Discursive alignment of trafficking, rights and crime control

Since the 1990s, human trafficking has become the battleground for competing discourses on human rights and penality. While rights solutions are generally presented as in opposition to crime-control measures, in the context of anti-trafficking interventions, rights-based initiatives and criminal governance are often linked together both discursively and in practice. Drawing on the findings of Discourse Analysis of 120 texts about trafficking, this paper explores how dominant discourses and alternative voices construct the relationship between human rights and penality. It is contended that penality is framed as a crucial tenet of human rights. Dominant discourses (the ‘law enforcement’ and the ‘victims first’ discourses) link human rights to state coercive action, seen as a necessary component of their effectiveness. Alternative voices (the ‘incompatibility’ and the ‘transformative justice’ discourses) reject the appropriateness of penal intervention, but they end up preserving what they denounce.


Introduction
Since the 1990s, human trafficking has become the battleground for competing discourses on human rights and penality. 1 Trafficking in human beings is generally framed as a human rights violation and a crime (O'Connell Davidson, 2010). However, the vagueness of these terms has left policy-makers, practitioners and scholars free to impose distinctive meanings on human rights and penality in relation to trafficking. The confrontation of different approaches has contributed to the fixation of dominant discourses, with other voices placing themselves in a position of criticism.
A dominant discourse, usually advanced by governments, courts and organisations interested in global security, maintains that criminal justice is the primary instrument to end trafficking. In this discourse, that I term the 'law enforcement' (LE) discourse, human rights are used to provide a justification for a broad range of criminal measures. Criminal law is necessary to punish 'organised criminals' and protect 'helpless' individuals, who can either be the 'genuine' victim of trafficking or the general population affected by the crime of trafficking. Prosecution of traffickers prevails over victims' protection. Trafficked people cease to be 'innocent' victims and become 'willing' lawbreakers once they commit a crime and pose a threat to the state's security.
Human rights advocates reject this discourse with a competing one that I call the 'victims first' (VF) discourse. While the emphasis is on victims' support, criminal justice plays a role in the fight against trafficking insofar as it complies with human rights. While human rights complement and moderate criminal measures by shifting the focus from the state to the victims, criminal law gives enforcement to human rights principles.
Critical academics and non-governmental organisations (NGOs), but also sex workers' grassroots movements, have questioned these two dominant discourses, by advancing alternative voices. There is a counter-discourse that insists on the incompatibly of human rights measures with penal policies. I call it the 'incompatibility' (INC) discourse. Another counter-discourse that I term the 'transformative justice' (TJ) discourse claims that both human rights and criminal law are inadequate for contending with the structural causes of human trafficking, and calls for a more transformative approach to challenge the vulnerabilities of a neoliberal world. Although counter-discourses have recently gained more space, especially within academia, they remain marginal compared to dominant discourses.
Several scholars have analysed the discourses that are produced around human trafficking. Much of this research, conducted from a feminist and sex-as-work perspective, has been concerned with the construction of narratives about gender, sex, agency and consent (Berman, 2003;Sanghera, 2005). For instance, Doezema (2010) has examined current debates surrounding sex trafficking by drawing historical comparisons to 'White slavery' at the end of the nineteenth century. Vance (2012), in her analysis of a series of documentaries about sex trafficking, has illustrated how melodramatic representations of male villains and female victims influence law and policy on the matter. Discourse Analysis (DA) has also been conducted from the perspective of migration (Ausserer, 2008;Dauvergne, 2008). Finally, O'Brien (2019) has focused on the victims, villains and heroes of trafficking stories, to show how the dominant trafficking discourse relies on cultural assumptions about gender and ethnicity, and wider narratives of border security, consumerism and Western exceptionalism. This paper continues the scholarly conversation about anti-trafficking discourses: my focus is on discourses about human rights and penality. The analysis of these discourses aims to unearth and problematise how, in the field of anti-trafficking, the relationship between rights protection and crime control is thought, discussed and put into practice.
Drawing on the findings of the DA of 120 texts, I show how, in initiatives against trafficking, human rights have become increasingly interwoven with penal agendas at both the domestic and international levels. Recent scholarship has noted an embrace of penality by human rightswhat Engle (2015) calls the 'turn to criminal law in human rights'. Human rights are no longer mainly a 'shield' against the excesses of criminal law (e.g. by providing due-process rights), but they are increasingly used as a 'sword', by making criminal law one of the main instruments for their promotion (Van den Wyngaert, 2006;Tulkens, 2011). Most commentatorsaccepting this trend as uncontroversialhave advocated the deployment of penal means to underwrite human rights (Neier, 2012;Sikkink, 2011). Other authors have expressed perplexity at this development (Engle et al., 2016;Pinto, 2020a). For some, human rights have been co-opted by a language of accountability from which they need to be rescued (Hannum, 2019); for others, human rights are prone to be 'governed through' (Sokhi-Bulley 2016) and to act as vehicles for extended securitisation and criminalisation (Lippert and Hamilton, 2020;Kapur, 2018).
While the use of human rights to achieve penal aims has also concerned the field of human trafficking (Pinto, 2020b;Bernstein, 2012), the literature has generally presented human rights solutions as in opposition to a dominant crime-control model to combat trafficking (Giammarinaro, 2020;Hathaway, 2008;Obokata, 2006). In other words, whereas much has been written about the differences between a human rights and a crime-control approach to trafficking, there have been few attempts to reflect on the commonalities between the two. Aiming to fill this gap, this paper shows how, from an analysis of anti-trafficking texts, different discourses emerge, each of which can be more focused on crime control or human rights. While even the LE discourse widely invokes human rights, the VF discourse is pervaded by penality language. Although the two poles remain human rights and penality, the current characterisation of the debate fails to acknowledge that both discourses, no matter which language they use, are committed to a similar premise, namely the necessity for penal intervention. The interesting division is not between crime control and human rights, but between discourses that are favourable to penality and counter-discourses that reject it. Yet, once we move from discourse to practice, we notice that penal intervention is hegemonic and counter-discourses have little to no impact on anti-trafficking action. The findings of this paper are also relevant beyond the field of trafficking and provide new insights into the examination of human rights not only as a source of resistance, but also as a means of criminal governance (Lippert and Hamilton, 2020;Armstrong, 2018;Sokhi-Bulley, 2016;Lippert, 2016). Whilst not the only factor, human rights are integral elements in demanding the state's penal intervention, whicheven when rights-compliantconstantly fails to protect the most marginalised.
The paper starts by presenting the DA method employed to retrace the discourses. I then examine the discourses, following three analytical dimensions: objects, subject positions and themes. The last part considers how knowledge produced through the analysed discourses relates to the reality beyond discourse.

DA
The method used in this paper is a DA of 120 texts about human trafficking. 2 DA is a method for critically examining how knowledge is discursively produced and attached to various social objects, subjects and techniques, enabling certain possibilities of action while precluding others (Dunn and Neumann, 2016, p. 12). To include the broad legal and political debate, I focused on four genres of texts: positive legal texts (both 'hard law' and 'soft law'); policy-makers' documents; reports from NGOs; and academic papers and books. Attention was given only to texts discussing human rights and penality (broadly conceived) in the context of trafficking. Texts were selected at three levels: the international level of the UN; the regional level of the Council of Europe and the EU; and the domestic level of England and Wales. Europe was selected for its advanced laws and policies on trafficking, which, more than any other region in the world, combine human rights elements along with penal measures. This can be evidenced by the 2005 Council of Europe Convention on Action against Trafficking in Human Beings (CoE Convention) and the work of its monitoring mechanism (GRETA); the EU Trafficking Directive 2011/36/EU; and the case-law of the European Court of Human Rights (ECtHR). England and Wales are an optimal case for my investigation, due to the lively debate on how to address human trafficking and the recent enactment of the Modern Slavery Act 2015a law enthusiastically supported by human rights actors. In terms of time, texts span from 1991 to 2020. Although trafficking is not a new phenomenon, it has only been in the last three decades that it has become a matter of international and national debate (Gallagher, 2010, p. 16).
Through an initial literature review, attention was given to texts that have a broader reception and are authoritative in terms of source or for their binding nature. Starting from these texts (Dunn and Neumann, 2016, p. 93), I gathered other documents taking into account their intertextuality (Hansen, 2006, pp. 77-78). I consulted various databases, including LexisNexis, Heinonline, Westlaw UK and ProQuest. Other texts have been gathered by searching on the websites of various bodies, organisations and NGOs. 3 The software NVivo was used on the thousands of documents initially collected to conduct a preliminary content analysis and reduce the number of texts to be analysed in detail. The choice of sampling was influenced by various factors. First, using the query function of NVivo, I selected only those texts in which 'trafficking', 'human rights' and various terms connected to 'criminal law' 4 were all mentioned. Second, I sought to identify those texts in which the relationship between human rights and penality was discussed at length. To this end, I reduced the corpus by focusing on texts with at least ten references to 'human rights' and 'criminal law' (or connected terms) and with the highest coverage of these words throughout the text. Based on the literature and a first reading of the documents, exceptions were made for texts that were often cited. Third, I sought to have a balance of documents of different kinds by considering the year of the texts, their author or source and whether they focused on trafficking at the international, regional or domestic levels. One hundred and twenty texts were sampled (thirty-five positive legal texts, thirty-five policy-makers' documents, twenty-five reports from NGOs and twenty-five academic texts). DA was conducted on these texts.
The DA consisted of a close reading of the statements in the texts forming the corpus, looking at their meaning and materiality (their relation with concrete realities) (Kendall and Wickham, 1999, p. 33). Statements were selected as the basic unit of analysis because they make social objects, subjects and techniques visible and, consequently, they become amenable to analysis. By exploring these statements, it was possible to unearth the matrices of power relations that make certain forms of knowledge sayable and others unsayable (Nicholls, 2009, p. 32). The DA also focused on regularities of statements (representations) forming a discourse and their correlation with other representations (Foucault, 1972, p. 31). Each text was examined in the same way. First, I located its context. Different dimensions of the context were considered: (1) the situation of utterance (Who speaks to whom? When? Where? About what?); (2) the socio-historical context (institutional, sociopolitical, positional, relational context of the documents); and (iii) the textual context (genre, paratext, intertext) (Alejandro et al., forthcoming). This process of contextualisation aimed to place each text within the larger structures of meaning of which it is a part (Dunn and Neumann, 2016, p. 106), rather than to imply a necessary correlation between a discourse and a particular author, institution or genre (Foucault, 1972, pp. 69-70). 5 Then, I looked closely at the various statements that mentioned either human rights or criminal law (or connected terms) to explore how particular knowledge about human rights, penality and their relationship was produced and, in turn, how this knowledge made certain actions in relation to trafficking possible. To this end, I focused on three analytical dimensions: (1) formations of objects; (2) formations of subject positions; and (3) formations of themes (Foucault, 1972, p. 116). Taken together, these dimensions formed a set of guiding principles that focused on what can be said and what cannot about trafficking, human rights and penality. In relation to formations of objects, I looked at the ways in which human trafficking, as a social phenomenon, formed the matter that a text dealt with, as well as whether and how trafficking was understood as a 'crime' and/or as a 'human rights violation' (cf. Foucault, 1972, p. 42). The second analytical dimension concerned how the speaking subject of a text created, privileged or marginalised certain subject positions (identities), namely ways of being and acting that individuals can assume, such as 'the victim' or 'the criminal' (Foucault, 1972, p. 95). With respect to formations of themes, I examined how different statements addressed and construed the role that human rights should play in relation to, and in the context of, penality (cf. Foucault, 1972, p. 64).
The next step consisted of combining statements with similar formations to identify different discourses, namely competing representational practices through which knowledge about human rights and penality in the context of trafficking is generated. In this way, I was able to identify the four basic discourses mentioned in the introductionthe 'law enforcement', the 'victims first', the 'incompatibility' and the 'transformative justice' discourses. These discourses advance diverse understandings of the relationship between human rights and penality, and, in turn, produce different social, legal and political orders (Foucault, 1970). Finally, bringing together existing academic discussions with the findings of the DA, I looked at 'the effects of power generated by what was said' (Foucault, 1978, p. 11). Here, I sought to retrace how the discourses under analysis have had certain real effects and how these effects have been set into practice in initiatives against trafficking (Hall, 2001, p. 76).
The methodology of this paper is subject to limitations. First, the DA was able to problematise how, in the context of trafficking, human rights have been aligned with the state's coercive action, but it could not answer the question of why this alignment has occurred and what can be done to avoid 5 It is also possible that in the same text some statements can be ascribed to a discourse and other statements to other discourses. Depending on the position or goal that an actor wants to take in a particular part of the text, they may articulate representations belonging to a certain discourse rather than to others. it. Second, only statements that employed the words 'human rights' and 'criminal law' (or connected terms) could be identified using the DA method. Certain voices that reject the deployment of penality but do not engage with the language of rights may have been overlooked. Third, besides the UN, the investigation was focused on Europe and England and Wales. Discussions of trafficking in other regions and countries of both the Global North and the Global South were not considered. While similarities cannot be excluded, the findings are not necessarily applicable to other contexts. Finally, it is acknowledged that the corpus is composed of a selection of texts. For instance, in relation to NGOs, only reports of large international organisationsmore likely to influence trafficking policywere explored, while documents produced by local NGOs or grassroots movements were not considered. The academic texts analysed are also a fraction of the large number of commentaries written on trafficking in the last three decades. Nonetheless, this analysis allowed me to identify four basic discourses that do not exactly correspond to the various approaches usually presented in the literature. Going beyond the simple division between a human rights and a crime-control approach, the analysis provided an opportunity to advance the understanding of how rights language is used to promote penal expansion.
Although my analysis centres on human rights and penality, it is important to bear in mind that in the past three decades, other discursive frames have been promoted in the anti-trafficking field. These additional frames include sex work and prostitution; migration, smuggling and border control; forced labour and modern slavery (Kotiswaran, 2019b, p. 55). Over the years, emphasis has shifted from certain frames to others, following both social change and dominant ideologies. The late 1990s and early 2000s were defined by a focus on sex trafficking and the association of trafficking with forced migration for sex work (Andrijasevic, 2007). Labour exploitation started being included in the dominant understanding of trafficking by 2009, due to the growing visibility of the International Labour Organisation's interventions on forced labour and the change of priorities (and presidency) in the US (Kotiswaran, 2019b, p. 63). This new phase rendered visible the competing frames of 'modern slavery' and 'forced labour'. Since 2014, trafficking interventions have been promoted explicitly in terms of slavery and forced labour both at the national and international levels (Kotiswaran, 2019b, pp. 64-68). Notwithstanding these developments, throughout the past three decades, trafficking has consistently been viewed as an issue of both human rights and criminal justice concern.
3 Formations of objects Since the 1990s, human trafficking has been the object of attention, debate and regulation from disparate groups, individuals and states. However, human trafficking remains an over-determined concept, around which everyone can mobilise while having a different understanding of the targeted phenomenon. Under Article 4 of the 2000 Protocol to Prevent, Suppress and Punish Trafficking in Persons Especially Women and Children (UN Trafficking Protocol), trafficking is composed of three elements: (1) an action -'the recruitment, transportation, transfer, harbouring or receipt of persons'; (2) certain means -'the threat or use of force or other forms of coercion, of abduction, of fraud, of deception, of the abuse of power or of a position of vulnerability or of the giving or receiving of payments or benefits to achieve the consent of a person having control over another person'; and (3) the purpose of exploitation. This definition is sufficiently broad to cover diverse situations and demand disparate responses (Chuang, 2014, p. 610). This section explores whether and how trafficking is framed as a 'crime' and/or a 'human rights violation' in the four discourses under analysis. It also examines how it relates to other phenomena and how understanding of trafficking is so constituted.

The 'law enforcement' discourse
The LE discourse understands trafficking as a criminal phenomenon, which, as such, may be a source of human rights violations. International treaties and legal instruments define human trafficking as a 'serious crime' (A/RES/73/146 2018). The UN Trafficking Protocol took the lead. While Article 3 provides for a definition of trafficking that is the definition of a criminal offence, Article 5 imposes on International Journal of Law in Context states a duty to criminalise. Subsequent legal instruments confirm the central role of criminal law. The EU Trafficking Directive describes trafficking as 'a serious crime, often committed within the framework of organised crime' (Recital 1). The Modern Slavery Act 2015 defines human trafficking as the offence of arranging or facilitating travel for the purpose of exploitation and punishes it with the maximum penalty of life imprisonment.
The LE discourse targets trafficking as a public moral wrong: its criminalisation relates to it being immoral, 'horrendous' and 'terrible' (Home Office and Scottish Executive, 2007, p. 4) and, as such, threatening the security of the state. From the 1990s to the 2010s, there has been a shift from an initial concern with prostitution to then including other forms of trafficking and associating it with the idea of modern slavery. Both sex trafficking and modern slavery appear as 'global health' risks (Pati, 2011, p. 139) that, through illegal immigration, corrupt the morality of Western culture (cf. Broad and Turnbull, 2019, p. 10). Accordingly, when the LE discourse addresses human trafficking as a crime that entails the 'violation or abuse of human rights' (A/RES/73/146 2018), the frame of 'human rights abuse' is only subsequent to the one of 'crime' and it is explicitly employed to highlight the moral repugnance of the offence (Kara, 2011).
In the UN Trafficking Protocol, trafficking in persons is not a simple offence but a form of transnational organised crime. Such conceptualisation seems obvious if we consider that the Protocol was adopted under the aegis of the UN Office on Drugs and Crime and supplementing the 2000 Convention against Transnational Organised Crime. In this way, not only is trafficking discursively associated with other global criminal trends (UN Global Plan of Action to Combat Trafficking in Persons 2010, Art. 43(c)), but its repression is made dependent upon the fight against other illegal phenomena, including smuggling of migrants, trafficking in arms and drugs, corruption, moneylaundering and terrorism (S/RES/2331 2016, para. 2(c); S/RES/2388 2017, para. 6). In the LE discourse, the element of 'organisation' is often related to the financial benefit traffickers may gain (European Commission, 2017). Various texts insist on how human trafficking is 'a high profit, low risk crime' (UN. GIFT, 2008) that needs to be fought 'just as vigorously' as other profit-driven offences 'are currently fought' (Pardo, 2009, p. 9).

The 'victims first' discourse
In the VF discourse, human trafficking is primarily a human rights issue (OHCHR, 2010, p. 4). The connection between trafficking and human rights is presented as something that occurs naturally through the 'definition' of trafficking (UN Secretary-General, 2003, para. 49). Both the LE and the VF discourses consider trafficking as a human rights abuse. While the former derives this characterisation from the seriousness of trafficking as a crime, the latter does so because it sees trafficking as caused by social conditions of poverty, discrimination, exploitation and inequality (European Commission, 2012, p. 2). The CoE Convention defines human trafficking as 'a violation of human rights and an offense to the dignity and the integrity of the human being'. In the words of the ECtHR in Rantsev v. Cyprus and Russia, 6 trafficking is a human rights abuse because, 'by its very nature and aim of exploitation, … [i]t treats human beings as commodities to be bought and sold' (para. 281). Within the VF discourse, there is also a tension between those who view sex work as a right (Pearson, 2000) and those who always view prostitution as a human rights abuse (Raymond, 2001).
In addition, the VF discourse treats trafficking as a criminal offence (CoE Convention,Art. 18). Yet the approach is opposite to the LE discourse. Human trafficking is a crime because it is '[o]ne of the most serious challenges facing human rights today' (UN Secretary-General, 2003, p. 2). Criminal law is here accessorial to human rights and used to provide further victims' protection (Piotrowicz, 2012, p. 184). Trafficking is not a criminal phenomenon because it affects public security and morality but because it causes private harm, namely violations of bodily, sexual and mental integrity of 6 (2010) Application no. 25965/04, ECtHR. other human beings. It is a 'crime against persons' (Skrobanek, 2000, p. 2) and an 'international crime' (Obokata, 2006, p. 37).

Counter-discourses
The frame of trafficking as both a crime and a human rights violation is criticised by counterdiscourses. The INC discourse treats human trafficking as a 'complex human rights problem' that should not be a matter for criminal law (Kaye et al., 2019, p. 16). In particular, it is suggested that trafficking 'has to be understood and addressed as part of the broader social, political and economic systems linked to migrants', women's and workers' rights' (GAATW, 2011, p. 5). This discourse also locates the core elements of trafficking in 'the presence of deception, coercion or debt bondage' (Foundation Against Trafficking in Women et al., 1999, p. 6), thus requiring a certain degree of force to be exercised against trafficked people. Accordingly, trafficking should not include sex workother than forced prostitutionand migration of individuals who are aware of their future work conditions (Smith and Mac, 2020, pp. 62-63).
For the TJ discourse, trafficking should be approached neither from a human rights nor from a criminal-law perspective. Rather, it should be understood 'as predominantly an issue of economic labor market exploitation' (Shamir, 2012, p. 81) or analysed through a 'postcolonial lens' (Sanghera, 2007, p. ix). The attempt 'to understand trafficking as a human rights violation'it is saidfails to capture the true 'harms and injustices of trafficking' that are connected to 'exploitative labor conditions, coercive processes of labor migration, global inequality, oppressive restriction on immigration …, and patriarchal gender roles and attitudes' (Christman, 2014, p. 321).

Formations of subject positions
Anti-trafficking discourses create knowledge about the people who are involved in the phenomenon of trafficking (Andrijasevic, 2014;Kapur, 2007). Both traffickers and trafficked people are rendered subjects of various forms of interventions and, in this way, provided certain identities that justify certain possibilities of action (Andrijasevic, 2007). Certain subject positions have an 'active' role: they are conferred 'the right to speak' and to define their role in relation to anti-trafficking initiatives. Other subject positions, such as 'the victim', have a 'passive' role: they are acted upon and determined by the speech and the practice of others. This section explores how the discourses under analysis frame the subjects they want to regulate, support or punish. family' (PK). 9 They come from underdeveloped or developing countries and are unaware of their prospective involvement in commercial sex work (Leidholdt, 1993, p. 135).
Although trafficked persons are presented as the bearers of human rights (UN Trafficking Protocol; EU Trafficking Directive), they are also portrayed as too helpless to exercise their rights independently or even to be aware of them ( International Journal of Law in Context trafficking -'[t]he consent of a victim of trafficking in person to the intended exploitation … shall be irrelevant' (UN Trafficking Protocol, Art. 3(b))but they are also deprived of political agency in relation to the enjoyment of their rights. In the LE discourse, victims of trafficking are conceived of as excluded from society and in need of reintegration by the state (UN Global Plan of Action to Combat Trafficking in Persons 2010, para. 26), even when they refuse to identify themselves as victims. Their voice is listened to only when it supports, and concurs with, the punitive action of the state, such as when victims act as witnesses in criminal proceedings. '[I]t seems implausible' -Williams J. says in O v. Commissioner of Police of the Metropolis 10that victims want 'no action against' their traffickers. Moreover, trafficking individuals do not enjoy protection and assistance just for the fact of being exploited but rather only once they are identified as victims (PK). 11 The process of identification has the performative role of transforming an individualseen as a prostitute or illegal migrant and treated as a criminalinto a victim of trafficking deserving state action to be rescued (cf. Lewis and Waite, 2019, pp. 232-236).

The criminal
If 'the victim' is the rights-holder, 'the trafficker' is 'the criminal'. While victims are presented as powerless, traffickers are described as organised and manipulative exploiters engaged in an immoral and illegal business. In the LE discourse, traffickers retain all agency: they are 'rational economic agents' (Kara, 2011, p. 141) who exercise 'control of the mind' over their victims (Centre For Social Justice, 2013, p. 32) with the purpose of financial gain or sexual gratification. Traffickers are also a threat to national and international security as they foster prostitution (QSA) 12 or illegal migration (A/RES/50/1995 1995) and 'have recourse to ever more sophisticated techniques, increasing financial resources and growing networks' (OSCE Action Plan to Combat Trafficking in Human Beings 2003).
The victim at times can become an offender, as it is not uncommon for people who have been trafficked to commit offences or become involved in trafficking themselves. The shift from 'victim' to 'criminal'and from protection to punishmentis constructed by conferring agency to the subject. As long as the trafficked person is a 'genuine' victim, 'she' is innocent and vulnerable. Once the same person commits a crime or is involved in illegal aspects of sex work, she is retroactively required to act as a 'reasonable person in the same situation' (Modern Slavery Act 2015, s. 45.1). In other words, through the commission of a crime, a trafficked individual ceases to be a 'credible' victim worthy of state protection and become a 'voluntary abuser' for whom prosecution is justified (R. v. LM). 13 4.2 The 'victims first' discourse 4.2.1 The victim In the CoE Convention, the category of victim of human trafficking is still circumscribed to 'especially women and children' (Art. 6). Victims are vulnerable due to human rights violations and in need of protection and assistance (Art. 1.1). They are also blameless, even when involved in unlawful activities, because they do not have control over their actions but are 'compelled' by overpowering criminals (Art. 26). The VF discourse also highlights the importance of victims' identification (VCL and AN v. United Kingdom, para. 160). 14 The phenomenon of trafficking is seen as involving the transformation of victims into criminals, such as prostitutes, illegal migrants, drug-mules and workers in cannabis farms (Whitehouse, 2013, p. 2). Only NGOs or the police can unmake such transformation when they rescue victims and recognise their suffering (OHCHR, 2002, Guideline 2.1). In so doing, they also make sure victims 'are treated as "rights holders"' (European Commission, 2017, p. 6 access to basic services, unemployed, dehumanised and penalised' (EOG). 15 Clearlythe VF discourse seems to sayhuman rights are mostly rights of victimised people, who have been subjected to cruelty and extreme suffering (cf. Aradau, 2008, p. 35). Yet the enactment of rights cannot be left to the victims themselves who 'are no longer free to decide their fate' (Pardo, 2009, p. 36) and often not even able to 'see themselves as "victims"' (Pearson, 2002, p. 32).
The VF discourse acknowledges that some trafficked individuals may be male adults (GRETA, 2012, p. 5) or even '"imperfect" victims' who cannot easily be distinguished from perpetrators (Ezeilo, 2012, para. 24). Nonetheless, the emphasis on vulnerability is maintained. The construction of victims as incapable of consent is explicitly fostered becauseit is saidit 'could improve the implementation of the anti-trafficking provisions and provide victims with greater confidence in self-reporting to NGOs and public authorities' (GRETA, 2018, p. 37). Agency can be restored once trafficked people decide to collaborate with authorities. Here they are supported 'to take an active and meaningful role in efforts to convict their exploiters' (Ezeilo, 2012, para. 91) or to 'bravely and generously' share their stories and experiences to inform and advise others (OHCHR, 2014, p. 2). Yet, ultimately, the discourse fixes them in their identity as vulnerable victims because 'the brutality and traumatization of trafficked sex slaves are unique and unrepairable' (Pati, 2011, p. 139).

The criminal
There are two types of criminals in the VF discourse: traffickers and corrupted public officials. Traffickers are perpetrators of human rights violations who act 'in flagrant violation of domestic laws and international standards' (A/RES/55/67 2000). They are seen as operating 'through secret and extremely well organized networks' (Warzazi, 2000, para. 90) and are described as 'very smart' because they change 'their behaviours as fast as countries change their laws to criminalise trafficking' (Anti-Slavery International, 2005, p. 20). Characterised by 'human greed and moral corrosion' (Pati, 2011, p. 140), they exercise control not only by using violence, but also by showing 'emotional attachment' to their victims (Kaye, 2003, p. 6). Public officials implicated in trafficking are deemed as dangerous as traffickers. These corrupted public officials are said to 'facilitate this trade through their inaction, inertia or occasional active involvement' (OHCHR, 2010, p. 76), but also to contribute to the victims' 'lack [of] confidence in the police and the judicial system' (OHCHR, 2002, Guideline 5).

Counter-discourses
The two counter-discourses are more conscious of their role in constituting subject positions. Accordingly, they shape the identities of people who are involved in the phenomenon of trafficking with a view to sustaining their critical agenda. Counter-discourses see trafficked people as essentially 'migrant workers' (Rijken and de Volder, 2009, p. 60) who 'make decisions about their lives, including the decision that working under abusive or exploitative conditions is preferable to other available options' (Foundation Against Trafficking in Women et al., 1999, p. 8). They are portrayed as individuals who 'combine multiple identities' (Sanghera, 2007, p. ix) and 'can both express consent and feel force in their migration decisions' (GAATW, 2011, p. 41). Trafficked people are 'anything but victims' (Sanghera, 2007, p. ix) and 'have the ability to regain control of their lives and make decisions based on their own interests and life projects' (Giammarinaro, 2020, para. 23). Consequently, they should not be rescued, but rather empowered so they can speak up 'for their own rights' (Wijers and Lap-Chew, 1999, p. 211).
In counter-discourses, trafficking is presented as a complex phenomenon. It is not always 'a case of criminal villains doing evil' (Shamir, 2012, p. 134), but rather a situation brought on by criminalisation of sex work, restrictive immigration rules, market forces and other structural factors that create vulnerability to exploitation. Counter-discourses present 'a common misunderstanding' that 'traffickers harm victims and governments rescue and protect them' (Pearson, 2000, p. 41 are not regarded as 'saviours, but oppressors' (Pearson, 2002, p. 33) that either subject trafficked people to serious rights violations (by criminalising sex workers and irregular migrants) or treat them as 'powerless pawns' (Dottridge, 2007, p. 1).

Formations of themes
The literature tends to distinguish between a crime-control and a human rights approach as solutions to trafficking. The former arguably addresses human trafficking as an issue of law enforcement. The latter appears to emphasise victim support over punitive measures. However, this distinction, though not unfounded, is somewhat misleading. In the context of anti-trafficking interventions, rights-based initiatives and criminal governance are often promoted and linked together both discursively and in practice. Governments use the language of rights to justify their law enforcement action; human rights actors rely on criminal sanctions to provide justice for victims. This section explores the roles that antitrafficking discourses ascribe to both rights-based interventions and penal measures.

The 'law enforcement' discourse
In the LE discourse, criminal justice is the primary instrument to end trafficking. Nonetheless, human rights have an important role to play. Acting as the 'sword' of criminal law (Van den Wyngaert, 2006), human rights provide a justification for crime-control measures and are rhetorically adduced as the goal of anti-trafficking strategies. The LE discourse constructs two relations between penality and human rights: (1) a far-reaching penality ensures greater human rights protection; (2) human rights provisions are incorporated into the existing punitive framework.

Penality as ensuring human rights protection
In the LE discourse, penality is construed as an essential element for achieving 'elevated human rights protections' (Kara, 2011, p. 123). In the UN Trafficking Protocol, the adoption of law enforcement provisions is explained with the need to protect victims' human rights. The 2010 UN Global Plan of Action to Combat Trafficking in Persons associates 'the promotion and protection of the human rights of victims' with 'the strengthening of the criminal justice response' (para. 3). In these and other documents, the deployment of the penal machinery appears as a core component of the state's human rights obligations towards victims: effectiveness in human rights protection is made dependent upon effective criminalisation, prosecution and punishment.
First, the creation of new offences criminalising a range of trafficking conduct is framed as a way to close the victims' 'vulnerability gap' between protection and exploitation (R. v. Connors). 16 In addition, Article 19 of the CoE Convention and Article 18.4 of the EU Trafficking Directive suggest that human rights are better protected when states criminalise the demand for victims' services. This approach is an extension of the Nordic model from the matter of prostitution (and the criminalisation of buyers of sexual services) (CATW, 2010) to other areas of human trafficking (GRETA, 2018, p. 55). Criminalisation of demand has at once a punitive and protective function. While those who buy and use victims' services are punished as part of the trafficking chain, human rights of the victims are reportedly upheld. Second, criminal prosecutions are regarded as 'remedies' for victims, which are 'triggered' by credible allegations of human rights violations (R. v. L). 17 To be successful, the 'maximum enforcement of the law' is required (UN. GIFT, 2008, p. 2). Conversely, when 'investigations are not launched', it is often claimed that 'victims do not receive the justice and support they need or deserve' (Independent Anti-Slavery Commissioner, 2017, p. 15). Finally, the LE discourse associates the promotion of human rights with a commitment to penal severity. proportionate and dissuasive sanctions' involving deprivation of liberty. The concept of proportionality is relevant here because it is a fundamental tenet of human rights law. In the context of human trafficking, this principleusually employed to anchor punishment within clear limitsis used to promote longer custodial penalties (OHCHR, 2002, Guideline 4.3).
In sum, crime-control measures are widely portrayed as having a rights-protective role. Such characterisation has a clear strategic purpose: it renders criminal measures less controversial and even desirable. By portraying human trafficking as 'a crime with a victim at the centre' (Centre For Social Justice, 2013, p. 13), the state's law enforcement machinery designed to contrast it is legitimised and strengthened.

Human rights in a punitive framework
The LE discourse establishes a second relation between human rights and penality: rights-oriented measures are generally embedded in a criminal justice paradigm. The UN Trafficking Protocol lays down various rights for trafficked individuals. However, victims' assistance and protection are mostly ensured by means of criminal proceedings. Moreover, while criminalisation provisions are obligatory, states are merely required to 'consider implementing' and 'endeavour to provide' measures for victimsand only 'in appropriate cases and to the extent possible under … domestic law' (Art. 6). 18 Pulling most assistance and protection into the criminal justice orbit has various consequences. First, it transforms crime-control measures into 'human rights enhancing measure[s]' (Home Office, 2014, para. 5). Take, for instance, the case of victims' detention, which in some cases is viewed as a way to protect victims from further harm. In TDT, 19 Underhill L.J. observes that trafficked people 'are at high risk of falling back into the control of their traffickers if released from detention'. Therefore, he continues, human rights obligations provide that victims 'should not' be 'released without proper protection against the risk of being re-trafficked'. Moreover, support for victims is made conditional on co-operation with law enforcement. In England and Wales, victims of trafficking are generally granted temporary residence permits and shielded from prosecution insofar as they can support criminal investigations. When this function ceases, they can be returned to their country of origin (MS) 20 or even tried and punished (R. v. L). 21 In this regard, the LE discourse accepts that trafficked individuals can be prosecuted: 'the status of a person as a victim of trafficking … does not automatically exempt him or her from criminal liability' (PK). 22 The LE discourse also has a tendency to assess the extent to which victims' rights are protected on the basis of conviction rates (Kara, 2011). The (relatively) small number of convictions is usually compared to the much higher number of trafficked people (Jarbussynova, 2015, p. 31). The low figures for trafficking-related criminal trials are deployed as evidence that state efforts to fight human trafficking and provide remedies to victims are unsatisfactory.
In sum, the LE discourse shows interest in human rights insofar as they serve crime-control goals. As stated in the background paper of a workshop organised by the UN Global Initiative to Fight Human Trafficking (UN. GIFT, 2008, p. 2): 'Protection and support measures for trafficked victims are not only necessary to respond to the violations of the victims rights, but also to support the law enforcement response to human trafficking.'

The 'victims first' discourse
The VF discourse supports a more 'holistic' approach that aims at safeguarding human rights through a solution encompassing prosecution, prevention and protection ('the three Ps'

International Journal of Law in Context
is primarily on victims' unconditional assistance and support, this discourse does not entirely diverge from the LE discourse and embraces penality as an essential element of human rights protection. The VF discourse uses human rights as both the 'shield' and the 'sword' of penality (Van den Wyngaert, 2006): (1) as a 'shield', human rights mitigate penal measures; (2) as a 'sword', criminal law enforces rights-oriented solutions.

Human rights as complementing penality
In the VF discourse, human rights are construed as tempering and correcting the state's deployment of penality. The UN Recommended Principles and Guidelines on Human Rights and Human Trafficking set out how human rights can be integrated within prevention and criminal-justice strategies. Criminal-law norms are also retained in the CoE Convention, but they are softened by prioritising '[m]easures to protect and promote the rights of victims, guaranteeing gender equality' (Chapter III) and by laying down a non-punishment provision for trafficked individuals (Art. 26). In the VF discourse, when trafficking is seen as a crime against public order, not only are victims less likely to receive redress or compensation (Pearson, 2002, p. 38), but they are treated as tools of law enforcement or deported as illegal migrants (Anti-Slavery International et al., 2003, p. 7). Conversely, by configuring trafficking as a human rights violation and a crime against the person, the VF discourse supports a 'victim-centred, gender-specific and child-sensitive' criminal law (European Commission, 2019). Criminal-justice measures are linked to victim support (OHCHR, 2010, p. 222), while ending impunity for traffickers and securing justice for victims become the main purposes of punishment (OHCHR, 2014, p. 17).
The VF discourse does not advocate for less criminalisation and punishment but for a different kind of criminalisation and punishment (Gallagher, 2010, p. 370). Confronted with the fact that traffickers are rarely arrested, investigated and punished, the discourse deploys state obligations under human rights law to put pressure on governments to mobilise their criminal law more effectively (Obokata, 2006, p. 35). The discourse still supports proactive prosecutions against traffickers but curbs criminalisation of victims (VCL and AN v. United Kingdom). 23 Simply put, a human rights approach 'ensure[s] that traffickers, rather than victims, are the ones put behind bars' (Annison, 2013, p. 14).
Integrating human rights in crime-control initiatives has another important role: it ensures that criminal-justice responses to trafficking are effective. The EU Trafficking Directive explains that the aim of victim protection is 'to safeguard the human rights of victims', but also 'to encourage them to act as witnesses in criminal proceedings' (Recital 14). In the words of the Inter-Agency Coordination Group Against Trafficking in Persons (ICAT) (2012, p. 8): 'in the course of a criminal investigation and prosecution, not only is the protection of victims right in principle but also right in practice as it is not effective to prosecute traffickers without placing the protection and assistance of victims at the heart of the intervention.' In conclusion, the VF discourse aims to demonstrate that 'the protection of the rights of trafficked people' and 'successful prosecution' are 'by no means contradictory' (Jarbussynova, 2015, p. 31). A human rights approach is regarded as 'a major precondition for effective investigation and prosecution' (Jarbussynova, 2015, p. 12) and as a way to correct the harshness of penalitynot to do without it.

Penality as complementing human rights
In the VF discourse, penality is just one element of a broader set of instruments to be applied in support of trafficked people (SM v. Croatia,para. 306;Rantsev v. Cyprus and Russia,para. 285). 24 Although the state's function does not end at criminalising the act of trafficking, penal measures 23 (2021) Application nos 77587/12 and 74603/12, ECtHR. still have an important role in the context of rights-oriented solutions. The VF discourse maintains that, by combining law enforcement with 'a rights-based, victim-centred approach' (Mann, 2011, p. 13), penality becomes acceptable and even desirable, while human rights are given weight and substance. While the LE discourse shows interest in human rights insofar as they serve crime-control goals, for the VF discourse the opposite is true. Penal measures are welcome to the extent that they comply with international human rights law. The UN Recommended Principles and Guidelines on Human Rights and Human Trafficking spell out the human rights standards applicable, in the contexts of criminal proceedings, to both trafficked people and those suspected of trafficking. The VF discourse also demands that 'trafficking cases are prosecuted and adjudicated fairly' and 'in accordance with international human rights' (OHCHR, 2010, p. 201).
Despite acknowledging the limits of penality, Chuang (2014, p. 641) argues that 'when pursed in a victim-centred, rights-protective manner, criminal justice interventions offer much needed accountability and restitution for egregious wrongs'. This statement summarises another function that the VF discourse assigns to penality, which is to enable states to use their methods of enforcement to vindicate human rights violations. The discourse in fact acknowledges 'the inherent political, legal, and structural weaknesses of the international human rights system' (Gallagher, 2009, p. 792), which alone has proven itself 'incapable of taking serious steps towards eliminating trafficking' (Gallagher, 2009, p. 847). First, the VF discourse recognises that trafficking would never have received the same level of attention from governments had it only stayed within the realms of the human rights system (Chuang, 2014, p. 641). Second, because human rights law is designed to hold states (rather than individuals) accountable, the VF discourse concedes that this framework is limited in addressing the harm done by traffickers (Knust and Lingenfelter, 2019, p. 4). Third, the discourse regards the human rights system as lacking proper enforcement mechanisms to uphold rights (Obokata, 2006, p. 36). Therefore, it accepts that the deployment of penality can help to compensate for these weaknesses. The government's willingness to use criminal law is seen as a demonstration of a serious commitment to dealing with trafficking (cf. Chacón, 2010Chacón, , p. 1626. Crime-control measures are regarded as tools which communicate that trafficking is a serious human wrong and, therefore, morally unacceptable (Mantouvalou, 2018(Mantouvalou, , p. 1019. Moreover, the prohibition of trafficking becomes directly enforceable at the domestic and international levels with the inclusion of individual criminal responsibility (Gallagher, 2009, p. 799).
A rights-oriented criminal law is also seen as a means of redressing the unequal distributions of power that underlie trafficking. The state's machinery of criminal justice compensates for trafficked individuals' vulnerability and 'change the equation of fear and power': 'when would-be traffickers are afraid of the consequences of their actions, potential victims start to become less vulnerable' (Centre For Social Justice, 2013, p. 150). The punitive machine is also discursively turned against the powerful, such as sex workers' clients and exploiters, public officials and multinational corporations who are implicated in trafficking. A rights-focused criminal law is said to pierce the shield of privilege and immunity of these subjects and to uphold equality before the law (Amnesty International and Anti-Slavery International, 2004, p. 10).
In sum, the VF discourse sees penality as a necessary component of a human rights response to trafficking. The underlying logic is that human rights regulate the states to ensure they take penal action, since human rights in themselves provide only for limited enforcement against trafficking.

Counter-discourses
Counter-discourses place themselves in a position of criticism vis-à-vis the relations between human rights and penality exposed so far. The INC discourse contends that penal policies are unsuitable for the nature of trafficking and are inevitably damaging the rights of trafficked people. The TJ argues that both human rights and criminal law are incapable of addressing the root causes of trafficking and thus seeks to advance a more transformative approach.

The 'incompatibility' discourse
According to the INC discourse, states generally address trafficking from a crime-control, rather than a human rights, perspective (Giammarinaro, 2020). The UN Trafficking Protocol, which was developed within a crime-prevention framework, is deemed to foster a 'law-and-order approach to trafficking' around the globe (Ertürk, 2009, para. 41). The INC discourse criticises immigration policies 'that restrict free movement', the 'criminalization of work in the sex sector' and the 'detention and deportation' of trafficked people (Ertürk, 2009, para. 40). Not only are criminal-justice responses deemed to be inadequate to 'cope with the systemic nature of exploitation' (Giammarinaro, 2020, para. 32), but they overshadow 'rights protections' (GAATW, 2011, p. 82). Prosecuting and punishing those who traffic or buy victims' services neither appear to 'offer protection to the victims, nor ensure that their human rights are respected' (Skrivankova, 2007, p. 224).
For the INC discourse, 'enforcing the law and upholding human rights' do not 'amount to the same thing' (Dottridge, 2007, p. 2). On the contrary, the discourse insists on going beyond 'law and order' (Ertürk, 2009, p. 39) and shifting 'the working paradigm from one of criminal sanction to human rights promotion' (Skrobanek, 2000, p. 2). The discourse calls for 'action to address the wider, more systemic processes or root causes that contribute to trafficking in persons, such as inequality, restrictive immigration policies, and unfair labour conditions' (Giammarinaro, 2015, para. 20). This 'authentic' human rights approach contrasts with the current models of protection offered to trafficked persons, whichit is said -'too often prioritise the needs of law enforcement over the rights of trafficked persons' (Pearson, 2002, p. 35). Such strategies are accused of 'treating the symptoms rather than the cause of the problem' (Kaye, 2003, p. 10), but also of unleashing a border control agenda 'in the name of human rights' (Sanghera, 2007, p. viii).

The 'transformative justice' discourse
The TJ discourse does not limit its critique to penality but extends it to human rights-based responses to trafficking. Unlike the INC discourse, it does not view the dominant approach as only limited to criminalisation and anti-immigration, but rather as a combination of a transnational crime framework with a rights-oriented approach. Yet the latter approach, far from correcting the limits of penality, is considered 'part of the problem' (cf. Kennedy, 2002).
The TJ discourse seems troubled that human rights language is deployed in such a way that disempowers the very same individuals whom anti-trafficking initiatives aim to help, whenever trafficked people are cast as vulnerable victims devoid of agency (Kapur, 2018, pp. 99-100). Human rights are also criticised for failing 'to deal with the economic, social, and legal conditions' that create people's exploitation (Shamir, 2012, p. 80). Despite the rhetorical power of the human rights framework - Shamir (2012, p. 94) observesthe latter 'helps few and, even for those few, to a doubtful extent'. The TJ discourse further notes that human rights promote an individualistic approach whereby social issues can be solved by tackling single acts of abuse and violence (cf. Kotiswaran, 2019b, p. 70). Moreover, any attempt to moderate penality through an insistence on victims' rights is said to 'reinforce, rather than challenge, the use of criminal justice framework' to address trafficking and sexual exploitation (Fudge, 2015, p. 20).
For these reasons, the TJ discourse seeks to advance new perspectives that would go beyond the language of rights and directly address the vulnerabilities of an unjust global economic order. Various proposals have been made, such as a labour (Shamir, 2012), a development (Kotiswaran, 2019a) and a post-colonial approach (Sanghera, 2007, p. ix). These responses to trafficking are presented as 'transformative' and better suited for addressing the structural conditions of exploitation.

Effects of anti-trafficking discourses
The form that anti-trafficking endeavours have taken in the past three decades has not been shaped by robust empirical studies but mainly by a discursive terrain made of assumptions about sex, exploitation, migration, (im)morality, crime and human rights (Kotiswaran, 2021, p. 46). This section explores how anti-trafficking discourses have materialised in concrete actions against trafficking. Of particular interest is how the relations that are discursively established between human rights and penality have been put into practice as part of the solutions to eradicate trafficking in human beings.

Dominant discourses
Anti-trafficking policy and practice are the product of the struggle and contestation between the competing discourses, with multiple goals and positions advanced at the same time. However, in this struggle, not all the discourses are instantiated in concrete anti-trafficking endeavours. The LE and the VF discourses establish between them an 'agonistic' engagement that presupposes one another's legitimacy. Conversely, they are in an 'antagonistic' relationship with the counter-discourses: each group treats the other as presumptively illegitimate (Mouffe, 2013). The dominant discourses have different aims and remain in contestation, but they accept and strengthen one another. The reason is because they are grounded on the same premises, namely the necessity for state intervention and the construction of penality as a fundamental tenet of human rights. Both discourses frame human rights as dependent upon state-control mechanisms and coercive action, which in turn are regarded as necessary conditions to achieve victims' rights and protection. As much as it is useful to try to disentangle the two dominant discourses, when they are translated into practice, they sustain each other and become hegemonic. As a result, their reiteration in different contexts and by different actors contributes to generating and perpetuating anti-trafficking endeavours that are imbued with a close connection between human rights and penality.
The call for action to address human trafficking is primarily accommodated by passing new statutes or adopting new treaties (OHCHR, 2002, Guideline 4;Centre For Social Justice, 2013). Far from being the result of a single mind and politics, anti-trafficking laws generally represent the combination of the LE and the VF discourses. This is true for the UN Trafficking Protocol and the CoE Convention, both of which originated from a debate between state representatives, troubled by organised crime and mass migration, and human rights activists, interested in protecting and securing justice for trafficked individuals (Ditmore and Wijers, 2003). Similarly, the Modern Slavery Act 2015 resulted from the lobbying of a wide range of human rights actorsfrom British and international NGOs to monitoring bodies like GRETAon the British government to establish a new anti-trafficking law (Van Dyke, 2019, pp. 66-67). The UK had already introduced the various offences of 'slavery, servitude, and forced or compulsory labour' in section 71 of the Coroners and Justice Act 2009. However, some decisions of the ECtHR, 25 together with criticism of the UK policy and its implementation, fuelled interest in the promotion of a more comprehensive act (Haynes, 2016, p. 37). The outcomethe Modern Slavery Act 2015is the perfect combination of the two discourses. It provides penal measures along with provisions for the protection of trafficked individuals, including a new defence for victims who commit an offence.
Demands for more state action have not only contributed to the enactment of new treaties and legislations, but have also resulted in heightened police control for the most marginalised, including sex workers and migrants (cf. Bernstein, 2018, p. 30). Emphasis on proactive policing as the solution to human trafficking does not only come from the LE discourse (EU Trafficking Directive, Recital 15). The VF discourse also supports increasing resort to covert surveillance, new technologies and hi-tech forensic tools (Ezeilo, 2012, para. 97), as well as the creation of police task forces 'specialised in the fight against trafficking and the protection of victims' (CoE Convention, Art. 29.1). In England and Wales, the police response was consolidated in November 2016, when the Home Secretary approved an investment of £8.5 million for the Modern Slavery Police Transformation Programme (NPCC, 2019, p. 8 improving investigative case work' (NPCC, 2019, p. 8). The critique that interception of communications, intensification of controls and intelligence-led investigations would result in further surveillance and criminalisation of the already over-policed is generally overlooked. '[T]he imagined victimization, rescue, and ultimately "empowerment"' of trafficked victims has in fact moralised anti-trafficking surveillance and made it human rights-compliant (Bernstein, 2018, pp. 145-146). In this way, human rights and surveillance have been co-constitutive. The latter has (at least rhetorically) provided better victim protection while human rights have served to moralise the extension of new modes of policing and legitimise these practices as human rights-oriented.
The construction of trafficking as an issue of human rights and penality not only allows the enactment of particular politics, but also constrains the prospect of different political actions (Aradau, 2008, p. 15). As the LE and the VF discourses have been widely adopted, ideas incompatible with their parameters have been set aside and dropped out of mainstream anti-trafficking initiatives. Casting the state's legal action as the solution justifies substantial allocation of resources to criminal-justice institutions and increased power to state officials. It also diverts attention from those exploitative practices the state's law and policy in fact enable, including criminalisation of sex work, policing of borders, unregulated labour markets and cuts to social protection programmes, such as public housing and welfare (Chuang, 2010(Chuang, , p. 1694. Moreover, positioning additional state regulation as the principal antidote discourages seeking (and funding) other non-state-led responses, including community interventions and long-term organising plans.

Counter-discourses
Counter-discourses argue that anti-trafficking legislation and related measures have had little to no impact on countering human trafficking. In many cases, the 'collateral damage' of anti-trafficking initiatives is said to be even greater than their success (GAATW, 2007). According to alternative voices, human trafficking is not (merely and only) what mainstream discourses say it is, namely an organised crime or a human rights abuse. Rather, human trafficking is a multi-faceted social issue (involving questions of migration, labour, race, gender, sexuality and political economy) that cannot be addressed (merely) by means of criminalisation and rights protection (cf. Nelken, 2010, p. 490). But is counter-discourses' critique ever translated into practice? While at the local level there is some evidence of virtuous examples, such as community-based initiatives to prevent sex trafficking (GAATW, 2018), counter-discourses remain generally unheard at the national and international levels (Gerasimov, 2019, p. 8).
Counter-discourses advance a powerful critique of state coercive intervention in the matter of trafficking, but they tend to remain a dead letter in terms of concrete actions to eradicate trafficking. States and international organisations rhetorically preach a human rights approach but continue neglecting what the INC discourse considers 'the true human rights abuses' (Leigh, 2015, p. 33). Transformative responses are praised by many academics but are rarely put into practice. As a result, human rights are not dislodged from the politics of crime and alternative models of justice are not implemented. Human rights remain as interwoven as ever with penal agendas (Bernstein, 2012). The reasons why counter-discourses have only a marginal role in practice are varied. Not only are alternative voices fewer in number than hegemonic ones, but their antagonistic and criticalrather than agonistic and normativeapproach reduces their ability (and, at times, willingness) to influence policy-making. They are also perceived as too radical or at odds with states' political and socio-economic interests or concerns. Moreover, as observed by Quirk (2020), the fact that critical voices are formulated as part of an antagonistic stance towards mainstream approaches renders the current anti-trafficking framework the foundational starting point of any critique and alternative vision. In this way, counter-discourses tend to operate within a symbiotic relationship to dominant discourses and practices, which, in turn, limits the ability of critical voices to generate authentic change.

Conclusion
Drawing on the findings of DA, this paper has highlighted how the relationship between human rights and crime control develops within anti-trafficking discourses. Dominant discourses have shaped the meaning of human rights as sources of penality, while alternative voices have questioned this relationship, albeit without succeeding in dismantling it. Human rights are thus constructed to align with the state's coercive action, which in turn becomes a necessary element for the fulfilment of rights.
The LE discourse uses human rights to legitimise penality and incorporates rights-oriented measures in a criminal justice paradigm. This approach is not (merely) a misapplication of human rights language to preserve the social-moral order of the state, but it is often accommodated by those very human rights law and principles. For instance, appeals to the principle of proportionality have led to the promotion of longer penalties on the basis that they are necessary both to reflect the severity of trafficking and to deliver justice to victims. In a different way, the VF discourse embraces penality as an instrument of rights protection. Criminal lawa criminal law that respects victims' rightsis relied upon for promoting human rights, saving victims and providing justice for the most vulnerable. Here, it is not the function of human rights that is reoriented, but the role of criminal law, which ceases to appear primarily as a social control institution and becomes an instrument for the expression of social and moral value (Garland, 1990).
Though apparently in opposition, the LE and the VF discourses become entangled and bolster each other. While the state is designated as the protector of both national security and victims' human rights, its crime-controlling arm is reasserted as the solution to human trafficking. In this context, human rights have not (only) been co-opted, but in fact have been integral ingredients in demanding the state's coercive intervention (Bernstein, 2018, p. 66). Either constructed as a language of legitimisation of state power or made dependent on it, human rights have increasingly served to facilitate, rather than counter, the strengthening of the criminal justice system (Musto, 2010, p. 387).