DOUBT THE DOUBTERS

In 1950 Austin Hill and Richard Doll wrote to every doctor in the UK

They wanted to know how much they smoked

And down the line how they died

40K doctors got back to them to take part

Doctors were a good subject group because they were easy to keep track of

And could be relied upon to have good post-mortems

By 1954 Hill & Doll had enough data to publish the first ever findings into smoking and cancer

Smoking made you sixteen times more likely to get lung cancer

In 1965 a US senate committee was considering putting a health warning on cigarettes

An expert witness called Darren Huff was presented

He began by telling a story about babies and storks 


He said he could use statistics to prove that storks delivered babies

He started pointing to charts and graphs

Across Europe he said there is an incredibly strong relationship between storks and babies

Where there are fewer storks there are fewer babies

More storks, more babies

And the pattern is strong enough to not be a coincidence

But of course, storks don’t deliver babies Huff concluded

Just because there is a correlation, doesn’t mean there’s causation

Huff then likened storks and babies to smoking and cancer

A statistical quirk, no more

In 1954 Darren Huff had written a book called How to Lie with Statistics

In which he argued all statistics were a con trick not to be believed

He had subsequently been paid by the tobacco industry to appear in front of the senate committee

And sell his very special brand of doubt

To make them doubt the evidence of Hill and Doll

It’s a tactic the tobacco industry used for decades

Calling for more research

Announcing tangential research into car exhausts and mad cow disease

The tobacco lobby acknowledged ‘doubt is our product’

Selling doubt is a tactic of those who want people to give up

To assume they can’t know anything and so accept the way things are  

A tactic Steve Bannon calls ‘flooding the zone with shit’

It involves questioning experts and branding everything unknowable

But things are knowable if you use proper statistics and robust methods

Like the fact that smoking causes cancer

The findings of Hill and Doll have saved millions of lives

It’s those selling doubt that we need to beware of