The Washington Post has been reconstructing the biggest news stories of 2020 through a range of open-source visual forensic techniques and it has more in store for 2021.
The visual forensic team was officially launched last September after informally working since March. They did a piece on how Iran is building a massive grave for coronavirus victims, covered the events surrounding the death of George Floyd, and more recently, the siege on the US Capitol on 6 January 2021.
The formal launch meant continual investment and training in the investigative skills required to explore these stories.
The team is lead by senior producer for The Post Nadine Ajaka, consisting of around half a dozen video reporters, plus a wider, rotating group of investigative journalists, collaborating closely with the rapid-response investigative team.
"It's a good marriage of traditional reporting and this new, digital armchair research," Ajaka told Journalism.co.uk.
In the series of around a dozen videos that have been released so far, it typically tries to pair up footage from the event with material sourced afterwards.
In the example of the crackdown on protesters before Trump’s photo op in Lafayette Square, The Post acquired police radio chatter and placed that upon user-generated content of demonstrations at the scene. Meanwhile, animated geolocation maps run next to the video footage to give an aerial view of key developments.
"It gives you a 360 view of what is going on, and adds context or nuance to an event that previously isn't there when you take all these things and look at them together," she continues.
Other videos have made use of facial recognition technology, publicly available maps, text message exchanges and their own procured interviews to piece these events together.
"We didn't start out knowing that we want to focus on the 41 minutes after the breach when the rioters came close to the lawmakers," Ajaka reveals.
"We were looking at a wide expanse of time and through that realised we had a video that we can map, and we can show that using geolocated video.
"We didn’t have as much video from the mob that was being cleared. That was important to know but that was not the visual forensic story we wanted to pursue."
All videos start with a detailed timeline of events. That is narrowed down into a tighter window, where visuals are pulled together across that time frame. Then it is a matter of identifying what exactly is happening, what can be geolocated, what is the source, has it been verified, and sometimes, selecting the best accompanying visuals to be engaging and provide the most clarity.
Besides fancy graphics, visual forensics can be impactful and way to hold powers to account. The project takes inspiration from Bellingcat and BBC Africa Eye, two examples which have demonstrated this well.
"We have a couple of questions that we ask [to determine] whether it's a visual forensic piece. Is it a visual story? What are we showing? For these stories, they play out on video and on social media," Ajaka explains.
"But we have other questions: are we holding a group, organisation or a person accountable? Is there new information that we can show using open-source investigative tools? Is there going to be impact?"
The timeline of Trump's photo op at Lafayette Square picked up a duPont award, recognising 'outstanding audiovisual reporting in the public interest'.
Ajaka recognised the efforts of her colleagues with this award, namely Eric Rich and Mary Pat Flaherty, leading the rapid response team, which paved the way for this style of investigation. Having a diverse team with a strong representation of women of colour, she added, has informed the range of storytelling too.
Looking ahead to 2021, the formalisation of the visual forensic team signals bigger and more ambitious stories to follow.
"It is a mistake to create this bifurcated world of text and visuals when so many of these huge news stories play out in a way where they are caught on camera.
"Visuals can add context, nuance and a lot of times, they are the story. I think we have realised that and now we provide revelatory investigations for our audience using visuals.
"There’s still a lot to do surrounding the riot on 6 January and we're curious about other ways we can reconstruct a scene using 3D modelling, or other ways to visually tell what has happened that is less reliant on visuals, but still stays within this reconstruction space."
Want to receive journalism news and job updates straight to your phone? Subscribe to Journalism.co.uk on Telegram on our jobs channel for latest job opportunities and our news channel for a weekly digest every Monday morning.
Free daily newsletter
- AI text-to-video technology can help newsrooms get more mileage out of their copy
- How do narratives spread, and what does this mean for how we report the news?
- 100 days of war: What has international media learned from covering Ukraine?
- The Washington Post's pandemic-born visual forensic team is here to stay (and still works remotely)
- Yusuf Omar, co-founder of Seen, on wearable technology in journalism