An Adversarial Approach to Campaign Ethics: Promises and Challenges

Some of the dilemmas facing candidates in competitive elections, we submit, are structurally similar to those faced by ethical businesspeople. As campaign ethicists have pointed out, deceptive and dishonest tactics such as lying, pandering, misdirection, waffling, and incivility can have serious negative consequences. Yet such behaviors may also be “profitable” in electoral terms: indeed, certain campaign environments will be ripe for arms races in which all candidates use increasingly “dirty” tactics to keep up with their rivals, while those who eschew these tactics are more likely to be left behind.^{Footnote 8} That is what previous approaches to campaign ethics have neglected to account for: not simply that destructive behaviors are sometimes rational but also that there may be circumstances where refusing to partake prevents ethical actors from remaining in the competition at all. Under such circumstances, only those who make unethical choices will even be presented with such choices again—and in this way, making ethical choices becomes actively self-undermining.^{Footnote 9}

We argue that adopting an adversarial approach can help us understand the responsibilities of candidates in such situations. More specifically, the responsibility of candidates to avoid profitable but destructive behaviors should be theorized on a continuum along with feasible ways of altering the circumstances that make those behaviors “profitable” in the first place. An ethical candidate should not deceive her potential constituents without reason, of course, but she should also not allow her opponents to use such tactics to their advantage. Instead, her first response should be directed at undermining their effectiveness. Perhaps she will only need to highlight their unethical nature, enforcing norms against them via informal means; or perhaps she will need to strike an agreement with her opponents, with the aim of preventing an arms race.^{Footnote 10} Just like the ethical businessperson, the ethical candidate should not only abstain from destructive tactics but should also seek to reduce the electoral “profitability” of those tactics.

At this point along the continuum, however, two differences emerge between elections and other adversarial institutions. First, in the context of sports, law, and business, there is an (ideally) unbiased third party to adjudicate disputes about infractions—a role with no direct analog in democratic politics. As such, where the ethical businessperson can lobby for centralized enforcement as a last resort—and can negotiate for self-regulation in the shadow of this possibility—the ethical candidate cannot. Centralized enforcement of campaign norms is not strictly impossible, but giving regulators the power to silence or disqualify candidates on the basis of campaign discourse could seriously endanger the open, pluralistic society that elections are supposed to protect. It is a cure for the excesses of competition that is worse than the disease. As a result, the last resort of athletes, lawyers, and businesspeople is unavailable to candidates in democratic elections: arms races must be averted without the help of referees, judges, and regulators.

The second difference between campaigns and other adversarial institutions is that a candidate’s “incentives” for winning may be moral, as well as material. If the attempts of an athlete or businessperson to implement informal norms or formal regulations against a destructive practice are unsuccessful, after all, we would still think it wrong to participate in that practice. As long as her reasons for doing so are primarily material (or self-interested), rather than moral (or other-regarding)—which is usually true of athletes and businesspeople—her moral imperative to quit or close up shop is fairly clear. In electoral campaigns, however, a candidate’s reasons for employing a particular tactic may themselves be moral. If the election has significant normative consequences, the question of whether or not to engage in the destructive behavior is a question internal to ethical theory. As a result, we cannot responsibly tell candidates to “quit” the competition, which we would usually advise athletes and businesspeople to do in similar circumstances.

Given these two differences, the key question that emerges regarding “dirty” campaign tactics is whether they are likely to be effectively constrained through informal, decentralized sanctions. Because formal rules are unavailable as a means of regulation, yet we cannot tell ethical competitors to quit, we must rely on such sanctions to prevent arms races. If they are at least plausibly effective, then the ethical responsibility of candidates is to ensure they are enforced; that is, make the dirty tactics unprofitable. If an arms race will ensue regardless, however, then candidates with important substantive goals may be excused for using some dirty tactics. If such tactics are effective, indeed, ethical candidates may have a responsibility to use them.

The Challenges of Informal, Decentralized Enforcement

In practice, there are two kinds of obstacles to effective informal sanctions against many destructive campaign behaviors. Some are inherent to the ambiguity of the norm in question, and others result from the costs of sanctions. Sometimes norms are difficult to enforce because it is genuinely hard to tell what counts as an infraction, in other words, and sometimes the challenge is that those meant to enforce the norm have significant incentives not to do so. In many cases, of course, it is the combination of these two challenges that renders norms particularly difficult to enforce via informal, decentralized means.^{Footnote 11}

Consider the difference between handshake conventions and norms prohibiting romantic advances in the workplace. The latter are ambiguous, first of all, in a way that the former are not. Barring exceptional circumstances, no one will disagree about whether Al has offered Betty his left or right hand, but there may be great disagreement about whether a particular action of Al’s counts as sexual harassment or harmless flirtation—even among people who agree about the wrongness of sexual harassment in general. This ambiguity does not preclude the possibility of decentralized enforcement altogether, but it clearly makes coordination more difficult. Second, the sanctions that accompany handshake conventions are not typically very costly. When they are—as is often the case with sexual harassment—even people who desire to enforce a norm in general may be reluctant to do so in particular cases. A senator, for instance, might sincerely believe that sexual harassment is wrong and may even enact sanctions against it in other circumstances, yet still look the other way when her partisan ally is accused of wrongdoing.^{Footnote 12}

Such straightforward hypocrisy is surely an important obstacle to decentralized enforcement of campaign norms as well. Imagine, for example, that a candidate clearly violates a norm. The only certain way to sanction him for doing so is to withdraw support or take action that will cause others to withdraw their support. Those who already support other candidates cannot withdraw their support from the violator, of course, and they already have ample incentive to induce others to do so. Among those who share the violator’s substantive goals, meanwhile, implementing sanctions will come at significant substantive cost. In many cases, this kind of sacrifice will not seem worthwhile.

Drawing on the example of Venezuela’s descent into authoritarianism, Milan Svolik (Reference Svolik2018) shows that this is a key reason why polarization threatens the maintenance of political norms. Svolik’s survey experiment with Venezuelan voters supports the predictions of his formal model (and of common sense): if there are more people in the ideological “center,” then there are more people—elites and ordinary citizens alike—who will be willing to sanction violators by switching their support. In a polarized society, however, the threat of this kind of punishment is less credible, because switching one’s support will typically require a greater sacrifice in substantive terms.

Of course, this account considers only the most straightforward mechanism through which polarization might make norm enforcement more difficult. Svolik assumes that everyone agrees a violation has occurred and that everyone considers the violation to be destructive. Polarization comes in only after these facts have been established, giving partisans incentives to consciously ignore the violation. In real political life, however, polarization is also likely to cause problems for norm enforcement well before this juncture, by intensifying entirely genuine disagreements about what counts as a norm violation in the first place (Ahlquist et al. Reference Ahlquist, Ichino, Wittenberg and Ziblatt2018).

According to a wealth of research about the phenomenon known as “motivated reasoning” (Kunda Reference Kunda1990) or “identity-protective cognition” (Kahan et al. Reference Kahan, Braman, Gastil, Slovic and Mertz2007), for instance, human beings systematically underconsume and undervalue information that contradicts our prior commitments and group identities (Bartels Reference Bartels2002; Lodge and Taber Reference Lodge and Taber2013). We do not seek out such information and, even if we are exposed to it, we are likely to ignore it, discredit it, or underestimate its importance—and then forget about it later. On the one hand, then, the information partisans receive will be quite different; and on the other hand, the way they interpret that information will also vary widely. Our prior commitments and group identities thus profoundly shape the way that we genuinely perceive the world—and in a polarized society, partisans of opposing camps are often likely to disagree in entirely sincere terms about whether norms have been violated (Bagg Reference Bagg2018a).

This outcome is all the more likely, of course, as the ambiguity of the norms in question increases, thereby making motivated reasoning more difficult to detect and counteract. Ambiguity also gives plausible deniability to those who are consciously underenforcing, like the senator who ignores her colleague’s misdeeds. Given the difficulty of norm enforcement even when the violation is clear, this problem is only exacerbated the more room there is for disagreement about what counts as a violation. As a result, norms governing campaign discourse will be extremely fragile. Commitment to them must outweigh powerful substantive concerns and identity loyalties, which distort our ability to enforce norms in both conscious and unconscious ways.

We do not mean to suggest that it is impossible to sustain any norms of political discourse in the context of campaigns. Norms against incitement to vigilantism and political violence, for instance, are effectively enforced in most stable democracies. Black activists outraged by racialized police violence and mass incarceration almost universally disavow the logic and rhetoric of extrajudicial revenge, as do the vast majority of antiabortion activists convinced that federal law authorizes a genocide of the unborn. Calls for illegal violence are rare, and when they do occur, those who make them are marginalized even by their allies. The same can usually be said for norms against threats to undermine the integrity of elections or the rule of law, as well as certain forms of discourse that explicitly target racial and ethnic minorities. In other words, some campaign norms are enforceable, even in difficult circumstances. Nevertheless, we should not underestimate the difficulty of enforcing them. Sustainable enforcement requires that people who disagree on nearly everything else must not only agree about what counts as a violation but also coordinate on effective sanctions. When enforcement works—which it does, for certain violations, in all functioning democracies—we should celebrate it as the fragile achievement it is.

Against an Abstinence-Only Ethics of Campaigns and Elections

We can now state more clearly and comprehensively our reasons for transcending the focus on honesty favored by extant accounts of campaign ethics. In brief, many of the deceptive practices they prohibit are so ambiguous that cross-partisan enforcement will likely prove impossible.^{Footnote 13} It will be too difficult for people across the political spectrum to agree on what counts as a violation, much less coordinate on costly sanctions. Wherever the incentives favoring certain deceptive practices are substantial, therefore, we can expect some noncompliance with norms against them, potentially yielding an arms race (or “race to the bottom”). Given a commitment to freedom of expression—and a healthy skepticism, in particular, of empowering regulators to silence political leaders—ethical candidates cannot turn to centralized enforcement of norms. Yet given the high normative stakes of many elections, they also cannot morally justify abandoning the competition altogether. If decentralized sanctions fail to undermine the profitability of certain forms of dishonesty, therefore, we conclude that ethical candidates ought to be excused for committing some forms of campaign deception.

To illustrate these claims more systematically, let us begin with the five deceptive practices prohibited by Eric Beerbohm—lying, bullshitting, pandering, mudslinging, and strategic ambiguity—which can fairly stand in for the range of honesty-focused guidelines offered to candidates by other campaign ethicists as well. We do not deny the prima facie ethical wrongs Beerbohm identifies with each of these practices. However, we submit that only a prohibition on lying is plausibly enforceable through decentralized sanctions.^{Footnote 14} If blatant and consequential enough, lies can be identified and condemned across ideological lines, just like calls to violence and certain racially discriminatory remarks. When these norms are well enforced, an ethical candidate faces no dilemma at all: she has ample reason not to lie. Even in cases where these norms have been weakened, however, and lying might help her achieve her other normative goals, the ethical candidate should still refrain from doing so. Since norms against lying are sometimes enforceable, her first responsibility is to revive them—and lying clearly undermines that goal.

In the case of the other four behaviors Beerbohm classifies as morally off-limits, however, coordination on sanctions will be much more challenging, and this changes candidates’ ethical calculus. Consider that it is difficult enough for two people with very similar substantive views to agree about whether a candidate has engaged in these practices. Generating enough agreement about such questions to induce coordinated sanctions across the political spectrum is very likely impossible. And although it is difficult to imagine circumstances where bullshitting would be essential to victory, it seems that candidates with valuable aims will sometimes be able to improve their chances of victory quite significantly by employing the other three strategies: strategic ambiguity, mudslinging, and pandering. As a result, the choice about whether to employ them looks quite different, from an ethical perspective, from the choice about whether to lie. Outright lying to voters is rarely if ever excusable, but ambiguity, mudslinging, and pandering might be.

We may consider the cases of ambiguity and pandering together. As a wealth of evidence demonstrates, ordinary voters are astonishingly poorly informed about both substantive issues and the track record of individual candidates (Achen and Bartels Reference Achen and Bartels2016; Somin Reference Somin2013). Their policy “preferences” exhibit little coherence across issues and only weak consistency over time, changing dramatically depending on the framing of the question. The problem of incoherence, meanwhile, only gets worse when considering dilemmas of social choice and the aggregation of preferences (Ingham Reference Ingham2018; Riker Reference Riker1982). As individuals and as a society, it seems, we want contradictory things—better services and lower taxes, more security and stronger civil liberties—and who can blame us? Yet politicians will ultimately be called on to make choices between these priorities.

On the one hand, candidates can be fully honest about the trade-offs and uncertainties they face in making policy, presenting the same platform in the same precise way for every audience. On the other hand, they can remain strategically ambiguous about the finer points of policy while expressing hopeful optimism about their ability to solve problems, emphasizing different aspects of these hopes and dreams for different audiences. Given the typical behavior of voters, candidates who adopt the first strategy risk losing many supporters to candidates who adopt the second (promising better services and lower taxes, for instance). Where candidates must appeal to multiple audiences in multiple stages of an election, moreover—as in the American primary system—the problem is even more pronounced: the claims that resonate with one audience will differ substantially, by design, from those that appeal to the next. Without appealing to ambiguity or pandering, a perfectly honest candidate in such a system would seem to be at a huge disadvantage.

Consider the common connotations of the words “politician” and “political” as meaning “slippery” or “ambiguous,” such that one who refuses to take a side is acting like a “politician” and dodging a question makes one’s answer “political.” This speaks not only to the frustrating quality of such tactics, of course, but also to their enduring effectiveness in democratic politics. Given what we know about voter behavior, it is unlikely that many voters will know or care about subtle differences between politicians’ utterances across various contexts, and this enables politicians to get away with significant ambiguity.^{Footnote 15} Given the divergent incentives they face in front of a wide range of different audiences, meanwhile, it is no surprise that many successful politicians maintain significant ambiguity or flexibility regarding their commitments.

Similar concerns can be articulated regarding prohibitions on mudslinging. As with the most blatant forms of pandering, there is some debate about whether and when negative campaigning is effective (Lau, Sigelman, and Rovner Reference Lau, Lee and Rovner2007). Where it seems clear that it will not be effective, however, there is no ethical dilemma to speak of. Meanwhile, its sheer prevalence again belies the supposition that it always fails to achieve candidates’ goals. Whenever it appears to be effective, then, the candidate faced with a mudslinging opponent must choose between remaining on the defensive—and thereby accepting the “frame” proposed by her opponent (Lakoff Reference Lakoff2004)—or going on the offensive herself, positing an alternative frame for voters to grab hold of. Again here, the correct choice in strategic terms is clear: by refusing to respond in kind to effective mudslinging by her opponents, she risks being uniquely tarred with damaging stigmas. As with ambiguity, refusing to engage in any mudslinging whatsoever could plausibly endanger a candidate’s chances of winning—and when it does, retaliation may be the most responsible course of action.

If it were possible to enforce norms against strategic ambiguity, pandering, and mudslinging—as it is with lying and calls to violence—this consequence might be averted. Unfortunately, this is not the case. Whether any given act counts as one of these deceptive practices, or as an acceptable form of rhetoric and opposition research, is often very ambiguous. Even among substantive allies, it will be very difficult to agree on whether violations have occurred. Coordination across party lines of mutual enmity and distrust will likely be impossible. We are left with a classic arms race, where those who eschew such strategies are placed at a significant disadvantage.

In other adversarial contexts, there is always one fail-safe option for maintaining one’s ethical scruples: abandoning the competition altogether. What is at stake in at least some democratic elections, however, is not a gold medal or a bump in share price but matters of great normative importance. For this reason, we cannot simply say that candidates with valuable substantive goals should categorically refrain from ambiguity, pandering, or mudslinging, nor should they refrain from competing altogether. Both responses would significantly increase the likelihood that political power would be exercised by the least scrupulous actors. When there are morally important consequences for victory, therefore, any blanket prohibition on all forms of dishonesty will sometimes make for irresponsible recommendations. Instead, we claim that at least some forms of campaign dishonesty, under at least some circumstances, must be considered morally permissible (if not, in some cases, required).

Which forms of dishonesty are these, and under what circumstances will they be rendered permissible? Most obviously, unsanctionable deception or mudslinging from opponents and their allies will typically make it permissible for candidates to respond in kind. However, there may be other excusing or justifying circumstances as well. In fact, we expect that most candidates most of the time will not be fully forthright—that is simply the way campaigns work—and it strikes us that even in the absence of specific, demonstrable instances of dishonesty by opponents, the demands of thriving within such an environment could plausibly serve as a general excuse for mild deceptive practices like pandering and ambiguity. Either way, the prevalence of such relatively trivial deception by candidates does not particularly concern us. On our “realist” account of democratic elections, after all, such commonplace deceptions represent minor distortions of a principal-agent relationship between citizens and officeholders that was largely fictional to begin with. As a result, they do not represent particularly serious wrongs, and they may be relatively easily excused. We think campaign ethicists—and candidates themselves—should focus their normative attention elsewhere. Rather than further analyzing the permissibility of various deceptions in relation to various hypothetical excusing circumstances, therefore, we turn now to articulate our alternative approach to the ethics of campaigns and elections, which emphasizes pluralism rather than honesty.

From Honesty to Pluralism: Standing Our Ground Where It Matters

Let us be clear that we do not suggest abandoning honesty altogether. Deceptive practices such as ambiguity, mudslinging, and pandering do degrade the relationship between candidates and citizens to some degree, and candidates should avoid them whenever possible. Nevertheless, these mild forms of dishonesty are not nearly as threatening to democracies as other campaign ethicists have assumed, and in the darker moments of competitive campaigns, we think there are far more important places for candidates to stand their moral ground. In particular, we think that the most pressing moral dilemma facing ethical candidates—the one that might actually keep them up at night—is how to weigh the demands of maintaining democratic institutions against the prospect of achieving normatively valuable substantive policies. As a result, we think that a theory of campaign ethics should address this dilemma head-on. In doing so, moreover, it should focus on preserving those things that are truly necessary to democracy.^{Footnote 16}

On theories assuming that the purpose of elections is to enable ordinary citizens to “authorize” the laws through representatives conceived as agents of the people, of course, dishonesty by candidates does strike at the foundations of democracy’s value (Beerbohm Reference Beerbohm2012). However, the principal-agent model of representation implied by this ideal—whereby representatives respond to exogenous citizen preferences—has long been understood by political scientists as highly unrealistic, if not incoherent (Disch Reference Disch2011; Mansbridge Reference Mansbridge2003). Countless surveys and experiments demonstrate that most citizens have no opinion at all on most policy issues and that the opinions they have are not well informed (Achen and Bartels Reference Achen and Bartels2016; Lodge and Taber Reference Lodge and Taber2013). Meanwhile, social choice theory has exposed profound difficulties with the very idea of a popular will, which would persist even if everyone did have well-formed opinions (Arrow Reference Arrow1951; Riker Reference Riker1982; Weale Reference Weale2018).

In light of these findings, we should cease trying to preserve an idealized agency relationship between citizens and their representatives that was deeply implausible to begin with. It is simply not the case that elections would enable the people to authorize the laws, were it not for the strategic ambiguity of candidates. Yet elections seem valuable nonetheless, and in our view, the reason for this has more to do with competition between parties than with the fantasy that representatives could ever be said to make laws “in our name.”^{Footnote 17} We should still demand honesty from candidates, then, but it need not be our first priority. Other excesses of adversarial campaign competition are far more dangerous to democratic institutions.

It is beyond the scope of this article to develop a comprehensive theory of campaign ethics, much less a complete picture of healthy democratic institutions. Building on recent scholarship across several distinct traditions, however, we observe that many of the most dangerous tendencies of electoral competition seem to share an “anti-pluralist” character. Acceptable forms of partisanship and competitive behavior cross a line and become dangerous to democracy, authors such as Nancy Rosenblum and Jan-Werner Müller argue, when they lose their fallibilistic character; that is, when they refuse to accept the legitimacy of disagreement. We suggest, therefore, that a focus on protecting social and political pluralism represents a more fruitful direction for campaign ethics: pointing out lines that might be tempting to cross in the name of valuable substantive goals, but that ethical candidates should nevertheless refrain from crossing.

In her study of the ethics of partisanship, for instance, Rosenblum acknowledges that all partisans think they are right. Those she calls “extremists,” however, display “a deliberate and celebrated… single-mindedness,” which ignores the fact of pluralism and excludes certain groups from consideration altogether (Rosenblum Reference Rosenblum2010, 401). In his analysis of populism, similarly, Jan-Werner Müller admits that all democratic politicians make claims on behalf of “the people.” Yet those he calls “populists” (in a necessarily pejorative sense) make an “exclusive” and “moralistic” claim to representation—that “they, and only they, represent the people” while opponents are necessarily “part of the immoral, corrupt elite” (Müller Reference Müller2016, 20). Although often claiming to represent the people, by contrast, nonpopulists concede “that representation is temporary and fallible, that contrary opinions are legitimate, that society cannot be represented without remainder, and that it is impossible for one party or politician permanently to represent an authentic people apart from democratic procedures and forms” (40). Other recent accounts have stressed a similar exclusionary, moralistic logic as central to populism, linking it both theoretically and empirically to the erosion of democratic norms and institutions (Galston Reference Galston2018; Levitsky and Ziblatt Reference Levitsky and Ziblatt2018; McKean Reference McKean2016).

Given that the normative valence of populism is seriously contested, we prefer to describe our target as “extremism” or “anti-pluralism,” as in Rosenblum’s account, rather than “populism,” as Müller and others would have it.^{Footnote 18} Regardless of the terminology employed, all of these authors agree that the phenomenon in question is both endemic to electoral competition—in Müller’s words, it is a “permanent shadow of representative politics” (2016, 101)—and seriously threatening to democracy. The logic of exclusive or direct representation suggests an all-too-recognizable set of polarizing strategies, which have proven frustratingly effective throughout the history of representative democracy and have been deployed by opportunists from a range of ideological backgrounds. The notion that one’s opponents and their supporters are illegitimate and morally corrupt—perhaps not even part of the “real” people—has too often served as a potent justification for dismantling independent checks on centralized power (Levitsky and Ziblatt Reference Levitsky and Ziblatt2018, 72–96; Müller Reference Müller2016, 44–49).

We propose that a more fruitful approach to campaign ethics would be oriented toward averting these dangers. Given its effectiveness in certain contexts, even candidates with valuable substantive goals may be tempted by hyperpolarizing, anti-pluralist rhetoric. Despite these temptations, a pluralist approach asks candidates to conscientiously address their arguments to the entire political community, thereby making themselves at least minimally accountable to all groups within that community. Some degree of polarization can be advantageous or even necessary in the fight for substantively valuable ends, but it can also make norm enforcement more difficult and, in some cases, slide into outright anti-pluralism. As a result, we maintain that ethical candidates should seek to minimize this polarization as much as they can and make conscious efforts to “depolarize” where possible.^{Footnote 19} Meanwhile, genuinely anti-pluralist strategies must remain off-limits, even when the substantive stakes are very high.

For instance, candidates must vigorously attack their rivals—not only criticizing their policy positions but also highlighting their character flaws, unearthing past indiscretions, and exposing corrupt backroom bargaining. In doing so, however, they should accept the legitimacy of opposing politicians and refrain from demonizing them, casting them as morally corrupt outsiders in contrast to a morally pure in-group. Moreover, this license to attack one’s political rivals does not extend to the ordinary people whose only vice is supporting those politicians. In particular, ethical candidates should not refer to their opponent or her supporters as “enemies” or label either as existential threats to the well-being of the political community—even when one suspects they might be.^{Footnote 20} Such claims are particularly problematic (and deserving of sanction) when the targeted person or group can be identified by ascriptive characteristics, such as race, ethnicity, gender, and sexual orientation. Furthermore, we argue that it is not enough to refrain from attacking an opponent’s supporters: even just ignoring a group can imply that its members do not deserve consideration as part of the polity. Instead, ethical candidates should work to include partisan opponents in their audience—answering their concerns and asking for their support—even if they recognize that such appeals are unlikely to be effective.^{Footnote 21} Finally, ethical candidates should not pursue electoral strategies based on demobilizing opposition voters or discouraging participation in the democratic process more generally.

These demands are fairly extensive, of course, and we might wonder whether they overstep the bounds of campaign ethics, which ought to govern how people pursue their goals rather than which goals they pursue. Following our guidelines would seem to prevent ethnonationalist candidates, for instance, from presenting their ideas at all: even if they were perfectly willing to abide by the results of elections and adhere to other norms of democratic life, our account would rule large swaths of their substantive platform out of bounds as a violation of campaign ethics. Even if we find their views repugnant in a broader sense, therefore, the objections we make may appear not to fit within the scope of campaign ethics.^{Footnote 22}

This objection, however, depends on a strict separation between values of social pluralism—concerning an inclusive society—and political pluralism—concerning open democratic procedures—which cannot be sustained. On the one hand, we think that the former is just as central to the value of democracy as the latter: the benefits of competition between parties are seriously undermined if some members of society are excluded from the audience, such that their interests may be entirely ignored.^{Footnote 23} On the other hand, the work of scholars such as Rosenblum, Müller, and Levitsky and Ziblatt has revealed a clear connection in practice between the two forms of anti-pluralism, which we would be unwise to ignore. Candidates who reject “social” pluralism often reject “political” pluralism as well, if and when they gain enough power to ban competing candidates or parties. This is no coincidence: as Müller argues, it is written into the “inner logic” of anti-pluralist forms of populism (2016, 10–11). Although we cannot predict the outcome with certainty, it seems reasonable to suppose that candidates who continually demonize a particular religious or ethnic group as enemies would curb its formal political power if they could.^{Footnote 24}

Three Advantages of a Pluralist Approach

This preliminary sketch of a pluralistic approach to campaign ethics is only suggestive, of course, but we now have a clear enough picture to appreciate its key advantages over alternatives focused on honesty. First, pluralist norms are easier to enforce through decentralized sanctions. Second, even if these sanctions fail, candidates who follow pluralist norms are still less likely to seriously harm their electoral chances, and for both of these reasons, following them will less often generate perverse results. Third and finally, upholding pluralist norms is far more important for democratic politics. Thus, even if candidates lose votes by refusing to engage in anti-pluralist tactics, their sacrifice serves a worthy cause, and not an ideal that was hopeless to begin with.

Let us begin with the question of perverse results. As we have seen, existing approaches to campaign ethics seek to prohibit a wide range of deceptive practices, yet only the most unambiguous violations—blatant lies and perhaps the most egregious forms of pandering—are effectively preventable through cross-partisan sanctions. Given the very significant advantages to be gained through subtler forms of dishonesty and ambiguity, we have argued that refraining from such behaviors in the absence of broader enforcement mechanisms will often do serious damage to the electoral chances of ethical candidates.

The norms of pluralism we propose do not have the same effect. On the one hand, our account licenses certain forms of dishonesty, which gives ethical candidates a fighting chance in even the must cutthroat campaigns. On the other hand, upholding pluralism is not always or necessarily costly in the first place. Whereas ambiguity has almost no strategic downsides, anti-pluralist tactics can carry their own risks. We do not deny that those tactics are sometimes effective, even for candidates with valuable substantive goals; otherwise, there would be no genuine dilemma. However, their efficacy is far from guaranteed, and especially in the face of an anti-pluralist opponent, affirming pluralist values is not an obviously inferior response. (Compare this to the hapless candidate discussed earlier, who frets openly about the trade-offs between tax cuts and service provision while her carefree opponent promises the world). Addressing one’s arguments to the entire political community rather than firing up one’s base by demonizing the enemy may come with costs, but it may also have benefits, and in the face of this strategic uncertainty, it is ethically responsible to err on the side of pluralism.

In addition, of course, many pluralist norms are more amenable to cross-partisan sanctions than norms of honesty. Consider, for instance, that during the 2012 US presidential race, Republican nominee Mitt Romney was criticized widely, even within his own party, for calling 47% of American citizens “takers.” The same thing happened four years later on the Democratic side, when Hillary Clinton sustained broad criticism—again, crucially, from many within her own party—after labeling many of her opponent’s supporters a “basket of deplorables.” Both Romney and Clinton backed away from their comments and did not make similar statements in the remainder of their campaigns. At least in these cases, decentralized, cross-partisan enforcement of pluralist norms seems to have worked—just as coalitions reliably mobilize within functioning democracies to marginalize actors who incite political violence.

To be sure, pluralist norms are hardly uncontestable—and as such, they will not always be enforceable by decentralized means. Indeed, many observers believe that we have recently witnessed a widespread breakdown of these norms in a number of countries around the world. Yet the very thesis of breakdown implies a prior state in which the norms were more reliably enforced. A lapse in enforcement is not evidence of structural unenforceability, but of a need to revive the conditions supporting more reliable enforcement—and that is precisely what our guidelines recommend. When sanctions fail—as they often do—this is not because enforcement is impossible, as is the case for mild forms of deception. Even in the name of valuable substantive goals, therefore, ethical candidates should not simply abandon these norms at the first sign of erosion. Given the threat that anti-pluralist tactics pose to democracy—and thus to a future in which substantively valuable goals may be safely pursued—candidates and their supporters must work to defend pluralist norms by abandoning any candidates who are responsible for that erosion, regardless of their substantive priorities. As demonstrated by the earlier examples, this can be an effective response in its own right. Even when such sanctions appear to be ineffective, however, this forbearance will not represent a meaningless sacrifice on the road to an inevitable arms race. Because the norms in question are at least potentially enforceable, an arms race is preventable—at least in the long run—and attempting to reinstate the norms is essential to its prevention.^{Footnote 25}

Even if there is deep and seemingly unending contestation about the meaning and scope of pluralism, this is not necessarily a fatal blow to our account. After all, it was motivated in the first instance by a desire to address a set of concerns that were more significant for the health of democracy, and in that context, continuing contestation can be a feature, not a bug. Put otherwise, our account directs attention to curbing the behaviors that are most harmful to democratic forms of government, and if what results from this increased attention is increased discussion and disagreement about what those behaviors are, then our goal will have been partially accomplished. Of the various advantages we have claimed for a pluralist theory, after all, this one is ultimately the most significant: it aims to preserve the real values of democracy.