Abstract
This article examines the role that the global television news agencies play in the handling of user generated content (UGC) video from Syria. In the almost complete absence of independent journalists, Reuters, Associated Press and Agence France-Presse are sourcing citizen videos from YouTube channels and passing it on to their clients. This article examines the verification processes that the agencies undertake to check on the veracity of this material and asks whether the agencies have abandoned independent journalism to activists. This article provides a comparative analysis of two months’ worth of UGC videos from Syria that were broadcast by the global news agencies after Russia joined the bombing campaign in Syria in late 2015. It analyses the content, verification processes and information that the agencies give their clients about this material. Through interviews with senior editors from the three organisations, questions of certainty versus probability are explored, along with ethical arguments about propaganda versus information transparency. The global news agencies are the engine drivers of international news coverage and their decisions and interpretation feed directly into the media ecology of mainstream and then alternative media.
Keywords
Introduction
This article examines the role that the global TV news agencies – Associated Press (AP/APTN), Reuters and Agence France-Presse (AFPTV) play in the sourcing, verification and republishing of user generated content (UGC) from Syria. The territories under Islamic State (IS) control are now almost completely cut off to either professional or local journalists, leaving the way open for the various participants in the war to put out their own versions of the truth. While scholars have analysed the way that IS ‘events’ find their way into mainstream and alternative media, little attention has been focused on the role of the main engine drivers of international news – the global news agencies. These organisations have taken on a new job of brokering the UGC material that comes out of Syria and is passed on to TV audiences worldwide. Apart from large public service media (PSM) companies like the BBC, few can afford the time or resources to independently source and verify the UGC that is constantly refreshed on dedicated YouTube channels or other social media platforms and the media therefore rely on the agencies to do this investigative work for them. This article will examine how the agencies carry out this work and what they tell their client broadcasters about the content.
Background
The so-called Islamic State (variously known as IS, ISIS, ISIL or Daesh) forced its way onto the world’s stage in 2014, after taking over large parts of Syria and Iraq and declaring a ‘caliphate’. Since then, there has been much interest in the public domain, as well as in scholarly circles, in the group’s alleged media prowess. In particular, the gruesome execution videos of hostages such as Peter Kassig and James Foley in 2014 have earned IS a reputation for a skilled grasp of the workings of the media and its production values. In an article in the UK’s Daily Telegraph newspaper (Freeman, 2014), it was noted that the Kassig video had been edited using sophisticated software called Avid, which was said to retail at the time for £200,000. These notorious execution videos are highly stage managed, with close-up shots of the hostages, often overlaid with a soundtrack of intoned group singing. Their rebroadcasting has led to calls from some quarters that the media should not assist in the propaganda efforts of IS. In 2016, a Lowy Institute Report by Lauren Williams called on the Australian Press Council to develop a new set of guidelines for the media, with the purpose of not aiding IS propaganda. ‘These should include recommendations on the use of Islamic State-supplied material and the terminology used to describe it’ (Williams, 2016: 14).
In this report, Williams argues that the main purpose of the videos is ‘to recruit supporters’, ‘generate fear amongst opponents’ and ‘assert legitimacy’ as a state (p. 3). As such, the IS media machine does not just deal in torture and death, but also in demonstrating how life can seemingly be fulfilling inside its territory. In a detailed study, Charlie Winter (2015) found that IS published many more media ‘events’ than had previously been realized. Between mid-July to mid-August 2015, IS put out ‘an average of 38.2 unique propaganda events a day from all corners of the Islamic State caliphate’ (p. 5). He found that the content fell into six categories: ‘mercy, belonging, brutality, victimhood, war and utopia’ with over 50 percent depicting everyday ‘life in the Islamic State-held territories’ (p. 6). However, the data collection for Winter’s study was carried out before the bombing campaign against IS and rebel positions was intensified later that year, when Russia (September 2015) and then the UK (December 2015) joined in the bombing. It is likely that Winter’s study regarding IS ‘events’ would have found that the percentage of ‘victimhood’ and ‘war’ content would have risen against ‘utopia’ if it had been carried out later.
Alongside questions related to the ethical use of propaganda material are those of verification and legitimacy. When content has been sourced by necessity from people or groups who are not your employees, how can you know it is what it says it is? News organisations had largely stopped sending their own staff to Syria by mid-2013 (Murrell, 2014; Pendry, 2015). ITN Channel 4 News International Editor Lindsey Hilsum summed up this problem with reference to Syria when she said, ‘Sometimes a clip doesn’t tell you the whole story. People putting footage out often have an agenda’ (Lees, 2016). According to Williams (2016: 6):
Islamic State has been able to exert significant control over the way in which it is depicted. There is almost no way to verify the information first-hand meaning reports about Islamic State often rely on material released by the group itself.
For busy journalists, verifying UGC can be costly in time and resources and truth is often elusive. Channel 4 News has only one person in charge of verification (Lees, 2016), whereas the BBC has had a ‘UGC Hub’ staffed by many journalists since 2005, whose job is to verify social media material that the organisation wants to put out. The UGC verification process has been covered with reference to the BBC’s online platforms by scholars such as Leon Barkho (2007, 2010; Barkho and Richardson, 2010) and Alfred Hermida (2009). On the broadcasting side, the BBC’s UGC work has been documented by other scholars including Claire Wardle and Andrew Williams (2008), Valerie Belair-Gagnon (2015) and Lisette Johnston (2016). The BBC UGC Hub has a history of sourcing video from Syria which stretches beyond IS to 2011 and the ‘Arab Spring’. UGC is being widely used these days to cover conflict as it is available on the internet like low-hanging fruit for the picking. However, companies that use it are not necessarily verifying it, nor are they necessarily sourcing or crediting the people or organisations that shoot it. A global study of UGC use in TV and online news found that all the 24-hour TV channels monitored by the authors ‘included UGC to report on the Syrian conflict. Indeed, for some news organisations Syria was the only story for which they included UGC’ (Wardle et al., 2014: 11).
Some of the same sourcing and checking techniques that the BBC reporters have been using are now employed by agency journalists on behalf of their broadcast clients. News agencies tend to be overlooked in research because television agencies, in particular, can be hard to penetrate. Nonetheless, over the years, a number of scholars have reminded us of the centrality of these news organisations to the global media cycle. Oliver Boyd-Barrett, 1 in work stretching from 1980 to 2014, has argued that agencies exercise ‘media imperialism’. He stresses that there is an ‘Anglo-Franco-American dominance of an international or systemic network of global, regional and national news agencies’ (Boyd-Barrett, 2014: 3). Many other scholars including Daya Thussu (2006, 2009), Thierry Rantanen 1 (2005, 2006, 2011; Boyd-Barrett and Rantanen, 1998; Rantanen and Boyd-Barrett, 2008), Kevin Williams (2011) and Chris Paterson 1 (1998, 2004, 2010, 2011) have also contributed to these debates, with the latter stressing that this global dominance affects diversity of coverage. The agencies themselves are quick to underline their dominance. As of 9 June 2016, Reuters’ website advertises the fact that it has 300 staff and 300 freelance television journalists around the world. Similarly, AP states on its website that ‘video captured by AP can be seen by over half of the world’s population on any one day.’ And AFPTV says it puts out 200 videos a day, in seven languages and nine formats. The latter agency has been less scrutinized by English-speaking scholars than the former two.
Methodology
For the purposes of this study, two months were selected that followed on from the start of the bombing of Syrian rebel and IS positions by Russia (from 30 September 2015). The data cycle went through to 2 December 2015; this day was added in order to include an IS hostage execution video, which enable exploration in more depth of the agencies’ different approaches to this phenomenon. During this time, access was secured to the online archives of Associated Press Television (AP) and to Reuters’ video newsfeeds. Using the AP archive service, many different search terms were trialled before finding the one that accessed the most relevant stories – ‘Syria and UGC’. After dismissing replicated items, such as ones that contained only archive or file, or items that were essentially the same but with a Spanish or English narration, 31 stories were selected from the two months. Using the Reuters video and text library, Reuters Media Express, similar search term trials were undertaken before choosing ‘Syria’ and then dismissing all the stories that did not contain UGC (variously called ‘social media’, ‘activist video’ or ‘amateur video’). At the end of this culling process, 65 videos were selected from Reuters. After negotiations with AFPTV, managers passed on to me the only four Syrian UGC TV stories that the agency put out during this period. They also made available to me the emails between staff members who were discussing the verification processes for each item.
The news items selected were short news pieces that were mostly less than 3 minutes in length. The videos were shot in ‘natural sound’ with any interviews being conducted in local languages and carrying a translation into English in the ‘dope-sheet’ (information sheet). A few had narration in English, French or Spanish on top of the natural sound, as long as they were not duplicates of other stories. The videos were analysed, and sourcing and verification data were noted down. The agency ‘dope-sheets’ that go out to clients, along with the videos, were also collected to ascertain sourcing and verification information. Most of the key information about sourcing is normally found in the shot-lists at the top of the dope-sheets, but for Reuters, which did not give much precise information in the shot-lists, there was occasionally some clarification given in the bulk of the news stories, which formed part of the dope-sheets. After collecting and reviewing the video and print data, qualitative interviews were carried out face-to-face with senior television editors of all three agencies in London. The AFPTV interview was conducted simultaneously with two senior journalists: one in London and one via Skype from Nicosia in Cyprus.
Content
A total of 58 out of the 65 videos broadcast by Reuters Television covered airstrikes or the aftermath of air strikes against Syrian rebel targets, including IS and the Al-Nusra Front (sometimes called al-Qaeda in Syria or al-Qaeda in the Levant). These videos included airplanes or helicopters in the sky, bombs dropping, explosions, smoke rising and at ground-level, and they invariably showed wounded or dead people in shops, streets, market-places and houses, or people being conveyed to hospitals in cars or trucks. The footage came from multiple towns across Syria, including from Damascus, Douma, Aleppo and Latakia. There were 6 stories featuring militant/s either warning of IS attacks or praising attacks against the West. Another category of content related to the execution of a Russian-Chechen prisoner.
The 31 AP videos, selected by using the words ‘Syria and UGC’, by a large margin also covered bombing or the aftermath of bombing against Syrian rebel targets (see Table 1, which shows that 25 videos concerned this type of content). The videos could be extremely grim with multiple dead bodies and blood flowing freely. In some stories, the filming was interrupted when bombs went off in the vicinity. In one, the person filming was knocked off his feet by an explosion (30 October 2015). There were occasionally interviews with victims of these events and very often the person filming intoned prayers to ‘Allah’ over the pictures. There were three other categories of story content among the videos: one portraying a militant or militants praising and/or warning of IS attacks (2), one related to an execution of a captured Russian-Chechen prisoner (1) and others depicting the destruction of the ancient city of Palmyra by IS fighters (4). From AFPTV, all four of the videos involved the bombing campaign and its aftermath. There were none featuring threats to the West nor were there any videos related to the destruction of Palmyra.
Content of video.
Sources and verification
In terms of passing on information to clients about sourcing, AP and Reuters differed sharply, both in the amount of information they put out and also in its transparency (see Table 2).
Types of source identified (Reuters).
Reuters television
After analysing the data, it was found that Reuters gave out the least amount of information to clients about the source of the UGC video in the dope-sheets. There were three broad categories of sourcing information (A, B and C) as can be seen in Table 3.
Types of source identified (AP).
Type (A) was present in all 65 items, and it emphasized that the material was picked up from a social media website and that the information was ‘said to be’ from a particular place on a particular date. For example, the shot-list to a story from 1 October 2015 read:
Said to be Kafranbel (Said to be 1 October 2015) (Social Media Website)
This headline for this story ran:
Mideast-Crisis/Syria-Idlib-Homs Airstrikes – amateur video.
The words ‘amateur video’ or ‘amvid’ were present in the headline to the majority of the stories, occasionally replaced by the words ‘activists’ video’.
The words ‘said to be’, ‘purported to show’ and ‘alleged’ were put into every new source of video clip. These gave scarce information along the lines of:
Thick black smoke rising from ground following what is said to be a Russian air strike. (13 October 2015)
And:
Unidentified Free Syrian Army (FSA) commander (right) being interviewed on alleged Russian air strikes. (30 September 2015)
Only occasionally were additional lines put into the body of the story, but they did not add much extra information. For example:
Syrians in rebel-held territory uploaded videos on Sunday (October 4) alleging to show the aftermath of Russian airstrikes. It was not clear how the Syrians in rebel areas who filmed the incidents knew the planes carrying out the strikes were Russians.
Sometimes information from Russia’s Ministry of Defence was added in print as a new angle or a source to a generalized bombing picture, such as in the above story where it claimed to have made ‘25 flights in Syria in the past 24 hours and hit nine Islamic State targets’ including ‘three Islamic State targets in Homs province’ (5 October 2015). However, these references did not confirm that the place being bombed directly corresponded to the Russian government statement.
Some groups or their buildings ‘allegedly’ being shown were occasionally given further explanations, e.g. ‘The Liwa Suqour al-Jabal, whose fighters have attended military training organised by the CIA in Saudi Arabia and Qatar’ (7 October 2015) but again no confirmation was given that these were the people we were seeing. In all of these instances, unlike in the bulk of the AP material, the social media websites were not disclosed by name.
Type (B) was found in 41 of the 65 stories. It read:
This edit contains User Generated Content that was uploaded to a social media website. It has been checked by Reuters’ Social Media Team and reviewed by a Senior Editor. Reuters is confident the events portrayed are genuine.
The reader of this information is asked to take it on trust as there is no mention of any forms of verification being carried out.
The third type (C) was used in 11 of the 65 stories. These stories did identify the source of the picture. These were mostly sourced from IS websites, For example:
The video was published with the onscreen logo of the Amaq News Agency by an account linked to the agency, which itself publishes Islamic State videos. (4 November 2015)
In choosing another example to illustrate this category, on 14 November 2015 Reuters put out a story headlined, ‘Islamic State video calls on Muslims to carry out attacks in France’. At the top of the news story it said in capital letters, ‘Islamic State distributes an undated video urging Muslims who are unable to travel to Syria to wage holy war to carry out attacks in France, a day after gunmen and bombers killed 129 people in Paris.’ In other instances, Reuters wrote: ‘A video posted online by the Shaam News Network, an opposition media outlet’ (1 November 2015) or ‘Video uploaded by the Revolutionary Council in Hama’ (30 November 2015).
Associated Press (APTN)
Regarding the AP footage, the source information was classed into five distinct categories (D–H). Most of the information was available at the top of the dope-sheet within the shot-list itself. The first category is titled ‘D’ and this information was evident in 15 stories:
Validated UGC Content – AP Clients Only ++User Generated Content: This video has been authenticated by AP based on the following validation checks: ++Video and audio content checked against known locations and events by regional experts ++Video is consistent with independent AP reporting ++Video cleared for use by al AP clients by ‘HOMS MEDIA CENTRE’ Talbiseh, Syria – 30 September 2015.
Sometimes, there was also further information about a particular ‘media centre’ in the body of the news story, and sometimes not. For example, on 29 November 2015, the news story said:
User generated video posted online by Ariha Today, a Facebook page that covers events in the town of Ariha in Idlib province.
The second category (E) repeats all the same information as in (D) but adds a line, informing clients they need to credit the source. There were 12 instances of this approach. Here is an example from 1 October 2015:
++Mandatory on-screen credit to ‘Revolutionary Command Council.’
In the body of this story, the client is told:
Video by a group calling themselves the Revolutionary Command Council – who are loosely affiliated with the Free Syria Army.
Quite a few of the stories are either supplied by or contain images of the ‘Syrian Civil Defence Units’ at work. According to AP, they represent ‘a group of paramedics that help evacuate people to hospitals in case of attacks’ (30 October 2015). Sometimes the shot-list also informs: ‘Video cleared for use by all AP clients by Syria’s Civil Defence Units (4 November 2015).
In sourcing example (F), the body of the message remains the same as (D) but above it, there is an urgent addition, as instanced on 5 October 2015:
++ Please note – This is not a recognised news gathering organisation ++
There were four instances of this being done. In all the examples it concerned ‘Ugarit News’, a fact that was taken up later in the interview with the AP editor.
Category (G) distances the organisation from being able to categorically say it knows who is behind the footage. Instead, it says simply ‘militant video’, as in this example from 16 November 2015:
Militant video Editors please note: ++ This militant video was not accessed from a recognised militant channel typically used for ‘official’ communications ++ ++ The AP has made every effort to identify and access the UGC from the content creator, however the AP cannot routinely communicate with militant sources ++ ++ AP Television has no way of independently verifying the content, location or date of this video ++ ++ Audio translated by regional experts ++
Finally, category (H) is not simply portrayed as ‘militant video’ but is very categorically sourced as IS. This example from 19 November 2015 informed:
Raqqa Media Centre of the Islamic State Group ++ Editors please note: These images were uploaded to a website reported to be the official media outlet of the Islamic State Group ++
A slight variation on this within the news story, said: ‘Islamic State Group Handout’.
Another example was:
The video, released by the pro-Islamic State Aamaq News Agency. (16 November 2015)
Agence France-Presse Television (AFPTV)
With regard to Syrian UGC, I expected to find that there was less coverage by AFPTV in comparison to the other agencies, as the global news director of AFP, Michèle Léridon, had said publicly in 2014 that she would not allow the agency to become a propaganda arm of IS (Léridon, 2014). In a blog on the AFP website she stated that:
Our challenge is to strike a balance between our duty to inform the public, the need to keep our reporters safe, our concerns for the dignity of victims being paraded by extremists, and the need to avoid being used as a vehicle for hateful, ultraviolent propaganda.
Interestingly, she noted that because this UGC was widely available on internet websites, it ‘relieves us of having to decide whether to provide them to our clients’. She wrote that she thought the job of the agency was to sort through the videos and select from the images as this was journalism. She added that just because this material was available everywhere, it did not mean that you had to publish it as it was not adding value. These videos were all related to the bombing campaign and its aftermath in Syria and there were none connected to militant threats, even after the Bataclan killings (13 November 2015), nor was there anything related to the execution of the Russian-Chechen hostage (2 December 2015).
The four videos went out on 30 September and 13, 15 and 16 October 2015. The first video was from the day the Russians joined the bombing campaign in Syria. The dope-sheet said that it was shot in Talbiseh by the ‘Homs Media Centre’. It noted that the pictures were of smoke ‘billowing over Talbiseh’, that they were posted on YouTube and a URL was given to the original site. The dope-sheet said: ‘AFP has been able to authenticate the city of Talbiseh.’ It also noted that they were posted on YouTube on 30 September 2015, ‘on the first day of Russian airstrikes’ but did not confirm that they were actually shot on that day. The other videos were similar in content and information, and were sourced to the ‘Daret Ezza Media Centre’ and the ‘Talbiseh YouTube Channel’. AFPTV also made available the emails that went between senior staff discussing the verification process. On 15 October 2015, two staff members discussed the bombing videos from Talbiseh; they noted that a water tank and a mosque with its minaret had been authenticated by Google Maps and the more senior staffer said to the other (translated from French) that it was fine to put out the video as long as the dope-sheet did not affirm in the shot-list that it was a Russian plane, as it was not possible to authenticate this. This person reiterated that the date could not be authenticated either, despite the Russian Ministry of Defence confirming that it had been bombing in the area and that ‘all the normal reservations about this type of video [will] be reflected in the dope-sheet.’
Hostage execution video (2 December 2015)
On 3 December 2015, AP covered the story of the IS execution of a Russian-Chechen hostage by putting out 13 seconds of a still of Magomed Khasiev. He was pictured in an orange jumpsuit, kneeling in front of a man in army fatigues who held a large knife. AP’s script said that it was posted on the internet by ‘Islamic State militants’. It added the usual provisos that the AP had no way of ‘independently verifying the content’ nor could it ‘routinely communicate with militant sources’. Then it added this strict additional instruction:
MILITANT VIDEO – GRAPHIC CONTENT; AP PROVIDES ACCESS TO THIS HANDOUT PHOTO TO BE USED SOLELY TO ILLUSTRATE NEWS REPORTING OR COMMENTARY ON THE FACTS OR EVENTS DEPICTED IN THIS IMAGE. THIS IMAGE MAY ONLY BE USED FOR 14 DAYS FROM TIME OF TRANSMISSION; NO ARCHIVING; NO LICENSING.
On the same day, Reuters put out 42 seconds of ‘a video released online by Islamic State’ in which the same Russian-Chechen hostage is paraded in the orange jumpsuit. The script said: ‘an Islamic State fighter … appears to cut the man’s throat and cut his head off.’ However, the various shots of the video did not show this – they were simply a collection of shots of the man speaking in Russian, conveying messages from IS threatening Russia and President Putin with attacks. At the end of the script, it said: ‘Reuters could not immediately verify the authenticity of the video.’ AFPTV did not put out any video or stills related to this story.
Interviews
In the interviews that were carried out with senior editors from the three agencies, a number of themes were canvassed, relating to the sourcing, verification and information published about the video content and also concerning the newsworthiness and acceptability of the content published. With regard to the provenance of the material, the interviews revealed that all the agencies were looking for material in the same places – YouTube channels, Twitter, Instagram and a range of other social media platforms.
Reuters
Reuters updates its social media policies regularly. In its policy document, ‘Reporting from the internet and using social media’ (Reuters, updated 2016) it reminds its journalists that they should not deviate from basic reporting behaviour:
Internet reporting is nothing more than applying the principles of sound journalism to the sometimes unusual situations thrown up in the virtual world. The same standards of sourcing, identification and verification apply. Apply the same precautions online that you would use in other forms of newsgathering and do not use anything from the Internet that is not sourced in such a way that you can verify where it came from.
However, these words do not really match the evidence in the Reuters TV dope-sheets pertaining to UGC video, which do not give out much information about ‘where it came from’. Reuters has a team of three producers in its Beirut office, who monitor these sites every day. According to Yann Tessier, Reuters’ Senior Editor of the Day:
They then pick their material. They are in contact with some of the groups. It is obviously a lot easier to speak to an FSA battalion that regularly posts material than it is to Islamic State which just doesn’t return our phone calls. Then it’s a question of what has become classical UGC verification – verify the video on the one hand, verify the information on the other. Go look for the oldest version you can find, use tools like the Amnesty International tools, and various other tools that will split it into key frames and do reverse images searches on the key frames. [Check on] Google Maps [and] triangulate the location.
Tessier explained in great detail how Reuters staff had checked all their material particularly carefully at the time that the Russians joined in the bombing. In particular, they looked for details such as whether or not a helicopter was Russian or Syrian as it was not enough that the camera footage included audio of people claiming it was Russian. Reuters took close-up grabs of the helicopter concerned and sent them to a Russian helicopter specialist at the International Institute of Strategic Studies so that they were finally able to clarify the country of origin, even though it had taken them a day to do so. Regarding this kind of UGC, Tessier said:
There is no certainty. Reuters was not involved in any part of that coverage. So all you can do is take the video that you have, and build the evidential picture that will then lead a jury of good journalists to declare it a Russian helicopter or not.
Tessier asserted that over the past two years, due to experience, the agency had come to learn which of the groups and ‘media centres’ were the more reliable. He added that a lot of the video had logos burnt on: ‘SMAART is generally more reliable than something with the black flag of Islamic State.’ When questioned about the Reuters scripts and shot-lists, which the data show is minimalist about the provenance of the material, he replied that they did not put out something generally
unless we are confident with it and it is proven beyond all reasonable doubt. Probably not quite that high, but getting there. We don’t tend to see much use in saying ‘we think it might be that, but make your own minds up’.
However, there was a sense that he thought there might be a need to be more precise:
It’s a good question – should we provide more, should we say more about the source, should we do a little thing like Storyful that gives the elements of verification … Our clients pay us to do the legwork. But it is a sort of very fluid situation and we have sort of discussed something like ‘verified by geo-location, tick; ambient weather, tick’ what have you – giving something a little bit more precise regarding the reliability of the source. But then, if we did not feel a source was relatively reliable – we would not include it. Our producers in Beirut – they have a dialogue with some, but not all, of these groups. (emphasis in original)
But even if the dope-sheets do not have the verification methods ticked off for clients to see, the same type of verification procedures are going on behind the scenes. Reuters also texts, tweets or emails forms to people to secure access to the footage. Tessier said that other broadcasters knew that Reuters did not ‘put something on the service lightly’ and they knew that a process of verification had taken place. Nonetheless, he thought some broadcasters were happy to pluck their own material off the internet unverified because ‘there’s often a kind of pack effect and you see “Oh the guys at Ruristan 24 have run it, bloody hell let’s run it too. If it’s good enough for them, it’s good enough for us”.’ But he thought the majority of broadcasters tried to act with as much due diligence as their resources allowed.
With regard to putting out hostage and execution videos, he was sanguine about the role of a news agency:
They’re very macabre debates. You want to give broadcasters enough to work with but – bearing in mind that our activity is business to business – we’re an agency. We are not, on the units that I’m on, taking decisions as to what the general public sees. I’m taking decisions as to what I think you at ITN would want to see, or the BBC or Argentinian TV and what have you. There’s two bits to that, there’s what they will end up airing, and then the little bit of context either side, so that they can understand, you know, that it really did end badly. We have in the past had requests to see the entire video, and people come to us and say ‘you know – we want to see the entire video because we want to report’, and in that case you just sort of send them the link.
However, he also said that he was aware of the growing risk of vicarious trauma, both to his own staff and to the journalists in the newsrooms of subscribers. He said, ‘We don’t run heads getting hacked off, we don’t run knives in their necks, we don’t run Jordanian pilots getting burnt to death.’
With regards to the Russian-Chechen hostage who was executed by ISIS on 2 December 2015, Tessier said one of the issues had been the man’s nationality.
We put out [a video] saying this is a man that ISIS said is a Russian. This is a man, who has a knife to his throat and who was later executed, and in the run-up to that they said he was a Russian, they said he was a Chechen. But our role as an agency is to be as transparent as we can with the sources, and indeed as transparent as we can with the elements of verification.
AP
AP also updates its social media policies regularly. Its ‘Social Newsgathering in Sensitive Circumstances’ document (AP, 2015) advises its employees to seek permission to access UGC and advises them not to precipitate coverage:
We should avoid asking individuals who are not employed by the AP to create content for us in hazardous situations. The AP does not use UGC without permission, so it is acceptable to seek permission to access content that’s already been created.
As we have seen from the earlier data, AP gave out more information to clients than Reuters about what was known regarding the sourcing and how much they could stand by it. When possible, they also named the source organisation and gave clients some information about the outlet’s political, religious or military leanings. AP’s Head of News Production (International Video) Beth Colson said the organisation had UGC editors in London, Cairo, Russia, Asia and the US. The Syrian UGC was found by the editors in either London or Cairo and discussed and verified between the two places. On top of this, they would call in ‘regional experts’ from their other offices to find out if anybody had expert knowledge of a particular issue or place. She said: ‘You know we don’t want to be in a position where we use words like “purported” because that’s not enough. So that’s why we do all the checks.’ Colson thought that the organisation had taken a clear decision to end obfuscation regarding what could or could not be confirmed.
Early on, a couple of years ago, we would use material that we would verify ourselves, to the point that we were certain regarding what we were looking at. But we would always – I think there was a line that we used to use in scripts that would give us a little bit of distance. But we decided well that’s not really what we want to do. If we’re going to put material out, then we need to have a process that it’s been through. So, is it verified? Yes it is. Who does it belong to? Have we asked for their permission? Yes. It brings clarity but it is also just being upfront about what you think you’ve got. (emphasis in original)
AP also used all the digital media tools available to check weather, geo-location, shot recognition and evidence from the Russian ministries about bombing targets. She noted that around 90 percent of UGC videos from Syria had the logo of the original source burned on them, so that helped with chasing down the verification. She added that media groups that always report from a certain location were generally more reliable because they ‘logically have the means to shoot in these locations’. She said they used a wide mixture of activist and mainstream media organisations as sources for videos, after asking permission. However, she added that some fall off the list. They used to take material from both UGARIT and the Shaam News Network, however they have since decided to ditch them as neither are ‘a recognized news gathering organisation’ but simply a collective that collates and publishes videos. AP now uses the information from these sites to help find the original videos. Some of the media centres give out blanket permission to AP to use their material but, with others, they ask permission each time. She said that another useful method of triangulation for verification was to check if, at a time a media centre put out a video of an airstrike, there were other nearby media centres/channels that had put out similar videos.
With regard to the message that sometimes goes out to clients telling them to credit the source, she said:
We think crediting is really important because it’s part of the transparency of showing where the video came from. In the same way as you’d credit on another story to show whose it is, where you got it from, the origin. Sometimes if somebody doesn’t want to be credited, then we don’t credit them. It’s always a conversation. It’s a case-by-case.
She confirmed that the organisation had very strict policies and forms for asking permission that had to be followed, and AP looked for written permission where possible. They had created ‘social media friendly versions of our permission forms. We’ve worked really hard to make them user-friendly because otherwise people can’t be bothered.’ She said that the instruction in the dope-sheets saying, ‘the AP cannot routinely communicate with militant sources’ meant that the source was IS. ‘Again, that’s an attempt at transparency.’ With reference to the times that they had written that they could not ‘independently verify’ the material, Colson argued these times were increasingly rare and, most often, this phrase meant the source was IS.
Sometimes media centres are not used if they are posting pictures of ‘fresh corpses’. And, regarding the hostage and execution videos, Colson was adamant about what AP would not show:
We wouldn’t use the moment of death. Anything that’s gratuitous. I think there’s always a fine line between telling the story and just grimness, and that’s a judgement with every story. I think we don’t use horribly graphic things. But as an agency we do use more graphic things than you would as a broadcaster because all broadcasters have a different tolerance depending on where they are in the world. So it’s about what’s useable.
As at Reuters, AP has long conversations about issues such as the Russian-Chechen hostage execution video on 2 December 2015. Colson said: ‘These conversations go to the top – it will go right to our heads of standards and various format leaders. They are lengthy and thorough.’ Also, she said they were decided on a case-by-case basis and on news judgement.
AFPTV
Policy decision making regarding social media UGC at AFP comes straight from the top, and therefore it relieves those lower down the hierarchy from having to make final decisions on coverage. At regional levels, the staff simply collect possible stories, conduct the verification and then pass the data up to Paris to finalize. On 3 November 2015, Florence Biedermann sent out an urgent email to all staff reminding them of their roles in checking material. This email was passed on to me. In part it read:
AFP’S clients are frequently aware of breaking news or significant UGC images through Twitter feeds, but they rely on AFP to verify the information and secure the image rights. Our clients rely on our core journalism principles to check and re-check the authenticity of content. At bureau and regional level, we need to develop our procedures to source and verify social network material quickly and efficiently. Any editorial question or doubt about a piece of eyewitness content should be rapidly escalated to the Central Redchef (Central Chief Editor).
According to Quentin LeBoucher, AFPTV’s Middle East and North Africa coordinator, the agency uses less UGC material from Syria because it has its own correspondents in the country. It retains a bureau on the government side in Damascus, with a photographer, a couple of text reporters and a video journalist. The company has also had a ‘regular contributor’ who was one of the founders of the Aleppo Media Centre, ‘so he was an activist at first, and so we treated all his videos as UGC at the beginning.’ After that, he was considered ‘a freelance correspondent’, who in 2015 won the Rory Peck Award. However, during the period monitored for material, this man was out of action due to injury. Other reasons cited for why AFP used less UGC than the other agencies were that they were more stringent with their verification methods and also more ‘practically cautious’ due to having lost a court case in 2013 in which AFP and Getty Images were fined $1.2m for the ‘unauthorized use’ of a photographer’s images, taken off Twitter (Recinos, 2013). Like the other agencies, they have template emails in multiple languages for securing permission to use material before publication. Wherever possible, they are in touch with the owners of YouTube channel sites, except for IS. When I asked if they were ever under pressure from clients to put out more material, Richard Sargent, AFPTV’s Europe Coordinator, said sometimes the TV stations just go ahead and use the material straight off the YouTube channels, and they ‘clearly don’t have the same policies as us in terms of authentication’. But mostly he thought the clients
appreciate the work we do because it takes a lot of time, as you can imagine, to make sure we authenticate. And that’s kind of what we sell to them because most of the time anybody is able to find a video on YouTube.
The finding and vetting of most UGC from Syria is done by the Nicosia office but the final notification to publish comes from the central chief editor in Paris. As previously mentioned, the verification is done by obvious methods – checking video details against Google Maps, previous footage, the number plates on cars, shop signs in the street, etc. to know that the place is correct. If people in the videos are talking, then Arab journalists at AFP might be called on to check their accents. Regarding bombing aftermath, this was checked against information from the Russian Ministry of Defence. But both Leboucher and Sargent admitted that it was not always possible to know the date on which something was shot. They both claimed that the policy was to explain what was and was not known, so that they were clear in the information that was passed on to clients, who could then decide in their own newscasts what to say about it. Sargent said: ‘It’s all very honest and transparent which is what we’re trying to do.’
Both journalists said that the AFP watched all the IS videos but they barely put them out any more and, when they do, it is usually just a grab of a video, used as a still. According to Sargent:
Obviously we don’t want to be a platform for Islamic State propaganda. And I mean the number of videos they produce is just huge. You know every day, almost every day, there are new videos. And most of them are just pointless you know. It’s just a terror message trying to be worse and worse, more horrible every time.
With regard to the IS execution video concerning the Russian-Chechen hostage (2 December 2015), Sargent said,
It was just really disgusting and you see some man being beheaded and usually they show the moment when they cut the head. And now they did and close ups on the veins and the neck. And after and obviously it was just horrible.
AFP did not put it out and reiterated that the man’s origins were uncertain and the Russians had not confirmed that they had a hostage taken. He said that regarding hostages more generally, it was AFP’s policy to only put out footage or photos if the hostage ‘is not in the situation where it’s embarrassing for him or humiliating’. He said they considered the family’s wishes first or the perspective of the hostage’s government.
Conclusion
The findings reveal that all three television agencies are grappling with the complications of handling UGC from Syria. They are all sourcing it from the same Syrian YouTube channels and they are all employing similar methods of verifying this material. The Reuters news agency put out the largest number of videos during the months of October and November 2015, while simultaneously giving out significantly less information to clients about the videos’ provenance. Reuters’ Yann Tessier believed the news agency was only sending out material that was checked and verified, and yet at the same time, the words purportedly, said to be and allegedly are in all scripts and shot-lists. He appeared to consider that the agency might move to a more clear-cut policy of ticking off verification methods, but he was making no promises. AP, on the other hand, put out half the material that Reuters did over the same period. Beth Colson admitted that AP had employed words in the past that gave the agency ‘distance’ from confirming the material’s origins. However, the agency had since changed this policy and now did more concentrated verification of the material and was being more transparent with clients about what they knew. AP also gives out names and information about the groups posting the picture and an explanation about their affiliations. AFP-TV put out video on only four days. It went even further in giving information to clients by also publishing the URL of the sites from which the original video was taken. Emails passed on to this author reveal that a number of checks and balances were done and finally the dope-sheets are quite clear about what was known and what was not confirmed.
This sourcing and verification show the enormous changes that have taken place in the production practices of news agencies due to the growth of social media and citizen journalism, along with the problem of not having their own personnel in place. ‘Amateur video’ has historically always played a part in agency life, but now it plays a much more significant role. Agencies pride themselves on being the first people into a combat zone and the last people to leave. But now this war is being covered third-hand and largely by activists. It is likely that the difficulties of being certain about the footage lessen its worth for clients. Perhaps it is used by broadcasters more as ‘wallpaper’ to cover general bombing references rather than as a valuable picture that can help explain what is happening in Syria. It is only rarely that a video or picture has become iconic in this war, and this is usually when it has been widely republished via social media. However, this can only be confirmed if interviews are followed up with the broadcast clients of the agencies. It would certainly be interesting to see if, in this ‘post-fact world’, broadcasters care less about certainty of provenance when using agency material. When news agencies, as prime drivers of the world’s media agenda use the words ‘purportedly’ and ‘said to be’ about material that they are carrying, then it should be worrying for all of us as viewers of these images further down the chain.
In terms of the graphic nature of some of the content, there is clearly a difference between AP and Reuters, on the one hand, and AFPTV on the other. Both Tessier and Colson were adamant that the Syria story had to be told somehow as it was newsworthy and this third-hand way was the only way it was possible to get access to fresh footage. They also both believed that it was the agencies’ prime job to make material available for clients and then for the clients to make the decisions about what they would or would not broadcast. They agreed that they would mark events like IS executions, but they would not put out the ‘moment of death’ picture. AFP, on the other hand, was clear that in limiting its IS and rebel footage, it is reflecting that it is not a propaganda arm of IS. But this is to negate the role of a news agency as a provider and enabler for broadcasters. When asked if clients ever chased them for missing material, Leboucher and Sargent did not seem to mind that the agency would be bypassed and that clients could simply go straight to the source on YouTube. This French importance placed on ‘media effects’ post terrorism outrages has since been taken up by other media outlets in France. On 27 July 2016, following a terrorist attack in Nice (14 July 2016) and the attack on a French church congregation in Rouen (26 July 2016), Le Monde’s editor Jérôme Fenoglio decided that the paper would no longer publish pictures of IS terrorists, to ‘avoid posthumous glorification’. Le Monde had already evolved its practices in recent times and no longer published extracts of IS documents or its threats. It said its practices were continuously under discussion to counter the effect of social networks and social media that reproduced IS’s strategy of hate.
In not putting out IS propaganda, AFP have not published material, such as the following soundbite from an IS supporter on the day after the Bataclan massacre in France on 14 November 2015, that was published by Reuters. Does the French public have a right to know about these threats, or should they be shielded from this knowledge, due to editors’ decisions about propaganda and a campaign of hate?
And if you are sincere to Allah in your worship and in your creed and are unable to travel, then operate within France. Terrorize them and do not allow them to sleep due to fear and horror. There are weapons and cars available and targets to be hit. Even poison is available, so poison the water and food of at least one of the enemies of Allah. Kill them and spit in their faces and run them over with your cars. Do whatever you are able to do in order to humiliate them, for they deserve only this.
Since this IS video warning was published by Reuters a number of new incidents that have involved terrorists using vehicles to kill people have taken place including in Nice in 2016 and in London in 2017. In research flowing on from this article, I will be exploring the comparative policies of different countries’ media organisations with regard to covering IS and its ‘events’. I will examine whether or not, in the world of social media, mainstream media organisations acting as ‘gatekeepers’ to knowledge is excusable or even viable.
Footnotes
Acknowledgements
I would like to thank Professor Charlie Beckett of the London School of Economics and Political Science, who hosted me in 2015–2016 as a visiting research fellow while I was carrying out the data gathering for this and another project.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Notes
Author biography
.