‘Hi, John, I’m the intern who’s been assigned to fact-check your article. I was hoping you could clarify how you determined that there are thirty-four strip clubs in the city while the source you’re using says thirty-one.’

So says Jim Fingal to John D’Agata in their co-written book, The Lifespan of a Fact, a text ostensibly based on Fingal’s fact-checking of a D’Agata essay on the suicide of a Las Vegas teenager, published in the Believer.

The book is built from their extensive correspondence about that essay. In it, Fingal presses D’Agata about various questionable claims. Were the dog-grooming vans pink or purple? Did the teenager fall for eight or nine seconds? D’Agata responds by aggressively asserting his identity as a literary essayist rather than a journalist, and thus his concern with emotional or artistic truth rather than brute reality.

‘It’s called art, dickhead!’ D’Agata snaps at one point.

Perhaps conscious of the market generated by creative writing pedagogy, The Lifespan of a Fact structures the debate in terms familiar to writerly academics, depicting a generic innovator pushing the boundaries of a traditionalist industry.

But how accurate is this?

In her New Statesman review of the book, Olivia Laing refers approvingly to ‘the grand American tradition of fact-checkers, the staff of newspapers and magazines who might spend a year ascertaining the truth status of every assertion in a writer’s work and who have the capacity to kill an article if it’s fudged or faked’. Yet note that Fingal introduces himself to D’Agata as an intern – that is, an unpaid volunteer.

The current state of publishing means that journals printing creative nonfiction must, out of necessity, decide on which grand editorial traditions (American or otherwise) they are prepared to skimp, with a small literary magazine like the Believer – or Overland for that matter – inevitably distinguishing between core and non-core tasks.

Once one knows that fact-checking has been assigned to an intern (along with opening the mail, making the coffee and reading the slush pile), Fingal’s pedantry becomes less representative of industry norms and more indicative of how those ‘norms’ now rely on the labour of workers not compensated for their diligence. To put it another way: read properly, The Lifespan of a Fact suggests that truth in nonfiction might be investigated through economics as much as (or more than) through aesthetics.

But rather than examining a literary magazine with a circulation of about 20 000, let’s consider BuzzFeed, a news and entertainment site that in November 2013 alone was read by an astonishing 130 million people.

That same month, on the Thanksgiving holiday, a television producer called Elan Gale tweeted of an encounter with a rude woman on a delayed flight. BuzzFeed’s Rachel Zarrell aggregated Gale’s tweets and photos, and published them under the heading ‘This Epic Note-Passing War on a Delayed Flight Won Thanksgiving’. In a notoriously slow news period, the story received nearly 1.4 million page views.

Yet, while the note-passing war might have been epic, it didn’t actually happen. When media pundits learnt Gale had fabricated the incident, some complained that BuzzFeed should have investigated before publishing. But had Zarrell waited while she checked Gale’s rapidly unfolding narrative, she could not have posted the story – and BuzzFeed would have lost all its page views. Instead, she ran the piece, and when alerted to its falsity, added a correction. This, in turn, brought more traffic.

There was, in other words, no down side to being wrong.

As Slate’s David Weigel quipped: ‘“Too good to check” used to be a warning to newspaper editors not to jump on bullshit stories. Now it’s a business model.’

The clickbait that circulates on Facebook and Twitter rarely develops organically. Sites like BuzzFeed deliberately build viral content; their posts are the result of extensive experimentation with various headlines, excerpts and graphics (hence the instantly recognisable formulas: ‘You’ll never believe …’ and so on).

Late last year, the Wall Street Journal featured a profile of a certain Neetzan Zimmerman, whose posts for Gawker singlehandedly generate more than 30 million hits per month.

Zimmerman maintains a feed of a thousand or so sources of incipient virality. Each morning, he scrolls through his incubators, assessing what he finds according to a scale of outrageousness, cuteness, hilarity and other desirable traits, and then deciding how each item might work with the memes of the day. In effect, he keeps a running mental list of what is – or is becoming – ‘hot’, and creates content based on these themes.

‘It might be that right now, people don’t care about stories about cats that much, and instead, sloths are more popular,’ he says. ‘So I’ll have a rule – cats are out, sloths are in, focus on sloths because that’s going to be your meal ticket.’

Zimmerman tells of a post he created around a firefighter rescuing a kitten from a burning building. Unfortunately, the cat later died. Including a mention of the animal’s demise was, he says now, a mistake, one that prevented the post from going viral.

‘People are just not going to share a cat video of a dead cat,’ he concludes.

Zimmerman does not invent stories. His point is that viral content should be true but not too true, for too much information kills traffic. You can report on the cat but not on its death, since no-one looks ‘to these stories for hard facts and shoe-leather reporting’.

Gawker, BuzzFeed, Upworthy and their many imitators might seem far from the serious journalism of conventional news outlets. But they are not: BuzzFeed recruited its foreign news editor from the Guardian, and now employs a team of investigative reporters that includes Mark Schoofs, former foreign correspondent with the Wall Street Journal.

That was why the launch of BuzzFeed Australia mattered – it was a rare instance of a media outfit hiring rather than firing. When most newspapers are shedding staff, BuzzFeed can pay for an ‘international women’s rights correspondent’ based in Nairobi because, for instance, the posts that explain the Egyptian revolution via Jurassic Park GIFs are so extraordinarily popular.

Even traditional news outlets are increasingly adopting the viral style, hoping to maximise their impact on social media. A prime example is CNN, a news outlet operating for over three decades, which recently tweeted: ‘14-year old girl stabbed her little sister 40 times, police say. The reason why will shock you.’

Yes, there is still a difference between a New York Times article and an Upworthy listicle. But every site administrator (whether at a newspaper or a literary journal) fanatically monitors traffic stats; every ad department craves the Facebook shares that generate page views. If, as Zimmerman argues, virality actually depends on a certain economy with the truth – and if, as the BuzzFeed example suggests, there is little penalty for being wrong – do we really expect editors to worry overly as to whether Las Vegas contains thirty-four or thirty-one strip clubs?

That being said, professional fact-checking has not disappeared. On the contrary, it’s enjoying a remarkable resurgence, albeit in a different form.

During the last federal election, Australians who wanted their information professionally verified found themselves spoilt for choice. All of a sudden, they could choose between an array of fact-checking units (FCUs): the ABC, the Conversation and Crikey all ran dedicated fact-checking operations, and a local version was launched of PolitiFact, a Pulitzer Prize-winning US site set up in 2007 to investigate and evaluate political statements.

Yet what we might call ‘new-school’ fact-checking differs from ‘old-school’ fact-checking in at least three important ways.

First, the old fact-checkers evaluated information in their own publications. From a magazine proprietor’s point of view, the ‘grand American tradition’ to which Laing refers was less about an ethical commitment to truth and more about minimising the potential for expensive defamation suits. By contrast, the new FCUs assess the claims made by others, usually politicians.

Second, old-school fact-checking belonged in the production process. It was deliberately invisible, with the reader only conscious of checkers if they had made a mistake. New-school fact-checking, on the other hand, takes place in public; it is undertaken to produce a story – most usually, a piece explaining whether or not a particular political claim can be verified.

Third, old-school fact-checkers examined all the content that a magazine published. New-school fact-checking focuses almost exclusively on parliamentary politics, which was why the new Australian FCUs emerged directly from the federal election coverage.

At first glance, the politicisation of fact-checking seems baffling. Why scrutinise political stories more than others? Why do facts take on particular significance during an election rather than in the reporting of a murder, a scientific discovery or a war?

The answer lies in the peculiar context out of which new-school fact-checking evolved. The local FCUs are all, more or less explicitly, imitations of US sites, most of which took shape in the disastrous presidency of George W Bush.

It was in the Bush era that the comedian Stephen Colbert coined the term ‘truthiness’ to describe a quality that Republican politicians recognised by gut instinct, generally in opposition to the claims of scientists and other experts. The gag reflected a widespread sense that facts themselves were under assault by the neoconservatives and their allies.

Famously, the journalist Ron Suskind described an encounter with an unnamed presidential aide (widely identified as Karl Rove):

The aide said that guys like me were ‘in what we call the reality-based community,’ which he defined as people who ‘believe that solutions emerge from your judicious study of discernible reality.’ […] ‘That’s not the way the world really works anymore,’ he continued.

In response, appalled US liberals proclaimed themselves members of the ‘reality-based community’, and the defence of truth became part of their opposition to the radical conservatism of Rove and his co-thinkers.

You can see the legacy still playing out in Australia today, where FCUs have been embraced by liberal outlets (Crikey, the Conversation, the ABC) rather than by, say, the Murdoch publications.

But if facticity has taken on particular political connotations, the new FCUs also evolved in a specific media ecosystem.

In the immediate aftermath of September 11, much of the media establishment joined in the hysterical nationalism of the time. The New York Times played (as much or more than FOX) a key role in Bush’s case for an Iraq invasion, through the now-discredited reporting of Judith Miller about Saddam’s alleged weapons of mass destruction.

It therefore fell to smaller, newer publications (many of which were online) to hold the administration to account. In retrospect, 2003 – the year of the Iraq invasion – probably marks the peak of the blogosphere’s influence; at this point, amateurs were scrutinising the Bush agenda far more closely than many traditional reporters at major outlets.

The subsequent rise of FCUs represents a restoration of journalistic authority; a mea culpa, perhaps, but also an attempt to reassert the importance of professionalism. As we saw with the Believer, smaller publications cannot afford paid fact-checkers. By its nature, a FCU makes a claim to the traditional heft once enjoyed by newspapers: it is presenting truths professionally assessed by experts.

You can see how that plays out in Australia with, in particular, Crikey’s regular Get Fact feature. Crikey originated as an email bulletin, specialising in political (sometimes anonymous) gossip and running unverifiable (though often correct) stories that the newspapers would not touch. Today, it’s grown into something much more like a traditional news outfit. On the one hand, fact-checking allowed Crikey’s mostly liberal writers to push back against what they decried as the ‘unhinging’ of federal politics; on the other, the creation of its own FCU lent Crikey an authority distinguishing it from smaller online rivals.

What’s wrong with all of this?

In some respects, nothing. But we must ask: what do hard facts actually achieve?

‘Fact-checking isn’t perfect,’ argues Greg Jericho, ‘but it aspires to provide voters with a better debate and clearer information, which can also be used by journalists when questioning politicians. Surely that is a good thing.’

That is fine, as far as it goes – but it doesn’t go nearly as far as one might think.

Fact-checkers might provide clearer information but, by itself, that information cannot achieve the twin aims with which FCUs are still associated.

Let’s first think about facts as a liberal project – an attempt to dispel overheated political rhetoric with hard data. The problem is that this simply does not work. There is very little evidence that the presentation of facts is, in and of itself, politically effective. A series of studies performed by the University of Michigan in 2005-06 revealed that people exposed to corrected information in news stories did not change their minds. On the contrary, they often became more convinced of their earlier beliefs; rather than dispelling misinformation, facts actually made the unhinging stronger.

As Joe Keohane argues in the Boston Globe:

Most of us like to believe that our opinions have been formed over time by careful, rational consideration of facts and ideas, and that the decisions based on those opinions, therefore, have the ring of soundness and intelligence. In reality, we often base our opinions on our beliefs, which can have an uneasy relationship with facts. And rather than facts driving beliefs, our beliefs can dictate the facts we chose to accept.

One must be careful with these formulations, because they can be taken as evidence that people are incorrigible idiots, incapable of grasping reality. The point is more that facts become meaningful because of the theories in which they’re embedded, because of the set of assumptions that makes them relevant. Philosophically, that is the traditional critique of crude positivism – ‘facts’ do not simply exist in the world, but are generated as discrete ‘objects’ by social and historical discourses. You can check how many refugee boats actually arrived last year, but such an exercise does not address the unspoken assumptions about immigration, borders and ethnicity implicit in the very need to publish the statistics at all.

It’s not that facts don’t matter. Rather, in the social sciences, ‘facts’ exist in a relationship with theory and political practice, a relationship that is fundamentally dynamic.

This is accurately – though terrifyingly – hinted at in the exchange between the Bush adviser and Suskind:

We’re an empire now, and when we act, we create our own reality. And while you’re studying that reality – judiciously, as you will – we’ll act again, creating other new realities, which you can study too, and that’s how things will sort out. We’re history’s actors […] and you, all of you, will be left to just study what we do.

In retrospect, that sounds like a pretty good account of what actually took place: the Bush administration reshaped reality in ways that previously would have seemed impossible.

Fifteen years ago, everyone knew that waterboarding – a technique employed by Pol Pot – was torture. Yet the Bush administration was able to radically alter many people’s understanding of the practice. Arthur Brisbane, former public editor of the New York Times, explains how ‘reality’ suddenly changed:

The Bush administration offered formal legal opinions that the ‘enhanced interrogation techniques’ it authorized were not torture under United States law. The Times adopted the view that labelling these as ‘torture’ in news articles could create the appearance of taking sides.

What about the second facet of the fact-checking project: the notion that FCUs might help the media recover its traditional authority? The unspoken assumption is that, by definitively pronouncing on facticity, respectable media outlets will regain the cultural and political stature they enjoyed throughout most of the twentieth century.

This framing entirely reverses the causal sequence. The newspapers of old did not possess authority because they could determine truth – on the contrary, they could determine truth because they possessed authority.

Our ideas of journalism are still very much shaped by the postwar boom, a period of political and economic stability in which newspapers, in particular, were an integral part of the liberal order. If you were a good citizen, you read one of the two or three papers available in your city, because in these papers matters of social import – from literary taste to public policy – were discussed and debated and settled. The media’s centrality to the public sphere – to good governance and the social order – underpinned its business model. A metropolitan newspaper could fund journalism with advertising because it possessed a captive market for advertisers: if a person wanted to participate in the social life of that city, he or she needed the paper.

For better or worse, that period is over.

In our era of uncertainty and instability, with the social infrastructure of the long boom degraded, a much more atomised citizenry has disengaged from – and, sometimes, become actively hostile towards – the old institutions of public life. The fragmentation of the traditional public has collapsed the old journalistic business models – which, significantly, were in trouble even before the internet. Without a locked-in readership, media outlets can only sell advertising if they generate truly huge numbers. Hence the BuzzFeed strategy: the site produces viral posts for sponsors, deliberately breaking down the traditional distinction between paid and unpaid content. The listicle entitled ‘15 Most Inspiring Moustaches in the Animal Kingdom’ might look identical to BuzzFeed’s other fare, but it’s not – it was created to promote Gillette razors.

The BuzzFeed model works because viral content so perfectly matches twenty-first-century conditions. The output from BuzzFeed and its imitators might be social – it relies on ‘sharing’, after all – but it’s not in any sense collective. On the contrary, it depends on a distinctly neoliberal sense of individuality, one in which an individual’s personality is expressed by the products he or she consumes. He’s the guy who likes LOLcats; she’s the one who shared that funny clip from a Japanese talk show.

Making minimal demands on readers, BuzzFeed memes are ideal for harried workers seeking what Zimmerman calls ‘fleeting instances of joy or comfort’ – that is, an escape from the routine of a dull cubicle job. By contrast, the old broadsheet newspaper, with its solemn pronouncements on matters of weight, was designed to be read at the breakfast table by the family’s breadwinner before he went off to work.

The difficulty in applying that model to the contemporary environment extends beyond format or word count or prose style. Traditional journalism depends upon a reader integrated into a liberal political culture, someone who knows and cares about the grand narratives of public life. Contrast that with the ‘true but not true’ sensibility of Zimmerman’s memes: entirely self-contained, his products are capable of delivering their payload of laughter or pathos without any reference to the downer of the real.

In the past, a major newspaper catching a prime minister out on a lie might have brought down an administration. Today, despite being repeatedly fact-checked, Abbott continues to call asylum seekers ‘illegals’, because he knows that tut-tutting admonitions from the broadsheets do him no damage and might even lend him some ‘anti-elitist’ credibility – as per Toronto mayor Rob Ford, who actually increased his popularity after footage of him on a crack-fuelled rant went viral.

In that context, let’s revisit the debate in The Lifespan of a Fact.

Fingal, the fact-checker, is surely correct to suggest there is something wrong with a theory of art that attempts to reveal the truth about the world by lying. If you are a serious nonfiction writer and your aesthetic goals rely upon falsification, well, there is probably something wrong with your aesthetic goals.

At the same time, Fingal’s objections do not get to the heart of the matter. No, nonfiction writers should not lie, but that does not mean truth consists simply of an aggregation of verified facts. The art of nonfiction – and there is such a thing – entails revealing the deeper truths behind the surface appearance – not by falsifying reality but by exploring the context in which facts come to be. The writer says what is, but more than that, he or she shows how what is now relates to what once was and, even, to what yet may come.

This is an aesthetic task but also a political one.

For, in a sense, the Bushites were right: those who act do create their own reality – not out of thin air but by changing the conditions they inherit from the past.

As Jess Whyte argued in Overland some time back:

The [Left’s] easy dismissals of the Bush adviser obscure the extent to which his comments expressed an idea once central to the Left: the possibility that political action can transform reality. Today, few on the Left still believe it possible to alter the entire political terrain. The portrayal of the aide’s comments as delusional illustrates the extent to which those who once believed in the transformative capacity of political action have succumbed to pragmatic realism.

If the proliferation of information in the digital age has not dispelled the prevailing dishonesty of our times, it’s not merely because we haven’t been writing well enough. It’s also because, in a sense, the era in which we write creates the conditions of reception for our critiques. We cannot re-create the circumstances that produced the old media models – and nor should we try. Indeed, the Left should not mourn the collapse of the newspaper giants’ monopoly over information.

We need, instead, to create new audiences, to build communities for whom information matters. In a previous Overland, Rjurik Davidson described this in terms of creating a counter-hegemonic public sphere, a counterculture of writers and readers, thinkers and activists, publications and campaigns, and an infrastructure that allows people to make news rather than merely consume it.

To put the matter another way, it’s not enough for nonfiction writers to simply describe the world. The point, as always, is to change it.