Fake But True Vs. Real But False
How to deal with accuracy, risk, and our messy statistical abilities?
A few months ago, when I wrote about the apparent suspension of New York City’s congestion-pricing plan, a whole adjacent discussion opened up in the comments. It started with a comment on crime in public transit, and the fact that the perception of crime is probably outsized compared to the risk.
One person commented: “I think we have to Do Something about the media and their constant voyeuristic ‘crime reporting.’”
Another replied: “Yikes. How dare the press make factual reporting available to the public? If we let that happen, how will we keep controlling the narrative that everything is fine????”
The exchange continued:
Over-reporting individual crimes (as opposed to broader statistics with proper context) makes people feel like crime (both crime in general and specific crimes) is far more frequent than it actually is.
And that has major negative social consequences, like how the over-reporting of incredibly rare stranger kidnappings resulted in the rise of the “helicopter parent”
And:
But is it “over-reporting” simply to record what happened and publish it in a newspaper? That’s just “reporting”. Almost all of the subway crime reports I read are very by-the-book acts of journalism. Not reporting those crimes feels like an attempt to sweep the problem under the rug.
Then:
If you report on a rare event so often that it distorts people’s sense of how often it happens, you are effectively lying to them about how common that event is.
And so on. In other words, there are two ideas of truth or accuracy or fact in tension here, because of how we process or extrapolate discrete facts into probabilities or risks.
I want to pull a bit from a political newsletter here, from Jonathan V. Last at the The Bulwark. This isn’t about the topic in that piece (he was writing about J.D. Vance amplifying the claim that immigrants were eating cats, which was not true), but about this idea of “fake but accurate,” or “fake but true”:
Two of the formative controversies of my journalism career were fights over Rigoberta Menchú and George W. Bush’s Texas Air National Guard service.
You probably remember the Bush-TexANG story, in which fake documents were planted supposedly showing that Bush served dishonorably. 60 Minutes got suckered by these fakes. It was kind of a big deal.
I covered this story at the time and I was pretty hung up on the fact that the documents were forgeries. Many people on the left argued that even if the docs weren’t legitimate, the story was “fake, but true,” because Bush was a terrible person who must have served dishonorably.
The story of Rigoberta Menchú is slightly more obscure. A Guatemalan human rights activist, Menchú published a testimony of her life in 1983. The book was called I, Rigoberta Menchú and it was widely acclaimed as witness of the terrible treatment of indigenous peoples. In 1992, she was awarded the Nobel Peace Prize on the strength of her activism, which was derived largely from the platform her influential book created for her.
In 1998 an anthropologist named David Stoll published a book documenting several major discrepancies between Menchú’s account of her life and other evidence.
Over the course of a few years, this controversy blew back and forth. There was a movement to revoke Menchú’s Nobel prize. (It failed.) I didn’t cover this story at the time—I only followed it. But the fight broke down along the same basic lines: Some people (like me) thought it was important (and wrong) that Menchú had lied about key details of her life. On the other hand, Menchú’s ideological allies maintained that even if the specific details of her story never took place, there had been terrible human rights abuses in Guatemala. So her story was “fake, but true.”
You’ve probably seen or heard some variety of this argument: My facts may have been wrong, but the general thing I claimed was basically true.
What interests me about this crime reporting thing is that there’s something almost opposite going on: not fake but true, but real but false.