Contact Us
317-278-8011
755 West Michigan Street
Indianapolis, IN 46202

News from the Center for Digital Scholarship

Serial. A Case for Crowdsourcing Criminal Trials?

Printer-friendly versionPrinter-friendly version

Like many others I’ve fallen into the rabbit hole that is Serial.  For those not yet hooked, it’s a spinoff of NPR’s This American Life.  A real life who-done-it or perhaps better, a real life are-we-sure-the-State’s-case-proves-that-the-guy-who got convicted-actually-done-it. Serial’s creators describe the podcast as, “we’ll follow a plot and characters wherever they take us. And we won’t know what happens at the end til we get there.” Its first season presents a riveting 12-episodes that examine past and new evidence for the now 15 year-old murder trial of Adnan Syed, with some of the new “evidence” surfacing only as a result of the podcast.

While the unfolding of the case, evidence, and reanalysis is in itself interesting, I found all the extra stuffs that unexpectedly (to myself and the creators) surfaced fascinating.  There’s the general popular media frenzy (Facebook, Twitter, blogs, TV spots, news stories, etc.), a Talking Dead-like follow up podcast from Slate, Serial Spoiler Special, which follows each Serial episode with analysis and critique of the new information revealed but also the story-telling and analysis methods of podcast’s producers Sarah Koening, Dana Chivvis, and Julie Snyder. Finally there’s a reddit community and, WOW is there ever.  Here you can find primary documents (potential evidence?) and in-depth analysis from average Joes to criminal lawyers. The producers admit that the podcast was often only one show ahead in its research, which suggests that all of this hub-bub played a part in how they thought about the case and presented the story.

As I sift through the podcast provided analysis but also the reddit community offered evaluation (some of which is really involved, see visualizations of cell phone evidence) my digital librarian self considers crowdsourcing.  That’s what’s happening here, passionate individuals, working for free (as is often but not always the case in crowdsourcing) towards a common goal.  Is this a better alternative to trials? What does a jury of thousands/millions look like?  What about big data? Would our understanding of the case alter with combined, culled, coded, and analyzed crowdsourced and police/lawyer gathered data?  As I’m want to do, I consider some negatives first:

  • Changing our publishing systems has been hard enough, change the legal system?! Whoa Nellie.
  • Enter new professions, Trial Lobbyists and Defendant Advertising Agents aka wealthy defendants have more opportunities, but that’s not new.
  • Gaming the system and Crowdturfing (people paid or with motive to make specific often inaccurate comments in online reviews)
  • Justice through popularity contest? 
  • Jury’s have 100% consensus. Not possible with a jury of thousands/millions.
  • Exacerbate issue of impartial tribunal. More people will know about the case.
  • Vocal “jurors” are known and therefore potentially targeted, and because of this,
  • There is no crowd to source for mob cases.
  • How do we evaluate evidence that is collected during the crowdsource phase?

But some positives:

  • Large amounts of evidence reviewed faster by lots more people.
  • Assuming most humans are reasonable ones (yeah I don’t know), the “right” decision will be found by majority.
  • Alleviate issue of impartial tribunal. So many people will know about the case that everyone is at a similar level of knowing.
  • Surfacing of witnesses and evidence that never would have surfaced in a traditional trial.

Serial producers and interviewees remind listeners that winning a trial is as much (if not more) about telling a credible story with the known evidence at hand as it is the truth.  This sometimes means ignoring evidence that doesn’t fit or support that story.   Would analysis of the big data collected from a crowdsourced murder trial eliminate the story, drill down to the truth? Is it a more data-driven, scientific method of justice?  I think not.  The data is still coming from story-loving fallible humans. BUT a verdict pointed to by thousands of stories, bits of evidence, gathered from numerous, disparate, unconnected, individuals with varying perspectives, that’s intriguing at least.



Last updated by andjsmit on 11/21/2013