2000 mules

a review

Juancito, the best mule of Argentina, courtesy of Darrio Urruty@Wikimedia

I watched Dinesh D’Souza’s latest documentary 2000 mules yesterday (courtesy of CitizenFreePress; the main website appears to be down today!).

I came in to the movie with an open mind, eager to see the convincing evidence of systemic 2020 election theft that it was touting. Still, I also, admittedly, had some reservations about yet another promise to “release the Kraken”, and also Dinesh D’Souza’s previous convictions for campaign finance fraud (subsequently pardoned by Trump, presumably due to the political nature of his prosecution).

The claim

As far as I can tell, D’Souza’s claim is based primarily on two types of evidence:

1. Video and audio evidence of apparently egregious wrongdoing. The former is in the form of video camera footage around voting dropboxes that the film crew was able to produce through legal channels. The latter is in the form of anonymous witnesses who discuss what they saw around election night.

2. A reconstruction of cellphone locations, based on private data, showing apparently suspicious behavior by presumed “mules” traveling to at least 10 different dropboxes each and stuffing an average of 5 ballots per box.

The film also discusses the legalities (or lack thereof) of different mechanisms for delivering ballots by proxy (e.g., ballot harvesting, paid delivery) in different states.

The problems

Unfortunately, I did not find the film convincing on its own.

Yes, the video shown appears quite damning - the centerpiece being a woman coming up to a drop box in blue gloves (and a short-sleeved shirt), apparently in the dead of the night, delivering a number of ballots, and then disposing of her gloves in a nearby trash can.

However, this video could also be explained, for example, by a person delivering her family’s ballots and being deathly afraid of potentially contracting COVID - this would explain both the gloves and the approach in the dead of night, when the box would be deserted. The proper way to address such uncertainty, as investigative reporters know, is to identify and investigate (in a journalistically responsible manner) the people who appear to be performing illegal activities.

Similarly, the cellphone location data appears to be quite damning as it is presented in the video … what person would take a walk by 10 different ballot dropboxes in the course of a night? However, again, we do not know many of the specifics of how the locations were determined, their error margins, etc.

If only cellphone tower signal stregth data was used to determine location, then the error margins could easily be tens of feet, meaning that it would not be surprising to find some number of people, among an entire population of a city, who traveled within range of a number of dropboxes. Even if the location data was based on GPS, it would still have error margins that may or may not explain the apparent behavior - for example, many of the dropboxes are deliberately placed in high-traffic areas.

I should stress that fine-grained cellphone location data is extremely private, and we should be suspect of any company that collects and distributes such data about its customers without their explicit and clear consent.

Potential remedies

There are various potential ways in which the film staff could buttress their (admittedly scandalous) claims:

• Do the investigative reporting legwork to identify potential mules. Find more convincing evidence that their behavior was malicious and make it known.

• Publicize the cellphone data used by the film.

  • Publish the algorithms that were used to produce mule travel traces.

  • Publish some sanity checks that can be used to verify the reliability and authenticity of the underlying data and the film’s conclusions.

  • Publish a range of statistics on the cellphone data (e.g., what was the average number of dropboxes by which people traveled that day).

• Subject the film’s analysis to a completely independent third-party evaluation.

None of these require the aid of the police or FBI (although both could certainly help). Yet the claim being made is significant enough that people should be rightfully skeptical without more convincing proof.