Conflict management in wild chimpanzees (Pan troglodytes)
Von der Fakultät für Biowissenschaften, Pharmazie und Psychologie
der Universität Leipzig
genehmigte
DISSERTATION
zur Erlangung des akademischen Grades
doctor rerum naturalium
Dr. rer. nat.
vorgelegt von
Diplom-Biologe Roman Wittig
geboren am 23.07.1968 in
Karlsruhe
Dekan: Prof. Dr. Kurt Eger Gutacher: Prof. Dr. Christophe Boesch Prof. Dr. Peter Kappeler Prof. Dr. Frans B. M. de Waal Tag der Verteidigung: 19.12.2003 Preface
to Fynn and Cathy
ii Preface
This thesis is partly based on the following publications and manuscripts:
1. Wittig, R. M. & Boesch, C. (2003a). Food competition and linear dominance hierarchy among female chimpanzees of the Taï National Park. International Journal of Primatology. 24: 847-867.
2. Wittig, R. M. & Boesch, C. (2003b). 'Decision-making' in conflicts of wild chimpanzees: An extension of the Relational Model. Behavioral Ecology and Sociobiology. 54: 491-504.
3. Wittig, R. M. & Boesch, C. (2003d). The choice of post-conflict interactions in wild chimpanzees (Pan troglodytes). Behaviour. 140: 1527-1559.
4. Wittig, R. M. & Boesch, C. (in review). How to repair relationships in wild chimpanzees (Pan troglodytes). Ethology.
iii Table of contents
TABLE OF CONTENTS
NATURE OF CONFLICTS AND CONFLICT MANAGEMENT IN ANIMALS - AN INTRODUCTION 1
1.1 THE NATURE OF CONFLICTS 2 1.2 CHANGING THE VIEW OF AGGRESSION 2 1.3 COSTS AND BENEFITS OF AGGRESSION 3 1.4 BENEFIT OF CONFLICT MANAGEMENT – COST REDUCTION 5 1.5 CONFLICT MANAGEMENT IN WILD CHIMPANZEES OF THE TAÏ NATIONAL PARK 6
BIOLOGY OF COMMON CHIMPANZEES (PAN TROGLODYTES) 8
2.1 DISTRIBUTION OF CHIMPANZEE 9 2.2 SOCIAL STRUCTURE AND MATING SYSTEM OF CHIMPANZEES 11 2.3 DIET AND ASPECTS OF FEEDING ECOLOGY OF CHIMPANZEES 13 2.4 RECONCILIATION (AND CONFLICT MANAGEMENT) IN CHIMPANZEES – A BRIEF HISTORY 14
GENERAL METHODS OF THE STUDY 16
3.1 DATA COLLECTION 17 3.1.1 STUDY SITE 17 3.1.2 OBSERVATION PROCEDURE 17 3.1.3 ETHOGRAM 18 3.2 OPERATIONAL DEFINITIONS 20 3.2.1 BEHAVIOURS 20 3.2.2 CONFLICT VARIABLES 20 3.2.3 CONTEXT VARIABLES 21 3.2.4 DOMINANCE VARIABLES 21 3.2.5 SOCIAL VARIABLES 22 3.2.6 VARIABLE ADAPTATIONS FOR STATISTICAL COMPARISON 23 3.3 MULTIVARIATE DYADIC APPROACH 23 3.4 STATISTICS 23 3.4.1 MULTIVARIATE ANALYSIS 24 3.4.2 DYADIC PERMUTATION TESTS 25 3.4.3 PROGRAMS 26
FOOD COMPETITION AND LINEAR DOMINANCE HIERARCHY AMONG FEMALES 27
4.1 INTRODUCTION 28 4.1.1 HIERARCHIES IN CHIMPANZEES 28 4.1.2 SOCIO-ECOLOGY OF DOMINANCE RELATIONSHIPS 28 4.1.3 RELATIONSHIPS IN CHIMPANZEES 29 4.2 SPECIFIC METHODS 31 4.2.1 DATA AND TEST CONDITIONS 31 4.2.2 OPERATIONAL DEFINITIONS 31 4.2.3 STATISTICS 32
iv Table of contents
4.3 RESULTS 33 4.3.1 FEMALE HIERARCHY 33 4.3.2 FOOD COMPETITION 34 4.3.3 CORRELATES OF THE LINEAR HIERARCHY 37 4.3.4 SOCIAL RELATIONSHIPS 37 4.4 DISCUSSION 39 4.4.1 SUMMARY OF RESULTS 39 4.4.2 DOMINANCE RELATIONSHIPS 39 4.4.3 CONTEST COMPETITION 40 4.4.4 REASONS FOR LINEARITY 40 4.4.5 COMPARISON AMONG STUDY SITES 41
DECISION-MAKING IN CONFLICTS: EXTENSION OF THE RELATIONAL MODEL 43
5.1 INTRODUCTION 44 5.1.1 THE RELATIONAL MODEL (RM)44 5.1.2 EXTENSION TO THE RELATIONAL MODEL 45 5.1.2.1 Likelihood of winning 45 5.1.2.2 Conflict duration and cost of aggression 46 5.1.2.3 Creation of social costs 46 5.1.3 TESTING THE EXTENDED RELATIONAL MODEL 46 5.2 SPECIFIC METHODS 48 5.2.1 DATA AND TEST CONDITIONS 48 5.2.2 ADDITIONAL OPERATIONAL DEFINITIONS 49 5.2.3 SPECIAL ANALYSIS 49 5.3 RESULTS 51 5.3.1 STEP 1: BENEFIT 51 5.3.1.1 Benefit of the resource 51 5.3.1.2 Likelihood of winning 52 5.3.2 STEP 2: COST OF AGGRESSION 53 5.3.2.1 Conflict intensity 53 5.3.2.2 Conflict duration 54 5.3.3 STEP 3: SOCIAL COSTS 55 5.3.3.1 Creating social costs 55 5.3.3.2 Reducing of social cost 57 5.4 DISCUSSION 59 5.4.1 SUMMARY OF RESULTS 59 5.4.2 GRABBING THE BENEFIT 59 5.4.3 ECONOMIC HANDLING OF THE COSTS OF AGGRESSION 60 5.4.4 MINIMISING SOCIAL COSTS 61 5.4.5 THE EXTENDED RELATIONAL MODEL 62
CHOICE OF POST-CONFLICT INTERACTIONS 64
6.1 INTRODUCTION 65 6.1.1 ADVANTAGES AND DISADVANTAGES OF POST-CONFLICT INTERACTIONS 65 6.1.2 POOL OF PCIS 66 6.1.2.1 Reconciliation 66 6.1.2.2 Consolation 66 6.1.2.3 Redirected aggression 67
v Table of contents
6.1.2.4 Renewed aggression 67 6.1.2.5 Avoidance of interactions (no PCI)68 6.1.3 TESTING THE CHOICE OF PCI 68 6.2 SPECIFIC METHODS 69 6.2.1 DATA AND TEST CONDITIONS 69 6.2.2 ADDITIONAL OPERATIONAL DEFINITIONS 70 6.2.3 SPECIAL ANALYSIS AND STATISTICS 70 6.3 RESULTS 71 6.3.1 ARE PCIS DEPENDENT ON THE CONFLICT?71 6.3.2 THE EFFECT OF CONFLICT DURATION 72 6.3.3 AVOIDANCE OF INTERACTIONS (NO PCI)73 6.3.4 NORMAL INTERACTIONS – BUSINESS AS USUAL 73 6.3.5 CHOOSING A POST-CONFLICT INTERACTION 74 6.3.5.1 Reconciliation 74 6.3.5.2 Offered consolation 77 6.3.5.3 Renewed aggression 77 6.3.5.4 Redirected aggression 77 6.3.5.5 Third party aggression 77 6.4 DISCUSSION 79 6.4.1 SUMMARY OF RESULTS 79 6.4.2 BUSINESS AS USUAL 80 6.4.3 AVOIDING FURTHER INTERACTIONS 81 6.4.4 RECONCILIATION 81 6.4.5 CONSOLATION 83 6.4.6 RENEWED AGGRESSION 83 6.4.7 REDIRECTED AGGRESSION 84 6.4.8 THIRD PARTY AGGRESSION 84 6.4.9 CHOICE OF PCI: GENERALISED RULES 85
HOW ARE RELATIONSHIPS REPAIRED ? 87
7.1 INTRODUCTION 88 7.1.1 WHY REPAIR RELATIONSHIPS?88 7.1.2 WHAT DAMAGES RELATIONSHIPS?88 7.1.3 HOW TO REPAIR THE DAMAGE? VARIATION IN RECONCILIATION 89 7.2 SPECIFIC METHODS 91 7.2.1 DATA AND TEST CONDITIONS 91 7.2.2 ADDITIONAL OPERATIONAL DEFINITIONS 92 7.2.3 MEASURING RECONCILIATION 93 7.3 RESULTS 94 7.3.1 OCCURRENCE OF RECONCILIATION 94 7.3.2 FUNCTION OF RECONCILIATION 94 7.3.3 VARIATION WITHIN LATENCY AND DURATION OF RECONCILIATION 97 7.3.4 VARIATION WITHIN COMPLEXITY OF RECONCILIATION 99 7.3.5 INITIATOR OF RECONCILIATION 99 7.4 DISCUSSION 100 7.4.1 SUMMARY OF RESULTS 100 7.4.2 FUNCTION OF RECONCILIATION 101 7.4.3 THE VARIATION WITHIN RECONCILIATION 101 7.4.4 REPAIRING THE DAMAGE 103
vi Table of contents
OPEN QUESTIONS AND POSSIBLE ANSWERS - A GENERAL DISCUSSION 106
8.1 CONTRIBUTION OF THIS STUDY TO OUR KNOWLEDGE IN CONFLICT MANAGEMENT 107 8.2 AVOIDANCE AND PREVENTION OF CONFLICTS 108 8.3 CONFLICT REGULATION IN A WIDER SOCIAL CONTEXT 110 8.4 EFFECTIVENESS OF THE MECHANISMS OF CONFLICT MANAGEMENT 111 8.5 SIMILARITY OF CONFLICT MANAGEMENT IN HUMANS AND CHIMPANZEES 113 8.6 PROSPECTS 114
SUMMARY 116
9.1 SUMMARY 116 9.2 ZUSAMMENFASSUNG 118
ACKNOWLEDGEMENTS 123
BIBLIOGRAPHY 125
APPENDICES 149
CURRICULUM VITAE 159
DECLARATION OF INDEPENDENCE 160
vii Chapter 11:
Nature of conflicts and conflict management in animals - an introduction
Figure 1.0 Male Marius is sitting on a tree-fall after fighting a competitor over a female in oestrous (Photo by R. Wittig).
1 The present study focuses on social (group-living) animals Chapter 1: Nature of conflicts and conflict management – an introduction
1.1 The nature of conflicts
Living in groups holds many advantages. Social partners may increase protection against predators (Hamilton, 1971; Alexander, 1974), they can provide support in competition (de Waal, 1978), they facilitate food acquisition (Janson, 1988a; Boesch & Boesch, 1989) and they may help in caring for ones’ offspring (O'Brien, 1993; Muroyama, 1994). To increase these advantages individuals build cooperative relationships (Wrangham, 1980), that usually reflect long-term investments and can change over time (Kummer, 1978; Hinde, 1979).
Social partners, however, can also be disadvantageous, as they compete for the same resources or they pursue different, even contrary, goals. An individual’s mating success, food acquisition and spatial patterns are often influenced by the behaviour of the social partners (Hamilton, 1967; Eisenberg et al., 1972; Parker, 1974; Clutton-Brock & Harvey, 1978). When competing for the same resource social partners face a problem: both claim the same resource, but often only one can possess it (e.g. food competition: Hand, 1986). When pursuing contrary goals social partners face another problem in as much that both try to achieve different aims. In order to do so, however, they may need to prevent the other from realising their aim (e.g. weaning conflict: Trivers, 1974, conflict of the sexes: Dawkins, 1976). Such situations are defined as conflicts of interest (McEnery, 1985; van der Dennen & Falger, 1990). A conflict of interest often escalates and results in aggression. Aggressive interactions are then called open conflicts, or for the sake of simplicity, just conflicts.
1.2 Changing the view of aggression
The traditional view of aggression implied a spontaneous, uncontrollable and antisocial nature to conflicts. This idea derived from the concept that aggression is an independent instinct, which is driven by a number of internal and external variables (Lorenz, 1963; Hinde, 1967; Tinbergen, 1968). This concept suggested that the motivation for aggression could increase up to a threshold where aggression would erupt spontaneously. The conditions for opening such a pressure relief valve would be situations in which aggression had not been vented for some time. Lorenz (1963) postulated that
2 Chapter 1: Nature of conflicts and conflict management – an introduction the function of intra-species aggression, in plain terms, is a spacing mechanism. The repulsive reaction of opponents to aggression would produce an even distribution of individuals over the habitat, which would serve the survival of the species by guaranteeing an optimal exploitation of all resources.
Over the last 30 years, however, the view of aggression has changed slowly as it became apparent that aggression seems to be primarily an adaptation to cope with competition (Maynard Smith, 1982; Huntingford & Turner, 1987; Moynihan, 1998). Competition appears to be ubiquitous and inevitable, as an animal’s environment does not consist solely of food supply, sleeping places and shelter from predators, but also of social partners. Therefore, each individual competes with others (e.g. kin, friends, rivals, mating partners, neighbours, other species) over the resources that contribute positively to their inclusive fitness. As resources are limited, mechanisms are required to regulate access to resources. Aggressive interactions, for example, regulate the control over resources. However, besides the advantage of controlling resources, aggression carries disadvantages. One of these drawbacks is that in attacking or fighting individuals run the risk of injury or even death.
Recently the view of aggression has changed again and aggressive behaviour has been recognised as a tool of negotiation in conflicts of interest among group-living animals (de Waal, 1996a; Silk, 1996; Aureli & Smucny, 1998; Aureli et al., 2002). Aggression – besides tolerance and avoidance of the social partner – is seen as one possible strategy to regulate a conflict of interest. Conflicts of interest, therefore, can be settled by sharing the resource among the competitors (tolerance towards the partner: de Waal, 1989a), or by the subordinate partner leaving the resource to the dominant (avoidance of the partner: Janson, 1985). A conflict of interest, however, can also escalate and result in aggression.
1.3 Costs and benefits of aggression
Whether or not a conflict of interest escalates into aggression depends on three variables of a cost-benefit estimation: (1) each competitor’s motivation to fight, which depends on the value the resource has for each of them, (2) each competitor’s likelihood of winning, which means each competitor’s likelihood of accessing the resource, and (3)
3 Chapter 1: Nature of conflicts and conflict management – an introduction the costs of aggression that each competitor is prepared to accept when the conflict of interest escalates into aggression (Maynard Smith, 1982; Huntingford & Turner, 1987; Dunbar, 1988; Preuschoft & van Schaik, 2000).
The value of a resource is related to how much benefit the resource provides for a competitor in terms of fitness increase. The limiting factor for fitness increase appears to be different for each sex. While in males the limiting factor is the accessibility of mates, food access seems to be the limiting factor in females (Crook & Gartlan, 1966; Emlen & Oring, 1977; Wrangham, 1980; Janson, 2000). Direct fighting over these resources is one possible way to access the benefit. However dominance relationships often regulate the access to resources. The dominance rank of a male, for example, may improve his access to fertile females. Many studies in the wild have found a positive correlation between dominance rank and reproductive success for male primates (Pope, 1990; de Ruiter & van Hoof, 1993; Dixson et al., 1993; Ohsawa et al., 1993; Altmann et al., 1996; Gerloff et al., 1999; Constable et al., 2001; Launhardt et al., 2001; Boesch et al., in prep.).
Injury and death are obvious risks of fighting. Injuries can have tremendous costs leading to infection, infirmity and disability, which can have detrimental consequences for food acquisition, travelling and predation risk (Jaeger, 1981; Tuttle & Ryan, 1981; Nishida et al., 1985; Mesnick & Le Boeuf, 1991). Despite injuries and death, there are other costs of conflicts. Aggressive interactions induce acute stress in opponents (Maestripieri et al., 1992). Stress is costly, as it may distract the attention of opponents from beneficial activities and it may be energetically expensive. Furthermore aggression disturbs the relaxed coexistence between opponents and thus interferes negatively with their relationship (Cords & Aureli, 1996). Relationships reflect the history of many interactions among social partners and negative consequences will be mirrored in all of their interactions (Kummer, 1978; Hinde, 1979). Thus, an out-competed cooperative partner may for example withhold support from the aggressor in the future. In short, the basic dilemma of group-living competitors may be that in initiating a fight the aggressor risks losing a cooperative partner. Such costs will be referred to as social costs.
As soon as individuals have several possibilities to resolve a problem situation, such as a conflict of interest, they are viewed as going through a decision-making process (Krebs & Kacelnik, 1991). The term decision does not necessarily imply any conscious reasoning process, but rather that behaviours can switch between options. The causal
4 Chapter 1: Nature of conflicts and conflict management – an introduction mechanisms underlying such switching as well as the functional significance are of main interest for understanding the evolution of decision-making. The theoretical perspective of decision-making is based on the assumption that individuals maximise the net benefit of a conflict of interest and, therefore in the long term, increase their fitness (Hamilton, 1964). Thus the decision can be analysed in terms of costs and benefits of alternative courses of action, while measuring costs and benefits in terms of a more immediate metric, such as access to food, energy expenditure, risk of injury, or number of copulations. When an individual decides to engage in aggression, which is not cost-free, the benefits that are accessible through the conflict must prevail over the costs of conflicts. In order to maximise the net benefit of the outcome of a conflict, opponents could either increase the benefits or decrease the costs of the conflict. Because benefits gained are fixed by the value of the resource, only costs of the conflict are variable. Therefore adaptive strategies are usually recognised when they decrease disadvantages (e.g. dominance relationships: Bernstein, 1981). However another possibility for an adaptive strategy would be finding an alternative way of gaining the advantages (e.g. food sharing: de Waal, 1992).
1.4 Benefit of conflict management – cost reduction
By conflict management one usually understands all actions, strategies and social agreements that serve to reduce the costs of conflicts to one or both opponents (Mason & Mendoza, 1993; Cords & Killen, 1998). Many actions of conflict management are post- conflict interactions, which are interactions of one or both conflict partners after their aggressive interaction. Reconciliation, the friendly post-conflict interaction between former opponents using affiliative behaviour, can repair the disturbance of the relationship of conflict partners in order to allow for future cooperation between them (Cords, 1992; Cords & Thurnheer, 1993), and seems to be able to reduce acute stress (Maestripieri et al., 1992; Aureli, 1997). Consolation, a friendly post-conflict interaction between a conflict partner and a third party using affiliative behaviours, and redirected aggression, when a conflict partner attacks a third party after an aggressive interaction, are both supposed to reduce the stress of conflict partners (Das, 2000; Watts et al., 2000).
In contrast to post-conflict interactions, dominance relationships manage conflicts of interests before any escalation by preventing a conflict of interest from escalating into aggression (Preuschoft & van Schaik, 2000). On the other hand, ritualised aggression
5 Chapter 1: Nature of conflicts and conflict management – an introduction enables aggression to be used in a conflict of interest whilst incurring minimal costs, as ritualised aggression reduces the risk of injuries. Thus strategies that regulate the outbreak or the level of escalation are included in conflict management.
Although conflict management reduces the costs of conflicts it does not necessarily resolve the conflict of interest. Dominance relationships, acceptance of ownership or ritualised aggression, for example, regulate the access of a resource with low or no costs of aggression, but they do not eliminate the incompatibility of the opponents’ goals. On the other hand, sharing resources can end the conflict of interest over sharable resources, since both opponents obtain their goal (reviewed in: Aureli & de Waal, 2000a).
1.5 Conflict management in wild chimpanzees of the Taï National Park
This study was conducted in order to understand the economics behind conflict management and the causes and functions of the decision-making” of wild chimpanzees in the Taï National Park (Pan troglodytes verus). Given the complexity of the topic, many detailed questions are of importance for the understanding of the conflict management, which are mentioned within the result chapters (4, 5, 6 and 7). Chimpanzees are of special interest for the study of conflict management (see chapter 22), since they are our closest living relatives (Ebersberger et al., 2002), and they were the first non-human species where reconciliation was observed (de Waal & van Roosmalen, 1979).
The conflict management of chimpanzees encompasses different mechanisms that act in conflicts of interest at different times. Avoidance of dominant conflict partners or tolerance of conflict partners by sharing are actions that are carried out before the conflict of interest escalates into aggression. This will be referred to as pre-conflict management. In order to understand pre-conflict management, I investigated whether or not chimpanzees dominance relationships are linear and based on contest competition (chapter 4; Wittig & Boesch, 2003a,c), and if the decision to initiate aggression is influenced by the resource benefit or the dominance relationship between opponents (chapter 5; Wittig & Boesch, 2003b). Using ritualised aggression instead of contact aggression or fighting non-friends instead of friends are strategies that can be used
2 see 2.4
6 Chapter 1: Nature of conflicts and conflict management – an introduction during the escalation to aggression – referred to here as peri-conflict management. Thus, in terms of peri-conflict management, I next investigated if opponents use strategies of cost reduction by varying the aggression and type of opponent (chapter 5; Wittig & Boesch, 2003b). Finally, many interactions of conflict management are reactions to costs that have been already created. These interactions are carried out after the conflict and are part of post-conflict management. For the post-conflict management, I examined the economic rules for choosing a post-conflict interaction (chapter 6; Wittig & Boesch, 2003d) and the way to repair the disturbance of usually relaxed relationships after the conflict (chapter 7; Wittig & Boesch, in review).
Conflict management is naturally a system of decision-making. Although triadic relations are also important (de Waal & van Hooff, 1981; Judge, 1991; Das et al., 1997), conflict management mostly contains dyadic interactions. Thus my analyses were based on dyadic interactions. Multivariate analyses were used to detect co-variances and interactions between influencing factors (see chapter 33). This study is the first to my knowledge that has concurrently investigated all aspects of conflict-management in a species. Moreover this study places the patterns of conflict management in an economic framework of costs and benefits. As Pan and Homo are sister-groups, knowing about the conflict management of chimpanzees may help us to understand our own conflict management and aspects of our own evolution (chapter 84).
3 see 3.4.1 4 see 8.5
7 Chapter 25:
Biology of common chimpanzees (Pan troglodytes)
Figure 2.0 Members of the North community in Taï are feeding of and resting around a fallen Palm-tree (Photo by R. Wittig).
5 This chapter provides the reader with information on the biology of chimpanzees that is relevant to this study. Chapter 2: Biology of chimpanzees
2.1 Distribution of chimpanzee
Chimpanzees (Pan troglodytes) are found in the forests and woodlands of tropical Africa (Figure 2.1a). They group in four sub-species (in order from west to east, Figure 2.1b): (a) the West African Chimpanzee (Pan troglodytes verus), ranging in the west of the Niger River, from Senegal to Ghana; (b) the Nigerian Chimpanzee (P.t. vellorosus), found between the rivers Niger and Sanaga in Nigeria and the northern part of Cameroon; (c) the Central African Chimpanzee (P.t. troglodytes), existing in the west of the Congo River between the rivers Sanaga and Ubangi in Gabon, Central African Republic (C.A.R.), Congo and northern part of Democratic Republic of Congo; and (d) the East African Chimpanzee (P.t. schweinfurthii), ranging between the rivers Congo and Ubangi and spread out east beyond the Great Rift Valley, into the triangle of Uganda, the eastern part of Democratic Republic of Congo and Tanzania (Gagneux et al., 2001; Gagneux, 2002). Chimpanzees and Bonobos (Pan paniscus) represent together the genus Pan, which is the phylogenetic sister-group to humans (Homo sapiens) (Kaessmann et al., 2001; Ebersberger et al., 2002).
Our knowledge about the biology of wild chimpanzees originates mainly from seven long-term studies (see Figure 2.1b). The two oldest long-term studies on chimpanzees are located in east Africa with Pan troglodytes schweinfurthii. One is in the Gombe Stream National Park, Tanzania (since 1960: Goodall, 1986) and the other is in the Mahale Mountains National Park, Tanzania (since 1965: Nishida, 1990), both on the shores of Lake Tanganyika. Two further long-term studies were established in the second half of the 1970s in west Africa on Pan troglodytes verus. One is located in Bossou, Guinea (since 1976: Sugiyama, 1984) and the other one in the Taï National Park, Côte d’Ivoire (Figure 2.1c,d; since 1979: Boesch & Boesch-Achermann, 2000; Wittig, 2003). Another three studies also investigate Pan troglodytes schweinfurthii, of which two are placed in the Kibale National Park, Uganda (since 1987: Kanyawara, Wrangham et al., 1991; since 1995: Ngogo, Watts, 1998) and one in the Budongo Forest Reserve, Uganda (since 1994: Newton-Fisher, 1999a). All of these long-term studies are located in eastern or western Africa and represent only two of the four subspecies (P.t. schweinfurthii and P.t. verus). However, recently another two study sites were established with the potential to become long-term studies. One study site is in the Gashaka Forest Reserve, Nigeria (Sommer et al., 2003) on Pan troglodytes vellorosus and the other one in the Goualougo Triangle, Nouabalé-Ndoki National Park, Republic of Congo (Morgan & Sanz, 2003) on Pan
9 Chapter 2: Biology of chimpanzees troglodytes troglodytes (Figure 2.1b). The present study was conducted in the Taï National Park, Côte d’Ivoire.
P.t. verus Senegal P.t. vellerosus P.t. troglodytes Guinea P.t. schweinfurthii Bossou Nigeria Côte Gahshaka Liberia d’Ivoire Ghana Taï C.A.R. Cameroon Goualougo Budongo Uganda Congo Gabon Democratic Kibale Republic of Congo Gombe Tanzania Mahale Africa
(b)
(a)
(5°52’ N, Study Area of 7°22’ W) the Taï Chimpanzee Côte d’Ivoire Taï Project
Taï National Park Yamoussoukro
Taï National Liberia Park Abidjan
(d) (c)
Figure 2.1 Approximate distribution of the four subspecies of chimpanzees (Pan troglodytes) over Africa (a+b), following the suggested distribution by Gagneux (2002), locations of the long-term studies ( ) and the precise location of the Taï Chimpanzee Project (c+d).
10 Chapter 2: Biology of chimpanzees
2.2 Social structure and mating system of chimpanzees
Chimpanzees live in multimale – multifemale communities (=unit-groups: Nishida, 1968; Goodall, 1983). These communities have a fission–fusion grouping pattern, as they travel around in temporary grouping units (= parties) with constantly fluctuating compositions (Sugiyama, 1984; Goodall, 1986; Nishida, 1990; Boesch, 1991; Wrangham et al., 1992). All party members seen together in various parties belong to the community. Communities consist of between 10 (middle community in Taï: Herbinger et al., 2001) and 150 individuals (Ngogo community: Mitani & Watts, 2001) of all age classes. Males are the philopatric sex in chimpanzees. They stay in their natal community, while females transfer permanently to another community usually at the end of adolescence (Nishida & Hiraiwa-Hasegawa, 1985; Watts & Pusey, 1993).
Chimpanzees inhabit a home range of between 9 km2 (Bossou: Sakura, 1994) and 35 km2 (Ngogo: Watts, 2002), with home range size increasing with number of males (Lehmann & Boesch, in press). As only about 10% of the home range area is exclusively used by the resident community, whilst the remaining area is also used by neighbouring communities, chimpanzees can face up to a mean of four inter-community encounters per month (Lehmann & Boesch, in press). The males of the resident community, sometimes supported by the females, defend the their home range by means of drumming displays accompanied by pant-hoot and long bark vocalisations (Crockford & Boesch, 2003). In case of direct contact between the communities, serious fighting can emerge, which can be lethal (Nishida et al., 1985; Fawcett & Muhumuza, 2000).
Chimpanzees have a promiscuous mating system, where females mate with several males and males with several females (Tutin, 1979; Hasegawa & Hiraiwa-Hasegawa, 1983; Matsumoto-Oda, 1999). Female chimpanzees show a sexual swelling of their perineum (Dixson, 1998). This sexual swelling (=oestrus swelling; Figure 2.2) lasts for about 10 days (the time period which I refer to as being in oestrus) during each fertile cycle in female chimpanzees (about 36 days: Wallis, 1982) and indicates for the probability of ovulation (Deschner et al., 2003). After conception female chimpanzees have a gestation period of about 8 months, following a lactation time of about 2 years and a total inter-birth interval of almost 6 years (Furuichi & Hashimoto, 2002; Wrangham, 2002). Only when a younger sibling is born do children stop travelling on their mother’s
11 Chapter 2: Biology of chimpanzees back, however they still stay together with her mother until they are adolescent (another five years).
Figure 2.3 Male Macho and female Venus are feeding on
Figure 2.2 Female Mystère with a full sexual swelling a carcass of a red colobus (Procolobus badius), while (Photo by T. Deschner). females Goma and Belle are watching the carcass (Photo by R. Wittig).
Social relationships play an important role in the social life of chimpanzees. In most communities chimpanzee are probably male bonded, as males are philopatric and seem to have stronger bonds than females (Mitani et al., 2002). Males, for example, build long lasting alliances, that can be sometimes strong enough to perform a coup, overturning the alpha male (Nishida, 1983). In Taï chimpanzees, however, relationships in females can become so strong that a female may adopt the children of another female who dies (Boesch & Boesch-Achermann, 2000). Chimpanzees of both sexes have many benefits to give to or to receive from their partners. Partners can support each other in agonistic situations (e.g. de Waal, 1984; Hemelrijk & Ek, 1991), they can share special foods (e.g. de Waal, 1989a; Mitani & Watts, 2001), or they can groom each other (e.g. Huffmann, 1990; Wrangham et al., 1992; Watts, 2000). Chimpanzees can trade benefits in order to built up or to maintain good relationships. Since chimpanzees can become 40 years and older (Hill et al., 2001), many of their relationships can be long lasting and may influence the structure of the community (e.g. grooming and association: Boesch & Boesch- Achermann, 2000; Watts, 2000).
12 Chapter 2: Biology of chimpanzees
2.3 Diet and aspects of feeding ecology of chimpanzees
Chimpanzees are omnivorous, although most of their diet consists of ripe fruits (Goodall, 1986; Yamakoshi, 1998; Newton-Fisher, 1999b; Boesch & Boesch-Achermann, 2000). Most fruits are eaten in large plentiful patches under and in the trees and therefore are not usually a limited food source. However apart from fruits chimpanzees feed on many other food items. They hunt other mammals, mostly smaller primates, but in some communities also small forest antelopes and bush pigs (Boesch, 1994a; Stanford et al., 1994; Mitani & Watts, 1999; Watts & Mitani, 2002). In Taï, group hunting is common and hunters collaborate by taking different roles6 (Boesch, 1994b). A carcass of a hunted monkey can be characterised as nutritious but monopolisibale (Figure 2.3). Moreover chimpanzees in Taï extract nut kernels from their shells using pieces of wood or stones as hammers and anvils (Boesch & Boesch, 1983; Boesch & Boesch, 1984a,b). Some nut species possess very hard shells and chimpanzees are obliged to use stone hammers to crack them (Figure 2.4). Since large stones are rare in the Taï forest, stone hammers are also considered to be a monopolisable resource, needed to extract nutritious food. Chimpanzees also feed on insects. For some types of insects they use tools, for example when extracting maggots out of dead wood (Figure 2.5) or dipping for ants (Figure 2.6). When dipping for driver ants (Dorylus spec.), the right position at the nest is important to avoid painful ant bites – also a situation of possible monopolisation. Despite fruits, meats, nuts and insects, chimpanzees feed also on mushrooms, leaves, terrestrial herbivorous vegetation and insect products, such as honey. Some of these food items can be rare and monopolisable, but most of them are widely spread and usually common.
Figure 2.4 Female Goma is cracking Panda-nuts (Panda Figure 2.5 Female Belle is extracting maggots from a oleosa) with a stone hammer, while her daughter Giselle dead tree with a stick that she is using like a fork (Photo is feeding on the left-over (Photo by R. Wittig). by R. Wittig).
6 coordination in time, space and function of behaviour
13 Chapter 2: Biology of chimpanzees
2.4 Reconciliation (and conflict management) in chimpanzees – a brief history
With such a social structure of long-term cooperative relationships and competition over food and mating partners, it is not surprising that reconciliation was discovered first in chimpanzees. In 1979 de Waal & van Roosmalen published the first detailed study on unexpected friendly interactions among former opponents after aggression in a group of chimpanzees in Arnhem Zoo, The Netherlands. They found that former opponents were more likely to be partners in the first friendly interaction after a conflict and that the time interval between the aggressive and the first interaction was shorter than between the first and the second interaction. De Waal and van Roosmalen called those first friendly interactions reconciliation and intuitively used the right functional label for these interactions (chapter 77; Cords & Thurnheer, 1993; Wittig & Boesch, in review). Although reconciliation has been studied in many other species after de Waal’s and van Roosmalen’s discovery (Aureli & de Waal, 2000a; Appendix A), it needed more than 10 years before the first comparison was conducted with another chimpanzee group in a zoo (Baker & Smuts, 1994). Since then several other studies have partly re-evaluated the results of de Waal’s and van Roosmalen’s pioneering work with data on other captive chimpanzees (Fuentes et al., 2002; Preuschoft et al., 2002).
Figure 2.6 Male Marius is ant dipping from a nest of driver ants (Dorylus spec.) on the ground (Photo by R. Wittig).
7 see 7.3.2
14 Chapter 2: Biology of chimpanzees
However in wild chimpanzees reconciliation was forgotten for many years. In 1986 (Goodall, 1986) mentioned reconciliation in a short paragraph and stated: “The gestures (note: that is reconciliation) used by the Arnhem chimpanzees were similar to those observed at Gombe (note: her study site), but it was obviously still more important in the captive situation (where the chimpanzees could not escape from one another) that quarrels be speedily settled.” It needed more than 20 years before the first study on reconciliation was published for wild chimpanzees (Arnold & Whiten, 2001). However, only with the present study were we are able to analyse all aspects of conflict management in wild chimpanzees.
15 Chapter 3:
General methods of the study
Figure 3.0 The author, Roman M. Wittig, fully equipped
(hand-held computer, watch, binoculars, pencil and paper) in the forest while observing chimpanzees in a tree (Photo by C.
Crockford). Chapter 3: General methods
3.1 Data collection
3.1.1 Study Site
Data was collected in the study area of the Taï Chimpanzee Project, Taï National Park, Côte d’Ivoire (West Africa, 5°52 N, 7°22 W). The study was conducted between October 1996 and April 1999 on the North-community, which has been observed continuously since 1979 (further information: Boesch & Boesch-Achermann, 2000). In October 1996 the community consisted of four males (three adults, one adolescent), 14 females (11 adults, three adolescents) and 13 juveniles and infants. Almost all individuals, except for the female Ricci and her adolescent son Nino, were fully habituated to humans and were followed from nest to nest. The end of 1997 Nino became fully habituated and was followed for full days. During the observation period five chimpanzees disappeared or died (one adult male, two adolescent females, two juveniles) and six infants were born.
3.1.2 Observation Procedure
I collected the following four types of data during all-day follows of the four males and 10 adult females. (1) I did all-day focal animal sampling (Altmann, 1974) of a target chimpanzee, recording activities, social interactions and vocalisations. (2) I recorded the target’s party composition by scanning the presence of individuals in visibility of the target every 10 minutes. (3) I documented the identity and number of females with genital swellings per day. (4) I noted specific information at each feeding site regarding the foods consumed by the target, food monopolisability and number of competitors present. All these data I entered directly into a Psion Organiser® hand-held computer via The Observer® (Noldus, 1989). Additionally I recorded quickly changing behaviours on a tape recorder and entered this information the same evening into the daily computer file (example for a daily computer file: Appendix B).
I aimed to change the target chimpanzee each day, observing females once and males twice per month. There was however, some variability in individual observation frequency due to the fission-fusion character of chimpanzee societies (chapter 28), death and the habituation level. The result was 80 all-day follows of males (Macho/Marius: 31 days each, Brutus/Nino: 9 days each) and 123 all-day follows of females (between 10 and 15 days per female). 1071 conflicts with complete information were collected. Of these,
17 Chapter 3: General methods
876 conflicts were analysed in chapter 59 (Wittig & Boesch, 2003b) and 610 (Wittig & Boesch, 2003d), as I excluded conflicts involving juveniles and infants. However in chapter 711 (Wittig & Boesch, in review), I analysed only conflicts with the complete information required for testing for reconciliation, which basically left 791 conflicts to analyse.
3.1.3 Ethogram
The ethogram, I used for my study, was a subset of the ethogram of the Taï chimpanzee Project (C. Boesch, unpublished). It included 11 activities of the focal animal, 6 behaviours related to the proximity between partners, 11 social behaviours related to food, 6 submissive social behaviours, 21 aggressive social behaviours, 12 affiliative social behaviours and 16 vocalisations and gestures (Table 3.1). Some of the behaviours had additional modifiers to separate subtypes of the same behaviour (e.g. Table 3.1, No. 62, french kiss vs. directed kiss). Names of food plants were taken from a list including all species consumed by the Taï chimpanzee (D. Anderson, Z. Gone Bi and C. Boesch, unpublished).
Table 3.1 Ethogram of the study on conflict management in Taï chimpanzees. Behaviours have a serial number and are ordered in the 7 categories: activity, proximity, food related social behaviour, submissive social behaviour, aggressive social behaviour, affiliative social behaviour and gestures/vocalisations. A represents the actor (or initiator) and B the reactor (or receiver) in the description of the behaviour. Occasionally either the French or the German name for the behaviour refers to the code, which is presented then in brackets.
No. Name Code Description activity 1 move (déplacement) DEP A is moving on the ground or in the trees 2 eat (manger) MAN A is harvesting food and/or ingesting food, nut-cracking, and/or travelling within feeding site incl. what A is eating fruits, meat, nuts, insects, leaves, mushrooms, THV (terrestrial herbal vegetation), others incl. how many competitors absolute number of competitors incl. monopolisability relative amount and distribution of food source in relation to the number of competitors 3 play (jeux) JEU A plays with social partner or object. Play contains either behavioural elements out of the aggressive range or object manipulation, but is always accompanied by play face. 4 hunt (jagen) JAG A is at the height of monkeys and/or is moving towards monkeys with the “intention” (trying) to catch one 5 wait WAI A interrupts travelling movement when others lag behind or do not follow, and continues to travel once others have appeared or start following 6 rest (repos) and sleep REP A is lying, sitting, standing or sleeping, not involved in other activities 7 groom GRO A grooms B incl. direction indicating whether A grooms B or grooming is reciprocal incl. duration of grooming indicating the end of grooming session 8 nest (nid) NID A constructs a nest to lie in 9 contact CON A rests in physical contact with B 10 handhold HAN A rests holding B’s hand 11 have HAV A possesses X, with X being a food item, a tool, or other object
8 see 2.2 9 see 5.2.1 10 see 6.2.1 11 see 7.2.1
18 Chapter 3: General methods
proximity 12 approach APP A enters at normal speed within 1 m of B 13 hesitant approach HAP A enters at normal speed within 1 m of B, interrupted by pauses 14 fast approach FAP A runs or walks fast directly over to B and enters within 1m of B 15 leave LEA A casually walks out of 1m range of B 16 retreat RET A makes sudden movement away from the vicinity (1m range) of B, but remains in visibility 17 pass PAS A enters B’s 1m range, continues walking and leaves B’s 1m range without changing direction food related social behaviour 18 stare at (fixer) FIX A approaches within 1m of B and stares at food B is possessing 19 beg (mendier) MEN A, who is within 1m of B, lightly touches the food B is holding, or touches B with his hand (when B is eating), or puts his hand, palm up, toward food, or puts cupped hand under the food, or attempts to take food from B 20 take (prendre) PRE A takes food belonging to B 21 take over (récupération) RE* A takes over food or object that B left on the ground incl. type of take over either after B left the spot, or when B is still around incl. type of food or object fruits, meat, nuts, insects, leaves, mushrooms, THV (terrestrial herbal vegetation), others, tools 22 food share (partage) passive PPA B takes bits of food or object which A controls. A neither defends food nor encourages B 23 food share active-passive PAP B takes bits of food from A and A makes a gesture to encourage B 24 food share active PAC A actively gives a piece of food to B 25 present (cadeau) CAD A gives B all or more than half of food source 26 refusal REF A moves food or himself away from B 27 take back TBA A takes back some or all the food taken by B 28 replacement (suppléant) SUP A approaches B who leaves, A takes over feeding/resting site or grooming partner submissive social behaviour 29 present PRS A turns rump to B exposing genital area 30 crouch present CPR A presents to B with flexed legs 31 crouch CRO A faces B with flexed legs 32 bobbing BOB A performs up and down movement by flexing and stretching arms and legs 33 intense greeting IGR A sequence of fast approach, eye contact seeking, bobbing and intense pant-grunting 34 hand reach HRE A stretches arm out to B without actually touching B aggressive social behaviour 35 piloerrection PIL A has erected hair 36 head tipping HTI A makes a slight upward and backward jerk of the head towards B 37 push away PAW A stretches arm out and shoves B in direction away from himself 38 arm wave AWA A is flinging-out his arm towards B, a hitting down, punching-down or sweeping upwards movement 39 throw THR A picks up an object and throws it unaimed (no target) or aimed (hitting a target) 40 shaking branch SBR A takes hold of a branch, twig or sapling and sways it from side to side or forward and backward 41 rocking ROC A rocks from side to side or forward and backward 42 hunch HUN A raises shoulders accompanied by hair erection, can be bipedal, quadrupedal or in sitting position 43 shuffling SHU A walks dragging feet on the ground making the movement noisy 44 bipedal swagger BSW A stands upright, sways from foot to foot, shoulders slightly hunched, arms out 45 slapping or stamping HIT A hits substrate or B with hands or feet 46 bite BIT A puts his mouth to B’s body and bites him (presses his teeth) 47 dragging DRA A pulls object or B over the ground 48 display DIS A shows a variety of conspicuous and exaggerated motor patterns (e.g. bipedal running, stamping ground, waving arms in air) including piloerection 49 short chase SCH A makes an movement of the whole body in the direction of B 50 chase CHA A runs after B who runs away from A 51 charge CHR A rushes towards B without B fleeing 52 attack ATT A suddenly, forcefully and deliberately contacts B with a series of aggressive contacts 53 fight FIG A attacks B while B attacks A 54 support SPP C supports A or B via aggressive behaviours in an aggressive interaction 55 flight FLI A runs away from B affiliative social behaviour 56 bluff over BOV A walks behind B putting his arms over him 57 reach touch TOU A reaches and touches any party of B’s body 58 finger in mouth FIM A puts his finger in B’s mouth 59 touch genitals TGE A touches B’s oestrous or scrotum 60 rump rub RRU A rubs his rump against B’s rump 61 mount MOU A mounts B’s body, A makes pelvic thrusts, but does not ejaculate 62 kiss KIS A presses his lips to B’s body incl. modifier french: kiss mouth to mouth; directed: kiss mouth to body part 63 embrace EMB A stands/squats in front/behind B and puts his arms around B’s body incl. modifier reciprocal, directional 64 under arm UAR A stands beside B and puts one arm around B’s body, while A and B look in the same direction 65 pat PAT A strokes or quickly caresses B’s body 66 copulation COP A approaches B, mounts, achieves penetration, performs pelvic thrusts 67 genital investigation GIN A touches, sniffs or makes a few quick grooming movements on B’s oestrous
19 Chapter 3: General methods
vocalisations, gestures & facial expressions 68 solicited grooming SGR A approaches B, stops in front and turns his side to B, presents an arm or scratches vigorously 69 solicited copulation SCO A presents penis erection and/or pelvic thrusts, bipedal swagger, knuckle knock to B 70 grin (fear) GRI A shows his teeth while maxilla and mandible are closed or almost closed 71 pant-grunt PGR A pants while inhaling and grunts with an open mouth while exhaling 72 bark BAR A produces a short and sharp vocalisation, that sounds like ‘waa’, ‘waaoo’ or ‘aaoo’ to the human ear 73 pant PAN A produces a panting sound whilst in- and exhaling with an open mouth 74 grunt GRU A produces a grunting sound with closed or open mouth 75 greeting-hooh GHO A produces a repeated and intense serious of ‘hoos’ 76 pant-hoot PHO A produces a series of alternating pants and hoots with increasing volume that usually ends in a climax scream 77 scream SCR A produces a loud and shrill vocalisation with an open mouth 78 crying (pleur) PLE A produces a series of short tonal hoos 79 drum (tambourinage) TAM A hits buttress of a tree with hands and/or feet 80 drumming sequence DSE A performs a sequence of behaviours starting with a warm up (rocking and/or leaf clipping in front of a tree), pant-hooting, drumming and climax scream 81 alarm call (cris d’alarme) CDA A produces a series of very loud bark-like vocalisations that sounds like ‘waaoou’ to the human ear 82 temper tantrum TTA A screams or cries and tumbles on the ground 83 smack SMA A makes a smacking sound with the lips while grooming B
3.2 Operational definitions
3.2.1 Behaviours
Aggressive behaviours are threats (e.g. barks, arm wave), non-contact aggression (e.g. displays) and contact aggression (e.g. bites, hits). Affiliative behaviours are friendly behaviours with body contact (e.g. kiss, genital touch, embrace, grooming). Submissive behaviours are submission (e.g. greeting, crouching) and flights (e.g. fleeing, retreating).
3.2.2 Conflict variables
A conflict was defined as a dyadic aggressive interaction that started with the first aggressive behaviour exchanged and ended with either submission, flight, reciprocal screaming, or a non-aggressive behaviour which was not immediately followed by further aggression. The conflict duration was measured from the first exchange of aggressive behaviour until the defined end of the conflict. Aggressive behaviours were classified into five ordinal intensities according to the likelihood of the aggression injuring the partner (Table 3.2). For each conflict the most intense aggressive behaviour used determined the intensity score for that conflict. The cross product of sex of both the initiator and the receiver of aggression are referred to as the sex combination of conflict partners in four categories (initiator-receiver: male-male, m-m; male-female, m-f; female-male, f-m; female-female, f-f).
20 Chapter 3: General methods
3.2.3 Context variables
I observed conflicts within three different contexts. In sex contexts, conflicts are with or over oestrus females either to copulate with them or to prevent others’ copulation attempts. Food conflicts are over food resources. In the social contexts, conflicts show or challenge dominance (e.g. subordinates approaching without greeting followed by aggression) or conflicts arise from competition for social partners. However, it is partly a default option (as conflicts can only occur with other social partners present), which other studies have summarised in an unknown category (e.g. de Waal & Hoekstra, 1980). Two further variables are the possibility to monopolise the resource as well as how many competitors are present. The resource monopolisation is a measure of the relative distribution and aquantity of the resource that is fought over. This is a relative scoring of the number of competitors needed to monopolise the resource. The competitor proportion is the proportion of competitors present in the party compared to all potential competitors for the specific resource.
3.2.4 Dominance variables
Hierarchies in chimpanzees are classically determined by greeting vocalisations (de Waal, 1978). I used pant-grunts among males and additionally greeting-hoo and greeting- pant among females to determine the hierarchy (chapter 412; Wittig & Boesch, 2003a). One rank change occurred in the males during the data collection period due to the death of Brutus, the -male. Since the two lower ranking males rose up in rank accordingly, conflicts between them and the -male had two rank differences each in the data set, depending on whether the conflict occurred before or after the rank change. The relative dominance of the initiator towards the recipient of aggression was scored in the initiator’s rank variable. The winner of the conflict was defined as the one able to access the resource. In food contexts the winner was the one possessing the food after the conflict. In sex contexts the winner was able to assert his or her choice (e.g. a female refused copulation, a male interrupted a copulation of another male). In social contexts the winner was the one who neither showed submission, screams nor flight, or, in cases of competition over the access to a social partners, the one who affiliated with this social partner afterwards. I defined conflicts as a draw when neither conflict partner won. Whether or not the initiator won was reflected in the initiator victory outcome, while
12 see 4.2.1
21 Chapter 3: General methods whether or not the dominant won the conflict was marked by the variable of the winner’s rank.
3.2.5 Social variables
Association is defined as being present in the same party (being within visibility). I used the dyadic association index (DAI) to measure how frequently two individuals were (A B) associated: DAIAB = , where A is the time individual A was seen, B A B (A B) is the time individual B was seen and A+B is the time A and B were seen together (Nishida, 1968). I gave the relationship benefit three categories according to the presence of food-sharing and support events within the dyads (Table 3.2). Dyads of high relationship benefit are called high value partners or friends.
Table 3.2 Main variables with definitions used in the present study on conflict management to describe the conflict situation. Parameters names of the variables are underlined for a better contrast. Some variables are adapted for comparisons (see 3.2.5) and thus same parameters have different ranges for e.g. different sex combinations.
Name Type Variable definitions and parameter scoring Conflict Duration continuous duration from start of first aggressive behaviour to end of the conflict (in seconds) Conflict Intensity ordinal intensity of the conflict scored by the most intensive single aggressive behaviour: 1 = multinomial aggressive vocalisation or gesture; 2 = non-contact aggression, no movement; 3 = non- contact aggression, including movement; 4 = physical contact aggression, one action; 5 = physical contact aggression, many actions Initiator’s Sex binomial sex of the initiator of aggression: m = male aggressor; f = female aggressor Recipient’s Sex binomial sex of the recipient of aggression: m = male aggressed; f = female aggressed Conflict Context multinomial context in which the conflict occurred: sex = conflict about oestrus females; food = conflict about food or possession of tool; social = conflict about hierarchy or social partners and default category Resource Monopolisation ordinal relative scoring of the monopolisability of the resource that is fought over: monopolisable by multinomial one = resource can be monopolised by one competitor; monopolisable by few = resource can be monopolised by some of the competitors; not monopolisable = resource is not monopolisable Competitor Proportion ordinal proportion (prop.) of competitors present in the party in relation to all potential competitors multinomial (in social context: adults in party / all adults; in sex context: males in party / all males; in food context: adults feeding / all adults in party): few for prop.?0.33; some for 0.33
22 Chapter 3: General methods
3.2.6 Variable adaptations for statistical comparison
Some variables are affected by the sex of chimpanzees (e.g. DAI: higher for males than females, Boesch & Boesch-Achermann, 2000; dominance-rank: males dominant over females, Bygott, 1979). For the multivariate analysis, I corrected for sex differences by calculating indices. The indices relate data to average values of sex combinations in order to enable a comparison between the sexes (detailed description: Table 3.2).
3.3 Multivariate dyadic approach
I employed a multivariate approach to detect the factors that affected the variation within conflict management. Table 3.2 displays variables (with definitions and scoring of parameters) that I considered either as dependent variables or as independent variables for variation within conflict management. In addition to the variables of Table 3.2, I tested some chapter specific variables and interactions of variables, which are explained in detail within the methods part of each chapter. Conducting a multivariate analysis I considered all independent variables simultaneously. As many of the variables (e.g. rank difference, association index, relationship benefit) were different for the same individual with different partners, I conducted the analyses on a dyadic level. However, since repeated measurements of individuals can inflate the Type I () error, I controlled for this and ruled out the influence of repeated measurements on variables (see statistical process). The strength of the results was therefore similar to an individual based analysis, but included the advantage that one is able to detect dyadic variability.
3.4 Statistics
Despite the usual parametric and non-parametric statistic, I conducted two less common types of statistical tests throughout almost all four chapters. One type of tests was multivariate analysis in two different versions (for categorical and continuous dependent variables) and the other type included dyadic permutation tests. When a
23 Chapter 3: General methods question required another unusual statistical approach (e.g. chapter 4: linearity tests13), then I have explained it in the methods specific to that chapter.
3.4.1 Multivariate analysis
To detect the variables that affect the variation found within conflict management, I executed multivariate analyses as follows:
1. In order to eliminate repeated measurements of the same conflict type per dyad, I summarised conflicts in one data point for identical cases. Identical conflicts had the same initiator and receiver of aggression, the same value of some chapter specific variables and were identical in all other independent variables, apart from conflict duration and conflict intensity, which were scored with mean and median values respectively. However each specific question of my thesis (each of the chapters 4, 5, 6 and 7) had a different set of independent variables. Thus the previously described balancing process lead for each question to a different final data set for the multivariate analysis, which is referred to in each chapter again.
2. I used a generalisation of the Logistic Regression called General Linear Models (GLM: McCullagh & Nelder, 1989; Agresti, 1996) to examine the functional relationship between the occurrence of one dependent variable and several independent variables. GLMs have been successfully applied in studies on reconciliation and dominance relationships (Call et al., 1999; Côté & Festa-Bianchet, 2001).
(a) Investigating the variation of categorical dependent variables I conducted a Generalised Linear Model (GLZ). For GLZs I chose an ordinal multinomial, or binomial respectively, error distribution and a logit link function (McCullagh & Nelder, 1989). The best model was selected by the best subset method, which is an iterative method based on maximum likelihood estimation (LR: likelihood statistics), and the Akaike’s Information Criterion (AIC), which penalises for the number of independent variables in the model (Akaike, 1973). The significance of the independent variables and their parameters was assessed using Wald statistics for the best model (Dobson, 1990). The estimate-coefficient is an indicator for the strength of the effect that an independent variable-parameter
13 see 4.2.3
24 Chapter 3: General methods
has on the occurrence of the dependent variable. The probability that the tested dependent variable occurred was e more likely with one unit increase of the independent variable, after adjusting for all other variables. This means that, for example, a positive indicates an effect in favour for the first parameter of the dependent variable that enters the model, while a negative indicates the opposite direction. The further away was from 0 the bigger the influence of the independent variable-parameter.
(b) Investigating the variation of continuous dependent variables I conducted a General Linear Model (GLM). The GLM is based on the least square method and can estimate the effect of several variables on one dependent variable. After calculating the all effect model, I calculated the effect size for the significant variables by excluding trivial effects (effect size ? 0.1) from the model (Cohen, 1988). For variable interactions I used Fisher LSD post-hoc test (equivalent to t- test) with Bonferroni correction.
3. Since I was analysing on a dyadic level, I controlled for a possible inflation of the Type I () error due to multiple measurements of the same actor. Therefore I included the identity of the conflict initiator as an additional independent variable and tested the best model again (similar procedure was used by Côté & Festa-Bianchet (2001)). When the significant explanatory independent variables remained significant, one is able to assume that the effect was not due to the replicated observations of the same individual (Bland & Altman, 1995). For the sake of simplicity I have presented only the remaining significant variables of the best models in the results, where p<0.05.
3.4.2 Dyadic permutation tests
1. In order to test differences of variables between different subgroups, I conducted a Permutation Test, which is an extension of the network analysis test by (Dow & de Waal, 1989). I created a matrix of the variable of all individuals and calculated the mean difference between the subgroups I wanted to compare. Afterwards I sampled the data set without replacements and calculated from this sample the mean difference between subgroups again. The procedure was repeated 5000 times and I checked if the original means were found inside or outside the 95% confidence interval of the approximate distribution of mean differences. The tests were
25 Chapter 3: General methods
conducted two-tailed with an significance level of P<0.05 and Bonferroni corrected when multiple tested.
2. In order to compare dyadic interactions with a baseline, I conducted a bootstrap test. I calculated the mean value of a certain interaction for each dyad and then sampled the means of this interaction with replacements (Manly, 1997). This bootstrap procedure was repeated 5000 times to estimate the distribution of the samples. Afterwards I calculated a bias corrected 95% confidence interval, which considers a skewed distribution of samples (Efron & Tibshirani, 1994). Finally I checked whether the baseline (=1) was found within or outside of the 95% confidence interval of the approximate distribution of relative latencies of dyads. This bootstrap test was conducted two-tailed with a significance level of P<0.05. Baselines were different and therefore I refer to them in the specific methods part of the chapters.
3.4.3 Programs
All multivariate analyses (GLZ and GLM) and parametric statistics were performed in STATISTICA© 99 edition (StatSoft, 1999). Common non-parametric statistics were carried out in STATXACT© 5 (CytelSoftware, 2001), using exact statistics for samples of N?15 (Mundry & Fischer, 1998). The bootstrap confidence intervals and the permutation tests were computed in S-Plus (Insightful Corporation, 2001). All tests were performed as two- tailed, apart of one test in chapter 614, which is indicated as one-tailed. In order to uniformly state all the tests statistics (e.g. N = complete sample, n = sample of variable), I used the statistical nomenclature of Vogt (1999), .
14 see 6.3.5.2
26 Chapter 415:
Food competition and linear dominance hierarchy among females
Figure 4.0 Conflict over food: Male Macho and the females Belle and Venus are feeding on the carcass of a red colobus monkey (Procolobus badius). Female Fossey (below) begs for meat by touching the carcass, while Venus screams and raises her arm to hit Fossey (Photo by R. Wittig).
15 Corresponding with: Wittig, R.M. & Boesch, C. (2003a). Food competition and linear dominance hierarchy among female chimpanzees of the Taï National Park. International Journal of Primatology. 24:847-867 Chapter 4: Food competition and linear dominance hierarchy
4.1 Introduction
4.1.1 Hierarchies in chimpanzees
Dominance relationships in many primates fit to a linear hierarchy, though in some species rank orders are unclear or ambiguous (Walters & Seyfarth, 1987). Both seem to be true for chimpanzees. While male chimpanzees rank either in linear hierarchies (Mahale K-group: Nishida, 1979; Mahale M-group: Nishida & Hosaka, 1996; Kibale Ngogo: Watts, 1998; Taï North-group: Boesch & Boesch-Achermann, 2000) or at least in narrow rank categories (Budongo Sonso: Newton-Fisher, 2002; Gombe Kasakela: Bygott, 1979; Goodall, 1986), linear hierarchies in female chimpanzees have not been detected. Dominance relationships among females were either ordered in broad rank categories (Gombe Kasakela: Pusey et al., 1997; Kibale Kanyawara: Wrangham et al., 1992), or at least two-thirds of the dyadic dominance relationships were unknown to researchers (Mahale M-group: Nishida, 1989; Budongo Sonso: Fawcett, 2000). Nevertheless high ranking females in Gombe had significantly higher infant survival, faster maturing daughters and more rapid production of offspring (Pusey et al., 1997). Thus female chimpanzees increase reproductive success in a similar way to male chimpanzees, wherein the alpha male or high ranking males are more successful in siring offspring (Gombe: Constable et al., 2001; Taï: Boesch et al., in prep.). Although the outcome might be similar in both sexes, the reason for the variance in reproductive success seems to be different. While males can monopolise mating partners and secure exclusive mating, high- ranking females may obtain access to the best foods. A better fed female can invest more energy in reproduction and thereby produce more offspring, or she can supply more food to her offspring.
4.1.2 Socio-ecology of dominance relationships
Ecological factors have far reaching consequences on the relationships formed among primate females (Wrangham, 1987). The socio-ecological model predicts that food distribution and predation risk shape the competitive regime and therefore the relationships formed among females of diurnal primate species (van Schaik, 1989; Sterck et al., 1997; Koenig, 2002). Species facing scramble competition or no competition should have egalitarian dominance relationships, in which hierarchies are unclear and non-linear, if distinguishable at all. In contrast, species facing contest competition should have despotic dominance relationships, in which dominance relationships are clearly established
28 Chapter 4: Food competition and linear dominance hierarchy and form usually linear hierarchies. Such despotic females have often formalised dominance relationships, which are expressed in ritualised signals wherein the direction is independent of the context (de Waal, 1986; de Waal, 1989b).
Contest-type competition occurs when food distribution allows some individuals to exclude others from accessing the resource. Therefore, contest competition should increase with the monopolisablity of the resource and with the number of competitors. A linear hierarchy should be adaptive when contest is so strong that the number of aggressive interactions needs to be reduced by clear dominance relationships among the competitors. Indeed, intra-specific and inter-specific comparisons have proved that females that face more contest competition have a more linear and formalised hierarchy, e.g. Saimiri oerstedi versus Saimiri sciureus (Mitchell et al., 1991), sympatric Presbytis thomasi and Macaca fascicularis (Sterck & Steenbeek, 1997), and 3 neighbouring groups of Semnopithecus entellus (Koenig, 2000).
Dominance relationships develop from repeated contests within dyads (Bernstein, 1981; Drews, 1993). Many group-living animals use ritualised signals to avoid aggression (de Waal, 1986). Such formalised submissions are one of the last behaviours to indicate rank changes and therefore serve as an indicator for acceptance of the relationship (Macaca fascicularis: de Waal, 1977; Macaca mulatta: de Waal & Luttrell, 1985; Papio cynocephalus: Walters, 1980; Pan troglodytes: de Waal, 1982; Wittig, 1997). De Waal & Luttrell (1985) concluded that social integration is a condition for a formal hierarchy as it clarifies the social status of an individual in the group. Clear relationships among all individuals should lead to an interaction pattern, where in all individuals can be accepted as partners for affiliative interactions, though interaction frequencies and intensities among dyads differ due to the quality of the relationship.
4.1.3 Relationships in chimpanzees
Chimpanzees of all study sites live in multimale-multifemale, fission-fusion societies, where the composition of parties can change frequently (Sugiyama, 1984; Goodall, 1986; Nishida, 1990; Boesch, 1991; Wrangham et al., 1992). When meeting each other, subordinates greet dominants by emitting pant-grunt vocalisations, which in chimpanzees serve the function of formalised submission, as they provide a unidirectional and context- free assessment of dominance relationships (de Waal, 1978; Bygott, 1979; Noë et al.,
29 Chapter 4: Food competition and linear dominance hierarchy
1980). However, as an exchange of greetings is only possible when associated, the greeting frequency should increase with association.
Most chimpanzee communities are probably male-bonded (Mitani et al., 2002), as males are philopatric in chimpanzees, while females emigrate from their natal community (Watts & Pusey, 1993). Furthermore, males create long-term alliances (Riss & Goodall, 1977; de Waal, 1982; Nishida & Hosaka, 1996), and female chimpanzees are usually characterised as being egalitarian (Sterck et al., 1997). However, genetically both Taï and Gombe males are not more related than females within the same community (Vigilant et al., 2001), though they are still the philopatric sex. Furthermore, it has been argued that Taï chimpanzees are bisexually bonded because females build co-operative long-term relationships, occasionally with stronger association than males (Boesch & Boesch- Achermann, 2000). As Taï females are nut-crackers (Boesch & Boesch, 1983; Boesch & Boesch, 1984,b) and have frequent access to meat (Boesch & Boesch, 1989; Boesch, 1994a,b), contest competition might be more important among them than in other communities. Therefore Taï females may benefit from forming a linear dominance hierarchy.
I investigated the existence of formal dominance relationships among females of a community of wild chimpanzees in the Taï National Park, Côte d’Ivoire. I fit the formal dominance relationships of the females to a hierarchy and test for linearity. I next quantify contest competition over food resources in females and hypothesise that females increase aggressive interactions when more competitors are present or with monopolisable food. Additionally, dominant partners are expected to possess the food after the conflict. I also investigated the relationship among dominance hierarchy, age and contest aggression. As females in other study sites did not establish linear hierarchies, though they had different ages, dominance rank should only correlate with contest aggression. Finally I assess the association and grooming relationships among the females and discuss possible explanations for different findings in hierarchies in other chimpanzee populations. Taï females should associate more frequently and have more grooming dyads, indicating a stronger integration of females in the social network. The same principles should work within the community and female dyads with unknown relationships should be less associated and have weaker grooming relationships.
30 Chapter 4: Food competition and linear dominance hierarchy
4.2 Specific methods
4.2.1 Data and test conditions
The data set consists of 123 full-day observations of focal females. I observed each female between 10 and 15 days during the observation period, which provided a total of 1028 h of female focal observation time. The death of Brutus in March 1997 changed the association pattern in the community, so I analysed the 2 periods separately for the questions related to association.
4.2.2 Operational definitions
I assess the dominance relationships of dyads by greetings, which are specific vocalisations emitted by the subordinate to the dominant, usually accompanied by an approach and submissive behaviour, such as crouching or bobbing, of the subordinate. Three vocalisations serve the function of greetings in Taï: (a) pant-grunt (PG), a repeated grunt exhaled with an open mouth; (b) greeting-hooh (GH), a repeated and intense hooh; and (c) greeting-pant (GP), a repeated pant accompanied by submissive behaviour.
Food is monopolisable when it was accessible only in one spot or by one tool in the feeding site, e.g., one stone hammer. One food spot is only one fruit, one water hole, one insect-nest or whole colobus monkey. All these situations gave one individual the possibility to feed on the food exclusively. The number of competitors are all the adult and independent subadult chimpanzees present in the feeding site (=feeding party), which excludes all members of the party that stay outside of the feeding site, e.g., resting under the food tree, but visible. I calculated the observation time for feeding parties and the time a type of food was present from the focal female observations. An aggressive interaction over food is a food conflict and was won by the individual that possessed the food after the conflict. I excluded food conflicts with infants and juveniles from the analysis.
The greeting rate (GR) is the number of greetings within dyad AB divided by the observation time of females A and B. The conflict rate is the number of aggressive interactions among females in a feeding party of size X divided by the total observation time of feeding parties of size X. The grooming rate is the duration of grooming of dyad AB divided by the observation time of female A and B. Generally, rates are the number of
31 Chapter 4: Food competition and linear dominance hierarchy events per total observation time; however, corrected rates are divided by the DAI, because females in fission-fusion societies can only interact with a partner when associated in the same party. For comparisons with other study sites I had to calculate the rates as number of events among all females divided by the total observation time of all females, due to lack of data for relationships among female chimpanzees.
4.2.3 Statistics
I tested for linearity of hierarchies via MATMAN© (Noldus, 1998), which provides several measures to describe the linearity. I implemented two tests:
1. The linearity test, including Landaus linearity index (h) and Kendall’s coefficient of linearity (K), provide a measure of the degree to which a dominance hierarchy is linear, and both range from 0 to 1, with 1 describing complete linearity. While h basically compares the number of dyads in which A dominates B to the total number of dyads (Landau, 1951; Chase, 1974), K basically compares the number of circular triads with the total number of dyads (Appleby, 1983). Linearity is evidenced when the proportion of circular triads is less than expected by chance based on a 2 distribution (Appleby, 1983). Both measurements give similar results but they are problematic when the matrix of the dominance relationships contains unknown relationships.
2. I conducted an improved linearity test when >10% of the relationships were unknown or tied. Unknown relationships are dyads with no greeting exchanged. Tied relationships are dyads with an equal number of greetings in both directions. For the improved linearity test I used the corrected linearity index (h’), which is the average of all h values calculated for the complete set of each possible dominance matrix. The complete set of dominance matrices is produced by systematically switching the direction of dominance for unknown relationships, while the tied relationships are always assigned half dominant and half subordinate status (de Vries, 1995). The significance of the linearity was obtained by a randomisation test, which compares the h value of 10000 randomly chosen matrices with h’. Linearity is documented when >95% of the randomly chosen matrices have a smaller h than the corrected linearity index (h’).
32 Chapter 4: Food competition and linear dominance hierarchy
To compare differences of interactions in dyads, I applied the permutation test. Kendall partial rank-order correlation coefficient was conducted using the procedure of (Siegel & Castellan, 1988).
4.3 Results
4.3.1 Female hierarchy
I observed 187 greeting interactions among females that included PG (62%), GP (30.5%) or GH (7.5%). Of the 55 female-female dyads, two thirds showed unidirectional greeting (67.3%), 12.7% of the dyads had bi-directional greetings and 20% of the dominance relationships remained unknown as no greetings were exchanged. This lack of ~ greetings might be due to the rather low greeting rate (GR) in female dyads (GR ff=0.011 greetings/hour). Greetings among females occurred sixteen times less frequently than ~ among males (GR mm=0.178 g/h; Permutation test: p<0.05) and 4 times less frequently ~ than in mixed sex pairs (GR mf=0.047 g/h; Permutation test: p<0.05).
dominant subordinate Receiver rank categories: Signaller Macho Brutus Marius Mystère Nino Loukoum Venus Ricci Goma Perla Belle Castor Narcisse Dilly Fossey Macho 0 0 0 0 0 0 00000 0 0 0 Brutus 130000 000000000 Marius 643 5 0 0 0 0 0 0 0 0 0 0 0 0 Mystère 96 19 20 0 1 0 00000 0 0 0 Nino 361 21 42 0 0 0 00000 0 0 0 Loukoum 51 4 13 7 4 0 00000 0 0 0 Venus 42 10 11 1 0 2 11000 0 0 0 Ricci 79 11 19 4 0 5 2 0000 0 0 0 Goma 44 11 14 2 1 2 12 000010 Perla 48 14 15 7 1 6 117 11020 Belle 28 9 15 1 0 0 4132 0000 Castor 94 12 11 19 3 5 2 1121 0 0 0 Narcisse 53 3 15 3 0 3 1 00002 0 0 Dilly 86 33 30 25 1 8 4 32110 4 4 0 Fossey 29 3 7 0 0 2 0 13110 0 0
Figure 4.1 Dominance relationships, based on greetings, among the adult chimpanzees in the North community in Taï. The signaller, presented vertical, is greeting the receiver, presented horizontal. The individuals are shown according to their rank in increasing order from left to right. The six narrow rank categories of females are indicated by different bars above the females names (note that the last category, marked white, includes the greetings of the subadult females). For dyads including males (italic–grey letters) pant-grunts are shown only (grey numbers), whereas among females greeting-pants and greeting-hoohs are inserted as well.
When analysing the direction of greeting events I detected a linearity in the hierarchy among the females (Figure 4.1; Improved linearity test: h’=0.67, p<0.01). Forming narrow rank categories (Figure 4.1; two rank-neighbouring females per category, and the three adolescent females in the lowest rank category), females showed a clear
33 Chapter 4: Food competition and linear dominance hierarchy
2 rank order (Linearity test: h=0.97; K=0.97; 30 =51, p<0.01), which has an equivalent strength to the linear hierarchy of the males (Figure 4.1; Linearity test: h=1; K=1). The hierarchy of the complete community (females and males) was linear with females subordinate to males (Improved linearity test: h’=0.80, p<0.001) except of the unknown dominance relationship between -female Mystère and subadult male Nino. Thus the dominance relationships of female chimpanzees in Taï fitted a linear hierarchy.
Table 4.1. Food conflicts among female chimpanzees in the North community in Taï. The number of conflicts is presented per initiating female in declining dominance rank order. While the left column shows the proportion of conflicts over monopolisable food, the right columns show the possessor of the food depending on the dominance relationships of the conflict partners. Average proportions and expected proportion for conflicts over monopolisable food are presented in the last rows.
Female Conflicts over Possessor of food after conflict conflicts monopolisable food* dominant subordinate 3rd individual Initiator n % nn n Mystère 9 22 9 0 0 Loukoum 35 63 27 5 3 Venus 5 60 5 0 0 Ricci 3 33 2 1 0 Goma 6 33 4 0 2 Perla 11 36 8 1 2 Belle 2 100 1 1 0 Castor 11 9623 Narcisse 1 100 0 1 0 Dilly 17 59 7 5 5 Fossey 3 33 3 0 0 mean 50% 66% 23% 11% expected 8% *conflicts over non monopolisable food complete proportions to 100%
4.3.2 Food competition
I observed 103 conflicts over food among females, giving a rate of about 0.1 food conflicts per hour of observation time and about 0.22 food conflicts per hour in a feeding party. Although only 8% of the community’s feeding time was on monopolisable food, monopolisable food was the reason for 50% of the food conflicts among females (Table 4.1). Comparing the indices of difference (observed minus expected frequency) for each female showed that females initiate conflicts over monopolisable food more frequently than over non-monopolisable food (Wilcoxon exact: T=66, N=11, p<0.01). Females fought over many different types of monopolisable food, such as meat, stone hammers
34 Chapter 4: Food competition and linear dominance hierarchy
(tools to crack nuts), water holes in trees, eggs of ants, honeycombs, fruits of Treculia africana and tree-mushrooms. Although only 42% of the feeding time on monopolisable food was attributed to meat, almost 70% (34 of 49 cases) of the conflicts on monopolisable food were over meat. Meat was the main reason for contests over 2 monopolisable food among female chimpanzees in Taï (Goodness of fit exact: 1 =15.04, p<0.001).
2
1.5
r 2=0.74, p <0.001
1 conflict rate [conflict / h] / [conflict rate conflict
0.5 r2=0.01, ns
0 non 1234567891011 12 13 14 competitors present in feeding party [N]
Figure 4.2 Conflict rate among females over food with different numbers of competitors present in the feeding party in Taï chimpanzees. Feeding parties attending non- monopolisable food () are considered separately from those attending monopolisable food (). Regression lines are calculated for non-monopolisable (solid line) and monopolisable food (dashed line). Test statistics see 4.3.2.
The conflict rate, when feeding on non-monopolisable food, strongly increased with the number of competitors present in the feeding party (Figure 4.2; Pearson: r=0.86, N=14, p<0.001). However, no correlation was found for monopolisable food (Pearson: r=0.12, N=14, ns), indicating no difference in strength of competition over food between
35 Chapter 4: Food competition and linear dominance hierarchy one and more competitors, when the food can be monopolised by one individual (Figure 4.2). Although females faced competition with between 1 and 17 chimpanzees in their feeding parties, I only included feeding parties up to 14 competitors, as cumulative feeding time reaches a plateau with 14 competitors and feeding time with more competitors was rare (<1 h).
On average the dominant female of the conflict dyad possessed the food in 66% after conflicts (Table 4.1). Being dominant over a competitor provided an advantage in female contests, as dominant conflict partners possessed the food significantly more frequently after conflicts than subordinates independent from the initiator (Table 4.1; Wilcoxon exact: T=53.5, N=11, p<0.01). However in 15 cases none of the females was successful (Table 4.1), as both lost the food to a third party (in five cases to a male, four to a dominant female, five to a group of females and in one case to a subordinate female). Hence, female chimpanzees in Taï faced contest competition over food, as females fought more for monopolisable food or with increasing numbers of competitors and dominant females possessed the food after a conflict.
Table 4.2 Correlates of dominance rank in Taï females. The females are ordered in declining dominance rank order and show the assigned dominance rank, contest rank and age.
Name Dominance rank Contest rank Age [years]* Mystère 1224 Loukoum 2127 Venus 3721 Ricci 4336 Goma 5526 Perla 6423 Belle 7823 Castor 8623 Narcisse 91016 Dilly 10 9 21 Fossey 11 11 20 *approximate age in 1999, as dates of birth are estimated except for Belle (Boesch and Boesch-Achermann, 2000)
36 Chapter 4: Food competition and linear dominance hierarchy
4.3.3 Correlates of the linear hierarchy
I tested two correlates of the linear ranking in Taï females. First, dominance rank order of females was related to age, with older females dominating younger ones (Table 4.2; Spearman Rank: r=-0.70, N=11, p<0.05). Second, I created a linear hierarchy based on contest aggression interactions over food (contest rank), with winners being assigned higher ranking (Table 4.2; Improved linearity test: h’=0.43, p=0.06), and found that contest rank correlates with dominance rank (Spearman Rank: r=0.86, N=11, p<0.001). However, when eliminating the influence of either contest rank or age, the correlation between dominance rank and age disappeared (Kendall Partial Correlation: T=0.05, N=11, ns), while the correlation between dominance rank and contest rank remained significant (Kendall Partial Correlation: T=0.54, N=11, p<0.05). Thus the linear hierarchy is related to the outcome of the contest, while it is independent from age.
4.3.4 Social relationships ~ Females had a median DAI ff of 0.27 (Appendix C). Although their association is very high compared to other study sites (Table 4.3), female-female association in Taï was ~ about one third less than male-male association ( DAI mm=0.42; Permutation test: ~ p<0.05), but more frequent than female-male association ( DAI mf=0.23; Permutation test: p<0.05). Before the death of Brutus the association among the females was about ~ 25% lower than after (Appendix C; with Brutus: DAI ff=0.21; Permutation test: p<0.05). If the lack of clear dominance relationships in other study sites is due to low greeting rates because of a lower association (Table 4.3), one would expect a lower corrected greeting rate (CGR) during the period of lower association. However there was no difference in the CGR between the periods of lower and higher association (Permutation test: ns). I even found a higher CGR during the period of lower association, when analysing the 15 female dyads that showed greetings in both periods (with Brutus: ~ CGR ff=0.171 greetings/associated hour (25%: 0.147, 75%: 0.311); without Brutus: ~ CGR ff=0.074g/ah (25%: 0.032, 75%: 0.128); Permutation test: p<0.05).
Finally I tested whether the 11 female dyads of unknown dominance relationships ~ were rare associates and non-groomers. The median DAI unknown of the 11 dyads was 0.27 (combine Figure 4.1 and Appendix C) and thus the females of unknown dominance relationships associated the same amount as females with unidirectional or bidirectional
37 Chapter 4: Food competition and linear dominance hierarchy
~ ~ relationships ( DAI unidirectional=0.29; Permutation test: ns; DAI bidirectional=0.31; Permutation test: ns). Female pairs had a median grooming rate of 2.71 seconds/hour (25%: 0.32s/h, 75%: 4.90s/h), spread over 30 bi-directional grooming, 18 unidirectional grooming and 7 non-grooming relationships. The 11 unknown dominance relationships were represented by five bi-directional, five unidirectional and one non-grooming dyads. Thus there was no difference in the grooming pattern between dyads of unknown and unidirectional 2 dominance relationships (Goodness of fit exact: 2 =0.83, ns). Hence I note that within the community an increase in association did not imply more greetings and that neither association nor grooming patterns among Taï females were able to explain the lack of dominance relationships in some dyads.
Table 4.3 Comparison of dominance relationships among female chimpanzees in five study sites in the wild. Parameters presented in the table describe the hierarchy, social relationships and contest competition of females.
proportion of non- mean female N adult rank unidirectional greeting grooming greeting rateU food conflict rateV study site observation time [h]R females categories dominance relationships DAIS patternT corrected rate W corrected rate W Budongo1 54 12 and not- 74% 0.09 --- 0.022 --- Sonso 0.24
Gombe2,3 4363 11-172 high, middle, --- 0.053 35/26/393 --- 0.0033 Kasakela low2 0.06
Kibale4 7 13 high and low --- 0.08 0/0/100 ------Kanyawara
Mahale5,6,7,8 635 135 1-135 (linear) 69%5 (0.39)6 45/558 0.0465 0.0165 M-group W0.057 0.92 0.32
Taï9 103 11 1-11 (linear) 33% 0.27 55/32.5/ 0.113 0.1 North-group 12.5 0.42 0.37 Note: R mean observation time of target females in hours. S average dyadic association index of females. T grooming pattern shows proportion of bi-directional / unidirectional / non grooming female dyads (Mahale: grooming / non-grooming). U greeting rate is all greetings exchanged by females per total female observation time. V food conflict rate is all food conflicts among females per total female observation time. W corrected rate (in italic numbers) are rates divided by the DAI. --- no data available. 1 Fawcett, 2000 2 Pusey et al., 1997 3 Goodall, 1986; observation time: 8 focal females for the two year period of 1978-79 from appendix c, DAI: from 1978-81 (appendix d), grooming pattern: calculated from in camp grooming sessions of 3 adult females within 31 dyads (Fig. 14.2), conflict rate: from Fig. 12.4 and multiplied with 0.66 as only two thirds of female conflicts are with other females (chapter: aggression). 4 Wrangham et al., 1992 5 Nishida, 1989; conflict rate: from Table 8. 6 Nishida, 1968; K-group. 7 Nishida & Hosaka, 1996; DAI: with a correction factor for DAI in not artificial feeding areas (correction method described in: Boesch & Boesch-Achermann, 2000, Table 5.9). 8 Huffmann, 1990; from focal data of females in 1985: Fig. 12.2, of 29 F-F dyads representing grooming dyads vs. non-grooming dyads. 9 this study (chapter 4; Wittig & Boesch, 2003a).
38 Chapter 4: Food competition and linear dominance hierarchy
4.4 Discussion
4.4.1 Summary of results
The dominance relationships of female chimpanzees in Taï fit to a linear hierarchy. While the linear hierarchy is independent of age, winning contest over food is related to the dominance rank order. Taï females face food competition with higher competition over monopolisable food or with more competitors in the feeding site. They maintain affiliative relationships with almost all other females in the community and show higher association patterns than females in other chimpanzee sites. Neither association nor grooming patterns explained the occurrence of unknown dominance relationships within the community.
4.4.2 Dominance relationships
Despite the fact that I found a linear hierarchy among the females in Taï, some dyads emitted greetings in both directions. This is similar to other study sites where females sometimes used mutual pant-grunts during greetings (Gombe: Goodall, 1986; Mahale: Nishida, 1979). Even among males, where unidirectionality of pant-grunts is generally reported in the literature, dominance reversed pant-grunts were found (Gombe: Goodall, 1986; Mahale: Nishida & Hosaka, 1996; Taï: Boesch & Boesch-Achermann, 2000), indicating that pant-grunts are not exclusively unidirectional. Even though I used a combination of different greeting vocalisations among the females, these greetings still showed strong directionality. Therefore these greetings provide a good indicator for females’ dominance relationships, showing that they form a stable and linear hierarchy.
Female chimpanzees in Taï appeared to show a more clear linearity in their hierarchy than females in the five other chimpanzee sites (Table 4.3). Only two dyads (Perla-Dilly and Dilly-Fossey) remained unclear when taking triadic relationships into account. In contrast more than two thirds of the dominance relationships were unknown in Mahale, about three quarters of the dominance relationships were unknown in Budongo (Table 4.3) and in Gombe, even though broad categories were implemented, 12% of the females did not fit to a rank (Pusey et al., 1997).
39 Chapter 4: Food competition and linear dominance hierarchy
4.4.3 Contest competition
Female chimpanzees in Taï used aggression to keep resources or to win them from other females. Unequal access, especially to monopolisable food such as meat, might be an explanation for the development of the linear hierarchy in Taï females. For example, dominant females may access 500 grams of meat per successful hunt, a substantial benefit compared to 80 grams of average females (Boesch & Boesch-Achermann, 2000). There was even unequal access to non-monopolisable food due to contests in large feeding parties (Figure 4.2). It seems that Taï females contest over feeding space in food patches (e.g crowns of fruit trees), as observed in Thomas langurs (Presbytis thomasi: Sterck & Steenbeek, 1997). Being dominant seems to enable females to have priority of access to a higher quality and quantity of food. However the question remains whether the frequency of contest is regular enough to actually cause the linear hierarchy. Comparing the individual aggression rates of Taï females attending a feeding site reveals that Taï females are in the upper range of primate species of whom several fit to linear hierarchies (individual female food contest rate [conflict/hour feeding]: e.g.: Cercocebus torquatus: 0.04c/h, Range & Noë, 2002; Chlorocebus aethiopis: 0.07c/h and Erythrocebus patas: 0.04c/h, Pruetz & Isbell, 2000; Saimiri sciureus: 0.28c/h, Mitchell et al., 1991; Pan troglodytes: 0.22c/h, this study; Wittig & Boesch, 2003a). Thus contest competition seems to be high and beneficial for the dominant females.
4.4.4 Reasons for linearity
It seems that being older enhances the chance to win the contest. Taï females, like females in other study sites, may increase their individual rank as they age (Gombe: Pusey et al., 1997; Mahale: Nishida, 1989). However the dominance rank order in Taï females is related to contest rank but not to age. Thus competition over food is the main determinant for the development of the linear hierarchy in Taï females.
In spite of contest competition, Taï females have affiliative interactions with almost all other female community members. Although neither association nor dyadic grooming relationships were able to explain the variability in the Taï dominance relationships, the comparison with other study sites reveals remarkable differences (Table 4.3). In addition to an almost five times higher association rate among females in Taï, they presented the highest number of dyads with bi-directional grooming and the lowest number of females that never groomed together compared to other study sites. Taï females build long-lasting
40 Chapter 4: Food competition and linear dominance hierarchy friendships including food-sharing and support (Boesch & Boesch-Achermann, 2000) and they seem to take in account such factors when initiating aggression (chapter 516; Wittig & Boesch, 2003b). Therefore the female society in Taï might be shaped through advantages of female bonding when access to food is affected by contest competition.
4.4.5 Comparison among study sites
Why did I find such differences in the female relationships among different communities of the same species? One argument might be that linear hierarchies have not been detected yet in the other study sites. On one hand, linearity is easier to find in small groups (Drews, 1993), but the number of females examined was similar across sites (Table 4.3). On the other hand longer observations increase the chance to observe greetings between females. Observation time is surely not the problem for Gombe, while it is perhaps a problem in the Kibale study (Table 4.3). However observation time in Taï, Mahale and Budongo is similar, and therefore differences among the female hierarchies should be independent of observation time. In Mahale and Budongo greetings among females occurred at least 50% less often than in Taï (greeting rate) which could explain the large number of non-unidirectional relationships (Table 4.3). However in Mahale, but not in Budongo, females actually greeted more frequently when they met each other than in Taï (Table 4.3; corrected greeting rate). This is a similar observation to our findings of a higher corrected greeting rate during the lower association period. It may be, that female chimpanzees have to confirm their submissive position more often the less frequently they meet when a dominance relationship does exist. Thus in Mahale a linear hierarchy might exist, but may require more data for detection, whereas the dominance relationships are ambiguous among the females in Budongo.
The other possibility might be that female dominance relationships do not fit to linear hierarchies in other study sites. As food distribution and predation risk are suggested to influence female gregariousness, which shapes the competitive regime of females, contest competition should depend on the association of females (Sterck et al., 1997). Although food conflicts occur 6 to 30 times less frequently in other study sites than in Taï (Table 4.3; food conflict rate), the actual rate of food conflicts among two females when associated is the same for Taï and Mahale (Table 4.3; corrected food conflict rate). Thus Mahale females face similar levels of contest competition as Taï females, whereas in
16 see Figure 5.5
41 Chapter 4: Food competition and linear dominance hierarchy
Gombe contest competition among females is lower. However females in Mahale and Taï cope differently with the contest competition. While Mahale females seem to disperse to reduce contest competition, Taï females build a formal linear hierarchy to endure the contest competition. This may reflect an higher advantage in staying together for Taï females – perhaps to reduce predation risk (Caraco et al., 1980; van Schaik et al., 1983; Boesch, 1991). Although mortality rates for Taï and Mahale chimpanzees are similar and the highest under the known chimpanzee communities (Hill et al., 2001), predation by leopards is only known to exist in Taï (Hiraiwa-Hasegawa et al., 1986; Boesch, 1991; Boesch & Boesch-Achermann, 2000). Hence females in Taï may have evolved a clearer linear dominance hierarchy than females in other chimpanzee study sites, as an adaptation to a stronger contest competition (compared to Gombe) and a higher predation risk (compared to Mahale).
The comparison among study sites (Table 4.3) revealed strong differences, not only in the females’ dominance relationships, but also in their competitive regimes, their level of association and affiliation, and the predation risk females and their offspring face. Although there are some limitations to the data used here for the cross-study comparison, the data having been collected with different methods or during different periods and adjusted post-hoc for comparison, our comparison supports the predictions of the socio- ecological model (van Schaik, 1989; Sterck et al., 1997). Data on general food distribution and abundance were not considered for the comparison, as such data are just becoming available (e.g.: Fawcett, 2000; Anderson et al., 2002).
Our findings of a linear dominance ranking in females and a stronger female integration in the social network, support the bisexual model of Taï chimpanzee social organisation (Boesch, 1991), as do other findings from Taï (Lehmann & Boesch, in prep.). This shows that social structure in chimpanzees is more flexible than previously thought and adapts to ecological circumstances.
42 Chapter 517:
Decision-making in conflicts: extension of the Relational Model
Figure 5.0 Alpha-male Macho (centre) displays a bipedal swagger with a shaking branch, just before he starts to charge male Marius (Photo by R. Wittig).
17 Corresponding with: Wittig, R.M. & Boesch, C. (2003b). “Decision-making” in conflicts of wild chimpanzees (Pan troglodytes): an extension of the Relational Model. Behavioral Ecology and Socioniology. 54:491-504 Chapter 5: Decision-making in conflicts
5.1 Introduction
5.1.1 The Relational Model (RM)
Group-living animals face the problem of competing over resources and mating partners with the same social partners that they cooperate with in protection against predators and in resource acquisition (Wrangham, 1980; Walters & Seyfarth, 1987; van Hooff & van Schaik, 1992; de Waal, 2000b). The dilemma is that an out-competed cooperation partner might withhold assistance in the next situation when cooperation is required. Therefore the use of aggression to resolve competitive situations carries advantages and disadvantages. This has led to a number of mechanisms being proposed which function as counter-disadvantage regulations (e.g. dominance hierarchy: Popp & de Vore, 1979; post-conflict management: Kappeler & van Schaik, 1992; meat-sharing: Boesch, 1994b). The most recent and comprehensive model for aggressive competition including counter-disadvantage mechanisms is the Relational Model (RM: de Waal, 1996a).
The RM considers aggression as well as tolerance and avoidance as possible reactions to a conflict of interest – a competitive situation of two individuals having incompatible goals (van der Dennen & Falger, 1990). The model predicts that whenever there is a conflict of interest, the tendency to take aggressive action increases with the number of opportunities for competition, as well as the resource value and the reparability of the relationship, while it decreases with the risk of injury and the value of the relationship (de Waal, 2000a). An individual is viewed as going through a decision-making process during a conflict of interest, weighing the costs against the benefits. Thus, I do not imply any deliberate intention when referring to decision-making18. Although the RM is a convincing model, it has never been tested as a complete model with data of one species. In this paper I assess this decision-making process in wild chimpanzees using a cost – benefit estimation based on the RM, where the net-benefit applies to benefits minus costs.
In the RM the benefit is taken as the value of the resource. According to the socio- ecological model, food is proposed as the most beneficial resource for females whereas a sexual partner holds the most benefit for a male (Janson, 2000). High rank, determining
18 see last paragraph of 1.3
44 Chapter 5: Decision-making in conflicts priority of access to females, is very beneficial to male primates (Dixson, 1998). In the RM cost of aggression is taken as the risk of injury. Injury obtained during fighting can entail tremendous costs through infection and disability, hampering food acquisition, travelling and defence against predators (Palombit, 1993). In the RM social costs are represented by the negative effects aggression produces by disturbing the relationship between partners (e.g. unwillingness to support or to share food). The more valuable the relationship, the more partners suffer from the disturbance (Aureli & Smucny, 2000). However, the disturbance can be reduced by reconciliation, such that relationships of former opponents return to their baseline level and these dyads seem to interact like before (Cords, 1992). In the RM the tendency to reconcile is referred to as the reparability of the relationship. The opportunity for competition, suggested as a predictor in the RM, seems to be more a predictor for the number of conflicts of interests individuals face rather than for their aggressive tendency. Although the aggression rate increases with competition (Janson, 1988b; Barton & Whiten, 1993; Sterck & Steenbeek, 1997), the individual’s aggressive tendency per conflict of interest should not be affected when the other factors stay constant. Therefore, the opportunity for competition is not considered as a predictor in our study.
5.1.2 Extension to the Relational Model
5.1.2.1 Likelihood of winning
Despite the fact that several costs and benefits are taken into account, the RM does not explicitly consider two other important factors. First, whilst predicting a higher tendency for aggression when the value of the resource is high, it only evaluates the potential benefit of the resource, without considering that the resource has to be accessed to gain the benefit. This requires winning the contest. Different fighting abilities of conflict partners influence the outcome of the contest (Huntingford & Turner, 1987). In stable social groups, where the same two individuals compete repeatedly, dyadic dominance relationships often emerge as linear rankings at group level (Jackson, 1988). Conflict partners with small rank differences have more equally matched fighting abilities. In consequence, the probability of winning should increase with the rank difference for the dominant initiator, but should decrease for the subordinate who fights a stronger individual. Except for the highest and the lowest ranking member of the community, individuals are facing conflict situations where each can be either dominant or subordinate to the conflict partner. In this manner, the likelihood of winning a contest, a precondition
45 Chapter 5: Decision-making in conflicts for gaining the value of the resource (the benefit), is an important component added to the RM.
5.1.2.2 Conflict duration and cost of aggression
Second, the RM incorporates the risk of injury as the only cost of aggression. Although most aggressive interactions end without injuries (Bernstein et al., 1983; Whitten & Smith, 1984), are they really free of cost? On one hand aggression requires energy expenditure and this energetic cost of fighting has been shown to increase with duration (Smith & Taylor, 1993; Hack, 1997; Neat et al., 1998). On the other hand, longer conflicts are usually followed by aggressive post-conflict interactions (chapter 619; Wittig & Boesch, 2003d). Thus, third parties might join the conflict the longer an aggressive interaction continues, which involves additional costs of aggression. As energy expenditure increases with conflict duration and long conflicts involve additional aggression costs I have added duration as second necessary component to the RM.
5.1.2.3 Creation of social costs
Furthermore the RM predicts that the tendency to initiate aggression increases with reparability of the relationship independent of whether there will be negative effects of the aggression. However, repairing a relationship is important only when there is something to repair. Cords & Thurnheer (1993) have shown that the occurrence of reconciliation is dependent on the value of the relationship of former opponents. Therefore, the value of the relationship and its reparability are dependent on their contribution to the RM. Both factors have to be considered together in the decision- making process, as repairing the relationship can counteract the social costs.
5.1.3 Testing the extended Relational Model
In general animals can accept higher costs as long as the net-benefit is positive (Parker & Rubenstein, 1981). This might lead to different strategies by the same individual depending on the conflict situation: the same resource provides a different benefit to different individuals and different opponents require different fighting intensities or incur different social costs. Thus the dyadic character of conflict situations has to be considered by a competitor when deciding about aggression.
46 Chapter 5: Decision-making in conflicts
Chimpanzees represent a good model to test the decision-making process. Chimpanzees live in multi-male multi-female communities and face intra- and inter-sexual competition with multiple partners. A formal linear hierarchy is found among all adult members of the community based on dyadic dominance relationships (chapter 420; Wittig & Boesch, 2003a). They have frequent bouts of aggression that occur in different contexts, with different partners and different intensities (Goodall, 1986). Alternatively, individuals can avoid aggression before it arises by just leaving the party due to the fission-fusion character of the chimpanzee society. Finally chimpanzees built and maintain long-term relationships and often reconcile after aggressive interactions (Arnold & Whiten, 2001; Preuschoft et al., 2002).
The aim of this chapter is to test the extended RM with aggressive interactions of chimpanzees in the Taï National Park, Côte d’Ivoire. This test investigates how chimpanzees handle the conflict situation, taking into account the conflict’s costs and benefits. The extended RM includes as benefit a combination of the benefit of the resource – assessed by the value of the resource – and the likelihood of winning, whereas costs include the cost of aggression and the social costs – assessed by the value of the relationship – minus the part of the social costs that gets reduced by relationship repair. My extended version of the RM predicts that aggression is expected when the net-benefit is positive.
I tested the extended model using a step by step procedure, adding more parameters in each step, to show schematically how the decision-making process may operate for chimpanzees. First, the resource arouses the individuals interest. The individual assesses its likelihood of winning and how much aggression to invest, taking into account the identity of the competitor. Finally the individual estimates the disturbance the aggression will cause to its relationship with the competitor and if this is repairable. Thus the potential gain in net-benefit may be estimated, allowing an individual to decide whether to initiate aggression in this particular situation and how to handle the conflict. Each decision is highly dependent on the identity of the individuals in the dyad. Therefore our analysis is based on dyadic data and not on individual data. For simplicity from now on the term conflict is used instead of aggressive interactions.
19 see Figure 6.2 20 see 4.3.1
47 Chapter 5: Decision-making in conflicts
Step1: Estimating the benefit a. The benefit of the resource: individuals are predicted to initiate conflicts more frequently in accordance with the value of the resource to them. I expect females to initiate more conflicts over food and males over oestrous females. b. The likelihood of winning: individuals are predicted to initiate conflicts more frequently the higher their likelihood of winning. I expect dominants to initiate and to win more conflicts than subordinates.
Step 2: Estimating the costs of aggression Individuals are predicted to risk more the higher the benefit of the resource is to them as long as they have a chance to access it. I expect chimpanzee initiators to have longer and more intense fights the higher the value of the resource is to them and when they are dominant in the dyad. However, although dominant initiators are able to risk more than subordinates, in order to minimise costs individuals are predicted to avoid escalation of aggression. I expect therefore short and less intense fights when the rank difference to the opponent is large.
Step 3: Estimating the social costs a. The social costs: individuals are predicted to keep the disturbance of their relationship with the conflict partner small. I expect chimpanzees to initiate less conflicts with high benefit partners. If they initiate conflicts, then they are shorter and less intense (to minimise the costs of aggression) and mainly for high benefit resources in dyads which they are likely to win (to maximise the benefit). b. Reducing social costs: individuals are predicted to reduce social costs more frequently when the social costs are high. I expect chimpanzees to have a high chance of reconciling when they initiate conflicts with high benefit partners (to reduce social cost).
5.2 Specific methods
5.2.1 Data and test conditions
The result of the balancing process to eliminate repeated measurements was a data set of 409 conflicts. I tested the predictions following a stepwise model design, which changed the number of variables in each step (Table 5.1). During step 1b (likelihood of
48 Chapter 5: Decision-making in conflicts winning), cases ending in a draw were not considered, as samplesize was too small (less than 4% of conflicts).
Table 5.1 Variables considered for the tests in the different steps of the stepwise model design.
independent variables step part factor of the extended RM dependent variable set W set X set Y set Z 1 a benefit of the resource Conflict Context ? b likelihood of winning Initiator’s Victory Outcome ? ? 2a4 Aggression Type ? ? b 5 cost of aggression Conflict Intensity ?? c 6 Conflict Duration ? ? 3 a social cost Relationship Benefit ? ? ? b reduction of social cost Corrected Conciliatory Tendency Index ? ? ? ? ?= set of variables considered as independent variables for the test set W: Initiator’s Sex, Receiver’s Sex, Sex Combination set X: Conflict Context, interaction (Conflict Context x Initiator’s Sex), Initiator’s Rank, Rank Difference, interaction (Initiator’s Rank x Rank Difference) set Y: Conflict Intensity, Conflict Duration set Z: Relationship Benefit, Association Index
5.2.2 Additional operational definitions
Reconciliation was defined as the first post-conflict interaction with socio-positive body-contact among former opponents. I used the corrected conciliatory tendency (CCT) for the probability that a dyad will reconcile after a conflict (Veenema et al., 1994). CCT is the quotient of the number of reconciliation events occurring sooner than the baseline time minus the number of reconciliation events occurring later than the baseline time, divided by number of conflicts. I calculated the baseline time as the average time interval between two consecutive friendly interactions of the same dyad on the same day. Since CCT is higher for males than females (Preuschoft et al., 2002), I corrected CCT for sex differences by calculating an index for comparative reasons (like for DAI and rank difference: see 3.2.5 and Table 3.2).
5.2.3 Special analysis
To calculate expected frequencies I corrected for the frequency of occurrence of each context. As conflicts could occur only when there was a partner present, I excluded data when chimpanzees were alone. Social conflicts could happen anytime (expected frequency: 1.0). However food conflicts could occur only when chimpanzees were feeding. Males like females showed an average feeding time of about 45% of their time budget (expected frequency=0.45). Conflicts in a sex context could occur only when at least one female had a genital swelling. For male focals, oestrous females were present on 48 of the 80 days and focal females were in oestrous on 23 of the 123 days. As a
49 Chapter 5: Decision-making in conflicts result, expected frequencies are different for male-male conflicts (0.6), mixed sex conflicts (0.35) and female-female conflicts (0), since only conflicts with the focal animal were recorded and competition over sex among females was not expected. To enable a comparison between expected and observed frequencies, I standardised the frequencies by setting the sum of expected frequencies within each sex combination to 1. When comparing among five and less individuals, standard non-parametric statistics is not applicable (step 1a for male initiated conflicts). Therefore, I applied the permutation test (see 3.4.2).
Table 5.2 Frequency (in %) of conflicts initiated by females and males per conflict context and sex of conflict partner in Taï chimpanzees.
females female – female conflicts female – male conflicts context n social food sex* n social food sex Belle 14 86 14 2 50 50 0 Castor 19 42 58 4 25 50 25 Dilly 26 35 65 3 67 33 0 Fossey 75743 2 0 0100 Goma 10 40 60 2 50 50 0 Kana 3 33 67 0 ------Loukoum401387 1839610 Mystère 19 53 47 3 33 67 0 Narcisse 6 83 17 4 100 0 0 Perla 15 27 73 1 0 100 0 Ricci 8 62 38 5 20 80 0 Venus 12 58 42 5 60 0 40 average4951404515 expected 69 31 56 25 19
males male – male conflicts male – female conflicts context n social food sex n social food sex Brutus 19 58 37 5 31 42 55 3 Macho 356 80 9 11 100 41 30 29 Marius 29 90 3 7 77 51 4 45 Nino 0 ------36 58 25 17 average 76 16 8 48 28 24 expected 49 22 29 56 25 19 expected are the standardised expected frequencies within each sex combination; *females did not compete over sex.
50 Chapter 5: Decision-making in conflicts
5.3 Results
5.3.1 Step 1: Benefit
5.3.1.1 Benefit of the resource
Did chimpanzees fight more in particular contexts (Table 5.1: step 1a)? Table 5.2 compares the frequency of initiated conflicts in the different contexts of each individual with the random expectation. The average frequencies suggest that females in same sex and mixed sex dyads used more aggression in competition over food. On the other hand aggression among males was more common in social contexts and males attacked females more in food and sex contexts. I compared the relative difference between observed with expected frequencies for each individual. Females fought more frequently over food than in social contexts with other females (Wilcoxon exact: T=10, N=12, p<0.05), and more frequently over food than in social and sex contexts when attacking
males (Friedman exact: T2=6.049, N=11, p<0.05). For males I applied the permutation test due to the low number of individuals (Table 5.2). The six male-male dyads fought more often in social contexts (Permutation test: p<0.01) and less often over food (Permutation test: p<0.05), while they fought like expected over sex (Permutation test: ns). On the other hand the 47 dyads of males attacking females did not show a preference in any context (Permutation test: social, ns; food, ns; sex, p<0.1). Thus females fought more over food and males fought males more in social contexts.
Table 5.3 Factors affecting the likelihood of winning a conflict in Taï chimpanzees (Table 2: step 1b). The table presents the significant variables and the estimate- coefficient () of the parameters.
independent variables dependent variable Initiators Victory Outcome Wald P parameters n Wald P
1 Initiator’s Rank 40.15 <0.0001 dom 302 2.53 40.15 <0.0001 sub 78 -2.53 2Initiator’s Rank x 16.76 <0.001 dom-small 110 -1.24 16.75 <0.0001 Rank Difference sub-small 45 1.24 dom-middle 108 -0.45 1.94 ns sub-middle 24 0.45 dom-large841.69 sub-large 9 -1.69 13.18 <0.001 Model: Winner vs. loser independent parameters: dom = dominant initiator, sub = subordinate initiator, small = small rank difference, middle = middle rank difference, large = large rank difference.
51 Chapter 5: Decision-making in conflicts
5.3.1.2 Likelihood of winning
How was the chance of winning a conflict affected by the dominance relationship and the context of the conflict (Table 5.1: step 1b)? Likelihood of winning a conflict was 2 significantly affected by two variables (GLZ: LR 5,380 =84.19, p<0.0001; Table 5.3). Dominant initiators were more likely to win conflicts (92%, =2.53) than subordinate initiators (60%, =-2.53). However the likelihood of winning decreased with rank difference for dominant initiators whereas the opposite was true for subordinates (Figure 5.1). The initiator’s likelihood of winning was greater the larger the relative dominance of the initiator (5. 1: from very subordinate to highly dominant). In consequence dominants initiated aggression more frequently with increasing rank difference (see n in Figure 5.1, as the categories dom-large and sub-large etc. represent dyads from two different perspectives). Thus, the more dominant an initiator was over his opponent, the higher was his likelihood of winning the conflict and the more conflicts he initiated.
100%
90%
80%
70%
60%
50%
conflicts 40%
30%
20%
10%
0% dom-large dom-middle dom-small sub-small sub-middle sub-large all conflicts (n=110) (n=108) (n=84) (n=45) (n=24) (n=9) (N=380) interaction: Initiator's Rank x Rank Difference
Figure 5.1 Likelihood of winning a conflict in Taï chimpanzees. The proportion of initiator winning ( ) compared to initiator losing ( ) the conflict is displayed as a function of the interaction between Initiator’s Rank and Rank Difference. The interaction is presented from the highest to the lowest likelihood of winning (dom = dominant initiator, sub = subordinate initiator, large = large rank difference, middle = middle rank difference, small = small rank difference, test statistics see Table 5.3).
52 Chapter 5: Decision-making in conflicts
5.3.2 Step 2: Cost of aggression
5.3.2.1 Conflict intensity
How was the intensity of the fight affected by the resource and the partner’s dominance relationship (Table 5.1: step 2a)? Aggression type was only significantly 2 affected by rank difference between opponents (GLZ: LR 4,409 =17.11, p<0.01; Table
5.4a). Dyads with a large rank difference were more likely to use non-contact aggression (large: 81%, =0.63) than dyads with closer rank conditions (middle: 63%, =-0.34; small: 64%, =-0.29), independent of the context. Did chimpanzees then prefer low levels of intensity when subordinate and high levels when dominant to their conflict partner (Table 5.1, step2b)? Conflict intensity was only significantly affected by initiator’s 2 rank (GLZ: LR 3,409 =23.62, p<0.0001; Table 5.4b). Looking at non-contact and contact aggression separately (Figure 5.2), I found that subordinates initiated conflicts with lower levels of intensity for both non-contact (threats, level 1) and contact aggression (single physical contact, level 4). Dominants, however, preferred to initiate conflicts with higher
45
40 non-contact aggression (1-3) contact aggression (4-5) 35
30
25
20 conflicts [%] 15
10
5
0 level 1 (n=95) level 2 (n=86) level 3 (n=96) level 4 (n=72) level 5 (n=60) single multiple threat display no locomotion display with locomotion contact aggression contact aggression Conflict Intensity
Figure 5.2 Intensity of conflicts in Taï chimpanzees. The percentage of conflicts for dominant ( ) and subordinate ( ) initiators as well as for all conflicts ( ) is presented for each level of intensity. Intensity levels 1–3 are non-physical contact conflicts, intensity levels 4+5 are physical contact conflicts (Table 3.2, test statistics see Table 5.4b). levels of intensity for both aggression types (non-contact aggression: displays, level 2+3;
53 Chapter 5: Decision-making in conflicts contact aggression: multiple physical contact, level 5). Thus both types of aggression occurred in the repertoire of subordinates and dominants, but subordinates mainly initiated the lower and dominants the higher intensity variants of both types. In addition both preferred non-contact aggression when the rank difference was large.
Table 5.4 Factors affecting intensity and duration of conflicts in Taï chimpanzees (Table 2: step 2, costs of aggression). The table presents the significant variables for three multivariate analyses (a: model of aggression type, b: model of conflict intensity, c: model of conflict duration). In addition it shows the GLZ estimate- coefficient () of the parameters and GLM post-hoc statistics.
independent variables dependent variables (a) Aggression Type Wald p parameters n Wald p 1 Rank Difference 8.41 <0.05 small 174 -0.29 2.87 <0.1 middle 138 -0.34 3.75 <0.1 large 97 0.63 8.40 <0.01
(b) Conflict Intensity Wald p parameters n Wald p 1 Initiator’s Rank 5.30 0.0213 dominant 314 -0.39 5.30 <0.05 subordinate 95 0.39
(c) Conflict Duration FpPost-hoc P 1 Initator’s Sex 8.89 <0.01 Fisher LSD <0.05 2Conflict Context 5.95 <0.05 Fisher LSD <0.01 Model a: Aggression type (non-contact aggression vs. contact aggression). Model b: Conflict intensity (1 – 2 – 3 – 4 – 5). Model c: GLM
5.3.2.2 Conflict duration
In addition to intensity, duration of the conflict was considered as a cost of aggression (Table 5.1: step 2c). Two variables significantly influenced the duration of a conflict (GLM: F10,347=3.44, p<0.001; Table 5.4c). Males initiated longer conflicts than females and conflicts in social contexts were significantly longer than in food contexts (Figure 5.3). Since males fought more in social contexts and females more in food contexts, but the interaction of both significant effects was not significant in the GLM
(F1=2.27, ns), I investigated the interaction separately again. I found that male initiated
54 Chapter 5: Decision-making in conflicts social conflicts were longer and had the largest range, whereas none of the other combinations were significantly different (Figure 5.3; Fisher LSD with Bonferroni correction: significance level of p<0.016; male-social: range=1-61s, male-food: range=1- 21s, female-social: range=1-22s, female-food: range=1-33s). Thus, the effects of initiator’s sex and conflict context was partly due to males initiating long conflicts in social contexts.
5
4
3
2
p <0.01 conflict[s±SE] duration 1 p<0.01
p <0.01
0 male female social food male male female female (n=176) (n=182) (n=193) (n=165) social food social food (n=108) (n=68) (n=85) (n=97) Initiator's Sex Conflict Context interaction: Initiator's Sex x Conflict Context
Figure 5.3 Duration of conflicts in Taï chimpanzees. The mean duration (sMSE) is presented for each parameter of the significant variables of the model of conflict duration and also for the interaction. Post-hoc statistic for the interaction is shown in the graph in case of significance (significance level was Bonferroni-corrected to p<0.016, test statistics see Table 5.4c and 5.3.2.2). 5.3.3 Step 3: Social costs
5.3.3.1 Creating social costs
How did chimpanzees distribute their social costs (Table 5.1: step 3a)? Conflict 2 induced social costs were significantly affected by three variables (GLZ: LR 8,409 =90.97, p<0.0001; Table 5.5a). Dominant initiators initiated more conflicts with low benefit partners (low: 37.9%, =0.48), while subordinate initiators initiated more conflicts with opponents of higher benefit (medium+high: 67.4%, =-0.48). Moreover, the interaction of dominance and rank difference had the strongest effect of all significant variables on
55 Chapter 5: Decision-making in conflicts social costs (Table 5.5a, -coefficients). Considering dominant and subordinate initiators separately showed that dominants initiated more aggression with friends when the rank difference was small and more with non-friends when rank difference was large (Figure 5.4). Conversely subordinate initiators attacked friends more when the rank difference was large and non-friends when rank difference was small (Figure 5.4). Thus dominant, like subordinate, individuals started conflicts with friends more frequently as their relative likelihood of winning decreased (dom-small: =-0.95, sub-large: =-0.94). In addition, the sex combination of opponents had an effect (Table 5.5a). Mixed pairs fought relatively less with low benefit partners and relatively more with partners of a high benefit relationship, whereas in conflicts among opponents of same sex it was the opposite (conflicts per sex combination with low benefit / high benefit partners: m- m=36.3%/15.9%; f-f=49.0%/6.9%; m-f=28.8%/24.7%; f-m=22.0%/19.5%; all conflicts=36.7%/16.1%).
80
70 dominant initiators subordinate initiators
60
50
40
conflicts [%] conflicts 30
20
10
0 dom - large dom - middle dom - small sub - small sub - middle sub - large all conflicts (n=85) (n=109) (n=120) (n=54) (n=29) (n=12) (N=409) interction: Initiator's Rank x Rank Difference
Figure 5.4 Relationship benefit of opponents representing the social costs of conflicts in Taï chimpanzees. The percentage of conflicts between high benefit ( ) and low benefit ( ) partners is displayed as a function of the interaction between Initiator’s Rank and Rank Difference. For simplicity medium benefit partners are not presented in the figure, but when including medium benefit partners conflicts total a 100%. Social costs are presented from high to low likelihood to win (dom = dominant initiator, sub = subordinate initiator, large = large rank difference, middle = middle rank difference, small = small rank, test statistics see Table 5.5a).
56 Chapter 5: Decision-making in conflicts
5.3.3.2 Reducing of social cost
When were chimpanzees more likely to reduce social costs (Table 5.1: step 3b)? Reduction of social costs was only significantly affected by relationship benefit of the 2 opponents (GLZ: LR 6,409 =58.81, p<0.0001; Table 5.5b). The tendency to reconcile between opponents increased strongly with relationship benefit, such that among friends more conflicts showed high CCT index (high relationship benefit: 63.6% with high CCT index, 36.4% with low CCT index, =-1.20) and among lower benefit partners more conflicts showed a low CCT index (medium relationship benefit: 23.8% high CCT index, 76.2% low CCT index, =0.57; low relationship benefit: 23.3% high CCT index, 76.7% low CCT index, =0.63). Thus social costs were more likely to be reduced among friends.
Table 5.5 Factors affecting social costs in Taï chimpanzees (Table 2: step 3). The table presents the significant variables for two multivariate analyses (a: model of social costs, b: model of reduction of social costs) and the estimate-coefficient () of the parameters.
independent variables dependent variables (a) Relationship Benefit Wald p parameters n Wald p
1Initiator’s Rank x 25.69 <0.0001 dom-small 120 -0.95 23.55 <0.0001 Rank Difference sub-small 54 0.95 dom-middle 109 0.01 0.01 ns sub-middle 29 -0.01 dom-large 85 0.94 16.60 <0.0001 sub-large12-0.94 2 Sex Combination 8.57 <0.01 m-m 77 0.46 f-f 145 0.46 8.57 <0.01 m-f 146 -0.46 f-m 41 -0.46 3 Initiator’s Rank 6.55 <0.05 dom 314 0.48 sub 95 -0.48 6.55 <0.05
(b) CCT Index Wald p parameters n Wald p 1Relationship Benefit 13.07 <0.01 low 150 0.63 4.83 <0.05 medium 193 0.57 4.73 <0.05 high 66 -1.20 13.03 <0.001 Model a: Relationship Benefit (low – medium – high) Model b: Corrected Conciliatory Tendency Index (low vs. high) independent parameters: dom = dominant initiator, sub = subordinate initiator, small = small rank difference, middle = middle rank difference, large = large rank difference, m = male, f = female
57 Chapter 5: Decision-making in conflicts
Figure 5.5 Schematic illustration modelling the factors affecting two fighting strategies R in Taï chimpanzees. The illustration follows frequency of conflictsof of initiation of our proposed order of the ‘decision-making’ benefit of resource process. Diagrams R-U show how the costs or benefits considered in each step S affect the variation in initiation frequencies frequency of conflicts of conflicts, while diagram V shows how of initiation likelihoodlikelihood of winning to social costs affect the variation in the strategy frequency of reduction of social costs. The dominant initiator subordinate initiator positive or negative slopes of the graphs A B are based on the observed frequencies of conflicts in Taï chimpanzees (see Figures 5.1-5.4), using a straight line for simplicity. T
Two strategies were found: Strategy A is fr eq uency of conflictsof of initiation of used by dominant initiators (fat white cost of aggression arrow) whereas strategy B is used by subordinate initiators (small black arrow). The outcome (W) of the two strategies is U frequency of conflictsof of in itiation represented by the net benefit in relation social cost to the relative initiation frequency of conflicts for dominant (A, ) or subordinate V initiators (B, ). The outcomes of both strategies are presented in a simplified frequency of reduction of of social cost form (the most likely outcome) and are social cost shown together in one graph to clarify the outcome dominant initiator subordinate initiator trade-off used by subordinate conflict partners: subordinate initiators trade their A W B lower frequency of winning a conflict for a A greater positive net benefit, when they are winning, and a smaller negative net- benefit, when they are losing, compared to B B dominant initiators (see 5.4). frequency of initiated conflicts initiated A negative 0 p ositive net benefit
58 Chapter 5: Decision-making in conflicts
5.4 Discussion
5.4.1 Summary of results
Taï chimpanzees consider costs and benefits when determining whom, when and how to fight – how to handle a conflict situation. All variables considered in the RM (value of the resource, risk of injury, value and reparability of the relationship) were found to significantly affect the decision-making process of Taï chimpanzees. Our results also revealed the limits of the RM. While the RM does not consider different aggression tendencies for dominant and subordinate initiators, I found that these differed in Taï chimpanzees, as they considered their likelihood of winning the conflict in their conflict decisions. Moreover, the RM could not directly explain the within individual variation in costs of aggression and social costs that I observed in Taï chimpanzees. The extension of the RM, I propose, remedies these shortcomings.
5.4.2 Grabbing the benefit
Adult chimpanzees fought more in contexts that had a high potential benefit to them. Female chimpanzees in Taï initiated more aggression over food than in social conflicts. This result corresponds with the findings in Gombe chimpanzees, where females also fought frequently over food (Goodall, 1986). The females’ focus on food is not surprising as better fed primate females have a higher overall fitness (e.g. baboons: Altmann et al., 1985; Japanese macaques: Mori, 1979; rhesus macaques: Small, 1981). In contrast, males preferentially engaged in social conflicts. Considering the fact that access to oestrus females represents the key resource for reproductive success of males, this result is surprising. However, males might achieve high ranks by winning conflicts in non- sexual contexts, which in itself might be sufficient to reach their reproductive goals, either through female choice (Matsumoto-Oda 1999) or because subordinate males are reluctant to challenge dominants in a sexual context. Therefore, engaging in social conflicts might be a more effective strategy to maximise the reproductive success in male chimpanzees, since high ranking males in Pan species sire more offspring than low ranking ones (Gerloff et al., 1999; Constable et al., 2001; C. Boesch & L. Vigilant, unpublished data). Thus individuals initiated more conflicts as the benefit of the resource increased (Figure 5.5, R).
59 Chapter 5: Decision-making in conflicts
In accordance with the general mammalian pattern, of dominant individuals being more aggressive than subordinates (e.g. carnivores: Creel et al., 1997; perissodactyla: Clutton-Brock et al., 1976; primates: Hrdy & Hrdy, 1976; ungulates: Côté, 2000), Taï chimpanzees of both sexes initiated more conflicts the higher their dominance position. Furthermore, Taï chimpanzees also won more conflicts the more dominant they were over their opponent. Neither results are surprising as they reflect the character of a dominance hierarchy. Thus individuals initiated more conflicts as the likelihood of winning increased. (Figure 5.5, S).
Contrary to expectation, subordinate initiators often won, suggesting that Taï chimpanzees use a sufficient decision-making process. This may be due to the egalitarian character of chimpanzee societies which enables subordinate individuals to compete successfully (van Schaik, 1989). Alternatively, subordinate winners may have had potential dominant supporters in the party, as the mere presence of supporters can change the outcome of a dyadic conflict (de Waal, 1982). A third explanation may be that sex related differences in fighting decisions help females to win food conflicts against males, since males seem to avoid food conflicts more often than females, although they are generally dominant over females. Thus, subordinate initiators may choose their opponents more carefully than dominants do by monitoring the situation more precisely. The factors enabling subordinates to win require further examination.
5.4.3 Economic handling of the costs of aggression
Taï chimpanzees handled their costs of aggression economically, their investment in duration and intensity of aggression being shaped by the potential benefit of the context and their likelihood of accessing the benefit. Individuals used non-contact aggression more frequently when the winner was predictable due to a large rank difference. Males fought for a longer time in contexts that were more beneficial for them. While it may seen questionable that fighting for an extra second represents a significant increase of energetic costs, the range of conflict duration shows that male chimpanzees fought up to one minute in social contexts, almost three times as long as the maximum conflict duration over food. A continuous performance of muscles in humans can lead to energetic problems within one minute due to usage of ATP reserves (Keul et al., 1967), phosphocreatine shortage (McKenna et al., 1999) or neuronal firing fatigue (Ross et al., 2001). In contrast, females did not fight much longer over food than in social contexts, even though food is the most important context for them. This might reflect female
60 Chapter 5: Decision-making in conflicts experience that longer fights draw more combatants (R.M. Wittig & C. Boesch, unpublished data). Thus a food specific risk arises, as food owners can lose control over the resource to a third party during a conflict, a risk that individuals seem not to face when competing for rank. Taï chimpanzees risk high costs of aggression only when this is required to achieve a positive outcome.
Subordinate initiators used lower levels of contact and non-contact aggression than dominants. It must be adaptive for subordinate initiators to minimise costs of aggression as they are less often able to offset their costs by accessing the benefit. (Crowley, 2000) detected a similar mixed evolutionary stable strategy (EES) for asymmetric contest situations. He investigated a general form of the hawk-dove game (Maynard Smith & Parker, 1976) which included the fighters’ knowledge about their own fighting ability and that of their conflict partner. The main EES combined a cautious strategy (dove) for subordinate partners with a daring strategy (hawk) for dominants (Crowley, 2000). Similarly dominant opponents in Taï chimpanzees initiated more conflicts with high costs of aggression, whereas subordinate opponents initiated more conflicts with low costs of aggression (Figure 5.5, T).
5.4.4 Minimising social costs
Taï chimpanzees risked different social costs depending on their likelihood of winning. Social costs originate from the disturbance of a cooperative relationship when one partner withholds cooperation (de Waal, 1996a). Thus, a disturbed relationship with non-cooperative partners is unlikely to create social costs. In Taï, the effect of the initiator’s rank on the relationship benefit of opponents clearly showed that subordinates initiated more conflicts with cooperative partners. Consequently subordinate initiators generally risked more social costs than dominant ones (Figure 5.5, U). This general trend was also reflected by the fact that both types of initiators, dominant as well as subordinate, risked more social costs the lower was their relative likelihood of winning. It was probably unnecessary for a dominant to fight a friend when the outcome of the conflict was predetermined, but dominants risked a fight when the outcome was less predictable. Why, however, did subordinate initiators also not fight more with friends with a small rank difference? A partnership does not provide the same value to each partner (Cords & Aureli, 2000). Dominant individuals are likely to provide better quality food and support to subordinates than the other way around. Therefore, subordinate initiators should have been more careful with highly dominant friends. Instead, however, they
61 Chapter 5: Decision-making in conflicts initiated less conflicts with closely ranked opponents. Opponents of a small rank difference fought harder and therefore those conflicts may have induced a stronger disturbance to the relationship. Subordinates thus were probably more careful with friends when facing the risk of a difficult relationship repair. Subordinate initiators generally risked more social costs than dominants, except when reparation of the relationship could be difficult – suggesting more gentile slopes of social costs (Figure 5.5, U) compared to other graphs in Figure 5.5.
Such a strategy is only adaptive if high social costs were reduced more frequently than low ones after the conflict. Indeed Taï chimpanzees were more likely to reduce higher than lower social costs (Figure 5.5, V). This higher reconciliation frequency with friends has also been seen in other primates (reviewed in: Aureli et al., 2002). Since social costs seemed to be easily reduced afterwards, why did Taï chimpanzees not fight friends more frequently? Reconciliation, however, is not free of risk, as individuals have to approach former opponents which may lead to further aggression (e.g. Aureli & van Schaik, 1991b; Cords, 1992). Moreover reconciliation might fail if the same relationship is repeatedly disturbed, with a bigger disturbance possibly requiring longer and more intense reconciliation.
5.4.5 The extended Relational Model
How did Taï chimpanzees decide to fight for a resource? They were more likely to initiate conflicts for resources that were highly beneficial to them, but they used less aggression when the chance of winning was lower. The likelihood of winning therefore served as the key factor in the decision-making process, and two different alternative strategies can be identified (Figure 5.5):
(A) Dominant initiators: fought harder, but they did not risk undue costs of aggression. Additionally they risked less social costs by fighting less with partners of higher benefit. Thus, dominant conflict partners fight over highly beneficial resources, investing more in aggression costs but less in social costs. Dominant initiators gain a medium positive net-benefit when winning and face a medium negative net-benefit when losing, since costs of aggression are non-reducible and have to be subtracted from the benefit. In consequence, as dominant initiators almost always win, their strategy has a predictable outcome: often obtained medium positive net-benefit (Figure 5.5, W)!
62 Chapter 5: Decision-making in conflicts
(B) Subordinate initiators: had lower costs of aggression, but took higher social costs, except when reparation of the relationship could be difficult. Hence subordinate conflict partners also fight over high benefit resources, but invest less in aggression costs and more in social costs. Subordinate initiators gain a high positive net-benefit when winning and face low negative net-benefit when losing, since they have low costs of aggression and the social costs are reducible. In consequence subordinate initiators trade their lower chance to win the conflict for a higher positive and a lower negative net- benefit compared to dominant initiators. This trade off is a risky strategy: rarely obtained high positive net-benefit (Figure 5.5, W)!
The extended RM describes the decision-making process in Taï chimpanzees well , although one can never be sure that all potential effects have been identified. I did not investigate situations directly when chimpanzees reacted non-aggressively during a conflict of interest. Nevertheless, I found that females initiated less social conflicts, whereas males initiated less conflicts over food. Furthermore, one partner initiated the conflict while the other one had not yet decided to initiate aggression. Therefore, Taï chimpanzees seem to avoid aggression when the resource is not beneficial to them and when their decision-making process does not offer a positive net-benefit. However, the decision-making process is not perfect as sometimes initiators misjudged the situation and lost the conflict. Still the overall net-benefit in both strategies was positive, as subordinate initiators showed a trade-off for their lower winning rate by going for a larger net-benefit (Figure 5.5, W).
Taï chimpanzees demonstrate a very flexible decision-making process before initiating a conflict. Clearly more factors are taken into account than initially suggested by the RM. Decisions are done in an economical fashion to increase the benefit for the initiator while diminishing the costs. The extended RM proposed here fits well with the social complexity observed in wild chimpanzees. The extended RM, providing a powerful way to investigate conflict management, may help to answer further questions about when aggression is avoided and how effective the mechanisms are for managing a conflict (de Waal & Aureli, 2000). Hence, comparative studies are requested, to test this model on other primate species, to make within and between species differences in conflict decisions apparent.
63 Chapter 621:
Choice of post-conflict interactions
Figure 6.0 Female Venus has the choice of several social partners in the party for a grooming interaction (Photo by R. Wittig).
21 Corresponding with: Wittig, R.M. & Boesch, C. (2003d). The choice of post-conflict interactions in wild chimpanzees (Pan troglodytes). Behaviour. 140:1527-1559 Chapter 6: Choice of post-conflict interctions
6.1 Introduction
6.1.1 Advantages and disadvantages of post-conflict interactions
Aggression disturbs the relationship of conflict partners and induces social tension within social groups (Cords, 1992; Aureli et al., 1999; Matsumura & Okamoto, 2000). Such costs of conflicts remain present even after the aggression has ended. Post-conflict management is believed to reduce the costs of conflicts for social living animals. Post- conflict interactions (PCI), which in brief is the first interaction of a conflict partner subsequent to aggression, can either reduce those costs (e.g. stress reduction: Cords & Aureli, 2000; limit damage to a disturbed relationship: Aureli & Smucny, 2000) or can prevent further aggression (Aureli et al., 2002). Yet, while PCIs represent an advantageous mechanism for social living animals, serious disadvantages may also be accrued (Watts et al., 2000; Aureli et al., 2002). Conflict partners may become trapped in further aggression when, for example, approaching former opponents for reconciliation (Aureli & van Schaik, 1991b; Cords, 1992). Conflict partners can choose from a pool of PCIs that consist of several affiliative or aggressive PCIs, carried out with former opponents or third parties (Aureli & de Waal, 2000a). Therefore selecting an optimal PCI requires an evaluation of both the advantages and disadvantages. Individuals choosing a PCI are viewed as going through a decision-making process. The terms decision and choice do not necessarily imply a conscious reasoning process, rather individuals can switch between different behavioural possibilities22.
Within group aggression erupts when individuals compete over food, mating partners, social partners or privileges of access to resources (Huntingford & Turner, 1987; Mason & Mendoza, 1993). Subsequently to aggressive interactions, which I will refer to as conflicts, PCIs can be implemented. Conflict partners can choose from a pool of PCI including reconciliation, solicited consolation, renewed aggression and redirected aggression, while in addition conflict bystanders can offer consolation or attack conflict partners (Aureli & de Waal, 2000a). Alternatively conflict participants may avoid any further interaction (no PCI). Each of the seven options (six PCIs + no PCI) has a potential pay-off, since they carry certain advantages and disadvantages.
22 see last paragraph of 1.3
65 Chapter 6: Choice of post-conflict interctions
6.1.2 Pool of PCIs
6.1.2.1 Reconciliation
Reconciliation, the affiliative PCI between former opponents, repairs the relationship of former opponents by restoring inter-opponent tolerance levels to baseline (Cords & Aureli, 1996). Partners of highly valuable relationships should restore tolerance levels to normal, since otherwise beneficial cooperation might be withheld (Cords & Thurnheer, 1993; de Waal, 1996a). Reconciliation also reduces stress indicators (Aureli, 1997; Aureli et al., 1999) and prevents further aggression (Aureli & van Schaik, 1991b; Watts, 1995a; Silk et al., 1996). Moreover approaching former opponents might offer another chance at accessing the resource, which precipitated the fight. Relaxed partners could, for example, share food after reconciliation has reduced stress levels. Although reconciliation appears to reduce most costs of conflicts, it certainly is not implemented after each conflict (Aureli & de Waal, 2000a). This may be due to the fact that reconciliation requires former opponents to approach one another thereby giving the opportunity of further aggression (Aureli & van Schaik, 1991b; Cords, 1992). Subordinate partners may thus profit disproportionately from relationship repair, as dominant partners are likely to share better quality food and better support than subordinates (Cords & Aureli, 2000). On the other hand subordinate cooperation partners may tip the scales in certain situations of social leverage (Lewis, 2002), that the need for relationship repair could be similar for both partners. Nonetheless subordinate partners incur comparatively higher risks by approaching former opponents, and might thus be less likely to initiate reconciliation when dominant partners are highly motivated to fight again.
6.1.2.2 Consolation
In addition to reconciliation, consolation, the affiliative PCI with third parties, is also proposed as a mechanism to alleviate stress and reduce the risk of further aggression (Das, 2000; Watts et al., 2000). Consolation is unlikely to repair the relationship of conflict partners, since former opponents do not interact directly, however it avoids an opportunity of further aggression. Consolation thus may substitute for reconciliation after conflicts with a high likelihood of further aggression or among low benefit partners, as both serve similar functions of reducing stress and preventing further aggression (Watts et al., 2000). Consolation may even restore tolerance levels of former opponents to baseline levels when carried out with kin of former opponents (Cheney & Seyfarth, 1989; Judge, 1991). Conflict partners profit from consolation by means of stress reduction, while
66 Chapter 6: Choice of post-conflict interctions third parties may profit from offering consolation when they are potential recipients of redirected aggression (Das et al., 1997; Das et al., 1998). Since there might be a functional difference, I have distinguished between offered consolation (Table 6.1), initiated by third parties, and solicited consolation, initiated by conflict participants (de Waal, 1993; de Waal & Aureli, 1996; Watts et al., 2000).
6.1.2.3 Redirected aggression
In contrast to affiliative PCIs, all aggressive PCIs include the complete spectrum of disadvantages that are associated with aggression such as risk of injury, energy use, disturbed relationships and potential loss of access to the resource (chapter 523; de Waal, 1996a; Wittig & Boesch, 2003b; Table 6.1). Redirected aggression, in which a conflict partner initiates an aggressive PCI with third parties, nonetheless carries advantages. It may reduce aggression-induced stress and frustration (Aureli & van Schaik, 1991a), as losers of conflicts with a high rate of redirected aggression showed lower levels of physiological (Levine et al., 1989; Sapolsky, 2000) and behavioural stress indicators (Aureli & van Schaik, 1991b; Maestripieri et al., 1992). Since winners of conflicts are also stressed by the conflict (Aureli, 1997), they might use redirected aggression to calm their own stress response. Furthermore, individuals may use redirected aggression to deflect the aggressive attention of an opponent to a third party (de Waal & van Hooff, 1981). Sometimes, when conflicts are long or intense, this might be one of the best possibilities for a victim to leave the focus of aggressive attention. In contrast to redirected aggression, third party aggression, in which a third party initiates an aggressive PCI, is neglected in most studies. Advantages for a third party to attack a conflict participant may be either to seize possession of an especially profitable resource (chapter 524; Preuschoft & van Schaik, 2000; Wittig & Boesch, 2003b) or to defeat a frequent opponent caught in a weak position (e.g. contra-intervention: de Waal, 1978).
6.1.2.4 Renewed aggression
In addition to these most often discussed PCIs, two additional types of post-conflict behaviour, renewed aggression and avoidance of further interactions, are of interest as well (Table 6.1). While renewed aggression, the aggressive PCI between former opponents, entails costs of conflict, it provides the loser of a conflict with a second opportunity to gain access to the disputed resource. Therefore it can be seen as a re-
23 see 5.1.1 & 5.1.2
67 Chapter 6: Choice of post-conflict interctions escalation of the same conflict. Nonetheless renewed aggression is only advantageous for losers with a sufficient likelihood of winning. On the other hand renewed aggression may help winners of conflicts in strengthening their dominant position against an already weakened opponent (Johnstone & Dugatkin, 2000).
6.1.2.5 Avoidance of interactions (no PCI)
Finally, conflict partners may avoid any interactions (no PCI) in order to prevent any further confrontation. However, avoiding the disadvantages of PCIs means that the advantages are also inaccessible, like stress reduction or a second attempt at accessing a resource. Since stress also reduces over time (Aureli & van Schaik, 1991b), conflict partners, that are only mildly stressed, might not need stress reducing PCIs. Nonetheless, avoiding interactions with party members makes monopolisable resources, that are possessed by them, inaccessible – a clear disadvantage of no PCI for the loser.
6.1.3 Testing the choice of PCI
Following a conflict social living animals can select a option from the pool of PCIs (six PCIs and no PCI). Since some advantages can be gained from several different options, e.g. conflict partners can reduce stress by means of at least three different PCIs (reconciliation, consolation and redirected aggression), advantages and disadvantages that are unique to a PCI should play a decisive role in the evaluation process for the best choice of PCI. Table 6.1 summarises the advantages and disadvantages of all PCIs and the conditions under which each PCI is expected. Chimpanzees represent an excellent model to investigate the decision-making process behind the choice of PCI. They are highly sociable (Goodall, 1986; Nishida, 1990; Boesch & Boesch-Achermann, 2000) and most of the described options in the pool of PCIs are known to exist in chimpanzees (de Waal & van Roosmalen, 1979; de Waal & van Hooff, 1981; de Waal, 1984; de Waal & Aureli, 1996; Arnold & Whiten, 2001; Preuschoft et al., 2002). However this study is the first to check on most of them for wild populations and in any case to consider all options of the pool of PCIs simultaneously. I investigate the selection process of PCIs in wild chimpanzees, in terms of the factors that influence the choice of PCI and who initiates the PCI. Our underlying hypothesis is that after a conflict chimpanzees use the PCI that provides the most advantages while carrying the fewest disadvantages. Our main questions are:
24 see Table 5.2
68 Chapter 6: Choice of post-conflict interctions
(1) Do wild chimpanzees apply all seven options from the pool of PCIs (six PCIs and no PCI) for post-conflict management? (2) Under which conditions do wild chimpanzees use each type of PCI and does this choice of PCI follow the proposed evaluation process that takes both advantages and disadvantages into account? (3) Which are the general rules that reflect the choice of PCI?
Table 6.1 The pool of post-conflict interactions (PCIs): Overview of advantages and disadvantages of PCIs and the conditions under which each
is expected.
Advantages (A) and Disadvantages (D) Conditions under which PCI is expected: Type of PCI for either conflict participants (CP) or third parties (TP) PCI implemented more often when - A* (CP) stress reduction --- A* (CP) prevent further aggression --- Reconciliation A (CP) relationship repair relationship benefit is high between CP (a) A (CP) access resource (non-aggressive attempt) loser only: resource is limited D (CP) possible renewed aggression renewed aggression is unlikely between CP A* (CP) stress reduction --- Consolation A* (CP) prevent further aggression --- (b) offered + A (CP) no approach of former opponent renewed aggression is likely between CP (c) solicited A (TP) calm tension third party only: TP possible target of redirected aggression D (CP) no relationship repair relationship benefit is low between CP conflict is a draw, resource is limited or highly valuable Renewed A (CP) access resource (second aggressive attempt) loser only: outcome of conflict is unexpected aggression A (CP) defeat frequent opponent winner only: CP is rank neighbour (d) D* (CP) costs of aggression --- A* (CP) stress reduction --- Redirected A (CP) defuse frustration loser only: conflict is strong (long and intense) aggression A (CP) deflect aggression non-aggressive PCI unlikely to stop aggression for CP (e) D* (CP+TP) costs of aggression --- Third party A (TP) defeat weakened frequent opponent conflict is strong (long and intense) aggression A (TP) access resource (first aggressive attempt) resource is limited or highly valuable (f) D* (CP+TP) cost of aggression --- No PCI A (CP) avoid risks of disadvantages of PCIs no advantages of PCIs are needed by CP (e.g. no stress) (g) D (CP) no possibility to access resource resource is not limited Conditions relating to the advantages or disadvantages are shown in the same row. Overlapping* advantages and disadvantages have no expected conditions (-- -), since they do not distinguish between the different types of PCI.
6.2 Specific methods
6.2.1 Data and test conditions
A data set of 595 conflicts resulted from the balancing procedure. I modelled the likelihood that a particular PCI occurs against the non-occurrence of this PCI (ˆ all other PCIs). To detect the variables that affects the choice of PCI, I executed multivariate analyses. All variables of Table 3.2, plus the interactions of initiator’s sex with recipient’s sex, rank difference with initiator’s rank and conflict context with competitor proportion, were considered simultaneously as independent variables in each of the multivariate analyses. Additionally, conflict duration was analysed separately on an individual level due to limitations of our binomial testing procedure on continuous predictors. A distinct mid- length conflict duration for a PCI would not be detected, since I tested one PCI against all
69 Chapter 6: Choice of post-conflict interctions the others (which would have longer and shorter conflict duarations). Non-parametric statistics was applied, for example, Friedman-ANOVA with missing values, which is basically a 10000 times permutated ANOVA (Mundry, 1999).
6.2.2 Additional operational definitions
A post-conflict interaction (PCI) was defined as the first interaction of the focal conflict partner with another individual subsequent to an aggressive interaction. Six possible PCIs were recorded: (a) reconciliation, affiliative PCI between former opponents (n=188); (b) offered consolation, affiliative PCI initiated by a third party (n=164); (c) solicited consolation, affiliative PCI with a third party initiated by a conflict partner (n=176); (d) renewed aggression, aggressive PCI between former opponents (n=174); (e) redirected aggression, aggressive PCI with a third party initiated by a conflict partner (n=88); and (f) third party aggression, aggressive PCI initiated by a third party (n=28). Avoidance of any further interaction was recorded as (g) no PCI, when the focal conflict partner did not interact with any other chimpanzee for the rest of the day (n=58). Affiliative PCIs consisted of friendly behaviours with body contact (e.g. kiss, genital touch, hand holding, embrace, grooming), while aggressive PCIs consisted of threats (e.g. barks, arm wave), non-contact aggression (e.g. displays) and contact aggression (e.g. bits, hits). In the results I distinguish between post-conflict reactions, which are all possible actions after a conflict (from a to g), and post-conflict interactions, which are only the PCIs related to post-conflict management.
6.2.3 Special analysis and statistics
I applied an altered version of the time-rule (Aureli & van Schaik, 1991a; Castles & Whiten, 1998a), to determine whether or not the occurrence of PCIs depended upon the preceding conflict. The time rule implies that a PCI needs to be initiated faster after a conflict than a control interaction. Therefore, I computed four different baselines that represent the normal interaction-intervals in our study group. Baselines for inter-opponent PCIs consisted of a value for each dyad, while baselines for PCIs with third parties had a value for each individual: (a) baseline for reconciliation is the mean interaction-interval of consecutive affiliative interactions for each dyad; (b) baseline of renewed aggression is the mean interaction-interval of any consecutive interaction for each dyad; (c) baseline of consolation is the mean interaction-interval of consecutive affiliative interactions of each individual with any other member of the community; (d) baseline of aggression with third
70 Chapter 6: Choice of post-conflict interctions parties is the mean interaction-interval of any consecutive interactions of each individual with any other member of the community. To compare the latencies of PCIs with their latency particular baselines, I calculated for each PCI event the relative latency (= ), with baseline a relative latency <1 representing PCI events occurring faster than baseline. For each PCI I calculated the mean relative latencies for each dyad (inter-opponent PCIs) or the mean relative latencies for each individual that participated in the conflict and the PCI (PCIs with third parties) and conducted the bootstrap test.
6.3 Results
6.3.1 Are PCIs dependent on the conflict?
Chimpanzees had the choice of six types of PCI and the option of no PCI. In order to determine if the PCIs were implemented as a consequence of the preceding conflict, I
2,5
ns 2
1,5
Baseline 1 * relative latency relative
* * 0,5 * *
0 Solicited Offered Redirected Third Party Reconciliation Renewed Consolation Consolation Aggression Aggression Aggression (=Normal Interaction)
Figure 6.1 Mean relative latencies for each of the six possible post-conflict interactions latency (PCIs) compared to baseline in Taï chimpanzee. Relative latencies ( ) <1 show baseline that the PCI occurred faster than baseline. Error-bars mark the two-tailed 95% confidence interval which was computed from the relative latencies of either dyads (reconciliation and renewed aggression) or individuals. When the baseline was placed outside the confidence interval of latencies, the PCI was significantly different from normal interactions (significance level * is p<0.05, test statistics see 6.3.1).
71 Chapter 6: Choice of post-conflict interctions tested the time-rule of whether or not baseline was included in the confidence interval (95% two-tailed) of the latencies of each PCI. Results revealed that all PCIs except solicited consolation were initiated more quickly than baseline (Figure 6.1). The occurrence of offered consolation, redirected aggression, third party aggression, reconciliation and renewed aggression was triggered by the preceding conflict (all bootstrap tests: p<0.05), while solicited consolation was not different from the baseline (bootstrap test: ns). Thus, solicited consolations were independent of the preceding conflicts. As they were indistinguishable from normal interactions, solicited consolations are no longer referred to as a PCI in this paper. The remaining five PCIs can however be considered as post-conflict management.
6.3.2 The effect of conflict duration
The duration of the preceding conflict, calculated on individual levels, varied among the seven different post-conflict reactions (five PCIs, no PCI and normal interaction; Friedman-ANOVA with missing values: k=7, N=90, permutation=10000, p<0.01). Therefore I ordered the seven possible reactions to a conflict with increasing mean conflict duration (Figure 6.2). I found a distinct relation between conflict duration and
18
] 16 SD M 14
12
10
8
6
4
2 duration of preceding conflict[s 0 Normal Reconciliation No PCI Offered Renewed Redirected Third Party Interaction (n=15) (n=10) Consolation Aggression Aggression Aggression (n=15) (n=15) (n=12) (n=12) (n=11)
Figure 6.2 Average individual conflict duration (sMSD) preceding the different types of post-conflict reactions in Taï chimpanzees. The post-conflict reactions are ordered from the shortest (left) to the longest (right) average duration of preceding conflicts (differences of conflict duration: a: among all post-conflict reactions **, b: same increasing effect for each individual from left to right **, c: affiliative vs. aggressive **, significance level ** is p<0.01, test statistics see 6.3.2 ).
72 Chapter 6: Choice of post-conflict interctions type of post-conflict reaction. This was not due to individual effects, as each individual that was involved in each type of interaction (5 PCIs and normal interaction) revealed the same effect between conflict duration and type of interaction (Page exact: L=651, k=6, N=8, p<0.01). This suggests that conflict duration influenced the choice of PCI. Comparison between aggressive and affiliative PCIs revealed that shorter conflicts were followed by affiliative PCIs while longer conflicts resulted in aggressive PCIs (Wilcoxon exact: T=14, N=15, p<0.01). Multivariate analysis confirmed this result on a dyadic level and revealed conflict duration to be the only predictor (of list in Table 3.2) that influenced 2 whether aggressive or affiliative PCIs followed the conflict (GLZ: LR 4,545 =15.1, p<0.01; Table 6.2a). Thus, the risk of further aggression increased with conflict duration and Taï chimpanzees preferred affiliative PCIs to manage shorter conflicts while aggressive PCIs were used to manage longer conflicts.
6.3.3 Avoidance of interactions (no PCI)
First, I investigated which conditions resulted in no PCI. Conflict partners were more likely to avoid interactions, when resources were not monopolisable by a single competitor, when conflict partners were rare associates or opponents had a small rank 2 difference (GLZ: LR 5,595 =15.9, p<0.01; Table 6.2b). Chimpanzees were thus more likely to implement interactions after conflicts over limited resources and after fighting familiar partners, while fighting over a resource, that was available somewhere else in the party, did not require following social interactions with community members. Subsequently I analysed under which conditions the different interactions were implemented.
6.3.4 Normal interactions – business as usual
Since the time-rule analysis showed that solicited consolations were indistinguishable from normal interactions, I tested under which conflict conditions conflict partners continued with business as usual. Multivariate analysis proved that conflict partners continued with normal interactions after very short conflicts or after conflicts 2 among same sex partners (GLZ: LR 5,545 =27.18, p<0.0001; Table 6.2c). Thus Taï chimpanzees were able to have business as usual after very short conflicts, while longer conflicts needed conflict management related PCIs. Subsequently I investigated the choice amongst the five PCIs that were dependent on the preceding conflict.
73 Chapter 6: Choice of post-conflict interctions
Table 6.2 Factors affecting the occurrence of aggressive versus affiliative post-conflict interactions, avoidance of interactions and normal interactions in Taï chimpanzees. Presented are the significant variables of the best models, their parameters (including the relative frequency ( f p =%) and estimate-coefficients ) and the Wald-statistics.
independent (a) Aggressive PCIs vs. affiliative PCIs variables Wald df p parameter n effect Wald p Conflict 6.02 1 <0.05 continuous 545 IR 0.83 6.02 <0.05 Duration
independent (b) Avoidance of interactions (no PCI) variables Wald df p parameter n [%] Wald p Resource 6.79 2 <0.05 by one 283 5.7 -0.54 6.08 <0.05 Monopolisation by few 159 9.4 0.14 0.38 ns by non 153 12.4 0.40 3.56 <0.1 Association 4.34 1 <0.05 rare 384 9.9 0.38 4.34 <0.05 Index frequent 211 5.7 -0.38 Rank 6.03 2 <0.05 small 211 10.1 0.55 5.78 <0.05 Difference middle 247 8.5 0.06 0.06 ns large 125 4.8 -0.61 3.92 <0.05
independent (c) Normal interaction variable Wald df p parameter n [%] Wald p Conflict 8.81 1 <0.01 continuous 545 KS -1.29 8.81 <0.01 Duration Sex 5.11 1 <0.05 m-m 148 23.6 0.29 Combination f-f 160 33.1 0.29 5.11 <0.05 m-f 194 17.0 -0.29 f-m 43 23.3 -0.29 Model a: Aggressive PCIs vs. affiliative PCIs; R longer conflicts preceding aggressive PCIs than preceding affiliative PCIs.
Model b: No PCI vs. all interactions; f no PCI=8.4%.
Model c: Normal interaction vs. PCIs, f normal interaction=24.0%; S shorter conflicts preceding normal interactions. % = frequency in percent of the dependent variable after conflicts with particular parameter f ( f p ); = overall frequency of dependent variable.
6.3.5 Choosing a post-conflict interaction
6.3.5.1 Reconciliation
Conflict partners were most likely to reconcile with opponents of opposite sex, with 2 high benefit partners or with frequent associates (GLZ: LR 4,414 =20.1, p<0.001; Table
74 Chapter 6: Choice of post-conflict interctions
6.3a). Reconciliation was equally initiated by winners and losers of conflicts (winner: 2 41.7%, loser: 58.3%; Goodness of fit: 1,115 =3.14, ns). After non-contact aggression losers significantly increased their proportion of initiation of reconciliation with conflict intensity, while after contact aggression (parameters 4 and 5) the proportion of losers as 2 initiators was not significantly different from average (GLZ: LR 7,115 =24.2, p<0.01; Table 6.3b). Furthermore, while losers were more likely than winners to initiate reconciliation after conflicts over food, winners showed a tendency to initiate more reconciliation in social and sex contexts (Table 6.3b). Reconciliation was thus preferred among mixed sex partners and among opponents with highly valuable relationships (Figure 6.3). Losers initiated reconciliation following conflicts over food, and their initiation frequency increased with conflict intensity, yet neither conflict partner was more initiative in reconciliation after contact aggression.
Reconciliation Consolation 60
50
40
30
20
10
0
-10
-20 relative differencefrequency in [%]
-30
m-m m-f f-m f-f low medium high rare frequent few some many -40 Sex Combination Relationship Benefit Association Index Competitor Proportion
Figure 6.3 Comparison between the tendencies to initiate reconciliation and consolation in f f Taï chimpanzees. The relative difference in frequency of each parameter ( p ) is shown f
for the significant variables, separately for reconciliation ( ) and consolation ( ). Positive relative differences indicate a positive effect of the parameter on the frequency of the PCI, while negative relative differences point out a negative effect (test statistics see Table 6.3).
75 Chapter 6: Choice of post-conflict interctions
Table 6.3 Factors affecting the choice of affiliative post-conflict interactions in Taï chimpanzees. Presented are the significant variables of the best models, their parameters (including the relative frequency ( f p =%) and estimate-coefficients ) and the Wald-statistics.
independent (a) reconciliation variables Wald df p parameter n [%] Wald p Association 4.64 1 <0.05 rare 267 25.5 -0.26 4.64 <0.05 Index frequent 147 36.1 0.26 Relationship 6.76 2 <0.05 low 142 23.2 -0.29 2.74 <0.1 Benefit medium 201 27.9 -0.22 1.85 ns high 71 45.1 0.51 6.76 <0.01 Sex 4.02 1 <0.05 m-m 113 26.5 -0.24 Combination f-f 107 19.6 -0.24 4.02 <0.05 m-f 161 34.8 0.24 f-m 33 42.4 0.24
(b) Initiator of reconciliation independent variables Wald df p parameter n [%] S Wald p winner Conflict 9.18 2 <0.05 social 57 49.1 0.66 3.74 <0.1 Context sex 26 50 0.70 2.95 <0.1 food 32 21.9 -1.36 8.93 <0.01 Conflict 10.66 4 <0.05 1 26 57.7 1.34 5.54 <0.05 Intensity 2 27 40.7 -0.22 0.22 ns 3 32 28.1 -1.44 7.97 <0.01 4 22 36.4 -0.08 0.02 ns 5 8 62.5 0.40 0.29 ns
independent (c) Offered consolation variables Wald df p parameter n [%] Wald p Competitor 7.78 2 <0.05 few 61 34.4 0.36 5.84 <0.05 Proportion some 209 24.4 -0.02 0.02 ns many 144 18.8 -0.34 6.98 <0.01 Relationship 7.54 2 <0.05 low 142 30.1 0.28 4.96 <0.05 Benefit medium 201 19.9 -0.24 3.77 <0.1 high 71 21.1 -0.04 0.08 ns Sex 3.96 1 <0.05 m-m 113 29.2 0.18 Combination f-f 107 26.2 0.18 3.96 <0.05 m-f 161 19.9 -0.18 f-m 33 18.2 -0.18
Model a: Reconciliation vs. other PCIs; f reconciliation=29.2%.
Model b: Winner vs. loser initiation of reconciliation; f winner=41.7%; f loser=58.3%; R f f loser = 100% - winner; S positive = effect in favour of winner.
Model c: Offered consolation vs. other PCIs; f offered consolation=23.9%. %= frequency in percent of the dependent variable after conflicts with particular parameter f ( f p ); = overall frequency of dependent variable.
76 Chapter 6: Choice of post-conflict interctions
6.3.5.2 Offered consolation
Taï chimpanzees received consolation more often after conflicts with same sex partners, with partners of low benefit or when only a few competitors were present (GLZ: 2 LR 5,414 =12.8, p<0.05; Table 6.3c). Offered consolation and reconciliation occurred after almost complementary conflict situations (Figure 6.3). Moreover, consolation was received after significantly longer conflicts as compared to reconciled conflicts (Wilcoxon exact one- tailed: T=93, N=15, p<0.05; Figure 6.2).
6.3.5.3 Renewed aggression
Renewed aggression did not show any distinct predictors in the multivariate analysis 2 (GLZ: LR 1,414 =2.68, p<0.2). However, after conflicts ending in a draw, conflict partners were more likely to initiate renewed aggression than any other PCI (9 of 21 draws resulted in renewed aggressions, while 96 times renewed aggression was chosen of 418 2 conflict partner initiated interactions; Goodness of fit: 1,21 = 4.76, p<0.05). Renewed aggression was more likely to be initiated by winners than losers of 2 conflicts (winner: 74.4%, loser: 25.3%, Goodness of fit: 1,87 = 21.25, p<0.0001). Losers, however, renewed aggression when they had initiated the conflict, or when they were the 2 dominant partner (GLZ: LR 10,87 =57.1, p<0.0001; Table 6.4a). Therefore, renewed aggression seemed to be the preferred PCI following draws, while dominant initiators renewed aggression when they unexpectedly lost a conflict.
6.3.5.4 Redirected aggression
Aggression was redirected slightly more often after initiators won the conflict (GLZ: 2 LR 2,414 =6.73, p<0.05; Table 6.4b). However this predictor was only marginally significant.
6.3.5.5 Third party aggression
Third parties reacted more often with aggression after very long or very intense 2 conflicts or when the conflicts were in a social context (GLZ: LR 11,414 =48.2, p<0.0001; Table 6.4c). However this PCI was less frequently observed after the most intense non- contact aggression.
77 Chapter 6: Choice of post-conflict interctions
Table 6.4 Factors affecting the choice of aggressive post-conflict interactions in Taï chimpanzees. Presented are the significant variables of the best models, their parameters (including the relative frequency ( f p =%) and estimate-coefficients ) and the Wald-statistics.
(a) Initiator of renewed aggression independent R variables Wald df p parameter n [%] S Wald p winner TInitiator’s 14.49 1 <0.001 winner 77 83.1 3.76 14.49 <0.001 Victory loser 10 10 -3.76 Outcome Winner’s 3.96 1 <0.05 dominant 70 84.3 1.35 3.96 <0.05 Rank subordinate 17 35.3 -1.35
independent (b) Redirected aggression variables Wald df p parameter n [%] Wald p Initiator’s 4.68 2 <0.1 winner 347 18.8 0.83 4.46 <0.05 Victory loser 42 9.5 0.05 0.01 ns Outcome draw 25 4 -0.88 1.56 ns
independent (c) Third party aggression variable Wald df p parameter n [%] Wald p Conflict 14.15 2 <0.001 social 203 10.3 1.19 9.22 <0.01 Context sex 73 1.4 -0.60 0.77 ns food 138 4.3 -0.59 1.56 ns Conflict 11.12 1 <0.001 continuous 414 IU 2.08 11.11 <0.001 Duration Conflict 10.07 4 <0.05 1 84 6.0 -0.01 0.01 ns Intensity 2 97 5.2 0.36 0.71 ns 3 101 2.0 -1.38 5.13 <0.05 4 67 7.5 0.04 0.01 ns 5 65 16.9 0.99 7.56 <0.01
Model a: winner vs. loser initiation of renewed aggression; f winner=73.9%; f f f loser=26.1%; R loser = 100% - winner; S positive = effect in favour of winner; T draws were excluded, since we tested winner vs. loser. f Model b: Redirected aggression vs. other PCIs; redirected aggression=16.9%.
Model c: Third party aggression vs. other PCIs: f third party aggression=6.8%; U longer conflicts preceding third party aggression. % = frequency in percent of the dependent variable after conflicts with particular parameter ( f p ); f = overall frequency of dependent variable.
78 Chapter 6: Choice of post-conflict interctions
CONFLICT DRAW
CONFLICT DURATION SHORT LONG
AFFILIATIVE AGRESSIVE POST-CONFLICT POST-CONFLICT MANAGEMENT MANAGEMENT
RELATIONSHIP BENEFIT ‘EMERGENCY VERY LONG UNEXPECTED HIGH LOW EXIT’: AND INTENSE LOSER STRESS FIGHT SEX COMBINATION REDUCTION MIXED SAME AFTER LONG FIGHT, LEAVE FURTHER AGGRESSION FOCUS OF UNLIKELY LIKELY AGGRESSION
REDIRECTED THIRD PARTY RENEWED RECONCILIATION CONSOLATION AGGRESSION AGGRESSION AGGRESSION
Figure 6.4 Schematic depiction of the evaluation process for post-conflict management in Taï chimpanzees. Several factors influence the optimal choice of PCI between the conflict (starting point: black, on top) and five post-conflict interactions (end points: black, at bottom). Affiliative PCIs are arranged on the left side, while aggressive PCIS are grouped on the right side of the figure (marked in grey). Some post-conflict interactions can be reached following several different effects. Dashed box indicates that effect is one possible conclusion.
6.4 Discussion
6.4.1 Summary of results
Taï chimpanzees applied five of the initially six identified PCIs and no PCI for post- conflict management. I have summarised the results of the choice of PCI in Figure 6.4, in order to provide an accessible overview of the post-conflict management related PCIs.
79 Chapter 6: Choice of post-conflict interctions
This suggests that Taï chimpanzees show a clear-cut evaluation process in selecting a PCI. Conflict participants avoided further interactions after conflicts over non- monopolisible resources or among rare associates. Very short conflicts did not require post-conflict management, as Taï chimpanzees continued with business as usual. Below, I discuss the influence of the advantages and disadvantages on the choice of PCI and compare the post-conflict management of Taï chimpanzees with those of other mammals. Finally I extract general rules for the evaluation process and discuss their validity.
6.4.2 Business as usual
Solicited consolation was to the only possible PCI that did not serve a function in the post-conflict management of Taï chimpanzees. As solicited consolation followed very short conflicts and it was indistinguishable from normal interactions it seems likely that any costs were minimal after very short conflicts. Moreover, conflict partners were able to interact non-aggressively after shorter conflicts while longer conflicts led to aggressive PCIs, and the longest conflicts usually resulted in attacks by third parties. In fact there was almost a linear relationship between increasing length of conflict and likelihood of further aggression as well as the level of escalation. On the one hand this may reflect the motivation for escalation to aggression within the dyad and therefore within the party. Competition over beneficial resources was more likely to lead to aggression than less beneficial ones (chapters 425; Janson, 1988a; Preuschoft & van Schaik, 2000; Wittig & Boesch, 2003a) and the length of the conflict was dependent on the value of the resource (chapter 526; Wittig & Boesch, 2003b). On the other hand this may also suggest that stress and tension created by the conflict increased with conflict duration. Although some studies have tried to find relationships between post-conflict stress and conflict intensity (contact vs. non-contact aggression), but failed to detect them (Macaca fascicularis: Aureli, 1997; Macaca fuscata: Kutsukake & Castles, 2001; Papio anubis: Castles & Whiten, 1998b), they all neglected conflict duration as a possible predictor. It remains unclear whether or not there is also a positive correlation between conflict duration and stress or tension, besides the correlation with escalation to aggression. I suggest that conflict duration is possibly a good predictor for the level of induced stress.
25 see 4.3.2 26 see 5.3.2.2
80 Chapter 6: Choice of post-conflict interctions
6.4.3 Avoiding further interactions
While Taï chimpanzees continued with business as usual after a seemingly negligible stress response, they avoided any further interaction after fighting over non-limited resources. Although non-limited resources do not usually cause conflicts, aggression can arise in situations where many competitors are present (chapter 427; Janson, 1988b; Wittig & Boesch, 2003a). Usually, with a non-limited resource, the possible disadvantages of engaging on a PCI are likely to out weigh the possible benefits gained. Therefore, moving to a different feeding spot or engaging in other activities (e.g. resting, travelling) seems to be a reasonable post-conflict reaction. Whether or not the avoidance of any further interaction was dependent on the preceding conflict, was not possible to test with the time-rule method. Thus I can neither definitely include nor exclude no PCI from the conflict management. Although Taï chimpanzees sometimes avoided interactions or continued with business as usual after conflicts, most of the conflicts required a PCI.
6.4.4 Reconciliation
Taï chimpanzees reconciled more often with cooperative partners and frequent associates. This is strong evidence that the choice for reconciliation was due to its advantage of relationship repair. Reconciliation with high value partners is common in many primate species, such as among cooperative partners (e.g. Cords & Thurnheer, 1993), alliance partners (e.g. Watts, 1995a), kin (e.g. York & Rowell, 1988; Cheney & Seyfarth, 1989; Kappeler, 1993; Castles & Whiten, 1998a), partners of affiliation (e.g. Cords & Aureli, 1993; Watts, 1995a; Castles et al., 1996; Schino et al., 1998; Call et al., 1999), and frequent associates (e.g. de Waal & Yoshihara, 1983; Aureli et al., 1989). Wild chimpanzees in Budongo, Uganda, preferably reconciled with mating partners (Arnold & Whiten, 2001). Taï chimpanzees might even consider potential mating partners as valuable since they reconciled more often in mixed sex dyads.
Reconciliation was also more likely when further aggression seemed to be less likely. The result suggests that the disadvantage of risking recurring aggression, when approaching the former opponent, also shaped the choice for reconciliation. Similar results were found in other studies, where low intensity conflicts (non-contact aggression) were reconciled more often than high intensity ones (e.g. Eulemur fulvus: Kappeler, 1993; Pan troglodytes (Budongo): Arnold & Whiten, 2001). Furthermore, when Budongo
27 see Figure 4.2
81 Chapter 6: Choice of post-conflict interctions chimpanzees accepted the outcome of conflicts by emitting greetings, they were more likely to reconcile afterwards (Arnold & Whiten, 2001). Tufted capuchins (Cebus apella) reconciled only after non-food conflicts although almost 90% of their conflicts were over clumped food (Verbeek & de Waal, 1997). This might show that tufted capuchins do not dare to reconcile when the chance of further aggression is high. Since macaques and baboons reconciled less after food than after non-food conflicts (Macaca fascicularis: Aureli, 1992; Macaca maurus: Matsumura, 1996; Papio anubis: Castles & Whiten, 1998a), Aureli et al. (2002) argued that food conflicts may not damage the relationship of opponents and therefore there is less need for reconciliation. However our multivariate analysis for Taï chimpanzees showed that the occurrence of reconciliation was not different between food and social context.
Losers initiated more reconciliation after food conflicts, while the initiation rates of winners and losers were equal in social and sex contexts. After food conflicts losers appeared to be the main profiteers of reconciliation. Taï chimpanzees, as well as Budongo chimpanzees (Arnold & Whiten, 2001), probably tried to access food with a second attempt after they calmed the former opponent. In contrast after social and sex conflicts the social leverage of subordinates seemed to be similar to those of dominants (Hand, 1986; Lewis, 2002). Since both potential mating partners can avoid copulation, and the support of a high value partner might be needed in future conflicts with other individuals, both partners should have an interest in repair the relationship after social and sex conflicts. However losers initiated more reconciliation with increasing conflict intensity, but after contact aggression neither conflict partner showed an initiation preference. This suggests that losers are more restricted by increasing conflict intensity than dominants. Since subordinate partners risk more in approaching the former opponent than their dominant partners, they may decrease their initiation rate when the dominant partners is highly motivated to fight again. A conflict partner using contact aggression, however, might signal a willingness to risk more to gain the benefit of the resource. Losers, therefore, may be more hesitant to approach the former opponent for reconciliation than after non-contact aggression. The results suggest that increasing conflict intensity increasingly disturbed the relationship of opponents in Taï chimpanzees. The advantages of relationship repair and of accessing the resource through affiliation as well as the disadvantage of further aggression shaped the proportion of loser to winner initiated reconciliations. Different proportions of victim initiated reconciliation among species might be explained by inter-specific differences in the risk of further aggression or the
82 Chapter 6: Choice of post-conflict interctions advantage of cooperation or affiliation (more aggressor initiated: e.g. Cebus capucinus: Leca et al., 2002; Papio anubis: Castles & Whiten, 1998a; Carpa hircus: Schino, 1998; no difference: e.g. Macaca fuscata: Aureli et al., 1993; Macaca silenus: Abegg et al., 1996; Papio papio: Petit & Thierry, 1994b; more victim initiated: e.g. Colobus guereza: Björnsdotter et al., 2000; Macaca arctoides: de Waal & Ren, 1988; Crocuta crocuta: Wahaj et al., 2001).
6.4.5 Consolation
Consolation was offered in almost the contrary conflict situations as compared to reconciliation (Figure 6.3). Taï chimpanzees seemed to offer consolation when reconciliation was either not beneficial or was too risky for conflict participants. Since low benefit partners were those that did not share food and did not support each other, opponents would gain little from relationship repair. Similarly same sex dyads would not provide much benefit in terms of potential mating partners. Additionally, same sex opponents probably have a higher tendency for further aggression than mixed sex opponents, as competing aggressively over the same resource again might be very unlikely when the resource provides different benefits to each opponent. Indeed same sex dyads in Taï chimpanzees fought more often than mixed sex dyads (chapter 528; Wittig & Boesch, 2003b). Since the risk of further aggression increased with length of conflict and consolation followed longer conflicts more than reconciliation, consolation was probably offered when approaching former opponents was too risky. Thus consolation may have been substituted for reconciliation when further aggression was more likely. In gorillas (Gorilla gorilla) consolation might also be a substitute for reconciliation for the same reason. Consolation mostly occurred after conflicts among gorilla females (Watts, 1995b), three quarters of which were conflicts ending without any submissive sign that showed the acceptance of the outcome (Watts, 1994). This preference for consolation when conflicts could be followed by further aggression might explain why female gorillas mainly sought consolation from males (Watts, 1995b).
6.4.6 Renewed aggression
Undecided conflicts in Taï chimpanzees were usually followed by renewed aggression. In addition losers of conflicts mostly initiated renewed aggression when they were dominant losers or initiators that lost, and thus they had a good chance of winning
28 see Table 5.2
83 Chapter 6: Choice of post-conflict interctions the new conflict (chapter 529; Wittig & Boesch, 2003b). The advantage of accessing the resource in a second aggressive attempt thus seemed to be a main factor for the choice of renewed aggression. However, why did so many winners renew the aggression? In contrast to Taï chimpanzees, redfronted lemurs (Eulemur fulvus) had a higher tendency to reconcile, when the conflict was undecided (Kappeler, 1993). Reconciliation is the only other PCI that allows possible access to the resource but in addition excludes costs of aggression. Redfronted lemurs might face a lower risk of further aggression than chimpanzees, which enables them to arrange undecided conflicts peacefully while chimpanzee winners might react aggressively when the former opponent, aiming to reconcile, approaches too early. For other mammals there is no evidence that undecided conflicts result in any specific PCI (e.g. Cercopithecus athiops: Cheney & Seyfarth, 1989; Lemur catta: Kappeler, 1993; Capra hircus: Schino, 1998).
6.4.7 Redirected aggression
Redirected aggression was the only PCI in the post-conflict management of Taï chimpanzees that showed an ambiguous pattern. Taï chimpanzees redirected aggression marginally more often when the initiator won, so after expected outcomes (see: renewed aggression). This casts doubts on the idea that individuals would redirect their frustration to third parties. However, redirected aggression is the only aggressive PCI with evidence for stress reduction (Aureli & van Schaik, 1991b). Since preceding conflicts of aggressive PCIs were longer than those of affiliative PCIs, Taï chimpanzees might use redirected aggression as a kind of emergency-exit to reduce their stress after long conflicts, where either reconciliation or consolation were too risky, and to sneak out of the focus of aggression. Thus the emergency-exit strategy pays for all dyads that engaged in long conflicts regardless of their relationship. Other primate species might also employ the emergency-exit, since redirected aggression is usually not affected by the relationship of opponents (not affected: Gorilla berengei: Watts, 1995b; Macaca fuscata: Aureli et al., 1993; Macaca sylvanus: Aureli et al., 1994; Papio anubis: Castles & Whiten, 1998a; but, affected: Cercopithecus athiops: Cheney & Seyfarth, 1989).
6.4.8 Third party aggression
Third party aggression occurred after the longest and the most intense conflicts in Taï chimpanzees. On one hand, individuals might support their coalition partners after the
29 see 5.3.1.2
84 Chapter 6: Choice of post-conflict interctions conflict has already been decided (winner-support: de Waal, 1978) or general high arousal within the party might induce general escalation of aggression (e.g. Aureli & van Schaik, 1991b; Cords, 1992; Preuschoft & van Schaik, 2000). On the other hand conflict partners, especially losers, were probably weakened after such long and intense fights. These vulnerable individuals would be easy targets of aggression for their frequent competitors, for example rank neighbours. Since third party aggression also occurred more often after social conflicts, I was unable to rule out any of these possibilities.
6.4.9 Choice of PCI: generalised rules
Following the post-conflict management found in Taï chimpanzees (Figure 6.4) I try to extrapolate generalised rules for the choice of PCIs: Reconciliation appeared to be chosen when a disturbed relationship is costly and further aggression is unlikely. However when initiation of reconciliation appears to be too risky, consolation seems to substitute for reconciliation. In situations where losing partners perceive a chance to access the resource by further aggression, they might renew aggression to gain the benefit of the resource in a second attempt. If the social tension has reached a level where attempting peaceful post-conflict management is too risky, conflict partners might take the emergency-exit and redirect aggression to deflect aggressive attention to third parties. When conflict partners seem to be weakened, third parties might seize the opportunity to defeat a frequent competitor.
The interaction of advantages and disadvantages of PCIs might explain why despotic macaque species reconcile less often than egalitarian ones (Thierry, 2000). As both select usually high value partners for reconciliation (Macaca arctoides: de Waal & Ren, 1988; Macaca fascicularis: Aureli et al., 1989; Macaca fuscata: Aureli et al., 1997; Macaca mulatta: de Waal & Yoshihara, 1983; Macaca nemestrina: Judge, 1991), relationship repair seems to be important in both hierarchy types. However the approach of former opponents in despotic macaques is likely to carry higher risks of further aggression, as tolerance levels in despotic societies are less clear-cut. Therefore approaching a former opponent can be easily mistaken as a counterattack and lead to further aggression. Additionally aggression in egalitarian macaques is milder and less costly for the victim (Thierry, 1986; Thierry, 2000). Therefore a failed reconciliation attempt in despotic macaques is more costly than in egalitarian ones. Following the generalised rules, one would expect consolation to substitute for reconciliation in despotic macaques. However consolation seems not to be part of the macaque repertoire (Watts
85 Chapter 6: Choice of post-conflict interctions et al., 2000), probably due to social or cognitive constraints (de Waal & Aureli, 1996). Despotic macaque species probably use redirected aggression instead to deflect the disadvantages of aggression to third parties and perhaps to reduce their own stress (e.g. Macaca fascisularis: Aureli & van Schaik, 1991b; Aureli, 1992; Macaca fuscata: Aureli et al., 1993). In egalitarian species, however, redirected aggression is either less frequent (Thierry, 1985; Thierry et al., 2000a) or is not observed at all (e.g. Macaca sylvanus: Aureli et al., 1994), as they reconcile more likely.
In consequence our results indicate that Taï chimpanzees have a clear-cut evaluation process as they seem to weigh carefully advantages against disadvantages to select the best PCI to the experienced conflict situation. Conflict management in Taï chimpanzees appears to take into account both cost and benefit before escalating a conflict of interest (chapter 530; Wittig & Boesch, 2003b) and selecting the best PCI to handle remaining costs. Since both conflict partners have the possibility to vary their post- conflict cost, conflict management opens the door for negotiation to lessen the consequences of conflicts. Thus post-conflict management seems to be an important negotiation tool for social mammals to enable advantageous social living despite the existing conflicts of interest.
30 see 5.4.5
86 Chapter 731:
How are relationships repaired ?
Figure 7.0 Strong relationships between siblings. Subadult male Lefkas embraces his younger brother Leonardo (Photo by R. Wittig).
31 Corresponding with: Wittig, R.M. & Boesch, C. (in review). How to repair relationships in wild chimpanzees (Pan troglodytes). Ethology. Chapter 7: How are relationships repaired ?
7.1 Introduction
7.1.1 Why repair relationships?
Aggressive interactions damage the relationship of conflict partners, since they create stress and interrupt tolerance levels between opponents (Maestripieri et al., 1992; Aureli et al., 2002). Damaging a relationship is costly as it jeopardises the benefits related to relationships (e.g. food-sharing, grooming, support). Reconciliation, the friendly post- conflict interaction among former opponents, appears to repair relationships by restoring tolerance to normal levels (Cords, 1992) and reduces stress induced by aggression (e.g. Aureli & van Schaik, 1991b; Castles & Whiten, 1998b; Kutsukake & Castles, 2001). Therefore reconciliation provides considerable benefit for high value partners. Although reconciliation is beneficial, it nonetheless entails costs, as for example the risk of further aggression (Aureli & van Schaik, 1991b; Cords, 1992). Such potential costs may even prevent former opponents from reconciling conflicts under certain circumstances (chapter 632; Wittig & Boesch, 2003d).
Although a consensus exists over the benefits of reconciliation, almost nothing is known about how opponents achieve the benefit that is provided by reconciliation. (Call et al., 1999) found two clusters of conciliatory behaviours, one that seemed to serve relationship repair between high value partners, while the other was used during reconciliation by opponents of unspecific relationships. Moreover other studies have reported that opponents preferred socially intense behaviours for reconciliation, though the repertoire of reconciliation also contained socially less intense behaviours (Petit & Thierry, 1994a; Abegg et al., 1996). These results suggest that socially intense behaviours are more effective in achieving reconciliation than others. Thus it could be that some conflict situations need more intense or complex repair than others and former opponents vary the behaviours used to achieve reconciliation accordingly.
7.1.2 What damages relationships?
When comparing the pattern of reconciliation within and between species, I found great variation (Appendix A). Intra-specific variation is of special interest and probably varies for several reasons. For example it may indicate that individuals reconcile differently under different levels of relationship damage. A strongly damaged relationship
88 Chapter 7: How are relationships repaired ? might require a long and intensive reconciliation in order to restore tolerance levels and reduce stress. Therefore, the investment of opponents in the repair of their relationship should positively correlate with the strength of the damage done by the preceding conflict.
Two factors, conflict intensity and relationship quality, potentially affect the damage to relationships. Rhesus macaques (Macaca mulatta) showed a stronger behavioural stress response to heavy aggression as compared to light aggression (de Waal & Yoshihara, 1983). Moreover, severe aggression in longtailed macaques (Macaca fascicularis) was reconciled more often than milder aggression (Koyama, 2001), and the intensity of conflicts affected the tendency to reconcile in Japanese macaques (Macaca fuscata: Schino et al., 1998). These results indicate that damage to a relationship increases with the intensity of the conflict and probably other conflict characteristics.
In contrast, other studies have failed to detect a relationship between conflict intensity and behavioural stress responses (Macaca fascicularis: Aureli, 1997; Macaca fuscata: Kutsukake & Castles, 2001; Papio anubis: Castles & Whiten, 1998b). Instead, opponents were more stressed either after conflicts with frequent as opposed to rare associates (Aureli, 1997) or after conflicts with kin as opposed to non-kin (Kutsukake & Castles, 2001). These results indicate that damage to a relationship increases with relationship quality or other relationship characteristics. Since the results are not consistent across studies as to which effects are causing damage to relationships, a multivariate approach is required.
7.1.3 How to repair the damage? Variation in reconciliation
Variation is also found in the timing of reconciliation, probably due to the fact that opponents sometimes have to compromise high benefits of reconciliation with accompanied high costs. For example opponents should quickly re-establish mutual tolerance as a precondition for beneficial interactions (Cords & Thurnheer, 1993), or minimise the time being exposed to the costs of stress (Sapolsky, 1998). Therefore reconciliation should be initiated soon after a conflict. However, since stress and tension decrease with time (Aureli & van Schaik, 1991b), aggression can revive when reconciliation is initiated too quickly after the conflict (chapter 633; Wittig & Boesch,
32 see 6.4.9 33 see 6.3.5.3
89 Chapter 7: How are relationships repaired ?
2003d). Thus both extremes of latency of reconciliation contain high costs and benefits such that the timing of initiation of reconciliation is likely to be a compromise. Therefore both costs and benefits need to be considered for the understanding of the variation within reconciliation.
Reconciliation has been studied in almost three dozen different mammal species. Conflict partners of most species reconciled preferably with high value partners (Aureli et al., 2002), and most species also showed evidence for intra-species variation within reconciliation in different features (Appendix A). First, for most species variability was found for the duration between conflict and reconciliation and some authors specifically stated that the latency of reconciliation was not equal (Gust & Gordon, 1993; Arnold & Whiten, 2001). Second, many species showed several typical behaviours for reconciliation (Appendix A), ranging from short touches to long bouts of allogrooming, indicating that the duration of reconciliation varies greatly (de Waal & Ren, 1988; Aureli et al., 1993). Second, opponents sometimes employed several different behaviours in the same reconciliation, therefore increasing what I refer to here as the complexity of reconciliation. Stumptailed macaques (Macaca arctoides), for example, combined allogrooming with contact sitting (Call et al., 1999), sooty mangabeys (Cercocebus torquatus) joined allogrooming and embracing in the same reconciliation (Gust & Gordon, 1993) and female gorillas (Gorilla gorilla) reacted with combinations of several behavioural elements to aggression by males (Watts, 1995a).
Finally, neither victims nor aggressors were exclusively initiating reconciliation in any of the conducted studies (Appendix A). Both partners can profit from reconciliation, since both face stress after aggression (Aureli, 1997). However, a subordinate opponent might profit more from a good relationship with a dominant partner than the reverse, as relationships have asymmetrical benefits for each partner (Cords & Aureli, 2000).
Wild chimpanzees are a good model to investigate the function of and the reasons for intra-species variation within reconciliation, as they use reconciliation (chapter 634; Arnold & Whiten, 2001; Wittig & Boesch, 2003d) and they optimise conflicts based on cost and benefit strategies (chapter 535; Wittig & Boesch, 2003b). The aim of this study is to test if the function of reconciliation in wild chimpanzees is about relationship repair.
34 see Figure 6.1 35 see Figure 5.5
90 Chapter 7: How are relationships repaired ?
Furthermore, I aim to understand how the variation within reconciliation, in terms of latency, duration, complexity and initiator of reconciliation, is influenced by the preceding conflict and the damage to relationships.
7.2 Specific methods
7.2.1 Data and test conditions
From the 1071 collected conflicts 791 were conflicts between adults and had complete information on the conflict and the latency, duration and complexity of post- conflict interactions. The resulting balanced data set of 146 out of a total of 178 conflicts with reconciliation was used for the multivariate analyses. For each balanced data point I calculated the mean of latency and duration of reconciliation and then I assigned each mean to the appropriate category of quartiles (Table 7.1). In addition I used the median for the complexity of reconciliation. All variables of Table 3.2, plus the interactions of initiator’s sex with recipient’s sex, rank difference with initiator’s rank and conflict context with competitor proportion, were considered simultaneously as independent variables in each of the multivariate analyses.
Table 7.1 Three variables of the variability of reconciliation including the parameters.
Variable Distribution Definition and parameters time(t) between the end of the conflict and the start of Latency of ordinal the reconciliation split into categories which Reconciliation multinomial corresponded with the four quartiles: very short for t?20s; short for 20s
time(t) between the start and the end of the Duration of ordinal reconciliation split with into categories which Reconciliation multinomial corresponded with the four quartiles: very short for t?4s; short for 4s
number of different behavioural elements used during Complexity of ordinal reconciliation: simple = one behavioural element; Reconciliation multinomial advanced = two behavioural elements; complex = three and more behavioural elements
91 Chapter 7: How are relationships repaired ?
7.2.2 Additional operational definitions
The first affiliative interaction between former opponents after a conflict was defined as reconciliation when no other interaction with participation of the focus individual was observed between conflict and reconciliation. I took three measurements to describe variation found within reconciliation: (a) the latency of reconciliation, which was the time elapsing between conflict and reconciliation, (b) the duration of reconciliation, which was the time between the start of the first affiliative interaction to the end of the last affiliative interaction of reconciliation, and (c) the complexity of reconciliation, which was the number of different behavioural elements used during the reconciliation. Since both latency and duration of reconciliation showed a strongly skewed distribution (Figure 7.1), I defined for each of them four ordinal categories that corresponded with their quartiles (Table 7.1).
(a) Latency of reconciliation 100 90 80 70 60 50 40 30 % reconciliation <= time <= % reconciliation 20 10 0 0 2 4 6 8 1012141618202224262830 time [min]
(b) Duration of reconciliation
100
90 80
70 60
50
40 30 % reconciliation <= time <= reconciliation % 20 10
0 0 12 24 36 48 60 72 84 96 108 120 132 144 156 168 180 time [s]
Figure 7.1 Distribution of latency and duration of reconciliation over time in Taï chimpanzees (N=178). Graph (a) shows the percentage of reconciliation that was initiated within a certain time, while graph (b) shows the percentage of reconciliation that continued for a certain time.
92 Chapter 7: How are relationships repaired ?
7.2.3 Measuring reconciliation
I described the occurrence of reconciliation using two methods. Firstly, the Corrected Conciliatory Tendency (CCT: Veenema et al., 1994), in order to show that reconciliation did not occur by chance. Therefore I computed a baseline that represents the mean interval among the consecutive affiliative interactions for each dyad. Then I A D calculated the CCT for each dyad using the following formula: CCT = , with A # conflicts
(attracted pairs) = #reconciliation with latency < baseline, and D (dispersed pairs) = #reconciliation with latency > baseline. The relative preponderance of attracted pairs (CCT>0) showed that the dyad had a tendency to reconcile.
Secondly, I applied an altered version of the time-rule (Aureli & van Schaik, 1991a; Castles & Whiten, 1998a), to determine whether or not the occurrence of reconciliation depended on the preceding conflict. The time rule implies that reconciliation needs to be initiated faster after a conflict than baseline. To compare the latencies of each dyad with latency their baseline, I calculated for each dyad the relative latency (= ). Relative baseline latencies of reconciliation smaller than one indicated that the occurrence of reconciliation was dependent on the preceding conflict.
To demonstrate the function of reconciliation I calculated relative latencies per dyad for two other types of interactions: (a) the first affiliative inter-opponent interaction after reconciliation (post-conciliatory interaction, n=113 in 43 dyads), and (b) the first affiliative inter-opponent interaction after unreconciled conflicts (post-non-conciliatory interaction, same 43 dyads), where for example consolation or redirected aggression occurred. Relative latencies of post-conciliatory interactions are expected to equal one, if reconciliation sets tolerance levels to normal. In contrast, relative latencies of post-non- conciliatory interactions should be greater than one, as disturbed tolerance levels were not repaired.
93 Chapter 7: How are relationships repaired ?
7.3 Results
7.3.1 Occurrence of reconciliation
I observed 791 conflicts in 90 of the 105 adult dyads, of which 48 dyads used reconciliation at least once, while the other 42 dyads never implemented reconciliation (Table 7.2). Almost one out of four conflicts (178 of 791) was followed by reconciliation. The mean conciliatory tendency (CCT) of dyads was 15.85% for Taï chimpanzees . Conciliatory tendencies varied greatly between the sexes. In general, dyads including
males showed a higher CCT than dyads including females (CCTm=25.34%; CCTf=15.94%;
Permutation test: nm=46, nf=84, p<0.05). Female-female dyads had a lower CCT than
male-female dyads (CCTff=5.92%; CCTmf=26.96%; Permutation test & Bonferroni: nff=44,
nmf=40, p<0.016). In contrast male-male dyads (CCTmm=14.56%) did not differ from
male-female dyads (Permutation test & Bonferroni: nmm=6, nmf=40, ns) and female-
female dyads (Permutation test & Bonferroni: nmm=6, nff=44, ns).
Table 7.2 Conciliatory Tendency (CCT) of adult Taï chimpanzees. males females CCT [%] Brutus Macho Marius Nino Belle Castor Dilly Fossey Goma Loukoum Mystere Narcisse Perla Ricci Venus a/d/n interactions* 9 38 13 0 100 0 33 33 100 0 67 Brutus ------1/0/10 3/0/5 2/0/14 0/0/1 1/0/0 0/0/1 1/0/2 1/0/2 1/0/0 0/0/2 2/0/1 12 1 38 20 0 25 50 44 23 0 0 17 22 Macho 46/16/190 9/8/87 6/0/10 2/0/8 0/0/10 2/0/6 1/0/1 4/0/5 4/1/8 1/1/0 0/0/3 2/1/3 2/0/7 15 -11 0 20 33 100 36 14 0 43 60 43 Marius 2/0/11 1/2/6 0/0/5 1/0/4 2/0/4 1/0/0 5/0/9 1/0/6 0/0/7 3/0/4 3/0/2 9/0/12 0 25 13 0 50 16 20 25 0 20 Nino --- 0/0/3 2/0/6 1/0/7 0/0/2 2/0/2 4/1/14 1/0/4 1/0/3 0/0/3 2/0/8 10 0 0 0 0 0 20 0 0 Belle --- 1/0/9 0/0/1 0/0/2 0/0/1 0/0/6 0/0/1 1/0/4 0/0/1 0/0/2 0 0 0 -33 0 0 Castor ------0/0/1 0/0/1 0/0/5 0/1/2 0/0/1 0/0/3 33 50 0 0 -25 0 33 Dilly --- 1/0/2 2/0/2 1/1/3 0/0/1 0/1/3 0/0/1 3/0/6 0 0 50 0 0 Fossey ------0/0/2 0/0/2 1/0/1 0/0/1 0/0/2 0 0 0 0 0 Goma --- 0/0/3 0/0/1 0/0/2 0/0/1 0/0/2 0 -11 0 0 33 Loukoum 0/0/5 0/1/8 0/0/7 0/0/3 1/0/2 100 0 0 0 Mystere 1/0/0 0/0/6 0/0/2 0/0/2 0 Narcisse ------0/0/1 0 0 Perla 0/0/2 0/0/2 Ricci --- Venus Top line (bold) = CCT. Bottom line = number of attracted (a), dispersed (d) interactions and conflicts without reconciliation (=neutral interactions, n) of each dyad for the observation period (see 7.2.3). --- = no conflict observed in this dyad.
7.3.2 Function of reconciliation
Reconciliation was demonstrated following the time-rule (Figure 7.2), since dyads of all three sex combinations were reconciling faster than baseline (Bootstrep tests: m-m,
94 Chapter 7: How are relationships repaired ? n=5, p<0.05; m-f, n=29, p<0.05; f-f, n=9, p<0.05). In contrast, post-reconciliatory interaction occurred as fast as baseline for dyads with male participation (Bootstrep tests: m-m, n=5, ns; m-f, n=29, ns), while for female-female dyads post-reconciliatory interactions were faster than baseline (Bootstrep test: f-f, n=9, p<0.05). Friendly interactions after unreconciled conflicts however occurred slower than baseline within each sex combination (Bootstrep tests: m-m, n=5, p<0.05; m-f, n=29, p<0.05; f-f, n=9, p<0.05). Hence, aggression disturbed the usual interaction pattern among opponents, since opponents of unreconciled conflicts needed longer than usual to interact again. Reconciliation reinstated normal interaction rates for dyads with males but not in female- female dyads (Figure 7.2).
reconciliation post-conciliatory interaction post-non-conciliatory interaction
8
7
6
5
4
3
2 Baseline 1
relative latency of interaction (interval/baseline) relative of interaction latency 0 male-male male-female female-female
Figure 7.2 Comparison between latency of reconciliation ( ), latency of the first post- conciliatory interaction ( ) and latency of the first unreconciled interaction (D) with baseline in Taï chimpanzees. Mean relative delays (interval between interactions divided by baseline) of the dyads are presented for each sex class, while relative baseline is one. Confidence intervals (95%) are shown for a two-sided distribution, calculated with bootstrap method (test statistics see 7.3.2).
Finally I checked whether conciliatory tendencies increased with the relationship benefit of opponents (Figure 7.3). I merged small and medium relationship benefit into one category of low benefit and compared it with high benefit partners. Male opponents of high relationship benefit showed higher CCTs than those of lower relationship benefit
(Permutation test: nlow=4, nhigh=2; p<0.05) and the same was true for female opponents
95 Chapter 7: How are relationships repaired ?
(Permutation test: nlow=38, nhigh=6; p<0.05). For mixed sex dyads, however, no difference was observed (Figure 7.3; Permutation test: nlow=32, nhigh=8; ns).
35 low benefit high benefit
30 * 25
20 * 15 verage dyadic CCT verage dyadic
a 10
5
0 m-m m-f f-f Sex class
Figure 7.3 Comparison of Conciliatory Tendencies between high ( ) and low (D) benefit partners within each sex class of Taï chimpanzees. Low benefit partners have a small and medium relationship benefit. *indicate significant differences between the samples (p-values see 7.3.2).
Table 7.3 Factors affecting (a) the latency and (b) the duration of reconciliation in Taï chimpanzees. The table shows the significant variables and the estimate-coefficient () of the parameters.
R (a) Latency of reconciliation independent very short short long very long Wald df p parameter n Wald p variables [%] [%] [%] [%] Receiver’s Sex 9.96 1 <0.01 male 63 11.1 25.4 25.4 38.1 -0.59 9.96 <0.01 female 83 36.1 24.1 24.1 15.7 0.59 Rank Difference 10.73 2 <0.01 small 76 30.3 28.9 21.1 19.7 0.70 9.41 <0.01 middle 46 19.6 17.4 26.1 36.9 -0.41 3.13 ns big 24 20.8 25 33.3 20.9 -0.29 0.97 ns Conflict Duration 4.61 1 <0.05 continuous 146 KGII-0.98 4.61 <0.05
(b) Duration of reconciliationS independent very short short long very long Wald df p parameter n Wald p variables [%] [%] [%] [%] Receiver’s Sex 7.68 1 <0.01 male 63 23.8 19.1 22.2 34.1 -0.61 7.68 <0.01 female 83 26.5 32.5 22.9 18.1 0.61 Rank Difference 12.75 2 <0.01 small 76 23.7 32.9 22.4 21.0 1.08 12.66 <0.001 middle 46 28.2 26.1 21.7 24.0 -0.24 0.49 ns big 24 25.0 8.3 25.0 41.7 -0.84 4.75 <0.05 Conflict Context 12.66 2 <0.01 social 81 22.2 18.5 23.5 35.8 -0.29 1.29 ns sex 29 6.9 55.2 27.6 10.3 -0.68 4.36 <0.05 food 36 47.2 22.2 16.7 13.9 0.97 12.61 <0.001 Model a: Latency of reconciliation (very short – short – long – very long); R very short (1st quartile): t?20s; short (2nd quartile): 20s
96 Chapter 7: How are relationships repaired ?
7.3.3 Variation within latency and duration of reconciliation
More than 50% of the reconciliations were initiated within 3 minutes (median=144s) after the conflicts (range=2s-251min; Figure 7.1a). Reconciliation was initiated more quickly after conflicts with female victims, after conflicts between conflict partners of a small rank difference and after short conflicts, while conflict partners took longer to initiate reconciliation after conflicts with male victims and long conflicts (GLZ: 2 LR 5,146 =30.0, p<0.0001; Table 7.3a).
duration of reconciliation [s] ( ) Median
-12-10-8-6-4-20246810121416182022242628303234363840420 12 24 36 48 22
20 food Conflict 18 sex context 16 social
14
12 Sex of female 10 receiver male Y Data 8
6 large Rank4 middle difference 2 small 0
-144-1080 -72 -36144 0 36 72 108 288144 180 216 252432 288 324 360 396 432576 468 504 latency of reconciliation [s] (D) Median
Figure 7.4 Median latency (D: bottom scale) and duration ( : top scale) of reconciliation for the influencing variables in Taï chimpanzees. Median latencies and durations of the variable parameters are shown as deviations of the overall median values for latency ~ ~ ( X =144s) and duration ( X =12s) of reconciliation, indicated by the median line. Deviations to the left mark that latency or duration for the parameter were shorter than median, while deviations to the right show latency or duration were longer than median (test statistics see Table 7.3).
97 Chapter 7: How are relationships repaired ?
About 50% of reconciliations were shorter than 12s (range =1s-1026s; Figure 7.1b). Reconciliation was shorter after food conflicts, after conflicts between opponents with a small rank difference and after conflicts with female victims, while reconciliation was longer after conflicts over sex (with medium duration), after conflicts between opponents 2 with a large rank difference and after conflicts directed to males (GLZ: LR 9,146 =38.3, p<0.0001; Table 7.3b).
Consequently either opponents used short reconciliation soon after the conflict, when they had a small rank difference and female victims, or opponents initiated long reconciliation with some delay, when they had a big rank difference and male victims (Figure 7.4). Additionally long fighting resulted in late reconciliation while food conflicts were reconciled more quickly than conflicts over sex.
gentle aggression (level 1) hard fighting (level 5) 20
15
10
5
0 simple advanced complex -5
-10
-15
-20 relative occurrence of reconciliation [%] reconciliation of occurrence relative -25
Figure 7.5 Complexity of reconciliation depending on the intensity of the preceding conflicts in Taï chimpanzees. The relative occurrence of reconciliation is shown for the two significant parameters of gentle aggression (level 1: ) and hard fighting (level 5: D). Positive values indicate higher frequencies and negative values indicate lower frequencies of reconciliation than average (test statistics see Table 7.4).
98 Chapter 7: How are relationships repaired ?
7.3.4 Variation within Complexity of reconciliation
Most reconciliation (60.3%) was carried out in a simple form using only one behavioural element for the interaction, while 30.1% were advanced (two behavioural elements) and 9.6% were complex reconciliations (three and more behavioural elements). Intensity of the preceding conflict was the only predictor of reconciliation complexity 2 (GLZ: LR 8,146 =19.6, p<0.05; Table 7.4). Gentle aggression (conflict intensity level 1) led to simple reconciliation, while hard fighting (conflict intensity level 5) was followed by advanced and complex reconciliation (Figure 7.5). Conflicts with medium intensity (levels 2-4) did not differ in their variability of reconciliation complexity from the average proportion, as expected for an ordinal variable with a linear influence. Hence, opponents increased the complexity of reconciliation with the intensity of the preceding conflict.
Table 7.4 Factors affecting the complexity of reconciliation in Taï chimpanzees. The table shows the significant variables and the estimate-coefficient () of the parameters.
Complexity of reconciliation independent simple advanced complex Wald df p parameter n Wald p variables [%] [%] [%] Conflict 14.87 4 <0.01 level 1 30 76.7 23.3 0.0 0.94 9.11 <0.01 Intensity level 2 31 67.7 19.4 12.9 0.32 1.33 ns level 3 42 50.0 40.5 9.5 -0.28 1.36 ns level 4 30 60.0 30.0 10.0 0.06 0.04 ns level 5 13 38.5 38.5 23.0 -1.04 7.82 <0.01 Model: Complexity of reconciliation (simple – advanced – complex); simple: one behavioural element; advanced: two behavioural elements; complex: three and more behavioural elements.
7.3.5 Initiator of reconciliation
Since the sex of the victim was an important determinant for latency and duration of reconciliation, I tested which kind of reconciliation (latency, duration and complexity) was initiated by aggressors (39.7%) and by victims (60.3%). Aggressors initiated more simple reconciliation, as the estimate coefficient showed a preference of victim initiators 2 for advanced and complex reconciliation (GLZ: LR 2,146 =8.13, p<0.05; Table 7.5a). When merging advanced and complex reconciliation to one category, victims preferentially 2 initiated advanced and complex reconciliation (GLZ: LR 1,146 =8.91, p<0.01; Table 7.5a).
For victim initiated reconciliation, sex of the victim (male: 47.7%; female: 52.3%) 2 played a role on the latency and duration of reconciliation (GLZ: LR 6,88 =20.71, p<0.01,
99 Chapter 7: How are relationships repaired ?
Table 7.5b). Female victims started reconciliation soon after the conflict while male victims waited longer and used long reconciliation.
Table 7.5 Factors affecting initiator of reconciliation in Taï chimpanzees. Part (a) distinguishes between aggressor and victim initiated reconciliation, while part (b) distinguishes male or female victim initiated reconciliation. The table shows the significant variables and the estimate-coefficient () of the parameters. In the far right the adapted estimate-coefficients (*) are shown for the parameters, which were merged.
(a) Initiator of reconciliation adapted parameters* independent aggressor victim Wald df p parameter n Wald p * Wald* p* variables [%] [%] Complexity of 7.60 2 <0.05 simple 88 48.9 51.1 0.73 6.64 <0.01 0.50 Reconciliation advanced 44 27.3 72.7 -0.21 0.24 ns 7.50 <0.01 -0.50 complex 14 21.4 78.6 -0.52 1.33 ns
(b) Victim initiates reconciliation independent male female Wald df p parameter n Wald p variables victim [%] victim [%] Latency of 11.94 3 <0.01 very short 19 21.1 78.9 -0.99 4.59 <0.05 Reconciliation short 20 45 55 0.13 0.12 ns long 21 42.9 57.1 -0.39 0.87 ns very long 28 71.4 28.6 1.25 10.34 <0.01 Duration of 7.98 3 <0.05 very short 24 45.8 54.2 -0.01 0.00 ns Reconciliation short 22 31.8 68.2 -0.68 2.63 ns long 16 37.5 62.5 -0.41 0.80 ns very long 26 69.2 30.8 1.10 7.19 <0.01 Model a: Initiator of reconciliation (aggressor versus victim). Model b: Victim initiator (male versus female).
7.4 Discussion
7.4.1 Summary of results
Aggression in Taï chimpanzees decreased the interaction rate of former opponents and reconciliation restored interaction rates to normal levels. A high relationship benefit increased the tendency to reconcile between same-sex partners, while reconciliation in mixed sex dyads was independent from the cooperative benefits of relationships. Characteristics of reconciliation varied widely in Taï chimpanzees. Latency of reconciliation increased with the duration of the preceding conflict and duration of reconciliation was short after food conflicts. Both latency and duration of reconciliation were short after conflicts among rank neighbours or when females were attacked. Furthermore, complexity of reconciliation increased with the intensity of the preceding conflict. Finally our analysis revealed that victims of aggression initiated more complex reconciliation compared to aggressors and female victims initiated shorter reconciliation and sooner after a conflict compared to male victims.
100 Chapter 7: How are relationships repaired ?
7.4.2 Function of reconciliation
While aggression among Taï chimpanzees seemed to disturb the tolerance levels between opponents, reconciliation seemed to restore the tolerance levels of conflict partners to normal. These results support the proposition that reconciliation functions to repair relationships following a conflict. This was also demonstrated by Cords (1992) experimental study which showed that tolerance to proximity with former opponents was higher after reconciled conflicts in longtailed macaques (Macaca fascicularis). Results indicating that reconciliation restores tolerance levels in a wild population have not previously been shown.
Reconciliation seems to be adaptive for same sex partners with highly cooperative relationships, as friendly interactions were remarkedly up to eight times less frequent than when no reconciliation occurred. Same sex dyads in Taï reconciled more with partners with whom they usually shared food and support. The reconciliation pattern of Taï chimpanzees shares similarities with that of longtailed macaques (Macaca fascicularis), where cooperation over food determined reconciliation (Cords & Thurnheer, 1993), and with that of mountain gorillas (Gorilla gorilla), where alliance partners reconciled more than others (Watts, 1995a). Mixed sex dyads showed the highest conciliatory tendencies of all dyads, but, contrary to same sex dyads, they did not reconcile more with high benefit partners. This indicates that mixed sex dyads gain different benefits from reconciliation compared with same sex dyads. Budongo chimpanzees had a higher reconciliation rate between mating partners compared to dyads that never copulated (Arnold & Whiten, 2001). A relaxed relationship with a potential mating partner could be beneficial for males, as females may be more willing to mate with them, and for females, as sexual harassment by males may be avoided (Niemeyer & Anderson, 1983; Paul, 2002).
7.4.3 The variation within reconciliation
Reconciliations in Taï chimpanzees were highly variable, with latency, duration, complexity and initiation of reconciliation varying according to the costs and benefits accompanied by reconciliation. Opponents, for example, delayed reconciliation after long conflicts. The risk of further aggression is known to increase with conflict duration (chapter 636; Wittig & Boesch, 2003d) and tension is known to reduce over time (Aureli &
36 see 6.3.2
101 Chapter 7: How are relationships repaired ? van Schaik, 1991b). This suggests that former opponents require some time after long conflicts before they can approach to reconcile without risking further aggression.
Moreover, reconciliation in Taï chimpanzees was shorter after food conflicts compared with sex and social contexts. When competing over limited food sources, delaying reconciliation is costly and conflict partners should resume feeding soon after the conflict before the food is depleted. Thus, the potential loss of time accessing limited resources appears to shorten the duration of reconciliation. In varying reconciliation, chimpanzees take into account not only the risk of further aggression – higher after longer conflicts – but also the cost of interrupting beneficial activities. So that in conflicts over food, which are often monopolisable sources (chapter 437; Wittig & Boesch, 2003a), individuals used shorter reconciliation than after conflicts over less restricted resources.
Taï chimpanzees showed a strong interaction between latency and duration of reconciliation, as they performed either a short reconciliation soon after the conflict or a long reconciliation after waiting for some time. The fact that females fight preferably over food while males usually engage in social conflicts (chapter 538; Wittig & Boesch, 2003b), could be the cause for female victims initiating shorter reconciliation and sooner after conflicts compared with male victims. Female victims appeared to accept an increased risk of further aggression when reconciliation carried the potential benefit of rapid access to a limited food source. This may explain why post-conciliatory interactions among females occurred faster than baseline. In contrast male victims that fight mainly in social contexts waited longer to reconcile with a former opponent. With no apparent benefits to off-set the risk of further aggression males seem to wait longer to initiate reconciliation. Males might only seek long-term benefits of reconciliation such as relaxed relationships (Koyama, 2001), while females might also take into account potential immediate benefit of reconciliation.
Close ranking chimpanzees initiated shorter reconciliations sooner after the conflicts than distantly ranked opponents, and they usually fight harder (with more intensity) than distantly ranked ones (chapter 539; Wittig & Boesch, 2003b). Similarly female capuchins (Cebus capucinus) were more stressed by the presence of close ranking individuals than by far ranking ones (Manson & Perry, 2000). Thus conflicts between close ranking
37 see Table 4.1 38 see 5.3.1.1
102 Chapter 7: How are relationships repaired ? opponents compared to distantly ranked opponents appeared to be more disturbing. Rank neighbours might not be able to accept subordination as easily as distantly ranked subordinates. In rats, for example, daily rhythmicity of heart rate and body temperature of subordinates were more disturbed the more individuals resisted the aggression of dominants (Meerlo et al., 1999). It seems that Taï chimpanzee opponents shortened the latency of reconciliation when relationships were strongly disturbed.
7.4.4 Repairing the damage
Taï chimpanzees reacted with more complex reconciliation after more intense aggression. This reaction suggests that the disturbance to the opponents’ relationship increased with the intensity of the preceding conflict. This would contradict the integrated hypothesis (Aureli, 1997), which proposes that damage to relationships is only dependent on the benefit of relationships and not on the intensity of conflicts. Three studies showed that behavioural stress indicators did not increase with intensity of conflicts (Macaca fascicularis: Aureli, 1997; Macaca fuscata: Kutsukake & Castles, 2001; Papio anubis: Castles & Whiten, 1998b). The two macaque species rather had increased stress indicators after conflicts with opponents with high affiliation rates or relatedness. Longtailed macaques (Macaca fascicularis), nonetheless required more reconciliation after more severe aggression, since they showed a greater proportion of reconciled conflicts after severer aggression (Koyama, 2001).
This discrepancy could have several explanations. Behavioural indictors of stress might not be the most reliable measures of stress, as for example there are studies that detected either stable or even decreasing rates of self-directing behaviours after aggression (e.g. Cebus capuchinus: Manson & Perry, 2000; Pan troglodytes: Arnold & Whiten, 2001). More direct measures of stress are needed to understand the interactions of conflict intensity, relationship quality and stress (e.g. heart rate: Aureli et al., 1999; hormones: Sapolsky, 1986; Creel, 2001). An alternative explanation is that categories of conflict intensities used in other studies may have been too broad to detect a correlation. In the present study the two extreme values out of five categories of conflict intensity were responsible for the significant influence on the complexity of reconciliation. Another explanation is that stress may not increase with the disturbance of the relationship, but rather with lost benefit from disturbed cooperations. If so there would be no discrepancy
39 see Table 5.4a
103 Chapter 7: How are relationships repaired ? between the integrated hypothesis and our results. A last possibility is that dominance style may affect stress levels. The two macaque species have despotic dominance styles (Thierry et al., 2000b), which imply a high degree of asymmetry in conflicts (Thierry, 2000). The mere presence of dominants may be enough for stress to increase (Castles et al., 1999) and may be as stressful as open aggression (Sapolsky, 1993; Abbott et al., 2003). In contrast chimpanzees have a more egalitarian dominance style allowing subordinate individuals to win some conflicts (chapter 540; Wittig & Boesch, 2003b). It is conceivable that victims in despotic groups perceive mild aggression (e.g. threats) as being as stressful as intense aggression, while victims in egalitarian groups are less stressed by threats than by contact aggression. Nonetheless in Taï chimpanzees an increased level of disturbance of relationships seems to require an increased complexity of reconciliation to repair the relationships of former opponents.
I conclude that conflicts in Taï chimpanzees disturbed the relationship between opponents and that reconciliation repaired the relationship. Reconciliation cleared up the disturbance with a more complex pattern for more intense conflicts. However, high value relationships were reconciled for all levels of disturbance. Therefore, I suggest that the function of reconciliation is to reduce disturbance in all relationships. Previous authors have suggested reconciliation occurs as a repair mechanism in high value relationships only, either through stress reduction or cooperation repair (Aureli, 1997; de Waal & Aureli, 1997). Our data has shown that disturbance resulting from strong fighting is also repaired by reconciliation independent from relationship quality of opponents. Repair of low value relationships has not previously been predicted, but its occurrence in primates could be beneficial in societies where out-competed partners may react with revenge and contra-intervention (de Waal & Luttrell, 1988; Drukker et al., 1991; Aureli et al., 1992; Silk, 1992). Since it might be of equal benefit to have a friend as it is costly to have an enemy, social living animals should prevent losing friends and making enemies. Thus reconciliation maintains the beneficial relationship of high value partners, but it also seems to prevent low benefit partners turning into enemies.
Based on the results of the variation within reconciliation of Taï chimpanzees I suggest a functional framework of reconciliation across species as follows: Reconciliation is likely to repair all relationships independent of the relationship value for the conflict partners. Since any disturbance is costly for high value partners, these partners should
40 see Figure 5.1
104 Chapter 7: How are relationships repaired ? reconcile many conflicts. In contrast low value partners only suffer costs from strong disturbances and therefore they should reconcile some but less conflicts than high value partners. Such a pattern is evident for almost all studies on reconciliation in primates (Cords & Aureli, 2000; Aureli et al., 2002), although reasons for it have rarely been discussed. Obviously we need to know more about long-term consequences of unreconciled conflicts on the relationship quality of conflict partners.
105 Chapter 8:
Open questions and possible answers - a general discussion
Figure 8.0 Male Marius is resting on a liana hang-matt, while he is waiting for females to finish off feeding on a fruit tree close by (Photo by R. Wittig). Chapter 8: Open questions and possible answers – a general discussion
8.1 Contribution of this study to our knowledge in conflict management
Taï chimpanzees used conflict management in situations of conflicts of interest before, during and after aggressive interactions. Conflict management in Taï chimpanzees was dependent upon the dominance relationships among conflict partners, which were based on contest and ranked in a linear hierarchy (chapter 441; Wittig & Boesch, 2003a,c). However the effect of dominance relationships was strongest during pre- and peri-conflict management (chapter 542; Wittig & Boesch, 2003b), with rank playing only a minor role in the choice of post-conflict interactions (chapter 643; Wittig & Boesch, 2003d). During pre- conflict management Taï chimpanzees considered their likelihood of winning and the benefit of resources when deciding whether to avoid or to initiate aggression (chapter 544; Wittig & Boesch, 2003b). Furthermore the quality of the relationship among the competitors as well as the potential costs of aggression influenced the decision-making during the pre-conflict management. Peri-conflict management mainly varied according to differences in the intensity of aggression, as Taï chimpanzees adjusted their conflict intensity to their likelihood of winning (chapter 545; Wittig & Boesch, 2003b). Conflict partners avoided hard fighting when it was not needed and they accepted more risk when they were likely to access the resource. Interactions showing the most variability were found during post-conflict management of Taï chimpanzees. Taï chimpanzees initiated one out of five possible different post-conflict interactions (PCIs), according to the potential advantages and disadvantages of each PCI (chapter 646; Wittig & Boesch, 2003d). Reconciliation, one of the possible interactions of post-conflict management, resolved conflicts of interest among Taï chimpanzees (chapter 747; Wittig & Boesch, in review). Conflict partners were more likely to initiate reconciliation when the relationship was of a high benefit to them, and they carefully invested more in reconciliation when the relationship was more disturbed.
Although the subject of conflict resolution in non-human primates is rather new, the topic is extensively discussed (Aureli & de Waal, 2000b, chapter 1, Appendix A). However, we are far away from understanding all functions and mechanisms of conflict
41 see 4.3.1 and 4.3.3 42 see Figure 5.5 43 see Figure 6.4 44 see 5.3.1.2 45 see 5.3.2 46 see Figure 6.4
107 Chapter 8: Open questions and possible answers – a general discussion management among animals (e.g. Aureli et al., 2002). The contribution of the present study to the field of conflict management is that it investigates decision-making before and during aggressive interactions, tests for the first time the Relational Model with data of one species and makes significant extensions to the model. It provides information on the economics of post-conflict management by describing the choice of post-conflict interactions in terms of an evaluation process. Besides answering the question of when relationship repair is applied, this study shows what is needed for the reparation, and it proves the shaping power of contest competition on dominance relationships. Furthermore several open questions were formulated by de Waal & Aureli (2000). I discuss below what my results can contribute to a better understanding of the remaining open questions. I compare the results from wild chimpanzees with our knowledge of conflict management in humans, in non-human primates and in non-primate animals. This approach may help to provide a better understanding of the evolution of conflict management and the forces behind its development.
8.2 Avoidance and prevention of conflicts
There is only fragmentary knowledge about the mechanisms that help individuals to avoid or prevent conflicts (de Waal & Aureli, 2000). In Taï chimpanzees a low likelihood of winning was usually found to prevent chimpanzees from escalating conflicts of interest into aggression (chapter 548; Wittig & Boesch, 2003b). This result is not surprising since dominance relationships are supposed to enhance the net-benefit of competitive situations for group-living animals by regulating the access to resources without or by means of low-cost fighting (Walters & Seyfarth, 1987; Drews, 1993). Many other primate species have conflict avoidance mechanisms, which are also based on dominance relationships (Preuschoft & van Schaik, 2000).
Investigation of human negotiation behaviour in conflicts of interest revealed the use of five different conflict handling behaviours (Kelly et al., 1970; Ruble & Thomas, 1976): avoiding, accommodating, compromising, competing and collaborating. These different conflict managing styles represent a two-dimensional model of conflict behaviour, that describes each style as more or less assertive and more or less
47 see 7.3.2 48 see Figure 5.1
108 Chapter 8: Open questions and possible answers – a general discussion cooperative. Thus, for example, the competing (also called forcing) managing style is highly assertive but not cooperative, while the avoiding style is neither assertive nor cooperative. Similar to other primates, dominance relationships play an important role when human conflict partners decide about the style they use to resolve their conflict of interest (Rahim, 1986; Kozan, 1991; Thomas, 1992). Jordanian managers mostly avoid conflicts with partners of the same dominance level, while they are more likely to compete forcefully with subordinate partners and collaborate with partners superior to them (Kozan, 1991). Cross-cultural comparisons revealed that although the frequencies of the managing styles are marginally different between different cultures (e.g. USA: Rahim, 1986; Jordan: Kozan, 1991), dominance relationships between conflict partners seem to be the main factor determining whether to avoid confrontation or to use the forcing management style to achieve a goal.
Dominance provides a wide range of benefits in primates and non-primate animals. Dominant individuals can exclude subordinates from certain feeding sites or resources (e.g. spotted hyena (Crocuta crocuta): Tilson & Hamilton III, 1984; chimpanzee (Pan troglodytes): chapter 449; Wittig & Boesch, 2003a), dominant males are the preferred mating partners of females (e.g. American cockroach (Periplaneta americana): Breed & Rasmussen, 1980; anemonefish (Amphiprion akallopisos): Fricke, 1979; fallow deer (Dama dama): McElligott & Hayden, 2000; savanna baboon (Papio cynocephalus): Bercovitch, 1995; Weingrill et al., 2000), or only dominant individuals reproduce in a group (e.g. clownfish (Amphiprion percula): Buston, 2003; naked mole-rat (Heterocephalus glaber): Reeve, 1992; wolf (Canis lupus): Bibikow, 1990; several species of marmosets and tamarins (Callithrichinae): Garber, 1997). Most of the time those benefits are achieved without aggression, which qualifies dominance relationships as a conflict avoidance mechanism.
The dominance relationships in fish are usually triggered by body size. In group- living clownfish (Amphiprion percula), individuals are ranked in a linear hierarchy according to body size and breeding is exclusive to the dominant pair (Fricke, 1979; Buston, 2003). When one group member disappears from the group, the individuals that are subordinate relative to the disappeared member enter a growth period while the size of the relatively dominant members remains constant (Buston, 2003). Thus, clownfish queue in the dominance hierarchy and wait for their time to reach the dominant position
109 Chapter 8: Open questions and possible answers – a general discussion before reproducing. This means that although a subordinate clownfish never challenges the position of a dominant, he is guaranteed of reaching the dominant breeding position, as long as he does not die before those dominant to him. This queuing mechanism resolves an evolutionary conflict of interest and qualifies therefore as a mechanism which prevents conflicts before they are created.
There are also many other strategies to avoid aggressive interactions. For example, individuals can increase tolerance levels towards each other (e.g. primates: Judge, 2000; humans: Fry, 2000; other mammals: Rasa, 1979). Furthermore territory defence by vocal (e.g. birds: Krebs, 1977; frogs: Wells, 1977), olfactory (e.g. isopoda: Linsenmair, 1972; mammals: Rasa, 1973) or behavioural (e.g. sharks: Johnson & Nelson, 1973) signals restrict the use of a certain area to the owner (one individual or a group), without needing to constantly fight against competitors from outside. Pre-conflict management thus seems to be a wide spread strategy throughout all animal classes (e.g. isopoda, insecta, amphibia, mammalia), in both intra- and inter-group situations and in both social and solitary living animals.
8.3 Conflict regulation in a wider social context
The involvement of third parties in conflicts raises the question of how conflicts are regulated in a wider social context (de Waal & Aureli, 2000). In Taï chimpanzees consolation seemed to substitute for reconciliation when approaching the former opponent was too risky (chapter 650; Wittig & Boesch, 2003d). As mentioned in chapter 6, one reason for individuals to offer consolation could be that they are either friends of the conflict partner or potential targets of redirected aggression. Another possibility may be that all community members, in chimpanzees, share the responsibility to maintain the benefits of community living (e.g. defence against neighbours and predators). Consequence of unrepaired disturbance among community members could be cheating or defecting behaviour, such as not indicating danger to others (e.g. Seyfarth et al., 1980) or not aiding in defence against neighbours (e.g. Nishida et al., 1985). Accordingly each chimpanzee would have to understand the type of conflict-induced disturbance to the relationships between other community members and the consequences of this
49 see Table 4.1 50 see 6.4.5
110 Chapter 8: Open questions and possible answers – a general discussion disturbance to the benefits of community living for everybody. Maintaining the benefits of group living for each member might have been one selective pressure for the evolution of consolation in mammal societies.
In contrast to offered consolation, solicited consolation in Taï chimpanzees did not qualify as post-conflict management (chapter 651; Wittig & Boesch, 2003d). Chimpanzees seem to be the only species, except for humans, that offer consolation (Watts et al., 2000). De Waal & Aureli (1996) concluded that the reason for the difference in consolation between macaques and chimpanzees may be the cognitive constraints of macaques. However, since macaques use their knowledge of triadic relationships when they redirect aggression towards kin of former opponents (Macaca fascicularis: Aureli & van Schaik, 1991a; Macaca fuscata: Aureli et al., 1992), their cognitive ability seems to cope with the recognition of triadic relationships. Therefore cognitive constraints in macaques seem to be restrictive only in terms of affiliative behaviours. Emotions, such as compassion and empathy, could be absent in the emotional repertoire of macaques while they may be present in chimpanzees. Empathy and compassion might be conditional for someone offering consolation (de Waal & Aureli, 1996). Whether or not the absence of such emotions is the reason for the absence of offered consolation in monkeys and whether or not this absence is constrained by cognitive abilities needs further research and discussion.
8.4 Effectiveness of the mechanisms of conflict management
When terms like reconciliation or consolation were introduced into the research of animal behaviour (de Waal & van Roosmalen, 1979), the functions of these behaviours were unknown, even though the terms imply certain functions. The effects of some post- conflict interactions are still not completely understood and one has to be careful applying such terms to a behaviour (de Waal & Aureli, 2000). However Taï chimpanzees have provided good evidence for the conciliatory function of reconciliation (chapter 752; Wittig & Boesch, in review). Reconciliation in Taï chimpanzees was so effective that former opponents were able to interact affiliatively up to eight times quicker after reconciliation than without. Under the condition, for example, where a female needs food to be shared
51 see 6.3.1 52 see Figures 7.2 & 7.3
111 Chapter 8: Open questions and possible answers – a general discussion to access it, reconciliation seems to have a major effect on her food budget. This result, together with the tolerance experiments on long-tailed macaques (Macaca fascicularis: Cords, 1992; Cords & Thurnheer, 1993), proves the effectiveness and the function of reconciliation in primates.
Evidence for the effectiveness of management during a conflict (peri-conflict management) was shown in the distribution of conflict intensity, since fights were significantly less intense when the opponents had a large fighting asymmetry (chapter 553; Wittig & Boesch, 2003b). Taï chimpanzees risked more contact aggression when conflict partners did not have a large rank difference. In other species, potential conflict partners are also assessing asymmetries in fighting abilities to avoid unnecessary and potentially harmful escalations of aggression (e.g. Red deer (Cervus elaphus): Clutton- Brock & Albon, 1979; toads (Bufo bufo): Davies & Halliday, 1979; spiders (Agelenopsis spec.): Maynard Smith & Riechert, 1984). Peri-conflict management, including the use of ritualised aggression, seems to be a wide spread and effective mechanism to manage conflict situations.
Ritualised submission may be an effective tool to terminate conflicts. Since greetings clearly indicate the submission of the sender towards the receiver in Taï chimpanzees (chapter 454; Wittig & Boesch, 2003a), they could point out the acceptance of the conflict outcome by the subordinate conflict partner. Thus submissive greetings after aggression may prepare the way for opponents to reconcile, as the subordinate might signal he has accepted his defeat. An effective preparation of reconciliation could be the reason for the correlation between pant-grunt vocalisations (see Table 3.1) and reconciliation events in chimpanzee (Budongo: Arnold & Whiten, 2001; Taï: R.M. Wittig & C. Boesch, unpublished data). Such a vocalisation, when used all the time in combination with conciliatory behaviours, may over time become effective enough to function as reconciliation itself. A functional change from a preparation behaviour for reconciliation to reconciliation itself could be a possible scenario for the evolution of vocal reconciliation (Cheney et al., 1995; Cheney & Seyfarth, 1997).
53 see 5.3.2.1 54 see 4.3.1 & Figure 4.1
112 Chapter 8: Open questions and possible answers – a general discussion
8.5 Similarity of conflict management in humans and chimpanzees
The comparison between human and chimpanzees forces itself into the discussion, since chimpanzees are humans’ closest living relatives (Ebersberger et al., 2002; Kaessmann & Pääbo, 2002). One of the most striking similarities between humans and other primates has already been mentioned by (de Waal & Aureli, 2000). They argued that one of the shared principles of conciliatory mechanisms is that they operate mainly between cooperative partners (more details for humans: Fry, 2000; Yarn, 2000; chimpanzees: chapters 655 & 756 Wittig & Boesch, 2003d; Wittig & Boesch, in review; other primates: Cords & Aureli, 2000). However, there may be many more similarities.
Ritualised non-contact aggression plays an important role when chimpanzees want to prevent conflicts from escalating into harmful fighting (chapter 557; Wittig & Boesch, 2003b). In humans, ritualised non-contact aggression is applied similarly in situations when individuals want to avoid getting into fights or keeping a fight from becoming serious (Fry, 2000). Humans have supposedly developed such fighting performances with non-contact aggression in order to control their harmful aggression. Thus self-control in humans may be seen as a mechanism to decrease costs of aggression. The same mechanism seems to be the reasons for chimpanzees to use non-contact aggression (chapter 558; Wittig & Boesch, 2003b).
The intervention of friendly peacemakers is one of the most important strategies in humans. Those third party interventions can function as judges (Nader, 1991; Yarn, 2000) or as mediators (Fry, 2000; Neu, 2000). While judges can enforce the end of the conflict, mediators have no enforcement possibilities and their intervention can function only when both sides accept the mediators suggestion. Although chimpanzees sometimes enforce the end of a conflict (usually when dominant individuals intervene aggressively in an ongoing conflict but without supporting any side), it is unknown whether or not the intervening individual judges the conflict. In contrast, mediation in conflicts seem to happen in chimpanzees. Chimpanzees offer consolation when direct contact among former opponents is risky (chapter 659; Wittig & Boesch, 2003d). Later consolation may
55 see 6.3.5.1 56 see Figure 7.3 57 see 5.3.2.1 58 see 5.4.3 & 5.4.5 59 see 6.3.5.2
113 Chapter 8: Open questions and possible answers – a general discussion lead to peace-making, when the mediator brings former opponents in proximity so that they can reconcile (humans: Pruitt et al., 1993; primates: de Waal, 1996b, Appendix D). Human mediators, like chimpanzee mediators, seem to intervene in situations when former opponents cannot meet again without further aggression. In humans, conflicts are mediated when they would lead to serious violence (e.g. Zapotec Indians: Fry, 2000) or they include serious violence already (e.g. in wars and political conflicts: Neu, 2000). Thus both humans and chimpanzees may employ the mechanism of mediation and seem to manage their conflicts in a similar way on a dyadic level (de Waal & Aureli, 2000). Testing how comparable the motivation is of humans and chimpanzees behind these apparently similar mechanisms, can be seen as the challenge for the future.
8.6 Prospects
Despite providing many answers, this study also offers several new questions. For example, would high benefit partners really withhold future beneficial cooperation because of disturbed relationships? One of the future research efforts should try to answer the question of what happens to relationships when they are not reconciled after a disturbing conflict. This includes not only the long-term consequences of non- reconciliation, but also of reconciliation, which Koyama (2001) has started to tackled in a study on Japanese macaques (Macaca fuscata).
Another important question is whether or not offered consolation is really an adequate substitute for reconciliation. Since there is the potential exchange of consolation for reconciliation, research may need to focus on the functional similarities of both and the relationship of the individual that offers consolation to both conflict partners. Consolation offered by friends (or kin) of the consoled conflict partner may have another function than consolation offered by friends (or kin) of the former opponent.
Furthermore it would be of interest to quantify the costs of aggression due to conflict duration. When concentrating on energetic costs physiological measurements may be needed to see whether or not there is a cost difference between short and long conflicts. A similar approach is required for measuring stress. It seems that behavioural stress indicators are not precise enough to reflect the complete pattern. However
114 Chapter 8: Open questions and possible answers – a general discussion physiological measurements started to be introduced in the research of stress production and stress reduction (e.g. Aureli et al., 1999).
One of the remaining open questions is how good comparisons are between humans and other primates. A comparative study of conflict behaviour between humans and apes, applying the same methods, seems to be overdue. First steps in this direction have been done by several scientists focusing on conflict management in pre-school children with controlled methods (Butovskaya et al., 2000). However the inclusion of juveniles in studies on conflict management seems to be critical, since correct application of decision-making requires experience. Therefore future research, and not only in humans, should concentrate on adults. The present study is one of the rare case of conflict management among adults and provides thus a strong data-set for comparisons between adult humans and chimpanzees.
What would social living animals do without conflict management? This rhetorical question may have now an easy answer. Social living animals would probably not live in social units, since the costs of competition due to social living would presumably prevail, for most of the group members, over the benefits of social living. Although conflict management is not an invention of social living animals (see: 8.2 last paragraph), the large diversity in social living animals would suggest a boost in evolving behaviours that manage conflicts. Thus conflict management possibly went through a behavioural radiation each time passing from solitary to social living. These patterns of conflict management seemingly make social living more advantageous than disadvantageous and therefore may allow animals to live social lives.
115 Summary
9.1 Summary
Besides many advantages, social living also holds several disadvantages. Social partners compete for the same resources or seek contrary goals. When facing such conflicts of interest, competitors go through a decision-making process of whether or not to fight over a resource. However aggressive interactions, which I will refer to here as conflicts, incur costs, which can be separated into costs of aggression (increased risk of injury, higher energy usage) and social costs. Social costs are created by the consequences for the social life, such as the disturbance of cooperative relationships or stress. Conflict management should diminish the costs of conflicts by avoiding escalation to aggression, regulating the intensity of the escalation or dealing with the social consequences (e.g. relationship disturbance or social stress) of the conflict. Thus conflict management can be used before (pre-conflict management), during (peri-conflict management) and after the conflict (post-conflict management). The underlying hypothesis for optimal conflict management is that the benefits prevail over the costs, meanwhile the net-benefit is maximised.
I investigated the conflict management of wild chimpanzees (Pan troglodytes verus) in the Taï National Park, Côte d’Ivoire, West Africa. Of the 1071 conflicts observed during full-day focal animal follows of adults (4 males, 11 females), I analysed 876 dyadic conflicts among adult chimpanzees of both sex. Multivariate analysis was carried out to detect the variables that influence the decision-making process, while dyadic statistics were usually conducted for mono-factorial testing.
Dominance relationships can regulate access to resources and thus help to avoid aggression (pre-conflict management). Although linear hierarchies are commonly found among male chimpanzees, they are believed to be absent among females. However, I detected a formal linear dominance hierarchy among the Taï females based on greeting behaviour directed from the subordinate to the dominant female. Females faced contest competition over food, which increased when either the food was monopolisable or the number of competitors increased. Dominant females usually possessed the food after the conflict. Winning contests over food, but not age, was related to the dominance rank. 9: Summary
Affiliative relationships among the females did not help to explain the absence of greetings in some dyads. However post hoc comparison among chimpanzee populations made differences in food competition, predation risk and observation time apparent, which may explain the difference in dominance relationships.
I also examined the decision-making process of whether or not to initiate aggression and how strong to fight. An extended version of the Relational Model (de Waal, 1996a) was developed to describe the dynamics of the decision-making process in Taï chimpanzees, such that the net-benefit determines the occurrence of conflicts. Both sexes fought more frequently for the resources that were most important to them: food for females and social contexts for males. Individuals used two different strategies according to their likelihood of winning the aggressive interaction, which was determined by the dominance relationship of the conflict partners. Dominant initiators had longer and more intense aggressive interactions, but they limited their social disadvantages by fighting non-cooperative partners. Subordinate initiators had shorter and less intense aggressive interactions, but risked more social costs, which they could reduce afterwards by reconciliation. Both strategies included a positive overall net-benefit. The extended Relational Model fits the complexity of wild chimpanzee conflicts and allows for more flexibility in the decision-making process compared to the original model.
Post-conflict management in social living animals can reduce costs that remain after aggressive interactions by means of a variety of interactions implemented after aggression (e.g. reconciliation, consolation, redirected aggression). Each post-conflict interaction (PCI) provides different advantages and disadvantages, although the functions may sometimes overlap. Individuals can therefore choose a PCI to achieve the most favourable outcome within a given conflict situation. I investigated which conflict- condition led to which type of PCI and related the choice of PCI to its advantages and disadvantages. Taï chimpanzees used reconciliation to resolve conflicts among high value partners and when approaching the former opponent was unlikely to entail further aggression. Consolation seemed to substitute for reconciliation, when opponents were low value partners or approaching the former opponent was too risky, such as when further aggression was likely. Taï chimpanzees renewed aggression after undecided conflicts and when losers were unexpected. They used redirected aggression after long conflicts, possibly because friendly PCIs were likely to fail. However, Taï chimpanzees continued with business as usual when conflicts were very short, and they avoided further
117 9: Summary interactions when the accessibility of the resource was unlimited. Taï chimpanzees appeared to follow a clear-cut evaluation process as they seemed to weigh advantages against disadvantages for the appropriate choice of PCI.
However reconciliation appears to be the only PCI that is able to repair the relationships of former opponents after being disturbed by aggressive interactions. Despite a consensus about the benefits of reconciliation, it remains unclear how former opponents achieve these benefits. Variation within reconciliation is evident in many species, but understanding what causes the variation has been mostly neglected until now. Therefore I investigated how Taï chimpanzees reconciled. This study provides evidence for the repair function of reconciliation, since aggression disturbed tolerance levels among former opponents and reconciliation restored tolerance to normal levels again. Partners with highly beneficial relationships reconciled more often compared with partners of low mutual benefit. Latency and duration of reconciliation varied in combination, such that short reconciliations were initiated soon after the conflict, while long reconciliations were initiated later. Latency increased with the risk of further aggression, while duration decreased when costs were incurred from interruption of beneficial activities. In contrast, the complexity of reconciliation varied according to the intensity of the preceding conflict, such that reconciliation was more complex after more intense conflicts. My results suggest that relationships between opponents are increasingly disturbed with increasing conflict intensity and reconciliation repairs all relationships independent of their relationship value. I propose that the function of reconciliation is to reduce the disturbance created by aggression, but that reconciliation occurs more frequently the more beneficial it is for former opponents.
Taï chimpanzees engaged in conflict management before, during and after the conflict. The decision-making process of Taï chimpanzees is based on economic rules in terms of costs and benefits. Conflict management provides Taï chimpanzees with a tool to minimise the disadvantages of group-living.
9.2 Zusammenfassung
Das Leben in Gruppen beinhaltet neben vielen Vorteilen auch zahlreiche Nachteile. Gruppenmitglieder konkurrieren über dieselben begrenzten Ressourcen oder verfolgen
118 9: Summary unterschiedliche Ziele. Während eines Interessenkonfliktes durchläuft jeder Konkurrent einen Entscheidungsprozeß, in dessen Zentrum die Frage steht, ob es sich lohnt für eine bestimmte Ressource zu kämpfen. Dabei muß einbezogen werden, daß aggressive Auseinandersetzungen Kosten verursachen. Diese Kosten können zum einen in Aggressionskosten, z.B. ein erhöhtes Verletzungsrisiko oder hohen Energieverbrauch, und zum anderen in Sozialkosten, z.B. die Störung kooperativer Beziehungen oder Streßreaktionen, aufgeteilt werden. Die unter dem Begriff Konfliktmanagement zusammengefaßten Verhaltensweisen helfen Konfliktkosten zu verringern. So können in Interessenkonflikten Aggression vermieden, deren Stärke gedämpft und soziale Konsequenzen verringert werden. Konfliktmanagement kann vor (pre-conflict management), während (peri-conflict management) und nach (post-conflict management) dem Auftreten von Aggression eingesetzt werden. Die Hypothese, die einem optimalen Konfliktmanagement zu Grunde liegt, ist daß der Nutzen eines Konfliktes seine Kosten übersteigen muß, wobei der Profit (Nutzen – Kosten) aus dem Konflikt maximiert wird. Vereinfachend nenne ich aggressive Auseinandersetzungen von nun an Konflikte.
Das Konfliktmanagement von freilebenden Schimpansen (Pan troglodytes verus) wurde im Taï National Park, Côte d’Ivoire (Westafrika), untersucht. Von 1071 beobachteten Konflikten, die ich während ganztägiger Beobachtungen an 4 männlichen und 11 weiblichen Fokustieren gesammelt habe, wurden 876 zwischen erwachsenen Schimpansen beider Geschlechter analysiert. Multivariate Analysemethoden wurden angewandt, um die entscheidenden Faktoren des Entscheidungsprozesses bei Konflikten aufzuspüren, während überwiegend Paarstatistik für einfaktorielle Analysen verwandt wurde.
Dominanzbeziehungen können den Zugang zu Ressourcen regulieren und damit den Ausbruch von Aggression verhindern (pre-conflict management). Frühere Studien haben gezeigt, daß häufig lineare Rangordnungen unter männlichen, nicht aber unter weiblichen Schimpansen bestehen. Die vorliegende Arbeit konnte hingegen eine lineare Rangordnung auch unter den Weibchen der Taï Schimpansen nachweisen, welche auf Grußlauten der untergeordneten Weibchen gerichtet an die Dominanten beruhte. Im Nahrungskontext waren die Taï Weibchen untereinander direkter Konkurrenz (contest competition) ausgesetzt. Dieser Wettstreit wurde intensiver, sobald eine Nahrungsquelle monopolisierbar war oder die Anzahl von Konkurrentinnen anstieg. Der Rang in der Hierarchie unter den Weibchen war abhängig vom Gewinn der Wettstreite aber
119 9: Summary unabhängig vom Alter. Warum zwischen einigen Weibchen keine Begrüßungen beobachtet wurden, konnte nicht mit dem Fehlen sozio-positiver Beziehungen erklärt werden. Ein Vergleich zwischen Populationen von Schimpansen zeigte Unterschiede in der Nahrungskonkurrenz, dem Raubdruck und der Beobachtungszeit. Diese Faktoren könnten der Grund für die unterschiedlichen Dominanzbeziehungen unter den Weibchen sein.
Anschließend untersuchte ich Variablen, die darüber entscheiden, ob und wie intensiv Individuen kämpfen. Dazu erweiterte ich das Relational Model (de Waal, 1996a), um die gesamte Dynamik des Entscheidungsprozesses bei Taï Schimpansen beschreiben zu können. Das erweiterte Relational Model basiert auf der Annahme, daß der zu erwartende Profit den Ausbruch von Aggression bestimmt. Schimpansen beider Geschlechter kämpften häufiger um Ressourcen, die von besonderer Bedeutung für sie waren: Nahrung für Weibchen und sozialer Rang für Männchen. Schimpansen benutzten zwei Strategien, die auf ihre Wahrscheinlichkeit diesen Kampf zu gewinnen zurückgeführt wurden. Die Wahrscheinlichkeit einen Kampf zu gewinnen wurde durch die Dominanzbeziehung der Gegner bestimmt. Dominante Angreifer initiierten längere und intensivere Kämpfe, aber bemühten sich Sozialkosten zu begrenzen, indem sie selten Kooperationspartner angriffen. Untergeordnete Angreifer kämpften kürzer und weniger intensiv, riskierten jedoch höhere Sozialkosten, die sie anschließend durch Versöhnungsmechanismen wieder zu verringern versuchten. Beide Strategien resultierten in einem positiven Profit für den Angreifer. Mit dem erweiterten Relational Model kann die gesamte Komplexität von Konflikten zwischen Taï Schimpansen beschrieben werden. Es erlaubt eine größere Flexibilität im Vergleich zur ursprünglichen Version des Models.
Das Post-conflict Management sozial lebender Tiere kann dazu eingesetzt werden, Kosten zu reduzieren, die am Ende des Konfliktes bestehen. Dazu werden eine Vielzahl von Verhaltensweisen angewandt, so z.B. Versöhnung (reconciliation), Trost (consolation) oder Weiterleitung von Aggression (redirected aggression). Jede dieser Interaktionen, die erst nach dem Konflikt initiiert werden (PCI = post-conflict interaction), bietet unterschiedliche Vor- und Nachteile, die gelegentlich überlappen. Um den bestmöglichen Vorteil aus einer Konfliktsituation zu ziehen, können Individuen unter verschiedenen PCIs wählen. Die vorliegende Arbeit untersuchte, welche Konfliktsituation bei Taï Schimpansen zu welchen PCIs führten, und überprüfte, ob die Vor- und Nachteilen der ausgewählten PCI mit den Bedürfnissen zur Kostenbegrenzung der Konfliktpartner übereinstimmte. Ehemalige Gegner versöhnten sich nach Konflikten, wenn ihre Beziehung wertvoll für sie
120 9: Summary war, und wenn eine Annäherung aneinander nur unwahrscheinlich zu erneuter Aggression geführt hätte. Das Trösten durch Dritte schien manchmal die Versöhnung zu ersetzen. Trost wurde von Dritten angeboten, wenn zwischen ehemaligen Gegnern keine wertvolle Beziehung bestand oder eine Annäherung der Gegner vermutlich wieder zu Aggression geführt hätte. Taï Schimpansen nahmen einen Konflikt wieder auf, wenn die vorherige Auseinandersetzung unentschieden war oder einen unerwarteten Verlierer aufwies. Nach lang anhaltenden Konflikten, oder wenn es wahrscheinlich ausging, daß friedliche PCIs fehlschlagen würden, leiteten Taï Schimpansen die Aggression häufig an Unbeteiligte weiter. Im Gegensatz dazu verhielten sich Taï Schimpansen nach kurzen Konflikten so, weiter als wäre nichts geschehen, und verweigerten jede Art von Interaktion (keine PCI), wenn die betreffende Ressource nicht an Ort oder Zeit gebunden war. Taï Schimpansen schienen Vor- und Nachteile klar gegeneinander abzuwägen, um die geeignetste PCI (Strategie) auszuwählen.
Insgesamt scheint Versöhnung die einzige PCI zu sein, mit der es möglich ist, die aggressionsbedingte Störung einer Beziehung zu beseitigen, d.h. eine Beziehung zu reparieren. Obwohl der Nutzen von Versöhnungen allgemein anerkannt ist, war annähernd keine Kenntnis darüber vorhanden, wie ehemalige Gegner eine solche Reparatur durchführen. Frühere Studien gaben Anhaltspunkte über unterschiedliche Längen, Latenzen und Verhaltensweisen von Versöhnungen innerhalb einer Art. Die Gründe für die Variabilität im Versöhnungsverhalten waren jedoch weitgehend unbekannt. Aus diesem Grund untersuchte ich besonders das Versöhnungsverhalten der Taï Schimpansen. Die Daten bestätigten, daß die Versöhnung eine Beziehung reparieren kann. Aggression störte die Toleranz zwischen den Gegnern, Versöhnung normalisierte diese wieder. Ehemalige Gegner mit wertvollen Beziehungen versöhnten sich häufiger als Partner mit weniger wertvollen Beziehungen. Die Latenz und Dauer der Versöhnung verändern sich in Abhängigkeit voneinander, da kurze Versöhnungen schnell nach einem Konflikt erfolgten. Lange Versöhnungen hingegen dauerten auch lange, bis sie zustande kamen. Hinzu kam, daß Taï Schimpansen eine lange Latenz wählten, wenn ein erneutes Aufflammen der Aggression wahrscheinlich erschien, aber nur kurze Zeit in die Versöhnung investierten, wenn die Zeit anderweitig vorteilhafter genutzt werden konnte. Dahingegen war die Komplexität der Versöhnung abhängig von der Stärke des Konfliktes. Je härter zuvor der Kampf geführt wurde, desto komplexer war die Versöhnung. Diese Ergebnisse deuten darauf hin, daß die Beziehung zwischen Konfliktpartnern um so stärker
121 9: Summary gestört wird, je heftiger der Konflikt ist, und daß alle Beziehungen (auch die weniger wertvollen) repariert werden können.
Taï Schimpansen benutzten Konflikt Management vor, während und nach aggressiven Auseinandersetzungen. Der Entscheidungsprozeß der Taï Schimpansen basiert auf ökonomische Regeln, die auf einer Kosten – Nutzen Abschätzung beruhen. Konfliktmanagement stellt den Taï Schimpansen ein Werkzeug zur Verfügung, das die Nachteile, die ein Leben in der Gruppe mit sich führt, auf ein erträgliches Maß reduziert, und die Vorteile des Gruppenlebens überwiegen läßt.
122 Acknowledgements
I am grateful to the ‘Ministère de la Recherche Scientifique’, the ‘Ministère de l’Agriculture et des Ressource Animales’ of Côte d’Ivoire, the director of the Taï National Park and the ‘Projet Autonome pour la Conservation du Parc National de Taï’ for permission to conduct this study.
I am very grateful to my supervisor Christophe Boesch, who accepted me as a PhD student. He helped me to develop this PhD project and to separate the wheat from the chaff in my ideas. Furthermore I am grateful to the Swiss National Foundation and the Max-Planck Society for their financial support.
Many thanks are due to the ‘Centre Suisse de la Recherche Scientifique’ (CSRS) in Abidjan, especially to their directors and families, Jakob and Maria Zinsstag and Olivier and Simone Girardin, for personal and logistical support in Côte d’Ivoire. To Odile Soublé, who took care of me during an period of illness. To Henry, Homer, Omarou and Jürg and all other staff-members and scientists of the CSRS for their welcoming friendliness.
I am very grateful to all staff-members of the ‘Projet Chimpanzé Taï’, especially Kpazahi Honora Néné and Nohon Gregoire Kohon, for their company and assistance in the forest. Furthermore to Patricia, Parfaite, Vivianne and Félicité for taking care of the camp, and to many students for their delightful company, especially Becky, Dean, Ilka, Lionel, Nick, Tobias and Yasmin.
Additionally many thanks to the staff-members and scientists of the ‘Projet Singes Taï’ and at the ‘Station du Centre de Recherche en Ecologie’. All of them contributed in one way or another to make my stay successful in Côte d’Ivoire. Here, I want to thank especially Germain, Mandy, Klaus, Pablo und Pierre for the good logistical cooperation and technical support.
Many thanks are due to Ousman and Henry, for the restful holidays at Farafina plage, and to the tailor of the “KM 17” for his nice work. 10: Acknowledgements
As data are useless without proper analyses, I am deeply indebted to Gunter Weiss, who made me familiar with multivariate analyses, and to Daniel Stahl, who advised me on any other statistical topic.
Many thanks are due to Filippo Aureli, Josep Call, Nadja Corp, Tobias Deschner, Frans de Waal, Diane Doran, Julia Fischer, Ilka Herbinger, Julia Lehmann, Elaine Madsen, Toshisada Nishida, Ulrich Reichard, Martha Robbins, Bernhard Thierry and Linda Vigilant for stimulating discussion on various parts of the thesis or different papers.
Also to Rocco Buchholz, Astrid Eckstein, Leila Kunstmann, Claudia Nebel and Silke Streiber for administrative support in Leipzig. To the people from the MPI EVA, which are not mentioned here, but have contributed in one way or the other to my work or simply to an enjoyable time in Leipzig. To my parents for their moral support.
Finally, I am very grateful to my wife Cathy, who I met in Côte d’Ivoire, for the wonderful time in the forest and for her never-ending support during our joint time in Leipzig – especially her positive influence on my English was most helpful indeed.
124 Bibliography
Abbott, D. H., Keverne, E. B., Bercovitch, F. B., Shively, C. A., Mendoza, S. P., Saltzman, W., Snowdon, C. T., Ziegler, T. E., Banjevic, M., Garland Jr., T. & Sapolsky, R. M. (2003). Are subordinates always stressed? A comparative analysis of rank differences in cortisol levels among primates. Hormones and Behavior. 43: 67-82. Abegg, C., Thierry, B. & Kaumanns, W. (1996). Reconciliation in three groups of lion- tailed macaques. International Journal of Primatology. 17: 803-816. Agresti, A. (1996). An introduction to categorical data analysis. New York: Wilely. Akaike, H. (1973). Information theory and an extension of the maximum likelihood principle. In: International symposium on information Theory (Ed. by B. N. Petran & F. Csàaki), pp. 267-281. Budapest: Akadèemiai Kiadi. Alexander, R. D. (1974). The evolution of social behavior. Annual Review of Ecology and Systematics. 5: 325-383. Altmann, J. (1974). Observational study of behavior: Sampling methods. Behaviour. 49: 227-267. Altmann, J., Hausfater, G. & Altmann, S. A. (1985). Demography of Amboseli baboons. American Journal of Primatology. 8: 113-125. Altmann, J., Alberts, S. C., Haines, S. A., Dubach, J., Muruthi, P., Coote, T., Geffen, E., Cheesman, D. J., Mututua, R. S., Saiyalel, S. N., Wayne, R. K., Lacy, R. C. & Bruford, M. W. (1996). Behavior predicts genetic structure in a wild primate group. Proceesings of the National Academy of Science of the USA. 93: 5797-5801. Anderson, D. P., Nordheim, E. V., Boesch, C. & Moermond, T. C. (2002). Factors influencing fission-fusion grouping in chimpanzees in the Taï National Park, Côte d'Ivoire. In: Behavioral diversity in chimpanzees and bonobos (Ed. by C. Boesch, G. Hohmann & L. F. Marchant), pp. 90-101. Cambridge: Cambridge University Press. Appleby, M. C. (1983). The probability of linearity in hierarchies. Animal Behaviour. 31: 600-608. Arnold, K. & Barton, R. A. (2001). Postconflict behavior of spectacled leaf monkeys (Trachypithecus obscurus). I. Reconciliation. International Journal of Primatology. 22: 243-266. 11: Bibliography
Arnold, K. & Whiten, A. (2001). Post-conflict behaviour of wild chimpanzees (Pan troglodytes schweinfurthii) in the Budongo Forest, Uganda. Behaviour. 138: 649- 690. Aureli, F., van Schaik, C. P. & van Hooff, J. A. R. A. M. (1989). Functional aspects of reconciliation among captive long-tailed macaques (Macaca fascicularis). American Journal of Primatology. 19: 39-51. Aureli, F. & van Schaik, C. P. (1991a). Post-conflict behaviour in long-tailed macaques (Macaca fascicularis): I. The social events. Ethology. 89: 89-100. Aureli, F. & van Schaik, C. P. (1991b). Post-conflict behaviour in long-tailed macaques (Macaca fascicularis): II. Coping with the Uncertainty. Ethology. 89: 101-114. Aureli, F. (1992). Post-conflict behaviour among wild long-tailed macaques (Macaca fascicularis). Behavioral Ecology and Sociobiology. 31: 329-337. Aureli, F., Cozzolino, R., Cordischi, C. & Scucchi, S. (1992). Kin-oriented redirection among Japanese macaques: An expression of revenge system? Animal Behaviour. 44: 283-291. Aureli, F., Veenema, H. C., van Panthaleon van Eck, C. J. & van Hooff, J. A. R. A. M. (1993). Reconciliation, consolation, and redirection in Japanese macaques (Macaca fuscata). Behaviour. 124: 1-21. Aureli, F., Das, M., Verleur, D. & van Hooff, J. A. R. A. M. (1994). Postconflict social interactions among Barbary macaques (Macaca sylvanus). International Journal of Primatology. 15: 471-485. Aureli, F. (1997). Post-conflict anxitety in non-human primates: The mediating role of emotion in conflict resolution. Aggressive Behavior. 23: 315-328. Aureli, F., Das, M. & Veenema, H. C. (1997). Differential Kinship Effect on reconciliation in three species of Macaques (Macaca fascicularis, M. fuscata,and M. sylvanus). Journal of Comparative Psychology. 111: 91-99. Aureli, F. & Smucny, D. (1998). New directions in conflict resolution research. Evolutionary Anthropology. 6: 115-119. Aureli, F., Preston, S. D. & de Waal, F. B. M. (1999). Heart rate response to social interactions in free-moving rhesus macaques (Macaca mulatta): A pilot study. Journal of Comparative Psychology. 113: 59-65. Aureli, F. & de Waal, F. B. M. (2000a). Natural conflict resolution. Berkeley: University of California Press.
126 11: Bibliography
Aureli, F. & de Waal, F. B. M. (2000b). Why natural conflict resolution? In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 3-10. Berkeley Los Angeles London: University of California Press. Aureli, F. & Smucny, D. (2000). The role of emotion in conflict and conflict resolution. In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 199-224. Berkley: California University Press. Aureli, F., Cords, M. & van Schaik, C. P. (2002). Conflict resolution following aggression in gregarious animals: A predictive framework. Animal Behaviour. 64: 325-343. Baker, K. C. & Smuts, B. B. (1994). Social relationships of female chimpanzees. In: Chimpanzee cultures (Ed. by R. W. Wrangham, W. C. McGrew, F. B. M. de Waal & P. G. Heltne), pp. 227-242. Cambridge: Harvard University Press. Barton, R. A. & Whiten, A. (1993). Feeding competition among female olive baboons, Papio anubis. Animal Behaviour. 46: 777-789. Bercovitch, F. B. (1995). Female cooperation, consortship maintenance, and male mating succes in savanna baboons. Animal Behaviour. 50: 137-149. Bernstein, I. S. (1981). Dominance: The baby and the bathwater. Behavioral and Brain Sciences. 4: 419-457. Bernstein, I. S., Williams, L. & Ramsay, M. (1983). The expression of aggression in old world monkeys. International Journal of Primatology. 4: 113-125. Bibikow, D. I. (1990). Der Wolf. Wittenberg: A. Ziemsen Verlag. Björnsdotter, M., Larsson, L. & Ljungberg, T. (2000). Post-conflict affiliation in two captive groups of black-and-white guereza (Colobus guereza). Ethology. 106: 289- 300. Bland, J. M. & Altman, D. G. (1995). Calculating correlation coefficients with repeated observations: Part 1-correlation within subjects. British Medical Journal. 310: 446. Boesch, C. & Boesch, H. (1983). Optimisation of nut-cracking with natural hammers by wild chimpanzees. Behaviour. 83: 265-286. Boesch, C. & Boesch, H. (1984a). Mental map in wild chimpanzees: An analyseis of hammer transports for nut cracking. Primates. 25: 160-170. Boesch, C. & Boesch, H. (1984b). Possible causes of sex differences in the use of natural hammers by wild chimpanzees. Journal of Human Evolution. 13: 415-440. Boesch, C. & Boesch, H. (1989). Hunting behaviour of wild chimpanzees in the Tai National Park. American Journal of Physical Anthropology. 78: 547-573.
127 11: Bibliography
Boesch, C. (1991). The effects of leopard predation on grouping patterns in forest chimpanzees. Behaviour. 117: 220-242. Boesch, C. (1994a). Chimpanzees - red colobus monkeys: A predator - prey system. Animal Behaviour. 47: 1135-1148. Boesch, C. (1994b). Cooperative hunting in wild chimpanzees. Animal Behaviour. 48: 653-667. Boesch, C. & Boesch-Achermann, H. (2000). The chimpanzees of the Taï Forest. Oxford: Oxford University Press. Boesch, C., Nohon, G., Siedel, H., Kpazahi, H. & Vigilant, L. (in prep.). Paternity and male competition among Taï chimpanzees. N.N. Breed, M. D. & Rasmussen, C. D. (1980). Behavioural strategies during intermale agonictic interactions in a cockroach. Animal Behaviour. 28: 1063-1069. Buston, P. (2003). Size and growth modification in clownfish. Nature. 424: 145-146. Butovskaya, M., Verbeek, P., Ljungberg, T. & Lunardini, A. (2000). A multicultural view of peacemaking among young children. In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 243-258. Berkeley: University of California Press. Bygott, J. D. (1979). Agonistic behavior, dominance, and social structure in wild chimpanzees of the Gombe National Park. In: The great apes (Ed. by D. A. Hamburg & E. R. McCown), pp. 405-427. Menlo Park, CA: Benjamin & Cummings. Call, J., Aureli, F. & de Waal, F. B. M. (1999). Reconciliation patterns among stumptailed macaques: A multivariate approach. Animal Behaviour. 58: 165-172. Caraco, T., Martindale, S. & Pulliam, H. R. (1980). Avian flocking in the presence of a predator. Nature. 285: 400-401. Castles, D. L., Aureli, F. & de Waal, F. B. M. (1996). Variation in conciliatory tendency and relationship quality across groups of pigtail macaques. Animal Behaviour. 52: 389-403. Castles, D. L. & Whiten, A. (1998a). Post-conflict behaviour of wild olive baboons. I. Reconciliation, redirection and consolation. Ethology. 104: 126-147. Castles, D. L. & Whiten, A. (1998b). Post-conflict Behaviour of wild olive baboons. II. Stress and self-directed behaviour. Ethology. 104: 148-160. Castles, D. L., Whiten, A. & Aureli, F. (1999). Social anxiety, relationships and self- directed behaviour among wild female olive baboons. Animal Behaviour. 58: 1207- 1215. Chase, I. D. (1974). Models of hierarchy formation in animal societies. Behavioral Science. 19: 374-382.
128 11: Bibliography
Cheney, D. L. & Seyfarth, R. M. (1989). Redirected aggression and reconciliation among vervet monkeys, Cercopithecus aethiops. Behaviour. 110: 258-275. Cheney, D. L., Seyfarth, R. M. & Silk, J. B. (1995). The role of grunts in reconciliation opponents and faciliating interactions among adult female baboons. Animal Behaviour. 50: 249-257. Cheney, D. L. & Seyfarth, R. M. (1997). Reconciliatory grunts by dominant female baboons influence victims' behaviour. Animal Behaviour. 54: 409-418. Clutton-Brock, T. H., Greenwood, P. J. & Powell, R. P. (1976). Ranks and relationships in highland ponies and highland cows. Zeitschrift für Tierpsychologie. 41: 202-216. Clutton-Brock, T. H. & Harvey, P. H. (1978). Mammals, resources and reproductive strategies. Nature. 273: 191-195. Clutton-Brock, T. H. & Albon, S. D. (1979). The roaring of red deer and the evolution of honest advertisement. Behaviour. 69: 145-170. Cohen, J. (1988). Statistical power analysis for behavioral sciences. Hillsdale, NY: Erlbaum. Constable, J. L., Ashley, M. V., Goodall, J. & Pusey, A. E. (2001). Noninvasive paternity assignment in Gombe chimpanzees. Molecular Ecology. 10: 1279-1300. Cords, M. (1992). Post-conflict reunions and reconciliation in long-tailed macaques. Animal Behaviour. 44: 57-61. Cords, M. & Aureli, F. (1993). Patterns of reconciliation among juvenile long-tailed macaques. In: Juvenile Primates: Life history, development, and behavior (Ed. by M. E. Pereira & L. A. Fairbanks), pp. 271-284. Oxford: Oxford University Press. Cords, M. & Thurnheer, S. (1993). Reconciliation with valuable partners by long-tailed macaques. Ethology. 93: 315-325. Cords, M. & Aureli, F. (1996). Reasons for reconciliation. Evolutionary Anthropology. 5: 42-45. Cords, M. & Killen, M. (1998). Conflict resolution in human and nonhuman primates. In: Piaget, evolution, and development (Ed. by J. Langer & M. Killen), pp. 193-218. Mahwah, N.J.: Lawrence Erlbaum Associates. Cords, M. & Aureli, F. (2000). Reconciliation and relationship qualities. In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 177-198. Berkeley: University of California Press. Côté, S. D. (2000). Dominance hierarchies in female mountain goats: Stability, aggressiveness and determinants of rank. Behaviour. 137: 1541-1566.
129 11: Bibliography
Côté, S. D. & Festa-Bianchet, M. (2001). Offspring sex ratio in relation to maternal age and social rank in mountain goats (Oreamnos americanos). Behavioral Ecology and Sociobiology. 49: 260-265. Creel, S., Creel, N. M., Mills, M. G. L. & Monfort, S. L. (1997). Rank and reproduction in cooperatively breeding African wild dogs: Behavioral and endocrine correlates. Behavioral Ecology. 8: 298-306. Creel, S. (2001). Social dominance and stress hormones. Trends in Ecology & Evolution. 16: 491-497. Crockford, C. & Boesch, C. (2003). Context-specific calls in wild chimpanzees, Pan troglodytes verus: analysis of barks. Animal Behaviour. 66: 115-125. Crook, J. H. & Gartlan, J. S. (1966). Evolution of primate societies. Nature. 210: 1200- 1203. Crowley, P. H. (2000). Hawks, doves, and mixed symmetry games. Journal of theoretical Biology. 204: 543-563. Cytel Software (2001). StatXact 5. Cambridge, MA: Cytel Software Corporation. Das, M., Penke, Z. & van Hooff, J. A. R. A. M. (1997). Affiliation between aggressors and third parties following conflicts in long-tailed macaques (Macaca fascicularis). International Journal of Primatology. 18: 159-181. Das, M., Penke, Z. & van Hooff, J. A. R. A. M. (1998). Post-conflict affiliation and stress-related behavior of long-tailed macaque aggressors. International Journal of Primatology. 19: 53-71. Das, M. (2000). Conflict management via third parties: Post-conflict affiliation of the aggressor. In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 263-280. Berkeley, CA: University of California Press. Davies, N. B. & Halliday, T. R. (1979). Competitive mate searching in male common toads, Bufo bufo. Animal Behaviour. 27: 1253-1267. Dawkins, R. (1976). The selfish gene. Oxford: Oxford University Press. de Ruiter, J. R. & van Hoof, J. A. R. A. M. (1993). Male dominance rank and reproductive success in primate groups. Primates. 34: 513-523. de Vries, H. (1995). An improved test of linearity in dominance hierarchies containing unknown or tied relationships. Animal Behaviour. 50: 1375-1389. de Waal, F. B. M. (1977). The organisation of agonistic relations within two captive groups of Java-monkeys (Macaca fascicularis). Zeitschrift für Tierpsychology (Journal of Comaparative Ethology). 44: 225-282.
130 11: Bibliography de Waal, F. B. M. (1978). Exploitative and familiarity-dependent support strategies in a colony of semi-free living chimpanzees. Behaviour. 66: 268-311. de Waal, F. B. M. & van Roosmalen, A. (1979). Reconciliation and Consolation among chimpanzees. Behavioral Ecology and Sociobiology. 5: 55-66. de Waal, F. B. M. & Hoekstra, J. A. (1980). Contexts and Predictability of aggression in chimpanzees. Animal Behaviour. 28: 929-937. de Waal, F. B. M. & van Hooff, J. A. R. A. M. (1981). Side-directed communication and agonistic interactions in chimpanzees. Behaviour. 77: 164-198. de Waal, F. B. M. (1982). Chimpanzee politics. London: Jonathan Cape. de Waal, F. B. M. & Yoshihara, D. (1983). Reconciliation and redirected affection in rhesus monkeys. Behaviour. 85: 224-241. de Waal, F. B. M. (1984). Sex differences in the formation of coalitions among chimpanzees. Ethol. Sociobiol. 5: 239-255. de Waal, F. B. M. & Luttrell, L. M. (1985). The formal hierarchy of rhesus macaques (Macaca mulatta): An investigation of bared-teeth display. American Journal of Primatology. 9: 73-86. de Waal, F. B. M. (1986). The integration of dominance and social bonding in primates. Quaterly Review of Biology. 61: 459-479. de Waal, F. B. M. (1987). Tension regulation and nonreproductive functions of sex in captive bonobos (Pan paniscus). National Geographic Research. 3: 318-335. de Waal, F. B. M. & Luttrell, L. M. (1988). Mechanisms of social reciprocity in three primate species: Symmetrical relationship characteristics or cognition? Ethology and Sociobiology. 9: 101-118. de Waal, F. B. M. & Ren, R. M. (1988). Comparison of the reconciliation behavior of stumptail and rhesus macaques. Ethology. 78: 129-142. de Waal, F. B. M. (1989a). Food sharing and reciprocal obligations among chimpanzees. Journal of Human Evolution. 18: 433-459. de Waal, F. B. M. (1989b). Dominance 'style' and primate social organisation. In: Comparative socioecology (Ed. by V. Staden & R. A. Foley), pp. 243-263. Oxford: Blackwell. de Waal, F. B. M. (1992). Appeasment, celebration, and food sharing in the two Pan species. In: Topics in Primatology (Ed. by T. Nishida, W. C. McGrew, P. R. Marler, M. Pickford & F. B. M. de Waal), pp. 37-50. Tokyo: University of Tokyo Press.
131 11: Bibliography de Waal, F. B. M. (1993). Reconciliation among primates: A review of empirical evidence and unresolved issues. In: Primate social conflict (Ed. by W. A. Mason & S. P. Mendoza), pp. 111-144. New York: State University of New York Press. de Waal, F. B. M. (1996a). Conflict as negotiation. In: Great ape societies (Ed. by W. C. McGrew, L. F. Marchant & T. Nishida), pp. 159-172. Cambridge: Cambridge University Press. de Waal, F. B. M. (1996b). Good natured: The origins of right and wrong in humans and other animals. Cambridge, MA: Harvard University Press. de Waal, F. B. M. & Aureli, F. (1996). Consolation, reconciliation, and a possible cognitive difference between macaques and chimpanzees. In: Reaching into Thought: the minds of great apes (Ed. by A. E. Russon, K. A. Bard & S. Taylor Parker), pp. 80-110. Cambridge: Cambridge University Press. de Waal, F. B. M. & Aureli, F. (1997). Conflict resolution and distress alleviation in monkeys and apes. In: The integrative neurobiology of affiliation (Ed. by C. S. Carter, B. Kirkpatrick & I. Lenderhendler), pp. 317-328. New York: Annals of the New York Academy of Science. de Waal, F. B. M. (2000a). The first kiss. In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 15-33. Berkeley: University of California Press. de Waal, F. B. M. (2000b). Primates - A natural heritage of conflict resolution. Science. 289: 586-590. de Waal, F. B. M. & Aureli, F. (2000). Shared principles and unanswered questions. In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 375-379. Berkeley Los Angeles London: University of California Press. Demaria, C. & Thierry, B. (2001). A comparative study of reconciliation in Rhesus and Tonkean macaques. Behaviour. 138: 397-410. Deschner, T., Heistermann, M., Hodges, K. & Boesch, C. (2003). Timing and probability of ovulation in relations to sex skin swellings in wild West African chimpanzees, Pan troglodytes verus. Animal Behaviour. DOI: 10.1006/anbe.2003.2210. Dixson, A. F., Bossi, T. & Wickings, E. J. (1993). Male dominance and genetically determined reproductive success in the Mandrill (Mandrillus sphinx). Primates. 34: 525-532. Dixson, A. F. (1998). Primate sexuality. Oxford: Oxford University Press. Dobson, A. J. (1990). An introduction to generalized linear models. New York: Chapman & Hall.
132 11: Bibliography
Dow, M. M. & de Waal, F. B. M. (1989). Assignment methods for the analysis of network subgroup interactions. Social Networks. 11: 237-255. Drews, C. (1993). The concept and definition of dominance in animal behaviour. Behaviour. 125: 283-313. Drukker, B., Nieuwenhuijsen, K., van der Werff ten Bosch, J. J., van Hooff, J. A. R. A. M. & Slob, A. K. (1991). Harassment of sexual interactions among stumptailed macaques, Macaca arctoides. Animal Behaviour. 42: 171-182. Dunbar, R. I. M. (1988). Primate social systems. Bekenham: Croom Helm Ltd. Ebersberger, I., Metzler, D., Schwarz, C. & Pääbo, S. (2002). Genomwide comparison of DNA sequences between humans and chimpanzees. American Journal of Human Genetics. 70: 1490-1497. Efron, B. & Tibshirani, R. J. (1994). An introduction to the bootstrap. Boca Raton: CRC Press LLC. Eisenberg, J. F., Muckenhirn, N. A. & Rudran, R. (1972). The relation between ecology and social structure in primates. Science. 176: 863-874. Emlen, S. T. & Oring, L. W. (1977). Ecology, sexual selection, and the evolution of mating systems. Science. 197: 215-223. Fawcett, K. A. (2000). Female relationships and food availability in a forest community of chimpanzees. PhD University of Edinburgh, Edinburgh, UK. Fawcett, K. & Muhumuza, G. (2000). Death of a wild chimpanzee community memeber: Possible outcome of intense sexual competition. American Journal of Primatology. 51: 243-247. Fricke, H. W. (1979). Mating system, resource defense and sex change in the anemonefish Amphiprion akallopisos. Zeitschrift für Tierpsychologie. 50: 313-326. Fry, D. P. (2000). Conflict management in a cross-cultural perspective. In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 334-351. Berkeley: University of California Press. Fuentes, A., Malone, N., Sanz, C., Meatheson, M. & Vaughan, L. (2002). Conflict and post-conflict behavior in a small group of chimpanzees. Primates. 43: 223-235. Furuichi, T. & Hashimoto, C. (2002). Why female bonobos have a lower copulation rate during estrus than chimpanzees. In: Behavioural diversity in chimpanzees and bonobos (Ed. by C. Boesch, G. Hohmann & L. F. Marchant), pp. 156-167. Cambridge: Cambridge University Press. Gagneux, P., Gonder, M. K., Goldberg, T. L. & Morin, P. A. (2001). Gene flow in wild chimpanzee populations: what genetic data tell us about chimpanzee movement
133 11: Bibliography
over space and time. Philosophical Transactions of the Royal Society of London, B. 365: 889-897. Gagneux, P. (2002). The genus Pan: Population genetics of an endangered outgroup. Trends in Genetics. 18: 327-330. Garber, P. A. (1997). One for all and breeding for one: Cooperation and competition as a Tamarin reproductive strategy. Evolutionary Anthropology. 5: 187-199. Gerloff, U., Hatung, B., Fruth, B., Hohmann, G. & Tautz, D. (1999). Intracommunity relationships, dispersal pattern and paternity success in a wild living community of bonobos (Pan paniscus) determined from DNA analysis of faecal samples. Proceedings of the Royal Society of London, Series B. 266: 1189-1195. Goodall, J. (1983). Population dynamics during a 15 year period in one community of free-living chimpanzees in the Gombe National Park, Tanzania. Z. Tierpsychol. 61: 1-60. Goodall, J. (1986). The chimpanzees of Gombe - Patterns of behavior. Cambridge, MA: The Belknap Press of Harvard Univerity Press. Gust, D. A. & Gordon, T. P. (1993). Conflict resolution in sooty mangabeys. Animal Behaviour. 46: 685-694. Hack, M. A. (1997). The energetic costs of fighting in the house cricket, Acheta domesticus L. Behavioral Ecology. 8: 28-36. Hamilton, W. D. (1964). The genetical evolution of social behaviour (I,II). Journal of theoretical Biology. 7: 1-32. Hamilton, W. D. (1967). Extraordinary sex ratios. Science. 156: 477-488. Hamilton, W. D. (1971). Geometry of the selfish herd. Journal of theoretical Biology. 7: 295-311. Hand, J. L. (1986). Resolution of social conflicts: Dominance, egalitarianism, spheres of dominance, and game theory. The quaterly Review of Biology. 61: 201-220. Hasegawa, T. & Hiraiwa-Hasegawa, M. (1983). Opportunistic and rstrictive matings among wild chimpanzees in the Mahale Mountains, Tanzania. J. Ethol. 1: 75-85. Hemelrijk, C. K. & Ek, A. (1991). Reciprocity and interchange of grooming and 'support' in captive chimpanzees. Animal Behaviour. 41: 923-935. Herbinger, I., Boesch, C. & Rothe, H. (2001). Territory characteristics among three neighbouring chimpanzee communities in the Taï National Park, Côte d'Ivoire. International Journal of Primatology. 22: 143-167.
134 11: Bibliography
Hill, K., Boesch, C., Goodall, J., Pusey, A., Williams, J. & Wrangham, R. W. (2001). Mortality rates among wild chimpanzees. Journal of Human Evolution. 40: 437-450. Hinde, R. A. (1967). Nature of aggression. New Society. 9: 302-304. Hinde, R. A. (1979). The nature of social structure. In: The grat apes (Ed. by D. A. Hamburg & E. R. McCown), pp. 295-315. Menlo Park, California: Benjamin/Cummings. Hiraiwa-Hasegawa, M., Byrne, R. W., Takasaki, H. & Byrne, J. M. E. (1986). Aggression towards large carnivores by wild chimpanzees of Mahale Mountains National Park, Tanzania. Folia Primatologica. 47: 8-13. Hrdy, S. B. & Hrdy, D. B. (1976). Hierarchical relations among female Hanuman langurs (Primates: Colobinae, Presbytis entellus). Science. 193: 913-915. Huffmann, M. A. (1990). Some socio-behavioral manifestations of old age. In: The chimpanzees of the Mahale mountains (Ed. by T. Nishida), pp. 237-255. Tokyo: University of Tokyo Press. Huntingford, F. A. & Turner, A. (1987). Animal conflict. London: Chapman and Hall. Insightful Corporation (2001). S-PLUS 6 for Windows. Seattle, WA: Insightful Corporation. Jackson, W. M. (1988). Can individual differences in history of dominance explain the development of linear dominance hierarchies? Ethology. 79: 71-77. Jaeger, R. G. (1981). Dear enemy recognition and the costs of aggression between salamanders Plethodon cinereus. American Naturalist. 117: 962-974. Janson, C. (1985). Aggressive competition and individual food consumption in wild brown capuchin monkeys (Cebus apella). Behavioral Ecology and Sociobiology. 18: 125-138. Janson, C. H. (1988a). Intra-specific food competition and primate social structure: A synthesis. Behaviour. 105: 1-17. Janson, C. H. (1988b). Food competition in brown capuchin monkeys (Cebus apella): Quantitative effects of group size and tree productivity. Behaviour. 105: 53-76. Janson, C. H. (2000). Primate socio-ecology: The end of a golden age. Evolutionary Anthropology. 9: 73-86. Johnson, R. H. & Nelson, D. R. (1973). Agonistic display in the grey reef shark Carcharhinus menisorrah and its relationship to attacks on man. Copeia. 1: 76-84.
135 11: Bibliography
Johnstone, R. A. & Dugatkin, L. A. (2000). Coalition formation in animals and the nature of winner and loser effect. Proceedings of the Royal Society London B. 267: 17-21. Judge, P. G. (1991). Dyadic and triadic reconciliation in pigtail macaques (Macaca nemestrina). American Journal of Primatology. 23: 225-237. Judge, P. G. (2000). Coping with crowded conditions. In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 129-154. Berkeley: Univeristy of California Press. Kaessmann, H., Wiebe, V., Weiss, G. & Pääbo, S. (2001). Great ape DNA sequences reveal a reduced diversity and expansion in humans. Nature Genetics. 27: 155-156. Kaessmann, H. & Pääbo, S. (2002). The genetical history of humans and great apes. Journal of Internal Medicine. 251: 1-18. Kappeler, P. M. & van Schaik, C. P. (1992). Methodological and evolutionary aspects of reconciliation among primates. Ethology. 92: 51-69. Kappeler, P. M. (1993). Reconciliation and post-conflict behaviour in ringtailed lemurs, Lemur catta and redfronted lemurs, Eulemur fulvus rufus. Animal Behaviour. 45: 901-915. Kelly, H. H., Shure, G. H., Deutsch, M., Faucheux, C., Lanzetta, J. T., Moscovici, S., Nuttin, J. M. J., Rabbie, J. M. & Thibaut, J. W. (1970). A comparative experimental study of negotiation behavior. Journal of Personality and Social Psychology. 16: 411-438. Keul, J., Doll, E. & Keppler, D. (1967). The substrate supply of the human skeletal muscle at rest, during and after work. Experientia. 23: 974-979. Koenig, A. (2000). Competitive regimes in forest-dwelling Hanuman langur females (Semnopithecus entellus). Behavioral Ecology and Sociobiology. 48: 93-109. Koenig, A. (2002). Competition for resources and its behavioral consequences among female primates. International Journal of Primatology. 23: 759-783. Koyama, N. F. (2001). The long-term effects of reconciliation in Japanese macaques Macaca fuscata. Ethology. 107: 975-987. Kozan, K. (1991). Interpersonal conflict management styles of Jordanian managers. In: Conflict resolution - Cross-cultural perspectives (Ed. by K. Avruch, P. W. Black & J. A. Scimecca), pp. 85-105. Westport, CT: Greenwood Press. Krebs, J. R. (1977). Song and territory in the great tit. In: Evolutionary Ecology (Ed. by B. Stonehouse & C. M. Perrins), pp. 47-62. London: Macmillan.
136 11: Bibliography
Krebs, J. R. & Kacelnik, A. (1991). Decision-making. In: Behavioural ecology - An evolutionary approach (Ed. by J. R. Krebs & N. B. Davis), pp. 105-136. Oxford: Blackwell Sciebtific Publications. Kummer, H. (1978). On the value of social relationships to nonhuman primates: A heuristic scheme. Soc. Sci. Info. 17: 687-705. Kutsukake, N. & Castles, D. L. (2001). Reconciliation and variation in post-conflict stress in Japanese macaques (Macaca fuscata fuscata): testing the integrated hypothesis. Animal Cognition. 4: 259-268. Landau, H. G. (1951). On dominance relations and the structure of animal societies: I. Effect of inherent chrarcteristics. Bulletin of Mathematical Biophysics. 13: 1-19. Launhardt, K., Borries, C., Hardt, C., Epplen, J. T. & Winkler, P. (2001). Paternity analysis of alternative male reproductive routes among the langurs (Semnopithecus entellus) of Ramnagar. Animal Behaviour. 61: 53-64. Leca, J.-B., Fornasieri, I. & Petit, O. (2002). Aggression and reconciliation in Cebus capucinus. International Journal of Primatology. 23: 979-998. Lehmann, J. & Boesch, C. (in prep.). Do chimpanzees (Pan troglodytes verus) in the Taï National Park (Côte d'Ivoire) exhibit sex-specific ranging patterns? Lehmann, J. & Boesch, C. (in press). Social influences on ranging patterns among chimpanzees (Pan troglodytes verus) in the Taï National Park, Côte d'Ivoire. Behavioral Ecology. Levine, S., Coe, C. & Wiener, S. (1989). The psychoneuroendocrinology of stress: A psychobiological perspective. In: Psychoendocrinology (Ed. by S. Levine & R. Bursh), pp. 181-207. New York: Academic Press. Lewis, R. (2002). Beyond dominance: The importance of leverage. Quaterly Review of Biology. 77: 149-164. Linsenmair, K. E. (1972). Die Bedeutung familienspezifischer "Abzeichen" für den Familienzusammenhalt bei der sozialen Wüstenassel Hemilepistus reaumuri Audouin u. Savigny (Crustacea, Isopoda, Oniscoidea). Zeitschrift für Tierpsychologie. 31: 131-162. Lorenz, K. (1963). Das sogenannte Böse (engl.: On aggression). Wien: Verlag Dr. G. Borotha-Schoeller. Maestripieri, D., Schino, G., Aureli, F. & Troisi, A. (1992). A modest proposal: Displacement activities as an indicator of emotions in primates. Animal Behaviour. 44: 967-979.
137 11: Bibliography
Manly, B. F. J. (1997). Randomization, bootstrap and Monte Carlo methods in biology. London: Chapman and Hall. Manson, J. H. & Perry, S. (2000). Correlates of self-directed behaviour in wild white- faced capuchins. Ethology. 106: 301-317. Mason, W. A. & Mendoza, S. P. (1993). Primate social conflict. pp. 419. Albany NY: State University of New York Press. Matsumoto-Oda, A. (1999). Female choice in the opportunistic mating of wild chimpanzees (Pan troglodytes schweinfurthii) at Mahale. Behavioral Ecology and Sociobiology. 46: 258-266. Matsumura, S. (1996). Postconflict affiliative contacts between former opponents among wild moor macaques (Macaca maurus). American Journal of Primatology. 38: 211-219. Matsumura, S. & Okamoto, K. (2000). Conflict, social costs, and game theory (Box 5.1). In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 79-81. Berkeley: University of California Press. Maynard Smith, J. & Parker, G. A. (1976). The logic of asymmetrical contests. Animal Behaviour. 24: 159-175. Maynard Smith, J. (1982). Evolution and the theory of games. Cambridge: Cambridge Univerity Press. Maynard Smith, J. & Riechert, S. E. (1984). A conflicting-tendency model of spider agonistic behaviour: Hybrid-pure population line comparison. Animal Behaviour. 32: 564-578. McCullagh, P. & Nelder, J. A. (1989). Generalized linear models. New York: Chapman & Hall. McElligott, A. G. & Hayden, T. J. (2000). Lifetime mating success, sexual selection and life history of fallow bucks (Dama dama). Behavioral Ecology and Sociobiology. 48: 203-210. McEnery, J. H. (1985). Towards a new concept of conflict evaluation. Conflict. 6: 37-87. McKenna, M. J., Morton, J., Selig, S. E. & Snow, R. J. (1999). Creatine supplementation increases muscle total creatine but not maximal intermitted exercise performance. Journal of Applied Physiology. 87: 2244-2252. Meerlo, P., Sgoifo, A., De Boer, S. F. & Koolhaas, J. M. (1999). Long-lasting consequences of a social conflict in rats: Behavior during the interaction predicts subsequent changes in daily rhythms of heart rate, temperature, and activity. Behavioral Neuroscience. 113: 1283-1290.
138 11: Bibliography
Mesnick, S. L. & Le Boeuf, B. J. (1991). Sexual behavior of male northern elephant seals. II. Female response to potentially injurious encounters. Behaviour. 117: 263- 280. Mitani, J. C. & Watts, D. P. (1999). Demographic influences on the hunting behavior of chimpanzees. American Journal of Physical Anthropology. 109: 439-454. Mitani, J. C. & Watts, D. P. (2001). Why do chimpanzees hunt and share meat? Animal Behaviour. 61: 915-924. Mitani, J. C., Watts, D. P. & Muller, M. N. (2002). Recent development in the study of wild chimpanzee behavior. Evolutionary Anthropology. 11: 9-25. Mitchell, C. L., Boinski, S. & van Schaik, C. P. (1991). Competitive regimes and female bonding in two species of squirrel monkeys (Saimiri oerstedi and S. sciureus). Behavioral Ecology and Sociobiology. 28: 55-60. Morgan, D. & Sanz, C. (2003). Naïve encounters with chimpanzees in the Goualougo Triangle, Republic of Congo. International Journal of Primatology. 24: 369-381. Mori, A. (1979). Analysis of population changes by measurement of body weight in the Koshima troop of Japanese monkeys. Primates. 20: 371-397. Moynihan, M. (1998). The social regulation of competition and aggression in animals. Washington: Smithsonian Institution Press. Mundry, R. & Fischer, J. (1998). Use of statistical programs for nonparametric tests of small samples often leads to incorrect P values: Examples from Animal Behaviour. Animal Behaviour. 56: 256-259. Mundry, R. (1999). Testing related samples with missing values: A permutation approach. Animal Behaviour. 58: 1143-1153. Muroyama, Y. (1994). Exchange of grooming for allomothering in female patas monkeys. Behaviour. 128: 103-119. Nader, L. (1991). Harmony models and the construction of law. In: Conflict resolution: Cross-cultural perspectives (Ed. by K. Avruch, P. W. Black & J. A. Scimecca), pp. 41- 59. Westport, CT: Greenwood Press. Neat, F. C., Taylor, A. C. & Huntingford, F. A. (1998). Proximate costs of fighting in male cichlid fish: The role of injuries and energy metabolism. Animal Behaviour. 55: 875-882. Neu, J. (2000). Interpersonal dynamics in international conflict madiation (Box 4.1). In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 65-67. Berkeley: University of California Press.
139 11: Bibliography
Newton-Fisher, N. E. (1999a). Association by male chimpanzees: A social tactic? Behaviour. 136: 705-730. Newton-Fisher, N. E. (1999b). The diet of chimpanzees in the Budongo Forest Reserve, Uganda. African Journal of Ecology. 37: 344-354. Newton-Fisher, N. E. (2002). Relationships of male chimpanzees in the Budongo Forest, Uganda. In: Behavioral diversity in chimpanzees and bonobos (Ed. by C. Boesch, G. Hohmann & L. F. Marchant), pp. 124-137. Cambridge: Cambridge University Press. Niemeyer, C. L. & Anderson, J. R. (1983). Primate harassment of matings. Ethology and Sociobiology. 4: 205-220. Nishida, T. (1968). The social group of wild chimpanzees in the Mahali Mountains. Primates. 9: 167-224. Nishida, T. (1979). The social structure of chimpanzees of the Mahale mountains. In: The great apes (Ed. by D. A. Hamburg & E. R. McCown), pp. 73-122. Menlo Park, California: Benjamin Cummings Publishing Company. Nishida, T. (1983). Alpha status and agonistic alliance in wild chimpanzees (Pan troglodytes schweinfurthii). Primates. 24: 318-336. Nishida, T. & Hiraiwa-Hasegawa, M. (1985). Response to a stranger mother-son pair in wild chimpanzee: A case report. Primates. 26: 1-13. Nishida, T., Hiraiwa-Hasegawa, M., Hasegawa, T. & Takahata, Y. (1985). Group extinction and female transfer in wild chimpanzees in the Mahale National Park, Tanzania. Zeitschrift für Tierpsychologie. 67: 284-301. Nishida, T. (1989). Social interactions between resident and immigrant female chimpanzees. In: Understanding chimpanzees (Ed. by P. G. Heltne & L. Marquardt), pp. 68-89. Cambridge, MA: The Chicago Academy of Science. Nishida, T. (1990). The chimpanzees of the Mahale Mountains: Sexual and life history strategies. Tokyo: University of Tokyo Press. Nishida, T. & Hosaka, K. (1996). Coalition strategies among adult male chimpanzees of Mahale Mountains, Tanzania. In: Great ape societies (Ed. by W. C. McGrew, L. F. Marchant & T. Nishida), pp. 114-134. Cambridge: Cambridge University Press. Noë, R., de Waal, F. B. M. & van Hooff, J. A. R. A. M. (1980). Types of dominance in a chimpanzee colony. Folia Primatologica. 34: 90-110. Noldus (1989). The Observer - Software for behavioural research. Wageningen: Noldus Information Technology b.v. Noldus (1998). MatMan. Wageningen: Noldus Information Technology b.v.
140 11: Bibliography
O'Brien, T. G. (1993). Asymmetries in grooming interactions between juvenile and adult female wedge-capped capuchin monkeys. Animal Behaviour. 46: 929-938. Ohsawa, H., Inoue, M. & Takenaka, O. (1993). Matings strategy and reproductive success of male patas monkeys. Primates. 34: 533-544. Palombit, R. A. (1993). Lethal territorial aggression in a white-handed gibbon. American Journal of Primatology. 31: 311-318. Parker, G. A. (1974). The reproductive behaviour and nature of sexual selection in Scatophaga stercoraria. IX. Spatial distribution of fertilisation rates and evolution of male search strategy within the reproductive area. Evolution. 28: 93-108. Parker, G. A. & Rubenstein, D. I. (1981). Role assessment, reserve strategy, and acquisition of information in asymmetric animal conflicts. Animal Behaviour. 29: 221-240. Paul, A. (2002). Sexual selection and mate choice. International Journal of Primatology. 23: 877-904. Pereira, M. E., Schill, J. L. & Charles, E. P. (2000). Reconciliation in captive Guyanese squirrel monkeys (Saimiri sciureus). American Journal of Primatology. 50: 159-167. Perez-Ruiz, A. L. & Mondragon-Ceballos, R. (1994). Rates of reconciliatory behaviors in stumptailed macaques: Effects of age, sex, rank and kinship. In: Current Primatology: Social development, learning and behaviour (Ed. by J. J. Roeder, B. Thierry, J. R. Anderson & N. Herrenschmidt), pp. 147-155. Strasbourg: Universite Louis Pasteur. Petit, O. & Thierry, B. (1994a). Reconciliation in a group of black macaques (Macaca nigra). Dodo. J. Wildl. Preserv. Trusts. 30: 89-95. Petit, O. & Thierry, B. (1994b). Reconciliation in a group of Guinea baboons. In: Current primatology. Social development, learning and behaviour (Ed. by J. J. Roeder, B. Thierry, J. R. Anderson & N. Herrenschmidt), pp. 137-145. Strasbourg: Universite Louis Pasteur Press. Pope, T. R. (1990). The reproductive consequences of male cooperation in red howler monkey: Paternity exclusion in multi and single male troops using genetic markers. Behavior, Ecology and Sociobiology. 27: 439-446. Popp, J. L. & de Vore, I. (1979). Aggressive competition and social dominance theory: Synopsis. In: The great apes (Ed. by D. A. Hamburg & E. R. McCown), pp. 317-338. Menlo Park, CA: Benjamin & Cummings.
141 11: Bibliography
Preuschoft, S. & van Schaik, C. P. (2000). Dominance and communication. In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 77-105. Berkeley: University of California Press. Preuschoft, S., Wang, X., Aureli, F. & de Waal, F. B. M. (2002). Reconciliation in captive chimpanzees: A re-evaluation with controlled methods. International Journal of Primatology. 23: 29-50. Pruetz, J. D. & Isbell, L. A. (2000). Correlations of food distribution and patch size with agonistic interactions in female vervets (Chlorocebus aethiops) and patas monkeys (Erythrocebus patas) living in simple habitats. Behavioral Ecology and Sociobiology. 49: 38-47. Pruitt, D., Peirce, R. S., Zubek, J. M., McGillicuddy, N. B. & Welton, G. L. (1993). Determinants of short-term and long-term success in mediation. In: Conflict between people and groups (Ed. by S. Worchel & J. A. Simpson), pp. 60-75. Chicago: Nelson-Hall Publisher. Pusey, A., Williams, J. & Goodall, J. (1997). The influence of dominance rank on the reproductive success of female chimpanzees. Science. 277: 828-831. Rahim, M. A. (1986). Referent role and style of handling interpersonal conflict. Journal of Social Psychology. 125: 79-86. Range, F. & Noë, R. (2002). Familiarity and dominance relations among female Sooty mangabeys in the Taï National Park. American Journal of Primatology. 56: 137-153. Rasa, O. A. E. (1973). Marking behaviour and its social significance in the Afrcan Dwarf Mongoose, Helogale undulata rufula. Zeitschrift für Tierpsychologie. 32: 293-318. Rasa, O. A. E. (1979). The effects of crowding on the social relationships and behaviour of the dwarf mongoose (Helogale undulata rufula). Zeitschrift für Tierpsychologie. 49: 317-329. Reeve, H. K. (1992). Queen activation of lazy workers in colonies of the eusocial naked male-rat. Nature. 358: 147-149. Ren, R., Yan, K., Su, Y., Qi, H., Liang, B., Bao, W. & de Waal, F. B. M. (1991). The reconciliation behavior of Golden monkeys (Rhinopithecus roxellanae roxellanae) in small breeding groups. Primates. 32: 321-327. Riss, D. & Goodall, J. (1977). The recent rise to the alpha-rank in a population of free- living chimpanzees. Folia Primatologica. 27: 134-151. Rolland, N. & Roeder, J. J. (2000). Do Ringtailed lemurs (Lemur catta) reconcile in the hour post-conflict?: A pilot study. Primates. 41: 223-227.
142 11: Bibliography
Ross, A., Leveritt, M. & Riek, S. (2001). Neural influences on sprint running: training adaptations and acute responses. Sports Medicine. 31: 409-425. Ruble, T. L. & Thomas, K. W. (1976). Support for a two-dimensional model of conflict behavior. Organizational Behavior and Human Performance. 16: 143-155. Sakura, O. (1994). Factors affecting party size and composition of chimpanzees (Pan troglodytes verus) at Bossou, Guinea. International Journal of Primatology. 15: 167- 183. Samuels, A. & Flaherty, C. (2000). Peaceful conflict resolution in the sea? In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 229-231. Berkeley: University of California Press. Sapolsky, R. M. (1986). Stress, social status, and reproductive physiology in free-living baboons. In: Psychobiology of reproductive behavior (Ed. by D. Crews), pp. 291- 322. Englewood Cliffs: Prentice-Hall, Inc. Sapolsky, R. M. (1993). The physiology of dominance in stable versus unstable social hierarchies. In: Primate social conflict (Ed. by W. A. Manson & S. P. Mendoza), pp. 171-204. Albany: State University of New York Press. Sapolsky, R. (1998). Why Zebras don't get ulcers: A guide to stress, stress-related diseases and coping. New York: W. H. Freeman. Sapolsky, R. (2000). Psychological correlates of individual dominance style. In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 114-116. Berkeley, CA: University of California Press. Schino, G. (1998). Reconciliation in domestic goats. Behaviour. 135: 343-356. Schino, G., Rosati, L. & Aureli, F. (1998). Intragroup variation in conciliatory tendencies in captive Japanese macaques. Behaviour. 135: 897-912. Seyfarth, R. M., Cheney, D. L. & Marler, P. (1980). Monkey responses to three different alarm calls: evidence for predator classification and semantic communication. Science. 210: 801-803. Siegel, S. & Castellan, N. J. J. (1988). Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill, Inc. Silk, J. B. (1992). The patterning of intervention among male bonnet macaques: Reciprocity, revenge, and loyality. Current Anthropology. 33: 318-325. Silk, J. B. (1996). Why do primates reconcile? Evolutionary Anthropology. 5: 39-42. Silk, J. B., Cheney, D. L. & Seyfarth, R. M. (1996). The form and function of post- conflict interactions between female baboons. Animal Behaviour. 52: 259-268.
143 11: Bibliography
Small, M. F. (1981). Body fat, rank, and nutritional status in a captive group of rhesus macaques. International Journal of Primatology. 2: 91-96. Smith, I. P. & Taylor, A. C. (1993). The energetic cost of agonistic behaviour in the velvet swimming crab, Necora (=Liocarcinus) puber (L.). Animal Behaviour. 45: 375-391. Sommer, V., Denham, A. & Little, K. (2002). Postconflict behaviour of wild Indian langur monkeys: Avoidance of opponents but rarely affinity. Animal Behaviour. 63: 637-648. Sommer, V., Fowler, A. & Abanu, J. (2003). The Nigerian chimpanzee (Pan troglodytes vellerosus) at Gashaka: Two years of habituation efforts. Folia Primatologica. 74: 179-230. Stanford, C. B., Wallis, J., Mpongo, E. & Goodall, J. (1994). Hunting decisions in wild chimpanzees. Behaviour. 131: 1-18. Stat Soft (1999). Statistica for Windows - computer program manual. Tulsa: StatSoft Inc. Sterck, E. H. M. & Steenbeek, R. (1997). Female dominance relationships and food competition in the sympatric Thomas langur and Long-tailed macaque. Behaviour. 134: 749-774. Sterck, E. H. M., Watts, D. P. & van Schaik, C. P. (1997). The evolution of female social relationships in nonhuman primates. Behavioral Ecology and Sociobiology. 41: 291-309. Sugiyama, Y. (1984). Population dynamics of wild chimpanzees at Bossou, Guinea, between 1976 and 1983. Primates. 25: 391-400. Swedell, L. (1997). Patterns of reconciliation among captive Gelada baboons (Theriopithecus gelada): A brief report. Primates. 38: 325-330. Thierry, B. (1985). Patterns of agonistic interactions in three species of macaques (Macaca mulatta, M. fascicularis, M. tonkeana). Aggressive Behavior. 11: 223-233. Thierry, B. (1986). A comparative study of aggression and response to aggression in three species of macaque. In: Primate ontogeny, cognition and social behaviour (Ed. by J. G. Else & P. C. Lee), pp. 307-313. Cambridge: Cambridge University Press. Thierry, B. (2000). Covariation of conflict management patterns across macaque species. In: Natural Conflict Resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 106-128. Berkley: University of California Press.
144 11: Bibliography
Thierry, B., Bynum, E. L., Baker, S., Kinnaird, M. F., Matsumura, S., Muroyama, Y., O'Brien, T. G., Petit, O. & Watanabe, K. (2000a). The social repertoire of Sulawesi macaques. Primate Research. 16: 203-226. Thierry, B., Iwaniuk, A. N. & Pellis, S. M. (2000b). The influence of phylogeny on the social behaviour of macaques (Primates: Cercopithecidae, genus Macaca). Ethology. 106: 713-728. Thomas, K. W. (1992). Conflict and negotiation process in organizations. In: Handbook of industrial and organizational psychology (Ed. by M. D. Dunette & L. M. Hough), pp. 651-717. Chicago: Rand-McNally. Tilson, R. L. & Hamilton III, W. J. (1984). Social dominance and feeding patterns of spotted hyaenas. Animal Behaviour. 32: 715-724. Tinbergen, N. (1968). On war and peace in animals and man. Science. 160: 1411-1418. Trivers, R. L. (1974). Parent-offspring conflict. American Zoologist. 14: 249-264. Tutin, C. E. G. (1979). Mating patterns and reproductive strategies in a community of wild chimpanzees (Pan troglodytes schweinfuthii). Behavioral Ecology and Sociobiology. 6: 29-38. Tuttle, M. & Ryan, M. J. (1981). Bat predation and the evolution of frog vocalizations in the neotropics. Science. 214: 677-678. van der Dennen, J. & Falger, V. (1990). Sociobiology and conflict. London: Chapman and Hall. van Hooff, J. A. R. A. M. & van Schaik, C. P. (1992). Cooperation in competition: The ecology of primate bonds. In: Coalitions and alliances in humans and other animals (Ed. by A. H. Harcourt & F. B. M. de Waal), pp. 357-389. Oxford: Oxford University Press. van Schaik, C. P., van Noordwijk, M. A., Warsono, B. & Sutriono, E. (1983). Party size and early detection of predators in Sumatran forest primates. Primates. 24: 211-221. van Schaik, C. P. (1989). The ecology of social relationships amongst female primates. In: Comparative socioecology. The behavioural ecology of humans and other mammals (Ed. by V. Standen & R. A. Foley), pp. 195-218. Oxford: Blackwell. Veenema, H. C., Das, M. & Aureli, F. (1994). Methodological improvements for the study of reconciliation. Behavioural Processes. 31: 29-38. Verbeek, P. & de Waal, F. B. M. (1997). Postconflict behavior of captive brown capuchins in the presence and absence of attractive food. International Journal of Primatology. 18: 703-725.
145 11: Bibliography
Vigilant, L., Hofreiter, M., Siedel, H. & Boesch, C. (2001). Paternity and relatedness in wild chimpanzee communities. Proceedings of the National Academy of Sciencies USA. 98: 12890-12895. Vogt, W. P. (1999). Dictionary of statistics & methodology. Thousand Oaks, CA: SAGE Publications, Inc. Wahaj, S. A., Guse, K. R. & Holekamp, K. E. (2001). Reconciliation in the spotted hyena (Crocuta crocuta). Ethology. 107: 1057-1074. Wallis, J. (1982). Sexual behaviour of captive chimpanzees (Pan troglodytes): pregnant versus cycling females. American Journal of Primatology. 3: 77-88. Walters, J. (1980). Interventions and the development of dominance relationships in female baboons. Folia Primatologica. 34: 61-89. Walters, J. R. & Seyfarth, R. M. (1987). Conflict and cooperation. In: Primate societies (Ed. by B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham & T. T. Struhsacker), pp. 306-317. Chicago: University of Chicago Press. Watts, D. (1994). Agonistic relationship between female mountain gorillas (Gorilla gorilla beringei). Behavioral Ecology and Sociobiology. 34: 347-358. Watts, D. (1995a). Post-conflict social events in wild mountain gorillas (Mammalia, Hominoidea). I. Social Interactions between opponents. Ethology. 100: 139-157. Watts, D. (1995b). Post-conflict social events in wild mountain gorillas. II. Redirection, side direction and consolation. Ethology. 100: 158-174. Watts, D. P. & Pusey, A. E. (1993). Behavior of juvenile and adolescent great apes. In: Juvenile Primates (Ed. by M. E. Pereira & L. A. Fairbanks), pp. 148-167. New York: Oxford University Press. Watts, D. P. (1998). Coalitionary mate guarding by male chimpanzees at Ngogo, Kibale National Park, Uganda. Behavioral Ecology and Sociobiology. 44: 43-55. Watts, D. P. (2000). Grooming between male chimpanzees at Ngogo, Kibale National Park. I. Partner number and diversity and grooming reciprocity. International Journal of Primatology. 21: 189-210. Watts, D. P., Colmenares, F. & Arnold, K. (2000). Redirection, consolation, and male policing. In: Natural conflict cesolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 281-301. Berkeley: University of California Press. Watts, D. P. (2002). Reciprocity and interchange in the social relationships of wild chimpanzees. Behaviour. 139: 343-370. Watts, D. P. & Mitani, J. C. (2002). Hunting and meat sharing by chimpanzees at Ngogo, Kibale National Park, Uganda. In: Behavioural diversity in chimpanzees and
146 11: Bibliography
bonobos (Ed. by C. Boesch, G. Hohmann & L. F. Marchant), pp. 244-255. Cambridge: Cambridge University Press. Weingrill, T., Lycett, J. E. & Henzi, S. P. (2000). Consortship and mating success in Chacma baboons (Papio cynocephalus ursinus). Ethology. 106: 1033-1044. Wells, K. D. (1977). The social behaviour of anuran amphibians. Animal Behaviour. 25: 666-693. Westlund, K., Ljungberg, T., Borefeld, U. & Abrahamson, C. (2000). Post-conflict affiliation in Common Marmosets (Callithrix jacchus jacchus). American Journal of Primatology. 52: 31-46. Whitten, P. L. & Smith, E. O. (1984). Patterns of wounding in stumptail macaques (Macaca arctoides). Primates. 25: 326-336. Wittig, R. (1997). Änderungen der Verhaltensprofile bei Schimpansen einer Gehegegruppe während der Integration zweier fremder Artgenossen. Mitteilungsblatt der Ethologischen Gesellschft e.V. 39: 54. Wittig, R. M. (2003). Die anderen Vettern aus dem Wald - 25 Jahre Taï Schimpansen Projekt, Côte d'Ivoire, Westafrika. Phyllodrom Journal. 2: 49-53. Wittig, R. M. & Boesch, C. (2003a). Food competition and linear dominance hierarchy among female chimpanzees of the Taï National Park. International Journal of Primatology. 24: 847-867. Wittig, R. M. & Boesch, C. (2003b). 'Decision-making' in conflicts of wild chimpanzees: An extension of the Relational Model. Behavioral Ecology and Sociobiology. 54: 491- 504. Wittig, R. M. & Boesch, C. (2003c). Linear dominance hierarchy due to contest competition among female chimpanzees (Pan troglodytes verus). Folia Primatologica. 74: 179-230. Wittig, R. M. & Boesch, C. (2003d). The choice of post-conflict interactions in wild chimpanzees (Pan troglodytes). Behaviour. 140: 1527-1559. Wittig, R. M. & Boesch, C. (in review). How to repair relationships in wild chimpanzees (Pan troglodytes). Ethology. Wrangham, R. W. (1980). An ecological model of female-bonded primate groups. Behaviour. 75: 262-300. Wrangham, R. W. (1987). Evolution of social structure. In: Primate Societies (Ed. by B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wrangham & T. T. Struhsaker), pp. 282-296. Chicago: University of Chicago Press.
147 11: Bibliography
Wrangham, R. W., Conklin, N. L., Chapman, C. A. & Hunt, K. D. (1991). The significance of fibrous foods fore Kibale Forest chimpanzees. Philosophical Transactions of the Royal Society of London, B. 334: 171-178. Wrangham, R. W., Clark, A. P. & Isabirye-Basuta, G. (1992). Female social relationships and social organisation of Kibale Forest chimpanzees. In: Topics in primatology: Human origins (Ed. by T. Nishida, W. C. McGrew, P. Marler, M. Pickford & F. B. M. de Waal), pp. 81-98. Tokyo: University of Tokyo Press. Wrangham, R. (2002). The cost of sexual attraction: is there a trade-off in female Pan between sex appeal and received coercion? In: Behavioural Diversity in chimpanzees and bonobos (Ed. by C. Boesch, G. Hohmann & L. F. Marchant), pp. 204-215. Cambridge: Cambridge University Press. Yamakoshi, G. (1998). Dietary responses to fruit scarcity of wild chimpanzees at Bossou, Guinea: Possible implications for ecological importance of tool use. American Journal of Physical Anthropology. 106: 283-295. Yarn, D. H. (2000). Law, love and reconciliation: Searching for conflict resolution in Homo sapiens. In: Natural conflict resolution (Ed. by F. Aureli & F. B. M. de Waal), pp. 54-70. Berkeley: University of California Press. York, A. D. & Rowell, T. E. (1988). Reconciliation following aggression in patas monkeys, Erythrocebus patas. Animal Behaviour. 36: 502-509.
148 Appendices
Appendix A. Studies on reconciliation. Overview of within- and between-species variation of three variables of reconciliation in primates and other mammals.
Type of Variation of Variation of Species Behaviour Latency InitiatorT Prosimii Eulemur fulvus 1 ?yesA Plathyrrhini Callithrix jacchus 3 gr, pl yes ? Cebus apella 4 co, gr, pl ? ? Cebus capucinus 5 hb, mo yes A>V Saimiri sciureus 6 ?yes? Catarrhini Cercocebus torquatus 7 co, em, gr ? A Hominidae Gorilla gorilla 32 em, to yes ? Pan paniscus 33 gc ? A>V* Pan troglodytes 34,35,36 co, gc, ki yes A=V, A other Mammalia Carpa hircus 38 ki ? A>V Crocuta crocuta 39 gc$, to yes A Note Appendix A: behaviour typically used in reconciliation, ap: appologies, br: body rubbing, co: contact (e.g. sitting, laying), em: embrace, gc: genital contacts, gr: grooming, gv: grunt vocalisation, hb: hold-bottom, hh: hold-hand, ki: kiss or mouth- mouth contact, ls: lip-smack, mo: mounting, pl: play, so: share or offer object, to: touch. variable delay of reconciliation observed within the same study (yes / no). T rough proportion of reconciliation initiated by aggressor (A) or victim of aggression (V), A 150 12: Appendices Appendix B. Observation file of November 1, 1997. The focal animal was female Castor. The first column (TIME) contains the starting time of the behaviour in seconds after the observation start. The second column (ACT) displays the name of the actor/initiator of the behaviour. The middle column (BEH) shows the behaviour. The forth column (REA) contains the name of the reactor/receiver of the behaviour, while the last column (MOD) shows sometimes modifications of the behaviour. On the far right side, explanations of the behaviours are given sometimes, referring to codes in the table by matching colours. TIME ACT BEH REA MOD 1 CAS DEP ARB 36 CAS DEP SOL 205 CAS MAN L33 NNN 285 CAS DEP ARB 314 CAS MAN L33 NNN 357 CAS REP 425 CAS DEP ARB 454 CAS REP 586 CAS MAN T33 K02 925 CAS MAN T33 K03 1048 CAS DEP SOL 1085 MYS HAV N11 K03 1086 CAS APP MYS FIX 1179 CAS REP 1324 CAS DEP SOL 1457 CAS MAN F33 K05 1530 CAS DEP SOL 1848 CAS REP 1878 CAS DEP SOL 1888 CAS DEP ARB 1922 MAC VOC ALL DSE 1928 NAR VOC ALL PHO 1935 ? VOC ALL PHO 1940 CAS DEP ARB 1957 NIN VOC ALL TAM 1966 CAS DEP ARB 2054 MAC VOC ALL DSE 2064 FEM VOC ALL PHO 2072 CAS MAN F33 NNN 2180 PER VOC ALL TAM 2188 CAS MAN F33 NNN 2502 CAS DEP ARB 2528 CAS DEP SOL 2553 CAS DEP SOL 2774 CAS DEP ARB 2825 CAS REP 2924 CAS DEP ARB 2940 END HAV ? ? 3335 CAS DEP ARB 3362 CAS DEP SOL 3455 CAS REP 3511 CAS DEP SOL 3581 CAS REP 3589 CAS DEP SOL 3659 CAS REP 151 12: Appendices 3706 CAS DEP SOL 4061 CAS REP 4082 CAS DEP SOL 4234 CAS REP 4410 CAS DEP SOL 4445 CAS REP 4453 CAS DEP SOL 4479 CAS REP 4508 CAS DEP SOL 4688 CAS REP 4752 CAS DEP SOL 4875 CAS DEP ARB 4971 END HAV ? ? 6740 CAS DEP ARB 6750 CAS WAI CAC 6782 CAS DEP SOL 6895 CAS REP 6962 CAS DEP SOL 7013 CAS MAN T33 K05 7102 CAS DEP SOL 7299 CAS REP 7330 CAS DEP ARB 7461 CAS MAN F33 K10 7727 CAS MAN F33 K14 7818 VEN SUP CAS FSI 7830 CAS SCR VEN 7835 CAS TBA VEN FSI 7839 CAS MAN F33 K14 7890 VEN SUP CAS FSI 7895 CAS CHA VEN RSC 7911 VEN FLI CAS RSC 7913 CAS MAN F33 K14 7955 VEN APP CAS FIX 7960 CAS SBR VEN SCR 7968 VEN LEA CAS ? 7970 CAS MAN F33 K14 Castor, being in 8550 CAS DEP ARB oestrous, leaves 8658 CAS REP 8695 CAS DEP SOL Macho 8809 CAS LEA MAC OES 8810 MAC ATT CAS Conflict 8822 CAS CRO MAC SCR Macho - Castor 8823 MAC LEA CAS 8824 CAS REP 8915 CAS DEP SOL 9100 CAS REP Castor solicits 9274 CAS GRO CAC DIR consolation of 9318 END GRO CAC ? Cacao and 9322 CAS REP receives it 9454 CAS DEP SOL 9502 FOS HAV I11 K03 9503 CAS APP FOS FIX 9609 CAS DEP SOL 10331 CAS DEP ARB 10387 CAS REP 10510 CAS DEP ARB 10548 CAS DEP SOL 10748 CAS REP 10797 CAS DEP SOL 10929 CAS REP 152 12: Appendices 10941 CAS APP MAC PGR 10944 CAS KIS MAC FRE 10947 CAS LEA MAC ? 10948 CAS DEP SOL 11058 CAS REP 11080 CAS DEP SOL 11265 MAC APP CAS ? 11266 CAS PGR MAC 11268 MAC LEA CAS ? 11269 CAS REP 11445 CAS DEP ARB 11485 CAS REP 11521 CAS MAN F33 K06 12483 END HAV ? ? 15082 CAS DEP ARB 15136 NIN APP CAS ? 15137 NIN LEA CAS ? 15138 CAS DEP SOL 15175 CAS REP 15244 NIN WAI CAS 15249 CAS DEP SOL 15295 CAS MAN L11 K02 15326 CAS DEP SOL 15482 CAS MAN F33 NNN 15582 CAS DEP SOL 15764 CAS REP 15814 CAS DEP SOL 15944 CAS REP 15956 CAS DEP SOL 16124 CAS REP 16148 CAS DEP SOL 16268 CAS DEP ARB 16283 CAS REP 16337 CAS DEP ARB 16348 CAS MAN T33 K04 16396 CAS DEP SOL 16675 CAS REP 16689 CAS MAN L33 NNN 16735 CAS DEP SOL 16857 CAS REP 16917 NIN APP CAS ROC 16919 CAS DEP SOL 17011 CAS REP 17029 CAS CPR NIN OES Castor has a genital 17043 NIN GRO CAS DIR swelling and 17097 NIN GRO END ? Nino grooms her 17106 CAS MAN L33 NNN 17145 CAS REP 17176 CAS DEP SOL 17251 CAS REP 17455 CAS MAN I11 K02 17557 CAS DEP SOL 17586 CAS DEP ARB 17691 CAS MAN F33 K04 17971 CAS MAN F33 K07 18037 CAS MAN F33 K08 18159 END HAV ? ? 20613 CAS DEP ARB 20736 CAS REP 20910 CAS GRO CAC DIR 153 12: Appendices 21131 END GRO CAC ? 21135 CAS DEP SOL 21169 CAS REP 21250 CAS DEP SOL 21464 CAS WAI CAC 21482 CAS DEP SOL 21652 CAS REP 21665 CAS DEP SOL 21827 CAS REP 21871 CAS DEP SOL 22099 CAS MAN L33 NNN 22130 CAS DEP SOL 22319 CAS MAN N33 K08 22358 CAC SCR FED 22359 CAS SPP CAC LOS 22360 CAS CHR FOS 22363 CAS ATT FOS 22366 FOS FLI CAS SCR 22368 CAS REP 22503 CAS CON CAC 22844 CAS REP 22960 CAS DEP SOL 23038 CAS REP 23107 CAS DEP SOL 23211 CAS REP 23260 MAC APP CAS ? 23261 CAS PGR MAC 23262 MAC LEA CAS ? 23272 MAC HAV N11 K03 23273 CAS APP MAC HRE 23287 CAS REP 23368 MAC HAV N11 K03 23369 CAS MEN MAC Conflict context: 23375 MAC SCO CAS OES sex 23382 CAS REP 23413 MAC BSW CAS 23418 CAS CRO MAC SCR Conflict between 23419 MAC ROC CAS ? Macho and Castor 23422 CAS CRO MAC SCR 23430 CAS APP MAC ? 23431 CAS PRS MAC OES Reconciliation 23432 MAC COP CAS K02 23446 CAS REP between Macho 23580 MAC HAV N11 K03 and Castor 23588 CAS MEN MAC 23666 CAS REP 23721 CAS JEU CAC 23750 CAS REP 23766 CAS DEP SOL 24278 CAS REP 24300 CAS DEP SOL 24618 CAS REP 24653 CAS DEP SOL 24672 CAS DEP ARB 24720 CAS REP 24809 CAS DEP ARB 24901 CAS MAN F33 K02 26126 END HAV ? ? 27207 CAS DEP ARB 27258 CAS DEP SOL 154 12: Appendices 27378 CAS DEP ARB 27389 CAS REP 27482 CAS DEP ARB Focal animal 27612 END HAV ? ? 29209 CAS DEP ARB lost: 29252 CAS REP no observation 29273 CAS DEP SOL 29422 CAS REP 29447 CAS DEP SOL 29522 CAS REP 29570 CAS DEP SOL 29648 CAS REP 29755 CAS DEP SOL 29956 CAS DEP ARB 30068 CAS MAN L33 K02 31031 CAS DEP ARB 31068 CAS DEP SOL 31112 CAS REP 31226 CAS DEP SOL 31349 CAS REP 31424 CAS DEP SOL 31567 CAS REP 31577 CAS DEP SOL 31598 CAS DEP SOL 31651 CAS REP 31848 CAS DEP ARB 31906 END HAV ? ? 32444 CAS DEP ARB 32498 CAS DEP SOL 32596 CAS REP 32611 CAS DEP SOL 32768 CAS REP 32843 CAS DEP SOL 33317 CAS MAN T33 K03 33985 CAS DEP SOL 34138 CAS MAN I11 NNN 34204 CAS DEP SOL 34505 CAS REP 34565 CAS APP MAC ? 34566 CAS VOC MAC GRU 34570 CAS KIS MAC FRE 34572 CAS TOU NIN ? 34573 NIN LEA CAS ? 34574 CAS DEP SOL 34655 SIR APP CAS PAN 34656 CAS FIM SIR ? 34659 SIR TGE CAS 34665 SIR LEA CAS ? 34666 CAS DEP SOL 34724 CAS APP SIR ? 34725 CAS KIS SIR FRE 34729 SIR VOC CAS PAN 34730 CAS DEP SOL Castor was eating fruits 34999 CAS DEP ARB (non-monopolisable) at 35139 CAS MAN F33 K10 37173 CAS REP the same time with10 37255 CAS DEP ARB competitors 37280 CAS DEP SOL 37387 LOU HAV F33 K12 37388 CAS APP LOU FIX 155 12: Appendices 37490 CAS REP 37549 CAS DEP SOL 37592 CAS JEU CAC 37604 CAS DEP SOL 37674 CAS REP 37697 CAS DEP SOL 37769 CAS REP 37902 CAS JEU CAC 37909 CAS DEP SOL 37984 CAS REP 38049 CAS DEP SOL 38070 CAS MAN L33 NNN 38128 CAS DEP ARB 38133 MAC VOC CAS PAN 38140 CAS REP 38671 ? NID End of observation 38843 CAS NID 39001 {end} after 39001 s 156 12: Appendices Appendix C. Dyadic association index (DAI) of males and adult females in the North community of Taï chimpanzees. DAIs for the period before the death of Brutus (10.1996 – 2.1997) are shown above the diagonal in italic numbers, while DAIs without Brutus are shown under the diagonal. Individuals are separated by sex (males in italic letters) and in alphabetical order. DAI Brutus Macho Marius Nino Belle Castor Dilly Fossey Goma Loukoum Mystère Narcisse Perla Ricci Venus Brutus 0.35 0.42 0.25 0.14 0.13 0.17 0.11 0.14 0.17 0.19 0.15 0.19 0.25 0.18 Macho 0.48 0.35 0.13 0.15 0.13 0.11 0.14 0.32 0.19 0.15 0.15 0.24 0.17 Marius 0.68 0.31 0.11 0.13 0.15 0.12 0.13 0.16 0.19 0.14 0.16 0.27 0.25 Nino 0.45 0.41 0.15 0.18 0.15 0.15 0.18 0.25 0.19 0.16 0.17 0.38 0.29 Belle 0.34 0.29 0.33 0.23 0.21 0.22 0.19 0.11 0.26 0.19 0.24 0.22 0.15 Castor 0.24 0.20 0.25 0.24 0.30 0.52 0.34 0.16 0.41 0.17 0.29 0.18 0.16 Dilly 0.18 0.18 0.21 0.23 0.31 0.31 0.47 0.11 0.30 0.16 0.29 0.18 0.22 Fossey 0.19 0.18 0.23 0.21 0.46 0.27 0.35 0.11 0.36 0.15 0.25 0.19 0.17 Goma 0.18 0.17 0.22 0.22 0.32 0.48 0.35 0.15 0.36 0.21 0.29 0.17 0.28 Loukoum 0.32 0.27 0.30 0.25 0.25 0.25 0.23 0.25 0.16 0.13 0.14 0.15 0.23 Mystère 0.24 0.23 0.28 0.26 0.39 0.35 0.30 0.37 0.26 0.19 0.31 0.23 0.17 Narcisse 0.20 0.18 0.24 0.25 0.29 0.27 0.25 0.31 0.24 0.31 0.21 0.21 0.15 Perla 0.24 0.22 0.24 0.26 0.31 0.38 0.27 0.36 0.29 0.36 0.30 0.21 0.16 Ricci 0.27 0.28 0.41 0.29 0.24 0.21 0.25 0.21 0.25 0.26 0.28 0.23 0.19 Venus 0.24 0.26 0.28 0.23 0.28 0.30 0.27 0.31 0.38 0.30 0.29 0.35 0.25 157 12: Appendices Appendix D. Peace making in Taï chimpanzees. I did this observations and wrote it down in terms of a story by reason of the retirement of my old professor Prof. Dr. Dr. Hubert Hendrichs from University of Bielefeld, Germany. „Frieden stiften“ bei Schimpansen Im November 1996 befanden sich drei erwachsene Männer in der von mir beobachteten Schimpansengesellschaft: Macho, der -Mann, der alte Brutus, an der - Position, und der junge Herausforderer Marius. Marius hatte schon die letzten Wochen versucht Machos Regeln immer wieder zu durchbrechen. Brutus, hatte sich in dieser Zeit meistens von den beiden fern gehalten. Jetzt waren sie wieder alle zusammen. Die Spannung zwischen Macho und Marius explodierte und führte zu einer minutenlangen Auseinandersetzung, die Macho letztendlich gewann. Marius war die daraus resultierende Anspannung anzumerken. Er schlich immer wieder um Macho herum und versuchte zaghaft Kontakt zu ihm aufzunehmen, aber Macho blieb resistent. Er groomte Brutus und wies Marius Annäherungsversuche zurück. Brutus wurde sichtlich nervöser. So ging es über eine halbe Stunde. Schließlich stand Brutus auf, mittlerweile wurde er nicht mehr von Macho gegroomt sondern lag neben ihm, und begann mit Macho zu spielen. Macho faßte Brutus Knöchel und Brutus lief los. Sie bildeten eine Formation mit Brutus an der Spitze und Macho folgend, immer noch mit beiden Händen die Knöchel von Brutus umfassen. Sie spielten „Eisenbahn“. Brutus führte Macho immer wieder um einen einzelnen Baum herum – minutenlang. Plötzlich veränderte Brutus die Richtung und bezog einen zweiten Baum in den Parcours mit ein, nämlich den Baum, neben dem Marius saß und den beiden zuschaute. Brutus schritt nun eine Acht um die beiden Bäume herum, immer noch mit Macho im Schlepptau. In der dritten Runde griff Brutus nach Marius Knöcheln, als sie bei ihm vorbei kamen. Nun hatte sich eine „Eisenbahn“ aus drei Schimpansenmännern gebildet, die mit „Lachen“ dem achter Parcours um die beiden Bäume folgte. Angeführt von Marius, der von Brutus an den Knöcheln gehalten wurde, der wiederum von Macho an den Knöcheln gehalten wurde, liefen sie Minute für Minute im Kreis. Doch auf einmal zog sich Brutus aus dem Spiel heraus. „Automatisch“ nahm Macho die Knöchel von Marius und die beiden setzten das Spiel alleine fort. Brutus saß am Rand des Parcours und sah den beiden zu, wie sie ihre „Versöhnung“ begingen, die von ihm initiiert worden war. Es sollte die einzige Situation bleiben, in der Brutus mir gezeigt hat, wie Schimpansen Frieden stiften. Er starb im März 1997, vermutlich an Altersschwäche. 158 Curriculum vitae Roman Martin Wittig Born 23/07/1968 in Karlsruhe / Germany Scientific education 1989 – 1996 Studies in biology at the University of Bielefeld 1996 Diploma degree in biology Topic of Diploma thesis: “Changes of behavioural patterns during the integration of strangers in a group of captive chimpanzees”. Supervision by Prof. Dr. Dr. Hubert Hendrichs. 1996 – 1999 Observations of wild chimpanzees in the Taï National Park, Côte d’Ivoire. Scientific member and project manager (for 1.5 years) of the “Projet chimpanzé Taï”. Project leader: Prof. Dr. Christophe Boesch. since 1999 PhD thesis at the Max-Planck Instuitute for Evolutionary Anthropology in Leipzig, under the supervision of Prof. Dr. Christophe Boesch. Leipzig, September 25, 2003 Roman Wittig Declaration of independence Herewith I declare that I have conceived and written this dissertation without any inadmissible help and/or material that has not been explicitly indicated. All sources of information that were used have been indicated. This disseratation has not been submitted elsewhere, neither inside nor outside this country. I have not previously attempted to complete this or any other PhD thesis. Leipzig, September 25, 2003 Roman Wittig 160