the iconic image on social media: a rapid research response to the death of aylan kurdi* Visual Social Media Lab 1 * We are aware that his name is Alan and not Aylan. We decided however to consistently use the name Aylan, rather than Alan, as that was the name the image was globally circulated and identified with. Published by the Visual Social Media Lab December 2015. Suggested citation: V is, F., & Goriunova, O. (Eds.). (2015). The Iconic Image on Social Media: A Rapid Research Response to the Death of Aylan Kurdi* This report is part of the ‘Picturing the Social: Transforming our Understanding of Images in Social Media and Big Data Research’ project, funded by the Economic and Social Research Council (ESRC). Grant reference: ES/M000648/1 Open Access. Some rights reserved. As the publishers of this work, the Visual Social Media Lab wants to encourage the circulation of this report as widely as possible while retaining the copyright. We therefore have an open access policy, which enables anyone to access our content online without charge. Anyone can download, save, perform or distribute this work in any format, including translation, without written permission. This is subject to the terms of the Creative Commons licence we have chosen for this work: Attribution-NonCommercial-NoDerivs (CC BY-NC-ND). This means: Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use. NonCommercial — You may not use the material for commercial purposes. NoDerivatives — If you remix, transform, or build upon the material, you may not distribute the modified material. ​For more information about Creative Commons licences, please see: https://creativecommons.org/licenses 2 Table of Contents 5 10 11 19 27 31 Introduction Olga Goriunova Part I - SOCIAL MEDIA RESPONDS 1. Journey of an Image: From a Beach in Bodrum to Twenty Million Screens Across the World Francesco D’Orazio 2. What Can Search Data Tell Us About How the Story of Aylan Kurdi Spread Around the World? Simon Rogers 3. Examining the Hundred Most Shared Images of Aylan Kurdi on Twitter Farida Vis 4. Undermining Aylan: Less Than Sympathetic International Responses Mike Thelwall 37 38 40 42 44 3 Part II - WHAT DID THE IMAGE DO? 5. Discussion and Action: Political and Personal Responses to the Aylan Kurdi Images Anne Burns 6. The Strength of Weak Commitment: A Norwegian Response to the Aylan Kurdi Images Lin Prøitz 7. Politics, Publics, and Aylan Kurdi Lucy Mayblin 8. #KiyiyaVuranInsanlik: Unpacking Artistic Responses to the Aylan Kurdi Images Holly Ryan 46 Part III – THE ICONIC IMAGE ON SOCIAL MEDIA 47 50 53 9. On the Iconology of Aylan Kurdi, Alone Ray Drainville 10. The Life of Images: The Iconography of the Photograph of Aylan Kurdi’s body and the Turkish Policeman Jim Aulich 11. Aylan Kurdi and the Movability of Images Simon Faulkner 56 Part IV – SHOWING / NOT SHOWING THE IMAGE 57 68 12. S hoes of Childhood: Exploring the Emotional Politics Through Which Images Become Narrated on Social Media Lisa Procter and Dylan Yamada-Rice 13. When Should We Share Distressing Images? Seeing Aylan Kurdi Sam Gregory 14. Journalism and Death on the Social Web: The Case of Aylan Kurdi Claire Wardle 15. Body Counts: Counting Aylan Kurdi Evelyn Ruppert and Funda Ustek 71 Conclusion 61 64 4 Olga Goriunova and Farida Vis Introduction O lga G oriunova On the 2nd of September 2015 the body of a three-year-old boy was found washed up on the beach at Bodrum, Turkey. The family of Aylan Kurdi, as he was identified, was trying to cross the Mediterranean to the Greek island of Kos on an inflatable boat that capsized. Aylan Kurdi, a Syrian of Kurdish origin, his brother, mother and other passengers drowned, becoming the latest additions to the already substantial list of casualties in the refugee crisis. Turkish journalist Nilüfer Demir took a series of photographs of the bodies, of which two became photographic sensations rapidly spreading around the globe. These photographs became iconic images that represented the atrocity of the Syrian refugee crisis in a way that instantly and substantially affected individuals and societies. The unprecedented degree of mobilization of people responding to the crisis was a direct effect of the images, and as their circulation gained momentum, it seemed that a change in political response was possible and forthcoming. This report is an attempt to understand the tragic photographs of Aylan Kurdi and associated imagery, specifically focusing on how they were discussed within a European context, mainly searching for information and data in English. The photographs became what we call iconic images; being featured by mainstream media, their rapid spread on social media coupled their iconicity with the new virality of the digital age. The power accrued by the images testified to this new regime of visuality, meaning-making through images coupled with the power of social media and their new role in publishing and associated changes in political agency. The decision to produce this report is based on a belief that change for the better is possible. Understanding how a unified global response that amassed substantial power was arrived at, we hope, can help us become more humane – by realizing the conditions for an atrocity to become visible and widely known and to acquire a political voice that leads to political change. We don’t discuss the photographs of a dead child from the point of view of their effect as a mere ‘example’, but start from an understanding that the event of Aylan Kurdi’s and others’ deaths can never be matched by description or analysis. The way in which we analyse these images is aimed at understanding the new conditions, visual, technical, social and political, in which the events of life and death become manifest, and are in turn recruited to perform, today. Analysing and tackling these conditions, which are partly informed and shaped by the technical infrastructures of social media, their functionality, practices and dynamics, may have an effect on future matters of life and death. And even if our findings provide an unsatisfactory diagnosis for, or of, us as part of humanity, all of us still need to further our understanding as agents in the visual-political and socio-affective regimes of today. The report asks a number of questions : • Why did these particular images accrue such power and manage to draw public attention to the refugee crisis? Many young children died trying to cross the Mediterranean with none of these events being able to attract wide public attention or solicit adequate response. • Where did the power of these images come from? • Which images were most forceful? What other images were drawn upon? • How were they presented and how did they spread? • How did technical networks of social media help the images amass their power? • What does their creation and spread tell us about national and global societies and political order? What were the responses? • Were there correlations between what people said and how they acted given the overall consensus? • How, and by what, were the responses shaped? Was there dissensus? • Did the images retain their power to enact responses and change? • What does this mean for our societies, for our political organisation, for our humaneness? 5 O verview of the report and the data we worked with The report has four sections. The first of these, Social Media Responds studies in detail the unfolding of the event on social networks and media both nationally and internationally including the spread of images on social media. The second section, What Did the Image Do? addresses personal, political and artistic responses to the images. The third, The Iconic Image on Social Media offers iconographic explorations of the photographs, looking at the iconography of suffering, war and press photography and at counter-narratives amongst visual cultural traditions. The last section, Showing/Not Showing the Image discusses the exposition of graphic images as ethical and publishing decisions as well as a technical capability, in relation to the practices of visibility and accountability. Some of the authors worked with data collected through Pulsar (part of ESRC funded Picturing the Social project). Historical Twitter data covering the period 1-14 September 2015 was used based on the search terms enclosed as a reference below.[1] It covered social media, news, forums, and blogs (amounting to 2,843,274 posts). The vast majority of the data gathered, however, was from Twitter. One author also worked with Google data, offering an additional perspective using a specific dataset. Most authors mainly engage with images shared through Twitter and Facebook, often offering a cross-media analysis. As a team, we decided that images would be included in the report only as hyperlinks; we also decided to consistently use the name Aylan, rather than Alan, as that was the name the image was globally circulated and identified with. We are aware however that the child’s actual name is Alan and not Aylan. Part I - S ocial M edia R esponds Francesco D’Orazio starts the report by covering the shift in vocabulary on social media – from migrant to refugee – that accompanied the spread of the image. The change of terms changes the debate. He then maps the spread of the image onto a timeline and adds further precision by tracking the authors of the first tweets of the image and their locations. Francesco offers an almost hourly tracking: from Turkey, through the countries of the Middle East, to the Emergency Director at Human Rights Watch, Peter Bouckaert, where the story goes viral. The virality itself seems to have degrees of intensity, expanding to become more and more global. Simon Rogers offers a parallel investigation of the same dynamic by looking at Google search data. Google search history confirms the shift in the terms of discussion (from migrant to the refugee) happening in September. Rogers looks at the history of searches by country to track the interest to and the change in international migration and movement. As the tragic news and the image spread, what people wanted to know and how they searched for it changed too, both reflecting and constructing nationally specific forms of knowledge and response. Farida Vis continues this section with a close analysis of the one hundred most shared images on Twitter, as collected in the Pulsar dataset. The paper explores what kinds of images of Aylan were shared. Specifically if there is a difference in showing Aylan dead or alive, speculating why this is done. Second, it explores national differences, focusing especially on image sharing in Europe and the UK. Reflecting on the typology, geography, the emotional and political charge of images and on a variety of responses exemplified in these exchanges, Farida, in parallel with previous texts, traces the change in classification and naming: the child acquires a name, moving from the category of ‘dead children’ to being identified and remembered. While diffusion of the story remained consistently ‘image-led,’ the responses to the spread of images of Aylan Kurdi and related imagery were not uniform. Mike Thelwall explores hostile and ‘less than sympathetic’ responses to the story of the boy, his family and the region they came from. Mike’s is an analysis through data-mapping focusing on the popularity of hashtags and keywords by language (Turkish, German, French and English) to identify sympathetic responses as opposed to the discussions of various allegations aimed at undermining the power of the image by shifting blame onto the victims themselves. 6 The authors agree that the image produced a ‘seismic shock’ in terms of change of opinion (from being inimical to ‘migrants’ to empathic to ‘refugees’) and that following the images means understanding this change in media, public opinion and civil society. Part II - What D id the I mage D o? The second section focuses in particular on reactions and responses to the images, with Anne Burns starting this section by investigating how public opinion was continuously shaped by discussions that arose in response to the image. Anne argues that acting as a ‘symbol’ of a larger human tragedy, this image effectively acted metonymically (i.e. representing the deaths of many by the death of one.) She suggests that it is through the symbolic power of including the death of many into the death of the child while performing a public outcry for action, that the circulation of these images justified itself. Depending on the geography of responses, and correspondingly, on national governments and locally specific political situations, seeing and being moved by the image didn’t always actualize into action or change. The affective and global here becomes specific and particular molded by on-the-ground conditions and different political histories that, following the initial outcries, start playing larger roles. While many authors condemn the UK’s lack of political response, Lin Prøitz’s exploration of the ad-hoc grass-roots campaign #RWTN (refugeeswelcometonorway) that emerged in Norway seems to offer an exception to the other texts in that it shows people being able to carry out significant political change. Its rapid growth seems to have had an effect on Norwegian local elections which is much more positive than the Cameron government’s response to the refugee crisis. Lucy Mayblin’s text reports that while the response in the UK was immediate and forceful, and the Refugees Welcome UK campaign was gathering momentum rapidly, the overall picture is very far from positive. Her findings suggest that discussions quickly became displaced, focusing on ‘extraterritorial contexts’ or indicating the ‘class politics of invisibility and deservingness’ in operation. While the response of civil society in the UK has been overwhelming, and that the Prime Minister, David Cameron, was arguably forced to respond to the crisis, albeit rather superficially, Mayblin’s disquiet is maintained by the fact that the power of the image and the tragedy of the child and the family did not manage to do enough to change the political situation or legislation. She finishes her paper by drawing our attention to a continuous toughening of asylum legislation and the newly proposed Immigration Bill that suggests deporting parents while putting children into care in the UK. This is a shared finding and concern: nearly all authors draw attention to the fact that though the initial reaction was strikingly forceful, it was often brief, with few long-lasting consequences – such as welcoming more refugees or developing policy adequate to the crisis – to be seen. Working her way through widely reposted artistic responses to the photograph of dead Aylan on the shore, Holly Ryan completes the second section by pointing towards a therapeutic function of creative response. Whereas an ability to rework traumatic experience creatively increases resilience, it also becomes a foundation for dealing with affective reaction, one ‘beyond words,’ both representing and preparing affect for being remembered and mobilizing affect for politics. Among the paradoxes of affect are non-directionality of response. In line with others, Ryan is left wondering whether the political mobilizations stimulated by the event and by the photographs are not misled, whether they would bring relief or even more suffering. Part III - The I conic I mage on S ocial M edia In the third section all three authors recognize the iconic resemblance of Aylan’s photograph to images of sacred death in the Christian visual tradition, with the Pieta motif identified as a key visual context of the image of the boy’s body carried by the policeman. Ray Drainville analyses the photograph, building a striking lineage of aestheticized death in the European visual tradition. As we as viewers are ‘primed,’ trained to perceive aestheticized corpses through the 7 Christian tradition, Aylan’s body is reworked into that of a sleeping angel. In a way, this tragedy becomes visible because its portrayal fits into our visual training. Jim Aulich notices the gesture of care and protection (a child in the hands of an adult) as that of the Pieta. We understand the world with the help of images, Aulich argues. The image aids and shapes cognition. If we weren’t trained to see it in this way, we would not be able to perceive the event. ‘The image constitutes the world’. Recognising this, in photojournalism, war photography and devotional photography, photographs are often not produced as reliable ‘visual depictions,’ but as images impregnated with interpretation, with references and frameworks assisting in assigning them meaning. The iconography of Pieta includes higher authority. Here, both depictions of Aylan resting in heaven and outrage at political leaders (all images circulated on Twitter as responses to the original photographs) are inscribed into the visual tradition of the image itself. Simon Faulkner’s article picks up on the analysis of imagery generated as responses to Aylan’s photographs. He focuses on the power the image acquired as it moved. Moving between mediums and media, the picture was also dissected and montaged to enable the movement of the image of the body on its own. Here, the body is lifted out of the image, turned into a motif and becomes nomadic. When the body became a motif, an element in a montage, its force was grounded in a ‘metaphoric versatility,’ whereupon, as Faulkner attests, guilt is produced by ‘visual association.’ Montage is used here to demonstrate or incur culpability and the main element of the original photograph possesses ‘transposable rhetorical power.’ Part IV - S howing /N ot S howing the I mage In the last section, Procter and Yamada-Rice engage with one specific element through which the affective engagement with the photograph took place: the boy’s shoes. These shoes are both depersonalised and simplified so that they could belong to any child. This agenthood of the shoes rests in the power of such an object as a symbol of childhood. Through the shoes, the innocence of childhood unfolds; it also created a figure of the carer who has been putting the shoes on, as well as other adults ultimately responsible for his death by not exercising care. Shoes indicate vulnerability and exemplify care both exercised and withdrawn (in the image of the seemingly sleeping boy still having his shoes on). Shoes here witness and demonstrate the atrocity and function as an agent to enact the emotional response. Sam Gregory’s piece focuses on Facebook’s auto-play function and platform control. Software and interface elements here come to play important roles in how the image spread and what force it acquired, as they facilitate and address peoples’ decisions. Such technical agency poses new sets of social, political and ethical questions. Building on this, Claire Wardle’s fascinating account of close engagement between the technological development of platforms and users’ practices in relation to the publishing of graphic images is full of material for thought. Wardle, who for a while worked as a Senior Social Media Officer at UNHCR, (the UN Agency for Refugees) reports on the history of circulation of graphic images on social media, especially in relation to her position as an advocate for refugees who for many years tried to engage people through various means, photography not the least of them. Wardle suggests that reposting Aylan’s image across wide audiences became an individual act of publishing, a decision on what should be seen. Aylan’s image is not graphic, and perhaps, as other contributions indicate, that is the root of its ‘success’. Twitter Moments, a new curated content feature on Twitter, which launched in September 2015 were quick to place warnings on graphic images (for instance of the Ankara bombings), but they would not have picked the image of Aylan. Such technical decisions (by a content review team) and by algorithms may now possibly become full actors in deciding what is to be seen and what not, together with individual users who become publishers. The debate about graphic images is not over, but it proposes questions of how to walk a fine-line between what people want to see, deciding to share (a publishing decision) and acting alongside newspapers in informing others whilst striving to become more humane. In the last paper in this report, Evelyn Ruppert and Funda Ustek talk about technologies and practices of counting and specifically of counting refugees. Refugees are the hardest population to capture: hard to track, to count. Highly dynamic and fluid, the category escapes usual systems of assigning numbers. The politics of numbering relates here 8 to humanitarian statistics and to the question of who counts as population. The habitual visualisation of refugees as swarms, movements and lines is, in the case of Aylan, interrupted by one singular image. Counting creates knowledge about lives and livelihoods, and it is in the counting of Aylan’s dead body that we learn more about the lives of others. References 1. S earch terms: # kiyiyavuraninsanlik · kiyiya vuran insanlik · #humanitywashedashore · humanity washed ashore · #refugeeswelcome · #refugeecrisis · #refugeescrisis · #aylankurdi · #alankurdi · #aylan · aylan · alan · kurdi · #nomoredrownings · no more drownings · #dyingtogethere · dying to get here · #syria boy · syria boy · #syrian boy · syrian boy · #syria child · syria child · #syrian child · syrian child · #syrianchild · #syrianboy · bodrum beach · drowned boy · drowned child · drowned toddler · refugees welcome · refugee crisis · refugees crisis Statement on ethics As this report was interested in uncovering broad trends and not in highlighting individual users, for Twitter we only identify Twitter users by name and handle where we felt it was clear that they are public figures tweeting in a public capacity. Names of private users have been blurred in visualisations. Google data was available only to one author, who worked with it in aggregate. 9 PART I SOCIAL MEDIA RESPONDS The opening section tracks the spread of the images of Aylan Kurdi starting with a question: ‘How did [these] images travel from a beach in Bodrum to the screens of almost 20 million people across the world in the space of 12 hours and thirty thousand tweets?’ Changing the terms of the debate from migrants to the refugees, the images were widely discussed online before the mainstream media picked up the story. Francesco D’Orazio, Simon Rogers, Farida Vis and Mike Thelwall analyse how the event unfolded through the images. Moving from an hourly tracking of tweets (D’Orazio) and Google search data (Rogers) to the data-mapping focusing on the popularity of hashtags and keywords by language (Thellwal), this section sheds light on nationally specific and global responses. Furthermore, Farida Vis offers a discussion of the top hundred most shared images, coming up with many surprising findings. The variety of imagery developed and circulated around the original photographs testifies to a wide range of emotional responses and attempts to act, as they become intertwined with cultural contexts, media coverage and political campaigns. 10 1. J ourney of an Image: From a Beach in B odrum Across the World Francesco D’Orazio to Twenty Million S creens I first saw the picture of  three-year old Syrian refugee Aylan Kurdi  lying lifeless on a beach in Bodrum, Turkey on September 2nd on Twitter. It felt like getting punched in the stomach. But while I was shocked by it, I didn’t realise how impactful it was going to be until the next day, when the same image started to appear in my Facebook newsfeed. While on Twitter I read the news headlines carrying the picture and reporting the tragedy on Facebook it was easier to stumble upon the social context of the image. This included the comments of people who were actively engaging with the image and were debating immigration. And that’s when I noticed that the comments seemed to have a different tone from the conversation we had been seeing in the press and social media until then. A lot more people now seemed to be talking about ‘refugees’ rather than ‘migrants’. The shift, if that was the case, could have been significant: a ‘migrant’ is someone who’s got a choice (and according to some, often an ‘economical’ reason to move) while a ‘refugee’ is someone who has no choice but to flee their country to survive. The term that the politicians, the media and the people would end up adopting to talk about the issue would inevitably have massive implications in terms of humanitarian aid and policy making. Was public opinion on Twitter changing as a result of exposure to the pictures of Aylan Kurdi? The answer is in the chart below. While for most of 2015 ‘migrants’ and ‘refugees’ are head to head in public opinion, accounting for pretty much the same volume of conversation over nine months (5.2M vs 5.3M), from September 2nd onward public opinion radically flips towards ‘refugees’ (2.9M vs 6.5M) and at the moment of writing, two months after the pictures were published, the new ratio remains unchanged. 11 But how did this happen? How did the image(s) of Aylan Kurdi travel from a beach in Bodrum to the screens of almost 20 million people across the world in the space of 12 hours and via thirty thousand tweets? What was the journey of that image? Let’s go back to the very beginning of that journey, the morning of September 2nd 2015. The Turkish press covers the story At 8.42AM, Turkish news agency DHA (Dogan Haber Ajansi) are the first to report the death of 12 Syrian refugees drowned while trying to reach the Greek island of Kos on a dinghy boat. The article, in Turkish, featured a gallery of 50 pictures taken earlier that morning on the beach of Bodrum where the bodies where first discovered. Amongst those 50 pictures, four featured Aylan Kurdi. The DHA article leads with one of the pictures of Aylan. Twitter is silent. At 9.10AM, another Turkish news agency, Diken, cover the story with another article in Turkish, leading again with a picture of Aylan Kurdi. Twitter is still silent. The first tweet carrying the picture At 10.23AM, less than two hours after the story has been reported, the first post carrying one of the pictures of Aylan Kurdi appears on Twitter. Michelle Demishevich, the author, is a Turkish journalist and activist. The post doesn’t link to any news source, it only features a caption, the picture and five hashtags including #Refugeeswelcome and #Syrianrefugees. Within an hour the tweet generates 33 retweets and a handful of other tweets carrying the picture start to crop up from Greece and Spain, but the audience is mostly Turkish (in red in the graph below). Diffusion graph A: 1 hour in from the appearance of the first picture on Twitter. Size of the nodes indicates impact on the audience (Visibility Score). Colour of nodes and edges indicates country of the user. Source: Pulsar. 12 At this point the Turkish press is starting to cover the story more heavily and between 11.00 and 11.30AM more than 15 publications start posting articles in Turkish on their websites. But whilst the press is spreading the news to a mostly Turkish audience, the images have now embarked on a very different journey on Twitter. The images spread through the M iddle E ast The Turkish press has now taken to Twitter with agency Diken and Haber Analyz expanding the local audience. Images are now starting to spread through the Middle East carried by tweets arriving from Lebanon, Gaza and Syria. Journalists, activists like Free Syria Media Hub (@Free_Media_Hub) and politicians such as the ex Minister of Health in the Hamas government of Gaza, Basim Naim (@basemn63), are getting involved on Twitter. Lebanon is particularly key in spreading the images in the region with Newsweek Middle East correspondent in Beirut, Martin Jay, suggesting for the first time that the picture might be going viral. Less than an hour later, Emergency Director at Human Rights Watch, Peter Bouckaert (@bouckap) from Geneva posts three pictures of Aylan Kurdi and comments on the need for an urgent plan to deal with the crisis. His tweet receives 664 retweets from a variety of countries including the US, the UK, Australia and Malaysia. The pictures have now left the Middle East. The story has gone viral. Diffusion graph B: 2 hour in from the appearance of the first picture on Twitter. Size of the nodes indicates impact on the audience (Visibility Score). Colour of nodes and edges indicates country of the user. Source: Pulsar 13 G oing Viral (& G lobal) If a handful of tweets with a few hundred retweets were enough to claim virality, what happens next will require a redrawing of the scale of the event completely. So far the pictures have been shared on Twitter less than 500 times in two hours (3% of the total diffusion of the images at the end of day 1) and have reached an audience of half a million Twitter users across 100 countries. That’s when, at 12.49PM, Washington Post Beirut Bureau Chief, Liz Sly (@Lizsly) weights in sharing a tweet that will end up being retweeted 7,421 times, becoming the most viral post in the dataset. Liz Sly changes the scale of the diffusion generating in the first 30 minutes of her post the same amount of tweets that had been generated in the previous two hours. And it’s not simply the scale that’s changing: the composition of the audience is becoming truly global. New posts are now coming from Lebanon, Iraq, Palestine, Syria, US, Switzerland, Spain and are getting retweeted by an audience that is now mainly out of the Middle East and from the US (17%), Spain (10%), UK (9%), India (4%), Netherlands (3.4%) and Greece (3.4%). Diffusion graph C: 3 hour in from the appearance of the first picture on Twitter. Size of the nodes indicates impact on the audience (Visibility Score). Colour of nodes and edges indicates country of the user. Source: Pulsar 14 At this point in time, almost five hours after the photos have been published online and three hours after they appeared on Twitter first, the Turkish press is the only one officially covering the story, no international news organisations are covering it. And in the void left by the media Twitter has managed to amplify the story to a global audience. But it hasn’t simply seeded the story globally. Twitter acted as a decentralized catalyst that has delivered the story to a highly relevant group of people (journalists, activists, politicians, aid workers interested in the region) along the edges of its interest graph. It is this dynamic that’s behind the extremely high engagement rate in the dataset (17 retweets for every single tweet posted). This doesn’t mean that journalists are not getting involved in this process. In fact, they are leading and engineering the viral spread of the images, but they are posting their content on Twitter exclusively and independently from their publications. For the first international publication to release an article on Aylan Kurdi we have to wait until 1.10PM when the Daily Mail publishes their first story on the matter: ‘Terrible Fate of a tiny boy who symbolizes the desperation of thousands’. G oing mainstream The Daily Mail story kicks off an entirely new phase of the diffusion characterized by the heavy involvement of the international press. Between the Daily Mail article and the end of the day more than 500 articles are published online and shared on Twitter of which 35% from the UK and the US. The Independent, Huffington Post, The Guardian, Mirror, Mashable, ITV, CBS, NBC, Aljazeera, NBC, Metro, El Mundo, Reuters all run their leading stories within the next five hours and a second wave, heavier in US news and focused on the next days front pages, is published from 7PM onwards. News Timeline Sept 2nd 2015. Source: Pulsar The huge influx of news content leads to the peak in mentions and reach between 9 and 10PM (6K tweets/hour carrying a relevant image and 20K tweets/hour talking about the story overall). But it also changes radically the shape of the diffusion of the images, featuring official accounts from editors and news publications pushing their own angle on the story rather than simply sharing the images. 15 Tweet Timeline, image-only (Sept 2nd, 08.00 to 23.59). Visualising the top tweet by number of retweets for every hour. Source: Pulsar Compared to the previous phase of the diffusion, the audience exposed to the story, sharing the images and talking about them is now 25 times bigger. However, by the end of the day, it is the news ecosystem that ‘owns’ the story and uses it to carry its own content rather than just the pictures. Diffusion graph D: 12 hours in from the appearance of the first picture on Twitter. Size of the nodes indicates impact on the audience (Visibility Score). Colour of nodes and edges indicates country of the user. Source: Pulsar 16 The trend continues the next day where the volume of news content keeps growing and fuels the overall peak of tweets, which is reached at 53K/hour at 8PM on September 3rd. After that, both Twitter activity and news content start to gradually decline for the following 10 days until they go back to 50% of the levels reached on September 2nd. Tweet Timeline vs News Timeline by the hour, all tweets (Sept 2nd, 08.00 – Sept 3rd 23.59). Source: Pulsar Although the appearance and growth of news content has radically changed the velocity of diffusion, the size of the audience reached and the composition of that audience, it hasn’t changed the key behavioural dynamic of the story. Throughout the first 12 hours and through the rest of the viral cycle, the diffusion of the story has remained consistently image-led. Image-Tweets Timeline vs Text-Tweets Timeline by the hour (Sept 2nd, 08.00 – Sept 3rd 23.59). Source: Pulsar 17 Within this dynamic, a visual narrative has emerged as the images going viral have changed over time. Whereas the original images of Aylan Kurdi have dominated the first 48 hours of the diffusion cycle, from September 4th onward they have been replaced by user-generated variations on the original images designed by illustrators and graphic designers. The need for sharing the impactful images and the concerns over hurting the sensibility of the audience have led, within the first 12 hours, to the creation and diffusion of surrogate Aylan Kurdi images designed to mitigate the brutality of the original images in order to sustain the visual narrative of the story and its diffusion. Aylan Kurdi Image-Tweets vs Variations Image-Tweets by the hour (Sept 2nd, 08.00 – Sept 3rd 23.59). Source: Pulsar This is a visual narrative deeply intertwined with Twitter’s ability to act as a catalyst, connect emerging stories and relevant audiences. Its ability to develop and connect those audiences to a global scale and make a story go mainstream before the international press has even started covering it. But we shouldn’t forget that it was the journalists on the ground that broke the story on Twitter and through the social platform put it in front of the right audience, facilitating a very effective and optimized diffusion dynamic. Tweet Timeline by the hour, image-only tweets (Sept 2nd, 08.00 – 23.59). Source: Pulsar 18 2. W hat Can S earch Data Tell Us A bout How S pread A round the World? S imon Rogers the Story of Aylan Kurdi What can search data tell us about the way the tragic story of Aylan Kurdi spread around the world? Did it change the terms of the debate, as well as the social conversation? We used Google Trends data to explore what happened. When the searches began Aylan’s body was discovered around 5.30AM GMT on September 2nd, according to reports, with the photographs taken in the hours after. This chart (above) looks at worldwide searches on Google for some of the main terms people used that first day: ‘Aylan Kurdi’, ‘drowned Syrian’, ‘Drowned Child’, ‘Syrian Child’ and ‘Syrian Boy’. For a search term to show up in this data, it has to be significant, and in this case there were a few searches for the more generic terms before the story even broke. The first serious search interest starts showing up at around 9.40AM GMT, for ‘Syrian boy’ — these searches are before the story took off on Twitter, according to the Pulsar analysis. But it wasn’t until 11.27AM when a few isolated searches start showing up for ‘Aylan Kurdi’ as his name became more widely-known. For the first day, ‘Syrian boy’ was much wider searched. The next day it all changed: As his full story became public, Aylan Kurdi was no longer another anonymous death; his name was widely known and increasingly became the dominant search term. 19 Related searches are a useful way to see Google data - what else do people search for in the same session? These were the top related searches for Aylan in the week after his death, and behind related searches for the photos, came those for his story, reflecting a greater desire to find out what happened. H ow it changed the agenda The biggest impact was in the way that refugees and migrants were searched for. In fact, September saw the highest ever global search volume for the topic of refugees in Google search history. Historically, the term ‘migrant’ and the term ‘refugee’ have seen very similar search volume. September was the month that all changed. Google data is organised by ‘entities’ in the Knowledge Graph which powers search: these are collections of search terms, often thousands of them across different languages, which allow us to search for ‘refugees’ in thousands of possible combinations and hundreds of languages. 20 Where did the story spread to? In the 24 hours after the story broke, the top countries searching for Aylan were a selection of European nations, with the addition of Canada, Argentina and New Zealand. Searches spiked in different countries at different times over the next two days. This chart reflects time differences as each country woke up to the news, but also relative interest ― the news spread like this: 21 • First to show a spike of search interest for Aylan Kurdi was Syria, followed by the UK, then Germany, Italy and lastly the US (where it was a relatively smaller proportion of searches). • The biggest spikes of search interest of these countries were in Germany and Italy. • Germany not only had the biggest spike but the highest search interest among these countries. But as the story spread, so did search interest in different places. Looking at searches for Aylan for the whole of September, the map of the world looks like this, with countries like Mauritius and Kosovo having higher proportional search interest than many of the places, which showed initial concern. Google data is also granular down to city level. We looked at the top cities searching for Aylan in the two days after his death, as the story became more widely known. We also wanted to know whether we could use Google search data to look at other aspects of international migration and movement. This [1] series of maps shows searches for the topic of ‘immigration’ combined with different countries over 2015. 22 M igration to … G ermany Afghanistan and Iran are the top countries in the world for search interest in migration to Germany, and Syria is higher here than for anywhere else in the G8. What the world wanted to know The questions that people ask Google reflect concerns and shifting attitudes. The Washington Post recently looked at how searches around gay marriage, for instance, reflect changing attitudes in the United States. So, we looked at the questions asked on Google for both Aylan and the wider issue of Syrian refugees. Globally, these were the top questions asked about Aylan over September: Questions around refugees and migrants became common after Kurdi’s body was found, and reflected the different politics in each country. We looked at searches in the weeks after the photograph was published to see if they would show different attitudes. 23 In the US, the issue was remote but concerning, despite heavy news coverage of the war in Syria. People took to Google to find out what was going on and why Syrians were migrating. In France, questions around migration focused more on Somalian refugees — and even what it would mean for tourism. In Canada, the death of Kurdi became an issue in the General Election and questions focused on trying to gain a greater understanding. 24 Germany had the highest-placed questions on how to help migrants, plus attempts to understand the difference between a migrant and a refugee. Also trying to understand why refugees would want to go to the UK at all. Italy is the only country to ask about adopting Syrian orphans, plus a desire to find out the Pope’s position. Question four, about Germany’s fingerprint policy, may reflect questions by migrants in Italy hoping to travel there instead. Questions in Hungary reflect a charged political atmosphere, which has seen migrants marching across Europe in response to new rules. The questions from there revolve around the ‘correct’ attitudes of Christians, the cost of migration and even whether Budapest has become dangerous. 25 When Aylan Kurdi died, his family’s tragedy spread around the world, creating a wave of sympathy. But it also changed that world too — altering what we wanted to know about the entire issue of refugees and migration, through the window on the world that is our search box. As soon as something happens, we take to search to find out what happened and what it means. And those ripples are felt across the world. What is G oogle Trends data? Google Trends probably represents the world’s biggest publicly available dataset. Every day there are more than 3 billion searches around the world on Google, and that’s not including Youtube or Google News. The data can be searched freely using the Google Trends explore tool and provides search volume and geographic data. That data is normalised – i.e. high search volume for a particular topic would represent high numbers of searches as a proportion of all of them. In addition, the Google Trends data journalism team is able to extract other data from the database. So, in total, we are able to explore: • Search volume • Top questions asked • Rankings of different search terms and topics • All of the above by geography at global, national, regional and city level There are some key things about the data , which make it powerful : 1. S ize : 3 billion+ searches a day takes you beyond the echo chamber of social media. 2. H onesty : there are no ‘dumb searches’ in Google - and as such it provides a unique insight into what people are really interested in or care about online. 3. Immediacy : as soon as something happens in the world, we take to Google to find out more about it. As a result there are identifiable spikes around news events, often within seconds of them becoming known and before the story has been reported in the mainstream media. Events can go viral on Google. F ootnotes [1] http://googletrends.github.io/iframe-scaffolder/#/s/Lkaayd 26 3. E xamining the Hundred Most S hared Images on Twitter Farida Vis of Aylan Kurdi This paper looks at the hundred most shared images of Aylan Kurdi on Twitter, as collected in the Pulsar dataset. The most shared image in this dataset (of another dead child washed ashore) was retweeted 18,487 times and the least shared (of Scotland’s First Minister Nicola Sturgeon urging the UK to do more about the refugee crisis) was retweeted 706 times. This paper aims to do two things: first, it explores what kinds of images of Aylan were shared. Specifically if there is a difference in showing Aylan dead or alive, speculating why this is done. Second, it explores national differences, focusing especially on image sharing in Europe and the UK. S haring different kinds of images of Aylan K urdi This first section discusses how many of the hundred most shared images on Twitter were of Aylan. It pays particular attention to whether these images showed him dead or alive, alone or with others or whether these were images responding to his death. Overall, half of these most shared images were of Aylan. Aylan, dead Just over one fifth (21%) of the hundred most shared images on Twitter were of Aylan dead. Most users circulated images of Aylan alone, lying face down in the surf (13%), followed by ones showing him with someone else. Mostly the image of the Turkish police officer picking him up or images of scenes that include one or sometimes two officers approaching his lifeless little body (4%). There are also a few images of the front pages of newspapers (4%), which include both these images: Aylan alone and with the Turkish police officer. These images are all front pages of UK newspapers, which all prominently covered this event. Aylan, alive Following calls from Aylan’s aunt to share images of Aylan alive rather than dead - as a more appropriate way to remember him - journalists in particular started to encourage this type of sharing. Overall images of Aylan alive were shared 17% across this dataset. There were five variations on this theme: an image that shows Aylan on his own (which was shared most), standing at the top of a blue slide in a playground smiling with his eyes closed. He is wearing a similar outfit to the one he died in, though it is unlikely that this is the same outfit. Visually however the viewer might be encouraged to imagine this scene as the child just before death: happy, smiling, full of life. The second image of Aylan alone shows him with a football, about to kick the ball. The outfit he is wearing is again similar to the one he died in: a red t-shirt and blue shorts. Once more, one might assume that this image was chosen because of how similar the outfit is to the one Aylan died in. Then there are two images of Aylan with his older brother Ghalib. One where they are sitting on a sofa with a white toy bear sitting in between them: both boys are smiling with Aylan showing a broad toothy smile, his eyes closed. The image easily evokes the sound of two small children giggling and laughing in happier times. A second image of Aylan with Ghalib shows just the boys: Ghalib with a protective arm around his little brother. Again both boys smile sweetly at the camera and the viewer. It’s a perfectly recognizable scene. There is only one image that shows Aylan with his brother and their father Abdullah Kurdi together. This is an image shared by the BBC and shows Abdullah with his two sons, one on either side of him, holding their hands. Aylan has a broad smile and his eyes half closed. One person who seems notably absent (and Mike Thelwall also notes this in his chapter) is the boys’ mother. She is not in any of the images. She is most strikingly absent in the image that includes the father. It is possible that she is 27 the photographer of this as well as some of the other images. Researchers like Gillian Rose have done extensive work on the role of women, mothers in particular, in terms of family photography; as the ones creating photographic records of the family through albums and photographs on display in the house. These personal images are often what the media gains access to as well (Rose, 2010). Whatever the reason, by not seeing Aylan’s mother in any image, she becomes a notably visible absence. She is also rarely mentioned in the newspaper coverage. Aylan, alive and dead There is only one image that presents a compilation image: showing Aylan both alive and dead in one overall frame. On the left Aylan is shown alive, using the image of him with the football, wearing a red t-shirt and blue shorts. In the image on the right he is dead; face down in the surf. By placing these two images together, with Aylan wearing a seemingly identical outfit in each photo and the indistinct landscapes further adds to this, it sets up a macabre before and after sequence: a child happy and alive playing football one minute, dead the next. Response images Nearly a fifth (17%) of the most shared images were response images, including cartoons, sand sculptures or other forms of in situ responses, such as children in Morocco dressed in similar outfits to Aylan lying face down on the beach in Rabat. Responses by artists are discussed in more detail elsewhere in this report, but it is worth discussing one image here as is not discussed in later chapters. This is an image of a sand sculpture created by Sudarsan Pattnaik, who describes himself as an ‘international sand artist’. Overall the image of the sand sculpture is shared eight times in the dataset. Pattnaik created the sculpture of Aylan in Puri beach in Odisha. It seems that he frequently creates pieces connected to events that have been in the news and then tweets images of his work, presumably to encourage retweeting and further online engagement and spread of the images. His response to Aylan’s death shows the body of Aylan carved in the sand with the words ‘Humanity washed ashore, shame, shame shame’ written out in front. Taken all together these individual shares of Pattnaik’s sand sculpture total nearly 20 thousand shares (18,800) altogether. The image is also included in newspaper coverage in the UK (for example in this article in The Mirror on 5 September). I mage sharing by country This next section discusses which countries these images were shared from, focusing explicitly on sharing from Europe, responsible for nearly half of the shares. The discussion then pays particular attention to images shared in the UK followed by a brief discussion of image sharing originating from North America, The Middle East and North Africa (MENA), Asia, and Latin America. Europe Nearly half (48%) of the hundred most shared images were shared from accounts located in Europe. Because of the varying responses to the migration crisis across Europe these images were divided further to highlight specific zones within Europe. This shows that most of the image sharing took place in Northern Europe (25%) - United Kingdom (24%) and Ireland (1%); followed by Western Europe (15%) - France (9%), The Netherlands (4%) and Germany (2%). Relatively few images were shared from Eastern Europe - Turkey (3%), Serbia (1%) and Hungary (1%). The same was true for Southern Europe - Spain (2%) and Greece (1%). It is perhaps surprising that so few images were shared in Germany given the central role this country has had in accepting refugees and the media coverage this has received. One explanation might be the language bias in the data collection, favouring those countries where English is the main language. More than that, Twitter use in Germany is relatively low; this Forbes article from July 2014 notes that there are only 4 million Twitter users in Germany compared to 15 million in the UK, which could further explain the low sharing figures for Germany. 28 Language bias could also explain the relatively low numbers for Turkey and Greece, two countries central to this event and subsequent images. Again these issues could be addressed in further research. The 24 images shared in the UK are worth highlighting in detail in that they include a wide range of different types. However, none are shares of the two original images depicting Aylan dead; showing him alone or in the arms of the Turkish police officer. Rather, these two images are shown indirectly through the sharing of images of newspaper front pages. This includes ‘tomorrow’s front pages’ ( The Guardian, The Independent) and compilations that highlight how individual newspapers have changed their stand on the refugee crisis. For example by presenting front pages of The Sun to illustrate the paper’s ‘U-turn’. Another strategy is to show a compilation of front pages from papers across the political spectrum to emphasise their shared position on the migration crisis. More than that, the text of the tweet accompanying this compilation urges UK Prime Minister David Cameron to ‘wake up’ and for the UK to play its part in responding to the refugee crisis. Other images continue this theme of urging the UK to do more, by presenting various infographics, image compilations or photographs of well-known landmarks and British icons such the London Underground to either emphasise that the UK could and should take on more refugees or to highlight how few have been allowed to come to the UK so far. A picture of Buckingham Palace suggests there is space for at least 1000 refugees in this building, while another image shows that all the Syrian refugees the UK has accepted so far could all fit into one London Underground train. Both make clever use of familiar objects to discuss these numbers, perhaps to make them more accessible and easier to understand. There are further images of texts mocking David Cameron’s position on the crisis, specifically noting his drastic U-turn on the situation in the aftermath of the publicity around Aylan’s death. Imitating the Prime Minister one text starts: ‘Britain has a moral responsibility to help refugees, even though it didn’t a couple of days ago’. There are also images that directly show politicians, featuring pictures of their faces next to quotations of them pledging to play their full part in finding solutions to this crisis (Scotland’s First Minister Nicola Sturgeon), or showing politicians posing with ‘Refugees Welcome’ signs (Green Party MP, Caroline Lucas, tweeted by herself), as well as an image of new Labour leader Jeremy Corbyn, again tweeted from his own account, showing him looking over a crowd gathered in Parliament Square. The text in the tweet highlights that these people have gathered because they want a government who can and will respond properly to this crisis. All these politicians presumably identify themselves as such politicians: who can and will do more. Linked to this theme, two final images are worth mentioning in that they both highlight moments in history where the UK did far more to help refugees from across Europe. The first image is an aerial image of the ‘The Great Polish Map of Scotland’ a large three-dimensional, outdoor concrete scale model of Scotland built as a thank you to the Scottish for giving Poles refuge during the Second World War. The second image is an old black and white photograph of a large steamship in the Southampton dock. The tweet states that it shows how the UK accepted 4,000 Basque children in 1937, asking: ‘would we have let them drown’? Perhaps noting that back then the UK was accepting refugees from Southern Europe, but is now unwilling to accept refugees from the Middle East to the same extent. Overall the images shared from the UK are highly diverse in terms of type, with a high degree of creativity in sharing various political views. North America Nearly a third (28%) of the hundred most shared Twitter images were shared from accounts located in North America. Most came from The United States (27%) and only one from Canada. Given the United States’ involvement in the wider geopolitics and military situation, particularly in relation to Syria, this number of shares from the United States is perhaps not surprising. It is worth noting that the most shared image overall is from the United States. It shows a dead child (not Aylan) and is shared on 4 September, presumably to highlight the general silence so far concerning dead children washing up on European shores. 29 Middle East and North Africa (MENA) Overall, relatively few (11%) of the hundred most shared images were shared from accounts located in the MENA region. Most were shared from Lebanon (4%) and Morocco (4%), with a single share each from Qatar, the United Arab Emirates and Saudi Arabia. Images from the MENA region included several shares of the image showing Aylan dead in the surf (all from Lebanon) and responses by artists: a cartoon of Aylan with angel wings (Saudi Arabia) and one of a white angel picking Aylan up from the surf (Qatar). An image from the UAE highlights the ‘Syrian exodus’ apparently showing a large group of refugees crossing a bridge, walking from Bulgaria to Austria. Two identical images from Morocco show a crying man getting off a boat holding a child. A quick search shows this is Laith Majid, a refugee who nearly drowned with his family earlier in the summer trying to reach Greece. The image was widely circulated at the time and was presumably shared again following Aylan Kurdi’s death to highlight the plight of refugees. Language bias almost certainly will have played a factor in the relatively low shares from the MENA region and is worth exploring in further research. Given that some of the response images highlight the failure of Arab leaders it would be interesting to examine how many of these images are shared across the whole dataset. It might be the case that whilst they do not make the top 100 in terms of volume of shares, they still play a significant role overall and within the MENA region and certain countries in particular. Further work would need to expand the search to include Arabic and draw on additional expertise from scholars from or familiar with the region, including in terms of differences in visual culture. News outlets within the MENA region frequently share images that within a North American, Northern and Western European tradition would be deemed too graphic for publication and this may also affect image sharing on Twitter. Asia Relatively few (7%) of the most shared images were shared from accounts located in Asia. This includes South East Asia (Malaysia (3%) and India (2%), as well as Asia more widely, China (1%) and Japan (1%). These images included the image of Aylan dead in the surf, the sand sculpture tribute (Malaysia and India), Aylan alive with his brother Ghalib (Malaysia) and destruction of buildings (Japan). Latin America and Other Few of the hundred most shared images were shared from accounts located in Latin America - Argentina (1%) and Venezuela (1%) and a few came from unknown locations (4%). It is worth nothing that the image from Argentina shows a heavily dehydrated 13-year old boy, who according to the text weighs only ten kilos, presumably to draw attention to a specific situation in Argentina. C onclusion This short paper has given an overview of different sharing practices on Twitter concerning the two main images of Aylan Kurdi following his death. It has highlighted differences in showing Aylan either dead or alive and when alive the apparent strategy of showing him in an outfit similar to the one he died in. The absence of his mother in these pictures was also highlighted. The chapter reported on a number of international trends, with a specific focus on the wide range of image types shared in the UK. This showed a range of interesting, nuanced and creative responses, urging the UK government to do more. Further work could build on this, extending search terms to get a better understanding of image sharing within the MENA region. This work would benefit from working in collaboration with researchers who have language and regional expertise. Such work could potentially produce valuable and rich comparative insight. References Rose, G. (2010). Doing Family Photography: The Domestic, The Public and The Politics of Sentiment. 30 Farnham: Ashgate. 4. Undermining Aylan: L ess Than Sympathetic International Responses Mike Thelwall The images of Aylan Kurdi that spread on social media on the day of his death, 2 September 2015 (Butler & Toksabay, 2015) created a seismic shock for opinion about Syrian refugees in Europe. The images depicted an indisputably innocent tragic victim of out of control events: a child in a position that many would instinctively recognise as typical peaceful sleep. The natural urge to protect this already dead child extended to sympathy for his family – brother Galib and mother Rehana, who died, and father Abdullah, who survived the accident. Perhaps in many countries there was also guilt about national military involvement in the tragedy in Syria but there was certainly a widespread burst of strong feeling that Europe should do more to help other families in the same situation. For example, opinion polls in France showed a large shift in public opinion towards refugees after the publication of the image (Love, 2015). Canada, which was the target destination of the Kurdi family, was directly affected by an apparent previous failure to grant them a visa to join a family member in Vancouver (Moyer, 2015). Not all people were touched by the image in the same way, however, and there were attempts in the press and elsewhere to undermine its central message, accompanied perhaps by national differences in its reception. One way in which this occurred was through prominent press coverage painting Aylan’s family as something other than innocent victims. Press ran stories claiming that Aylan’s father Abdullah Kurdi was driving the boat and that he was therefore a criminal, ‘a people smuggler’ (Stanton, 2015; Oliphant, 2015), implying that he was partly responsible for his son’s death. The allegations against Abdullah Kurdi seem to originate from a Reuters reporter or source, Saif Hameed in Baghdad (Reuters, 2015). Some news stories also claimed that he may have abandoned the boat and passengers to their fate and that his wife and children had been reluctant to travel from a relatively secure place where they had been living (Lee, 2015). This strategy undermines the central message of the images of Aylan Kurdi. Although it would be impossible to claim that he was anything other than an innocent victim, the story would complicate the sympathy for his family and hence undermine the overall power of the image, as ‘the story behind his tragic death is more complicated than it first appeared’ (Lee, 2015). This issue grew to the extent that naturally sympathetic press, such as The Independent, directly refuted the allegations (Griffin, 2015) and published a more sympathetic version starting with the term tragic, ‘Tragic Aylan Kurdi’s father was “people smuggler who led boat journey which killed his sons” - survivor claims’ (Wheatstone, 2015). Outside of the UK, other press also covered the allegations against Aylan Kurdi’s father, including in Australia (Booker, 2015), which would not be directly affected by Syrian refugees. A right wing Australian politician also went further by accusing Abdullah Kurdi of organising the trip in order to get dental treatment (Mann, 2015). Russian press, with a national interest in Syria and presumably hostile to the migrants, also published the allegations against (RT, 2015). The allegation that Abdullah Kurdi drove the boat also received wide coverage internationally, including in Ireland (Irish Times, 2015), India (Hindustan Times, 2015), and Canada (Toronto Sun, 2015). The Aylan Kurdi image was also attacked by ISIS supporters, because allegedly his family was ‘pretending’ to escape from ISIS, with ISIS seeing the existence of the refugees as an accusation against them (hence branding them as economic migrants) (Admin, 2015). Two simultaneous strategies to indirectly undermine the impact of the image were to post pictures of former ISIS or Al Qaida members arriving in Europe as refugees (e.g., September 10, September 9) with the implication that either they are infiltrating Europe for terrorism or that they are undeserving because they caused the conflict that they are fleeing. While this implication is sometimes explicit and in any case vocal; at least some of the images are fake while others are misleading (Dearden, 2015). The remainder of this paper will explore international reactions to Aylan’s images that are hostile or at least less than sympathetic. The main focus is on social media content that mentions his father in a way that provides a ‘logical’ angle from which to undermine the impact of the image and a more subtle strategy than the obvious direct attack. 31 The data from the analysis is 2,843,274 social media, blog and news posts from September 1 to 14, 2015 gathered through Pulsar. Almost all (99%) of the posts are from Twitter so the analysis is essentially an analysis of tweets containing a keyword related to Aylan Kurdi or the refugee crisis during this key time period (see Figure 1). Figure 1. Keywords used to search social media for relevant posts. To give context to the social media analysis, tables 1-4 lists the 20 most popular terms in a number of different countries for English, French, German and Turkish. In most countries, the hashtag #refugeeswelcome is the most popular, except in Turkish, confirming the overall very strongly positive sentiment towards refugees (except in Turkey). None of the hashtags are associated with an anti-migrant theme in English or German content. In French, however, the term ‘father’ appears as a top keyword in France, Canada and Switzerland, and in Turkish the keyword ‘father’ appears in Germany and Turkey. An examination of tweets containing the Turkish word for ‘father,’ babas, found that some contained sympathetic stories, but many others discussed the allegations against him. This suggests that the overall reaction to the incident in Turkey may have been different to many countries in Europe. This perhaps relates to the ethnic origin of Aylan and the political issues in Turkey relating to the Kurdish minority, although there was no specific evidence of this. Table 1a. The most common keywords and hashtags in social media posts in English related to Aylan Kurdi, by country. Rank UK USA Canada Australia India 1 #refugeeswelcome crisis crisis #refugeeswelcome refugee 2 refugee refugee refugee refugee crisis 3 crisis #refugeeswelcome syrian crisis syrian 4 #refugeecrisis #refugeecrisis #refugeeswelcome #refugeecrisis aylan 5 refugees syrian #refugeecrisis #auspol drowned 6 syrian help #cdnpoli syrian boy 7 help drowned kurdi aylan #refugeeswelcome 8 aylan @savechildrenuk drowned refugees #aylankurdi 9 people stand canada #aylankurdi kurdi 10 kurdi boy boy kurdi world 11 boy refugees #elxn42 drowned #refugeecrisis 12 drowned aylan refugees help europe 13 response kurdi help #refugees sand 14 europe @harry_styles family people help 15 welcome europe aylan #lightthedark refugees 16 cameron children #aylankurdi boy photo 17 today washed harper australia beach 18 world #aylankurdi world world indian 19 child world alan europe artist sign dead stand #syria sculpture 20 32 Table 1b. The most common keywords and hashtags in social media posts in English related to Aylan Kurdi, by country. Rank Turkey France Spain Netherlands 1 #refugeeswelcome Germany syrian #refugeeswelcome #refugeeswelcome #refugeeswelcome 2 refugee refugee refugee crisis crisis 3 crisis crisis crisis refugee refugee 4 #refugeecrisis #refugeeswelcome #refugeecrisis #refugeecrisis #refugeecrisis 5 refugees #refugeecrisis #aylankurdi syrian syrian 6 syrian aylan help help help 7 #refugees #aylankurdi syrian boy aylan 8 help boy aylan @savechildrenuk refugees 9 europe drowned @savechildrenuk stand @savechildrenuk 10 boy kurdi stand refugees boy 11 aylan #refugees refugees aylan stand 12 people refugees boy drowned kurdi 13 welcome europe photo @harry_styles drowned 14 #aylankurdi world kurdi europe people 15 @savechildrenuk #syria @harry_styles #aylankurdi europe 16 drowned child drowned syria #aylankurdi 17 stand photo europe today world 18 #refugeescrisis help #aylan kurdi @harry_styles 19 germany stop world beach photo 20 photo migrant #refugees image stop Table 2. The most common keywords and hashtags in social media posts in French related to Aylan Kurdi, by country. Rank France Canada Switzerland 1 #aylankurdi aylan aylan 2 aylan petit #aylankurdi 3 #aylan kurdi #aylan 4 petit #aylankurdi petit 5 #refugeeswelcome mort photo 6 photo #aylan mort 7 kurdi père #refugeeswelcome 8 mort photo daylan 9 père canada kurdi 10 daylan famille père 11 monde #refugeeswelcome plage 12 ans daylan aux 13 réfugiés réfugiés ans 14 plage ans famille 15 enfant monde #syrie 16 #migrants noyé monde 17 hommage plage gt 18 cest été hommage 19 famille syrien guerre 20 paix enfants noyé 33 Table 3. The most common keywords and hashtags in social media posts in German related to Aylan Kurdi, by country. Rank 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Germany #refugeeswelcome flüchtlinge #trainofhope #flüchtlinge heute jetzt refugees #trainofhopedo hier menschen welcome danke #fluechtlinge mal deutschland helfen aylan #aylankurdi @matthiasmeisner #marchofhope Austria #refugeeswelcome #trainofhope #westbahnhof #marchofhope flüchtlinge heute jetzt #hbfvie #refugeecrisis bitte menschen danke gebraucht refugees hier wien #wien dringend #infotrain #nickelsdorf Switzerland #refugeeswelcome flüchtlinge heute aylan #trainofhope #flüchtlinge jetzt hier #syrien refugees menschen #aylankurdi #marchofhope #aylan europa welcome mal deutschland #refugeecrisis danke Table 4. The most common keywords and hashtags in social media posts in Turkish related to Aylan Kurdi, by country Rank 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 34 Turkey aylan #aylankurdi ın bir için kurdi bu kurdinin minik kıyıya #alankurdi vuran dünya ailesi (family) babası (father) suriyeli onu demiş ile nin Germany aylan #aylankurdi ın için bir kurdi bu kıyıya #alankurdi kurdinin minik vuran #refugeeswelcome ile ailesi (family) #aylan galip babası (father) nin olan There were some international differences in the proportion of posts mentioning negative or potentially unsympathetic issues by country. In the USA (California), Canada and Australia, these themes seemed to attract a relatively high number of posts. Overall, however, the level of interest in most countries was low. Figure 2. The percentage of English posts containing the keywords “ISIS”, “smugglers” and “Abdullah” in ten countries with most posts in English. C onclusion Overall, although there was some evidence of themes that were less sympathetic to Aylan Kurdi’s family in social media during this nearly two-week period, and particularly in California and Canada, by far the dominant theme everywhere investigated was support for refugees. The only exceptions found were French speaking countries or areas: France, Switzerland and Canada, as well as Turkish language content from Germany and Turkey. In Turkey, for example, Aylan Kurdi’s father was a major theme of relevant tweets and although some were sympathetic stories, others discussed the smuggling allegations against him, which may have had the purpose of undermining support for refugees. These conclusions are tentative because the data collection primarily used English keywords and did not include French and Turkish hashtag equivalents of #refugeeswelcome. References Admin (2015, September 4). Aylan Kurdi’s family had FREE housing in Turkey, while father’s story is full of holes. Retrieved from https://themuslimissue.wordpress.com/2015/09/04/family-of-drowned-toddler-aylan-kurdihad-been-given-free-housing-in-turkey-while-fathers-story-is-full-of-holes/ Booker, C. (2015, September 11). Aylan Kurdi’s father is a people smuggler, woman claims. The Sydney Morning Herald. Retrieved from http://www.smh.com.au/world/migrant-crisis/aylan-kurdis-father-is-a-people-smugglerwoman-claims-20150911-gjkt2m.html Butler, D. & Toksabay, E. (2015, September 3). UPDATE 5-Twelve Syrians drown heading from Turkey to Greek island. Reuters. Retrieved from http://in.reuters.com/article/2015/09/02/europe-migrants-turkey-idINL5N1180RA20150902 Dearden, L. (2015, September 16). The fake refugee images that are being used to distort public opinion on asylum seekers. Independent. Retrieved from http://www.independent.co.uk/news/world/europe/the-fake-refugeeimages-that-are-being-used-to-distort-public-opinion-on-asylum-seekers-10503703.html Griffin, A. (2015, September 13). Aylan Kurdi’s father denies claims he was a people smuggler and driving boat that capsized and led to son’s death. Independent. Retrieved from http://www.independent.co.uk/news/world/ 35 middle-east/aylan-kurdi-s-father-denies-claims-he-was-a-people-smuggler-and-driving-boat-that-capsizedand-led-10498798.html Hindustan Times (2015, September 13). Aylan Kurdi’s father is a people smuggler, refugee woman alleges. Retrieved from http://www.hindustantimes.com/world/aylan-kurdi-s-father-is-a-people-smuggler-refugee-womanalleges/story-QxUU70vSj7DcbXH3hAkp3J.html Irish Times (2015, September 11). Aylan Kurdi: Syrian toddler’s father drove boat, say passengers. Retrieved from http://www.irishtimes.com/news/world/europe/aylan-kurdi-syrian-toddler-s-father-drove-boat-saypassengers-1.2349012 Lee, A. (2015, September 15). Migrant crisis: The truth about the boy on the beach Aylan Kurdi. Express. Retrieved from http://www.express.co.uk/comment/expresscomment/604590/Migrant-crisis-the-truth-about-the-boy-thebeach-Aylan-Kurdi Love, B. (2015, September 10). French opinion rapidly swings in favor of refugees, poll shows. Reuters. Retrieved from http://www.reuters.com/article/2015/09/10/us-europe-migrants-france-idUSKCN0RA1MD20150910 Mann, S. (2015, September 8). Aylan Kurdi died because his father wanted dental treatment, claims Australian politician Cory Bernardi. Evening Standard. Retrieved from http://www.standard.co.uk/news/politics/aylan-kurdi-diedbecause-his-father-wanted-dental-treatment-claims-australian-politician-cory-a2942576.html Meers, D. & McClellan, B. (2015, September 11). Aylan Kurdi: Drowned boy’s father accused of being people smuggler in charge of boat that crashed denies claims. The Daily Telegraph. Retrieved from http://www.dailytelegraph.com. au/news/aylan-kurdi-drowned-boys-father-accused-of-being-people-smuggler-in-charge-of-boat-thatcrashed-denies-claims/story-fni0cx4q-1227523270638?sv=4d19a328df0eb1692a58f672b0dd7346 Moyer, J. W. (2015, September 3). Aylan’s story: How desperation left a 3-year-old boy washed up on a Turkish beach. Washington Post. Retrieved from http://www.washingtonpost.com/news/morning-mix/wp/2015/09/03/adesperate-refugee-family-a-capsized-boat-and-3-year-old-dead-on-a-beach-in-turkey/ Reuters (2015, September 11). Syrian toddler Aylan’s father drove capsized boat, other passengers say. Retrieved from http://www.reuters.com/article/2015/09/11/us-europe-migrants-turkey-iraq-idUSKCN0RB2BE20150911 RT (2015, September 11). Capsized boat survivors claim Aylan Kurdi’s father is people smuggler. Retrieved from https://www.rt.com/news/315116-passengers-reveal-kurdi-smuggler/ Oliphant, V. (2015, September 12). Father of Aylan Kurdi accused of driving boat that killed his family. Express. Retrieved from http://www.express.co.uk/news/world/604730/Father-Aylan-Kurdi-accused-of-driving-boat-that-killedfamily SkyNews (2015, September 11). Claims Aylan Kurdi’s father “people smuggler” Retrieved from http://www.skynews.com.au/news/top-stories/2015/09/11/claims-alan-kurdi-s-father-a-people-smuggler.html Stanton, J. (2015, 14 September). Revealed: SECOND passenger claims Aylan Kurdi’s father was working with people smugglers and was driving the boat on which his son died. Mail Online. Retrieved from http://www.dailymail. co.uk/news/article-3233802/Second-passenger-claims-Aylan-Kurdi-s-father-driving-boat-son-died-workingpeople-smugglers.html Toronto Sun (2015, September 11). Iraqi couple allege Alan Kurdi’s father was captain of the boat that capsized. Retrieved from http://www.torontosun.com/2015/09/11/iraqi-couple-allege-alan-kurdis-father-was-captain-of-the-boatthat-capsized Wheatstone, R. (2015, September 11). Tragic Aylan Kurdi’s father was “people smuggler who led boat journey which killed his sons” - survivor claims. The Mirror. Retrieved from http://www.mirror.co.uk/news/world-news/tragic-aylan-kurdis-father-people-6426179 36 PART II WHAT DID THE IMAGE DO? This section is concerned with registering and analysing the variety of responses that the circulation of images of Aylan Kurdi evoked. The papers try to comprehend what these responses meant, how effective the actions that were triggered by emotional engagement with the images were and whether the outcome of this engagement became politically meaningful and lasted. Aylan Kurdi’s images triggered strong personal responses, mobilised civil society into organising networks of support and lobbying for the refugees, including various Refugees Welcome campaigns, and became the subject of artistic work. The findings of Anne Burns, Lin Prøitz, Lucy Mayblin and Holly Ryan vary. Political efficiency turns out to largely depend on local context, and while Norway, one of the case studies, seems to have mobilised in a productive way, in the UK, after the initial peak of affective responses, matters for the refugees keep darkening. While the image of Aylan Kurdi acted as a symbol that represented the deaths of many by the death of one and it generated affect that was used for political mobilization, whether this tragedy is able to transform public opinion, public action and legislation to deal with the worst refugee crisis since the Second World War remains to be seen. 37 5. Discussion and Action: P olitical and P ersonal Responses to the Aylan K urdi Images A nne B urns As Francesco D’Orazio’s Twitter data analysis demonstrates, the Aylan Kurdi image prompted a shift in the conversation concerning migrants and refugees. Studying online discourse, at both the level of terminology and the broader concepts expressed, enables us to explore how public opinion is shaped at both a local and global level. Here, I will be examining responses reported in the media from political figures and private individuals and analysing how the discussion of photographs can potentially act as a catalyst for wider action, and for the formation of consensus. A search of international news sites from the two weeks following the image’s publication (from the 2nd to the 16th of September 2015) demonstrates two strong themes, relating to the effect on viewers, and to the symbolic role of this tragedy. The image was termed a ‘wake up call’ ‘lightbulb moment’ that ‘sparks an outcry’ and serves as a ‘switch to the world’s conscience’. Commentators asserted that Kurdi’s death was representative of a wider crisis, stating that the image ‘captures human tragedy’, and serves as a ‘tragic symbol’. The metonymic role of the image – in which Kurdi’s body is interpreted as representing the deaths of many others – enables the image to become symbolically powerful, and to function as shorthand for multiple related conversations about refugees. This sense of a wider message and impact is emphasised by the assertion that the photograph’s message ‘echoes around the world’, makes ‘everyone become human’ and sits alongside other images that ‘change history’ and ‘shook the world’. These news stories describe not just a seismic moment of reaction – evoking images of sparks, light bulbs and switches – but also an aftereffect, in terms of global and historical consequences. Such terms divide time into a before and an after, in which ignorance and inattention has been replaced with awareness and concern. This suggests that things cannot be the same now, that what has been seen cannot be unseen, and that the world has been irrevocably altered. Yet these statements of the image’s power originated very soon after its emergence and therefore represent an initial peak (albeit an enormous one) in interest, rather than a sustained change in public discourse. Asserting that this image has a lasting role in – and effect upon – global politics may therefore be premature. So why would so many reports talk about the Kurdi image in this way? I would argue that this assumed political function is cited in order to redeem the image’s circulation, in that it is constructed as being acceptable to look at and share a photograph of a dead child if that is perceived to do something, and to somehow improve a situation that otherwise seems overwhelmingly complex. But when Kurdi’s death functions as a ‘tragic symbol’, it becomes depersonalised – he is no longer just a 3-year old, but is a representative of the thousands of lives lost in the last few years. And this not only places a great responsibility on a single image, it also begins to obscure what it is we are actually looking at, as it seems that we then see the symbol and not the life that was ended much too soon. The sentiments expressed by global political leaders in response to the Kurdi image echo the wider media response, in that they demonstrate a unified sense of dismay. UK Prime Minister David Cameron stated that he was ‘deeply moved’ his (then) Canadian counterpart Stephen Harper described Kurdi’s death as a ‘heartbreaking situation’ and Scotland’s First Minister Nicola Sturgeon was ‘reduced to tears’. Although other politicians responded in different ways – either by using it to criticise an opponent’s policies and argue for increased military intervention, or to promote their own message of intolerance regarding refugees – the wider message seemed to be one of grief and shock. But was there a connection between these expressions of sorrow, and a change in policy? Initially, there did seem to be a correlation, in that Cameron and Sturgeon both announced an increase in assistance. In a speech delivered a month after Kurdi’s death, Cameron stated that he found it “impossible to get the image of that poor Syrian boy Aylan Kurdi out of [his] mind”, suggesting that the image continues to influence his decision-making. Yet he also spoke of the need to avoid being “overwhelmed” by refugees, and later rebuked the Church of England for prompting him to increase assistance for those who have already made the journey to Europe. Harper, who had displayed sympathy at Kurdi’s death as the little boy reminded him of his own son, was later found to be obstructing immigration officials from 38 processing Syrian asylum claims, in order to ‘prevent potential militants entering Canada’. And when asked if Sturgeon would still be prepared to offer a home herself for refugees – as she had suggested in an interview a few days after Kurdi’s death ­– less than three weeks later a spokesman stated that there were no plans for this to happen. Therefore it would seem that the political figures joined in with the wider discussion about the Kurdi photograph, as not to do so appeared politically unacceptable, but that their responses were framed in relation to the public outcry, rather than an accurate reflection of their own shift in attitudes. Alongside politicians’ responses, media reports concerning the widespread reaction to the Kurdi image also focused on private individuals. Across the UK, a similar narrative emerged of people who were prompted upon seeing the photograph into action of some kind, whether starting a campaign, a petition, collecting aid, coordinating shelter for refugees or volunteering. The Telegraph acknowledged this specific correlation between seeing the image, and the wish to help refugees, by collating a number of options for readers to donate or volunteer. It is this relationship between viewing and doing – that is the most striking element of the public discourse about the Kurdi image, in that private individuals have no requirement to comment on the subject as politicians do, and are not obligated to make a display of sympathy of their will to act, but yet this is what emerged in the weeks following September 2nd. There is one further element of this discussion that merits examination, in that responses were often framed as being from the vantage point of a parent. Boxing champion Amir Khan stated that he was spurred into collecting aid for refugees ‘because he was a dad’.The Mirror reported on a popular blog post that compared Kurdi to the writer’s own son, as being ‘the same age, and a similar size’. And as mentioned above, both Harper and Cameron liken Kurdi to their own children. This process of personalising the death of Kurdi, of likening him to a child that the commentator knew, or of responding ‘as a parent’, demonstrates how deaths can be reframed as more significant, and more shocking, once the victim has been ‘un-othered’. But yet only specific victims seem eligible for this process, in that no one came forward to express sympathy for Kurdi’s mother, as someone they could relate to ‘as a mother’. Following on from Ray Drainville’s exploration of the symbolism of the child, the online discussion of the Kurdi image demonstrates a cultural focus on not only the value of children, but on the vantage point of parenthood. Mothers and fathers cited their own life experience as reason for their compassion in this case, rather than a general concern for human suffering. Therefore although individuals’ responses appeared to be more genuine than politicians, in that they were unprompted and aligned with heartfelt (if misplaced) plans for action, empathy nevertheless seems to extend only to those that can be somehow imagined, and related to, by the viewer. Additionally, a You Gov report demonstrates that even these kinds of responses are not necessarily representative of a shift in public opinion, in that only 9% of those who reported seeing the Kurdi image stated that they believed more Syrian refugees should be allowed into the UK. A further You Gov report suggests that public sympathies are shortlived, in that similar stories from last year did not effect a lasting change in public opinion towards refugees. The language used to describe the Aylan Kurdi image expressed a widely felt sense of shock, and a wish for change. At both a political and individual level, responses focused on the tragic nature of the little boy’s death and on the need to turn sorrow into positive action. Yet, within only a few weeks, politicians were reneging on their promises and it is predicted that public opinion will shift again too. Therefore despite the narrative of ‘photographs that changed the world’, progress does not necessarily result from tragedy. But it would be my hope that public opinion was indeed changed by this photograph – if not greatly, as might have been expected by the media coverage, then incrementally. After all the hyperbole has died down, this image will have taken its place in the public imaginary, and in the public moral conscience, and as a result I would hope that it will never be possible to demonise and dismiss refugees in quite the same way again. 39 6. The Strength of Weak C ommitment : A Norwegian Response Kurdi Images L in P røitz to the Aylan In this article, I map out the mobilisation in Norway in September 2015, using the Facebook-group Refugees Welcome to Norway (#RWTN) as a case study. I examine the momentum around this engagement, and how the sheer force of number of people involved affected the Norwegian local government election and influenced the political participation by young Norwegians in particular. In Norway September 2015, there were two specific occurrences that gained publicity in the media: one concerned the Norwegian local government election that is held every second year in September (this year it was on the 14th), and the other was related to the migrant crisis. Particularly it was related to the image of Aylan Kurdi and the engagement with refugees that arose out of its spread. #RWTN – R efugees Welcome to N orway In July 2015, a young man stopped a woman passing by at Tøyen in Oslo, and asked for the direction to the Police Immigration Unit (PU). The woman decided to walk him to the station, not anticipating what she was about to see. The PU office is situated in an anonymous brick building and it is a place where all newly arrived refugees are registered before they are sent to different asylum and refugee reception centres all over the country. The small PU office, not designed to host people overnight or to manage the great number of refugees that arrived in August and September, has literally become a shelter with two to four hundred refugees arriving daily. All over, PU, people were sitting, standing and, in the backyard, lying directly on the ground. Only one toilet was in service. The woman, who witnessed these inhuman conditions, went straight out and bought food for thousands of Norwegian Krones, which she shared among the refugees at the PU. As a direct response to what she witnessed and what she later termed as an issue that had been shrouded in the ‘the cold political debate’, she spontaneously founded #refugeswelcometonorway - #RWTN, an ad hoc issue-based social media group (Vromen, Xenos & Loader 2015). The aim of the group and its Facebook page was to inform the public about what was actually happening right then, right there in Norway. The page was a space where people could share experiences and stories as well as being a tool for organising meetings, mobilising volunteer actions – and, as the founder said - a space where people could come together and do what politicians did not: engage and care about humanity. The Facebook group rapidly grew to 200 members and in August started collaboration with one of Oslo’s most distinguished restaurant owners, entrepreneurs and chefs, who initiated distribution of restaurant-quality food directly to refugees from his many restaurants, whilst encouraging other restaurants to follow. While carrying out these activities, #RWTN developed a more thorough and better-structured organisation. Shortly after the media publicity wave regarding the chef’s initiative, images of Syrian refugees carrying their children on their shoulders whilst walking through Europe started to fill the television news. #RWTN received a noticeable number of inquiries from people wanting to become members and help. When the image of Aylan Kurdi went viral, however, the #RWTN-Facebook-group exploded. From being a Facebook group that was starting to raise awareness, #RWTN has grown to a group of nearly 90,000 members almost overnight. All of a sudden, masses of citizens volunteered, impatient to help. People wrote on the Facebook wall asking the same questions: Where can I deliver food? Where can I hand in clothes? Where can I hand in toys? How can I help in other ways? Where and how do I register as a volunteer? At the same time, the images of Aylan Kurdi as well as emotions around the image were shared widely; and among all the posts, comments from refugees expressing their gratitude for the Norwegian engagement could occasionally be seen. 40 After approximately one and a half week, more critical voices were raised in the Facebook group. Some claimed that most of the group members were ignorant citizens who lacked knowledge and real commitment, emphasising that this crisis is not just something that started in September, that there were not only Syrian refugees and that people did not just start walking across Europe all of a sudden. Other voices started to debate whether or not the group should be politically engaged. Some claimed that by not linking the group to politics, there were better chances to maintain the wonderful engagement of all those who ‘watched television, and just wanted to donate their sweater’. Others argued that this crisis is by all means political and that people had to start differentiating between politics and party politics. Regardless of people’s strong or weak commitment and despite internal disagreements and battles, the group of nearly 90,000 members existed due to one common concern: the desire to contribute in Europe’s largest refugee crisis since the Second World War. A Facebook group with nearly 90,000 members is a large social media group in the Norwegian context. The group achieved substantial publicity. In addition to this group, several locally anchored groups emerged, both on and offline. At the same time as the image of Aylan Kurdi started being widely shared and the activity of #RWTN skyrocketed, the Norwegian Government local election campaign entered into its final phase. D id one photo influence the N orwegian G overnment local election ? Although some researchers and commentators claimed that the civic engagement in Norway was more about emergencies and humanitarian aid than politics and long-term integration, mobilisation and change in political rhetoric was clearly noted. The Social Left Party, the Left Party, and the Christian People’s Party all argued that the refugee crisis and the Norwegian people’s obligation to help overshadowed and should overshadow local matters. The change in rhetoric was mostly notable in numerous televised election debates between the Conservative and the Progressive Parties. Traditionally, the Progressive Party would gain attention and support when asylum and immigrant politics dominated public media. This year, petitions of households that signed up to host and otherwise contribute to settling the refugees mobilised the nation – and made it difficult for the Progressive Party to gain support as they were proposing to reduce or refuse refugees’ settlement altogether. In this year’s TV election campaign, whenever politicians from the Progressive Party tried to speak out against immigration and refugees, they would encounter significant resistance from the audience and many other parties. Overall, the Labour Party, the Center Party and the Green Party were the winners of this year’s election (Torset, Færaas 2015). The election was devastating for the right-wing parties. The Progressive Party achieved the worst results in a governmental local election in the last 24 years. So, did the image of Aylan Kurdi influence the Norwegian Government local elections? The image of Aylan Kurdi and #RWTN indeed raised awareness of the refugee crisis among the public in general and with young people in particular. During these weeks, the ad hoc issue-based social media group #RWTN represented a space where people - through their every day political talk, expressions of emotions, concerns, debates and storytelling - engaged politically. #RWTN became a public space, where young people, through their many debates, attempted to conceptualise, but also redefine what political engagement is and can be. The broader study that has informed this short article will require more detailed, systematic investigation in order to thoroughly answer the questions raised here, but preliminary, my initial findings indicate that the sheer force of numbers – or, the strength of weak commitment – must be taken into consideration. This insight is valuable for researchers and policy makers who would like not only to see an increase in young people’s civic engagement, but would also like to better understand how young people can be involved in politics in the future. References Torset, N.S. and A. Færaas (2015, 15 September). Slik stemte Norge. Aftenposten. Retrieved from http://www.aftenposten.no/nyheter/iriks/politikk/Slik-stemte-Norge-8162898.html Vromen, A., Michael a. Xenos & Brian Loader (2015). Young people, social media and connective action: 41 from organisational maintainance to everyday political talk, Journal of Youth Studies, 18:1, 80-100 7. P olitics , P ublics , and Aylan Kurdi L ucy M ayblin On 1st and 2nd September 2015 I was at an academic workshop in London where experts on asylum and immigration from around the world came together to discuss key issues in the field. The unfolding refugee crisis was obviously a central topic of discussion and German academics described how what they were seeing in Germany was astonishing and unprecedented, at least in recent memory. After decades of anti-immigrant sentiment German people from all walks of life were opening their doors to Syrian refugees, greeting them at train stations, donating their belongings, volunteering, campaigning for more state support, more aid. Nothing like this was happening in the UK, they observed, nothing would break through the strength of hostility to asylum seekers and refugees that existed in Britain. Then, on the evening of 2nd September 2015 a picture appeared on my Twitter feed, which catapulted Europe’s most recent refugee crisis from another news story to an issue of global moral concern. Though the crisis had been building for several years, deaths in the Mediterranean had been intensifying for months, and large numbers of largely Syrian refugees had been arriving in Europe since the spring, it was this image that apparently took the issue, especially in the UK, from mere news item to issue of personal political concern. Suddenly, everything had changed. The speed and intensity of the civil society response was intense. The Charities Aid Foundation (CAF) claimed that one in three Britons had in some way contributed to the relief effort. This involved such activities as donating money, dropping off goods at local collection points, volunteering, or offering to house displaced Syrians. One third of those surveyed said that they were moved to act after seeing the image of Aylan Kurdi. Groups collecting clothing and other essentials for migrants living in the Calais Jungle camp were inundated and ran out of storage space. There were online pleas for people to stop going themselves and instead go to drop off points so that NGOs could sort donations first. Public figures such as Yvette Cooper MP, chair of Labour’s Refugee Taskforce, commented that “the outpouring of sympathy and generosity from the British public towards refugees fleeing conflict, violence and persecution has been incredible”. Stephen Hale, the chief executive of Refugee Action, said it was “wonderful to see this outpouring of support for welcoming refugees to the UK”. I had several conversations with NGOs in my home city of Sheffield where they expressed amazement at the sudden interest in asylum seekers and refugees expressed by citizens – a level of interest that they did not always have the capacity to harness. While these activities are undoubtedly positive actions of human compassion within a wider public and political discourse that is often dehumanising and morally suspect, the response has not always been straightforwardly positive. What is particularly curious is the external focus of many grass roots responses. This is especially the case in the UK, which is very difficult for asylum seekers to reach, and which has resisted all but a token quota of refugees –none from the European ‘crisis zone’. While numerous newspaper articles and Facebook groups, such as ‘Refugees Welcome UK’ have identified activities such as donating time or money to local asylum and refugee charities as immediate actions that people can do, there has been a tendency to focus on two extra-territorial contexts: the Syrian crisis in central, Eastern and Southern Europe, and the migrant camp at Calais. For example, on the 16th of September this message was posted on the Refugees Welcome UK Facebook page: ‘Dear all, I am really lucky as a UK citizen, because I live in Vienna, and it takes me 15 minutes to walk in either direction and I can help refugees immediately. I really wish I could find a good way of helping those living in the UK to help the people that are just out of their reach’. But of course, refugees and asylum seekers are not out of reach, there are thousands of them living in the UK. Most cities have destitute failed asylum seekers living in them, some rendered stateless, all in a purgatory of mere existence and absolutely dependent on charitable hand-outs to survive. When the University of East London pledged to provide ten scholarships for refugees to study at the institution, the scholarships were aimed at Syrians only, not others who already live in the UK. There is a class politics of visibility and deservingness to this, which is difficult to fully unpick while we are still in the midst of the crisis. At the same time as the Hungarian Prime Minister announces that his country cannot take Syrians because they are Muslim, this effort at othering is diminished in Western Europe by the fact that most of the Syrians arriving in the EU are clearly middle class – well dressed, articulate, educated. The quality of Aylan Kurdi’s trainers has not gone unnoticed. 42 Often, the most helpful actions are mundane, boring and long term. They are things like setting up a direct debit to an asylum support organisation, joining a local anti-destitution or anti-deportation group, lobbying your MP, volunteering to sort clothes in a charity shop round the corner from your house. It is incredible that people are wanting to do things to help refugees, but the key question is how this energy might be sustained, and how it might be converted into both mundane, often boring, actions closer to home, and then pressure for high level political change. There are therefore two questions to ask about the political effects of this image. The first is: has the energy created in the wake of the image changed public opinion on asylum seekers and refugees in the long term? The second is: has it had any impact on policy - will the energy created translate into something tangibly beneficial for asylum seekers? Because however many vans of toiletries and clothes are sent to the Jungle camp at Calais, what is really needed to help those living in a squalid refugee camp in the heart of one of the wealthiest continents in the world is a political commitment to help. The camp exists because it is allowed to exist. It has emerged from within, not outside of, the European asylum regime. So, for me, the second question is pivotal. In the UK the situation of asylum seekers has been worsening since 1993. Successive rafts of legislation have made it increasingly difficult to apply for asylum or to get refugee status at the same time as making it illegal for asylum seekers to work, limiting the amount of financial support they are given, introducing enforced dispersal, reporting to immigration authorities, and indefinite detention. In short, it is extremely difficult to reach the UK and exercise your human right to claim asylum, but if you are lucky enough to succeed you will be permitted to exist (physically, not politically or economically) and little more. For those who research British asylum policy, the picture truly is dystopian. And yet, for decades the public have supported increased sanctions on asylum seekers and MPs have passed legislation that has delivered. What this moment of sudden sympathy for asylum seekers and refugees offers is both an education in the reality of the cruel and punitive asylum system that we have allowed to develop, and the chance to demand change. Which brings me back to my second question. Has the response to the photograph of Aylan Kurdi brought policy change? Superficially, yes: on the 4th of September David Cameron announced that the UK would take 20,000 refugees over 5 years –not from Europe but from camps in Syria, Turkey, Jordan and the Lebanon- and would offer £100 million in humanitarian aid for those in the camps. Yvette Cooper said there was “no doubt” public pressure was to thank for forcing David Cameron to change his position. She, and others, went on to urge the government to go further, but to no avail. Just weeks later the Home Secretary, Theresa May gave a speech on immigration in which she argued that immigration of all kinds must be reduced and suggested that migrants in Calais are the “wealthiest, fittest and strongest” and are therefore undeserving of sympathy. Days later, a new Immigration Bill began to make its way through parliament. This bill aims to do such things as make refused asylum seekers absolutely destitute with no recourse to public funds of any kind, and to legalise the separation of families so that parents can be deported and their children put into care in the UK. Yet contestation of this new legislation – boring, hard to follow and understand, right here in rainy Britain and not at all exotic - is not being rolled into ‘refugees welcome’ on a large scale. Whether the current moment will therefore give rise to a truly transformative politics which alters the fundamental conditions under which access to the right to asylum is offered, is uncertain. 43 8. #K iyiyaVuranInsanlik: Unpacking A rtistic Responses to the Aylan K urdi Images H olly R yan On Thursday 3rd September, the North American web-based news and entertainment company Buzzfeed published a series of artistic responses to the photograph of 3-year old Syrian refugee, Aylan Kurdi, washed up on a Turkish beach. Seventeen artistic responses were published, with iterations in English, Spanish and Portuguese, whilst one further image was republished in the French version. Around the same time, Bored Panda, which describes itself as an online platform for ‘artists, designers and enthusiasts to create highest level editorial stories that drive millions of people to see their works and ideas’ opened a tribute page. For two weeks, readers were invited to respond directly to the tragic image of Aylan by submitting their own compositions. The webpage closed with 97 submissions, ranked according to popularity, and - as of October 20th - has achieved some 215,000 views. Shared widely across Twitter, the Bored Panda and Buzzfeed pages spawned an even wider web of paintings, illustrations, photographic manipulations and other graphic rejoinders. The written commentary accompanying the Bored Panda tribute page was short and the message direct: ‘These touching responses range from grief to rage, and regardless of where you stand on the Syrian refugee crisis and Europe’s response, one thing is certain – children like Aylan and [his brother] Ghalib should not be dying like this’. Yet, this simple framing obscures a number of more complex questions, namely: why did artists (or would-be artists) feel compelled to respond in the first place? What does a pictorial response offer that words alone cannot? Moreover, what do these artistic expressions achieve when shared online? This brief article attempts to shed some light on these questions. Over the course of history, there has been on-going debate about what constitutes art and what functions - if any - art should serve. Some hold the view that only the ensemble of institutions that make up the artworld – museums, galleries, universities and so on - have the power and authority to confer the status of arthood on an object. Others hold the view that an artwork may only be defined as such if it serves a purpose such as expressing emotion, imitating nature or celebrating status (Belfiore and Bennett 2008). In 1947, Georges Braque suggested that ‘art is meant to disturb’ and there is certainly a strong lineage of interventions that have tested the boundaries of existing cultures, norms and political practices. But when it comes to the artistic responses to the photograph of Aylan Kurdi, there is something else going on. Rather than testing art’s ability to disturb, these individuals seem to be using art as a means of dealing with the disturbing. Raymond Drainville, in the following chapter in this report, discusses the iconography of the photograph, and hints at some of the reasons that it unsettled audiences so, eliciting responses ranging ‘from grief to rage’. The notion of a therapeutic effect in art production was first articulated in the work of Aristotle (See Poetics), who placed value in the artist’s ability to both bring about and facilitate a purgation of negative feelings such as ‘pity and fear’. For Aristotle, this purgation constituted an integral part of the tragedy genre, by supplying a catharsis and relief from the most disquieting of emotions. Fast forward by a couple of millennia and we can see some similar arguments today in the work of psychologists and trauma therapists who have referred to creativity as a possible ‘resilience factor’. Post Traumatic Stress Disorder expert, Athena Phillips (2012) for example suggests that, ‘[b]arriers in discussing dark, shameful, or disturbing information can be softened with creativity; in fact it seems to be a conduit for undigested [or indigestible] material’. Meanwhile, Saveet Talwar’s research into the role of the arts in psychotherapy suggest that expressive modalities such as art, dance, music, poetry and drama are powerful tools for calling forth preverbal responses – the things that we feel, but find hard to express adequately in words (Talwar 2007). One artist, Peter de Wiel, [Bored Panda submission, ranked #46] seems to put it exactly this way, using multiple repetitions of the phrase “ik heb er geen woorden voor” (I have no words for this) to render a figure that recalls the body of Aylan, lying face down on the shoreline. Another quite common practice was the use of art to reject the actuality of death and instead project alternative, happier endings to Aylan’s story. Amongst the most popular image submissions were those that depict the toddler ‘sleeping’ in/on his bed 44 (See #2 ‘How this story should have ended’, submitted by Steve Dennis), or happy at play. In submission #94: ‘I[t] should end this way’, by El Vigia de Ponce, young Aylan dribbles a football toward the viewer. He is surrounded by characters from the popular children’s comic Peanuts. Both Phillips’ and Talwar’s comments suggest that producing artistic responses can provide us with insights into our own feelings about the drowning of infants at Europe’s borders. So, not only are these artworks an example of creative resilience, they also play a role in generating forms of knowledge and understanding that are at once both personal and political. Roland Bleiker (2009:12) summarises this well. He states that, ‘while art cannot tell us how to stop wars or prevent terrorism and genocide, it can give us insights into these experiences and the feelings we have about them. In so doing, art can shape the way we understand and remember past events and, in consequence, how we set ourselves the challenges we face in the future’. Moreover, as Jen Webb (2008) writes, ‘pure silence, or pure unmediated experience is available only to the dead, who no longer have the capacity to see, hear or feel anything, or to make sense of their environment.’ For the rest of us, death - including Aylan Kurdi’s - is always mediated by practices of representation, which in turn become productive of what we know. Twitter data pulled from social data intelligence platform Pulsar suggests that #KiyiyaVuranInsanlik (Humanity Washed Ashore) tweets containing visuals spread much faster than text-only messages. It also shows that the Bored Panda and Buzzfeed tribute pages and individual images were widely shared and influential. For these reasons, we should be particularly attentive to the nature and content of the circulating representations. Where do they take us? What political discourses and policy responses do they (wittingly or unwittingly) constitute and amplify? Particularly prevalent among the artistic responses were images that juxtaposed Aylan’s lifeless body with those of living political leaders – the people specifically tasked with protecting civilians and representing their interests. In the submission that ranked in at #1, ‘Do You See It Now?’, Aylan’s corpse - cut out and magnified - is displayed right at the centre of an Arab League meeting. Umm Talha’s submission, ‘The Leaders Are Watching’, ranked #8, depicts a group of figures clad in the traditional Arabian thawb. They stand by with spades, ready to bury the toddler. Can Doyum’s submission ‘Proud Ride’ ranked #33) which transposes another controversial and widely recognized image – that of world leaders marching together in the wake of the Charlie Hebdo attacks – onto an inflatable raft, out at sea. There is inevitably a lot going on in these artistic responses that is challenging, if not problematic, to try and decode. However, what does seem clear is that these pictures charge political leaders with passivity. In so doing, they build momentum for a policy response. Yet, as the UK government considers dropping bombs on ISIS targets in Syria, I have to wonder if this is quite the policy response that was hoped for. To borrow some words from the British Comedian, Frankie Boyle: are we to be convinced that somehow, the plight of this strategically significant state has touched the hearts of British political leaders? After all, Britain is so concerned about refugees that it will do anything – except, that is, take in refugees... References Aristotle (350 B.C.E). Poetics. [Translated by S. H. Butcher] Retrieved from http://classics.mit.edu/Aristotle/poetics.html Belfiore, E., & Bennett, O. (2008). The Social Impact of the Arts: An Intellectual History. London: Palgrave Macmillan. Bleiker, R. (2009). Aesthetics and World Politics. London: Palgrave Macmillan. Boyle, F. (2015, October 20). Britain clings to its bombing addiction with the weary rationale of a junkie. The Guardian. Comment is Free. Retrieved from http://www.theguardian.com/commentisfree/2015/oct/20/ britain-clings-bombing-addiction-weary-rationale-junkie-frankie-boyle Phillips, A. (2012, March 7). Art and Trauma: Creativity As a Resiliency Factor. Good Therapy. Retrieved from http://www.goodtherapy.org/blog/art-and-trauma-creativity-as-a-resiliency-factor-0307124 Talwar, S. (2007). Accessing traumatic memory through art making: An art therapy trauma protocol (ATTP). The Arts in Psychotherapy, 34(1), 22-35. Webb, J. (2008). Understanding Representation. London: Sage. 45 PART III THE ICONIC IMAGE ON SOCIAL MEDIA An image is not just a tool that allows us to see something it is pointing at. Images teach and train us to see, while providing cognitive frameworks that help us interpret what we are seeing. What we become able to see depends on many factors, not the least of them our visual cultural traditions. The images of Aylan Kurdi became iconic, acquiring a capacity to be seen by audiences that had previously been blind to the refugee crisis. The authors of this section, Ray Drainville, Jim Aulich and Simon Faulkner, unpack the iconicity of these images. They all agree that the photographs of Aylan’s body being carried by the policeman as well as lying on the shore conform to the Christian tradition of depicting sleeping angels and aestheticized death in the iconography of Pieta. Evoking, furthermore, photojournalism and war photography, the authors argue that the images include the notions of failed protection and withdrawn care, calling for a response by those currently in power. Montages that include only the cut out image of the body help spread the image as it acquires rhetorical power of transposing culpability and producing guilt in different contexts. 46 9. On the I conology of Aylan Kurdi , A lone R ay D rainville Over a century ago, Aby Warburg called for ‘an extension of the methodological borders of our study of art, in both material and spatial terms,’ one that could ‘range freely, with no fear of border guards’ and ‘cast light on great and universal evolutionary processes in all their interconnectedness’ (Warburg, 1999: 585). The scope of this plea can be extended beyond to the examination of pictorial motifs in general, including photojournalism and social media imagery. In all of these, the reproduction of motifs and concepts provide evidence of people mentally processing powerful images. Given the range of source material displayed in the Mnemosyne Atlas and his enthusiastic use of cutting-edge technology, I think that Warburg would approve of examining the images people share on the Internet (Michaud, 2004). A particularly powerful image shared recently is that of the body of Aylan Kurdi, photographed by Nilüfer Demir. I will focus on the representation of Aylan alone, lying face down in the surf (Figure 1 and Figure 2), as it drew particularly strong response from social media. Two factors regarding this photo matter particularly: the position of the body and its physical condition. He does not look dead. He had been in the sea for only a few hours, and escaped the indignities to which the drowned body is exposed. His body is not bloated; there are no visible wounds; his skin is not peeling off; sea scavengers have not violated him. This is important: by his position, it is easy to see the image as that of a sleeping child. Most parents would recognise the awkward sleeping positions of children at this age. I recall encountering my own son sleeping like this, the macabre thought “he looks like he’s dead” occurring, unbidden, to me. This is the ‘sleep of death’ in Hamlet, where the one is viewed as unnervingly similar to the other. In contrast, the photo of Aylan’s brother Ghalib (Figure 3) does not offer the viewer an opportunity for a ‘safe’ projection of sleep. His right leg is bent at an unnatural angle. His shirt is pulled up strangely high. It is abundantly clear that something is terribly wrong. We cannot project any happy, protective fantasies on him. The pictures of Aylan and Ghalib show the alternating aspects of Barthes’ punctus: Aylan, whose code-switching of sleep/death can break through the studium and touch the viewer; and Ghalib, whose image is so raw and uncompromising that it can drive the viewer to madness (Barthes, 1981). When an image with variations - such as those taken by photojournalists - becomes so popular, it is worth asking: why did this image become popular, and not that image? Social media users can provide us with clues. Images are often joined by comments or hashtags; and people may share modified versions of the original image, all of which can provide clues to their interpretations. In Figure 4 - attributed to Azzam Daaboul - we see Aylan tucked under a blanket and pillow, in an attempt to normalise the grim image. Daaboul left the surrounding context still recognisable: the surf is visible, but has been blurred, its impact lessened by the use of a spotlight filter to sharpen attention on Aylan, and away from it. It is as if Azzam is trying to deny the reality of the situation, and indeed, the tweet accompanying the picture confirms this, highlighting that Aylan is ‘sleeping’, that he is ‘in heaven and safe’. Daadoul created additional images mobilising this idea, see for example his inclusion in this list on Buzzfeed (number 6). The effect is more pronounced in Figure 5 — originally attributed to Steve Dennis by Buzzfeed, but more plausibly attributed to Omer Tosun by Huffington Post Canada. Save for the body, Tosun has thoroughly domesticated the scene: and the text accompanying the image notes that the author dreams of ‘how his story should have ended’. The surf that approached the child’s face in the photograph has been turned into a pillow. The glass of water set by the bedside is rendered considerably brighter than the rest of the scene and thus beckons for the viewer’s attention, creating an ad hoc symbol that obliquely references the manner of Aylan’s death. The association of sleep with death may be fairly stated as a common interpretation of Nilüfer’s images. But why did the images of Aylan capture the European imagination so? I think because it is a photogenic and cleansed image of death: indeed, by being removed from any larger context, aestheticised. This is important in a Western cultural context. For the past two millennia, the most common image by far has been that of the aestheticised corpse of Christ, whose body has been depicted overwhelmingly with few of the abuses subject to it during the Passion (Figure 6). 47 It should be made clear: I do not think one should see in Aylan a reference to Christ, although the extreme oblique angle in Figure 1 superficially recalls Mantegna’s Lamentation over the Dead Christ (Figure 7). It is clear from some images that, if anything, people view Aylan more as a sleeping angel (Figure 8), of which there are several variations. The modified versions recognise a power inherent in the original image, many of which further aestheticise it. Figure 9 is evidence that the image of Aylan is turning into its own icon. So Aylan is not Christ; both are aestheticised corpses. And we have been primed to view such images in this manner. We are surrounded by a tradition of images, twenty centuries long, depicting aestheticised death. They are familiar to everyone in Europe, not only for religious people or those who visit cathedrals and art collections, but also for those who merely walk past a church or graveyard, or whose pious relatives keep Christian images on the walls of their homes. Edgar Wind said: ‘every act of seeing is conditioned by our circumstances’ (Wind, 2009: 191); those circumstances include the repeated forms that surround us. We do not need to accept the conscious ideologies of such images in order to absorb the cultural connections I have suggested above. We absorb images daily, unconsciously, and they prime us to navigate whole classes of images in prescribed ways. Warburg considered this a ‘social memory’ of images (Gombrich, 1970: 242ff); we might today call it a ‘visual habitus’, following the work of Pierre Bourdieu (Bourdieu, 1984). They may function beyond traditional iconographic categorisation or the encompassing narrative schemes posited by Jan Białostocki (Białostocki, 1981), although they naturally include them as well: they constitute a potent mixture of discrete symbols, meaningful poses, and repeated themes. The photographs of Aylan Kurdi have literally changed the terms of the debate about refugees (D’Orazio, 2015). Thanks to the power of these images, politicians from the UK, Germany, and other countries have been roused from indolence to their desperate plight (Wintour, 2015; Smith and Tran, 2015). It is another tribute to the power of virality in social media to direct social change: disparate, loosely networked groups of people highlighted the human side of this international catastrophe to such an extent that political leaders could not longer ignore the situation (Nahon and Hemsley, 2013: 108–111). That tribute seems short-lived. Since the initial outcry, the British government has announced a scaling-back of relief efforts (Dearden, 2015). Germany appears on the brink of refusing more refugees (Paterson, 2015). This volte-face amounts to a cynical bet that we will become distracted enough to forget them, passing on to other subjects like the Prime Minister’s university misadventures (Mason and Phillips, 2015; Rentoul, 2015). Thus, the images of Aylan Kurdi pose a challenge for Europe and the long-term effects of social media as a medium for social change. Will the images return to prick the consciences of European leaders and citizens? Or do they constitute another example of a viral phenomenon that spikes, and then disappears, leaving little permanent change in its wake? References Barthes, R. (1981). Camera Lucida: Reflections on Photography. London: Cape. Białostocki, J. (1981). Stil und Ikonographie: Studien zur Kunstwissenschaft. Köln: DuMont. Bourdieu, P. (1984). Distinction: A Social Critique of the Judgement of Taste. Cambridge, Mass.: Harvard University Press. D’Orazio, F. (2015, September 30) SnapChart: How a single image changed the debate on immigration #KiyiyaVuranInsanlik. Retrieved from http://www.pulsarplatform.com/blog/2015/how-a-single-imagechanged-the-debate-on-immigration-kiyiyavuraninsanlik/ Dearden, L. (2015, October 8). UK Withdraws Last Ships Dedicated to Rescuing Refugees After 15 Children Drown in the Aegean Sea. The Independent. Retrieved from http://www.independent.co.uk/news/uk/home-news/ uk-withdraws-last-ships-dedicated-to-rescuing-refugees-after-15-children-drown-in-the-aegeansea-a6685066.html 48 Gombrich, E. H. (1970). Aby Warburg: An Intellectual Biography. [S.l.]: Warburg Institute. Hariman, R. and Lucaites, J.-L. (2007). No Caption Needed: Iconic Photographs Public Culture, and Liberal Democracy. Chicago: University of Chicago Press. Mason, R., & Phillips, T. (2015, September 21). Cameron Biography: Ashcroft Makes New Debauchery Claims About Student Days. The Guardian. Retrieved from http://www.theguardian.com/politics/2015/sep/21/davidcameron-pig-head-lord-ashcroft-unofficial-biography-drug-taking-claims Marrow, J. H. (1979). Passion Iconography in Northern European Art of the Late Middle Ages and Early Renaissance: A Study of the Transformation of Sacred Metaphor into Descriptive Narrative. Kortrijk: Ghemmert. Merback, M. B. (1999). The Thief, the Cross, and the Wheel: Pain and the Spectacle of Punishment in Medieval and Renaissance Europe. London: Reaktion. Michaud, P.-A. (2004). Aby Warburg and the Image in Motion. New York: Zone Books. Nahon, K. and Hemsley, J. (2013). Going Viral. Cambridge, UK: Polity Press Cambridge Paterson, T. (2015, October 7). Refugee Crisis: Germany’s ‘Welcome Culture’ Fades as Thousands Continue to Arrive. The Independent. Retrieved from http://www.independent.co.uk/news/world/europe/refugee-crisisgermanys-welcome-culture-fades-as-thousands-continue-to-arrive-a6685361.html Rentoul, J. (2015, October 10). Refugee Crisis: Surge of Sympathy for Refugees in Britain ‘Likely to be Fleeting,’ Study Says. The Independent. Retrieved from http://www.independent.co.uk/news/uk/home-news/refugeecrisis-surge-of-sympathy-for-refugees-in-britain-likely-to-be-fleeting-study-says-a6689316.html Sashin, D. (2015, September 4). Artists Depict Boy’s Body on the Beach. CNN. Retrieved from http://edition.cnn.com/2015/09/03/world/syrian-boy-artist-tributes/index.html?sr=tw090415toddlerdrowncart oon1130aStoryGalPhoto Smith, H., & Tran, M. (2015, September 8). Germany Says it Could Take 500,000 Refugees a Year. The Guardian. Retrieved from http://www.theguardian.com/world/2015/sep/08/germany-500000-refugees-a-year-clasheslesbos Warburg, A. (1999). ‘Italian Art and International Astrology in the Palazzo Schifanoia, Ferrara.’ In The Renewal of Pagan Antiquity: Contributions to the Cultural History of the European Renaissance. Los Angeles, Calif.: Getty Research Institute for the History of Art and the Humanities; [Garsington : Windsor] [distributor], pp. 563-592. Wind, E. (2009). Warburg’s Concept of Kulturwissenschaft and its Meaning for Aesthetics.’In D. Preziosi (Ed.), The Art of Art History: A Critical Anthology. 2nd ed. (pp. 189-194). Oxford: Oxford University Press. Wintour, P. (2015, September 7). UK to Take up to 20,000 Syrian Refugees over Five Years, David Cameron Confirms. The Guardian. Retrieved from http://www.theguardian.com/world/2015/sep/07/uk-will-accept-up-to-20000syrian-refugees-david-cameron-confirms 49 10. The L ife of Images: The Iconography of the P hotograph of Aylan K urdi’s body and the Turkish P oliceman J im A ulich Aylan Kurdi’s body was photographed and filmed as it lay in the waves on the shore close to Bodrum in Turkey. Along with 12 others he had not survived the 5 km crossing from Turkey to the Greek island of Kos in the early hours of 2nd September 2015. Photographs of the consequences of this fatal voyage appeared on front pages across the world, on news websites and circulated on social media platforms (Figures 1-3, image collection 1). Such was the effect of the photographs that they stimulated its first Wikipedia entry on 12 September 2015 where the details of the incident and reactions of Heads of State from around the world were recorded. The photojournalist Nilüfer Demir took the photographs as she was waiting on the beach to document refugees as they came ashore (Rossington, 2015). The News and Views blog ‘Video of dead bodies of Syrian Refugees at Turkish Shore’ shows that they document the moments when a Turkish Police Crime Scene Investigator lifted the child from the shoreline and carried him across the beach. Simultaneously, as he takes him to the shelter of some rocks and away from prying eyes, he turns his back slightly towards the direction of the cameras, in a gesture of protection and respect for the dead child. Aylan was three, his brother aged five and his mother also died, his father survived: this important familial dimension to the story was reported but not pictured. Another photograph from Anadolu Agency (caption: A Turkish gendarmerie soldier moves the body of a young migrant washed ashore after a boat sank off the coast of Bodrum), taken by a different photographer is from a 1300 word article published by CNBC headlined: ‘Dead child on Turkey beach a reminder of EU’s migrant crisis’ (Reuters, 2015). It made a singular appearance at the moment when the better-known pictures began to circulate. The article makes no reference to the drowned children. Because it depicts the same event, it gives rise to the question of why the photographs of the policeman carrying the body across the beach generated such interest. Arguably, their ‘success’ led to a particular understanding of the wider implications of the drownings and gave the body politic and the international community a firmer grasp of the refugee crisis that other images, such as this one, failed to do. An injured, dying or dead body straddled across the lap or cradled in the arms of someone offering succour, help, protection or refuge, or allowing the intimacy of grief, is a natural enough pose and one we find commonly pictured in the press (Figures 5-10 image collection 2). However, as a depiction it is more than mere nature, the simple record of a basic human act. Looking at the picture from a position informed by an understanding of the migratory life of images derived from the art historical scholarship of Aby Warburg (Lindberg, 1999), Erwin Panofsky (1970) and Edgar Wind (1969) the imagery carries a huge potential charge as it connects to a heritage in western life and culture. Essentially, it is a Pieta, originally coined in 11th century Byzantium the iconography developed in Northern Europe from about 1400 as a stage in the depiction of the Passion of Christ (Figure 11). As such we could tag it with the words: Family; Sacrifice; Martyrdom; Displacement; Flight; Survival; Protection; Resurrection; Redemption; Salvation; Pity; Mercy; Sorrow; Piety and ultimately Victory over Death. Crucially, the body is held forward by an authority figure (Mother of God, father, soldier, policeman, fireman). The body is held in a gesture of offering up, as if to bear witness and to implore for the intercession of an even higher authority. It is interesting to note that accounts of the event from non-Western sources favour the image of the body alone to signal a different visual tradition. The complex iconographical lineage shape-shifts in gender, age profile and context. Of course, not all photojournalism functions in this way, and many images, which have become iconic such as representations of the collapse of the Twin Towers do not, perhaps, because the denotation provides its own connotation independent of historical precedent in the visual tradition. One of the most potent iconographically charged examples is found in Don McCullin’s reportage from the Vietnam war published by The Sunday Times Magazine 24 March 1968 (Figure 8). In this ‘objective’ medium we can find a Pieta; Nativity; Flagellation; a ‘suffer the children to come unto me’; and a Deposition to impose a more familiar moral universe that undoes the moral inversions of war. 50 The photograph of Aylan Kurdi carried by the officer belongs to a complex typology of images. The suggestion is that they contribute to the making of the world in cognitive terms. They are not simply interpretations of facts but are constitutive of the world as we understand it. According to Lambros Malafouris (2013), human cognitive processes in the shape of material images extend beyond the internal mind to comprise the world we live in. They close the gap between mind and world and give the image its value. The picture does not simply reproduce an event within the ‘objective’ discourses and institutions of photojournalism, but draws upon and contributes to the process of cognition. So, ultimately, the imagery belongs to an established or recognisable typology or group of typologies. As Belting (2011:9) has remarked, ‘We live with images, we comprehend the world in images. And this living repertory of our internal images connects with the physical production of external pictures that we stage in the social realm.’ This has the effect of projecting human cognitive processes in the shape of the image into the material environment so that they are not an interpretation of a world, but a making of it (Malafouris, 2013:227). In other words, the role of the image and the photograph takes on significance beyond what Zelizer (2010) terms the imaginative and the subjunctive, driven by contingency and emotion, to construct the world it partly constitutes. The image constitutes the world on which reason acts: it is no mere representation. The Sunday Express clumsily recognized the power of the effect of this process in the article headed ‘Will heartbreaking image of dead Syrian child define our era like Vietnam napalm girl?’ (Virtue, 2015). The juxtaposition of Nick Ut’s photograph of Phan Thi Kim Phuc, Napalmed Children Flee Attack, June 8th, 1972, struggles to make the point, but only succeeds in remarking on its success as a stimulus for world opinion. As the pieta image travels through space and time so it is continually written over like some palimpsest, always carrying with it a residue of its previous existences. Memories of where readings were determined in their own times by institutional frames and the shifting sands of ideology persist. Every occurrence is linked intertextually and synchronically, ensuring that the image can never be definitive in meaning or prised from the baggage carried by the receiver. Barbie Zelizer has argued the same point within the narrower context of the self-referentiality of war photography where the appeal of the conventional has important implications for the body politic because of the tendency to forgo ‘reliable visual depictions’ and to use instead those that already rest in established narratives of nation, patriotism, heroism, sacrifice and redemption (Zelizer, 2004, 115-16). For western eyes the photograph under discussion takes part in the construction of a world where something might be done, that there might be life after death, and the guilt arising from a double sense of responsibility and impotence might be lifted. The reference to the Passion of Christ gives it the significance of a higher authority to obscure the political and military realities of the war in Syria and the ensuing refugee crisis. The image is not immaculately conceived and is created, reproduced and received within a complexity of institutional and ideological frames. The interpretation and reception of the photographs of Aylan Kurdi are dependent upon the babble of discourses encountered by the iconography of the pieta throughout its history in photojournalism and in devotional pictures and is therefore contingent on their meaning and cognitive functions in the world. References Belting, H. (2011). An Anthropology of Images. Picture – Medium – Body. Princeton & Oxford: Princeton University Press. Lindberg, S. (Ed.), (1999). Aby Warburg. The Renewal of Pagan Antiquity. Los Angeles: The Getty Research Publications Programs. Malafouris, L. (2013). How Things Shape the Mind. A Theory of Material Engagement. Cambridge, Massachusets & London, England: The MIT Press. Panofsky, E. (1970). Meaning in the Visual Arts. Harmondsworth: Peregrine. 51 Wind, E., (1969). Giorgione’s Tempesta with Comments on Giorgione’s Poetic Allegories. Oxford: Oxford University Press. Reuters. (2015, September 3). Dead child on Turkey beach a reminder of EU’s migrant crisis. Retrieved from http://www.cnbc.com/2015/09/03/12-migrants-drown-off-turkey-others-halt-eurostar-in-eu-crisis.html Rossington, B. (2015, September 4). Photographer who took picture of drowned toddler Aylan Kurdi says she had to “make this tragedy heard”. Daily Mirror. Retrieved from http://www.mirror.co.uk/news/world-news/photographer-who-took-picture-drowned-6380991 Unknown. (2015, September 2). Video of dead bodies of Syrian Refugees at Turkish Shore. Retrieved from http://newsmyviews.blogspot.co.uk/2015/09/video-of-dead-bodies-of-syrian-refugees-child-red-t-shirtturkey-turkish-shore.html Virtue, B. (2015, September 4). Will heartbreaking image of dead Syrian child define our era like Vietnam napalm girl?’ Sunday Express. Retrieved from http://www.express.co.uk/news/uk/602720/migrant-crisis-dead-Syrian-boy-Napalm-Girl Zelizer, B. (2004). When war is reduced to a photograph. In S. Allen & B. Zelizer (Eds.), Reporting War. Journalism in Wartime (pp.115-135). London & New York: Routledge. Zelizer, B. (2010). About to Die. How Images Move the Public. New York: Oxford University Press Inc. 52 11. Aylan Kurdi and the Movability of Images S imon Faulkner This article will address images of Aylan Kurdi’s dead body lying on Bodrum beach in relation to the ways that these images were subject to movement. [1] The concern here will be with how the circulation of these images can be considered from the perspective of the movability of images in general. The assumption underpinning this approach is that it is the movement of images that gives them a good deal of their affective power. Thought about in the most obvious way, the movability of images relates to their status as portable physical artefacts that can be moved from one geographical location to another. Most images, frescos painted as part of the architecture of buildings aside, can be moved in this way. The circulation of images on the Internet also involves their translocation from one physical place to another. This is simply because people who upload or observe these images are necessarily situated somewhere in offline space, often at a distance from each other. Some social media platforms, such as Twitter, are also designed to enable the extended virtual movement of images through acts of retweeting that also, necessarily, entail the further movement of these images between points in physical space. Thought about in a different way, the movability of images also refers to the way that images can be transferred from one medium to another. Although it is generally not possible to physically move a fresco painted in tempera on plaster in a Florentine church, it is possible to transform such an image into photographic images that can then be moved round (Berger, 1972). Press photographs, such as Joe Rosenthal’s picture of the raising of the flag on Iwo Jima, can be translated into posters, jigsaw puzzles, cartoons, paintings, cinematic scenes, and so on (Hariman and Lucaites, 2007). In that such processes of remediation involve the further distribution of the iconic content of images, they also involve the movement of images through space. But it is not only whole images that can be moved from medium to medium. Visual motifs that have their origin in particular images can be transcribed across mediums and reused in different pictorial contexts for quite different purposes. This movement of motifs can occur over extended periods of time and large geographical distances through processes that Rudolf Wittkower defined as the ‘migration of symbols’ (Wittkower, 1977). Contemporary digital technologies allow for an accelerated version of this migration of motifs from one medium and pictorial context to another. What does this initial discussion of the movability of images mean for the consideration of the photographic images of the body of Aylan Kurdi on Bodrum beach? A simple response to this question is to state that these images were subject to all the forms of movement identified in the preceding paragraphs. The images were literally moved from the photographer’s camera to a variety of mainstream and social media destinations. The images were also remediated in multiple ways as drawings, cartoons, graffiti-murals, sand-sculptures, and direct-action-style performances. All of these remediations of the original photographic images were themselves circulated on social media extending the reach and versatility of the image in terms of meaning. Like most iconic images, the photographs of Aylan’s body on the beach are relatively simple in terms of their formal content, constituted as they are by the body lying at the meeting point between the sea and the sand. This visual simplicity lent itself to an easy legibility that appears to have been a crucial motivating factor for those people who contributed to the circulation of the images. These images were subject to further simplification through the removal of Aylan’s body from its original setting. The body (with its easily recognisable clothing of red t-shirt, blue shorts, and plimsolls) became a movable icon that could be repurposed in line with quite different concerns. In this sense, the movability of the body as a visual motif was conjoined with its metaphoric versatility. Much commented upon has been the re-contextualisation of the body from the beach to a child’s bedroom. But the movement of the motif of the body went much further than this, resulting in the generation of a whole range of unexpected associations often involving the establishment of a relationship of moral responsibility between the dead child and some other party. Such relationships of responsibility were generally set up as some form of culpability for the death of the child. However this culpability was not always the same. Sometimes it involved the insinuation of responsibility for the conditions that caused the Kurdi family to become refugees in the first place and by extension for Aylan’s 53 death. For example, in one image his body was montaged into a photographic scene of the Assad family sitting at a dinner table. Members of the family appear to be blowing out candles on a birthday cake, which has been replaced by Aylan’s corpse. This montage was intended to affect a shocking conjunction between a family celebration and the harsh political reality that had resulted in the death of a child. In this sense, its effect was the opposite of other relocations of the body to domestic settings in ways that were apparently meant to assuage the upsetting vision of the body on the beach in the original photographs. In other instances, parties were identified as culpable for not acting to prevent Aylan’s death. One cartoon circulated on Twitter involved the relocation of his body to the centre of a large round table situated in a room decorated with the European Union emblem. Around this table sit a number of EU officials who are facing away from the body. The implication is straightforward: the officials have failed to respond to the refugee crisis that caused Aylan’s death and are now refusing to see his corpse. This format was repeated in a montaged image that involved the relocation of the body to the United Nations Security Council chamber. Although this time the council members face the body, they appear to not see its blatant and shocking presence on the floor. The theme of these images is not only culpability, but also the apparent contradiction between the high visibility of the image of Aylan’s body enabled by its mass circulation and the indifference of official bodies. The presence of Aylan’s body has a particularly haunting quality in the image of the UN Security Council chamber in that the council members are shown going about their normal business of voting in the presence of harrowing death. But it is not only the EU and the UN who were identified as culpable through the visual relocation of the body to their meetings. Arab leaders and the Gulf States in particular were also found to be failing in the face of the Syrian refugee crisis. Thus Aylan’s body was relocated in composite images featuring the luxurious skylines of Qatar and Dubai. The body was also montaged into the middle of the Arab League meeting chamber and drawn in front of a set of Arab leaders who appear to have dug a grave for the boy. All of these images function through a form of guilt by visual association and attest to the transposable rhetorical power of the motif of Aylan’s body. There are also a number of images that involve the repositioning of Aylan’s body into the arms of another figure, mimicking the equally iconic pieta-style photograph of the body held by a Turkish police officer on the beach. For example, one tweeted cartoon depicted the body in the arms of a weeping crocodile that sits next to a wall topped with barbed wire, suggesting hypocrisy on the part of those who feign concern for Aylan’s death, while blocking the movement of other refugees. In another example, the body has been montaged into the arms of Saddam Hussein. The suggestion here being that Aylan has joined and is now being protected by Saddam in heaven. From the perspective of someone like myself, who was encouraged for decades to view Saddam as an evil figure, this is a striking conjunction that was obviously generated in a context within which Saddam was viewed in a more positive light. Whatever one might think of this re-contextualisation of the motif of Aylan’s body, it attests to the sense that this motif and the original images rapidly became utterly nomadic as things that could be reframed in relation to very different cultural and political imaginaries. As such they presented a particularly strong example of the complex movability of images that has been a crucial aspect of their cultural functions for millennia, but which in its new digitally driven intensity is all the more urgent to describe and understand. This movability is crucial to understand, not only because the images of Aylan Kurdi were widely and multiply circulated, and through this movement were encountered by many diverse spectators, but also because this very movement contributed significantly to the affective force of these images. In this sense, the force of the images is not just to do with their iconic content, but also with how their affective power accrued through their circulation. Simply, the more they circulated the more they were loaded, as iconic images, with a performative force that became the condition of their further reproduction, distribution, and adaptation. 54 F ootnotes [1] The images in question involve three photographs taken by Nilüfer Demir from different positions around the body. Two of these photographs involve close-ups of the body, while the other includes more of the beach as well as two Turkish police officers next to the body and two fishermen in the background. These images were cropped in different ways when used. References Berger, J. (1972). Ways of Seeing. London: Penguin Books Ltd. Hariman, J., & Lucaites, L. (2007). No Caption Needed: Iconic Photographs, Public Culture, and Liberal Democracy. Chicago and London: The University of Chicago Press. Wittkower, R. (1977). Allegory and the Migration of Symbols. New York: Thames and Hudson. 55 PART IV SHOWING / NOT SHOWING THE IMAGE Social media create unregulated spaces in which the decisions on what is to be published and seen are partially about the mere technical capacity to share images and the ease with which they are encountered when auto-play mode is on. Such images are often graphic images of disasters and deaths that can’t be published in mainstream media. However, apart from technical allowances and regulations, the decisions to share images on social media also become individual publishing decisions, with ordinary people becoming publishers. Lisa Procter and Dylan Yamada-Rice, Sam Gregory, Claire Wardle and Evelyn Ruppert and Funda Ustek focus on the attempts to construct forms of visibility, accountability and affective engagement. Procter and Yamada-Rice explore the single object from the photograph: the boy’s shoes as a symbol of the vulnerability of childhood, while both Gregory and Wardle discuss histories of engagement and nonengagement with the photographs of atrocities and the techno-social politics of depicting the refugee crisis. Ruppert and Ustek conclude the report by looking into practices of accountability, and specifically, understanding refugees as numbers, swarms, lines and ruptures. 56 12. S hoes of Childhood: E xploring the E motional P olitics Through Which Images Become N arrated on S ocial Media L isa P rocter and D ylan Yamada-R ice There is a long history of iconic graphic images of children in poverty, war and famine that have become accredited with changing widespread public opinion towards large-scale political events. The recent photograph of the refugee Aylan Kurdi lying dead at the edge of the sea is the most recent of such examples. In particular the dissemination of the photograph of Aylan on Twitter can be thought of as a kind of emotional politics (Ahmed, 2004). Specifically users’ emotions appear to have played an important role in contributing towards a collective political narrative that is displayed in the tweets they produced and shared. To unpack this notion further we first examine how some depictions of children have changed political history. This we argue occurs through the visual representation of material objects of childhood, which appear to be used to symbolise an idealised space of the child. The second part presents the findings of tweets that were analysed using Pulsar and the keywords ‘his shoes’. The findings show how Twitter users who mentioned Aylan’s shoes expressed narratives of helplessness and care in response to his image. Such narratives seem to use the materiality of his shoes to conjure up notions of childhood as an idealised period of time. In turn such discourse seems to have been used to express support for Syrian refugees more widely. P ictures that have changed history Throughout history certain images of childhood have intentionally or inadvertently been utilised to change public perception of political events. For example, the 1891 painting entitled ‘The Doctor’ by Luke Fildes depicts the image of a doctor looking over what the viewer assumes is a very sick child lying on two wooden chairs in a dark room. The image appears to depict the poverty in which the child lives that is symbolised by the lack of a bed. It is perhaps no coincidence then that in 1949 the image was used ‘by the American Medical Association in a campaign against a proposal for nationalised medical care. The image was used in posters and brochures along with the slogan, “Keep Politics Out of this Picture” implying that involvement of the government in medical care would negatively affect the quality of care. 65,000 posters of The Doctor were displayed, which helped to raise public scepticism for the nationalised health care campaign’ (Family Matters: the Family in British Art, Exhibition Catalogue). Another iconic image is a black and white photograph of a girl running naked with burns on her body as the result of a Napalm attack taken by Nike Ut in 1972. Like Fildes’ painting, this photograph has also been credited with changing the American public’s perception- this time of the Vietnam War. An exhibition shown at the Millennium Gallery, Sheffield in 2012 explored the representation of childhood in art. The exhibition catalogue Family Matters: the Family in British Art asks ‘How has the role of children in the family, and in the eyes of adults, changed?’ It states that we can find answers in artists’ depictions of children, which lets us trace changing ideas of childhood. However, it could be argued that the notion of the child as innocent and in need of protection is a discourse that transcends time. This can be seen in the work of the artist Gainsborough who was instrumental in creating images of children that showed them as archetypes of innocence (Higonnet, 1998). In academia, Alanen (1988:54) says childhood is often defined as a period of time that ‘separates children from adults and defines the ideal family as a nuclear unit consisting of protected children and protecting adults’. This is a notion that has been challenged by the artist/photographer Jo Spence whose work is critical of the notion of childhood as an idealised time. Spence recreates a family album to record events in childhood that often go un-photographed. For example Spence’s reconstructed photograph of her adult-self clutching a teddy bear and crying. Her work forces the viewer to question the privileging of some moments of childhood above others. 57 Additionally, ‘children’s games and toys have been used by artists as metaphors for the expectations, achievements and fallibility of adult life throughout the history of Western Art’ (Langmuir, 2006: preface). It is this aspect of the materiality of childhood that led us to decide what to focus on when asked to write about the colour photograph of the refugee child, Aylan Kurdi after he drowned when he and his family attempted to escape war. Undoubtedly like many readers of this paper we asked ourselves what was it about the image that brought about such an emotional response from so many people. A month prior to the writing of this report we were walking down a road when a friend that was with us started talking about a Guardian newspaper article about images that have changed the world (Jack, 2015). In response to this we reflected on why the image of Aylan has captivated so many people and we discussed why we felt this image was so easy to feel drawn into. Other writers in this report have talked about the ‘everyday-ness’ of the image of Aylan, the way he was laying and the resemblance of his position to a sleeping child. Another example of everyday childhood objects are Aylan’s shoes, which seemed like they could be a pair belonging to any young child known to us. The plainness of the shoes and the fact that you cannot see Aylan’s face are perhaps important features of the photograph. McCloud (1994) writes that when images are reduced to the most simplified form possible, the less it resembles something specific and becomes a more generalizable image for the viewer to look at and identify with it in relation to their own experiences and emotions. In other words, Aylan’s body, its position and clothing become symbolic of childhood in general. Indeed new materialism studies have argued that humans have a fundamental desire to account for the ‘“nonhuman” as well as “human” forms of agency’ (Barad, 2003:810). This reflects how viewers of the image can be as much affected by the agency of the shoes as the child and in doing so they become a signifier for the need for human improvement. Perhaps one of the big differences about the photograph of Aylan in comparison to other images described so far in this section, is its sharing on social media. The digital nature of the image has allowed it to be manipulated, shared socially and also commented on. Thus unlike the paintings described previously or the black and white photograph taken in Vietnam it is possible to find out first-hand viewers’ impressions and interpretations of the image. In relation to this the next section focuses on tweets that specifically made reference to Aylan’s shoes and the extent to which those tweets can be seen as a representation of idealised views of what childhood should be. Below we present a discussion of tweets that include the phrase “his shoes” to demonstrate the role of emotion in the production of socio-emotional and political narratives related to Aylan’s death. E xploring narratives of helplessness and care in relation to the photograph of Aylan One Twitter user notes that the specific detail they find so distressing, are ‘his little shoes’. Tweets such as this illustrate how Aylan’s shoes seem to have become symbolic of idealised notions of adult-child relations, where adults should provide care and protection to children (James, 2013). His shoes are representative of the people or person who placed the shoes on his feet. As another Twitter user notes that someone must have helped Aylan “tie on his shoes that morning”. Users connect with the everydayness of the act of helping a child tie on his/her shoes. Another users comments: “It’s those shoes”, again highlighting that someone put them “on his little feet and did the straps up that morning”. In these ways, Twitter users begin to collectively give life to a perceived relationship of care between Aylan and the significant adults in his life, adults who must have prayed that he would survive this journey. There is a sense that these Twitter users want to reach out and offer care and protection to children in similar situations to Aylan but do not have the capacity to do so. They reflect a sense of trust that the adults in these children’s lives want to offer this too. The shoes are symbolic of hope; users seem to see the shoes as demonstrative of the tender care that Aylan received on the morning of his death. Parents drew on Aylan’s shoes in their tweets to make connections to their own experiences of being a parent. As one parent tweeted that it could have been their son “in his brown shoes”. The shoes signify a tangible means through which parents emotionally connected with the image, for example one user notes that on that Wednesday morning “someone like you was helping Aylan put on his shoes”. The examples above show how users seem to see Aylan’s 58 shoes and were reminded of their own children’s lives. As another parent comments that every night they check on their own two-year old son, in bed, observing that sometimes he lay exactly like Aylan, but “without shoes”. This tweet also reflects other readings of the image as a boy sleeping, explored in earlier sections of this report. It is the shoes that make the interpretation of the photograph difficult for this user. When parents put their own toddlers to bed they would take off their shoes. The shoes become reflective of the tensions presented in the image, which draw on particular notions of what childhood should and should not be. The shoes interfere with readings of a boy asleep: emphasising that because Aylan was wearing socks and little shoes he was ready “to try life again” and that seeing his cheek on the sand made them “choke”. Instead Aylan’s shoes symbolise the meeting of an idealised version of childhood with the reality of the atrocities faced by him and other children in a similar situation. Twitter users describe their sense of Aylan’s carers being put in a helpless situation, as reflected in sympathetic tweets showing concern for his family’s feelings. Users noting that seeing the image had “ended” them, because his parents put on his shoes “not realising it’s the last time”. Such tweets show how Aylan’s shoes become reflective of his relationship to adults. The tweets show how children’s lives have a tendency to be understood in terms of how they are networked within wider familial relationships (Rose, 1999). Aylan’s “little shoes” also become a marker of his social position, seemingly a visual symbol of his helplessness, his need for protection (James, Jenks and Prout, 2007), as a young child: his little shoes still on his feet. This sense of Aylan’s powerlessness enacts Twitter users to reflect on their own lives in relation to those seeking asylum, highlighting they can’t get the picture of this little boy out of their head, especially “his little shoes”, stressing that “we have no idea how lucky we are”. One user describes Aylan’s shoes as “tiny”, further evoking a sense of his fragility as a young child. In one tweet his shoes become symbolic of powerful feelings of sadness and ultimately inspiring calls for action to be taken to support others, as one user notes: they will never look at a pair of tiny shoes like Aylan’s “in the same way”. They describe seeing the shoes as “gut wrenching”, hoping that “his legacy be real action to stop these atrocities”. These narratives seem to begin to work to unsettle negative discourses about those seeking asylum and instead show how people experienced a stronger emotional connection and will to support those, especially children, escaping conflict. C onclusion This paper has explored the emotional meanings that Twitter users produced in response to photographs of Aylan on Twitter. We have explored the ways that emotional narratives are constructed by and between users in response to the photograph. To do so we drew specifically on the material agency of Aylan’s shoes. Tweets about Aylan’s shoes seemed to show how they became symbolic of his helplessness, due to his young age, and at the same time the care he received from parents or other significant adults in his life particularly adult-constructed politics that caused the war he was fleeing and the new life his family were seeking for him. These two narratives of helplessness and care intersected to reveal how idealised notions of childhood are reflected through Twitter users’ expressions of their emotional connectedness to Aylan’s death. A collective sense of Aylan’s helplessness seemed to strengthen users’ sympathy towards those adults who have and continue to care for him. These narratives, we argued, seem to destabilise negative discourses in relation those seeking asylum and instead show how people were able to show sorrow for those fleeing very difficult circumstances in their home countries. Through these examples we were able to show that although the image of Aylan draws on a long history of iconic images of children in hardship, social media allows the unique opportunity to understand viewers’ interpretations of such images first-hand and on a large scale. 59 References Ahmed, S. (2004) The Cultural Politics of Emotion. Edinburgh: Edinburgh University Press. Alanen, L. (1988) Rethinking Childhood. Acta Sociologica, Vol. 31, No. 1 , p. 53-67. Barad, K. (2003) Posthumanist Performativity: Toward an Understanding of How Matter Comes to Matter. Signs: Journal of Women in Culture and Society, Vol. 28, No.3, p.801–831. Higonnet, A. (1998) Pictures of Innocence: The History and Crisis of Ideal Childhood. New York & London: Thames & Hudson. Jack, I. (2015) Can Images change History? The Guardian. Friday 4th September 2015. Retrieved from http://www.theguardian.com/commentisfree/2015/sep/04/images-aylan-kurdi-syria James, A. (2013) Socialising Children. London: Palgrave. James, A., Jenks, C., and Prout, A. (2007) Theorizing Childhood. Cambridge: Polity Press. Langmuir, E. (2006) Imagining Childhood. New Haven & London: Yale University Press. McCloud, S. (2001) Understanding Comics: The Invisible Art. William Morrow Paperbacks. Rose, N. (1999) Governing the Soul: Shaping of the Private Self. London: Free Association Press. 60 13. When S hould We S hare Distressing Images? S eeing Aylan Kurdi S am Gregory At WITNESS we see hundreds of images of distressing events and we know that there are millions of images of horror out there in the world. It is a function of our ‘cameras everywhere’ world that there is a greater volume of images that show the dark side of life. Some are shot with the best of intentions by citizen eyewitnesses, others are shot by journalists and some are shot by perpetrators. We often grapple professionally with how we understand what images to share and how to share them. Sometimes a particular image breaks through and hits me particularly directly and personally. I cried when I saw the image of Alan Kurdi (initially identified as Aylan Kurdi), a young Syrian Kurdish boy dead on a Turkish beach, and it made me think through a very personal lens about the choices I and others make to view and to share images like this and the constraints upon those choices. Last year I wrote about this topic in a blog post on ‘Images of Horror: Whose Roles and What Responsibilities?’ in which I explored the sharing of images of the killings of hostages by ISIS. I looked at our roles as sharers, for example how individuals shared the images – or counter images of the hostages with their families – via their social networks via our social media networks. And I discussed the roles that those platforms, on which we were sharing the images, played. The debate around this question has been reignited in the social media and professional media spheres in the US and globally around two recent but separate incidents. First, the distribution on social media of a video shot by a man of the moments in which he murdered news reporter Alison Parker and her cameraman Adam Ward. The second, the photos shared of Alan Kurdi, who died at sea while trying to reach Europe with his family and whose body washed-up on a popular beach in Turkey. Much of the debate has focused on the role of the media in sharing these images (for a good survey see Robert Mackey’s post in the New York Times), but equally important is the question of our role as individuals and communities in watching, sharing and using them (and ultimately, most important will be the question of how our policies, laws and actions change in response to these images). The A uto -P lay E ffect and P latform C ontrol In the first scenario in Virginia, the murders of Ms. Parker and Mr. Ward have queasy parallels to the global challenges around sharing perpetrator-shot footage that I explored in my earlier post. There I noted that: ‘[B]y circulating this imagery we can play into the propaganda needs of human rights violators, we can justify what has emerged in some settings as a commercial market for terrible imagery of violations, and we can re-victimize and re-violate for a third time the dignity of people who have already faced a direct abuse, and the humiliating violence of having this captured on camera.’ All of these as equally apply to a murder on camera, as a war crime on camera. An additional technical aspect that has emerged is the way in which videos autoplay on social media platforms such as Facebook. So, like it or not, we can be scrolling through a social media feed, and be exposed to someone’s life being taken in front of our eyes. There are no technical fixes here if auto-play continues to become the norm, other than to make sure individuals can choose to switch it on or off. As with other contexts much of this relies on a rapidly-reacting content review team at the social media outlets that can either make choices about when content should be taken down or to ensure there is a graphic violence warning and that auto-play is disabled on that particular video. Many applauded Facebook and Twitter’s fast action to take down the killer’s accounts in the Virginia shooting case. However, the power vested in these companies is fraught with challenges. So far, we have not seen the images of Aylan Kurdi deleted or removed from Twitter or Facebook (though they come with a graphic content warning) but this may only be because of the attention given the image on both social media and in the mainstream media. 61 A contrasting experience faced the artist Khaled Barakeh, who, on August 29, 2015 posted a set of seven photos to Facebook (probably culled from media reports) that showed children and youth who had drowned in a shipwreck off the coast of Libya the day before. As Nicholas Mirzoeff who blogged about it, describes: ‘The photographs are elegiac, mournful and devastating. They were shared over 100,000 times, reaching me on August 30 via an Indian friend living in the U.K. I was moved to write a blog post “The Drowned and the Sacred” that has been read and shared thousands of times’. However, on August 31 many of the Facebook community, who had shared and discussed the photographs and my post, noticed that the link to Barakeh’s album had disappeared from their timelines and activity logs, including myself. His photos had been taken down by content review processes in Facebook, presumably for violations of guidelines on graphic imagery (though no information was apparently shared to Khaled Barakeh on the rationale). The P ower of C ounter -S peech and A mplifying B eyond J ust S haring We also need the types of more conscious conversation about what we choose not to watch and not to share that have been taking place around these images. This discussion is one about human dignity. Of course, at WITNESS we believe in the right to free expression, but also the importance that human rights discourse places also on individual dignity and integrity and how we protect people who have suffered violence already from further re-victimization. But it is also a conversation about the power of these images to break through the chatter, incite discussion and mobilize change, and how we balance these imperatives. Just as in earlier examples of how people responded to ISIS images of killings, one element that we saw in both the cases of the murders in Virginia and Aylan Kurdi is the power of amplifying counter-speech and symbolic speech from each of us an individual sharing online. In terms of amplifying counter-speech, people shared images of the two young people killed in Virginia in happier times. Similarly in the case of Alan Kurdi, individuals made a choice to amplify his story with images that expanded the reach of the original image. There are only so many times one can see a heart-breaking image of a toddler dead in the surf, but there are alternatives that amplify the reach of the moment. One of the most powerful I saw in my timeline was the choice of VJ Um Amel to share her own image of a toddler’s clothing arranged just as Alan lay in the sand. Other choices included sharing family photos of Aylan and his older brother (who also died) in a happier moment, and sharing images of one’s own children enjoying the beach in the way children should enjoy a beach. And cartoonists and remixers have taken the image and turned it into a range of poignant, beautiful and angry images that transmute Alan’s image to an idyllic beach with a bucket-and-spade, a peaceful night-time’s sleep, or the floor at the center of the UN Security Council Chamber (warning: graphic image) and multiply its reach and advocacy impact. When We D o S hare the I mage ? But many of us – including myself – did share the image of Alan Kurdi dead in the surf (as did many European newspapers). This decision asks us at the personal and collective levels to balance out imperatives of individual dignity and re-victimization with a recognition of the iconic power of this image. Such a discussion is not new: we have seen how iconic images take an individual act of degradation or pain – such as Kim Phuc, who was photographed as a young girl napalmed in Vietnam in 1972 or Neda Agha-Solta, shot in Iran in 2009 – and turn it into a representation of a systemic problem. As someone who works within an organization dedicated to video advocacy storytelling for human rights, we are familiar with how an emblematic story takes an individual’s narrative and makes it stand in for the whole. It is a paradox of human rights that most of the time as Tom Keenan has put it “the claim is meaningless if it is not universalizable, but it is effective only if it is rooted concretely.” All claims for human rights recognition are inherently a “plea on behalf of everyone, passing through someone in particular.” 62 It is a sad reality that to bring home the reality of the claim of refugee status that Syrians fleeing conflict are entitled to under international law, and the horrors of their attempts to reach Europe, we have to root that plea for everyone in the death of a young boy from Kobane in Syria. And even for someone as close to the situation and as involved as the experienced Human Rights Watch researcher Peter Bouckaert, who posted about his choice to share the image, he acknowledged the power of individual, relatable emotional and experiential connectivity that I too felt: “I thought long and hard before I retweeted the photo It shows a lifeless toddler, lying face down on a popular Turkish beach, one of eleven Syrians who have almost certainly died as they tried to reach safety in Europe by boarding a smuggler’s boat. Instead they ended up as the latest victims of Europe’s paltry response in the face of a growing crisis. What struck me the most were his little sneakers, certainly lovingly put on by his parents that morning as they dressed him for their dangerous journey. One of my favorite moments of the morning is dressing my kids. They always seem to manage to put something on backwards, to our mutual amusement. Staring at the image, I couldn’t help imagine that it was one of my own sons lying there drowned on the beach.” The power of homophily (the tendency to bond with similar people – in this case that both Peter and myself relate Alan’s loss to our own children of a similar age) cuts both ways. Often times we are asked whether we would want our own child or relative to be the one seen in these images. It is a legitimate query that goes to the heart of the question of dignity and representation, and there is no easy answer. As CNN describes it in a recent profile of Kim Phuc:: “In the beginning, says Phuc, she hated the photo. It embarrassed her. And she struggled with the publicity that surrounded it. For her it was personal: It captured a moment of torment — her face frozen in an agonizing wail moments after a napalm attack burned and disfigured her for life.” Later she came to be at peace with the image, and its impact on the conflict in Vietnam. In the case of the little boy from Syria, Alan Kurdi, we have our initial answer in the words of his father as he left the morgue to visit his son: Abdullah Kurdi collapsed in tears after emerging from a morgue in the city of Mugla near Bodrum, where the body of his three-year old son Aylan washed up on Wednesday. A photograph of the boy’s tiny body in a bright red t-shirt and shorts, face down in the surf, appeared in newspapers around the world, prompting sympathy and outrage at the perceived inaction of developed nations in helping refugees. Aylan’s 5-year old brother Galip and mother Rehan, 35, were among 12 people, including other children, who died after two boats capsized while trying to reach the Greek island of Kos. “The things that happened to us here, in the country where we took refuge to escape war in our homeland, we want the whole world to see this,” Abdullah told reporters. “We want the world’s attention on us, so that they can prevent the same from happening to others. Let this be the last,” he said. This paper first appeared on WITNESS blog and has been republished here with permission. 63 14. J ournalism and Death on Aylan Kurdi Claire Wardle the S ocial Web: The Case of “I don’t know about you, but I’ve had to stop looking at Twitter this week. I don’t want to see all of this horror”. This was a comment made by a pretty hardened UN worker who had spent her early career in Rwanda during the genocide. It was March 2015, a week when videos emerged of ISIS pushing homosexual men off buildings. As a communications officer who followed large numbers of journalists, she couldn’t prevent seeing these images that were automatically viewable in her Twitter timeline. In many countries, regulation controls the inclusion of graphic content in news output either broadcast on television or published in newspapers. As a result, particularly graphic material is not seen, especially ‘moment of death’. It is important to note, graphic content is very culturally specific. What would be prohibited in the US or UK could lead an evening news bulletin in the Arab world. The web, and therefore social networks are not regulated in the same way as television or print. Facebook, Twitter and YouTube have Terms of Service, and any content that breaks these, particularly around illegal activities is removed. While algorithms play a role in automatically removing certain types of content (which can then be challenged by the user who uploaded it), for the most part, the system works by users flagging content to moderation teams, who decide whether or not to keep the content up on the site. This chapter explores the moments of tension, which have emerged in the last three years as technical developments on the main social networks have exposed challenges in relation to graphic content. Aylan Kurdi was one of these moments, but it’s worth considering the context to the conversations that played out around the image. The history of graphic images on the social web In early 2013, Twitter changed their search results page, so at the very top of the page, just above the chronological list of keyword tweets, a montage of related images popped up automatically. Remember, in 2013, you had to click on a link to see anything visual on Twitter, there were no in-timeline automatically-opened images. People applauded the move, particularly journalists who could quickly find images related to breaking news events. And then the Boston marathon bombing happened in April 2013. As the news spread, many Twitter users turned to the service knowing it provides some of the quickest eyewitness reports during events like this, but the harmless search #Boston returned results pages with very graphic images at the top of the page. With no warning, a collection of images emerged of very injured people lying in pools of blood, many with missing limbs. This moment marked the start of an increasingly vocal debate about the role of social networks and graphic imagery. At that point, the networks were very clear that they were simply platforms; they were not editorial entities and therefore did not make editorial decisions. While the complaints about the Boston marathon were heard, no platform changes were made. In August 2014 the US journalist James Foley was killed by ISIS and video of his execution was disseminated on social media. Journalists, covering the story, had to watch the video in order to verify the content, but some retweeted the video, and it was quickly being shared widely, well beyond ISIS channels. However, within a few hours the hashtag #ISISMediaBlackout emerged, reminding people that by sharing, they were actively supporting the objectives of the extremist group. Significantly, Twitter and Facebook removed the video, making a clear editorial decision to take down content that did not explicitly break their Terms of Service. This was the moment when the networks transitioned from simply communication platforms to publishers. 64 In December 2013, Facebook rolled out on mobile an auto-play function on videos uploaded to their platforms. While users can turn off this feature the default setting is switched to ‘on’. As Twitter learned with the Boston marathon bombings, the challenges posed by technological advancement only become clear through use. For Facebook and the autoplay feature it was the shooting of the police officer Ahmed Merabet shortly after the offices of Charlie Hebdo had been targeted. Jordi Mir captured Merabet’s death by pointing his phone out of his apartment window. He uploaded the video to Facebook, but 15 minutes later, changed his mind and removed the video. But in that time, the video had been downloaded by one of his friends and re-uploaded to YouTube and the video took on a life of its own. It was used by almost all news outlets either in its original form, or as part of a package. These were broadcast or shared on news websites, but they were also pushed out by news organizations via their Facebook pages, entering people’s feeds, auto-playing before users were prepared for what they then saw. In focus groups carried out as part of a research project exploring audience attitudes towards viewing eyewitness media in news content, many participants, unprompted, discussed the amount of graphic content they were seeing in their feeds, arguing that much of it is eyewitness media. The research took place a couple of months after the Charlie Hebdo shootings and it was still top of people’s minds. Many people discussed how upsetting it was to see Merabet killed, arguing that many were not expecting to see that. As one participant explained: “I couldn’t avoid it. I pick up the Metro every morning and I wish I hadn’t seen it, but I saw someone before they died. It was horrible.” When pushed by the moderator about whether they were seeing this content because it was being shared by friends, they explained that this wasn’t the case, that it was actually because they following news brands on Facebook or Twitter. In April 2015, Walter Scott was shot and killed by a police officer in South Carolina as he ran away. Feidin Santana, who was passing by at the time, captured the shooting on his mobile phone. The video was shared very widely, and with Facebook’s auto-play feature embedded on desktop and mobile the issue of users inadvertently seeing graphic imagery they couldn’t unsee was being raised again. This piece by The New Yorker, which starts with the line, “Want to watch a video of a man being killed?” is an example of how these debates were moving from the water cooler mainstream coverage. In August 2015, a bomb exploded at a shrine in Bangkok. An eyewitness wandered onto the scene immediately afterwards. He pulled out his phone and started live-streaming on Periscope. Over the course of three separate streams, he walked around the scene streaming what was in front of him, narrating what he was seeing and responding to live comments he was receiving on his stream from people watching all over the world. As this [1] detailed analysis highlights audiences were torn by the graphic nature of what they were seeing, with half requesting to see more, and the other half arguing that the stream be switched off to protect the dignity of the victims. Later that day the stream was ‘featured’ on the Periscope home screen (with a graphic warning added) but over 100,000 saw it. Later that month, the social networks faced the first major backlash about auto-play videos when the Virginia journalists Alison Parker and Adam Ward were killed live on television as they undertook an interview for their morning news show. Two videos emerged. One was captured by the killer who had filmed the events and then uploaded them to his Facebook and Twitter accounts. The social networks moved fast and removed the films very quickly but news organisations and other users had downloaded copies and shared stills from the videos throughout the day. And then some viewers who had been watching the events on their televisions, rewound their DVRs, captured the moment and distributed it via social media. In those early confused minutes, many people saw these incredibly upsetting videos, and as the day unfurled, and news organisations around the globe pieced together the available images, huge numbers saw a woman killed in front of their eyes as she looked directly into the camera. Although the reaction of the social networks to quickly remove the content, the debate around auto-play raged. A piece by Mashable explaining how to turn off auto-play settings was very widely shared. Some commentators argued that people needed to see the hard reality of gun violence, but it was a minority argument. Interestingly, the decision by the New York Daily Post, to publish stills from the video as a full page image on their front page led to significant backlash, but played into the wider debate of people wanting choice about whether or not to see an image. In the same vein as 65 videos auto-playing, or images popping up already opened in a timeline, people complained loudly about having such graphic images on a front page viewable on newsstands in the street, in shops and at subway stations. I mages of refugee children on the social web For transparency, from August 2014 to April 2015, I was the Senior Social Media Officer at UNHCR (the UN Agency for Refugees), responsible for sharing stories, images and videos with our audience on social channels. We would tweet hourly, post 2-3 times per day on Facebook, share regular images on Instagram, and publish professionally produced videos on YouTube twice a week at least. As a team we would regularly review ‘what had worked’, looking for the content that produced the most engagement. We knew that any images of children received the most engagement, but mostly when the children were smiling. On social media, people wanted to share uplifting and inspiring stories and images. We were therefore constantly juggling telling the necessary story of the daily struggles faced by refugees but also recognizing that a daily digest of hardship would not create the engagement we needed to keep us visible in people’s timelines. We had to work with the algorithm. Remember that the refugee story was certainly not mainstream, even earlier this year. Almost 800 refugees and migrants lost their lives off the island of Lampedusa, there was an uptick in coverage, but it was minimal. In just the year I was working in this role, I was regularly amazed at the images that would ‘take off’ on social for us and those that didn’t. We shared so many incredibly moving stories, illustrated by images taken by some of the world’s best photographers, but they would often hardly register. Our ‘fans’ would dutifully like, comment and share with their networks, but we struggled to move beyond the communities of people who had already shown an interest in refugee issues. On the weekend of August 30, two days before the image of Aylan Kurdi emerged, I saw a friend of a friend on Facebook post a collection of images from a Libyan beach showing a group of dead babies that had washed up after their boat had capsized. They were face up, their bodies bloated and their eyes rolled back. They were still wearing nappies. I saw the images, and was taken back to the daily sadness I had felt at UNHCR as we saw and heard about the realities faced by refugees on a daily basis. I remember thinking that this was posted by someone who didn’t normally share information about the refugee cause (he was a technology journalist) and wondering whether this showed that more people were starting to appreciate the reality of these Mediterranean crossings. A few hours later, I saw this post from the same person complaining that the images he had shared had been reported by someone to Facebook, and the photographs had been removed from his timeline. A fascinating conversation developed under his status update with people discussing graphic imagery, censorship, the need to see versus the right to not have to see. This same conversation exploded two days later when the image of Aylan Kurdi emerged. On the evening of September 3 I left work and opened Facebook. Every single status update in my timeline referred to the image. As someone who had worked formally or as a consultant with UNHCR since 2011, I would always notice when non-UNHCR friends posted information or images related to refugees. It always felt like a success when that happened. To see every one of my friends, and all news organizations discussing a subject so close to my heart was astonishing. Aylan K urdi I’m not going to write about the image itself. That has been covered in great detail in this report. What I’m going to write about was how users themselves framed the image. As mentioned, immediately after the shooting of the two journalists, Alison Parker and Adam Ward in Virginia, some anti-gun activists were very vocal about the need for people to see the shooting as a way of acknowledging the reality of gun violence, but it was a limited point of view. With the image of Aylan Kurdi, the social web was divided. In 66 my feed at least, half of my timeline referenced the image but said they didn’t want to share, as it was so upsetting, and also recognizing that most people would have already seen it. The rest of my timeline were posting the image and justifying doing so, stating – everyone must see this image. For me, it was the first time I’d seen users on social networks recognize the power they have as individuals to decide what they want their own audiences to see. The conversations playing out on social media around the Aylan Kurdi image were echoing those that take place in newsrooms every day. What should people see? Why do we think our audience needs to see this? Is it appropriate for us to show it? Are there other ways we should be showing this? Will children see this image? Is that appropriate? And so on. This warning from Buzzfeed shows that they wanted to warn their users before they clicked on the image, but this position was a rare one. Furthermore when Aylan Kurdi was identified and his name emerged, users argued strongly that his name should be referenced so that he be identified as an individual, not simply one of the thousands of nameless refugees attempting to flee every day. The fact that so many people know Aylan Kurdi’s name is quite incredible. When compared with other iconic images, particularly of children over history, whether it is the photograph of ‘Napalm girl’ (Phan Thi Kim Phuc), the image of ‘vulture and the little girl’, or Colombian girl trapped in the mudslide (Omayra Sanchez), their names are unknown or barely recognizable. Rather than seeing the image of Aylan Kurdi on a nightly news bulletin and reading about him in an article in the morning newspaper, people saw the image and read his name multiple times, because friends and social connections argued that you should. C onclusions Despite the furore over auto-play shortly after the Virginia shootings, it continues to exist. It is almost inconceivable that it would now be turned off, although product design on Facebook and Twitter might evolve so users are informed more clearly about how to turn it off. In October 2015, in the US, Twitter launched Moments, a new product run by a small editorial team which curates the ‘best’ tweets around any particular subject. Rather than scroll through a timeline, the user swipes through individual tweets and related images and videos. Two days after launch, a peaceful protest in Ankara was rocked by two large explosions killing almost 100 people. Very quickly, graphic images emerged of the bombings and the immediate aftermath. The Twitter Moments team reacted quickly, placing a graphic warning image before you swiped into images or videos from the scene. The image of Aylan Kurdi was incredibly upsetting, but it wasn’t graphic. An algorithm would never have automatically picked up that image and flagged it as graphic. For me, the image raised for many people the challenges faced by publishers around important stories. Where do you draw the line? There is now much discussion about the responsibilities of social networks, as they move from simply platforms to publishers. And while they are building systems to automate the removal of illegal and extremely graphic content, this example of the image of Aylan Kurdi highlights the very human aspect of social networks. Individuals making particular decisions about what they think is appropriate or not. Previously, this power lay in the hands of a very small number of editors. That has changed fundamentally. F ootnotes [1] Brown, P (2015, August 26). “OMG I can’t ever unsee that”: What happened when the aftermath of the Bangkok bomb blast was streamed live on Periscope. Medium. Retrieved from https://medium.com/1st-draft/omg-ican-t-ever-unsee-that-what-happened-when-the-aftermath-of-the-bangkok-bomb-blast-was-7a3f39ee2b0 67 15. B ody C ounts: C ounting Aylan Kurdi E velyn R uppert and F unda U stek How Aylan Kurdi is counted and comes to count testifies to the force of what Cornelius Castoriadis (1997) calls social imaginaries. Spirits, gods, God, polis, citizen, nation, population, money, capital, interest, taboo, virtue, sin, and so forth are imaginaries not because they fail to correspond to recognisable and specific phenomena or experiences but because they require acts of imagination. So too does a single body lying on a Bodrum beach. It is the force of this image that calls upon individuals to act. It is also a force that calls upon political authorities to act including those assembled at what is claimed to be the first International Conference on Refugee Statistics. Both the timing and location of the conference (one month after Aylan Kurdi’s death and in the Mediterranean city of Antalya, a major route for refugees in their attempt to travel to nearby Greek islands) marked an important turn of events with respect to the politics of counting refugees and migration movements in Europe and beyond. Although the conference agenda clearly focused on counting as a technical issue and did not directly discuss Aylan Kurdi, the discussions highlighted the contested politics of counting migrating bodies. The conference was international in its scope; with 28 countries[1] and 10 international organisations[2] being represented. However, because the majority of countries and organisations represented Europe, most often the discussion of refugees concerned the recent migratory movements of Syrians. Often the presentations started with ‘thanking Turkey’ for ‘doing such a great job’ by hosting the highest number of Syrian refugees in the world. The UNHCR cited estimates that currently there are 2 million Syrians in Turkey, 1.1 million in Lebanon and 630 thousand in Jordan. But such estimates belie the normative and political judgments about who counts and how migrating bodies are counted. Who is a refugee ? All counting practices begin with a definition. The 1951 Refugee Convention defines a refugee as someone who “owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality, and is unable to, or owing to such fear, is unwilling to avail himself of the protection of that country” (UNHCR 2011). Although this was the working definition at the conference, it emerged that it far from answers the question of who counts as a refugee and who does not especially for registering migrating people ‘on the ground.’ Should they be counted as part of the population or assigned a separate category? Does a person’s refugee status (ever) end? Should counting differentiate between refugees in need or not in need? What of stateless persons, internally displaced populations (IDPs), asylum applicants and people whose applications have been rejected and who have different legal statuses in international law? Here we can ask a question that was not asked: how will the body of Aylan Kurdi be counted? Or the other thousands of bodies and corpses that wash up on beaches and are usually unidentified, interred in mass unmarked graves and whose stories are not being told?[3] Clearly, not only is there a politic of counting moving bodies but also of counting when they stop and meet their end. These questions attest to definitional, political and legal struggles, which at the conference came at least provisionally to one consensus: the current definition of refugees captures a ‘stock population’ but is unable to capture ‘refugee flows’. Indeed, it is moving and migrating bodies that drive the political imaginary and have also been the hardest and most difficult historically for states to count and monitor. It is such flows that have stoked political rows such as when Prime Minister David Cameron referred to attempts by migrants to cross the English Channel from Calais as ‘a swarm of people coming across the Mediterranean, seeking a better life’ (Dearden 2015). Such imaginaries are also fed by visualisations of migration movements. One international consortium has produced an Interactive Map on Migration (i-Map) to inform intergovernmental dialogues on migration.[4] The visualization enacts 68 migrants as a series of lines crisscrossing states and in flows and volumes that dynamically change from 2000-2013. Yet it is the image of one body in transit washed upon a shore that has turned such lines on a map into living and dying people. Why count refugees? For some countries, debates about refugees invariably turn to whether they should be counted using the same method as that of enumerating the ‘usual population’ or through a separate method such as a special register for refugees. In other words, are they part of a population or should refugees be separated categorically, methodologically and thus politically? The UNHCR’s response to this question is revealing of the politics of counting refugees. They noted that counting is not a purely statistical matter, as refugees are not just numbers; rather, counting is the first step in restabilising an individual’s identity after displacement by tying a date of entry and a name to a person. Registration is thus beyond the question of counting but as a means of identifying and then entitling refugees to assistance from the time they are in need of protection to the time they no longer need it. That response highlights another politic of counting: it is very much connected to specific needs, rights and entitlements and also the distribution of resources. This goes beyond hosting and counting refugees as the object of concern and casts attention to their rights, needs and futures. But the response also misses that counting refugees remains grounded in an imaginary of who does and does not count as part of a population, an imaginary that feeds definitions and counting practices. But even if included in the definition of a population, estimates of the number of refugees were not left unquestioned at the conference. Discussions often quickly turned into concerns about the lack of reliable data and that the making of humanitarian statistics responds to different objectives than those of counting the ‘usual population.’ To meet humanitarian purposes agencies in the field start with different interests and objectives, work with different definitions and specific management requirements, and need to respond rapidly to crisis situations. This inevitably leads to different estimates and worries about double counting or missing people. In addition, many counts do not include social characteristics of a person such as gender, age and country of origin. Such data are either not collected at all or not collected in a systematic way that would allow comparison or developing targeted responses. But such missing data on demographic characteristics are crucial for translating a social imaginary into numbers and flows of refugees that are socially meaningful and, as international organisations reminded, are also necessary for knowing and allocating specific services and assistance: ‘It is not about wanting to know these details, but needing to know’ the UNHCR emphasised. Counting refugees is thus not only about counting ‘heads’, but knowing about social positions, lives and livelihoods. It is possible that in his death we know more about Aylan and his family than we know about the lives and social characteristics of other migrants. C ounting one body at a time The foregoing describes one way that refugee numbers, counts and flows are politically charged and become socially meaningful. But what the First International Conference on Refugee Statistics also revealed is that even when there is an internationally accepted definition of refugees, its implementation encounters numerous normative, political and practical questions. Nevertheless the conference will lead to the drafting of a ‘Handbook for Refugee Statistics,’ which will be presented at a United Nations Statistical Division Conference in March 2016. Yet it remains to be seen whether visualisations of flows of migrants or the image of one body will come to count more. 69 F ootnotes [1] Afghanistan, Azerbaijan, Bulgaria, Canada, Cote d’Ivoire, Denmark, Ecuador, Egypt, Ethiopia, Germany, Ghana, Greece, Jordan, Kenya, Kosovo, Mongolia, Morocco, the Netherlands, Niger, Norway, Palestine, Poland, Serbia, Thailand, Turkey, Uganda, the United States. [2] EUROSTAT, EFTA (European Free Trade Association), EASO (European Asylum Support Office), United Nations High Commissioner for Refugees (UNHCR), United Nations Relief and Works Agency for Palestine Refugees in the Near East (UNRWA), United Nations Population Funds (UNFPA), United Nations Children’s Fund (UNICEF), World Bank (WB), World Food Programme (WFP), Households International Migration Surveys in the Mediterranean Countries (MED-HIMS). [3] See one extensive account by Eric Reidy, ‘The Secret Mass Graves of the Refugee Crisis’ [4] i-Map is described as a support instrument that aims to enhance and facilitate intergovernmental information exchange, and to support the development and implementation of knowledge-based co-operation initiatives. It builds upon an online interactive platform serving a wide range of users, principally governments of participating states, as a source and exchange instrument for information on migration matters. ICMPD, Europol and Frontex initially developed it in 2006 for the Dialogue on Mediterranean Transit Migration (MTM). Current donors include the European Union, European Commission, and Netherlands, Norway, Switzerland and Turkey. References Castoriadis, C. (1997) World in Fragments: Writings on Politics, Society, Psychoanalysis and the Imagination, ed. D. A. Curtis. Stanford: Stanford University Press. Dearden, L. (2015, August 15) David Cameron says he did not dehumanise migrants with “swarms” comment, The Independent, Retrieved from http://ind.pn/1RyZrWN UNHCR (2011) Convention and Protocol Relating to the Status of Refugees. Geneva: United Nations High Commissioner 70 for Refugees. Retrieved from http://www.unhcr.org/3b66c2aa10.html C onclusion O lga G oriunova and Farida Vis A rapid response report is an unusual genre for academics. It promises a more immediate mode of engagement where time has not yet lapsed to offer a distance. This distance is where the regimes of recursive evaluation and conclusive explanation operate. What a rapid response does is diagnose a condition that can change tomorrow, affected by new events or old developments. The events leading to or occurring after the death of Aylan haven’t stopped happening, people haven’t stopped suffering. Our kind of writing therefore is both intense and porous. On the one hand, it engages with the pain and the forces of the moment, and on the other hand, it cannot offer a solution, a closure. On the one hand, we, the authors, hope that we can do something, with words, and on the other hand, we feel quickly overpowered, with flows of events and political forces attempting to wrench what has been our focus out of sight. This is the nature of the rapid response: two months after the spread of the images of the body of Aylan on the Turkish beach shocked the world - leading to the large-scale, even though momentary, mobilisation of publics to welcome and help those that were now addressed as refugees instead of migrants - new stories began to arise in the aftermath of the Paris terrorist attacks. Finding a passport of a Syrian refugee who had travelled to Europe through Greece on the site of the Parisian attack prompted many a UK newspaper to run a headline that swayed the rhetoric once again towards ‘defence of the borders’. Indeed, it looks more likely that the softening of attitudes to refugees which the image of Aylan Kurdi gave rise to will be to a greater or lesser extent undone. What we do know is that attitudes to asylum seekers in the UK, and Europe more broadly, have been very negative for at least two decades. The image of Aylan Kurdi created a social media event that looked capable of becoming a critical juncture in changing attitudes for the better. The aftermath of the Paris attacks likely mean that this promise will not be realized. The report focuses on the power of a few images, and the events behind them, to transform global opinion and affect political decisions. One of the important lessons clearly heard throughout contributions to this report is that the tragic images of Aylan Kurdi were and continue to be used by different actors to achieve different ends. What we learned is that as forcefully shocking as they are in themselves, the images were efficient because they fit the visual canon we’re trained in and move efficiently between contexts. The report unearthed our envelopment in certain national visual cultures and practices, as well as our dependency on the specific local politics and national histories. Overall, the report testifies to the immense power digital media, social networks and technological devices hold, transforming us into publicists, exposing us to the necessity to make moral decisions that are rendered technical or intertwined with technical affordances, and expanding the realm of public discussion, emotional or creative reaction and practical response. Throughout the report, we called the dead child Aylan Kurdi, which is factually incorrect (his name is Alan Kurdi). The problem is that the images spread as the ones depicting Aylan, and Google search data is only significant in relation to Aylan. The events of the transformation triggered by the images are related to the wrong spelling of the child’s name both in human and in technical memory, much of which is now hard to change. As we were analyzing the spread of the images and the response to them, where all archived data are related to Aylan, we decided to continue using this name in the report. The images are still active in discussing the Syrian conflict and drawing attention to the devastation it brings. Journalist Liz Sly who was key to the image going viral on Twitter before being picked up by mainstream media talks about the ruins of the city of Kobane where the family was from. The images also keep being used in different political contexts, such as elections. One could even say that the images continue their circulation, becoming instruments in various kinds of campaigns and attention trading, leaving the surviving father at the centre of an atrocity panopticon. The social media life of these images is such that there is a total loss of control around their circulation and use, both for the father and family and for the photographer. Yet, certainly, the viral character of social media and the force of affective engagement that visual imagery can provide don’t make them the only avenue through which to create awareness of the war and empathy towards those affected. Images of Aylan allowed a certain type of framing and media coverage, the one ‘speaking for the dead.’ 71 Covering the refugee crisis from and through the eyes of ‘still alive’ refugees is difficult but clearly possible. Social media are also not only used to do something about the refugees, but are actively used by refugees themselves, becoming their lifeline to the world. [1]. Understanding this and many other complex intertwinings between images, social media, technical networks, politics, public opinion, publishing, war, deaths and lives requires a coming together of researchers from many disciplines and opening up of possibilities of new ways of working across sectors. It also seems that finding ways of doing this promptly and genuinely contributing to the debate as it is happening with more large-scale research following on will become imperative. We hope that the next steps following on from this rapid response will involve further research, expanding our scope as well as reflection on the mode of working we trialed. F ootnotes [1] An Xiao Mina (2015, October 29). The Problem with #FirstWorldProblems. Our global imaginations are failing us and how we talk about tech. Medium. Retrieved from https://medium.com/@anxiaostudio/failures-of-our-globalimagination-8648b2336c2c 72 Biographies Jim Aulich is Professor at the Manchester School of Art. Jim is part of the Visual Social Media Lab. Anne Burns is a Research Associate in the Visual Social Media Lab, based in the Information School at the University of Sheffield. Ray Drainville is a PhD student at the Manchester School of Art and part of the Visual Social Media Lab. He is also the Director of ArDes, based in Sheffield. Francesco D’Orazio is Vice President of Product at social analytics firm Pulsar and Chief Innovation Officer at innovation consultancy FACE. Francesco is part of the Visual Social Media Lab. Simon Faulkner is Senior Lecturer in Art History at the Manchester School of Art. Simon is part of the Visual Social Media Lab. Olga Goriunova is Senior Lecturer in Digital Culture, Department of Media Arts at Royal Holloway, University of London. Sam Gregory is the Program Director at WITNESS and teaches at the Harvard Kennedy School. Lucy Mayblin is an ESRC Future Research Leaders Fellow, based in the Department of Politics, at the University of Sheffield. Lisa Procter is Lecturer in Early Childhood Education in the School of Education, at the University of Sheffield. Lin Prøitz is a Marie Skłodowska Curie Fellow in the Visual Social Media Lab, based in the Information School at the University of Sheffield. Simon Rogers is the Data Editor at Google News Lab. Evelyn Ruppert is a Professor in the Department of Sociology, Goldsmiths College, University of London. She is Principal Investigator of the project, Peopling Europe: How data make a people (ARITHMUS) funded by the European Research Council. Holly Ryan is a Research Associate at The Crick Centre, based in the Department of Politics, University of Sheffield. Mike Thelwall is Professor of Information Science at the University of Wolverhampton. Mike is part of the Visual Social Media Lab. Funda Ustek is a Post-Doctoral Researcher on the ARITHMUS project, Goldsmiths College, University of London. Farida Vis Farida Vis is Director of the Visual Social Media Lab and a Faculty Research Fellow, based in the Information School, at the University of Sheffield. Claire Wardle is the Research Director at the Tow Center for Digital Journalism, Columbia University. Dylan Yamada-Rice is Lecturer in Early Childhood Education in the School of Education, at the University of Sheffield. 73 the iconic image on social media: a rapid research response to the death of aylan kurdi* Visual Social Media Lab