Becoming a Smuggler: Migration and Violence at EU External Borders

ABSTRACT Migrants’ involvement in smuggling increases alongside restricted cross-border movement and violent borders, yet this dynamic is usually examined from migrants’ position as clients. In this article, we move away from migrants and smugglers as two separate roles and question migrants’ aspirations to and experiences of resorting to smuggling networks as workers in the context of EU land borders, where direct violence is used daily to fight cross-border crime. By doing so, we move further the examination of fluid relations in smuggling provisions and the way they are intertwined with care and exploitation, as shaped and circumscribed by violent borders. The article illustrates the intersections between border violence and migrants’ active involvement in smuggling by drawing on the case study of an anonymised Border Town and multi-site, multi-author fieldwork from Serbia and Bosnia. By questioning migrants’ experiences of shifting roles from clients to service providers, and by taking into account their work in smuggling provision, we show that, in a situation of protracted vulnerability orchestrated by border violence, state and law enforcement, the categories – “migrant” and “smugglers” – can blur.


Introduction
'I am a smuggler. Without smugglers, no people would reach Europe, not even me', said Mula, while sitting in a train station. Mula, like dozens of other people around him, travelled through the 'Balkan Route' to attempt his journey across the increasingly violent borders with the EU states, such as Croatia. However, Mula had no money, and the only way to pay for clandestine transport was using his own labour as a smuggler, as he called himself.
A blurring of clients and perpetrators in organised crime is not a new phenomenon (Lo Iacono 2014). Same patterns of fluid relations also take place in smuggling, an activity recognised in policy terms as impersonal organised crime; a feature that, however, often lacks in human smuggling that is considered to be organised by social and family kin (Brachet 2018;Kyle and Scarcelli 2009;Maher 2018) and is driven by moral and ideological aspirations to provide help rather than mere greed (Mandić 2017;Sanchez 2015;Vogt 2016). Migrants 1 increasingly opt to work for smugglers in exchange for waived cross-border transport fees (Kyle and Scarcelli 2009;Sanchez 2015); partly due to a perception that this type of work will provide quicker and safer border crossings. However, little is known about the conditions and the extent to which migrants are more likely to engage with smuggling activities as service providers rather than mere clients and how this dynamic evolves.
Existing research suggests that migrants, as clients, are reliant on smugglers due to the lack of legal and safe routes (Antonopoulos and Winterdyk 2006;Maher 2018;Van Liempt and Doomernik 2006) and that their need to seek help from smugglers increases in violent border contexts (Ayalew Mengiste 2018). The diversity in violence against migrants at borders is not a new phenomenon, and has been recorded at EU external borders, within EU territory, as well as beyond it in Niger, Libya, Turkey (Andersson and Keen 2019). However, direct use of violence by EU state authorities has been central to border deterrents along land borders in South East Europe, most recently Croatia (Isakjee et al. 2020), where state authorities commonly use beatings and torture when pushing migrants back to non-EU states (Amnesty International 2019; BVMN 2020; Medecins Sans Frontieres 2017). Violence, in its more structural forms, is equally present in the restriction of movement through the housing of migrants in precarious and overcrowded makeshift camps (Bird et al. 2020;Isakjee et al. 2020).
Authors including Brachet (2018), Maher (2018) Mandić (2017) and Van Liempt and Doomernik (2006) argue that these tough border policies force migrants into criminal activity and create complex negotiations and relations between migrants and smugglers. Yet, the point of departure of these studies commonly makes a distinction between migrants as clients and smugglers as service providers. In this article, we move away from migrants and smugglers as two separate roles and objects of study. Instead, we question migrants' aspirations to resort to smuggling networks as workers and highlight how "migrants" and "smugglers" are not two mutually exclusive categories, but are co-constitutive, and produced by violent borders.
We explore migrants' aspirations to shift their role from client to worker in smuggling organisations in the context of the EU's land borders where direct violence is used to produce and control cross-border crime. We further ask, how the context of the EU's land borders and violent measures shapes relations between smugglers and migrants when cooperating in smuggling provision. By exploring these questions and moving the analysis of fluid relations in human smuggling further, we contribute to the literature approaching human smuggling as a social reality that consists of constant bargaining and negotiations (Baird 2016;Maher 2018;Liempt & Doomernik, 2006). Our examination of a narrow line between being a client and a service provider in smuggling also adds to the literature on human smuggling as an activity intertwined with migrants' need to gain protection (Baird 2016;Kyle and Scarcelli 2009;Vogt 2016), as driven by the context of EU violent crossborder crime-fighting activities. Yet, we show how care and violence are inextricably intertwined for migrants working in smuggling provision, as generated by political conditions along borders rather than smuggling organisations per se (Maher 2018).
This article examines migrant smuggling by focusing on the land border of non-EU states with Croatia as this case study helps us understand the relationship between smuggling and violent border enforcements, and the ways in which they are embedded in broader dynamics of EU external security governance. Countries such as Bosnia-Herzegovina, Serbia and other former Yugoslav states have been one of the major transit routes for migrants to and through the EU. Simultaneously, the countries have been subjected to international intervention in areas of border management and counterorganised crime, especially through the EU Common Foreign and Security Policy, and most recently through the External Dimension of the Area of Freedom, Security and Justice, with the first activities dating back to 1994 with the EU administration of the city of Mostar (Juncos 2013). Yet the intersections between migrant-smuggler fluid roles and border measures remain under-examined in this geographical area, which this paper aims to address.

Methodological Reflections
This article is based on multi-site and multi-author fieldwork carried out between 2015 and 2019 in several locations across Bosnia-Herzegovina and Serbia, with the longest fieldwork trip being eight months. During this time, authors compiled an archive of over 100 interviews and used participant observations. Our main participants were young adult male migrants who are most commonly subjected to violent border measures (Augustová 2021) and become subsequently reliant on cooperation and complex negotiations with smugglers, not only as their clients but also as workers, as we will show.
The interviews were conducted either in English, or in participants' native languages and translated into English by their friends or other camps' inhabitants, due to the unavailability of professional translators. We also interviewed non-governmental organisations, aid workers, social workers, camp management staff, EU officials and local and national authorities. We spoke to our participants about several key themes including: border violence suffered by migrants at EU external borders, arrangements (including smuggling) used for crossing borders without authorisation, as well as -in our interviews with NGO workers and local/national officials -discussions about support networks, accommodation and implementation of local and national policies.
We have anonymised the location our case study focuses on, by calling it Border Town, in order to minimise any identifying details pertaining to our participants and location itself, since some of the practices described here are viewed from a state perspective as illegal. The only information we will provide is that it is located near the Croatian border with one of its non-EU neighbours. We also do not wish to link a specific locality to the practices and networks of smuggling. For the same reason, all names used in the article are pseudonyms, and we avoid pointing to migrants' nationalities or any other identifying features to not to reveal information that could be considered potentially compromising to a specific person. As we discuss below, whilst broader, geographic, and contextual data about the region is relevant, the specific location of the Border Town is not -in the sense that the practices we describe, are commonplace across the 'the Balkan Route' (and indeed, as existing literature suggests, across the globe). Border Town is a part of an 'illegality industry' (Andersson 2014, 13-15), and captures dynamics replicated across similar regional border towns: the proximity of an EU border, migrant destitution and visibility, presence of smugglers, and the broader implications of EU and regional-level anti-smuggling efforts.
Like most transit locations along the Balkan Route, the Border Town lies along the EU's land borders where fight against cross-border crime is based on the daily use of direct police violence, alongside more subtle and structural violence, that gives rise to fluid cooperation between migrants and smugglers. This contrasts with sea borders, where most injuries and deaths come from a hostile environment and drownings, as interconnected with state authorities push-backs and enclosure of migrants (Kyle and Scarcelli 2009;Schindel 2019), rather than the use of direct force against migrants' bodies and possessions.
Lastly, we wish to note that methodologically, our aim in this paper is not to excavate factual details about how these networks operate -we take our cue from Malkki (1995) who notes that ethnographers are not 'police detectives' searching for 'the' truth -but instead to show how migrants stuck between EU and non-EU borders experience, often embedded in violence, negotiate and account for the process of smuggling and working for smugglers.

Understanding the Smuggler-client Relationship in the Context of Direct Border Violence
Migration, at EU policy level, is often understood through the lens of societal insecurities and criminal activities (Dobreva and Radjenovic 2018). Within this context, migrant smugglers have been labelled as ruthless criminal organisations (Albahari 2018;Vogt 2016) who send derelict and overcrowded dinghies across the Mediterranean (Triandafyllidou 2018). Smuggling of migrants has become a central element in the fight against international organised crime (Perkowski & Squire 2018). The EU, in particular, promotes cooperation on anti-smuggling in its wider neighbourhood (Little & Vaughan-Williams 2016), including land borders in South East Europe, as we detail below.

Migrant Smuggling along the 'Balkan Route'
The Croatian borders with other former Yugoslav states are one part of multiple converging migratory routes through Europe that has popularly become known as 'The Balkan Route' (Bird et al. 2020). As El-Sharaawi and Rasza (2019) argue, the conflation in the popular imaginary of the contemporary migration route with the pre-existing smuggling routes  creates symbolic resonances between the region's imagined 'criminality', and allows for the creation of policies and intervention strategies based on these assumptions. South East Europe, in particular, Bosnia, Serbia, North Macedonia and Kosovo, have long been a focal point for EU intervention, often, based on both the EU's governance of its neighbourhood and assumptions about 'The Balkans'.
Although smuggling, illicit trade, organised crime, corruption, and the 'shadow' or 'gray' economy have been especially visible in the region since the 1990s wars, scholars argue that their interpretation has often been put down to simplistic and 'apolitical analytical categories', such as elite greed (Strazzari 2007, 205-206), the wars themselves, ethnic grievances or 'cultural assumptions' about the region's character (Hozic 2004). Whilst the 1990s wars did make illicit trade and smuggling more prominent and accessible, Hozic (2004), , Belloni and Strazzari (2014) and Pugh (2017) all point out that the illicit activities (including smuggling and corruption) are not created simply due to state absence or weakness, but rather, are embedded in deeper historical, international and capitalist structures and developments. Hozic notes that smuggling and 'gray' economy links date back to the region's position as a transit zone during the Ottoman Empire, caught between fluid trade routes and taxation systems, which facilitated both licit and illicit trade (Hozic 2004), whilst the growth of the post World War II 'Balkan Route' for drug smuggling was partly facilitated by the geopolitics of Yugoslavia and its positioning between the two power blocs . During the 1990s, illicit trade and smuggling became a core part of 'survival economies' and often the only way to obtain certain goods -both licit and illicit -and eventually, as Belloni and Strazzari note (2014, 867) the informal war economies became an integral part of post-war transitions to power and embedded within governing structures.
Crucially, as Hozic (2004, 36) argues, the 'pervasive informal and illicit trade' in the region should be seen rather as 'an integral part of the world economy, a local response to the rise of global merchant networks', and as such, embedded within an international states system with 'zones of law and lawlessness' and reinforced through practices of international intervention. Organised criminal activity including smuggling has often been associated with, or has been the direct result of international interventions in the region, such as embargoes, or the arrival of peacekeepers (Hozic 2006, 243): the 1991 arms embargo for instance, is seen as the 'grand entree of organised crime on to the Yugoslav stage' (Strazzari 2007, 189).
Present-day migrant smuggling through the Balkans appears to have little operational connection to historic and existing smuggling regional networks, according to what we have been able to observe on the ground. Local networks and actors are involved in contemporary migrant smuggling, but often in marginal ways. However, what does link the historical and war-time smuggling in the Balkans and the contemporary migrant smuggling today is that both of these dynamics exist as 'integral parts' of global economies (Hozic 2004), as well as the 'illegality industry' produced by EU border regimes (Andersson 2014, 12-14). Movement through the EU is restricted for both residents of the Balkans (for instance, a visa regime is still in place for Kosovo citizens) as well as for the migrants they now host. Such restrictions, Andersson (2014, 14-15) argues produce the 'illegality industry', an assemblage that draws together disparate groups of actors, technologies and environments, including border restrictions, '"management" of irregular migration', funding, development aid, 'security solutions', smugglers, activists and NGOs.

Impact of the EU's Externalisation on Migrants and Their Experiences of Violence
In addition to the importance of the historical smuggling context of South East Europe we would also like to highlight the role played by the EU and its borders measures in constituting smuggling dynamics in this region and impacting migrant-smuggler cooperation.
Scholars focusing on the strategies and techniques used by the EU to govern its external borders have long identified systematic attempts to shift internal security concerns, which traditionally fall under national security matters, to the domain of external relations and foreign policy (Lavenex and Wichmann 2009;Monar 2004;Rees 2008;Trauner and Carrapico 2012). The EU has externalised its border policies and responsibilities to countries outside of its external borders (Norman 2020), including South East Europe which it sees as the 'Western Balkans' a policy area to which initiatives, such as border security externalisation, Frontex cooperation and additional funding for migration issues, are diffused and implemented. Such cooperation is seen in broad terms as a part of the EU integration process (Geiger 2016;Rexhepi 2018) in Serbia, Bosnia, North Macedonia, Kosovo and Albania, and prior to its accession, Croatia.
Although countries, such as BiH and Serbia began their EU integration processes long before 2015, and with it, the alignment of their migration policies and border security with those of the EU (Stojić-Mitrović and Vilenica 2019), the responsiveness of both countries to EU migration policies and funding became more evident after this time. For instance, both countries were able to open additional (and refurbish existing) asylum and reception centres with EU funds (Obradovic-Wochnik 2018). We also found during our fieldwork that locally, specific initiatives, such as hiring of additional social workers for unaccompanied minors in Serbia, were only possible due to additional EU funding.
EU counter-smuggling efforts in the SEE region have been undertaken in the context of the active Stabilisation and Association Process (SAP) since 2000 (Dahlman 2016;Hajdinjak 2002). Anti-smuggling efforts are a focal point of the 'Schengen and External Borders' section of the Commission's (2020a, 44) progress report on Bosnia, for instance. The report notes that the Border Police has seen a higher number of detections related to human smuggling and that it requires additional surveillance equipment at the border and better monitoring of 'green borders' (land border areas between 'legal' border checkpoints often used by smugglers) (European Commission 2020a). The EU's report on Serbia is less positive, noting that the country is 'yet to establish a convincing track record of investigations' and that it needs to 'increase efforts to detect and prevent the smuggling of migrants' (European Commission 2020b, 40).
This externalisation process significantly impacts cross-border journeys in terms of migrants' exposure to violence, and subsequently their increased reliance on smugglers. As Anderson and Keen (2019) argue, the EU's externalisation confines migrants in various countries in the Middle East, Africa and Europe, where they are stranded between state authorities who get political legitimacy, financial support and impunity for abusive behaviour.
For instance, existing research shows that EU policy and financial support for border militarisation has resulted in the confinement of migrants in camps across Greece, Serbia and Bosnia, where they often live with minimal or no aid in both formal and informal camps (Ahmetašević and Mlinarević 2018;Bird et al. 2020;Zaragoza-Cristiani 2017). Whilst there are wide variations in accommodation and services available to migrants in the region (Bird et al. 2020; Obradovic-Wochnik 2018) many migrants are destitute or stuck in legal or living conditions, which place them in even more precarious and desperate conditions and reliant on smugglers.
Groups working directly with migrants and monitoring violence at EU external borders have noted an increase in the use of physical force to push migrants back into non-EU countries since 'border closures' in 2015 (BVMN 2020;Nedoh 2017). Recorded instances of physical violence against bordercrossers have included the use of batons, kicks, punches, electronic devices, shootings, and detainment in vans with a severe lack of oxygen; removal, theft and/or destruction of personal belongings including clothes, shoes, food, money and mobile phones (Augustová and Sapoch 2020). Currently, collected testimonies and evidence suggest that the majority of such incidents happen at EU borders and appear to be perpetrated by EU countries border staff (Croatia, Hungary, Greece, Slovenia), with migrants pushed back into non-EU countries including BiH, Serbia and Turkey (BVMN 2020).
Border violence follows a broader trend Europe-wide, in which increasing border restrictions are framed as activities that counter smuggling and therefore supposedly help migrants by eliminating the criminal organisations that they rely on. Increased border security, however, often targets the migrants themselves rather than the smugglers (Mandić 2017;Simpson 2020), and the military operations that aim to destroy smuggler vessels result, instead, in the blocking of legal pathways for migration and asylum claims (Garelli and Tazzioli 2018). Tightened migration controls and border violence make migrants more reliant on smugglers when seeking protection from this violence (Antonopolous & Winterdyk 2006; Baird 2016; Kyle and Scarcelli 2009;Mandic & Simpson 2017;Vogt 2016). Violence levels at borders thus provide an important context for understanding human smuggling and its internal relations (Kyle and Scarcelli 2009). Against this background in which border security, border violence and an increase in smuggling all intersect, we argue that smuggler-migrant cooperation and co-option becomes fluid alongside violent border externalisation measures, which has been under-researched in the literature.

Smuggler-migrant Negotiations along the 'Balkan Route'
Whilst the existing literature has advanced our understanding of human smuggling, its root causes, economic rationale, operating mechanisms, and social structures (Achilli 2018;Ahmad 2011;Maher 2018;Majidi 2018;Müller 2018;Triandafyllidou 2018), there is less knowledge of how migrants get involved in smuggling as operatives in the context of external pressures and precarious situations, such as avoiding border violence. Yet Baird (2016) argues that migrants are active agents in smuggling. Similarly, Zhang, Sanchez, and Achilli (2018) have suggested that migrants are not just passive actors in smuggling, but rather active in vetting and producing smuggling services and evaluating their reliability. This also includes findings on smugglers with a migration background (Sanchez 2017;Van Liempt and Doomernik 2006).
Furthermore, empirical findings from Greece, Turkey (Achilli 2018) and Eritrea (Ayalew Mengiste 2018), indicate that migrants occasionally work for human smugglers as recruiters, guides, or intermediaries. Active engagement in smuggling businesses both as clients and service providers has been observed especially in migrants' home states where migrants and smugglers are often related through kinship and strong social and ethnic ties and local histories, in addition to economic relations (Kyle and Scarcelli 2009;Stone-Cadena and Velasco 2018). These ties contribute to the creation of symbiotic smuggling relations and give raise to a complex cooperation between migrants and smugglers built on trust (Brachet 2018;Van Liempt and Sersli 2012).
This context contrasts with that of transit states where socio-cultural ties decrease as migrants move away from their home communities (Stone-Cadena and Velasco 2018), and migrant-smuggler cooperation and negotiations become more dependent on border regimes and social realities at borders. For instance, the tighter and more violent the border control is deemed to be, the higher the smugglers' prices (Vogt 2016). The same rationale applies to the 'Balkan Route', where we observed that prices varied seasonally and temporally, and depended on the level of difficulty and violence associated with specific borders, especially crossing from non-EU to EU zones. However, ethnic and social ties do not dissolve fully when migrants often -but not always -use smugglers from their own community and rely on recommendations of smugglers from their friends and relatives along the Balkan Route, as also pointed out by Mandic (2017).
Price is at the centre of negotiations between service providers and (potential) clients (Maher 2018), but it has been mostly viewed as financial capital rather than migrants' own labour. As Simpson (2020) and Tinti and Reitano (2016) suggest, smuggling services are mostly based on how much money migrants have at their disposal, with the 'higher end' services including flights and fake documents. Generally, migrants using the 'Balkan Route' have fewer resources at their disposal, and rely on more ad hoc, less 'professional' and far more precarious smuggling operations (with prices that are nonetheless exploitatively high (see Müller 2018;Simpson 2020), and journeys that can take months or even years, as we detail in the sections below. The prices paid to smugglers can vary depending on the border being crossed. However, those without funds or connections are often left to find various ad-hoc arrangements as they reach a town or a border.
When exploring accounts of smugglers and smuggling as experienced by migrants in the context of the EU's land borders, we found repeating themes across the multiple types of experience, including, notably, the recruitment of migrants into smuggling networks as operators. Previous research by Mandic (2017) found that migrants who lacked funds for smuggling were offered the possibility to work for their smuggler in hairdressing, hospitality, or construction after they reached their destination in order to pay off their debt. Similarly, a research by Sanchez (2015) reveals that migrants can enter smuggling networks to reduce smuggling fees. However, the context in which migrants are being offered the opportunity to work for smugglers around EU (land) borders, the diverse aspirations of these individuals to do so, as well as when and how this dynamic evolves remain under-examined.
Yet this fluid relationship echoes broader findings elsewhere, notably that the migrant-smuggler relationship is complex and often blurs lines between migrants as "victims" and smugglers as "perpetrators" (Triandafyllidou 2018). Smugglers can be perceived as facilitators of protection from violence (Ayalew Mengiste 2018; Brachet 2018; Maher 2018; Vogt 2016), as well as agents of exploitation and violence themselves, as they may kidnap or rob migrants (Müller 2018;Peterka-Benton 2011;Simpson 2020). The smuggler-migrant relationship thus intertwines care and control. However, the point of departure of existing analyses of smuggling relationships commonly follows static roles of smugglers as service providers and migrant clients (Van Liempt and Doomernik 2006). In what follows, we shift our attention to the context of violence along the Croatian (EU) borders with non-EU states and question how it impacts smuggler-migrant cooperation, and promotes extensive fluidity and contestation in this relationship.

Avoiding Border Violence with Smugglers
'It is not possible to [to cross the border to Croatia] without [taking] a car arranged by a smuggler. If you walk you get caught, beaten and you are back [in the Border Town] . . . I need a good smuggler to make it', Rashed told us after describing his unauthorised border crossing and push-back from Croatia back to the Border Town. Rashed's comment points to the reality of pushbacks from EU states to its wider neighbourhood (see Augustová and Sapoch 2020; Isajkee et al 2020). Rashed stresses how fear of this violence made him reliant on smugglers, which illustrates how violence around borders is a common impetus for hiring a smuggler (Kyle and Scarcelli 2009). Similarly, many other migrants told us that they feared walking across unknown terrains without the help of smugglers, and risking being easily detected by drones, helicopters and police patrols, which could lead to more instances of pushbacks and violence. From conversations with migrants, it became apparent that many believed that smugglers had knowledge of the terrain and commonly organised their clandestine transit with the support of bribed border authorities. To some extent, this belief gave migrants the perception of being protected from physical abuses by state authorities and push-backs when cooperating with smugglers, in comparison to having to navigate the journey alone.
We use Rashed's comment to begin examining how security measures, used within the EU's external border governance at the Croatian border with non-EU states, impact migrants' negotiations in smuggling and their relations. As observed along borders in wider European context, including our case study, the brunt of the military techno-scientific approach (Jones and Johnson 2016) deployed to fight cross-border crime falls on migrants rather than smugglers (Garelli and Tazzioli 2018;Mandić 2017;Simpson 2020), who are routinely subjected to direct physical violence. In our research, migrants primarily report using smugglers to avoid violence by border guards, as the following interview excerpt highlights: The last time I went on the game [unauthorised border crossing], they [Croatian border police] broke my finger and humiliated me. I am not young anymore. All my life is in Austria, so even if they [Croatian border police] keep beating me, I will keep going. I will now try [to cross the border] with a smuggler by car because it is safer (Haji).
Attempting the route alone or with groups of friends with only mobile phones and Google maps -often the only tools available -is complicated, not only due to the threat of border violence but also due to unexpected and often littleknown terrain and dangers, such as mine fields left over from the 1990s wars, which remain un-cleared. We observed that missing migrants, deaths and injuries whilst crossing difficult terrain or inflicted by border authorities against those trying to cross borders into Croatia were not uncommon. Migrants thus seek smugglers and resort to cooperation with them to gain protection and assistance (Tinti and Reitano 2018;Vogt 2016), which is also significant in the context of EU land borders.
The threat of border violence also impacts negotiations regarding the price of smuggling services. For almost the same period as BVMN (2020) reports an increase in violence on the part of border authorities, we also note an increase in smuggling prices. In spring 2018, an increasing number of migrants in the camps told us they paid 1.500 euros per person for unauthorised transport leading from the border town to Italy, and mostly no deposit was needed. In contrast, in January 2019, several migrants referred during the interviews that they were asked by smugglers to pay 3.000 euros in advance for the same route and means of transport. Those unable to pay the higher-end smuggling services were more physically exposed -by walking, rather than, for instance, being smuggled in a car -to border violence, the difficult terrain, and longer enclosure in poor living conditions of makeshift camps.
Tougher conditions in terms of navigating border controls and more expensive journeys, produce a context in which working with and for smugglers comes to be understood as a necessity. This became apparent to us when meeting migrants who had extensive dealings with a smuggler named 'Yusuf', who appeared to be at the centre of smuggling operations in the Border Town. Migrants lacking the necessary funds to pay for smuggling services commonly undertook tough and risky jobs for Yusuf around borders to pay for their own future journeys. In what follows, we ask when, how, and why migrants opt to enter smuggling businesses as operatives around borders and discuss the negotiations and fluidity of the relationship between migrants and smugglers, which goes beyond a reductive client-service provider relationship.

Becoming a Smuggler
Ahmad was sitting on the bed and trying to remove thorns from his skin and clean small wounds on his legs. A few hours before, he had returned from a two-day long walk from the Border Town to Croatia and back. 'I walked twenty people to the point in Croatia, from there they were picked up by a car. Also, women and children were with me. Now, they are in Europe, safe, and I am back here [Border Town] [crying]. . . . One man had a problem with his kidney and could not walk, so I carried him for hours on my back. I did not sleep for two days, but this is my job. I am a smuggler. It is risky and hard, but I do not have another option. I will go ten times more, and after that, it will be my turn'. (Excerpt from field diary).
The above narrative refers to Ahmad, who was describing his working routine in the smuggling business. Ahmad was trying to transit to the EU, but had no money to pay for a clandestine transfer by car. Two months following his arrival to the Border Town, Ahmad was making weekly attempts to cross the border using only Google maps. However, he always returned to the camp with new marks of violence on his body from police attacks in Croatia. Then, Ahmad said that he got a job offer from Yusuf that would help him to travel to his destination by car, free of charge.
Ahmad's experience of shifting roles from a client to a smuggling service provider suggests an essential intersection between border violence at the EU external and internal borders and migrant smuggling. The impact of restricted freedom of movement and violent border regimes goes beyond migrants' dependence on smuggling services as clients (Tinti and Reitano 2018;Vogt 2016) and the production of a 'few' smugglers with social and economic networks of relational proximity (Brachet 2018). Violent policies outsourced from EU areas (Croatia) to the EU neighbourhood put migrants, who become stranded in unknown transit terrains with little social connections, in a situation of violence and precarity. In this context, migrants' aspirations to join smuggling organisations as workers become a necessity in order to cross the border with no financial means and a lack of social ties. Thus, the context of fear and violence whilst border crossing allows to shift the transit fee -that is at the centre of smuggling -from financial capital (Maher 2018;Simpson 2020;Tinti & Reitano 2016) to migrants' labour in smuggling provision. Whilst Ahmad described himself as a smuggler, his role in smuggling operations was different from that of his boss. Ahmad led clients across the border to Croatia and delivered them to specific 'pick-up points'. From there, they would continue their journeys onwards by car, while Ahmad returned to the Border Town to rest before the next job. Ahmad was seeing others crossing the border and moving closer to safety thanks to him, which was an aspect of his role as a "smuggler" that he was proud of. However, he was then returning to insecurity and violence in the Border Town, causing him further distress. Seeing Ahmad's tired and injured body also indicated that his job was physically demanding when he had to walk for hours through forests and carry others who struggled to continue the journey. Many other migrants said they were engaged in the provision of smuggling services and expressed aspirations similar to Ahmad's: to avoid further violence and push-backs along the borders, to compensate for their limited finances, improve their chances of being smuggled to their destination, and to obtain diverse amenities in transit spaces.
Migrants-turned-operatives work in smuggling networks as guides, drivers, recruiters of clients, price negotiators, and builders of clandestine routes, with their roles being allocated on the basis of individuals' skills, age, nationality, and remuneration. For instance, Isak who had driving experience with large vehicles in his home country, said that his job consisted of driving clients in a van from the 'pick-up points' along the border with Croatia to the Croatian-Slovenian border. In contrast, those who could speak several languages were recruiting clients of various nationalities, negotiating prices, and acting as intermediaries between smugglers and clients.
It is essential to note that we also met a few migrants who rejected the opportunity to work for smugglers in exchange for free passage and preferred to keep trying to cross the border in a group of friends for moral reasons. "Yusuf is Dracula. He [financially] exploits people. I would never want to engage with such person", explained Karam. For others, caring and family responsibilities in transit did not allow them to take risks associated with an active engagement in smuggling provisions, namely being arrested. However, our daily encounters with individuals with fluid client/service provider roles indicate that this type of relation in smuggling was significant in the Border Town.
What these encounters show is that some migrants are pushed into smuggling organisations as operatives in the context of violent borders. We thus agree with Stone-Cadena and Velasco (2018) that smuggling is a continuum of migrants' decision-making in which migrants play active roles. This cooperation, nevertheless, goes beyond occasionally recruiting clients, guiding them over the border (Achilli 2018;Ayalew Mengiste 2018) or reviewing their services (Zhang, Sanchez, and Achilli 2018). Once migrants are trapped living destitute in transit countries like Serbia and Bosnia, experiencing violence each time they cross the border alone, a new form of dependency is developed, which pushes them towards increasingly complex and extensive involvement in the smuggling business. Migrants' aspirations to actively engage in smuggling provisions in violent transit contexts, thus, differ from those of migrants in the context of their home states, where symbiotic relations in smuggling are primarily conditioned by social, ethnic and economic kin (Maher 2018) rather than mere border regimes. Nevertheless, social relations remain fundamental for migrants in transit when they provide smuggling services for others, especially their friends, and help them cross the border with reduced or no fees. Yet, working around borders is not only about help to be smuggled but also about numerous risks involved in navigating violent policies.

Being Helped and Harmed in the Smuggling Business
Migrants' smuggling work around borders is often fraught with contradictions between on the one hand what they expected and perceived this work to consist of, and the 'safety' measures it would bring, and on the other hand the reality of what happens once they start smuggling. In this section, we examine the discrepancy between expectations and the reality of smuggling and show the fluidity between being helped and exploited in the migrantsmuggler continuum.
The first set of contradictions comes from perceptions linked to the provision of amenities and services by smugglers, to those working for them. Smugglers often provided the men working as leaders, drivers and middlemen a flat with heating, hot water and food. Migrants framed these amenities as a significant level of help, given that most migrants lived in makeshift camps, more specifically in shelters made of wood and tarpaulin, with no electricity, limited access to sanitation and food. Many people in the camps reported skin diseases, hepatitis, lack of medical care, cold and hunger. Whilst the existing literature shows that smugglers provide their clients with care and shelter (Maher 2018;Mandic 2017;Vogt 2016), we found this to be a 'luxury' available mainly to those whose roles blurred the categories of client and operative in the Border Town. This can be explained by the fact that only a limited number of people can afford these services. In addition to being provided decent accommodation and food, a few migrants working around borders as operatives also reported that they were being paid 10% of the client fee (200-300 euros), which they used to buy food, clothes and support their families in their countries of origin. Most importantly, the promise of being smuggled across the border 'free of charge' (once they had worked a specific length of time) gave the migrants, who were left with no other crossing options, new hope to move across the border faster and safer by car by the smugglers, who reportedly had knowledge of the journey and connections with border guards. Consequently, some participants said that they felt increased safety and control, not only over their own migratory journey but also of that of other smuggled people: 'When I am done and ready to leave, I can choose one person to go with me for free. . . . If I am not a smuggler, this would not be possible and I would get stuck in the Border Town in camp for a long time' (Ahmad).
'I can give discounts to people who I know if I talk to my boss. I want to help people, not like some smugglers who trick people and steal their money' (Faled). Therefore, migrants do not always experience smuggling as a predatory transaction, which also includes those working around borders as operatives. The role of a "smuggler" is perceived as that of a helper (Bilger, Hofmann, and Jandl 2006;Maher 2018) and a facilitator of survival ); a position empowered by repressive border management (Brachet 2018;Mandić 2017). Whilst this association has been examined from the perspective of migrants as clients, we argue that it is even more significant in a situation where migrants decide to work for smugglers as the result of border violence. Indeed, an MSF (2017, 4) report on violence against unaccompanied minors along the 'Balkan Route', shows that although some violence was perpetrated by the smugglers, the vast majority of violence was attributed to state authorities. This report echoes findings elsewhere (Maher 2018) that abuse at the hands of smugglers is rare in contrast to abuse by state authorities along the EU's borders (BVMN 2020). Entering smuggling networks as operatives therefore presents another layer of negotiations through which migrants aim to reduce the vulnerabilities associated with their clandestine journeys (Sanchez 2017).
Yet when we consider migrants' narratives of their day-to-day work, we also repeatedly find themes of abuse and exploitation, which highlight the distinction between the migrants' hopes and intentions to enter smuggling organisations versus their experiences on the ground. We found working for a smuggler to be rarely a one-off event, which often turned into a protracted involvement of months or even a year. Further, several men said that they felt underpaid for their hard manual labour, such as cutting bushes in forests or facilitating crossings, given that the money they received was not enough to cover their food costs in the Border Town.
Whilst migrants generally believe that they will be safer from border violence and police interceptions whilst crossing with a smuggler, as smugglers are thought to work together with border guards whom they pay off (a dynamic found elsewhere in the migrant smuggling business, see e.g. Achilli 2019, 201), the reality is that many migrants still experience violence and arrest when working around borders. Migrants working as path-finders across the border and drivers encountered police interceptions and border violence whilst crossing the Croatian and Slovenian borders with their clients. Ali, for instance, led the same group of clients to the pick-up point on three separate occasions, but always got apprehended, detained and physically abused by the border guards before being pushed back to the Border Town: They took me to a police station. They told me that they had seen me crossing the border many times and started shouting at me: 'You are a smuggler!' Afterwards, they took me to a closed room and beat me for half an hour until they got tired . . . They did not have any proof in the end and let me go. I was lucky I did not end up in jail for seven years (Ali).
Although migrants working as smugglers were frequently intercepted and accused of being smugglers, the same suspicion was directed at migrants who were trying to cross the border alone, often based on things like having an open phone map, being repeatedly apprehended in the same border locations, or their ability to speak multiple languages. All those suspected by border authorities of being smugglers (whether they were involved in smuggling as operatives or not) described being detained without due process, and being physically abused before being pushed back to the Border Town. Similar reports of authorities confusing migrants with smugglers and using violence against them were also reported in the Mediterranean (Simpson 2020). Thus, smugglers' promises of help and migrants' hope to be protected in fact often result in further experiences of violence and arrests.
The above risks can be seen as re-delegated and displaced onto migrants from their bosses, whose visibility and risks around the Border Town decreased alongside more fluid relations with migrants. Indeed, using migrants as modus operandi allows smugglers coordinating transactions to be distant from the border and operate behind closed doors (Simpson 2020), which is the case when migrants work around borders. Migrants were also kept in the dark about the detailed arrangements between authorities and their bosses, which is another common risk mediation strategy of smuggling organisations (Bilger, Hofmann, and Jandl 2006;Tinti & Reitano 2016). The majority of our participants believed that their bosses would not expose them to such risks, even though they were apprehended and attacked by border authorities during their work. Sometimes, the risks of apprehension and violence were accepted as the price 'worth' paying for increasing their chances to move to the EU. The risks of working for smugglers were generally seen as lower and more acceptable than the risks associated with crossing borders alone without a smuggler and facing the prospect of border violence. Once exposed to risks, migrants rarely stop working for a smuggler before they make it across the border.
However, the simultaneous occurrences of care and violence are not caused by smuggling networks alone. These everyday interactions in smuggling are generated by political conditions of border crossing routes (Maher 2018). We cannot forget that migrants opt to resort to smuggling networks for protection in the first place, as the result of the international fight against cross-border crime (Brachet 2018). This is particularly significant in the context of the Balkan Route where the international involvement to manage "organised crime" (smuggling) was imminent during the war and post-war interventions (Hozic 2004) and later increased by the EU exporting its migration and border security policies to South-Eastern Europe (Stojić-Mitrović and Vilenica 2019), which was characterised by the common use of violence (Isakjee at al. 2020). As Sanchez (2017) reminds us, intersections between violence and smuggling should be understood in the context of migration enforcement systems, which heighten risk and precarities among those who are left with no other options than travel with smugglers, and as our research shows, work for smugglers.
The fluidity and contradictions of the migrant-smuggler relationship problematise both the exploitation of migrant precarity by smugglers, as well as the increased border security and violence that have become a part of the migrants' everyday. Eventually, some of the participants we interviewed, reached their destinations as arranged with their bosses, but at the expense of having been involved in an activity that is stipulated as illegal by the States involved and having repeatedly experienced harm and abuse related to smuggling work. However, many others still remain in the Border Town, trapped by the limited options of either facing the violent borders or relying on facilitators of clandestine transits.

Concluding Comments
In this article, we examine the relationship between the increasingly violent deterrents along the EU land borders and the recruitment of migrants into smuggling networks as operatives. The experiences of Ahmad and other migrants stranded in the Border Town highlight that, in a situation of protracted vulnerability by border violence, state and law enforcement, the categories of "migrant" and "smuggler" -blur when migrant and smuggler become the same person. By exploring migrants' aspirations to work as facilitators of clandestine transits and their experiences of it, we move the analysis of relations and negotiations in smuggling further and contribute to scholarly findings on care and violence in smuggling, which we argue are two faces of the same coin. We show that the relationship between smugglers and smuggled in transit is not only diverse (Van Liempt and Doomernik 2006) and symbiotic due to social and economic ties (Brachet 2018). Smuggler-smuggled relations can get fluid and often blurry, as border controls and direct violence are outsourced from the EU (Croatia) to non-EU states, leading migrants to seek protection while in transit.
Focusing on migrant-smuggler experiences reveals that border violence increases smugglers' bargaining power, enabling them to charge higher prices for their services and allowing to negotiate with migrants, whose lack of funds and social ties make them more open to the possibility of paying smuggling fees with their own labour. Migrants are thus pushed to enter the smuggling industry as they believe that smugglers offer protection from violent pushbacks and increase chances of a successful border crossing, which is a dynamic previously examined from the perspective of migrants as clients (Baird 2016;Kyle and Scarcelli 2009;Vogt 2016). However, migrants-turned-smugglers often narrate accounts of exploitation and further violence and arrests by state authorities while undertaking risky tasks. The threat of border violence can thus determine who engages in smuggling provision whilst migrating, as well as further shape the experience of their smuggling provision work around borders.
We argue that it is vital to shed light on migrants' experiences of smuggling in diverse roles -both clients and smugglers -that increasingly overlap as the result of tougher EU violent border policies in this wider region. By doing so, we problematise exporting and externalising justice and home affairs norms and practices addressing cross-border crime to neighbouring states, such as the SEE region and elsewhere (Council of the European Union 2009). Violent tools used along EU borders increase visible and direct forms of violence by border authorities and shape migrants' vulnerabilities, which lead to a flexibilization of the smuggler-client relation and push migrants to work for smugglers. Border violence and its direct effects on bargaining between smugglers and migrants thus must be brought to the debate on fighting cross-border -crimes within the EU's governance in and outside its borders.