In the Eyes of the Beholder: Border enforcement, suspect travellers and trafficking victims

Over the past decade, the border and border policing has figured as central to identifying and responding to trafficking. This article draws on original research into immigration officers’ decision-making — both at the border and within the nation — to identify the persistent preoccupation with suspect travellers. Examining research in Australia and Thailand that spans seven years, the article brings together research that demonstrates the predominance of the binary category of victim of trafficking/unlawful migrant worker and highlights the ambiguity of daily decision-making processes that categorise women who come into contact with immigration authorities. While the policy rhetoric is based on categories and risk profiles for identifying suspected victims of trafficking or those deemed at risk, we contribute to the growing body of work that has highlighted the presence of gendered and racialised stereotypes in immigration decision-making and consider implications this may have on women’s mobility across and within borders.


Introduction
While traditionally border control has been conceived in relation to the physical act of crossing, it is now recognised as mobile and performative; it is enforced and (re)asserted both within and outside border lines. 1 Yet, in international law the border remains a key site for the negotiation and assertion of state rights, that is, the right to determine who may enter a nation state and upon what conditions, even within the context of specific regional arrangements (for example, the European Union).While states increasingly require significant screening and approval of travellers prior to embarking on their journey (resulting in much border control occurring ahead of mobility), state agencies working at the border continue to exercise considerable power in the process of determining the status of an individual, a power that is exercised within the context of daily routine decisions that attract limited scrutiny and/or accountability.Attending to the border and the performance of the border regime, we must attend to these daily practices to better understand the everyday implications of border enforcement for the human rights of migrants.This article draws on two research projects that included interviews with immigration officials and specifically focusses on how officers operationalise anti-trafficking information in their daily decision-making.The first project was conducted in 2006 and 2007.It examined the implementation of antitrafficking policy efforts in Australia and Thailand.The second project, conducted in 2012 with immigration officers in Australian airports, examined decision-making on the arrival of suspect travellers.While seemingly disparate projects in focus and timelines, it is the resonance of key findings across 1 L Weber, '"It sounds like they shouldn't be here": Immigration checks on the streets of Sydney ', Policing & Society, vol. 21, no. 4, 2011, pp. 456-467; L Weber and S Pickering, Globalisation and Borders: Death at the global frontier, Palgrave Macmillan, London, 2011; L Weber, 'The Shifting Frontiers of Migration Control' in S Pickering and L Weber (eds.)Borders, Mobility & Technologies of Control, Dordrecht, Springer, 2006; N Wonders, 'Global Flows, Semi-Permeable Borders and New Channels of Inequality' in S Pickering and L Weber (eds.)Borders, Mobility & Technologies of Control, Dordrecht, Springer, 2006. DOI: 10.14197/atr.20121323these two projects that is the focus here -in particular, the potency of the discretionary decision-making made in the assertion of the border regime and the importance of attending to the consequences of these decisions whether they are assessments for the potential for victimisation (i.e.trafficking, as per the first project) or the intention to breach one's visa conditions that restrict or prohibit work (as per the second project).
We draw attention to the active use of simplistic, opportunistic stereotypes in decision-making processes, within the context of the administrative demands of implementing a mass migration surveillance system.By focussing on immigration officers' reflections on the use of stereotypes in discretionary decisionmaking, we hope to highlight the active processes of racialisation that occur in migration as well as counter the notion that stereotypes are passively reproduced constructs.We consider the consequences of these practices in relation to agency, illegality, victimisation and the upholding of human rights.In doing so, the article examines how the concern about trafficking has consistently overlapped with the sorting of nontrafficked migrants, and to interrogate the highly subjective decision-making practices that are part of Australia's national efforts to counter human trafficking.

Methodology
The two research projects that provide the foundation for this discussion involved extensive semi-structured interviews with immigration officials, in addition to other key stakeholders relevant to each project.The two projects together enable an examination of the border performance at the border and within the nation.The first project, conducted in 2006 and 2007, involved semi-structured interviews (n=50) with police and immigration authorities, international and local nongovernmental organisations and victims in Australia and Thailand.The second project examines travellers' entry at the airport and draws on 2012 semi-structured interviews (n=15) with Australian immigration officers at two major international airports in Australia.With the permission of the Australian Department of Immigration and Citizenship (DIAC), two airport sites were visited for extended periods over a nine-month duration.These visits included observation of all parts of the immigration process -(1) from the management of Advanced Passenger Information while planes were still en route to Australia, (2) to the identification of risky travellers at 'the line' or once people disembark an aircraft, (3) from observing initial conversations with those identified to ascertain whether there is data error that has erroneously flagged the person as being of concern, (4) through to formal interviews.Immigration officials were asked to describe all parts of how they came to identify risky travellers ('passenger of concern') with a specific focus on gender and were specifically asked to talk about what made for an easy or difficult 'risky' traveller to process.Both projects involved interviewing immigration officials as part of the broader project within which they were conducted, and questions pertained to immigration decision-making practices.These projects were not undertaken with the intention of being analysed together; rather, it is the parallel findings from the independent analysis of each that has sparked our effort to bring together data from both projects. 2We emphasise that this article brings analysis undertaken and reported elsewhere, and the presentation of this analysis and the data is indicative rather than exhaustive, reflecting both the confines of space and the intention of this piece as commentary.The discussion that follows below focusses on the two projects independently in order to examine decision-making in different contexts: at the border and within the nation.As the discussion progresses, however, we bring to the fore our concerns regarding the implications of these findings.

At the Point of Entry: Identifying potential victims and potential offenders
The use of trafficking in persons as a specific and unique issue connected to wider issues of illegal cross-border activities has been adopted by state agencies involved in border control.This means that immigration officers are charged with assessing, on entry, both the potential of someone to become a victim of trafficking as well as assessing the likelihood that they may intend to work illegally (i.e. they have entered on a tourist or visitor visa without work rights but intend to work while in the country). 3Even those who attempt to cross the border legitimately (i.e. with a valid visa) are screened as to whether they present with suspect characteristics, travel patterns, behaviour or background.
We have found that dominant stereotypes about sex work, trafficking, and victims of trafficking play a critical role in the process of determining potential victims and potential offenders at the border crossing.This finding is consistent with Weber's earlier findings in relation to immigration officers' decisions to detain asylum seekers in the UK. 4 Such stereotypes are relied upon in an administrative environment where officers are tasked with predicting travellers' vulnerability (to trafficking) and intentionality (of working illegally).The working cultures among immigration staff interviewed at Australian airports in 2012 reflected those working on compliance within the Australian and Thai nations in 2006, where decisions were based on administrative dichotomies (i.e.allow entry or reject entry), where the emphasis was focussed less on the vulnerability or protection of the individual (as the result is often turnaround), and more on minimising risk to the state and avoiding the difficulties of detecting, investigating and prosecuting trafficking and In this article, we use the term agency to refer to the ability to act in a given context.See L M Ahearn, 'Agency', Journal of Linguistic Anthropology, vol. 9, no. 1-2, 1999, pp. 12-15.victimisation post-entry.This results in individual officers possessing considerable discretionary power and often relying upon subjective assessments to make determinations.This was evidenced in discussions with immigration officers at two major international airports in Australia regarding how potential victims and potential offenders might be identified.Overwhelmingly (n=14/15 or 93%), interviewees identified sex workers as the main example of the 'problem women traveller'.Unlike men, women suspected of coming to Australia to be involved in sex work represented two risks: risks of victimisation (trafficking) and/or the risk of working illegally (i.e.breaching tourist visa conditions).For the airport immigration officials interviewed, identifying potential trafficking victims or potential (illegal) sex workers ultimately involved scrutinising women's agency and sexuality at the border.As detailed below, this involved scrutinising women's luggage for sexy clothing and judging if women were perceived to be sufficiently involved in arranging their travel and have knowledge of their intended destination.This assessment was often further complicated by the legalisation and decriminalisation of sex work in the two jurisdictions studied, and the ways immigration officers understood the legality and morality of sex work as work.

Agency in Border Crossing
Immigration officers' explanations of their assessment processes demonstrated attempts to judge 'appropriate' expressions of women's agency, 5 particularly the extent of agency in one's migration, as critical to determining whether they were future sex workers or future victims of trafficking.A lack of curiosity or knowledge about their travel or who were ill-informed of their plans were suspected of being potential trafficking victims, as one officer explained in relation to women who enter Australia on student visas: DOI: 10.14197/atr.20121323 [T]hey could be traffick[ed]… into the sex industry… 'cause I guess I always think… as a human being you are curious.[You would expect a person to ask]… a few questions, [such as] which school am I going to [and] you would be excited about it….[T]he fact that they are not told anything at all, not even the name of a course… like nothing at all, they have no idea….
[T]he fact that they know absolutely nothing that makes me think that they really have no idea or no active decision making.(Australian airport immigration officer, 2012) Evident across both research projects was the precarious balance in determining agency, where too little is an indicator of potential victimisation, and too much agency is an indicator of potentially working in breach of visa conditions: [W]e search the phones as part of evidence.…Wehave seen sort of like sexy type of messages on them, which leads us to believe that they are in the sex industry and if that was the case, there would normally be, we would think it was by choice.(Australian airport immigration officer, 2012) Agency was assessed according to travel organisation (where organisation by a third party was interpreted as evidence of possible trafficking and independence in organising travel was likely to raise suspicion of involvement in sex work) and plans upon arrival (women perceived to be ill-informed of their plans or who articulated open-ended plans were suspected as vulnerable).While it could be argued that many travellers have travel arranged by third parties (travel agents, fellow travellers) and many may arrive with open-ended plans, these characteristics within the context of particular risk profiles (gender, race and visa) were deemed suspicious and warranted further investigation.

Sexuality at the Border
Often risk assessments relied on what women were wearing and/or whether the type of clothing in a woman's luggage was perceived to be in accord with the visa on which she was travelling.Immigration officials (n=8/15) noted that via referral from another agency that had searched the luggage of a traveller, or following their own request to search the luggage of a suspect traveller, the inclusion of 'sexy' clothing (e.g.underwear, lingerie) advanced their questioning of a woman as a potential victim of trafficking or unauthorised sex worker: But we have to find evidence….[so when] we are doing a baggage search [the question is]… what are their motives.If you're coming here for a holiday, why do you bring some sexy lingerie and so many, like, the sex worker?…Why do you bring those items?(Australian airport immigration officer, 2012) However, assessments did vary depending on individual officers' attitudes towards sex work (as work), knowledge about trafficking and attitudes about profiling.A few interviewees recognised the sex industry as a work sector, and the need to be critical of risk profiles based on stereotypical assumptions of sex work as illegal or immoral: Customs sometimes will say, 'But they are sex workers,' and I am like, 'I don't really care how they make their dollar, all I am concerned about do they have work rights, if they have work rights.That is how she wants to make her money, power to her and she probably, you know, makes more money than you and I combined.So like, you know, who is the idiot here, us or her [laughs]?(Australian airport immigration officer, 2012) The officer quoted above was one of only two immigration officers interviewed in 2012 that challenged moralising discourses around the desirability or otherwise of sex work as work (although one of the two still reported sex workers as a category of concern).
For the majority of airport immigration officers, identifying potential sex workers was both a highly gendered process (in all interviews references to sex workers were references to women) and a racialised one.
Research on both sex trafficking and migrant sex workers in Australia has challenged public assumptions that large numbers of Asian women are routinely trafficked into Australia for the purpose of prostitution.6However, immigration judgments about women travelling with sexy underwear suggests that race and gender are key considerations when predicting victimisation or the intention to work illegally, with Asian women forming the bulk of women suspected of entering Australia to work in the sex industry.The connection between sexuality and risk was evident not only in discussing women's risk of victimisation but also in explanations regarding the circumstances that warrant suspicion and therefore further inquiry: If we are looking at possible impostors: [a profile such as] a Chinese born person travelling on a Hong Kong passport, they are in an age group of maybe 20 to 40, first time arrival to Australia, they have got a recently issued visa and they are maybe coming out of either Hong Kong or another… port, that tends to fit all our boxes [and we will]… want to have a look at them….In most cases they are genuine, but it's worth asking those questions as to what they intend to do.(Australian airport immigration officer, 2012) The suspect-traveller approach operationalised in Australian airports can be connected to broader perceptions that sex workers are a 'problem category' of migrant, likely to disrupt order within the nation through breaching visa prohibitions against work or by being a potential victim of trafficking.We examine the continuity of these concerns with practices within national borders, drawing on interviews conducted with antitrafficking stakeholders in Thailand and Australia in 2006.

At the Point of Enforcement: Identifying potential victims and potential offenders
Trafficking is difficult to detect and prove within the destination country.This is despite increasing efforts around compliance and enforcement of migration status and work rights undertaken by a range of agencies. 7The precarious legal status of victims -as non-citizens who may be working unlawfully -translates into a complex situation where a person may be at once an unlawful non-citizen and a victim of criminal exploitation.Immigration control is increasingly occurring within the community through an expansion of border control in various forms of internal compliance and enforcement. 8  Immigration officers play a critical role in identifying potential victims of trafficking, and it is in the context of conducting immigration compliance raids that initial contact between potential victims and authorities is often made.As an Australian law enforcement officer explained:  ', Women's Studies International Forum, 32(4), 2009, pp. 251-260. 8 L Weber, '"It sounds like they shouldn't be here": Immigration checks on the streets of Sydney ', Policing & Society, vol. 21, no. 4, 2011, pp. 456-467; K Aas, '"Crimmigrant" bodies and bona fide travelers: Surveillance, citizenship and global governance ', Theoretical Criminology, vol. 15, no. 3, 2011, pp. 331-346.that all the investigators there have and they're aware of indicators of trafficking.(Australian law enforcement officer, Australia, 2006) Thus immigration officers enforcing the border regime within the nation play a critical role in the identification of potential victims of trafficking.While authorities made reference to a list of characteristics/factors that give rise to suspicion of trafficking (i.e. an internal departmental or authoritydeveloped checklist which was not publicly available in Australia or Thailand), the interpretation of the criteria for identification is also important.In the assessment of suspected trafficking cases, as a support worker identified, authorities tend to rely upon how women behave in order to interpret the situation: I think that [they] are a victim [of]… trafficking… but the authorit[ies] will not recognise [this], they think that the women lie and some women don't cry and don't seem vulnerable enough to be victim… [When I reported a case to an authority, he] said [to me] 'I don't feel she's a victim she didn't cry'… and then 'oh she has a mobile phone'.(Thai NGO, Thailand, 2006) Echoing the sentiments of airport immigration officers at international Australian airports in 2012, the assumptions held by immigration officers in Australia and Thailand are a critical component of decision-making contexts, particularly in relation to race/ethnicity and in relation to sex workers.Gendered narratives of 'real' victimisation were evident in the way interviewees explained the challenges in identifying potential victims. 9The participants in this project indicated that those who are more likely to be identified as potential victims are those perceived by authorities to be closest to the 'ideal' 9 N Christie, 'The ideal victim' in E Fattah (ed.), From Crime Policy to Victim Policy: Reorienting the justice system, St. Martin's Press, New York, 1986.image of innocent and passive victims, subject to extremely exploitative conditions. 10This was evident in descriptions of women who are not identified as potential victims; like suspect travellers, they are the most proactive, independent, experienced sex workers: Most of the Thai women going out are fairly organised, they know what they're getting into… they are party to human smuggling, they are party to the fact that they'll be using false documents, that they're having to pay off middle men in order to get into Japan, they are party to that.They may not be party to the fact that once they've got to Japan that they're ending up being pushed around by thugs….[T]he Thai embassy… feels that so much pressure is being put on them to treat these women as victims of trafficking, whereas they feel that the majority are completely aware of what they're getting into….some of them have changed their passport several times and the Thai embassy is saying you know, 'look we realise that we have to be looking out for victims but with all due respect the majority coming through are pretty hardcore'.(Government-funded organisation staff, Thailand, 2006) In addition to being identified as a 'knowing' or 'willing' sex worker, the status as non-citizen also has an impact on assumptions about 'real' victims.As one police officer explained, the line between potential victim and illegal noncitizen was often perceived to be very unclear: [I]t comes back to… are they a victim or [not]….and that's where the water's really muddy… because they're sort of in the middle of both worlds… Most of them only really become a victim when things go wrong for them.You know there are very few people who… go to 10 See chapter two, M Segrave, S Milivojevic and S Pickering (2009), op.cit.DOI: 10.14197/atr.20121323Australia… not knowing that they're going to be involved in the sex trade [and]… it's only when that money is not forthcoming that they either decide to make a complaint or… report it to the authorities.(Australian law enforcement officer, Thailand, 2006) The above statement reveals a consistent finding regarding the circumstances in which illegal migrant sex workers will claim victimisation.Such statements offer some insight into the perception of a 'victim' as an identity rooted in a broader social context, rather than within the context of harm perpetrated against a person.In part, many participants indicated that identifying victims according to the administrative and legal criteria was challenging, as these criteria had little relevance to the circumstances within which women lived, and (most often) did not resonate with their experiences on the ground, for example: It's difficult… it's really confusing… it's confusing because most of the time the victims are well aware and consented to be trafficked I guess….it's very rare that they don't know what's going on…. [So] it's confusing, it's confusing because often the victims are well aware of what's happening.(Thai law enforcement officer, Thailand, 2006) For immigration officers the dilemmas in identifying potential victims of trafficking are related to competing pressures.They are at once required to identify potential victims, while also ensuring they fulfil their obligations in removing illegal noncitizens and upholding the border regime.Yet we know that trafficking offences are often tangled in diverse situations that may involve degrees of criminality and/or victimisation.The findings above illustrate the challenge of identifying what takes place in the decision-making process, while also revealing the subjective assumptions and stereotypes in operation in this context.

Implications
In part, these findings are not surprising; they add to the well-established research literature that individual officers use stereotypes to assess risk in immigration. 11The day-to-day decision-making of border and compliance immigration officers offers a valuable site to consider how national policies and commitments to stop trafficking are challenged, reinforced or performed in practice.These assumptions adopted by immigration officers in the sorting of legitimate and suspect travellers, reveal the dependence on women's behaviour, demeanour and appearance to confirm the feasibility of women's victimisation and/or their intention to breach visa conditions (rather than work conditions or women's accounts of their experiences) and their reasons for travelling and working.As researchers actively undertaking examination of border crossings and the official and unofficial policing of borders, this article is an opportunity to note that this remains unchanged in spite of the significant developments in countertrafficking strategies and discourse internationally.Both projects highlight the importance of recognising the connection between the identification of 'genuine' victimisation and the border.Although airport immigration officers are tasked with identifying cases of trafficking, in everyday practice this decision-making is limited to predicting rather than identifying victimisation and intention.In the second research project, immigration officers were wary when suspected victims or illegal workers' behaviour (or 'performance') didn't match the antitrafficking narrative (or script) of the abject victim.These findings have serious rights implications.There is evidence of racial and gender-based discrimination, covered under both the Convention on the Elimination of All Forms of Racial Discrimination (CERD) and the Convention on the Elimination of All Forms of Discrimination against Women (CEDAW).Yet, the rights framework is limited.It requires signatory status in the nation, but also requires motivation and 11 L Weber and L Gelsthorpe (2000), op.cit.DOI: 10.14197/atr.20121323momentum to ensure rights are upheld.This is particularly true in relation to the role of the International Convention on the Protection of the Rights of All Migrant Workers and Members of Their Families, which is largely ignored in countertrafficking discussions and which remains off the agenda as a priority Convention for many nation states, including Australia. 12The projects that form the basis of this discussion are indicative of the ways in which the assertion of individual agency tends to be interpreted as entrepreneurial activity that becomes problematic for the least 'desirable' migrants.The recognition of agency as suspect and potentially 'illegal' is at odds with the commitment to preventing discrimination on the basis of race or gender.The findings presented in the discussion also suggest there are potentially serious implications for certain groups who, as reported by airport immigration officers, are subject to more scrutiny based on gender, sexuality, and race in particular.The topic of identifying sex workers at the border has received little attention in anti-trafficking research 13 but is receiving more attention in online media. 14The extent to which the case can be made that these practices are a breach of specific human rights conventions is yet to be determined, but these projects are indicative of the need to examine this in greater detail through a more thorough review of immigration decision-making processes.

2
For further detail of the projects and findings, see: M Segrave, S Milivojevic and S Pickering, Sex Trafficking: International context and response, Willan, Devon, 2009.DOI: 10.14197/atr.20121323

…
the majority of [victims]… come to the AFP [Australian Federal Police]… via the Department of Immigration… where they are doing a compliance investigation, [or] a compliance raid at a brothel and they'll come across a person and they have documents 7 M Segrave, S Milivojevic and S Pickering (2009), op.cit.; M Segrave, 'Order at the Border: The repatriation of victims of trafficking