Frames and consensus formation in international relations: The case of trafficking in persons

Abstract

This article examines the process of consensus formation by the international community regarding how to confront the problem of trafficking in persons. We analyze the corpus of United Nations General Assembly Third Committee resolutions to show that: (1) consensus around the issue of how to confront trafficking in persons has increased over time; and (2) the formation of this consensus depends upon how the issue is framed. We test our argument by examining the characteristics of resolutions’ sponsors and discursive framing concepts such as crime, human rights, and the strength of enforcement language. We conclude that the consensus-formation process in international relations is more aptly described as one of ‘accommodation’ through issue linkage than a process of persuasion.

Keywords

Human rights international consensus trafficking in persons transnational crime United Nations

Introduction

The past three decades have seen two interesting global trends: renewed attention to human rights and an increased effort to combat transnational crime. Trafficking in persons raises issues precisely at the intersection of these two trends. It involves both serious violations of the rights of persons being trafficked — the right to security of the person, the right to be paid and to work under decent conditions, even potentially the right to life — as well as the commission of serious transnational crimes, from kidnapping, to money laundering, to fraud. Despite the complexity of the issue, there has been a fairly swift convergence among states and other actors that trafficking in persons is a serious problem that should be addressed both in international law and domestic criminal statutes.

How do we explain the recent rapid agreement by many countries to criminalize the exploitative trans-shipment of people across borders? Our central hypothesis is that issue framing has been critical in the process of consensus formation. We examine evidence of issue framing and normative convergence in the one forum in which every state on earth has a right to participate, and on which non-governmental actors also have some influence: the United Nations General Assembly (UNGA). Successive UNGA resolutions expose the process of consensus formation that makes binding agreements possible and sustainable.¹ We use resolutions of the UNGA Social, Humanitarian and Cultural Affairs Committee (the ‘Third Committee’) to show that: (1) consensus on the best approach to confront trafficking in persons has increased over time; and (2) the development of this consensus has depended upon how the issue was framed, with a ‘crime-fighting’ frame generating more support than a ‘human rights’ frame, especially in the crucial period leading up to the adoption of the Protocol to Prevent, Suppress and Punish Trafficking in Persons (TIP Protocol) in 2000.²

The argument is tested by literally watching consensus grow over time in the language of 12 resolutions on trafficking in women and girls adopted by the Third Committee between 1994 and 2012.³ Since there is no voting record on these resolutions, the formal record of which countries co-sponsored and endorsed each resolution is used to trace how the changes in content correlate with patterns of endorsement over time. We first analyze the evolution of these 12 resolutions and then estimate logistic models of support patterns for 150 countries for a total of 1800 country-year observations. We find that framing trafficking in persons as a common threat (linked to transnational crime) encouraged more states to support stronger resolutions than framing the issue in terms of common responsibilities (to protect/respect human rights). This suggests that the consensus-formation process is more aptly described as one of accommodation of differing values rather than a process of persuasion toward a single normative approach. Even in an area as imbued with normative significance as the trafficking of human beings, consensus is not necessarily achieved by changing minds or values, but by accommodating the values of a broad range of states.

Our findings contribute to a significant strand of research that explores the extent to which issue framing contributes to policy formation (Bleich, 2002). While our conclusion will hardly surprise those who view ‘securitization’ as a powerful driving force in world politics (Buzan et al., 1998), it is a useful reminder that international consensus may have more to do with how issues are framed rather than persuading others to do the right thing. The analysis may help shed light on the development of international consensus on other policy issues, from migration and human smuggling, to food security and HIV/AIDS.

The article begins with a description of trafficking in persons and the context of rapid globalization in which it has been discussed over the past two decades. The next section theorizes consensus formation, and develops a model of how consensus is achieved internationally. This section focuses on issue framing, which is especially critical under conditions of uncertainty and fluidity, such as those facing states in the early post-Cold War period. The third section presents evidence from UNGA resolutions that consensus on a new framework to confront trafficking in persons developed rather quickly, and that this can largely be attributed to framing the issue in terms of its links to transnational crime. Evidence at both the level of the resolution and at the level of each state’s decision to support a resolution suggests that while human rights concerns are important, crime frames have elicited more and stronger support from states. The final section concludes with the implications for reaching consensus on complex and contentious international problems.

The international response to trafficking in persons

Trafficking in persons is a multidimensional issue involving most countries of the world to some degree. The United Nations (UN) claims that it is the third most profitable sector of organized crime, after drug smuggling and illicit arms transfers.⁴ While internal trafficking is also very common, victims trafficked from over 136 countries were found in 118 countries worldwide between 2007 and 2010 alone.⁵ The extent of the problem has only compounded the potential diversity of approaches to confront it.

Trafficking in persons is hardly a recent phenomenon. States have tried to cooperate on this issue for over a century. A series of early international agreements, including the International Agreement for the Suppression of White Slave Trade (1904)⁶ and the Convention for the Suppression of the Traffic in Women and Children (1921),⁷ were primarily aimed at coordinating efforts among states to enforce national anti-prostitution laws. The former called for the notification of authorities of ‘persons in charge of women and girls destined for an immoral life.’⁸ It did not, however, mandate much positive action to insure their protection from broader harm. The 1921 agreement has similar purposes (and explicitly refers to the earlier agreement), though it drops the racist language and apparently broadens its focus beyond prostitution, calling for some regulatory oversight of employment agencies ‘to ensure the protection of women and children seeking employment in another country.’⁹

The Second World War brought these codification efforts to a standstill, but in 1949, the UN passed a resolution that entered into force in 1951 as the Convention on the Suppression of Trafficking in Persons and the Exploitation of the Prostitution of Others.¹⁰ Without the overt racism of the early 20th century, this agreement again posed prostitution of women and children as the central problem. Transnational trafficking — moving persons via criminal networks from one place to another — was secondary. Despite the massive displacement of the war years, international support was tepid: a decade after opening for signature and ratification, it had garnered only 25 state parties.

Nearly half a century would pass before the international community made another serious effort to tackle trafficking in persons, this time reaching a sweeping agreement in only a few years’ time. The TIP Protocol (2000) defines trafficking in persons as:

the recruitment, transportation, transfer, harboring or receipt of persons, by means of 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, for the purpose of exploitation.¹¹

Furthermore, it views ‘the recruitment, transportation, transfer, harboring or receipt of a child for the purpose of exploitation’ as trafficking in persons, regardless of the means used.¹² It obligates state parties to criminalize trafficking in persons in their national statutes,¹³ and to protect victims’ privacy and identity ‘in appropriate cases and to the extent possible under domestic law.’¹⁴ As of June 2013, more than 150 states have ratified the TIP Protocol and 140 countries have criminalized sex and labor trafficking in their national law.¹⁵

Something had clearly changed by the 1990s — but what? One answer lies in the normative development in the field of human rights. From the 1970s to the 1990s, international legal instruments guaranteeing human rights — both ‘hard’ and ‘soft’ law — burgeoned in a range of areas, from torture prohibition, to women’s rights, to the rights of children (Simmons, 2009). Rights advocacy groups bloomed and influenced norms, law, and practice worldwide (Clark, 2001; Korey, 1998). Rights claims have been credited with creating difficult-to-resist spirals (Risse et al., 1999) and boomerangs (Keck and Sikkink, 1998), which eventually contributed to holding governments more accountable for human rights violations than had ever been the case in the past. Indeed, some scholars have written about the latter decades of the 20th century in terms of a veritable ‘rights revolution’ (Epp, 1998).

Another critical contextual change has been the economic liberalization and, for many, dislocation associated with the end of the Cold War and, indeed, globalization writ large (Bhattacharyya, 2005). The breakdown of the Soviet Union opened borders to the flow of people and goods, both legal and illicit, at unprecedented rates. New crime routes evolved and fueled networks involving trafficking in drugs, people, and weapons, and associated crimes like money laundering, corruption, and possibly even terrorist financing (Thachuk, 2007). Transnational crime — estimated at USD730 billion a year worldwide¹⁶ — was increasingly viewed as a threat to legitimate local economies, with the potential to undermine law and order. Developing and transitioning states were particularly vulnerable to this wave as they often lacked the capacity to police, prosecute, or secure borders against large criminal networks (Lloyd et al., 2012).

The global organizational response to trafficking in persons reflects these two contexts. As early as 1985, the Seventh United Nations Congress on the Prevention of Crime (which meets every five years) began to focus discussions on the global threat presented by organized crime. By the Eighth Congress in 1990, both newly independent and established states were demanding new international institutions to confront it. The first step in this direction was the establishment of a Commission on Crime Prevention, followed by a World Ministerial Conference on Organized Transnational Crime, convened in 1994. This conference attracted an unprecedented number of attendees — 142 state delegations (86 at the ministerial level) — in addition to intergovernmental and non-governmental organizations. The conference agreed to establish a group of experts with the mandate to draft an international instrument on transnational crime. Operating out of the United Nations Office on Drugs and Crime (UNODC) in Vienna, this group held 11 sessions over a two-year period, from which the draft of the first-ever ‘international crime bill,’ the United Nations Convention against Transnational Organized Crime (UNCTOC), emerged.¹⁷

Parallel to these discussions (referred to as the Vienna Process), a series of debates on trafficking in persons took place in a number of UN bodies, but particularly in the Third Committee of the UNGA. Between 1994 and 2012, the Third Committee debated and adopted a series of 12 resolutions focused on Trafficking in Women and Girls (1994–2012) and four resolutions on the coordination of efforts against trafficking in persons (2006–2012).

These developments reflect an extraordinarily broad and speedy (by diplomatic standards) move toward a legally binding response to human trafficking. In about five years, the international community moved from 45 years of silence to a binding treaty obligating states to prevent human trafficking, to protect its victims, and, most especially, to criminalize trafficking and prosecute traffickers. The consensus is fragile, yet, as of this writing, has been maintained. This raises a question central to world politics: how has this consensus been possible, given the sensitive and complicated nature of the problem of trafficking in persons? The answer will provide insights into how consensus among heterogeneously situated actors develops at the international level.

A theory of consensus formation: The importance of framing

Issue framing and normative consensus formation

We hypothesize that issue framing has had a great deal to do with consensus making on trafficking in persons — and potentially many other international issues as well. Framing can be understood broadly as ‘a way of selecting, organizing, interpreting, and making sense of a complex reality to provide guideposts for knowing, analyzing, persuading, and acting’ (Rein and Schön, 1993: 146). Whether employed consciously or not, frames allow individuals to filter information to make some of its aspects more salient than others.

A growing literature addresses the role that issue framing plays in social and political processes, at all levels of human interaction. Political psychologists have long recognized that individuals’ attitudes are quite susceptible to issue framing (Tversky and Kahneman, 1981). Framing experiments demonstrate that the ‘lens’ through which a respondent views a question torques his or her attitudes on a broad range of questions (Chong and Druckman, 2007). Framing effects are said to occur when ‘[in] describing an issue or event, a speaker’s emphasis on a subset of potentially relevant considerations causes individuals to focus on these considerations when constructing their opinions’ (Druckman, 2001: 1042). Frames have the potential to change the way individuals deliberate and what they believe to be important (Nelson and Oxley, 1999). They are especially influential when promulgated by sources the individual views as credible (Druckman, 2001) and when they are taken up by the modern media (Iyengar and Kinder, 1987; Scheufele, 1999).

Framing has also been a prominent concept in sociology. Sociologists have documented the role of framing, especially during times of change as actors struggle to come to terms with how the world ‘works.’ Actors draw on framing to create a shared understanding of events that both legitimate and motivate action (Gamson and Modigliani, 1989; McAdam et al., 1996). Since issue frames can be critical to the success or failure of social movements, they are sometimes deployed strategically by groups vying to organize supporters and to motivate action by challenging existing frames and replacing them with galvanizing ways to view the world (Benford and Snow, 2000). Snow and Benford (1988) identified ‘frame alignment’ — frames that are linked in congruency and complementariness — as an important element in social mobilization, and argue that social change is more likely when multiple frames resonate widely to inform and motivate behavior. The incorporation of both prosecution and victim protection into the ‘3P’ (protection, prosecution, and prevention) framework of the TIP Protocol, for example, can be interpreted as a result of bringing together the highly salient frames of social threat and human rights.

Issue framing is also important in understanding relations between states. Security and foreign policy research is rife with studies of the ways in which adversaries are framed so as to rouse domestic audiences to support aggressive policy positions (Mintz and Redd, 2003). At a deeper level, constructivist theorists stress that interpretation is an essential aspect of the construction of reality, and therefore focus much of their analytic attention on how this reality is socially constructed (Adler, 2002). The securitization literature is consistent with these themes, but stresses ‘the process through which an issue is presented as an existential threat’ as a way to motivate extraordinary behavior (Buzan et al., 1998: 24). Although the securitization literature in International Relations (IR) and framing literatures in psychology and sociology have developed largely in isolation from one another, these approaches all perceive actors as (often strategically) constructing problems through discursive practices, and emphasize the role of ‘audience, communicator and culture’ in the discursive process (Watson, 2012: 284).

Overall, literature from several disciplines fairly conclusively supports the idea that framing influences individuals’ attitudes and possibly even their political and social behavior. While frames have been criticized as shallow efforts to market policies and political ideas (Oliver and Johnston, 2000), they are also powerful tools for shaping the process of consensus building. If the framing of issues can have a significant impact on people’s attitudes, their political behavior, and their policy preferences, there are ample reasons to believe that framing can have a strong effect on the formation of policy consensus. How an issue is framed, represented, and discussed can have a powerful effect on the likelihood of reaching agreement among actors who may have very different initial preferences and perspectives.

Framing trafficking in persons: Human rights and transnational crime

Several frames are possible contenders for interpreting the problem of trafficking in persons (Pajnik, 2010), and a range of frames have gained in salience over time (see Locher, 2007). Historically, the highly gendered victim protection frame linked trafficking to prostitution, as illustrated by international treaties of the first half of the 20th century. Championed by religious and women’s advocacy groups and some states, this frame views trafficking in persons as a matter of protecting the vulnerable from exploitation, typically of a sexual nature, and (compared to the human rights frame, discussed next) de-emphasizes (though does not deny) the autonomy and rights-bearing nature of the trafficked individual.¹⁸

This frame predominated through at least the 1950s, but is visible today in the TIP Protocol, which gives considerable attention to victim protection.¹⁹ Nonetheless, by the late 1990s, the victim protection frame was weakened somewhat by divisions over the acceptability of ‘sex work’ — as reflected in debates over whether prostitution is an individual choice of profession (Doezema, 1999; Farrell and Fahy, 2009) or an inherently abusive activity from which individuals should be protected. These divisions were most clearly demarcated between conservative and liberal non-governmental organizations (NGOs) (Kempadoo and Doezema, 1998; Scarpa, 2008), but also characterized differences among states, as the Netherlands legalized and Sweden criminalized prostitution (Di Nicola, 2009). The frame was also weakened by the lack of distinction in some cases between trafficking victims deserving of protection and illicit immigration. Support for trafficking victims was lessened to the extent that they could themselves be cast as lawbreakers deserving of immediate repatriation.

Debates in the 1990s reflected the growing salience of another frame: the human rights frame. Along with globalization, the rights revolution in international and national law from the 1960s provided a compelling new way to think of trafficking in persons. Through the lens of human rights, trafficked persons are not simply ‘vulnerables’ to be protected; they are individuals with agency that need to be respected. While rights framing continues to view women and children as especially vulnerable, it moves beyond the focus on prostitution and gender to draw attention to the full spectrum of human rights violations that trafficking in persons involves: from labor violations, to violations of freedom of movement, to inhumane treatment, to (in the extreme) the right to life itself. The human rights frame emphasizes the coercive aspects of trafficking in persons and even the slave-like conditions in which a good many trafficked individuals are held.

The language of human rights was, and continues to be, a highly salient way to think about the problem of trafficking in persons. The Special Rapporteur on Violence against Women is an exemplary voice for the human rights frame. Building on the dense international legal structure of the past 40 years, in 1997, the Special Rapporteur criticized the narrow definition of trafficking found in the 1949–1951 agreement, referring to its understanding of trafficking in persons as ‘ill-defined’ and ‘uniquely abolitionist.’²⁰ She chose to describe the plight of persons trafficked in terms redolent of human rights abuses that had gleaned widespread support internationally, including slavery and torture. Referring to the practice in terms of ‘modern-day slavery’ (Bales, 2005; Bales and Soodalter, 2009) is an especially graphic way to emphasize that trafficking in persons is a serious violation of human rights.

The human rights frame understands trafficked persons as bearing rights to work and migrate without loss of dignity.²¹ In common with the victim protection frame, the human rights frame focuses on the plight of the individual and obligates states. It holds state officials responsible for preventing coercive bondage, but also for fully respecting the rights of individuals in every aspect of prevention and law enforcement, drawing heavily on international human rights law.

Non-state actors have been among the most determined to frame trafficking in persons as a human rights abuse. In addition to the Special Rapporteur, the human rights frame has been articulated by several non-state actors, including Amnesty International,²² the Office of the High Commissioner for Human Rights (OHCHR), and international agencies including The United Nations Children’s Fund (UNICEF) and the International Organization for Migration (IOM) (Gallagher, 2001). These organizations have urged states to see the problem from a rights perspective, calling on them — with limited success — to assure the protection of the rights of trafficked persons to remain in the destination country rather than forcibly returning them to their country of origin (Gallagher, 2001). However, as the Special Rapporteur herself has noted, states have few incentives to genuinely embrace protection or rights frames, as source countries have economic incentives to encourage remittances and destination countries often find it easiest to simply repatriate foreign individuals who have been exploited on their soil.²³

The (often gendered) victim protection frame, like the human rights frame more generally, focuses on the trafficked individual, and we therefore include both of these as elements of a human rights frame in our empirical work. In contrast, a third frame — the transnational organized crime frame — emphasizes traffickers as a challenge to state authority and societal well-being. This frame situates trafficking in persons firmly within the broader problem of criminal networks that transcend national borders, linking it to globalization, illicit labor migration (Hughes, 2000; Salt, 2000), and money laundering or migrant smuggling networks.

The crime frame views trafficking as corrosive of state authority, and even sees it as a potential national security threat (Farrell and Fahy, 2009; Thachuk, 2007; Vlassis, 2000). Early treaties tapped this frame, but only lightly, referring to victim ‘protection against the criminal traffic known as the “White Slave Traffic”’ (1904).²⁴ Later agreements refer to the ‘offence’ (1921) of trafficking for purposes of prostitution and call for ‘punishment’ (1949), without linking this offense to broader transnational threats. Emphasis clearly changed in the 1990s, even at the highest policy level. Former US President Bill Clinton first raised the concept of transnational organized crime as a global security threat in 1995 (Van Dijk, 2011) and the major European countries endorsed his analysis at a Group of 8 (G8) meeting in Lyon shortly thereafter. More than a decade later, Deputy National Security Adviser Denis McDonough noted that trafficking in persons ‘fundamentally endangers [the] international security’ of the US, by fueling transnational crime and potentially financing terrorist activities.²⁵ The US media has increasingly framed trafficking in persons as a matter for crime-fighting, law enforcement, and prosecution, with mixed consequences (Farrell and Fahy, 2009). There is evidence that the global media follows the same trend, as Figure 1 demonstrates.

Figure 1.

Framing of trafficking in persons in the global media.

The transnational crime frame resonates in Eurasia as well. From Europe to Asia, the fear of unwanted migration has raised concerns generally about migrants and asylum seekers as challenges to state sovereignty and potential criminal elements (Curley and Wong, 2008; Huysmans, 2000; Ibrahim, 2005), which, in turn, has led to enhanced policing capacities against ‘the other’ throughout Western Europe (Loader, 2002). Links between crime and migration have played into fears in countries of origin as well. Some campaigns to discourage unwanted migration from Eastern Europe, for example, have purposefully sought to associate it with trafficking in persons (Nieuwenhuys and Pécoud, 2007). As a result, some have argued that the predominant security paradigms have gradually shifted from war-fighting to crime-fighting (Andreas and Price, 2001). Some scholars even contend that this shift marks a qualitative change in the conduct of foreign and security policy (Kaldor et al., 2007).

Developing countries also embraced the transnational crime frame, and initiated early discussions to address it. Well before the TIP Protocol was finalized, representatives from Guinea claimed that the ‘trafficking of women had been integrated into all forms of organized crime,’ and called for ‘strict enforcement measures,’²⁶ while Ghana called for law-enforcement cooperation through regional organizations and Interpol.²⁷ More recently, leaders of the Dominican Republic have lumped human trafficking together with gun trafficking as an international cooperative law-enforcement priority.²⁸ Many developing countries desperately need resources to bolster their authority, but also to stem the social harms resulting from organized crime. National security concerns have thus created additional incentives for devoting resources to combat trafficking for a broad range of states. Trafficking in persons has been easy to interpret through the crime and security frames of the post-Cold War world, in a way consistent with much of the literature on ‘securitization’ (Balzacq, 2005; Buzan et al., 1998; McDonald, 2008; Stritzel, 2007; Wæver, 1995).

What produces consensus? Expectations about the power of specific frames

By the 1990s, the international community had not spoken authoritatively on trafficking in persons in 45 years, when it had cast the problem largely in terms of prostitution. The Cold War was over, international human rights had a firm legal footing and advocacy network worldwide, and transnational crime was increasingly understood as a serious challenge to state authority and even security. States varied significantly in their exposure to trafficking in persons, their commitment to human rights, their cultural orientation toward such things as sex work and child labor, and certainly in their economic and legal capacity to enforce border controls and prosecute crime. Given this heterogeneity, how has it been possible to form and maintain some degree of consensus on trafficking in persons?

We argue that state consensus regarding this complex issue was largely driven by framing the problem in terms of crime-fighting. Criminalization asserts the authority of the state versus transnational challenges; it justifies resourcing the police power of the state, beefing up border controls, and even seeking international aid to do so (as many developing countries emphasize). Human rights guarantees, on the other hand, create obligations toward individuals — often, foreign individuals. Many states perceive that they are saturated with rights obligations and eschew any additional explicit rights obligations, especially toward non-citizens. This is not to deny that a good many states take human rights seriously; rather, we argue that the inherently state-empowering crime-fighting approach is much more likely to appeal to a broader coalition of states than are human rights obligations toward foreigners.

We do not claim that human rights justifications for strong anti-trafficking norms were irrelevant to consensus formation; far from it. By the 1990s, NGOs, some progressive states, and the OHCHR were strong advocates for the rights of trafficked victims, while many states balked at efforts to impose yet another set of legal, institutional, and financial obligations. In order to get a consensus, the crime-fighting frame was crucial. It has brought more states on board, and made possible an integrative approach that accommodates both rights and crime-fighting. The evidence suggests that concerns about victims (rights and protection) were accommodated into the integrative ‘3P’ approach primarily because of the growing commitment to crime-fighting. The crime frame literally helped grow the coalition of states against trafficking in persons. We demonstrate this in the following through an analysis of the adopted UNGA resolutions leading to, and following the adoption of, the TIP Protocol.

Consensus formation: Evidence from the UN

Data and expectations

Since frames are often presented in communicating texts (Entman, 1993: 52) and ‘issues are securitized through discursive “speech acts”’ (Jackson, 2006: 301), our empirical investigation focuses on resolutions of the UN — the major global forum in which states have the opportunity to express a broad range of attitudes. ‘Consensus’ is a collection of attitudes and is therefore difficult to observe directly, but expressed support in a formal public forum such as the UN is a useful proxy for attitude formation and convergence. These resolutions are especially useful because consensus formation is a process that unfolds over time, as states struggle to define the nature of the problem they are facing and the alternatives for confronting it. State support for various communicative texts in this series represents an observable and roughly comparable indicator of attitudes over time expressed in a controlled setting. Unlike studies of treaty ratification (Lloyd et al., 2012), these agreements reflect the process that produced the adopted treaty in the first place.

Our database is comprised of the universe of resolutions adopted in the UN Third Committee from 1994 to 2012 on ‘trafficking in women and girls.’²⁹ These resolutions do not relate exclusively to females, despite this label; beginning in 1995, they also recognize the victimization of males. We focus on these resolutions to keep the data-generating process constant, as these 12 documents represent the longest sequence of trafficking resolutions in the Third Committee.

Unfortunately, these resolutions were adopted without recorded votes. The best indicator we have of consensus formation is the change in the content of resolutions over time and the record of which countries introduced and formally endorsed each resolution for consideration by the Third Committee. Endorsement is quite meaningful: it requires bureaucratic attention to an issue and therefore is not a costless gesture. States that support the introduction of a resolution go on record as supportive of its general purposes and its specific priorities. Unfortunately, there is no record of resolutions that failed or that were deterred from being introduced. Nonetheless, we are able to analyze the language associated with resolution support and observe endorsement patterns over time.

To preview our findings, crime language is associated with more sponsors, more diverse supporters, and stronger language in general than is the human rights frame. Crime language is also more likely than human rights language to elicit state endorsement. These results are remarkably robust across a diverse range of states, although we do find some evidence that destination and transit states, in particular, are attracted to a strong law-enforcement approach. Taken together, the evidence of state support suggests that consensus on how to respond to trafficking in persons was facilitated by framing the issue as challenging the authority and security of states and their societies. Human rights could be accommodated into the integrative ‘3P’ approach because of the growing emphasis on crime-fighting.

Evidence of consensus formation

We begin by reviewing the evidence that international consensus regarding trafficking in persons has been building over time. After a 45-year hiatus, beginning in 1994, a resolution on trafficking in persons has been adopted roughly every one to two years. Also adopted after the TIP Protocol entered into force in 2003 are resolutions focused on improving the coordination of efforts against trafficking. This is a basic indicator that states agree, at a minimum, that the problem of trafficking in persons belongs near the top of the international agenda. Moreover, the number of states sponsoring resolutions has increased since 2000 and remains at a high level, hovering around 100 co-sponsors since 2006. That is, by the 2000s, a majority of UN members have been willing to sponsor resolutions that address the issue of trafficking in persons, as Figure 2 demonstrates.

Figure 2.

Sponsorship of resolutions on women trafficking over time.

The diversity of states that support anti-trafficking resolutions has also increased over time. Figure 3 plots the number of resolution sponsors and endorsers by World Bank income category. Anti-trafficking resolutions have clearly not been the pet project of the wealthier destination states. These trends indicate that the consensus against trafficking has generally been high and has retained its diversity as the movement has gathered steam.

Figure 3.

Sponsors of women-trafficking resolutions by income groups.

Another critical indicator of consensus is the content of the resolutions themselves. When international consensus is low, we should expect a series of resolutions representing quite disparate views, reflecting the variety of ideas and concerns. Low consensus should also reflect frequent language shifts from one resolution to another. As consensus begins to form, we should expect resolutions to converge on similar language, reflective of the shared ideas, interpretation of the problem, and the best solution to the problem.

To test this idea, we analyzed changes in the language of resolutions over time,³⁰ comparing the use of word stems from one resolution to the next (disregarding word order), and plot the differences chronologically in Figure 4. Resolution language changes less and less over time, as relatively minor adjustments are necessary to glean support. This is especially true after the passage of the 2000 TIP Protocol. Language similarity has increased even as resolutions are getting longer and more detailed, which suggests convergence on their increasingly precise content over the past two decades.³¹

Figure 4.

Cosine similarity of women-trafficking resolutions.

But what, exactly, are these resolutions converging on? One indicator of growing consensus might be the strength of language that calls on states to take particular actions. As a first cut, we examined the resolutions for ‘pressure language’ — clauses that are a call to action. Figure 5 graphs the frequency of two phrases used to identify such clauses over time: ‘urges’ and ‘calls upon.’ There is a nearly linear increase in the frequency of these words over time in these resolutions.

Figure 5.

Pressure language in women-trafficking resolutions.

Interestingly, stronger language has apparently not resulted in less support. Quite the contrary: there is an unmistakable positive correlation between the strength of language and the number of states supporting a resolution over time (see Figure 6). This is consistent with our claim that there has been a significant increase in the international consensus about how to confront trafficking in persons among states over the last two decades.

Figure 6.

Pressure language and the number of sponsors of women-trafficking resolutions.

To summarize, the history of resolutions passed by the UNGA Third Committee demonstrates that the international consensus to oppose trafficking in persons has grown over time in terms of the number of resolutions and the number and diversity of sponsors of such resolutions. Moreover, there is evidence of both increasingly minor language adjustments and the growing use of language that expresses urgency to take action, especially after the landmark 2000 agreement. Moreover, the more urgent the resolution, the more support it tended to attract. All of these constitute evidence of a developing consensus among states on the need to address the problem and the way to do so. We now return to our primary question: how has consensus been possible among heterogeneous state actors in such a relatively short period of time?

Explaining consensus formation

Documenting frames

Our central argument is that international consensus is swayed by the way an issue is framed; in particular, frames that securitize an issue are more likely to attract state support than those that do not. To make this point, we analyze how the content of these resolutions has evolved from 1994 to 2012. If our central claim is correct, states should support resolutions that frame trafficking as an issue of combating transnational crime more readily than resolutions adopting alternative interpretations.

To assess the framing content, we hand-coded the resolutions using traditional techniques of content analysis. While this is time-intensive, human judgment and expert knowledge improve validity, and allow us to derive more insights from the complex and highly stylized language employed in the UNGA resolutions. In our coding of trafficking resolutions, we follow a standard approach by focusing on individual clauses (or ‘phrases’).³² Clauses represent complete ideas and are easy to identify in trafficking resolutions, which follow a rigid format.

In particular, all resolutions include: a heading (specifying the body issuing the resolution); preambular clauses (indicating the framework through which the problem is viewed); and operative clauses (outlining recommendations for the course of action). All clauses generally start with verbs or adverbs (such as ‘concerned,’ ‘convinced,’ ‘welcoming,’ ‘reaffirming,’ ‘encouraging,’ ‘recognizing,’ ‘urges,’ ‘welcomes,’ ‘calls upon,’ ‘encourages,’ ‘invites’) and convey one idea or call for a specific action. We coded each separate clause as adopting a human rights frame³³ or a crime frame,³⁴ or conveying other information.³⁵

It is important to realize that states apparently do care about the language of these UN resolutions, especially when their own citizens are victimized. For instance, when in 1995, resolution RES 50/167 was discussed in the Third Committee, states disagreed over including the phrase ‘internationally recognized human rights standards.’³⁶ Spain, a country of transit and destination for trafficking, strongly objected to the phrase, while representatives of the Philippines and India, both countries of origin, preferred its inclusion, but accepted the proposed change. States press hard for specific expressions, and sometimes refuse to support resolutions that contain language with which they disagree. An examination of the Trauvaux Preparatoires of the TIP Protocol shows that the obligation to criminalize (Article V) elicited much less debate than sections on rehabilitation of victims.³⁷

Figure 7 illustrates how the salience of human rights and crime frames in Third Committee resolutions on trafficking in persons has changed over time. While the resolutions started with predominantly rights framing, over time, the proportions of human rights and crime clauses per resolution nearly converge. Prior to the 2000 TIP Protocol, more than 40% of the content was focused on the rights aspect, and only 10% on the crime aspect. By 2005, two years after the Protocol entered into force, rights clauses dropped to about 25% of the content while the total number of crime clauses nearly doubled, representing a clear shift in emphasis. Sponsorship, urgency, and language uniformity also increased with the growing dominance of the crime frame.

Figure 7.

Rights and crime framing in women-trafficking resolutions.

Framing and consensus: Resolution as the unit of analysis

So far, we have shown that the number of crime clauses in these resolutions has increased over time. But is the crime frame more powerful than the human rights frame in spurring consensus on a programmatic response at the UN? If so, then the more crime-oriented the resolution, the broader the support and the stronger we might expect the language of each resolution to be.

Table 1 reveals that more crime words are correlated with more total support: every additional crime clause attracts nearly 13 more sponsors. In addition, the emphasis on crime is positively correlated with stronger and more urgent language in the resolutions overall. There is also some evidence — though our coefficient does not quite reach statistical significance — that crime language is positively correlated with the diversity of resolution support, as measured by the Herfindahl Index calculated across World Bank income categories (see Table 1). This effect is mild largely because, as Figure 3 demonstrated, support for these resolutions was fairly diverse in the first place.

Table 1.

Framing and the number of sponsors of women-trafficking resolutions.

	No. sponsors	No. of pressure clauses	No. of total clauses	Herfindahl Index by income
No. crime clauses	12.754 (3.509)**	1.535 (.467)**	3.561 (1.788)+	.008 (.005)
No. rights clauses	−1.079 (4.677)	.256 (0.623)	2.598 (2.348)	−.012 (.006)+
Constant	5.253 (36.027)	−4.185 (4.802)	−9.144 (18.360)	.832 (.050)***
Adjusted R-squared	.836	.853	.795	.102
F-statistic	29.05***	32.98***	22.29***	1.623

Notes: Regression coefficients (robust standard errors). +p < .10; *p < .05; **p < .01; ***p < .001. N = 12.

Table 1 reveals very different results for human rights clauses. These clauses do not result in additional supporters (the coefficient suggests that each additional clause may actually reduce support by one state, but it is not statistically significant), and they are not correlated with additional pressure clauses. There is also a good possibility that rights clauses slightly reduce the diversity of supporters. Although our sample is small (12 resolutions), Table 1 does provide some evidence that crime rather than rights clauses drive the growing support for the resolutions on trafficking in women and girls.

Framing and resolution support: State decisions to sponsor as the unit of analysis

Do states prefer to support resolutions with a greater human rights emphasis or those with a stronger crime-fighting emphasis? In this section, we analyze each country’s decision to sponsor a trafficking in women resolution, examining support patterns for 150 countries and 12 resolutions for a total of 1800 country-year observations. We begin by estimating a logistic model with a fixed-effects estimator to explore how framing affects the likelihood of sponsorship. In this model, country indicators control for country-specific effects (i.e. the omitted country-level variables that are constant over time). We also estimate a fixed-effects logistic model with an interaction term between human rights and crime clauses to explore whether there is a trade-off between choosing one emphasis over the other. As Figure 7 shows, the number of crime and human rights clauses has converged over time, which suggests that accommodating both concerns may facilitate consensus.

Second, we examine the hypothesis that the changes in resolution framing have contributed to the changes in sponsorship. Here, we employ the conditional logit model, and rely only on information from those countries that change their sponsorship behavior (e.g. they sponsor a resolution after having abstained from sponsorship of a similar resolution in the previous year). Since the conditional logit model is a difference estimator, statistical results will measure the degree to which changes in the language of the resolutions influence changes in sponsorship behavior.³⁸

In all regressions, our dependent variable is binary: coded 1 if a country sponsors (endorses) a resolution and 0 if it does not. Our main explanatory variables are the human-coded number of crime and human rights clauses, which proxy for the emphasis on crime-fighting or human rights. We control for the prevalence of pressure language and for the total number of clauses per resolution, and explore the role of income and salience of trafficking in the decision to sponsor a resolution.³⁹

Table 2 summarizes results for the logistic regression of resolution sponsorship from three models.⁴⁰ The coefficient on the number of crime clauses is positive and significant — statistically as well as substantively — across all models, even when controlling for the total number of clauses per resolution. The coefficient on the number of rights clauses is much smaller and not significant in Models 1 and 3. This large positive coefficient on the number of crime clauses is consistent with our hypothesis that adopting the crime frame has increased state support for combating trafficking in persons.⁴¹ Including an interaction term in Model 2 substantially increases the coefficients on both crime and rights clauses, and the coefficient on the number of rights clauses becomes significant. The coefficient on the interaction term is negative and statistically significant, albeit small. The interaction term suggests an interesting trade-off between the relative strength of crime and rights framing: the effect of the number of crime clauses on the probability of sponsoring a trafficking resolution decreases as the number of rights clauses grows, and vice versa. Figure 8 presents the interaction effect graphically.

Table 2.

The decision to sponsor a resolution on the trafficking of women and girls.

DV: 1(0) = sponsor(not)	Model 1 (fixed effects)	Model 2 (fixed effects)	Model 3 (conditional logit)
No. crime clauses	.573 (.114)***	1.347 (.287)***	.473 (.107)***
No. rights clauses	.113 (.074)	.528 (.164)**	.103 (.113)
Crime clauses*rights clauses	−	−.082 (.028)**	−
No. pressure clauses	.227 (.091)*	.307 (.101)**	.217 (.081)**
No. total clauses	−.086 (.023)***	−.090 (.023)***	−.081 (.021)***
Crime clauses*high income			.315 (.176)
Rights clauses*high income			.080 (.211)
Crime clauses*low income			−.104 (.142)
Rights clauses*low income			−.053 (.181)
Constant	−4.728 (.594)***	−8.761 (1.561)***	−
Wald chi²	−	−	396.380
P > X²	−	−	.000
Observations	1800	1800	1800
Countries (groups)	150	150	150

Notes: Logistic regression results (robust standard errors). Country dummies omitted to save space. *p < .05; **p < .01; ***p < .001.

Figure 8.

Probability of sponsorship: Interaction between the number of rights and crime clauses.

The influence of the crime frame is strong across state income levels (see Table 3). The coefficient on the number of crime clauses is positive and significant in all three subsets of countries (and is largest for middle-income countries). This suggests that crime language is likely to elicit an endorsement across a very diverse range of countries. By comparison, the coefficient on the number of human rights clauses changes sign across regressions and is not statistically significant.

Table 3.

The effect of frames on the likelihood of resolution sponsorship, by income level.

DV: 1(0) = sponsor(not)	High income	Middle income	Low income
No. crime clauses	.517 (.257)*	.591 (.168)**	.510 (.213)*
No. rights clauses	−.101 (.150)	.156 (.105)	.169 (.148)
No. pressure clauses	.220 (.223)	−.221 (.141)	.270 (.143)
No. total clauses	.034 (.057)	−.101 (.033)**	−.132 (.046)**
Observations	420	888	492
Countries (groups)	35	74	41

Notes: Logistic regression with country fixed effects (robust standard errors). Country dummies omitted to save space. +p < .10; *p < .05; **p < .01; ***p < .001.

We also test for the nature of a country’s exposure to trafficking, hypothesizing that destination and transit states are most affected by the negative externalities of trafficking and, therefore, should be more likely to sponsor resolutions that adopt the crime frame. In contrast, countries of origin may be more interested in protecting the rights of their citizens victimized by trafficking and, thus, more likely to sponsor resolutions that focus on rights. We use the UN Report on Trafficking in Persons: Global Patterns⁴² to document the incidence of reporting of origin, destination, and transit countries at three levels of intensity — very low, low, medium, high, and very high (coded 1 through 5). We separate countries into two groups: primarily countries of origin with a low incidence of transit, and primarily destination and transit countries.⁴³ Results in Table 4 support our hypothesis that countries of destination and transit are more concerned with a focus on crime, as evidenced by the much larger coefficient on the number of crime clauses. In comparison, rights clauses and pressure clauses are large and significant only for the states of origin.

Table 4.

The effect of frames on the likelihood of resolution sponsorship, by trafficking ‘type.’

DV: 1(0) = sponsor(not)	States of origin	Transit and destination states
No. crime clauses	.463 (.172)**	.694 (.178)***
No. rights clauses	.315 (.168)+	.012 (.163)
No. pressure clauses	.431 (.153)**	.060 (.159)
No. total clauses	−.172 (.040)***	−.029 (.042)
Constant	−3.200 (1.338)*	−3.036 (1.231)*
Observations	480	504
Countries (groups)	40	42

Notes: Logistic regression with country fixed effects (robust standard errors). Country dummies omitted to save space. +p < .10; *p < .05; **p < .01; ***p < .001.

Finally, we examine the substantive impact of the independent variables on the likelihood of sponsorship from our primary regression analysis using the estimates of the fixed-effects logit model in Figure 9. The results show that the probability of sponsorship increases much faster with the number of crime clauses than with the number of rights clauses. In an average resolution, with six crime clauses and 11 rights clauses, the probability of sponsorship is 0.13. When the number of crime clauses increases by one, the probability of sponsorship changes to 0.21 (i.e. by 0.08). In contrast, when the number of rights clauses increases by one, the probability changes to 0.14 (i.e. by 0.01).

Figure 9a.

Probability of sponsorship and crime framing: Model 1 (fixed-effects logit).

Figure 9b.

Probability of sponsorship and rights framing: Model 1 (fixed-effects logit).

To summarize, statistical analysis shows that the probability of sponsorship increases in the number of clauses that reflect the crime framing, but not in the number of clauses that reflect the rights framing. The effect of crime language is substantively large and statistically significant and holds for countries in different income groups and at different levels of exposure to trafficking.

Discussion and conclusions

Consensus among states is often critical to effectively handle the world’s most pressing collective problems, from global warming, to the spread of HIV, to nuclear security. Trafficking in persons is one such problem. States have been grappling with this issue for more than a century, but until recently, broad consensus on the nature of, and solutions to, the problem has been elusive. In the first half of the past century, geographically limited international agreements fixated on prostitution were the dominant approach. Trafficking in persons became even more challenging in the second half of the century: how could nearly 200 highly heterogeneous independent states reach agreement on how best to deal with a complex problem stemming from globalization, lagging local opportunities, and economic disruption in many parts of the world? How would this be possible among states that were variably sources, transit routes, and destinations for trafficking in persons — each with their own set of social and political challenges? And how would this be possible with very different cultural, legal, and institutional values and capacities?

The framing of the issue was critical for reaching a broad consensus, as it often is in policymaking settings. We have emphasized the distinction between the human rights and crime frames to show how the latter garnered broad support among the community of states. This crime frame had deep historical roots — as seen in the numerous references to ‘punishment’ in early agreements — but the forces of globalization in the early post-Cold War years made transnational crime seem more widespread, threatening, and urgent than in earlier periods. Governments were increasingly concerned with protecting their borders and their very sovereignty from organized crime that potentially threatened local economies, societies, and the integrity of national governance. This framing made human trafficking easy to ‘securitize’ and could be used to justify stronger law-enforcement and border regimes, and perhaps even increase foreign aid for these purposes.

Human rights — respect for the individual as a human being, including their safety, autonomy, and dignity — offered another compelling way to view the problem of trafficking in persons. A ‘rights revolution’ had rather thoroughly altered the international normative atmosphere between the 1960s and the 1990s, and it was no longer possible politically or morally to approach trafficking in persons as a simple matter of international cooperation to enforce national vice laws. Global NGOs (from Amnesty International, to Stop the Traffic) and local NGOs (from the Philippines, to Mongolia, to Costa Rica) added their voices to those of the Special Rapporteur on Violence against Women and other officials to demand attention to the plight of the trafficked. They had plenty of legal and normative ammunition in the form of widely accepted international agreements, such as the Convention on the Elimination of Discrimination Against Women, the Convention on the Rights of the Child (especially the optional protocol on child exploitation), and, more recently, the International Convention on the Protection of the Rights of All Migrant Workers and Members of Their Families. Bolstered by three decades of international human rights agreements, these actors made a compelling case that, first and foremost, trafficking in persons is a problem of rights, and should be declared and dealt with as such.

As important as human rights are to the global normative environment, the evidence analyzed here suggests that consensus on trafficking in persons has been facilitated by foregrounding the crime frame. Our focus on Third Committee resolutions provides a window into the consensus-building process in the UN, the forum in which every state has a right to participate. The years 1994–2012 are especially revealing because they represent a ‘new look’ at an issue that had been largely ignored by the international community since the late 1940s. These resolutions reveal the building blocks of consensus formation in an especially useful way: by allowing us to literally compare the language that attracts state support from that which does not.

The patterns in the texts reveal a fairly clear trend. Crime framing is associated with more endorsers, more diverse supporters, and stronger language in general than is the human rights frame. Crime language is also more likely than human rights language to elicit state endorsement for a particular resolution. Destination and transit states — those who are impacted by the presence of organized crime networks but whose citizens are less likely to be at risk — are especially attracted to a strong law-enforcement approach.

Our findings do not imply that human rights are insignificant to the human trafficking regime. On the contrary, our research shows that concern for human rights was strong in the early resolutions of the 1990s. However, to broaden the coalition against trafficking in persons, stronger and more consistent attention had to be paid to states’ security concerns, which meant framing the issue in terms of its linkages to all of the dangers associated with transnational crime. Consensus could be reached because rights advocates and actors concerned about national security accommodated one another’s concerns. Initially based on a fragile consensus, the ‘3P’ framework that emerged in 2000 strengthened over time and eventually embraced victim protection, prosecution, and prevention. The TIP Protocol itself requires states to develop domestic laws criminalizing trafficking in persons, but it also gives them the flexibility to prioritize protection of victim rights, as the Legislative Guide on the treaty makes clear.⁴⁴

This research has important implications for how states reach consensus on contentious issues in international politics. While much recent research has concentrated on how and why states ratify international agreements, it is important to understand how such agreements become possible in the first place. Traditional histories are useful in this regard, but it is also instructive to take seriously the public record of state support for various interpretations of a problem and how this shapes eventual solutions.

We have argued that frames can be chosen selectively in order to increase the probability of getting a broad agreement on an issue. Our account emphasizes strategic framing, but also accommodation of alternative values and viewpoints. While other scholars have emphasized socialization (Goodman and Jinks, 2004; Johnston, 2002) or persuasion (Deitelhoff, 2009; Hawkins, 2004; Risse, 2000) as a critical feature of world politics, we stress coalition building and the process of accommodation. Rather than hammering away at the human rights travesty that trafficking in persons certainly represents, states were drawn to support crime-fighting and the broader threat that illicit trans-border flows represent. The 3P approach appealed primarily to those worried about security, while continuing to acknowledge important human rights issues. Advancing a framework of criminalization and law enforcement encouraged a broad consensus through what might be thought of as a soft ‘log-roll’ of rights concerns and crime-fighting — a combination of aims that many observers have noted are not always fully compatible.

Much work remains to be done to explain why one frame trumps the other and why the dominance of a given frame changes over time across issue areas and historical time periods. A vast array of documents await exploration to add nuance to our understanding of framing and consensus formation — draft resolutions, verbatim records, country statements, and media framing of the issues. And while trafficking in persons is a useful starting point for studying framing and consensus formation, the opportunities are vast in other areas of international cooperation. This research captures an important part of international consensus formation that is foundational to international coordination and cooperation — especially on transnational issues such as crime, of which trafficking in persons is a particularly egregious example.

Footnotes

Funding

Beth Simmons received general support from the Weatherhead Center for International Affairs, Harvard University.

Notes

Author biographies

Volha Charnysh is a PhD candidate in the Department of Government at Harvard University, USA. Her work explores how ideas, identity, and culture shape patterns of political attitudes and behavior within and across states. Her dissertation focuses on the long-term effects of forced migration in the wake of the Second World War on economic and political development in Europe.

Paulette Lloyd is an American Association for the Advancement of Science fellow and foreign affairs officer at the US Department of State. The views expressed in this article are her own and do not necessarily reflect those of the Department of State or the US government.

Beth Simmons is Clarence Dillon Professor of International Affairs in the Government Department at Harvard University, USA. Her research relates to human rights, international law, and international political economy.

References

Adler

(2002) Constructivism and International Relations. In: Carlsnaes

Risse-Kappen

Simmons

(eds) Handbook of International Relations. London and Thousand Oaks, CA: SAGE Publications, 95–118.

Andreas

Price

(2001) From war-fighting to crime-fighting: Transforming the American national security state. International Studies Review 3(3): 31–52.

Bales

(2005) Understanding Global Slavery: A Reader. Berkeley, CA: University of California Press.

Bales

Soodalter

(2009) The Slave Next Door: Human Trafficking and Slavery in America Today. Berkeley, CA: University of California Press.

Balzacq

(2005) The three faces of securitization: Political agency, audience and context. European Journal of International Relations 11(2): 171–201.

Baumgartner

Green-Pedersen

Jones

(2007) Comparative Studies of Policy Agendas. London: Routledge.

Benford

Snow

(2000) Framing processes and social movements: An overview and assessment. Annual Review of Sociology 26(1): 611–639.

Bhattacharyya

(2005) Traffick: The Illicit Movement of People and Things. London: Pluto.

Bleich

(2002) Integrating ideas into policy-making analysis: Frames and race policies in Britain and France. Comparative Political Studies 35(9): 1054–1076.

10.

Budge

(2001) Mapping Policy Preferences: Estimates for Parties, Electors, and Governments, 1945–1998. Oxford and New York, NY: Oxford University Press.

11.

Buzan

Wæver

Wilde

(1998) Security: A New Framework for Analysis. Boulder, CO: Lynne Rienner Publishers.

12.

Chong

Druckman

(2007) Framing theory. Annual Review of Political Science 10: 103–126.

13.

Clark

(2001) Diplomacy of Conscience: Amnesty International and Changing Human Rights Norms. Princeton, NJ: Princeton University Press.

14.

Curley

Wong

(2008) Security and Migration in Asia: The Dynamics of Securitisation. London and New York, NY: Routledge.

15.

Däubler

Benoit

Mikhaylov

. (2012) Natural sentences as valid units for coded political texts. British Journal of Political Science 42(4): 937–951.

16.

Deitelhoff

(2009) The discursive process of legalization: Charting islands of persuasion in the ICC case. International Organization 63(1): 33–65.

17.

Di Nicola

(2009) Prostitution and Human Trafficking: Focus on Clients. New York, NY: Springer.

18.

Doezema

(1999) Loose women or lost women? The re-emergence of the myth of white slavery in contemporary discourses of trafficking in women. Gender Issues 18(1): 23–50.

19.

Druckman

(2001) On the limits of framing effects: Who can frame? The Journal of Politics 63(4): 1041–1066.

20.

Entman

(1993) Framing: Toward clarification of a fractured paradigm. Journal of Communication 43(4): 51–58.

21.

Epp

(1998) The Rights Revolution: Lawyers, Activists, and Supreme Courts in Comparative Perspective. Chicago, IL: University of Chicago Press.

22.

Farrell

Fahy

(2009) The problem of human trafficking in the U.S.: Public frames and policy responses. Journal of Criminal Justice 37(6): 617–626.

23.

Gallagher

(2001) Human rights and the new UN Protocols on Trafficking and Migrant Smuggling: A preliminary analysis. Human Rights Quarterly 23(4): 975–1004.

24.

Gamson

Modigliani

(1989) Media discourse and public opinion on nuclear power: A constructionist approach. The American Journal of Sociology 95(1): 1–37.

25.

Goodman

Jinks

(2004) How to influence states: Socialization and international human rights law. Duke Law Journal 54: 621–701.

26.

Hawkins

(2004) Explaining costly international institutions: Persuasion and enforceable human rights norms. International Studies Quarterly 48(4): 779–804.

27.

Hughes

(2000) The ‘Natasha’ trade: The transnational shadow market of trafficking in women. Journal of International Affairs 53(2): 1–18.

28.

Huysmans

(2000) The European Union and the securitization of migration. JCMS: Journal of Common Market Studies 38(5): 751–777.

29.

Ibrahim

(2005) The securitization of migration: A racial discourse. International Migration 43(5): 163–187.

30.

Iyengar

Kinder

(1987) News that Matters: Television and American Opinion. Chicago, IL: University of Chicago Press.

31.

Jackson

(2006) International organizations, security dichotomies and the trafficking of persons and narcotics in post-Soviet Central Asia: A critique of the securitization framework. Security Dialogue 37(3): 299–317.

32.

Johnston

(2002) The social effects of international institutions on domestic (foreign policy) actors. In: Drezner

(ed.) Locating the Proper Authorities: The Interaction of Domestic and International Institutions. Ann Arbor, MI: University of Michigan Press, 145–196.

33.

Kaldor

Martin

Selchow

(2007) Human security: A new strategic narrative for Europe. International Affairs 83(2): 273–288.

34.

Keck

Sikkink

(1998) Activists beyond Borders: Advocacy Networks in International Politics. Ithaca, NY: Cornell University Press.

35.

Kempadoo

Doezema

(1998) Global Sex Workers: Rights, Resistance, and Redefinition. New York, NY: Routledge.

36.

Klingemann

H-D

Hofferbert

Budge

(1994) Parties, Policies, and Democracy. Boulder, CO: Westview Press.

37.

Korey

(1998) NGOs and the Universal Declaration of Human Rights: A Curious Grapevine. New York, NY: St. Martin’s Press.

38.

Lloyd

Simmons

Stewart

(2012) Combating transnational crime: The role of learning and norm diffusion in the current rule of law wave. In: Nollkaemper

Zurn

Peerenboom

(eds) The Dynamics of the Rule of Law. Cambridge, MA: Cambridge University Press, 153–180.

39.

Loader

(2002) Policing, securitization and democratization in Europe. Criminology and Criminal Justice 2(2): 125–153.

40.

Locher

(2007) Trafficking in Women in the European Union: Norms, Advocacy-Networks and Policy Change. Wiesbaden: VS Verlag für Sozialwissenschaften.

41.

McAdam

McCarthy

Zald

(1996) Introduction: Opportunities, mobilizing structures, and framing processes — toward a synthetic, comparative perspective on social movements. In: McAdam

McCarthy

Zald

(eds) Comparative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings. Cambridge and New York, NY: Cambridge University Press, 1–22.

42.

McDonald

(2008) Securitization and the construction of security. European Journal of International Relations 14(4): 563–587.

43.

Mintz

Redd

(2003) Framing effects in international relations. Synthese 135(2): 193–213.

44.

Nelson

Oxley

(1999) Issue framing effects on belief importance and opinion. The Journal of Politics 61(4): 1040–1067.

45.

Nieuwenhuys

Pécoud

(2007) Human trafficking, information campaigns, and strategies of migration control. American Behavioral Scientist 50(12): 1674–1695.

46.

Oliver

Johnston

(2000) What a good idea! Ideologies and frames in social movement research. Mobilization: An International Quarterly 5(1): 37–54.

47.

Pajnik

(2010) Media framing of trafficking. International Feminist Journal of Politics 12(1): 45–64.

48.

Rein

Schön

(1993) Reframing policy discourse. In: Fischer

Forester

(eds) The Argumentative Turn in Policy Analysis and Planning. Durham, NC: Duke University Press.

49.

Risse

(2000) ‘Let’s argue!’ Communicative action in world politics. International Organization 54(1): 1–39.

50.

Risse

Ropp

Sikkink

(1999) The Power of Human Rights: International Norms and Domestic Change. Cambridge and New York, NY: Cambridge University Press.

51.

Rubenzer

(2011) Campaign contributions and U.S. foreign policy outcomes: An analysis of Cuban American interests. American Journal of Political Science 55(1): 105–116.

52.

Salt

(2000) Trafficking and human smuggling: A European perspective. International Migration 38(Supplement 1): 31–56.

53.

Scarpa

(2008) Trafficking in Human Beings: Modern Slavery. Oxford: Oxford University Press.

54.

Scheufele

(1999) Framing as a theory of media effects. Journal of Communication 49(1): 103–122.

55.

Schloenhardt

Bevan

(2011) To ratify or not to ratify? Exploring the barriers to wider ratification of the trafficking in persons protocol. The New Zealand Yearbook of International Law 9: 161–184.

56.

Simmons

(2009) Mobilizing for Human Rights: International Law in Domestic Politics. New York, NY: Cambridge University Press.

57.

Snow

Benford

(1988) Ideology, frame resonance, and participant mobilization. International Social Movement Research 1(1): 197–217.

58.

Stratmann

(2002) Can special interests buy congressional votes? Evidence from financial services legislation. Journal of Law and Economics 45(2): 345–373.

59.

Stritzel

(2007) Towards a theory of securitization: Copenhagen and beyond. European Journal of International Relations 13(3): 357–383.

60.

Thachuk

(2007) Transnational Threats: Smuggling and Trafficking in Arms, Drugs, and Human Life. Westport, CT: Praeger Security International.

61.

Tversky

Kahneman

(1981) The framing of decisions and the psychology of choice. Science, New Series 211(4481): 453–458.

62.

Van Dijk

(2011) Transnational organized crime, civil society, and victim empowerment. In: Letschert

Van Dijk

JJM

(eds) The New Faces of Victimhood: Globalization, Transnational Crimes and Victim Rights. New York, NY: Springer, 99–126.

63.

Vlassis

(2000) The global situation of transnational organized crime, the decision of the international community to develop an international convention and the negotiation process. Annual Report for 2000 and Resource Material Series No. 59.

64.

Wæver

(1995) Securitization and desecuritization. In: Lipschutz

(ed.) On Security. New York, NY: Columbia University Press, 46–86.

65.

Watson

(2012) ‘Framing’ the Copenhagen School: Integrating the literature on threat construction. Millennium — Journal of International Studies 40(2): 279–301.