To receive a free full-text email of The Zeitgeist whenever we publish to the website, please sign up here. You’ll get two or three of these emails every week, and your email will not be shared with anyone. Ever.
Mestema: In a future reality I shall destroy you!
Paul: I reject your reality and substitute my own!The Dungeonmaster (1984)
We write a lot about how governments and businesses desperately want Official Numbers to be treated as synonymous with the real-world feature they are used to describe. You might say it’s kind of our thing here at Epsilon Theory. We even have a word we use for it: cartoons.
We have written about COVID-19 cartoons being promoted by the Chinese Communist Party.
And pension obligation cartoons.
And top-line revenue cartoons.
And labor statistic cartoons.
And daily stock price attribution cartoons.
And risk-model cartoons.
And user / subscriber count cartoons.
The air of legitimacy offered by facts and figures makes them impossibly seductive for leaders in need of a friendly way to frame what is going on in the world. The problem, of course, is that numbers are not always a true reflection of the thing they are being said they are or what they represent. That may be because of how they are measured. It may be because of simplifying assumptions, abstractions and priors in their calculation. It may be because they are based on a model. It may be because they are biased or otherwise influenced in ways that confound the relationship. It may be because they represent only a dimension of the thing and not the thing itself. It may be because of mistakes. It may be because of outright fraud.
It isn’t hard to understand why sensitivity to cartoons around COVID-19 is high. Some of those reasons are good, and some are bad. All the same, it should not be surprising that the article below both topped our Zeitgeist of the most linguistically influential articles AND was by far the most shared article about COVID-19 over the last two days. It certainly helps when the President retweets it, and even more when the self-appointed truth arbiters at Twitter decide to censor it.
New CDC report shows 94% of COVID-19 deaths in US had underlying medical conditions [Various Fox Affiliates]
I have my own views on whether the popularity of this article was driven more by serious, well-informed questions about medical coding methodologies or pre-existing political conditions. That’s my diplomatic way of saying that I suspect the number of people capable of having a reasoned discussion on the implications of reported comorbidities falls somewhat short of the number of shares this article got on Facebook.
Still, confirmation biased click-bait or not, the common knowledge implied by this article sitting at the top of the Zeitgeist is that COVID data is cartoonified. So what’s the verdict? Should citizens be worried that strident representations about what COVID case and mortality data mean might be cartoonified?
I mean, of course they should. And considering that we have multiple official statements from the White House indicating an explicit preference for the reported data to look a certain way, I think most of that evidence points to cartoonification in the direction of underreporting of most statistics. Beyond that, however, there are significant methodological differences between and among states, agencies and individual institutions. There are fog of war issues in hospitals and clinics. There are testing abnormalities. There are high-risk financial incentives for reporting non-COVID pneumonia treatments as COVID-related, and low-risk political incentives for minimizing the number of cases, hospitalizations and deaths under various local and regional political leaders’ watches.
Major media outlets, likewise, have so thoroughly erased whatever line existed between their editorial and news practices that the number of stories they can concoct by simply cherry-picking start and end dates, taking advantage of reporting irregularities, discussing ratios or rates of changes is limitless. If you want to create a COVID-19 cartoon that will suit your political sensibilities, you will have no difficulty doing so. If you are not on cartoon-watch each and every time you see anyone quoting COVID statistics of any kind, you are nuts.
But y’all. Seriously?
The existence of cartoons does not give us a put option on thinking.
The point of being aware of numbers that are prone to manipulation is not to permit us to turn off our brains and pretend that there is no underlying real world feature being measured. The point of being aware of cartoons is not to permit us to write off entire issues as being clouded by the complication of measurements or the attempts by political powers to manipulate their relationship to features of the real world. When you find Wittgenstein’s Ruler, you don’t throw your hands up and say, “Oh well!” The fact that the world of narrative is powerful and messy and fraught with emotional pulls and unknown intentions doesn’t give us carte blanche to reject reality and substitute our own.
The impulse should be the complete opposite.
When we see cartoons, it is a warning that our focus must be on seeking out facts and measurements that are less vulnerable to abstraction. Things like models for excess deaths that predate COVID-19 that can shed light on the aggregate marginal effect of the pandemic on deaths in America. Things like actual nationwide ICU and hospital utilization rates. Things like the actual, demonstrable activities of medical professionals at hundreds of medical facilities. Things like historical data on comorbidities and conditions contributing to death for comparable diseases.
[Editor’s note … I’m sorry, Rusty is trying to be nice here. I won’t.]
If you actually read the actual CDC report and you still think that it is at all damning to the seriousness of this pandemic that deaths from a novel coronavirus that definitionally manifests in cardiopulmonary distress are being coded alongside pneumonia (42%), respiratory failure (34%), ARDS (14%), other types of respiratory failure (9-14%), cardiac arrest and arrhythmia (13-36%), and renal failure/sepsis most frequently reported to be primarily related to direct viral damage rather than co-infection with bacteria (8-17%), then you are wrong [Ed. Note: You are a fool].
If you actually read the actual CDC report and you still think it is at all damning to the level of concern that “normal” Americans should have about COVID-19 that 22% of deaths were coded alongside a condition experienced by roughly half of the adult population (hypertension), or that 16% of deaths were coded alongside a condition experienced by about a third (diabetes and pre-diabetes), or that 11% to 15% of deaths were coded alongside one or more conditions experienced by as many as half of American adults above the age of 85 (i.e. dementia and Alzheimer’s), then perhaps it is worth considering whether how the Widening Gyre is influencing your humanity [Ed. Note: You are a sociopath].
If you think that the rate of comorbidities and conditions contributing to death associated with COVID-19 somehow makes its numbers less “real” than other diseases, like, say, the actual flu (which similarly often has a coded comorbidity in more than 90% of deaths, consisting of the same laundry list of cardiopulmonary and sepsis/infection-related conditions contributing to death) or the dozens of other viral infections that make you die in specific ways for which an ICD-10 code exists, then it may be worth spending more time trying to understand the feature of the world the data represents and less time letting people tell you what it means [Ed. Note: You are sociopathic fool].
The point of being aware of cartoons and other abstractions isn’t to reject someone else’s reality and substitute it with our own.
It is to reject someone else’s reality and substitute it with…reality.