By Gordon Farrer, RMIT University

A posse of fact-checkers has been riding the boundary of the federal election. Not happy with the standard of honesty in political discourse, the ABC, this website and PolitiFact.com.au, a localised version of a US format, staffed mostly by ex-Fairfax journalists, set up operations to check facts in statements made by politicians and others during the campaign.

Isn’t fact-checking what journalists are meant to do already?

Of course it is. And they do. Facts are the building blocks of good reportage, the substance upon which a true and full record of history is built. They are gathered, checked and double-checked before being published in print, on television and radio, and online. At least, that’s the theory.

Journalism has changed. The conversation with the media audience has changed. The competition to be first with news means there is less time to check and confirm every line of a public figure’s statements. The multitude of new avenues for politicians to deliver their unfiltered message to its audience by going around the traditional gatekeepers of the media have changed the nature of information and of the political conversation.

Politicians know this and take advantage of these changes. Facts are spun, taken out of context, cherry-picked or cunningly applied to create a false impression. The fact-checkers’ challenge is how to strip away the noise, lay bare how facts are distorted and to expose the deceit built into the rhetoric of politics.

People think they know a fact when they see one. People should think again.

The truth is that “facts” can be tricky, elusive things.

Here are three facts most would accept on face value:

1+1=2

Clive Palmer is overweight

The unemployment rate is 5.7%.

The equation 1+1= 2 is self-evident; simple observation and experience tell us it is true. But some cheeky mathematicians take delight in proving that 1+1=1 is also true.

They conjure this surprise result by using a numerical sleight-of-hand known as a “mathematical fallacy”. This is, essentially, a well-camouflaged false step, and if you don’t spot the false step or know how to go through the mathematical working to pinpoint where it was introduced, you might be tempted or feel compelled to accept that 1+1=1.

A mathematical fallacy can be created on purpose, as a party trick to impress or challenge fellow numbers geeks at mathematics soirees. A mathematical fallacy can also be accidental, a simple, subtle miscalculation buried in the working that leads to an incorrect result. If not discovered, such mistakes could have potentially fatal consequences – for example, if the error is made by a designer of nuts and bolts used in bridges or space shuttles.

So, it is possible to believe that 1+1=1 is a fact if you don’t think to look for an error, don’t know how to look for an error, or if you don’t know there’s an error to be found. Everyone will know it is wrong, but only a few have the skills to know how to prove it is wrong.

Hold that thought.

Mining magnate Clive Palmer is a larger-than-life character whose physique matches his personality. Even a casual observer can see that he is overweight. But we don’t have to trust the observation of casual observers to know that the statement “Clive Palmer is overweight” is a fact because medical science gives us a definitional tool for the classification of body weight: the body mass index.

The BMI correlates height and weight to arrive at a number. A person is considered to be underweight, overweight or to have a healthy body weight depending on where that number sits on a spectrum.

It’s safe to say that Clive Palmer’s BMI would categorise his weight as
above the ideal for his height. He would sit in the overweight or (according to my dietitian) the obese section of the spectrum.

But consider this: at 193 centimetres and 96 kilograms, Nick Riewoldt –
captain of St Kilda AFL club, superb athlete and fine specimen of a human
being – has a BMI of 25, categorising him as overweight. “Nick Riewoldt is overweight” is as much a fact as “Clive Palmer is overweight” is a fact. Crazy, I know.

Hold that thought, too.

According to the government department responsible for measuring unemployment, the current jobless rate is 5.7%. The statisticians in
the Australian Bureau of Statistics are experts, independent of political
influence, so we have good reason to trust that they know how to measure
unemployment in Australia. The 5.7 figure should be one we can accept as
fact.

But what is being measured? There is considerable debate about the value of unemployment figures. According to a recent column in the Fairfax press the rate does not include roughly 100,000 people who have been moved from the unemployed queues into training schemes. It also does not include those who have given up looking for work, those who work for a family business, or those who do just one hour of paid work each week. Include these categories and you get an unemployment rate of 6.2%.

These three examples help us understand that no fact is an island. Facts are constructed and constrained by social, historical, cultural, scientific and economic factors and cannot exist or be understood outside the context and connections created by those factors. Change the context or the connections and you change the fact.

Fact-checking operations know this and so parse context and connecting factors to arrive at their shades-of-truth rulings, with the tested fact sitting on a spectrum from True through to False via a range of incremental stages (for example PolitiFact’s ratings Half True/False, Partly True/False and Mostly True/False).

Epistemicism is the sub-branch of philosophy that deals with the question of vagueness and inexactness, that border area in which something is going from being one thing to being another. It considers such questions as: At what point does a thin thing become a not-thin thing? Is there a tangible, identifiable definitional line that separates these states?

If there is, we might ask, is there also a line between non-physical states such as “fact” and “not-a-fact” (or between “fact” and “not the fact supposedly being presented”)? That is, is there a “truth mass index” we can turn to for help, a version of the BMI that can be applied to fact?

And if there isn’t an easily defined line between “fact” and “non-fact”, on what basis do the fact-checkers think they can make judgements about factual accuracy?

The fact-checkers operate in this zone of vagueness and, in practice, they do an effective job. As experienced journalists they know how to examine and expose the rhetorical equivalents of mathematical fallacies. They can identify how definitions and assumptions around, say, unemployment figures have been warped or constructed to achieve a desired result.

Of course, there is argument about the nuances of fact-checkers’ rulings; in the real world that is where subjectivity enters proceedings, and there is no hard and fast way to calculate the impact of personal preference or opinion.

But even without a truth mass index, the checkers could rule that Nick Riewoldt is as healthy a specimen of a human being as you will find. They could also rule that Clive Palmer should stop eating hamburgers.

Because sometimes facts speak for themselves.

Gordon Farrer was a Fairfax journalist for 13 years.
Fairfax also holds a stake in his current employer, Metro Media Publishing.

This article was originally published at The Conversation.
Read the original article.

A piece I wrote for The Conversation