Serial and the perils of success
Also: fake documentaries, assignments in the age of AI, I talk about Tombstone on the radio
November is the dumbest month. First the end of daylight saving time, which coincides with the onset of Oregon winter, six months of low gray skies and drizzle. I grew up in Arizona, the only state besides Hawaii with the sense to resist this mishegoss; four-thirty sunsets, my ass. Then there was the election, which I’m not going to talk about except to say it didn’t make the vibes around here better.1
In retrospect, early November may not have been the best time to schedule a multi-class viewing of In Cold Blood. It’s a good movie, well-made and -acted, but it’s grim and dreary, and not just because of the content—the black-and-white doesn’t help, nor does the fact that the murder scenes were filmed at the actual murder scene.2 It’s also more than two hours long and ends with (spoiler alert?) a man being hanged.3
Bleakness notwithstanding, it was interesting to teach. I’d seen the film before, but had never taught it, and you never really know a work until you teach it. I don’t get to teach movies much; I’ve only taught one full-length film before, Werner Herzog’s Grizzly Man, a movie I love and have subjected multiple classes to. But that’s a documentary. In Cold Blood isn’t. It just tries to seem like one.
The film came out in 1967, two years after the book, and the filmmaker, Richard Brooks, took a novel approach to the challenge of how to adapt an unprecedented book to the screen. One of the main points we discuss in class is that the movie is basically one long reenactment, an approach that has since become a cliche in true crime texts. We talk about how hard the film tries to seem real: shot on location, anonymous (at the time) actors, the black-and-white itself.4 And yet. The actors are actors, not the real people. The actions are scripted and staged. It’s not the actual events; it’s a carefully crafted illusion of the actual events, constantly trying to convince you it’s the thing itself.5
The second half of the course is about how true crime changes once it’s translated to non-textual media. In Cold Blood is sort of the bridge, since it’s the same story (more or less) told in two forms. From it, we move onto YouTube videos, a TV series, and podcasts. I change up those a little bit every time I teach the class. Other than In Cold Blood (the book), the only thing I always teach is Serial.
As I’ve said ad nauseum during class discussion this last month, Serial was the In Cold Blood of podcasts. Both were genre-defining works in multiple genres, invented a new form of storytelling, were first published in serial form in major American media outlets, and captured to an uncanny degree the America of their era. There are a whole lot of other similarities I’m skipping for the sake of brevity.
As I write this, I’m about to finish listening to the first, canonical season of Serial for what I think is the fourth time. The first was ten years ago this fall, when everybody was listening to it. I have a hard time conveying how much of a phenomenon it was to my students, who were too young to remember. Serial was the first real mainstream podcast, in large part because it came out just at the right time; by 2014 everyone had an iPhone, and the Apple Podcasts app was added to iOS the same month Serial debuted. I show my class interviews with the host about its massive success, and the SNL parody. But I’m not sure if they get how big Serial was. You sort of had to be there.
That success is also why Serial is such a mess. On a fast re-listen, its failures are very obvious. In fairness, that’s not how it was intended to be listened to; the show was designed to air once a week, a decision that informs pretty much every aspect of its making. But the show started airing before it was finished, before the creators even knew where or how the story would end, and the last few episodes were recorded after it had already become a nationwide sensation. The result is that they’re still learning new information through the last episode, many of the principals are either still coming out of the woodwork or changing their stories, and it has no real ending. Serial spends ten hours re-investigating an old murder, casting doubt on the case, presenting a confusing jumble of evidence and reenactments. Then it just hands it all over to the audience and tells them to figure it out.
That was a big part of its legacy: it was among the first major true crime texts to place the audience in the roles of both jury and detective. The Serial subreddit is a rich tapestry of psychosis to this day; Adnan Syed’s Wikipedia page was the subject of so much controversy and vandalism that it had to be blackholed. Pretty much every contemporary true crime podcast, web series, or video channel emulates Serial’s invitation to audience participation, and now whenever there’s a high-profile murder, the “citizen journalists” aren’t far behind. Last night, on a flight back from Thanksgiving travel, I perused the in-flight true crime options and wound up watching half an episode of #cybersleuths, a Paramount TV “docuseries”—that prefix means less with every passing year—about the Idaho student murders. Can’t say I recommend the show, but, like most contemporary true crime, you can find Serial’s fingerprints all over it.
Anyway. The class is almost over now. The last week, we’re watching an episode of American Vandal, a juvenile Netflix true crime satire from 2017 that is, somehow, a masterpiece.
The last time I taught this big true crime survey course was in early 2023, the infancy of ChatGPT. I don’t think a single student used it then. Less than two years later, I’d guess half use it (or one of its siblings) to some degree for every assignment. You can still tell. Or at least writing teachers still can. But give it another year.
So far, it seems like the upshot of most official academic advice on AI has been to recommend that faculty embrace it. More specific suggestions, when they exist, usually boil down to reimagining decades of writing pedagogy on the fly, mostly by assigning a bunch of busy work. Of course, in practice that lowers expectations to almost infantile levels and jacks up grading loads, but not to worry—soon, presumably, AI will be able to do the grading, too.6 Further complicating the situation is the fact that in most cases, you can’t fail an assignment for using AI because it’s almost impossible to definitively prove. (And the scale of its use makes that impossible, anyway—no professor’s going to pursue academic misconduct cases against half of their students.)
Instead I’ve been fighting the inevitable, rewriting assignments to make AI use harder, or at least to make it result in lower grades. A couple of the more obvious tells of AI-generated writing right now is how much it struggles with concision and specificity. It’s great at writing long, vague passages that sound OK if you don’t read them carefully. So all quarter I’ve been revising assignments to be shorter and to rely more on class discussion or students’ personal response to a text, just to make them harder to automate.
For that reason and others, I think the perfect assignment in 2024 is a short presentation, either in person as part of a group (if the class size allows it) or as videos. It develops directly translatable skills, requires teamwork and diplomacy, sounds easy but is actually hard, develops critical thinking, and is probably more fun to do than essays. They’re a lot more fun to grade. Plus I get to learn about all kinds of wild true crime media I’d never heard of.
A few recommendations, true crime and otherwise:
I finally got around to reading the Argentine writer Rodolfo Walsh’s Operation Massacre, a book often described as a precursor to In Cold Blood. It’s a less compelling read than Capote’s book, partly because it’s more factual, and also because it’s steeped in the politics of Argentina in the late 1950s, which are mostly opaque to me despite the translator and editor’s best attempts to explain them in endnotes. But it’s a bracing read, a feat of reportage, and—despite what every memoir blurb would have you believe—one of the few books that deserves to be called courageous. The edition I read also includes Walsh’s last work, a 1975 open letter directed to the military junta at the time, which got him assassinated the next day, as he seems to have known it would. That letter ends with what was likely the last line Walsh ever wrote:
“These are the thoughts I wanted to pass on to the members of this Junta on the first anniversary of your ill-fated government, with no hope of being heard, with the certainty of being persecuted, but faithful to the commitment I made a long time ago to bear witness during difficult times.”
By coincidence, I also watched the first part of Patricio Guzmán’s documentary trilogy The Battle of Chile, which captures the political situation just across the border from Walsh at around the same time. (It’s also sort of a historical bookend to Nostalgia for the Light, which I watched a couple of months ago, and looks back on the tragic legacy of the Pinochet regime that has not yet seized power when The Battle of Chile begins.) It’s a great film—like Walsh’s book, more concerned with reportage and witness than art—but this was, in retrospect, perhaps not the ideal month to spend so much time with nonfiction documenting the rise or reach of fascism. In the first film, Salvador Allende is still alive and trying to fight off the CIA-funded right-wing opposition. It ends—shockingly—with a cameraman capturing his own murder on film, as a member of the Argentine military shoots him in the street. I’d read about the moment once, long ago, in some work of documentary film theory, but was not prepared to see the camera fall. It was another unexpected entry in the list of filmed murders I seem to be unwittingly working my way down.
Speaking of—ahem—docuseries, my interview with Phoenix NPR affiliate KJZZ about my hometown, the new Wyatt Earp Netflix show, the movie Tombstone, frontier violence, etc., aired last week. Some of my Val Kilmer fanboying wound up on the cutting room floor, but that’s probably for the best.
Oddly enough, it turns out that the person I was emailing with to set up that interview grew up in Winslow, Arizona, and once made a short film called The Slow Escape, about a murder in her hometown, that I wound up watching on Criterion and really loving.
With the possible exception of my new mayor, about whom I’m cautiously optimistic.
The filmmakers rented out the Clutter house from the people who bought it. There’s a long and fascinating sequence in Joe Berlinger’s Cold Blooded where some of the principals in the film discuss how unsettling filming was, and their sense that the film was cursed. Among other eerie coincidences, the actor who played Perry, a then-unknown Robert Blake, was famously later charged with murdering his wife.
This is one of a few key differences between it and the book that we talk about in class; Capote made up a fake ending set at the Clutter graves, and said he did it because wanted the book to end peacefully.
Somewhat counterintuitively, the latter was chosen partly to convey a sense of realism; by 1967, roughly half of feature films were in color, but most documentaries weren’t, and the filmmakers wanted it to feel like a documentary.
Whether the content of a documentary is “real events” is another question, one you shouldn’t ask me unless you want to be bored half to death.
I’d guess we’re a few years away from some intrepid university offering online classes with no professors. I wish I was kidding.
re: 6. I think first we'll see courses with a single instructor using an AI assistant to teach multiple courses for less pay.