“Journey into hell […where] migrants froze to death”; a critical stylistic analysis of European newspapers’ first response to the 2019 Essex Lorry deaths

In the early hours of October 23rd, 2019, 39 people were found dead in a refrigerated lorry in Grays, Essex, UK. This case attracted media interest across the world; in the 48-h period after the story broke, reporting on this discovery extended to newspapers not just in the UK, but also across Europe. This study uses elements of Critical Stylistics (Jeffries 2010) to analyse and compare first response articles published by European dailies in relation to the event at Grays, to address the nature of this reporting. We found that linguistic choices tend to dramatise what happened, criminalise victims, and even presume the driver’s innocence, with the international criminal network he is presupposed to be part of remaining only speculated on. Though there are attempts to distribute some accountability to governments and policies, as well as structural systemic factors such as war and poverty, responsibility for these factors tends to be diffused, and hence unallocated, this helping ultimately justify draconic law enforcement and border security policies. By highlighting linguistic trends and underlying ideologies which we in turn question, we address the need to tend to the structural causes of such transnational people movement-related crime (i.e. trafficking and smuggling) and shift accountability to governments.


Introduction
In the early hours of October 23 rd , 2019, 39 people were found dead in a refrigerated lorry in Grays, Essex, UK. Given the victims' supposedly non-British nationality and hence the crime's transnational nature, the press assumed that an organised criminal enterprise engaged in either transnational human trafficking or people smuggling was behind it. Though both crimes involve people movement and exploitation (the extent of the latter being influenced by the length of the journey's 'movement'), the two are legally defined differently. Trafficking is a crime against an individual who gets treated as a commodity, while smuggling is a crime against the state into which an immigrant wilfully gains illegal access (though the circumstances leading to the individual making the journey often resemble those of trafficked individuals). As also noted in Gregoriou and Ras (2018a), the reality of trafficking/smuggling is not quite so clear and the two crimes are intertwined; consent may be blurry or absent at various stages of either process, and those who are smuggled may, if exploited, become victims of trafficking in a legal sense (Wylie 2016: 6;Coghlan and Wylie 2011). Doonan's (2016) analysis of the US discourse(s) around trafficking and migration indicates that the legal and discursive distinction between these two acts persists also as a result of anxieties relating to national security and a patriarchal notion of the United States' duties and global role; one may presume that this distinction is maintained in other states' discourse(s) for similar reasons (see also O'Neill 2001 for a discussion of the blurred boundary between trafficking syndicates and social networks, marked by state definitions of the legitimacy of migration). Such interconnectedness no doubt contributes to the media's difficulties in classifying criminal events such as the one at Grays, particularly during the early stage of the police investigation we chose to focus on, when little is known about what actually happened. At that point, journalists shaping their story may have to rely on what they perceive to be similar incidents, as well as what they 'know' about trafficking and smuggling; focusing on the first 48-h of reporting on this event allows us to uncover the nature of that knowledge.
The power of the media lies in their ability to naturalise meanings and to frame, order and name our social reality (Couldry 2000: 57). Media representations of modern slavery and human trafficking, in particular, tend to reproduce and reinforce existing and persisting societal myths (Birks and Gardner 2019: 72). Previous research on British newspaper reporting on such cases over a 17-year period has found not only a conflation of 'human trafficking' and 'people smuggling', but also that the stereotype of trafficking is one in which criminal organisations, often suggested to be Eastern European, (violently) force and detain young girls in forced prostitution, the latter being idealised victims who these narratives overfocus on at the expense of other, real, victims (Gregoriou and Ras 2018b). The suggestion made by these newspapers that these criminal organisations are 1) foreign and 2) generally mono-ethnic draws on broader myths regarding criminal organisations (see also Leiva and Bright 2015, though they found limited reference to mono-ethnic criminal organisations in the Australian newspapers they examined). Other common, yet problematic, assumptions include a belief that such crimes are notable because they are supposedly uncommon in developed nations such as Britain, are associated with sex work or irregular migration, and that an element of consent or payment by those moved means that trafficking has not occurred (Birks and Gardner 2019: 72). Even more so, such narratives deny victims agency and overlook the problem's structural causes; '[t]he failure to highlight these root causes removes the call for structural reforms from the public imagination and shifts accountability and responsibility away from governments' (Sharapov et al 2019: 8-9). Indeed, as Rawlinson (1998) argues, there is a tendency in the media, but also more generally, to see organised crime as somehow separate from society, rather than a response to various systemic factors that create a supply and a demand to which criminal organisations can, and do, respond. Our prior research into media representations of human trafficking and modern slavery focused on, primarily, the narrative of these crimes that persists over time, and the stereotyping of victims, offenders and other parties (Gregoriou andRas 2018a, 2018b;Muždeka 2018). In contrast, this paper focuses on reporting over a very short timeframe, comparing continental European responses to each other and to the UK responses, thereby also allowing comparisons between newspapers. By qualitatively examining the first responses of European daily newspapers to the event at Grays, we expect to uncover the 'go-to' narrative, thereby illuminating the stereotypes that underlie such reporting. This is important given that the lorry in question appears to have been driven all around Europe. Comparing British reporting (where the discovery was made) with non-Anglophone newspaper coverage from across Europe, this paper adds to existing research on (primarily Anglophone) representations of human trafficking and people smuggling. In this regard, this paper also identifies those elements of the human trafficking and people smuggling narratives that are pervasive across cultures, and languages. As noted in Gregoriou (2018: v), communicating stories of this kind in translation is not without ideological challenges, some of which this paper highlights.
We next outline our data collection method before giving an overview of the nature of the critical stylistic approach undertaken to analysing this data.

Data collection
Given the discovery of the bodies in the early hours of October 23 rd , and the need to analyse (inter)national case reporting, we opted to collect data generated throughout the 23 rd , and also the 24 th , of that month, allowing the press outside the UK time to respond to UK reports. We settled on a manageable collection of 78 big national daily newspaper articles so as to qualitatively explore texts representative of the relevant national newspaper (and related political) landscape. Doing so would give us the (inter)national 'go-to', or first, response to this major incident.
Though we decided on a collection of texts from a variety of European languages and countries, we were limited to languages we are CEFR C2 + (native or nearnative) users of; to gain an impression of texts in Greek, English, Serbian, Croatian, and Dutch, we explored the case's early coverage in Cyprus, Greece, UK, Serbia, Croatia, Flanders and Brussels, and the Netherlands. An overview of the newspapers that published relevant articles within those first 48-h and that have been included in this study can be found in Table 1. Note that there may be other newspapers in these countries that may be considered as national newspapers; if they have not been included in this study, it is due to their not having published articles on this event within the relevant time frame.
To collect the articles published by mainstream UK papers, the database of Lexis Nexis UK was used; the parameters for the search were set to only show articles published on October 23 rd and 24 th in all national British papers, using the search term 'Grays', this being the town in which these people were found. This search produced 16 articles, published by 12 newspapers, both broadsheet and popular.
To trace the case's Greek language coverage online, we opted for data from countries where Greek is spoken as a first language, i.e. Greece and Cyprus. Due to the lack of an easily accessible archive through which to trace hardcopy paper articles in Greek, we found ourselves limited to those Greek newstexts published online. We used 'google.gr' and 'google.com.cy' through which to respectively search for Greek and Cypriot 'News' pages listing the words 'φορτηγό' and 'νταλίκα' (Greek The Guardian Left 2 The Independent Left 3 The London Evening Standard Centre-right 1 The MailOnline Right 1 The Sun Right 1 The Times Right 1 322 for 'lorry' and 'trailer' respectively, given the victims' discovery in a lorry trailer), then limiting the searches to those significant Greek language online newspaper texts published on, primarily, the 23 rd , and then, if major newspaper articles were not secured, the 24 th of October 2019. Using this method, we were able to locate articles from seven (of twenty) major Greek and four (of six) major Cypriot newspapers. Given also the need to trace first-response pieces from this two-day period, the resulting Greek language corpus altogether came to consist of nineteen variedlength news pieces from the eleven newspapers; the corpus consisted of multiple articles by some newspapers as these published several pieces over the given period online, as more case detail came to rise. For Dutch-language reporting, data were collected from European territories in which Dutch (or a variety thereof) is a main language; these are the Netherlands, in which Netherlandic-Dutch is the standard variety; Flanders, the northern region of Belgium, and Brussels, the capital region of Belgium, in both of which Belgian-Dutch is the standard variety of Dutch used. To examine Netherlandic-Dutch newspapers, use was made of the Nexis Uni (NL) database. The date-parameters were set to only include articles published on either October 23 rd or 24 th , with all duplicate articles grouped to focus on the mainstream national newspapers. More regionally focused newspapers such as Het Parool and De Provinciale Zeeuwse Courant (PZC), despite their sizeable readerships, were excluded, in the same way that The Yorkshire Evening Post is not included in the English corpus. The search term used was 'Grays'; this search produced fourteen articles published by nine different newspapers, both quality and popular newspapers. For Belgian-Dutch language news articles, only articles published online were included, taken from the websites of mainstream national newspapers, again using the search term 'Grays' and limiting the results to articles published on October 23 rd and 24 th . This search produced seven articles published by six different newspapers.
For the selection of the Serbian language articles, we used EBART, the most extensive Serbian online news media database. The search terms used were expressions 'kamion smrti u Engleskoj' (the death lorry in England) and 'hladnjača smrti u Engleskoj' (the refrigerated lorry of death in England), since these specify both the circumstances and the location of the event that took place. The location criterion was especially useful since it helped us distinguish between this particular incident and similar past cases that occurred in similar circumstances (i.e. dead bodies found in a lorry trailer) elsewhere. The search produced fifteen articles from eleven different newspapers, both broadsheet and tabloid, published on October 23 rd and 24 th . In the cases where two articles from the same outlet were used, the first was always a report containing very little information, while the subsequent report provided a more detailed account. Due to the unavailability of an extensive online database for Croatian news articles, these were found via 'google.hr', using the same search terms as for the Serbian sources. Seven news texts were collected, all published in the specified 48-h window.
Note that these corpora reflect what these newspapers write on this topic in the first 48-h; this is not what their readers read. Whilst, for instance, the UK and Netherlandic-Dutch left-wing papers generally generate more and longer articles on this topic, the right-wing papers (e.g. The Sun, The MailOnline, De Telegraaf) have substantially larger readerships.

Method
Though stylistics is, strictly speaking, defined as the study of literary texts, critical stylistics (see Jeffries 2010) is a field defined by the use of linguistic methodology so as to critically engage with the style of literary as well as non-literary texts. Critical stylistics (henceforth CS) is, as explained by Jeffries (2010Jeffries ( , 2014, a response to the lack of a defined toolkit of critical discourse analysis (henceforth CDA), thereby situating it on the more linguistics-focused end of the spectrum of critical discourse analyses. In that regard, CS is more concerned with the 'description' and 'interpretation'-aspects of CDA than 'traditional' critical discourse analyses might be considered to be; 'traditional' CDA tends to focus more on the 'explanation' (Jeffries 2010). We specifically focus on the (linguistic) strategies used to refer to the actors mentioned in media texts, to analyse how the text creator understands their role in the described situation and broader context, but also to examine how the text creator judges these actors. Such linguistic strategies include names and descriptions, metaphors, grammatical allocation of agency, possibility and so on (see also Jeffries 2010, for a full description of the critical stylistic toolkit).
We start with an investigation of the framing of the event, before investigating metaphors alongside transitivity patterns. Transitivity relates to grammatical aspects concerning responsibility, accountability and blame, which the metaphors also touch on. We then turn to look at the naming and describing of the case's victims. For instance, newspapers' tendency to refer to the deceased as 'migrants' (with all the connotations this word carries, albeit with various strengths in various countries) when no other details are yet known, highlights that particular aspect of these people's identity. An initial label such as this must be linked to earlier research findings which suggest that human trafficking, modern slavery and people smuggling are framed as issues of national (border) security (see Farrell and Fahy 2009), as opposed to, for instance, issues to do with personal hardship, or death/crime on a large scale. Lastly, we consider the naming and describing of the crime's perpetrators. Unknown as these were at this stage, we consider the focus on the lorry driver, but also any hints as to the nature of his and others' involvement to the crime, even at this early stage of the investigation and of reporting.

Framing
Through framing, which focuses on 'how news content affects and influences news consumers' (Johnson-Cartee 2005: 25), politicians, issue advocates and stakeholders use news texts to 'communicate their preferred meaning of events' (D'Angelo and Kuypers 2010: 1). For example, all European texts frame the event primarily as a great tragedy involving a big police investigation (quoting politicians and police officials as sources), resulting from the possible involvement of organised crime in the form of trafficking and smuggling. This framing is reminiscent of the 'crime' frame described by Farrell and Fahy (2009), that they link to the exponential rise in reporting on trafficking in the US in the early 2000s. Whilst this is a problematic frame, as it legitimises harsh law enforcement responses that are, ultimately, a cause of secondary victimisation (see Gregoriou and Ras 2018a;Desyllas 2007), this 'crime' frame has also been shown to be hugely important in creating and sustaining international support for anti-trafficking resolutions and conventions (Charnysh, Lloyd and Simons 2015). Sobel (2014) similarly shows that the 'crime' frame is the predominant one in US, Thai and Indian reporting on human trafficking; Pajnik (2010) also shows 'criminalisation' as a key frame in Slovene reporting.
Both Serbian and Croatian reports conflate trafficking and smuggling and maintain focus on this event having to do with a 'serious and organised crime' ('teški i organizirani kriminal' in the Croatian Dnevnik), with the Serbian Danas calling this 'one of the most tragic cases of human trafficking in Great Britain' ('jedan od najtragičnijih slučajeva trgovine ljudima u toj zemlji, tj. Velikoj Britaniji'). The Greek and Cypriot newspapers frame the event as (mass) murder related to most possibly organised crime to do with immigration ('οργανωμένο έγκλημα μετανάστευσης', as in 'organised immigration crime'), and specifically trafficking and/or smuggling of immigrant victims. Given this uncertainty, some wording is in reference to people/body trading ('σωματεμπορίας'), slave trading ('δουλεμπόριο'), not to mention the English word 'trafficking' itself; some refers to people movement ('διακίνηση ανθρώπων'); and some to people smuggling ('λαθρεμπόριο'). Netherlandic-Dutch papers all over the political spectrum, in particular the Calvinist-orthodox Het Nederlands Dagblad and the left-wing De Volkskrant, are interested in the question of whether this is people smuggling or human trafficking.
The Belgian-Dutch De Morgen's emphasis on the role of a 'highly intelligent professional criminal gang' [that smuggles people] ('een heel erg professionele misdaadbende') exemplifies how this framing must also be linked to Christie's (1986) 'narkohai', i.e. the spectre of a big, scary, powerful perpetrator, which justifies farreaching law enforcement interventions. The continuous focus on these victims being foreign, migrants, this being a case of transnational smuggling/trafficking, and having to do with organised crime serves to further justify far-reaching border control and anti-migration regulations and interventions. Even the route taken by the lorry -as one of the main focuses of priming -is presented as a direct result of the fact that the borders in Dover and Calais are under strict control ('rigorozna kontrola', the Croatian 24 sata). The question of whether the refrigerated trailer did pass through the Belgian port of Zeebrugge, and whether controllers at Zeebrugge could have detected these people, is a central concern for Belgian-Dutch newspapers. The issue of the transit route is also raised by the Greek, Croatian and Serbian newspapers, though there is no focus on the controllers. While all reports, through ethnification of victims, suggest that this is an imported issue, some newspapers, especially those on the left side of the political spectrum, acknowledge that this is a systemic issue (yet still fail to signal who or what is responsible for creating and maintaining that system). The Independent, in particular, suggests reasons such as fleeing 'war' and 'persecution', while The Daily Mirror and the Netherlandic-Dutch Het NRC similarly suggest that these may have been people from 'war-torn' countries ('ze probeerden te vluchten, wellicht van oorlog en armoede'). The Serbian newspaper Danas also says that they presumably 'wanted a "better life" for themselves in Great Britain' ('according to BBC') ('želeli da se domognu "boljeg života" u Velikoj Britaniji' ('navodi Bi-Bi-Si')). The wording in Serbian is of importance here, as, literally translated, the clause 'to grab a better life' is used, which is the standard expression for someone who wants to attain something desperately. Sobel (2014), in comparison, shows that when (in those relatively rare cases) a cause is mentioned in narratives of this kind, this is often speculated to be socioeconomic, with 'war' mentioned by the Thai and Indian papers examined (albeit much less often), and 'policy' by the US, Thai and Indian papers examined (also less often than socioeconomic reasons). This acknowledgement of these people's agency in attempting to improve their conditions is important, as it somewhat negates the prevalent portrayal of victims as solely acted-upon (Gregoriou andRas 2018a, 2018b; see also Choi 2014; Desyllas 2007 for explorations of the complexity of victims' agency and the lack of consideration thereof in anti-trafficking and -smuggling policy and legislation, with the caveat that these articles primarily focus on women who are sexually exploited). The Independent and The Daily Mirror also focus on the role of government anti-migration and border control regulations in driving these people to seek out incredibly dangerous routes to that better life, emphasising 'the urgent need for the government to create safe and legal routes to the UK for people fleeing war and persecution' (The Independent). These border control regulations are questioned in the Danas as well ('you can't turn Great Britain into a fortress', says one of the sources quoted). Some of the articles openly comment on the British 'open borders' politics (the Croatian Jutarnji). The use of this frame is a step in the direction of acknowledging the role of systemic global inequalities in driving trafficking/smuggling.
The Netherlandic-Dutch, Greek, Serbian and Croatian articles, through priming, focus on the unknown origin of the victims and the need for identification. Where the right and centre-right wing UK newspapers frame the event as one related to mass murder of 'immigrants', the independent and left-wing ones refrain from overfocusing on where it is that these individuals are from, unknown as their origins are at this stage anyway. Metro, The London Evening Standard, The Independent and The Guardian instead offer a more matter-of-fact frame through presenting the event as a vehicle-related 'tragedy' that involved people-exploitation and resulted in their murder.
In some tabloids (the UK The Daily Mail, the Serbian Alo), the events are sensationalised through an over-focus on the circumstances surrounding the victims' death. Though the relevant February 2020 post-mortems gave a combination of hypoxia and hyperthermia as the provisional causes of the 39 victims' death, and even though the media does not, at the early stage of the investigation we chose to focus on, know whether the refrigerated lorry's refrigeration had been switched on, some of the tabloids over-focus on this possibility. Since events as here described are ones that did not happen, but are nevertheless discussed as ones that might have happened, they can be described as 'disnarrated' (Lambrou 2019). Originally coined by Prince (1988: 2), 'disnarration' describes events which are narrated in the negative or hypothetical mode, so that their existence within the ontology of the storyworld, though potentially transiently affirmed, is ultimately uncertain or denied. Further to sensationalising the story, such counter-factual excerpts contribute to its tellability (i.e. what makes a story worth telling in the first place) and newsworthiness. We elaborate on other aspects of linguistic sensationalism and its links to dramatization next.

What happened is a tragedy!
Unsurprisingly, most European newspapers sensationalise the events, for instance by referring to the lorry as a frozen coffin, hearse, tomb, or by referring to an item deemed to be ironic/poignant, the driver's dream catcher bearing links to the victims' own dreams for a better future. The Croatian Slobodna Dalmacija provides a particularly imaginative account; it treats the event as a gruesome story to tell in as many gory details as possible. References to people 'dying in terrible conditions, the temperature plummeting to -25 °C' ('umirali u jezivim uvjetima, temperatura se spuštala do -25 °C') focus on the chilling, terrible, frightening and nauseating aspects of the story, regardless of the need to adhere to news' truthfulness value. What adds to the sensationalisation is the use of alliteration (see the /f/s -'φ' -of the Greek 'φορτηγό-φέρετρο', Ta Nea, as in 'coffin-lorry' and 'φορτηγόνεκροφόρα', Eleftheros Typos, as in 'lorry-hearse'). Notice references to a so-called horror lorry bearing a macabre/shocking and horrific sighting/scene in the Greek press, for instance (with the again /f/ Greek alliterative 'φορτηγό της φρίκης' and 'μακάβριο/σοκαριστικό και φρικτό θέαμα/φρικτή σκηνή', in To Vima and Efimerida Sintakton). These highlighted elements contribute to the story's newsworthiness and tellability (see Muždeka 2018).
However, indirectly, through highlighting these aspects of events, the metaphorical conceptualisation is one of a fatal dramatic narrative, whereby the event is transplanted out of reality and into the fictional realm. Much like with references to these individuals' ethnicity, the event dramatization distances readers from these (non-us) others' 'story'.
The nouns that encapsulate this event also suggest a lack of human control over something that merely 'occurred'. Besides, 'disasters' and 'tragedies' are events that can be described as accidental or natural, and hence ones that no human need be blamed for. Furthermore, as these are nouns that encapsulate these events, there is no human action or agency in these labels; i.e., descriptions of these events have been nominalised. Through these metaphorical, nominalised descriptors, the making of accusations, but also the ascription of responsibility, is avoided. As a 'disaster', this event is unexpected, distant and disturbing, rather than expected, given systemic factors and developments in governmental migration policies. These metaphors generating the given effects confirm the role that metaphors play in the framing of crime news and in the influencing of readers' beliefs about crime, as explained by Kahneman (2011; see also Hobbs 2021 for more examples of how framing affects people's moral judgments about violence and crime).
Furthermore, in places, this event is actually presented as the cause of further events, actively reminding of similar cases e.g. 'invoked comparisons' (the UK's The Financial Times), or generating emotions e.g. 'shocked and saddened' (the UK's The Sun), suggesting that the problem is not those who caused this event, but the event itself. This grammatical construction again shifts the focus away from the systemic factors that underlie trafficking, smuggling and migration, and hence the event's political, economic and ideological causes. Put differently, the nominalization of these events obscures the line of action, and places focus on the outcome and process rather than the agent/doer behind it, who is backgrounded.
The following section examines the use of intransitive clauses and agentless passives, which too contribute to the impression of there being a general lack of someone to blame.

Who is to blame?
Similar instances of ignoring or shifting blame are done through the very common use of the agentless passive voice. Throughout all articles, we find references to these people as being trafficked or smuggled into the UK, being put or placed into the lorry, and to the lorry having been registered, taken places and rented. Such agentless passives again allow a lack of focus on those responsible for the crimes, even if these individuals are yet unknown.
The same can be said for the texts' intransitive clauses which, despite mentioning that the driver was arrested on suspicion of murder or manslaughter, allow the suppletion of agentless passives, through references to the victims, as agents, having suffocated, perished, died or lost their lives. Besides, people can lose their lives because of any number of causes and indeed even by accident or without any other's fault, as in, for instance, the Dutch-language nominalisations 'doodsoorzaak' (cause of death) and 'het overlijden' (… van deze mensen) (the dying (…of these people)), which function similarly as the metaphorical 'disaster' described in the previous section. These individuals' deaths, however, were caused by others. This tendency, to refer implicitly to these (potential) perpetrators through agentless passives, is persistent across cultures: Gleason, Baker and Maynard (2018) found this same tendency in interviews on human trafficking with local stakeholders in Hawaii.

Does responding to trafficking require violence?
The official government response is, especially by British papers, framed in terms of fighting/war (with such The MailOnline references as that of 'Britain was perceived by organised crime as a potentially easy target for traffickers'), and the overall focus remains on police investigation (their discovering of bodies, screening the 328 containers, making arrests and so on). This suggests that the only thing that matters is enhanced policing/the action on the part of the authorities, tapping into the 'heroic saviour' aspect of the stereotypical narrative (on the detrimental nature of such policy, see Vandenberg 2007). As argued in Gregoriou and Ras (2018b), the war metaphor oversimplifies responding mechanisms, potentially legitimises police violent reaction to the problem, and focuses on a response to the symptoms of the trafficking problem, as opposed to its structural factors that in fact demand attention. Such metaphors tend to be absent from papers in other languages, except in translated, quoted material. Where violent metaphors are avoided, a matter-of-fact reporting style is effected, which is perhaps preferable to the highly emotive style of some British tabloids. Nevertheless, an unemotive style is not without its own problems; reporting these events as though committed by inanimate or unreported agents, and maintaining an uncritical focus on the actions of law enforcement and customs agencies, enables these writers to resist attempts to examine the underlying inequalities and structures that drive migration, trafficking and smuggling.

Government (Action) is a killer
Interestingly, we see a similar elision of specific responsibility through the presentation of government actions that have contributed to these deaths as inanimate actors of these deaths, e.g. 'government policy […] has deliberately closed down safe and legal routes into Britain' (The Independent, UK). Whilst it is laudable that some papers exceptionally employ this structure, and mention government policy as a serious contributory factor, the grammar of these sentences still lays the responsibility for the consequences of this policy at the feet of that policy, and not those responsible for creating and carrying out this policy. The Independent's personification of government ('government had "blood on its hands"') and the metaphorical conceptualisation of the UK government as a murderer or accessory to murder does emphasise the role of the UK government in this event, but, through the personification, makes it impossible to pinpoint any particular person who may be held responsible, thereby still diffusing that responsibility.

Lorry personification and focus
Personification is favoured throughout, important as it is to try and explain abstract and complex circumstances, entities and concepts, as if they were simple, relatable and human. All European newspapers examined describe the lorry undertaking various parts of the journey. Instead of the driver himself, it is the lorry cab and its trailer that are personified as beings that transport, travel, pass, stay, and come into and out of places, follow a certain route ('welke route de vrachtwagen aflegde', the Netherlandic-Dutch Metro), with the animated truck picking up the refrigerated container itself ('werd de container opgehaald door een vrachtwagen', the Netherlandic-Dutch De Volkskrant, Het Nederlands Dagblad), given the need for some objectivity and the lack of absolute knowledge regarding who it was that had control of the vehicle at any one point. Despite such constructions being perhaps somewhat (legally) necessary, these also present the inanimate as if these were animate, and suggest that the lorry moved unaided and on its own accord, without a human driving it. By doing so, these papers refrain from ascribing much responsibility for these deaths to any party, much like in Strega et al.'s (2014: 16) discussion of murdered sex workers, in which it is the personified 'street' that is designated as responsible for these women's murder. Again, the traffickers behind the operation are unknown but also effectively backgrounded altogether. When the lorry is not the agent of the sentence, it is the subject of an agentless passive: 'de oplegger is in Zeebrugge op een schip geladen' ('the trailer was loaded onto a ship in Zeebrugge', Belgian-Dutch De Tijd).

Basic demographics
All newspapers, regardless of language, first refer to the size of the victim group, then to the fact that these people are dead, sometimes in a sensational and graphic manner: 'corpses', as in the Greek 'πτώματα/σορούς', 'leševi' in the Croatian Jutarnji, Slobodna Dalmacija, Dnevnik, Telegram and Večernji, as well as the Serbian Glasnik, Novosti and Informer, 'lijken' in Dutch; bodies 'devoid of souls', as in the Greek 'άψυχα κορμιά'; 'cadaver', as in 'truplo' in the Croatian Slobodna Dalmacija; 'frozen bodies', as in 'smrznuta tela', in the Serbian Glasnik, Novosti and Informer. Newspapers also all mention the fact that at least one of these people is not an adult, which serves to enhance the sense of tragedy, raising the implication that the young are the least deserving of this fate.
Some newspapers also indicate the group's gender split ('31 men and eight women', The MailOnline). The Serbian Politika and Danas -contrary to the convention of naming men first, women second -indicate that the victims were 'eight women and 31 men', whilst the Netherlandic-Dutch De Volkskrant and Het Nederlands Dagblad suggest that a stereotypical case of trafficking concerns 'primarily young women who end up in prostitution', thereby perhaps highlighting the supposed 'oddness' of the case, or suggesting that this case is less serious than a 'real' (in the sense of 'ideal') trafficking case. Pajnik (2010) has found a similar distinction, between irregular migrants and people who have 'actually' been trafficked, made in Slovene reporting of such cases. In the case of these papers, the deaths of (migrant) women, who may be perceived as a more vulnerable gender than men, are perhaps more newsworthy; this may be linked to the responses to trafficking/smuggling survivors, whereby men are routinely criminalised and women are more readily accepted as victims (see Musto 2009). In this regard, it is also noteworthy that the Croatian Dnevnik and Novosti claim that all the victims were male.

Victimhood, nationality, migration status and guilt
Though it is widely agreed that these people are victims in a legal sense, having died of unnatural causes, they are not generally also granted the victim-status in a social sense.
Firstly, many references to these people in these papers have an effect of dehumanisation. Many newspapers refer to trafficking as a business or trade (the UK's The Independent makes reference to the 'modern equivalent of the slave trade'), with wording that contributes to the trafficked being viewed as currency or cargo/ animals/objects to manipulate/sell and traffickers/ing as a machine. An example of this is the UK's The Sun's implied comparison of these people with the lorry's 'usual load'. Equally dehumanising are the Greek press's references to transport rings that infest ('κυκλωμάτων διακινητών που λυμαίνονται', Eleftheros Typos) or nominalisations, such as that of the British press's 'illegal crossings' (The Guardian)/border 'infiltration' (The MailOnline), grammatical structures that also metaphorically reduce individuals to (unwanted) non-human masses.
Secondly, these people (insofar their agency is recognised), are blamed for their misfortune. Despite not actually knowing the nationality of these people, many papers do not hesitate to immediately label them as migrants ('migrants', 'migranti' in Serbian, 'migranten' in Netherlandic-Dutch). It must, however, be noted that in most Croatian and Belgian-Dutch articles, they are named 'migrants'/ 'immigrants' only by association, when mentioning other, previous cases of migrants/immigrants being found dead en route. Implicit to the label of 'migrants', in all languages, is the aspect of agency: these people chose to move abroad. This agency, which has a victim-blaming effect, is even more explicit in other references to these people: British tabloids The Daily Star and The MailOnline describe them as 'stowaways', as a 'group' who potentially 'sneaked inside the container', or, more indirectly, through a former Immigration Enforcement officer's description of the victims as ones that criminals 'provide services to'. Similarly, Greek papers refer to a comparable case of '58 chinese stowaways' ('58 κινέζοι λαθρεπιβάτες'). The Croatian Večernji names these people 'passengers' ('putnici'). The Belgian-Dutch De Morgen includes a description of these people as ones who willingly and knowingly do not want to request asylum and only intend to reach the United Kingdom as quickly as possible and then fall into the hands ['claws'] of people smugglers, and as climbers-in ('inklimmers').
This agency is not only emphasised through nouns in which the exercising of agency is part of the meaning, but also through active verbs relating to these people voluntarily undertaking this risky journey: the Croatian Slobodna Dalmacija claims that 'the police are trying to determine when the migrants entered the lorry' ('migranti ušli u kontejner'), with the Croatian Dnevnik stating that 'illegal migrants have been trying to reach Great Britain for years' ('ilegalni migranti već godinama pokušavaju dospjeti do Velike Britanije'). Serbian papers also note that it is 'yet unknown at which point the migrants entered the lorry' ('migranti ušli u hladnjaču'); the Netherlandic-Dutch Het NRC mentions migrants that climb into containers to come ashore here ('migranten die in containers klimmen om hier aan land te komen'), and the Serbian newspaper Danas making reference to migrants taking huge risks, paying 10 000 lb and more ('migranti snose ogroman rizik, plaćaju 10 000 funti ili više') to be transported into the UK. Through these references, these papers suggest that these people voluntarily undertook this journey, and were thus smuggled, thus are criminal. Much like in Strega et al.'s (2014: 14) previously mentioned discussion of murdered sex workers, the dominant discourse here too precludes the movement of those murdered into 'the subjective space reserved for innocent victims'.
On the other hand, however, as also mentioned in Sect. 3.1, it must also be acknowledged that some newspapers do offer sympathetic, and legitimating, hypotheses as to why these people might have willingly boarded the lorry. For instance, The Guardian suggests the possibility of these people having been 'refugees'. In Greek papers, there are references to the victims as 'most likely'/ 'presumed to have been' ('κατά πάσα πιθανότητα'/ 'εικάζεται ότι ήταν'), 'unfortunate' ('άτυχοι'), and 'irregular' ('παράτυπ[οι]') 'immigrants' ('μετανάστες'), with then Labour leader Corbyn describing them as 'desperate' ('απελπισμέν[οι]'). The Serbian Danas also references the number of 'refugees killed in transport since 2014' ('broj poginulih izbeglica u transportu od 2014. godine'). The word 'refugees', and the description of these people as 'desperate', indicates that, according to these journalists/newspapers at least, these people are fleeing desperate conditions and are attempting to migrate for reasons that are seen as acceptable to the readership.
In this regard, the ongoing discussion in most papers with respect to the nationalities of these people is also pertinent; beyond offering a basic demographic detail, such discussion also does ideological work. For instance, rather than omitting this detail, papers immediately note that 'nationalities are unknown'; 'άγνωστη η προέλευση τους και η εθνικότητα τους'/ 'unknown nationality and origins' in Greek; 'nije poznata nacionalnost'/ 'nationality unknown', in Serbian; 'hun nationaliteit is nog onduidelijk'/ 'their nationality is still unclear' in Netherlandic-Dutch and Belgian-Dutch. Writers also speculate on this nationality through a process of elimination, noting that these people 'are unlikely to be Bulgarian', which might be expected given that the lorry was registered in Bulgaria ('malo je verovatno da su Bugari', in Serbian; 'het is zeer onwaarschijnlijk dat de slachtoffers Bulgaren zijn', in Netherlandic-Dutch), although, strangely, the Netherlandic-Dutch De Telegraaf does suggest that they were ('het gaat waarschijnlijk om Bulgaarse ingezetenen'/ 'this likely concerns Bulgarian residents'). The victims are, however, erroneously described as, or presumed to be, Chinese ('truck migrants were Chinese'; 'Κινέζοι'/ 'κινέζοι', as in 'Chinese'/ 'chinese'; '39 kineskih migranata'/ '39 Chinese migrants', 'all Chinese nationals'/ 'kineske nacionalnosti', the Serbian Politika and Danas; 'Chinese migranten' in Dutch). The fact that these people, who turned out to be Vietnamese, were initially identified as Chinese, also raises questions about potentially highly problematic, indeed racist, assumptions made by either the British police, or by journalists. In any case, this emphasis on establishing a nationality or country of origin (regardless of accuracy) suggests some level of pre-occupation on the part of these papers, and possibly ties into the discussion as to whether these people had 'acceptable' (as refugees) or 'unacceptable'(as irregular migrants) reasons for wanting to relocate.
Finally, the question of these people being human trafficking victims is raised. The Greek communist paper Rizospastis refers to others' presumption of their being slave trader victims ('θύματα δουλεμπόρων'). The Serbian Danas compares it to 'the dealings of a Bulgarian-Hungarian human trafficking group' ('akcija bugarskomađarske grupe za trgovinu ljudima'). The Netherlandic-Dutch De Volkskrant and Het Nederlands Dagblad both mention that this possibility remains a question; these are 'illegal migrants that paid to try and reach Great Britain or […] victims of human trafficking' ('illegale migranten die tegen betaling probeerden Groot-Britannië te bereiken of […] slachtoffers van mensenhandel'); the Netherlandic-Dutch De Telegraaf also suggests that these people may have been victims of people smuggling ('slachtoffers van mensensmokkel'). Having said that, speculation that this might have been a case of human trafficking is largely absent from most articles.
In short, whilst there is some acknowledgement that these people may be 'legitimate' victims, the general tendency of these newspapers, across languages, is to suggest that these people had a hand in their own misfortune, thereby mostly ignoring those systemic factors that drive migration and the risk of migration-related victimisation. It is precisely this 'selective description and omission of the features of a situation' (Entman 1993: 54) that leads to the framing in which character, causes, and consequences of an event become changed (also see Edelman 1993 and Kahneman and Tversky 1984 for more insight into how frames call attention to particular aspects of the reality described).

The driver
All newspapers, across all examined countries, describe the driver in terms of his gender, his profession and role in this event, his age, and his origins. Of these details on the driver, descriptions of his name, role in the event and personal circumstances are of particular note.
Of some interest is the way in which this driver has been explicitly named. Most papers simply name the driver as 'Mo Robinson' at first mention, and 'Robinson' at subsequent mentions insofar the naming of suspects is permitted and accepted given national journalistic practices and guidelines.
His role in these events, and the description of these events, is remarkably consistent across papers and countries. Most UK, Dutch and Greek papers indicate that he has been arrested on suspicion of murder ('verdenking van moord', 'φόνο'/ 'δολοφονία'). That said, some Greek papers instead opt for manslaughter ('ανθρωποκτονία') in places. One interpretative difficulty here is that Serbian and Croatian reporting generally does not differentiate between 'murder' and 'manslaughter', but simply refers to the fact that the driver has been arrested on charge of 'ubistvo'/ 'ubojstvo', which in this short form (i.e. without further qualification) can refer both to murder ('ubistvo prvog stepena'/ 'ubojstvo prvog stepena') and manslaughter ('ubistvo bez predumišljaja'/ 'ubojstvo bez predumišljaja'). However, the most likely implication and the most likely interpretation on the part of the readers is 'murder', not 'manslaughter'. The interpretation here is then that regardless of the role of the victims in these events, their deaths were criminal and not accidental. However, the focus is still on the acts of one man that lead to these deaths, and not on systemic factors.
The more sensationalist papers, including The MailOnline, the i, The Daily Mirror, and Slobodna Dalmacija, go into some detail on Robinson, including on the cost of his house, his family situation, and his social media communications. The MailOnline¸ The Daily Mirror, but also De Morgen, also include quotes by Robinson's friends and family, all of which suggest that he could not possibly be guilty. One could argue that such sensational human interest-details normalise the driver, and thus increase reader-sympathy for this person. This may also be achieved by the right-wing The MailOnline's, The Daily Telegraph's, and The Times's use of a respectful and distancing 'Mr Robinson'. Relatedly The Daily Telegraph refers to Robinson as having been 'arrested in connection with the death of the migrants', and The Times notes that Robinson has been arrested, without mention of the charges.

Gangs
Newspapers in all languages suggest some involvement of 'gangs' or 'organised crime', both directly and through noting the convictions of gang members in comparable cases.
Few details about these (hypothetical and/or comparable) gangs are known; their nationality remains a central discussion point. Where the tabloid The MailOnline describes the perpetrators as 'Northern Ireland-based' or 'Irish', the Daily Star opts for 'migrant', and The Daily Mirror gives us a list including 'Afghan', 'Bulgarian', 'Iraqi', 'Georgian', and 'Vietnamese'. The Netherlandic-Dutch De Volkskrant and Het Nederlands Dagblad further speculate that these networks are often from Eastern Europe ('Oost Europa'). It is noteworthy that Pajnik (2010) also finds a tendency for Slovene reporting to suggest the perpetrators being Eastern European.
Furthermore, all papers question whether these are 'smuggling' or 'trafficking' gangs. This suggests a conflation between trafficking and smuggling, at least in places (see, for instance, one Daily Star reference to the victims 'feared to [have been] trafficked into the UK by people smugglers') which, as shown in Gregoriou and Ras (2018b), is common in British newspapers, and contributes to the criminalisation of human trafficking victims. Similarly, the Croatian Slobodna Dalmacija refers to 'smuggling', and Nacional, Telegram, Jutarnji and 24 sata mention 'human trafficking' and 'traffickers'. The Serbian Blic does not differentiate between 'human traffickers' ('trgovci ljudima') and 'human/people smugglers' ('krijumčari ljudima'), again suggesting some conflation. Greek papers instead do distinguish between the two crimes: 'What the police is firstly looking into is whether the dead were victims of organised crime connected to either human trafficking or people smuggling', as in 'Αυτό που εξετάζουν πρωτίστως οι αρχές είναι το αν οι νεκροί είχαν πέσει θύμα ενός οργανωμένου εγκλήματος το οποίο σχετίζεται είτε με το δουλεμπόριο είτε με το λαθρεμπόριο' (Simerini). The Netherlandic-Dutch Het Nederlands Dagblad offers an in-depth exploration of the differences between these crimes. Belgian-Dutch newspapers only refer to this crime as one of 'smokkelen' (smuggling); trafficking (which would be 'mensenhandel') is not mentioned.

Conclusions
Figurative language (metaphors), grammar (nominalisations, transitivity) and phonetics (alliteration) help portray what happened in dramatic, fictionalising and sensationalistic terms. Over-focusing on the event's effects backgrounds doers and helps writers avoid the ascription of responsibility. Where business language somewhat normalises the 'trading' of people, metaphors of infiltration and infestation abnormalise these transactions, with metaphors of fighting, war and violence used to oversimplify responding mechanisms and divert attention away from dealing with trafficking's underlying causes. Where reporting is more matter-offact and unemotive, it remains mostly uncritical of the actions of law enforcement and customs agencies, and -along with agentless passives, intransitive clauses and personification -enables writers to steer away from examining what actually drives trafficking and smuggling. To heighten the event's newsworthiness (Fowler 1991), all newspapers, regardless of language, mention the number and rough ages of victims and focus on the fact that they have been found dead. Important is the question of what these people are the victim of, which in this case translates to a concern as to the extent to which they can be held responsible for their own misfortune (as potentially being clients of smugglers). That said, some (other) perpetrators are also mentioned; all papers generally focus on the person of the driver, although confusion remains as to whether he has been arrested for murder or for manslaughter. Other (hypothetical) perpetrators are big, international networks, who either specialise in trafficking or smuggling; there seems to be some lack of clarity as to which. As mentioned, suggesting the involvement of such perpetrators is comparable to Christie's (1986) 'narkohai'. Through these linguistic choices, the newspapers (perhaps unintentionally) help to justify draconic law enforcement and border security policies (see also Pajnik 2010). Though we do notice attempts to distribute some responsibility to governments and policies, as well as structural factors such as war and poverty, especially in left-wing papers, responsibility for these factors tends to be diffused, and hence unallocated, through nominalisations and personification. In short, the majority of these papers fail to reflect on the role of systemic factors. These findings are not unique to European papers, or even this event; Gulati (2011) has found similar results in reporting by The Washington Post and The New York Times across multiple decades. As Stephens (2019: 87-8, 89) rightly notes, '[t]he mainstream's response -calling for harsher borders, criminal justice for 'greedy and unscrupulous' traffickers and safe passage for 'genuine' refugees -fails to interrogate the global conditions that lead people to risk dangerous travel, and the deadly effects of border controls on all migrants'; indeed, perhaps it is 'borders [that] kill'. Even more so, 'the greater moral and physical risk to migrants may come from attempts to stop them from leaving the country at all' (O'Neill 2001: 162). As Le (2019) notes in her blog, Vietnamese responses sometimes similarly dehumanised these people, with arguments suggesting that 'poverty shouldn't be an excuse to break the law', and also called these 'traffickers' 'evil'. A further exploration of Vietnamese, as well as Chinese reporting of this incident over the same 48-h period would be of interest in comparison, as would undertaking a bigger study of this case's reporting over a longer period of time, allowing an analyst to trace changes in language use over time, particularly as more information as to the case's specifics came to rise.

Conflicts of interest
The authors have no conflicts of interest to declare that are relevant to the content of this article. The research doesn't involve human participants and/or animals and there are no issues of consent to account for.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.