GRADUATE SCHOOL OF COMMUNICATION
Master’s programme Communication Science:
Erasmus Mundus Journalism, Media and Globalisation (joint degree)
Master’s Thesis
Associating trafficking to sex work:
How media covers trafficking in human beings in countries with different legal frameworks of sex work.
Julia L.C. Natri
13999222
Supervisor: Dr. R. (Rachid) Azrout
Amsterdam, 24/06/2022
Abstract
Trafficking has growingly become a subject on the public and political agenda. The present study shines a light on how the trafficking of human beings is framed in print media in 7 different countries. Specifically, it is interested in the link framing of trafficking has with the legal framework of sex work (decriminalization, legalization, swedish model, or prohibition) in those countries. Through a quantitative content analysis, this study compares the coverage of human trafficking across the four legal frameworks. The major finding is that countries that are more favorable to sex workers’ rights (decriminalization and legalization) tend to be less biased towards sex trafficking. Indeed, non-favorable countries (prohibition and swedish model) rather associate trafficking with sex trafficking, and mention more often women and children as victims, while decriminalization and legalization countries tend to discuss more equally other forms of trafficking, such as labor trafficking, which is the most common form of trafficking occurring. This has important implications, as the association between trafficking and sexual exploitation is often used to further criminalize sex work.
Keywords: trafficking in human beings, sex work, framing, content analysis, media coverage
Introduction
Human trafficking has created energetic debates regarding the sex industry and the increased recruitment and obtainment of individuals for profit. However, the overall public perception of trafficking in human beings remains incomplete and repeatedly inaccurate about its nature, its causes, and how to prevent it (Sharapov et al., 2019). This is due to a specific representation of trafficking that is contructed and reproduced in mass media, including movies, television and newspapers, as well as in the public discourse (Andrijasevic & Mai, 2016). Social movement theorists have previously used the concept of framing in order to discuss the way social movement actors can shape arguments in the best way to win supporters to their cause. This implies a selection of concepts and language to influence political debate as well as decision-making (Bacchi, 2009). It is therefore appropriate to analyze frames when researching issues related to gender inequality, its construction, and performance.
When it comes to the framing of human trafficking in mainstream media, the focus has been on the sexual exploitation of foreign women: trafficking is commonly framed as a crime issue, and related to the sex industry (Guilati, 2010; Pjanik., 2010; Sobel, 2014). The plot as presented in the majority discourse rarely varies: it beggins with deception, followed by coercion into prositution, continues on to the tragic story of (sexual) slavery, to end with finding a solution through the victim being rescued by the police or an NGO (Doezema, 2010). Yet, figures show that trafficking for other purposes, such as child domestic work or other labor abuses, are more widepsread and affect a larger number of individuals than sex trafficking (Brysk, 2009). This common untruthful representation influences the perception of the average citizen, who associates human trafficking with the smuggling of (white) women for sexual slavery (Bonilla & Mo, 2018). Moreover, previous studies focusing on the
coverage of trafficking in human beings have shown that the coverage of trafficking in media is largely episodic, focuses on crime and policy rather than human rights and public health, and neglects the voice of survivors or organizations that work with them (Jonhston et al, 2014, 2015; Sobel, 2014). This reoccurring narrative serves a purpose, namely legitimizing state policies that criminalize activities undertaken by minorities. Indeed, the framing of trafficking in the media helps argue for continuing to develop stricter migration laws (Gulati, 2010; Pajnik, 2010). Furthermore, the media representation of trafficking legitimizes the criminalization of sex work (Andrijasevic & Mai, 2016; Jonhston et al., 2015). To emphasize, the framing of human trafficking in the media is used to confirm or continue in the direction of already existing policies or legal frameworks.
Sex workers' rights organization have raised their concern on the current trends in human trafficking research and literature, which is failing to analyze and comprehend human trafficking act as a whole, and to understand the variety of actors involved. A biased conception that focuses only on one aspect of human trafficking can lead to policies and legislations that are not effective enough to respond to the different forms of exploitation that occur. Even worse, it can lead to policies that have a negative impact on minorities such as sex workers or migrant workers. Therefore, sex workers’ rights activists insist on the urge to stop conflating trafficking with sex work, and on the need to listen to sex workers in order to effectively combat trafficking of human beings, and enable individuals who work in the sex industry to freely choose.
Since the framing of trafficking is used to reaffirm government measures and policies (Soderlund 2005; Pajnik, 2010) the present study is interested in shedding light on the link made between trafficking in human beings and sex work in the media, and on how the
coverage of trafficking is influenced by the legal framework of sex work. More specifically, it attempts to answer the following question:
How does the framing of trafficking in human beings differ in the media when comparing countries with different legal models concerning sex work?
Through a content analysis of 7 countries, the coverage of trafficking is analyzed in four different legal models of sex work. The present analysis includes the United States and South Africa, in which sex work is prohibited by law (prohibition model); Canada and France, where clients of sex workers are criminalized (Swedish model); Switzerland and Luxembourg, where sex work is legal but strictly regulated by the state (legalization model);
and New Zealand, the only country to have decriminalized sex work, and hence legally recognizes sex work as work and treats it as any other service sector (decriminalization model).
To summarize, the present research explores, for what appears to be the first time, how the legal model of sex work influences the way trafficking in human beings is framed in the media. This complements the already existing knowledge on the ways political and legal contexts affect the media coverage of a topic, focusing precisely on sex work and human trafficking. Besides filling this existing research gap, understanding the link between the framing of trafficking and the legal framework of sex work is the first step towards less biased news on the subject, and a more accurate and complete coverage of trafficking in human beings. Indeed, with more knowledge of the dominant dynamics within the coverage of human trafficking, journalists working on the topic can consciously take into consideration the abovementioned link, and work accordingly. This knowledge gives them the possibility to
balance the dynamic and main narrative, for example by including in their coverage voices that are usually less visible.
Theoretical framework
Framing
Entman (1993) defines framing as a process that “essentially involves selection and salience.
To frame is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation and/or treatment recommendation for the item described”
(p.52). He also proposes an operationalization of framing looking at their four functions.
Frames “define problems - determine what a causal agent is doing with what costs and benefits, usually measured in terms of common cultural values; diagnose causes - identify the forces creating the problem; make moral judgments - evaluate causal agents and their effects;
and suggest remedies - offer and justify treatments for the problems and predict their likely effects” (Entman, 1993, p.52). A frame in a text does not always include all four functions, while one sentence can achieve several of these framing functions, or many sentences in a text can complete none (Entman, 1993).
Framing of trafficking in the media
By comparing different official numbers concerning several forms and victims of human trafficking, Brysk (2009) explains that certain forms of labor exploitation and abuse, such as child domestic work, are far more widespread than international sex trafficking, and affect a greater number of people. Yet, trafficking related to sexual exploitation receives a much
higher level of mass media attention than trafficking for other forms of exploitation, whether it is in romanticized movies or in the news (see Brysk, 2009; Denton, 2010; Guilati, 2010;
Pjanik., 2010; Sobel, 2014; Soderlund, 2005). This phenomenon is widespread. Pajnik (2010) analyzed the different frames present in Slovene media when covering trafficking, and found that these frames “limits our perception of trafficked people to those women enduring abusive experience of forced sex (..) [and] conflates trafficking with sex-trafficking” (p.57).
Framing trafficking only related to sexual abuse doesn’t allow the exploration of distinct dimensions of trafficking such as the trafficking of men, or trafficking to other industries than sex work, such as domestic work or farming (Pajnik, 2010). In the United States, Soderlund (2005) notes that stories about Asian or Latin American women being illegally trafficked into western Europe to work in the sex industry began to multiply in newspapers in late 1980’s. In the 2000’s, the focus shifted to rural-to-urban movement, and to the trafficking of foreign women or children into the US. According to her analysis, stories about trafficking for
“modern-day sexual slavery” are “frequently selected by journalists because of their sensationalistic qualities rather than their status as exemplary stories of women in global economy” (Soderlund, 2005, p.71). Empirical research confirms this: when looking at American, Canadian, and British newspaper coverage of trafficking, Gulati (2010) found that over half of the articles refer to it in terms of “sex trafficking, prostitution or pornography (54%) while only one-fifth referred to the issue in terms of labor exploitation (20%)” (p.15).
Another common finding is the fact that trafficking is often discussed in the media with a criminal angle in the coverage, rather than with a human rights approach (Gulati, 2010;
Jonhston et al., 2014; Jonhston et al., 2015; Sobel, 2014; Pajnik 2010). Pajnik (2010) operationalized it as a frame by its own, that “tend to portray dark, omnipotent criminals in contrast to the white female victims who have fallen into the traps of middle-men” (p.52).
Sobel (2014) found that most stories in her study on the news coverage of human trafficking in Thailand, India and the USA are framed as a crime story or for discussions about policy/legislation changes. Jonhston et al. (2014) and Jonhston et al. (2015) studied specifically the coverage of sex trafficking but come to the same conclusion: sex trafficking is most commonly framed as a crime issue.
This has an effect on other aspects of the coverage. Indeed, stories where human trafficking is dominantly framed as a crime issue give different causes, consequences, solutions and sources than stories with another dominating frame or issue. Jonhston et al. (2014) analyzed sex-trafficking related stories in major print news sources, and looked at which issues dominated the articles while comparing it to other framing dimensions. They found that crime was the dominant issue in 67% of the articles, long before policy/legislation (16%), followed by human rights (11%) and activism/awareness (5%). In the articles where trafficking was framed as a crime issue, more than half did not mention a remedy, and the second most mentionned solution to the problem of human trafficking was increased punishment for traffickers or the clients using the victims services. These stories also mostly featured law enforcement or political sources. As a contrast, the articles framing trafficking as a human rights issue cited most frequently advocacy groups or victims of trafficking and their representatives, and the most frequent remedy brought up was educating the public on human trafficking and increasing awareness on the topic (Johnston et al., 2014). The sources used differ depending on the dominant issue frame, which leads to the majority of human trafficking stories featuring government officials or law enforcement sources rather than victims and their representatives, directly affected but largely absent in the coverage (Gulati, 2010; Jonhston et al., 2014; Jonhston et al., 2015; Sobel, 2014).
Soderlund (2005) and Pajnik (2010) note that the framing of human trafficking in the media is used to further disseminate a political message that confirms government actions. In particular, articles about trafficking in the Slovene media use certain frames to legitimize a call for stricter laws and tighter border regimes. Moreover, other frames are employed “to help the state reinforce its eroded sovereignty, promoting it as arbiter and securer of the national community” (Pajnik, 2010, p. 54). In the United States, around the years 2000, trafficking of women became a common divisor political issue, used to bridge people across the religious and political spectrum, uniting them against an apparently indisputable act of exploitation (Soderlund, 2005). The disproportionate attention of the media on sex trafficking, and the excessive framing of trafficking as a crime issue follows a political agenda. It serves the purpose of legitimizing state policies that criminalize activities undertaken by minorities. It clears the state of the responsibility of deeper rooted inequalities and social injustices. As Soderlund (2005) writes, “an obsessive focus on sex trafficking ultimately distracts from drawing connections between gendered poverty and forced prostitution and presumes a moralistic approach that is unlikely to consider poverty, hunger, and low wages as equally pressing forms of violence against women” (p. 70). Instead, the framing of trafficking in the media helps argue for the further implementation of stricter migration laws (Gulati, 2010; Pajnik, 2010) and to legitimize the criminalization of sex work (Andrijasevic & Mai, 2016; Jonhston et al., 2015).
To emphasize, previous literature indicates that news media help support the dominant view, and mainly echo the narratives of the major actors in the policy process (Gulati, 2010;
Larson, 2006). Gulati’s (2010) study of the framing of human trafficking in the American, British and Canadian media shows that there is almost no questioning of the positions and statements of official sources, which are very present in the coverage. The study also shows
that the few academic or policy analyst sources cited actually tend to be associated with the leading policymaking community and that it is only on rare occasions that alternative voices are cited to propose another perspective than the dominant one (Gulati, 2010). The same goes for the mention of causes or solutions related to human trafficking: the alternative views are being marginalized by the coverage, while it further legitimizes the official perspective (Gulati, 2010). Gulati (2010) concludes that “the media has had an important role in anti-trafficking policy by limiting how many competing views on the causes of the problem are discussed and the range of recommended policy alternatives considered” (p.22).
Furthermore, other scholars argue that the dominant discourse concerning human trafficking goes hand in hand with sex work related discourse and legal framework. In the United States, they condemn anti-trafficking policies from a “sex-positive” feminist point of view and denounce the false and problematic claims made by leading activists and organizations in their discourse about sex trafficking, and sex work to a larger extent (Chapkis 2003; Doezema 2000; Weitzer, 2007). According to these scholars, the high and flawed estimates on human trafficking, particularly sex trafficking, are mainly advanced by religious conservatives who are on a moral crusade against female sexual liberation. Joined by “radical” feminists, who view prostitution as the ultimate form of male dominance and violence against women, these religious conservatives use a specific discourse on trafficking to provoke indignation among Americans. The final aim is to build more drastic solutions to fight sex work and criminalize it (Chapkis 2003; Doezema 2000; Weitzer 2007).
Considering the above, I argue that the legal framework of sex work will have an influence on the media framing of trafficking. In countries where sex work is prohibited or partly criminalized (prohibition and Swedish model), the dominant political discourse is that sex
work is an activity related to criminality that needs to disappear, which can be achieved with laws punishing the criminals and continuous police interventions. Hence, in those countries, I presume the framing of trafficking will be linked to sex work, and predominantly negative, calling for further criminalization (of both sex work but also border crossing). In those countries, I expect official sources to be quoted more often, and the discussion to revolve around policy/legislation. On the other hand, in a country from the decriminalizaiton model, such as New Zealand, sex work is considered as work and not something to eliminate (ICRSE, 2018). The major political discourse concerning sex work is different from countries with other legal frameworks. Logically, as the main narrative is not leaning towards the criminalization of sex work, I believe the coverage will not necessarily link trafficking to sex work. Additionally, New Zealand has proved its capacity to acknowledge the perspective of sex workers when building policies centering on their rights (ICRSE, 2018). Following that trend, I assume the media will discuss trafficking more from the perspective of survivors, in order to call for greater consideration of their rights. Therefore, in country with the decriminalization model, I expect newspapers to use more frequently a human rights, alternative or awareness approach. Moreover, I anticipate that sex workers and their representatives, as well as victims of trafficking and their representative, will be more cited.
Finally, in countries with a legalization model, the dominant political perspective is that sex work is tolerated but an occupation that ultimately needs to disappear because of its supposedly inherently exploitative and patriarchal aspect. Accordingly, I am expecting a framing of human trafficking that is “in-between” the two previously cited ones.
H1 - In countries from the decrminialization and legalization model, the media will frame trafficking of human beings as less related to sex work than in countries from the swedish and
prohbition model
H2 – In decriminalization and legalization countries, the media will frame trafficking more as a human rights, alternative or awereness issue while swedish model and prohibition
countries will frame it more frequently as a crime or policy issue
H3 - Decriminalization and legalization countries will use more sources representing the voices of sex workers and of survivors of trafficking, while swedish model and prohibition
countries will use more official sources
Methods
To answer the research question, the present study analyses data quantitatively. In particular, the method chosen is content analysis, defined by Maier (2017, p. 2) as “a systematic, quantitative process of analyzing communication messages by determining the frequency of message characteristics”.
When studying the framing of human trafficking in the media, many scholars have done content analyses of newspapers or news sites (see Pajnik, 2010; Jonhston et al., 2015;
Jonhston et al., 2014 Gulati, 2010; Denton, 2010). Indeed, evidence shows that new media sources, such as bloggers, link back to legacy media, namely newspapers sources, for information and politics (Pew Research Center, 2010). Hence, studying how newspapers frame the issue is appropriate considering they are an originating source of information for other media.
Research design
As the present study is interested in comparing the framing of trafficking in countries with different legal frameworks, the sample is based on the sex work’s legal framework
classification made by the International Committee on the Rights of Sex Workers in Europe (2018), that present four main legal frameworks: prohibition, swedish Model, legalisation and decrimininalization.
Countries in which sex work is prohibited consider sex work illegal, see sex workers as a threat to society, and conduct policies aiming to decrease the presence of sex workers in public spaces, such as police raids of sex workers workplaces or homes (ICRSE, 2018). The Swedish model is a legal framework where sex workers are not legally punished for their work, but it is illegal to buy sexual services. Sex work is seen as the ultimate proof of violence committed by men against women (ICRSE, 2018). In legalization model countries, sex work is considered legal, but sex workers are strictly regulated, through various forms such as mandatory registration, specific imposed conditions (for example mandatory medical examination) that control who is allowed to work as a sex worker, or location limitations (ICRSE, 2018). Finally, in the decriminalization model, which is the legal framework that many sex workers’ rights group advocate for, sex work is legally preceived as work, treated as other service sectors, and sex workers are granted labor rights (ICRSE, 2018). The only country that is commonly agreed upon as having decriminalized sex work is New Zealand.
Sample
In order to get a general picture of the media environment in each of the four legal frameworks, and considering the accessibility in terms of written language and NexisUni database, the present study includes the following countries: New Zealand for decriminalization; Luxembourg and Switzerland for legalization; France and Canada for the Swedish model; the United States and South Africa for prohibition. In order to have an equivalent number of articles in each legal framework, different amounts of newspapers were analyzed in each country. The most popular newspapers, considering highest distribution rate
or readership, were included in the analysis. The sample includes The New Zealand Herald, The Press and The Dominion Post (New Zealand); 24Heures, La Tribune de Genève (Switzerland) and Luxembourg Times (Luxembourg); The Globe and Mail, The Toronto Star (Canada), Le Figaro and Ouest France (France); The New York Times, USA Today (United States), Sunday Times, Cape Times and Daily News (South Africa). Articles between the 1st of January 2017 and the 14th of April 2022 (beginning of the coding) were included in the analysis, and retrieved through a NexisUni advanced search of the terms “trafficking in human beings” and “human trafficking” being mentionned at least twice. Due to lack of article in the time frame, for the Luxembourg Times, an exception was made: articles from 2016-2015 were also included. Considering the large coverage of the refugee crisis in European media in 2015, I am aware that this might have an influence on the result of the anlysis.
Operationalization
Previous studies focusing on the framing of trafficking in the media have built up on Entman’s four-part typology to operationalize their content analysis (Pjanik, 2010; Sobel, 2014; Jonhston et al. 2015). In particular, Jonhston et al. (2015, p. 241) study the framing of sex trafficking in US newspapers between 2008 and 2012 by “coding for the solutions, causes, and remedies as they were expressed by sources or exposition in the stories”. Sobel (2014) analyzes the coverage of trafficking in human beings in 3 English-language newspapers (from USA, India and Thailand) by coding articles looking at how the problem of human trafficking is defined, if and which causes are suggested to explain why it occurs, if and who is blamed for the issue of trafficking occurring, as well as if and which remedies are suggested to end trafficking in human beings.
Variables
The present study follows a similar methodology and looks precisely into the way the articles present and discuss the problem of trafficking (Hypothesis 1). The causes and solutions are analyzed to determine the issue frame of the articles (Hypothesis 2), and I also study which sources are cited (Hypothesis 3). In order to test for the reliability of the study using content analysis, an intercoder reliability test was conducted (Nuendorf, 2002), with 10% of the total number of articles being coded by myself and a co-coder. The Krippendorff’s Alpha of each variables is showcased in Table 1, 2 and 3 of Appendix 1.
To start with, in order to determine the way trafficking is defined to the audience and hence test hypothesis 1, I coded for the form of human trafficking discussed. Sobel (2014) includes the categories ‘sex trafficking’ and ‘forced labor’ to which I added ‘criminal exploitation’ in my coding, but removed after the analysis due to its very rare appearance. As shown in Table 1, the Krippendorffs’s alpha (kalpha) is of .81 for the variable measuring the presence of sex trafficking in the articles (G1), and of .75 for the variable measuring the presence of labor trafficking (G2). Hence, the reliability of those variables is high. During the data collection, I also coded for the prevalent form of trafficking. The variables are not mutually exclusive: the coders were told that several different forms of trafficking could dominate in one article.
Table 1 shows that the variable measuring the prevalence of trafficking for another purpose (P4) has a kalpha of 0, which can be due to a coding error, even though G4 also has a very low kalpha. Indeed, in one article, the co-coder coded 1 in P4, but 0 in the G4 of the same article, which is illogical. Considering the limited time and resources, another round of coding to assure better reliability was not possible. Hence, it is important to keep in mind that it will affect the reliability of my measures.
In order to test hypotheses 2, I looked at the presence or absence of 4 different issue frames:
crime, policy/legislation, human rights, and alternative. Following the coding and analysis process, a fifth frame appeared: awareness. The coder was asked to subjectively code for the presence of the issue frames. In the codebook, the crime frame was for example presented this way: “The crime frame is present if the article mentions the criminal aspect of human trafficking and the criminalization of activities related to trafficking. This includes articles about trials of “criminals” involved in human trafficking”. For more information on how each issue frame was defined, see section 2.4 in Codebook 1 in Appendix 2. I also coded for which frames felt dominant in the article. As Table 2 shows, most variables, except the variable measuring the presence of another issue frame (L5), are reliable. To ensure a more objective measurement of the issue frames, I also looked at the causes and solutions cited in each article, to classify them in one of the abovementioned issue frames. From that I created five objective issue frames: each cause and solution was considered to belong to one or more issue frame. The presence of those issue frames was then determined by the presence of causes and solutions associated to them.
In order to analyze the causes presented in the articles, I drew once again on previous studies.
I included, amongst others, Sobel’s (2014) categories: ‘socioeconomic conditions’, ‘victims’
ignorance’, ‘pressure from significant other’, ‘result of legislation or policy’, combined with Jonhston et al.’s (2015) ‘demand and access’ and ‘internet facilitating’. Moreover, as this study looks at different countries depending on the legal framework of sex work, ‘sex work legal framework’ were added. This was linked with who/what the blame for the trafficking occuring is being placed on, as Sobel (2014) codes: ‘family or environment or upbringing’,
‘lack of protective legislation’, ‘border patrol or immigration’, ‘trafficker’ or ‘person benefitting from the exploitation. As the present study is interested in the link between sex work and the framing of trafficking, ‘sex worker’s rights organizations’ were added.
The solutions presented in the articles to deal with human trafficking were also classified, influenced by previous studies. Jonhston et al. (2015) code for the following remedies:
‘increased punishment for traffickers and sex buyers’ - in the present study, ‘increased punishment for traffickers’ is a different category than ‘increased punishment for sex workers’’ or for ‘the customers’ - , ‘coordinating efforts and promoting NGOs’, ‘educating public and increasing awareness’, ‘legislation and policy changes’ - in the present study, there is a distinction between ‘border policy change’, looking at whether the articles mentions
‘tightening’ or ‘loosening’ of the borders; and ‘sex work policy change’, looking at whether the articles mentions ‘harder regulations on sex work’ or ‘looser regulations on sex work’.
For more specific information on which causes and solutions were coded for, see Codebook 1 in Appendix. Please note that the Codebook also includes variables measuring the consequences related to trafficking. However, as the consequences were rarely present in the articles, difficult to code for, and not included in most of the previous studies on the same topic, they were excluded from the analysis.
Based on previous findings, including Gulati (2010), Jonhston et al., (2014, 2015) and Sobel (2015), it was expected before the analysis that certain causes and solutions would be part of specific frames. Following the coding period and for the analysis, based on my experience as a coder and previous literature, the causes and solutions were classified in order to create obejctive issue frames. Certain causes or solutions were included in several issue frames. For example, ‘victim hoping for a better life/being fooled by trafficker’ is a cause that often appears in articles focusing on the trial of the trafficker, or on the police investigation. In that specific case, the cause is presented from the perspective of the criminal taking advantage of a victim's ‘foolishness’, and therefore related to the crime frame. However, in other articles, this same cause can be presented from the perspective of a person that, because of their socioeconomic and cultural conditions, will accept a risky situation that puts them in danger
in order to flee poverty or abuse. In that specific case, the cause is related to the alternative frame. For more information on which causes/solutions I analyzed as part of a certain issue frame, please see Codebook 2 in Appendix 2. Once the causes and solutions were classified as belonging to a particular issue frame, I looked at their frequency of appearance in each issue frame (See Appendix 3). Considering that, and to ensure a similar scale for each objective issue frame, I decided to code for the objective presence of the frames in the following way: if 0 causes or solutions from that frame appear in the article, the frame is considered absent; if 1 or 2 appear, the frame is considered present; if 3 or more of its causes or solutions appear, the frame is considered very present. As shown in Table 2, the objective frames are overall more reliable than the subjective ones. The objective crime frame (OF1), however, has a rather low kalpha. This can be due to the fact that some of the variables used to create OF1 scored very low in their kalpha, because of the rareness of the variable in the data: one difference between the coders strongly reduces the reliability.
Finally, to test hypothesis 3, the use of sources was analyzed in each articles. Inspired by a combination of previous studies while adding the angle of sex worker’s representation, I coded for the presence, amongst others, of the following sources: ‘law enforcement’,
‘governement/political source’, ‘advocacy group’ - with a distinction between ‘sex workers’
rights advocacy group’ and ‘human rights advocacy group’, ‘sex worker’ and
‘academic/expert’. For more information, see Codebook 1 in Appendix 2. Table 3 shows the reliability score of each source variable.
Results
The aim of this study is to determine how the framing of trafficking in human beings differ in the media when comparing countries with different legal models concerning sex work.
During the analysis process of the data, the comparison was therefore always made across the different legal frameworks. This is also due to the nature of the variables. As the independent variable is categorical, while the dependent variable is dichotomous, I looked at the proportion of dependent variable per category. I then tested the difference using Chi-square and the strength with Cramer’s V. The tables showcasing the results (Table 4 to Table 7) can be found in Appendix 1.
Form of trafficking discussed
Across all legal frameworks analyzed, half of the articles (50.3%) mention sex trafficking.
The second most mentionned form is labor trafficking, followed by no specific form mentionned. As shown in Figure 1a, the p-value for the measurement of sex trafficking is lower than .001, which implies that the amount of sex trafficking mentioned differs significantly depending on the legal framework, but the association is rather weak as Cramers’V equals .25. Sex trafficking is much more mentionned in articles from the swedish (61.3%), legalization (58.8%) and prohibition model (52.5%) than in articles from the decriminalization model (28.9%). The percentage of articles mentioning sex trafficking is only lower than the total (50.3%) in the decriminalization model. On the contrary, for labor trafficking, the trend is inversed: it is present in 65.8% of the articles from the decriminalization model, while only mentionned in 26.3% of the swedish model articles, and in 25% of the prohibition articles. The legalization model stands in between, with 36.8% of its articles refering to labor trafficking. We see the same patterns with the prevalent form of trafficking (Figure 1b). Consequently, hypothesis 1 is confirmed: countries with sex work decriminalized tend to associate human trafficking with sex work less often than countries where sex work is still criminalized. Finally, it is also interesting to note that the swedish and
prohibition model tend to mention only women and children as victims in a greater percentage of articles - respectively 33.8% and 35% - than the two other legal frameworks:11,8% in decriminalization, 20.6% in legalization model.
Figure 1a: Form of trafficking mentioned per legal framework
Figure 1b: Prevalent form of trafficking per legal framework
Framing of the issue
In order to text for hypothesis 2, the presence of the different issue frames were analyzed both subjectively and objectively. The results were then compared. For further information on the difference between subjective and objective issue frames, please see the Methods section.
For the subjective frames, both the presence of the frames, and the most prevalent frames in each article were analyzed. See results in Figure 2a/2b. The policy frame was found much more present in the decriminalization (57.9%) and prohibition model (41.3%) than in the legalization (26.5%) or swedish model (26.3%). As the p-value is lower than .001, the difference is significant, but the association rather weak as Cramer’s V is of .27. Same goes for the human rights frame: it is significantly more present in the decriminalization (43.4%) and prohibition model (40%) than in the legalization (27.9%) or Swedish model (17.5%).
Unfortunately, the reliability of the variable measuring the presence of the human rights frame is rather low. For the other frames, the differences are significant and the variables moderately related. The awareness frame is the most common other frame, and it appears in 23.5% of the article from the legalization model, and in 10% of the articles from the swedish model, in contrast with only 1.3% for decriminalization and prohibition. It is also interesting to note that it is only in the Swedish model that the sex work frame appears, in 3.8% of the articles. However, it is important to keep in mind that the reliability of this variable is low.
As for which frames are the most prevalent, it is interesting to note that the crime frame is significantly dominant in all 4 legal frameworks. It is particularly prevailing in articles from the Swedish model, followed by prohibition, legalization, and finally decriminalization. On the contrary, the policy framework is much more prevalent in the decriminalization model (35.5%), followed by legalization (11.8%), prohibition (10%) and Swedish model (8,8%).
Accordingly, hypothesis 2 cannot be completely confirmed: in the decriminalization model, the media does not necessarily frame trafficking in human beings more as and alternative or awareness issue. However, they do frame it more frequently as a human rights issue than in the swedish model. Also, countries with legal frameworks unfavorable to sex workers' rights (prohibition and Swedish model) frame trafficking more frequently as a crime issue than in the decriminalizaiton model. However, in the decriminalization model, human trafficking is framed more often as a policy issue than in the swedish and prohibition model. These findings suggest the need for additional research that looks into the policy frame, making a differenec between policies that are pro sex workers’s rights and policies that are calling for further criminalization.
Figure 2a: Presence of subjective issue frame per legal framework
Figure 2b: Prevalence of subjective issue frame per legal framework
Using the causes and solutions as indicators, the objective frames were determined. See results in Figure3a/3b. None of the differences across the legal frameworks are significant, and Cramers’V is very low, which demonstrates a noticeably weak correlation between the variables. Yet, it is interesting to note that the difference across the legal frameworks is much smaller than with the subjective issue frames. When the frames are determined in a more
“objective way” - determined by the presence of indicators rather than by the “feeling” of the coder after reading the article- the results are more homogeneous and less sweeping, or perhaps, exaggerated. The difference probably has to do with my culture as a European, which might have affected my coding. Indeed, when looking at the correlation between subjective and objective issue frames, the value of Cramer's V varies strongly from country to country, which means that they are associated in certain countries, while not in others. This leads me to believe that there could be a cultural bias from me as a coder: I might not have noticed certain subjective frames in certain countries, as I do not have the cultural tools to understand them. Similarly, I might have missed the presence of certain causes or solutions because they were not presented in a way that made them noticeable to me.
Figure 3a: Presence of objective issue frame per legal framework
Figure 3b: Prevalence of objective issue frame per legal framework
Sources mentioned
The sources mentionned across the articles also vary depending on the legal framework of sex work. As shown in Table 7, law enforcement officials are more present in articles from the swedish (45%) and prohibition model (41.3%), compared to the decriminalization (26.3%) and legalization model (26.5%). The effect of framework on visibility of law enforcement officials is significant, as the p-value is of .023, but the correlation between the variables is very weak, with a Cramers’ V of .18. Politicians and state officials are present in 60.5% of the articles from the decriminalization model and 53.8% of the prohibition model while only in 30% of the swedish model. Sex workers and the sources advocating for their rights are slightly more represented in the decriminalization model, in comparison to the swedish or prohibition model. However, these differences are not significant, and the correlations between the variables are very weak. On the contrary, advocates for human rights and for the rights of survivors of trafficking are more present in articles from the swedish or prohibition model, than in articles from the decriminalization or legalization model. But again, the p-value shows that the differences across legal frameworks are non significant, and the correlation is very weak. Survivors of trafficking and their representatives are the least cited in articles from decriminalization, compared to the three other legal frameworks.
However, the difference is non-significant. Finally, even though the results are not significiant and with a very low Cramers’V, it is interesting to note that there are less experts or academic reports cited in the articles from the swedish model (8.8%) than in articles from the decriminalization (18.4%), legalization (16.2%) or prohibition model (17.5%). To conclude, more research is needed in regards to the third hypothesis. Indeed, the results from the current study confirm that some official sources, such as law enforcement officials, are
more cited in countries with legal frameworks non-favorable to sex worker’s rights. However, it is not possible to confirm whether this is the case for all official sources. Moreover, more extensive research is needed to determine whether countries with the decriminalization or legalization model use more sources representing the voices of sex workers and of survivors of trafficking than the prohibition or swedish model.
Discussion and conclusion
The aim of this study has been to determine how the framing of trafficking in human beings differ in the media when comparing countries with different legal models of sex work, through a content analysis of newspapers in 7 countries. I hypothesized that countries with a legal framework more favorable to sex worker’s rights would frame human trafficking as less related to sex work, use more often a human rights, awareness and alternative approach to the topic, and cite more frequently sources representing the voices of survivors of trafficking and sex workers. By doing so, this study seems to be one of the first to examine the influence the legal framework of sex work has on the framing of human trafficking in newspapers.
The first major finding is that decriminalization and legalization countries tend to discuss labor trafficking to a greater extent than countries whose laws are non-favorable to sex workers' rights (prohibition and swedish model). On the contrary, prohibition and swedish model countries tend to excessively mention sex trafficking in comparison with the decriminalization model. Moreover, there is a bigger tendency to mention only women and children in countries from the prohibition and swedish model. These findings confirm the first hypothesis, by showing that countries that are favorable to sex workers’ rights are less biased towards sex trafficking in their coverage. Similarly, the white female sexual victim, as
mentioned by Doezema (1999), Pajnik (2010), and Soderlund (2005), is portrayed in New Zealand to a smaller extent than in prohibition or swedish model countries. As sex trafficking is often used by abolitionnists or religiours conservatives as an argument to further criminalize sex work (see Chapkis 2003; Doezema 2000; Ward, 2020; Weitzer 2007), it is important to acknowledge that newspapers in prohibtion and swedish model countries tend to frame trafficking as largely related to sex work, which is not necessarily representative of the realities of the current human trafficking trends. This misrepresentation of human trafficking, and the resulting laws, have serious consequences on the working and living conditions of sex workers. It also influences the way countries fight human trafficking overall. A consequence of this bias towards sex trafficking, at the expense of for example labor trafficking, might be policies and legislations that are not effectively responding to the different forms of exploitation that occur.
When it comes to the way human trafficking is framed as an issue, the policy/legislation frame is much more present in the decriminalization model than in other models. These results can be due to the legal framework of migration in New Zealand. While the country decriminalized sex work for its citizens and people holding permanent residency in 2003, migrant sex workers were excluded from that legal protection (Bennachie et al., 2021).
Recently, with the influence of the United-States, debates on anti-trafficking policies in New Zealand have shifted towards a law-and-order framing instead of taking a gender equity, human rights, and harm-minimization approach (Showden, 2017). A natural progression of the present study would be to look more in-depth into the distinction between policies that are pro-sex workers’ and migrant rights, versus policies that are leaning toward the criminalization of sex work and migration.
Furthermore, in this study, the awareness frame appears as more present in the legalization, prohibition and swedish model. Further research is needed to establish a greater degree of accuracy on this matter. Especially, it would be interesting to make a distinction between different forms of awareness messages. Is it awareness in the line of criminals-kidnapping-innocent-young-girls, or rather awareness campaigns targeting vulnerable communities and acknowledging that certain portions of the population are more at risk of being trafficked due to their socio-economic conditions?
Overall, considerably more work needs to be done on the matter of issue frames, adding more indicators for each frame in order to enhance their precision. For example, including sources as indicators of a specific issue frame. To emphasize, the present study found law enforcement officials to be more present in the prohibition and swedish model, while government officials were cited further in the decriminalization model, which probably has to do with the dominant issue frames. Indeed, the crime frame is more prevalent in prohibition and swedish model countries, while the policy frame is dominant to a larger extent in New Zealand. On another note, surprisingly, human rights defenders appeared more in prohibition and swedish model newspapers. However, the results are non-significant, and the variables measuring the presence of those sources are non-realiable. For a better understanding of the presence of these sources, further research is needed. Finally, even though the results are non-significant, I found that sex workers are more cited in the decriminalization model than in any other. This probably has to do with the fact that it is easier for sex workers to speak up in New Zealand than in coutnries where they risk legal punishments if they reveal their profession.
Apart from the limitations mentioned above, the major limitation of this study is the size of the sample and the lack of several coders. Future studies on this specific topic should include
more countries, with a wider range of newspapers to fully represent the dominant narrative of the country. Moreover, having several coders would require greater effort to develop a clear codebook for better reliability of the measurements. More broadly, qualitative research is also needed to compare more in depth the way trafficking is discussed in decriminalization versus prohibition countries. In the present study, more precised and nuanced observation are lacking, partly due to the method chosen. Finally, it would be interesting to look at how the survivors of trafficking are portrayed, as well as which numbers and sources are cited to discuss human trafficking. Indeed, more insight on what is presented as the “truth” on human trafficking in different countries would be very interesting.
To conclude, the present study suggests there is a lot to research on the way the legal framework of sex work influences the coverage of human trafficking in the media. Indeed, in countries where sex work is decrimininalized, or leaning towards it, there is less association of sex work with human trafficking. Instead, other more common forms of trafficking, such as labor exploitation, receive a bigger part of the coverage than in countries where sex work is criminalized. Hence, future research should dive into this, by continuing to study if the coverage of human traffikcing in decrimiminalization countries is less biased towards cartain frames or issues, than in countries that consider sex work as something inhenrently criminal.
References
Andrijasevic, R., & Mai, N. (2016). Editorial: Trafficking (in) Representations:
Understanding the recurring appeal of victimhood and slavery in neoliberal times.
Anti-Trafficking Review, 7. https://doi.org/10.14197/atr.20121771
Bacchi, C. (2009). The issue of intentionality in frame theory: the need for reflexive framing [E-book]. Dans E. Lombardo, P. Meier, & M. Verloo (Éds.), The Discursive Politics Of
Gender Equality Stretching, Bending, And Policymaking (1st ed., p. 19‑35). Routledge.
https://doi.org/10.4324/9780203881330
Bennachie, C., Pickering, A., Lee, J., Macioti, P., Mai, N., Fehrenbacher, A., Giametta, C., Hoefinger, H., & Musto, J. (2021). Unfinished Decriminalization: The Impact of Section 19 of the Prostitution Reform Act 2003 on Migrant Sex Workers’ Rights and Lives in Aotearoa New Zealand. Social Sciences, 10(5), 179. https://doi.org/10.3390/socsci10050179
Brysk. (2009). Beyond Framing and Shaming: Human Trafficking, Human Security and Human Rights. Journal of Human Security, 5(3), 8–21. https://doi.org/10.3316/JHS0503008
Chapkis. (2003) Trafficking, Migration, and the Law. Gender & Society, 17 (6), 923-37. DOI:
10.1177/0891243203257477
Denton. (2010) International News Coverage of Human Trafficking Arrests and Prosecutions:
A Content Analysis, Women & Criminal Justice, 20 (1-2), 10-26, DOI:
10.1080/08974451003641321/
Doezema, J. (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.
https://doi.org/10.1007/s12147-999-0021-9
Doezema, J. (2010) Sex Slaves and Discourse Masters: The construction of trafficking, Zed Books, London.
ICRSE - International Committee on the Rights of Sex Workers in Europe. (2018, march).
Diverse, Resilient, Powerful - Intersectional Activism Toolkit for Sex Workers and Allies.
https://www.eswalliance.org/intersectional_activism_toolkit_for_sex_workers_and_allies
Johnston, Friedman, & Sobel. (2015). Framing an Emerging Issue: How U.S. Print and Broadcast News Media Covered Sex Trafficking, 2008-2012. Journal of Human Trafficking, 1(3), 235-254. https://www.tandfonline.com/doi/abs/10.1080/23322705.2014.993876
Johnston, Friedman & Shafer. (2014) Framing the Problem of Sex Trafficking, Feminist Media Studies, 14:3, 419-436, DOI: 10.1080/14680777.2012.740492
Larson, S. G. (2006) Media & Minorities: The Politics of Race in News and Entertainment.
Lanham, MD: Rowman & Littlefield Publishers, Inc.
Neuendorf, K. A. (2002). The content analysis guidebook. Thousand Oaks, CA: Sage.
Pajnik. (2010) Media Framing of Trafficking, International Feminist Journal of Politics, Vol.
12, 45-64, https://doi.org/10.1080/14616740903429114
PEW RESEARCH CENTER. 2010. “New media, old media: How blogs and social media agendas relate and differ from traditional press.”, May 23. Accessed March 3rd, 2022.
http://pewresearch. org/pubs/1602/new-media-review-differences-from-traditional-press.
Sharapov, K., Hoff, S., & Gerasimov, B. (2019). Editorial: Knowledge is Power, Ignorance is Bliss: Public perceptions and responses to human trafficking. Anti-Trafficking Review, 13 (1–11). https://doi.org/10.14197/atr.201219131
Showden. (2017). From human rights to law and order : the changing relationship between trafficking and prostitution in Aotearoa/New Zealand policy discourse. Women’s Studies Journal, 31(1), 5–21.
Sobel. (2014) Chronicling a crisis: media framing of human trafficking in India, Thailand, and the USA, Asian Journal of Communication, 24(4), 315-332, DOI:
10.1080/01292986.2014.903425
Soderlund. (2005). Running from the Rescuers: New U.S. Crusades Against Sex Trafficking
and the Rhetoric of Abolition. NWSA Journal, 17(3), 64–87.
https://doi.org/10.2979/NWS.2005.17.3.64
Weitzer, R. (2007). The Social Construction of Sex Trafficking : Ideology and Institutionalization of a Moral Crusade. Politics & Society, 35(3), 447‑475.
https://doi.org/10.1177/0032329207304319
Ward. (2020). “Framing figures” and the campaign for sex purchase criminalization in Ireland: A Lakoffian analysis. Irish Journal of Sociology : IJS, 28(3), 314–332.
https://doi.org/10.1177/0791603520951754
Appendix 1 - Tables
Purpose of variable Name of
variable
Krippendorffs’
s alpha The article mentions: Sex trafficking
Labor trafficking
Trafficking for another purpose No specific form of trafficking Only women and children as victims
G1 G2 G4 G5 G7
.81 .75 .28 .57 .67
The prevalent form of trafficking in the article is:
Sex trafficking Labor trafficking
Trafficking for another purpose No specific form of trafficking
P1 P2 P4 P5
.71 .64 0 .48 Table 1: Reliability of variables used to determine the way trafficking is defined to the audience
Purpose of variable Name of
variable
Krippendorffs’
s alpha In the article, the
following subjective issue frame is present:
Crime
Policy/Legislation Human Rights Alternative Other frame
L1 L2 L3 L4 L5
.81 .69 .6 .77 .55
In the articles, the
following subjective issue frame is prevalent:
Crime
Policy/Legislation Human Rights Alternative Other frame
M1 M2 M3 M4 M5
.62 .72 1 .79 .65
In the article, the
following objective issue frame is present:
Crime
Policy/Legislation Human Rights Alternative Awareness
OF1 OF2 OF3 OF4 OF5
.53 .77 .72 .87 .84 Table 2: Reliability of variables used to determine the issue frame of the article
Purpose of variable Name of variable
Krippendorffs’s alpha
The article contains the following source:
Law enforcement officer, agency, police report Judge, prosecutor, or court document
Politicians/government/state officials
Human rights organizations/advocate for human rights Advocate for rights of victims of trafficking
Sex workers’ rights advocate
Victim of trafficking or their representatives Sex worker or their representative
Academic/expert/research report
N1 N2 N3 N6 N7 N8 N10 N11 N15
.81 .76 .73 .71 .79 1 .77 1 1 Table 3: Reliability of variables measuring the presence of sources
P- value
Cramers
’V
Decriminal -ization
N = 76
Legalization N = 68
Swedish model N = 80
Prohibition N = 80
Total N = 304
The article mentions:
Sex trafficking Labor trafficking
Another form of trafficking No specific form of
trafficking
Only women and children as victims
The prevalent form of trafficking is:
Sex trafficking Labor trafficking
Another form of trafficking No specific form of
trafficking
,000 ,000 ,187 ,151
,002
,008 ,000 ,360 ,098
,25 ,34 ,23 ,13
,22
,20 ,39 ,13 ,14
28,9%
65,8%
9,2%
18,4%
11,8%
22,40%
42,10%
2,60%
29,30%
58,8%
36,8%
11,8%
26,5%
20,6%
33,80%
14,70%
4,40%
42,60%
61,3%
26,3%
16,2%
22,5%
33,8%
48,80%
8,80%
7,50%
35%
52,5%
25%
20%
33,8%
35%
36,30%
5%
6,20%
47,50%
50,3%
38,2%
14,5%
25,3%
25,7%
35,50%
17,40%
5,30%
38,60%
Table 4: Form of trafficking mentionned per legal framework
P- value
Cramers
’V Decriminal -ization
N = 76
Legalizatio n N = 68
Swedish model N = 80
Prohibitio n N = 80
Total N = 304
The following subjective frame is present:
Crime Policy Human rights Alternative Awareness
The following subjective frame is prevalent:
Crime Policy Human rights Alternative Awareness
,101 ,000 ,002 ,254 ,006
,044 ,000 ,765 ,564 ,011
,14 ,27 ,22 ,11 ,54
,16 ,30 ,06 ,08 ,50
64,5%
57,9%
43,4%
31,6%
1,3%
30,3%
35,5%
10,5%
6,6%
1,3%
48,5%
26,5%
27,9%
25%
23,5%
39,7%
11,8%
10,3%
10,3%
19,1%
67,5%
26,3%
17,5%
18,8%
10%
52,5%
8,8%
6,3%
7,5%
1,3%
60%
41,3%
40%
30%
1,3%
40%
10%
10%
12,5%
10%
60,5%
38,2%
32,2%
26,3%
8,6%
40,8%
16,4%
9,2%
9,2%
7,6%
Table 5: Presence of subjective issue frame per legal framework
P- value
Cramers
’V Decriminal -ization
N = 76
Legalization N = 68
Swedish model N = 80
Prohibitio n N = 80
Total N = 304
The following obejctive frame is present:
Crime Policy Human rights Alternative Awareness
The following objective frame is prevalent:
Crime Policy Human rights Alternative Awareness
,470 ,256 ,331 ,956 ,725
,470 ,256 ,331 ,956 ,725
,20 ,11 ,11 ,05 ,08
,20 ,11 ,11 ,05 ,08
48,7%
21,5%
23,7%
34,2%
15,8%
6,6%
15,8%
2,6%
11,8%
1,3%
36,8%
23,5%
19,1%
33,8%
16,2%
5,9%
11,3%
0%
7,4%
0%
45%
20%
18,8%
33,8%
21,3%
7,5%
3,8%
6,3%
7,5%
1,3%
37,5%
20%
26,3%
36,3%
22,5%
12,5%
11,3%
2,5%
10%
0%
42,1%
21,2%
22%
34,5%
19,1%
8,2%
9,5%
3%
9,2%
0,7%
Table 6: Presence of obejctive frame per legal framework
P- value
Cramer
s’V Decriminal -ization
N = 76
Legalizatio n N = 68
Swedish model N = 80
Prohibitio n N = 80
Total N = 304
The sources present are:
Law enforcement officials Judge, prosecutor
Politicians/state officials Advocate for human rights Advocate for rights of survivors of trafficking Sex workers’ rights advocate Survivor of
trafficking/representative Sex worker/representative Academic report/expert
,023 ,103 ,001 ,224 ,124
,197 ,653
,219 ,310
,18 ,14 ,23 ,12 ,14
,12 ,07
,12 ,11
26,3%
32,9%
60,5%
15,8%
13,2%
7,9%
17,1%
6,6%
18,4%
26,5%
50%
45,6%
16,2%
11,8%
1,5%
19,1%
0%
16,2%
45%
33,8%
30%
27,5%
18,8%
3,8%
25%
3,8%
8,8%
41,3%
33,8%
53,8%
22,5%
25%
2,5%
20%
5%
17,5%
35,2%
37,2%
47,4%
20,7%
17,4%
3,9%
20,4%
3,9%
15,1%
Table 7: Percentage of sources mentionned per legal framework
Appendix 2 - Codebooks Codebook 1
Coverage of human trafficking in newspapers
Introduction
This codebook was designed to help you in the process of coding the content of news stories (i.e. the unit of analysis is the news story). Specifically, it will be used to quantitatively analyze the use of certain frames in the coverage of trafficking in human beings in 7 different countries.
General Information
In the first part of this codebook, you are asked to code for some general information on the articles.
Start by briefly skimming through the article and answering questions A.
A. Article ID
For every coded article, there needs to be a possibility to come back to it. Please indicate the 3 first words of the Title.
B. Coder ID
◯ Julia (1)
◯ Sandra (2)
C. Publishing Date
The date the article was published is coded in the format DD-MM-YYYY (e.g. January 20th 2022 is coded as 20-01-2022). Please note that there might be 2 dates on the article: the
‘publishing’ date and the ‘last edited’ date. Write down the publishing date, not the ‘update’
or ‘last edited’ date. If only one date is present in the article, write that one down. If no date is to be found, leave a blank space.
D. Newspaper
The Newspaper of the article is coded as follows:
New Zealand
- The News Zealand Herald, NZ (1) - The Press, NZ (2)
- The Dominion Post, NZ (3) United States of America
- The New York Times, US (4) - USA Today, US (5)
South Africa
- Sunday Times, ZA (6)
- The Times (South Africa), ZA (7) - Cape Times, ZA (8)
- Daily News (South Africa), ZA (9) - Pretoria News, ZA (10)
Luxembourg
