Sbagliando s'impera

Gardner, Dan (2008). Risk: The Science and Politics of Fear. London: Virgin Books. 2009.

goodreads.com

Gardner – al suo esordio in quest’opera – è un giornalista canadese. Questa sua opera si colloca in un filone che potrei definire quello del neo-ottimismo quantitativamente fondato, e il cui manifesto è The Rational Optimist di Matt Ridley, che abbiamo recensito qui. Non è un caso se l’ultimo capitolo si intitola There’s never been a better time to be alive.

La speranza di vita è la più elevata da quando l’homo sapiens sapiens ha conservato qualche informazione sulle sue condizioni di vita. Perché allora ci sembra di vivere in un’epoca sempre più piena di rischi, si chiede Gardner? Perché le nostre ansie e le nostre paure, invece di diminuire, aumentano?

In parte perché nelle nostre menti, in situazioni critiche, la parte irrazionale prende il sopravvento su quella razionale: nello spiegare questo, Gardner fa un buon lavoro di illustrazione delle ricerche di Paul Slovic, Daniel Kahneman e Gerd Gigerenzer. Ma in parte anche perché i giornalisti fanno male il loro mestiere: e Gardner, pur appartenendo alla categoria, porta molti esempi ben documentati.

Una combinazione micidiale, che porta a privilegiare l’evidenza aneddotica con un forte contenuto emotivo sulla fredda analisi statistica. Non restano molte speranze per chi, come me, è convinto della necessità della crescita della cultura quantitativa e dell’esercizio dell’analisi critica. Non penso assolutamente (nonostante qualche oscillazione e qualche eccezione da parte mia) che la soluzione possa essere lo statistical storytelling, quanto meno nel senso approssimativo e corrivo in cui viene inteso, cioè come ricerca del sensazionalismo anche a scapito del rigore nella presentazione dei dati (su questo si veda, ad esempio, la polemica di Donato Speroni sui numeri della disoccupazione giovanile). Neppure Gardner – mi pare – ha antidoti convincenti da proporre, né scorciatoie da seguire, tranne quella della “immane fatica del concetto” (die Anstrengung des Begriffs) di hegeliana memoria.

* * *

Una delle storie di disinvoltura ai limiti della disonestà intellettuale, ancorché a fin di bene, raccontate da Gardner dovrebbe suonarci familiare, perché le statistiche sulla povertà (“le famiglie che non arrivano alla fine del mese”) sono tra le preferite dai nostri giornali e, anche nel nostro caso, le differenze tra povertà assoluta, povertà relativa, deprivazione materiale e percezione del disagio economico sono colpevolmente trascurate.

Leaving my neighbourhood grocery store one afternoon, I came across a poster featuring a sad-eyed boy wearing a t-shirt emblazoned with the words ‘I’m hungry.’ The caption read: ‘One in five Canadian children lives with hunger.’ It was an appeal for donations to ‘The Grocery Foundation’ […]. The cause is irreproachable. But I’d never heard that statistic before and I couldn’t believe the situation was that dire. The wording was also odd. What does it mean that a child ‘lives with’ hunger? Does that mean that they experience it every day? Once a week? How is hunger defined and measured? I wanted to know more so I e-mailed the executive director of the foundation, John McNeil.
[…] he sent me an excerpt from a letter written by Sue Cox, the former head of the Daily Bread Food Bank and ‘an acknowledged authority on hunger and poverty,’ according to McNeil. Cox’s case for the one-in-five statistic went like this: First, ‘child hunger and child poverty are inextricably linked’; second, Statistics Canada says the ‘current rate of child poverty is one in six’; third, the real number is likely closer to one in five because the telephone survey used to come up with the one-in-six number would not catch very poor people who can’t afford telephones.
What Cox didn’t mention is that Statistics Canada has no data on ‘child poverty’ or any other kind of poverty. What the agency has is something called the ‘low Income Cut-off,’ or LICO. That’s where the one-in-six number came from. But the LICO is not a ‘poverty’ number, as Cox claimed. It is a measure of relative deprivation only, intended to identify ‘those who are substantially worse off than the average,’ in the words of Ivan Fellegi, the head of Statistics Canada. If the income of the top 10 per cent in the country doubled tomorrow, the number of people who fall below the LICO would soar – even though alla the people who suddenly dropped below that line would have exactly the same income they had before. The statistics agency has repeatedly stated that it does not consider LICO to be a measure of poverty. ‘Statistics Canada does not and cannot measure the level of “poverty” in Canada,’ wrote Fellegi.
So the basis for the claim that ‘one in five Canadian children lives with hunger’ is this: A number that Statistics Canada says is not a measure of poverty was used as a measure of poverty; the word ‘poverty’ was changed to ‘hunger’; and the number was arbitrarily reduced from one in six to one in five.
[…]
It’s understandable that honourable people pursuing a worthy cause would not be terribly concerned about the strict accuracy of the information they use in the pursuit of their cause. But it is also unfortunate. And unfortunately common. [pp. 172-173]

* * *

‘People can come up with statistics to prove anything, Kent. Forty per cent of all people know that.’ [Homer Simpson, citato a p. 127]

Language is one of the most basic means of medicalizing a problem, the critical first step to getting people to ask their doctors for a pill. So impotence’ becomes ‘erectyle disfunction’, an impressive medical-y phrase that pushes away consideration of factors like stress and anxiety as causes of impotence that cab be cured without a pill. Numbers are also key. People will be more likely to conclude they have a condition if they think it’s common, and so drug companies push statistics like ‘more than half of all men over 40 have difficulties having or maintaining an erection’ […] [p. 159]

Call it ‘denominator blindness.’ The media routinely tell people ‘X people were killed’ but they rarely say ‘out of Y population.’ The ‘X’ is the numerator, ‘Y’ is the denominator. To get a basic sense of the risk, we have to divide the numerator by the denominator – so being blind to the denominator means we are blind to the real risk. An editorial in The Times of London is a case in point. The newspaper had found that the number of Britons murdered by strangers had ‘increased by a third in eight years.’ That meant, it noted in the fourth paragraph, that the total had increased from 99 to 130. Most people would find this at least a little scary. Certainly the editorial writers did. But what the editorial did not say is that there are roughly 60 million Britons and so the chance of being murdered by a stranger rose from 99 in 60 million to 130 in 60 million. Do the math and the risk is revealed to have risen from an almost invisible 0.0001 per cent to an almost invisible 0.00015 per cent. [pp. 194-195]

According to the RAND-MIPT terrorism database – the most comprehensive available, there were 10,119 international terrorist incidents worldwide between 1968 and April 2007. Those attacks took the lives of 14,790 people, an average annual death toll of 379. […] Terrorism is hideous, and every death it inflicts is a tragedy and a crime. But still, 379 deaths worldwide annually is a very small number. In 2003, in the United States alone, 497 people accidentally suffocated in bed; 396 were unintentionally electrocuted; 515 drowned in swimming ppols; 347 were killed by police officers. [pp. 299-300]

‘I do not know why attacks didn’t occur’ in the years after 9/11, [George Tenet, former director of the CIA] wrote. ‘But I do know one thing in my gut: al-Qa’ida is here and waiting.’
And that’s just what terrorists want the gut of George Tenet and every other American to think. ‘America is full of fear from its north to its south, from its west to its east,’ Osama bin Laden said in a 2004 video. ‘Thank God for that.’ [p.337]

Jeffrey Sachs, a renowned economist and development guru, estimates malaria could be controlled at a cost of between $2 billion and $3 billion a year, so here is a case where millions of lives could be saved and billions of dollars saved for an annual cost equivalent to about 5 per cent of the money the United States budgeted for counter-terrorism in 2007. [p. 347]