The Epistemology of a Podcast

Unless you’ve been living under a rock for the past few months, you’ve probably heard of Serial, the podcast sensation taking the internet by storm. Hosted by Sarah Koenig, the podcast is a serialized account of the 1999 murder of Hae Min Lee, a 17 year-old senior at Baltimore’s Woodlawn High School. In the style of a true crime drama, Serial revolves around the fundamental question: whodunit. But in this case, there is also a possibility of wrongful conviction. Koenig’s entry into the story comes through the family of Adnan Syed, Lee’s ex-boyfriend, who was convicted of 1^st degree murder in her death. Koenig set out on a year-long investigation of the case, pouring over trial records, interrogation transcripts, even the prosecutor’s evidence files. Was Syed wrongfully convicted? If he didn’t do it, than who did? 

Serial, a spinoff of the popular NPR podcast This American Life, has attracted an incredible amount of media attention. Time, New York Times, and The Atlantic have all covered the podcast. Slate began its own meta-podcast to discuss Serial each week. Some of this coverage has focused on allegations that racial prejudice pervades Koenig’s reporting. Because the case focuses on the murder of a Korean-American teenager by her Pakistani-American boyfriend, Koenig (as a white journalist) is reporting on communities in which she is a cultural outsider. Others have criticized Koenig for making herself, as narrator and amateur detective, the protagonist of someone else’s story.

I discovered Serial after a few episodes had already aired. I have listened to all of the available episodes thus far, and for the most part have enjoyed listening to the show. Today, the highly anticipated final episode of Serial will go live. Before I listen to the final episode, I wanted to share a #histstm perspective on the show and its surprising success.

Reasonable Doubt?

At its core, Serial is a reflection of the nature of truth. The show is fueled by uncertainty, as Koenig brings the listener through a series of “buts,” “what-ifs,” and “wtfs” that will make your head spin. Adnan was primarily convicted based on the testimony of Jay, his friend and alleged accomplice in disposing Hae’s body. Jay’s testimony is riddled with inconsistencies. It doesn’t help that Jay provided his testimony on four separate occasions: two pre-trial interrogations and two times on the stand. Koenig meticulously pokes at these inconsistencies, in hopes that by picking apart Jay’s testimony, the case against Adnan will unravel.

What struck me about this was how little the testimony’s inconsistencies bothered me. It seemed intuitive to me that if you told a story four times, under great pressure and in extraordinary circumstances, that it would change a little each time. Yet, it is these inconsistencies that drive Koenig’s uncertainty, as well as the entire Serial narrative. Her discomfort seems to grow out of the fact that testimony used to convict someone of first-degree murder shouldn’t be full of holes, and potentially, full of lies. Such uncertainty appears to be an assault on our standard of justice, and in particular the belief that we must prove guilt beyond reasonable doubt.

The question then becomes: how much doubt is “reasonable” in the context of the criminal justice system? There are piles of scholarly work on this topic, but as layperson it was not something I had given much thought. In Episode 8, “The Deal With Jay,” Koenig interviews Jim Trainum, a private investigator, former homicide detective, and expert on false confessions. I think that their conversation really gets at the heart of the issue. Trainum, while admitting that the inconsistencies were troubling, also acknowledges that investigators were “better than average” in handling the evidence. He explains that the detectives in the case didn’t push on Jay’s inconsistencies, the way that Koenig is, out of fear of creating “bad evidence.” Jay was the prosecution’s star witness: the entire case hinged on his testimony. If the investigators pushed too hard, there would be nothing left for them to use, and their case would fall apart.

Koenig balks at Trainum’s use of the expression “bad evidence.” “All facts are friendly!” she exclaims, “You can’t pick and choose.” Trainum responds by explaining that for prosecutors, the goal was not to get at absolute truth, but to build a strong case. Koenig doesn’t back down: “How can you build a good case, how can he be a good witness, if there is stuff that is not true or unexplained?” Trainum concludes by suggesting that in any case, there will always be things that are unexplainable. He also points to the possibility of confirmation bias: the prosecutors were looking for facts to support the theory they already believed to be true.

In a weird way, Trainum’s explanation brought me back to Kuhn’s explanation of normal science: that as evidence accumulates, the underlying assumptions of a scientific theory go unquestioned.  It is only when enough anomalies accumulate that scientists will begin to question the tenets of their current paradigm. For Koenig, any anomaly or inconsistency should be enough to throw the conviction out the window. But as Trainum points out, there are important consistencies that check out, and give credence to the theory that Syed is guilty. In the absence of alternative explanation, it becomes a compelling case.

The CSI Effect 

The CSI Effect! Source: https://sites.google.com/site/effectofcsi/

Another unsettling feature of the case is the absence of physical evidence. In Episode 7, Koenig interviews lawyers in the “Innocence Project” at University of Virginia School of Law. The students reviewing the case are unanimously unconvinced of Syed’s guilt. One student claims that there are “mountains of reasonable doubt.”

The sticking point for these students is the absence of physical evidence. Although a liquor body found near Hae’s body was scraped for epithelial cells, they were never tested. Fibers found in the soil around her body were only tested against a very small number of fabric samples. There were no DNA tests performed on the body itself. Some of the forensic reports from the case appear to be missing from the records.

Could these students be suffering from the “The CSI Effect?” This expression is used to describe an increasing demand among jurors and the wider public for forensic evidence, which is attributed to the popularity of television crime shows like CSI. Studies have shown that frequent viewers of CSI may place a lower value on circumstantial evidence, and many lawyers argue that there has been a significant shift in the behavior of juries. Could Koenig (as well as the listener’s) uncertainty be a side effect of 15 years of television that glorifies forensic science? Is one person’s testimony enough to put someone away for 1^st degree murder?

Interestingly, there was actually a cutting-edge technology that was introduced into the trial: the cell phone. Syed’s case was one of the first in Baltimore Country to use cell phone records and tower pings as evidence in a criminal trial. Syed himself had only purchased a cell phone three days before the crime occurred, a fact that was used to throw suspicion onto him. The ways in which the cell phone records and cell phone tower pings confirmed Jay’s testimony lent significant strength to the prosecution’s case.

Reddit and the Radio

Lastly, I want to reflect on why Serial’s surprising popularity. After all, as a genre serialization is nothing new. Literary serials date back to the 17^th century, and surged in popularity during the Victorian era with Dickens’ The Pickwick Papers. More recently, serialized stories were the bread and butter of television networks everywhere. Daytime soaps, prime-time dramas, even some reality shows are versions of the serial form. And not that long ago, before Netflix allowed us to mainline entire seasons of the Gilmore Girls in a weekend (no judgment), viewers had to wait with bated breath for the next installment. Maybe Serial’s charm derives from being an old-school radio show in an era of immediacy and on-demand entertainment.

Another interesting phenomena to pop up in Serial’s wake is the creation of a subreddit dedicated to the show. As of the time of writing, the subreddit has almost 30,000 followers. Posters propose alternative theories, map cell phone pings, and share detailed timelines of the crime. Important figures in the case, including Adnan’s brother and Jay, are rumored to be posting in the group. The group’s activity reminds me a little of citizen science, where the collection or analysis of large amounts of data is crowd-sourced by non-professionals. Perhaps this collision of an old-school serial and the universe of social media can help explain the show’s popularity. The pleasure of waiting for the next installment is further intensified by discussion and speculation within an online community.  

For weeks, listeners have expressed anxiety about the ending of the show (it even inspired a Funny or Die parody). After all, Koenig is investigating a real case, not reciting a script. And in real life, we don’t always succeed in finding the truth. Koenig has acknowledged this pressure, but insists she has no special responsibility to provide listeners with a satisfying conclusion. We will soon find out!