(featured image via Dall-E 3)
We can be misled by false information, but we can also be misled by information that is the 100% unadulterated truth. And we ourselves are doing the misleading…
These days, there is a lot of buzz about misinformation – not least because several countries will be holding important elections this year. In fact, we even hear misinformation about misinformation. Some misinformation consists of outright untruths, manifestly false information. Other common forms include bullshit, a general term that philosopher Harry Frankfurt gave a more formal definition: "speech intended to persuade without regard for truth"; and spin and distortion of the truth in order to serve a certain agenda. Even though, here, the responsibility lies with those who spread the misinformation, a critical attitude on our part might help us detect the attempt to mislead us and avoid being taken in.
Unfortunately, that is not all there is to misinformation. It is possible to be misinformed, even when we are looking at accurate facts. No lies, no bullshit, just the plain truth – and still we are being misled.
Looking at the facts
Alice is a young mother with four-year-old twins. Two days ago, she was at a local playground with a friend who has a preschooler, Theo, of the same age. At some point, the little dude was showing off to Alice's kids, and banged his leg against one of the posts of the swings, causing a sizeable bruise on his shin, and some loud wailing for a while. Today, as Alice and her twins return to the playground for another afternoon of fun, they see Theo running around without seeming at all troubled by the earlier mishap. Alice sits down on the bench next to his mum, and remarks how quickly bumps and bruises seem to heal in young kids. "Ah, but I gave him some arnica tablets, a homeopathic remedy, Friday evening and yesterday!", Theo's mother replies. Alice wonders whether she should get some of that stuff, too. After all, the facts are clear: Theo injured himself, got given some tablets, and two days later he is better. That must mean that the remedy is efficacious, mustn't it?
During the lunch break at work, Bob is sitting with a colleague, and the conversation turns to the cost of living. "Did you hear," his companion says, "that it's been thirty years since we paid as large a share of our income on food?" To underscore this bombshell, he produces his smartphone and shows Bob a tweet referring to a Wall Street Journal article. It contains a chart that does indeed indicate a fairly deep trough, with a sharp rise back to the proportion of food spend in the early 1990s. Bob agrees with the conclusion that this is pretty bad – 30 years of progress wiped out. That's what the chart shows, isn't it?
Chris is out for a drink with friends, one of whom has just got a new electric car. Someone remarks that the demand for electric power is likely to soar as more and more people will swap petrol or diesel for electrons, and that a lot more renewable generation will be needed. Another person adds, "Yes, but I saw an American congressman claiming that more people have died from wind turbines than from nuclear power, stressing this as a fact – I looked it up, and it's true! Shocking, isn't it?" Later. Chris verifies the claim and does indeed find that the claim is correct: a third more deaths. Clearly rather shocking, no?
Looking for meaning
The three conclusions – arnica ointment works to heal bruises, food has become proportionately a lot more expensive, and wind power is a bigger killer than nuclear – would seem to be sound, and based on the facts. But let's slow down a bit: we know the facts are correct, but our conclusions go beyond what the facts tell us. They ascribe meaning to the facts. They are interpretations of the facts. And that is where things might go awry.
We know, from experience, about causes and consequences. If we do, or fail to do, something, then it will often have consequences later. If we forget to put the milk back in the fridge after breakfast on a hot day, we will likely find it has gone off when we return in the evening. If a little boy runs full speed into a steel bar, he is likely to experience some bruising. Causes always precede their consequences, but not everything that precedes an event is its cause. There needs to be a clear process that links the two. For example, bacteria in milk start multiplying rapidly at room temperature, and convert the lactose in the milk into lactic acid, which in turn makes the proteins solidify into curds. Leg tissue – muscle, bone and skin – hitting a solid object causes cell damage and ruptures small blood vessels. This leads to swelling and discolouration, which tends to heal naturally in the vast majority of cases. There is, however, no credible evidence for the efficacy of treatments based on dilutions of an active ingredient so extreme (C30 means 1 part per 10 to the power 60!) that not a single molecule of it is left. Believing that the administration of the arnica tablets caused the healing is a fallacy, sometimes referred to by its Latin name "Post hoc, ergo propter hoc" – "after this, therefore because of this".
The things you see when you embiggen the picture
Closer inspection of the chart Bob's colleague showed him tells us that the trough is perhaps not as deep as it appeared, given the way the Y-axis, which does not show the zero, serves to inflates the difference. But it's not just the Y-axis that might lead us to misinterpret the chart. When we go back further in time, we notice that the trough is in fact just a small dip in a pretty flat trendline. There is another question, too: what does 'food' really represent? It comprises two categories – food at home, and food away from home. It seems that the sharp(ish) increase in 2022 is the result of a rapid return to pre-pandemic levels of eating out, and that the dip itself largely coincided with the pandemic lockdowns, when a drastic drop in eating out was naturally not compensated by buying more food to eat at home. The proportion of disposable income going to food at home is lower than ever. The interpretation of the chart as illustrative of a retreat to 1990 may be compatible with the facts, but it is not the only possible one.
Chris's friend who made the praiseworthy effort to verify the congressman's claim was not quite right when confirming it was correct. The data show that the death rates per TWh (one billion kWh) are 0.03 and 0.04 respectively for nuclear and wind energy generation. In 2022, the electric power consumption in the US was 4,240 TWh, of which around 18% and 10% came from nuclear and wind, respectively. On this basis, nuclear would be responsible for about 31 deaths per year, and wind for about 13 (which is fewer deaths in absolute numbers). However, before drawing a conclusion about how shocking this is, and about the desirability of different modes of power generation based on the corresponding death rates, it is instructive to also look at coal, gas and solar sources. The death rates for these generation methods are, respectively, 24.62, 2.82 and 0.02, and their respective contributions to the overall production are 20%, 40% and 3.5%. They would thus be responsible for around 20,000 (coal), 4,800 (gas), and 3 (solar)deaths per year. The most meaningful conclusion to draw is that fossil fuel generated electricity is vastly more harmful than nuclear, wind and solar. The difference between the latter is barely significant, and not large enough to use respective death rates as a major factor in deciding what mix between different generation methods is optimum.
Misinformation is not only a matter of accurate data. When we are presented with true facts, the work is not done: we need to interpret the facts and decide what they mean. And it is precisely in this meaning that misinformation manifests itself. There are many ways in which we can get this interpretation wrong; the examples above illustrate just a few of them. If we see causation without verifying that there is indeed a plausible causal mechanism, or if we draw conclusions without considering the wider context, we risk being misled – misled by ourselves.
We then become our very own source of misinformation, despite being led by the facts.
