In an ideal world, the people who recover video evidence from devices for legal purposes would be fully conversant with how to perform this task in a forensically sound manner so that all parties and the court could benefit from the value of the images. In the real world, this approach is neither universally affordable nor practical. This article, while not intended to be a comprehensive treatment of the subject, discusses the challenges involved in collecting video evidence for use in investigations and the courtroom, both from devices and social media.
Sources of video evidence
Perhaps a useful starting point is to consider the many sources of video evidence. At the domestic level, video evidence may be recorded by commercial, residential, or municipal CCTV systems. It may be recorded by dashcams in private and commercial vehicles, including police vehicles. Video evidence may be recorded by dedicated documentarians such as crime scene unit members. Additionally, drones are becoming a common source of aerial video evidence for investigative or personal purposes. Last though certainly not least, smartphone video has become a constant source of visual evidence that is frequently uploaded to social media and other open source sites (as of 2023, there are an estimated 7.33 billion smartphone users in the world).
In international investigations, though there are other sources, smartphone video recordings are a prevalent source of image-based evidence, and they are recorded by a broad range of civil society members, human rights investigators, investigators from the International Criminal Court’s Office of the Prosecutor (OTP), victims, and combatants. This is well represented by the images emanating from Russia’s invasion of Ukraine.
Authentication steps
Given the vast volume of video and other image-based evidence that exists, we must consider how to best acquire it so that it can be profitably used in the search for the truth. To properly authenticate video evidence, the party tendering it in court should lead evidence to cover three important authentication topics:
a) Details of the camera, CCTV system, or other device that recorded the images. This should include the brand name of the device; model number; the number of cameras (CCTV); whether the cameras record automatically, are motion activated, or are manually operated; infrared capability; frame rate; audio track; time offset, if any; etc.
b) How the video was exported from the original source (device). This should include the method used (there are many methods that can be used to export video from devices), which cameras were downloaded in a multiple camera CCTV system and why, and the time frame that was downloaded. Where an audio component was present, details as to downloading that component of the evidence are essential. Further details should include whether the video was recovered in a lossless or lossy format (eg., whether the evidence was digitally compressed during exporting).
c) Whether the video recording was edited or altered in any way during the exporting process or otherwise after acquisition.
When video evidence is recovered by qualified people promptly in this manner, there can be justified confidence that the available video evidence has been captured in a forensically sound manner that can later be viewed, analyzed, authenticated, and presented in the courtroom. However, only a small percentage of available video evidence is recovered in this robust manner.
Problems acquiring video evidence
In practice, video evidence is typically obtained by one or more of the following methods.
A common scenario is where private citizens or business owners download video from their phone or DVR and provide it to the police either as a physical exhibit (flash drive, disc, hard drive) or via electronic submission to a dedicated portal provided by a police or other agency. Since the police cannot export all the video evidence that exists, downstream users are dependent upon the choices made by the person doing the downloading. Absent specific instruction, it is unlikely that the three points noted above will be adequately covered. If the submitted video evidence is reviewed early on by competent people, it may be possible to address any shortcomings by obtaining a statement from the evidence provider or re-downloading the video evidence in a forensically sound manner, particularly when the evidence may be pivotal in a case. This would have to be done promptly before the evidence is lost through overwriting or manual deletion.
Some police agencies have dedicated portals that they invite private citizens, business owners, and others to use to submit video evidence to the police for consideration. Evidence.com is a commonly used domestic portal, though there are others. While receiving some video evidence is better than receiving none, the problem with relying upon non-investigators to gather and submit video evidence is that most people have no training in how to properly capture video evidence in an uncompressed manner and in the original file format, what cameras to download, what time frame is relevant, etc. Nor should the police expect them to do so, yet that is perhaps the implicit message given in offering evidence portals to the public.
The OTP recently (May 2023) unveiled a new portal for the global submission of video and other image-based evidence for matters within ICC jurisdiction. The OTP provides little guidance and virtually no restrictions on the kind of media that can be submitted, thus facilitating the problematic submission of evidence that will likely not be able to address the points raised above and may therefore fail for want of authentication. In my upcoming book, Image-Based Evidence in International Criminal Prosecutions: Charting a Path Forward (to be published by Oxford University Press in early 2024), I recommend a more robust process for receiving public evidence submissions. These recommendations include establishing file format parameters, requiring adequate photographer identification and authentication details, and a direction for retention of the original media when safe to do so.
A more commonly used method of making video recordings available to the public is where people upload recordings to social media and other open source sites. If the police or other interested parties happen to discover such recordings, they will typically face authentication challenges. Social media and open source recordings are usually stripped of their metadata during the uploading process making authentication and reliable content analysis problematic. Many such submissions are anonymous which virtually eliminates any hope of obtaining access to the original recording. Further, the longevity of uploads on sites is subject to the whims of the person who uploaded the media and decisions made by the site itself (YouTube, Facebook, TikTok, etc.) as to whether they will purge certain submissions. A great deal of valuable video evidence of war crimes and other atrocities has been removed by hosting sites, resulting in the loss of important evidence.
Solutions
What then is the best way to secure video evidence that will meet authentication requirements? This requires that a person with the requisite technical knowledge export the video files in a forensically sound manner and document the three authentication steps noted above. Using this method effectively eliminates the need for the homeowner or business owner to testify regarding the authenticity of the recordings as that can be covered by the person who does the video recovery. This is a win-win for both the evidence provider and those that will rely on the video evidence in legal proceedings.
Since the ability of policing agencies and others with the requisite technical knowledge to personally export video evidence is limited, the second-best method is for private citizens and business owners to submit video recordings to the police either via physical exhibit or through an evidence portal with full documentation of the three authentication steps noted above. This documentation is essential when determining whether the media can be authenticated and if reviewed early enough, further steps may be undertaken to shore up any shortcomings. In addition to these steps, a caution to retain the original recording would be advantageous.
The democratization of evidence gathering and the plethora of cameras recording so much of what occurs in the world is both a blessing and a curse. While it provides vast amounts of potential evidence that can be instrumental in determining the truth, it often poses significant authentication challenges such that in many cases the evidence cannot be used in court. Therefore, while the public submission of video evidence is not something to be discouraged, providing some parameters regarding technical specifications and accompanying authentication documentation will go a long way to making these submissions of more evidential value to downstream users.