“I am always disappointed by the media coverage on my research area,” writes Sabine Hossenfelder, thus summarizing the feelings of just about every expert in any field who has talked to a journalist, ever. Hossenfelder, a professor at the Nordic Institute for Theoretical Physics in Stockholm, drew the above graph to analyze the causes of her frustration.
Journalists who write about science (or medicine or public policy or finance) must translate the difficult, technical language of their sources into the language of their readers. What’s lost in translation angers scientists, and often, the result of the process is nonetheless too difficult to understand for the average reader.
There are a few things to note about Hossenfelder’s graph. First of all, if Hossenfelder correctly estimates the relationship between readership and accuracy, then the scientists are simply wrong to think that more accurate journalism is more useful to society. That is, if we describe the total amount of knowledge transferred as the number of readers multiplied by the number of accurate facts, then less accurate journalism transfers more knowledge. The blue curve, representing the opinions of journalists on this question, better represents total knowledge transferred as a function of accuracy than does the orange curve, which represents the opinions of scientists.
Hossenfelder hopes that technology might be a solution to this problem. If readers could somehow customize the articles for their level of background knowledge, then accuracy would no longer constrain readership. For example, technical footnotes or marginalia might supplement a main text that is written with the least common denominator in mind. Instead of reading left to right and down the page, readers could take detours and wander through the information as their interest guides them. Then lay readers’ ignorance of scientific topics would not prevent experts from getting the information that they want. Something like this is Know More’s goal. Each post here gives you, dear reader, an accessible little summary of some area of human knowledge, and if you decide you’d like more information, you can click below to keep reading. It is an (admittedly minor) improvement over the print model, where the reader only gets what’s on the front page and then what’s after the jump.
Know More, however, does not believe that less accurate journalism is more socially useful. The problem is in the assumption that readership decreases as accuracy increases. This might be true in general – but really good journalism allows people to understand something they did not understand before. That shifts the peak of the readership curve to the right, allowing readers to appreciate more accurate and complicated reporting.
I was grateful, because up to that moment I had absorbed many hundreds of reports about “subprime lenders in trouble” but had not understood a single one of them. It wasn’t that these reports were uninformative. Rather, I was not informable because I lacked the necessary background knowledge to grasp what was being sent to me as news. On the other hand there was no easy way for me to get that background and make myself informable because the way our news system works, it’s like the updates to the program arrive whether you have the program installed or not!
The Giant Pool of Money was an extremely popular program. Hardly anyone outside of finance – not even a voracious reader of news such as Rosen – understood anything about the mortgage crisis before listening to it, but the piece created its own audience.
The problem is not so much that the conventions inherited from older media – print, radio – limit the amount of accurate information that journalists can convey to an audience of a given size. The problem is that explaining complexity is just really, really hard. Figuring out how to divide up a subject up into comprehensible bits and then putting those bits in an order that can be understood takes a lot of reporting and a lot of careful thought. Making the bits interesting, so that readers won’t get bored along the way, is even harder. If journalism about science tends to fall to the left on Hossenfelder’s graph, that’s more a result of a lack of effort on the part of journalists than anything else.
In fact, articles don’t really exist on a continuum of accuracy. A journalist can write an accurate article that only scientists can understand, an inaccurate article that attracts a wide readership, or an accurate, thorough explanation of an important subject that also attracts a wide readership. There’s not much in between. Human knowledge is in chunks, not in bits. That is, we think and learn in terms of forests, not trees; oceans rather than the droplets that fill them; and people, not their gut bacteria. Representing the connection between accuracy and readership as a smooth curve might not be possible.
Thanks to Hossenfelder for the use of this image. Click below for more of her graphs showing the relationships among scientists, journalists, and readers.