Preferences Under Pressure

Eric Skoog

Conflict, Threat Cues and Willingness to Compromise Dissertation presented at Uppsala University to be publicly examined in Zootissalen, EBC, Villavägen 9, Uppsala, Friday, 13 March 2020 at 10:15 for the degree of Doctor of Philosophy. The examination will be conducted in English. Faculty examiner: Associate Professor Thomas Zeitzoff (American University, School of Public Affairs).

Abstract Skoog, E. 2020. Preferences Under Pressure. Conflict, Threat Cues and Willingness to Compromise. Report / Department of Peace and Conflict Research 121. 66 pp. Uppsala: Department of Peace and Conflict Research. ISBN 978-91-506-2805-0.

Understanding how preferences are formed is a key question in the social sciences. The ability of agents to interact with each other is a prerequisite for well-functioning societies. Nevertheless, the process whereby the preferences of agents in conflict are formed have often been black boxed, and the literature on the effects of armed conflict on individuals reveals a great variation in terms of outcomes. Sometimes, individuals are willing to cooperate and interact even with former enemies, while sometimes, we see outright refusal to cooperate or interact at all. In this dissertation, I look at the role of threat in driving some of these divergent results. Armed conflict is rife with physical threats to life, limb and property, and there has been much research pointing to the impact of threat on preferences, attitudes and behavior. Research in the field of evolutionary psychology has revealed that threat is not a singular category, but a nuanced phenomenon, where different types of threat may lead to different responses. I argue that by taking a more nuanced approach to threat, drawing on theories from the field of evolutionary psychology, some of the variance in outcomes can be explained. In particular, some commonly observed features of protracted conflicts, such as seemingly indivisible issues and parochialism may be moderated by threat. In the four essays, I address this from both a theoretical and empirical point of view. In Essay I, I illustrate in a formal bargaining setting how threats can lead actors to prefer risky all-or-nothing gambles to division schemes, preventing bargaining solutions to be found. In the second essay (Essay II), I show that willingness to make compromises with members of other groups are not contingent on group affiliation alone, but rather on the expected reciprocity of that group. Furthermore, characteristics of others beyond group can also affect pro-sociality. Based on a threat management perspective, me and my coauthors in Essay III show that non-threatening social categories, such as women or the elderly, are shown higher levels of altruism, also when there individuals are outgroup members. Exposure to violence can even increase altruism across group lines, but only to these non-threatening groups. Finally, in Essay IV, I show that those who experience post-traumatic growth as a result of traumatic events reverse the standard loss aversion people generally display, even showing gain-seeking preferences. Together, these results point to the importance of bringing in a more nuanced conceptualization of the role of threat in the study of peace and conflict. Even in the most brutal and destructive conflicts, humans are able to cooperate, also across group lines. However, for this cooperation to function, managing the threats in conflict is of central importance.

Keywords: Evolutionary psychology, Conflict, Threat perception, Social preferences

Eric Skoog, Department of Peace and Conflict Research, Box 514, Uppsala University, SE-75120 Uppsala, Sweden.

ISSN 0566-8808 ISBN 978-91-506-2805-0 urn:nbn:se:uu:diva-403234 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-403234)

To those who came before me

List of Essays

This thesis is based on the following essays, which are referred to in the text by their Roman numerals.

I Skoog, Eric (2020). Divide or Conquer? Deep Rationality and Seemingly Indivisible Issues as a Cause of Conflict. Unpublished manuscript. An earlier version was presented at the 58th Annual Convention of the International Studies Association, Baltimore MD, USA, 22-25 February 2017. II Skoog, Eric (2020). Indirect reciprocity and tradeoff paradigms in the wake of violent intergroup conflict. Unpublished manuscript. An earlier version was presented at the 59th Annual Con- vention of the International Studies Association, San Francisco, CA, USA, 4-7 April 2018. III Hall, Jonathan, Eric Skoog and Dogukan Cansin Karakus (2020). Kindness in the aftermath of cruelty? The effect of exposure to war-time trauma on altruism across social categories. Un- published manuscript. An earlier version was presented at the 51th Peace Science Society International North American Meet- ing, Tempe, AZ, USA, 2-4 November 2017. IV Skoog, Eric (2020). The Vagaries of Valuation. How post-traumatic growth impacts psychological responses to gains and losses. Unpublished manuscript. An earlier version was presented at the 60th Annual Convention of the International Studies Association, Toronto, ON, Canada, 27-30 March 2019.

Contents

Introduction ...... 13

Definitions of the central concepts ...... 16 Preferences, pro-sociality and compromise ...... 17 Threat, threat perception, and threat cues ...... 18 Intergroup armed conflict: ...... 20

Previous literature ...... 21 Conflict and its effects on society ...... 21 Individual level effects of conflict exposure...... 23 Parochialism in conflict ...... 25 The role of threat in parochial attitudes ...... 27

Theoretical framework ...... 30

Methodology and empirical strategy...... 38

Presenting the essays...... 45 Essay I...... 46 Essay II ...... 47 Essay III ...... 47 Essay IV ...... 48

Conclusion ...... 49 Avenues for future research ...... 51 Policy implications ...... 52 References ...... 54

Acknowledgements

This dissertation would scarcely amount to a thing were it not for my super- visors, Magnus Öberg and Hanne Fjelde, who have been there as support, sounding boards – and indeed disciplinarians when needed to – throughout the entire process. Whatever issue, whether milestone or minor inconvenience, you have been excellent mentors and guides into the mystical land that is the academic world. Magnus – your advise when I first started to always read broadly and not get stuck in an intellectual silo has in no small part contributed to the direction this dissertation has taken. Your calm, down-to-earth and prag- matic approach has been invaluable for a mildly manic-depressive character such as myself, while still encouraging me to address the Big Questions, and seek out new and untrodden paths of scientific inquiry. Hanne – no matter which tangential (at best) or outlandish (at worst) line of reasoning I have me- andered down, your comments have always been clear-headed, to-the-point, impeccably structured and relevant. Knowing all too well my own tendency to sometimes go tightrope-walking over the chasm of chaos, I am greatly in- debted to your ability to offer staunch support for my many stunts, yet always being able to pull me back up and set me on the right path as soon as my stride got too wobbly. Jonathan Hall has in many ways been my ”unofficial third supervisor”, since he first invited me into the project that would also form the basis for the empirical part of my dissertation, back when I had just started my PhD. Through our multiple collaborations, I have been able to work on many and varying topics and issues, and through successes and struggles (remember Heroku?), your enthusiasm, energy and support has been truly inspiring. The dissertation has also benefitted immensely from the persistent scrutiny, ques- tioning and encouragement of the members of the Monday group (lately on Wednesdays): Håvard Hegre, Nina von Uexkull, Nynke Salverda, Colin Walch, Niklas Karlén, Karin Johansson, Sophia Hatz, Ida Rudolfsen, Stefan Döring, David Randahl, Mihai Croicu, Kristina Petrova, Gudlaug Olafsdottir and (briefly) Annkatrin Tritschoks. On a more personal note, it has been a pleasure and a privilege to have been part of the DPCR PhD community over the last five years. While I am hardly known as an introverted or reserved person, it may nevertheless come as a

surprise to some that I have always been better at making friends, than keeping them. This (I hope) is not primarily due to some inherent flaw in my personality, but just a natural result of constantly moving between contexts and places, rarely staying for more than a few years in any one. Therefore, had someone asked me before I joined the department five years ago, I would likely have assumed that my stay here would be no different. However, unlike the predictions of my gloomy 2015 alter ego, I would count many of those I have met here among my closest friends. In particular, I want to salute the original 4th floor crew (A.K.A. the fab 5, or class of ‘15), Anni Tritschoks, Charlotte Grech-Madin, Kyungmee Kim and Stefan Döring. Stefan shared an apartment with me in Ann Arbor in 2016; a traumatic enough experience, I’m sure, even without me wandering off at random times for my “little walks”. That he remains my friend nevertheless says plenty about his character. – I still owe you a steak! I shared an office with Kyungmee when I first started my PhD, and she was thus the first person I got to know at the department. Coming as a complete stranger to Uppsala and the department, it was very nice to share some of that initial confusion with someone! Charlotte, through having the office next door, has often been forced to bear the brunt of my singing, talking out loud to myself, and throwing curses at my computer. That she, despite this, is always endlessly helpful, caring and kind is a true testament to her magnanimity. Anni, in a saint-like manner, has put up with me and humored me when I time and again barge into her office, sometimes to rant, sometimes to just shoot the breeze, even when she clearly has better things to do. Her contagious optimism, generous provision of deli- cious baked goods, (and love for the “Merry Season”) has made my years at the department all the more enjoyable. Honorary 4th floorer Susanne Schaf- tenaar should of course not be left out of this illustrious group. Whether going out for a sumptuous burger lunch, or just bantering in the hallway or staff room, our conversations have always been pleasant breaks from the daily chores of PhD life. While our collective journey is quickly coming to an end, and we one by one cross over to the “other side”, wherever that may lead us, I am very grateful for your company and support for the last five years, and look forward to future shenanigans! Of course, many others, too numerous to mention, all deserve my thanks and gratitude, but I will nevertheless shout out to Niklas, Florian, Karin, Sebastian, Mihai, Anne, Gulla and Kristina! Even with the sometimes arduous journey from Stockholm, you have all made it a joy to come to the office! Outside of the department, I want to extend my thanks to Magnus Johan- nesson, Örjan Sjöberg and Erik Lindqvist who inspired me to start the academic journey that eventually led me here, while I was still at the Stockholm School of Economics (SSE). There are many others, who in various ways have contributed greatly to making this dissertation come together. Dogukan Cansin Karakus, who coordinated the data collection rounds in very challenging environments, and without whom the empirical side of this dissertation

would never have been possible. Elisabeth Larsson at MC², who helped me with setting up the mathematical modelling in MATLAB, thus saving me many hours of cursing at my screen. Many others, who have commented on my work, inspired and encouraged me in the research journey over the years, deserve my thanks. In particular Christopher Butler, James Morrow, Douglas Kenrick, Steven Neuberg and Scott Atran. I would also like to mention Patrik Olsson, with whom I wrote my Master’s thesis back at the SSE, which in many ways proved the starting point for this journey. I like to believe that our lengthy discussions, never swerving from any topic, and turning for-granted taken truths on their head, has made me a better researcher, and a better thinker in general. Furthermore, I am very grateful to Anna Maria Lundins Stipend- iefond, for funding my travels to conferences and courses in various locations across the Atlantic Ocean. Lastly, I bow most humbly before the esteemed “Father of History”, Herodotos of Halicarnassus, for being an inspiration in what true curiosity, open-mindedness, and the quest for knowledge should look like – still after almost 2500 years, the Historiai is, in my opinion, the most enjoyable piece of academic writing ever put on paper (parchment or papyrus). Finally, I want to thank my parents, Barbro Skoog Eriksson and Bo Eriks- son for everything, and my grandparents, Maj and Carl Axel Skoog, and Signe and Sven Eriksson, for everything. I want to thank Sweden for paying me for five years to basically be curious, I want to thank all of human history for being a constant inspiration and source of wonder, and I want to thank God that I have finally finished this dissertation!

Eric Skoog Uppsala, January 2020

Introduction

Understanding how preferences are formed is a key question in the social sciences. Agents’ preferences decide what actions they will take, how they evaluate the outcomes of these actions, and how the social contexts in which they operate are factored into these decisions. The importance of preferences holds the same weight on the level of interpersonal and intergroup interactions as on the macro level of interactions between states and empires. The ability of agents to associate with each other, interact in economic tradeoffs, and make compromises over contested issues is a prerequisite for efficient market allo- cations, political processes, intergroup interactions, and the interpersonal relations central to well-functioning societies. Indeed, the willingness of individuals, groups and nations to interact with one another and find means of mutually acceptable exchanges is a central tenet of human cooperation at any level. This has particularly profound implications in the context of conflict, and in conflict-afflicted societies. For instance, failure to find compromises over contested issues may lead to escalatory contentious spirals, ultimately resulting in violence and war (Pruitt and Kim 2004). Furthermore, in the aftermath of armed conflict, the ability of former enemies to cooperate and have functional social, political and economic interactions, even when they have divergences of interest, is necessary for reconstruction, reconciliation and avoiding the recurrence of violence. It is evident that the preferences actors have will greatly impact what decision they chose to make in these situations, yet much remains to be learned about how conflict impacts preferences in the first place. Since Fearon (1995), the discussion on bargaining in conflict has relied heavily on rational choice theory. This approach rests on the assumptions of neoclassical economics (e.g. von Neumann and Morgenstern 1944). Under these assumptions, rational utility maximizers with stable preferences interact to solve coordination problems. In classic bargaining models, failure to reach solutions result from factors such as imperfect information (e.g. Fearon 1995; Skaperdas 2006) and commitment problems (Fearon 1995; R. Powell 2006)1. Preferences and preference formation, on the other hand, have often been considered either fixed or black-box. However, preferences are central in deter- mining the bargaining space, and changes in the preferences of actors could

1 Interestingly, even within more cognitive approaches, the focus has often been primarily on information processing (Stein 2013).

13 substantially alter the possible bargaining space and thereby influence whether parties are able to find a mutually acceptable solution to a conflict of interest, or if they prefer to engage in contentious behaviors (Pruitt and Kim 2004). Likewise, preference changes that impact social interactions in post-conflict settings will in turn substantially impact actors’ ability to build the social capital necessary to reconstruct war-torn countries and turn them into functional societies (Knack and Keefer 1997; Knack and Zak 2003). This reconstruction hinges on actors being willing to interact, trust, and make tradeoffs with one another. Hence, investigating preference formation in conflict settings is of relevance from a bargaining perspective, as well as a peacebuilding perspective. A growing literature within psychology, economics, sociology, and political science has started to investigate how circumstances matter for preference formation. Importantly, this literature has pointed out that preferences are not fixed, but are functions of context and motivations (Gintis et al. 2005; Kenrick et al. 2009; Voors et al. 2012). If this is the case, then it is of great relevance to explore the underlying mechanisms of preference formation and change (Druckman and Lupia 2000). In this dissertation, I will address one such mechanism, which I argue is very important in conflict settings, namely threat and threat perceptions. Threat is a central component of any conflict, and the psychology of threat has been studied thoroughly, indicating that it can affect attitudes and behaviors, such as willingness to compromise and the ability to see mutually beneficial solutions to conflicts of interest (Canetti-Nisim, Ariely, and Halperin 2008; Canetti et al. 2015; Maoz and McCauley 2005). In the four papers of this dissertation, I look at the role played by threat in shaping social preferences, that is preferences that take into consideration social interactions and relationships, in a conflict setting. I argue that by taking a more nuanced approach to threat, drawing on theories from the field of evolutionary psychology, we can gain a better understanding of how social preferences are formed. Based on this approach, I investigate how factors relating to managing affordances, i.e. the costs and benefits of associating with others (Brown, Neel, and Sherman 2015; Neuberg, Kenrick, and Schaller 2011), impact preferences relevant for the willingness to interact and engage in exchanges with others. Associating with others may provide potential benefits such as cooperative opportunities or trade-offs, but also costs such as the risk of free-riding or hostility. From this perspective, threats are essentially signals of potential costs. Thus, the presence of a threat can change the cost-benefit analyses of various outcomes, thereby in practice altering preferences over them. The psychological research on threat has developed substantially in recent years (Neuberg, Kenrick, and Schaller 2011; Neuberg and Schaller 2016; Stroessner et al. 2015). Recent developments within the field of evolutionary psychology have strongly impacted how psychologists think of social interactions regarding conflict and cooperation, including threats (Barkow 2006;

14 Cosmides and Tooby 2013; Tooby and Cosmides 2010). However, these developments have, with a few notable exceptions (see for instance Hall et al. 2018; Landau-Wells 2018; McDermott and Davenport 2017) remained on the fringe of peace and conflict research. In this dissertation, I argue that a more nuanced approach to threat, drawing upon the theories of threat management and fundamental motivations from evolutionary psychology, may explain some of the variability in the willingness to associate and interact with others that we observe in conflict situations, through its effect on social preference formation. In particular, while I argue that the role of threat in shaping preferences has broad implications for the study of conflict, within this dissertation I will focus on preferences relevant for a number of social outcomes that are particularly important in the conflict setting, namely willingness to compromise and pro-sociality. Willingness to compromise here means the ability to find mutually acceptable solutions to divergences of interest. Pro-sociality denotes the willingness to cooperate with, and benefit, others, even when this is personally costly. These are social preferences relevant for economic interactions, i.e. interactions that concern the allocation of goods and values. Inability to engage in successful economic interactions with others will have profound implications, not only for the individuals concerned, but also for society as a whole. In societies where social trust is low and market functions are not op- erating properly, impediments to economic efficiency will result in reduced growth, lower standards of living, and also in an increased probability of the occurrence (or recurrence) of social strife and conflict (Braithwaite, Dasandi, and Hudson 2016; Collier, Hoeffler, and Söderbom 2008). This dissertation makes a theoretical contribution by bringing the nuanced conceptualization of threat from evolutionary psychology, and insights from behavioural economics, into the field of peace and conflict research. It also makes empirical contributions by studying the role of these more nuanced threat categories in actual intergroup conflict settings. The empirical parts of the dissertation rely on data from a number of unique large-N survey experiments, carried out in populations of refugees from ongoing2 conflicts in the Middle East. By having large samples of individuals hailing from these conflict settings, the empirical material addresses some of the issues brought up by the recent replication crisis in the social sciences (Camerer et al. 2018). By taking the theories developed in the lab into the field, in a non-WEIRD3 setting (Henrich, Heine, and Norenzayan 2010a, 2010b) where the threat of conflict is actually present in the environment, this dissertation provides a unique investigation of the actual effects of threat on preferences. This makes the find-

2 The latest UCDP data (Pettersson, Högbladh, and Öberg 2019) reports active conflicts in both Syria and Iraq in 2018, and the violence was ongoing in 2019 (UCDP Bulletin 2019). 3 That is a setting that is Western, Educated, Industrialized, Rich and Democratic, as is common in many economic and psychological experiments.

15 ings more valuable both from a scientific point of view through increased external validity, as well as from a policy perspective with relevance for the reconciliation and reconstruction processes in conflict-affected societies. A more in-depth discussion of the field context and the samples is provided later in the methods section. In the four essays of this dissertation, I address the effects of threat on social and economic preferences at the group and individual level, employing a variety of approaches. Essays I through to III address willingness to compromise and pro-sociality as outcomes directly. Essay IV does not look directly at these outcomes, but rather at a preference which will have a big impact on willingness to compromise and pro-sociality, namely valuation. In Essay I, I present a theoretical argument for the role of threat in causing bargaining failure, ar- guing that certain threats make agents act as though issues are indivisible. Es- says II and III investigate how the types of threats posed by others impact the willingness to interact and make compromises with them. Using unique data from a survey experiment among refugees in Turkey, Essay II looks at the role of previous actions as signals for future reciprocity when making tradeoffs. Using similar data, Essay III looks at how characteristics of others which may be cues of threat affects pro-sociality, and how these effects are moderated by exposure to violence. Finally, using data from another survey experiment with refugees in Turkey, Essay IV looks at how the pro-social threat and trauma coping mechanism of post-traumatic growth (PTG) (Tedeschi and Calhoun 2004) impacts the valuation of resources. This introduction proceeds as follows. First, I will outline some of the central concepts that I use in this dissertation, in particular social preferences, threat, and the conflict context. I then provide a review of, and situate the dissertation within, existing literature of peace and conflict research and related fields. The subsequent sections give a brief background on theory and methodology, including research ethics, while the final section concludes the dissertation, pointing to the policy relevance of the findings and future avenues of research.

Definitions of the central concepts The main concepts of interest in this dissertation are preferences and threat perceptions. It is generally difficult to directly observe an individual’s preferences. Instead, what we can observe are behaviors, attitudes and outcomes indicative of a preference. Thus social outcomes, behaviors and attitudes are indicative of an individual’s underlying preference. Through actions, the individual shows his or her revealed preference (Samuelson 1938, 1948), which gives us information about the underlying utility function. Thus, while all the essays of this dissertation are concerned, in one way or another, with the role of conflict-induced threat on the formation of social preferences, the observed

16 outcomes and operationalization vary between essays. The definitions here are kept at a general level, while the specific details of each operationalization will be found in the respective essay. Furthermore, as the studies all relate to the context of conflict, I will elaborate briefly on the particular role of conflict in the dissertation.

Preferences, pro-sociality and compromise The concept of preferences is by its very nature broad, and there are multiple existing definitions, as well as similar and overlapping concepts (interests, tastes, values, etc.) that are often used as alternatives, or even interchangeably. Druckman and Lupia (2000, 2) define a preference as “a comparative evaluation (i.e. a ranking over) a set of objects”. That is, a preference entails a valuation of a set of objects of interest, in relation to the valuation of other sets of objects (including the empty set). This neither stipulates where these preferences come from, nor identifies the nature of the objects of interest and the factors of comparison. In this dissertation, I will primarily look at preferences in the economic sense, distinguishable from other types of preferences (such as political preferences, religious preferences etc.) in that they generally concern more tangible values. However, most of the theoretical rationale is appli- cable to any type of preferences, and I will thus use this more general term. The nature of preferences, and their role in utility calculations, has been thoroughly formalized and described by consumer choice theory, based on the neoclassical assumptions (see for instance Jehle and Reny 2001)4. However, in this dissertation, I particularly consider preferences related to social outcomes, that is, social preferences. The term social preferences comes from the field of behavioral economics, and refers to preferences that diverge from the standard selfish and strictly materialistic, utility maximizing preferences of Homo Economicus, in that they take into consideration social contexts and relations (Camerer and Loewenstein 2004). They are thus social preferences in the sense that they are dependent upon social relations to other individuals and groups. People not only care about their own outcomes, but also about the outcomes of other actors (Loewenstein, Thompson, and Bazerman 1989). This does not necessarily mean that such preferences are normatively good. Some are often perceived as such, for instance altruism and trustworthiness. However, phenomena like spite and envy are also social preferences, in that they are formed in relation

4 The main assumptions are that preferences are complete (i.e. the actor can compare and evaluate any pair of object sets), transitive (i.e. the actor can consistently rank any finite number of object sets) and continuous (i.e. that the preference sets are closed). In addition, assumptions of strict monotonicity (i.e. that more valuable object sets are preferred to less valuable sets) and finally convexity (i.e. that indifference curves are convex) are often added, although the first three are generally sufficient to formulate a continuous utility function (Jehle and Reny 2001).

17 to the outcomes of another actor. Thus, a social preference has the same characteristics as a standard preference, but also includes evaluation of the social context and other individuals’ outcomes in addition to their own. While the dependent variables vary between the essays of this dissertation, they are generally operationalizations that pertain to two outcomes of particular interest for the study of social preferences in conflict situations, namely pro-sociality and willingness to compromise. The term pro-sociality here relates to the types of social preferences that favor positive interactions with others, i.e. the antithesis of anti-social preferences. Acting pro-socially is essentially behaving in a non-selfish way, such as acting in an altruistic, trusting and trustworthy manner and contributing to public goods (e.g. civil society engagement). That is, actions that benefit others while being personally costly to the performer (Gintis 2003). For example, in Essay III the outcome variable is the Welfare Trade-off Ratio (WTR), which is a measure of how individuals trade off their own welfare against other individuals’ welfare. By willingness to compromise, I here refer broadly to the ability to make decisions in social interactions with others about the allocation of scarce resources5. In some cases, compromises have been defined more narrowly as a solution to a divergence of interest, where parties concede to some middle ground solution, in order to differentiate it from what is often called integrative solutions, which seeks to integrate the interests of the parties rather than make a split (Pruitt and Kim 2004). While this is a relevant distinction, I will here use the more broad definition of compromises as tradeoffs over contested resources that are mutually acceptable to the concerned parties. For example, in Essay II the outcome is acceptance of a trade-off over a contested resource. These concepts are closely linked to one another, as the willingness to compromise with others can be contingent on pro-sociality (such as willingness to associate with others), but also on other less pro-social (or even anti-social) preferences such as spite, as well as valuations and risk preferences over the issue(s) at stake. Thus, even though the essays have different outcome variables, they all relate to each other, and the overarching question of how threats affect social preferences in a conflict setting.

Threat, threat perception, and threat cues Threat is another broad concept. In general terms, a threat in this dissertation relates to someone or something which is likely to be a source of danger (physical or symbolic) to an actor, or the group to which it belongs. Threat perceptions are the beliefs of an actor that someone or something poses a threat (real or imagined). The perception of threat, of course, does not necessarily corre-

5 In some of the Essays, I also use the term trade-off to refer to the same concept.

18 spond to factual conditions, and an entity perceived as threatening could potentially in reality pose no threat at all (and vice versa). However, as decision making is generally based on the perceptions and beliefs of the decision maker, rather than the potentially unobservable state of the factual situation, it is the threat perceptions that are of interest in studying the effects of threat on behavior. While threat perceptions are argued to be important factors in the formation of attitudes and behaviors, there is still much we do not know, in particular about how more nuanced categories of threats impact preferences in conflict situations (Canetti-Nisim, Ariely, and Halperin 2008). In the essays of the dissertation, I will mainly focus on a more specific contribution of introducing a threat management approach (Neuberg, Kenrick, and Schaller 2011; Neuberg and Schaller 2016; Schaller and Neuberg 2012) to the effect of threat on social preferences in conflict. However, there is a substantial literature on threat in other related fields, for which the results of the dissertation may also be relevant, and which I will briefly touch upon in the section on previous literature. The conceptualization of threat used in this dissertation draws heavily upon research from psychology, and in particular, evolutionary psychology (Cottrell and Neuberg 2005; Neuberg, Kenrick, and Schaller 2011; Neuberg and Schaller 2016). According to this perspective, threat perceptions and the responses to them will vary depending on the type, source and context of the threat. There are of course innumerable factors in our environments that may be the source of danger. However, for the purposes of delimiting the scope of this dissertation, I will focus on the types of threats that are most likely to occur in social situations such as conflicts, i.e. threats posed by other humans (Duntley 2015). The more detailed nuances of the conceptualizations of threat arising from this literature will be developed further in the later theory section. In brief, the fundamental takeaway from this literature is that threat perceptions, and the reactions thereto (including preference formations) are functional. That is, they are aimed at solving reoccurring adaptive problems. As a result, the human psychology will have a number of heuristics, or rules of thumb, which shape preferences and behaviors differently, depending on what type of adaptive problem they are trying to solve (Kenrick, Griskevicius, et al. 2010; Kenrick, Neuberg, et al. 2010). It is worth noting that while threat is the main proposed mechanism, what I am actually measuring in the essays is not the threat perception per se, but rather the cues that are theoretically predicted to elicit threat perceptions and motivations of self-protection according to evolutionary theory. This is partly because threat perceptions can be difficult to measure in a survey in the field (but see Cottrell and Neuberg 2005 for a useful methodology), and also because the theoretical predictions do not necessarily hinge on there being actual physical threats present in the current situation. Rather, they depend on situations and cues that in ancestral environments would have been likely to pose

19 threats, and which as a result would elicit the adaptive responses and motivations.

Intergroup armed conflict: While threat would reasonably impact preferences under any circumstances, the essays in this dissertation focus on threats that are associated with conflict. In armed conflict, there are many different types of threats present, but perhaps the most important is the threat of violent action by others. Conflict has been defined as a social situation where two parties have a perceived divergence of interest (Pruitt and Kim 2004), where the social aspect comes from the fact that conflict involves two parties that interact with each other. My arguments generally extend to all such social situations, but I will primarily look at armed intergroup conflict. As such, the armed intergroup conflict setting is in essence a scope condition for all the essays of the dissertation (especially Essays II and III). Since group affiliation, and the threats associated with ingroups and outgroups, is one of the main factors I investigate in this dissertation, the intergroup dynamic is an important component. Of course, on some level, basically all conflicts except strictly interpersonal conflicts are intergroup. However, I here focus on conflict where the group identity (whether ethnic, religious or ideological) is a central component of the incompatibility itself. This is important since, as we will see in the literature review, much of the previous literature has focused on group identity as a central factor driving threat (Canetti-Nisim, Ariely, and Halperin 2008; Schaub 2017; Tajfel and Turner 2001). If, as I hypothesize, other threat-related factors may actually moderate the effects of group identity, then the intergroup armed conflict context is a relevant scope condition for testing these hypotheses. The empirical material for this dissertation was collected among refugees that had fled the conflicts in Syria and Iraq. These are both intrastate armed conflicts of an internationalized nature (Pettersson, Högbladh, and Öberg 2019). While the governments of various countries are involved, they are not interstate conflicts between governments, but the main cleavages are rather based on group identities. In these cases, the identities are often ethic and religious in nature, making the group dynamic important for both these conflicts. As such, they are relevant cases for studying the role of threat in intergroup conflict. Furthermore, since preference changes can impact the ability to compromise and find mutually acceptable exchanges, conflict is also to some degree an outcome. This is primarily investigated in the first essay of the dissertation. That said, while this focus further emphasizes the relevance of the research in general, I do not explicitly look at this in any of the empirical essays.

20 Previous literature While the four essays focus on threats and preferences underlying the willingness to compromise, this dissertation speaks to a broader literature, both concerning the onset of conflict through its effect on the available bargaining space, but also concerning the effects of violence and threat on societies in general, and individual attitudes and preferences in particular. While the findings of this dissertation are of importance for both bodies of literature, the literature review will primarily focus on the latter, as this is where the main empirical contributions of the dissertation lie. However, as noted above, these factors – individual preferences and bargained outcomes – are related and will impact one another. I will here briefly outline the general literature on the societal effects of conflict, before zooming into the more specific literature on the effects of intergroup conflict on attitudes and behavior, and particularly, social preferences. I will focus primarily on large-N field studies (surveys and experiments), but will also discuss some lab and standard observational studies where relevant.

Conflict and its effects on society Armed conflict has many detrimental effects on societies (Blattman and Miguel 2010; Gates et al. 2012; Grosjean 2014). They stymies development (Blattman and Annan 2010; Gates et al. 2012) and international trade (Anderton and Carter 2001; Bayer and Rupert 2004), negatively affects public health and well-being (Ghobarah, Huth, and Russett 2003; Iqbal 2010) and reduces social trust while increasing polarization (Wood 2008) and crime (Deglow 2016) in the societies that it ravages. That said, this literature has also revealed that, somewhat counterintuitively, not all the effects of armed conflict are negative. Under some circumstances, the aftermath of civil war may lead to more normatively positive outcomes as well. As an example, studies have shown that there is a tendency towards fast economic recovery once conflicts end (Chen, Loayza, and Reynal-Querol 2008; Collier 1999). In principle, even though conflict itself damages the economy, many societies experience a high rate of economic growth in the post-conflict period, sometimes bringing them back to their pre-war “steady state”.6

6 One oft-proposed explanation for this rapid economic growth, observed for instance in postwar West Germany and Japan, is that while the war destroyed productive capital, it also destroyed much of the existing rigidity and red tape imposed by pre-war institutions and interest groups. Thereby, wars may remove many of the distortions in the market place hindering growth (Blomqvist and Lundahl 2002; M. Olson 1983). This gives these post-war economies an advantage over those economies burdened by the agendas of interest groups, allowing them to out-perform the latter in the short run. Likewise, the destructiveness of conflict may change the ratios of the production factors in the economy, increasing relative productivity in the less exposed factors which drive increased growth (Barro and Sala-i-Martin 1995; Chen, Loayza, and Reynal-Querol 2008).

21 That a complex social phenomenon such as warfare can have divergent effects is not puzzling in and of itself. However, identifying the circumstances under which these less negative effects materialize is a much more challenging question. In an attempt to address the causes of the contradictory and divergent effects of war, and the more nuanced effects of armed conflict on social outcomes, a growing literature has turned from the macro-level to the individual level, looking at the effects of wartime violence on individuals (Blattman and Miguel 2010)7. War does not only affect societies as a whole; individuals who have experienced war also suffer its consequences, both psychologically and materially. That conflict affects social behavior has been well established for a long time. The classic “Robber’s Cave” experiment (Sherif et al. 2010) showed with some clarity that the mere presence of other groups, but in particular the existence of groups with competing goals, can cause behavioral shifts both on an individual and group level. These behavioral and attitudinal shifts are important to study, as they may substantially impact people’s ability to interact or find mutually acceptable solutions to divergences of interest, which is also important for the reconstruction, reconciliation and avoiding the recurrence of violence after conflicts have ended. Psychological responses to conflict may hinder conflict resolution and peacemaking efforts through a number of varying mechanisms, such as values, beliefs and ideological convictions, as well as through emotional and motivational states (Bar-Tal and Halperin 2011). In their model, Bar-Tal and Halperin (2011) integrate these factors to show how contextual, emotional and cognitive factors come together to create barriers to successful conflict resolution efforts. In addition, they point out that threatening contexts have been suggested to make these barriers to conflict resolution more rigid. That such factors do indeed reduce openness to compromises, thereby precluding mutually acceptable solutions, has also been empirically supported in a study in Israel (Halperin and Bar-tal 2015). As a result, there is a vast literature discussing the impact of war on individual attitudes and behavior, far too expansive to be thoroughly covered here. I will thus primarily focus on those studies that deal with economic and political outcomes that are relevant from a social preference point of view. Meth- odologically, these studies have often directly imported methods and outcomes from behavioral economics such as trust games (Bauer, Fiala, and Levely 2017; Cassar, Grosjean, and Whitt 2013; Gilligan, Pasquale, and Samii 2014; Gneezy and Fessler 2012), public goods games (Gilligan, Pasquale, and Samii 2014; Hopfensitz and Miquel-Florensa 2015; Mironova and Whitt

7 This development has gone hand-in hand with similar developments within the field of economics, where for instance Gintis (2009) has argued for the inclusion of findings from other behavioral sciences in the study of economics and strategic decision making.

22 2016a) and dictator games (Gilligan, Pasquale, and Samii 2014; Mironova and Whitt 2016a; Whitt and Wilson 2007).8

Individual level effects of conflict exposure Individuals’ attitudes and behaviors are evidently affected by the experiences of conflict. However, also on the micro-level, there are contradictory findings on how conflict actually impacts people. Some of the research in this vein has painted a somewhat more optimistic picture than expected, suggesting that experience of conflict may lead to an increase in pro-sociality, civil society engagement and trust as a result of conflict (Bauer, Fiala, and Levely 2017; Gneezy and Fessler 2012; Voors et al. 2012). For instance, in an experimental study with heads of households in rural Burundi, a country afflicted by ethnic conflict, Voors et al. (2012) found that individuals with greater exposure to wartime violence displayed more altruistic tendencies towards neighbors. Similarly, Bauer, Fiala and Levely (2017) found that individuals who had been forcefully recruited by the Lord’s Resistance Army (LRA) in Uganda were more trustworthy and more likely to engage themselves in civil society. Due to the LRA’s indiscriminate forceful recruitment, individuals did not self-se- lect into the experience of soldiering, making assignment of this treatment “as- if random”. The results thus suggest that soldiering, and especially soldiering at an early age, had a causal and positive effect on trustworthiness and civil society engagement. Bellows and Miguel (2009) likewise found a positive effect of wartime exposure to violence on participation in civil society organizations in Sierra Leone. These findings are in line with the idea of “altruism born of suffering” (Staub and Vollhardt 2008; Vollhardt and Staub 2011), which suggests that the experience of trauma makes individuals more empathic to others who have been through similar events. As a result of this empathetic reaction, individuals are argued to become more altruistic and open towards others in the same situation, even including members of other groups. Warfare may also help strengthen social cohesion and trust, which may also help post-conflict reconstruction by building social capital. Exposure to wartime violence was found to be positively related with pro-social attitudes such as trust and altruism in Nepal (Gilligan, Pasquale, and Samii 2014), willingness to host refugees in Liberia (Hartman and Morse 2018), and participation in religious activities

8 Some of the political effects, such as community engagement and civil society participation, are in a sense also pro-social (it is the real-world equivalent of contributing in a public goods game) and hence such outcomes will also be discussed where relevant. For instance, there are discussions on effects of wartime violence, such as tendencies towards increased religiosity (Henrich et al. 2019) and voting in democratic elections (Blattman 2009), which both can be seen as pro-social tendencies, indicative of changes in social preferences. However, as these are somewhat beside the main theoretical argument of this dissertation, I will generally not delve deeper into that literature.

23 and groups in Uganda, Sierra Leone and Tajikistan (Henrich et al. 2019). This is in line with findings from psychology pointing to not only negative, but also positive effects of trauma and distress, where individuals grow as a result of their life experiences (S. Powell et al. 2003; Tedeschi and Calhoun 2004). All of the aforementioned tendencies seem to indicate the accumulation of social capital; the resource necessary for functional societies (Knack and Keefer 1997; Knack and Zak 2003). On the other hand, further studies have pointed out that under most conditions, violence has a detrimental rather than positive impact (Cassar, Grosjean, and Whitt 2013; De Juan and Pierskalla 2016; Kijewski and Freitag 2018). In a study with over 17,000 observations from Central and Eastern Europe, it was found that when controlling for ob- jective measures of wartime victimization during World War II, the effects of reported victimization on pro-social attitudes such as political participation, civil society engagement and trust were negative (Child and Nikolova 2018). Individual war exposure was also found to be negatively related to social trust in a study in Kosovo (Kijewski and Freitag 2018), suggesting that the positive effects of warfare on pro-social attitudes are highly conditional. A similar pattern was observed in another study by Grosjean (2014) using a large nationally representative survey from 35 countries. Exposure to conflict-related violence was associated with a reduction in the perceived legitimacy and effectiveness of state institutions and political trust across all nations. However, the study also observed an increase in participation in civil society associations among exposed individuals. While this might seem in line with the more positive strands of literature, Grosjean (2014) observes that although conflict exposure did increase the ability for local collective action, this may not indicate that conflict fosters social capital, as there was a simultaneous decrease in trust. Instead, the interpretation is that conflict exposure makes people associate more closely with their respective groups, whether ethnic, geographic, religious or otherwise. A similar pattern with a simultaneous increase in ingroup social capital, and decrease in general societal trust, was found in a study of conflict-exposed populations in rural Burundi (Voors and Bulte 2014). After conflict, the social capital that is built is of the “bonding” type, with people retreating into the identity groups that they feel can protect them. Meanwhile, the “bridging” social capital is eroded, with people losing faith in society and other people in general. For instance, individuals victimized by crime in Latin America displayed lower levels of trust in institutions, but this did not necessarily, or at least not as strongly, translate into decreased trust in others as friends or business partners (Corbacho, Philipp, and Ruiz-vega 2015). This is also in line with the findings by De Luca and Verpoorten (2015). They noted that following conflict in Uganda, in areas where conflict had been particularly intense, there seemed to be an increase in certain social activities, such as engaging in political discussion and attending community meetings. However, this did not translate into for instance increased voter turnout. The authors argue that this may in part be because while violence makes people engage

24 with their local community to solve pressing issues, they at the same time lose faith in the social institutions. While exposure to violence creates bonds between individuals in the local environment, it burns bridges at a broader societal level.

Parochialism in conflict In particular, the increased pro-sociality, trust and societal engagement observed in the literature outlined above seems to mainly be directed at members of one’s own group, something often termed parochial altruism (Bauer et al. 2014, 2016; Rusch 2014). That is, when conflicts are fought along group lines, the increased pro-sociality is not of a general nature, but focused on one’s own ethnic, social, religious or political group. The concept of parochial altruism stems from evolutionary theories regarding the co-evolution of group conflict and ingroup altruism and pro-sociality (Bowles 2008, 2009; Choi and Bowles 2007; Gintis et al. 2008; Rusch 2014). According to this theory, ingroup fa- voritism and ingroup bias evolved as a response to competition between groups of humans in the ancestral environment. In this environment, multiple bands of humans would have competed for scarce resources in the form of water, foraging patches, and game for hunting. As individual survival would have been very contingent on the performance of the group or band, so the argument goes, members of more efficient groups would have substantial fit- ness9 benefits (Bowles 2008, 2009; Choi and Bowles 2007). Cooperative groups that could better utilize the synergies and economies of scale of working in concert outperformed less cooperative groups, leading to a selection pressure on traits favorable for cooperating with ingroup members. As a result, humans have a tendency towards favoring ingroup members, which is trig- gered by intergroup conflict, just as observed in the Robber’s Cave example (Sherif et al. 2010). While there may be other underlying mechanisms that drive parochial altruism in addition to this (Rand and Nowak 2013), the co- evolutionary (multi-level selection) argument has been among the most prev- alent in the literature. These parochial tendencies have received ample support from studies both in the lab and field (Bauer et al. 2014, 2016; Bernhard, Fischbacher, and Fehr 2006; Fehr and Fischbacher 2003; Romano et al. 2017). Individuals are more likely to be altruistic towards ingroup members (Bernhard, Fischbacher, and Fehr 2006; Whitt 2007), trust them more (de Dreu 2010; Fershtman and Gneezy 2001), and share information with them (Lyall, Shiraito, and Imai 2015). The most important component of this parochialism is the favorable treatment and primacy of the ingroup (de Dreu 2010; de Dreu, Aaldering, and Saygi 2014). While this type of positive ingroup bias does not necessarily lead

9 I here use the word ”fitness” in the Darwinian sense, i.e. the ability of a gene to survive and reproduce.

25 to outgroup hostility (Brewer 1999), empirical evidence suggests that implicit in this tendency towards “groupishness” is also a potential for discriminatory or hostile behavior towards outgroups. As an example, in an experimental study in eastern Europe, it was found that hostile attitudes and behaviors towards an ethnic outgroup (the Roma) were much more contagious than hostile behavior towards co-ethnics (Bauer et al. 2018)10. The tendency of outgroup hostility can be particularly exacerbated by experiences of conflict, as follows from the co-evolutionary argument of altruism and war. For instance, in a study among youths competing in a football tournament in Sierra Leone, those more exposed to violence were not only more generous towards ingroup members in a classic dictator game. They also exhibited much stronger preferences for competition against outgroup members, and were more likely to accrue foul cards in the football matches (Cecchi, Leuveld, and Voors 2016). In another study, it was shown that when northern Sudanese individuals were exposed to violent riots enacted by southerners, this increased support for southern secession, but also made northerners more negative to southerners remaining as citizens in the north (Beber, Roessler, and Scacco 2014). That is, exposure to violence from the outgroup made group distinctions more important, and increased preferences for separation and avoidance. Intensity of conflict can also decrease social trust and increase ethnic fractionalization (Rohner, Thoenig, and Zilibotti 2013). Hence, parochialism may become a self-ful- filling prophecy, whereby conflict leads to distrust and hostility, which leads to less functional social relations and unwillingness to find mutually acceptable solutions to divergences of interest, which in turn can lead to escalation or resurgence of conflict in a vicious cycle. This suggests that under some circumstances, violence leads to much more negative outcomes than under others (Grosjean 2014), and the effects can even vary depending on what measures we use (Child and Nikolova 2018). As Grosjean (2014) points out, it seems as though war, and in particular civil war, is more likely to build bonding social capital, than bridging social capital. When this occurs on the nation level, it can increase national social cohesion, which may have contributed to the development of strong and functional nation states (e.g. Tilly 1992). However, at the same time, this tendency is of a strongly parochial nature, essentially focused on strengthening the ingroup at the outgroup’s expense. For instance, civil society organizations, it is claimed, played an important role in the growth of the Nazi Party (NSDAP) in Germany in the interwar period (Satyanath, Voigtländer, and Voth 2017). Thus, in that case the overtly pro-social activity of civil society engagement proved to build bonding social capital and increase investment in a clearly parochial endeavor (the Nazi party). After interstate conflict ends, the negative effects on social

10 However, as suggested by the study by Whitt and Wilson (2007), the negative attitudes towards outgroups following war sometimes might revert towards normal levels as time goes by, and norms of fairness may persist even in the face of parochialism (Whitt 2014).

26 trust and legitimacy of state institutions are not present in the same way as after intrastate conflict. This is because in interstate conflict the ingroup/outgroup distinction clearly follows national boundaries. Therefore, the nation becomes the relevant ingroup with which individuals identify. In intrastate conflict, on the other hand, the ingroup is the relevant ethnic, religious or political group with which the individual identifies, and towards which parochial altruism will be directed (Grosjean 2014). While substantial support for the parochial argument in this corpus of research is growing, there are still many gaps in our knowledge (Bauer et al. 2016; Blattman and Miguel 2010). Under what circumstances the pro-social effects of war are able to cross group boundaries remains an open question (Bauer et al. 2016). In much of the literature, group identities are assumed fixed, with conflict resulting in individuals grouping together with ingroups to protect themselves against threatening outgroups. Essentially, the conceptualizations of group are drawn from social identity theory (Tajfel and Turner 2001), where intergroup hostility is driven by the group identities per se, with individuals basing their self-esteem on the value of the group identity. When group lines are vague and group distinctions are unclear, as exem- plified by the conflict in Tajikistan, parochial effects are less evident. In the Tajik conflict, the group dimensions relevant for the conflict crosscut villages and ethnicities, making the relevant ingroup and outgroup boundaries less evident. As a result of this, pro-social effects of exposure to violence are less evident in the Tajik setting (Cassar, Grosjean, and Whitt 2013, 2014). Simi- larly, exposure to wartime violence in Bosnia in general leads to more retrib- utive justice preferences, however more interdependence with perpetrators leads to more restorative justice preferences (Hall et al. 2018). Hence, it seems as though the parochial tendencies do not neatly follow pre-determined group lines, but rather, that the parochial pro-sociality follows from threats posed. When everyone is potentially threatening, there is no pro-sociality, but when there is interdependence and potential coalition building, pro-sociality may extend across group lines, even towards members of “enemy” groups.

The role of threat in parochial attitudes The parochialism argument rests on an assumption that intergroup hostility is the result of intergroup threat. Armed conflict is rife with physical threats to life, limb and property, and there has been a lot of research pointing to the impact of threat on preferences, attitudes and behavior. Going back to Al- temeyer (1988), threat, or as it was then termed “belief in a dangerous world”, has been connected to exclusionary attitudes, authoritarianism and fascism. This connection between threat and authoritarian and group-based values has received further scrutiny, implying that authoritarian values are an essentially functional response to specific types of threats (Feldman and Stenner 1997;

27 Stenner 2005). When there are threats to one’s group, society and values, feel- ings of sameness and unity will help people get together and fight off the threat to their group. As a result, individuals who feel threatened may support policies that focus on more authoritarian values or strengthening the ingroup (Getmansky and Zeitzoff 2014). This, of course, is exactly the rationale ech- oed by the coevolution of parochial altruism and war argument (Bowles 2008, 2009; Choi and Bowles 2007). Beyond authoritarianism, however, threat has been shown to have many more nuanced effects on individual attitudes and preferences. A literature more specifically addressing threats and preferences has developed, drawing to a large extent upon data from surveys following the 9/11 terrorist attacks in New York City in 2001, but also further field experimental research. These studies have revealed that emotional responses to threats, such as the threat of terrorism, alter the evaluations of risk, as well as preferences over public policy (Lerner et al. 2003). In particular, different emotional responses, such as fear or anger, can lead to divergent preferences and behaviors (Lerner and Keltner 2001; Lerner, Small, and Loewenstein 2004; Zeitzoff 2018). Like- wise, in an experiment among Israelis exposed to Palestinian rocket attacks, Zeitzoff (2014) shows how anger aroused by threats in intergroup conflict leads to an increased willingness to engage in negative reciprocity and to punish non-cooperative ingroup members. However, this effect was conditional on whether the individuals were heavily exposed to the threat or not in reality. Among individuals who had experienced the threat of rocket attacks, anger instead led to more ingroup cohesion, and a reduced willingness to punish. Meanwhile, among those who did not experience the threat of rocket attacks, anger lead to higher degrees of punishment. This, Zeitzoff argues, is because the effect of the anger emotion is different under these two circumstances. Whereas individuals in the highly exposed area responded to anger with a focus on ingroup cohesion and a strengthening of local bonding capital (something also shown by a stronger reaction to non-cooperative play by the other individual in the experiment), the individuals in the non-exposed area only responded with a reduction of bridging capital. It has also been argued that the threat posed by territorial civil wars results in more severe social intolerance between groups (Tir and Singh 2015). While not directly testing the threat perceptions, the authors show that the effect is strongest among individuals who are more likely to be sensitive to threat cues (in this case operation- alized as individuals on the political right, holding more authoritarian values, cf. Altemeyer (1988)). This all points to threat and the responses thereto being a central factor in forming social preferences. As a result, what type of threat is present in a situation may impact willingness to interact with others due to its impact on social preferences. That the presence and legacy of intergroup threats can lead to a rejection of potentially mutually acceptable compromises has been shown by several studies (Canetti- Nisim et al. 2009; Canetti-Nisim, Ariely, and Halperin 2008). Much of this

28 research regarding threat has been based on the standard models of intergroup threat going back to realistic conflict theory (Sherif et al. 2010) and social identity theory (Tajfel and Turner 2001), with some more recent developments from integrated threat theory (Stephan and Stephan 2000). However, as evidenced by the varying effects described above, the role of threat seems to be more nuanced and conditional than simply relating to ingroup/outgroup distinctions. Depending on how the groups are distinguished (Cassar, Grosjean, and Whitt 2014) or whether the threat is direct or not (Zeitzoff 2014), the resulting responses may differ greatly. Perhaps, then, the variation in outcomes that we observe is a result of variations in the type and perceptions of threats present in the conflict context. I argue that in order to better understand under what conditions the parochial tendencies resulting from armed conflict are likely to appear or not, thereby helping or hindering the ability to interact and find mutually acceptable exchanges, we need to use a more nuanced conceptualization of threat. By doing so, it may be possible to disentangle some of the diverging effects observed, as suggested by for instance Zeitzoff (2014). Thereby, we may gain a better understanding of the conditions under which individuals may be more or less prone to suffer those “barriers to conflict resolution” suggested by the literature (Bar-Tal and Halperin 2011). The concept of threat has received a lot of attention within the field of social psychology, and it has been shown that threat is not a singular category, but rather a nuanced phenomenon where different types of threat may lead to very different responses (Cottrell and Neuberg 2005; Neuberg and DeScioli 2015; Neuberg and Schaller 2016). Much of this new understanding of threat has developed out of the field of Evolutionary Psychology. In this dissertation, I draw upon these theories to show how a more nuanced conceptualization of threat based on evolutionary theories can help our understanding of when, and with whom, we cooperate and compromise in conflict situations. In the following section, I will outline my theoretical argument for why these more nuanced threat concepts may matter for preferences, and the willingness to interact with others.

29 Theoretical framework

Figure 1. Outline of the dissertation. The observed independent variables (observable cues), dependent variables (social preferences) and the proposed mechanism (threat perception). Social preferences are of central importance for willingness to accept compromises and reach bargained outcomes, as well as for pro-sociality and costly cooperation.

Threat perceptions are the main theoretical mechanism in the dissertation. However, I do not directly observe the threat perceptions in any of the empirical essays. Rather, the main independent variables in the empirical tests are cues that would elicit threat perception, such as previous actions signaling in- tent to hurt (Essay II), or individual characteristics that may be cues of such intentions (Essay III), as well as mechanisms for coping with threat such as Post Traumatic Growth (Tedeschi and Calhoun 2004) (Essay IV). Figure 1 illustrates the proposed relationship between observable cues (i.e. characteristics of others, characteristics of the environment etc.), threat perceptions, social preferences, and their impact on social outcomes. While the theoretical arguments and operationalizations vary between the four essays of this dissertation, they all draw upon some common theoretical underpinnings regarding

30 the nature and function of threat. These theories hark from the field known as Evolutionary Psychology. One central difference of an evolutionary approach from a rationalist approach is that whereas the agent in a rationalist model seeks to maximize utility, agents in evolutionary models seek to maximize fitness, which denotes the ability of a gene to survive and reproduce and thus make its way into subsequent generations. An adaptive fitness maximizing behavior is thus a behavior causing the genes driving it to increase in the population over generations (Tooby and Cosmides 2015), whether carried there by the individual display- ing the behavior, or by their genetic kin (Hamilton 1964). Natural selection will therefore have given our human psychology adaptations that regulate behaviors, emotions and responses in the way that would best serve to maximize fitness, given the circumstances (Tooby and Cosmides 2008)11. While one component of fitness is of course reproductive efforts, this is far from the only, or even the most important component thereof. For a slow life history strategy animal such as humans (Kaplan and Gangestad 2005), survival is at least as important a factor as reproduction. As such, humans should have many adaptions for dealing with the recurrent problems relating to the successful survival and reproduction of themselves and their genetic kin that would have been encountered in ancestral environments12. This includes navigating the complex and challenging social world of an eusocial species such as Homo Sapi- ens (Wilson 1975, 2012). As the social animals that we are, a non-negligible share of these adaptations would therefore be directed at managing the affordances (Gibson 1979; Neuberg, Kenrick, and Schaller 2011), i.e. the opportunities and threats – benefits and costs – posed by other humans. The affordance management perspective suggests that there are potential benefits as well as costs of associating with others (Brown, Neel, and Sherman 2015; Gibson 1979; Neuberg, Kenrick, and Schaller 2011). For instance, associating with others may provide potential fitness benefits such as being coalition partners and allies against hostile groups, exploiting economies of scale in accruing resources, or as being potential mating partners. However, there may also be costs associated with engaging with others. An adaptive affordance management system would need to take both into account, making a cost-benefit analysis of interacting with other individuals. In this dissertation, I look primarily at the cost side of this equation, and the management of threats (Neuberg, Kenrick, and Schaller 2011; Neuberg and Schaller 2016).

11 An example of this is what is referred to as internal regulatory variables (Tooby and Cosmides 2008). That is, something that adjusts emotions, preferences or attitudes of an individual in a way that would serve to maximize fitness. Anger is one clear example of this (Sell 2011). 12 This environment is often known as the Environment of Evolutionary Adaptation (EEA), i.e. the environment of the Pleistocene era in which humans and their ancestors lived for the ma- jority of their evolutionary history (and which some hunter-gatherer societies, to all intents and purposes, still inhabit today).

31 Evolutionary theories of threat perception Over the evolutionary history of the human species, other conspecifics have posed many different types of dangers. They may be competitors for scarce resources (such as food, social status, and mates) (Duntley 2005), they may carry dangerous and unknown pathogens (Schaller 2011), and they may also be the perpetrators of various cost-inducing hostile actions such as violence (Pinker 2011), rape (Thornhill and Palmer 2000) and homicide (Daly and Wilson 1988). As a result, we have a lot of psychological adaptations for dealing with the threat of danger posed by other humans (Duntley 2015). For these systems to be adaptive, they would need to be attuned to the different types of threats other individuals might pose. This is the fundamental realization of threat management theory (Neuberg, Kenrick, and Schaller 2011; Neuberg and Schaller 2016), i.e. focusing on the cost side of affordance management. The implications of such a perspective are that different threat types are inferred from different cues and the psychological responses to these cues are different, depending on the type of threat inferred (Gibson 1979; Neuberg and DeScioli 2015). Thus, the systems would need to consider who is likely to pose a threat, and under what circumstances. While it is not generally possible to directly observe the types of threats posed by other individuals, a functional mechanism must use whatever cues are available. As suggested by Neuberg and DeScioli (2015, 706): “One cannot directly perceive another’s pathogens or intentions to harm. Rather, people must rely on cues – features of morphol- ogy, behavior, or reputation – that correlate (even if only weakly) with threats.” For instance, certain characteristics that in ancestral environments would have been correlated with threat, such as outgroup identity, being young, male, and physically formidable, would be more likely to act as cues for physical threat, than being old or female (Neuberg and DeScioli 2015; Neuberg and Sng 2013). Environmental factors, such as being in a dark area (Schaller, Park, and Mueller 2003), or the relative size of one’s coalition relative a potentially hostile coalition (Schaller and Abeysinghe 2006), can likewise impact threat perceptions. Since failing to perceive a threat in the presence of an actual danger is generally – in fitness terms – a much more costly error than the alternative (perceiving a threat when there is actually none), an adaptive system would also be geared towards being oversensitive to threat cues. Just like a smoke detector (Nesse 2005), threat perception programs are thus likely to be biased to make the least costly error and react even to weak cues (Haselton, Nettle, and Andrews 2005; Johnson et al. 2013). Furthermore, the best response to a threat of physical harm might not be the same as to a pathogen threat or economic threat. An adaptive threat management system would result in behavioral and attitudinal responses that best served to maximize the individual’s fitness in response to the specific threat. Hence, depending on what type of threat is posed in the situation, or by the

32 particular individual or group, different psychological mechanisms will be engaged (Cottrell and Neuberg 2005; Neuberg and Cottrell 2006). Research has further shown that depending on what goals and motivations are salient, different behavioral responses are engaged (Kenrick, Griskevicius, et al. 2010; Kenrick, Neuberg, et al. 2010; Schaller et al. 2017). This includes both changes in information processing and the formation of cognitive biases (Haselton et al. 2009; Haselton, Nettle, and Andrews 2005), as well as preference changes, and even preference reversals (Kenrick et al. 2009, 2012; Li et al. 2012). One central motivation is self-protection from threats (Kenrick, Griskevicius, et al. 2010). Threats to survival and cues of resource scarcity have also been argued to affect risk preferences in both humans as well as non- human species (McDermott, Fowler, and Smirnov 2008; Stephens 1981; Stephens and Krebs 1987). All of this points to preferences being highly sus- ceptible to the cues and contexts in which individuals find themselves, and in particular to the threats present in those contexts. Overall, the literature within evolutionary psychology has convincingly made the case that preferences are not fixed or a black box, but are instead functions of social contexts. Observable cues in the social environment increase or decrease threat perceptions, which in turn changes the preference rankings of observers in predictable ways. Thus, threat perceptions act as a kind of internal regulatory variable, which adjusts preference formation (Tooby and Cosmides 2008). For instance, cues of physical threats posed by outgroup members may decrease willingness to associate or interact with them, as well as decrease altruism and increase spitefulness. As preference rankings are central for defining the possible bargaining space in conflicts, what bargaining outcomes are considered acceptable or not will also vary depending on the threat-relevant cues in the environment. By exploring how the threat cues suggested by evolutionary theory impact social preferences, this dissertation seeks to explore the role they play in the observed variability in the willingness and ability of actors to interact and find mutually acceptable exchanges and compromises to conflicts of interest.

Evolutionary psychology and criticisms of biological explanations of human behavior Peace and conflict research is by nature an interdisciplinary field. Since its origins, the field sought to combine a multiplicity of theoretical and method- ological approaches to understand the nature of human conflict and cooperation. Indeed, much of the seminal work within the field has come out of the application of new approaches to understanding the causes and resolution of conflict. In his groundbreaking “Why Men Rebel”, Ted Robert Gurr (1970) drew upon the contributions from group psychology, and Fearon’s seminal

33 bargaining theory of war is based upon rational choice theory and the assumptions of neoclassical economics (Fearon 1995). Tying on to this tradition, I seek to bring some of the recent developments from evolutionary psychology into peace and conflict research. Evolutionary psychology, although originating already in the writings of Darwin (1859, 1871), has gained a lot of traction in the last few decades. Pre- vious research on the social-psychological foundations of human social behavior has often relied on “minitheories” (Kenrick 2011). That is, theories are often developed to address a certain empirical phenomenon, but say little about other, sometimes even closely related, phenomena. In its quest to look at the ultimate reasons, rather than proximate reasons, for behaviors and attitudes, evolutionary psychology endeavors to create a unified theory of the mind based on an evolutionary reasoning (Cosmides and Tooby 2013; Tooby and Cosmides 2015). This approach has shown its appeal also to researchers in other fields of social science, such as economics and political science (Sidanius and Kurzban 2003). This development has been encouraged by the evolutionary psychology community, even going so far as to state that “…scientifically psychology (along with the other social and behavioral sciences) is a subbranch of evolutionary biology and can no longer be defensibly divorced from it” (Tooby and Cosmides 2015, 11). Nonetheless, with a few notable exceptions (Johnson 2015; Johnson and Toft 2014; Landau-Wells 2018; Little and Zeitzoff 2017; McDermott and Davenport 2017), its introduction into the fields of international relations and peace and conflict research has so far been slow. One potential reason for this is that, despite its growing appeal to scholars from numerous scientific fields, the evolutionary approach has over the years experienced a substantial amount of criticism. Much of this has been addressed thoroughly by numerous authors (see for instance Barkow 2006; Hagen 2015; Kurzban and Haselton 2006; Pinker 2002; Segerstråle 2006). Indeed, it has become standard for books focusing on evolutionary or biological explanations of human behavior to come with a chapter addressing some of these cri- tiques (e.g. Hatemi and McDermott 2011; von Hippel, Haselton, and Forgas 2013; Pinker 2002; Thornhill and Palmer 2000). While these authors do a thorough job of addressing them, I will briefly address three of the most common criticisms of the evolutionary approach in the social sciences. The first two are scientific objections (see Hagen 2015 for a more thorough review), whereas the third is of a more philosophical, or even moralistic, nature (see Barkow 2006 for an in-depth discussion). The first, and perhaps most substantial, critique against an evolutionary adaptationist view is what is often called “hyperadaptationism”, or in a more colloquial language, “just-so stories”. This critique, raised perhaps most elo- quently by biologists Stephen J. Gould and Richard C. Lewontin in their paper The spandrels of San Marco and the Panglossian paradigm (Gould and

34 Lewontin 1979), point out that it is always possible to create a functional explanation for any feature of an organism. The spandrels (the triangular spaces between cross-arched vaults) of the cathedral of San Marco in Venice look like they were intended design features due to their exquisite decorations, but are in fact a by-product of a vaulted architectural design. So too, behaviors and traits that may seem functional may in fact be a by-product of some other factor or constraint. Adaptationists, Gould and Lewontin (1979) argue, look at the spandrels and come up with a post-hoc functional explanation, instead of looking at the underlying factors – the arched vault design of which they are an unintended by-product. This argument can be made for essentially any evolutionary explanation. A classical example is physical sexual dimor- phism13. While females are larger than males in many animal species, which can be explained by a selection pressure on females to have larger bodies to be able to produce eggs, something already noted by Darwin (1871), the opposite is true among most mammals. Despite mammalian females often also having to bear the extra energetic demands of nurturing their young after birth (Lindenfors, Gittleman, and Jones 2007), males are nevertheless on average bigger. This has been explained by a selection pressure on males in species with lower male parental investment to instead invest in competition with other males, which requires strength and size. That is, two completely opposite outcomes can both be explained with the same paradigm. As a result, the critics argue, the theory becomes non-falsifiable. Every trait or feature of an organism must be an adaptation resulting from selection pressures, and a log- ically consistent just-so story that explains its functionality can always be cre- ated post-hoc. This is the Panglossian paradigm14 that Gould and Lewontin point out (1979). While this critique is indeed valid, and it can be argued that the spandrels paper has forced evolutionary psychologists to be more thorough in their application of adaptational arguments, to characterize the field of evolutionary psychology as hyperadaptationist is in many cases a misnomer. Evolutionary psychologists rarely assume that everything is necessarily an adaptation (Kurzban and Haselton 2006). The adherents of evolutionary psychology would instead point out that an adaptationist approach is mainly a theoretical reasoning to generate hypotheses which can be tested against data (Andrews, Gangestad, and Matthews 2002; Segerstråle 2006), rather than a post-hoc ra- tionalization of empirical observations. What most evolutionary psychologists do, as well as any scientist employing an adaptationist approach, is look at psychological features that are potential functional adaptations, try to discover the function they might serve and implications thereof, and finally generate

13 That is, differences in physical appearance and size between males and females of the same species. 14 So named after the character Professor Pangloss in Voltaire’s Candide, who argues that everything is the way it is because we live in the best possible world.

35 testable hypotheses based on this (Andrews, Gangestad, and Matthews 2002). In the case of sexual dimorphism, the adaptationist explanation would be that given certain circumstances, selection would favor larger females than males, whereas other circumstances would favor the opposite outcome. Based on this reasoning, we can set up and test specific hypotheses, e.g. species with lower parental investment for males also have more competition among males for mates, which favors a male-biased sexual size dimorphism (Lindenfors, Gittleman, and Jones 2007). Thus, evolutionary theories are not able to explain everything, and neither are they meant to. However, rather than in a Panglossian manner suggesting just-so stories for every possible phenomenon, the approach actually allows us to generate hypotheses based on ultimate, as opposed to proximate, reasons. As argued by Hagen (2015), Gould and Lewontin (1979) have it the wrong way around. The adaptationist program is about identifying potential functional traits from other, non-adapted traits. Given that support for such a hypothesis is found, it is the supposition that something is not an adaptation, or a “spandrel”, which needs to be proven (Andrews, Gangestad, and Matthews 2002). Besides just-so stories, critics also point out that it may be difficult to formulate hypotheses about why a specific trait is a functional adaptation, if we do not know the environment in which it developed. It should be evident that the modern society in which most humans today live is far removed from the environment in which our ancestors evolved over millennia (Giphart and Van Vugt 2016). This environment, often termed the Environment of Evolutionary Adaptation (EEA), is often equated with the environment of the Pleistocene era (Hagen 2015). This was the environment in which early hominids, including the first modern humans, spent most of their evolutionary history, and so any adaptations are likely to be functional responses to reoccurring adaptive problems in this environment (Tooby and Cosmides 2015). The criticism often raised is that there is obviously no written record of the Pleistocene era. Fur- thermore, since the archaeological evidence, while substantial, is insufficient to make a definitive case of what the EEA would have actually looked like, any argument based on adaptation to this environment is hypothetical at best (see Hagen 2015 for a summary and rebuttal of this critique). Just like the critique against the use of just-so stories, the above critique is warranted, and has forced evolutionary psychologists to be more careful in specifying the particulars of the contexts in which the functional mechanisms that they propose would have evolved. However, such an argument would then apply even more so to the environments of other species, such as various extinct species, for which the archaeological record is more scant. Indeed, we know substantially more about the environment of the ancestors of modern humans than we know about the environment of the ancestors of many other animals, for which evolutionary theories are nevertheless seen as appropriate (Daly and Wilson 1999). In addition, anthropological research from modern

36 hunter gatherer societies ,which exist in an environment reminiscent of that which our ancestors inhabited in the Pleistocene, can provide us with further clues to the environment and the adaptive problems that early humans faced (Giphart and Van Vugt 2016; Tooby and Cosmides 2015). Even when there is less than 100 per cent certainty, we do in fact know a lot about the EEA, and thus it can still help us derive better hypotheses about functional behavior than no assumptions at all (Tooby and Cosmides 2015). The final critique differs from the former in that it is not based on question- ing the scientific assumptions of evolutionary psychology, but instead questions it at a more philosophical level. In particular, it relates to an issue called the naturalistic fallacy, or the “is-ought” conflation, i.e. the belief that if something is natural, that means that it must also be good or justified. If something is a certain way, that is because it ought to be that way. The fallacy comes from applying almost Rousseauian ideals of “the noble savage” on the natural state of humanity (Pinker 2002). In the natural state, the argument goes, man- kind was happier, but societal factors (e.g. patriarchal, colonial or authoritarian structures) alone are to blame for the iniquities we now suffer. If morally reproachable behaviors have a basis in biology, rather than in changeable social structures, these critics argue, this means not only that social engineering projects are doomed to fail, but also that individuals engaging in these im- moral behaviors cannot be blamed for their actions. Individuals who have committed criminal acts can say “it’s not me, it’s in my genes!”, and thereby shift culpability to a factor outside their control. Much of this critique has been based on moralistic arguments (Segerstråle 2006), stating that by trying to find the “natural” causes of various behaviors and phenomena that are generally not socially acceptable, evolutionary psychology is essentially justifying or legitimizing them. This is far from the truth, and often indeed quite contrary to the stated intentions of evolutionary psychologists, who much like other social scientists have an interest in their findings being able to influence policy for the better (e.g. Campbell 2006; Thornhill and Palmer 2000). Evolutionary psychology makes no normative assumptions about the morality of the natural state. Natural selection does not care about morals, only fitness. Hence, some behaviors may have been se- lected for because of their fitness enhancing effects, which are morally repre- hensible, e.g. rape (Thornhill and Palmer 2000), homicide (Daly and Wilson 1988), and racism (Kenrick 2011). In no way does the fact that evolutionary psychologists study the biological origins of these social phenomena mean that they endorse them or consider them desirable, any more than a physician studying the biological origins of cancer considers cancer desirable. Rather, by studying the ultimate reasons for these behaviors, it may in fact be possible to find ways of mitigating them, by using the same evolved human psychology to appeal to what Pinker (2011) has aptly termed “the better angels of our nature”.

37 Methodology and empirical strategy In this dissertation, I address the role of threat in affecting preferences in conflict, both from a theoretical and empirical perspective. For this purpose, the essays herein utilize somewhat different methodologies. While the first essay relies on a game theoretic analysis of bargaining theory, the other three essays are based on survey experiments carried out in a field setting among refugees in Turkey. In this section, I will discuss the main benefits and drawbacks of using an experimental method, as opposed to a more standard observational data approach. The data collection, including several large-scale survey experimental rounds that form the basis of the empirical material of this dissertation, will also be described thoroughly below.

Field experiments Preferences, as well as threat perceptions, are difficult concepts to successfully measure in observational studies of conflict (Hugh-Jones, Ron, and Zultan 2019). Actual perceptions and preferences only exist in people’s minds, and cannot normally be observed. Furthermore, social interactions and compromises do not occur randomly, and it seems plausible that their occurrence may be correlated with preferences and perceptions, meaning that any causal relationships will be impossible to isolate. For the empirical part of this dissertation, I thus rely on an experimental method. Experimental methods developed from the practices of psychology and economics have become more frequent also in other fields of social science, including political science and peace and conflict research (McDermott 2013). The main benefit of the experimental approach is that it allows for more reliable causal inference. When explanatory factors are randomly assigned to individuals, regardless of other individual characteristics, this overcomes the problem of omitted variable bias that might otherwise contaminate results. Furthermore, it allows us to create situations which, although artificial, get us closer to the theoretical concepts we are in- terested in. Experimental studies have often relied on WEIRD samples, mainly con- sisting of undergraduate students at American or European universities (Henrich, Heine, and Norenzayan 2010b, 2010a). While these samples may be suitable for some research questions, and provide an easily controlled environment, they may not always be representative of how humans in general think or behave. In particular, when investigating the effect of threat in a conflict setting, the context may be difficult to credibly replicate in a purely laboratory setting, and the external validity of such studies would suffer as a result. We do not know to what degree the results from laboratory experiments on group conflict (e.g. de Dreu, Dussel, and Ten Velden 2015; Reimers and Diekhof 2015; Yuki and Yokota 2009) travel to a field setting. To remedy this, the experiments in this dissertation were carried out in a field setting among

38 refugees from the conflicts in Syria and Iraq. The experiments conducted in this dissertation are therefore of the type Harrison and List (2004) term “arte- factual field experiments”. That is, experiments that use a standard lab methodology but a nonstandard subject pool. While the types of situations that individuals are asked to evaluate in experiments (save natural experiments) are generally somewhat artificial by design, the field setting allows a stronger external validity claim than what is usually possible in laboratory settings (McDermott 2013). There are of course other costs, such as reduced control over the setting and selection issues introduced by this method, that need to be weighed against the benefits. I will develop some of these considerations later. A field experimental approach is appropriate given the research question and theory for several reasons. Firstly, micro-level data on individuals in conflict zones is still fairly rare, and we thus know a lot less about the attitudes and behaviors of individuals under such circumstances than we do under more “normal” settings. If, as theory and empirics suggest (Voors and Bulte 2014), people’s preferences, attitudes and behaviors are qualitatively different in a conflict setting, then we need to study these phenomena in an actual conflict setting. Certainly, some issues are more easily addressed in a laboratory setting. For instance, the literature on minimal groups has pointed out the role that group affiliation plays for negative attitudes. However, while some laboratory experiments have attempted to incorporate a more realistic conflict context (Abbink 2012), the situation still remains highly artificial. Hence, being able to run controlled experiments in a sample from a conflict-exposed population, and with individuals who have experienced actual group-based conflict, is a major contribution of this dissertation. Secondly, an experimental setting allows me to randomly assign the relevant threat cues. This is especially important for the case of group affiliation. The decision to interact with others will not be independent of their group affiliation, and hence it would be impossible to disentangle the (causal) effects of group affiliation on preferences. An experimental approach allows me to do this, while still maintaining the relevant and realistic conflict groups. By running these experiments in the field, with naturally occurring and conflict relevant groups, we can thus overcome both the difficulty of observing these kinds of interactions in natural data, as well as the problem of disentangling causal effects. Of course, in a setting like the one studied here, many of the relevant variables such as conflict exposure and threat coping are not reasonably possible to randomly assign, both for ethical and practical reasons. This, of course, limits the possible internal validity claims that can be made. However, I believe that the gains in external validity that can be made by measuring these real factors, as opposed to the more artificial measures that have been used in laboratory studies to simulate motivations and threats, outweigh this limitation.

39 The empirical context – The conflicts in Syria and Iraq While the topic of threat and willingness to cooperate does not by itself have a geographical or regional focus, the empirical essays are based on survey experiments carried out in the context of ongoing15 conflict in the Middle East. The participants in the studies were recruited from refugees of these conflicts currently residing in Turkey. These conflicts are among the most complex conflicts of recent times, and as such an exhaustive discussion would be far too lengthy to be covered here. However a brief overview of the Middle East- ern conflict complex, and the conflicts in Syria and Iraq in particular, is warranted to contextualize the empirical material. Syria and Iraq alike have suffered from protracted armed conflict – in Syria since the beginning of the “Arab Spring” anti-government uprisings in 2011, and in Iraq since the intervention of the US-led coalition that toppled Saddam Hussein’s regime in 2003. In Syria, the uprising was against President Bashar al-Assad. Al-Assad’s support is based in the Alawi community, and primarily aligned with the Shia Arab communities of Syria. The opposition is divided between the Free Syrian Army, Jihadi insurgents (IS, the Al-Nusra front etc.), and other opposition groups primarily based in the Kurdish and Sunni communities of Syria. In Iraq, there has been ongoing violence since the US-led intervention to topple the regime of Saddam Hussein in 2003. In Iraq, the government is mainly Shiite, with a predominantly Sunni opposition, with groups of both a Jihadist (IS), and more secularly secessionist (Kurdish) orientation. Both countries rank among the top five countries in the world in terms of conflict related fatalities since 1989 (Pettersson and Eck 2018). Due to the recent and ongoing nature of the conflict, and the relatively few experimental and survey studies (but see Kao and Revkin 2018 for an example) carried out in the setting since the resurgence of sectarian violence following the Arab Spring, there is a great interest in studying how the conflict has affected those exposed to it. This has been further heightened in light of the refugee crisis in both adjacent countries and in Europe, and also because these are the individuals upon whom the burden of rebuilding society will fall once the conflict ends. In addition, and more specifically for the purposes of this dissertation, the conflict complex in the Middle East is a relevant case for studying threat and groups, as it is a context where the conflicts follow largely sectarian lines. It has been argued that a large part of the escalating hostilities between primarily Sunni and Shia groups is because of a discourse of threat being preva- lent among sectarian groups (Darwich and Fakhoury 2016). While the schism between the Sunni and Shia denominations within Islam goes back to the death of Ali in the year 661 AD, it has been at the center of many subsequent conflicts within Islam, with Umayyids, Abbasids, Seljuks, Safavids and Ottomans

15 As of the time this is written (UCDP Bulletin 2019).

40 vying for control over the fertile lands of Mesopotamia throughout the centuries (Gonzales 2013). Following the resurgence of violence after the Arab Spring, this schism has once again come to the fore in the conflicts in both Syria and Iraq (Abdo 2017; Darwich and Fakhoury 2016). A recent study indicated that exposure to sectarian conflict in this setting has had an impact on political and religious preferences (Mousa 2018). How- ever, the impact of sectarian and ethnoreligious group conflict on social preferences in this high-threat environment has yet to be investigated. Further- more, if there are in fact other threat-relevant factors that moderate this effect, even in this highly polarized environment, that would have very strong scientific as well as policy implications. Relying on data from refugees from these conflicts, currently residing in Turkey, I thus investigate the effects of threat on social preferences in a highly relevant setting, while also making an empirical contribution by providing novel data from this particular conflict setting. In essence, the field setting with refugees from the conflicts in the Middle East is a “hard case” for studying factors that may moderate group affiliation. In conflicts such as the Syrian and Iraqi ones where the conflict cleavages, and so much of the violence perpetrated, have gone along ethnoreligious lines, we would expect the effects of group affiliation to be substantial, and potentially overtake any other effects. Thus, this setting presents a particularly relevant and difficult test for identifying threat-relevant factors that would moderate the standard parochial tendencies (Bauer et al. 2016). Nevertheless, in two of my essays (Essay II and Essay III), I show how factors relevant for threat perceptions actually moderate the effects of group affiliation.

Data collection Doing research with human subjects in any field setting comes with its own set of challenges and considerations, and particularly so in a conflict setting. This relates both to the access and reliability of the data, and to the ethical considerations of research dealing with human subjects in highly sensitive situations.

Practical considerations The data collection for this dissertation was conducted within a larger data collection effort using surveys and survey experiments in populations of war- exposed and displaced individuals of the Middle East. This data collection was led by Jonathan Hall, and funded by the Swedish Research Council. As the samples are not standard convenience samples, there were a number of practical considerations in collecting the data that had to be considered. In particular, the population considered is generally difficult to access. Refugees living outside camps mixed with the general population are difficult to sample

41 using standard randomized methods. For this purpose, a different type of sampling strategy had to be used in some cases. While the data for Essay IV was based on a sample representative of the Syrian Sunni Arab population of the Midyat refugee camp in Mardin, near the Syrian border of Turkey, the samples in Essays II and III were of a different nature. In this case, a purposive, community based sampling was employed (Teddlie and Yu 2007). To be able to effectively interact with individuals in the refugee community, and build trust, the enumerators recruited to carry out the experiments were individuals who had experience working with refugees, and many of them had themselves a refugee background. The enumerators were trained in situ by the fieldwork coordinator, with assistance from the research team in Sweden. Since door-to-door recruitment would have been very difficult, such a sampling strategy instead relied on building trust with key members of the community. By doing so, it was possible to get in touch with individuals who were willing to participate in the survey experiments. The enumerators in the field worked through their networks to find prospective participants, and based on their contacts, a snowball sampling approach (Goodman 1961) was used, but without putting explicit restrictions on the number of referrals. Such an approach comes with some potential issues, particularly with regard to potential bias in the sample; for instance, there may be correlations between being in a particular social network and variables of interest (Biernacki and Waldorf 1981). This was somewhat mitigated by the purposive approach where enumerators strove to get a sample as representative of the population as possible in terms of background variables, and the potential drawbacks were considered to be outweighed by the benefits of being able to access this unique population. A field setting of course also has a number of limitations. For instance, there may be differences between individuals who end up in camps and those who end up in cities. Likewise, there may be a selection bias in who chooses to participate in the experiments. These problems are by no means unique to the surveys from this dissertation, but they do to some degree limit the external validity of the sample. While the enumerators strove to create as balanced and representative sample as possible with regards to important background factors such as gender, age, and country of origin, this means that the samples are not standard random samples. This was a calculated tradeoff. While we were able to access a very relevant and non-WEIRD (Henrich, Heine, and Norenzayan 2010a, 2010b) sample that is otherwise difficult to sample, some representativeness had to be sacrificed. In particular, the samples were slightly biased towards younger and more highly educated participants. This may have been a result of the social networks of the enumerators, who were themselves university students. As a result, their social networks would have had a bias towards persons with similar characteristics as themselves, which would come through in the data. Of course it is also possible that younger and more highly educated individuals are more likely to be willing to participate in the survey.

42 Particularly since it was implemented digitally, which even though it was de- signed to be as easy as possible to do, may exclude some digitally inexperi- enced individuals. Nevertheless, I believe that even though this may have an effect on external validity in terms of the representativeness of the sample, the added benefit of being able to access the unique and highly relevant population makes the tradeoff acceptable. Hence, the sampling approach used in the studies is warranted.

Ethical considerations As the subject populations in my studies are primarily civilian refugees from the conflicts in Syria and Iraq, they have all been through a variety of difficult and potentially traumatic events. Indeed, it is evident from the data that many of them have experienced a plethora of traumatic events, many of which are hard to imagine for most Western individuals born in the post-war era. In such a situation, it is of great importance to make sure that the research is conducted in such a way that it, as much as possible, does not risk posing harm (whether physical or emotional) to prospective participants (e.g. Wood 2006). As some of the questions that the subjects are asked to answer relate to sensitive topics, such as their experiences during the conflict, the manner in which subjects are approached with these questions requires special consideration. For instance, in a conflict-exposed population, revisiting traumatic experiences in an intense and too evocative way could under some circumstances lead to retraumatization (e.g. Brounéus 2008). Such risks always have to be considered when de- signing studies, by setting them up in such a way that these risks are mitigated as much as possible, and weighing any remaining risks against the potential social and scientific benefits of the study. All the field surveys had ethical approval from the regional ethics review board in Uppsala, (Dnr. 2016/189 for Essay III and Dnr. 2016/189/1 for Es- says II and IV). Even though care was taken to mitigate the risks, research in these kinds of situations can never be made completely risk-free, and hence the potential risks have to be weighed against the benefits of the research. In this case, the questions are of importance as they pertain to how individuals from conflict-exposed populations can get along and potentially build a future together after the violence has ceased. To be able to properly address these issues, it is imperative to study the actual contexts where this matters, i.e. among individuals who have themselves experienced conflict and who are also the ones that will be rebuilding the war torn societies after conflict ends. Likewise, even for those refugees who chose to remain in the host countries, it is important to study how the experiences of war affect preferences, as this may have a great impact on their ability to carve out a new life in the host country. Studying these issues in an artificial lab-setting, with standard convenience samples can only get us so far, and to properly address these issues, we must go to the field setting (McDermott 2013). In addition, my research

43 also has implications for policy matters. If, as my results suggest, there are other factors beside group identity that drive people’s willingness to interact and cooperate, even after violent intergroup conflict, this is an important les- son for anyone working to build up relations in war-torn societies. Thus, by better understanding how threats impact pro-sociality and willingness to compromise, we may be able to discover new tools which may help foster reconstruction and reconciliation in societies that have been torn apart by war. For instance, if it is not the group identity per se, but the threats which it is perceived as posing, that matters for willingness to interact with others, then by managing threat perceptions, there may be ways to get former enemies to cooperate and interact. Thus, the risks also have to be weighed against the value of these findings in being able to help rebuild societies after conflict, and in the long run, perhaps being able to prevent conflicts from occurring in the first place. The two main risks usually associated with this type of survey research in conflict-exposed populations are: (i) the risk of a breach of integrity; and (ii) psychological discomfort. In the surveys underlying the empirical material for this dissertation, a number of precautions were undertaken to mitigate such risks. I will briefly describe some of these measures below. Firstly, there was a thorough development and testing period, throughout which the researchers worked closely with the field team to identify unclear, sensitive or otherwise problematic components, that would need revision. As many members of the field team had a refugee background, they had a good sense of what types of issues would be potentially sensitive to the subject population. For instance, some questions, such as any items relating explicitly to sexual violence or abuse, were removed or rephrased (using more general phrases such as physical abuse), as such experiences are traumatizing and stig- matizing enough in a Western context, but could have even more severe repercussions for individuals in the more religiously conservative setting where the studies were conducted. For the same reasons, any questions relating directly to the political situation were carefully vetted and phrased in such a way that any risk to participating subjects as well as enumerators was minimized. In addition, this approach helped with clarity by making sure that questions were phrased in a colloquial way that all subjects would be able to understand, making it more accessible (while also reducing the risk of misunderstandings introducing noise into the data). The enumerators’ refugee background also helped with access to the population, as their knowledge and ability to understand the experiences of the subject population helped them to build trust and connect with prospective participants. Anonymity of all answers is a further important aspect of the design. A lot of the negative effects of the truth telling processes noted by Brounéus (2008) came as a result of harassment or repercussions against those who had chosen to speak up. Thus, despite the abovementioned removal or revision of ques-

44 tions that could be of particularly sensitive nature, limiting the risk that a participating individual’s answers can be observed is of great importance. For this purpose, no detailed personal information was stored that could directly link an individual to a specific answer. Furthermore, the implementation uti- lized digital assistants (pads) on which the participants responded to the survey. This meant that the participants were not asked the questions face to face by an enumerator, but filled them in individually on the pads. The only excep- tion to this was some individuals who were illiterate, but who still chose to participate. These individuals completed the survey as an interview with the enumerator. Just as with the normal procedure, no identifying information was recorded for these individuals either. The data filled in was never available to the enumerators, who could only see whether the participants had answered the survey or not (but not what they had answered, or how many or which questions). Only the responsible researchers in Sweden had direct access to the data. As a further note regarding the ethical considerations, it should be pointed out that none of the survey experiments included any form of decep- tion. Neither did any of them actually manipulate the level of fear or harm by trying to induce these types of emotions directly. Rather, the cues were only identity markers of the other individual, which although potentially indicative of threats, are not likely to by themselves cause any significant emotional discomfort. Of course, there are additional challenges in the modern digital environment, such as the risk of digital surveillance, or losing access to the data (van Baalen 2018). These risks were mitigated by using a well-known and reliable survey tool (in this case the online survey software Qualtrics), which has strong encryption and secure data storage ensuring researcher access. Thus, some of the main problems with field data, such as transferring the data safely out of the field location, can be circumvented. More details on the individual surveys and what they entailed can be found in the respective essays and ap- pendices in the dissertation.

Presenting the essays The dissertation consists of four essays that in various ways address the relationship between threat, social preferences and willingness to compromise. Each essay is a self-contained contribution, and each speaks to slightly different literatures and topics, but all relate to the overarching theme and question of the dissertation, i.e. the effect of threat on social preferences relating to pro- sociality and willingness to compromise.

45 Essay I: Divide or Conquer? Deep Rationality and Seemingly Indivisible Issues as a Cause of Conflict In Essay I, I look at the impact of preference change on willingness to compromise over contested issues. In particular, I show how threat may result in changes of preferences over valuation and risk leading to agents acting as though issues were indivisible, thus preventing standard bargaining outcomes from being reached. Unlike the three following essays, which all are empirical investigations, the first essay is a theoretical contribution, illustrating within the framework of a standard bargaining model (Hirshleifer 2001) how threat may impact willingness to compromise. In the essay, I argue that taking an evolutionary approach to economic preferences over risk and valuation can help resolve the puzzle of seemingly indivisible issues as a cause of conflict. Evolutionary models have shown that under situations of survival threat and strong self-protection motivations, the preferences of prospect theory (risk seeking and loss aversion (Kahneman and Tversky 1979)) can be adaptive (Li et al. 2012; McDermott, Fowler, and Smirnov 2008). If this is the case, we would assume that in conflicts categorized by such (real or symbolic) survival threat, individuals would display these preferences. In a formal bargaining model, I then show that prospect theory preferences can lead agents to prefer risky all-or-nothing gambles to equivalent divisions, and to overvalue the issues to the point where any side-payment would be prohibitively large. How- ever, I argue that these preferences will primarily come to the fore in conflicts where a group’s (or individual’s) survival is threatened. Hence why we are much more likely to see these impediments in for instance sons-of-the-soil conflicts (Fearon and Laitin 2011), or conflicts where the issue at stake is of importance to group survival, identity and cohesion. However, we would not expect to see this to the same degree when sovereign states disagree over land, even when such land has sacred connotations (Hensel and Mitchell 2005). The conditions will thus affect the degree of loss aversion and risk seeking, under some circumstances making issues appear indivisible. Based on this, I argue that by taking an evolutionary “deep rationality” approach (Kenrick et al. 2009), we can gain some understanding of the puzzling observation of seemingly indivisible issues causing bargaining failure. This is important as it may help us find ways of overcoming such impediments to conflict resolution that are often seen as driving conflict intractability. If issues are not really indivisible, but agents act this way as a function of threats in the environment, we may be able to make parties see possible solutions to their incompatibilities, by managing threats.

46 Essay II: Indirect reciprocity and tradeoff paradigms in the wake of violent intergroup conflict In the second essay, I focus on the individual level, and look at how threat cues affect the willingness to compromise with others. In a survey experiment carried out among Syrian and Iraqi refugees in the Central Anatolia region of Turkey, I show that the effect of group affiliation on willingness to compromise is in fact strongly conditioned by the physical threat posed by these groups in the preceding conflict. I hypothesize that indirect reciprocity (Nowak and Sigmund 2005), i.e. an individual’s willingness to cooperate with others, based on their previous reputation for cooperation, may drive group biases after violent conflict. I argue that the previous actions of a group are used as a proxy for cooperative reputation, where hostile acts perpetrated by members of a group are used as signals of that group’s members being un- likely reciprocal partners. In essence, the previous actions act as inputs for a “group cooperation heuristic” (Yamagishi and Kiyonari 2000), used to evaluate the appropriate approach to interacting with others. In the experiment, subjects evaluated a range of trade-offs over a contested issue at different prices, with an ingroup or outgroup member. The analysis uses a multilevel logistical regression, with decisions (tradeoffs at different prices) nested within individuals. Results show that people in general are more price sensitive, and employ different trade-off paradigms (A. P. Fiske and Tetlock 1997; Rai and Fiske 2011; Tetlock et al. 2000) depending on the identity of the other party. As expected, they employ a so called “Market Pricing” paradigm when trading with outgroups, but a so called “Communal Sharing” paradigm when trading with ingroups (Kenrick and Griskevicius 2013, 60–73). However, the effects are not driven mainly by the group affiliations per se, but by whether the other group posed a physical threat in the conflict or not. This is suggested by the fact that only those who reported high degrees of harm by members of the outgroup applied strict Market Pricing with outgroups. Conversely, those who had been harmed by the ingroup employed a Market Pricing paradigm also with ingroup members. Hence, I argue that indirect reciprocity in intergroup interactions hinges on the expectation of functioning reciprocal relationships with members of that group in the future, and wartime victimization by members of a group is used as a cue for low likelihood of future reciprocation. The paper thus adds to our understanding of threat and its effects by showing that group biases are in fact driven by previous actions by the groups signaling physical threat, and not necessarily by the group identity in itself.

Essay III: Kindness in the aftermath of cruelty? The effect of exposure to war-time trauma on altruism across social categories The third paper, coauthored with Jonathan Hall and Dogukan Cansin Karakus, further nuances the threat concept with regard to groups. It shows that social

47 categories beyond group affiliation, which are relevant for threat evaluations, have a substantial impact on social preferences and willingness to associate with others. Furthermore, these effects are moderated by individual level exposure to violence in conflict. Primary social categories beyond group affiliation are used to form beliefs about others (S. T. Fiske 2000; Neuberg and Sng 2013). These categories can be used as more detailed signals of the types of threat posed than just group affiliation alone, and should therefore impact social attitudes and preferences (Neuberg and Schaller 2016). Hence, this essay adds further support to the notion that a more nuanced conceptualization of threat can explain the great variance we see in willingness to associate with others following violent conflict. In this experiment, we show that pro-sociality following wartime exposure is not only based on ingroup/outgroup distinctions, but on a more nuanced set of social categories based on group, age and sex. Subjects conduct a so-called welfare trade-off task (WTT) (Kirkpatrick et al. 2015), with a hypothetical other with randomly assigned traits (group, age, sex). Analyses are run using ANOVAS. Results show that altruism follows the threat level of the particular social category of the target. Groups that are perceived as less threatening, such as ingroups, women and the elderly, are given higher appraisals than more threatening groups. In particular, young outgroup men receive the lowest appraisal of all groups. These effects are exacerbated by exposure to wartime violence and trauma. Thus, the paper shows the need to look more specifically at the threat types posed when trying to predict intergroup attitudes and pro-sociality, particularly in conflict-exposed populations. People in conflict-affected communities do not only base their pro-sociality on group affiliation, and exposure to violence does not necessarily lead to parochialism. Under some circumstances, it can actually increase altruism, also across group lines, but only when these outgroup members are perceived as non-threatening.

Essay IV: Valuation and vulnerability: How post-traumatic growth impacts loss aversion The final paper looks at a slightly different part of the proposed causal chain. I argue that different threat and trauma coping strategies represent essentially different fundamental motivational states (Schaller et al. 2017). These states have profound implications for preferences over valuation, one of the precon- ditions for most economic interactions. Previous studies have argued that trauma coping mechanisms such as Post-traumatic Stress (PTS) and Post-traumatic Growth (PTG) affect preferences (Callen et al. 2014; Gilligan, Pasquale, and Samii 2014), but few have explicitly measured them in a survey. Thus, the paper contributes by showing how a motivational state, in this case PTG (Tedeschi and Calhoun 2004), aimed at dealing with threat can have a strong

48 effect on preferences, to the point of reversing the standard behavioral economics finding of loss aversion (Kahneman and Tversky 1979; Tversky and Kahneman 1992). The essay uses a survey experiment in the Midyat refugee camp, near Mardin (in Turkey, close to the Syrian border). I use the standard measures for PTG and PTS to measure trauma coping. I also use a standard measure for emotional responses to gains and losses (Harinck et al. 2007; Kenrick et al. 2012). I show that while conflict-exposed individuals with low levels of PTG display some degree of loss aversion, increases in PTG reverse this tendency, whereby those with the highest PTG even have stronger emotional responses to gains than to losses. Thus, the paper shows that different psychological mechanisms for dealing with threat and trauma can have substantial implications for preference formation. In particular, the more pro-social coping mechanism of PTG can reverse the standard tendency towards loss aversion, which could make people more prone to engage in potentially beneficial exchanges with others.

Conclusion The main puzzle that I have addressed in this dissertation is why we see such diverging effects of wartime exposure in different situations. Under some circumstances, individuals, groups and nations are willing to cooperate and interact even with former enemies, and extend altruism and pro-sociality also across group lines. However, under other circumstances, we see strong group biases, hostility and outright refusal to cooperate or even interact at all with outgroups. Previous research has addressed various such empirical phenomena and suggested some explanations for this divergence (Bauer et al. 2016; Beber, Roessler, and Scacco 2014; e.g. Cassar, Grosjean, and Whitt 2014; Grosjean 2014; Zeitzoff 2014). Building on this research, this dissertation draws upon recent developments in evolutionary psychology – a field that seeks to develop a unified theory of the mind (Kenrick 2011; Tooby and Cosmides 2015) – to help explain some of the observed divergence in outcomes. It is becoming more and more evident that whether individuals, groups and states are willing to interact, make compromises, and find mutually acceptable solutions to conflicts of interest, is a function of more than just information sets alone, as assumed in standard bargaining models. Preferences are not fixed, but are malleable and fluctuating. Whether we chose to find a mutually acceptable compromise over a contested issue, or if we want to interact and cooperate with another individual or not, will depend on a number of factors, such as characteristics of the other individual, the context, and the issue at stake. In this dissertation, I have focused on one factor of importance, namely threat. From an evolutionary affordance management perspective (Brown,

49 Neel, and Sherman 2015; Gibson 1979; Neuberg, Kenrick, and Schaller 2011), humans will have needed systems to deal both with the potential fitness benefits, but also the potential costs, of associating with others. I have argued that throughout our evolutionary history, threats posed by other individuals has been a recurring adaptive problem, to which complex responses will have been developed by a process of natural selection (Duntley 2005, 2015). As a result, perceptions of threats will impact important internal regulatory variables, changing the preferences in such ways that would have been fitness maximizing in the ancestral environment (Tooby and Cosmides 2008). However, these functions are with us still today, impacting the decisions we make, for better or for worse (Giphart and Van Vugt 2016). Thus, studying how the cues that signaled threat in these ancestral environments affect social preferences today is an important contribution. By drawing upon this more nuanced conceptualization of threat that has developed out of evolutionary psychology (Cottrell and Neuberg 2005; e.g. Neuberg and Schaller 2016), I have shown in this dissertation some of the ways in which threat may play a role in the divergent effects of conflict. Thus, including an affordance management perspective in the study of peace and conflict can open up new and exciting avenues of research, as well as answer some of the puzzling questions that have challenged scholars studying human conflict and cooperation. Firstly, in Essay I, I have pointed to the role threats may play in preventing conflict resolution, or even conflict onset, by making agents act in a risk seeking manner, overvaluing outcomes in the loss domain, and hence preferring risky all-or-nothing gambles to objectively preferable division schemes. This, I argue, will make actors behave as though issues were, in fact, indivisible, thus preventing bargained solutions to be found. Furthermore, I have shown that the often-emphasized parochial tendencies and the associated primacy of the ingroup is not necessarily as strong as is often assumed. It is not the group identity per se, but rather the underlying threat perceptions that matter. While group affiliation can be a cue of threat, this effect is moderated by other characteristics such as sex or age (Essay III). And indeed, outgroups need not always cue a threat, even in a conflict context. As I show in Essay II, it is the actions of group members signaling threat and reciprocal potential that drives willingness to compromise, not the group affiliation in and of itself. This, I argue, has important implications for how we think about the role of group identity in conflict resolution and post-conflict reconstruction. Finally, I have shown that the way in which individuals cope with threats and trauma can have a strong impact on preferences. In Essay IV, I show that the more pro- social coping mechanism of post-traumatic growth can lead to a reversal of the commonly observed phenomenon of loss aversion. Together, these results point to the importance of bringing in a more nuanced conceptualization of the role of threat in the study of peace and conflict. My dissertation shows that threats can substantially limit the possibility of cooperation and compromise, but also that by mitigating perceived threats,

50 other oft-suggested hurdles to cooperation, such as apparently indivisible issues, or parochialism, may in fact be overcome. Thus, while this is hardly the last word on the debate, the dissertation sparks some light in the often oppres- sive darkness of intergroup armed conflict. Even in the context of the most brutal and destructive conflicts, the ability of humans to cooperate across group lines – one of the main factors that have allowed humans to build complex civilizations and become the dominant species on our planet (Wilson 2012) – is still present to be called upon when needed. However, for this cooperation to function and bridging social capital to be accumulated, rather than parochial bonding social capital, managing the threats and threat perceptions in a conflict situation is of central importance. As such, the nuanced conceptualizations of threat coming out of the field of evolutionary psychology should be of great interest for academics and policy makers alike.

Avenues for future research The essays in this dissertation only scratch the surface of this rich vein of scientific inquiry opened up by taking an evolutionary psychology approach to threat perception in conflict. I argue that in the study of peace and conflict, we need to improve our conceptualization of threat, and in particular, take into consideration the evolutionary underpinnings of threat perceptions (c.f. Landau-Wells 2018). By allowing for these more nuanced conceptualizations of threat, I have pointed to instances where differing threat perceptions lead to differing outcomes. When threat is present, bargaining solutions may become untenable due to loss aversion and risk seeking, making actors behave as though issues were indivisible (Essay I). Variations in individual coping may also lead to great variation in subjective valuations, which will change preference rankings over outcomes (Essay IV). In addition, by using more nuanced threat cues, I have shown that group affiliation is not as important a predictor of willingness to compromise and associate as it may have been thought to be. Instead, more nuanced threat cues actually moderate the oft- observed parochial tendencies after violent conflict (Essays II and III). Within the limited scope of this dissertation, many new questions arise. For instance, what are the scope conditions of the findings? Even though the experimental subjects are all refugees that have escaped the conflict, the violence is still ongoing in adjacent locales. It would be interesting to see if the legacy of the threats posed in the conflict remain even once the violence ceases, or if a process of normalization of intergroup attitudes can begin (Whitt and Wilson 2007). Furthermore, maybe the fact that there is no direct threat in the present context means that individuals are more attentive to other cues beside group affiliation. For those who are in the midst of an intergroup conflict, perhaps the primacy of the group comes to the fore, crowding out more subtle cues. Thus, replication of these results among populations directly in the ongoing conflict area would provide an important contrast. Of course, the practical and

51 ethical complications in accessing that subject pool makes such an endeavor difficult, although not impossible, to achieve. Future research would also benefit from further exploring the underlying mechanism of how different types of threat may affect preferences in finer detail. Since different types of threats may have very different adaptive responses, more effort should be put into looking at exactly what type of threat other individuals and groups pose in a conflict setting. While I have here focused mainly on physical threats, conflicts settings are contexts where other types of threats may also be relevant. Members of other groups can be threats also to economic resources and social values (Neuberg and Cottrell 2006; Neuberg and Schaller 2016). Furthermore, in conflict situations with ill-functioning infrastructure and health care, the potential effects of perceived pathogen threats should also be considered (Aarøe, Osmundsen, and Petersen 2016; Huang et al. 2011; Schaller and Park 2011). A threat management approach to managing conflict and post-conflict reconstruction would need to take all such factors into consideration, and future studies will need to look specifically at what role these different threat types play in conflict settings. Most importantly, future studies should seek to incorporate measures of actual threat perceptions. While the essays in this dissertation have relied on theoretically relevant cues, some previous studies within psychology have be- gun exploring actual threat perceptions (e.g. Cottrell and Neuberg 2005). Us- ing similar methods, it would be possible to model and study the underlying mechanism, to see the pathways through which threat cues lead to preference shifts.

Policy implications While the results in this dissertation speak to more fundamental questions about human psychology, conflict and cooperation, some important insights for policy can nevertheless be gleaned from them. Particularly, since the empirical studies were carried out in a realistic field context, this gives my results a stronger external validity. The first, and perhaps most important takeaway, is that when looking at situations of intergroup conflict, while group dimensions are still important, they are far from the only components driving how people interact. This is a very important finding, particularly in contexts such as the conflicts in Syria and Iraq, where many of the main cleavages in the conflict have followed sectarian and ethnoreligious group lines. After many intergroup conflicts, there is a tendency to build reconstruction efforts around separation or complex power sharing schemes, which may not necessarily be to the liking of either party. The situation in Kosovo is a case in point. After the end of the war in the 1990’s and early 2000’s, the system was built largely upon a separation of ethnic groups, with separate systems for everything. As has been shown, this has not been a complete success (see for instance Franks and Richmond 2008; Krampe 2017). Nevertheless, the assumption that the

52 history of conflict renders ethnic groups unable to successfully build a future together seems to be present in several other conflicts as well, such as for instance the Israel-Palestine conflict (c.f. Ginges et al. 2007; Sheikh, Ginges, and Atran 2013). Nevertheless, previous research has shown that even group biases can, with time and effort, be overcome (Mironova and Whitt 2016b; Whitt and Wilson 2007). Building on these findings, I show in this dissertation that it is not necessarily the group identities per se that drive attitudes and behaviors, but rather the threats posed. In cases where there is little or no threat, group biases may be negligible. As shown in Essay III, other factors beside groups determine people’s willingness to associate, or interact pro-socially with others. Essay II even shows that in cases where there has been violence also within groups, such strategies of separation may even backfire as victims are unwilling to cooperate also with ingroup members. The results thus provide an important critique of the often argued-for primacy of the group as a factor in conflicts (e.g. de Dreu, Aaldering, and Saygi 2014; Tajfel and Turner 2001), and also support the strand of literature which suggest that basing such policies squarely upon the group may not be the best solution (Mironova and Whitt 2016b; Sambanis 2000). Indeed, partition need not at all be the best solution to ethnic conflict (Sambanis 2000). Instead of relation management, peace making may be well served by looking more at threat management. If perceptions of threat can be overcome, people from different groups may be able to interact and make compromises over contested issues almost as easily as with members of their own group. By starting with the threat, and looking at how the negative effects of conflict legacies can be mitigated, and the potential positive effects harnessed, it may be possible to build better strategies for post-conflict reconstruction and reconciliation. Such strategies would need to be much more context specific, and sensitive to the potential threat legacies, and their psychological effects on individuals and groups. The evolutionary perspective provides a novel and interesting angle on questions of “nudging” (Sunstein 2014) and social engineering to build functional and peaceful societies (c.f. Pinker 2011). While the concept of a human nature even partially determined by biological factors might seem at first to limit the ability of projects to reform societies, the evolutionary approach actually just replacing proximate factors with ultimate factors. Thereby, it can provide policy makers with a potentially powerful new toolbox for conflict resolution and management. By seeking to find the root causes of many of the problems that plague humanity, such as intergroup armed conflict, this approach is also revealing to us a new set of “levers and controls”, which can be used to nudge individuals, groups and states towards the behaviors that could potentially lead to a more peaceful world.

53 References

Aarøe, Lene, Mathias Osmundsen, and Michael Bang Petersen. 2016. “Distrust as a Disease Avoidance Strategy: Individual Differences in Disgust Sensitivity Regulate Generalized Social Trust.” Frontiers in Psychology 7(JUL). Abbink, Klaus. 2012. “Laboratory Experiments on Conflict.” In The Oxford Handbook of the Economics of Peace and Conflict, eds. Michelle R. Garfinkel and Stergios Skaperdas. New York: Oxford University Press, 532–53. Abdo, Geneive. 2017. The New Sectarianism: The Arab Uprisings and the Rebirth of the Shi’a-Sunni Divide. Oxford: Oxford University Press. Altemeyer, Bob. 1988. Enemies of Freedom: Understanding Right-Wing Authoritarianism. San Francisco: Jossey-Bass. Anderton, Charles H., and John R. Carter. 2001. “The Impact of War on Trade: An Interrupted Times-Series Study.” Journal of Peace Research 38(4): 445–57. Andrews, Paul W, Steven W Gangestad, and Dan Matthews. 2002. “Adaptationism – How to Carry out an Exaptationist Program.” Behavioral and Brain Sciences 25(04): 489-504; discussion 504-553. van Baalen, Sebastian. 2018. “‘Google Wants to Know Your Location’: The Ethical Challenges of Fieldwork in the Digital Age.” Research Ethics 14(4): 1–17. Bar-Tal, Daniel, and Eran Halperin. 2011. “Socio-Psychological Barriers to Conflict Resolution.” In Intergroup Conflicts and Their Resolution: A Social Psychological Perspective, ed. Daniel Bar-Tal. New York: Psychology Press, 217–40. Barkow, Jerome H. 2006. “Introduction: Sometimes the Bus Does Wait.” In Missing the Revolution: Darwinism for Social Scientists, ed. Jerome H. Barkow. Oxford: Oxford University Press, 302. Barro, Robert J., and Xavier I. Sala-i-Martin. 1995. Economic Growth. Cambridge: The MIT Press. Bauer, Michal et al. 2016. “Can War Foster Cooperation?” Journal of Economic Perspectives 30(3): 249–74. Bauer, Michal, Jana Cahlíková, Julie Chytilová, and Tomáš Želinský. 2018. “Social Contagion of Ethnic Hostility.” Proceedings of the National Academy of Sciences 115(19): 4881–86. Bauer, Michal, Alessandra Cassar, Julie Chytilová, and Joseph Henrich. 2014. “War’s Enduring Effects on the Development of Egalitarian Motivations and In-Group Biases.” Psychological Science 25(1): 47–57. Bauer, Michal, Nathan Fiala, and Ian Levely. 2017. “Trusting Former Rebels: An Experimental Approach to Understanding Reintegration after Civil War.” The Economic Journal 128(August): 1786–1819. Bayer, Reşat, and Matthew C. Rupert. 2004. “Effects of Civil Wars on International Trade, 1950-92.” Journal of Peace Research 41(6): 699–713.

54 Beber, Bernd, Philip Roessler, and Alexandra Scacco. 2014. “Intergroup Violence and Political Attitudes: Evidence from a Dividing Sudan.” The Journal of Politics 76(3): 649–65. Bellows, John, and Edward Miguel. 2009. “War and Local Collective Action in Sierra Leone.” Journal of Public Economics 93(11–12): 1144–57. Bernhard, Helen, Urs Fischbacher, and Ernst Fehr. 2006. “Parochial Altruism in Humans.” Nature 442(7105): 912–15. Biernacki, Patrick, and Dan Waldorf. 1981. “Snowball Sampling: Problems and Rechinques of Chain Referral Sampling.” Sociological Methods and Research 10(2): 141–63. Blattman, Christopher. 2009. “From Violence to Voting: War and Political Participation in Uganda.” American Political Science Review 103(02): 231– 47. Blattman, Christopher, and Jeannie Annan. 2010. “The Consequences of Child Soldiering.” Review of Economics and Statistics 92(4): 882–98. Blattman, Christopher, and Edward Miguel. 2010. “Civil War.” Journal of Economic Literature 48(1): 3–57. Blomqvist, Hans C., and Mats Lundahl. 2002. The Distorted Economy. Basingstoke: Palgrave MacMillan. Bowles, Samuel. 2008. “Conflict: Altruism’s Midwife.” Nature 456(November): 326–27. ———. 2009. “Did Warfare among Ancestral Hunter-Gatherers Affect the Evolution of Human Social Behaviors?” Science 324(5932): 1293–98. Braithwaite, Alex, Niheer Dasandi, and David Hudson. 2016. “Does Poverty Cause Conflict? Isolating the Causal Origins of the Conflict Trap.” Conflict Management and Peace Science 33(1): 1–22. Brewer, Marilynn B. 1999. “The Psychology of Prejudice: Ingroup Love or Outgroup Hate?” Journal of Social Issues 55(3): 429–44. Brounéus, Karen. 2008. “Truth-Telling as Talking Cure? Insecurity and Retraumatization in the Rwandan Gacaca Courts.” Security Dialogue 39(1): 55–76. Brown, Nicolas A., Rebecca Neel, and Ryne A. Sherman. 2015. “Measuring the Evolutionarily Important Goals of Situations: Situational Affordances for Adaptive Problems.” Evolutionary Psychology 13(3): 1–15. Callen, Michael, Mohammad Isaqzadeh, James D Long, and Charles Sprenger. 2014. “Violence and Risk Preference: Experimental Evidence from Afghanistan.” American Economic Review 104(1): 123–48. Camerer, Colin F., and George Loewenstein. 2004. “Behavioral Economicsa: Past, Present, Future.” In Advances in Behavioral Economics, eds. Colin F. Camerer, George Loewenstein, and Matthew Rabin. Princeton and Oxford: Princeton University Press, 3–51. Camerer, Colin F et al. 2018. “Evaluating the Replicability of Social Science Experiments in Nature and Science between 2010 and 2015.” Nature Human Behaviour 2(September): 637–44. Campbell, Anne. 2006. “Feminism and Evolutionary Psychology.” In Missing the Revolution: Darwinism for Social Scientists, ed. Jerome H. Barkow. New York: Oxford University Press, 63–100.

55 Canetti-Nisim, Daphna, Gal Ariely, and Eran Halperin. 2008. “Life, Pocketbook, or Culture: The Role of Perceived Security Threats in Promoting Exclusionist Political Attitudes toward Minorities in Israel.” Political Research Quarterly 61(1): 90–103. Canetti-Nisim, Daphna, Eran Halperin, Keren Sharvit, and Stevan E. Hobfoll. 2009. “A New Stress-Based Model of Political Extremism: Personal Exposure to Terrorism, Psychological Distress, and Exclusionist Political Attitudes.” Journal of Conflict Resolution 53(3): 363–89. Canetti, Daphna et al. 2015. “Exposure to Violence, Ethos of Conflict, and Support for Compromise: Surveys in Israel, East Jerusalem, West Bank, and Gaza.” Journal of Conflict Resolution: 1–30. Cassar, Alessandra, Pauline Grosjean, and Sam Whitt. 2013. “Legacies of Violence: Trust and Market Development.” Journal of Economic Growth 18(June): 285– 318. ———. 2014. “Social Preferences of Ex-Combatants: Survey and Experimental Evidence from Postwar Tajikistan.” In The Economics of Conflict: Theory and Empirical Evidence, ed. Karl Wärneryd. Cambridge: The MIT Press, 231–62. Cecchi, Francesco, Koen Leuveld, and Maarten Voors. 2016. “Conflict Exposure and Competitiveness: Experimental Evidence from the Football Field in Sierra Leone.” Economic Development and Cultural Change 64(3). Chen, Siyan, Norman V Loayza, and Marta Reynal-Querol. 2008. “The Aftermath of Civil War.” The World Bank Economic Review 22(1): 63–85. Child, Travers Barclay, and Elena Nikolova. 2018. War and Social Attitudes. Maastricht. Choi, Jung-Kyoo, and Samuel Bowles. 2007. “The Coevolution of Parochial Altruism and War.” Science 318(5850): 636–40. Collier, Paul. 1999. “On the Economic Consequences of Civil War.” Oxford Economic Papers 51(1): 168–83. Collier, Paul, Anke Hoeffler, and Måns Söderbom. 2008. “Post-Conflict Risks.” Journal of Peace Research 45(4): 461–78. Corbacho, Ana, Julia Philipp, and Mauricio Ruiz-vega. 2015. “Crime and Erosion of Trust : Evidence for Latin America.” World Development 70: 400–415. Cosmides, Leda, and John Tooby. 2013. “Evolutionary Psychology: New Perspectives on Cognition and Motivation.” Annual Review of Psychology 64: 201–29. Cottrell, Catherine A, and Steven L Neuberg. 2005. “Different Emotional Reactions to Different Groups: A Sociofunctional Threat-Based Approach to ‘Prejudice’.” Journal of personality and social psychology 88(5): 770–89. Daly, Martin, and Margo Wilson. 1988. Homicide. New Brunswick and London: AldineTransaction. Daly, Martin, and Margo I. Wilson. 1999. “Human Evolutionary Psychology and Animal Behaviour.” Animal Behaviour 57(3): 509–19. Darwich, May, and Tamirace Fakhoury. 2016. “Casting the Other as an Existential Threat: The Securitisation of Sectarianism in the International Relations of the Syria Crisis.” Global Discourse 6(4): 712–32. Darwin, Charles. 1859. On the Origin of Species. ———. 1871. The Descent of Man, and Selection in Relation to Sex.

56 Deglow, Annekatrin. 2016. “Localized Legacies of Civil War: Postwar Violent Crime in Northern Ireland.” Journal of Peace Research 53(6): 786–99. de Dreu, Carsten K.W. 2010. “Social Value Orientation Moderates Ingroup Love but Not Outgroup Hate in Competitive Intergroup Conflict.” Group Processes and Intergroup Relations 13(6): 701–13. de Dreu, Carsten K.W., Hillie Aaldering, and Ozum Saygi. 2014. “Integroup Conflict and Negotiating Settlement.” In Social Conflict within and between Groups, ed. Carsten K.W. de Dreu. London and New York: Psychology Press, 1–18. de Dreu, Carsten K W, D. Berno Dussel, and Femke S. Ten Velden. 2015. “In Intergroup Conflict, Self-Sacrifice Is Stronger among pro-Social Individuals, and Parochial Altruism Emerges Especially among Cognitively Taxed Individuals.” Frontiers in Psychology 6(MAY): 1–9. Druckman, James N., and Arthur Lupia. 2000. “Preference Formation.” Annual Review of Political Science 3: 1–24. Duntley, Joshua D. 2005. “Adaptions to Dangers from Humans.” In The Handbook of Evolutionary Psychology, ed. David M Buss. Hoboken: Wiley, 224–49. ———. 2015. “Adaptions to Dangers From Humans.” In The Handbook of Evolutionary Psychology Volume 1: Foundations, ed. David M Buss. Hoboken: Wiley, 264–86. Fearon, James D. 1995. “Rationalist Explanations for War.” International Organization 49(3). Fearon, James D., and David D. Laitin. 2011. “Sons of the Soil, Migrants, and Civil War.” World Development 39(2): 199–211. Fehr, Ernst, and Urs Fischbacher. 2003. “The Nature of Human Altruism.” Nature 425(6960): 785–91. Feldman, Stanley, and Karen Stenner. 1997. “Perceived Threat and Authoritarianism.” Political Psychology 18(4): 741–70. Fershtman, Chaim, and Uri Gneezy. 2001. “Discrimination in a Segmented Society : An Experimental Approach.” The Quearterly Journal of Economics 116(1): 351–77. Fiske, Alan Page, and Philip E. Tetlock. 1997. “Taboo Trade-Offs : Reactions to Transactions That Transgress the Spheres of Justice Author.” Political Psychology 18(2): 255–97. Fiske, Susan T. 2000. “Stereotyping, Prejudice, and Discrimination at the Seam between the Centuries: Evolution, Culture, Mind, and Brain.” European Journal of Social Psychology 30(3): 299–322. Franks, Jason, and Oliver P. Richmond. 2008. “Coopting Liberal Peace-Building: Untying the Gordian Knot in Kosovo.” Cooperation and Conflict 43(1): 81– 103. Gates, Scott, Håvard Hegre, Håvard Mokleiv Nygård, and Håvard Strand. 2012. “Development Consequences of Armed Conflict.” World Development 40(9): 1713–22. Getmansky, Anna, and Thomas Zeitzoff. 2014. “Terrorism and Voting: The Effect of Rocket Threat on Voting in Israeli Elections.” American Political Science Review 108(3): 588–604.

57 Ghobarah, Hazem Adam, Paul Huth, and Bruce Russett. 2003. “Civil Wars Kill and Maim People-Long after the Shooting Stops.” The American Political Science Review 97(2): 189–202. Gibson, James J. 1979. The Ecological Approach to Visual Perception. Boston: Houghton Mifflin. Gilligan, Michael J., Benjamin J. Pasquale, and Cyrus Samii. 2014. “Civil War and Social Cohesion: Lab-in-the-Field Evidence from Nepal.” American Journal of Political Science 58(3): 604–19. Ginges, Jeremy, Scott Atran, Douglas Medin, and Khalil Shikaki. 2007. “Sacred Bounds on Rational Resolution of Violent Political Conflict.” Proceedings of the National Academy of Sciences of the United States of America 104(18): 7357–60. Gintis, Herbert. 2003. “Solving the Puzzle of Prosociality.” Rationality and Society 15(2): 155–87. ———. 2008. “Strong Reciprocity and the Roots of Human Morality.” Social Justice Research 21(2): 241–53. ———. 2009. The Bounds of Reason: Game Theory and the Unification of the Behavioral Sciences. Princeton and Oxford: Princeton University Press. Gintis, Herbert, Samuel Bowles, Robert Boyd, and Ernst Fehr. 2005. “Moral Sentiments and Material Interests: Origins, Evidence, and Consequences.” In Moral Sentiments and Material Interests: The Foundations of Cooperation in Economic Life, eds. Herbert Gintis, Samuel Bowles, Robert Boyd, and Ernst Fehr. Cambridge: The MIT Press, 3–40. Giphart, Ronald, and Mark Van Vugt. 2016. Mismatch. London: Robinson. Gneezy, Ayelet, and Daniel M. T. Fessler. 2012. “Conflict , Sticks and Carrots: War Increases Prosocial Punishments and Rewards.” Proceedings of the Royal Sociaety B 279 (June 2011): 219–23. Gonzales, Nathan. 2013. The Sunni-Shia Conflict: Understanding Sectarian Violence in the Middle East. Mission Viejo, CA: Nortia Press. Goodman, Leo A. 1961. “Snowball Sampling.” The Annals of Mathematical Statistics 32(1): 148–70. Gould, Stephen J., and Richard C. Lewontin. 1979. “The Spandrels of San Marco and the Panglossian Paradigm: A Critique of the Adaptationist Programme.” Proceedings of the Royal Society of London - Biological Sciences 205(1161): 581–98. Grosjean, Pauline. 2014. “Conflict and Social and Political Preferences: Evidence from World War II and Civil Conflict in 35 European Countries.” Comparative Economic Studies 56: 424–51. Gurr, Ted Robert. 1970. Why Men Rebel. Princeton, NewJersey: Princeton University Press. Hagen, Edward H. 2015. “Evolutionary Psychology and Its Critics.” In Handbook of Evolutionary Psychology Volume I: Foundations, ed. David M. Buss. Hoboken, N.J.: Wiley, 136–60. Hall, Jonathan, Iosif Kovras, Djordje Stefanovic, and Neophytos Loizides. 2018. “Exposure to Violence and Attitudes Towards Transitional Justice.” Political Psychology 39(2): 345–63.

58 Halperin, Eran, and Daniel Bar-tal. 2015. “Social-Psychological Barriers to Peace Makng: An Empirical Examination within the Israeli Jewish Society.” Journal of Peace Research 48(5): 637–51. Hamilton, William D. 1964. “The Genetical Evolution of Social Behaviour.” Journal of Theoretical Biology 7: 17–52. Harinck, Fieke, Eric Van Dijk, Ilja Van Beest, and Paul Mersmann. 2007. “When Gains Loom Larger Than Losses: Reversed Loss Aversion for Small Amounts of Money.” Psychological Science 18(12): 1099–1105. Harrison, Glenn W., and John A. List. 2004. “Field Experiments.” Journal of Economic Literature 42(4): 1009–55. Hartman, Alexandra C, and Benjamin S Morse. 2018. “Violence , Empathy and Altruism : Evidence from the Ivorian Refugee Crisis in Liberia.” British Journal of Political Science: 1–25. Haselton, Martie G. et al. 2009. “Adaptive Rationality: An Evolutionary Perspective on Cognitive Bias.” Social Cognition 27(5): 733–63. Haselton, Martie G., Daniel Nettle, and Paul W. Andrews. 2005. “The Evolution of Cognitive Bias.” In The Handbook of Evolutionary Psychology, ed. David M Buss. Hoboken: Wiley, 724–46. Hatemi, Peter K., and Rose McDermott. 2011. “Evolution as a Theory for Political Behavior.” In Man Is by Nature a Political Animal: Evolution, Biology, and Politics, eds. Peter K. Hatemi and Rose McDermott. Chicago: The University of Chicago Press, 317. Henrich, Joseph et al. 2019. “War Increases Religiosity.” Nature Human Behaviour 3(February). Henrich, Joseph, Steven J Heine, and Ara Norenzayan. 2010a. “Most People Are Not WEIRD.” Nature 466(July): 29. ———. 2010b. “The Weirdest People in the World ?” Behavioral and Brain Sciences 33: 61–135. Hensel, Paul R, and Sara Mclaughlin Mitchell. 2005. “Issue Indivisibility and Territorial Claims.” In GeoJournal, , 275–85. von Hippel, William, Martie G. Haselton, and Joseph P. Forgas. 2013. “Evolutionary Psychology and Social Thinking.” In Evolution and the Social Mind, eds. Joseph P. Forgas, Martie G. Hasleton, and William von Hippel. New York: Routledge, 324. Hirshleifer, Jack. 2001. The Dark Side of the Force: Economic Foundations of Conflict Theory. Cambridge: Cambridge University Press. Hopfensitz, Astrid, and Josepa Miquel-Florensa. 2015. 14 How Forced Displacement Flows Affect Public Good Contributions: The Social Consequences of Conflict in Colombia. Huang, Julie Y., Alexandra Sedlovskaya, Joshua M. Ackerman, and John A. Bargh. 2011. “Immunizing Against Prejudice: Effects of Disease Protection on Attitudes Toward Out-Groups.” Psychological Science 22(12): 1550–56. Hugh-Jones, David, Itay Ron, and Ro’i Zultan. 2019. “Humans Reciprocate by Discriminating against Group Peers.” Evolution and Human Behavior 40(1): 90–95. Iqbal, Zaryab. 2010. War and the Health of Nations. Stanford, California: Stanford University Press.

59 Jehle, Geoffrey A., and Philip J. Reny. 2001. Advanced Microeconomic Theory. Second edi. Boston, MA: Addison Wesley. Johnson, Dominic D. P. 2015. “Leadership in War: Evolution, Cognition, and the Military Intelligence Hypothesis.” In Handbook of Evolutionary Psychology Volume II: Integrations, ed. David M. Buss. Hoboken: John Wiley & Sons, Inc., 722–44. Johnson, Dominic D. P., and Monica Duffy Toft. 2014. “Grounds for War: The Evolution of Territorial Conflict.” International Security 38(3). Johnson, Dominic D P, Daniel T. Blumstein, James H. Fowler, and Martie G. Haselton. 2013. “The Evolution of Error: Error Management, Cognitive Constraints, and Adaptive Decision-Making Biases.” Trends in Ecology and Evolution 28(8): 474–81. De Juan, Alexander, and Jan Henryk Pierskalla. 2016. “Civil War Violence and Political Trust: Microlevel Evidence from Nepal.” Conflict Management and Peace Science 33(1): 67–88. Kahneman, Daniel, and Amos Tversky. 1979. “Prospect Theory: An Analysis of Decision under Risk.” Econometrica 47(2): 263–92. Kao, Kristen, and Mara Redlich Revkin. 2018. To Punish or to Pardon ? Reintegrating Rebel Collaborators After Conflict in Iraq. Kaplan, Hillard S., and Steven W. Gangestad. 2005. “Life History Theory and Evolutionary Psychology.” In The Handbook of Evolutionary Psychology, ed. David M Buss. Hoboken: Wiley, 68–95. Kenrick, Douglas T., Steven L. Neuberg, et al. 2010. “Goal-Driven Cognition and Functional Behavior: The Fundamental-Motives Framework.” Current Directions in Psychological Science 19(1): 63–67. Kenrick, Douglas T. 2011. Sex, Murder and the Meaning of Life. New York: Basic Books. Kenrick, Douglas T., and Vladas Griskevicius. 2013. The Rational Animal: How Evolution Made Us Smarter than We Think. New York: Basic Books. Kenrick, Douglas T., Vladas Griskevicius, Steven L. Neuberg, and Mark Schaller. 2010. “Renovating the Pyramid of Needs: Contemporary Extensions Built Upon Ancient Foundations.” Perspectives on Psychological Science 5(3): 292–314. Kenrick, Douglas T., Yexin Jessica Li, Andrew E. White, and Steven L. Neuberg. 2012. “Economic Subselves: Fundamental Motives and Deep Rationality.” In Social Thinking and Interpersonal Behavior, eds. Joseph P. Forgas, Klaus Fiedler, and Constantine Sedikides. New York: Psychology Press, 23–43. Kenrick, Douglas T et al. 2009. “Deep Rationality: The Evolutionary Economics of Decision Making.” Social Cognition 27(5): 764–85. Kijewski, Sara, and Markus Freitag. 2018. “Civil War and the Formation of Social Trust in Kosovo: Posttraumatic Growth or War-Related Distress ?” Journal of Conflict Resolution 62(4): 717–42. Kirkpatrick, Matthew, Adrew W Delton, Harriet de Wit, and Theresa E Robertson. 2015. “Prosocial Effects of MDMA: A Measure of Generosity.” Journal of Psychopharmacology 29(6): 661–68.

60 Knack, Stephen, and Philip Keefer. 1997. “Does Social Capital Have an Economic Payoff? A Cross-Country Investigation.” The Quarterly Journal of Economics 112(4): 1251–88. Knack, Stephen, and Paul J Zak. 2003. “Building Trust: Public Policy, Interpersonal Trust, and Economic Development.” Supreme Court Economic Review 10: 91–107. Krampe, Florian. 2017. “Water for Peace? Post-Conflict Water Resource Management in Kosovo.” Cooperation and Conflict 52(2): 147–65. Kurzban, Robert, and Martie G. Haselton. 2006. “Making Hay Out of Straw? Real and Imagined Controversies in Evolutionary Psychology.” In Missing the Revolution: Darwinism for Social Scientists, ed. Jerome H. Barkow. New York: Oxford University Press, 149–62. Landau-Wells, Marika. 2018. Dealing with Danger: Threat Perception and Policy Preferences. Boston, MA. http://iibraries.mit.edu/ask. Lerner, Jennifer S., Roxana M. Gonzalez, Deborah A. Small, and Baruch Fischhoff. 2003. “Effects of Fear and Anger on Perceived Risks of Terrorism: A National Field Experiment.” Psychological Science 14(2): 144–50. Lerner, Jennifer S., and Dacher Keltner. 2001. “Fear, Anger, and Risk.” Journal of Peronality and Social Psychology 81(1): 146–59. Lerner, Jennifer S., Deborah A. Small, and George Loewenstein. 2004. “Heart Strings and Purse Strings: Carryover Effects of Emotions on Economic Decisions.” Psychological Science 15(5): 337–41. Li, Yexin Jessica, Douglas T. Kenrick, Vladas Griskevicius, and Steven L. Neuberg. 2012. “Economic Decision Biases and Fundamental Motivations: How Mating and Self-Protection Alter Loss Aversion.” Journal of Personality and Social Psychology 102(3): 550–61. Lindenfors, Patrik, John L. Gittleman, and Kate E. Jones. 2007. “Sexual Size Dimorphism in Mammals.” In Sex, Size and Gender Roles: Evolutionary Studeies of Sexual Size Dimorphism, eds. Daphne J. Fairbairn, Wolf U. Blanckenhorn, and Tamás Székely. Oxford: Oxford University Press, 16–26. Little, Andrew T, and Thomas Zeitzoff. 2017. “A Bargaining Theory of Conflict with Evolutionary Preferences.” International Organization 71(Summer 2017): 523–57. Loewenstein, George, Leigh Thompson, and Max H. Bazerman. 1989. “Social Utility and Decision Makig in Interpersonal Contexts.” Journal of Personality and Social Psychology 57(3): 426–41. De Luca, Giacomo, and Marijke Verpoorten. 2015. “Civil War and Political Participation : Evidence from Uganda.” Economic Development and Cultural Change 64(1). Lyall, Jason, Yuki Shiraito, and Kosuke Imai. 2015. “Coethnic Bias and Wartime Informing.” Journal of Politics 77(3): 833–48. Maoz, Ifat, and Clark McCauley. 2005. “Psychological Correlates of Support for Compromise : A Polling Study of Jewish-Israeli Attitudes toward Solutions to the Israeli-Palestinian Conflict.” Political Psychology 26(5): 791–807. McDermott, Rose. 2013. “The Ten Commandments of Experiments.” Political Science 46(03): 605–10.

61 McDermott, Rose, and Christian Davenport. 2017. “Toward an Evolutionary Theory of International Relations.” Oxford Research Encyclopedia of Politics (January): 1–16. McDermott, Rose, James H. Fowler, and Oleg Smirnov. 2008. “On the Evolutionary Origin of Prospect Theory Preferences.” The Journal of Politics (2): 335. Mironova, Vera, and Sam Whitt. 2016a. “Social Norms after Conflict Exposure and Victimization by Violence: Experimental Evidence from Kosovo.” British Journal of Political Science: 1–17. ———. 2016b. “The Evolution of Prosociality and Parochialism after Violence.” Journal of Peace Research 53(5): 648–64. Mousa, Salma. 2018. The Causal Effect of Sectarian Violence: Political Preferences, Religiosity, and Iraq’s Sunni – Shi’a Divide. Nesse, Randolph M. 2005. “Natural Selection and the Regulation of Defenses. A Signal Detection Analysis of the Smoke Detector Principle.” Evolution and Human Behavior 26(1): 88–105. Neuberg, Steven L., and Catherine A. Cottrell. 2006. “Evolutionary Bases of Prejudices.” In Evolution and Social Psychology, eds. Mark Schaller, Jeffery A. Simpson, and Douglas T. Kenrick. Psychology Press, 163–89. Neuberg, Steven L., and Peter DeScioli. 2015. “Prejudices: Managing Perceived Threat to Group Life.” In Handbook of Evolutionary Psychology Volume II: Integrations, ed. David M. Buss. Hoboken: John Wiley & Sons, Inc., 704–21. Neuberg, Steven L., Douglas T. Kenrick, and Mark Schaller. 2011. “Human Threat Management Systems: Self-Protection and Disease Avoidance.” Neuroscience and Biobehavioral Reviews 35(4): 1042–51. Neuberg, Steven L., and Mark Schaller. 2016. “An Evolutionary Threat- Management Approach to Prejudices.” Current Opinion in Psychology 7: 1–5. Neuberg, Steven L, and Oliver Sng. 2013. “Perceiver Effects - A Life History of Social Peception: Stereotyping at the Intersections of Age, Sex, Ecology (and Race).” Social Cognition 31(6): 696–712. von Neumann, John, and Oskar Morgenstern. 1944. Theory of Games and Economic Behavior. Princeton: Princeton University Press. Nowak, Martin A., and Karl Sigmund. 2005. “Evolution of Indirect Reciprocity.” Nature 437(7063): 1291–98. Olson, Mancur. 1983. The Rise and Decline of Nations. New Haven and London: Yale University Press. Pettersson, Therése, and Kristine Eck. 2018. “Organized Violence, 1989–2017.” Journal of Peace Research 55(4): 535–47. Pettersson, Therése, Stina Högbladh, and Magnus Öberg. 2019. “Organized Violence, 1989–2018 and Peace Agreements.” Journal of Peace Research 56(4): 589–603. Pinker, Steven. 2002. The Blank Slate: The Modern Denial of Human Nature. New York: Penguin. ———. 2011. The Better Angels of Our Nature. New York: Penguin. Powell, Robert. 2006. “War as a Commitment Problem.” International Organization 60(1).

62 Powell, Steve et al. 2003. “Posttraumatic Growth After War : A Study with Former Refugees and Displaced People in Sarajevo.” Journal of Clinical Psychology 59(1): 71–83. Pruitt, Dean G., and Sung Hee Kim. 2004. Social Conflict: Escalation, Stalemate and Settlement. Boston: McGraw-Hill. Rai, Tage Shakti, and Alan Page Fiske. 2011. “Moral Psychology Is Relationship Regulation: Moral Motives for Unity, Hierarchy, Equality, and Proportionality.” Psychological Review 118(1): 57–75. Rand, David G., and Martin A. Nowak. 2013. “Human Cooperation.” Trends in Cognitive Sciences 17(8): 413–25. Reimers, Luise, and Esther K. Diekhof. 2015. “Testosterone Is Associated with Cooperation during Intergroup Competition by Enhancing Parochial Altruism.” Frontiers in Neuroscience 9(MAY): 1–9. Rohner, Dominic, Mathias Thoenig, and Fabrizio Zilibotti. 2013. “Seeds of Distrust : Conflict in Uganda.” Journal of Economic Growth 18(August): 217– 52. Romano, Angelo, Daniel Balliet, Toshio Yamagishi, and James H Liu. 2017. “Parochial Trust and Cooperation across 17 Societies.” Proceedings of the National Academy of Sciences 114(48): 12702–7. Rusch, Hannes. 2014. “The Evolutionary Interplay of Intergroup Conflict and Altruism in Humans: A Review of Parochial Altruism Theory and Prospects for Its Extension.” Proceedings of the Royal Society B: Biological Sciences 281: 1–9. Sambanis, Nicholas. 2000. “Partition as a Solution to Ethnic War: An Empirical Critique of the Theoretical Literature.” World Politics 52(July): 437–83. Samuelson, Paul A. 1938. “A Note on the Pure Theory of Consumer’ s Behaviour.” Economica 5(17): 61–71. ———. 1948. “Consumption Theory in Terms of Revealed Preference.” Economica 15(60): 243–53. Satyanath, Shanker, Nico Voigtländer, and Hans-Joachim Voth. 2017. “Bowling for Fascism : Social Capital and the Rise of the Nazi Party.” Journal of Political Economy 125(2): 319–61. Schaller, Mark. 2011. “The Behavioural Immune System and the Psychology of Human Sociality.” Philosophical Transactions of the Royal Society B: Biological Sciences 366(1583): 3418–26. Schaller, Mark, and A. M. N. D. Abeysinghe. 2006. “Geographical Frame of Reference and Dangerous Intergroup Attitudes : A Double-Minority Study in Sri Lanka.” Political Psychology 27(4): 615–31. Schaller, Mark, Douglas T. Kenrick, Rebecca Neel, and Steven L. Neuberg. 2017. “Evolution and Human Motivation: A Fundamental Motives Framework.” Social and Personality Psychology Compass 11(6): 1–15. Schaller, Mark, and Steven L Neuberg. 2012. “Danger, Disease, and the Nature of Prejudice(S).” In Advances in Experimental Social Psychology, eds. James M Olson and Mark P Zanna. Burlington: Academic Press, 1–54. Schaller, Mark, and Justin H. Park. 2011. “The Behavioral Immune System (and Why It Matters).” Current Directions in Psychological Science 20(2): 99–103.

63 Schaller, Mark, Justin H. Park, and Annette Mueller. 2003. “Fear of the Dark: Interactive Effects of Beliefs about Danger and Ambient Darkness on Ethnic Stereotypes.” Personality and Social Psychology Bulletin 29(5): 637–49. Schaub, Max. 2017. “Threat and Parochialism in Intergroup Relations: Lab-in-the- Field Evidence from Rural Georgia.” Proceedings of the Royal Society B: Biological Sciences 284(1865). Segerstråle, Ullica. 2006. “Evolutionary Explanation: Between Science and Values.” In Missing the Revolution: Darwinism for Social Scientists, ed. Jerome H. Barkow. New York: Oxford University Press, 121–47. Sell, Aaron N. 2011. “The Recalibrational Theory and Violent Anger.” Aggression and Violent Behavior 16(5): 381–89. Sheikh, Hammad, Jeremy Ginges, and Scott Atran. 2013. “Sacred Values in the Israeli-Palestinian Conflict: Resistance to Social Influence, Temporal Discounting, and Exit Strategies.” Annals of the New York Academy of Sciences 1299: 11–24. Sherif, Muzafer et al. 2010. The Robber’s Cave Experiment: Intergroup Conflict and Cooperation. 3rd Editio. Middletown: Wesleyan University Press. Sidanius, Jim, and Robert Kurzban. 2003. “Evolutiuonary Approaches to Political Psychology.” In Oxford Handbook of Political Psychology, eds. David O. Sears, Leonie Huddy, and Robert Jervis. Oxford: Oxford University Press, 146–81. Skaperdas, Stergios. 2006. “Bargaining versus Fighting.” Defence and Peace Economics 17(6). Staub, Ervin, and Johanna R. Vollhardt. 2008. “Altruism Born of Suffering: The Roots of Caring and Helping after Victimization and Other Trauma.” The American journal of orthopsychiatry 78(3): 267–80. Stein, Janice G. 2013. “Threat Perception in International Relations.” In The Oxford Handbook of Political Psychology, , 364–94. Stenner, Karen. 2005. The Authoritarian Dynamic. Cambridge: Cambridge University Press. Stephan, Walter G., and Cookie White Stephan. 2000. “An Integrated Threat Theory of Prejudice.” In Reducing Prejudice and Discrimination, ed. Stewart Oskamp. Mahwah, NJ: Lawrence Erlbaum Associates, Inc, 23–45. Stephens, David W. 1981. “The Logic of Risk-Sensitive Foraging Preferences.” Animal Behaviour 29(2): 628–29. Stephens, David W., and John R. Krebs. 1987. Foraging Theory. Princeton: Princeton University Press. Stroessner, Steven J., Abigail A. Scholer, David M. Marx, and Bradley M. Weisz. 2015. “When Threat Matters: Self-Regulation, Threat Salience, and Stereotyping.” Journal of Experimental Social Psychology 59: 77–89. Sunstein, Cass R. 2014. Why Nudge?: The Politics of Libertarian Paternalism. New Haven and London: Yale University Press. Tajfel, Henri, and John Turner. 2001. “An Integrative Theory of Intergroup Conflict.” In Key Readings in Social Psychology. Intergroup Relations: Essential Readings, eds. M. A. Hogg and D. Abrams. New York: Psychology Press, 94–101.

64 Teddlie, Charles, and Fen Yu. 2007. “Mixed Methods Sampling : A Typology.” Journal of Mixed Methods Research 1(1): 77–100. Tedeschi, Richard G., and Lawrence G. Calhoun. 2004. “Posttraumatic Growth: Conceptual Foundations and Empirical Evidence.” Psychological Inquiry 15(1): 1–18. Tetlock, Philip E. et al. 2000. “The Psychology of the Unthinkable: Taboo Trade- Offs, Forbidden Base Rates, and Heretical Counterfactuals.” Journal of Personality and Social Psychology 78(5): 853–70. Thornhill, Randy, and Craig T. Palmer. 2000. A Natural History of Rape. Cambridge, MA: The MIT Press. Tilly, Charles. 1992. Coercion, Capital, and European States AD 990-1992. Cambridge, MA & Oxford, UK: Blackwell Publishing. Tir, Jaroslav, and Shane P. Singh. 2015. “Get off My Lawn: Territorial Civil Wars and Subsequent Social Intolerance in the Public.” Journal of Peace Research 52(4): 478–91. Tooby, John, and Leda Cosmides. 2008. “The Evolutionary Psychology of the Emotions and Their Relationship to Internal Regulatory Variables.” In Handbook of Emotions, eds. Michael Lewis, Jeanette M. Haviland-Jones, and Lisa Feldman Barret. New York: The Guilford Press. ———. 2010. “Groups in Mind: The Coalitional Roots of War and Morality.” In Human Morality and Sociality: Evolutionary and Comparative Perspectives, ed. Henrik Høgh-Oelsen. New York: Palgrave MacMillan, 191–234. ———. 2015. “The Theoretical Foundations of Evolutionary Psychology.” In Handbook of Evolutionary Psychology Volume I: Foundations, ed. David M. Buss. Hoboken: John Wiley & Sons, Inc., 3–87. Tversky, Amos, and Daniel Kahneman. 1992. “Advances in Prospect Theory: Cumulative Representation of Uncertainty.” Journal of Risk and Uncertainty 5: 297–323. UCDP Bulletin. 2019. Islamic State - Defeated? 100 Days after the Declared Victory. Vollhardt, Johanna R., and Ervin Staub. 2011. “Inclusive Altruism Born of Suffering: The Relationship between Adversity and Prosocial Attitudes and Behavior toward Disadvantaged Outgroups.” American Journal of Orthopsychiatry 81(3): 307–15. Voors, Maarten J et al. 2012. “Violent Conflict and Behavior : A Field Experiment in Burundi.” The American Economic Review 102(2): 941–64. Voors, Maarten J, and Erwin H Bulte. 2014. “Conflict and the Evolution of Institutions : Unbundling Institutions at the Local Level in Burundi.” Journal of Peace Research 51(4): 455–69. Whitt, Sam. 2007. “The Dictator Fairness and Ethnicity Game , in Postwar Bosnia.” American Journal of Political Science 51(3): 655–68. ———. 2014. “Social Norms in the Aftermath of Ethnic Violence Ethnicity and Fairness in Non-Costly Decision Making.” Journal of Conflict Resolution 58(1): 93–119. Whitt, Sam, and Rick K Wilson. 2007. “The Dictator Game, Fairness and Ethnicity in Postwar Bosnia.” American Journal of Political Science 51(3): 655–68. Wilson, Edward O. 1975. Sociobiology. Cambridge: Harvard University Press.

65 ———. 2012. The Social Conquest of Earth. New York: Liveright Publishing Corporation. Wood, Elisabeth J. 2006. “The Ethical Challenges of Field Research in Conflict Zones.” Qualitative Sociology 29(3): 373–86. ———. 2008. “The Social Processes of Civil War: The Wartime Transformation of Social Networks.” Annual Review of Political Science 11(1): 539–61. Yamagishi, Toshio, and Toko Kiyonari. 2000. “The Group as the Container of Generalized Reciprocity.” Social Psychology Quarterly 63(2): 116–32. Yuki, Masaki, and Kunihiro Yokota. 2009. “The Primal Warrior: Outgroup Threat Priming Enhances Intergroup Discrimination in Men but Not Women.” Journal of Experimental Social Psychology 45(1): 271–74. Zeitzoff, Thomas. 2014. “Anger , Exposure to Violence , and Intragroup Conflict : A ‘ Lab in the Field ’ Experiment in Southern Israel.” Political Psychology 35(3). ———. 2018. “Anger, Legacies of Violence, and Group Conflict: An Experiment in Post-Riot Acre, Israel.” Conflict Management and Peace Science 35(4): 402– 23.