Problem Sets: What challenges are you facing in using video documentation as evidence?

6 posts / 0 new
Last post
Problem Sets: What challenges are you facing in using video documentation as evidence?

What challenges have you faced? How did you address those challenges? What lessons did you learn? Share your experiences, resources and ideas by replying to the comments below.

For help on how to participate in this conversation, please visit these online instructions.


Should Evidentiary Interviews be on Camera?

The answer to the question, “Should we videotape interviews of witnesses and survivors of human rights abuses for advocacy?” seems to have a clear answer: Yes, so long as it’s within the confines of safety, security and ethics. A primary reason being that personal stories propel our advocacy efforts.

I have also asked a number of investigators and attorneys, “Should we videotape interviews of witnesses and survivors of human rights abuses for evidentiary purposes?” The answer is not so clear. As a matter of fact, the answers are all over the board. So I wanted to set out an initial list on when it might make sense to use a camera when interviewing for evidentiary purposes and when it doesn’t hoping others will add in their thoughts too.

One key note first: Whether or not we should or shouldn’t videotape an interview in some ways doesn’t matter as activists on-the-ground will, absolutely, collect video testimony. Knowing this, WITNESS would like to provide the activists we work alongside with good solid guidance to:

  • Help them think about whether to pick up a camera or not; and
  • If they do, help them interview safely, ethically and effectively.

So here’s the first question we discuss at trainings when talking about whether to videotape interviews for evidentiary purposes and the list of why or why not.

QUESTION: Is video the most appropriate choice to document an interview for evidentiary purposes or is it better to take a written statement or make an audio recording?

Initial Thoughts on Reasons to Use Video:

  • Timing: In many human rights situations, time to talk with witnesses is limited. Video can serve as an efficient way to capture information about an incident that can be analyzed later. For instance, UN investigators had 7 1/2 hours to investigate the alleged sarin gas attacks in Ghouta, Syria in August of 2013. Given the limited access to witnesses – and the limited chances of relocating witnesses at a later date - video served as an efficient way to capture interviews to be thoughtfully analyzed later.
  • Credibility Assessment: A video taped interview will more easily allow members of an investigation or legal team who are not present in the field to assess the credibility of a witness. Videos show people’s physical features, voice and background that add valuable information to the factual information provided by the video.
  • Limited Interview Opportunity: Sometimes a key witness may not be available for trial because of imminent death, relocation or other extra-ordinary circumstances. In some jurisdictions, and in specific circumstances, portions of an interview may be allowed in court without the possibility of cross-examination and therefore, a videotaped interview statement may prove very helpful.
  • Video is Often Good Lead Evidence: Since video taped interviews are not likely to find their way into the courtroom, there is often no harm to a case caused by interviewing witnesses because the initial on-the-ground interviews tend to provide good leads to investigate versus being key courtroom evidence.

Initial Thoughts on Reasons Not to Use Video:

  • Timing: An investigator, analyst, lawyer can quickly scan written interview summaries or transcripts to find relevant information. With video, they must review the entire video to find the information they need. This can prove to be very time consuming so if you use video, it is best to: have clear and complete metadata; and include a summary of the interview and transcriptions when possible. [Note: Transcriptions of video taped testimony tend to be very time consuming, and usually costly, and therefore should only be done when valuable.]
  • May Undermine Credibility: If an interview is recorded, and then weeks, months or years later, the interviewee is asked to testify in court, inconsistencies between the recorded interview and in-court testimony could undermine the witnesses credibility.
  • Security: A video file of a person’s testimony may increase the risk of security threats.
  • Ethics: Some people are either uncomfortable providing information about human rights violations on video or experienced extensive trauma and the recording of an interview will only enhance the re-vicitimaztion they experience. Take special care and seek advice from psychosocial support experts when interviewing witnesses and survivors of human rights abuses.

After discussing the above, we then ask activists, “If you decided that a video camera is the most appropriate recording method for the interview, how can you do it safely, effectively and ethically?” This is always a dynamic conversion.

I would love to hear anyone’s thoughts on this follow-up question too. I will also post our thinking on both questions our blog later this year.

Recording of our Google Hangout on questions/challenges

Thank you Abby, Kelly, Wendy, Mayss, Enrique and Frances for joining our Google Hangout today! (my apologies for any confusion I caused by accidently creating more than one Hangout - still trying to figure out this tool)

The discussion on the Hangout focused on 3 questions/challenges that are currently facing practitioners:

  1. When you've captured the video, how do you store and manage it before it is uploaded to a secure archival system?
  2. Should activists be interviewing witnesses on camera or off? Or, what should be considered when making this decision? Kelly posted her thoughts on this question above.
  3. Are there situations in which authentication of a particular video isn't necessary?

You can watch the recording of the call below and add your thoughts, ideas and advice by replying to this comment. Thanks!

- Kristin Antin, New Tactics Online Community Builder



One question raised in the hangout yesterday related to data management systems for videos collected, both in the short/medium-term (when the data has first been collected) and the long-term (when the data is archived for later use).  This is an issue we are exploring at the eyeWitness to Atrocities Project in relation to footage we collect.  I think one reason it is challenging to find an appropriate management system is that the system needs to accomplish multiple different objectives.

1.  Store and organize:  The system needs to allow the user to efficiently catalogue, find, and retrieve videos.  It also needs to safeguard the integrity of the data and maintain the chain of custody by allowing the organization to limit and keep a log of access.

2. Analyze:  The system also needs to facilitate user analysis of the videos.   The type of analysis to be conducted will vary depending on the situation and types of violations.  The user may want to aggregate footage to show, for example, everything collected in a specific geographic region over time or on a specific date, related to a specific party or perpetrator, or targeting certain categories of victims.  The user may also want to identify if there are patterns in the footage, for example, links between individuals or repetition of specific violations.

3.  Present and share:  When offering videos as evidence, the user needs to be able to share the footage with investigators or prosecutors.  Therefore, the management system needs to store and export the information in a way that is technically compatible with whatever evidence management system the investigators or lawyers are using.  Additionally, related to the analysis point, the information needs to be compiled and presented in a manner that makes it meaningful to investigators and lawyers.  In other words, it has to tell them what they need to know about the data.  This too may vary by case and jurisdiction.

There are many commercial products available that address some of these components.  Law enforcement agencies and law firms use various evidence management or litigation software that facilitate cataloguing, organizing, and retrieving videos.  However, they include numerous features that are not relevant, and they can be limited in their analytic capacity.   Similarly, there are other important tools that facilitate analysis or visualization of data, such as Sentinel Visualizer, Ushahidi, or the Rashomon Project.  These are very useful platforms, but also tend to focus on specific aspects of analysis.

I’d be very interested in hearing what management systems or data analysis tools others are using and have found effective.

Hi Wendy

Hi Wendy

For an example from WITNESS, check out the comment I posted in the Managing Video for Evidence thread (NB: We aren't collecting video for legal evidence and analysis, but more for advocacy and production use). 

As you mention, there are many commercial (and, I might add, free and open source) products available. They all serve different industries and purposes, so one has to do a lot of research to identify the one that best suits one's needs (or build it from scratch).  At WITNESS, we went the latter route, by building a database using an off-the-shelf database application (FileMaker Pro Server) and a slew of custom scripts that incorporate a lot of open source tools that enable us to manage our video collection.

We don't do analysis on our files in the way you describe, although via human descriptive cataloging, we would be able to identify and retrieve videos in our collection based on location, date, etc.

In terms of the "present and share" issue, I think the key thing is to use existing metadata standards. In a way, it doesn't matter what management system you use that works for you, so long as it can export your data in according to a known structure and in a form that is readable by other systems. So for example, the metadata in our database is compliant with PBCore, a standard for audiovisual production assets developed by public broadcasters in the US, and MODS, a descriptive standard developed by libraries. We can export records that comply with those standards in XML format, a text-based mark up language that is pretty universal.  

I'm not aware of what metadata standards are employed by the evidence management community (if someone knows, I'd love to hear about it!). If you can map your data to an existing standard, it makes it easier to exchange data with others in your community. Even if everyone uses different standards (or a different combination of standards), the fact that they are documented means that you could potentlally "cross-walk" between different standards in order to exchange information. 

Staying Safe: One Option from Amnesty

This is slightly off-topic, but the issue of personal security remains ever-present in this work.  Videre est Credere has done a lot of work in this area in terms of training citizen journalists and human rights documentarians about how to take video clandestinely but also safely and what to do in the event of discovery.

Just a short plug for an important new app by Amnesty International - the panic button.

The app has activists designate three emergency contacts. In the event the individual gets into trouble, s/he can rapidly press the phone's power button, which sends an SMS distress call to the person's network. When the GPS function is enabled, the message includes a map showing the user's coordinates.  The location updates can be sent every few minutes as well. 

This mobile app runs on Android phones:

This is a great example of activists working with tech companies in order to harness the power of technology to enhance human rights protections.