Tropes and evidence
How content cues aid evidence-fitting for open arguments
Quite some time ago I talked about the importance of tropes. The word “trope” can have many meanings but here we mean it in the manner of TV Tropes: a reused “narrative device or convention used in storytelling or production of a creative work.” The “ticking time bomb”, the “falsely framed man”. The sort of repeated plot point or character element you see in any narrative work of art.
For me, the question of tropes — both back then and now — focused around a simple question. I mentioned in a previous post that when faced with a new event, a primary question people ask online is “What is this thing in front of me evidence of?” A few years back, I had noticed how quickly people make those determinations. Given a white van behind a polling location on voting day, people immediately concoct a story about smuggled-in ballots. Given a bus out front, it’s immediately a story about out-of-town voters.
My belief is that people — seeing an event or piece of supposed evidence — recall similar events in the past based often on surface similarities. Tropes — repeated, concrete elements of story — assist people in mapping events to evidence and evidence to claims.
Car lot on fire
A good example of this is from yesterday, when this video went up, without any context other than a “peeking through fingers” emoji:
First note that this is very popular — to the extent the number can be trusted, it has had 1.2 million views.
To look at how people were interpreting this, I pulled (manually, b/c there’s no longer a Twitter research API) the 55 English language quote tweets sharing this. Here’s a sample:
EVs are the future!
Triggered
#France
When a strategy to limit the fire spread among vehicles is missing.
The riots in France are manifestation of deep-rooted socioeconomic inequalities and injustice and the fact that how fragile so-called strong systems are.
Wow - FRANCE in total war
Daaaaaaaamn!🙆🏽♂️
I then classified each by topic. About 14 were general (“WTF?”, emoji, “OMG”, etc.) A couple were undecipherable. One was a joke about Padres fans. Two were asking what the video was of.
Of the ones using the video as evidence of something, three theories emerged:
This is a video of the riots in France, and evidence of how bad things are there (21)
This is a video of electric vehicles and a sign of how dangerous they are (7)
This is a video of an insurance scam, because no one is buying new cars so dealers resorting to arson (1)
This is a video showing the perils of capitalism (1)
If you want those in a chart, here you go:
Tropes help people categorize (not always correctly)
Keep in mind the poster said nothing about what the fire was. In actuality the fire wasn’t in France, it wasn’t this year, and it wasn’t EVs. It was from a car lot of conventional vehicles in Perth, Australia, last year.
Tropes often help people get to good interpretations of evidence, but here the opposite is the case. Here, the two dominant misinterpretations here are trope-informed.
First, cars on fire are a trope of riots and protests. And in France this doubly so — there’s a long history in France of mobs setting cars on fire. It’s sort of a French thing. So given there are protests going on in France and the video matches tropes of French protests, the obvious (though wrong) interpretation is that it must be video from the protest.
The second trope — an EV fire — is more interesting. This has been a trope quite popular in many climate change denial circles. It is based on a real issue — when EVs catch fire they can be difficult to extinguish. Weirdly, scientists are not even fully sure why they are so hard to put out. That said, among certain communities every time there is video of such a fire (or a photo of the aftermath) it is widely circulated, creating an impression of a world where there is a burning Prius on every block.
The video had actually circulated earlier as a (fake) EV story, before the protests. But now, with the protests more salient, the riot trope became a more dominant frame.
Note that once you know the trope, the argument is pre-made for you. A car on fire at a riot is evidence that the riots are out of control, and law and order has been lost. An EV lot on fire is evidence that governments taking climate change seriously by promoting EVs are actually pushing us into a dangerous dystopia (other purported evidence might include power brown-outs, malfunctioning windmills, and claims about lithium strip-mining).
With tropes, all this happens in an instant. You see the thing, you (think) you know what it means, and you know how it advances an argument. Sometimes you’re right. But often this way of seeing the world leads you astray.
Postscript: A SIFT connection
Incidentally, this is one of the reasons that in SIFT the first question we ask is not “Is this true?” but rather, “Do I know what I’m looking at here?” From my forthcoming book with Sam Wineburg:
Looking at the post, text, or update you have received, you ask the most important initial question. That question is not “Is this true?” but rather “Do I know what I’m looking at here?”
Tropes are powerful interpretative devices, but like a lot of shortcuts of interpretation they can cause you to see things that aren’t there. It’s one of many reasons why the first step of verification is to get the context for what you are looking at, to make sure you haven’t made too many assumptions.
Update:
RT joins in many hours later because of course
