Patrick Thaddeus Jackson| Renewal 29/2
The day after Donald Trump’s inauguration as President of the United States in January 2017, then-press secretary Sean Spicer claimed that the crowd had been larger than the crowd that had assembled for Barack Obama’s second inauguration, and offered as evidence a set of what appeared to be estimates of crowd sizes:
… from the platform where the President was sworn in, to 4th Street, it holds about 250,000 people. From 4th Street to the media tent is about another 220,000. And from the media tent to the Washington Monument, another 250,000 people. All of this space was full when the President took the Oath of Office’.1
All of these numbers were bogus. A pair of sociologists specialising in crowd size estimates placed the carrying capacity of the entire National Mall – the expanse of green space in front of the US Capitol building, where crowds gather for events like these – at only 524,103, almost a quarter of a million fewer people than Spicer claimed were present.2 Aerial photographs also clearly showed that the space was less full for Trump’s inauguration than for Obama’s.3
The problem is not simply that Spicer was lying, but that Spicer, and the rest of the administration, kept on repeating these claims even after they had been debunked, and suggested that they were simply offering – in the infamous words of Kellyanne Conway, a senior advisor to Trump – ‘alternative facts’ to those presented by official agencies and scholarly specialists. Combined with a dismissal of any media outlet that criticised Trump as ‘fake news’, this strategy drew a protective barrier around the Trump administration that effectively allowed it to make whatever claims it wanted to, and have those claims repeated and recirculated by media outlets sympathetic to and supportive of Trump. And Trump supporters gained their own separate set of supposed facts that they could draw on in discussions with other people – a set which collapsed the distinction between partisan talking-points and truthful representations of shared situations.
To simply focus on the psychological level – did Spicer and Conway know that what they were saying was untrue? – fails to grasp the broader significance of the embrace of ‘alternative facts’. That untrue claims sometimes have great political valence is nothing new. But the appropriation of the epistemic authority of facts for purely partisan purposes is a new and disturbing development, and it’s not just confined to the United States or to the Trump administration. The problem is that these claims are taken to be based in facts, even when they have been clearly shown not to be. This in turn indicates a very profound and wide-spread confusion about what it means for something to be a fact in the first place.
Facts and their contexts
To illustrate, consider a different example. We’ve all heard it said, often on romantic occasions, that a person’s truest love match is their ‘soulmate’. But we also know from experience that not all relationships work out, and that sometimes people whose ‘soulmate’ did not turn out to be their forever match end up finding a measure of happiness with someone else afterwards. In his book What If?, Randall Munroe subjects the idea of the ‘soulmate’ to a statistical analysis, calculating the precise odds of a person’s ever meeting their soulmate and concluding that you would need 10,000 lifetimes to have a chance of finding ‘the one’ – and that’s assuming meeting ‘a few dozen new strangers each day’ in the quest for true love.4
The absurd humour of the situation Munroe describes comes from a particular kind of epistemic confusion: taking a statement that isn’t usually understood to be a testable empirical claim and treating it as though it were. Certainly most people who talk about soulmates don’t mean it in a factual sense; they are instead expressing a sentiment or reporting an experience, neither of which are generally amenable to systematic empirical analysis. What makes a romantic sentiment ‘valid’ is that it captures the way that people feel about a situation, or perhaps the way that they believe that they are supposed to feel. In the context of a wedding, say, or an anniversary card, the claim serves to reinforce those feelings and the broader narrative of romantic love in which they are embedded. The claim does not derive its power and authority in that setting from its factuality, but (so to speak) from its authenticity: the claim is recognisable to the participants, and is repeated and recirculated just so long as it helpfully frames and summarises a situation.
So the very same words can be taken as a statement of fact, as an expression of sentiment, or a number of other things. To determine what a claim means, we need to make some decisions about the context within which it is being made, and therefore the intention of the speaker; having done so we can then bring to bear the appropriate techniques for evaluating it and seeing whether and in what sense it is valid. The context, and therefore the intention, of a claim is thus something that we the listeners or readers produce or generate by how we engage with the claim in the first place – and by how others react to our construal of the claim. If at a wedding you stand up to protest the factuality of a claim about soulmates, you are likely to be shouted down or to have something tossed at you, but not because you have failed to understand the claim; instead you have simply understood it inappropriately, treating it as something different from the ways that others treat it in that setting.
This is not just a conflation of the ‘literal’ and the ‘figurative’ meanings of a sentence. Consider a sentence like ‘this vaccine is effective against Covid-19’. Certainly one way to understand that claim is to bring to bear a battery of empirical tests, and to regard ‘effective’ as verbal shorthand for ‘statistically significant at the .05 level’. But is that the literal meaning of the words used? A quick glance at the Oxford English Dictionary reveals that words like ‘effective’ (to say nothing of ‘against’) have a plethora of meanings. Indeed, one meaning of ‘effective’ is something that cannot be measured directly! The very opposition between literal and figurative meanings is itself a product of privileging one way of making sense of claims, emphasising the empirical, the tangible and the factual – and dismissing alternatives as less worthy. Here again it is not the words of the sentence that tell us how to make sense of them; and insisting that there is only one ‘literal’ way to construe a claim cannot resolve the conundrum.
While we could call the empirical/tangible/factual way of construing claims ‘scientific’, doing so obscures rather than clarifies the issues at play. In ordinary speech, ‘science’ designates a number of different but related things: an approach, an institution, a body of knowledge, sometimes an anti-religious source of truth. Disputes abound over what is and is not properly considered a ‘science’, and the stakes of such disputes can be quite high: if a field is regarded as ‘scientific’, that comes with both a certain priority in the competition for funding (‘STEM fields’ are nowadays closely linked to the enterprise of providing youth with gainful future employment, and to national issues of global competitiveness) and a certain epistemic authority for the conclusions reached by researchers: listen to us not because we are powerful interest-groups, but because we are correct. With so much at stake, labelling an approach ‘scientific’ is fraught with controversy.
‘Factic’ claims
At this point I will take the liberty of introducing a novel term to designate that approach to construing claims that emphasises the empirical and the tangible meanings of words and looks to evaluate claims by comparing them with systematic evidence. That word is ‘factic’.5 Much as the word ‘aesthetic’ designates a way of construing claims that emphasises their artistic qualities – the sonorous aspect of language, the beauty and harmony of verbal expression – the word ‘factic’ designates a way of construing claims that treats them as, so to speak, aspiring to factuality. We can’t call this way of construing claims ‘factual’ because that word, applied to a claim, suggests that it is already factually true, when what we are after is a name for a way of grasping claims as candidates for being either factually true or factually false.
To treat a claim factically is to construe it as a statement that should be evaluated with empirical evidence, as opposed to other ways we might treat it – aesthetically, say.
Now, because the words of a sentence do not tell you how to construe a claim, you have to look for other contextual clues in making sense of what is said. Claims appearing in the pages of scholarly journals can, for the most part, be safely understood as factic claims, because the process of peer review should have weeded out assertions of personal preference, or at least insisted that expressions of the author’s subjective bias be clearly labelled as such. Claims in the newspaper (other than in the opinion section) similarly purport to be factic, backed up by the journalistic procedures of fact-checking and the occasional retraction (the very existence of such retractions further reinforces the point that the claims on the newspaper page are factic without always being factual.) These clues are not determinate; one can, for example, apprehend a piece of scholarly or journalistic writing aesthetically, without regard for its factual truth or falsity. But ordinarily we do not do so, because of the customs and conventions in which we have been trained and socialised: we read the newspaper to get facts, and we go to the museum to see art. Armed with these presuppositions, we construe claims in the way that we presume that they were intended, and apply the appropriate standards (appropriate to the context, that is) in evaluating them.
There is no such thing as a ‘literal’ meaning which would definitely say what kind of claim a sentence delivers. There are only distinct ways of apprehending and construing claims, each of which comes with a different set of standards and procedures for evaluating a claim. It is a matter of sociological and historical interest that we nowadays often regard the factic construal of a claim as its ‘literal’ meaning; this privileging of the factic tells us something about our post-industrial societies and the importance we place on empirical observation and systematic evaluation. But custom is not, in itself, a justification for something.
Weaponising facticity
Actually evaluating a factic claim, and concretely producing a fact, is quite a bit more difficult than we often suppose. In the absence of a widespread understanding of that difficulty, the epistemic authority of a fact can be all too easily appropriated for political purposes. We thus get claims that appear, or purport, to be factic, when they are actually operating with a very different intention: to support a desirable political outcome or goal. The best evidence for this is that such claims, if actually regarded as factic, quickly fall apart – but this matters not at all to their adherents.
Consider the claims repeatedly advanced by the Trump campaign in the United States regarding the 2020 presidential election. The election was stolen, Trump and his allies claimed: Trump actually received more votes and then state and local officials changed vote totals so that it looked like Biden had won. And we have proof, the campaign claimed, and proceeded to file at least 63 separate lawsuits about the issue. If we treat these claims as factic claims, then the proper way to evaluate them would have been to produce evidence, and the court cases promised to do just that. The campaign did not succeed in any of these lawsuits; despite being given ample opportunity to produce evidence, the Trump campaign failed to convince a single court that their claims were factually true. In a process intended to evaluate whether claims were supported by empirical evidence, the claims had been judged and found wanting, and could and should be dismissed.
Except that this is not what happened. Polls showed that a majority of Republicans – in some polls as high as 75 per cent of Republicans – believed that Trump had won the election, and that Biden’s apparent victory was the result of electoral fraud. The campaign’s claims continued to circulate on social media, and were repeated by Trump and his supporters at every possible opportunity – a process that culminated (at least for the moment) in the 6 January 2021 attack on the US Capitol, as scores of people ransacked the building in the name of opposing a fraudulent election. The use of the campaign’s claims as justification for violent action suggests that even potential doubters among the rioters thought that the audiences to which they were appealing would regard the claims as factual, as does the continued circulation of these claims on social media to this day. Once again, we see the disconnect between processes that treated the claims as factic and rejected them, and processes that utilised the claims because they supported a preferred outcome – in this case, a Trump victory and a Biden defeat.
To treat a claim as factic is to treat it as aspiring to factuality, as intending to represent a state of affairs in a way that is impersonally true to any observer. Facts are not personal possessions; they are not the property of any particular observer or speaker. ‘This glass is half-full of water’ is a fact just so long as everyone who speaks our language – everyone who uses the words ‘water’, ‘glass’, and ‘half-full’ in the same ways that we do – would agree with the claim, because to disagree would be to demonstrate that they don’t know how to use those words correctly.
If we are dealing with an apparent disagreement over a fact, there are actually only three options. First of all, two people might disagree over whether the glass was ‘half-empty’ or ’half-full’. Here, however, there is actually no disagreement over the facts – ‘half-empty’ and ‘half-full’ mean the same thing, factically construed – but only a disagreement over whether to put an optimistic or pessimistic slant on the facts. Second, the speakers are talking about different things, and thus talking past one another: one says that the glass is half-full while the other says that the glass is wasteful (because clearly someone failed to drink all of their beverage): both can be true at the same time, and once again the apparent disagreement evaporates. Third, one of the parties is simply incorrect: ‘I thought the glass was half-full, but on closer inspection I see that it is actually only a quarter-full.’ Factic construal of a claim does not mean that what we think are facts are somehow immune to criticism; on the contrary, those claims that we think are factual stand perpetually in danger of being overthrown if the evidence no longer supports them. To engage in the factic evaluation of claims always means holding out the possibility of being incorrect in our assertions, and reserving the label of ‘fact’ for those statements which, were someone to disagree with them, would lead us to wonder what language our interlocutor was speaking.
All of which means that facts are much rarer than we sometimes think that they are. Of course we want our cherished beliefs to be facts; we want them to be not merely things that we hold to be true, but things that are true, empirically and impersonally. But the realm of statements that every competent speaker of the relevant language would agree with is quite a small one, once we get past the sorts of descriptions that we use in everyday life to orient ourselves towards objects and towards one another. Even there, we encounter ambiguities and difficulties: does ‘bring me the large spoon’ mean bring me the spoon with the longest handle or the spoon with the biggest liquid capacity? In such situations it is fair to say that there are no facts, because conventional usage is insufficient to provide a clear answer (and in order to get you to bring me the spoon I want, I may have to describe it in some other way, some factual way – such as ‘bring me the spoon on the left-hand side of the drawer’).6
Once we get out of the practical context of everyday life, conceptual and technical ambiguities abound to an even greater degree. Physicists disagree about what constitutes evidence for or against string theory; economists disagree about how to measure growth; international relations scholars point to different bodies of evidence about the relationship between democracy and war and come to different conclusions about whether there is a ‘democratic peace’ or not; philologists and theologians disagree about the correct translation of Galatians 2:16. All of these scholarly controversies take place within a factic approach, so to speak, as long as they are concerned with establishing which claims are best supported by empirical evidence. Although there are different methodologies between, and sometimes within, scholarly fields and disciplines, most scholarly endeavours – those that we call the ‘natural’ and ‘social’ sciences, as well as a good proportion of what we call in English the ‘humanities’ but which in German are called the Geisteswissenschaften, the ‘human sciences’ – fall broadly into the factic realm. The heart of substantive scholarly contention is precisely the quest to establish a claim as factual, presenting one’s case in a way that is impersonally compelling to the relevant group of one’s scholarly peers: the speakers of the relevant language. To have secured just one fact would be an immense accomplishment, and the ongoing scholarly controversies about virtually everything show just how little has been settled factually to date.
Facts are difficult to establish, but once established, carry epistemic authority: because no one can sensibly disagree with them, they can serve as an impersonal basis on which to proceed. This is almost certainly why people want to establish facts. But having the facts on one’s side can also be a powerful weapon in a political debate, and without a clear understanding of what facts are, the process of establishing facts is open to being abused.
Putting facts in their proper place
Over a century ago, in his famous lectures on the vocations for scholarship and politics Max Weber identified the difference between those two realms as consisting precisely in the distinction between factic claims and claims intended to produce effective change in the world. If you speak about democracy in a public meeting, he argued, you have a ‘damn duty’ to make your value-laden perspective on the topic known, but if you are discussing democracy in a seminar room or a lecture hall, you need to confine yourself to factual issues like an analysis of the causes of different systems of governance or the consequences of different electoral and administrative procedures.7 Although this depiction of scholarship is often caricatured as ‘value neutrality’ – something that Weber actually thought was impossible – what is often lost in translation is the reason that Weber advocated this separation of the political and scholarly realms. Only by confining scholarship to a focus on factic claims, Weber claimed, could the epistemic authority of facts be preserved, allowing scholarly conclusions to serve as an input that sets the parameters of the ensuing political debate. Because facts would emanate from a place outside of politics, they could serve as a neutral basis on which conflicting parties could agree – and this in turn would allow political debate and discussion to focus on questions of which outcomes are desirable. This doesn’t mean that scholars either don’t have value commitments or shouldn’t make them known to anyone else. But the value of scholarship for politics, in Weber’s view, lay not in the value commitments of scholars, but in those factual claims that should be compelling to everyone, regardless of their value commitments.
In the absence of the clear distinction between factic and partisan claims, we have something especially pernicious: the circulation of claims in the political realm – perhaps especially by unscrupulous media outlets – that purport to be factual, but which are grounded in no process other than the ordinary political calculus of partisan advantage. If we looked at claims like those made by the Trump campaign without construing them to be factic claims, that might attune us to the way that such claims resonated emotionally with people who feel alienated from the political process, and perhaps to the way that those claims served to direct anger and motivate action. We would, in effect, be suspending the question of whether such claims were true or false, and focusing instead on what they revealed about the experiences of the speakers. The sentences in question would become signs of a system of meanings quite distinct from our own, and ethnographic evidence of a different kind of common sense in which different things were taken for granted.
But key to the political effect of these claims is that they were taken to be factually true by the people acting to undermine the possibility of a Biden presidency – they were publicly construed as factic claims, and true claims at that, despite the complete absence of evidence supporting them. We might call them counterfeit facts: claims that look like they are supported by a reliable process of factic evaluation, but are not. Just like counterfeit money is passed off as a common store of value when it is actually just a ruse, a counterfeit fact is treated as though it were impersonally true and carried epistemic authority when it does not. The ordinary circulation of money, and facts, presumes trust and good faith on the part of the parties to the transaction; counterfeit items in either realm are an abuse of that trust. Counterfeit facts are not merely lies; they result from the deliberate confusion of different ways of construing a claim. Instead of mere errors, they are subversions of the whole process of the factic evaluation of claims. When a fact becomes simply a strongly held belief, we have lost the entire notion that any claim could ever be impersonally, factually true.
The solution, or at least the beginning of a solution, is to strive for more clarity and specificity not just in the words that we use, but in the means of construal and evaluation that we bring to bear on claims in different domains. Scholars need to avoid overreaching, and to confine their inquiries to the realm of the factic: causes and consequences of likely courses of action, disclosure of the tacit rules for appropriate action, and the epistemically adequate support of claims by empirical evidence. That means both being intellectually honest about the limits of scholarly inquiry – it cannot factically settle disputes that are actually about divergent values – and using the scholarly comparative advantage in factic evaluation to take on counterfeit facts and expose them for what they are. The latter sometimes means construing claims as factic even while suspecting that this isn’t how the political actors intended them, and then demonstrating that they are factually false; the former means, in part, refraining from efforts to back even desirable value orientations with facts. Both parts of this principled refusal to allow the purported epistemic authority of counterfeit facts to continue unchallenged are a necessary, but not sufficient, component of a campaign to put facts back into their proper place.
Complementing this scholarly strategy, political actors need to be much more careful in their use of factual claims, only using those that are actually factual as a basis for action and being honest about the absence of settled facts in many areas. The absence of settled facts doesn’t by any means make action impossible, but it does call for judgment and commitment: what Weber called an ‘ethic of responsibility’. The responsible politician holds the facts in one hand and moral purpose in the other, and acts without losing sight of either. Rather than giving in to the temptation to base one’s political actions on convenient, counterfeit ‘facts’, a responsible political actor is honest about the uncertainty of the future while accepting accountability for what transpires. Consider the Biden administration’s framing of the facts about the rate of vaccination in the United States in terms of the hope that families will be able to engage in the traditional ‘backyard barbecue’ on Independence Day this July: this is presented as a possibility, not contradicted by the facts but not guaranteed by them either. Such a stance respects the distinction between factic and political evaluation, and avoids the absurdity of misapplying standards where they do not belong. And it avoids wreaking even more damage on the tattered boundary between factic and other forms of construal and evaluation.
As for the public: we must assume that people are savvy enough to recognise the difference between a claim that is meant to be evaluated factically and one that is not, as long as both scholars and politicians provide appropriate contextual clues. While we can and should work to ensure that our educational systems are teaching students to operate with different ways of understanding and evaluating claims – not just factic and political, but aesthetic, ethical, technical, and so on – in appropriate contexts, we should also recognise that people already make these kinds of switches all the time. No one lives in a completely factic world. After all, despite the statistical unlikelihood, people still look for their soulmates, and sometimes even find happiness. That, to be sure, is a fact.
Patrick Thaddeus Jackson is Professor of International Studies in the School of International Service at American University, Washington, DC.
Notes
For helpful comments on and discussion of an earlier draft I thank Cathy Elliott, Lori Younissess, Ian Reynolds, Lucas Dolan, Adam Lerner, and the students in my ‘Knowing and the Known’ ECPR course in March 2021.
[1] Chris Cillizza, ‘Sean Spicer held a press conference. He didn’t take questions. Or tell the whole truth’, Washington Post, 21 January 2017.
[2] Clark McPhail and John McCarthy, ‘Who counts and how: Estimating the size of protests’, Contexts, Vol 3 No 3, 2004, pp16-17.
[3] David Z. Morris, ‘D.C. transit stats show weak demand during Trump inauguration’, Fortune, 22 January 2017.
[4] Randall Munroe, What If?: Serious scientific answers to absurd hypothetical questions, New York, Houghton Mifflin Harcourt 2014, p24.
[5] This is not a completely new word, but a resurrection of an archaic English form and usage.
[6] Any resemblance of this example to Ludwig Wittgenstein’s discussion of builders in the opening pages of Philosophical Investigations, Oxford, Blackwell 1958 is completely intentional.
[7] Max Weber, The Vocation Lectures, Indianapolis, Hackett Press 2004.