University of Birmingham ‘They were going to the beach, acting like tourists, drinking, chasing girls’

The United Nations Stabilization Mission in Haiti (MINUSTAH) has been marred by reports of sexual exploitation and abuse (SEA) perpetrated against local women/girls. However, there is very limited empirical evidence on the community’s perceptions regarding these sexual interactions. Through a mixed-methods approach, this article examines community experiences and perceptions of SEA, with three prominent themes arising: peacekeepers as tourists, peacekeepers as sexual exploiters and abusers, and peacekeepers as ideal partners. Uruguayan (n = 107, 28.1 per cent) and Brazilian personnel (n = 83, 21.8 per cent) were most commonly named in SEA narratives. We explore how these perceptions of MINUSTAH peacekeepers under-mine the purpose and legitimacy of UN peace support operations, and propose strategies to prevent and address peacekeeper-perpetrated SEA.


Introduction
While UN peace support operations (PSO) have proven successful in facilitating longlasting peace (Fortna 2008;Hultman, Kathman, and Shannon 2013, 2014Quinn, Mason, and Gurses 2007), they have also been tarnished by persistent reports of sexual exploitation and abuse (SEA) of vulnerable and innocent civilians (Karim and Beardsley 2016). Since the first reports of the early 1990s, almost all PSO have been associated with sexual misconduct to some degree of magnitude and severity (Lee 2017). Different missions have raised different concerns, and recent academic scholarship has led to some understanding of the factors underlying SEA within the context stability of peacekeeping economies around bases (Jennings and Bøås 2015;Jennings 2015;Rodriguez and Kinne 2019) including the economic power differentials between members of the local population and PSO personnel (Bell, Flynn, and Machain 2018).
The UN has long been aware of sexual misconduct allegations. Following the establishment of the Inter-Agency Standing Committee (IASC) Task Force on Protection from Sexul Exploitation and Abuse in Humanitarian Crises, the Secretary General of the UN, in October 2003, published the bulletin 'Special Measures for Protection from Sexual Exploitation and Abuse,' which detailed what has since become known as the 'zero-tolerance policy' (United Nations Secretariat 2003). While this bulletin defined and prohibited SEA, a more pronounced change in attitude and policy occurred in the wake of the so-called 'Zeid report,' commissioned by the UN Secretary-General to provide a 'Comprehensive Strategy to Eliminate Future Sexual Exploitation and Abuse in United Nations Peacekeeping Operations' (Prince Hussein Zeid 2005). In December 2007, the UN General Assembly adopted two resolutions: one on criminal accountability (Resolution 62/63) (UN General Assembly 2007b) and one on victim support (Resolution 62/214), the 'Comprehensive Strategy on Assistance and Support to Victims of Sexual Exploitation and Abuse by United Nations Personnel and Related Personnel' (UN General Assembly 2007a). The survivorcentred, rights-based approach was further underscored with the 2017 appointment of the first Victims' Rights Advocate in an attempt to ' elevate the voice of the victims' and to 'put their rights and dignity at the forefront' of the UN efforts (UN 2017a). Thus, while the UN has arguably done more than most national or international organizations to put SEA survivors at the core of SEA mitigation measures, it is the dichotomy between the apparent desire on the part of the UN to address SEA perpetrated by PSO personnel on the one hand (Women's International League for Peace & Freedom n.d.), and the evident failure to achieve meaningful justice for survivors on the other hand, that has been critiqued widely and is evident, too, in the case of Haiti (Joseph and Wisner 2019;Freedman 2018).

Gender and PSO
The gendered nature of PSO has long been recognised (Smith and Skjelsbaek 2001;Stiehm 2001;Carey 2001;Mazurana, Raven-Roberts, and Parpart 2005). Since the landmark UN Security Council Resolution 1325 in October 2000, the UN has started gendermainstreaming its mandates, albeit selectively, through the Inter-agency Taskforce on Women, Peace and Security (established by the Inter-Agency Network on Women and Gender Equality [IANWGE]) and subsequent Security Council resolutions (Kreft 2017). Concurrently, scholarship, particularly feminist scholarship, has engaged critically with gendered aspects of conflict and PSO, building theoretical frameworks for the study of military masculinities (Kronsell 2012;Chinkin, Kaldor, and Yadav 2020;Duncanson 2015;Chisholm and Tidy 2017;Henry 2017;Henry 2019) and evaluating PSO policies and practice (Pruitt 2016;Higate 2007;Kronsell and Svedberg 2011;Jennings 2019). While the gendered culture in PSO is no longer controversial (Carreiras 2010), the improved understanding of gendered determinants and outcomes for both the host country and the troop and police contributing countries (TPCC) has, by and large, not been translated into greater effectiveness (Karim 2017). Peacekeepers' discursive construction of the local context within which they operate and specifically of the local female population has compromised the ability to fulfil the mandate to protect (Jennings 2019). While we are theorising about gender (Shepherd 2010), there is still little room for gender when theorizing inequalities as a barrier to peace (Shepherd 2017). In the tradition of feminist scholarship, our research moves away from the ' colonial model of seeing men and not hearing women' (Heathcote 2014: 62), by seeking to understand the experiences of Haitian women living in a UN host community.
Throughout its 13-year history, the United Nations Stabilization Mission in Haiti (MINUSTAH), historically one of the most controversial PSO, was marred by allegations of wrongdoings, both by peacekeepers themselves and the institutions monitoring their practices. For instance, MINUSTAH personnel (uniformed and non-uniformed) have been linked to a variety of human rights abuses, including sexual exploitation and rape (Yearman 2007), as well as offering minors food and small amounts of cash in exchange for sex (BBC News 2006;Kolbe 2015). Often, reports were not investigated thoroughly, and even where allegations were substantiated, little action was taken to deliver justice for victims. Arguably, the most prominent example was the 114 Sri Lankan peacekeepers who were repatriated but never even charged following allegations of a sex ring involving Haitian children (Snyder 2017). This culture of impunity exists in part because UN personnel enjoy functional immunity in the host country, i.e., immunity from prosecution for crimes committed while on duty. Although peacekeepers do not enjoy full immunity, i.e., immunity from prosecution for crimes committed off duty, the UN has generously interpreted the PSO 'line of duty' and also determined that only TPCC can prosecute military members of the mission (UN General Assembly 2011). Thus, uniformed peacekeepers are not subject to the host country's jurisdiction, and victims have little access to justice in their home countries. For their part, few TPCC have been willing to investigate allegations, and the UN has little, if any, power to force TPCC to comply with delivering justice for alleged victims (UN Human Rights Council 2008; Boon 2016; Republic of Haiti 2016).

Purpose of the study
There is limited empirical evidence (Kolbe 2015) on the perceptions of local community members regarding peacekeeperperpetrated SEA. The current research was designed to address this knowledge gap by using Haiti as a case study to analyse how community perceptions and experiences of SEA undermines the purpose and legitimacy of PSO, and by extension the UN. From our thematic analysis, we present three prominent themes of how Haitians experienced and perceived peacekeeper-perpetrated SEA in their communities. Each theme is supported by a variety of illustrative quotes and descriptive statistics for differential patterns, where relevant. This analysis leads to an examination of how the host community's perceptions and experiences of SEA impacts a PSO's ability to successfully fulfill its mandates. The article closes with proposed strategies for preventing and addressing SEA in PSO.

Methods
We used Cognitive Edge's SenseMaker® app (Cognitive Edge 2017) as a narrative capture tool to collect self-interpreted stories from community members about the experiences of women/girls interacting with members of MINUSTAH. From this larger dataset, we present a subset of narratives focusing on SEA.

Location and participant selection
Ten UN bases across seven locations were selected and interviews were conducted within a 30 km radius of each base. Prospective participants were approached in public locations such as market areas, shops, public transportation depots, etc. Diverse participant subgroups (aged 11 and older) were intentionally included: women/girls who had experienced SEA, family members and friends of women/girls who had experienced SEA, and community members more generally.

Survey
SenseMaker® extracts meaning from micronarratives shared by participants on a topic of interest (in this case, interactions between women/girls and MINUSTAH personnel) (Cognitive Edge 2017). Participants were presented with three prompting questions (Appendix) and asked to share a brief narrative based on the story prompt of their choice. Narratives were audio-recorded, and participants then responded to a series of predefined interpretation and multiplechoice questions. The survey, which did not prompt for stories about SEA, was written in English, translated to Haitian Creole, and then independently back-translated to check for accuracy. We pilot tested the survey in Haiti with 54 participants and refined it based on the results and feedback.

Study implementation
Twelve Haitian research assistants from two local partners, Enstiti Travay Sosyal ak Syans Sosyal (ETS) and Komisyon Fanm Viktim pou Viktim (KOFAVIV), collected the data in June to August 2017. The ten ETS research assistants (six female and four male) were undergraduate students in social work, and the two female research assistants from KOFAVIV were volunteers who had experience working with survivors of gender-based violence. All research assistants completed four days of training immediately prior to data collection, covering topics such as research ethics, use of an iPad, a detailed question-by-question review of the survey, data upload, as well as management of adverse events and program referrals. All interviews were conducted in Haitian Creole, and audio recorded stories were later transcribed and translated to English by native Creole speakers.

Definitions
Using the UN definition, 'sexual exploitation' was taken to be ' any actual or attempted abuse of a position of vulnerability, differential power, or trust, for sexual purposes, including but not limited to, threatening or profiting monetarily, socially, or politically from the sexual exploitation of another' (UN 2017b). Sexual abuse was used to refer to 'the actual or threatened physical intrusion of a sexual nature, whether by force or under unequal or coercive conditions' (ibid.). Sexual interactions with a child, defined as under the age of 18, were always considered abusive. For this analysis, any individual referred to as a 'girl,' referenced to be in school, or described as being under the care of parents or other guardians was assumed to be a minor under the age of 18.

Ethics
All interviews were conducted confidentially, and no identifying information was recorded. Informed consent was reviewed in Creole and indicated on the handheld tablet. Since the survey did not ask direct questions about SEA and participants could choose to talk about whatever experience they wanted, the study was deemed to involve minimal risk and written consent was waived. Participants as young as age 11 were included because we knew anecdotally that they were affected by SEA, and we believed it would have been unethical to exclude their voices. Since the survey did not ask about SEA or about sexual interactions (i.e. the girls could talk about whatever experience they wanted) it was deemed acceptable to include adolescents aged 11 and older. Furthermore, the KOFAVIV research assistants were present to provide immediate support to participants if needed, and all participants were offered a referral card for counselling from KOFAVIV as well as legal counsel from Bureau des Avocats Internationaux. No monetary or other compensation was offered. The Queen's University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board approved this study (protocol # 6020398).

Analysis
A database of 1,221 transcribed and translated narratives was searched for the following keywords: sex, rape, sleep, girl, women, pregnant, violence, and violate, giving a sample of 632 stories. Results of an independent cross-check by another team member who had read and screened all 1,221 transcripts added another 78 narratives, giving a total of 710. CK and SB independently read and screened the 710 narratives retaining any story about or mentioning SEA. Stories that had only a single phrase or sentence about SEA, without any further detail that would make it useful for qualitative analysis, were omitted. Screening discrepancies were resolved through consensus, yielding a final sample of 381 narratives for analysis (see Figure 1).
A thematic analysis was conducted according to Braun and Clark (2006), using inductive coding and latent theming. After familiarisation with the data, CK and SB independently coded transcripts line by line in Dedoose 8.2.14 using an approach outlined by Saldaña (2012). Coding inter-rater reliability was measured with a pooled Cohen's kappa (0.82), and codes were organized into conceptual themes. Triangulation between and critical dialogue among CK, SL, and SB was maintained throughout the analysis.
Quantitative analysis was performed using SAS® Studio Release 3.8. Chi-squared tests were conducted, with the exception of tables with cell counts <5, where Fisher's exact test was used. Results with a p-value of less than 0.05 were considered statistically significant. Art. 10, page 6 of 22 A number of study limitations must be considered. Male participants significantly outnumbered female participants, reducing the number of first-person stories from women/girls who had experienced SEA directly. Additionally, the convenience sample was not representative of the population, and therefore the results are not generalizable. Furthermore, on average, SenseMaker narratives are shorter and less detailed than narratives collected with more traditional qualitative research. Finally, the researchers recognise their positionality and note that as non-Haitians and academics, the results are interpreted with their own biases and perceptions. Table 1 provides demographic data for study participants. In summary, participants were predominantly male (71.9 per cent), under the age of 35 (71.7 per cent), had partially or fully completed secondary school (59.6 per cent), and reported that their household income was average (60.6 per cent). Participants were most frequently located in Port Salut (28.1 per cent) and Cité Soleil (19.2 per cent).

Results
Across locations, consistent patterns described community experiences and perceptions of SEA perpetrated by MINUSTAH members. Figure 2 provides a breakdown of stories by base, TPCC, and SEA involving underage children. Of the 381 narratives, 93.2 per cent (n = 355) were about soldiers, in comparison to other roles within MINUSTAH, with Uruguay (n = 107, 28.1 per cent) and Brazil (n = 83, 21.8 per cent) being the most frequently identified TPCC. Overall, three predominant themes were identified: peacekeepers as tourists, peacekeepers as exploiters and abusers, and peacekeepers as desirable partners. Each theme is explored below with a variety of illustrative participant quotes.

Theme 1: Peacekeepers as Tourists
Peacekeepers socializing in the community with local civilians was a common theme throughout the analysis. For instance, there was a pattern of peacekeepers going to the beach for leisure and ten narratives clearly described SEA occurring at beach areas. Four sites in particular were named: Port Salut (50.0 per cent), Cité Soleil (30.0 per cent), Saint-Marc (10.0 per cent), and Cap-Haïtien (10.0 per cent), staffed predominantly by Uruguayan, Brazilian, Nepalese, and Chilean peacekeepers, respectively. Of narratives from Cité Soleil, Saint-Marc, and Cap-Haïtien about SEA on the beach, all described transactional sex, apart from one encounter in Saint-Marc that identified rape in the context of transactional sex. In contrast, in Port Salut, narratives at the beach described rape (60.0 per cent) or consensual sex (40.0 per cent).
Data included first-person narratives from women/girls who described meeting UN peacekeepers at the beach, as was the case in the following narrative.
The MINUSTAH had their base here, they always had time off and time to go to the beach. So, I met with one, we talked, then we became friends, afterward, he came to my home often, then the friendship went further, and after a lot of time we were in love. I was 17 years old, and he had a party for me, and we started a sexual relationship. So, I became pregnant, and then my parents found out. They put him in jail for a month, and he went back to his country. (Single female, aged 18-24 with a primary school education talking about Uruguayan MINUSTAH soldiers in Port Salut) Her report of peacekeepers always having time off to go to the beach is noteworthy as is the statement that the peacekeeper came to her house often. This participant was 17 years old at the time she was sexually abused by a member of MINUSTAH and became pregnant.
There were also wider community perceptions that peacekeepers freely spent their leisure time engaging with members of the host community. For instance, the following male Ongoing close engagement with the members of the host community is highlighted by the participant's description of peacekeepers  The participant indicated that the company had closed the beach to other visitors, allowing MINUSTAH personnel to ' enjoy themselves' with local women on the beach. That this SEA was so well known in the community, that a local company had conspired to facilitate it, and that it had all been documented by a camera likely led to a perception in the community that the UN must also be aware of this behaviour, and either didn't care or was unable to address the issue.

Theme 2: Peacekeepers as Exploiters and Abusers
A second theme highlighted how MINUSTAH personnel initiated exploitative and abusive sexual interactions (n = 64). These patterns, described across seven of the eight UN bases surveyed and involving peacekeepers from a variety of TPCCs, were more widespread than the tactics highlighted above. A majority of these stories (57.8 per cent) identified underage children as victims.
Several women were sexually exploited by MINUSTAH personnel while employed by the UN, as discussed in the following narrative.
We observed the result where MINUSTAH contributes to the women's achievement at the professional level, especially when they get a job. I think in our country, our institutions could either offer a comfortable salary due to these women's qualifications, but when they work directly for the MINUSTAH, their salaries are doubled. We also observed that the MINUSTAH often misbehaved very badly vis-à-vis of the women in various regions of the country. They sexually harass them, which is very wrong, and I think these are negative behaviours. (Single male, aged 18-24 with some post-secondary education living in Cap-Haïtien and talking about Uruguayan MINUSTAH soldiers) Acknowledging the peacekeeping economy in Haiti, which has been previously described (Vahedi, Bartels, and Lee 2019;Kolbe 2015), this participant pointed out that salaries paid by MINUSTAH were double that paid by many other local employers, thus making MINUSTAH jobs financially attractive. However, not only does this employment put women/girls at risk of SEA but it does so under circumstances that diminish mechanisms for recourse, given the informal nature of many of the jobs.
Another participant spoke of the financial instability that contributes to women working for MINUSTAH while also highlighting the power differential between the MINUSTAH employer and women who work as maids or housekeepers. The presence of comparatively wealthy male peacekeepers in Haiti increased the demand for 'women's labour' (i.e., waitressing, domestic labour, sex work) as others have also documented (Jennings and Nikolić-Ristanović 2009). There is no mistaking the power differential, with the participant remarking that maids must ' do as they are told.' There is also likely an element of racism and classism inherent in this power differential (Henry 2013: 127). There were also multiple descriptions of women/girls being waved into UN bases for the purpose of having sex. For instance, one male participant in Hinche reported seeing MINUSTAH personnel waving individuals into the base late at night, stating 'you know what these kinds of gestures mean.' The final pattern noted was around the initiation of SEA with children and adolescents, and often involved luring them with small amounts of money or food. The following female participant clearly articulated how children were easily enticed by MINUSTAH peacekeepers for a few pennies. This participant's description of peacekeepers attracting children by giving them small amounts of money and/or food has been described in other PSO (Alexandra 2014). In fact, earlier research has suggested that some UN personnel believe they are being altruistic when they buy sex because they are helping impoverished women to survive and support their families (Westendorf 2020). This rationalization has also been extended to underage children.
In several cases, sweets (which would likely be a rarity for many Haitian children) were offered by MINUSTAH personnel as a way to initiate conversations and engage with girls in the community.
The way that the environmental conditions are, and because of poverty, MINUSTAH abused our teenage girls. Like they attract them in a sense… some parents who have children but they are unable to take care of them and just leave them in the street so that they may see the child fit, he does not need to know their ages since it fit, only their breasts are a bit large, open a dialogue, speaks to them, gives them chocolate. (Single male, aged 18-24 with some secondary school education talking about Brazilian MINUSTAH soldiers in Cité Soleil) The participant singled out children whose parents could not provide for them. Other research has similarly highlighted how orphans are sometimes strategically targeted by UN personnel because they lack a support network (Alexandra 2014). The participant commented on peacekeepers not caring about the age of girls who were being sexually abused, as long as they had reached puberty.
It was evident that some peacekeepers made attempts to hide their prohibited/illegal King et al: 'They Were Going to the Beach, Acting like Tourists, Drinking, Chasing Girls' Art. 10, page 11 of 22 sexual behaviour -either from community members, from their commanders, or both. For instance, UN personnel applied a variety of tactics to covertly get women/girls into UN bases to engage sexually with them (n = 17). These stories were most often about Cité Soleil (n = 12, 70.6 per cent; p < 0.0001, Fisher's exact test) in comparison to all other bases and implicated MINUSTAH soldiers from Brazil (52.9 per cent), Argentina (17.7 per cent), and Canada (5.9 per cent), as well as unidentified TPCC (17.7 per cent). Narratives about such tactics were shared predominantly by younger participants who were 11-17 years old (17.7 percent), 18-24 years (35.3 per cent), and 24-34 years (35.3 per cent). Additionally, 76.5 per cent of these narratives also described transactions, suggesting that peacekeepers who used covert tactics were likely engaging in transactional sex with women/girls (n = 13, p < 0.0001, Fisher's exact test).
The most commonly used tactic for smuggling women/girls onto UN bases was a bucket-and-pulley system (n = 14, 82.4 per cent), whereby UN personnel physically hoisted women/girls over the base wall. This approach was often described in relation to adolescent girls, who, on average, weigh less, and are therefore easier to lift over the wall; 64.3 percent (n = 9) of stories describing the use of a bucket-and-pulley system identified underage children. Most such stories were told from Cité Soleil (n = 12, 85.7 per cent; p < 0.0001, Fisher's exact test) and were more likely to identify soldiers from Brazil than other TPCC (64.3 per cent, n = 9; p = 0.0006, Fisher's exact test). The following narrative detailed the SEA of adolescent girls by Brazilian soldiers. This participant's description of the soldiers' 'routine at night' suggested it was a common occurrence and implied that it was likely known widely by community members in Cité Soleil. The perceived sexual objectification of girls is evident from the chosen language, 'they used teenage girls.' Some participants were first-hand witnesses to the use of ropes and buckets to get women/girls inside the UN bases. The following participant added that UN personnel were strategic in where the hoist system was set up and took extra precautions such as building hatchways in an attempt to conceal that women/girls were being smuggled into the base.
Because I was eye witnessing when they used to take the ladies into a bucket so they can have sex with them. You know that the MINUSTAH's gate is in front of the public. They have seen that it will be too noticeable, they built a small hatchway on the top, if you could see where they were throwing, and they tie a bucket to a rope, so they can throw it to the ladies, which means that sometimes they may ride four, or five and pass to the above, then after they give them a forty dollars or thirty dollars depending on the quantity. front gate and building a structure so that the bucket and pulley would be more hidden from public view. The covert tactics demonstrate that the peacekeepers were aware that their behaviour was wrong and/or was against UN policies. Women/girls also crawled through holes in the walls to get into UN bases (n = 3). Such stories were told about the UN bases in Hinche and Saint-Marc, with Argentinian (n = 2) and 'Other' soldiers (n = 1) identified in those narratives. In the following story, the young male participant indicated that it was illegal for women/girls to enter the base in this way.
The white soldiers created an opening in their concrete protective walls to allow women to enter the base illegally and offered them money for sex. The girl that I was referring to earlier, her name was XXXX, she was the one who got pregnant. (Single male, aged 11-17 who had completed primary school living in Hinche, did not identify the nationality of soldiers in the story) Despite his young age, the participant knew that women/girls were not permitted to enter the UN bases and that soldiers deliberately tried to cover these illegal actions by knocking holes into solid walls allowing them to smuggle women/girls into the compound.
In two of the three narratives describing holes in the base wall to facilitate SEA, the sex involved underage children. For instance, one participant in Hinche described holes being dug to 'sneak children in to have sexual relations.' He concluded by saying that he did not like that young Haitians were losing their virginity in this way.
The following story illustrates superiors disapproving of such engagement with local women/girls and of this contributing to MINUSTAH personnel resorting to such measures as making holes in the base walls.
Then when the commander-in-chief does not want, they made a hole in the wall, and the girls appear with their buttocks in the holes to have sex. Then after leaving, the agents gave them US$5. (Single male, aged 11-17 with some primary school education living in Saint-Marc and talking about Argentinian MINUSTAH peacekeepers) In this example, the hole in the wall was apparently a strategy used to hide the sexual interactions from the contingent's commander. Again, despite his young age, the participant recognised that UN personnel were hiding their behaviour from superiors, highlighting an awareness among the peacekeepers that they were defying UN policy and making it obvious to the community that there were few ramifications for doing so.

Theme 3: Peacekeepers as Ideal Partners
A third theme illustrated that perceived consensual sexual relations were common (n = 63), across all UN bases, and involved contingents from a variety of countries. In comparison to all other TPCC, stories about consensual relations more often identified Brazilian soldiers (31.8 per cent, X 2 (1, n = 63) = 4.40, p = 0.04). Figure 3 provides a breakdown of these stories by base and TPCC.
It is important to note that although the UN and many scholars consider all sexual interactions between UN PSO personnel and local host community members to be exploitative and/or abusive, we present them here as they were perceived by the participants, which was more consensual in nature, in an effort to be true to the data and to highlight that local perceptions sometimes differ from official UN policies and academic perspectives.
In several first-person narratives, participants described MINUSTAH personnel being in love with them, as was the case with the following woman in Port Salut. This participant did not state that she had reciprocal feelings of love for the peacekeeper and her choice of language, 'I agreed to date him,' could be taken to imply that perhaps she did so reluctantly. The repatriation of peacekeepers who have fathered children with local women has been documented in more detail elsewhere .

I met a MINUSTAH because he used to come to my house where I was selling
In other consensual relationships, women and their families benefited from the financial resources provided by MINUSTAH personnel. In the following example, the MINUSTAH peacekeeper was providing material goods to both the woman and her family.  (2016) -'sex and material expressions of love.' Within this paradigm, male provision was conceptualised as 'love,' with the degree to which the relationship was motivated by financial or status motivations occurring along a continuum. There were also several narratives highlighting the perception that local women desired foreign partners. In some cases, these narratives indicated that the interactions were initiated by Haitian women who were seeking out MINUSTAH personnel. For instance, one male participant reported that the young girls of Cité Soleil ' always like white/foreigners' and frequently approach them in the community. This highlights that some women were active agents who pursued sexual relations with peacekeepers.
Some descriptions of consensual sexual relations also shed light on the ways in which peacekeepers incorporated dating into their daily lives and how they worked around UN regulations (such as not being out at night) in order to maintain relationships with local women/girls.
Every day, as soon as they have time, depending on what time they take their break, you see them walking around with the girl, they go to the girl's house, they are very comfortable, there's no problems, they lay down and then sleep. They sleep during the day because they know they are not supposed to spend the night out, they spend the night in their base. However, during the day, when they have time, they sleep with women, they sleep with young women in the woods. (Married male, aged 25-34 who had completed secondary school education talking about Uruguayan MINUSTAH soldiers in Port Salut) Narratives such as these suggest that some peacekeepers had more open and visible relationships with Haitian women/girls and as long as they worked within the constraints placed on them (e.g., restricted mobility at night), they were able to maintain these partnerships.

Discussion
This study provides new insights into Haitian community members' perceptions and experiences of peacekeeper-perpetrated SEA across seven locations in Haiti. It adds to our understanding of peacekeeping economies by adding the voices of the local communities to the discourse analyses from within the PSO themselves (Ferreiro 2018;Jennings 2019), thus allowing for a more sophisticated understanding of the complexities and contradictions in everyday lives in peacekeeping economies (Henry 2015a). The research has several notable strengths, including a detailed and nuanced analysis of how exploitative and abusive sexual relations occurred between local women/girls and UN peacekeepers. The survey's intentionally broad story prompts helped gather a wide range of Haitian experiences with MINUSTAH, with less researcher bias. Additionally, a relatively large sample size (n = 381), in comparison to conventional qualitative research studies, and inclusion of diverse participant subgroups including women/girls with firsthand experience of SEA, family members and friends of women/girls with first-hand experience, as well as community members, are also strengths.
Collectively, the qualitative theming identified predominant community views of MINUSTAH personnel. The central perception, that of the MINUSTAH peacekeeper as a tourist in Haiti, overlapped with two other identities as outlined below. The view of peacekeepers as tourists was prevalent and painted the picture of MINUSTAH personnel having lots of time off, able to move around the community freely, enjoying days at the beach, etc. In the second identified perception, MINUSTAH peacekeepers were viewed as exploiters and abusers who took sexual advantage of female employees as well as children in the community who were highly vulnerable due to extreme economic deprivation. Abject poverty was again identified as a key underlying factor contributing to peacekeeper-perpetrated SEA and was consistent with other research in Haiti (Vahedi, Bartels, and Lee 2019;Kolbe 2015;Henry 2015b). This perception also included the measures taken by MINUSTAH peacekeepers and contingents to conceal prohibited, and most often illegal, behaviour, particularly when it came to SEA involving underage girls. In the final perception, MINUSTAH peacekeepers were seen as ideal partners in consensual relationships that were often longer in duration and overlaid with material expressions of love in exchange for sex.
Regardless of the accuracy of each individual shared story, belief that MINUSTAH peacekeepers sexually exploited and abused local women/girls was widespread and deeply rooted. Peacekeepers are deployed to stabilise conflicts, oversee peace agreements, support transition to stable governance, and protect and promote human rights (UN, n.d.). However, in the current study, MINUSTAH personnel were instead viewed as tourists who sexually exploited and abused women/girls, made attempts to cover up the SEA, and also engaged in longer-term more consensual relationships with adult women, which, although not necessarily perceived as SEA by host community members, is nonetheless against UN policy. This contradiction, between PSO mandates on the one hand and host community perceptions on the other, is highly problematic for the reputation and legitimacy of UN PSO. For instance, ongoing SEA within PSO belies UN Security Council resolutions that call for training to prevent SEA and aim to end conflict-related sexual violence (e.g., Resolutions 1820Resolutions , 1960Resolutions , and 2106. According to Kent (2005), a failure of the UN to hold perpetrators responsible may actually worsen SEA against women and children, as the lack of accountability may fuel more pronounced discrimination and violence by UN personnel.
Furthermore, continued peacekeeper-perpetrated SEA destroys credibility in Security Council resolutions aimed at addressing impunity such as Resolution 2272 (UN Security Council 2016). When trust is diminished in this way, it can be deleterious to the success of PSO as demonstrated by Caplan (2014). As argued by Westendorf (2020), SEA is also damaging to the mission and to the UN because it compromises impartiality, which is a basic founding principle of UN PSO. It has also been proposed that SEA in PSO undermines peacebuilding at the individual, family, and community level by compounding existent human rights abuses and poverty (ibid.). Recognition of the direct links between existent poverty and SEA was clear in the perspectives presented above, although a perception by community members that SEA was a human rights abuse was not as evident. Finally, Howard (2019) has maintained that SEA undermines peacekeeping's -and the whole UN system's -ability to employ the power of persuasion, which has proven essential to the success of other PSO, such as the United Nations Transition Assistance Group (UNTAG) in Namibia.
Other academics have made similar but less SEA-focused arguments. Gordon and Young (2017) have demonstrated that local civilian cooperation is dependent on the degree to which local community members perceived peacekeepers positively, and Talentino (2007) found that resentment towards peacekeepers was likely to lead to obstructionism as a result of broken promises and mistrust. It is reasonable to assume, based on the narratives presented here, that SEA would have led to negative perceptions of and resentment towards MINUSTAH. This, in turn, is likely to have had a potentially detrimental impact on the ability of the PSO to successfully fulfill its mandate.
Efforts on behalf of peacekeepers to conceal sexually exploitative and abusive behaviours implies recognition on their part that sex with host community members was against the UN's policies and regulations. It also suggests that messaging around the 'zero tolerance' policy had been heard but raises concerns that training, in its current form, does little to change UN personnel's abuse of power, lack of respect for women, and so on, within the host community. Training that fails to recognise the complexities of behavioral change, including the social context in which identities are developed, will have little effect (Allais 2011). Best practices from behavioural change psychology and the field of knowledge mobilization need to be adapted in order to create the cultural and behavioral shifts necessary to address SEA by PSO personnel.

Conclusion
The UN has itself recognised the potential impact of SEA on its mission, reputation, and legitimacy. In fact, a 2013 UN investigation identified SEA as 'the most significant risk to UN PSO above and beyond other key risks including protection of civilians' (Awori, Lutz, and Thapa 2013). While the UN cannot change the events of the past, in an attempt to salvage the reputation of PSO, to rebuild faith in PSO's mandates, and to do right by Haitian women/girls who experienced SEA, it needs to become more transparent and accountable. As Westendorf (2020) proposes, the 'institutionalizing of impunity' for SEA is a key mechanism through which SEA undermines PSO. We single out this particular mechanism since unlike the SEA itself, which occurred in the past, the UN can, and urgently needs to, bolster its response to sexual misconduct. As a first step, the UN should immediately instate external investigators to handle allegations of SEA and paternity claims, making all existing evidence and data available to an external investigative team, and actively work to remove barriers, both perceived and real, to reporting SEA. While accountability continues to exist in theory rather than in practice, investigations that are independent of the UN are a critical first step in improving transparency. That the UN's role is limited to serving as a liaison between TPCC and host countries is recognised. However, not enough pressure has been exerted on TPCC with concerning track records of accountability, and the UN has not adequately leveraged national militaries' desire to engage in future PSO (for financial and/or other motivations) to improve answerability for previous SEA allegations. To this end, the UN should pressure TPCC to be accountable for acts of SEA by making future participation in PSO contingent on demonstrating that previous and existing allegations of SEA have been taken seriously, and that appropriate action has been taken. Security Resolution 2272 was passed, in part, to support such initiatives by altering the way the UN and TPCC relate, and to enable sanctions for entire contingents and TPCC as opposed to individual perpetrators (UN Security Council 2016).
Under the Status of Forces Agreement, military members of national PSO contingents are granted immunity for crimes committed during official duties in the host country (Allais 2011;Blau 2016;Simm 2013;Freedman 2018), although the UN has historically interpreted the line of duty quite loosely. The current data clearly show that MINUSTAH personnel were perceived as tourists in Haiti, that they organised their time off in order to go to the beach, and to meet local women/girls for sexual purposes, and so on. Such behaviour could never be within official PSO duties, and thus there can be no argument for freedom from prosecution on the basis of functional immunity. Furthermore, the UN has a strict non-fraternization policy and makes great efforts to provide PSO personnel with sports and recreational activities to fill leisure time, in addition to mandating frequent leaves, which allow military, police, and civilian staff to have down time for rest while also allowing them to stay connected with family. At least during the time periods referred to by participants in the current study, there appeared to be very little, if any, implementation and/or enforcement of the non-fraternization policy.
From a preventative perspective, earlier research found strong support that peacekeeper training was useful in reducing SEA cases within the context of PSO in the Democratic Republic of Congo (Neudorfer 2015). However, the current research on Haiti suggests that the UN needs to provide more context-specific training that leads to behaviour change rather than just raised awareness. This revised training must include increasing cultural awareness about existing poverty and other vulnerabilities in the host community to ensure that the inherent power differentials between local women/girls and foreign peacekeepers are appreciated. It is imperative that PSO personnel understand how a lack of education/employment opportunities and local socioeconomic pressures sometimes lead women/girls to exchange sex for food, goods, or money. It is equally crucial to communicate clearly the constraints that such socioeconomic pressures put on women/girls to give consent freely, and it is also vital that PSO personnel understand how transactional sex, particularly when negotiated from a position of relative disempowerment, can be detrimental to a woman/girl's health, including sexually transmitted infections, HIV/AIDS, and unplanned pregnancies. Because SEA is rooted in unfair gender norms, it is critical that such training be delivered with an equity lens that addresses deeply ingrained gender inequalities (Caparini 2019). Furthermore, it is important to consider how earlier experience and training in the TPCC may affect peacekeepers while on mission. For instance, as described by Harig (2015), Brazilian troops were accustomed to intense interactions with civilians, and it may be important to raise awareness about how UN PSO policies differ from those of military and police within the TPCC.
Future research ought to also include the perspectives of peacekeepers themselves in order to understand their perceptions about current SEA training programs and why they may not be as impactful as intended, in addition to collecting data that can inform future SEA prevention and mitigation strategies. Future evaluations to examine the effectiveness of redesigned, contextually specific SEA-prevention training would also be valuable, allowing curricula to be adapted and improved in an iterative process.

Additional File
The additional file for this article can be found as follows: • Appendix. Survey questions. DOI: https://doi.org/10.5334/sta.766.s1

Acknowledgements
We are grateful to all the participants who shared their perspectives about the experiences of women and girls living in communities that host a UN PSO. Our sincere thanks are extended to the three local collaborating partners -Bureau des Avocats Internationaux (BAI), Enstiti Travay Sosyal ak Syans Sosyal (ETS), and the Commission of Women Victims for Victims (KOFAVIV) -and, in particular, to the team of twelve interviewers who travelled around the country to collect data. This work was