Most magazines and newspapers employ “fact checkers”, whose job it is supposedly to ensure that the content of articles is accurate and truthful. The nasty little secret however, is that the purpose of such people is not so much to ensure that the readers of the magazine receive articles that are accurate, but to protect the editors and owners of the magazine from libel law. Therefore, a lot of the time what is actually checked is the accuracy of human sources rather than the accuracy of facts and the internal consistency of articles. If an article says that “Joe Bloggs said that the moon is made of blue cheese” then it is likely to be checked that Joe Bloggs actually said this. If it is merely stated that “The moon is made of blue cheese” then this is less likely to be checked. After all, the moon is unlikely to sue.
As a consequence of this, one finds a great many factual errors in the general media, particularly about scientific and technical information. And one finds dreadful innumeracy – which is a shame given the fact that a basic knowledge of the modern world is pretty much impossible without a decent understanding of the workings of the modern world and a basic understanding of the modern world.
However, this varies by publication, or course. In the British media, The Guardian is far better at getting factual information on technical subjects right than any other paper with the possible exception of the BBC. The Times and Telegraph are worse, and in the electronic media the BBC is usually dreadful. (This wasn’t always so. There used to be a strong pro-enlightenment wing of the BBC, but the decline of this is just one general symptom in the moonbat ascendancy in the BBC that has happened in recent decades).
In any event, an example. Last week I had a long flight in front of me, and as a consequence I grabbed a couple of magazines to get me through the flight. The July/August edition of Foreign Affairs had series of articles entitled “The Next Pandemic”, which considered the possibilities as to what might happen if the world faced an outbreak of a new, nasty, influenza strain. Foreign Affairs is the trade journal of a certain kind of pompous, overly statist Washington D.C. Policy wonk. In any event, it is read by what in D.C terms are “serious” people. I find this slightly distasteful, but I have a certain morbid fascination for the subject of contagious diseases and ways of coping with them, so I bought the magazine.
The lead article in the section (and one of the others) was written by Laurie Garrett, author of The Coming Plague: Newly Emerging Diseases in a World Out of Balance and Betrayal of Trust: : The Collapse of Global Public Health”. I have read the first of these books but not the second. I found it interesting in that it gave lots of historical information that I was not previously aware of, but I found its central argument – that standards of public health in the world is in decline and a consequence we are newly vulnerable to emerging diseases – to be unconvincing. I particularly disagree with the semi-stated corollary that the solution is the expenditure of vast amounts of public money. Certainly there are places in the world where standards of public health have declined (eg in British and Canadian public hospitals) but in a global sense sanitation has never been better, and global best practice (ie that of the United States) has clearly improved. On top of this we are in the midst of a biotechnology revolution of astonishing speed. Fifty years ago biology was largely taxonomy and medicine was largely “Try this and see if it work”, whereas today we have real understanding of how biological systems works and how diseases work, and as a consequence a much more basic understanding of how to attack them. In any event, despite all this I actually agree that the world is vulnerable to a dangerous new influenza strain, and I am interested in details. The possibility of the H5N1 influenza strain that has been sloshing around in the chicken populations in parts of Asia swapping genes with a strain that is highly contagious amongst humans is indeed scary. So I bought the magazine, and started to read the leading article. It started with an overview of the last time the world was hit by a really nasty influenza strain, in 1918-19. Fair enough, although the world was obviously struck at a particularly vulnerable time. Infectious diseases are opportunistic in that sense. So I read on.
On the third page of the article I read the following:
Nearly half of all deaths in the United States in 1918 were flu related. Some 675,000 Americans – about six percent of the population of 105 million and the equivalent of 2 million American deaths today – perished from the Spanish flu.
I read “six percent” and my first though was that “I knew it was bad, but had no idea it was that bad”. However, 675,000 is point six percent, not six percent, which is indeed absolutely terrible, but not that bad. For this error, I was willing to forgive the author. I suspect that the Foreign Affairs style guide (if there is one) states that words (ie “six”) should be used for numbers below a certain point, and numerals (ie “6″) should be used for numbers above that point. My suspicion was that Garrett had written “0.6 percent of all deaths…” and that an innumerate fact checker or sub editor had attempted to comply with this policy without being capable of comprehending the nonsensicality of the resulting sentence. Bad, but sadly common. So I pressed on.
… although doctors then lacked the technology to test people’s blood for flu infections, scientists reckon that the Spanish flu had a mortality rate of just less then one percent of those who took ill in the united States.
Okay, “just less” than one percent. How much less. I am sure that the scientists in question gave actual numbers in their papers. Let’s be charitable and assume that “just less than 1%” means 0.9%. (It may not, but how am I to tell? There are no footnotes) With 675000 deaths this means that 75 million people were infected, an infection rate of 69% of the previously stated population of 109 million. I can just about conceive of a disease with a 69% infection rate, but it flags my “Is that reallytrue?” sensor. I keep going, to the fourth page of the article….
The official estimate of 40-50 million total deaths is believed to be a conservative extrapolation of European and American trends. In fact, many historians and historians believe that a third of all humans suffered from influenza in 1918 – and of these 100 million died
Disregarding everything else, that last sentence is astonishingly frustrating.
“A third of all humans suffered”. What does “suffered” mean? I will assume it means “were infected” or perhaps “took ill”. (I’m not sure what that means either, but it is what the author said last time). “100 million died” is fine, but without telling us how many people were in the world at that point it doesn’t tell us what percentage of the human race died, and it makes it impossible to compare the first half of the sentence with the second, which is an elementary mistake for anyone with basic numeracy skills.
And if a third of the human race was infected and (from earlier) 69% of Americans were infected, is this likely? Given that even in 1918 America had first world standards of hygiene and most of the rest of the world didn’t, it seems most unlikely that the infection rate in the US was twice that for the world as a whole. given that infection rates from most infectious diseases can be removed dramatically by better hygiene. If this is true, it is so remarkable that it requires comment, but it receives none. I am sure that these different numbers have come from different sources and are based on different assumptions, but simply stating them without any attempt to reconcile them (or any apparent grasp of the fact that they need reconciliation) is simply dreadful.
At this point I came close to throwing the magazine across the aircraft in disgust, and I stopped reading the article and went back to the new Harry Potter, which was at least acknowledged as being set in an imaginary world. (I promised myself I would finish the article at some point, but have not yet forced myself to do so). I had gone way beyond the point where I could blame editors and fact checkers of the magazine fof not detecting the contradictions, and I had to conclude that the author’s methodology and numeracy were so flawed as to render her argument useless. (Which does not excuse the editors and fact checkers fpr publishing the article in the first place- somebody should have flagged this before it went to press).
Why do I care about this? Because, when you are dealing with questions as to whether public health is improving or getting worse, then you are taking improvements or declines of a couple of percent or less (usually less) a year, and a proper understanding of when these kinds of things can and cannot be extrapolated matters. If you cannot make your numbers add up in more basic contexts, then the more subtle judgements that are necessary in order to contrive good policy are obviously beyond you.
And if this is the sort of analysis that “serious” people in Washington are paying attention to in order to make decisions and health and environmental issues facing America and the world as a whole, then we are in trouble. If the sorts of people in charge of these things are incapable of reading such an article without raising the sorts of issues I have just raised, then we are really in trouble.
As it happens, I agree with Garrett that the risk of a global influenza epidemic is serious, although I don’t think the effects are likely to be quite as catastrophic as she does. But if it happens, an intelligent responce requires an intelligent analysis. And my faith that we might get one has just been seriously reduced