The other day, I told a friend that Knoxville is the capital of Tennessee. Five seconds and a blur of fingers later, he said: ‘No, it’s Nashville.’
My statement was obviously not true. But since I sincerely believed in the accuracy of what I was saying, I was nonetheless being truthful. I was mistaken, not mendacious. This distinction between truth and truthfulness is vital, but in danger of being lost in debates over ‘post-truth’ politics and ‘fake news’.
Most of us probably inadvertently share trivial untruths quite frequently. Nowadays, there’s almost always somebody, smartphone in hand, ready to set us straight. If it really matters, we’ll likely take the trouble to check our facts.
The stakes of getting it wrong are much higher for the institutions of media, academia or government. So they try hard – or at least should try hard – to get things right. My Knoxville mistake wouldn’t have made it into The New York Times, unless several factcheckers and copy-editors in the editorial process were negligent.
But even honest journalists and careful scholars will sometimes get things wrong. Honest mistakes are made. Once flagged, these errors will be immediately corrected and acknowledged; there might be some hard questions asked about process failures, too. But there’s a very big difference between an error and a lie – and between ‘fake news’ and ‘false news’. A fake is always false, and was intended to be. But a falsehood is not always a fake; it could simply be a mistake.
Political partisans try to blur this critical distinction. In 2018, the then US president Donald Trump and the Republican National Committee handed out 11 ‘Fake News Awards’ to various media outlets. But most of the news items had been noted and immediately corrected and/or retracted with full explanations. They were false statements, but they were not fake.
When it comes to public health information, the stakes are higher still. When the COVID-19 pandemic hit, we all wanted instant, accurate advice on what to do and what not to do. But the virus was, as the interim name given to it reminds us, novel. Scientists were scrambling to figure out what it was, how it spread, and how to defeat it. The honest answer to many of our most urgent questions was: ‘We don’t know yet.’ Guidance had to be given with incomplete information. Many errors were made. Some of the early advice was wrong – it turned out that masks were even more important than handwashing, and that outside was very different to indoors, and so on. Many pieces of official advice were false, but not fake. The most important question for citizens is not whether public health advice is always right, it’s whether public health officials are consistently trying to get it right – and communicating what the sociologist Zeynep Tufekci called the ‘the full painful truth’, honestly and clearly. Trust is built on truthfulness rather than truth.
Faced with an urgent need for information, we just want the truth. In response to the question ‘Which way to the emergency room?’, all that really matters is the accuracy of the response. But much of the time, it’s more important that a person is speaking truthfully than that they’re speaking the truth, especially when the answers are not yet clear. Most of us feel very differently towards the friend who makes an honest mistake, perhaps based on inadequate information, and the one who tells a deliberate lie. And we also know that nobody can be 100 per cent right about everything 100 per cent of the time, and forgive inadvertent errors made along the way. The same should hold for our institutions.
At the end of 2020, the former US president Barack Obama told The Atlantic:
If we do not have the capacity to distinguish what’s true from what’s false, then by definition the marketplace of ideas doesn’t work. And by definition our democracy doesn’t work. We are entering into an epistemological crisis.
I think Obama is right about the epistemological crisis. But I think it goes deeper than he suggests. The problem is not simply one of being able to discern true from false. The problem is being able to discern who is even attempting to present the truth, even if they don’t always succeed. The question is not ‘Where is the truth?’ It is: ‘Who is being truthful?’
The truth of a statement can be empirically tested: that’s what factcheckers are for. Various media organisations rank claims in terms of their veracity. In the US, the Washington Post project ‘The Fact Checker’ gives the claims of politicians a Pinocchio score from one to four; and the Poynter Institute’s PolitiFact has a six-scale Truth-o-meter that goes from ‘True’ to ‘Pants on Fire’.
But truthfulness is harder to assess, since it requires us to know what the speaker knows – to peer into their heads. Was it a lie, or an honest mistake? One way to tell is how a person reacts to evidence that their claim is false. If they continue to repeat it regardless, they’re clearly not being truthful. An honest mistake made on Monday and corrected on Tuesday becomes a lie if repeated on Wednesday.
As Bernard Williams, in his prescient final book, Truth and Truthfulness: An Essay in Genealogy (2002), wrote:
if what one believed turns out to be false, it does not follow that one ought not to have believed it. What does follow is that if one recognises the falsehood, one does not carry on having the belief …
Truth is empirical, but truthfulness is ethical. Truth is the end product; truthfulness a vital element in its production. The epistemological crisis has been blamed on populist politicians, and on technological platforms, and on profit-seeking celebrity trolls. These are all important. But the real problem is a loss of virtue, specifically the virtue of truthfulness. The epistemic crisis is an ethical crisis; and will require ethical solutions.
Making more effort often means turning to a higher-quality news source than your Facebook feed
Our patron saint in this effort could be Nathanael, who appears in the Gospel of John, and has a good claim to be the patron saint of truthfulness. When told about Jesus, he scoffed: ‘Can anything good come out of Nazareth?’ But Christ, knowing he had said this, exclaimed: ‘Behold, an Israelite indeed, in whom is no deceit!’ Christ was clearly not applauding Nathanael for the truth of his statement, but for his willingness to speak his mind – for his truthfulness.
Williams argued that truthfulness rests on two basic virtues: accuracy and sincerity. The virtue of accuracy requires that ‘you do the best you can to acquire true beliefs’. This doesn’t mean that we all need to try to become world experts in absolutely everything. There is, as Williams puts it, a necessary ‘epistemic division of labour’. We usually trust others to know how to ensure that our drinking water is safe or how to remove our appendix. The point is to do what we can, especially with regard to more important matters, within the limits of our knowledge and competence.
Accuracy is, Williams adds, ‘the virtue that encourages people to spend more effort than they might have done in trying to find the truth, and not just to accept any belief-shaped thing that comes into their head.’ My friend was demonstrating this virtue when he checked my claim about Knoxville. A person who reads the public health literature on wearing a mask or taking a vaccine, rather than simply believing what their neighbour tells them, is doing the same. Making more effort can often mean nothing more than turning to a higher-quality news source than your Facebook feed for, say, election news or Covid coverage.
Having tried to get accurate information, we should then share it fully and honestly. This is the virtue of sincerity, which demands, as Williams puts it, that ‘what you say reveals what you believe’. This might sound easy. But there could be circumstances in which we’re tempted to hide our beliefs, or at least some portion of them, and when it will therefore take some courage to do so. Perhaps you have come to believe that it’s important for everyone in your church to wear a mask during services, but most of your fellow congregants think that mask-wearing is politically correct nonsense. When the subject comes up, it might be easier, at least from a social point of view, to remain silent. But to be sincere means to speak up. If courage is sometimes required of the speaker, trust is asked of the hearer. Williams again: ‘Truthfulness is a form of trustworthiness, that which relates in a particular way to speech.’
There can, of course, be occasions when it is morally justified to be more sparing, as Edmund Burke argued. ‘Falsehood and delusion are allowed in no case whatever,’ he wrote in Two Letters on the Proposals for Peace with the Regicide Directory in 1796. ‘But, as in the exercise of all the virtues, there is an economy of truth.’
Burke stressed that ‘being economical with the truth’ is justified only in special circumstances. These might include the withholding of a trivial truth in order not to cause unnecessary offence; during political or international negotiations; or in cases where withholding the truth will save lives. As a general principle, being sincere means not holding out on the listener. This is why witnesses are required to swear to tell not just the truth, but the whole truth (at least the whole truth as relevant to the question at hand).
These two virtues of truthfulness – accuracy and sincerity – are most precious among those who occupy the institutions of research, teaching and communication. As Williams notes, the authority of academics is rooted in their truthfulness in both these respects: ‘they take care, and they do not lie.’ The same can (or at least should) be said of journalists and judges.
The point is not to disprove one claim and prove another. It is to cast doubt over all claims
These professions lie at the heart of what the writer Jonathan Rauch describes as the ‘constitution of knowledge’. This constitution operates according to certain rules, in particular, the freedom to hypothesise and the responsibility to undergo review and scrutiny. As Rauch writes:
The community that follows these rules is defined by its values and practices, not by its borders, and it is by no means limited to scholars and scientists. It also includes journalism, the courts, law enforcement and the intelligence community — all evidence-based professions that require competing hypotheses to be tested and justified. Its members hold themselves and each other accountable for their errors.
Working against this constitution are the forces of what Rauch labels ‘troll epistemology’. Trolls seek not the truth, but the destruction of an enemy, ideological or personal. Trolls not only fail to display the virtues of sincerity and accuracy, they work in precisely the opposite direction, deliberately offering up distorted visions of reality, based on cherrypicked information. The 2020 US election was a masterclass in troll epistemology. Trump, his lawyers and army of supporters took tiny, isolated incidents of errors or even a few cases of genuine fraud to paint an overall picture of a ‘stolen election’. Trump himself cast doubt on voting machines, citing a miscount in Antrim County, Michigan. It turned out that there had, in fact, been an error – but it was an official’s mistake made when setting up tabulation equipment, quickly corrected, and nothing to do with the voting machines.
Trump and his acolytes hyped up similar stories about dead people voting, or people voting twice, or votes being ‘found’, of poll-watchers being barred, and so on. In each case, there might have been just a microscopic germ of truth, then wildly exaggerated, and hence all counter-evidence was systematically ignored. It might seem odd to point to several different ways in which the election was stolen, rather than actually manage to prove a single one, but that’s to miss the point. The point is not to disprove one claim and prove another. It is to cast doubt over all claims, to make the idea of truth itself unstable, by, as Steve Bannon put it, ‘flood[ing] the zone with shit’.
There has been a catastrophic corrosion in the virtue of truthfulness, and of course many of the populist leaders and movements of recent years must take much of the blame. But there are other forces at work, many of which helped fuel populism in the first place.
Social media platforms have acted as accelerators and amplifiers of untruths, creating frictionless environments for information, misinformation and disinformation to spread rapidly. Rather than encouraging more engagement with reliable sources, organisations such as Facebook, Twitter, YouTube and others have facilitated the creation of epistemic bubbles, filled with people who mutually confirm their existing prejudices.
Social media companies are wrestling with this dilemma, since customers want clickbait (we’re only human after all), which drives revenue (so providers want this even more than customers). Right now, sincerity and accuracy seem to make for unappealing bait, and so can go unclicked. The business models of these companies rely on engagement, and the kind of content that maximises engagement is of the truth-diluting rather than truth-enhancing kind, as Tufekci and others have argued. These firms are now more aggressively flagging and downgrading false or misleading content, and policing hate speech more tightly. But the problem is structural. Should they try to make money or try to make truth? They can’t do both.
The virtues of truthfulness are also under pressure in university departments, think-tanks and newsrooms, perhaps in less dramatic ways but, in the long term, equally dangerous ones. Political polarisation incentivises even good people not only to pick a side, but to start cherrypicking their data to support it.
Scholars with ideological agendas can easily present data in a way that confirms their priors, even if that form of presentation isn’t the most robust – or at least, is only one of many ways in which it could be presented. Let me give a concrete example. The economic condition of the middle class in the US is an important question. So, what happened to the US median income between, say, 1979 and 2014? Well, it increased by 51 per cent. Or possibly 37 per cent, 33 per cent, 30 per cent, or 7 per cent. Or maybe it fell by 8 per cent. All of these answers are correct. It just depends on which study you read, and which methodologies the various scholars chose to use.
The world is a complex place, and the search for simplicity is very often what gets us into trouble
Say you have done as Bernard Williams urges and invested some of your own precious time on this question, only to come up with all these competing answers. Frustrating, perhaps. You might well ask: ‘Well, which is it?’ The point is that none of them are wrong, in the sense of being false. They are simply arrived at by different methodologies.
The danger is that Left-wing scholars adopt approaches to produce a particular result, and Right-wing scholars do the opposite, and each present their findings as ‘facts’. A partisan media can then amplify one of these ‘facts’ to suit their priors. Before you know it, people have starkly different views on the matter, and their views are based on perfectly solid research. The point here is not that we can’t know anything, it’s simply that the world is a complex place, and that the search for simplicity is very often what gets us into trouble. For scholars, the most important thing is to strive to present their work in a way that’s as objective as possible (accuracy), and to present a range of reasonable results wherever possible, giving the fullest possible picture (sincerity).
Scholarly truthfulness is especially important when it comes to assessments of public policy. It’s all too easy to construct evaluation studies in ways that produce positive results. This is understandable. Few funders, public or philanthropic, are thrilled to learn that their billion-dollar initiative failed to achieve any of its goals. The pressure then, even when it comes to well-conducted studies, is always to highlight any positive findings and downplay the disappointing ones. Rather than evidence-based policymaking, we end up with policy-based evidence-making.
In all of these cases, there’s a need both for institutional and individual responsibility. Media companies, universities, think-tanks and political parties will all have to work much harder to maintain the norms of openness, fallibility and exchange that facilitate the production and dissemination of knowledge. But we can’t just abdicate responsibility to institutions. This is a question of our own personal ethics too, and of our commitment to living and acting in a truthful manner.
A liberal to his bones, Williams was highly sensitive to the dangers that had accompanied the rationalist revolution of the Enlightenment, above all any attempt to order society around a scientifically grounded ‘truth’ about justice or human nature. That way lies tyranny. But the great gift of Enlightenment liberalism is the individual and collective striving to learn, to know more, about ourselves, each other, and the world. ‘We have something to fear from Enlightenment programmes for the advance and application of truth, but a lot to cherish in its concern for truthfulness,’ he wrote.
The blame for our epistemic crisis has been laid at the door of politicians, scholars and journalists, and the organisations they inhabit. And there’s plenty of blame to go around. But if Williams is right, and I think he is, then the root of our problem is ethical. The solution is for us to do better, to be better. Certainly, being truthful is a hard task. But without it, free societies cannot function. And nobody said freedom was easy.