20)
Perception of Range 8 to 20 4 to 16 4 to 16 6 to 18 2 to 20 4 458 31.58 0.86

jaguar impact Mean ± SD 15.84 ± 2.49a 12.23 ± 2.82b 12.59 ± 2.64b 11.82 ± 2.37b 10.29 ± 3.57c p < 0.001
on human
safety
(0 to 20)
75
Table 3.4. Summary statistics for experiences, attitudes and knowledge about jaguars, perception of change in jaguar abundance and perception of the
economic situation, and results of general linear models, Tukey tests and χ2test comparing these variables among sample groups (p < 0.05; groups
Group (n)
Urban Rural
Variable (unit or scale) (n=50) (n=52) (n=92) (n=220) (n=48) dffactor dferror F / χ2 α
Experience: Livestock Range - - - 0 to 10 0 to 10 1 267 31.19
loss % 0.87 2.21 p < 0.001
(% of holdings)
Experience: Yes % 4.0 9.6 8.7 8.6 55.0 4 - 0.012

Acquaintance with p < 0.001
injured person (yes/no)
Experience: Sighting Yes% 8.0 19.2 14.1 24.0 89.6 4 - p < 0.001
(yes/no)
Attitude Range -2 to 10 -4 to 10 -2 to 8 -10 to 9 -5 to 8 4 458 10.64 0.92

(-10 to 10) Mean ± SD 3.72 ± 2.62a 4.35 ± 3.06a 2.37 ± 2.12b 1.82 ± 3.12b 2.83 ± 3.32b p < 0.001
Knowledge Range 1 to 6 1 to 4 1 to 6 1 to 7 3 to 7 4 458 99.74 0.76

(0 to 7) Mean ± SD 2.51 ± 1.05a 2.06 ± 0.78ab 1.86 ± 1.17b 3.30 ± 1.22c 5.51 ± 0.87d p <0 .001
Perception of change Range - - -2 to 0 -2 to 1 0 to 2 2 358 480.68

in jaguar abundance Mean ± SD -1.18 ± 0.37a -0.72 ± 0.44b 1.32 ± 0.69c p < 0.001
(-2 to 2)
Perception of the Range - - -2 to 2 -3 to 2 -4 to 4 2 358 14.04 0.81

economic situation Mean ± SD -0.03± 0.54a -0.58 ± 1.08b -1.22 ± 2.54c p < 0.001
(-4 to 4)
76
The effect of region, land use, and place of residence
Negative experiences with jaguars
Reported livestock loss was greater in the Pantanal (2.2% of cattle holdings; Table 3.4) than
in Amazonia (0.9%). Not all of the reported livestock loss to jaguars was associated with
depredation. Twelve per cent of the ranchers in Poconé believed that jaguars caused cattle to
die by scaring them out of the „capões‟ (dry forest patches where cattle find refuge during
floods), into flooded areas, where they either drowned or became stuck in the mud
and starved to death. This purported indirect jaguar-induced mortality of livestock was not
confirmed by field observation.
Accounts of people who had been injured by a jaguar were most frequent in the Pantanal
(Table 3.4). Fifty five percent of the ranchers in the Pantanal claimed to know personally
someone who had been injured by a jaguar, against 8.6 and 8.7% of the ranchers and farmers
in Amazonia, respectively. Among urban residents, 9.6% in the Pantanal had acquaintance of
an injured person against 4% in Amazonia. It seemed that the great majority of accounts
referred to just one particular story of injury caused by a jaguar in each study region, both
happened over 5 years earlier. In Amazonia, the victim would have been attacked during the
night, while sleeping in his hammock, whereas in the Pantanal he would have been attacked
when unwittingly approaching a carcass upon which the jaguar was feeding. SM personally
met and interviewed the two victims who confirmed the stories.
Psychological variables
Table 3.4 displays the differences in experiences, attitudes and knowledge about jaguars, as
well as perception of change in jaguar abundance and perception of the economic situation
across groups. Attitudes towards jaguars differed significantly between places of residence,
77
with urban citizens having more positive attitudes towards the species. In rural Amazonia,
ranchers and farmer did not differ in their attitudes to jaguars, in support of Hypothesis 2.
Ranchers in the Pantanal knew more about jaguars and depredation than did ranchers in
Amazonia. Amazon ranchers were, in turn, more knowledgeable about jaguars and
depredation than farmers in the same region. Urban residents in Amazonia and the Pantanal
had lower levels of knowledge than ranchers and did not differ from each other in their levels
of knowledge.
A far greater percentage of people reported seeing a jaguar in the wild in rural Pantanal
(90%) than anywhere else. More urban residents in the Pantanal had seen a jaguar than had
urban residents in Amazonia. In rural Amazonia, more ranchers had seen a jaguar in the wild
than farmers. While 27% of the cattle ranchers in the Pantanal had lost count of the number
of times they had seen a jaguar, 82% of those people in other groups who had ever seen a
jaguar, had seen one only once (and typically only a glimpse). Sighting was therefore recoded
into a binary variable.
Perceptions regarding both changes in jaguar abundance and the regional economic situation
differed significantly across the three rural groups. Cattle ranchers in the Pantanal reported
the largest increases in jaguar numbers and the gloomiest economic situation, whereas
Amazon farmers held the least negative perceptions of the economic situation and believed
that jaguar numbers were declining in the region.
Socio-economic variables
As shown in Table 3.1, the sample of Pantanal ranchers differed from the other groups for
78
being older and composed exclusively by men. In contrast, in urban Pantanal more than 40%
of the respondents were female. Urban residents had higher levels of education than rural
residents. Ranchers in the Pantanal were more educated than ranchers in Amazonia. In rural
Amazonia, ranchers and farmers did not differ significantly in their levels of education.
Property size varied greatly across groups: the average ranch in the Pantanal was much larger
than the average ranch in Amazonia, which in turn was larger than the average farm in the
region.
Predicting perceptions of jaguar impact on human livelihood
Ranchers in Amazonia
Table 3.5 shows that, as expected, reported livestock loss was a significant predictor of
perceived impact of jaguars on livestock (Model 1). In support of Hypothesis 4, cattle
ranchers who reported higher losses of cattle to jaguars held stronger perceptions of the
impact of jaguars on livestock. However, in accordance with Hypothesis 5, the inclusion of
psychological variables (Model 2) accounted for 25% of the variation (R2 = 0.444 against
0.188) over and above reported livestock loss. Attitude to jaguars and knowledge about
jaguars and depredation emerged as significant predictors of perception of jaguar impact on
livestock. Respondents who had more positive attitudes to jaguars and were more
knowledgeable about the species perceived a lower impact of the jaguar on livestock.
Attitudes interacted with knowledge in their effect on perception of jaguar impact on
livestock: the higher the knowledge score, the smaller the effect of attitude on the perception.
At higher knowledge scores (>5), attitudes did not predict perceived impact on livestock
(GLM; β = -0.08, p = 0.399). Figure 3.2 illustrates the interaction between attitudes and
knowledge at lower knowledge levels. Table 3.5 also shows that socio-economic variables
were not significant predictors of perception of jaguar impact on livestock and their inclusion
79
to the model (Model 3) accounted for only 0.6% of the variation in perception.
Figure 3.2. Relationship between perception of jaguar impact on livestock and attitudes to jaguars at
different levels of knowledge (scores 1 and 2: R2 = .460 and scores 3 and 4: R2 = .198) among cattle
ranchers in Amazonia. The relationship was not significant for knowledge scores 5 to 7.
Table 3.5. General linear model predicting perception of jaguar impact on livestock in Amazonia.
Results shown in bold are significant.
Model Predictor included R2 B SE t p

1 Reported livestock loss 0.188 1.393 0.196 7.110 < 0.001
Acquaintance with injured people -0.044 0.566 -0.078 0.938
2 Reported livestock loss 0.444 0.833 0.175 4.764 <0.001
Attitude -0.674 0.100 -6.704 <0.001
Knowledge -0.559 0.136 -4.115 <0.001
Attitude x knowledge 0.106 0.031 3.475 0.001
Sighting -0.245 0.318 -0.770 0.442
Perception of change in abundance 0.283 0.306 0.924 0.356
Perception of economic situation -0.023 0.124 -0.184 0.854
Attitude -0.658 0.100 -6.581 <0.001
Knowledge -0.557 0.139 -3.996 <0.001
Age 0.010 0.011 0.922 0.358
Gender 0.208 0.312 0.668 0.505
Education 0.184 0.182 1.012 0.313
Property size <-0.001 <0.001 -0.017 0.987
80
As displayed in Table 3.6, negative experiences did not affect perception of jaguar impact on human
safety in Amazonia (Model 1), in contrast to Hypothesis 4. Perception of jaguar impact on human
safety was predicted only by attitude and knowledge (Model 2): the more negative the attitudes
towards jaguars and the lower the knowledge about jaguars and depredation, the stronger the
perception of the impact of jaguars on human safety, in support of Hypothesis 5. Socio-economic
variables were not significant predictors of the perception of jaguar impact on human safety and their
inclusion to the model (Model 3) accounted for only 0.2% of the variance in perception.
Table 3.6. General linear model predicting perception of jaguar impact on human safety in Amazonia.

1 Reported livestock loss 0.011 0.241 0.197 1.221 0.223
Acquaintance with injured people 0.521 0.569 0.915 0.361
2 Attitude 0.110 -0.221 0.111 -1.984 0.049
Knowledge -0.352 0.152 -2.311 0.022
Sighting 0.199 0.365 0.546 0.586
Perception of change in abundance 0.352 0.352 -0.124 0.902
Perception of economic situation 0.135 0.142 0.948 0.344
3 Attitude 0.112 -0.191 0.051 -3.757 <0.001
Knowledge -0.332 0.137 -2.415 0.017
Age -0.010 0.012 -0.821 0.412
Gender -0.200 0.357 -0.559 0.577
Education -0.157 0.207 -0.759 0.448
Property size <0.001 <0.001 0.387 0.699
Nonetheless, Table 3.7 shows that the socio-economic variables age, gender and property size
were significant predictors of attitude and knowledge, supporting Hypothesis 6. Younger
respondents held more positive attitudes towards jaguars, and male respondents who owned
larger properties had more knowledge about jaguars and depredation.
81
Table 3.7. General linear models of the effect of socio-economic variables on the factors that
significantly affected perceptions of the impact of jaguars on human livelihoods in Amazonia. Results
shown in bold are significant.
Dependent Predictor R2 B SE t p
variable
Reported Age 0.036 0.004 0.004 0.970 0.333
livestock loss Gender 0.069 0.129 0.537 0.592
Education 0.105 0.075 1.404 0.162
Property size <0.001 <0.001 0.496 0.620
Attitude Age 0.061 -0.054 0.016 -3.313 0.001
Gender 0.728 0.484 1.541 0.125
Education -0.488 0.278 -1.792 0.074
Property size <0.001 0.000 0.575 0.566
Knowledge Age 0.176 -0.006 0.006 -0.935 0.351
Gender 0.694 0.173 4.0.24 <0.001
Education 0.032 0.101 0.317 0.752
Property size <0.001 <0.001 5.148 <0.001
Ranchers in the Pantanal
As expected, livestock loss significantly affected perceptions of jaguar impact on livestock
also in the Pantanal (Model 1; Table 3.8). Perceived impact of jaguars on livestock was
stronger among ranchers who had reportedly lost more cattle to jaguars, which supports
Hypothesis 4. However, attitude and perception of economic situation emerged as significant
predictors of perception of jaguar impact on livestock, in accordance with Hypothesis 5.
Attitude and perception of economic situation (Model 2) accounted for 40.6% of the variation
in perception (R2 = 0.668 against 0.262) over and above reported livestock loss. The inclusion
of age, education and property size (Model 3) accounted for only 0.6% of the variation in
perception of jaguar impact on livestock and none of the socio-economic variables was a
significant predictor of perception.
82
Table 3.8. General linear model predicting perception of jaguar impact on livestock in the Pantanal.

Attitude -0.285 0.088 -3.256 0.002
Knowledge 0.114 0.309 0.370 0.713
Sighting 0.859 0.893 0.961 0.342
Attitude -0.237 0.098 -2.425 0.020
Age 0.031 0.025 1.213 0.232
Education 0.072 0.188 0.385 0.703
Property size <-0.001 <0.001 -0.273 0.786
Regarding the perceptions of jaguar impact on human safety, negative experiences with jaguars were
not significant predictors (Model 1; Table 3.9), in contrast to Hypothesis 4. The only significant
predictor among the psychological variables was knowledge about jaguars and depredation (Model 2).
In partial support for Hypothesis 5, ranchers who were more knowledgeable about jaguars perceived a
lower impact of jaguars on human safety. Socio-economic factors (Model 3) did not affect perceptions
of the impact of jaguars on human safety. However, in partial support of Hypothesis 6, age affected
attitude to jaguars (Table 3.10): older respondents had more negative attitudes towards jaguars.
83
Table 3.9. General linear model predicting perception of jaguar impact on human safety in the
Pantanal. Results shown in bold are significant.

1 Reported livestock loss 0.005 -0.054 0.495 -0.109 0.914
2 Attitude 0.412 -0.186 0.142 -1.307 0.198
Knowledge -1.981 0.501 -3.956 <0.001
Sighting -1.161 1.443 -0.804 0.426
Perception of economic situation 0.301 0.204 1.472 0.148
3 Knowledge 0.416 -2.249 0.504 -4.462 <0.001
Age 0.049 0.036 1.365 0.180
Education 0.359 0.302 1.919 0.241
Property size <0.001 <0.001 0.344 0.732
Photograph interpretation experiment
In the Pantanal, respondents assigned between four and eight of the ten photographs to jaguar
depredation (mean 5.53 ±SD 1.187, n = 15), while in Amazonia, respondents assigned
between zero and ten photos (mean 5.07 ±SD 1.704, n = 56). The number of photographs
assigned to jaguar predation in the Pantanal was not predicted by reported livestock loss,
attitudes towards jaguars, or knowledge of the species. In contrast, in Amazonia, while the
number of photographs was also not predicted by livestock loss (β = .015, p = 945, Eta =
.000) or knowledge (β = .055, p = .764, Eta = .002), it was predicted by attitudes (β = -.47, p
< .001) [Adj. R2 = .31, F = 7.201]. Respondents who had more negative attitudes towards
jaguars assigned a larger number of photographs to jaguar predation. There was a significant
interaction between attitudes and knowledge in Amazonia (β = .047, p = .046); the effect of
attitude on the number of photographs assigned to jaguar predation was stronger when
knowledge about jaguar and depredation was lower, with attitudes influencing the number of
photographs assigned to jaguar depredation for knowledge scores 1 and 2 only.
84
Table 3.10. General linear models of the effect of socio-economic variables on the factors that
significantly affected perceptions of the impact of jaguars on human livelihoods in the Pantanal.
Dependent variable Predictor R2 B SE t p

Reported livestock loss Age 0.004 -0.017 0.014 -1.258 0.215
Education -0.065 0.113 -0.571 0.571
Property size <-0.001 0.000 -1.173 0.247
Atttitude Age 0.057 -0.095 0.038 -2.489 0.017
Education 0.117 0.322 0.364 0.717
Property size 0.000 0.000 1.869 0.058
Knowledge Age 0.003 0.006 0.011 0.515 0.609
Education 0.028 0.091 0.306 0.761
Property size <0.001 0.000 1.538 0.131
Perceptions of change in Age 0.007 -0.002 0.009 -0.257 0.798
jaguar abundance Education 0.022 0.074 0.298 0.767
Property size <-0.001 0.000 -0.627 0.534
Perceptions of economic Age 0.000 -0.030 0.031 -0.968 0.338
situation Education -0.048 0.262 -0.181 0.857
Property size 0.000 0.000 1.728 0.091
We examined the drivers of perceptions of jaguar impact on human livelihood and identified
two distinct dimensions in our sample: impact on livestock and impact on human safety.
These can best be addressed separately. We found that perceived jaguar impact on human
livelihoods are hugely influenced by region and place of residence, with perceived impact on
livestock being greater in the Pantanal, and perceived impact on human safety being greater
in the Amazon, and in urban than rural areas. Reported livestock loss is not the only predictor
of perceived jaguar impact on livestock. Rather, experience of losses may act in combination
with attitudes, knowledge and perceptions of the economic situation, to determine how
people perceive the impact that jaguars have on their livestock. Perhaps surprisingly,
professed acquaintance with somebody injured by a jaguar was not a factor in perceptions of
jaguar impact on human safety, and urban residents were more afraid of jaguars than ranchers
85
and farmers. Again perceptions of impact were governed by knowledge about jaguars and
depredation and by attitudes towards the species.
In some cases, a negative impact or threat is perceived even in the absence of any livestock
loss to jaguars or jaguar attack on people. Attitudes and knowledge, in turn, are influenced by
a person‟s age, gender, property size (reflecting relative wealth), and place of residence. This
is evidence that ecological, cultural and socio-economic context all play roles in the
relationships between reality and perceptions behind the conflicts between people and
jaguars.
As expected, perception of jaguar impact on livestock was predicted by livestock loss.
However, the perceptions held by ranchers regarding trends in jaguar abundance and in the
regional economic situation were also related to their perception of jaguar impact on
livestock in the Pantanal. According to Crawshaw (2002), jaguar abundance has increased in
the Pantanal since the 1980s, following the retraction of cattle ranching in areas that were
recurrently hit by severe floods that began in 1974. Hearing that jaguar numbers are growing,
and perhaps even seeing evidence of them more frequently, is likely to cause ranchers to
perceive an increasing threat to their cattle. As for the perceived decline in the regional
economic situation, over recent decades, growing competition within the cattle industry,
higher taxes, and generational land-splitting have rendered cattle-ranching less profitable in
the Pantanal. A perceived decline in the financial margins of profits from cattle ranching may
decrease ranchers‟ tolerance of jaguar depredation on their cattle.
Attitudes to jaguars were important predictors of conflict over livestock in both Amazonia
and Pantanal, whereas knowledge appeared to be relevant in Amazonia only. While it is
86
obvious that livestock loss to jaguars is likely to engender negative attitudes towards the
species, our photograph interpretation experiment indicated that, in Amazonia, attitudes
towards jaguars, and poor levels of knowledge about them, prejudice subsequent judgments
in ways that have the potential to become self-reinforcing. Ranchers who had more negative
attitudes towards jaguars and less knowledge about them ascribed jaguar depredation as the
cause of death to a larger number of photographs, strongly suggesting that their judgments
would be similarly prejudiced in the field. Such a phenomenon could in turn affect the
number of reports of livestock lost to jaguars, leading to a vicious circle of exaggerated
perceptions. An interesting question might be whether this could work the other way round, if
conservation-minded people underestimated depredation.
The owners of smaller properties in Amazonia had worse attitudes towards and poorer
knowledge about jaguars and these in turn affected their perceptions of jaguar impact on both
livestock and on human safety. The hostility might be partly attributable to smaller holdings
having more marginal economic viability, but the finding that attitudes did not differ between
small landowners who did or did not raise cattle suggests that their heightened hostility is
more likely associated with their poorer education. Furthermore, the owners of small
properties tended to believe that depredation problems were more serious on neighbouring
ranches than on their own. In a landscape dominated by small properties, a patchy
distribution of depredation may give disproportionate weight to rumours, especially if a few
properties may be severely and repeatedly affected. Certainly, the impact of a few episodes
that take on widespread perceptions was vividly illustrated in the context of the threat posed
by jaguars to human safety: in each study region there was one instance of a (non-fatal)
jaguar attack – in both cases more than five years earlier – and these were repeatedly cited as
the basis of negative perceptions. Macdonald (1987) describes similarly how accounts of
87
serious attacks by foxes, Vulpes vulpes, on lambs in the north of England often turned out to
refer to the folkloric experiences of other people long ago. In sum, perceptions of conflict
with jaguars on small properties are shaped primarily by what is heard from other people and
not by personal experience.
By contrasting Amazonia and Pantanal, rural and urban residents, cattle ranchers and crop
farmers, and by surveying a wide range of property sizes, this study is the first attempt to
understand what is universal in human-jaguar relationships and what is context-specific.
Attitudes to jaguars, for instance, affected perceptions of human-jaguar conflict in both
Amazonia and the Pantanal, while perceptions of the economic situation were relevant only
among ranchers in the Pantanal. Insights of this sort can guide conservation policy regarding
the likelihood that interventions found to be successful in one region or socio-economic and
cultural context are likely to be transferable elsewhere.
Further implications for conservation
Efforts to protect jaguars have generally been based on the assumption that people are
motivated to kill jaguars in retaliation for livestock-raiding. According to this rationale, if
ways could be found effectively to reduce jaguar predation on cattle (e.g., adoption of certain
husbandry practices, use of electric fences, aversive conditioning and translocation), then
persecution by cattle ranchers should subsequently decline (Hoogesteijn 2000). Interventions
to decrease depredation combined with monetary compensation to ameliorate the financial
costs of lost livestock, are intended to alleviate the economic burden imposed on ranchers by
sharing the land with jaguars. The assumption is that this compensation will reduce the
motivation for ranchers to kill them. However, the finding that perceived impact of jaguars on
human livelihood is neither always nor only determined by livestock loss, and that opinions
88
are much influenced by attitudes towards jaguars and knowledge of the species, suggest that
conservation interventions should not assume a simple cause and effect relationship. Hitching
conservation interventions too simply to the logic of a financial cost-benefit analysis may
miss the deep-seated, and sometimes prejudiced, human dimension. Kruuk (2002) suggests
that hatred of big predators has an evolutionary basis, and it is clear that people worldwide
invest disproportionately to both the costs and the evidence in killing predators worldwide.
Fox control on British farms seems to be related to historical damage, fears of what might
happen if control ceased, and tradition, rather than to recent losses to foxes (Macdonald
1984). A study of mounted foxhunting in Wiltshire, United Kingdom, found widespread
belief among farmers that foxes should be controlled everywhere, because they were too
numerous, even among farmers for whom the fox was not a personal pest (Baker and
Macdonald 2000). In the case of cheetahs, Acinonyx jubatus, in Namibia, for example, the
numbers killed by farmers were reduced markedly by both mitigation and education, but they
nonetheless persisted in killing some (Marker et al. 2010). Macdonald et al. (2010) review
cases of deep-seated and often misplaced loathing of big cats, and Dickman (2008) explores
the consequences for conservation of cultural relativism. Our results add to the evidence that
conservation interventions should take account of these human dimensions. The social
sciences offer various techniques for understanding these human dimensions (discussed by
Macdonald et al. 2010). For example, the theory of planned behaviour (TPB – Ajzen 1985)
and attitude-to-behavioural process models (ABPMs – Fazio et al. 1982) have been proposed
to explain these relationships (see CHAPTER 4).
This study reveals the relatively great importance of attitudes and knowledge in the perceived
conflict with jaguars on smaller properties (where owners are generally poorer and less
89
educated). Small properties have been relatively overlooked in jaguar research and
conservation in Brazil. Approximately two thirds of publications on jaguar ecology and
conservation produced in Brazil report studies conducted in the Pantanal, where the average
ranch is several thousand hectares in size. In Amazonia, however, properties up to 100
hectares are responsible for 30% of the deforestation (Fernside 2003). Small landowners in
Amazonia are likely less prone to comply with the law that prohibits the clearance of more
than 20% of the property area. Collectively, small properties form expanses of highly
degraded jaguar habitat, where low availability of native prey and high human density can
increase the chances of a jaguar taking cattle or encountering people, which enhances the
perception of conflict (and certainly decreases the jaguar‟s prospects of survival). These
landscapes merit higher priority in conservation strategy, and would benefit particularly from
education campaigns focusing on the realities of risk and using the leverage of influential
community institutions and informal social networks – in short, the very conduits through
which inaccurate perceptions of jaguars are currently fuelled.
90
Chapter 4 – From perceptions to persecution
CHAPTER 4
FROM PERCEPTIONS TO PERSECUTION: CONFLICTS
BETWEEN PEOPLE AND JAGUARS
All human actions have one or more of these seven causes: chance, nature, compulsions,
habit, reason, passion, desire.
Aristotle
Abstract
The killing of jaguars by ranchers in Amazonia and the Pantanal is a major threat to the
species. We used the Theory of Planned Behaviour (TPB; Ajzen 1985) to examine the role of
ranchers‟ perceptions, norms, attitudes and intentions concerning jaguar killing, in
determining their jaguar-killing behaviour. We also investigated the influence of: 1)
descriptive norm and social identity on ranchers‟ intention to kill jaguars on their properties;
and 2) the effect of perceptions of jaguar impact on human livelihoods (livestock and human
safety), and of property size, on the variables that influence intention to kill. Results based on
interviews with 268 cattle ranchers indicated that the perceived impact of jaguars on livestock
is not the only predictor of a rancher‟s intention to kill jaguars. Fear, personal and social
motivations, and internal and external barriers to killing jaguars can also influence jaguar
killing. The relative importance of these factors in determining intention to kill varies with
region and affluence. We recommend ways of deterring jaguar killing behaviour through
communication interventions. In addition to the economic and legal incentives that have
already been considered, effective strategies to protect jaguars on privately owned land will
need to address the social and psychological factors that determine the killing of jaguars by
ranchers. Conservationists need to find and support ways to make jaguar killing not only
unprofitable and illegal, but also socially and personally unacceptable.
91
Introduction.
The most urgent issue in jaguar (Panthera onca) conservation is the killing of jaguars by
humans (Nowell and Jackson 1996; Zeller 2007). Because jaguars are renowned for preying
on livestock (Hoogesteijn 2000), one might assume that people kill jaguars in an effort to
reduce economic losses associated with livestock predation. Conservationists have
approached jaguar killing within the framework of human-wildlife conflicts (Rabinowitz
2005) and to date research and conservation efforts have focused on the ecological (Azevedo
and Murray 2007; Cavalcanti 2008) and economic (Silveira et al. 2006) dimensions of the
conflict respectively. Little has been done to understand the link between jaguars killing
livestock and people killing jaguars. The killing of jaguars may not be strictly retaliatory and
might have motivations besides the economic. If we are to curb the killing of jaguars, first we
have to understand the underlying causes of this behaviour. We used an approach that went
beyond the usual human-wildlife conflict framework to examine the reasons – both related
and not related to perceived direct impacts – for ranchers killing jaguars in Amazonia and
Pantanal.
Investigations of human-wildlife conflict from a human perspective have focused mainly on
attitudes (Karlsson 2007; Manfredo 2008). Attitude studies are popular because they are
relatively easy to conduct and interpret, and they help anticipate behaviour and identify tools
for changing it. Unfortunately, in a review of 19 papers on this topic, Browne-Nuñez and
Jonker (2008) found only one that met their criteria for professionalism, which included
defining the attitude concept, describing a theoretical framework to explain the relationship
of attitudes to behaviour, and addressing the reliability and validity of the findings, and only
nine papers reported a pre-test of the instrument. The few studies that have attempted to
address human-jaguar conflict from the human side have assessed attitudes towards jaguars
92
among people that were directly involved in conflict with jaguars over livestock (Conforti
and Azevedo 2003; Palmeira and Barrela 2007; Zimmermann et al. 2005). However, the
relationship between attitudes and behaviours may not always be strong or direct.
Researchers attempting to use attitudes to predict and prevent unwanted behaviours often do
so without considering exactly how attitudes relate to behaviours (McCleery et al. 2006).
Moreover, by limiting the study of attitudes to the context of human-wildlife conflict, factors
involved in the decision to kill a certain species, but not directly related to the impact of that
species on human livelihood, may be overlooked. Theoretical frameworks have been
developed by social scientists to predict human behaviour from attitudes (Ajzen 1985; Fazio
1986; Fishbein and Ajzen 1975), and these could provide a useful structure for studying the
killing of jaguars. Key aspects of such an approach are the proper measurement of attitudes
and inclusion of additional explanatory variables. In this study, we develop a framework
based on the Theory of Planned Behaviour (TPB) (Ajzen 1985) to explore the relationships
between landowners‟ perceptions of jaguar impact on human livelihood and their own jaguar
killing behaviour. We incorporate factors not related directly to the impact of jaguars on
human livelihood, such as social motives for killing jaguars and perceived barriers to doing
so.
Linking attitudes to behaviour: the Theory of Planned Behaviour
The TPB proposes that human behaviours are governed not only by personal attitudes, but
also by social pressures and perceived control over one‟s own behaviour. According to the
TPB, the most proximal determinant of a person‟s behaviour is their intention to engage in
that behaviour (Figure 4.1). In turn, behavioural intentions are influenced by three main
factors: attitudes towards the behaviour, subjective norms and perceived behavioural control
(PBC). Attitude scores reflect an individual‟s overall positive or negative evaluation of
93
performing the behaviour. Subjective norms represent an individual‟s perception of whether
people important to them would approve of them performing the behaviour. Perceived
behavioural control reflects the extent to which an individual perceives the behaviour to be
under their volitional control. Thus, according to TPB, people who have positive attitudes
towards killing jaguars, think that there is normative support for killing jaguars, and perceive
that they can easily kill jaguars (or pay someone else to do it for them), should have strong
intentions to kill jaguars. In addition, to the extent that PBC is a proxy for the actual control
(accurately accounts for both the internal factors [e.g. knowledge, skills, courage] and
external factors [e.g. legal barriers, money, equipment, help from others]) needed to perform
the behaviour, it may also have a direct impact on behaviour (dotted arrow in figure 4.1).
While the great majority of attitudinal studies conducted in wildlife or conservation research
have assessed attitudes about the species in question (Bath et al. 2008; Bruskotter et al. 2007;
Lindsey et al. 2005), the TPB recognizes that attitudes will not predict behaviour unless they
are measured with corresponding levels of specificity: attitudes about objects (such as
jaguars) will not necessarily predict behaviours (such as killing jaguars). In order for attitudes
to predict behaviour, the attitude and behaviour must correspond on four levels of specificity:
action, target, context and time. In this study, attitudes (and other TPB variables) are related
to the specific rancher‟s behaviour of killing (action) jaguars (target) on their properties
(context) in the near future (time). Attitude specificity has arguably been one of the most
significant refinements in improving the applicability of the attitude concept (Manfredo,
2008).
A multitude of variables can be related to or influence attitude, subjective norm and PBC:
personality, mood, emotions, general attitudes, age, gender, education, socio-economic status,
94
nationality, religion, knowledge, past experience and so forth. The TPB recognises the
potential significance of such background factors (Figure 4.1). Inclusion of additional
influential variables may complement the TPB model, thereby deepening our understanding
of what determines performance of a particular behaviour. However, there is a vast array of
potentially relevant background factors, making it difficult to know which should be included
in the model without a theory to guide selection in the behavioural domain of interest.
Figure 4.1. Diagrammatical representation of the Theory of Planned Behaviour.
Extending the Theory of Planned Behaviour
Although reviews and meta-analyses have demonstrated broad support for the basic TPB
(e.g., Armitage and Conner 2001), it is acknowledged that for some behaviours and contexts,
the inclusion of other variables may increase the predictive utility of the model. Descriptive
norms and social identity are examples of such variables.
Descriptive norms reflect an individual‟s perception of whether other people perform the
behaviour in question (Cialdini et al. 1990). Descriptive norms describe what is typical or
normal, and motivate action by indicating what is likely to be effective, adaptive, and
appropriate action (White et al. 2009). For instance, a rancher who believes that all his
95
neighbours kill jaguars will feel motivated to do the same. In contrast, the subjective norm
component of the TPB is a social injunctive norm, reflecting an individual‟s perceptions of
whether significant others approve of the individual performing the behaviour – in other
words, social pressure. Social injunctive norms motivate action by offering potential social
rewards or punishments for either engaging or not engaging in a particular behaviour.
Social identity is that component of an individual‟s concept of himself that is derived from
his knowledge of group membership, and the value and emotion attached to that membership
(Tajfel 1981). According to the social identity theory, people define and evaluate themselves
in terms of distinct social categories (e.g. rancher, pantaneiro). By allocating himself a
particular social identity an individual is encouraged to accentuate both the similarities
between himself and other group members, and the differences between himself and people
outside the group (Fielding et al. 2008). A social identity approach assumes that if a certain
behaviour, for example killing a jaguar, is centrally linked to a social identity, then that
behaviour will be influenced by the norms of that social group rather than by the expectations
and desires of generalized others.
To date, the TPB and extensions of the model have been used successfully to understand the
factors involved in producing a range of behaviours such as recycling (e.g., Cheung et al.
1999), composting (Taylor and Todd 1995), energy conservation (Harland et al. 1999), water
conservation (Harland et al. 1999), adoption of sustainable agriculture practices (e.g., Beedell
and Rehman 2000; Fielding et al. 2005, 2008) and sport hunting (Hrubes et al. 2001).
However, to our knowledge there has been no research employing the TPB to understand the
determinants of persecution in a human-wildlife conflict context.
96
Modelling jaguar persecution
This study assessed an extended TPB model as a framework for understanding and predicting
the intention to kill jaguars among livestock owners on the Amazon deforestation frontier and
in the Pantanal. We assessed perceptions of conflict with jaguars, and the standard TPB
constructs of attitude, subjective norm and PBC - all in relation to killing jaguars. Then we
examined their effect on the decision to kill jaguars. Perceptions of conflict with jaguars
(more specifically, perception of jaguar impact on livestock, and on human safety), were
expected to affect intention: people who perceive that their livestock, or safety, is threatened
by jaguars are more likely to intend to kill jaguars (Hypothesis 1). However, we hypothesized
that the killing of jaguars by ranchers is not strictly retaliatory, and that by taking into
account factors that are not related directly to livestock depredation, or threat to human safety
(i.e. subjective norm and PBC), the TPB would offer a more predictive model of intention to
kill jaguars (Hypothesis 2).
During the preliminary stages of this study we gathered anecdotal evidence of a social
dimension to jaguar killing. Evidence included photographs of hunted jaguars and jaguar paw
trophies displayed prominently in ranchers‟ living rooms, boastful stories told by ranchers
about their bravery in hunting jaguars, and repeated mention that killing jaguars is something
that everybody does, or have done for generations, as an important element of the local
tradition. Therefore, including descriptive norm and group identity in the model was expected
to increase significantly the predictive utility of the TPB model (Hypothesis 3): ranchers who
believe that other ranchers kill jaguars and identify themselves with those ranchers will have
stronger intentions to kill jaguars themselves.
97
Attitude towards a particular behaviour is measured using experiential items (i.e. how it feels
to perform the behaviour e.g. unpleasant-pleasant) and instrumental items (i.e. whether the
behaviour achieves something e.g. detrimental-beneficial and worthless-useful) (Ajzen and
Fishbein 1980). Retaliatory killing of jaguars is instrumental in nature: it is expected to bring
the benefit of decreased livestock loss or increased human safety. Therefore, we predicted
that attitudes towards killing jaguars would be affected by perceptions of jaguar impact on
livestock and on human safety. Indeed, in the TPB, background factors are assumed to
influence intentions and behaviour indirectly by affecting attitude, norms and PBC: we
predicted therefore that perceptions of conflict with jaguars would influence jaguar
persecution indirectly by affecting attitude towards the killing of jaguars (Hypothesis 4).
Commercial, sport, and recreational hunting are prohibited in Brazil. In 1998 the killing of
jaguars was rendered a crime under Brazil‟s Environmental Crimes Act. Nonetheless, in
remote areas of rural Amazonia and Pantanal people do not expect to be penalised for
breaking the law. None of the respondents in this study had ever heard of a case of
prosecution for killing jaguars. Expectation of law enforcement and perceived risk associated
with breaking the law are probably lower among large, affluent, and consequently, more
influential ranchers. Furthermore, jaguars are more likely to be present on larger properties
and their landowners more likely to have the resources for killing jaguars. In summary, the
larger the property the more volitional control its owner should have over killing jaguars.
Therefore, we hypothesized that perceived behavioural control increases with property size
(Hypothesis 5).
While hosting relatively large populations of both cattle and jaguars (Cavalcanti et al. 2010),
the two study areas differ distinctly in some socioeconomic aspects. The Amazon frontier
98
was colonized recently, starting at most 35 years ago, by people from different parts of
Brazil, including areas where jaguars had long been extirpated. Small family-owned farms
coexist with large commercial cattle ranches on the Amazon frontier. In contrast, Northern
Pantanal is home to a few traditional families that own large ranches and have been raising
cattle in the region for generations. In communities where residents come from different parts
of a country, for example, in frontier areas or along the forest margins, there may be no
consensus on what are acceptable or unacceptable behaviours towards wildlife. In other
words, descriptive norms are unclear. Even if there is some agreement, many people may not
care about the opinions of others, and so the subjective norm and group identity are less
likely to influence behaviour. In contrast, where traditional communities are intact and there
is collective thinking and articulation about what constitutes acceptable practice, subjective
norms and group identity may be sufficiently powerful to maintain practices that are illegal
but socially encouraged. Therefore, we hypothesized that the relative importance of the social
determinants of jaguar killing and perceived control over this behaviour would differ between
Amazonia and Pantanal (Hypothesis 6).
Methods
Study areas and participants
The Amazonia survey was carried out in Alta Floresta, in the north of the State of Mato
Grosso. The Pantanal study was conducted in the districts of Cáceres and Poconé, also in the
State of Mato Grosso.
The sample unit was the property, with either owner – but preferably the husband - from each
surveyed (in a few large ranches, the ranch manager was interviewed when neither of the
owners was available). The association of rural workers of Alta Floresta, Cáceres and Poconé
99
provided a list of members, with landowners‟ names and contact information, from which we
could randomly select ranches. However, the lists were not comprehensive, so we used a
combination of sample selection techniques to maximise randomness and representativeness.
Google Earth imagery (Google Inc 2009) was used to map all roads within the study areas
(available road maps do not include minor roads). Then we used a systematic sampling
strategy, selecting every other property along randomly chosen roads. In Amazonia, ranches
were easily accessible by land and all interviews were carried out on site. In the Pantanal
some properties near the Transpantaneira highway could be visited but many others were
inaccessible overland and so some Pantanal ranchers were interviewed in their second
residences in town.
Procedures and measures
Interviews were conducted between March and October 2007 in Amazonia and between
February and May 2008 in the Pantanal. The killing of jaguars is illegal and can be a sensitive
issue for some ranchers. Therefore, some measures were taken to avoid response and
interviewer biases. All interviews were conducted by SM, who is Brazilian and acquainted
with the local culture and parlance in both study areas. SM was always accompanied by a
male local field assistant during the interviews. SM was as objective and neutral in
appearance and behaviour as possible. We anticipated that some ranchers would feel inclined
to omit their negative perceptions, attitudes, intentions and behaviours towards jaguars. Given
that people attempt to provide answers consistent with the ones they have already given in the
survey (Sudman et al. 1996; Tanur 1992), questions were asked in the order in which we
expected the likelihood of a deceiving answer to increase: questions of perceptions and
attitudes first, then intention and, finally, behaviour. All TPB components were measured
according to the recommendations of Ajzen and Fishbein (1980) and Ajzen (1985). Where a
100
TPB component was assessed using multiple questions, we created average scales to
summarise that component. The variables used in the analyses are described in more detail
below.
Socioeconomic variables. Data on age and property size were obtained through fill-in-the-
blank questions. Property size served as a proxy for affluence. Gender was also recorded.
Perceived impact of jaguars. Perceptions of jaguar impact on livestock and on human safety
were each measured using four questions. Responses to each question were recorded on a 6-
point scale coded 0 to 5 (no impact to high impact) according to the size of the impact.
Questions were designed to assess beliefs about past damage and perceptions of the risk of
future damage at different scales (i.e. ranch, neighbourhood, district, and personal,
household, community). Perceived impact on livestock was assessed by the questions
„Damage associated with depredation ever caused by jaguars to you‟, „Damage associated
with depredation ever caused by jaguars to your neighbours‟, „Risk of any damage associated
with depredation to you in the next 12 months‟, and „Risk of any damage associated with
depredation to your neighbours in the next 12 months‟ and perceived impact on human safety
by the questions „Number of people ever hurt by a jaguar in the neighbourhood‟, „Number of
household members ever hurt by a jaguar‟, „Risk of you being hurt by a jaguar in the next 12
months‟ and „Risk of a household member being hurt by a jaguar in the next 12 months‟. We
summed the response scores for each of the jaguar impact scales („perceived impact on
livestock‟ and „perceived impact on human safety‟) for each respondent separately, thus
producing two impact scores ranging from 0 to 20 for each respondent.
101
Attitude to jaguar killing. Respondents‟ attitudes towards killing jaguars were assessed by
asking them to indicate their attitude towards killing the next jaguar that appeared on their
properties, using five-point evaluative semantic differential scales. Five items were used
(bad-good; beneficial-harmful; useless-useful; exciting-boring; unenjoyable–enjoyable). We
coded each answer -2, -1, 0, +1, or +2. Some questions were phrased in favour of jaguar
killing, and others against, in an effort to balance any potential bias associated with question
direction. However responses that were favourable to jaguar killing were always coded
positively, and responses that were unfavourable were always coded negatively, so that we
could construct a summated (averaged) rating scale to summarise an individual respondent‟s
overall attitude towards jaguar killing. The attitude scale, therefore, ranged from -2 to 2.
Subjective norms. Four questions and statements were used to assess subjective norms: „How
many of the people important to you would disapprove of you killing jaguars?‟ (1-5, none to
all); „Most people important to me think that killing jaguars is admirable‟ (1-5, strongly
disagree to strongly agree); „Among your neighbours, how much agreement would there be
that it is a good thing to kill jaguars?‟ (1-5, none to total); „In your household, how much
agreement would there be that it is a good thing to kill jaguars?‟ (1-5, none to total).
Descriptive norms. Descriptive norms were assessed using two measures („How many of
your neighbours do you think kill jaguars?‟ (1-5, none to all); „Think of the landowners in
Alta Floresta/Cáceres/Poconé - what percentage of them do you think kill jaguars?‟ (1-5,
none to all).
Group identification. Three items assessed group identification: „How much do you identify
with the landowners in Alta Floresta/Cáceres/Poconé?‟ (1-5, not at all to completely); „How
102
much do you feel strong ties with the other landowners in Alta Floresta/Cáceres/Poconé?‟ (1-
5, not at all to very much); „In general, how well do you feel you fit into the group of
landowners in Alta Floresta/Cáceres/Poconé?‟ (1-5, not at all to very well).
Perceived behavioural control (PBC). We used three statements to assess perceived
behavioural control in relation to killing jaguars: „Whether I kill the next jaguar that appears
on my property is up to me‟ (1-5, completely false to completely true); „For me, to kill the
next jaguar that appears on my property would be:‟ 1-5 (very difficult to very easy); „The
number of factors outside my control which could prevent me from killing the next jaguar
that appears on my property are‟; 1-5 (very many to none). We also asked the respondent to
list the factors that could prevent them from killing the next jaguar to appear.
Behavioural intention. Intention to kill jaguars was measured using two statements („I intend
to kill the next jaguar that appears on my property‟ and „If a jaguar appears on my property, I
will try to kill it‟, both 1-5 (very unlikely to very likely)). We explained clearly to the
respondents that by killing jaguars we meant using any of the methods and approaches used
to kill jaguars in Amazonia and Pantanal (e.g. finding and pursuing a jaguar using dogs,
creating a baiting station at which to shoot a jaguar, poisoning, and acting either alone or with
others including hired professional hunters).
Behaviour. Three approaches were used to assess jaguar-killing behaviour: (i) respondents
were asked if they had ever killed a jaguar and, if so, when they had last done this; (ii)
respondents were asked to say which of their neighbours had killed jaguars within the
previous 5 years); and (iii) in Amazonia, two independent local informants – a professional
hunter and a veterinarian - indicated ranchers who were believed to have killed jaguars in the
103
previous 5 years. One year after the main survey we conducted a follow-up survey in
Amazonia; we: 1) revisited a randomly-selected sub-sample of 40 respondents to ask them
whether they had killed any jaguar since our last visit; and 2) asked the informants whether
any of the study participants were supposed to have killed jaguars during the year since the
survey. We used three different approaches (self-reporting, neighbour reporting and
informant reporting) to guard against the likelihood that some respondents would fail to
report their own killing behaviour and, to a lesser degree, that of their neighbours.
Data analysis
All analyses were conducted in SPSS, version 16. We used general linear models (GLM) to
examine broad differences between Amazonia and Pantanal. For each study area separately
we used χ2 tests to test for a relationship between recent jaguar killing and intention to kill
jaguars in the future (intention was recoded as a binary variable for this analysis, using 0 for
no intention to kill jaguars and 1 for any level of intention). This provided a way of validating
the measurement of intention. Pearson correlation coefficients were used to examine
correlations between predictor variables. The internal consistency of scales was examined
using Cronbach‟s alpha reliability coefficients (Vaske 2008). Two sets of GLMs were
performed to develop a predictive model of jaguar persecution for each study area. The first
GLM analysis examined the prediction of intention to kill jaguars. Instead of an automated
stepwise procedure, we carried out a structured comparison of specific models defined a
priori based on different explanations of jaguar persecution. Because the models included
multiple predictors that were somewhat correlated, we used sequential sum of squares (type I)
models so that each effect in an individual model adjusted only for preceding effects in the
model. The following models were assessed: (Model 1) perceptions of jaguar impact on
livestock and on human safety; (Model 2) TPB variables; (Model 3) significant predictors of
104
the TPB plus descriptive norm and group identity; and (Model 4) significant predictors of the
extended TPB plus perceptions of jaguar impact on livestock and on human safety. We
compared Models 1 and 2 to test the hypothesis that the TBP offers a more predictive model
for jaguar persecution than the usual approach that assumes strictly retaliatory killing. By
comparing Models 2 and 3 we examined whether adding descriptive norm and group identity
to the TPB resulted in an increase in the predictive utility of the model. Finally, by
contrasting Models 3 and 4 we assessed the contribution of the perceptions of conflict to the
extended TPB. The second GLM analysis examined the effect of certain background factors
((i) perceived impact of jaguars on livestock, (ii) perceived impact of jaguars on human
safety and (iii) property size) on those variables from the extended TPB that significantly
affected intention to kill jaguars in the first GLM analysis.
Results
Differences between Amazonia and Pantanal
On average, respondents in the Pantanal (n=48) were older (mean = 54.8 years, SD = 11.8,
range = 27-77) than those in Amazonia (n=220, mean 42.4 years, SD = 13.1, range = 17-82)
(GLM; β = -12.313; SE = 2.048, p < 0.001). In the Pantanal, all respondents were male,
whereas 26.7% of Amazonian respondents were female. Table 4.1 presents additional
information about the variables used in the analyses and comparisons between Amazonia and
Pantanal. All variables differed significantly between the two regions, except for attitudes
towards killing jaguars. The perceived impact of jaguars on human safety was greater in
Amazonia, whereas perceived impact on livestock, subjective norm and descriptive norm
(both regarding killing jaguars), perceived behavioural control, group identity, and intention
to kill jaguars, were greater in the Pantanal. The proportion of respondents intending to kill
the next jaguar that appeared on their property was greater in the Pantanal (46%) than
105
Amazonia (15%) (Tables 4.8 and 4.9).
Table 4.1. Comparison of variables between Amazonia and Pantanal; means, standard deviations (SD)
and GLM results.
Variable Amazonia Pantanal
Mean SD Mean SD F SE p
Property size (ha) 298.8 1,581.1 5,524.0 5,342.4 209.30 368.63 < 0.001
Perceived impact on livestock 2.52 2.61 7.58 2.89 127.79 0.43 < 0.001
(0-20)
Perceived impact on human 11.82 2.37 10.29 3.57 13.54 0.42 < 0.001
safety (0-20)
Attitude (-2 - +2) 0.25 0.91 0.42 1.24 1.10 0.16 0.296
Subjective norm (1-5) 1.85 0.71 2.70 0.91 50.30 0.12 < 0.001
Perceived behavioural control 2.26 1.13 4.09 0.73 116.22 0.17 < 0.001
(1-5)
Descriptive norm (1-5) 2.17 0.78 3.98 0.67 223.57 0.12 < 0.001
Group identity (1-5) 1.88 0.96 3.50 0.95 130.93 0.15 < 0.001
Intention (1-5) 2.40 1.13 3.47 1.25 33.92 0.18 < 0.001
Predicting intention to kill jaguars in Amazonia
Table 4.2 displays correlations between variables. Intention to kill jaguars was positively
correlated with perceived impact of jaguars on livestock and on human safety and also with
attitude towards killing, perceived behavioural control and descriptive norm. Subjective norm
was not correlated with the other variables. Attitude, perceived behavioural control and
descriptive norm were positively correlated with each other. Perceived behavioural control, in
turn, was greater among large ranchers, whereas group identity was stronger among small
landowners.
106
Table 4.2. Pearson correlations between variables and Cronbach‟s alpha coefficients (shown on main
diagonal) for respondents in Amazonia. Results shown in bold are significant.* p  0.05; ** p ≤ 0.01.
NB: A= Cronbach‟s alpha not computed (single measurement).
1 2 3 4 5 6 7 8 9
1. Perceived impact on (0.80)

livestock (0-20)
2. Perceived impact on 0.129 (0.86)

human safety (0-20)
3. Attitude (-2 - +2) 0.350** 0.372** (0.90)

4. Subjective norm (1-5) 0.012 -0.023 0.001 (0.85)
5. Perceived behavioural 0.038 0.097 0.161* 0.030 (0.88)

control (1-5)
6. Descriptive norm (1-5) 0.119 0.173* 0.217** 0.028 0.213** (0.90)
7. Group identity (1-5) -0.163* 0.065 0.080 0.026 -0.265** -0.063 (0.76)
8. Property size (ha) -0.059 -0.041 -0.085 0.014 0.158* -0.013 -0.132* A
9. Intention (1-5) 0.228** 0.257** 0.437** 0.016 0.243** 0.330** 0.086 -0.043 (0.92)
As shown in Table 4.3, both perceived impact of jaguars on livestock and on human safety
were significant predictors of behavioural intention (Model 1). In support for Hypothesis 1,
landowners who perceived a greater impact of jaguars on their livestock or their safety were
more likely to intend to kill jaguars. However, in support for Hypothesis 2, the TPB model
(Model 2) explained over twice as much of the variation in intention (R2 = 0.22 compared
with 0.10), even though subjective norm did not emerge as a significant predictor in the
analysis. Landowners with more positive attitudes towards killing jaguars and a greater sense
of control were more likely to intend to kill jaguars on their properties. In support of
Hypothesis 3, the addition of descriptive norm (Model 3) accounted for a further 4.5% of the
variation in intention to kill jaguars, over and above that explained by attitude, subjective
norm and PBC. Respondents who perceived that other landowners killed jaguars had stronger
intentions to kill jaguars.
107
Neither perceived impact on livestock nor perceived impact on human safety had a
significant impact on intention to kill jaguars in the extended TPB model (Model 4), and
inclusion of these variables accounted for only 1% of the variation in intention. Nonetheless,
both perceived impact on livestock and on human safety had a significant effect on attitude
(Table 4.4), and therefore an indirect effect on intention to kill jaguars. These two findings
provide support for Hypothesis 4: the perceived impact of jaguars on livestock and on human
safety determine intentions indirectly by affecting attitude towards the killing of jaguars.
Table 4.3. Results of GLM predicting intention to kill jaguars in Amazonia.
Model Predictor included R2 B SE p

1 Perceived impact on livestock 0.105 0.083 0.027 0.002
Perceived impact on safety 0.110 0.031 <0.001
2 Attitude 0.222 0.510 0.076 <0.001
Subjective norm 0.016 0.096 0.865
Perceived behavioural control 0.178 0.061 0.004
3 Attitude 0.267 0.458 0.075 <0.001
Descriptive norm 0.322 0.088 <0.001
Group identity 0.008 0.075 0.911
4 Attitude 0.278 0.389 0.084 <0.001
Descriptive norm 0.305 0.088 0.001
Perceived impact on livestock 0.033 0.026 0.207
Perceived impact on human safety 0.039 0.030 0.198
108
Table 4.4. Results of GLM examining the effect of perceived impact of jaguars on livestock, and on
human safety, and of property size, on the extended TPB variables that influenced intention to kill
jaguars in Amazonia.
Dependent variable Predictor R2 B p
Attitude Perceived impact on livestock 0.234 0.101 <0.001
Perceived impact on safety 0.126 <0.001
Property size 0.000 0.375
Perceived behavioural Perceived impact on livestock 0.037 0.015 0.605
control Perceived impact on safety 0.047 0.142
Descriptive norm Perceived impact on livestock 0.039 0.028 0.148
Perceived impact on safety 0.053 0.058
Property size was a significant predictor of perceived behavioural control (Table 4.4). In
support of Hypothesis 5, ranchers who owned large properties had a stronger sense of control
over killing jaguars on their land. The main barriers to killing jaguars pointed out by the
small landowners who perceived that killing jaguars was not under their volitional control
were: fear of jaguars (43%); lack of skills (31%); lack of equipment (15%); and illegality
(10%).
The above findings are summarised in Figure 4.2. Landowners within the Amazon
deforestation frontier were more likely to intend to kill the next jaguar that appeared on their
property if they: 1) had more positive attitudes towards killing jaguars; 2) had a greater sense
of control over their jaguar-killing behaviour; and 3) perceived that other landowners killed
jaguars. Landowners with stronger perceptions of jaguar impact on livestock, or on human
safety, had more positive attitudes towards killing jaguars. Larger landowners had a greater
sense of control over their own jaguar-killing behaviour.
109
Figure 4.2. Final model showing intention to kill jaguars in the Amazonia deforestation
frontier.
Predicting intention to kill jaguars in the Pantanal
Intention to kill jaguars in the Pantanal was positively correlated with perceived impact of
jaguars on livestock, attitude, subjective norm, descriptive norm and group identity (Table
4.5). Perceived impact of jaguars on human safety and on property size were not correlated
with the other variables. Property size and perceived impact on human safety were not
correlated with the other variables. Attitude, subjective norm, and intention were positively
correlated with one another, as were subjective norm, descriptive norm, group identity, and
intention.
110
Table 4.5. Pearson correlations between variables and Cronbach‟s alpha coefficients (shown on main
diagonal) for respondents in the Pantanal. Results shown in bold are significant.* p  0.05; ** p ≤
0.01. NB: A= Cronbach‟s alpha not computed (single measurement).
1 2 3 4 5 6 7 8 9
1. Perceived impact on
(0.80)
livestock
2. Perceived impact on
-0.080 (0.75)
human safety
3. Attitude 0.376** 0.242 (0.88)
4. Subjective norm 0.380** 0.191 0.408** (0.81)
5. Perceived
0.215 -0.052 0.287* 0.189 (0.95)
behavioural control
6. Descriptive norm 0.179 -0.015 0.114 0.340* 0.092 (0.79)
7. Group identity 0.180 0.220 0.237 0.575** 0.178 0.320* (0.77)
8. Property sizea -0.067 -0.085 -0.238 -0.151 0.207 -0.090 -0.026 A
9. Intention 0.410** 0.145 0.554** 0.676** 0.250 0.522** 0.645** -0.232 (0.90)
As shown in Table 4.6, perceived impact of jaguars on livestock was a significant predictor
of intention to kill jaguars (Model 1). Perceived jaguar impact on human safety had no
significant impact on intention. In partial support for Hypothesis 1, landowners who
perceived a greater impact of jaguars on livestock had stronger intentions to kill jaguars. The
TPB (Model 2) offered a more predictive model, in support of Hypothesis 2. Even without a
significant effect of perceived behavioural control, the TPB accounted for 55% of the
variation in intention to kill jaguars. Ranchers with more positive attitudes towards killing
jaguars and a greater perception of others‟ approval of jaguar-killing behaviour had a stronger
intention to kill the next jaguar that appeared on their properties. Descriptive norm and group
identity (Model 3) emerged as significant predictors of intention, in support for Hypothesis 3.
Ranchers who perceived that other ranchers killed jaguars and identified themselves with the
other ranchers were more likely to engage in jaguar killing. Descriptive norm and group
identity accounted for an additional 17% of the variation (over and above attitude and
111
subjective norm), providing support for the inclusion of these variables in the model (Model
3). Including perceived impact on livestock and perceived impact on human safety in the
extended TPB model (Model 4) did not improve the model significantly further (accounting
for just 0.8% more of the variation in intention), and their effect on intention was not
statistically significant. However, perceived impact of jaguars on livestock did have a
significant effect on both attitude and subjective norms (Table 4.7), and therefore an indirect
effect on intention to kill jaguars. Perceived impact of jaguars on livestock seems indirectly
to determine intention by affecting attitude (and subjective norms), which supports
Hypothesis 4.
Table 4.6. Results of GLM predicting intention to kill jaguars in the Pantanal.
Model Predictor included R2 B SE p
1 Perceived impact on livestock 0.200 0.193 0.061 0.003
Perceived impact on safety 0.063 0.047 0.187
2 Attitude 0.552 0.323 0.115 0.007
3 Attitude 0.722 0.345 0.089 < 0.001
4 Attitude 0.730 0.330 0.096 0.001
Perceived impact on livestock 0.039 0.042 0.359
Perceived impact on safety -0.014 0.031 0.645
112
Table 4.7. Results of GLM examining the effect of perceived impact of jaguars on livestock, and on
human safety, and of property size, on the extended TPB variables that influenced intention to kill
jaguars in the Pantanal.
Dependent variable Predictor R2 B p
Attitude Perceived impact on livestock 0.251 0.172 0.005
Property size -0.001 0.152
Subjective norm Perceived impact on livestock 0.205 0.129 0.006
Descriptive norm Perceived impact on livestock 0.038 0.042 0.250
Perceived impact on safety -0.002 0.957
Group identity Perceived impact on livestock 0.088 0.068 0.176
Figure 4.3 summarizes the above findings. Intention to kill jaguars in the Pantanal was
greater among ranchers who: 1) had more positive attitudes towards killing jaguars; 2)
believed there was normative support for performing such behaviour; 3) perceived that other
ranchers killed jaguars; and 4) identified themselves with the other ranchers. Ranchers who
had a stronger perception of the impact of jaguars on livestock also had more positive
attitudes towards killing jaguars and a stronger sense of social pressure to kill them.
113
Figure 4.3. Final model showing intention to kill jaguars in the Pantanal.
Relationship between intention and behaviour
No jaguar kills were observed or reported during the 12 months following the survey.
However, by cross-referencing information provided by landowners and by independent
informants, we identified 11 landowners in the Amazonia sample who had killed at least one
jaguar on their property in the previous five years (six of these landowners had not reported
these kills). This information was used to examine the relationship between landowners‟
declared intention to kill jaguars and their past jaguar-killing behaviour. Most of the
landowners that were confirmed to have killed jaguars in the previous five years said that
they intended to kill the next jaguar that appeared on their property, whereas most of those
that had not killed jaguars recently did not intend to do so in future (χ2= 40.5, df: 1, p < 0.001,
Table 4.8). This relationship between measured behavioural intention and actual past
behaviour suggests that intention to kill jaguars may be a valid proxy for future killing
behaviour in Amazonia.
114
Table 4.8. Proportion of Amazonia respondent who had and had not killed jaguars in the previous five
years, who did and did not intend to kill them in the future.
Intended Did not intend
to kill to kill Total % (n)
Past killing not confirmed (%) 11% 89% 95% (209)
Past killing confirmed (%) 82% 18% 5% (11)
Total % (n) 15% (33) 85% (187) 100% (220)
All seven ranchers in the Pantanal that reported killing at least one jaguar on their property in
the previous five years said that they intended to kill the next jaguar that presented a similar
opportunity. In contrast most of those that had not killed jaguars did not intend to do so in
future (2= 8.9, df: 1, p = 0.003, Table 4.9).
Table 4.9. Proportion of Pantanal landowners who had and had not killed jaguars in the previous five
years, who did and did not intend to kill them in the future.
Intended Did not intend
to kill to kill Total % (n)
Past killing not confirmed (%) 36% 64% 85.4% (41)
Past killing confirmed (%) 100% 0% 14.6% (7)
Total % (n) 46% (22) 54% (26) 100% (48)
This study shows that the perceived impact of jaguars on livestock is not the only predictor of
a cattle rancher‟s intention to kill jaguars. Intention to kill jaguars is also determined by fear,
personal and social motivations, and internal and external barriers to killing jaguars. The
relative importance of these factors in determining intention varies with region and affluence.
Conservation measures that aim to decrease jaguar persecution by ranchers will be more
effective if these factors are taken into account.
115
Fear is an important factor in determining Amazonian landowners‟ intentions to kill jaguars.
Fear affects intention through two distinct and opposing mechanisms. On one hand, fear of
jaguar impact on human safety can increase the likelihood of a positive attitude towards
killing jaguars, while, on the other, fear arising from landowners‟ own inability to kill jaguars
(poor perceived behavioural control) may act as an internal barrier that limits their intention
to engage in jaguar killing. A fear of jaguars is indeed common among the Amazonian
frontiersmen (see Chapter 3), who are largely immigrants with little experience of jaguars and
the forest, but fear of jaguars was also reported by some ranchers in the Pantanal. Jaguars are
potentially dangerous to people, and a phobic response to the species might be instinctive. It
has been argued that our evolutionary roots were shaped by a world in which humans were
primarily a prey species, mostly to lions and tigers (Hart and Sussman 2009). Anti-predator
adaptations to large cats may have been selected for in humans as a direct result of the threat
which they present to human survival; this may account in part for fear of jaguars among
humans. However, fear of jaguars varies with knowledge. The better informed people are
about jaguars the less they fear them (Chapter 3). After all, the strong fear of jaguars among
Amazonian ranchers is not grounded in fact. Attacks on people by jaguars are extremely rare
and have occurred almost invariably either when hunted jaguars are cornered or injured, or
when jaguars are defending cubs or carcasses. The only documented, unprovoked, fatal attack
by a jaguar on a human in Brazil occurred on 24 June 2008, when a young fisherman was
believed to have been killed by a jaguar while sleeping in his tent on a bank of the Paraguay
River in the Pantanal. This unique incident may have occurred as a result of jaguars
becoming habituated to people around baits used to attract them to be photographed by
tourists.
116
Attitude towards killing jaguars predicts landowners‟ intention to kill them in both Amazonia
and the Pantanal. Attitude towards killing jaguars is shaped by both instrumental and
experiential expectations. For example, landowners may favour killing jaguars because they
consider such behaviour beneficial and useful, as well as exciting and enjoyable. The thrill of
the hunt may be an important additional motivation for retaliatory killing and may in some
situations be the only reason for killing jaguars. Indeed before hunting was prohibited
(Brazilian Wildlife Protection Act 1967) and jaguars listed in the Convention on International
Trade in Endangered Species (CITES) in 1973, hunting jaguars for sport was popular in the
Pantanal (Almeida 1978; Roosevelt 1914; Siemel 1953). Ranch hands that do not own
livestock are known to kill jaguars on other people‟s ranches in the Pantanal, where absent
owners have specifically banned the practice (Cavalcanti 2008). This provides further
evidence that jaguar killing may be explained by personal motivations unrelated to the impact
of jaguars on human livelihoods.
Social motivations are important determinants of the intention to kill jaguars in both
Amazonia and the Pantanal. In the Pantanal, some ranchers justified their perception of the
social approval of jaguar killing on the grounds of tradition. Ranchers in the region often
refer, with apparent pride, to the „Pantaneiro culture‟ and the conviction that jaguar hunting
has been passed from generation to generation as an element of that culture. The subjective
norm among Pantanal landowners is affected by their perception of jaguar impact on
livestock; ranchers who perceive a strong impact of jaguars on livestock also perceive a
strong peer pressure to kill jaguars. The economic and cultural centrality of cattle ranching in
the region doubtless accounts for the correlation observed here between these factors (and
between perceived jaguar impact on livestock and attitude towards jaguar killing). It is widely
perceived that jaguar killing is common in both Amazonia and the Pantanal, and the
117
perception that neighbours often kill jaguars (descriptive norm) encourages landowners to
intend to do the same. Despite it being an illegal practice, jaguar killing is not something
people strive to keep secret. On the contrary, some perpetrators talk openly with other
ranchers, about killing jaguars, often with the support of photographs of the hunt. Jaguar
hunting is a favoured topic of conversation among ranchers; stories of jaguar hunting spread
quickly among the community, are told repeatedly, and are often remembered vividly as a
consequence. The ease with which stories of jaguar hunting are brought to mind among
landowners could account for the strong descriptive norm concerning jaguar killing (Tversky
and Kahneman 1974). A vicious circle may exist in which hunts are readily remembered and
repeated, often reaching other landowners, and creating the perception that jaguar killing is
common and acceptable, which in turn causes more jaguars to be killed. Finally, social
identity (group identity) also influences jaguar killing in the Pantanal. In our Amazon sample,
many small landowners were settlers belonging to small, organized communities. Although
group identity was greater among smaller settlers than among large landowners in the region,
jaguar killing was not perceived to be an in-group behaviour, and so group identity did not
affect their intention to kill jaguars in Amazonia. In contrast, ranchers in the Pantanal have a
strong group identity and perceive that jaguar killing is part of Pantaneiro culture. If
perceived association of jaguar killing with in-group members is coupled with a perceived
association of jaguar conservation with out-group members, conservation efforts considered
to emanate from outside the group may be poorly accepted. The role of conflicts between
human groups in resolving human-jaguar conflicts (Herda-Rapp and Goedeke 2005) deserves
further investigation.
118
Limitations
This study has two major limitations. First, it adopts a strictly cognitive model to explain
jaguar killing. In the TPB, behaviour is held to be thoughtfully planned, based on one‟s
attitudes toward the behaviour, in a deliberative process. However, some incidences of jaguar
killing do not result directly from a plan or intention. A couple of respondents reported killing
a jaguar in an unexpected, close encounter. In such cases jaguar killing behaviour results
from a spontaneous process in which thoughtful evaluations of the personal and social
outcomes of performing the behaviour (attitudes and social norms concerning the killing of
the jaguar, respectively) play no role. For such situations, Fazio‟s (1986) model of the
attitude-to-behaviour process may offer a more predictive model. Fazio (1986, 1990)
distinguishes between attitudes towards objects or targets, and attitudes toward behaviours,
arguing that the former are important in determining behaviour via spontaneous processing.
People that have negative attitudes towards jaguars, and a firearm at hand, are more likely to
kill them in the improbable event of an unexpected encounter.
A second limitation is that current behaviour could not be measured. Consequently, the utility
of the model to predict future behaviour could not be verified. Jaguar killing is a particularly
difficult behaviour to measure for two reasons: (i) it is illegal and therefore some respondents
might fail to report it; and (ii) it depends upon a contextual factor that is highly variable in
time and space: an encounter with a jaguar. While most behaviours that have been addressed
in TPB studies can be observed on a daily basis (e.g. exercising, recycling) or every month at
most (e.g. reducing energy use), killing jaguars is something a rancher typically does when
the opportunity arises, i.e. when a jaguar appears, e.g. a few times in one year and then not
for several more years. Robust validation of the intention-behaviour relationship would
require the measurement of behaviour either over several years for individual ranchers, or
119
over larger areas encompassing a larger number of ranches than surveyed in this study.
However this study provides some evidence that our measurement of intention to kill jaguars
was valid: (i) observed relationships between intention and its predictors were consistent with
expectations based on theory (Vaske 2008); and (ii) there was a significant relationship
between intention to kill jaguars and recent past killing behaviour. Meta-analyses of the TPB
indicate that intentions and perceived behavioural control account for 34% of the variation in
behaviour (Sutton 1998). Given that the model proposed in this study includes additional
explanatory variables, it might explain a greater proportion of actual jaguar killing by
ranchers. While the percentage of variance explained may give an overly pessimistic view of
the utility of our model, we propose that this model can make a substantial contribution to the
design of strategies to prevent jaguar killing.
Implications for conservation
Our model is not only useful to the extent that it can predict jaguar killing. It can also be used
to assess the relative importance of the different factors that motivate or deter such behaviour.
Efforts to increase people‟s tolerance of jaguars and discourage jaguar killing have focused
largely on economic incentives (e.g., monetary compensation for livestock loss; Silveira et al.
2006) and legal prohibitions and sanctions (e.g. establishment of protected areas). The role
and importance of social and psychological factors have been far less considered. This study
proposes a broader approach for understanding and preventing the killing of jaguars, an
approach that goes beyond the usual framework of human-jaguar conflicts, and considers all
the motivations and barriers – social and psychological as well as legal and economic -
concerning jaguar killing.
120
Human behaviour is partly determined by external, contextual factors (Clayton and Myers
2009). The measures mentioned above aim to prevent jaguar killing behaviour by changing
the economic and legal contexts. Our findings, however, highlight the importance of social
context. Social norms (descriptive norm and subjective norm) and social identity also affect
jaguar killing and should be taken into account by conservationists. A number of techniques
can be used to change social context. For instance, a persuasive communication campaign
using models, case studies and examples of coexistence with jaguars, combined with
information about the negative consequences of killing them, might help to create or redefine
a social norm by explaining that the community condemns rather than accepts the killing of
jaguars. Arranging for conspicuous and respected group members or community institutions
(e.g. cooperatives and rural schools) to promote tolerance of jaguars may be one way to
influence other group members. Communicating information about the behaviour and
practices of group members via informal social networks is another possible strategy. Our
findings also suggest that recommendations that are seen to come from within the group may
be more readily accepted than those that stem from outside the group. Social marketing can
offer other useful tools, i.e. using commercial marketing techniques to promote an idea or
behaviour that benefits either the individual or society. Examples of social marketing tools
include obtaining a commitment, prompts, personalized communication, norm appeals, word-
of-mouth and feedback (McKenzie-Mohr and Smith 1999). Each of these tools is associated
with an extensive literature suggesting that they may help to change behaviour; combinations
of these tools may be more effective than any one in isolation (McKenzie-Mohr and Smith
1999).
Human behaviour is also determined by internal, psychological factors (Clayton and Myers
2009). This study revealed the importance of perceptions, knowledge, skills, attitudes and
121
emotions such as fear and excitement in inciting the killing of jaguars. Education and
information-intensive campaigns might be used to influence these factors in order to deter
jaguar killing. The role of information in changing behaviour is complex. While researchers
agree that information alone will not motivate someone to adopt a new behaviour (e.g. to
tolerate jaguars) (Hungerford and Volk 1990; Stern 2000), it is equally clear that a lack of
information can be a barrier to changing behaviour (Schultz 2002; Kaplan 2000; DeYoung
2000). Education campaigns addressing livestock predation by jaguars might help ranchers
correctly to identify a depredation event, implement preventive measures, and adjust
exaggerated perceptions of jaguar impact on livestock to reality. Campaigns conveying the
true probability of a jaguar attack on humans might be particularly effective in preventing
jaguar killing among small landowners in Amazonia. In the longer term, formal and informal
education could help to accelerate the current shift away from traditional materialist values
(focused on physical security and economic well-being, and use and management of wildlife
for human benefit) towards post-materialist values (focused on quality of life, self-expression
and self-esteem, and mutualistic attitudes towards wildlife) that has been observed in western
societies (Inglehart 1990; Manfredo et al. 2003). A shift from utilitarian to mutualistic
attitudes towards jaguars would mean fewer people deriving pleasure from recreational
hunting or supporting the lethal control of livestock-raiding jaguars, and more people
supporting and enjoying coexistence with the species.
In neither Amazonia nor Pantanal would it be possible to construct an effective jaguar
protection strategy based on a single field of influence. Jaguar killing results from the
interplay between external incentives (economic/market, legal/administrative/coercive, and
social/normative) and internal influences (such as knowledge, skills, emotions, norms,
attitudes and intentions), and effective strategies to prevent jaguar killing should therefore be
122
based on the social and psychological aspects of human-jaguar relationships as well as on
legal and economic considerations. Conservationists need to find and support ways to make
jaguar killing not only illegal and unprofitable, but also socially and personally unacceptable.
123
Chapter 5 – Improving perceptions
CHAPTER 5
SCHOOL-BASED APPROACHES FOR IMPROVING
PERCEPTIONS OF JAGUARS AMONG CHILDREN AND THEIR
FATHERS ON THE AMAZON DEFORESTATION FRONTIER
People must learn to hate, and if they can learn to hate, they can be taught to love, for love
comes more naturally to the human heart than its opposite
Nelson Mandela
Abstract
Landowners in Amazonia are more likely to kill jaguars if they approve of the behaviour and
believe that their neighbours commonly kill jaguars. These attitudes and social norms are
affected by perceived impact of jaguars on human livelihoods, and ultimately by attitudes
towards, and knowledge of, the species. We conducted a school-based experiment on the
Amazon deforestation frontier to examine the effects of information and elaboration on these
psychological measures among pupils, and the effects of information (communicated via
printed materials including children‟s homework) on these measures among pupil‟s fathers.
We found that school-based communication interventions can have a powerful effect on
perceptions of jaguars among pupils, and that pupils can effectively transfer that effect to
their fathers. Books distributed via local schools successfully decreased fathers‟ perceptions
of descriptive norms, but the same books distributed via a conservation organisation did not.
This suggests that fathers were influenced not only by the information explicitly conveyed in
the content of books and their children‟s homework, but also by the implicit message that
jaguar conservation was supported by a community institution (e.g. the local school), and
probably also therefore by other community members. Information alone caused significant
positive changes in knowledge, attitudes and perceptions among pupils and fathers, but some
of those changes were not enduring. Elaboration alone produced more persistent effects, but
negative attitudes towards jaguars were reinforced, even among pupils who had been exposed
124
to information about jaguars. Information and elaboration combined created stronger and
more enduring effects than either intervention alone. These findings may be important in
designing conservation communication interventions for jaguars and other charismatic
species worldwide.
Introduction
The killing of jaguars by farmers and ranchers in rural Amazonia is determined by their
attitudes towards jaguar killing and their perceptions of how common such behaviour is
among their neighbours (Chapter 4). Attitudes towards jaguar killing are affected by
perceptions of the impact of jaguars on human livelihoods (Chapter 4), which are in turn
affected by attitudes towards jaguars and knowledge about the species (Chapter 3).
Interventions designed to deter jaguar killing can be directed at one or more of these
determinants. Changes in these factors should produce changes in the intention to kill jaguars
and, given adequate control over the behaviour, the new intention (i.e. to tolerate jaguars
instead of persecute them) should be carried out under appropriate circumstances (Ajzen
1985). The extensive literature on education, communication and persuasion (Jacobson et al.
2006; Monroe 2003; Petty and Cacioppo 1996) presents numerous approaches that have been
used successfully to increase knowledge, change social norms and improve attitudes.
Reaching landowners in rural Amazonia and effectively improving their perceptions of
jaguars, however, remains a challenge. As well as economic incentives (e.g. monetary
compensation for livestock loss; Silveira et al. 2006) and legal prohibitions and sanctions
(e.g. establishment of protected areas) to dissuade people from killing jaguars, there have
been a few attempts elsewhere to foster tolerance to the species by increasing knowledge or
changing attitudes and social norms. These efforts have focused on providing landowners
with factual information – mostly through printed manuals (e.g., Hoogesteijn 2000; Leite-
Pitman et al. 2002; Marchini and Luciano 2009a) – on the importance of jaguar conservation
125
and on how to prevent predation problems. However, the cost-effectiveness of a
communication campaign based on print media is limited in Amazonia because of the low
human density, difficulty of access due to poor road conditions, high cost of printing
attractive communication materials, illiteracy and the resistance of some landowners to
outsiders. A communication strategy whereby school children act as catalysts of perception
change among their parents and other community members could be an effective means of
improving tolerance of jaguars in rural Amazonia. In this study, we examined a school-based
approach to change perceptions of jaguars among pupils and their parents on the Amazon
deforestation frontier. We conducted an in-classroom experiment to compare the effect of
information versus elaboration on the perceptions of jaguars among 5th and 6th graders (11-15
year olds), and investigated whether, and how most effectively, school children can influence
their parents‟ perceptions of jaguars.
Changing perceptions: information and elaboration
Information
In the context of human-wildlife conflicts, information can improve perceptions of the
species in question in a number of ways, by: 1) changing the values people place on the
species (Kellert 1996); 2) reducing to more realistic levels their perceptions of the threat
posed by the species (Marker et al. 2003); and 3) empowering them to cope with any damage
caused and to find guidance on conflict mitigation. However, there have been cases in which
information-based interventions aimed at improving attitudes toward predators, by increasing
knowledge, proved ineffective, sometimes even reinforcing negative attitudes among those
already holding strong views (Hook and Robinson 1982; Kellert 1986). Knowledge of
jaguars is poor among children (and adults alike) in rural Amazonia, where the jaguar is
predominantly perceived as threatening (Chapter 2), although jaguar attacks on people are
126
extremely rare and have occurred almost invariably when hunted jaguars are cornered or
injured, or when jaguars are defending cubs or carcasses. In this study we provided school
pupils with information (lectures, activity books) about the jaguar (its ecological, economic
and cultural importance, its impact on human livelihoods in relation to how it has been
affected by human actions such as deforestation and persecution, and how to prevent
livestock depredation problems).We predicted that by providing such information we would
significantly increase pupils‟ knowledge of the species, decrease their perceptions of the
impact of jaguars on their livelihoods, and improve their attitudes towards jaguars.
Elaboration
Traditional school lessons allow teachers to disseminate large amounts of pre-planned
information with minimal effort. At least until recently this has been the favoured method of
teaching from primary schools right through to universities across the world. In a typical
classroom setting, students are involved only passively in learning, e.g. listening to the
teacher and reading the blackboard or text book when required. Analysis of the research
literature (Bonwell and Eison 1991; Chickering and Gamson 1987), however, shows that
such passive involvement in education generally leads to limited retention of information,
and suggests that in order to learn well students need to be actively involved during a lesson,
e.g. writing, discussing, or solving problems. The more elaborate mental processing
associated with active learning makes novel information both easier to remember and more
personally meaningful (Bonwell and Eison 1991). In a similar vein, Petty and Cacioppo
(1996) propose that attitude change depends upon the amount of thoughtful consideration
(cognitive elaboration) that occurs in response to a persuasive communication. According to
their Elaboration Likelihood Model of Persuasion (ELM), there are two types of attitude
change: 1) central route; and 2) peripheral route attitude change. These two routes of attitude
127
change differ in the amount of elaboration involved. Central route attitude change can occur
when there is significant cognitive elaboration (high elaboration likelihood), whereas
peripheral route attitude change can occur in the absence of cognitive elaboration (low
elaboration likelihood). Central route attitude change requires the motivation, and the ability,
to process cognitively a persuasive communication. For example, a person‟s attitudes are
more likely to change by the central route if they are presented with new information on a
subject they find stimulating, and this is done in a thought-provoking way. Attitudes are more
likely to change by the peripheral route if new information is presented on a subject they find
uninspiring, and without their active involvement. Central route attitude change is more
difficult to achieve than peripheral route change, but it also lasts longer, influences behaviour
more, and is more resistant to counter-persuasion than peripheral route change.
Jaguars have a stronger presence in the hearts and minds of Brazil‟s children and adolescents
than does any other native mammal, generally eliciting strong, but mixed, feelings (the
cognitive and affective prominence of the jaguar is highest among 12 and 13 years old, for
this reason we selected 5th and 6th graders as subjects in this study) (Chapter 2). Many pupils
find classes more interesting and enjoyable when they include active learning experiences
(Murray 2002). Therefore, we expected that children would be motivated to engage in a
discussion about jaguar issues and to process cognitively the related information. Lack of
knowledge about jaguars, however, could hinder their ability to elaborate. We predicted,
then, that elaboration would cause a greater impact on pupils who had been provided with
information about the species.
128
Disseminating perceptions: intergenerational learning and modelling
Jaguars are generally killed by adult male landowners (Chapter 4). Therefore any school-
based approach to jaguar conservation will be effective only if pupils can influence
perceptions among their fathers or other men in the community. Children are known to share
environmental learning and attitudes with parents (Ballantyne et al. 2001) and to influence
their parents‟ environmental attitudes (Ballantyne et al. 1998; Duvall and Zint 2007). Men in
rural Amazonia and elsewhere in the jaguar‟s range tend to have strong feelings towards the
species (Chapter 3 and 4), which might enhance intergenerational learning about jaguar-
related issues. We predicted that pupils participating in a jaguar education programme at
school would transfer to their fathers knowledge gained from lectures and educational
materials and, in so doing, cause a change in their father‟s perceptions of jaguars.
Another way schools might contribute to conservation is by acting as a conduit for the
distribution of communication materials. Communication effectiveness depends on the
credibility and trustworthiness of the information source (Petty and Cacioppo 1996).
Community institutions such as schools, cooperatives and church are arguably more credible
and trustworthy to rural Amazonians than are outside institutions. As a result they can act as
role models by demonstrating attitudes and behaviours which the community can readily
identify with and imitate (Bandura 1997). We predicted that communication material – more
specifically an illustrated book on jaguar ecology and predation problems – would be more
effective in changing landowners‟ perceptions of jaguars if it reached landowners through the
local school, with clear endorsement by the school, rather than through a non-governmental
conservation organization.
129
Methods
Experimental design
This study was conducted in Alta Floresta, on the deforestation frontier in northern Mato
Grosso (southern Amazonia). This experiment tested ways of improving perceptions of
jaguars among rural Amazonians. The work took place at six rural public schools and was
divided into two parts: I) influencing school pupils‟ perceptions directly via information and
elaboration interventions; and II) influencing fathers‟ perceptions indirectly via school-based
communication interventions with their children.
Part I. Influencing school pupils‟ perceptions via information and elaboration interventions
At four of the six schools (schools 1 to 4; Figure 1), one fifth of all pupils in the 5th and 6th
grades (ages 11 to 14) were randomly assigned to one of the following treatments: 1) control;
2) information; 3) elaboration; and 4) information plus elaboration. The remaining fifth of
pupils were assigned to Part II („book via school‟ treatment) (Figure 1). The information
treatment consisted of three 90-minute lectures (one per week over three weeks), about jaguar
issues. Lectures were based on two activity books (Appendix IV) and covered topics such as
jaguar ecology, impact of jaguars on human livelihoods, jaguar conservation status, reasons
to conserve jaguars, and measures to minimise the impact of jaguars on livestock. Lectures
and activity books focused on factual knowledge and did not attempt to convey ideas on
whether certain attitudes or behaviours towards jaguars would be morally right or wrong. For
each lecture, some sections of the activity books were assigned as homework. The lectures
were given by a local teacher (who had taught previously at one of the participating schools)
under the supervision of SM. The elaboration treatment consisted of a structured group
discussion in which pupils raised, shared and debated all their beliefs and perceptions relating
to jaguars (Appendix V). The discussion was approximately 2 hours long and was moderated
130
by SM with the assistance of the same local teacher that taught the lectures for the
information treatment. SM and his assistant were as objective and neutral in appearance and
behaviour as possible and did not provide pupils with any factual or judgemental information
about jaguars during the discussion. Pupils in the „elaboration‟ treatment of Part I did not use
activity books, and were not assigned homework. Pupils in the information and elaboration
group were given their third lecture one week before taking part in the elaboration exercise.
Pupils in the control group continued their education as usual.
Part II. Influencing fathers‟ perceptions indirectly via school-based communication
Apart from the fathers of children participating in the elaboration treatment of Part I, fathers
received information about jaguars from one or both of two sources (Figure 1): their
children‟s homework and the book „People and Jaguars: a Guide for Coexistence’ (Marchini
and Luciano 2009a) (Appendix VI). Pupils involved in the information treatment (and
attended lectures) of Part I were instructed to ask their fathers to sign their homework. Their
fathers received a letter from the school‟s director explaining that the school was
implementing a jaguar education programme directed at pupils, parents and community
members.
A Guide for Coexistence is an easy-to-read, highly illustrated colour book. It is divided into
the following sections: „Jaguars; what they are and how they live‟, „Jaguars: are they a
problem for us?‟, „People and jaguars: when we are the problem‟, „Reasons to coexist with
jaguars‟, „How to coexist with jaguars‟ and „Beyond coexistence: learning more about and
enjoying jaguars‟. As with the activity books, the guide focuses on factual information and
does not convey opinions about what may be morally right or wrong in the relationships
between people and jaguars. Fathers received the book directly from their children, together
131
with an endorsement letter from the school explaining that the book was part of broader
jaguar education programme. Another group of fathers, randomly selected from a list of
fathers of 5th and 6th graders at two other schools (schools 5 and 6; Figure 5.1), received the
book from SM or his assistant instead of their children. SM and his assistant visited each of
these fathers at home, introduced themselves as representatives of a local non-governmental
conservation organization (Fundação Ecológica Cristalino - FEC), interviewed the father
(see below), and handed him one copy of the book on behalf of FEC. These fathers were not
informed about the involvement of their children‟s school in providing parents‟ contact
details for the study.
Figure 5.1. Diagram showing assignment of pupils and fathers to experimental treatments (and sample
sizes). Dashed lines indicate that fathers were not aware of the involvement of their children‟s school
in the study.
Measurement of knowledge, attitudes, and perceptions
Pupils were evaluated three times: before and immediately after exposure to the treatment
(pre-test and post-test, respectively) and again three months later (delayed post-test) to test
132
for retention of treatment effects. Fathers were evaluated using a pre-test and a post-test only.
The experiment was conducted between July and November 2009.
Pupils were evaluated in class by the means of self-completion questionnaires. SM
introduced himself and his assistant as representatives of an educational project from outside
the school and gave the students a brief introduction. The introduction was devoid of value
statements and merely invited them to participate in a project involving questions about their
opinions about wildlife. SM stressed that there were no right or wrong answers, that no grade
would be assigned, and therefore pupils could feel at ease about offering their views
transparently. SM also emphasised the importance of each pupil completing their
questionnaires individually. Teachers were given the choice of staying or leaving the room; a
few stayed but did not interfere with the survey. Questionnaires and pencils were distributed
and questions were read out one at a time. We checked that students understood each
question and numbered their responses correctly. This approach allowed us to repeat
questions and provide further explanations where necessary. Questionnaires were completed
in approximately 30 minutes and returned directly to SM.
Fathers were evaluated at home using face-to-face interviews conducted by SM. SM was
always accompanied by a local field assistant during the interviews. SM and his assistant
were as objective and neutral in appearance and behaviour as possible. SM introduced
himself to the fathers of pupils in schools 1 to 4 as a representative of an educational project,
rather than a conservation organisation. SM introduced himself to the fathers contacted
through schools 5 and 6 as a representative of Fundação Ecológica Cristalino. He explained
to respondents that the purpose of the survey was to collect data on people‟s perceptions of
wildlife, and that these would ultimately contribute to a conservation project.
133
Separate questionnaires were designed for evaluating pupils and fathers but both types
examined the following: (1) knowledge about jaguars and depredation problems; (2) attitude
towards jaguars; (3) perception of the impact of jaguars on human safety; (4) perception of
the impact of jaguars on livestock; (5) attitudes towards killing jaguars; (6) descriptive norm
regarding the killing of jaguars (Appendix II). The variables are described in more detail
below.
Knowledge about jaguars and livestock predation. For knowledge questions all answers were
coded as binomial variables, using 1 for a correct answer and 0 for an incorrect answer or “do
not know”. Knowledge was measured by ten questions so that the scale ranged from 0
(minimum knowledge) to 10 (maximum knowledge) and had a reliability (Cronbach‟s alpha)
of 0.77. Questions were posed as statements, five of which were correct and five incorrect.
The statements were: (i) The jaguar generally begins to consume its prey from the front while
the puma consumes the areas from the ribs backwards (correct); (ii) A jaguar‟s prey is usually
hidden and covered with leaves, while a puma‟s prey isn‟t (incorrect); (iii) The jaguar‟s prey
generally presents a bite mark at the base of the neck, while the puma‟s prey generally has a
bite on the throat (correct); (iv) The jaguar‟s footprint is longer than wide with thinner and
pointed toes, while the puma‟s footprint is slightly wider than long, with round toes
(incorrect); (v) A female jaguar produces on average 1 or 2 cubs every other year (correct);
(vi) Black jaguars are more dangerous to cattle than yellow jaguars (incorrect); (vii) The
heaviest jaguar ever captured weighed approximately 150 kilos (correct); (viii) Jaguars kill
more people every year in Brazil than do domestic dogs (incorrect); (ix) Where cattle are
more abundant than native prey, jaguars take more cattle than native prey (correct); (x)
Calves kept closer to the forest edge have in general a smaller chance of being killed by
jaguars (incorrect).
134
Attitude towards jaguars. We assessed respondents‟ attitudes towards jaguars, according to
(Ajzen and Fishbein 1980; Ajzen 1985)‟s definition of attitudes, “the tendencies people have
when they view an entity with some degree of favourability or unfavourability”. Attitudes
were measured using a combination of Likert scales (e.g., strongly disagree, disagree, neither
agree nor disagree, agree, strongly agree) and semantic differential statements (where
respondents are offered a choice of two polar opposite responses e.g. good-bad) (Oppenheim
1998). Each question consisted of a statement about jaguars, and two extreme response
options describing the potential range of attitudes to the statement, e.g. very sad – very
happy. Respondents were asked to select a point on a five-point scale between those options
that reflected their view. We coded each answer -2, -1, 0, +1, or +2. Some questions were
phrased in favour of jaguars, and others against, in an effort to balance any potential bias
associated with question direction. However responses that were favourable to jaguars were
always coded positively, and responses that were unfavourable were always coded
negatively, so that we could construct a summated rating scale to summarise an individual
respondent‟s overall attitude towards jaguars. Five questions were used so the summated
rating scale ranged from -10 to +10 (Cronbach‟s alpha = 0.89). The questions were as
follows: (1) You would like the jaguar population in the region to: decrease a lot - increase a
lot; (2) If all the jaguars disappeared forever from the region, you would feel: very sad - very
happy; (3) What you feel towards jaguars is better described as: dislike a lot - like a lot; (4)
The jaguar has its value, even if it does not generate any income to you: strongly disagree -
strongly agree; (5) If you had to walk alone in a forest where there are jaguars, you would
feel scared: strongly disagree - strongly agree.
Perception of jaguar impact on human safety and livestock. Perceptions of jaguar impact on
livestock and on human safety were each measured using two questions. Responses to each
135
question were recorded on a 6-point scale coded 0 to 5 (no impact to high impact) according
to the size of the impact. Perceived impact on livestock was assessed using the questions
„How would you rate the damage associated with predation ever caused by jaguars to your
father‟ and „How would you rate the risk of any damage associated with predation to your
father in the next 12 months (pupils only), „How would you rate the damage associated with
predation ever caused by jaguars to you‟ and „How would you rate the risk of any damage
associated with predation to you in the next 12 months (fathers only). Perceived impact on
human safety was assessed using the questions „Number of people ever hurt by a jaguar in the
neighbourhood‟ and „How would you rate the risk of you being hurt by a jaguar in the next
12 months‟. We summed the response scores for each of the jaguar impact scales („perceived
impact on livestock‟ and „perceived impact on human safety‟) for each respondent separately,
thus producing two impact scores ranging from 0 to 10 for each respondent (Cronbach‟s
alpha = 0.83 and 0.91).
Attitude towards jaguar killing. Respondents‟ attitudes towards killing jaguars were measured
using five-point evaluative semantic differential scales. Two items were used (useless-useful
and exciting-boring). We coded each answer -2, -1, 0, +1, or +2. One question was phrased in
favour of jaguar killing, and the other against, in an effort to balance any potential bias
associated with question direction. However responses that were favourable to jaguar killing
were always coded positively, and unfavourable responses were coded negatively, so that we
could construct a summated rating scale to summarise an individual respondent‟s overall
attitude towards jaguar killing. The attitude scale, therefore, ranged from -4 (most
unfavourable of jaguar killing) to 4 (most favourable of jaguar killing) (Cronbach‟s alpha =
0.87).
136
Social norm concerning jaguar killing. Descriptive norms were assessed using two measures:
„How many of your neighbours do you think kill jaguars?‟ (0-4, none to all); „Think of the
landowners in Alta Floresta - what percentage of them do you think kill jaguars?‟ (0-4, none
to all). The descriptive norm scale ranged from 0 (perception that no landowner in Alta
Floresta kills jaguars) to 8 (perception that all landowners in Alta Floresta kill jaguars)
(Cronbach‟s alpha = 0.80).
The above tests and treatments were conducted in the following sequence: 1. Pre-test of all
pupils and fathers; 2. Book via school and book via NGO (fathers); at least two weeks until
next step; 3. Post-test of fathers (book via school and book via NGO); 4. Information (pupils)
and child‟s homework (fathers); 5. Post-test of fathers (child‟s homework and control); 6.
Elaboration (pupils); 7. Post-test of pupils; three months until next step; 8. Delayed post-test
of pupils.
Data analysis
All analyses were conducted in SPSS, version 14. We used paired sample t-tests to compare
the six composite measures – knowledge about jaguars and predation problems, attitude
towards jaguars, perception of the impact of jaguars on human safety, perception of the
impact of jaguars on livestock, attitudes towards killing jaguars and descriptive norm
regarding the killing of jaguars – between pre-test, post-test and delayed post-test. General
Linear Model (GLM) was used to examine possible interactions between the treatments.
Results
A total of 151 pupils (68 females and 83 males) completed both pre and post-tests, 145 of
which also completed the delayed post-test, and 172 fathers were interviewed (Figure 5.1).
137
Mean pupil age was 12.63 years (SD = 1.18, range = 11-18) and mean father age was 46.38
year (SD = 5.89, range = 36-63).
Information and elaboration
Both information and elaboration had significant effects on pupils (Table 5.1 and Figure 5.2).
The comparison between pre and post-tests showed that information alone significantly
improved knowledge about jaguars, improved attitudes towards them, reduced the
perceptions of their impacts on both human safety and livestock, and made people less
favourable towards jaguar killing. Only descriptive norm was not affected by information.
Elaboration alone changed attitudes to jaguars, perceived impact on human safety and
attitudes to jaguar killing. Combined, information and elaboration significantly changed
knowledge, attitudes to jaguars, perceived impact on human safety and livestock and attitudes
to killing. However, the delayed post-test revealed that some of these effects were not
completely retained. Comparisons between post and delayed post-tests showed that
knowledge, attitudes to jaguars and perceived impacts on human safety and livestock among
pupils who had been exposed to information alone, and knowledge and perceived impact on
human safety among pupils who had been exposed to information and elaboration combined,
changed in the opposite direction than in the pre-post-test comparison (nonetheless,
differences between pre-test and delayed post-test were still significant). The effect of
elaboration alone on perceived impact on safety and attitudes to killing was retained and for
attitudes to jaguars it significantly increased between post and delayed tests.
138
Table 5.1. Mean scores  standard deviations for, and paired t-tests between, pre-test, post-test and delayed post-test on pupils‟ perceptions of jaguars.
Pre-Post Post-Delayed Pre-Delayed

Treatment Measure Pre Post t df p Delayed t df P t df p
Control Knowledge 4.07±1.12 3.82±1.06 1.42 27 0.165 4.00±1.19 -0.895 27 0.379 0.348 24 0.731
Attitudes to jaguars 2.68±2.31 2.75±2.06 -0.348 27 0.731 2.82±1.96 -0.372 27 0.713 -0.583 24 0.565
Perception of impact on safety 4.79±1.68 5.11±1.42 -1.880 27 0.071 5.11±1.22 0.000 27 1.000 -1.560 24 0.130
Perception of impact on livestock 1.03±0.88 1.07±0.85 -0.297 27 0.769 1.11±0.83 -0.273 27 0.787 -0.626 24 0.537
Attitudes to killing -0.64±1.97 -0.54±1.86 -1.000 27 0.326 -0.41±1.84 1.544 27 0.134 0.528 24 0.602
Descriptive norm 5.11±1.31 5.14±1.43 -0.328 27 0.745 5.07±1.33 -0.626 27 0.537 0.570 24 0.573
Information Knowledge 4.08±1.38 6.63±1.19 -11.129 39 <0.001 6.00± 1.24 4.038 39 0.000 -7.980 38 <0.001
Attitudes 2.55±2.49 3.62±2.15 -4.611 39 <0.001 2.97±2.57 2.579 39 0.014 -3.597 38 0.001
Perception of impact on safety 4.67±1.84 1.72±1.34 15.312 39 <0.001 2.50±1.80 -3.740 39 0.001 9.618 38 <0.001
Perception of impact on livestock 1.05±1.21 0.37±0.92 4.970 39 <0.001 0.67±0.92 -2.762 39 0.009 2.564 38 0.014
Attitudes to killing -0.35±1.96 -1.77±1.74 8.924 39 <0.001 -1.65±1.76 -1.403 39 0.168 9.313 38 <0.001
Descriptive norm 4.95±1.93 5.05±1.62 -1.071 39 0.291 5.05±1.83 0.255 39 0.800 -1.356 38 0.183
Elaboration Knowledge 3.91±1.04 3.96±1.11 -0.298 44 0.767 4.13±1.16 -1.034 44 0.307 -1.279 42 0.208
Attitudes 2.60±2.05 3.22±3.67 -2.213 44 0.032 3.56±3.53 -2.236 44 0.030 -3.473 42 0.001
Perception of impact on safety 4.78±1.63 3.09±3.07 5.703 44 <0.001 3.11±3.24 -0.172 44 0.864 5.093 42 <0.001
Perception of impact on livestock 1.11±1.19 1.18±1.28 -0.684 44 0.497 1.11±1.17 0.596 44 0.554 0.000 42 1.000
Attitudes to killing -0.27±1.65 -0.71±2.31 2.379 44 0.022 -0.84±2.35 1.633 44 0.110 2.857 42 0.007
Descriptive norm 4.80±1.87 4.69±1.86 1.530 44 0.133 4.82±2.02 -1.354 44 0.183 -.330 42 0.743
Information Knowledge 3.95±1.27 7.08±1.38 -12.493 37 <0.001 6.63±1.42 2.340 37 0.025 -8.889 37 <0.001
+ Attitudes 2.82±2.29 4.52±4.00 -4.865 37 <0.001 4.68±3.76 -1.356 37 0.183 -5.676 37 <0.001
elaboration Perception of impact on safety 5.03±1.55 1.76±1.44 15.619 37 <0.001 2.26±1.97 -3.452 37 0.001 10.991 37 <0.001
Perception of impact on livestock 1.18±1.13 0.24±0.63 7.267 37 <0.001 0.26±0.53 -0.329 37 0.744 6.034 37 <0.001
Attitudes to killing -0.50±1.75 -1.66±1.88 7.333 37 <0.001 -1.81±1.95 1.639 37 0.110 6.963 37 <0.001
Descriptive norm 4.60±2.04 4.71±2.15 -0.813 37 0.422 4.71±2.05 0.000 37 1.000 -1.434 37 0.160
139
Figure 5.2. Variation in pupils‟ perceptions of jaguars between pre-test, post-test and delayed
post-test in response to the following treatments: control (◊), information (●), elaboration
(▲), and information plus elaboration (■). * denotes p < 0.05, ** denotes p <0.01.
140
As shown in figure 5.3, the direction of the effect of elaboration on attitudes to jaguars and
attitudes to jaguar killing depended on pupils‟ initial attitudes. The interaction between
attitudes to jaguars and test (pre versus post) was significant for elaboration alone (GLM: F =
17.236, df = 1, p < 0.001) and elaboration and information combined (GLM: F = 19.738, df =
1, p < 0.001). The interaction between attitudes to jaguar killing and test was significant for
elaboration alone (GLM: F = 4.190, df = 1, p = 0.047), but not for elaboration and
information combined (GLM: F = 0.780, df = 1, p = 0.383). Among pupils who had positive
attitudes towards jaguars (attitude scores 1 or above; mean±SD = 3.15±1.27) in the pre-test,
attitudes became significantly more positive after elaboration alone (mean±SD = 4.12±2.62)
(t = -3.723, df = 39, p = 0.001), or information and elaboration combined (pre-test: mean±SD
= 3.35±1.63; post-test: mean±SD =5.50±2.83) (t = -6.977, df = 33, p < 0.001). Among pupils
who had neutral or negative attitudes towards jaguars (attitude scores 0 or below; mean±SD =
-1.80±1.79) in the pre-test, attitudes became more negative after exposure to elaboration
alone (mean±SD = -4.00±2.74) (t = 4.491, df = 4, p = 0.011), or elaboration and information,
although this latter change was not significant (pre-test: mean±SD = -1.75±2.06; post-test:
mean±SD =-3.75±2.75) (t = 2.828, df = 3, p = 0.066).
Likewise, in pupils who had negative attitudes to jaguar killing (attitude scores 0 or below;
mean±SD = -1.14±1.41) in the pre-test, attitudes became significantly more negative after
elaboration alone (mean±SD = -1.83±1.65) (t = 3.839, df = 28, p = 0.001), or information and
elaboration combined (pre-test: mean±SD = -1.22±1.55; post-test: mean±SD =-2.40±1.52) (t
= 6.150, df = 26, p < 0.001). However, among pupils who had positive attitudes to jaguar
killing (attitude scores 1 or above; mean±SD = 1.31±0.45) in the pre-test, attitudes did not
change after exposure to elaboration alone (mean±SD = 1.31±1.96) (t = 0.000, df = 15, p =
141
1.000), and became less positive after exposure to information and elaboration combined pre-
test mean±SD = 1.27±0.47; post-test: mean±SD =0.18±1.33) (t = 3.833, df = 10, p = 0.003).
Inter-generational learning
Fathers were affected significantly both by books and by their children (Table 5.2 and Figure
5.4). Books alone increased fathers‟ knowledge and decreased their perception of the impact
of jaguars on human safety, regardless of the means through which they received the book i.e.
from an NGO or the local school. However, the significance of the effect was greater among
those who received the book from the school (p < 0.001 against 0.036 for knowledge and p =
0.006 against 0.026 for perception of impact on human safety). Besides, books received from
the local school caused a decrease in fathers‟ perception of descriptive norm related to jaguar
killing. Children‟s homework improved their fathers‟ attitudes towards jaguars and decreased
their perception of descriptive norm, but did not significantly affect other variables. In
contrast receiving the book from the school combined with seeing their child‟s homework
increased fathers‟ knowledge, improved their attitudes towards jaguars, decreased their
perceptions of the impact of jaguars on both human safety and livestock, and decreased their
perception of descriptive norm. Only attitude towards killing remained unaffected – and this
was the main metric which it would be good to change.
142
Figure 5.3. Variation in attitudes towards jaguars and jaguar killing between pre-test and
post-test among pupils exposed to elaboration alone (▲ and Δ) and elaboration and
information combined (■ and □). Solid shapes denote positive initial attitudes while hollow
shapes denote neutral and negative initial attitudes. * denotes p < 0.05, ** denotes p <0.01.
143
Table 5.2. Mean scores  standard deviation for, and paired t-tests between, pre-test and post-
test on fathers‟ perceptions of jaguars.
Treatment Dependent variable Pre Post t df p

Control Knowledge 5.07±1.65 5.03±1.63 0.297 29 0.769
Attitudes to jaguars 1.70±1.93 1.63±1.85 1.000 29 0.326
Perception of impact on safety 3.63±1.79 3.60±1.85 0.441 29 0.662
Perception of impact on livestock 1.20±1.06 1.30±1.09 -1.140 29 0.264
Attitudes to killing -0.43±2.02 -0.30±1.90 -1.439 29 0.161
Descriptive norm 5.63±1.40 5.60±1.30 0.297 29 0.769
Book via Knowledge 5.10±1.60 5.42±1.52 -2.177 39 0.036
conservation Attitudes to jaguars 1.75±1.86 1.85±1.76 -1.433 39 0.160
organization Perception of impact on safety 3.90±1.76 3.57±1.79 2.314 39 0.026
Perception of impact on livestock 1.22±1.05 1.27±1.11 -0.628 39 0.534
Attitudes to killing -0.17±1.93 -0.30±1.71 1.152 39 0.256
Descriptive norm 5.10±1.90 5.15±1.80 -0.495 39 0.623
Book via Knowledge 4.90±1.51 6.00±1.69 -5.284 39 <0.001
school Attitudes to jaguars 1.95±2.26 2.20±2.14 -1.818 39 0.077
Perception of impact on livestock 1.20±0,92 1.00±0.93 1.433 39 0.160
Child‟‟s Knowledge 4.97±1.24 5.03±1.24 -0.701 29 0.489
homework Attitudes to jaguars 1.43±4.85 2.57±4.14 -4.753 29 <0.001
Perception of impact on livestock 1.20±0.66 1.13±0.94 0.626 29 0.536
Attitudes to killing -0.30±2.19 -0.43±2.06 1.000 29 0.326
Book via school + Knowledge 4.8±11.73 6.28±2.05 -6.679 31 <0.001
Child‟s Attitudes to jaguars 1.43±3.20 4.15±2.54 -8.718 31 <0.001
homework Perception of impact on safety 3.97±1.38 2.75±1.52 4.978 31 <0.001
Perception of impact on livestock 1.22±0.83 0.50±0.76 5.578 31 <0.001
Descriptive norm 5.28±1.90 2.84±1.87 7.012 31 <0.001
144
Figure 5.4. Variation in fathers‟ perceptions of jaguars between pre-test and post-test in
response to the following treatments: control (◊), book via conservation organization (○),
book via child‟s school (●), child‟s homework (▲), and book via child‟s school plus child‟s
homework (■). * denotes p < 0.05, ** denotes p <0.01
145
This study shows that school-based education and communication interventions can have a
powerful impact on pupil‟s perceptions of jaguars, and those of their fathers; this process
could be used to positive conservation effect. The finding that pupils can influence their
fathers‟ perceptions of jaguars suggests that conservationists can use rural schools to reach at
once tens of students in a classroom, or hundreds on the school‟s soccer pitch, who will in
turn transfer the conservation message to their fathers. Given the logistical challenge of
visiting landowners one-by-one at home in rural Amazonia, this strategy might be relatively
cost-effective. The exceptionally strong prominence of the jaguar in people‟s hearts and
minds (Chapter 2), combined with the relatively high rate of primary school enrolment in
Brazil (97.5%) (Unesco 2007) and the willingness of public school directors and teachers to
cooperate with conservationists, renders school-based intergenerational learning a
particularly promising approach for jaguar conservation.
Attitudes and descriptive norms related to jaguar killing are the most immediate determinants
of jaguar killing behaviour among farmers and ranchers in Amazonia (Chapter 4). Therefore
perhaps the most relevant effect detected among fathers in this study was the decrease in their
perceptions of how common the killing of jaguars is among their neighbours.
Conservationists have used education and communication to increase knowledge and
improve attitudes among target stakeholders and, in so doing, change stakeholders‟
behaviours for the benefit of conservation (Jacobson et al. 2006). However, attempts to
influence social norms regarding conservation-orientated behaviours have been far less
considered. We found that books distributed via local schools changed descriptive norms
among fathers, while [the same] books distributed via a conservation organisation did not.
This result suggests that parents‟ perceptions can be influenced not only by the information
146
explicitly conveyed in the content of books and their children‟s homework, but also by the
implicit message that a community institution (and therefore other community members)
supports jaguar conservation more than they had realised. The use of role-models, case
studies, and examples of coexistence with jaguars, could conceivably enhance the power of
school-based communication campaigns to create or redefine social norms concerning
conservation-orientated behaviours.
This study also demonstrated the potentially detrimental effect of prejudiced communication
compounding negative attitudes towards species involved in human-wildlife conflict. Pupils‟
attitudes towards jaguars – both positive and negative – were reinforced by the elaboration
exercise; even the negative attitudes held by a few pupils, towards jaguars, were
strengthened. Possible explanations for this polarisation of attitudes are learning, self-
persuasion and reactance. Pupils may have learned selectively the new information raised and
shared by the group, about the benefits and drawbacks of coexisting with jaguars, to support
their schemas concerning the species, i.e. pre-existing biases that provide a framework or
structure for their beliefs regarding jaguars (Brewer 1981). Attitudes can also become more
extreme or polarized as a result of thinking, a phenomenon called self-persuasion. The
attitude change that occurs is not the result of information that originates externally but rather
the result of thoughts, ideas and arguments generated within oneself (Petty and Cacioppo
1996). Finally, reactance might also account for our results. During the elaboration
intervention, pupils were encouraged to share their opinions with the rest of the group
without knowing that they would subsequently have to justify their views. Only a few pupils
in each group expressed negative opinions about jaguars and their views were challenged by
the majority of their classmates. By defending their opinions under peer pressure, those
pupils became more convinced that they were right or developed even more extreme views, a
147
phenomenon known as the boomerang effect or reactance (Brehm 2000). Reactance can
occur in anyone who believes that his or her freedom to choose freely how to think, feel, or
act may be or has been limited. Conservationists should become acquainted with
stakeholders‟ views and attitudes, and consider carefully the implications of exposing or
challenging them, for fear of causing an unintended, perverse strengthening of a negative
view. Failure to consider reactance can cause communication interventions to backfire,
producing the opposite effect to that intended. Where group identity is strong, as is the case
among landowners in organized settlements on the Amazon deforestation frontier (Chapter
4), reactance is likely to be enhanced if the persuasive message is seen to stem from outside
the group. When reactance is likely to be an issue, communication campaigns would arguably
be more effective if they rely on descriptive norms (involving perceptions of which
behaviours are typically performed) rather than injunctive norms (involving perceptions of
which behaviours are typically disapproved of) (Cialdini 2003), and strive to promote
positive conduct (with prescriptive messages) rather than demote negative conduct (with
proscriptive messages) (Winter et al. 2000).
Information alone significantly improved knowledge, attitudes towards jaguars, and
perceptions of jaguar impacts on human safety and livestock among pupils and fathers.
However some of these changes did not endure. Because knowledge of jaguars among
children and adults in rural Amazonia is typically poor, and most people in the region over-
estimate the real impact of jaguars on human livelihoods (Chapters 2 and 3), information-
based interventions (e.g. lectures and books) were indeed expected to improve knowledge
and perceptions regarding the species. Although such interventions did improve pupils‟
knowledge, attitudes and perceptions in post-tests conducted immediately afterwards, these
effects (while still significantly different) were not sustained at the same level three months
148
later. Information and elaboration together had a stronger effect than information alone on
knowledge, attitudes towards jaguars and perceived impact on livestock. As expected, these
changes were more enduring than those produced by knowledge alone. Information may have
moderated the effect of elaboration on pupils‟ attitudes towards killing jaguars: while positive
attitudes towards killing remained unchanged after elaboration alone, they decreased
significantly when elaboration was preceded by information. In contrast, however, the same
moderating effect of information was not observed on pupils‟ negative attitudes towards
jaguars per se. This finding suggests that attitudes towards jaguars as a species are less
rational, and may be more entrenched, than attitudes towards killing them. While attitudes
towards killing jaguars may be manipulated using information-based interventions, education
and communication approaches aimed at ameliorating strongly negative attitudes towards
jaguars themselves may be more effective if they capitalise particularly on the strong, mixed
emotions elicited by the species (Chapter 2).
149
Chapter 6 – Summary and recommendations
CHAPTER 6
GENERAL SUMMARY AND RECOMMENDATIONS FOR FUTURE

RESEARCH
In the end we will conserve only what we love,

we love only what we understand, and we will understand only what we are taught
Baba Dioum
This study has provided new insights into the human dimensions of conflicts between people
and jaguars, exploring approaches to understand, predict and influence people‟s perceptions
of jaguars in the species‟ two largest strongholds, namely the Pantanal and Amazonia.
Negative perceptions about jaguars ultimately lead to jaguar killing, and killing jaguars is a
major threat to the species. Therefore understanding perceptions regarding jaguars among
landowners in the Pantanal and Amazonia, and identifying effective interventions to improve
them, are key elements to the future safeguarding of jaguar populations.
Findings
Our study showed that perceptions behind human-jaguar conflicts in Brazil are more strongly
related to social and psychological factors than previously understood, and may therefore be
more readily influenced through communication and education than has been thought. While,
not surprisingly, we found that perceptions of the impact of jaguars on livestock are
important elements in human-jaguar conflict, they are not the only predictors of the intention
to kill jaguars. Fear, personal and social motivation, and internal and external barriers to
killing jaguars can also influence jaguar killing behaviour. The relative importance of these
factors in determining intention to kill varies with region and affluence. Livestock loss acts
150
in combination with attitudes, knowledge and perceptions of the economic situation to
determine how people perceive the impact of jaguars on their livelihoods. Attitudes and
knowledge, in turn, are influenced by age, gender and whether respondents live in urban or
rural areas.
Our study also revealed that jaguars are the most prominent of all native mammals in the
perceptions of young Brazilians. This prominence can not be fully explained by knowledge
about jaguars or by direct experiences with the species, such as household livestock predation
problems. Young people predominantly perceive the jaguar as threatening, but also as
beautiful and endangered. Indeed, most adjectives attributed to jaguars are emotive. Drawing
on these findings, we designed communication and education tools, including an illustrated
manual and activity books, to improve perceptions of jaguars among school children and their
fathers on the Amazon deforestation frontier. We found that school-based communication
interventions can have a powerful effect on perceptions of jaguars among pupils, and that
pupils can successfully transfer that effect to their fathers. In our communication experiment,
fathers were influenced not only by the information conveyed explicitly in the content of
books and their children‟s homework, but also by the implicit message that a community
institution (i.e. the local school) - and therefore other community members - support jaguar
conservation more than they may have realised.
Implications
Most research, policies and programs related to jaguar conservation have focused narrowly
on the biology of the species or the immediate economic losses caused by jaguars to people
through livestock predation. Technical solutions for the problem of livestock predation
(Hoogesteijn 2000) have received considerable attention, and efforts to increase people‟s
151
tolerance of jaguars and discourage jaguar killing have focused largely on economic
incentives (e.g., monetary compensation for livestock loss; Silveira et al. 2006) and legal
prohibitions and sanctions (e.g., establishment of protected areas). The role and importance of
social and psychological factors have been far less considered.
Our findings, however, highlight the importance of social factors in jaguar killing. Social
norms (descriptive norm and subjective norm) and social identity affect jaguar killing
behaviour and should be taken into account by conservationists. A number of techniques can
be used to change social incentives and barriers. For instance, the use of role-models, case
studies and examples of coexistence with jaguars, combined with information about the
negative consequences of killing them, might help to create or redefine a social norm by
explaining that the community condemns rather than condones or glamorises the killing of
jaguars. Arranging for conspicuous and respected group members or community institutions
(e.g., cooperatives and rural schools) to promote tolerance of jaguars may be one way of
influencing other group members. Communicating information about the behaviour and
practices of group members via informal social networks is another possible strategy.
Conservation communicators should draw upon the exceptional social significance of the
jaguar, and the current societal shift in values from materialism (focused on the use and
management of wildlife for human benefit) toward post-materialism (focused on mutualistic
orientations towards wildlife) (Inglehart 1990; Manfredo et al. 2003), to promote the social
benefits of protecting, rather than killing, jaguars. Brazilian society is becoming increasingly
environmentally conscious. As a result people associated with conservation efforts enjoy
increased status, especially when these involve charismatic species, such as jaguar, or
habitats of worldwide concern, such as those of the Pantanal and Amazonia.
152
This study also revealed the importance of perceptions, knowledge, skills, attitudes and
emotions such as fear and excitement in inciting the killing of jaguars. Education and
information-intensive campaigns might be used to influence these factors in order to deter
jaguar killing. Education campaigns addressing livestock predation by jaguars might help
ranchers to identify correctly a predation event, implement preventive measures, and adjust
exaggerated perceptions of jaguar impact on livestock to realistic levels. Campaigns
conveying the true probability of a jaguar attack on humans might be particularly effective in
preventing jaguar killing among small landowners in Amazonia. Knowledge of jaguars
among children and adults in rural Amazonia is typically poor, and most people in the region
over-estimate the real impact of jaguars on human livelihoods. Therefore information-based
interventions should significantly improve knowledge, attitudes towards jaguars, and
perceptions of jaguar impacts on human safety and livestock, as they did among the school
pupils and fathers that participated in our education experiment. The educational effects of
information in our experiment were more persistent when combined with thoughtful
consideration about the messages conveyed, i.e. cognitive elaboration. Therefore in order to
achieve more powerful and enduring results, conservationists should consider using a
combination of passive and active learning techniques, while being aware of the fact that
elaboration can reinforce pre-existing negative attitudes. In our experiment, strong pre-
existing negative attitudes towards jaguars among pupils were reinforced by the elaboration
exercise. A possible explanation for this observation is „reactance‟, a phenomenon in which a
person, who believes that his or her freedom to choose freely how to think, feel, or act may
be or has been limited, becomes more convinced that he or she is right, or develops even
more extreme views (Brehm 2000). When reactance is likely to be an issue, communication
campaigns would arguably be more effective if they relied on descriptive norms (involving
perceptions of which behaviours are typically performed) rather than injunctive norms
153
(involving perceptions of which behaviours are typically disapproved of) (Cialdini 2003).
Such campaigns should strive to promote positive conduct (with prescriptive messages)
rather than demote negative conduct (with proscriptive messages) (Winter et al., 2000).
Because jaguar killing results from the interplay between external incentives
(economic/market, legal/administrative/coercive, and social/normative) and internal
influences (such as knowledge, skills, emotions, norms, attitudes and intentions), effective
strategies for preventing jaguar killing must be based on the social and psychological aspects
of human-jaguar relationships as well as on legal and economic considerations.
Conservationists need to find and support ways of making jaguar killing not only unprofitable
and illegal, but also socially and personally unacceptable. In order to do so, we will have to
broaden our scope and look beyond the biology of jaguars and the short-term economics of
livestock predation on individual ranches.
154
Future directions: zooming out for a broader perspective
Jaguars versus pumas
Future studies on people‟s perceptions of jaguars should also address the puma. Like jaguars,
pumas are big cats known for causing damage to livestock (Mazzolli et al. 2002) and
attacking and killing people (Quigley and Herrero 2005). In Brazil, the two species have the
same popular name, onça. The distinction between them is made by the second part of their
compound names: onça-pintada (spotted onça) for jaguar and onça-parda (brown onça) for
puma. This biological and etymological proximity can lead people to think of jaguars and
pumas as the same animal, or at least to confuse them, transferring attributes from one
species to the other. There is indeed room for confusion, more so in Brazil than anywhere
else, and this has important implications for research and management. For instance, when a
landowner says onça in his answer to a survey question, he can be referring to the jaguar, to
the puma or to both. Likewise, the loss of livestock to pumas can be used to justify the
retaliatory killing of jaguars and vice-versa. Nonetheless, the close association between
jaguars and pumas may offer a unique opportunity for researchers to investigate the role of
perceptions in the conflicts between people and big cats. Despite the similarity between the
two species, jaguars have a much greater cultural, social and psychological significance than
pumas. Witness the conspicuous occurrence of the jaguar in art, literature and folklore in
Brazil, as opposed to the scant presence of the puma [for instance, the jaguar features as the
main character in more books for children and adolescents than any other native species
while the puma is virtually absent from children‟s books (Appendix I)], and the prominence
of jaguars relative to pumas in the hearts and minds of young people (Chapter 2). The
significance of the puma has most likely been overshadowed by the exceptional prominence
of the jaguar. This difference may affect the way people relate to the two species. Comparing
information about the real impact of jaguars versus pumas on human livelihoods, and
155
examining differences in the way people react to those impacts might help to disentangle
rational, objective reasons for lethal control from emotional, deep-seated motivations for
persecution.
Conflicts between humans over jaguars
Conflicts involving wildlife are, ultimately, often conflicts between groups of people with
different views on how wildlife should be managed (Madden 2008; Marshall et al. 2007).
When people feel that their own needs are being subordinated to those of wildlife, or that
wildlife conservationists exclude them from decisions that affect their interests, they may
retaliate against or reduce cooperation with conservation organisations and authorities. The
influence of group identity on people‟s intention to kill jaguars, identified in our study,
reveals the potential for conflicts between human groups over jaguars. If there is a perceived
association between jaguar killing and in-group members (i.e. ranchers), and this is coupled
with a perceived association between jaguar conservation and out-group members, then two
results may occur. First conservation efforts considered to emanate from outside the group
may be poorly accepted, and second conflict between ranchers and jaguars may be replaced
by conflict, about jaguars, between ranchers and conservationists. Conflicts between people
over jaguars can be aggravated by the urban-rural divide. There is an exceptionally high rate
of urbanisation (1.8%) among the world‟s already highly urbanised countries, and Brazil is
among these, with 86% of the total population living in urban areas (CIA World Factbook).
Ranchers and farmers are a minority group in Brazil, and their numbers are dwindling. As a
result they may associate jaguar conservation with urban values that are increasingly imposed
on them, and might view the continuation of jaguar killing as part of their resistance to this
and their struggle to preserve their rural heritage. The recent growth of jaguar-based tourism
in the Pantanal also warrants attention. Differing views between wildlife biologists and
156
tourist operators on jaguar management in the region have been aggravated by the first-ever
documented, unprovoked, fatal attack by a jaguar on a human in Brazil. This occurred on 24th
June 2008, when a young fisherman was killed by a jaguar while sleeping in his tent on a
bank of the Paraguay River in the Pantanal. One speculation is that this unique incident
occurred because jaguars in the area had become habituated to people by the use of baits to
attract jaguars for photo-tourism. If local landowners fail to distinguish between the interests
of wildlife biologists and tourist operators, the negative impacts of poorly planned tourism
may damage the credibility and trustworthiness of jaguar conservation organisations among
them. If effective jaguar management strategies are to be identified and met, we need first to
understand the social relationships behind jaguar-related conflicts.
From biological to multidisciplinary to interdisciplinary approaches
Conservation problems are caused by human behaviour. Human-Wildlife Conflicts (HWC)
are conservation problems that have their human element explicitly acknowledged in their
name. Despite the diversity of situations and species that spawn HWC, there is one common
thread: the thoughts and actions of humans ultimately determine the course and resolution of
the conflict. Because humans are the constant in HWCs, approaches that aim at
understanding and influencing human behaviour at both individual and group levels are
critical for dealing with them. Nonetheless, efforts to understand and resolve HWC have
focused mainly on biological issues, rather than on psychological, anthropological,
sociological, educational and communication perspectives. Our finding that social and
psychological factors may be important in determining people‟s perceptions of jaguars
corroborates the view that resolving human-jaguar conflicts requires input from both the
natural and social sciences. In this research, we explored human-jaguar conflict from a
cognitive psychology perspective.
157
Cognitive psychology draws on psychology and neuroscience. Both cognitive psychologists
and conservationists are interested in the practicalities of effecting changes in human
behaviour. Therefore conservationists may have more in common with cognitive
psychologists than they do with branches of social science that are preoccupied only with
understanding human behaviour. In order to understand and change the way people relate to
jaguars, the following challenges are faced: 1) engaging social scientists in jaguar research
and conservation; 2) overcoming the traditional lack of communication and mutual
understanding between natural and social scientists (Mascia et al. 2003; Campbell 2005); and
in so doing, 3) setting the stage for a shift from the broader, plural perspectives of a
multidisciplinary approach to the common ground achieved by the integration of knowledge
and techniques from multiple disciplines that characterises true interdisciplinarity
(Kokelmanns 1979).
From ranch to region to range
Jaguar conservation has been implemented or at least considered at difference spatial scales,
ranging from a single ranch to the entire species‟ range (Silveira et al. 2006; Rabinowitz and
Zeller 2010). Concrete efforts to resolve human-jaguar conflicts specifically, however, have
concentrated mostly on individual ranches. The emphasis on technical solutions to livestock
predation problems, and compensation schemes to alleviate jaguar-livestock conflict, explain
this focus on the ranch as the unit of analysis and intervention. The use of electric fencing,
aversive conditioning, changes in livestock husbandry practices (Hoogesteijn 2000), and
small-scale compensation programmes (Silveira et al. 2006), are examples of ranch-scale
interventions. However it is increasingly acknowledged that analysis and interventions should
be implemented at broader scales. For instance, Cavalcanti (2008) suggests that in the
Pantanal tax benefits and special credit agreements for ranchers are required, together with a
158
regional increase in beef prices to compensate for losses associated with grazing cattle in an
area of high jaguar density. Rabinowitz and Zeller (2010) propose a holistic model to identify
least-cost corridors connecting the 90 known jaguar populations across its range. Our study
revealed the centrality of perceptions in conflicts between people and jaguars, and our
understanding of perceptions of jaguars may require three levels of analysis: ranch, region
and range. At the level of individual ranches, the analysis focuses on potentially relevant
psychological attributes of landowners such as knowledge, skills, attitudes and feelings, and
on local contextual factors such as livestock losses to jaguars. Information obtained at the
ranch level suggests how to influence the behaviour of ranchers to the benefit of jaguar
conservation. Regional level analysis addresses the social phenomena that influence people‟s
perceptions of jaguars, such as social norms and social identity. Information obtained at the
regional scale can contribute to the design of effective communication interventions. Finally,
range-level factors account for differences in perceptions across social and cultural
conditions. At the range-scale, we examine the similarities and differences between countries
or regions in order to assess the degree to which findings from one place may be extrapolated
to others, or what is culture-specific in people‟s perceptions of jaguars (Manfredo 2004).
After all, can people and jaguars coexist?
The quotation at the beginning of this chapter comes from a speech made by the Senegalese
environmentalist Baba Dioum in 1968, in New Delhi, to the general assembly of the
International Union for Conservation of Nature. We use Dioum‟s quotation to structure the
answer to the question above.
In the end we will conserve only what we love
Manfredo (2008) places emotions at the heart of human attraction to, and conflict over,
wildlife. This component of human-wildlife interactions has largely been neglected, probably
159
because conservationists emphasise rationality as a means of combating emotional
arguments, and because, for academics, emotions are both difficult to measure and
conceptually complex. Emotion and cognition involve different parts of the brain, but interact
in decision-making. Following this argument, we will coexist with jaguars only if we „love‟
them. But according to Maslow (1954), love is a higher order emotion most fully expressed
when more basic needs (for safety, shelter and sustenance) have already been met. Unlike
wolves and bears in Europe and North America, jaguars share the land mostly with people
whose basic needs have not been met, and who are likely therefore to be more concerned
with their own physical security and economic well-being than with the status of jaguar
populations. Whether people will love and conserve jaguars, or hate and kill them, depends to
a great extent on the prospects for socioeconomic development and poverty reduction in rural
Latin America.
We love only what we understand
People will „love‟ more than „hate‟ jaguars when they understand their true value. The
aesthetic value of the jaguar is already widely recognized. Science will be crucial in revealing
and quantifying the jaguar‟s ecological and economic values, as well as the cost-effectiveness
of coexisting with jaguars by contrasting its economic value with the costs associated with
livestock predation. Coexistence between people and jaguars is more likely to be achieved,
though, when the social, cultural and emotional values of the jaguar are also fully
appreciated. However, the value people place on jaguars will depend upon their own values.
[Rohan (2000) draws a distinction between the term value when used as a verb (assigned on
the basis of meaning, goodness and worth) and as a noun (a stable and enduring cognitive
measure), such that the process of valuing (verb) something is based on the enduring values
(noun) that a person holds.] Coexistence between people and jaguars may depend not only on
160
the effectiveness of economic instruments that make people value (verb) them more, but also
on the speed with which the luxury of post-materialist values (noun) (basically, quality of life
and mutualistic orientations towards wildlife; Inglehart 1990) are distributed among those
who share the land with jaguars.
We will understand only what we are taught
Most of what we know about and feel towards wildlife is learned. Jaguars may elicit
instinctive, genetically determined fear and aggression. Hart and Sussman (2009) argue that
our evolutionary roots were shaped by a world in which humans were primarily a prey
species (mostly to lions and tigers) and the threat to survival imposed by large cats would
have selected for anti-predator adaptations in humans. Shepard (1973) proposes that the
desire to hunt and pursue animals is an innate characteristic of our species and Wilson‟s
biophilia hypothesis (1984, 1993) posits that humans have innate tendencies to focus on and
respond positively to natural objects, including wildlife. Nonetheless, it is clear that whether
we love animals, kill them, or both, in specific contexts represents an interaction of our
evolved tendencies and learned thoughts and behaviours (Manfredo 2008). While there is
nothing we can do about our genetic heritage, we can influence values, attitudes, feelings and,
ultimately, behaviours, especially of young people, through education. Whether people and
jaguars can coexist in the longer term will depend, more than anything else, on our ability to
educate the younger generation of Brazil and Latin America.
161
Literature cited
LITERATURE CITED
Ajzen, I. 1985. From intentions to actions: A theory of planned behavior. Pages 11-39 in J.
Kuhl. and J. Beckman, editors. Action-control: From cognition to behaviour.
Springer, Heidelberg, Germany.
Ajzen, I. and M. Fishbein. 1980. Understanding attitudes and predicting social behaviour.
Prentice-Hall, New Jersey.
Almeida, A. de. 1976. Jaguar hunting in the Mato Grosso. Stanwill Press, London, UK.
Almeida, A. de. 1986. A survey and estimate of jaguar populations in some areas of Mato
Grosso. Pages 80-89 in B. des Clers, editor. Wildlife Management in the Neotropical
Moist Forest: Conservation status of thejaguar (Panthera onca). Conseil International
de la Chasse et de la Conservation du Gibier, Paris, France.
Andersone, Z. and J. Ozoliš. 2004. Public Perception of Large Carnivores in Latvia. Ursus
15(2):181-187.
Armitage, C. J. and M. Conner. 2001. Efficacy of the theory of planned behaviour: A meta-
analytic review. British Journal of Social Psychology 40:471-499.
Azevedo, F.C.C. and D.L. Murray. 2007. Evaluation of potential factors predisposing
livestock to predation by jaguars. The Journal of Wildlife Management 71:2379-2386.
Baker, S.E. and D.W. Macdonald. 2000. Foxes and foxhunting on farms in Wiltshire: a case
study. Journal of Rural Studies 16:185-201.
Ballantyne, R., S. Connell and J. Fien. 1998. Students as catalysts of environmental change:
A framework for researching intergenerational influence through environmental
education. Environmental Education Research 4:285-298.
Ballantyne, R., J. Fien and J. Packer. 2001. Program effectiveness in facilitating
intergenerational influence in environmental education: lessons from the field. Journal
of Environmental Education 32(4):8-15.
Bandura, A. 1977. Social Learning Theory. General Learning Press, New York.
Barros, A. L. 1998. Gente Pantaneira: Crônica de sua História. Nova Aguilar Press, Rio de
Janeiro, Brazil.
Bath, A., A. Olszanska and H. Okarma. 2008. From a Human Dimensions Perspective, the
Unknown Large Carnivore: Public Attitudes Toward Eurasian Lynx in Poland,
Human Dimensions of Wildlife 13(1):31 - 46
Beedell, J. and T. Rehman, 2000. Using social-psychology models to understand farmers'
conservation behaviour. Journal of Rural Studies 16(1):117-127.
Beier, P. 2001. Cougar attacks on humans in the United States and Canada. Wildlife Society
Bulletin 19:403-412.
Bjerke, T., T.S. Odegardstuen and B.P. Kaltenborn. 1998. Attitudes toward animals among
Norwegian adolescents. Anthrozoos 11(2):79-86.
Boggs, D.L. 1991. Civic education: an adult-education imperative. Adult Education Quarterly
42(1):46-55.
Bonwell, C. and J. Eison. 1991. Active Learning: Creating Excitement in the Classroom
AEHE-ERIC Higher Education Report No.1. Jossey-Bass, Washington, D.C.
Brazilian Institute of Geography and Statistics. 2007. Pesquisa nacional por amostra de
domicílios: Volume Brasil 2007. Rio de Janeiro, Brazil.
Brazilian Institute of Geography and Statistics. 2009. Pesquisa nacional por amostra de
domicílios. Rio de Janeiro, Brazil.
Borgerhoff-Mulder, M. and P. Coppolillo. 2005. Conservation: Linking Ecology, Economics,
and Culture. Princeton: Princeton University Press.
Brehm, J. W. 2000. Reactance. Oxford University Press, Oxford, UK.
162
Literature cited
Brewer, W. F. and J.C. Treyens. 1981. Role of schemata in memory for places. Cognitive
Psychology 13:207-230.
Browne-Nuñez, C. and S.A. Jonker. 2008. Attitudes Toward Wildlife and Conservation
Across Africa: A Review of Survey Research, Human Dimensions of Wildlife, 13:1,
47-70
Bruskotter, J.T., R.H. Schmidt and T.L. Teel. 2007. Are attitudes toward wolves changing? A
case study in Utah. Biological Conservation 139:211-18.
Campbell, L. 2005. Overcoming obstacles for interdisciplinary research. Conservation
Biology 19(2):574-577.
Capobianco, J.P.R. A. Veríssimo,A. Moreira, D. Sawyer and I. Santos, I. 2001.
Biodiversidade na Amazônia Brasileira: avaliação e ações prioritárias para a
conservação, uso sustentável e repartição de benefícios. Estação Liberdade, São
Paulo, Brazil.
Caso, A., C. Lopez-Gonzalez, E. Payan, E. Eizirik, T.G. Oliveira, R. Leite-Pitman, M. Kelly,
and C. Valderrama. 2008. Panthera onca. In: IUCN 2009. IUCN Red List of
Threatened Species. Version 2009.2.
Cavalcanti, S.M. 2008. Predator-prey relationships and spatial ecology of jaguars in the
southern Pantanal, Brazil: implications for conservation and management. PhD
Thesis. Utah State University.
Cavalcanti, S.M., S. Marchini, A. Zimmermann, E.M. Gese and D.W. Macdonald. 2010.
Jaguars, livestock and people: reality and perceptions behind the conflicts in Brazil.
Pages 383-402 in D.W. Macdonald and A. Loveridge, editors. The Biology and
Conservation of Wild Felids. Oxford University Press, Oxford, UK.
Central Intelligence Agency. The World Factbook. Retrieved 20 April 2010, from
https://www.cia.gov/library/publications/the-world-factbook
Chavez, A.S. and E.M. Gese. 2005. Food habits of wolves in relation to livestock
depredations in northwestern Minnesota. American Midland Naturalist 154:253-263.
Chavez, A.S., E.M. Gese and R.S. Krannich. 2005. Attitudes of rural landowners toward
wolves in northwestern Minnesota. Wildlife Society Bulletin 33:517-527.
Cheung, S.F., D.K.S. Chan and Z.S.Y. Wong. 1999. Reexamining the theory of planned
behaviour in understanding wastepaper recycling, Environment and Behavior 31:587-
612.
Chickering, A. W. and Z.F. Gamson. 1987. Seven Principles for Good Practice. AAHE
Bulletin 39:3-7.
Cialdini, R.B. 2003. Crafting normative messages to protect the environment. Current
Directions in Psychological Science 12(4):105-109.
Cialdini, R.B., R.R. Reno and C.A. Kallgren. 1990. A focus theory of normative conduct:
Recycling the concept of norms to reduce littering in public places. Journal of
Personality and Social Psychology 58:1015-1026.
Clayton, S. and G. Myers. 2009. Conservation psychology: Understanding and promoting
human care for nature. Wiley-Blackwell, Chichester, West Sussex, UK.
Conforti, V.A. and F.C.C. de Azevedo. 2003. Local perceptions of jaguars (Panthera onca)
and pumas (Puma concolor) in the Iguacu National Park, south Brazil. Biological
Conservation 111:215-221.
Conover, M.R. 2001. Resolving human-wildlife conflict: the science of wildlife damage
management. Lewis Publishers, New York.
Crawshaw, P.G. Jr. 2002. Mortalidad inducida por humanos y conservation de jaguares: el
Pantanal y el Parque Nacional Iguazy en Brazil. Pages 451-464 in R.A. Medellin, C.
Equihua, C.L.B. Chetkiewicz, P.G. Crawshaw, A. Rabinowitz, K.H. Redford, J.G.
163
Literature cited
Robinson, E.W. Sanderson and A. Taber, editors. El Jaguar en el Nuevo Milenio.

Ediciones Cientificas Universitarias, Mexico.
Crawshaw, P.G. Jr. and H.B. Quigley. 1991. Jaguar spacing, activity, and habitat use in a
seasonally flooded environment in Brazil. Journal of Zoology 223:357-370
Cronbach, L.J. 2004. My current thoughts on coefficient alpha and successor procedures.
Educational and Psychological Measurement 64(3):391-418.
Cunha, E. 1957. Rebellion in the Backlands. Translated from Portuguese Os Sertões.
University Of Chicago Press, Chicago.
Cunha, P. 1919. Viagens e Caçadas em Matto Grosso, Três semanas em companhia de Th.
Roosevelt. Francisco Alves Press, Rio de Janeiro, Brazil.
Czech, B., P.K. Devers and P.R. Krausman. 2001. The Relationship of Gender to Species
Conservation Attitudes. Wildlife Society Bulletin, 29(1):187-194
Dalponte, J. C. 2002. Dieta del jaguar y depredación de ganado en el norte del Pantanal,
Brasil. Pages 209-221 in R. A. Medellin, C. Equihua, C. L. B. Chetkiewicz, P. G.
Crawshaw, Jr., A. Rabinowitz, K. H. Redford, J. G. Robinson, E. W. Sanderson and
A. B. Taber, editors. El Jaguar en el Nuevo Milenio. Ediciones Cientificas
Universitarias, Mexico.
Decker, D.J. and K.G. Purdy. 1988. Toward a concept of wildlife acceptance capacity in
wildlife management. Wildlife Society Bulletin 16:53-57.
Deurbrouck, J. and D. Miller. 2001. Cat Attacks: True Stories and Hard Lessons from Cougar
Country. Sasquatch Books, Seattle, WA.
DeVellis, R.F. 2003. Scale Development: Theory and Applications. Sage Publications,
Thousand Oaks.
DeYoung, R. 2000. Expanding and evaluating motives for environmentally responsible
behavior. Journal of Social Issues 56(3):509-526.
Dickman, A.J. 2005. An assessment of pastoralist attitudes and wildlife conflict in the
Rungwa-Ruaha region, Tanzania, with particular reference to large carnivores. M.Sc.
dissertation, University of Oxford, UK.
Dickman, A.J., 2008. Investigating the key determinants of human-large carnivore conflict in
Tanzania. PhD dissertation, University College London, UK.
Duvall, J. and M. Zint. 2007. A review of research on the effectiveness of environmental
education in promoting intergenerational learning. Journal of Environmental
Education 38:14-23.
Eagles, P.F.J. and S. Muffit. 1990. An analysis of children's attitudes toward animals. Journal
of Environmental Education 21:41-44.
Fazio, R. H. 1986. How do attitudes guide behavior? Pages 204-243 in R. M. Sorrentino and
E. T. Higgins, editors. Handbook of motivation and cognition: Foundationsof social
behaviour. Guilford, New York.
Fazio, R. H. 1990. Multiple processes by which attitudes guide behavior: The MODE model
as an integrative framework. Pages 75-109 in M. P. Zanna, editor. Advances in
experimental social psychology. Academic, San Diego, CA.
Fazio, R.H., J. Chen, E.C McDonel, and S.J. Sherman. 1982. Attitude accessibility, attitude
behaviour consistency, and strength of attitude evaluation association. Journal of
Experimental Social Psychology 18:339-357.
Fearnside, P. M., 1993. Deforestation in Brazilian Amazonia: The Effect of Population and
Land Tenure. Ambio 22(8):537-545.
Fielding, K.S., D.J. Terry, B. Masser and M.A. Hogg. 2005. Explaining landholders'
decisions about riparian zone management: The role of attitudinal, normative, and
control beliefs. Journal of Environmental Management 77:12-21.
164
Literature cited
Fielding, K.S., D.J. Terry, B.M. Masser and M.A. Hogg. 2009. Integrating social identity
theory and the theory of planned behaviour to explain decision to engage in
sustainable agricultural practices. British Journal of Social Psychology 47:23-48.
Fishbein, M. and I. Ajzen. 1975. Belief, attitude, intention and behaviour: An introduction to
theory and research. Addison-Wesley, Reading, MA.
Flurry, I. and A. Burns. 2005. Children's influence in purchasing decisions: A social power
theory approach. Journal of Business Research 58:593-601.
Gardner, G.T and P.C. Stern. 2002. Environmental problems and human behaviour. Pearson
Custom Publishing, Boston.
Google Inc. 2009. Google Earth (Version 5.1.3533.1731) [Software]. Available from
http://earth.google.com/
Harland, P., H. Staats and H.A.M. Wilke. 1999. Explaining proenvironmental intention and
behaviour by personal norms and the theory of planned behaviour. Journal of Applied
Social Psychology 29:2505-2528.
Hart, D. and R.W. Sussman. 2009. Man the hunted: Primates, predators, and human
evolution. Westview, New York.
Hazzah, L., M.B. Mulder and L. Frank. 2009. Lions and Warriors: social factors underlying
declining African lion populations and the effect of incentive-based management in
Kenya. Biological Conservation 142(11): 2428-2437.
Herda-Rapp, A. and T.L. Goedeke. 2005. Mad About Wildlife. Brill, Boston.
Herrero, S. and A. Higgins. 1995. Fatal injuries inflicted to people by black bears. Pages 75-
81 in J. Auger and H. Black, editors. Proceedings of the 5th Western Black Bear
Workshop. Brigham Young University Press, Provo.
Hoogesteijn, R. 2000. Manual on the problem of depredation caused jaguars and pumas on
cattle ranches. Wildlife Conservation Society, NY. [Portuguese, Spanish and English
versions]
Hoogesteijn, R. and E. Mondolfi. 1992. The Jaguar. Armitano Editores, Venezuela.
Hoogesteijn, R., A. Hoogesteijn and E. Mondolfi. 1993. Jaguar predation and conservation:
cattle mortality caused by felines on three ranches in the Venezuelan llanos. Pages
391-407 in N. Dunstone and M.L. Gorman, editors. Mammals as predators. Oxford
University Press, Oxford, UK.
Hook, R. A. and W. L. Robinson. 1982. Attitudes of Michigan citizens toward predators.
Pages 382-394 in F. H. Harrington and P. C. Paquet, editors. Wolves of the world.
Noyes Publications, Park Ridge, New Jersey.
Hrubes, D., I. Ajzen and J.J. Daigle. 2001. Predicting Hunting Intentions and Behavior: An
Application of the Theory of Planned Behavior. Leisure Sciences 23:165-178.
Hungerford, H.R. and T.L. Volk. 1990. Changing learner behavior through environmental
education. Journal of Environmental Education 21(3):8-22.
Inglehart, R. 1990. Culture shift in advanced industrial societies. Princeton University Press,
Princeton, NJ.
Instituto Socioambiental. 2009. Unidades de Conservacao na Amazonia Legal.
<http://www.socioambiental.org/uc/quadro_geral>. Accessed 27/01/2010.
Jacobson, S.K., M.D. McDuff and M.C. Monroe. 2006. Conservation education and outreach
techniques. Oxford Univesity Press, Oxford, UK.
Kaltenborn, B. P. and T. Bjerke. 2002. The relationship of general life values to attitudes
toward large carnivores. Human Ecology Review 9: 55-61.
Kaltenborn, B.P., T. Bjerke, J.W. Nyahongo and D.R. Williams. 2006. Animal preferences
and acceptability of wildlife management actions around Serengeti National Park,
Tanzania. Biodiversity and Conservation 15:4633-4649.
165
Literature cited
Kaplan, S. 2000. Human nature and environmentally responsible behavior. Journal of Social
Issues 56(3):491-508.
Karlsson, J. 2007. Management of Wolf and Lynx Conflicts with Human Interests. PhD
dissertation. Swedish University of Agricultural Sciences, Uppsala, Sweden.
Kellert, S. R. 1986. The public and the timber wolf in Minnesota. Pages 152-161 in
Transactions of the 51st North American Wildlife and Natural Resources Conference.
Wildlife Management Institute, Washington, D.C.
Kellert, S.R. 1996. The Value of Life, Biological Diversity and Human Society. Island Press.
Washington, DC.
Kellert, S.R. 1976. Perceptions of animals in American society. Transactions of North
American Wildlife and Natural Resources Conference 41:533-546.
Kellert, S.R. and M.O. Westervelt. 1983. Children's attitudes, knowledge, and behaviours
toward animals. Fish and Wildlife Service, Washington, D.C.
Kokelmanns, J. J. 1979. Why interdisciplinarity? Pages 125-160 in J. J. Kokelmanns,
Interdisciplinarity and higher education. Pennsylvania State University Press.
Kruuk, H. 2002. Hunter and Hunted: Relationships between Carnivores and People.
Cambridge University Press, Cambrigdge, UK.
Kuran, T. and C.R. Sunstein. 1999. Availability cascades and risk regulation. Stanford Law
Review 51(4):123-134.
Lauber, T. B., B.A. Knuth and J.D. Deshler. 2002. Educating Citizens About Controversial
Issues: The Case of Suburban Goose Management. Society and Natural Resources
15:581-597.
Leader-Williams, N. and J.M. Hutton. 2005. Does extractive use provide opportunities to
offset conflicts between people and wildlife? In R.Woodroffe, S. Thirgood and A.
Rabinowitz, editors. People and wildlife: conflict or coexistence. Cambridge
University Press, Cambridge, UK.
Leite-Pitman, R.L., T.G. Oliveira, R.C. Paula and C. Indrusiak. 2002. Manual de
identificação, prevenção e controle de predação por carnívoros. Ibama, Brasília,
Brazil.
Lindsey, P.A., J.T. du Toit and M.G.L. Mills. 2005. Attitudes of ranchers towards African
wild dogs Lycaon pictus: conservation implications for wild dogs on private land.
Biological Conservation 125:113-121.
Linnell, J.D.C., J. Loe and H. Okarma. 2002. The fear of wolves: a review of wolf attacks on
humans. Norwegian Institute for Nature Research Oppdragsmelding 731: 1-65.
Lobato, M. 1933. As caçadas de Pedrinho. Editora Nacional, Rio de Janeiro, Brazil.
Loe, J. and E. Roskaft. 2004. Large carnivores and human safety: a review. Ambio 33(6):
283-288.
Loveridge A.J., S.W. Wang, L.G. Frank and J. Seidensticker. 2010. People and wild felids:
conservation of cats and management of conflicts. In D.W. Macdonald and A.J.
Loveridge, editors. Biology and Conservation of Wild Felids. Oxford University
Press, Oxford, UK.
Macdonald, D.W. 1987. Running with the Fox. Unwin Hyman, London.
Macdonald, D.W. 1984. A questionnaire survey of farmers' opinions and actions towards
wildlife on farmlands. Pages 171-177 in D. Jenkins, editor. Agriculture and the
Environment. ITE Publications, Cambridge, UK.
Macdonald, D.W. 2001. Carnivore conservation: science, compromise and tough choices. In
J.L. Gittleman, S.M. Funk and D.W. Macdonald, editors. Carnivore Conservation.
Cambridge Unirversity Press, Cambridge, UK.
166
Literature cited
Macdonald, D.W., A.J. Loveridge and A. Rabinowitz. 2010 Felid futures: crossing
disciplines, borders and generations. In D.W. Macdonald and A.J. Loveridge, editors.
Biology and Conservation of Wild Felids. Oxford University Press, Oxford.
Madden, F.M. 2008. The Growing Conflict Between Humans and Wildlife: Law and Policy
as Contributing and Mitigating Factors, Journal of International Wildlife Law &
Policy 11(2):189 - 206.
Maffei, L., E. Cuéllar and A. Noss. 2004. One thousand jaguars (Panthera onca) in Bolivia's
Chaco? Camera trapping in the Kaa-Iya National Park. Journal of Zoology 262: 295-
304.
Manfredo, M., T. Teel and A. Bright. 2003. Why Are Public Values Toward Wildlife
Changing? Human Dimensions of Wildlife 8(4):287-306.
Manfredo, M.J. 2008. Who Cares About Wildlife? Social Science Concepts for Exploring
Human-Wildlife Relationships and Conservation Issues. Springer, New York.
Manfredo, M.J. and A. A. Dayer 2004. Concepts for exploring the social aspects of human-
wildlife conflict in a global context. Human Dimensions of Wildlife 9(4): 317-330.
Marchini, S. 2003. Pantanal: opinião pública local sobre meio ambiente e desenvolvimento.
Wildlife Conservation Society and Mamiraua Institute, Rio de Janeiro, Brazil.
Marchini, S. and Luciano. R. 2009a. People and jaguars: guidelines for coexistence.
Amazonarium Press, Piracicaba, Brazil. [Portuguese, Spanish and English versions]
Marchini, S. and Luciano. R. 2009b. Onças-pintadas: o que são e como vivem. Amazonarium
Press, Piracicaba, Brazil.
Marchini, S. and Luciano. R. 2009c. Gente e onças: conflito e convivência. Amazonarium
Press, Piracicaba, Brazil.
Marker, L., A.J. Dickman, M.G.L. Mills and D.W. Macdonald. 2010. Cheetahs and ranchers
in Namibia: a case study. In D.W. Macdonald and A.J. Loveridge, editors. Biology
and Conservation of Wild Felids. Oxford University Press, Oxford, UK.
Marker, L., M.G.L. Mills and D.W. Macdonald. 2003. Factors influencing perceptions and
tolerance toward cheetahs (Acinonyx jubatus) on Namibian farmlands. Conservation
Biology 17:1-9.
Marker, L.L., M.G.L. Mills and D.W. Macdonald. 2003. Factors influencing perceptions of
conflict and tolerance towards cheetahs on Namibian farmlands. Conservation
Biology 17:1290-1298.
Marshall, K., R. White and A. Fischer. 2007. Conflicts between humans over wildlife
management: on the diversity of stakeholder attitudes and implications for conflict
management. Biodiversity and Conservation 16:3129-3146.
Mascia, M. B., J. P. Brosius, T. A. Dobson, B. C. Forbes, L. Horowitz, M. A. McKean, and
N. J. Turner. 2003. Conservation and social sciences. Conservation Biology 17:649-
650.
Maslow, A. 1954. Motivation and Personality. Harper, New York.
Mazzolli, M., M.E. Graipel and N. Dunstone. 2002. Mountain lion depredation in southern
Brazil. Biological Conservation 105:43-51.
McCleery, R.A., R.B. Ditton, J. Sell and R.R. Lopez. 2006. Understanding and improving
attitudinal research in wildlife sciences. Wildlife Society Bulletin 34:537-541.
McDougal, C. 1989. Leopard Attacks in Nepal. Cat News 9: 11.
McDougal, C., M. Cotton, A. Barlow, S. Kumal and D.B. Tamang. 2001. Tigers clam more
human victims in Nepal. Cat News 35: 2-3.
McKenzie-Mohr, D. and W. Smith. 1999. Fostering Sustainable Behavior: An Introduction to
Community-Based Social Marketing. New Society Publishers, Gabriola Island,
British Columbia, Canada.
167
Literature cited
McNay, M.E. 2002. Wolf-human interactions in Alaska and Canada: a reviewof the case
history. Wildlife Society Bulletin 30: 831-843.
Medellin, R. 2009. The conservation of jaguars: a top priority for Latin America. Jaguar
News, 35. Jaguar Conservation Fund, Brazil.
Michalski, F. and C.A. Peres. 2005. Anthropogenic determinants of primate and carnivore
local extinctions in a fragmented forest landscape of southern Amazonia. Biological
Conservation 124: 383-396.
Michalski, F., R.L.P. Boulhosa, A. Faria and C.A. Peres. 2006. Human-wildlife conflicts in a
fragmented Amazonian forest landscape: determinants of large felid depredation on
livestock. Animal Conservation 9:179-188.
Miranda, E.E., A. Gambarini. 2003. Natureza, conservação e cultura: ensaio sobre a relação
do homem com a natureza no Brasil. Metalivros, São Paulo, Brazil.
Monroe, M.C. 2003. Two avenues for encouraging conservation behaviors. Human Ecology
Review 10 (2), 113-125.
Murray, E. 2002. Challenges in educational research. Medical Education 36:110-112.
Naughton-Treves, L., R. Grossberg, and A. Treves. 2003. Paying for tolerance: rural citizens‟
attitudes toward wolf depredation and compensation. Conservation Biology 17:1500-
1511.
Nepstad, D.C ., C.M. Stinkler and O.T. Almeida. 2006. Globalization of the Amazon soy and
beef industries: opportunities for conservation. Conservation Biology 20(6) 1595-
1603
Nowell, K. and P. Jackson. 1996. Wild cats, status survey and conservatio action plan. UICN,
Gland, Switzerland.
Nyhus, P., H. Fisher, F. Madden and S. Osofsky. 2003. Taking the bite out of wildlife
damage: the challenges of wildlife compensation schemes. Conservation Biology
4:37-40.
Nyhus, P., S. Osofsky, P. Ferraro, F. Madden and H. Fisher. 2005. Bearing the costs of
human-wildlife conflict: the challenges of compensation schemes. Pages 107-121 in
R. Woodroffe, S. Thirgood, and A. Rabinowitz, editors. People and wildlife: conflict
or coexistence? Cambridge University Press, Cambridge, UK.
Oppenheim, A.N. 1998. Questionnaire Design, Interviewing and Attitude Measurement.
Pinter, London.
Palmeira F. and W. Barrella. 2007. Conflitos causados pela predação de rebanhos domésticos
por grandes felinos em comunidades quilombolas na Mata Atlântica. Biota
Neotropica 7: 119-128.
Palmeira, F.B., P.G. Crawshaw, C.M. Haddad, K.M.P.M. Ferraz, L.M. Verdade. 2008. Cattle
depredation by puma (Puma concolor) and jaguar (Panthera onca)in central-western
Brazil. Biological Conservation 141:118-125
Palmeira, F.B.L. 2004. Predação de bovinos por onças no norte do estado de Goiás. MSc
Thesis, University of Sao Paulo, Brazil.
Patterson, J.H. 1907. The Maneaters of Tsavo. Fontana, London.
Paula, R.C.; S.M. Cavalcanti, A. Desbiez and L. Jerusalinsky. (In press). Action Plan for the
Conservation of the Jaguar in Brazil. National Centre for the Conservation of Wild
Predators - CENAP/ICMBio, Atibaia, Brazil.
Petty, R. E., and J.T. Cacioppo. 1996. Attitudes and persuasion: classic and contemporary
approaches. Westview Press, Oxford.
Polisar, J., I. Maxit, D. Scognamillo, L. Farrell, M. Sunquist and J.F. Eisenberg. 2003.
Jaguars, pumas, their prey base, cattle ranching: ecological interpretations of a
management problem Biological Conservation 109:297-310
168
Literature cited
Prochaska, J. O., C.C. DiClemente and J.C. Norcross. 1992. In search of how people change.
American Psychologist 47:1102-1114.
Prodes. 2007. Monitoring of the Brazilian Amazon by Satellite. Brazil‟s National Institute for
Space Research. Ministry of Science and Technology, Brasília, Brazil.
Purdy, K. G. and D. J. Decker. 1989. Applying wildlife values information in management:
The wildlife attitudes and values scale. Wildlife Society Bulletin 17:494-500.
Quigley, H. and S. Herrero 2005. Characterization and prevention of attacks on humans. In
R.Woodroffe, S. Thirgood and A. Rabinowitz, editors. People and wildlife: conflict or
coexistence. Cambridge University Press, Cambridge. UK.
Quigley, H. and S. Herrero. 2005. Behavioral Characterization and Prevention of Carnivore
Attacks on Humans. Pages 27-48 in R. Woodroffe, S. Thirgood and A. Rabinowitz,
editors. People and Wildlife: Conflict and Coexistence. Cambridge University Press,
Cambridge, UK.
Quigley, H. B., and P. G. Crawshaw Jr. 1992. A conservation plan for the jaguar (Panthera
onca) in the Pantanal region of Brazil. Biological Conservation 61:149-157.
Rabinowitz, A. 1986. Jaguar predation on domestic livestock in Belize. Wildlife Society
Bulletin 14: 170-174.
Rabinowitz, A. 2005. Jaguars and livestock: living with the world's third largest cat. Pages
278-285 in R. Woodroffe, S. Thirgood and A. Rabinowitz, editors. People and
Wildlife: Conflict or Coexistence? Cambridge University Press, Cambridge, UK.
Rabinowitz, A. and K. Zeller. 2010. A range-wide model of landscape connectivity and
conservation for the jaguar, Panthera onca. Biological Conservation 143:939-945.
Roedder-John, D. 1999. Consumer socialization of children: a retrospective look at twenty-
five years of research. Journal of Consumer Research 26:183-213.
Rogers, E. M. 1995. Diffusion of innovations, 4th ed. Free Press, New York.
Rohan, M.J. 2000. A rose by any other name? The value construct. Personality and Social
Psychology Review 4(3):255-277.
Roosevelt, T. 1914. Through the Brazilian wilderness. Scribners, New York, New York.
Rosa, G. 1968. Meu tio, o Iauaretê, in Rosa. In G. Rosas, editor. Estas Estórias. José Olympio
Press, Rio de Janeiro, Brazil.
Røskaft, E., T. Bjerke, B.P. Kaltenborn, J.D.C. Linnell and R Andersen. 2003. Patterns of
self-reported fear towards large carnivores among the Norwegian public. Evolution
and Human Behaviour 24:184-198.
Røskaft, E., B. Händel, T. Bjerke and B.P. Kaltenborn. 2007. Human attitudes towards large
carnivores in Norway. Wildlife Biology 13(2):172-185.
Saberwal, V. K., J. P. Gibbs, R. Chellam, and A. J. T. Johnsingh. 1994. Lion-human conflict
in the Gir forest, India. Conservation Biology 8:501-507.
Saenz, J.C. and E. Carrillo. 2002. Jaguares depredadores de ganado en Costa Rica. In R.A.
Medellin, C. Equihua, C.L.B. Chetkiewicz, P.G. Crawshaw, A. Rabinowitz, K.H.
Redford, j.G. Robinson, E.W. Sanderson and A. Taber, editors. El Jaguar en el Nuevo
Milenio. Ediciones Cientificas Universitarias. Mexico.
Sanderson, E.W., K.H. Redford, C.L.B. Chetkiewicz, R.A. Medellin, A. Rabinowitz, J.G.
Robinson and A. Taber. 2002. Planning to save a species: the jaguar as a model.
Conservation Biology 16(1):58-72
Santos, F.R., A.T.A. Jacomo and L. Silveira. 2008. Humans and jaguars in five Brazilian
biomes: same country, different perceptions. Cat News, Special Issue 4:21-25.
Saunders, N.J. 1994. Predators of Culture: Jaguar Symbolism and Mesoamerican
Elites.World Archaeology 26(1):104-117.
169
Literature cited
Scarce, R. 2005. More than mere wolves at the door: reconstructing community amidst
wildlife controversy. In A. Herda-Rapp and T.L. Goedeke, editors. Mad About
Wildlife. Brill, Boston.
Schiaffino, K., L. Malmierca and P.G. Perovic. 2002. Depredacion de cerdos domesticos por
jaguar en un area rural vecina a un parque nacional en el noreste de Argentina. In
R.A. Medellin, C. Equihua, C.L.B. Chetkiewicz, P.G. Crawshaw, A. Rabinowitz,
K.H. Redford, j.G. Robinson, E.W. Sanderson and A. Taber, editors. El Jaguar en el
Nuevo Milenio. Ediciones Cientificas Universitarias, Mexico.
Schultz, P.W. 2002. Knowledge, information, and household recycling: Examining the
knowledge-deficit model of behavior change. Pages 67-82 in T. Dietz and P.C. Stern,
editors. New Tools for Environmental Protection: Education, Information, and
Voluntary Measures. National Academy Press, Washington, DC.
Scognamillo, D., I. Maxit, M. Sunquist and L. Farrell. 2002. Ecologia del Jaguar y el
problema de la depredacion de ganado en un hato de Los Llanos venezolanos. Pages
139-151 in R.A. Medellin, C. Equihua, C.L.B. Chetkiewicz, P.G. Crawshaw, A.
Rabinowitz, K.H. Redford, J.G. Robinson, E.W. Sanderson and A. Taber, editors. El
Jaguar en el Nuevo Milenio. Ediciones Cientificas Universitarias, Mexico.
Seidl, A. F., J. S. V. da Silva and A. S. Moraes. 2001. Cattle ranching and deforestation in the
Brazilian Pantanal. Ecological Economics 36: 413-425.
Shepard, P. 1973. The Tender Carnivore and the Sacred Game. Scribners, New York.
Siemel, S., 1953. Tigrero! Ace Books. A.A. Wyn Inc., New York.
Sillero-Zubiri, C., R. Sukumar and T. Treves. 2007. Living with conflict: the roots of conflict
and the solutions. Pages 253-270 in D.W. Macdonald and K. Service. Key Topics in
Conservation. Oxford University Press, Oxford, UK.
Silveira, L., A. T. Jácomo, C. K. Kashivakura, M. M. Furtado, E. Freitas, N. M. Tôrres,C.
Ferro, and M. Marcondes. 2006. Conservação da onça-pintada em propriedades rurais
no Pantanal. Summary report on findings to the National Research Center for the
Conservation of Natural Predators (Cenap), Atibaia, Brazil.
Silver, S.C., L.E.T. Ostro, L.K. Marsh, L. Maffei, A.J. Noss, M.J., Kelly, R.B. Walllace, H.
Gomez and G. Ayala. 2004. The use of camera traps for estimating jaguar Panthera
onca abundance and density using capture/recapture analysis. Oryx 38(2): 148-154.
Skogen, K., 2001. Who's Afraid of the Big, Bad Wolf? Young People's Responses to the
Conflicts Over Large Carnivores in Eastern Norway. Rural Sociology 66(2):203-226.
Soisalo, M.K. and S.M.C. Cavalcanti. 2006. Estimating the density of a jaguar population in
the Brazilian Pantanal using camera-traps and capture-recapture sampling in
combination with GPS radio-telemetry. Biological Conservation 129: 487-496.
Soto, J.R. 2008. Patterns and determinants of human-carnivore conflicts in the tropical
lowlands of Guatemala. MSc Thesis, University of Florida.
Stern, P.C. 2000b. Toward a coherent theory of environmentally significant behavior. Journal
of Social Issues 56(3):407-424.
Sudman, S., N.M. Bradburn and N. Schwartz. 1996. Thinking about questions: the
application of cognitive processes to survey methodology. Jossey-Bass Publishers,
San Francisco.
Sutton, S. 1998. Predicting and explaining intentions and behavior: How well are we doing?
Journal ofApplied Social Psychology 28(3):17- 1338.
Swarts, F. 2000. The Pantanal in the 21st century: for the planet‟s largest wetland, an
uncertain future. Waterland Research Institute, Goulsboro, USA.
Swenson, J. E. and H. Andrén. 2005. A tale of two countries: large carnivore depredation and
compensation schemes in Sweden and Norway. Pages 323-339 in R. Woodroffe, S.
170
Literature cited
Thirgood, and A. Rabinowitz, editors. People and wildlife: conflict or coexistence?

Cambridge University Press, Cambridge, UK.
Tajfel, H. 1981. Human groups and social categories: Studies in social psychology.
Tanur, J.M. 1992. Questions about questions: inquiries into the cognitive bases of surveys.
Russell Sage Foundation, New York.
Taylor, S. and P. Todd. 1995 An integrated model of waste management behaviour: A test of
household recycling and composting intentions. Environment and Behavior 27(5):
603-630.
Thornton, P.K., R.L.Kruska, N. Henninger, P.M. Kristjanson, R.S. Reid, F. Atieno, A.N.
Odero and T. Ndegwa. 2002. Mapping poverty and livestockin the developing world.
International Livestock Research Institute, Kenya.
Torres, N.M., P. De Marco Jr., J.A.F. Diniz Filho, L. Silveira. 2008. Jaguar distribution in
Brazil: past, present and future. Cat News Special Issue 4:4-8.
Treves, A. and U. Karanth. 2003. Human-carnivore conflict and perspectives on carnivore
management worldwide. Conservation Biology 17:1491-1499.
Tunnicliffe, S.D. and A. Scheersoi. 2009. Engaging the interest of zoo visitors as a key to
biological education. International Journal of Zoo Education 4:18-20
Tversky, A. and D. Kahneman. 1974. Judgement under uncertainty: Heuristics and biases.
Science 185:1124-1130.
United Nations Educational, Scientific and Cultural Organization. 2007. Education for all by
2015: will we make it? Paris, France.
United Nations Educational, Scientific and Cultural Organization. 2008. The state of
education in Latin America and the Caribbean: guaranteeing quality education for all.
Santiago, Chile.
Vaske, J.J., 2008. Survey Research and Analysis: Applications in Parks, Recreation and
Human Dimensions. Venture Publishing, Pennsylvania.
Wallace, R.B., H. Gomez, G. Ayala and F. Espinoza. 2003. Camera trapping capture
frequencies for jaguar (Panthera onca) in the Tuichi Valley, Bolivia. Mastozoologia
Neotropical 10(1): 133-139.
White, K.M., J.R. Smith., D.J. Terry, J.H. Greenslade and B.M. McKimmie. 2009. Social
influence in the theory of planned behaviour: the role of descriptive, injunctive and in-
group norms. British Journal of Social Psychology 48:135-158.
Wilson, E.O. 1984. Biophilia: the human bond with other species. Harvard University Press,
Cambridge.
Wilson, E.O. 1993. Biophilia and the conservation ethic. Pages 31-41 in S.R. Kellert and
E.O. Wilson, editors. The biophilia hypothesis. Island Press, Washington, DC.
Winter, P.L., B.J. Sagarin, K. Rhoads, D.W. Barrett and R.B. Cialdini. 2000. Choosing to
encourage or discourage: perceived effectiveness of prescriptive versus proscriptive
messages. Environmental Management 26(6):589-594.
Woodroffe, R. and J.R. Ginsberg. 1998. Edge effects and the extinction of populations inside
protected areas. Science 280: 126-2128.
WWF-UK. 2000. WWF-UK report: Europe's Carnivores: a Survey of Children's Attitudes to
Wolves, Bears and Otters. UK.
Yamazaki, K. And T. Bwalya. 1999. Fatal lion attacks on local people in the Luangwa valley,
eastern Zambia. South African Journal of Wildlife Research 29:19-21.
Zeller, K. 2007. Jaguars in the New Millennium Data Set Update: The State of the Jaguar in
2006. Report. Wildlife Conservation Society, New York.
Zimmermann A., M.J. Walpole and N. Leader-Williamns. 2005. Cattle ranchers' attitudes to
conflicts with jaguar Panthera onca in the Pantanal of Brazil. Oryx 39:406-412.
171
Appendix I – Jaguars in books for children and adolescents
APPENDIX I
THE JAGUAR IN BOOKS FOR CHILDREN AND ADOLESCENTS
IN BRAZIL
The number of books for children or adolescents with selected animal names in the title and
available from one of Brazil‟s largest book sellers (Livraria Cultura). These were identified
using a search on the bookstore‟s website (www.livrariacultura.com.br) on 2nd April 2010.
Search criteria were: (i) children‟s book + Portuguese language + animal synonyms; and (ii)
books for adolescents + Portuguese language + animal name. Taxa that are not
representatived in the Brazilian fauna are shown in italics.
Domestic animal Wildlife in general Brazilian Wildlife: top 10

Animal Children‟s Books for Animal Children‟s Books for Animal Children‟s Books for
books teenagers books teenagers books teenagers
Cat 176 37 Bear 117 4 Monkey 85 5
Dog 130 8 Lion 81 16 Bird 59 11
Duck 123 4 Wolf 75 16 Fish 64 3
Rabbit 112 1 Monkey 85 5 Turtle 41 1
Chicken 73 5 Bird 59 11 Butterfly 32 8
Cow 53 5 Fish 64 3 *Jaguar 33 5
Pig 56 0 Elephant 58 8 Ant 30 1
Horse 39 6 Turtle 41 1 Caiman 27 2
Donkey 22 1 Butterfly 32 8 Bee 26 0
Goat 14 0 Jaguar 33 5 Snake 19 2
Sheep 11 1 Fox 35 1 Dolphin 19 1
Ant 30 1
Caiman 27 2
Bee 26 0
Giraffe 19 3 * The word jaguar refers to a
Tiger 17 5 single species, while the other
Snake 19 2 names on this list refer to taxa
Dolphin 19 1 with multiple species
Zebra 18 1
Cicada 18 1
Armadillo 18 0
Whale 14 4
Hippo 12 1
Rhinoceros 9 0
Cockroach 7 1
Deer 7 0
Beetle 6 0
Maned- 5 1
wolf
Anteater 4 0
Tapir 2 0
Giant river 2 0
otter
Leopard 2 0
Coati 2 0
Sloth 1 0
Puma 0 0
172
Appendix II – Questionnaires and interview schedule
APPENDIX II
Questionnaire used to evaluate children and adolescents (Chapter 2)
1. Name: __________________________________________ 2. Date: ____/_____/_____

3. Sex: male ( ) female ( ) 4. Age: _______ years old.
5. School: __________________________________________ 6. Grade: __________
7. Place of residence: rural ( ) urban ( )
Cognitive and affective prominence

8. Cite three animals that you know to occur in the nearest forest:
1.______________________________________________
2.______________________________________________
3.______________________________________________
9. What animal do you like most?_____________________________________________________
10. What animal do you dislike most? __________________________________________________
Experience of jaguars
11. Has your family (i.e. household) ever lost livestock to jaguars? No ( ) Yes ( )
When did that happen?_____________________________________________________________
12. Do you personally know anyone who has been injured by a jaguar? No ( ) Yes ( )
Who was it and how did that happen?____________________________________________
__________________________________________________________________________
13. Have you ever seen a live jaguar in the wild? No ( ) Yes ( )
Where? ________________________________________________________________________
Knowledge about jaguars and depredation

Incorrect
Correct
Do not
know
14. The jaguar generally begins to consume its prey from the front while the
puma consumes the areas from the ribs back.
15. A jaguar‟s prey is usually hidden and covered with leaves, while a
puma‟s prey isn‟t.
16. The jaguar‟s prey generally presents a bite mark at the base of the neck,
while the puma‟s prey generally has a bite on the throat.
17. The jaguar‟s track is longer than wide with thinner and pointed toes,
while the puma‟s track is slightly wider than long, with round toes.
18. A female jaguar produces in average 1 or 2 cubs every other year.
19. Black jaguars are more dangerous to cattle than yellow jaguars.
20. Jaguars kill more people every year in Brazil than do domestic dogs.
173
General perception about jaguars

21. Complete the sentence:
“Of all the animals, the jaguar is the most_______________________________________________”.
Concern for safety/property and attraction towards jaguars

22. If I had to walk in the forest, I would be afraid of encountering a jaguar
strongly disagree( ) disagree( ) neither disagree nor agree( ) agree( ) strongly agree( )
23. I would feel more comfortable if there were no jaguars in the region.
24. I would have concerns about sharing the land with jaguars because they attack people.
25. I would have concerns about sharing the land with jaguars because they kill livestock.
26. I am afraid of jaguars.
27. It would be fun if jaguars were seen more often.
28. Jaguars are interesting animals.
29. I am not interested in knowing anything more about jaguars.
Source of information about jaguars

30. What I know about jaguars I have learnt mostly from:
( ) Magazines
( ) Zoos
( ) Internet
( ) Books
( ) Teachers/school
( ) Nature/travel
( ) Friend/neighbour
( ) Television
( ) Parents
( ) Other
174
Interview schedule (Chapters 3 and 4)

1. Questionnaire# 2. Date 3. Interviewer [4. Name of property]
5. Location
GPS: S___________ W___________ District:
Interviewee and property.

6. Name 7. Owner/Employee:
O( ) E( )
8. Sex 9. Age 10. Place of birth 11. Length of time lived in:
the property:_____ years
M( )F( )
the region:______years
12. Formal education
None ( ) Incomplete Primary Incomplete Secondary University
primary ( ) ( ) secondary ( ) ( ) or above ( )
13. Property size: 14. Number of people/kinship living in the property

__________ alq ( ) ha ( )
15. Economic activity Main Other Details
Cattle ranching: calving ( ) ( )
Cattle ranching: rearing and fattening ( ) ( )
Cattle ranching: dairy ( ) ( )
Crop growing ( ) ( )
Ecotourism ( ) ( )
Other ( ) ( )
16. Herd size 17. Smallstock 18. Pets
Cattle: Horse/Donkey: Pig: Poultry: Other: Dog: Other:
Perception of the regional economic situation.

19. The economic situation of (respondent‟s main economic activity) in the region is
currently:
very bad( ) bad( ) neither bad nor good( ) good( ) very good( ) no opinion( )
20. The regional economic situation is currently:
declining fast( ) declining( ) stable( ) improving( ) improving fast( ) no opinion( )
Perception of the impact of jaguars on human livelihood.

No opinion
Very high
Very low
Medium
None
High
Low
21. Damage associated with depredation ever caused by

jaguars to you.
jaguars to your neighbours.
jaguars in the district.
24. Risk of any damage associated with depredation to
you in the next 12 months.
175

your neighbours in the next 12 months.
anyone in the district in the next 12 months.
27. Number of people ever hurt by a jaguar in the
neighbourhood.
28. Number of family members ever hurt by a jaguar.
29. Risk of you being hurt by a jaguar in the next 12
months.
30. Risk of a family member being hurt by a jaguar in the
next 12 months.
Experience of jaguars.
Comments
31. What percentage of your herd was lost to jaguars in the last
12 months? (cattle ranchers only);
32. How many people do you personally know that have been
injured by a jaguar?
33. How many times have you seen a live jaguar in the wild?
Knowledge about jaguars and depredation.
Incorrect
Correct
Do not
know
35. A jaguar‟s prey is usually hidden and covered with leaves, while a puma‟s
prey isn‟t.
37. The jaguar‟s track is longer than wide with thinner and pointed toes, while
the puma‟s track is slightly wider than long, with round toes.
Perception of change in jaguar abundance.

41. Jaguar abundance on your property is currently:
decreasing a lot( ) decreasing( ) stable( ) increasing( ) increasing a lot( ) no opinion( )
42. Jaguar abundance in the region is currently:
decreasing a lot( ) decreasing( ) stable( ) increasing( ) increasing a lot( ) no opinion( )
Attitude towards jaguars.

43. You would like the jaguar population in the region to:
decrease a lot ( ) decrease ( ) remain unchanged ( ) increase ( ) increase a lot ( )
44. If all the jaguars disappeared forever from the region, you would feel:
very sad ( ) slightly sad ( ) indifferent ( ) slightly happy ( ) very happy ( )
176
45. What you feel towards jaguars is better described as:

dislike a lot ( ) dislike ( ) indifferent ( ) like ( ) like a lot ( )
46. The jaguar has its value, even if it does not generate any income to you:
47. If you had to walk on your own in a forest where there are jaguars, you would feel
scared:
Attitudes towards jaguar killing

In your opinion, jaguar killing is something:
48. very bad( ) bad( ) neither bad nor good( ) good( ) very good( )
49. very beneficial( ) beneficial( ) neither beneficial nor harmful( )
harmful( ) very harmful( )
50. very useless( ) useless( ) neither useless nor useful( ) useful( ) very useful( )
51. very exciting( ) exciting( ) neither exciting nor boring( ) boring( ) very boring( )
52. very unenjoyable( ) unenjoyable( ) neither unenjoyable nor enjoyable( )
enjoyable( ) very enjoyable( )
Subjective social norms re. jaguar killing.
53. How many of the people important to you would disapprove of you killing jaguars?
none( ) few( ) about half( ) many( ) all( )
54. Most people important to me think that killing jaguars is admirable.
55. Among your neighbours, how much agreement would there be that it is a good thing to
kill jaguars?
none( ) little( ) medium( ) much( ) total( )
56. In your family, how much agreement would there be that it is a good thing to kill
jaguars?
none( ) little( ) medium( ) much( ) total( )
Descriptive social norms re. jaguar killing
57. How many of your neighbours do you think kill jaguars?
58. Think of the landowners in Alta Floresta/Cáceres/Poconé, what proportion of them do
you think kill jaguars?
none( ) less than half( ) about half( ) more than half( ) all( )
Social group identification.
59. How much do you identify with the landowners in Alta Floresta/Cáceres/Poconé?‟
not at all( ) little( ) medium( ) much( ) completely( )
60. How much do you feel ties with the other landowners in Alta Floresta/Cáceres/Poconé?
not at all( ) little( ) medium( ) much( ) very much( )
61. In general, how well do you feel you fit into the group of landowners in Alta
Floresta/Cáceres/Poconé?
very well( ) fairly well( ) medium( ) little( ) not at all( )
Perceived behavioural control re. jaguar killing.
62. Whether I kill the next jaguar that appears on my property is up to me.
completely false( ) mostly false( ) neither false nor true( )
mostly true( ) completely true( )
177
63. For me, to kill the next jaguar that appears on my property would be:
very easy( ) easy( ) neither easy nor difficult( ) difficult( ) very difficult( )
64. The number of factors outside my control which could prevent me from killing the next
jaguar that appears on my property are:
none( ) few( ) medium( ) many( ) very many( )
65. What factors prevent you from killing the next jaguar that appears in you ranch?
Behavioural intention re. jaguar killing.

66. I intend to kill the next jaguar that appears on my property
very unlikely( ) unlikely( ) maybe( ) likely( ) very likely( )
67. If a jaguar appears on my property, I will try to kill it,
very likely( ) likely( ) maybe( ) unlikely( ) very unlikely( )
Behaviour: jaguar killing.
68. Have you ever killed a jaguar? No( ) Yes( ).
If so, when have you last done this?
69. Has any of your neighbours killed jaguars in the last 5 years? No( ) Yes( ).
If so, which of them?
178
Questionnaire used to evaluate pupils and their fathers (Chapter 5)
1. Name: ________________________________________________ 2. Date: ____/_____/_____

3. Sex: male ( ) female ( ) 4. Age: _______ years old.
5. School: ________________________________________________ 6. Grade: __________
Knowledge about jaguars and depredation
Incorrect
Correct
Do not
know
8. A jaguar‟s prey is usually hidden and covered with leaves, while a puma‟s
prey isn‟t.
10. The jaguar‟s track is longer than wide with thinner and pointed toes,
while the puma‟s track is slightly wider than long, with round toes.
14. The heaviest jaguar ever captured weighed approximately 150 kilos
15. Where cattle are more abundant than native prey, jaguars take more cattle
than native prey
16. Calves kept closer to the forest edge have in general a smaller chance of
being killed by jaguars
Attitudes towards jaguars

17. You would like the jaguar population in the region to:
decrease a lot ( ) decrease ( ) stay the same ( ) increase ( ) increase a lot ( )
18. If all the jaguars disappeared forever from the region, you would feel:
very sad ( ) sad ( ) indifferent ( ) happy ( ) very happy ( )
19. What you feel towards jaguars is better described as:
dislike a lot ( ) dislike ( ) indifferent ( ) like ( ) like a lot ( )
20. The jaguar has its value, even if it does not generate any income to you:
21. If you had to walk alone in a forest where there are jaguars, you would feel scared:
179
Perception of jaguar impact on human safety and livestock

22 (pupils only). How would you rate the damage associated with predation ever caused by jaguars to
your father?
none ( ) very small ( ) small ( ) medium ( ) high ( ) very high ( )
23 (pupils only). How would you rate the risk of any damage associated with predation to your father
in the next 12 months. (pupils only)
22 (fathers only). How would you rate the damage associated with predation ever caused by jaguars to
you? (fathers only)
23 (fathers only). How would you rate the risk of any damage associated with predation to you in the
next 12 months (fathers only).
24. Number of people ever hurt by a jaguar in the neighbourhood
none ( ) one ( ) two ( ) three ( ) four ( ) five or more ( )
25. How would you rate the risk of you being hurt by a jaguar in the next 12 months.
none ( ) very low ( ) low ( ) medium ( ) high ( ) very high ( )
Attitudes towards jaguar killing

In your opinion, jaguar killing is something:
26. very useless( ) useless( ) neither useless nor useful( ) useful( ) very useful( )
27. very exciting( ) exciting( ) neither exciting nor boring( ) boring( ) very boring( )
Social norm
28. How many of your neighbours do you think kill jaguars?
29. Think of the landowners in Alta Floresta - what proportion of them do you think kill jaguars?
none( ) less than half( ) about half( ) more than half( ) all( )
180
Appendix III – Photographs of dead livestock
APPENDIX III
PHOTOGRAPHS OF DEAD LIVESTOCK
These photographs1 were used in the photograph interpretation experiment.
1
Most of the photographs belong to Rafael Hoogesteijn, who kindly authorized their use.
181
Appendix III – Photographs of dead livestock
182
Appendix IV – Activity books
APPENDIX IV
A GUIDE FOR COEXISTENCE: ACTIVITY BOOKS
183
184
185
186
187
188
189
190
Appendix V – Elaboration exercise
APPENDIX V
ELABORATION EXERCISE
Schedule
1) Chairs were rearranged in the classroom to form a circle so that pupils could face each
other; pupils were given paper and pencil.
3) Each pupil wrote down the name of the animal he/she disliked most and, one by one,
showed and spoke to the group his/her choice and one reason for choosing it.
4) The group was divided into two sub-groups: the ones who disliked jaguars and the others
(most of whom disliked snakes, so they all represented snakes thereafter).
5) Each sub-group listed on a flipchart all the possible arguments to support their choice; as a
way of motivation, SM announced that the sub-group that raised the largest number of
arguments would win that stage of the game.
6) Sub-groups presented their results to each other; all the members of the sub-group stood
beside the flipchart and one or two representatives of the sub-group presented the results
orally; SM encouraged the debate between sub-groups by disputing some of the arguments;
group debated.
8) Sub-groups switched sides: the sub-group that had raised arguments against snakes had to
add arguments to the list of arguments against jaguars and vice-versa; as a way of motivation,
SM announced that the sub-group that raised the largest number of additional arguments
would win that stage of the game.
10) Sub-groups presented the results to each-other; SM encouraged the debate between sub-
groups by disputing some of the arguments: group debated.
11) Pupils returned to their chairs; SM presented to the group the flipchart with all the
arguments against jaguars and orally synthesized the results.
12) SM invited the pupils for an exercise of imagination: “Imagine that you have the power to
make all the jaguars disappear forever by simply pressing a button. Would you press that
button?”.
13) Pupils answered individually by writing yes or no on a piece of paper; one by one, they
showed and spoke to the group their decision and explained why.
14) The group was divided into two sub-groups: the ones who would press the button and the
ones who would not.
15) Each sub-group listed on a flipchart all the possible benefits they would derive from
maintaining or exterminating the jaguars; as a way of motivation, SM announced that the
sub-group that raised the largest number of arguments would win that stage of the game.
16) Sub-groups presented their results to each other and debated.
17) Pupils were given the chance of switching sides.
191
Appendix VI – A guide for coexistence
APPENDIX VI
PEOPLE AND JAGUARS: A GUIDE FOR COEXISTENCE
192
193
194
195
196
197
198
199
200
201
202
203
Appendix VII – Jaguars, livestock and people
APPENDIX VII
JAGUARS, LIVESTOCK AND PEOPLE IN BRAZIL: REALITIES

AND PERCEPTIONS BEHIND THE CONFLICT.
Sandra Cavalcanti*, Silvio Marchini*,

Alexandra Zimmermann, Eric M. Gese and David W. Macdonald
In: Biology and Conservation of Wild Felids. Editors: D.W. Macdonald and A. Loveridge.
Oxford Press, UK. 2010.
* Sandra Cavalcanti and Silvio Marchini are joint first authors
204
Introduction
The jaguar (Panthera onca) is the largest predator in the Neotropics and arguably the most
charismatic species for conservation in Central and South America. Regrettably, the jaguar is
also the carnivore that is least compatible with humans in 21st century Brazil. The
fundamental incompatibility is the jaguar‟s need for abundant, large prey, as well as
expansive, undisturbed habitat. Humans - also large, top predators - have competed directly
with jaguars for food for as long as they have co-existed (Jorgenson and Redford 1993), and
lately threaten them directly and indirectly through deforestation and habitat fragmentation.
Moreover, jaguar predation on livestock – particularly cattle – provokes retaliatory
persecution by humans.
Persecution looms as the coup de grace to jaguar populations outside of protected areas
(Nowell and Jackson 1996) and, due to their wide-ranging movements, threatens jaguars
within protected areas as well (Ginsberg and Woodroffe 1998). Although disentangling the
contributions of persecution and habitat loss may be difficult, jaguar distribution (Figure 1)
and abundance have drastically declined in recent decades. Our case study focuses on Brazil,
where the jaguar is a threatened species (Machado et al. 2005), although internationally it is
classed as near-threatened (i.e., it may be threatened with extinction in the near future) (Cat
Specialist Group 2002).
Efforts to protect jaguars by curbing persecution by humans have been based on what might
be termed a “bio-rational” understanding of the problem. Insofar as the root of the problem is
livestock-raiding then, so this rational goes, if we can find ways to effectively reduce jaguar
predation on cattle (e.g., use of electric fences, aversive conditioning, translocation), then
persecution by cattle ranchers should subsequently decline (Hoogesteijn 2000, Cavalcanti
2003). Mitigative actions combined with monetary compensation to ameliorate the financial
costs of lost livestock, are aimed at alleviating the economic burden on ranchers imposed by
having jaguars, on the assumption this will reduce the motivation for ranchers to kill them.
Although this bio-rational thinking may be plausible to a scientifically trained

conservationist, we hypothesized that human persecution of jaguars may be less related to
livestock depredation than previously believed, and the economic justification for killing
jaguars may be equally unclear. We argue that the ultimate motivator of retaliatory
persecution may not be the actual impact of jaguars on human safety or livestock, but rather
the cultural and social perceptions of the potential threat of jaguars attacking humans and
killing livestock. In conflicts between people and carnivores, the perceived impacts often
exceed the actual evidence (Conover 2001, Chavez et al. 2005, Chavez and Gese 2005, 2006,
Sillero-Zubiri et al. 2007). In addition, factors not directly related to the jaguars impact on
human livelihoods (e.g., perceived social status in the community, thrill of the chase) may
also be involved in the persecution of jaguars, making evaluations for the economic rationale
for killing jaguars more unclear. Such imprecise linkages between reality and perception
could prove perilous to a threatened species, adding a potentially lethal element to already
significant risks posed by retributive killing, and rendering irrelevant many biologically-
based conservation actions and mitigation measures.
To explore this „perception blight‟, this chapter addresses the realities and perceptions behind
the conflicts involving jaguars, livestock, and cattle ranchers in Brazil. We use the Pantanal
region to quantify the importance of cattle in jaguar ecology, and techniques adapted from the
205
social sciences to examine the ranchers‟ perceptions about jaguar depredation on cattle and
other perceptions about jaguars and jaguar hunting that may be relevant in dealing with
conflicts between ranchers and jaguars. We investigate then how these social and cultural
perceptions translate into the persecution of jaguars. Finally, we discuss how information on
the ecological, economic, social and cultural dimensions of a human-carnivore conflict can
fruitfully be integrated into a strategy spanning from individuals to populations (of both
jaguars and humans) in an attempt to promote coexistence between jaguars, livestock and
people.
Jaguars, livestock and people
The jaguar occurs from the southwestern United States to northern Argentina, across an area
of 11.6 million km2 and occupys a diverse array of habitats, including xeric shrublands, dry
forests, montane grasslands, moist lowland forest, wet savannahs and mangroves (Zeller
2007). Even though 36% of jaguar distribution overlaps protected areas (Zimmermann and
Wilson, in prep), very few of these areas offer true protection for jaguars and their prey.
Indeed the edges of protected areas often become hotspots for human-wildlife conflict
(Woodroffe and Ginsberg 1998). The human geography outside these protected areas, too, is
varied so jaguars may co-exist with people holding a range of different perceptions and levels
of tolerance for wildlife. Outside of protected areas, the most common land-use form
occupied by jaguars is livestock ranches, followed by logging areas, forest matrix lands,
agricultural areas and other forms of land use (Zeller 2007). On a continental scale, jaguars
occur mostly in areas with a low Human Footprint Index (95% of jaguar range is in areas of
<35HFI) and low cattle densities (96% in areas of  7.5 cattle/km2)(Zimmermann & Wilson
in prep). Nevertheless, hunting of jaguar prey and direct human persecution of jaguars (most
often in retaliation for livestock depredation) are, according to 130 jaguar experts, the most
serious threats to the survival of the jaguar (Zeller 2007).
Human-jaguar conflicts occur in many different socio-economic and cultural contexts and
vary in severity, but appear to be most extensive in regions with large cattle ranches, where
human densities are low, cattle densities are moderate, and small areas of wilderness with
natural prey still persist. There are several such vast rangelands in South America, most
notably the Pantanal, Llanos, Beni and Chaco regions of Argentina, Brazil, Bolivia,
Colombia, Paraguay and Venezuela. The best studied among the above regions of Brazil, the
Pantanal, is the focus of our chapter.
Conflicts between ranchers and jaguars over livestock are widespread and have been
documented throughout jaguar range, (e.g.: Belize: Rabinowitz 1986; Brazil: Crawshaw and
Quigley 1991, Dalponte 2002, Conforti and Azevedo 2003, Michalski et al. 2006, Azevedo
and Murray 2007, Palmeira et al 2008; Costa Rica: Saenz and Carrillo 2002; Argentina:
Schiaffino et al. 2002; Venezuela: Polisar et al. 2003, Scognamillo et al. 2002). Nevertheless,
some ecological, socio-economical, cultural, and historical aspects of the relationships
between people and jaguars in Brazil have made the country particularly important for jaguar
research and conservation.
206
The conflict in Brazil
Brazil covers 40% of the land area of Latin America. Even though jaguar abundance estimates are as
scarce for Brazil (Almeida 1986, Quigley and Crawshaw 1992, Soisalo and Cavalcanti 2006) as for other
parts of the range (cf Wallace et al. 2003, Maffei et al. 2004, Silver et al. 2004), Brazil contains the two
largest population strongholds for jaguars (Sanderson et al. 2002): the wetlands of the Pantanal
(140,000 km2) and the rainforests of Amazonia (3,400,000 km2). The southern Pantanal of Brazil has the
highest density of jaguars recorded (estimates range from 6.7 to 11.7 individuals/100 km2)(Soisalo and
Cavalcanti 2006). The Pantanal is also home to the largest jaguars, with males weight averaging 100 kg
(females are typically 10-20% smaller than males) and the largest males reaching 158 kg (Kindersley
2005). Jaguars were once widely distributed throughout Brazil until 1500, but are now
extirpated from entire regions (Sanderson et al. 2002)(Figure 1). Some jaguars remain in
fragments of the Atlantic forest and the Cerrado, but large jaguar populations are present only
in Amazonia and the Pantanal, where human population density has historically been low.
Brazil is also home to the world‟s largest commercial cattle herd (>200 million head) and is
the world leader in beef exports (Nepstad et al. 2006). For ecological and historical reasons,
there is overlap between areas where beef production flourishes and jaguars survive, namely
the Pantanal and the agricultural frontier of southern Amazonia (Thornton et al. 2002). Cattle
ranchers have never rejoiced in this overlap, and have a long tradition of killing jaguars
(Hoogesteijn and Mondolfi 1992). In essence, jaguars kill livestock and ranchers kill jaguars
in an attempt to prevent further losses.
Cattle ranching also threatens jaguars indirectly, insofar as it is the major driver for the level
of deforestation in Amazonia, being the primary reason for >66% of the habitat loss in the
region (Nepstad et al 2006). Between 1987 and 2006 an average of 18,000 km2 of prime
jaguar habitat was lost in this region every year, mostly from the Amazonian agricultural
frontier (PRODES 2007). In the last two decades, Brazil has lost larger areas of jaguar habitat
than any other country.
In 1967, the Brazilian Wildlife Protection Act prohibited commerce in wildlife and products
derived from their capture, pursuit or destruction. The Convention on International Trade in
Endangered Species (CITES) of 1973 made it illegal to trade jaguar skins or parts for
commercial gain. The CITES listing, in combination with the Brazilian legislation and anti-
fur campaigns, brought about a sharp decline in the fur trade, helping to reduce the pressure
on jaguar populations in the wild. However, jaguar persecution continues (Crawshaw 2002,
Michalski et al. 2006) – now very rarely for the illegal trade, but because of their perceived
threat to people and their livelihoods. This indiscriminate killing of jaguars is one of the most
serious threats to their survival across all of Latin America (Zeller 2007).
Jaguars, livestock and people have co-existed in Brazil for many decades across a wide range
of ecological, cultural, and socio-economic settings. From small family-run farms in the dry
Caatinga to commercial large-scale ranches in the wetlands of the Pantanal, from old
traditional cattle ranches in the Atlantic rainforest to recent settlements on the Amazon
agricultural frontier, Brazil is the perfect test tube in which to explore the interacting
chemistry of jaguars, livestock and people.
207
The Pantanal
The Pantanal is located in the geographic centre of South America and spans the borders of
Brazil, Bolivia and Paraguay (Figure 2). With a highly seasonal climate, the Pantanal receives
an average of >1.2 m of rainfall annually which causes vast amounts of areas to be flooded
and a subsequent flush of green grasses available for both native and domestic ungulates. The
Pantanal is characterized by savannas interspersed with isolated islands of secondary forest,
which are an important refuge for both predators and prey. Gallery forests border temporary
and permanent rivers and provide long corridors for wildlife.
Almost a third of published scientific articles on jaguar biology and conservation concern
Brazil. While these topics have been addressed in the Brazilian Amazon (Michalski et al.
2006, Oliveira 2002), Cerrado (Silveira and Jacomo 2002, Palmeira et al. 2008) and Atlantic
Rainforest (Conforti and Azevedo 2003, Crawshaw et al. 2004, Cullen et al. 2005, Garla
2001, Leite et al. 2002), the Pantanal accounts for the greatest portion of publications about
jaguars (e.g. Schaller and Vasconcelos 1978, Schaller 1979, Schaller and Crawshaw 1980,
Schaller 1983, Crawshaw and Quigley 1984, Crawshaw 1987, Quigley 1987, Crawshaw and
Quigley 1991, Quigley and Crawshaw 1992, Dalponte 2002, Crawshaw 2002, Quigley and
Crawshaw 2002, Zimmermann et al. 2005, Soisalo and Cavalcanti 2006, Azevedo and
Murray 2007a, 2007b).
In this landscape mosaic, cattle have been ranched for >200 years (Wilcox 1999). The
Pantanal consists almost entirely of large cattle ranches (e.g. average ranch size 12,950 ha 
SE 22,444 ha, Zimmermann et al 2005). Cattle are raised extensively in the region, with an
average cattle density of 16 head/km2 (Mourão et al. 2002). People and jaguars, however,
have coexisted uneasily. Jaguars have long been blamed for killing cattle and, in the past,
ranch owners employed men solely to hunt jaguars. The extent of retaliation by ranchers was
considerable. For example, in the early 1980s, 68 jaguars were killed over 8 years on one
ranch alone (Crawshaw in IUCN/SSC 1986). Whether as a result of legislation or the
economic crisis in cattle ranching caused by the severe flood of the 1970s, the rate at which
jaguars are killed appears to have lessened and jaguar abundance in the Pantanal appears to
be increasing (Crawshaw 2002) Nonetheless, as ranchers own 95% of this vast region, the
future of jaguars in the Pantanal is inextricably linked to the ranchers‟ perceptions and
attitudes towards them.
Assessing the realities and perceptions behind the conflict
In this chapter we weave together several studies conducted by the authors between 2000 and
2008 which explored human-jaguar conflict in the Pantanal and the Amazon from various
angles: data regarding jaguar predation rates on a cattle ranch, perceptions and attitudes of
ranchers towards jaguars and livestock losses, and the various factors that may shape human
beliefs and behaviour.
To document the realities of jaguar predation on livestock and native prey, the Pantanal
Jaguar Project quantified kill rates, composition of prey killed, characteristics of prey killed,
208
and patterns of predation on a ranch in the southern Pantanal (“Fazenda Sete” in Figure
2)(Cavalcanti and Gese, in review). In addition, GPS telemetry provided information on
jaguar movements (Cavalcanti and Gese, accepted) and facilitated analysis of habitat use and
spatial patterns of predation (on both domestic and native species) in relation to the type and
distribution of vegetation and other landscape attributes. Ten jaguars were equipped with
Global Positioning System (GPS) radio-collars (Televilt, Sweden), which recorded their
locations at two-hour intervals, enabling us to identify kill sites and thereby to find and
document 438 carcasses of prey (including the identity of the predator, the date and
approximate time of death, the period for which the predator stayed by the carcass, and the
vegetation cover at the kill site; Cavalcanti and Gese, in review, Cavalcanti et al., in prep).
Meanwhile, the Coexistence Project studied factors determining people‟s perceptions of

jaguars and how these perceptions translated into human persecution of jaguars. Interviews
with ranchers were used to assess the following: (i) socio-demographic variables, (ii)
description of the property, (iii) respondents‟ knowledge about jaguars and depredation
problems and perceptions of the jaguars‟ impact on (iv) livestock and (v) human safety,
together with perceptions of (vi) an increase in jaguar abundance, (vii) degeneration of
economic situation, (vii) the social acceptability/desirability of persecuting jaguars, including
the importance of traditional jaguar hunting, (viii) the ease or difficulty of this persecution,
(ix) attitudes towards jaguars, and (x) intention to persecute jaguars. Answers in either a
binary “yes/no” or in 3- or 5-point scale formats enabled us to construct measurement scales
(0 to 10 for knowledge and perceptions and -10 to 10 for attitude) and combine responses into
an additive score for each variable (the higher the score the greater the knowledge or
perception and more positive the attitude; Figure 6). In order to assess the degree to which the
findings from the Pantanal can be extrapolated to other regions or what is culture-specific in
human-jaguar conflicts, we replicated this study on an agricultural frontier area in southern
Amazonia (municipality of Alta Floresta). Like the Pantanal, the Amazon site hosts relatively
high densities of both jaguars and livestock, but as a recently established agricultural frontier
it differs in many social and cultural aspects from the Pantanal. Unlike the Pantanal, habitat
loss is a major threat to jaguars on the Amazon frontier. This study involved 45 ranchers in
two subregions of the northern Pantanal – Cáceres and Poconé - and 103 in Amazonia
(Figure 2; Marchini and Macdonald, in prep a).
We also examined the attitudes and conservation values of ranchers from an earlier study
involving 50 ranchers in the three subregions of the northern Pantanal, namely Cáceres,
Poconé and Barão de Melgaço (Figure 2), in which we investigated the association between
attitudes and some socio-economic variables such as age, ranch size, cattle herd size and
density, reported cattle losses, and involvement in tourism. Attitudes were explored using a
series of suggested statements regarding jaguars and conservation, and responses
were recorded on a five-point Likert scale so that they could be combined into an
additive score, and the relationships between the combined score and potential explanatory
variables were analyzed (Zimmermann et al 2005).
The realities of jaguar foraging ecology
The radio-tracking study (Cavalcanti and Gese, in review) revealed that native species
comprised 68.3% of the prey killed, with the remainder being cattle (31.7%). For individual
209
jaguars, the amount of cattle killed varied widely (Figure 3). Individuals also differed in the
species diversity of their diets; although collectively the ten jaguars killed 24 prey species,
some killed few prey species, others killed many species (Table 1). Jaguars killed ungulates
predominantly, but they also killed and consumed other predators, such as maned wolves
(Chrysocyon brachyurus), crab-eating foxes (Cerdocyon thous), coati (Nasua nasua), and
raccoons (Procyon cancrivorous).
Based on kills reported by ranch-hands, Crawshaw and Quigley (2002) calculated that cattle
comprised 46% of jaguar kills in the southern Pantanal. In their data, small prey was probably
under-represented insofar as they may be killed and consumed in secluded sites (see also
Schaller 1979). This bias might also affect our findings, although a small proportion of the
biomass killed and consumed was of small prey (e.g., birds, caiman lizard, coati, small
anaconda, armadillo). Homing in on radio-collared jaguars, Crawshaw and Quigley (2002)
found 17 prey items of which 29% were cattle and 41% were white-lipped peccaries – a close
match to our overall finding of cattle accounting for 31.7% of jaguar kills, varying
seasonally between 19.2% and 48.9%, respectively for wettest and driest periods of the 4-
year field study (Cavalcanti and Gese, in review).
Calves (<1 year old, <174 kg)) accounted for 69% of cattle killed by jaguars (Cavalcanti and
Gese, in review), which is higher than Crawshaw and Quigley (2002) reported (43%) in their
study in the same area in the southern Pantanal; perhaps again due to carcass detection bias.
These findings from the Pantanal are broadly consistent with those reported elsewhere. In
Venezuela, jaguars attacked young cattle (weaned calves and heifers 1-2 years of age) more
often than they did adults (Hoogesteijn et al. 1993, Farrell 1999, Scognamillo et al. 2002). In
northeast Argentina, cattle between 1-3 years comprised the majority of jaguar kills (Perovic
2002). Younger calves of 3-9 months of age comprised the majority of jaguar kills in
northern Goiás, central-western Brazil (Palmeira et al. 2008). Azevedo and Murray (2007)
found that in the southern Pantanal predation risk was higher among calves up to 12 months
of age.
Although jaguars can kill mature bulls (Hoogesteijn et al. 1993), we documented not a single
jaguar attack on an adult bull, and only one instance of scavenging on a bull carcass. Contrary
to the beliefs of ranchers, the GPS data indicated that jaguars scavenged a proportion of their
prey (we found six instances, involving three individuals, of feeding substantially from cattle
that had died from other causes) (see also Lopez-Gonzales and Piña 2002). Scavenging
complicates the interpretation of diet analyses based on undigested remains in faeces.
At 19 kill sites located by GPS-tracking the remains of two different prey species were found
(Cavalcanti and Gese, in review). We deduced this might have occurred when a jaguar killed
a species scavenging from the original kill, and in 79% of these occasions this was a plausible
explanation (e.g. one of the carcasses was of a potential scavenger, such as feral hog, peccary,
armadillo, raccoon, or caiman). This „scavenger-trap‟ hypotheses seemed inappropriate for
the remaining 21% of double kills, insofar as neither of the victims was a scavenger (e.g.,
calf, brocket deer, giant anteater, lesser anteater).
Jaguars are often considered nocturnal predators. However, we found the time of day in
which jaguars killed s was evenly distributed throughout the 24 hour period, even when
examining individual prey species (Cavalcanti and Gese, in review). Jaguars appear to be
210
adaptable to the movement and activity patterns of various prey species and readily exploit
these species when they are active or vulnerable to predation.
When examining the seasonality of predation by jaguars, we found the average number of
cattle, caiman and peccaries (the three major prey species) killed by radio-collared jaguars
each season indicated a peak of predation on cattle in the dry seasons of each year (Figure 4;
Cavalcanti and Gese, in review). The frequency of jaguar predation on caiman appeared to
be constant throughout all months of 2002, while predation appeared to peak during the wet
seasons (February-March) of 2003 and 2004. There may be an inverse relationship between
jaguar predation on cattle and caiman; as water levels recede in the Pantanal, caiman move
with these levels and predation declines, conversely as water levels recede cattle are moved
into these areas for grazing and jaguar predation on cattle increases. The fluctuation of water
levels is the major driver in this ecosystem dictating the availability and vulnerability of prey
species, including cattle. The frequency of jaguar predation on peccary also appeared to be
constant throughout 2002, then increased in 2003 and 2004. Seasonally, the mean number of
peccaries killed each month by jaguars appeared to be lowest during the wet seasons
(February-March; Figure 4). However, despite an apparent tendency for the number of cattle
killed each month to have declined over the 4-year study, the actual seasonal predation rates
of jaguars on cattle did not statistically decline between 2002 and 2004 (Figure 5).
Conversely, while the data suggest an increase in the number of caiman killed by jaguars
each month, the observed seasonal predation rates of jaguars on caiman did not statistically
increase over the seasons. Jaguar predation rates on peccaries did increase significantly
between the wet season of 2001-2002 and the dry season of 2004 (Figure 5). The increase in
jaguar predation rates on peccary during the study occurred during a period of relatively high
peccary densities (9.63 individuals/km2; Keuroghlian 2003). This increased predation rate by
jaguars on peccary during the study suggest the availability of alternative prey could reduce
predation rates of jaguars preying on cattle and could serve as a buffer species.
Because jaguars are ambush predators, an obvious prediction would be that kills were
associated with dense vegetation. Cavalcanti et al. (in prep.) found that while the 10 GPS
collared jaguars used forests and shrublands preferentially, kills were made in habitats in
proportion to their availability. Cattle, caiman, and peccaries killed by jaguars (n = 327) were
distributed in the various habitat classes according to their availability, except during the dry
season when caiman and peccaries were mainly killed in shrublands and forests, respectively.
Male and female jaguars consistently selected shrublands during both wet and dry seasons.
Although there was little evidence that particular species were killed in particular habitats,
there was a tendency for cattle to be killed further than expected from water.
Some authors have hypothesized that jaguar predation on cattle is a function of the
distribution, availability, or proximity to forest habitat or forest edges (Hoogesteijn et al.
1993, Rabinowitz 1986, Michalski et al. 2006, Palmeira et al. 2008). Hoogesteijn et al.‟s
(1993) comparison of three ranches in Venezuela led to the conclusion that jaguars killed
cattle closer to forested areas. Rabinowitz (1986) reported jaguars readily killed domestic
livestock that entered forested areas, but not when cattle were in open pastures. Quigley
(1987) reported cattle were killed only in gallery forests and forest patches, although some
might have been dragged there from the forest-edge. Cavalcanti et al. (in prep.) reported that
the 10 GPS-collared jaguars neither selected forested areas nor avoided open fields when
killing cattle. Rather, they killed cattle in all habitat types. During the wet season, cattle were
killed by jaguars significantly closer to forest edges than in the dry season. During the wet
211
season cattle were able to forage in chest-deep water, but they needed dry ground on which to
spend the night. Therefore, they might spend more time closer to forests, which are typically
associated with higher and drier ground. Several authors have suggested keeping cattle herds
away from forested areas as a strategy to minimize jaguar attacks (Rabinowitz 1986,
Hoogesteijn et al. 1993, Michalski et al. 2006, Palmeira et al. 2008), but we recorded jaguar
attacks on cattle in other habitats as well (Cavalcanti et al., in prep.).
Individual variation in jaguar diets: do “problem animals” exist?
Since jaguars differed individually in their diet (Cavalcanti and Gese, in review), we
examined whether some jaguars contributed more than others to the levels of domestic stock
losses (cf Linnell et al. 1999). Indeed, prey remains of individual jaguars indicated that while
cattle comprised >50% of the diet for some individual jaguars, for others it did not exceed
5%. Nevertheless, each of the 10 radioed jaguars killed cattle. Whether or not killing the
predominant cattle-killers would ameliorate the problem (e.g. as suggested by Rabinowitz
1986, Hoogesteijn and Mondolfi 1992) depends on the causes of this individual variation
(i.e., causes may include availability and vulnerability of prey, preference of particular prey
species, or cultural learning from their mother). However, we also found that for some
individuals that had >50% of their kills comprised of cattle in 2002 (a dry year), these same
jaguars exhibited an appreciable decline in cattle kills in 2003 (a wet year). Again, water
levels, and the consequent movement of both caiman and cattle, likely played an important
role in the availability of these two key prey species within individual jaguar home ranges
and therefore influenced encounter rates (Cavalcanti and Gese, in review).
Previous analyses of the variation in the level of livestock depredations suggest that males are
more likely to kill cattle than are females (e.g., Rabinowitz 1986, Stander 1990, Chellam and
Johnsingh 1993). However, results from the 4-years of study found no differences between
males and females in the level of predation on cattle (Table 1; Cavalcanti and Gese, in
review). Jaguars, especially females, may kill cattle in excess of their needs and might be
considered a mechanism for teaching their young to hunt (A. Silva, V. Correia, A. T. Neto, B.
Fiori, pers. comm.). While amongst the Carnivora, surplus killing is almost universal (Kruuk
1972), neither in our study, nor any other, has it been documented amongst jaguars. In
general, the time interval between kills, and the time spent at each kill, was related to prey
size (Cavalcanti and Gese, in review). After killing and consuming a small prey item, a
jaguar generally killed again in a shorter time (3.0 days before making another kill) as
compared to when they killed larger prey (5.1 days before making another kill). Similarly, the
length of time jaguars stayed at a carcass site significantly increased with increasing body
mass of prey; 16.0 hours were spent on small prey increasing to 27.9 hours on larger prey
(Cavalcanti and Gese, in review). Some authors have speculated that livestock-depredation is
more prevalent amongst subadult than adult jaguars (Rabinowitz 1986, Stander 1990,
Saberwal et al. 1994), whereas others conclude adults are more likely to kill cattle than
younger individuals (Bowns 1985, Esterhuizen and Norton 1985). In our study, stock-killing
occurred at a rather constant rate among individuals. On average, jaguars killed 1 calf every
13.3 ± 15.5 days, while adult cows were killed at a lower rate of 25.5 ± 18.4 days between
kills, but these rates varied annually (Cavalcanti and Gese, in review). The level of rainfall in
any given year appeared to be the most influential factor affecting individual jaguar predation
rates on cattle by determining the availability of cattle on the landscape (Cavalcanti and Gese,
in review). During wet years, cattle were held on higher drier ground for a longer period,
thereby reducing the number of jaguar territories exposed to cattle. Conversely, during dry
years, cattle were more dispersed over the landscape, thereby being exposed to more jaguar
212
territories and increasing encounter rates being individual predators and domestic prey.
Husbandry practices likely also had a large influence on jaguar predation. In the Pantanal,
calves were generally born over several months, increasing the time period over which
vulnerability to jaguar predation was prolonged. In addition, pregnancy rates of cows are
generally well below optimal, often between 60-75%. Native ungulates usually flood a
predator by having a short birth pulse, thereby decreasing the length of time that young are
exposed or vulnerable to predation. Shortening the birth pulse and increasing the number of
pregnant cows within a cattle operation could, in theory, reduce overall predation losses
within individual jaguar territories where calving grounds are located by flooding an
individual cat with far more prey than can be killed; assuming a Type III functional response
where beyond a certain level of prey density, satiation of the predator causes an asymptote in
the kill rate.
A common hypothesis in terms of jaguar predation on livestock is that it is prevalent amongst

wounded or sick predators, and this idea has been mooted for jaguars (Rabinowitz 1986, Fox
and Chundawat 1988, Hoogesteijn et al. 1993). Indeed, two studies in Venezuela both
revealed that the majority of the jaguars (75% and 53%) killed as part of predation control
had previously sustained severe wounds precluding them from hunting normally (Hoogesteijn
et al. 1993) although the condition of jaguars not killed could not be confirmed. In our study,
all ten radio-collared jaguars were in excellent physical condition at the time of capture
(Cavalcanti and Gese, in review), as too were those documented by Schaller and Crawshaw
(1980) and Hopkins (1989). Furthermore, we could detect no difference in predatory
behaviour between older or younger animals. The oldest individual radio-tracked - a male
estimated to be >11 years old - had two missing canines (a broken lower canine on his first
capture and a second broken upper canine on his second capture) and killed white-lipped
peccaries, feral hogs, and marsh deer at a similar rate (7.1 ± 5.6 days between kills) as did a
young jaguar in the sample (3.6 ± 3.4 days between kills; no significant difference between
these two cat‟s kill rates).
Perceptions about depredation and persecution
Depredation problems caused by jaguars have been reported by 82% of the landowners in the
northern Pantanal (Zimmermann et al. 2005, Marchini and Macdonald, in prep a). Not
surprisingly, jaguars are considered the most detrimental species to human livelihoods by
73% of 110 ranchers and ranch-hands interviewed in both the southern and northern Pantanal
(Marchini 2003). Reported losses to jaguars range from 0 to 11% of their livestock holdings,
with greater proportional losses among smaller ranches and smaller herds (r = -0.590 and -
0.716, both P < 0.001)(Zimmermann et al. 2005) with losses averaging between 2.1%
(Marchini and Macdonald. in prep a) to 2.3% (Zimmermann et al. 2005) of their livestock
holdings. In absolute terms, the greatest reported loss was 80 calves in one year from a herd
of 2,000 head on a 13,200-ha ranch. Given the average price of a calf in the region
(approximately US$ 228 in 2008), this case translates into a monetary loss of US$ 18,240
(Marchini and Macdonald. in prep a). Over one-third of the respondents (38%) ranked
jaguars as a greater problem affecting their income from cattle than floods, droughts, rustling
and disease (Zimmermann et al. 2005).
Most ranchers (62%) reported that jaguar attacks show no seasonal pattern (Zimmermann et
al. 2005). As for variation among years, 24% of the respondents believed the frequency of
213
attacks within their ranches was currently increasing, 35% believed it was declining and 41%
stated it was not changing (Marchini and Macdonald, in prep a). Seventy-two percent
believed that jaguars varied in their dietary preferences and thus believed that “problem
jaguars” were the ones killing cattle (Marchini and Macdonald, in prep a).
These findings suggest that perceptions of jaguar depredation might sometimes exceed
reality, as ecological studies addressing jaguar depredation in the Pantanal and elsewhere
revealed lower losses of livestock holdings (0.83% in two ranches of northern Pantanal; 0.3%
in one ranch of southern Pantanal; 1.26% in southern Amazonia)(Dalponte 2002; Azevedo
and Murray 2007b, Michalski et al. 2005, respectively). However, in the study in which
predation rates were estimated from GPS collared jaguars (Cavalcanti and Gese, in review),
the ranch foreman reported the ranch lost on average 70 head of livestock annually to jaguar
predation out of 6000 head (1.2% of livestock holdings). During a dry year (2002), a jaguar
killed an average of 2.1 calves/month and 0.6 adult cows/month, for a total of 2.7 head of
cattle/month. Extrapolating this kill rate to half (not all jaguars have equal access to cattle)
the estimated resident (80%) population of jaguars on the ranch (6.7 jaguars/100 km2 (Soisalo
and Cavalcanti 2006) would generate an estimated loss of about 390 head of livestock.
Conversely, during wet years (2003) on the study, a jaguar killed an average of 0.5
calves/month and 0.3 adult cows/month, for a total kill rate of 0.8 head of cattle/month.
Again extrapolating to half the resident jaguar population on the ranch generates an estimated
118 head of livestock lost. Data from a wet year placed cattle loses on the ranch similarly
between the perceived loss (70 head; 1.2% of cattle) and the estimated loss from actual jaguar
predation rates (118 head; 1.9% of cattle). However, during the dry year, predation rates
indicated over 5 times more cattle were lost (390 head; 6.5% of cattle) than the ranch
foreman perceived (70 head). Therefore, the level of rainfall is very influential on the
number of cattle lost annually and is directly related to the level of access the cattle have to
the landscape and the number of jaguars to which they are exposed. In addition, we generally
found cattle killed by the radio-collared jaguars that were never found by the ranch hands and
unreported losses are likely higher than previously believed. Ranch hands easily found cattle
kills in open fields and pastures, while missing most kills in the dense cover of shrublands
and forests.
We emphasize that these extrapolations on predation rates are from only one study and may
not be representative of all ranches in the Pantanal. However, it does raise the issue that
accurate and unbiased documentation of jaguar kill rates on livestock and native prey are
needed to lend credibility to claims on both sides of the argument regarding losses sustained
by livestock operations. In a study examining wolf (Canis lupus) predation on livestock in
central Idaho, USA, researchers reported that ranchers found only one in eight of the actual
kills documented (Oakleaf et al. 2003). During the years of wolf reintroduction into the
United States, agency personnel consistently agreed that a rapid response time and accurate
documentation of actual losses were critical to any compensation program proposed for
ranchers and can often lead to heated debate as to the actual level of losses sustained by a
ranching operation. Some ranchers were very diligent in keeping track of losses, while others
were less accurate and blamed predators for more losses than actually occurred.
In addition to the perceptions about the level of jaguar depredation on livestock, other beliefs
and perceptions about jaguars and jaguar hunting may be relevant in dealing with conflicts
between ranchers and jaguars. Thirty percent of the ranchers held the perception that jaguar
abundance was currently increasing (4% perceive it as decreasing; Marchini 2003). In the
214
Pantanal subregion of Cáceres, 80% believed jaguar abundance was increasing and there was
a widespread perception that jaguar numbers were now abnormally - and unbearably – high
(Marchini and Macdonald, in prep a).
Fifteen percent of the ranchers in the subregion of Poconé believed that jaguars caused cattle
mortality even without preying on them (Marchini and Macdonald, in prep a). The rationale
was that jaguars scared cattle out of the "capões" (dry forest patches where cattle find refuge
during floods), from which the cattle then ran to flooded areas, where they drowned or get
stuck in the mud and starved to death. This belief in “indirect predator-induced mortality” of
livestock needs further investigation.
Some people perceived jaguars as a threat to human safety. Fifty-three percent of the
respondents agreed that jaguars attacked people even when not provoked (Marchini and
Macdonald, in prep a). A rural school in Cáceres closed its doors in 2008 because the pupils
refused to attend classes after several sightings of jaguars in the vicinity. This episode
occurred, and the foregoing data were gathered prior to an incident on 24 June 2008 when a
young fisherman was killed by a jaguar while sleeping in his tent on a riverbank of the
Paraguay River, in the subregion of Cáceres. This was the first officially documented
unprovoked fatal attack of a jaguar on a human in Brazil, and was widely covered by the
national media. Prior to this incident, attacks were almost invariably associated with hunting
situations in which the jaguar was cornered or injured. Jaguars are also known to attack to
defend their cubs or the carcass upon which they are feeding. The impact of the above event
on people‟s perceptions of the risk that jaguars pose to human safety is currently being
assessed.
Many ranchers unashamedly admit the perception that killing jaguars is socially acceptable.
Only 15 percent of the respondents believed their neighbors or family would disapprove their
killing jaguars (Marchini and Macdonald. in prep. a). It is considered one of the traditions of
the Pantaneiro culture. Additionally there is the general view that all aspects of this culture
should be cherished and preserved. Indeed, a prevailing opinion is that hunting jaguars is an
act of bravery and a test of horsemanship among cowboys. Shooting a jaguar enhances a
cowboy‟s reputation, to the extent that the hired hands continue to do it even where
(sometimes absentee) owners have specifically banned the practice (S. Cavalcanti, pers. obs.,
amongst others).
The extent to which a complainant perceives it to be difficult to kill a problem jaguar may
affect the likelihood of pursuing this option. The general approach is to use dogs to find and
pursue the jaguar. Either the jaguar climbs a tree or turns at bay on the ground, whereupon
the hunters come up and kill it. In the Pantanal, hiring a professional hunter who owns a pack
of trained dogs can be relatively easy and affordable (sometimes a cow is offered in exchange
for the service), but in other regions, the difficulty and cost of hiring a hunter may discourage
small ranchers from killing jaguars. Several small landowners on the Amazon agricultural
frontier, for instance, told us they had never killed a jaguar but would have done so had they
had the means (Marchini and Macdonald. in prep a).
Jaguars also elicit positive feelings among ranchers. All our respondents considered the
jaguar a “beautiful” or “very beautiful” animal (Marchini and Macdonald in prep a), and 16%
would choose the species to be the symbol of the Pantanal (only the jabiru stork, the official
symbol of the region, ranks higher; Marchini 2003). Although we met ranchers who
215
confessed hatred of jaguars, irrespective of their behaviour, the average attitude score value
of ranchers in the region, based on questions that assessed the individual‟s like or dislike for
jaguars (e.g. how would you feel if all jaguars disappeared?) and unfavourability or
favourability towards jaguar persecution (e.g. Killing any jaguar that shows up in your
property this year will improve your livelihood?), was positive (Figure 6)(Marchini and
Macdonald in prep a).
Finally, the economic decline in the region may exacerbate the conflict between ranchers and
jaguars. In recent decades, growing competition within the cattle industry, higher taxes, and
generational land-splitting has rendered cattle-ranching less profitable in the Pantanal (Swarts
2000). Indeed, 95% of the ranchers believed their economic situation is worse now than in
the past (Marchini and Macdonald, in prep a). A decline in the financial margins of profits
from cattle ranching may decrease their tolerance of jaguar depredation on their cattle. The
expressive growth of ecotourism in the region has brought the hope of better days to some
ranchers (and conservationists as well), although ecotourism alone seems unlikely to be the
complete solution.
Variation in perceptions and its determinants
In order to understand how and why the above perceptions vary, we examined correlations
between perceptions and some socio-economic and demographic variables (Table 2). Details
of these analyses are in Marchini and Macdonald (in prep a).
The perceived impact of jaguars on livestock, which was measured using questions about
recent and past depredation events, magnitude of the loss in the evaluation of the rancher
(from none to very large) on his ranch as well as neighbouring and relatives‟ ranches, and
current trends in the jaguar depredation problem (decreasing, unchanged, increasing), was
positively correlated with the perception of increasing jaguar abundance (r = 0.41, p < 0.02)
and declining economic situation (r = 0.47, p < 0.04). It was also negatively correlated with
attitude to jaguars (r = -0.61, p < 0.0001) and years in school (r = -0.49, p < 0.0001): ranchers
holding stronger negative attitudes to jaguars and with fewer years in school (education
varied greatly among respondents, from 33% of them being illiterate to 22% with higher
education) had greater perceptions of impact. Attitude to jaguars was also negatively
correlated with the respondents‟ perception of the deterioration in the economic situation (r =
-0.57, p < 0.04) and positively correlated with years in school (r = 0.36, p < 0.0001), which in
turn was negatively corrrelated with age (r = -0.50, p < 0.001). The rationale for using
different questions to assess the perception of impact on livestock is that a rancher‟s
evaluation of the impact is not based solely on his recent losses to jaguars. The same loss can
be seen as small or large by different ranchers, depending on their background and socio-
economic situation. The perceived impact of jaguars on human safety, measured using
questions about the belief in unprovoked attacks by jaguar on humans and man-eating habit
among jaguars, first- or second-hand reports of jaguar attacks on people (fatal or not),
magnitude of threat to human safety and fear of jaguars (none to very large), was positively
correlated with perceived increase in jaguar abundance (r = 0.45, p < 0.02) and negatively
correlated with an index of the respondent‟s knowledge of jaguar ecology and depredation
problems (r = -0.53, p < 0.0001). In sum, if perceptions of jaguars‟ impact on livestock and
human safety determine retaliatory persecution, then the foregoing correlated factors may
also play a role in human-jaguar conflicts in the Pantanal.
216
Some differences between the perceptions in the Pantanal versus the Amazonia region are
relevant to this discussion (Figure 6)(Marchini and Macdonald, in prep a). The perception of
the jaguars‟ impact on livestock was more serious in the Pantanal than in the Amazon (t = -
9.966, p < 0.0001, df = 149) whereas the perceived threat to human safety was greater in the
Amazon than in the Pantanal (t = 2.919, p = 0.004, df = 149). Even though attitude scores
were similar in the two regions (t = -1.112, p = 0.268, df = 149), in the Pantanal, attitude was
correlated with the perceived impact on livestock (see above), whereas in the Amazon it was
correlated with the perceived impact on human safety (r = -0.39, p < 0.01). We also found
differences in the perception of social acceptability of jaguar hunting, which was assessed by
questions as to whether the respondent felt his family and neighbours would approve or
disapprove if he killed jaguars and his willingness to comply with them. The acceptability of
killing jaguars was higher in the Pantanal than in Amazonia (t = - 2.962, p = 0.004, df = 149)
and so was the perceived ease of persecuting jaguars (t = -13.044, p < 0.0001, df = 149).
People in the Pantanal were more knowledgeable about jaguars and depredation problems
than were people on the Amazon frontier (t = -7.684, p < 0.0001, df = 149). For instance,
whereas 89% of the respondents in the Pantanal could tell the difference between jaguar and
puma tracks, only 7% of the respondents in the Amazon were correct in their identification
skills.
From perceptions to persecution
From a conservation standpoint, what ultimately matters in conflicts between people and
jaguars is the level of persecution and its impact on a carnivore population. To investigate
the relationship between perceptions, attitudes, and persecution, we used a hierarchical
cognitive model based on the correlations mentioned above and adapted from the Theory of
Planned Behaviour (TPB, Ajzen 1985). This is an influential theory in social psychology
attempting to predict a person‟s behaviour (see also Macdonald et al., this vol, Chapter 29).
In the vocabulary of the TPB, a person‟s behaviour is explained by behavioural intention,
which is preceded by attitudes towards the behaviour. Intention also depends upon
subjective norms, which is a person‟s perception of the social acceptability or desirability of
the action in question, and perceived behavioural control, which is the actor‟s perception of
the ease or difficulty of performing the specific action (Ajzen 1985). In our model, the
perceptions of a jaguars impact on livestock and human safety affects attitudes towards
jaguars. The intention to persecute jaguars is also preceded by subjective norms and
perceived behavioural control concerning persecution of jaguars (Figure 7). This approach
allows us to assess the relative importance of the different components of the causal chain of
jaguar persecution so that more effective interventions can be devised to decrease
persecution. Background factors such as age, education, wealth, occupation, culture and
knowledge about jaguars may influence these perceptions, but are not incorporated in the
causal model.
Marchini and Macdonald (in prep a) assessed the intention to persecute jaguars via the
question: Would you kill any jaguar that shows up in your property? The answer to this
question was expressed in the form of a dichotomy: a person either intended to persecute or
not. Evidence of recent persecution of jaguars was found in 27 ranches (8 in the Pantanal and
19 in Amazonia), which facilitated validation of this measurement. Most (81%) of the people
who had killed jaguars in the previous two years said that they intended to persecute any
217
jaguar that showed up on their ranch, whereas 20% of the people who had not killed any
jaguar expressed the intention to persecute (χ2 = 35.301, df = 1, P < 0.000). This seeming
association between declared intentions and actions strengthens our belief that we should take
seriously the statements other ranchers made to us about their intentions towards jaguars.
Almost 60% of the landowners in the Pantanal declared the intention to kill any jaguar that
showed up on their land, whereas in the Amazon about 20 % of the landowners did so.
Regression analysis revealed that attitudes and subjective norms significantly explained the
variation in the intention to persecute jaguars in the Pantanal: more negative attitudes towards
jaguars and a greater perception of the social acceptability of jaguar hunting were associated
with a greater intention to kill jaguars (β = -0.259, p = 0.01 and β = 0.497, p = 0.024,
respectively; -2log likelihood = 46.722). Several ranchers in Marchini and Macdonald‟s (in
prep a) sample also expressed the view that killing jaguars was appropriate on the grounds
that it was a tradition passed from generation to generation. The important influence of norms
was unsurprising considering that many ranchers in northern Pantanal were inter-related, with
a network of family bonds linking ranches. Attitude, in turn, was correlated with the
perceived impact of jaguars on livestock (R2 (adjusted) = 0.364, F = 26.138, p < 0.0001).
Cattle ranching is an icon of the Pantaneiro culture. The few traditional families that together
own a substantial portion of the lands in the northern Pantanal have been raising cattle in the
region for generations, and this has been the only viable economic activity. The economic
and cultural centrality of cattle ranching in the region doubtless affects the high correlation
between perceptions of the jaguars‟ impact on livestock and the attitudes towards jaguars.
Although cattle ranching in the Pantanal is generally undertaken at such a large scale that the
loss of a few cattle is unlikely to seriously impact the ranchers‟ livelihoods, for the majority
of ranchers, such losses are unacceptable and may be higher than actually realized given
detection rates of kills. In addition, while losses may only be 1-3% of an operation, these
losses could be substantial in monetary form for any individual rancher. The rancher having
<1% losses may be able to absorb those costs, while the neighboring rancher having 3%
losses may be in an economic situation threatening the viability of that operation (i.e., it is
those operations with high losses that are most imperiled)
In contrast, in the Amazon, the intention to persecute jaguars was significantly explained by
attitude and perceived behavioural control (β = -0.481, p < 0.0001 and β = 0.663, p = 0.011,
respectively. -2log likelihood = 66.831). Indeed, a significant proportion of the landowners,
and particularly those with smaller properties, favored the idea of killing jaguars, but did not
intend to engage in this activity because they believed that they lacked the means (or were not
brave enough, as they told us) to do so. In the Amazon sample, norms did not significantly
affect the intention to persecute jaguars, which might reflect the reality that in this frontier
area people typically have little interaction, or shared background, with their neighbours.
However, their attitudes were heavily associated with the perceived risk of jaguars on human
safety (R2 (adjusted) = 0.488, F = 18.385, p < 0.0001). Fear of jaguars is common among the
frontiersmen, who were largely immigrants with little experience with jaguars and the forest.
Finding solutions for the future of jaguar-human coexistence
Direct persecution of jaguars by people, together with hunting of jaguar prey, is the most
significant threat to the long-term survival of jaguars throughout their range (Sanderson et al
2002, Zeller 2007). Most persecution is directed at jaguars living near or within areas of
218
livestock-raising. Jaguars kill livestock and this creates a conflict with ranchers from an
economic perspective. Aspects of jaguar ecology and behavior elucidated by our study have
direct implications for this economic aspect of jaguar conservation. The obvious, and
traditional, response has been attempting to curtail jaguar depredation on livestock through
preventive measures. A radical, but evidence-based, alternative would be for all stakeholders
to recognize the reality that cattle are routinely a component of jaguar diet in the region.
Under the Biodiversity Impacts Compensation Scheme (BICS) model proposed by
Macdonald (2000)(elaborated with respect to carnivore conflict by Macdonald & Sillero-
Zubiri, 2004), the approach would be to refine management interventions to reduce negative
impacts (stock losses), and then find other mechanisms to offset irreducible damage – in this
case, alternative mitigation measures to make bearable the residual stock losses to jaguars.
Additionally, while kills of domestic stock may be related to a lack of natural prey (Saberwal
et al. 1994, Vos 2000), insofar as leaving predators no alternative choice of food, this
chicken-and-egg logic can be reversed insofar as domestic stock adds to the carrying capacity
of the environment for predators. Schaller (1972) found that the more abundant a preferred
species was, the more likely it was to fall prey to lions. By extension, in the Pantanal, cattle
are both the most abundant and the most vulnerable prey, so one might recognize some level
of jaguar predation as inevitable and a natural part of ranching, like drought or soil fertility
(Soisalo and Cavalcanti 2006). By analogy, there are limits to the feasibility of mitigating
such environmental affects on agriculture, and limits to what society deems an acceptable
cost of environmental intervention – for example, the latter is clearly illustrated in Europe by
payments to farmers for custody of nature under the Common Agricultural Policy (Dutton et
al. 2008). To the extent that irreducible damage by jaguars to cattle ranchers must be off-set
(rather than tolerated as an inevitable consequence of farming in jaguar country), solutions
might lie in financial instruments such as tax benefits, favourable credit, or a regional
increase in beef prices. The significance of losses to jaguars will be proportionally diminished
by ranchers improving other aspects of rudimentary herd husbandry that currently often
account for much more significant losses than do jaguars (Hoogesteijn et al. 1993). That said,
eventually the quest for efficiency will bring the farmer into head-on collision with those
losses to jaguars that are unavoidable, and society will need to decide who is to bear these
costs.
Recently, there has been an effort in the Pantanal to alleviate jaguar-livestock conflict in the
form of a compensation program (Silveira et al. 2006). Such programs have been explored
worldwide (Saberwal et al. 1994, Wagner et al. 1997, Vos 2000, Naughton-Treves et al. 2003,
Swenson and Andrén 2005) but their effectiveness is debated (Bulte and Rondeau 2005,
Nyhus et al. 2003, 2005; Maclennan et al., submitted). Unverifiable losses, fraudulent claims,
overly bureaucratic procedures and associated time lags in payment, payments below market
values, lack of sustainable funding, high administrative costs, and moral hazard are some of
the drawbacks associated with compensation programs (Bulte and Rondeau 2005, Nyhus et
al. 2005, Zabel and Holm-Müller 2008). Ideally, such schemes would be closely monitored,
but in the Panatanal this is challenging because retaliatory, illegal killing of jaguars is often
clandestine.
An alternative to compensation involves „performance payments‟ (Nyhus et al. 2005, Zabel

and Holm-Müller 2008). By analogy with agri-environment schemes elsewhere, payments
would be conditional on some measure of effective jaguar conservation in an area (Ferraro
and Kiss 2002, Zabel and Holm-Müller 2008). As with all environmental payments (and
219
compensation schemes) it would be essential to have effective monitoring, robust regulation

and care to avoid unintended consequences.
However, the results of our studies demonstrated that the problem goes beyond the
economics and into the realms of culture - depredation on stock and retributive killing turn
out to be more loosely linked than is often supposed. But although prejudices against jaguars
are deeply ingrained within the culture of cattle ranching, attitudes can change over
generations. Wolves were eradicated from the Rocky Mountain region of the U.S. by the
1930‟s, but are now making a dramatic comeback after reintroduction efforts in 1995. It may
have taken decades, but policies towards wolves slowly changed over time as ecological
studies and social attitudes reflected an increasing appreciation for the role that top predators
play in ecosystem dynamics. In the case of the Pantanal, given that cowboys are ultimately
the ones whose behaviour will directly impact jaguar conservation, one priority would be to
make them stakeholders in jaguar conservation, and this could be a potent ingredient of any
performance-related scheme. Examples from the Amazon and Africa illustrate the potential
of community-based resource management in wildlife conservation (Lewis et al. 1990,
Castello 2004, Frost and Bond 2008). It will require ingenuity to formulate, and then regulate,
a scheme that delivers benefits to both land-owners and local communities from successful
custody of „their‟ jaguars. For example, mechanisms might be sought to channel payments
both to landowners and into wider community benefits (e.g. education, health, economic
development) to encourage, ideally in ways that even fosters, peer pressure against those
acting against jaguars.
Our synthesis reveals that while jaguars do indeed kill livestock in the Pantanal, this is not the
only, nor perhaps even the most important reason, why people kill jaguars. Therefore, in
many cases jaguar conservation may need to be approached in a different way. As described
here in our case studies from the Pantanal and the Amazon, the motivations for killing jaguars
include not only traditions and social rewards, but the fear and misconceptions of the threat
that jaguars pose to humans, the social incentives for persecution, and the economic viability
of ranchers as well. These insights may lead us towards approaches to decrease persecution
that rely on gradual changes in the values, attitudes and social norms concerning jaguars and
jaguar persecution and that are tailored for the specific region. For example, whereas in the
Pantanal communication campaigns to influence the social norms concerning jaguar hunting
may significantly contribute to decrease persecution, in the Amazon education to increase
knowledge and improve perceptions about jaguars‟ threat to human safety might be more
effective. Although the Pantanal is very important for jaguar conservation in the long-term
(Sanderson et al. 2002), it would be unwise to generalize too readily from this particular
situation to other parts of the jaguar‟s range. Nonetheless, conditions in the Pantanal are
similar to those in, for example, the tropical-wet savannahs of the Venezuelan Llanos and the
Bolivian Beni, so there is scope for an international analysis of cross-regional patterns in
jaguar conflict (Zimmermann and Macdonald in prep).
Unquestionably, practical conservation must be underpinned by sound science. The

illuminating power of data to allow for informed discussions and dispel misconceptions is
illustrated by the findings we report on jaguar predatory behaviour in the Pantanal. However,
the tensions between people and wildlife are so complicated that while ecological science is
necessary as a foundation for solutions, it is not sufficient to deliver them. Drawing on
methodologies from the social sciences we have shown that the link between jaguar
depredation on cattle and retaliatory persecution is only part of the story. To change peoples‟
220
actions will thus require a more far-reaching involvement that examines and understands their
perceptions and traditions.
Acknowledgements
The Pantanal Jaguar Project was supported by the Wildlife Conservation Society, the U.S.
Department of Agriculture (USDA) - National Wildlife Research Center at Utah State
University, Logan, Utah, the National Scientific and Technological Development Council in
Brazil (CNPq), the Conservation, Food, and Health Foundation, the Mamirauá Civil Society,
and Brazilian Foundation for Sustainable Development. The People and Jaguars Coexistence
Project, in collaboration with the Wildlife Conservation Research Unit (Oxford University),
has been supported by Cristalino Ecological Foundation, Instituto HSBC Solidariedade,
Anglo American Brazil, Rainforest Concern, Cleveland Metroparks Zoo, North of England
Zoological Society (Chester Zoo), Woodland Park Zoo, O Boticário Foundation, Fauna &
Flora International and Kevin Duncan. The global survey of human-jaguar conflicts project
is funded by North of England Zoological Society (Chester Zoo) in collaboration with
Wildlife Conservation Research Unit (Oxford University).
References
Ajzen, I. 1985. From intentions to actions: A theory of planned behavior. In J. Kuhl. & J.
Beckman (Eds.), Action-control: From cognition to behavior (pp. 11-39). Heidelberg:
Springer.
Almeida, A. 1986. A survey and estimate of jaguar populations in some areas of Mato
Grosso. In B. des Clers, ed. Wildlife Management in the Neotropical Moist Forest, pp.
80-89. International Council for the Conservation of Game, Paris and Manaus.
Azevedo, F. C. and Conforti, V. A. 1998. Predation dynamics of wild carnivores on livestock
ranches surrounding Iguaçu National Park: evaluation, impact, and implementation of
preventive methods. Unpublished report. Curitiba, Brazil: Boticario Foundation.
Azevedo F.C.C. and Murray, D.L. 2007a. Spatial organization and food habitats of jaguars
(Panthera onca) in floodplain forest. Biological Conservation, 137: 391-402.
Azevedo, F.C.C. and Murray, D.L. 2007b. Evaluation of potential factors predisposing
livestock to predation by jaguars. The Journal of Wildlife Management 71: 2379-2386.
Bodmer, R. 1990. Responses of ungulates to seasonal inundations in the Amazon floodplain.
J. of Tropical Ecology 6:191-201.
Bohner, G., and M. Wanke. 2002. Attitudes and attitude change. Taylor and Francis, New
York, New York, USA.
Bowns, J. E. 1985. Predation-depredation. Pages 204-215 in J. Roberson and F. G. Lindzey,
editors. Proceedings of the Second Mountain Lion Workshop, Salt Lake City, Utah,
USA.
Bulte, E. H., and D. Rondeau. 2005. Why compensating wildlife damages may be bad for
conservation. Journal of Wildlife Management 69:14-19.
Campos, Z., M. E. Coutinho, and W. Magnusson. 2004. Estivação de jacarés no Pantanal
sul. Comunicado Técnico 39, Embrapa Pantanal, Corumbá, MS. [In portuguese].
Castello, L. 2004. A method to count Pirarucu Arapaima gigas: fishers, assessment, and
management. North American Journal of Fisheries Management 24:379-389.
221
Cat Specialist Group (2002). Panthera onca. In: IUCN 2007. 2007 IUCN Red List of
Threatened Species. <www.iucnredlist.org>.[Acessed 6 March 2008].
Cavalcanti, S.M.C.. 2003. Manejo e controle de danos causados por espécies da fauna. Pp.
203-242 (In: Cullen, L., Jr., Rudran, R. and Valladares-Padua, C. (eds.). Métodos de
Estudo em Biologia sa Conservação e Manejo da Vida Silvestre. Editora UFPR. Curitiba,
PR.
Cavalcanti, S.M.C. and E.M. Gese. Kill rates and predation patterns of jaguars (Panthera
onca) preying on livestock and native prey in the southern Pantanal, Brazil. Journal of
Wildlife Management (in review).
Cavalcanti, S.M.C. and E.M. Gese. Spatial ecology and social interactions of jaguars
(Panthera onca) in the southern Pantanal, Brazil. Journal of Mammalogy (accepted).
Cavalcanti, S.M.C., E.M. Gese and C.M.U. Neale. Jaguar habitat use in the southern
Pantanal, Brazil – Landscape attributes and their influence on predation of livestock and
native prey. (in prep).
Cascudo LC (1972) Dicionário do Folclore Brasileiro. INL, Rio de Janeiro, Brazil. (THIS
GOES BEFORE CAT)
Chavez, A.S., and E.M. Gese. 2005. Food habits of wolves in relation to livestock
depredations in northwestern Minnesota. American Midland Naturalist 154:253-263.
Chavez, A.S., and E.M. Gese. 2006. Landscape use and movements of wolves in relation to
livestock in a Wildland-agriculture matrix. Journal of Wildlife Management 70:1079-
1086.
Chavez, A.S., E.M. Gese, and R.S. Krannich. 2005. Attitudes of rural landowners toward
wolves in northwestern Minnesota. Wildlife Society Bulletin 33:517-527.
Chellam, R., and A.J.T. Johnsingh. 1993. Management of Asiatic lions in the Gir forest,
India. Symposium of the Zoological Society of London 65:409-424.
Conforti, V.A. and F.C.C. de Azevedo. 2003. Local perceptions of jaguars (Panthera onca)
and pumas (Puma concolor) in the Iguacu National Park, south Brazil. Biological
Conservation, 111:215-221
Conover, M.R. 2001. Resolving human-wildlife conflict: the science of wildlife damage
management. Lewis Publishers.
Crawshaw, P.G. Jr. 1987. Top cat in a Brazilian marsh. Animal Kingdom 90:12-19.
Crawshaw, P.G. Jr. 2002. Mortalidad inducida por humanos y conservation de jaguares: el
Pantanal y el Parque Nacional Iguazy en Brazil. Paginas 451-464 em R.A. Medellin, C.
Equihua, C.L.B. Chetkiewicz, P.G. Crawshaw, A. Rabinowitz, K.H. Redford, j.G.
Robinson, E.W. Sanderson e A. Taber (eds). El Jaguar en el Nuevo Milenio. Ediciones
Cientificas Universitarias. Mexico.
Crawshaw, P.G. Jr. and H.B. Quigley. 1984. A ecologia do jaguar ou onça-pintada no
Pantanal. Instituto Brasileiro de Desenvolvimento Florestal (IBDF). Brasília.
Crawshaw, P.G. Jr. and H.B. Quigley. 1991. Jaguar spacing, activity, and habitat use in a
seasonally flooded environment in Brazil. Journal of Zoology 223:357-370
Crawshaw Jr., P. G., and H. B. Quigley. 2002. Hábitos alimentarios del jaguar y el puma en
el Pantanal, Brasil, con implicaciones para su manejo y conservación. Pages 223-235 in
R. A. Medellin, C. Equihua, C. L. B. Chetkiewicz, P. G. Crawshaw Jr., A. Rabinowitz, K.
H. Redford, J. G. Robinson, E. W. Sanderson, and A. B. Taber, editors. El jaguar en el
nuevo milenio. Fondo de Cultura Económica, Universidad Nacional Autónoma de
México, Wildlife Conservation Society, Mexico.
Crawshaw Jr., P.G., J.K. Mahler, C. Indrusiak, S.M.C. Cavalcanti, M.R.P. Leite, and K.
Silvius. 2004. Ecology and conservation of jaguar (Panthera onca) in Iguaçu National
Park, Brazil. In: Silvius, K. M., Bodmer, R. E., and Fragoso, J.M.V. (eds.): People in
222
Nature: Wildlife Conservation in South and Central America. Columbia University Press,
New York, New York, USA. Pp. 286-296.
Cullen, L., Jr., R. E. Bodmer, and C. Valladares Pádua. 2000. Effects of hunting in habitat
fragments of the Atlantic forests, Brazil. Biological Conservation 95:49-56.
Cullen Jr., L., Abreu, K.C., Sana, D., and Nava, A.F.D. 2005. Jaguars as landscape detectives
for the upper Paraná River corridor, Brazil. The Brazilian Journal of Nature Conservation
3:147-161.
Dalponte, J. C. 2002. Dieta del jaguar y depredación de ganado en el norte del Pantanal,
Brasil. El Jaguar en el Nuevo Milenio. (eds R. A. Medellin, C. Equihua, C. L. B.
Chetkiewicz, P. G. Crawshaw, Jr., A. Rabinowitz, K. H. Redford, J. G. Robinson, E. W.
Sanderson, & A. B. Taber), pp. 209-221. Fondo de Cultura Economica, Universidad
Nacional Autónoma de Mexico, Wildlife Conservation Society, Mexico, D.F.
Dutton A. E., Edwards-Jones, G, Strachan, R., Macdonald, D.W. 2008. Ecological and social
challenges to biodiversity conservation on farmland: reconnecting habitats on a
landscape scale. Mammal Review 38:205-219.
Esterhuizen, W. C. N., and P. M. Norton. 1985. The leopard as a problem animal in the
Cape Province, as determined by the permit system. Bontebok 4:9-16.
Farrell, L. E. 1999. The ecology of the puma and the jaguar in the Venezuelan llanos. Ms.
Thesis. University of Florida, Gainesville. 118 pp.
Ferraro, P. J., and A. Kiss. 2002. Direct payments to conserve biodiversity. Science
298:1718-1719.
Fox, J. L., and R. S. Chundawat. 1988. Observations of snow leopard stalking, killing, and
feeding behavior. Mammalia 52:137-140.
Fragoso, J. M. V. 1998. Home range and movement patterns of white-lipped peccary
(Tayassu pecari) herds in the northern Brazilian amazon. Biotropica 30:458-469.
Fragoso, J. M. V. 2004. A long-term study of white-lipped peccary (Tayassu pecari)
population fluctuations in northen amazonia. People in Nature: Wildlife Conservation in
South and central America. (eds. K. M. Silvius, R. E. Bodmer, & J. M. V. Fragoso. pp.
286-296. Columbia University Press, New York.
Frost, P. G., and I. Bond. 2008. The CAMPFIRE programme in Zimbabwe: payments for
wildlife services. Ecological Economics 65:776-787.
Garla, R. 2001. Jaguar (Panthera onca) food habits in Atlantic rain forest of southeastern
Brazil. Biotropica 33(4) 691-696.
Ginsberg, J.R and R. Woodroffe. 1998. Edge effects and the extinction of populations inside
protected areas. Science 26(280): 2126-2128.
Hoogesteijn, R. and Mondolfi, E. 1992. The Jaguar. Armitano Editores, Venezuela.
Hoogesteijn, R., A. Hoogesteijn and E. Mondolfi. 1993. Jaguar predation and conservation:
cattle mortality caused by felines on three ranches in the Venezuelan Llanos. Symposiaof
the Zoological Society of London 65:391-407.
Hoogesteijn, R. 2000. Manual on the problem of depredation caused jaguars and pumas on
cattle ranches. Wildlife Conservation Society, NY.
Hopkins, R. A. 1989. Ecology of the puma in the Diablo Range, California. Dissertation,
University of California, Berkeley, USA.
IUCN/SSC. (1986). Symposium on Jaguar and Wildlife Conservation in Neotropical Moist
Forest. Cat News. IUCN Cat Specialist Group. 5: 11-17.
Jorgenson, J. P. and Redford, K.H. 1993. Humans and big cats as predators in the Neotropics.
Symposia of the Zoological Society of London 65: 367-390.
Keuroghlian, A. 2003. The response of peccaries to seasonal fluctuations in the Pantanal of
Rio Negro. Earthwatch Institute Annual Report, CRI-Pantanal. Pp. 55-59.
223
Kiltie, R.A. and Terborgh, J. 1983. Observation on the behavior of rainforest peccaries in
Peru: why do white-lipped peccaries form herds? Z. Tierpsychol. 62:241-255.
Kruuk, H. 1972. Surplus killing by carnivores. Journal of Zoology 166: 233-244.
Leite, M.R.P., R.L.P. Boulhosa, F. Galvao and L. Cullen Jr. 2002. Conservacion del jaguar en
las areas protegidas del bosque atlantico de la costa de Brasil. in R.A. Medellin, C.
Equihua, C.L.B. Chetkiewicz, P.G. Crawshaw, A. Rabinowitz, K.H. Redford, j.G.
Robinson, E.W. Sanderson e A. Taber (ed.). El Jaguar en el Nuevo Milenio. Ediciones
Lewis, D., G. B. Kaweche, and A. Mwenya. 1990. Wildlife conservation outside protected
areas – lessons from an experiment in Zambia. Conservation Biology 4:171-180.
Linnell, J. D. C., J. Odden, M. E. Smith, R. Aanes, and J. E. Swenson. 1999. Large
carnivores that kill livestock: do problem individuals exist? Wildl. Soc. Bull. 27:698-705.
Lopez-Gonzales, C. A. and Piña, G. L. 2002. Carrion use by jaguars (Panthera onca) in
Sonora, Mexico. Mammalia 66:603-605.
Macdonald DW. 2000. Bartering biodiversity: what are the options? In: Helm D, editor.
Economic policy: objectives, instruments and implementation. Oxford: Oxford University
Press. p 142-171.
Macdonald DW, Sillero-Zubiri C. 2004. Conservation: from theory to practice, without
bluster. In: Macdonald DW, Sillero-Zubiri C, editors. The biology and conservation
of wild canids. Oxford, UK: Oxford University Press. p 353-372.
Machado, A.B.M., C.S. Martins and G.M. Drummong. 2005. Lista da fauna brasileira
ameaçada de extinção. Fundação Biodiversitas. Belo Horizonte, Brazil.
Maclennan, S.D., Groom, R.J., Macdonald, D.W. and L.G. Frank (submitted) Evaluation of a
compensation scheme to bring about pastoralist tolerance of lions. Biological
Conservation
Maffei, L., E. Cuellar and A.J. Noss. 2004. One thousand jaguars in in Bolivia‟s Chaco?
Camera trapping in the Kaa-Iya National Park. Journal of Zoology, 262(3):295-304.
Marchini S (2003) Pantanal: opinião pública local sobre meio ambiente e desenvolvimento.
Wildlife Conservation Society and Mamiraua Institute, Rio de Janeiro, Brazil.
Marchini, S. and D.W. Macdonald. Conflicts between people and jaguars in Brazil: from
perceptions to persecution. (in prep a).
Marchini, S and Macdonald DW. Improving people‟s perceptions of jaguars: the potential
role of ecotourism. (in prep b).
Michalski, F., R.L.P. Boulhosa, A Faria, C.A. Peres. 2006. Human-wildlife conflicts in a
fragmented Amazonian forest landscape: determinants of large felid depredation on
livestock. Animal Conservation 9(2): 179-188
Mourão, GM, Coutinho ME, Mauro RA, Tomás, WM, Magnusson W. 2002. Levantamentos
aéreos de espécies introduzidas no Pantanal: porcos ferais (porco monteiro), gado bovino
e búfalos. Embrapa Pantanal. Corumbá. Brazil.
Naughton-Treves, L., R. Grossberg, and A. Treves. 2003. Paying for tolerance: rural
citizens‟ attitudes toward wolf depredation and compensation. Conservation Biology
17:1500-1511.
Nepstad, D.C ., C.M. Stinkler and O.T. Almeida. 2006. Globalization of the Amazon soy and
beef industries: opportunities for conservation. Conservation Biology 20(6) 1595-1603
Nowell, K. and P. Jackson. 1996. Wild cats, status survey and conservatio action plan. UICN,
Gland, Switzerland.Nyhus, P., H. Fisher, F. Madden, and S. Osofsky. 2003. Taking the
bite out of wildlife damage: the challenges of wildlife compensation schemes.
Conservation Biology 4:37-40.
224
Nyhus, P., S. Osofsky, P. Ferraro, F. Madden, and H. Fisher. 2005. Bearing the costs of
human-wildlife conflict: the challenges of compensation schemes. Pages 107-121 in R.
Woodroffe, S. Thirgood, and A. Rabinowitz, editors. People and wildlife: conflict or
coexistence? Cambridge University Press, Cambridge, UK.
Oakleaf, J.K., C. Mack, and D.L. Murray. 2003. Effects of wolves on livestock calf survival
and movements in central Idaho. Journal of Wildlife Management 67:299-306.
Oliveira, T.G. 2002. Evaluacion del estado de conservacion del jaguar en el este de la
Amazonia y noreste de Brasil. in R.A. Medellin, C. Equihua, C.L.B. Chetkiewicz, P.G.
Crawshaw, A. Rabinowitz, K.H. Redford, j.G. Robinson, E.W. Sanderson e A. Taber
(ed.). El Jaguar en el Nuevo Milenio. Ediciones Cientificas Universitarias. Mexico.
Palmeira, F.B.L., P.G. Crawshaw Jr., C.M. Haddad, K.M.P.M.B Ferraz, L.M. Verdade. 2008.
Cattle depredation by puma (Puma concolor) and jaguar (Panthera onca) in central-
western Brazil. Biological Conservation 141: 118-125.
Peres, C. A. 1996. Population status of white-lipped Tayassu pecari and collared peccaries T.
tajacu in hunted and unhunted Amazonian forests. Biological Conservation, 77(2-3):
115-123.
Perovic, P.G. 2002. Conservación del jaguar en el noroeste de Argentina. El Jaguar en el
Nuevo Milenio. (eds R. A. Medellin, C. Equihua, C. L. B. Chetkiewicz, P. G. Crawshaw,
Jr., A. Rabinowitz, K. H. Redford, J. G. Robinson, E. W. Sanderson, & A. B. Taber), pp.
183-197. Fondo de Cultura Economica, Universidad Nacional Autónoma de Mexico,
Wildlife Conservation Society, Mexico, D.F.
Polisar, J., I. Maxit, D. Scognamillo, L. Farrell, M. Sunquist and J.F. Eisenberg. 2003.
Jaguars, pumas, their prey base, cattle ranching: ecological interpretations of a
management problems. Biological Conservation 109:297-310
PRODES. 2007. Monitoring of the Brazilian Amazon by Satellite. Brazil‟s National Institute
for Space Research. Ministry of Science and Technology.
Quigley, H.B. 1987. Ecology and Conservation of the Jaguar in the Pantanal Region, Mato
Grosso do Sul, Brazil. PhD Dissertation. University of Idaho, USA.
Quigley, H.B. e P.G. Crawshaw Jr., 1992. A conservation plan for the jaguar (Panthera
onca) in the Pantanal region of Brazil. Biological Conservation, 61: 149-157.
Quigley, H.B. and P.G. Crawshaw Jr. 2002. Reproduccion, crecimiento y dispersion del
jaguar en la region del Pantanal de Brasil. Paginas 289-302 em em R.A. Medellin, C.
Equihua, C.L.B. Chetkiewicz, P.G. Crawshaw, A. Rabinowitz, K.H. Redford, JG.
Robinson, E.W. Sanderson e A. Taber (ed.). El Jaguar en el Nuevo Milenio. Ediciones
Rabinowitz, A. 1986. Jaguar predation on domestic livestock in Belize. Wildlife Society
Saberwal, V. K., J. P. Gibbs, R. Chellam, and A. J. T. Johnsingh. 1994. Lion-human conflict
in the Gir forest, India. Cons. Biol. 8:501-507.
Saenz, J.C. and E. Carrillo. 2002. Jaguares depredadores de ganado en Costa Rica. in R.A.
Redford, j.G. Robinson, E.W. Sanderson e A. Taber (ed.). El Jaguar en el Nuevo Milenio.
Ediciones Cientificas Universitarias. Mexico.
Sanderson, E.W., K.H. Redford, C.L.B. Chetkiewicz, R.A. Medellin, A. Rabinowitz, J.G.
Robinson and A. Taber. 2002. Planning to save a species: the jaguar as a model.
Coonservation Biology, 16(1):58-72
Schaller, G. B. 1972. The Serengeti lion – a study of predator-prey relations. The University
of Chicago Press, Chicago, IL, USA.
225
Schaller, G.B. 1979. On the status of jaguar in the Pantanal. Instituto Brasileiro de
Desenvolvimento Sustentável (IBDF). Brasília.
Schaller, G. B. 1983. mammals and their biomass on a Brazilian ranch. Arquivos de
Zoologia, São Paulo 31, 1-36.
Schaller, G.B. and P.G. Crawshaw Jr. 1980. Movement patterns of jaguar. Biotropica,
12:161-168.
Schaller, G.B e J.M.C. Vasconcelos. 1978. Jaguar Predation on Capybara. Zeitschritt fur
Säugetierk, 43:296-301.
Schiaffino, K., L. Malmierca and P.G. Perovic. 2002. Depredacion de cerdos domesticos por
jaguar en un area rural vecina a un parque nacional en el noreste de Argentina. in R.A.
Redford, j.G. Robinson, E.W. Sanderson e A. Taber (ed.). El Jaguar en el Nuevo Milenio.
Ediciones Cientificas Universitarias. Mexico.
Scognamillo, D., I. Maxit, M. Sunquist, and L. Farrell. 2002. Ecología del jaguar y el
problema de la depredación de ganado en un hato de los llanos venezolanos. El Jaguar en
el Nuevo Milenio. (eds R. A. Medellin, C. Equihua, C. L. B. Chetkiewicz, P. G.
Crawshaw, Jr., A. Rabinowitz, K. H. Redford, J. G. Robinson, E. W. Sanderson, & A. B.
Taber), pp. 139-150. Fondo de Cultura Economica, Universidad Nacional Autónoma de
Mexico, Wildlife Conservation Society, Mexico, D.F.
Seidl, A. F., J. S. V. da Silva, and A. S. Moraes. 2001. Cattle ranching and deforestation in
the Brazilian Pantanal. Ecological Economics 36:413-425.
Sillero-Zubiri, C., Sukumar, R., and Treves, T. (2007). Living with conflict: the roots of
conflict and the solutions. In: Key Topics in Conservation. Ed. D.W. Macdonald & K.
Service. Oxford University Press. pp. 253-270
Silver, S.C., L.E.T. Ostro, L.K. Marsh, L. Maffei, A.J. Noss, M.J. Kelly, R.B. Walllace, H.
Gomez and G. Ayala. 2004. The use of camera traps for estimating jaguar Panthera onca
abundance and density using capture/recapture analysis. Oryx, 38(2):148-154.
Silveira, L. and A.T.A. Jacomo. 2002. Conservacion del jaguar en el centro del Cerrado de
Brasil. in R.A. Medellin, C. Equihua, C.L.B. Chetkiewicz, P.G. Crawshaw, A.
Rabinowitz, K.H. Redford, j.G. Robinson, E.W. Sanderson e A. Taber (ed.). El Jaguar en
el Nuevo Milenio. Ediciones Cientificas Universitarias. Mexico. (THIS GOES BEFORE
SILVER)
Silveira, L., A. T. Jácomo, C. K. Kashivakura, M. M. Furtado, E. Freitas, N. M. Tôrres, C.
Ferro, and M. Marcondes. 2006. Conservação da onça-pintada em propriedades rurais
no Pantanal. Summary report on findings to the National Research Center for the
Conservation of Natural Predators (Cenap). [In portuguese.]
Smith, N.J.H. 1976. Spotted cats and the Amazon skin trade. Oryx, 13:362-371.
Soisalo, M.K. and S.M.C. Cavalcanti. 2006. Estimating the density of a jaguar population in
the Brazilian Pantanal using camera traps and captura-recapture sampling in combination
with GPS radio-telemetry. Biological Conservation 128(4): 487-496.
Stander, P.E. 1990. A suggested management strategy for stock-raiding lions in Namibia. S.
Afr. J. Wildl. Res. 20:37-43.
Swarts, F. 2000. The Pantanal in the 21th Century: for the Planet‟s Largest Wetland, an
Uncertain Future. Waterland Research Institute.
Swenson, J. E., and H. Andrén. 2005. A tale of two countries: large carnivore depredation
and compensation schemes in Sweden and Norway. Pages 323-339 in R. Woodroffe, S.
Thirgood, and A. Rabinowitz, editors. People and wildlife: conflict or coexistence?
226
Thornton, P.K., R.L.Kruska, N.Henninger, P.M.Kristjanson, R.S.Reid, F.Atieno, A.N.Odero

and T.Ndegwa. 2002. Mapping poverty and livestockin the developing world.
International Livestock Research Institute, Kenya.
Vos, J. 2000. Food habits and livestock depredation of two Iberian wolf packs (Canis lupus
signatus) in the north of Portugal. Journal of Zoology 251:457-462.
Wagner, K. K., R. H. Schmidt, and M. R. Conover. 1997. Compensation programs for
wildlife damage in North America. Wildlife Society Bulletin 25:312-319.
Wallace, R.B., H. Gomez, G. Ayala and F. Espinoza. 2003. Camera trapping capture
frequencies for jaguar (Panthera onca) in the Tuichi Valley, Bolivia. Mastozoologia
Neotropical 10(1): 133-139.
Wilcox, R.W. (1999) The law of the least effort – cattle ranching and the environment in the
savanna of Mato Grosso, Brazil 1900–1980. Environmental History, 4, 338–368.
Woodroffe, R. & Ginsberg, J.R. (1998) Edge effects and the extinction of populations inside
protected areas. Science, 280, 2126-2128.
Zabel, A., and K. Holm-Müller. 2008. Conservation performance payments for carnivore
conservation in Sweden. Conservation Biology 22:247-251.
Zeller, K. (2007). Jaguars in the New Millennium Data Set Update: The State of the Jaguar
in 2006. Report. Wildlife Conservation Society. USA. 78pp.
Zimmermann, A. & Wilson, S. A Spatial Model of Human-Jaguar Conflicts in Latin
America. In prep.
Zimmermann, A., M.J. Walpole and N. Leader-Williams. 2005. Cattle ranchers' attitudes to
conflicts with jaguar Panthera onca in the Pantanal of Brazil. Oryx, 39: 406-412.
Figures & Tables
Figure 1. Historical and current jaguar distribution range
227
Figure 2. (A) Map of Brazil showing major biomes, the Amazon study site (Alta Floresta)
and the Pantanal (highlighted by the box), and (B) map of the Pantanal showing its
subregions and boundary between Northern and Southern Pantanal. This study was conducted
in the three subregions of Northern Pantanal, namely Cáceres, Poconé and Barão de Melgaço,
and a ranch in Southern Pantanal (“Fazenda Sete”).
228
100
90
80
% of carcasses found
70
60
50
40
30
20
10
0 #4
#2
#3
#2
#3
#4
#5
#6
#7
e
e
e
e
al
al
al
al
al
al
al
al
al
m
m
M
Fe
Fe
Fe
Collared individuals
Livestock
Native species
Figure 3. Distribution of native prey species and livestock killed by collared jaguars,
November 2001 to April 2004, southern Pantanal, Brazil (from Cavalcanti and Gese, in
review).
229
Table 1. Distribution of prey species (n, (% of kills)) detected at kill sites for 10 individual jaguars, November 2001 to April 2004, southern Pantanal,
Brazil (from Cavalcanti and Gese, in review).
Adult Adult Adult Adult Adult Adult Adult Adult Adult Subadult
female female female female male male male male male male
Prey
#1 #2 #3 #4 #1 #2 #3 #4 #5 #6
(n=80) (n=123) (n=22) (n=5) (n=47) (n=36) (n=18) (n=40) (n=36) (n=27)
tapir 0 0 0 0 0 0 0 1 (2.50) 1 (2.78) 0
birdsa 0 1 (0.81) 0 0 1 (2.13) 1 (2.78) 0 0 0 0
calf 30 (37.5) 18 (14.6) 0 3 (60.0) 24 (51.1) 3 (8.33) 2 (11.1) 7 (17.5) 1 (2.78) 6 (22.2)
capybara 4 (5.0) 1 (0.81) 0 0 0 0 0 1 (2.50) 0 3 (11.1)
marsh deer 3 (3.75) 2 (1.63) 1 (4.54) 0 4 (8.51) 1 (2.78) 0 2 (5.0) 0 3 (11.1)
maned wolf 2 (2.50) 1 (0.81) 0 0 0 0 0 0 0 0
land turtle 1 (1.25) 0 0 0 0 0 0 0 0 0
caiman 10 (12.5) 52 (42.3) 9 (40.9) 1 (20.0) 4 (8.51) 8 (22.2) 7 (38.9) 3 (7.50) 8 (22.2) 5 (18.5)
crab eating fox 0 1 (0.81) 0 0 0 0 0 1 (2.50) 1 (2.78) 0
racoon 0 0 1 (4.54) 0 0 1 (2.78) 0 0 1 (2.78) 0
feral hog 3 (3.75) 4 (3.25) 1 (4.54) 0 6 (12.8) 1 (2.78) 1 (5.55) 2 (5.0) 1 (2.78) 0
coati 0 0 1 (4.54) 0 1 (2.13) 1 (2.78) 0 0 0 2 (7.40)
peccaryb 7 (8.75) 23 (18.7) 5 (22.0) 0 4 (8.51) 11 (30.5) 6 (33.3) 14 (35.0) 20 (55.6) 2 (7.40)
anaconda 0 0 0 0 0 0 0 0 0 1 (3.70)
giant anteater 7 (8.75) 2 (1.63) 0 0 1 (2.13) 3 (8.33) 0 0 1 (2.78) 0
lesser anteater 1 (1.25) 1 (0.81) 0 0 0 3 (8.33) 0 2 (5.0) 0 0
armadilloc 2 (2.50) 0 3 (13.6) 0 0 0 0 1 (2.50) 0 0
adult cattle 9 (11.2) 16 (13.0) 1 (4.54) 1 (20.0) 2 (4.25) 2 (5.55) 2 (11.1) 5 (12.5) 0 4 (14.8)
brocket deerd 1 (1.25) 0 0 0 0 1 (2.78) 0 1 (2.50) 2 (5.55) 1 (3.70)
caiman lizard 0 1 (0.81) 0 0 0 0 0 0 0 0
a
Includes an egret (Egretta alba) , a jabiru stork (Jabyru mycteria), and a boat-billd heron (Cochlearius cochlearius).
b
Although collared peccaries (Tayassu tajacu) were present, the vast majority killed by jaguars were white-lipped peccaries.
c
Includes two species of armadillos present in the sudy area, Euphractos sexcinctus (n=4) and Dasypus novencinctus (n=1).
d
Includes both species, Mazama americana and Mazama gouazoubira.
- 230 -
9
2001 2002 2003 2004
8
7
6
5
4
3
2
1
0
Mean # killed per jaguar
nov dec jan feb mar apr may jun jul aug sep oct nov dec jan feb mar apr may jun jul aug sep oct nov dec jan feb mar apr
9
8
7
6
5
4
3
2
1
0
9
8
7
6
5
4
3
2
1
0
Figure 4. Distribution of the mean number of cattle, caiman, and peccary killed per month by
collared jaguars, November 2001 to April 2004, southern Pantanal, Brazil (from
Cavalcanti and Gese, in review).
231
Wet 2001-2002 Dry 2002 Wet 2002-2003 Dry 2003 Wet 2003-2004 Dry 2004
0
Predation rate (# days/kill)
10
15
caiman
domestic
peccary
20
25
30
35
Wet 2001-2002 Dry 2002 Wet 2002-2003 Dry 2003 Wet 2003-2004 Dry 2004
Figure 5. Seasonal variation in jaguar predation rates of caiman, peccary, and domestic cattle,
November 2001 to April 2004, southern Pantanal, Brazil (from Cavalcanti and Gese, in
review).
Table 2. Zero-order correlations among perception, knowledge and attitude scores and
demographic and socio-economic variables in the Pantanal
1 2 3 4 5 6 7 8 9 10 11
1. Age 1
2. Years in school -.50** 1
3. Property size .28 .08 1
4. Knowledge about
-.15 -.28 -.17 1
jaguars and depredation
5. Perception of increase
.05 -.12 -.08 .18 1
in jaguar abundance
6. Perception of decline
.43 -.25 .38 .08 .26 1
in economic situation
7. Perception of impact
-.08 .27 -.11 -.53** .45* .10 1
on human safety
8. Perception of impact
.18 -.49** -.14 .16 .41** .47** .19 1
on livestock
9. Perception of ease of
-.14 .09 -.12 .22 .13 -.11 .13 -.15 1
hunting jaguars
10. Perception of social
acceptability/desirability -.10 .19 -.04 -.25 .39 .29 .07 -.18 -.07 1
of jaguar hunting
11. Attitude to jaguars -.16 .36* .14 -.06 -.33 -.57** -.04 -.61** .04 -.05 1
** Correlation is significant at the 0.01 level (2-tailed)

* Correlation is significant at the 0.05 level (2-tailed)
232
Figure 6. Graphs showing the measurement scales, distribution of average scores values and
differences between Amazon and Pantanal. The box indicates the median, 25% and 75% quartiles and
whiskers are the largest values that are not outliers, while circles mark outliers.
233
Figure 7. Hierarchical cognitive model of jaguar persecution adapted from the Theory of Planned
Behaviour (Ajzen 1985).
234
Appendix VIII – Contemporary views
APPENDIX VIII
CONTEMPORARY VIEWS OF HUMAN-CARNIVORE

CONFLICTS ON WILD RANGELANDS
Alexandra Zimmermann1, 2 Nicholas Baker3, Chloe Inskip1, John D.C.Linnell4, Silvio

Marchini2, John Odden4, Greg Rasmussen2, Adrian Treves5
In: Wild Rangelands: Conserving Wildlife While Maintaining Livestock in Semi-Arid

Ecosystems. Editors J. Du Toit, R. Kock & J Deutsch. Blackwells, UK. 2009.
1)
North of England Zoological Society, Chester Zoo, CH2 1LH, UK
2)
Wildlife Conservation Research Unit, University of Oxford, OX13 5QL, UK
3)
University of Queensland, Brisbane, 4072 Australia
4)
Norwegian Institute for Nature Research, N-7485 Trondheim, Norway
5)
Nelson Institute for Environmental Studies, University of Wisconsin-Madison, Wisconsin,
53706, USA
* corresponding author a.zimmermann@chesterzoo.org
Introduction
Conflicts between wildlife and people pose a challenge of increasing concern to conservation
scientists (Woodroffe et al., 2005). Large carnivores in particular, require spaces and
resources that are increasingly compromised by the human dominance of landscapes that
were once prime habitat for wildlife. Human-carnivore conflicts occur in temperate as well as
tropical regions, in rangelands as well as in forests, wetlands and many other habitats.
Rangelands support carnivores and their prey, as well as people and their livestock, and as a
result have always been prone to human-carnivore encounters and conflicts. For rangelands,
already threatened by conversion to agriculture, desertification, invasive species and other
factors, human-wildlife conflicts add yet another dimension of complexity to the challenge of
their conservation (Sinclair & Schaller, this volume).
The order Carnivora contains 11 families and around 260 species (Macdonald 2001). It is the
terrestrial, large-bodied families in particular which come into conflict with people: the cats
(Felidae), dogs (Canidae), bears (Ursidae) and hyaenas (Hyaenidae). Conflicts with some of
the smaller-bodied species among the raccoons (Procyonidae), weasels (Mustelidae), skunks
(Mephitidae), civets (Viverridae) and mongooses (Herpestidae) are not widely known or
documented, with the exception of the wolverine, which is a predator of sheep and semi-
domestic reindeer in northern Eurasia. Even within the cat, dog, bear and hyaena families
there are some species that are more prone to come into conflict with people than others. For
carnivores, likelihood of conflict appears to be a function of body size and proximity to
human-dominated landscapes.
Several factors appear to predispose carnivores to conflicts with people, most importantly the
depletion of natural prey (where livestock provide a good alternative food resource); habitat
235
disturbance dynamics (e.g. edge effects of protected areas which often increase
animal/human encounter probabilities and create populations sinks) (Woodroffe & Ginsberg,
1998) and livestock management practices (levels of protection implemented, as well as
general condition of livestock) (Hoogesteijn, 2003; Ogada et al., 2003). A predator‟s
individual condition, including health, age and territory has also been correlated with
depredation, but not proven to predispose carnivores to prey on livestock (cf. Rabinowitz
1986, Linnell et al., 1999; Miquelle et al., 1999; Wydeven et al., 2004).
When livestock-raisers experience loss of income from predation on their sheep, goats, cattle,
horses, pigs, poultry, and losses of crops in some cases (e.g. bears); when hunters face
competition over game species, or ordinary villagers live with a fear of being attacked by a
predator, as can occur in predatory, territorial or defensive attacks by carnivores (Conover,
2002) retaliation inevitably ensues. Tolerance of carnivores in human-dominated areas is
influenced by the economic impacts of coexistence as well as the perceptions and attitudes
held by local communities (Woodroffe & Ginsberg, 1998; Sillero-Zubiri & Laurenson 2001;
Hussain, 2003; Treves & Naughton-Treves 2005; Zimmermann et al., 2005,), and many large
carnivore species have suffered significant population declines from persecution by people.
This is not only detrimental to carnivore populations which play a key ecological role in the
regulation of prey species (Treves & Karanth, 2003), but also for the wider challenges of
conservation efforts, such as those for rangelands, which require tolerance, positive attitudes
and participation on the part of the communities involved.
Understanding the complex dynamics that shape the nature of a conflict situation is
challenging in itself; resolving conflicts is even more difficult and requires a balance of
practical solutions, community outreach, and the best available information on both the
ecology of the species involved and the social psychology of the people affected. In this
chapter we aim to illustrate some of the diversity of pressures, responses and dynamics that
shape human-carnivore conflicts. As an introduction to the topic, we present six different
conflict case studies from around the globe to discuss a variety of contexts from economically
advanced as well as developing regions.
Hunter versus predator: Wolves in North America
Thirty-two years ago, the last grey wolves (Canis lupus) in the lower 48 United States were
declared endangered by the federal government. They had been eliminated from all but a tiny
area around Lake Superior because they competed with humans for livestock and game.
Today, the western Great Lakes Region of the USA holds around 3,500 wolves across
human-transformed areas of Minnesota (MN), Wisconsin (WI) and Michigan (MI) (USFWS,
2007). In 2007, the US government proposed removing federal endangered/threatened
species protections from this grey wolf population. The scenario above is not entirely rosy.
Returning management of wolves to range states is raising concerns about renewed
extirpation campaigns. For example, proposals to hunt wolves in WI are alarming wolf
preservationists. Transfer of authority is more than a shift in institutional responsibility for
wolf management, it signals a change in emphasis from restoration to coexistence. Yet
wolves continue to attack domestic animals and the many interest groups‟ definitions of
coexistence differ.
In the western Great Lakes region, wolf predation on domestic animals - mainly beef cattle
and hunting dogs - kept pace with population growth through 2000 (Fritts et al., 1992; Treves
236
et al., 2002). But such incidents may have begun to outpace wolf population increase in the
past seven years. Domestic animal depredation and other real and perceived conflicts
generate opposition to wolf conservation, especially in rural areas with high proportions of
livestock producers and hunters (Naughton-Treves et al., 2003; Treves et al., 2007).
Opponents of wolf recovery may retaliate illicitly. More than half of Wisconsin's adult wolf
mortality is caused by people, much of it apparently intentional (Wydeven et al., 2001). More
commonly, opponents call for stricter control of suspected problem packs and for regulated,
public hunting of wolves.
Neither selective lethal control nor public hunting has a strong record of effectiveness in
preventing future conflicts or eliminating culprits selectively (Treves & Naughton-Treves,
2005). Wolf predation on livestock continues despite ~20 years of legal, lethal control in MN
and several years in WI and MI, as well as continuous illicit killing (Wydeven et al., 2001).
Yet perceptions have changed about problem carnivores.
The traditional view that any carnivore will kill livestock if given the opportunity, has given
way in the face of scientific evidence (Treves et al., 2002; Wydeven et al., 2004). An
alternative explanation is that problem individuals arise spontaneously, so large carnivore
populations will always have some problem individuals. Opinions differ about the genesis of
such problem carnivores. Some maintain that exposure to livestock carcasses, garbage or
other human food sources leads to problem behaviors (Andelt & Gipson, 1979; Jorgensen,
1979). Another conjecture is that non-culprit carnivores are being killed at rates too high to
exert efficient selection against problem individuals, so we see no diminution of conflicts. A
recent review found high error rates in lethal control operations on wolves, bears and coyotes
(Treves & Naughton-Treves, 2005). In sum, for various reasons, we should not expect a
decline in livestock predation by wolves or any other large carnivore in the western Great
Lakes Region, until managers and stakeholders act in a systematic manner to understand and
prevent factors that may underlie conflicts.
Even if livestock predation by wolves could be reduced to a minimum, hunters might still
object in force to the presence of this competitor (Hammill, 2005). Hunters may find less
game in their habitual shooting grounds. Preventing a large carnivore from pursuing its wild
prey is more complex than preventing predation on livestock (Treves & Naughton-Treves,
2005). Hence interventions to reduce conflicts between hunters and wolves will take
ingenuity and changes in human behavior. Some groups address the issue through efforts at
hunter education, while others advocate a change in status of wolves from protected species
to game species. Numerous challenges lie ahead before wolves are subject to regulated,
public hunting but in the Lake Superior Region with its strong, popular tradition of hunting,
this intervention may gather momentum now that states have authority.
Finally, when one discusses the future of wolf recovery, managers often lower their voices
and allude to drastic changes if a wild wolf attacks a human. Other regions' experiences warn
us to prepare for such an attack (Rajpurohit, 1998; Linnell & Bjerke, 2002). It could
precipitate a dramatic change in wolf management policy. Swift justice would be demanded
and the response might not be limited to the suspected culprit or its pack. Pre-empting
disproportional retaliation should be an explicit management objective. In addition to a better
understanding of the genesis and behavior of problem carnivores, we need greater effort and
support for social scientific research aimed at understanding tolerance, anthropogenic
mortality and the events that precipitate political backlashes against large carnivores.
237
Developing a risk assessment protocol: Wild dogs in Africa
Colonialism in Africa ensured that the European, traditionally negative, perception of the
wolf was transferred to the African wild dog (Lycaon pictus); wild dogs were declared
vermin and a bounty put on their heads in both ranch lands and national parks (Bere, 1956;
Childes, 1988). Justification for this was self-defence, preservation of wild game, and
protection of domestic stock, though rarely was any damage verified or quantified. In
Zimbabwe, systematic persecution by ranchers and government occurred from 1904 to 1988,
when it was recognized that wild dogs were at risk of being extirpated and not a real threat to
the cattle industry, and government-sponsored persecution stopped. However, wild dogs
were still regarded by ranchers and general public as ruthless cattle and game killers. Their
resurgence into the cattle ranching regions of Nyamandlovu, Zimbabwe in 1993 revived the
conflict but resulted in better understanding and a structured approach for dealing with the
situation.
The primary objective of the approach to deal with the conflict, was the safety and welfare of
the wild dogs and, given their wide dispersal ability (Fuller et al. 1992) the importance of this
for metapopulation viability (Mills et al. 1998). To understand how to tackle the situation,
and most importantly determine whether the conflict was “real” or “imagined”, field ground-
truthing was used, as questionnaires were deemed unsuitable because ranchers‟ answers
mostly carry a bias (Rasmussen, 1999) Here, apart from verifying the nature and extent of the
conflict, ground-thruthing served to evaluate the ranchers‟ points of view, established
personal contact with aggrieved parties, and demonstrated willingness on behalf of the
conservation side to listen, which is an important step in ameliorating conflict.
The ground truthing then led to a three-part “risk assessment phase”. First, the probability of
the dogs being extirpated was assessed using criteria including: resident status, home range
utilisation, denning status, habituation, and vulnerability to being killed. Second, real risk to
stock was evaluated since, due to prejudice, whilst no stock losses do not equate to zero-
killing of predators, genuine high losses will result in increased killing. Stock risk assessment
evaluated actual losses, based on method of stock management, and the presence of natural
prey, as studies have demonstrated that this lessens stock losses (Rasmussen, 1999, Fritts et
al., 2003). Third, and arguably most important, public relations between ranchers and the
wider community were evaluated as public peer pressure relating to the acceptability of
killing predators can ameliorate ranchers killing predators.
This risk assessment phase led to the first action phase of circulating findings locally and
nationally, to ensure transparency and rationality, as well as promote positive public opinion
toward predators and their plight, thus securing assistance in tackling the situation.
Circulation of information regarding the situation, gained the predator a league of support,
whilst at the same time reduced ranchers‟ support for unnecessary predator control initiatives.
To determine the next phase of the conflict management strategy, the first primary
consideration was whether the dogs were resident in an area as determined by breeding. If
not, there was realistically no action that could be taken apart from maintaining contact and
relations with the land owners. If the dogs were resident, then from a welfare position,
vulnerability of the predator to control measures was an important criterion. For example, in
this study when the dogs were denning they were vulnerable, and when not they were so wide
ranging and behaviourally cryptic as to render the possibility of retaliatory killing slight.
238
When using a „welfare of the predator comes first” approach to conflict management, three
options arise. First, doing nothing: this may allow the dogs more time to disperse and breed,
but will most likely result in extirpation of dogs and loss of public support. Second, humane
culling of dogs: this is not only an ethically thorny tactic that can alienate public support,
particularly in the case of an endangered species, but also presents a minefield of legal
implications. Third, translocation: there can be difficulties in acquiring the large amount of
funding necessary for translocation initiatives and in also finding a relocation site.
Translocation can also be stressful for the dogs, but arguably less stressful and potentially
more productive than allowing them to be shot. In the Zimbabwe case study, as the desired
goal was to maximize the safety and welfare of dogs and their ability to contribute to the wild
population via dispersals, option three was considered the most appropriate. To maximize on
the potential of the pack to disperse, the decision to move the packs was not were taken until
the “11th hour”, when if the dogs were not moved they would have been shot. Two
translocations in Zimbabwe have been deemed successful (Hartwig and Rasmussen 1999)
with long-distance dispersers emanating from the translocation site. Furthermore, well the
dogs bred at all their new locations (Rasmussen unpublished data). Most importantly, a
rancher/conservationist entente cordiale‟ was maintained, so when, as expected, dogs filled
the vacuum caused by the relocation, they were tolerated for three years after their detection
until they were moved. Delayed translocation is therefore a valuable solution that increases
the metapopulation, minimizes conservation costs and assists to maintain or create an
essential „entente cordiale‟ between ranchers and conservationists.
Protecting a ‘pest’: Dingos in Australia
The dingo (Canis lupus dingo) is the largest terrestrial predator in Australia, and has managed
to radiate into virtually all habitats across the country since its introduction from Asia around
4,000-6,000 years ago (Corbett, 1995; Savolainen et al., 2004). It is a major predator of
livestock (the primary cause of conflict with humans), causing millions of dollars of damage
each year, as well as the costs associated with its control (Allen & Sparkes, 2001). This
includes the maintenance of one of the world‟s longest human-made structures, the dingo
barrier fence, at over 5,000km in length, poison baiting, trapping and shooting programs.
Dingoes also carry disease, predate on pets (Dickman & Lunney, 2001; Fleming et al., 2001)
and, in some isolated circumstances, have attacked humans, with two instances of human
death recorded.
The management of dingoes in Australia is a complex and expensive exercise, and is

complicated by the fact that the species is at the same time both a declared pest (requiring
that it must legally be controlled) and a protected native species in conservation lands
(Fleming et al., 2006). Public opinion about the species is similarly divided. The situation is
further complicated by the recent listing of dingoes on the IUCN Red List as Vulnerable, with
the main threatening processes being hybridisation with domestic dogs and lethal control
methods employed by humans (Johnson et al., 2007).
Dingoes are normally generalist in their approach to survival, and are able to exploit most
habitats and food sources. This greatly enhances the potential for conflict with humans as the
number of ways in which they can interact also increases. Many „traditional‟ dingo habitats
have been inhabited by humans, providing competition for space and other resources. There
is also a tendency for human settlement to increase the availability of some resources (such as
food) for dingoes.
239
While human-dingo conflict is increasing dramatically in urban and semi-urban areas in

Australia, large landowners in the rangelands are often the most affected (or at least,
perceived to be) by these conflicts in terms of the personal financial and psychological costs
borne. Financial costs include lost livestock (mainly sheep), opportunity costs of reduced
yield, and control costs. Psychological costs often arise from the trauma of witnessing
predation impacts upon stock and pets, as well as stress arising from financial strain. The
dingo has been persecuted as a result, and in many cases, this persecution has led to an
increase in the impacts that the control measures have tried to address. Increased poison
baiting campaigns have in many cases led to an increase in predation rates on livestock. It is
believed that dispersing individuals not part of stable packs tend to attack livestock more
frequently, and the removal of local stable packs allows these dispersers to invade areas they
would not normally be able to (Allen, 2000; Allen & Sparkes, 2001). A potential secondary
effect of local dingo removal is an increase in numbers of mesopredator and competing native
herbivore (e.g. macropods) species, further impacting on livestock industries through loss of
stock and reduced yield.
For the Australian livestock industry to remain viable there needs to be a change in the way
dingoes are viewed and the focus of control programs. To some extent this is already
happening with many abandoning the traditional sheep industry in favour of cattle, which are
less likely to be impacted by dingoes. Many are also reducing baiting campaigns and using
alternative non-lethal methods such as guard animals to reduce the impacts of dingoes on
their stock. As the benefits of these alternative approaches become more apparent, their
uptake is expected to increase. The question that remains to be answered is whether it is, as
some have suggested (e.g. Daniels & Corbett, 2003), too late to ensure the continued survival
of the dingo as we know it today.
Culture and conflict: Jaguars in South America
As the dingo is to Australia, the jaguar (Panthera onca) is the largest terrestrial predator of
Central and South America. Renowned for its power, the jaguar has always been feared and
hunted, yet they also evoked admiration for its exceptional strength, elegance and beauty,
which explains the strong cultural significance of the species. However, an additional
dimension evolved in the relationship between humans and jaguars when domestic livestock
was introduced to South America (Arnold, 1968) and jaguar predation on cattle resulted in
hostility of ranchers toward jaguars.
Attacks by jaguars on humans are extremely rare, but jaguar predation on cattle has long been
documented in the Brazilian Pantanal (Roosevelt, 1914) and since the late 1970s scientific
studies conducted in that region, and later in the Venezuelan Llanos and other sites, have
contributed to our understanding of jaguar feeding ecology and the relationship of jaguars to
livestock in the rangelands of South America (Mondolfi & Hoogesteijn, 1986; Hoogesteijn &
Mondolfi, 1992; Crawshaw & Quigley, 2002; Dalponte, 2002). More than 85 species have
been recorded in the jaguar‟s diet (Seymour, 1989) from frogs to tapirs, but in some areas
where cattle are ranched on prime jaguar habitat, such as the Pantanal and the Llanos, cattle
can become the most frequent prey species for jaguars (Hoogesteijn & Mondolfi, 1992;
Crawshaw & Quigley, 2002, Dalponte, 2002).
240
Some factors increase the likelihood of cattle depredation, including scarcity of wild prey
(Aranda, 2002, Nunez et al., 2002, Polisar et al., 2003), persecution of jaguars, which can
result in wounded individuals whose injuries make it difficult for them to capture wild prey
(Rabinowitz, 1986), and poor livestock husbandry practices (Schaller, 1983, Schaller &
Crawshaw, 1980). In the Pantanal and Llanos cattle are typically left unattended, often near
or even inside forested areas, which makes them more vulnerable to predation (Crawshaw &
Quigley 2002, Hoogesteijn 2002). This leads to livestock losses that are unrelated to jaguars,
but often blamed on them. Indeed, studies have shown that in the large ranches of the
Pantanal and the Llanos predation by jaguars is a minor cause of cattle mortality compared to
other causes, such as disease, abortion, malnutrition, attacks by vultures on newborn calves,
and puma depredation (Schaller, 1983, Hoogesteijn et al., 1993). The relative economic
damage associated with jaguar predation also varies. However, livestock losses to felids are
generally low and less than 1–3% of total stock per year (Jackson et al., 1994; Farrell, 1999),
reaching 6% in the worst cases (Hoogesteijn et al., 1993).
These findings have led to a number of recommendations for resolving the conflict by
attempting to decrease and compensate for economic damage. Electric fences (Saenz &
Carillo, 2002; Schiaffino et al., 2002; Scognamillo et al., 2002), translocation of "problem
jaguars" (Linnell et al., 1997; Rabinowitz, 1986; Vaughan & Temple, 2002), the introduction
of water buffalos (Hoogesteijn, 2002), and improved cattle husbandry and management
practices (Weber and Rabinowitz, 1996; Crawshaw & Quigley, 2002; Hoogesteijn et al.,
2002) are examples of interventions. Compensation payments for the cattle killed by jaguars
(Nowell & Jackson, 1996; Hoogesteijn et al., 2002; Perovic, 2002; Vaughan & Temple,
2002; Conforti & Azevedo, 2003), the implementation of wildlife-based tourism (Weber &
Rabinowitz, 1996; Dalponte, 2002; Miller, 2002; Conforti & Azevedo, 2003) and trophy
hunting of problem jaguars (Swank & Teer, 1992) have also been tried or proposed.
However, so far the evidence that any of these interventions can effectively reduce hostility
toward jaguars is scarce and conservation efforts focused solely on the ecological and
economic dimensions of human-jaguar conflicts have still not resulted in any noticeable
change throughout much of jaguar range.
Indeed, the above interventions will have little effect if jaguar persecution is found to be
socially and culturally ingrained. Recent research on ranches and farms in Brazil suggest that
rural attitudes to jaguars are not dictated by material loss caused by the animal (Conforti &
Azevedo, 2003, Zimmermann et al., 2005) and negative attitudes to the species are also found
among farmers who do not raise cattle (Marchini, 2006). Therefore, factors other than the
economic also determine negative attitudes, and resulting hostility, to jaguars. Fear,
prejudice, the social significance or simply the excitement of hunting a large predator, among
other socio-cultural phenomena, also explain why people persecute jaguars. It is imperative
then that we broaden the scope of our current approaches to resolve human-jaguar conflicts.
Future research should turn to the human side of the conflict in order to (1) understand
people‟s behaviour toward jaguars and identify the factors that determine hostility toward the
species, (2) develop education and communication interventions to engage landowners in the
effort to resolve the conflict, improve their perceptions of the issue and increase their
tolerance to jaguars, taking advantage of the exceptional socio-cultural significance of the
species, and (3) compare approaches for human-jaguar conflict mitigation in a broader
spectrum of ecological, economic, political, social and cultural circumstances, from large
cattle ranches in the Pantanal and Llanos to small farms in other parts of the continent. Our
coexistence with jaguars in the rangelands of South America depends upon the recognition
241
that the origin – and therefore the resolution – of the conflicts between humans and jaguars
lies on the human side of the conflict.
When people become prey: Tigers in Asia
The tiger (Panthera tigris) has suffered a severe decline in its population distribution over the
last decade (Biswas & Sankar, 2002). Fewer than 5,000 individuals remain in the wild
(Jackson, 1999), inhabiting only 7% of its original range (Sanderson et al 2006). Although
now predominantly associated with tropical and temperate broadleaf forests, tigers also
inhabit grassland and shrubland habitats, such as in the of the Terai-Duar landscape which
includes, for example, the Chitwan and Royal Bardia National Parks (Nepal), and Jim Corbett
National Park (India).
Much like the jaguar, the tiger is both feared and admired across its range, and, conflict with
people has undoubtedly contributed to its population decline. Unlike its Latin American
counterpart, however, tigers also occasionally attack people, a dimension that significantly
complicates this case of human-carnivore conflict. In areas where attacks on people occur,
fear inevitably exacerbates communities‟ animosity towards tigers. The resultant retaliatory
killing of tigers is widespread in Asia and occurs even within protected areas. As most of
Asia‟s protected areas are relatively small (Wikramanayake et al., 1999), tigers often range
into adjacent, human-dominated habitats, where conflicts also arise. Human-tiger conflict is
therefore often most severe in moderately disturbed habitat areas, such as buffer zones around
protected areas (Nyhus & Tilson, 2004a,b).
The scale of human-tiger conflict in rangelands, in particular the number of attacks on

humans per location, varies. Attacks on humans have been reported in both the area
surrounding Bardia National Park and, in Chitwan National Park, where, in 1998-99 eleven
deaths were reported, marking a sudden departure from the Park‟s prior average of 1.3 attacks
per year (McDougal, 1999). Attack rates also vary throughout the rest of the tigers‟ range. In
areas of low human and tiger density, attacks on people are rare (Nowell & Jackson, 1996).
In the Russian Far East, for example Miquelle et al., (1999) report only six unprovoked
attacks on humans since 1970. Conversely, in India‟s Dudhwa National Park, over 200
people were killed between 1978 and 1996 (Nowell & Jackson, 1996), and in Sumatra, over
170 people were killed or injured in tiger attacks over a 20 year period (Nyhus & Tilson,
2004a). The Sundarbans of India and Bangladesh are most notorious for „man-eating‟ tigers,
with an official figure of an average of 24 (and an unofficial estimate of approximately 100)
people killed per year (Reza et al., 2002). Attacks are generally attributed to people entering
reserves to harvest natural resources (Nowell & Jackson, 1996; Reza et al., 2000; Mukherjee,
2003), cultivation of land around reserves (Nowell & Jackson, 1996) or to hunters wounding
tigers (Miquelle et al., 1999).
More often than attacking people, tigers kill livestock and a number studies have quantified
these predation losses. For example: livestock constitutes 0.45–12% of tigers‟ diets in four
Indian protected areas (Biswas & Sankar, 2002; Bagchi et al., 2003; Reddy et al 2004;
Andheria et al., 2007). In the Russian Far East it has been estimated that up to 100 livestock
are killed each year by tigers (Miquelle et al., 2005), while in Sumatra, at least 870 head of
livestock were reportedly killed over a 20 year period (Nyhus & Tilson, 2004a) and across
Indochina livestock losses to tigers are common (Johnson et al., 2006). Livestock depredation
242
tends to be particularly severe in those areas where tigers‟ wild prey base has been
significantly reduced (Johnson et al., 2006; TIGRIS Foundation, 2007)
Various methods have been implemented or suggested in order to reduce human-tiger

conflicts, including compensation schemes (Madhusudan, 2003; TIGRIS Foundation, 2007),
changes to cattle husbandry practices (Johnson et al., 2006), village eco-development (Bagchi
et al., 2003) and the use of face masks worn on the back of the head (Nowell & Jackson,
1996). As with jaguars, few quantitative data are available on the effectiveness of such
mitigation techniques, although observations have been recorded. While in the Sundarbans no
attacks on people wearing face masks have been reported (Mukherjee, 2003), compensation
schemes have generally been found to be inefficient and ineffective (Madhusudan, 2003).
More detailed and up-to-date data describing the frequency, distribution and determinants of
attacks on humans by tigers in rangelands, and in other habitat types, are urgently required if
we are to understand the dynamics of human-tiger conflicts range-wide. The extent of
livestock depredation must also be examined extensively. Monitoring and evaluation of
existing mitigation strategies is required to complement this research, as it is fundamental to
the development of effective mitigation techniques. Emphasis must also be placed on
exploring the applicability and feasibility of new mitigation and land management techniques
for reducing human-tiger conflict. For example, the potential of community-based work and
alternative livelihood schemes to reduce the number of people entering reserves should be
investigated, as should the effectiveness of economic incentive and insurance schemes in
comparison to existing compensation schemes. Buffer zones in particular, are associated with
a high probability of conflict, yet if managed specifically with this in mind they can be used
to reduce human-tiger conflict in a wider area, and the effectiveness of multiple land use
practices in reducing conflict must therefore also be investigated (Nyhus & Tilson, 2004b).
Habitat corridors and meta-population management of tiger habitats is becoming increasingly
important as habitat becomes progressively fragmented (Wikramanayake et al., 2004) and the
maintenance of suitable habitat and wild prey base must remain a priority for the
conservation of tigers in human dominated landscapes (Reza et al., 2000). Finally,
understanding the human, dimension of human-tiger conflict is essential.
Re-emerging conflict: Lynx in Europe

The previous case studies highlight that although conflicts between large carnivores and
people and their livestock are a global phenomenon, the extent of the conflict varies greatly.
In this final case study we discuss what is possibly the greatest depredation conflict, which
occurs in the rangelands of Norway. Norway is a rugged country with the lowest human
population density in Europe and where less than 5% of the land area is suitable for
cultivation. The rest of the area consists of boreal forest (almost entirely used for intensive
commercial forestry) and alpine tundra above the tree line.
With so little farm land it has been the tradition for centuries to graze livestock in the forests
and alpine tundra habitats. In the past, grazing animals were accompanied by shepherds to
protect them from predators. However, by the early 20th century large carnivore populations
had been so reduced (wolves were actually exterminated) that it became possible to adopt a
form of free-grazing where livestock (mainly sheep) were released in early summer and
gathered in autumn. The animals were released into unfenced forest and alpine-tundra habitat
and only received occasional supervision. In the absence of large carnivores, losses for the
243
approximate four-month grazing season were remarkably low, in the region of 1-2% for ewes
and 5% for lambs.
However, since the 1980s the populations of all four large carnivore species in Norway
(brown bear (Ursos arctos), wolf, Eurasian lynx (Lynx lynx) and wolverine (Gulo gulo)) have
begun to recover due to more conservation-orientated legislation. This recovery has been
associated with a dramatic increase in depredation losses of livestock to the large carnivores.
Eurasian lynx are the most widespread of the four large carnivore species and are, together
with wolverine, responsible for the greatest losses of livestock. Lynx have been the subject of
intensive radio-telemetry based studies during the last decade (Scandlynx 2007). Between
1996-2005 lynx numbers have fluctuated between 300 and 500 individuals (the fluctuations
are caused by hunter harvest), and compensation has been paid for between 6,000 and 9,000
sheep, mainly lambs killed each year. It should be pointed out that not all losses are
documented, but a combination of documented losses combined with a wide range of studies
where sheep have been radio-collared to identify the cause of losses has confirmed that these
numbers are realistic. This works out to around 20 sheep per lynx per year. The results of an
ongoing study in which lynx have been radio-collared and intensively monitored have
confirmed that the calculated kill rates are realistic. The project has also shown that virtually
all lynx will kill sheep at some stage, but that adult males and juvenile lynx kill far more
sheep than adult female lynx, and that while the diet of lynx is dominated by wild ungulates,
sheep do constitute a significant proportion of a summer diet (26%), which, given that
unguarded livestock far outnumbered wild ungulates in our study area, was a surprisingly low
proportion. In contrast to depredation by tiger and hunting dog, the data also indicated that
the probability of a lynx killing sheep increased with the density of wild prey in the area.
These results combined to build a picture of a conflict situation where lynx are only
incidentally killing livestock that they encounter while pursuing their wild prey. However,
because unguarded sheep are found at high density throughout the natural habitats exploited
by lynx and their preferred prey, lynx–sheep encounters occur frequently, resulting in
depredation. The data further indicated that this is not due to specific problem individuals –
but that rather all of the lynx at some stage kill livestock.
Therefore, it is not surprising that the only way that lethal control reduces depredation is in
circumstances where it reduces the lynx population. The modern Norwegian “tradition” of
placing free-ranging and unguarded sheep directly into forest habitats where large carnivores
occur is a recipe for maximum conflict. Unfortunately, the high labour costs in Norway make
the restoration of the original shepherding tradition virtually impossible, and even fencing
sheep requires an expensive and radical restructuring of an industry that is already highly
subsidised. The only other alternatives for lynx are to either accept losses or phase out sheep.
However, in areas where other large carnivores such as bears, wolves and wolverines also
occur, the combined depredation pressure on livestock holdings will probably force change.
Comparative studies from France, Switzerland and eastern Europe have shown that confining
sheep on fenced fields or alpine pastures (out of the forest) dramatically reduces depredation
losses per lynx and that depredation becomes more associated with specific individuals.
Adopting mitigation measures such as electric fences, night-time enclosures or livestock
guarding dogs can basically reduce the problem to zero (Odden et al., 2002, 2006; Linnell &
Brøseth, 2003; Herfindal et al., 2005; Moa et al., 2006).
244
Conclusion
Our set of case studies has illustrated that although human-carnivore conflicts occur in a great
range of geographic and socio-economic contexts, on a global scale the challenges and
experiences are remarkably similar. Compromises to the ecological needs of carnivores
predispose them to livestock-killing and the severity of each conflict is inextricably linked to
the attitudes and beliefs of the communities affected. Perceptions of conflict are often formed
by a few memorable or catastrophic events and reflect not only those events experienced
first-hand, but historical events and often the stories from other people and communities
(Naughton-Treves & Treves 2005, Treves 2007). As such, perceptions of conflict severity
may not accurately reflect real conflict severity.
Similarly, the results of scientific research may not provide an entirely accurate picture.
Conflict incidents tend to be randomly distributed within and between communities, and the
few individuals, households, or communities that suffer the most devastating losses may be
masked by the regional or community averages commonly used in scientific literature to
describe losses. Scientific research is also rarely able to capture more extensive geographic
and historic perspectives of a conflict situation due to restricted study areas and short time
frames (Treves, 2007). Consequently when devising mitigation strategies both the „perceived‟
and the „scientific‟ views of conflict must be considered, as when viewed together they
provide a more comprehensive insight into conflict severity.
Globally, conflict mitigation techniques are plentiful, but efforts to evaluate these
systematically are lacking and mechanisms for easy exchanges of lessons-learnt are only
beginning to be established. To manage conflicts with greater success world-wide, we need to
understand better the spatial and ecological dynamics of human/wild interfaces; focus on the
importance of the human dimension of these conflicts, compare mitigation results and tailor
mitigation approaches to the characteristics of individual cases. Although there are species-
by-species differences among these conflicts, the core issues as well as the principles that
have been outlined in this chapter apply widely and are crucial to the successful management
of human-wildlife conflicts on wild rangelands.
References
Allen, L., (2000). Measuring predator control effectiveness: reducing numbers may not
reduce predator impact. In Proceedings of the 19th Vertebrate Pest Conference. San
Diego. (Eds T. P. Salmon & A. C. Crabb) pp. 284-289. (University of California Davis)
Allen, L., & Sparkes, E.C., (2001). The effect of dingo control on sheep and beef cattle in
Queensland. Journal of Applied Ecology 38: 76-87.
Andelt, W.F., & P.S. Gipson., (1979). Domestic turkey losses to radio-tagged coyotes.
Journal of Wildlife Management 43:673-679.
Andheria, A.P., Karanth, K.U., & Kumar, N.S., (2007). Diet and prey profiles of three
sympatric large carnivores in Bandipur Tiger Reserve, India. Journal of Zoology
273(2):169-175
Aranda, M., (2002). Importancia de los pecaries para la conservacion del jaguar en Mexico.
In El Jaguar en el Nuevo Milenio, eds. R.A. Medellin, C. Equihua, C.L.B. Chetkiewicz,
P.G. Crawshaw, A. Rabinowitz, K.H. Redford, J.G. Robinson, E.W. Sanderson & A.
Taber. Mexico: Ediciones Cientificas Universitarias, pp. 101-107.
245
Arnold, O., (1968). The story of cattle ranching. Irving-on-Hudson, NY: Harvey House Inc.
Bagchi, S., Goyal, S.P, & Sankar, K., (2003). Prey abundance and prey selection by tigers
(Panthera tigris) in a semi-arid, dry deciduous forest in western India. Journal of the
Zoological Society of London. 260: 285-290
Bere, R.M., (1956). The African wild dog. Oryx 3: 180-182.
Biswas, S., & Sankar, K., (2002). Prey abundance and food habit of tigers (Panthera tigris
tigris) in Pench National park, Madhya Pradesh, India. Journal of the Zoological Society
of London. 256: 411-420
Childes, S.L., (1988). The Past History Present Status and Distribution of the Hunting Dog
Lycaon-Pictus in Zimbabwe. Biological Conservation 44:301-316.
Conforti, V.A. & Azevedo, F.C.C. de., (2003). Local perceptions of jaguars (Panthera onca)
and pumas (Puma concolor) in the Iguacu National Park, Southern Brazil. Biological
Conservation, 111: 215-221.
Conover, M., (2002). Resolving human-wildlife conflicts: the science of wildlife damage
management. Lewis Publishers, New York.
Corbett, L.K., (1995). 'The dingo in Australia and Asia.' (University of New South Wales
Press: Sydney).
Crawshaw, P.G. Jr. & Quigley H.B., (2002). Habitos alimentarios del jaguar y el puma en el
Pantanal, Brasil, con implicaciones para su manejo y conservacion. In El Jaguar en el
Nuevo Milenio, eds. R.A. Medellin, C. Equihua, C.L.B. Chetkiewicz, P.G. Crawshaw, A.
Rabinowitz, K.H. Redford, J.G. Robinson, E.W. Sanderson & A. Taber. Mexico:
Ediciones Cientificas Universitarias, pp. 223-236.
Dalponte, J.C., (2002). Dieta del jaguar y depredacion de ganado en el norte del Pantanal,
Brasil. In El Jaguar en el Nuevo Milenio, eds. R.A. Medellin, C. Equihua, C.L.B.
Chetkiewicz, P.G. Crawshaw, A. Rabinowitz, K.H. Redford, J.G. Robinson, E.W.
Sanderson & A. Taber. Mexico: Ediciones Cientificas Universitarias, pp 209-222.
Daniels, M.J., & Corbett, L.K., (2003). Redefining introgressed protected mammals: when is
a wildcat a wild cat and a dingo a wild dog? Wildlife Research 30: 213-218.
Dickman, C.R., & Lunney, D., (Eds) (2001). 'A symposium on the Dingo.' (Royal Zoological
Society of New South Wales: Mosman).
Farrell, L.E., (1999). The ecology of the puma and the jaguar in the Venezuelan Llanos.
Master of Science Thesis. Univ. of Florida, Gainesville, FL.
Fleming, P.J.S., Allen, L.R., Lapidge, S.J., Robley, A., Saunders, G.R., & Thomson, P.C.,
(2006). A strategic approach to mitigating the impacts of wild canids: proposed activities
of the Invasive Animals Cooperative Research Centre. Australian Journal of
Experimental Agriculture 46: 753-762.
Fleming, P.J.S., Corbett, L.K., Harden, R., & Thomson, P.C., (2001). 'Managing the impacts
of dingoes and other wild dogs.' (Bureau of Rural Sciences: Canberra).
Fritts, S.H., Stephenson, R.O., Hayes, R.H., & Boitani, L., (2003). Wolves and Humans. In
Wolves: Behavior, Ecology, and Conservation, eds. D.L. Mech & L. Boitani. University
of Chicago Press, Chicago, pp 289-316
Fritts, S.H., Paul, W.J., Mech, L.D., & Scott, D.P., (1992). Trends and management of wolf-
livestock conflicts in Minnesota. US Fish and Wildlife Service, Resource Publication 181,
Washington, DC.
Fuller, T. K., M. G. L. Mills, M. Borner, M. K. Laurenson, and P. W. Kat. 1992. Long
distance dispersal by African wild dogs in East and South Africa. Journal of African
Zoology 106:535-537.
Hammill, J. (2005). Is a hunter backlash likely? The Midwest Wolf Stewards Meeting,
Hinckley, MN.
246
Hartwig, S., and G. S. A. Rasmussen. 1999. Observations on the integration of a captive-

raised African Wild Dog (Lycaon pictus) into a wild-caught pack and their adaptation
to the wilderness. Zoologische Garten 69:324-334.
Herfindal, I., Linnell, J.D.C., Moa, P.F., Odden, J., Austmo, L.B. & Andersen, R. (2005).
Does recreational hunting of lynx reduce depredation losses of domestic sheep. Journal of
Wildlife Management 69(3): 1034-1042.
Hoogesteijn, R., (2002). A Manual on the Problems of Depredation Caused by Jaguars and
Pumas on Cattle Ranches. New York: Wildlife Conservation Society.
Hoogesteijn, R., (2003). Manual on the problem of depredation caused by jaguars and pumas
on cattle ranches. WCS Report. Wildlife Conservation Society, USA.
Hoogesteijn, R. & Mondolfi, E., (1992). El Jaguar, Tigre Americano. Caracas, Venezuela:
Ediciones Armitano.
Hoogesteijn, R., Hoogesteijn, A., & Mondolfi. E., (1993). Jaguar predation and conservation:
cattle mortality caused by felines on three ranches in the Venezuelan Llanos. In Mammals
as Predators, Symposia Zoological Society of London, Oxford, UK: Clarendon Press, pp.
391-407
Hoogesteijn, R., Boede, E.O., & Mondolfi, E., (2002). Observaciones de la depredacion de
bovinos por jaguares en Venezuela. In El Jaguar en el Nuevo Milenio, eds. R.A.
Redford, J.G. Robinson, E.W. Sanderson & A. Taber. Mexico: Ediciones Cientificas
Universitarias, pp. 183-198.
Hussain, S., (2003). The status of the snow leopard in Pakistan and its conflict with local
farmers. Oryx, 37: 26–33.
Jackson, P., (1999). The tiger in human consciousness and its significance in crafting
solutions for tiger conservation. In Riding the tiger: Tiger Conservation in human-
dominated landscapes, eds. J.Seidensticker, S.Christie, & P.Jackson. Cambridge, UK:
Cambridge University Press, pp 50-54.
Jackson, R., Wang, Z.Y., Lu, X.D. & Chen, Y., (1994). Snow leopards in the Qomolangma
Nature Reserve of the Tibet Autonomous Region. In Proceedings of the seventh
international snow leopard symposium, eds. Fox, J.L. & Jizeng, D. Seatle: International
Snow Leopard Trust.
Johnson A, Vongkhamheng C, Hedemark M, Saithongdam T., (2006). Effects of human-
carnivore conflict on tiger (Panthera tigris) and prey populations in Lao PDR. Animal
Conservation 9: 421-30.
Johnson, C.N., Isaac, J. L & Fisher, D.O., (2007). Rarity of a top predator triggers continent-
wide collapse of mammal prey: dingoes and marsupials in Australia. Proceedings of the
Royal Society B: Biological Sciences. 274: 341-346.
Jorgensen, C. J., (1979). Bear-sheep interactions, Targhee National Forest. International
Conference on Bear Research and Management 5:191-200.
Linnell, J.D.C. & Brøseth, H., (2003). Compensation for large carnivore depredation on
domestic sheep in Norway, 1996-2002. Carnivore Damage Prevention News 6: 11-13.
Linnell, J.D.C., & Bjerke, T., (2002). Frykten for ulven. En tverrfaglig utredning. (Fear of
wolves: an interdisciplinary study.) NINA oppdragsmelding 722: 1-110.
Linnell, J.D.C., Odden,J., Smith, M.E., Aanes, R., Swenson, J.E., (1999). Large carnivores
that kill livestock: do "problem individuals" really exist? Wildlife Society Bulletin 27(3):
698-705
Linnell, J.D.C., Aanes, R., Swenson, J.E., & Smith, M.E., (1997). Translocation of carnivores
as a method for managing problem animals: a review. Biodiversity and Conservation, 6:
1245-1257.
247
Macdonald, D. W., (2001). The New Encyclopedia of Mammals. Oxford University Press.
Oxford, UK.
Madhusudan, M.D., (2003). Living amidst large wildlife: Livestock and crop depredation by
large mammals in the interior villages of Bhadra Tiger Reserve, South India.
Environmental Management. 31(4): 466-475
Marchini, S., (2006). Dimensões Humanas dos Conflitos entre Gente e Grandes Felinos na
Fronteira Agrícola da Amazônia: Resultados Preliminares. 7th International Conference
for the Management of Wildlife in Amazonia and Latin America, Ilheus, Brazil.
McDougal, C., (1999). Tiger attacks on people in Nepal. Cat News 30: 9-10
Miller, C.M., (2002). Jaguares, ganado y humanos: un ejemplo de coexistencia pacifica en el
noroeste de Belice. In El Jaguar en el Nuevo Milenio, eds. R.A. Medellin, C. Equihua,
C.L.B. Chetkiewicz, P.G. Crawshaw, A. Rabinowitz, K.H. Redford, J.G. Robinson, E.W.
Sanderson & A. Taber. Mexico: Ediciones Cientificas Universitarias, pp 477-492.
Mills, M. G. L., S. Ellis, R. Woodroffe, D. Macdonald, P. Fletcher, M. Bruford, D. Wildt et
al. 1998, African Wild Dog (Lycaon pictus) Population and Habitat Viability
Assessment. M. G. L. Mills, Ellis, S., Woodroffe, R., Macdonald, D., Fletcher, P.,
Bruford, M., Wildt, D., Seal, U.S., Rasmussen G.S.A., ed. IUCN/SSC Conservation
Breeding Specialist Group Apple Valley, Minnesota.
Miquelle, D., Nikolaev, I., Goodrich, J., Litvinov, B., Smirnov, E., Suvorov, E., (2005).
Searching for the coexistence recipe: a case study of conflicts between people and tigers
in the Russian Far East. In People and Wildlife: Conflict or Coexistence? eds. R.
Woodroffe, S. Thirgood, A. Rabinowitz. Cambridge University Press, UK, pp 305-322
Miquelle, D.G., Merrill, T.W., Dunishenko, Y.M., Smirnov, E.V., Quigley, H.B., Pukinov,
D.G., & Hornocker, M.G. (1999). In Riding the tiger: Tiger Conservation in human-
dominated landscapes, eds. J.Seidensticker, S.Christie, & P.Jackson. Cambridge, UK:
Cambridge University Press; pp 273-295.
Moa, P.F., Herfindal, I., Linnell, J.D.C., Overskaug, K., Kvam, T. & Andersen, R. (2006).
Does the spatiotemporal distribution of livestock influence forage patch selection in
Eurasian lynx? Wildlife Biology 12: 63-70.
Mondolfi, E., & Hoogesteijn, R., (1986). Notes on the biology and status of the jaguar in
Venezuela. In Cats of the world: biology, conservation and management, eds. S.D. Miller
and D.D. Everett. Washington, DC: National Wildlife Federation, pp 85-124.
Mukherjee, S., (2003). Tiger human conflicts in the Sundurban Tiger Reserve, west Bengal,
India. Tigerpaper. 30(2): 3-6
Naughton-Treves, L., and A. Treves. 2005. Socioecological factors shaping local support for
wildlife in Africa. Pages 253-277 in R. Woodroffe, Thirgood, S., Rabinowitz, A.,
editor. People and Wildlife, Conflict or Coexistence? Cambridge University Press,
Cambridge, UK.
Naughton-Treves, L., Grossberg, R. & Treves, A. (2003) Paying for tolerance: rural citizens‟
attitudes toward wolf depredation and compensation. Conservation Biology, 17: 1500–
1511.
Nowell K & Jackson P., (1996). Wild Cats: Status Survey and Conservation Action Plan.
IUCN / SSC Cat Specialist Group, Gland, Switzerland.
Nunez, M., Miller B., & Lindzey, F., (2002). Ecologia del jaguar en la reserva de la biosfera
Chamela-Cuixmala, Jalisco, Mexico. In El Jaguar en el Nuevo Milenio, eds. R.A.
Nyhus PJ, & Tilson R. (2004a.) Characterising human-tiger conflict in Sumatra, Indonesia:
implications for conservation. Oryx 38(1):68-74.
248
Nyhus PJ, & Tilson R., (2004b). Agroforestry, elephants, and tigers: balancing conservation
theory and practice in human-dominated landscapes of Southeast Asia. Agriculture,
Ecosystems and Environment 104:87-97.
Odden, J., Linnell, J.D.C. and Andersen, R., (2006). Diet of Eurasian lynx, Lynx lynx, in the
boreal forest of southeastern Norway: the relative importance of livestock and hares at
low roe deer density. Euro J Wildl Res. 52: 237-244.
Odden, J., Linnell, J.D.C., Moa, P.F., Herfindal, I., Kvam, T. & Andersen, R., (2002). Lynx
depredation on domestic sheep in Norway. Journal of Wildlife Management 66(1): 98-
105.
Ogada, M.O., Woodroffe, R., Oguge, N.O. & Frank, L.G. (2003). Limiting depredation by
African carnivores: the role of livestock husbandry. Conservation Biology, 17, 1521–
1530.
Perovic, P.G., (2002). Conservacion del jaguar en el noreste de Argentina. In El Jaguar en el
Nuevo Milenio, eds. R.A. Medellin, C. Equihua, C.L.B. Chetkiewicz, P.G. Crawshaw, A.
Rabinowitz, K.H. Redford, J.G. Robinson, E.W. Sanderson & A. Taber. Mexico:
Ediciones Cientificas Universitarias, pp. 465-476.
Polisar, J., Maxit, I., Scognamillo, D., (2003). Jaguars, pumas, their prey base and cattle
ranching: ecological interpretation of a management problem. Biological Conservation,
109: 297-310.
Rabinowitz, A.1986. Jaguar predation on domestic livestock in Belize. Wildlife Society
Rajpurohit, K. S., (1998). Child lifting wolves in Hazaribagh, India. Ambio 28:163-166.
Rasmussen, G., (1999). Livestock predation by the painted hunting dog Lycaon pictus in a
cattle ranching region of Zimbabwe: a case study. Biological Conservation .88 (1): 133-
139
Reddy, H.S., Srinivasulu, C., & Rao. K.T., (2004). Prey selection by the Indian tiger
(Panthera tigris tigris) in Nagarjunasagar Srisailam Tiger Reserve, India. Mammalian
Biology 69(6): 384-391
Reza, A.H.M.A., Chowdhury, M., & Santiapillai, C. (2000). Tiger Conservation in
Bangladesh. Tigerpaper. 27(1): 1-5
Reza, A.H.M.A., Feeroz, M.M., & Islam, M.A. (2002). Man-tiger interaction in the
Bangladesh Sundarbans. Bangladesh Journal o Life Science. 14(1&2): 75-82
Roosevelt, T., (1914). Through the Brazilian Wilderness. New York: Charles Scribner‟s
Sons.
Saenz, J.C. & Carrillo, E., (2002). Jaguares depredadores de ganado en Costa Rica. In El
Jaguar en el Nuevo Milenio, eds. R.A. Medellin, C. Equihua, C.L.B. Chetkiewicz, P.G.
Crawshaw, A. Rabinowitz, K.H. Redford, J.G. Robinson, E.W. Sanderson & A. Taber.
Mexico: Ediciones Cientificas Universitarias, pp. 127-138.
Sanderson, E., Forrest, J., Loucks, C., Ginsberg, J., Dinerstein, E., Seidensticker, J.,
Leimgruber, P., Songer, M., Heydlauff, A., O‟Brien, T., Bryja, G., Klenzendorf, S., &
Wikramanayake, E., (2006). Setting Priorities for the Conservation and Recovery of Wild
Tigers: 2005-2015. The Technical Assessment. WCS, WWF, Smithsonian, and NFWF-
STF, New York - Washington, DC.
Savolainen, P., Leitner, T., Wilton, A. N., Matisoo-Smith, E., & Lundeberg, J., (2004). A
detailed picture of the origin of the Australian dingo, obtained from the study of
mitochondrial DNA. PNAS 101, 12387-12390.
Scandlynx (2007). http://scandlynx.nina.no [accessed 01.04.2007]
Schaller, G.B., (1983). Mammals and their biomass on a Brazilian ranch. Arq. Zool. Sao
Paulo 31:1-36.
249
Schaller, G.B. & Crawshaw, P.G., (1980). Movement Patterns of Jaguar. Biotropica, 12:161-
168.
Schiaffino, K., Malmierca, L., & Perovic, P.G., (2002). Depredacion de cerdos domesticos
por jaguar en un area rural vecina a un parque nacional en el noreste de Argentina. In El
Jaguar en el Nuevo Milenio, eds. R.A. Medellin, C. Equihua, C.L.B. Chetkiewicz, P.G.
Scognamillo, D., Maxit, I.E., Sunquist, M., & Farrell, L., (2002). Ecologia del jaguar y el
problema de la depredacion de ganado. In El Jaguar en el Nuevo Milenio, eds. R.A.
Seymour, K.L., (1989). Panthera onca. Mammalian Species 340:1-9.
Sillero-Zubiri, C & Laurenson, M.K., (2001). Interactions between carnivores and local
communities: conflict or coexistence?. In: Carnivore conservation, eds. J.L. Gittleman,
S.M. Funk, D.W. Macdonald & R.K. Wayne. Cambridge University Press, Cambridge.
pp 282–312.
Swank, W. & Teer, J., (1992). A proposed program for sustained jaguar populations. In
Felinos de Venezuela. Caracas, Venezuela: FUDECI. Raúl Clemente Editores, pp. 95-
107.
TIGRIS Foundation (2007). Compensation for livestock predation.
http://www.tigrisfoundation.nl/cms/publish/content/showpage.asp?pageID=40#top
[Accessed 05/02/2007]
Treves, A. (2007). Balancing the needs of people and wildlife: When wildlife damage crops
and prey on livestock. Unpublished report. Land Tenure Centre, University of Wisconsin,
USA.
Treves, A. & Karanth, U., (2003). Human-Carnivore Conflict and Perspectives on Carnivore
Management Worldwide. Conservation Biology. 17(6): 1491-1499.
Treves, A., & Naughton-Treves, L., (2005). Evaluating lethal control in the management of
human-wildlife conflict. In People and Wildlife, Conflict or Coexistence? eds. R.
Woodroffe, S. Thirgood, & A. Rabinowitz. Cambridge University Press, Cambridge, UK,
pp 86-106
Treves, A., Naughton-Treves, L., Schanning, K., & Wydeven, A., (2007). Appendix H2:
Public Opinion of Wolf Management in Wisconsin, 2001-2005. Wisconsin Wolf
Management Plan. Wisconsin Department of Natural Resources, Madison, WI.
Treves, A., Jurewicz, R.R., Naughton-Treves, L., Rose, R.A., Willging, R.C., & Wydeven,
A.P., (2002). Wolf depredation on domestic animals: control and compensation in
Wisconsin, 1976-2000. Wildlife Society Bulletin 30:231-241.
USFWS. (2007). Endangered and Threatened Wildlife and Plants; Final Rule Designating the
Western Great Lakes Populations of Gray Wolves as a Distinct Population Segment;
Removing the Western Great Lakes Distinct Population Segment of the Gray Wolf
From the List of Endangered and Threatened Wildlife. Federal Register 72: 6051-
6103.
Vaughan, C. & Temple, S., (2002). Conservacion del jaguar en Centroamerica. In El Jaguar
en el Nuevo Milenio, eds. R.A. Medellin, C. Equihua, C.L.B. Chetkiewicz, P.G.
Weber, W. & Rabinowitz, A., (1996). A global perspective on large carnivore conservation.
Conservation Biology, 10-1046-1055.
250
Wikramanayake, E., McKnight, M., Dinerstein, E., & Joshi, A., (2004). Designing a
conservation landscape for tigers in human-dominated environments. Conservation
Biology 18(3):839-44.
Wikramanayake, E.D., Dinerstein, E., Robinson, J.G., Karanth, K.U., Rabinowitz. A., Olson,
D., Mathew, T., Hedao, P., Connor, M., Hemley, G., & Bolze, D. (1999). In: Riding the
tiger: Tiger Conservation in human-dominated landscapes, eds. J.Seidensticker,
S.Christie, & P.Jackson. Cambridge, UK: Cambridge University Press; pp 255-272.
Woodroffe, R., Thirgood, S. & Rabinowitz, A. (2005). The impact of human-wildlife
conflicts on human lives and livelihoods. In: People and Wildlife: Conflict and
Coexistence, eds. R. Woodroffe, S. Thirgood & A. Rabinowitz. Cambridge University
Press, Cambridge, UK. pp 13-26
Woodroffe, R. & Ginsberg, J.R. (1998). Edge effects and the extinction of populations inside
protected areas. Science, 280: 2126–2128.
Wydeven, A. (2006). Status of Wolves in Wisconsin. The Midwest Wolf Stewards Meeting,
Watersmeet, MI.
Wydeven, A.P., Treves, A., Brost, B., & Wiedenhoeft, J.E., (2004). Characteristics of wolf
packs in Wisconsin: Identification of traits influencing depredation. In People and
Predators: From Conflict to Coexistence, eds. N. Fascione, A. Delach, & M. E. Smith
Island Press, Washington, D. C., pp 28-50
Wydeven, A.P., Mladenoff, D.J., Sickley, T.A., Kohn, B.E., Thiel, R.P., & Hansen. J.L.,
(2001). Road density as a factor in habitat selection by wolves and other carnivores in the
Great Lakes Region. Endangered Species Update 18:110-114.
Zimmermann, A., Walpole, M.J., & Leader-Williams, N., (2005). Cattle ranchers' attitudes to
conflicts with jaguar Panthera onca in the Pantanal of Brazil. Oryx, 39: 406-412.
Acknowledgements
The authors would like to thank Richard Kock and Claudio Sillero-Zubiri for their reviews of
the manuscript and their valuable comments and assistance.
251
Appendix IX – Other communication materials
APPENDIX IX
OTHER COMMUNICATION MATERIALS
(a) (b) (c)
(d) (e)
(f) (g)
Example of tools used in the communication campaign. (a) logo of research project, (b)
poster of the dramatic perfomance (“Sassá, the jaguar” is the character staring the campaign
toward students), (c) poster of the project, (d) the “Pro-Jaguar Stamp” used in a partnership
with local handcrafters to promote palm seed and fiber handcrafting (a “jaguar-friendly”
economic activity, relative to cattle-ranching and agriculture), (d) key chains with project‟s
logo (f) stickers and (g) cartoon magazine of Sassá, the Jaguar
252

Human Dimensions of The Conflicts Between People and Jaguars (Panthera Onca) in Brazil

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Human Dimensions of The Conflicts Between People and Jaguars (Panthera Onca) in Brazil

Transféré par

Droits d'auteur :

Formats disponibles

HUMAN DIMENSIONS OF THE CONFLICTS BETWEEN

PEOPLE AND JAGUARS (PANTHERA ONCA)

Photo by Arne Sucksdorff

A thesis submitted for the degree of

Wildlife Conservation Research Unit, Department of Zoology

Linacre College, University of Oxford

Hilary Term 2010

Silvio Marchini Doctor of Philosophy

This work is dedicated to

Manoel „Manolo‟ Pacheco (in memoriam),

for his friendship and unconditional support to my work,

and for taking me to the most stunning place on Earth: Amazonia.

CHAPTER 1 AN INTRODUCTION TO THE HUMAN DIMENSIONS OF THE 1

CHAPTER 2 PERCEPTIONS OF JAGUARS AMONG CHILDREN AND 25

Concern and attraction 47

CHAPTER 3 PERCEPTIONS BEHIND THE CONFLICTS BETWEEN 56

CHAPTER 4 FROM PERCEPTIONS TO PERSECUTION: CONFLICTS 91

CHAPTER 5 SCHOOL-BASED APPROACHES FOR IMPROVING 124

CHAPTER 6 GENERAL SUMMARY AND RECOMMENDATIONS FOR 150

LITERATURE CITED 162

APPENDIX I THE JAGUAR IN BOOKS FOR CHILDREN AND 172

APPENDIX II QUESTIONNAIRES AND INTERVIEW SCHEDULE 173

APPENDIX III PHOTOGRAPHS OF DEAD LIVESTOCK 181

APPENDIX IV A GUIDE FOR COEXISTENCE: ACTIVITY BOOKS 183

APPENDIX V ELABORATION EXERCISE 191

APPENDIX VI PEOPLE AND JAGUARS: A GUIDE FOR COEXISTENCE 192

APPENDIX VII JAGUARS, LIVESTOCK AND PEOPLE IN BRAZIL: 204

APPENDIX VIII CONTEMPORARY VIEWS OF HUMAN-CARNIVORE 228

APPENDIX IX OTHER COMMUNICATION MATERIALS 252

aware of the involvement of their children‟s school in the study.

AN INTRODUCTION TO THE HUMAN DIMENSIONS OF

jaguars and, as a result, jaguars are disappearing.

the usual framework of human-wildlife conflicts (Rabinowitz 2005), in which conflict is

understand the underlying causes of this behaviour.

ecological, economic, social and cultural dimensions of a human-carnivore conflict can be

Conflicts between people and jaguars

important for jaguar research and conservation.

Conflicts between people and jaguars in Brazil

158 kg (Hoogesteijn and Mondolfi 1992).

Jaguar impact on people

northern Pantanal (Zimmermann et al. 2005). Sometimes, a major factor predisposing a

distinguish predation by jaguars and pumas.

Impact on personal safety

In contrast to the historical and contemporary record of unprovoked attacks on humans by

documented, unprovoked, fatal attack by a jaguar on a human occurred on 24 June 2008,

baits used to attract jaguars to be photographed by tourists.

Human impact on jaguars

The jaguar is currently considered as near threatened in Brazil (Caso et al 2008).

Figure 1.1. Historical and current jaguar distribution range

Mitigating human-jaguar conflicts by changing jaguar behaviour

night) repellents, translocation of problematic animals, and changes in husbandry practices

by the ranchers in Brazil.

The human dimensions approach to understanding and predicting conflicts

Human dimensions of wildlife

and demographic characteristics of wildlife user-groups, with emphasis placed on

incorporating information on such factors into wildlife management decisions.

Human dimensions of the conflicts between people and jaguars

they are at the same time feared and admired.

would keep the indigenous term jaguar).

provides detailed descriptions of a jaguar attack on cattle and on Indians.

1976) to contemporary movies and soap-operas (e.g. “Amazônia” by Globo Network). In

internet, magazines and TV documentaries.

interviewed in northern Pantanal by Zimmermann et al. (2005), attitudes towards jaguars