Bullshit in Politics Pays

Abstract

Politics is full of people who don't care about the facts. Still, while not caring about the facts, they are often concerned to present themselves as caring about them. Politics, in other words, is full of bullshitters. But why? In this paper I develop an incentives-based analysis of bullshit in politics, arguing that it is often a rational response to the incentives facing different groups of agents. In a slogan: bullshit in politics pays, sometimes literally. After first outlining an account of bullshit, I discuss the incentives driving three different groups of agents to bullshit: politicians, the media, and voters. I then examine several existing proposals to combat bullshit in politics, arguing that each will fail because they ignore the relevant underlying incentives. I conclude somewhat pessimistically that a certain amount of bullshit in politics is inevitable.

Chaos, rage, headlines front page,

Break out the spotlight, bullshit centre stage. (Pase Rock)

Introduction

Politics is full of people who don't care about the facts. They vote in elections, run for office, report the news, write opinion pieces, share clickbait, and more. Rather than caring about the facts, they're content instead to merely present themselves as caring. But why do politicians play fast and loose with the facts? Why are voters so often vocal on matters they know so little about? Why do people create (and consume) fake news? In short, why is there so much bullshit in politics?

In this paper I develop an incentives-based analysis of bullshit in politics, arguing that it is often a rational response to the incentives facing different groups of political agents. In a slogan: bullshit in politics pays, sometimes literally. This has important implications, both for how to understand bullshit in politics as an empirical phenomenon and how to address it as a practical problem. As we'll see, existing interventions to reduce the amount of bullshit in politics, as well as similar proposals to combat the spread of related phenomena such as fake news, fail to recognize the extent to which it is a product of widespread incentives.¹ This failure renders the success of such interventions highly unlikely. Indeed, these interventions might even exacerbate the problems they're supposed to fix.

I begin in section 1 by first outlining what I mean by ‘bullshit’. Although there are many different things we pick out in ordinary language with this expression, I restrict my attention to a phenomenon wherein people communicate without regard for the truth of what they communicate.² Next, in section 2, I discuss the sort of incentives that commonly drive three different groups of political agents to bullshit, focusing on politicians, the media, and voters. After highlighting the many ways in which it can be individually rewarding to bullshit, I explain how bullshit in politics can be collectively harmful. Among other things, the more agents in politics bullshit, the more likely the spread of harmful misinformation. This sets the stage for section 3, where I examine several existing proposals for how to combat bullshit in politics: reliance upon personal responsibility, media literacy training, fact-checking organizations, social media reform, and governmental regulation. Each is likely to fail, I argue, since they ignore the underlying incentives that drive agents to bullshit in the first place. I conclude in section 4 on a pessimistic note. The incentives to produce bullshit emerge whenever large groups of agents with conflicting aims and ambitions, and each with finite cognitive and attentional resources, interact with one another. Accordingly, a certain amount of bullshit in politics is inevitable.

1. What is Bullshit?

In ordinary language, people use the expression ‘bullshit’ in many ways. For instance, speakers often use ‘bullshit’ dismissively. We call bullshit when we want to dismiss or reject what another says (Bergstrom and West 2020). We're also happy to call habitual liars bullshitters because of their habitual lies, thus viewing bullshitters as a kind of liar. In this paper, however, bullshit and lying are understood as distinct (though not mutually exclusive) phenomena.³ Lies characteristically involve an agent uttering something they believe to be false in order to get their addressee to believe the content of the utterance.⁴ Bullshit, though, involves agents communicating without regard for the truth of what they communicate. So understood, lying and bullshitting can overlap – someone can tell lies while being fundamentally indifferent towards the truth – but they are nonetheless distinct and often pull apart. One can lie without bullshitting, and one can bullshit without lying.

To illustrate the difference, consider two competing politicians running for office, Jessie and James. Both are willing and able to use underhanded tactics to gain advantage over the other. Predictably, both want to diminish the standing of their opponent in the eyes of the electorate. To do this, Jessie fabricates vicious rumors about James’ personal life, all while believing them to be false. When she tweets about his conduct to her millions of followers, her tweets contain lies. (Indeed, her statements are lies even if, pace her expectations, the rumors turn out to be true.)⁵ James, meanwhile, does not fabricate rumors about Jessie, opting instead to opportunistically share rumors about her that were already circulating online. Importantly, he knows neither their origin nor whether they are true. In fact, he does not care to know. The truth of these rumors is of secondary importance to their usefulness as a political weapon. In short, James is a bullshitter.⁶

Several additional features of this account of bullshit are worth clarifying. First, when thinking about bullshit, the focus is on how an agent's mental state relates to what they communicate rather than the truth or falsity of what they communicate. Bullshitters can sometimes say true things. But when they get things right, they do so accidentally, for the truth is not something that concerns them. In other words, something (whether an utterance, written statement, news report, and so on) counts as bullshit because of the attitude(s) of the agent(s) who produced it and not because it is false, misleading, or otherwise dubious (although much of what bullshitters communicate will in fact be false, misleading, or otherwise dubious).

Second, when using “communicate without regard for the truth of what one communicates”, I intend to pick out two different ways of relating to the truth. On the one hand, one might not care whatsoever about the truth of what one communicates. On the other hand, one might doubt the truth of what one communicates and yet still choose to communicate it (Mukerji 2018: 934). Additionally, the comprehensiveness of one's indifference towards the truth can vary. For instance, some agents might not care about the truth in general, while others might not care about the truth of a specific claim (or series of claims).⁷ In this paper, the focus is on agents in politics who lack regard for the truth of specific claims (or series of claims), whether this means they do not care whatsoever about the truth of such claims or they doubt such claims while communicating them nonetheless.⁸ Such agents may subsequently end up caring about the truth of what they communicate in the sense that they go on to hope, for whatever reason, that they get things right. But at the time of utterance (or of writing, etc.), bullshitters lack regard for the truth of what they communicate insofar as they either possess no attitude whatsoever towards the truth or they do not care enough to make any effort to dispel whatever doubts they may feel.

Third, while other accounts of bullshit focus on saying (or uttering) things without regard for their truth, I focus on the much broader phenomenon of communicating without regard for the truth of what one communicates. This is largely because political bullshit is multimodal. It can be spoken, written, signed, tweeted, and more. As the example involving James shows, the bullshit one communicates often has its origin elsewhere, and needn't involve the bullshitter saying anything. One can instead transmit a message of some kind first produced by another agent. Indeed, what is transmitted by a bullshitter may be something both true and first produced by someone who cared about the facts. To return to the above example, James is still bullshitting even if the incriminating stories about Jessie are true, having been first uncovered by a whistle-blower greatly concerned with their veracity.

Of course, bullshit so understood doesn't exhaust the range of phenomena that may aptly be described as ‘bullshit’.⁹ This paper offers no definition in terms of severally necessary and jointly sufficient conditions, and other accounts aiming for a more comprehensive picture of bullshitting in all its varieties may pick out other phenomena. But the frequency with which such bullshit appears in politics (as well as its possible harms) warrants special attention. In politics, with so much at stake so often, a disregard for the facts can have serious consequences. Why, then, do so many people in politics have such a casual relationship with the truth?

The explanation defended in this paper is that such behaviors are often an instrumentally rational response to the incentives political agents face.¹⁰ Instrumentally rational agents adopt means that, given their beliefs, are suitable to their ends (Kolodny and Brunero 2020). The more suitable the means, the more rational they are, even if the ends strike us as outlandish or outright immoral. For example, a fact-checker who cares about properly discharging her duties acts rationally when she engages in painstaking research on the relevant subject, but so too does the fact-checker who, because she is too deeply involved in party politics to be unbiased, selectively fact-checks in ways flattering to her preferred party. While one warrants praise and the other blame, both act rationally in the intended sense.

Going forward, I will write often of the costs and benefits of bullshit. This should always be read as elliptical for costs and benefits that are relative to an agent's preferences. Benefits are outcomes that are broadly consistent with one's preferences, while costs will hinder the satisfaction of one's preferences. Since agents frequently possess different preferences, what constitutes a cost for one agent might constitute a benefit for another. Continuing our previous example, the biased fact-checker faces costs when forced to fact-check impartially. Relatedly, she benefits to the extent that she is free to act in a biased manner. In contrast, the dutiful fact-checker faces no such costs.

Seen this way, the decision to bullshit is just one strategy among many that political agents can choose in order to minimize their costs (or maximize their benefits), with political agents often deliberately bullshitting as a means to their varied ends.

2. Rational Bullshit

In this section I explore the incentives driving three groups of political agents to bullshit: politicians, the media, and voters. Though the details differ in each case, a common pattern emerges – namely, agents typically have a plurality of ends with which truth-directed inquiry competes. Given these competing ends, the costs of caring about the truth frequently outweigh the benefits. Moreover, merely presenting oneself as caring about the truth of what one communicates is significantly less costly than sincerely caring. Accordingly, a certain degree of bullshit in politics is rational.

2.1. Bullshitting Politicians

Let's begin with politicians. The claim that politicians are occasional bullshitters is not a hard sell, and cases of bullshitting politicians are likely what springs to mind for most when thinking about political bullshit. Still, it is worth asking why politicians are often such flagrant purveyors of bullshit.

In general, politicians must decide between competing courses of action when deliberating about how to pursue their ends. If they are thinking strategically about how to act, they will assess the overall costs and benefits of these competing courses of action to determine which will, in expectation, provide the most benefits and the least costs. Some campaign strategies, for instance, may offer greater expected benefits than others. For another, entering some coalitions rather than others may yield greater benefits. Decisions regarding how to communicate are no different. Communicating in a clear, honest, and well-informed manner might bring benefits that are inferior to communicative strategies that do not require one to be clear, honest, or well-informed. Accordingly, a politician deciding whether or not to care about the truth of what they communicate must determine the overall costs and benefits from either course of action.

Consider, then, the costs and benefits of not bullshitting (that is, the costs and benefits of caring about whether one communicates the truth). The most significant and common costs are related to the acquisition of information itself. Simply put, actively attempting to get things right is in many cases an extraordinarily demanding endeavor, bringing with it steep procedural costs. In politics one often grapples with complex issues where the facts are difficult to ascertain, and where significant time and effort must be devoted in pursuit of the truth. In short, exerting such effort is costly, and these costs push politicians in the direction of caring less about the facts than they would otherwise. Moreover, such efforts have high opportunity costs, for many politicians will have aims not best served by undertaking them. Time spent devoted to truth-directed inquiry is time that could be spent on other, more valuable things (Hardin 2009: 2–3).

Naturally, communicating the truth often either yields no benefits or actively harms politicians. Telling the electorate something that they don't want to hear – for instance, that their preferred economic policies are in fact bad for the economy – is a risky choice for self-interested politicians seeking (re)election, even if what they say is true. Of course, politicians looking to avoid such electoral costs are also incentivized to engage in more straightforward lying; simply tell the electorate what they want to hear, even if you believe otherwise. But one can tell the electorate what it wants to hear without even knowing or caring about the truth. Such disregard for the truth is the hallmark of bullshit, not lying. In short, then, both caring about and communicating the truth can be costly.

The costs of bullshitting are notably different. At first glance, one might think that bullshit has very few procedural costs, taking no more effort to engage in than the effort required to communicate in general. But this is inaccurate. For example, some bullshit is produced with the provocation and excitement of voters in mind (Jacquemet 2020). Consider again the case involving Jessie and James. James may attempt to rile up his supporters as much as possible by packaging the relevant rumors in a way that angers them, motivates them to continue opposing Jessie, and the like. But for politicians like James to know what animates their base, they must first expend some effort. Nonetheless, the procedural costs of determining what animates one's base are in many cases lower than the costs associated with doggedly pursuing the many facts relevant to policymaking and the like. The latter requires one to seriously engage with multiple, complex policy-relevant fields (or to delegate such engagement to subordinates), while the former will sometimes require only a series of well-executed public opinion polls, or to possess some general knowledge of the sorts of key issues that are reliably emphasized by voters (such as the economy, healthcare, or education). Even strategic bullshit designed to gain favor with the electorate, then, frequently has lower procedural costs than not bullshitting.

Politicians may suffer reputational harms if they are known bullshitters. It's plausible to think that such behavior will frustrate some of the electorate, and this frustration will in many cases translate to electoral setbacks for the relevant politicians. With that said, voters will likely level charges of bullshitting in a partisan manner, consistent with the level of partisan bias and hostility we find in politics more generally.¹¹ Voters themselves will often bullshit about whether politicians are bullshitting, not caring about whether the politicians they support are actual bullshitters, or about whether the politicians they dislike are engaged in honest pursuit of the truth. This suggests that the reputational harms associated with being labelled a bullshitter arise to a large extent regardless of what one does. In other words, one's reputation may be harmed even if one is falsely believed to be a bullshitter, and such reputational harms stem in part from simple partisan bias.

Additionally, the extent to which one suffers reputational harms because of one's bullshitting may vary with one's ability to bullshit in a convincing manner. Some politicians may lack the skill to bullshit in ways that accurately mimic the behavior of agents sincerely interested in the facts, while others may be better able to bullshit in non-obvious ways. Reputational harms can thus be more readily reduced by members of the latter group, skilled as they are at presenting themselves as concerned with the truth, while members of the former group need to tread more carefully.

Moreover, since one's base will often not care if one is a bullshitter, politicians can safely benefit from a certain amount of bullshit. Free from a commitment to getting things right, one can speak for all sorts of reasons. Politicians can engage in grandstanding that makes them appear attractive to voters (Tosi and Warmke 2016, 2020).¹² They can save face when asked difficult questions, speaking as if they possess expertise on complex matters with which they lack acquaintance entirely, safe in the knowledge that their supporters will either not know or not care about the bullshit.¹³ They can evade questions they would prefer not to answer (Carson 2016: 56–7). They can disparage and denigrate their political opponents without the need to find something worth disparaging. One can bullshit for a host of reasons, and it will often be beneficial and easy to do so. When paired with the high costs and low benefits of not bullshitting, it all too often becomes rational for politicians to bullshit.

2.2. Bullshitting Media

Let's now consider the extent to which the media bullshits rationally, whether traditional mass media connected to large and powerful news networks or smaller online platforms. First, media platforms can sometimes function as conduits for state-manufactured bullshit. Simple cases where the media is directly controlled by the state collapse the distinction between bullshitting politicians and bullshitting media, with the rational bullshit of the state being the rational bullshit of the media. If the media is a mere extension of the state, we can assume that many of the underlying incentives driving media bullshit are identical to those outlined in the previous section. News reports may be designed to maximize the provocation of citizens rather than to accurately present the facts, headlines written to discredit political opposition, propaganda produced to control the citizenry, and the like (Stanley 2015: 125–77).

Different cases arise when we consider more indirect mechanisms by which the media comes to distribute the state's bullshit. In some cases, the media might be coerced to spread bullshit at the behest of the state because of its control over licencing and regulation of interest to the media (Chomsky and Herman 1988: 13). Capitulating to the state's demands, though often regrettable, is instrumentally rational for agents if it helps them to avoid sufficiently high costs (in this case, the potential loss of relevant licences). In this case, once more the bullshitters are agents of the state since they are the source of the content. Additionally, though, media outlets may also lack regard for the truth of what they spread, being primarily concerned with acceding to the state's demands. If so, the media simply bullshits about state-manufactured bullshit.

Other cases lack this coercive element, with the state and the media voluntarily maintaining close ties while being nominally independent. Paradigmatic instances of such voluntary ties involve members of the media and agents of the state forming mutually beneficial relationships. For example, members of the media may capture whatever regulatory bodies are tasked with regulating their conduct, shaping legislation to their ends.¹⁴ But the direction of influence may run in both directions as closer ties are established between the pair. If the price of favorable legislation is the dissemination of bullshit, many will be willing to pay if the benefits of favorable legislation outweigh any relevant costs.

Of course, the media does not distribute political bullshit solely at the behest of the state. There are possibly cases where media platforms share certain content at the urging of corporate sponsors, financial partners, and other private entities while not caring whether the stories are accurate. Such content may be tailored to depict the relevant private organizations in a flattering light, to spread misinformation that politically benefits these private organizations, or for a host of other reasons.¹⁵ Whatever the case, expected financial costs from a failure to acquiesce to the demands of financial partners and sponsors might incentivize self-interested members of the media to treat the facts as of secondary importance. Similarly, and much like earlier cases involving the state and the media, such bullshit may be entirely collaborative, involving parties in mutually beneficial partnerships.

Naturally, the media often bullshits about politics even when facing no pressure from external sources. Financial incentives loom large in explaining the propensity of various media platforms to bullshit in different ways.¹⁶ These financial incentives can exist for media organizations in the absence of any ties to other groups, or even in the presence of pressure from other groups to not produce such bullshit. Consider first some incentives related to the differential costs and benefits of consistently aiming for accuracy as against bullshitting. Accurate reporting is a labor-intensive, costly endeavor that will not always yield compensatory benefits since potential customers will not always pay to read the boring truth. Eye-grabbing clickbait, on the other hand, is comparatively easy to generate, even if it's wildly inaccurate. If producing clickbait without regard for its truth is what sells, and if media outlets are sometimes concerned with maximizing profits, media outlets will sometimes produce bullshit in the form of clickbait.¹⁷

Closely related concerns revolve around fake news. Some of what people call fake news is bullshit – stories produced without regard for their truth that also happen to be substantially false.¹⁸ Some of this fake news is produced by instrumentally rational agents seeking to maximize profits (Rini 2017: 45). Much like ‘bullshit’, though, ‘fake news’ is used in many ways by different speakers. Some fake news involves lying, some of it involves non-bullshitting, non-lying misleading, and so on.¹⁹ Still, whenever fake news constitutes bullshit, and whenever such fake news is fake political news, we have a potential source of rational political bullshit.

Lastly, we shouldn't forget that members of the media, like everybody else, have their own political biases. Though in more traditional settings there may be journalistic norms prescribing a certain degree of impartiality, for many the temptation to bullshit on matters they feel strongly about will be too much.²⁰ Needless to say, the bewildering variety of online blogs and websites from which many get their daily news may lack even those partly effective norms recommending impartiality that are (sometimes) internalized by members of mainstream media sources. As far as bullshitting about politics is concerned, the result is predictable. It can be psychologically taxing for partisan individuals to sincerely engage with ideologically incongruent viewpoints. Correspondingly, it is much easier and less costly to engage in biased reporting, expressing largely congruent viewpoints.

Some citizens will react negatively to such reporting, and this may harm the reputations of the relevant media outlets. But as Gordon Tullock once wrote, “inaccuracy in your information is only important if the person whom you are trying to persuade knows or is likely to find out that you are in error” (Tullock 1972: 128). So long as there are sufficiently many who will reward the media even when they bullshit (whether by tuning in, paying for subscriptions, or what have you), bullshitting will remain rational. The question, then, is whether media outlets can expect their audience to be habitually ignorant in the requisite manner. If so, the reputational harms of having their bullshit discovered will often be outweighed by the expected benefits of continuing to bullshit. As we shall see in the following section, there are indeed reasons to think that audiences will often remain ignorant of the media's bullshit.

2.3. Bullshitting Voters

It is a commonplace among political economists that citizens in democracies are often rationally ignorant (Downs 1957). Becoming well-informed is just too hard while offering few rewards, and so the costs of becoming politically well-informed typically far outweigh the benefits.²¹ Consequently, most citizens are ignorant of even basic political matters. The underlying rational choice analysis also generalizes to political irrationality – it is often instrumentally rational to be epistemically irrational about politics (Caplan 2007). Again, epistemic rationality is costly, requiring serious time and effort. When its costs outweigh its benefits, we find widespread irrationality in politics.

How does this bear on the prevalence of bullshit in political discourse? Many citizens are both rationally ignorant and rationally irrational, but they are also regularly pressed into social settings that incentivize making one's voice heard on political matters. In short, they face pressure to speak while not knowing or caring about the facts (Frankfurt 2005 [1986]: 63; Cappelen and Dever 2019: 55).

Why might citizens be incentivized to speak without knowing or caring about the facts? First, political discussions provide frequent opportunities to grandstand. Speaking in ways that boost one's level of esteem within one's social group incentivizes saying things that earn the approval of the relevant group, even if one doesn't know or care whether what one says is true (Tosi and Warmke 2016; Simler and Hanson 2018: Ch. 16). An agent embedded in an online community characterized in part by, say, adherence to the claim that the Earth is flat, is motivated to publicly disavow scientific orthodoxy on this matter, even if the agent is unfamiliar with any of the relevant scientific evidence. The public disavowal acts as a signal to their in-group about the propriety of their beliefs, boosts their credibility and status, and helps to avoid penalties associated with believing the “wrong” things (e.g., scientific orthodoxy regarding the ellipsoidal shape of the Earth).²²

Secondly, and relatedly, political discussions provide ample opportunity to dismiss, discredit, and disparage one's political opponents. One can earn the esteem of one's peers by casting aspersions on the right people (members of opposing parties, ideologically opposed peers, and the like).²³ In short, one earns esteem by attempting to lessen the esteem of those perceived to be in opposition. Independently of such grandstanding motives, there are plausibly benefits from the mere expression of disapproval of politically opposed individuals alone. Sometimes one just wants to express how one feels; it can feel good to express anger, venting one's negative attitudes towards others. These are occasionally substantial psychological benefits, and it is important to note that one can attain these benefits without a serious regard for the truth of what one communicates.

Thirdly, and setting aside such tribalistic motives, there are more general pressures to appear well-informed about important political matters (Petrocelli 2018). One can lose face by appearing uninformed, and so one may speak in ways that mimic being well-informed without incurring the costs required to become genuinely well-informed.²⁴ More generally, one might fear losing face even if admitting one's political ignorance would not in fact earn the disapproval of one's peers. Consider a case where an individual consistently engages in preference falsification to avoid social sanctions from their community – presenting themselves as knowing that prevailing ideologies are sound while in fact being wracked by doubt regarding their veracity – all while, unbeknownst to them, their doubts are shared by the wider community and would be praised rather than punished.²⁵ In such cases, one's false beliefs about the social costs of public political ignorance incentivize face-saving bullshit, since bullshitting allows one to avoid the expected social costs associated with having one's ignorance exposed.

We should take care to avoid depicting citizens in an overly unflattering manner. First, although it is hard to be precise on this matter, not all citizens are bullshitters. Some are intellectually and morally serious people who care deeply about the truth, at least some of the time. Second, much bullshit results from merely wanting to avoid the costs of acquiring political information, rather than the largely partisan and reputational motives discussed above (Pennycook and Rand 2018, 2021; McBrayer 2021: 40).²⁶ The former type of bullshit is arguably more excusable than partisan bullshit, being a natural consequence of the time-consuming and effortful nature of the acquisition and evaluation of political information. Still, when the costs of acquiring political information is paired with the sort of partisan social incentives discussed above, the result is a considerable amount of bullshitting on the part of citizens. We don't all bullshit all the time, but we do it enough that politics is seemingly filled with it. As we'll see below, this has important implications.

2.4. The Collective Harm of Bullshit in Politics

One might be tempted to dismiss bullshit as a mere annoyance rather than something seriously harmful. Given the language used to characterize the phenomenon, one might even view discussions of it (such as this one) as flippant or somehow tongue-in-cheek. But one shouldn't be misled by the language used. Bullshit in politics is in fact a tremendous source of harm. The harmfulness of bullshit becomes more obvious when one recalls that we are first and foremost concerned with political agents lacking regard for the truth. When one lacks regard for the truth of relatively unimportant matters in one's private life, the consequences are trivial. Indeed, even when faced with important decisions in one's private life, one frequently bears the costs of any poor decision-making alone. But in politics where we must interact with others and make decisions collectively, others bear the costs of our bullshit. Agents pursuing local gains by rationally bullshitting contribute to widespread collective harms. In effect, the problem of rational bullshit in politics constitutes a tragedy of the epistemic commons (Hardin 1969; Joshi 2021).

Consider first the bullshit of politicians. Speaking without regard for the facts is an easy way to spread inaccurate information, especially when done – as is frequently the case – in highly visible ways in public fora. Bullshitters will sometimes stumble across the truth, but they'll often get things wrong too. When we, as voters, believe what bullshitting politicians say, we often form false beliefs. Specifically, when we believe bullshitting politicians who get the facts wrong, their mistakes become ours. When falsehoods circulate too readily, the epistemic health of democracies is threatened.²⁷ We can't vote in optimal ways if consistently wrong about the facts, whether about the character of political candidates, the content of their policies, or about politically relevant facts more generally.

In general, bullshit in politics inhibits our ability to make reliably accurate decisions on politically important matters. As we have seen, when there exists an incentive to produce fake news or bullshit clickbait, organizations will gladly supply them. But when the media is saturated with bullshit and fake news, we again become increasingly more likely to form false beliefs on important issues. When we're continuously reasoning from false beliefs, we're going to make bad decisions. Moreover, as the volume of fake news and bullshit becomes noticed by consumers of online media, we become increasingly distrustful of sources of news in general (Jaster and Lanius 2021). Indeed, fostering mutual distrust among already polarized voters might be the primary aim of some creators and distributors of fake news; and in conditions of heightened distrust, we might not even trust generally accurate and reliable sources of information. We ignore those who mostly get things right, or who at least make a sincere effort to get things right, unfairly dismissing them as bullshitters. If we don't become outright skeptics, we at least become epistemic cynics, viewing everybody as out to sell us their self-serving bullshit.²⁸

Of course, bullshitting citizens play a central role in this process. While some people are driven to create fake news because of financial incentives, far more are driven to consume and spread fake news. When we encounter an inaccurate story online, the first impulse of many of us is to share it – provided, naturally, that it vindicates our preconceptions. By sharing it, we can signal to others about our political beliefs, express our affiliation, and so on. Sharing content with the click of a button is far less costly than taking the time to fact-check its content. Even when we don't share fake news, the article may be skimmed, with its content subsequently being taken on board. In this way, misinformation and bullshit spread throughout online communities. As before though, the more bullshit we share online, the more diminished the quality of information we have access to. We partly create the conditions for epistemically defective political environments, for it is easier to do so than it is to create epistemically healthy ones.

Although the focus of this section has been squarely on potential epistemic harms of bullshit, one should bear in mind that these epistemic harms often lead to serious non-epistemic harms. When voters are not acquainted with the facts, they may vote for candidates they wouldn't support if they were fully informed. Additionally, candidates and political representatives more generally are incentivized to respond to the preferences of an electorate constituted by frequent bullshitters. Pairing such an electorate with self-serving politicians (who, let's recall, are also themselves frequent bullshitters) is a recipe for improperly designed laws and policies that can adversely impact the lives of millions of people: economic policies that impoverish us, public health policies that make us sick or unsafe, geopolitical policies that put our lives in danger, and more.

3. What Not to Do About Bullshit in Politics

All things considered, bullshitters and their bullshit cause much harm in politics. While some agents secure localized benefits for themselves through their bullshit, it is hard to see how we could collectively, in aggregate, gain more than we lose through such behavior. Naturally enough, then, some may wish to take action to safeguard against this bullshit. If bullshit is so harmful, isn't it worthwhile to try to reduce its prevalence? And if so, how might we go about this?

First, one might think that if the severity of the problems created by political bullshit were more adequately communicated to the public, one could rely on people to make more of an effort to do what they can to prevent its spread. In short, people could be relied upon to exercise personal responsibility, thus tackling the problem from the bottom up. For instance, Christopher Blake-Turner argues that we should stress the importance of holding others accountable whenever they share or create fake news (Blake-Turner 2020: 13). Perhaps by reproaching each other, we can effectively disincentivize the creation and transmission of fake news and other forms of bullshit.²⁹

Second, we could train people to be better able to spot political bullshit. More broadly, we can strive to inculcate in as many people as possible various epistemic virtues that render it more likely they make a sincere attempt to ascertain the facts, more likely that they can successfully navigate epistemic environments saturated with misinformation, and so on (Pritchard 2021). Such efforts would involve vigorous attempts to improve overall media literacy among the electorate so that citizens possess the tools they need to identify and manage misinformation created, disseminated, and shared by bullshitters (Holcombe 2017; Orlando 2017; Rini 2019; McBrayer 2021: 175–6).

Third, we could rely on fact-checkers to flag content that contains falsehoods, misleading claims, and other epistemically unhelpful statements (Amazeen 2013, 2015; Rini 2017). Perhaps if such content is consistently flagged, people would be less likely to uncritically accept it and, importantly, less likely to share it. Fact-checkers, of course, are not in the business of identifying bullshit per se; their quarry is a much broader category. Still, bullshit in politics is harmful to the extent that it enables the spread of misinformation, and if fact-checkers can reliably identify such misinformation (some of which has its roots in bullshitters), then they can help with this problem nonetheless.

Fourth, and closely related to the previous proposal, we could reform social media in such a way that the spread of harmful misinformation becomes less likely (Woolley and Howard 2019: 243–4). For example, social media platforms could highlight content that has been flagged by independent fact-checkers, providing to its users a warning that they are about to read and/or share potential misinformation (Rini 2017: 56–7). Alternatively, social media platforms could conduct their own fact-checking operations. Social media platforms could also assign reputation scores to users who repeatedly share flagged content (Rini 2017: 57–8). Such reputation scores would in principle act as a signal to other users that the relevant agents are epistemically unreliable, being prone to sharing fake news and political bullshit more generally. By undermining the credibility of habitual bullshitters, we make it more likely that they're ignored by their peers.

Fifth, and lastly, we could go as far as to empower lawmakers to enact legislation that attempts to curb the spread of misinformation generally (Sim 2019; Brown 2021; Fritts and Cabrera 2022a). Legislation could be targeted at the sources of bullshit themselves, in much the same way that we already prevent certain industries from advertising their products in deliberately misleading ways (O'Connor and Weatherall 2019). Alternatively, we could proceed indirectly, incentivizing social media platforms to take greater care to curate their content and keep it as free as possible from bullshit (Rini 2019). Instead of approaching the problem of political bullshit from the bottom up, we work from the top down, entrusting our political representatives and lawmakers more broadly with the task of preserving and improving the epistemic health of our democracies.

Although an overview of this length naturally overlooks various details regarding these five broad strategies, it suffices to highlight one major complication that harms the prospects of each – namely, each ignores to varying degrees the extent to which the production and transmission of political bullshit is incentivized. Unfortunately, though, we overlook the relevant incentives at our peril. While some of the proposals would be relatively harmless (even if ineffective), others create additional incentives to bullshit in new, potentially more harmful ways. Let's consider how each fails to account for the underlying incentives.

First, reliance upon personal responsibility overlooks the fact that, without the appropriate incentives, it's hard to see what could motivate people to hold each other accountable in the right way. One could grant that if we each held each other accountable fairly and diligently, then the creation and transmission of political bullshit might be curbed. If we each became far more epistemically vigilant, and if we each became more receptive to reprobation from those who criticize our epistemic conduct, then personal responsibility would constitute a plausible, bottom-up approach to the problem. But if political agents are instead incentivized to be epistemically negligent and even outright biased, such an approach will be unsuccessful. Of course, we have already seen that agents in politics can often secure rewards by bullshitting. These same rewards are present in contexts where we could hold others accountable. Accurately holding others accountable will be an occasionally difficult, time-consuming effort.³⁰ Additionally, an honest assessment of our peers might force us to hold our ideologically likeminded allies to account for their epistemic vices. Biased political agents looking to avoid such costs, whether voters or members of the media or politicians, will rationally bullshit about the bullshit of others.

Second, improved media literacy and training may be ineffective in the absence of appropriate incentives to use such training in a consistent, unbiased manner. On the one hand, such training may occur in relatively cloistered environments free of the sort of pressures and incentives that political agents face in quotidian political settings. For instance, even if people are educated about the identity of consistently egregious producers of fake news and characteristic hallmarks of bullshitting politicians, they may fail to apply what they have learned in cases when they are faced with political bullshit that meshes well with their political convictions, or when it is shared by a trusted friend, and so on (Martens 2010). On the other hand, those who provide the relevant training may do so in a biased fashion, influenced by powerful incentives to bullshit about the degree to which various political factions are prone to bullshitting, prone to producing and sharing fake news, and more. Were we to implement media literacy programs on a large scale, those who organize and implement them will suddenly find themselves in a position where they can potentially influence the character of the political informational landscape in a serious way. While there would no doubt be some who discharge their duties in an impartial manner, some will be unable to resist the temptation to abuse their position.

Third, the use of fact-checkers may be ineffective for several reasons. Most obviously, we need unbiased fact-checkers who are willing and able to set aside their partisan commitments and discharge their duties in a consistent and fair manner. But fact-checkers will often find it difficult to refrain from disproportionately scrutinizing political figures they dislike while subjecting claims made by favored figures to less intense scrutiny. Fact-checkers may not even be aware that they are guilty of focusing differentially on different figures, instead being driven by largely unconscious biases. Still, whether their biases are conscious or not, it can be difficult for people to question others with whom they largely agree. Correspondingly, it is much easier and more rewarding to critique one's political opponents. These twin pressures, as we have seen, incentivize a biased mode of political cognition that is characteristic of the habitual bullshitter. Additionally, the effectiveness of fact-checkers relies upon a citizenry willing to accept the verdicts of such fact-checkers. Even if fact-checkers are generally reliable and unbiased, citizens may disregard their findings when they are incongruent with their political beliefs (Walter et al. 2019; Fritts and Cabrera 2022b). Naturally, this sort of knee-jerk dismissal would be even more predictable were instances of genuine bias among fact-checkers to be documented, with bullshitting citizens erroneously tarring all fact-checkers with the same brush.

Similar concerns plague the suggestion that social media reform could meaningfully halt the spread of political bullshit. First, there is no guarantee that social media platforms who conduct their own fact-checking will do so without bullshitting about which political figures or media outlets are themselves guilty of bullshitting. Second, if they forego conducting their own fact-checking and choose to rely on independent fact-checkers, this simply raises the sort of problems that could arise with independent fact-checkers. Third, even if social media platforms hire agents willing and able to conduct these tasks without bullshitting, users of social media may bullshit about the relevant mechanisms, whether fact-checking verdicts, reputation scores, or both. One can even imagine social media users wearing their negative reputation scores as badges of honor, earned by bravely standing up to the biased verdicts of overly powerful and influential social media organizations.

Lastly, we should proceed cautiously before ceding power to political representatives and legislators to decide who is bullshitting, who is guilty of manufacturing and spreading fake news, and the like.³¹ Such proposals expose a curious short-sightedness and naivete among many – specifically, a naivete regarding the degree to which we can trust politicians with obvious incentives to abuse their position for political gain to fairly and accurately hold creators and sharers of political bullshit accountable. As we saw in section 2.1, politicians are incentivized in various ways to bullshit. With the power to punish those they deem guilty of bullshitting, we are in effect giving an even greater amount of institutional power to people who are likely to bullshit about when political opponents are bullshitting, bullshit about when stories critical of them are fake news, and so on. Such an outcome may be even worse than the status quo where, at the very least, few people are being punished by agents backed by the power of the state for their alleged bullshit.

In short, by failing to address the underlying incentives, these proposals are likely to be ineffective. Even worse, some of them will afford new, dangerous ways for bullshitters to bullshit. Nonetheless, one might maintain that all that matters from a practical vantage point is whether any given intervention would produce, on balance, a positive impact. And whether an intervention would do so or not is clearly an empirical question. But as we have seen, there is little reason to think that these interventions will be successful. More worryingly, we have some strong reasons to think that not only will these interventions fail, at least some of these interventions will create even worse problems than those they are supposed to fix. Absent strong countervailing evidence, then, we should err on the side of caution before implementing some of these more radical proposals, and proponents of such interventions shoulder an evidentiary burden which they have yet to discharge.

4. The Inevitability of Bullshit in Politics

That so many people could care so little about the facts is in part a consequence of powerful incentives that reward them for bullshitting. By failing to consider these incentives, existing proposals for reducing the harm of political bullshit are unlikely to succeed. Worse, they may simply make things worse by providing new avenues for bullshitters to bullshit. However, one might think that recognizing the fact that perverse incentives breed bullshitters can help us do better. Perhaps if we can figure out how to create different incentives more conducive to an epistemically well-functioning polity, we can reduce the prevalence of political bullshit, or at least mitigate its harms. This, no doubt, will be a difficult task. But it won't be an impossible one.

How might this work? Plausibly, successful anti-bullshit interventions will focus on making it costly for agents to disregard the facts, reward those who make a serious effort to determine the facts, or both. In other words, successful interventions will create incentives directly conflicting with existing incentives that, inter alia, make it costly to acquire political information and rewarding to bullshit. For instance, an ambitious strategy would be to strive to inculcate in children from a young age an urge to pursue the facts even when doing so is difficult or time-consuming. Perhaps this could be achieved either by portraying truth-directed inquiry as intrinsically rewarding or by stressing the instrumental benefits of such inquiry (or both). In either case, children raised to find truth-directed inquiry rewarding will, one hopes, be less likely to consider serious attempts to figure out the truth as costs to be avoided. Put another way, children raised in such a manner are less likely to be epistemically insouciant (Cassam 2018). In effect, by socializing children to internalize the benefits of pursuing the truth, we could in principle raise a generation of people who, by disposition, are strongly averse to bullshit in politics.³²

To take another less ambitious proposal, we could ensure that fact-checking organizations are visibly non-partisan, ideally being endorsed, supported, or otherwise associated with salient figures from a range of political parties, interest groups, and more. Partisan agents may be more willing to relinquish their beliefs if figures they trust are the ones either suggesting such beliefs be relinquished or publicly backing those who do make such suggestions (Margolin et al. 2017). When fact-checking processes are associated with trustworthy and ideologically likeminded people, changing one's mind is less likely to be seen as a cost to avoid in order to save face.

Relatedly, if important, charismatic public figures are seen to reliably pursue the facts, change their minds in light of the facts even if it is inconvenient for them, or otherwise strive to be epistemically virtuous, a healthy regard for the facts may come to be seen as something to emulate (Mohseni and Williams 2019).³³ If careful, diligent deliberation were seen less as a cost and more as something that popular, well-respected people regularly engage in, people could begin to secure social rewards for behaving similarly. We could exploit our tendency to conform to widespread practices and norms by making an effort to popularize figures who are known to reliably resist the temptation to bullshit. In the same way that in current political settings one can secure social rewards by toeing the party line, one could in principle secure similar rewards by conforming to increasingly widespread and popular practices of virtuous epistemic conduct.

However, it is much easier to outline such proposals than it is to successfully implement them, let alone implement them in ways that would achieve our ambitious aims, for the very same incentives that hamper existing attempts to reduce the amount of bullshit in politics reappear. Will those who inculcate in our children the relevant attitudes do so fairly and without bullshitting? Will the internalized rewards for properly regarding the facts outweigh the potential rewards of bullshit? How can we incentivize the formation of truly non-partisan fact-checking organizations? What incentives are there for popular public figures to avoid bullshit, and what incentives are there for the public to follow their lead? Do these incentives outweigh conflicting incentives to bullshit? If not, even those interventions that attempt to create the right sort of incentives will likely fail. In a sense, we would need to already possess sufficiently strong beneficial incentives for the appropriate interventions to succeed.

One might object that there must be some interventions that would have a beneficial impact. Even if existing interventions fail, why assume that none can succeed? Additionally, proponents of such interventions are likely aware that eliminating bullshit from politics is not feasible. Still, merely reducing its prevalence or mitigating its impact is much more realistic. If so, we shouldn't conclude that no intervention could succeed where others have failed.

The claim that there are possible interventions that could fare better than existing proposals should be granted. With that said, we need to be realistic about what anti-bullshit interventions can achieve. The incentives that drive agents to bullshit inevitably arise in the sort of conditions that characterize large political communities. In politics, people will always have a plurality of ends, only some of which are best served by serious truth-directed inquiry. For agents like us with finite cognitive and attentional resources, the acquisition and evaluation of political information will always have its costs.³⁴ Meanwhile, presenting oneself as knowledgeable will often yield benefits. Right away, such facts already incentivize a certain amount of bullshit. Politics always involves agents with competing aims and ambitions, and factions inexorably arise, whether we like it or not. When factions arise and the agents constituting them develop and intensify their partisan commitments, the various social incentives discussed earlier emerge, and such incentives drive the production of much bullshit in politics. Whenever there is a profit to be made from pandering to one's audience, and wherever costs can be cut by shirking one's duties to report the facts accurately, there will be bullshit in the media.

It's plausible that there is no escape from such conditions in politics. Consequently, while we may yet be able to reduce the amount of bullshit in politics through clever design or happy accident, it seems that we'll never be rid of it entirely. In politics, bullshit will always pay.³⁵