The Lifespan of a Fact records the correspondence between writer John D'Agata and fact-checker Jim Fingal. D'Agata, who teaches creative writing at the distinguished University of Iowa, has submitted a magazine piece about a suicide in Las Vegas. He is not a journalist but an essayist, he tells us, and this allows him some leeway with the details. Why get hung up on petty accuracy, he asks, when you can sometimes get closer to the truth by bending an inconvenient fact?
Fingal was hired by a magazine to scrutinise the piece and he is obsessive and pedantic (in other words, good at his job). The correspondence plays out over seven years, a joust between two views of creative, nonfiction writing. If it is nonfiction, how creative can you be? If it reads better to say something is purple instead of red, does it matter? If it flows rhythmically to say someone is 60 rather than 59, who cares? You have to allow for some poetry, the writer says. But the reader expects you to be honest and you cannot break that contract, the fact-checker responds.
Provoked, Fingal sets out to test every checkable fact, and a few uncheckable ones. This includes a visit to Los Angeles to see if the brick paving that broke the suicide's fall is indeed herring-boned. It is, you will be relieved to know. But was this at the main entrance to the casino, he asks, or just one of many? What defines a main entrance?
As you can imagine, D'Agata becomes a little exasperated.
A similar conundrum arose in the recent court action over whether the Democratic Alliance (DA) could call President Jacob Zuma a thief. Or should we say a "thief"? The African National Congress (ANC) went to court to argue that this broke the law prohibiting the publication of false information with the intention of influencing an election. Another clause makes it illegal for a political party to publish false or defamatory allegations about a candidate.
These are all good rules intended to elevate the election debate. But, as astute legal commentator Pierre de Vos pointed out, if you interpret this law literally and narrowly, it would have a drastic effect on debate. Nobody could promise to create 6-million jobs; you couldn't call another party racist, without providing the incontrovertible evidence.
We all know that parties make extravagant claims in an election, and can be exuberant in their denunciation of rivals, but our assumption is that voters can make their own judgments.
It would be hard to banish all unprovable claims without falling foul of the constitutional protection for free speech. (See our own fact-checking site at www.africacheck.org for a close look at the dubious claims made in election manifestos.)
The High Court ruled that the DA's description of Zuma was "fair comment", leaning towards the view that what mattered was not whether a claim was provably true, but whether it was an honest opinion and a legitimate part of public debate. The High Court favoured open public discussion over tight control.
The ANC appealed to the Electoral Court, which overturned the judgment, arguing that the statement purported to be fact, not comment, and it was not borne out by the Public Protector's Nkandla report, on which it was based.
It took the narrow view of what one can say in an election, which opens the door to a contestation over any election rhetoric that makes unprovable claims.
If only there were a neat distinction between facts and comments. If only we could ensure all election debate was based on provable fact. If only we could make politicians account for the accuracy of their claims and promises.
Until we can, I hope the Constitutional Court - where this dispute is likely to end up - supports the notion that we do not want to contain political debate or try to legislate the competing truths of our politicians. Truth comes through argument and exchange, and we need more of that, not less.
Now was that a fact, or fair comment?
