Last updated on September 8th, 2021

If you or a loved one works on a podcast that deals in facts, from a WNYC fact-checker to somebody reciting Wikipedia articles in a closet, I beg all of you: please stop cosplaying as a journalist from the turn of the previous century when you screw up. It’s the digital age; we have the tools to do better. It’s time to hang up the newsie cap and acknowledge just saying “whoops” isn’t going to cut it in an online world where memes posted to the Facebook page of a local radio station are redpilling your uncle into believing there’s a shadow president.

The road to podcasting hell is paved with good intentions and complacency, neither of which should be left unchecked. Reporting facts is a difficult job. Even before misinformation became easily weaponized in recent years, the most seasoned of journalists were not immune to reporting false information. A fair few podcasters are fulfilling a pseudo-journalistic role with little to no experience vetting sources, quick turnaround times, and a lackadaisical approach to deciding what websites qualify as a valid source. The end result is an environment where, inevitably, said podcasters will have to deal with the aftermath of publishing bad information. 

Unfortunately, the podcasting public has fallen into a bad habit of cosplaying as a warped, outdated imitation of a seasoned journalist when owning up to mistakes. I genuinely believe the intentions of most podcasters who have done this were motivated by the idea they were doing the right thing, but how information is communicated has evolved and it’s time to evolve with it.

Print media, especially newspapers printed on cheap paper that ages like milk, is ephemeral. Podcasts are functionally infinite as long as the hosting fees are paid every month.

Speaking of old journalism, I want to briefly clarify a term used in this piece as there’s a bit of gray area for what it can mean to different people. I will be using the journalistic definition of the words contraction and retraction as defined by the Digital Media Law Project (DMPL). Their definition lines up with what I learned in my five years tinkering around collegiate newspapers, and also partially because their page on corrections and retractions is a good resource for someone starting out. To paraphrase the DMLP: correction and retraction tend to be used interchangeably, but generally a correction tells one’s audience about a factual error that’s tangential to the main point of a piece, while a retraction corrects an error that impacts the main point.

What you’re reading now is attempting to address how podcasters sloppily handle corrections in podcasting. Rarely do we hear about retractions, and if they do occur it makes waves. The New York Times having to essentially throw Caliphate under a bus and giving back their Peabody award is the most scorched-earth example of a full retraction in recent memory. It (hopefully) will be a long time before anyone fucks up as royally as NYT did with Caliphate. (Update: this paragraph originally claimed Caliphate was outright deleted based off speculative reporting done when the controversy first broke. It has since been updated, and this is me telling you that in a 21st century manner).

But what about the smaller gaffes? A slip of the tongue, a source that outright lies, or a simple misremembered fact all turn an episode into something that can potentially use one’s platform to cause damage. The current go-to strategy for most is identical to the one recently laid out in Dr. Sydnee McElroy’s chapter of the podcast how-to guide Everybody Has a Podcast (Except You). In said chapter, she recalls an early episode of Sawbones: A Marital Tour of Misguided Medicine in which she didn’t notice multiple sources she found in her research were themselves sourcing one older, much-debunked source. The Sawbones audience pointed out the error and she deployed what is now the de facto response to mistakes on Sawbones: “Often, I will start our next episode off with a correction, so that the misinformation does not persist.”

Read more: My Podcast, My Guide Book, and Me: The McElroys Venture Into Podcasting How-To

I’m singling out Sawbones and quoting Dr. McElroy instead of being vague for three reasons:

  1. Sawbones is a successful venture that has proven itself to be operating in good faith, and I can be confident any critical commentary made here won’t have an actual impact on the show. The same cannot be said for if I were to directly name a smaller podcast. 
  2. Both in her chapter of Everybody Has a Podcast and passively on Sawbones, Dr. McElroy has made a point of stressing that she takes great strides to use quality sources and directly address whenever a concept or source isn’t concrete. A vital goal for a podcast that frequently debunks pseudoscience grifts.
  3. Dr. McElroy’s explanation of why she publishes a correction directly addresses the flaw in handling corrections like this in the podcasting age. 

The goal of filing a correction since the first time a newspaper made a whoopsie has been to preserve the integrity of the publication and stop any lies that might be spread. Except, when podcasts do this, the misinformation still exists. In May of 2020, episode 323 of Sawbones, “Cabin Fever,” opens with the McElroy couple correcting a misrepresentation of how COVID-19 social distancing measures were handled in Sweden. They crack wise about the volume of response they received from Swedish listeners, clarify the point they attempted to make with the bad info, and move on. 

When a newspaper prints a correction in a future issue, it’s done so because the physical limitations of printing something mean it’s impossible to recall the version already purchased by readers. Pre-television newspapers could also operate under the safe assumption that the majority of those who read the previous issue would also see the correction. Given their turnaround time, any copies of the erroneous issue are themselves a non-issue as they’re quickly disposed of to make room for the next paper. 

In the digital age infinite copies of the erroneous issue can sit on the shelf for years, are free to consume, and it’s likely a random person finding it might not go read the next issue in chronological order. Print media, especially newspapers printed on cheap paper that ages like milk, is ephemeral. Podcasts are functionally infinite as long as the hosting fees are paid every month. The Sawbones episode just before Cabin Fever, “COVID Lies, Darned Lies and Statistics,” still remains as-is when originally uploaded, as does the transcript and episode description. There is nothing to indicate “COVID Lies, Darned Lies and Statistics” has a slip-up besides a brief aside in a later, unrelated upload. 

Modern journalists have moved on. The “so about last time…” correction method does still exist in print form, but the digital version of the article is where the misinformation rubber meets the road. While podcasters are still LARPing as if it’s the 1920s, modern journos can just go update articles. It’s too easy to edit articles at this point. Back for its second appearance this article, The New York Times is so edit-happy they change article content, headlines, or even the url slugs after publication to future-proof the piece. There’s a Twitter bot dedicated to posting every instance of NTY editing an article on their main page. It posts a lot. 

Even the fuddy-duddiest of podcast hosting services allow some form of hot-swapping the .mp3 file of an episode that has already been uploaded without having the re-publish. There’s all sorts of dark ethical corridors this function can lead to, but in this specific case it means one can simply put the correction in the episode that needs correcting. There are too many tools at a given podcaster’s disposal now to justify the sole “fix” for misinformation being a shrug and chaste apology in the next episode. 

Podcast audiences are spoiled for choice and limited in how much time they can spend listening in a given day. The next-episode apology operates under the assumption the vast majority of people who listen to one episode will also listen to the next. Multi-episode coverage of a single topic can safely assume anyone listening to episode two will move on to three. Fiction podcasters with long arcs can also make this assumption. Podcasters who make episodic talk-about-a-topic shows akin to Sawbones, 99% Invisible, or Twenty Thousand Hertz cannot. Audiences faced with a huge backlog can and will cherry-pick what they listen to based on title alone. 

[Podcasters] have the ability to suck it up, admit fault, and re-upload from scratch.

We live in a time when there’s an episode of Welcome to Night Vale dealing with multiple realities that uses dynamic ad insertion software to serve random ending to the episode each time a listener downloads it. Even if the average podcaster never gains access to magical dynamic ad insertion abilities, we all have the ability to edit out bad information and hot-swap the file. Failing that, we have the ability to suck it up, admit fault, and re-upload from scratch. It’ll screw up statistics, for sure, but a weird download bump and some confused comments on Reddit is a small price to pay for integrity and doing the right thing. 

It’s time to recognize that while admitting one’s mistakes is a good thing to do. Podcasters with an educational or journalistic slant to their programs have a responsibility to combat the spread of misinformation as much as humanly possible.