Democratic Consent Under False Pretences

I. CONSENT TO WHAT? HOLDING ELECTORAL OFFICE

Political philosophers are most familiar with consent in relation to political obligation and the duty to obey the law. My approach in this article is completely different. I will focus instead upon how electoral consent authorizes someone to occupy electoral office in a democracy. As I will show, the most telling objection to consent as a ground for political obligation does not apply to consent as a ground for occupying electoral office.

Here is what I take to be the most telling objection to grounding political obligation in consent conferred at the ballot box.⁶ Political obligation is supposed to be universal within each state’s domain. Everyone within a state’s jurisdiction⁷ is supposed to have at least a prima facie duty to obey the laws of that state.⁸ But grounding the duty to obey the law in electoral consent would not deliver such a universal result. It may yield a duty to obey the law for those who voted (or perhaps only just those who voted for the winning candidate). But those who did not vote (and perhaps those who voted for a losing candidate) cannot have any duty to obey the laws on those grounds. Neither could those who are legally barred from voting. Resident aliens would, on this logic, have no obligation to obey the laws of the state in which they are resident. Neither, under many states’ electoral laws, would convicted felons. That is standardly, and rightly, regarded as a reductio ad absurdum of grounding political obligation on electoral consent. We clearly need to find some other basis for any political obligation to obey the law.

Even if electoral consent cannot ground a duty to obey the law, however, it can ground a permission for certain individuals (duly elected ones) to occupy electoral office. Democratic elections are devices by which voters confer consent to a particular party or person holding electoral office for a specific period of time.⁹ A person or party thereby acquires a special permission to do something that would otherwise have been impermissible for them to do (namely, to occupy electoral office without having been elected to it).¹⁰ Equivalently,¹¹ electoral consent authorizes those who are duly elected to occupy the office to which they have been elected for the period for which they have been elected.¹² Philosophers, political theorists,¹³ political scientists and indeed political practitioners all share this understanding of elections.¹⁴

States, like collective agents in general, are constituted by their internal decision procedures.¹⁵ Those procedures may pertain to many different things. Preeminent among them, however, is how to determine who is authorized to act on behalf of the collective. In a democracy, that authorization comes through the consent of voters. The fact that it relies on people’s consent may or may not be the best way to justify democracy itself.¹⁶ There are myriad other ways of justifying democracy if that will not do. My point here is a narrower one. It is a simple analytic truth that electoral consent plays this central role in the operation of a representative democracy.¹⁷

The fact that a state is a representative democracy merely entails that in that state representatives hold electoral office with the consent of voters conferred at elections.¹⁸ Nothing follows, analytically, about what proportion of voters must consent in order for the authorization to occur. Electoral law – the internal decision procedure that constitutes the state as a representative democracy – varies. The important point to notice, for present purposes, is that the consent of literally every voter is never required (and there are good normative grounds for thinking it should not be¹⁹) to trigger the internal decision procedure that authorizes representatives to occupy electoral office.²⁰

What I will, for short, call ‘collective consent’ emerges by aggregating, according to the collective’s internal decision procedure, votes expressing valid individual consent to their chosen candidate occupying office.²¹ Thereby the collective grants permission to the winning candidate under the internal decision procedure to occupy office—and it does so without all voters individually consenting to that person’s occupying office.

When discussing electoral consent as grounds for political obligation, non-voters formed the basis for a reductio. But non-voters thus pose no problems for using electoral consent as the grounds for a representative occupying electoral office. True, non-voters have not given their own consent to that person’s occupying office; neither have those who voted against that person, nor those not entitled to vote at all. But universality is not required in this matter in the same way in which it is with regard to political obligation. All that is required is that enough (or a large enough proportion) of qualified voters have given their valid consent at the ballot box to that person’s occupying the office to trigger the internal decision procedure by which the representative democracy in question authorizes representatives to hold electoral office.

In order to trigger that internal decision procedure of a representative democracy, however, the consent of individual voters must indeed be morally valid. As I will now go on to show, certain background conditions must be fulfilled for consent to be morally valid at elections just as in any other situation in which consent is conferred. If an individual’s electoral consent is morally invalid, it cannot count toward authorizing a person or party to hold electoral office. If the consent of enough voters is morally invalidated in this way to erase a candidate’s margin of victory, then that candidate will not have valid permission to occupy office under the internal decision procedure of that representative democracy, despite seeming to have won the election.

II. PREMISES

Like any argument, mine starts from certain assumptions, ones about the function of voting and the nature of electoral consent. The discussion above explains the two first premises upon which my argument rests:

Premise 1: Electoral consent is morally transformative.

The consent of voters at the ballot box makes it permissible, under the internal decision procedure of a representative democracy, for the winning candidate to occupy the office to which she has been elected.

Premise 2: Collective electoral consent derives from individual.

Elections are mechanisms by which, under their internal decision procedures, democratic collectives constitute themselves and authorize political leaders to occupy elected office. Since collective consent supervenes on the consent of individual voters, the moral validity of collective consent hinges on the validity of the requisite number or proportion of voters’ individual consent.

To these two premises, I will now add two more.

Premise 3: Voters are fact-sensitive.

My argument about how political misinformation undermines electoral consent assumes that enough voters are often enough motivated in ways that are fact-sensitive to make false factual claims potentially pivotal to the election’s outcome.

Electoral campaigns are conducted in ways that presuppose such voters constitute a significant portion of the electorate. Through electoral manifestoes and campaign promises, parties and candidates regularly describe what policies they would pursue in office and what benefits they would yield. The decision of voters of whether to consent or not to a candidate occupying electoral office—and therefore collective consent as well—often hinges on such beliefs.

There is abundant evidence that much of the electorate is indeed fact-sensitive in one way or another. That is to say, their beliefs about facts determine how they vote and, through that, the outcome of the election. One way is as a rational response to the issues of the election.²² But perhaps this may not be all that common.²³ Maybe people more typically vote instead on the basis of emotions, identities, or group loyalties based on partisanship, race, religion, ethnicity, or geography.²⁴ Even such voters, however, are still acting on the basis of factual claims that trigger their emotions, identities, or loyalties. Voters who care about group affiliations will be nonetheless sensitive to facts relevant to the question of whether ‘this candidate really is one of us’ (or instead a ‘RINO – Republican in Name Only’ – as moderate US Republican politicians are alleged to be these days). Voters who are emotionally driven will be nonetheless sensitive to alleged facts that serve as emotional triggers (anti-Semites to claims about the authenticity of the ‘Protocols of the Elders of Zion’; white supremacists to claims about ‘the Great Replacement’; Brexit supporters to claims that immigrants are massing at the borders set to swamp them). Importantly, even those who are driven by emotion or loyalty when voting would be presuming the truth of those factual claims that trigger their emotion, identity, or loyalty. And those claims may, of course, be either true or false. When parties and candidates win elections by appealing to such voters with factual falsehoods along those lines, the validity of individual voters’ consent—and potentially collective consent—is compromised just as much as by false factual appeals to rational issue-oriented voters.

Premise 4: Conditions of valid consent are the same across spheres.

Finally, I assume that the same conditions are required to make consent morally transformative in the political sphere as in the other spheres of life. There is no reason the moral validity of electoral consent—as expressed through citizens’ votes—should hinge on criteria different from those applying to consent in private matters.²⁵

For any consent to be morally valid, three things need to be clear: that one consented, to what exactly one consented, and that why they consented did not include any invalidating conditions. The consent-giver must be aware that what she is doing counts as consent. She must be aware of what she is consenting to.²⁶ And, she must not consent as a result of duress, impaired capacities, or fraudulently-induced defective beliefs.

My particular focus is on the last of these conditions. We have no reason to respect ‘consent’ obtained under false pretences.²⁷ A person’s consent is morally invalid if the consent-seeker’s deception created impactful²⁸ false beliefs in the prospective consent-giver about either what one is consenting to or one’s reasons for consenting. Both invalidate consent.²⁹ In both cases, the deceitful consent-seeker interferes with the consent-giver’s autonomous will. Insofar as the consent-giver would not have consented had she known the truth, the deceit constituted a difference-maker and the individual’s fraudulently-obtained consent is morally invalid.³⁰ While it is standardly assumed that intention is required for deceit to vitiate consent (of any kind), below I will go one step further to argue that unintentional misinformation can also invalidate electoral consent in certain circumstances.

Conclusion: Deception invalidates electoral consent

From the above premises, it follows that impactful deception invalidates electoral consent. Elections make permissible actions—occupying electoral office—that would be impermissible otherwise (Premise 1). Collective electoral consent derives from enough voters individually validly consenting to a certain candidate occupying electoral office under the internal decision procedure (Premise 2). But the same moral conditions that make consent morally transformative in private life must be fulfilled when citizens collectively consent to a party or candidate occupying office (Premise 4). Thus, an individual’s ‘consent’ procured through impactful deception at the ballot box is no consent at all. It should not count toward triggering the collective’s consenting to a candidate occupying electoral office under the collective’s internal decision procedure.³¹

That is to say, political misinformation vitiates democratic collective consent if the misinformation was a difference-maker for enough voters.³² It need not have been a difference-maker for every member of the majority, merely for enough of them to have made a difference to the electoral outcome: a number equal to or greater than the winning margin.

Insofar as the reason we care about voters’ consent is that we want to respect voters’ autonomous will, we have no reason to respect electoral ‘consent’ that was fraudulently obtained. Under the influence of a fraudulently-induced false belief, a voter’s ‘consent’ is no longer the expression of her autonomous will. Under such an influence, their individual consent is null and void. And if enough voters’ individual consent is invalidated in that way to deprive the winning candidate of her winning majority, collective consent to her occupying office is also rendered null and void.

III. IMPACTFUL DECEPTION

Insofar as collective consent at an election is what confers permission upon a party or candidate to occupy office, then serious consequences should in principle follow if that consent is morally invalid because fraudulently obtained. But first we would need to establish that the falsehoods in question are substantially more likely than not to have been difference-makers in the election. I will now discuss this ‘impact’ condition in more detail.

There are several ways in which political falsehoods can be impactful. One is through their propositional content. That occurs where false propositions (i) concern issues that were difference-makers in the election and (ii) the false propositions made a difference to voters’ beliefs about those issues. Alternatively, the falsehoods might have been impactful precisely in turning issues that would not otherwise have been difference-makers into difference-makers for enough voters to form a majority. Electoral consent would be morally invalid where political agents’ misinformation was ‘impactful’ in either of these ways.

How might we know whether either was the case? Sometimes we may have strong a posteriori evidence of impact. Survey data and exit polls can show what topics mattered most for voters. If we know that influential falsehoods about those topics circulated heavily, we can draw inferences as to whether they likely influenced the majority’s vote. Polls and surveys may indicate citizens’ reasons for voting one way or another (for example, immigration policy, the economy, the health system, and so on), and what proportion of voters each reason mobilised. If highly salient misinformation clearly surrounded the most prominent reasons behind citizens’ votes, then that misinformation was highly likely to have induced citizens’ electoral consent. Take for example, the false claim—immediately repudiated by the UK Statistics Authority—that the UK was paying £350 million a week to the EU.³³ That appeared prominently on the side of the ‘battle bus’ on which leaders of the Brexit campaign (including Boris Johnson, subsequently Prime Minister) travelled around the country. Post-referendum polling suggested that the potential savings from exiting the EU were the third most important issue for voters in the Brexit referendum.³⁴

In the absence of such a posteriori evidence, we can sometimes rely on a priori reasoning to surmise whether any given falsehood likely influenced citizens’ consent. Employing the ‘reasonable man test’, courts standardly use a priori evidence to establish whether one party’s consent was sufficiently likely to have been induced by another’s fault.³⁵ So too politically, we can use an analogous test to determine whether the misinformation sufficed to induce the consent of a reasonable person. Then we may judge it likely to have influenced the collective’s electoral consent as well.

Deception about the character of political agents can also be an impactful difference-maker to an election.³⁶ The sheer fact of political agents lying to voters might itself have mattered enough to have been a difference-makers for enough of them. Even if politicians’ lies (that is, their propositional content) concerned peripheral matters that were not themselves difference-makers in the election, the fact that those agents lied (even about peripheral matters) may in itself have constituted a potential difference-maker in the election. Had citizens been aware that those politicians were liars, they might not have voted for them even if they did not much care about the things politicians were lying about. Concealment of pertinent facts about character is a form of deception that could invalidate consent too, although it may be harder to prove due to the absence of a posteriori evidence.

IV. INTENDED OR UNINTENDED?

Political deception that is impactful in any of those ways can render electoral consent morally void. But does it matter whether or not the deception was intentional? While the standard answer is ‘yes’, next I argue in favour of a different answer in the case of electoral consent, mainly, that is can become morally invalid not just when it was obtained through impactful, intentional deception but also when it was obtained through impactful unintentional, yet negligent, deception.³⁷

Consider first impactful political falsehoods that were intentional. By ‘intentional’ I mean that the person uttering them knew them to be false and uttered them with the intention of deceiving the hearers. That intentional deception invalidates consent is widely agreed, so I shall not belabour the point.³⁸

Yet, sometimes politicians and parties may mislead voters in electorally consequential ways without deliberately intending any deception. They may have been honestly mistaken. Would citizens’ political consent be morally valid then?

According to classical debates about consent, it may seem so at first brush since those consent-seekers did not intend to deceive. But that would be jumping to conclusions especially in the case of electoral consent. First, in comparison to other more trivial instances in personal life where consent obtained through someone’s honest mistake may not matter so much, electoral consent has serious consequences, legitimizing political agents to occupy office, which can gravely affect people’s everyday life. The stakes being much higher in the case of electoral consent, we may argue, even honest mistakes and not just intentional deception may be enough to cast doubt on the moral validity of this consent. Second, consent-seekers who intentionally deceive are in the wrong, but so may be consent-seekers who are genuinely mistaken, insofar as they were under a duty not to be mistaken about certain matters.³⁹

Thus, while culpability may not extend to all cases where the consent-seeker unintentionally misleads the consent-giver, it may still apply to those cases where the consent-seeker has a duty to know the facts being stipulated when consent is being sought. These are cases where the consent-seeker could and should know the facts in question. If she does not, then we are not simply talking of ‘innocent’ mistakes, but rather of culpable ignorance on the consent-seeker’s part.⁴⁰ In these cases, we can talk of unintentional deception.

Take for example, the case of a doctor who genuinely, but wrongly, believes that the medical procedure he is recommending to his patient had been proven safe and effective. Surely, we may doubt that the patient’s consent to the procedure was in fact morally valid and entitled the doctor to administer it. Ex hypothesi the doctor could and should have known the readily available medical evidence concerning that procedure. If he did not, he was culpably ignorant; he failed and was blameworthy for failing to discharge his duty toward his patient. His negligence (even without any malevolent intention) put the patient in a situation where the latter’s consent was based on mistaken beliefs. We can thus say that the patient was deceived (albeit unintentionally) by the doctor who negligently betrayed the patient’s justified trust in him.

Doctors are under an obligation (professional, legal, and moral) both to know and to impart a full and accurate account of all available information to patients about the treatments that they prescribe. The doctor’s duty is not just not to misinform, but positively to inform patients about all matters relevant to their consent. Salespersons, on the other hand, are morally permitted to convey information selectively in trying to sell their wares. They have no duty to provide a comprehensive and impartial overview of all competing products. Nevertheless, they do have a duty not to provide false information, either intentionally or negligently, about the items that they sell.

Political parties and candidates are more like salespersons than doctors in that respect. Certainly, they are not subject to the equivalent of medical malpractice suits for failing to provide full information to those whose consent they seek. In a competitive democracy, politicians, like salespersons, may convey information selectively, focusing on certain matters or policies. But like salespersons, politicians should nonetheless be under a moral duty not to misinform, knowingly or negligently making false statements when trying to ‘sell’ themselves and their policies to voters.⁴¹

One reason that politicians are under a duty to know and not to misrepresent relevant facts is that government, as much as (or even more so than) a medical procedure, is an ultra-hazardous activity.⁴² A government’s decisions can impose severe risks upon its population. Those, whether physicians or politicians, who seek others’ permission to engage in such ultra-hazardous activities have a moral obligation to go to great lengths to be trustworthy communicators, insofar as they can, when seeking people’s consent.⁴³

Some might say that politicians campaigning for office do not yet have any such duties because they are not yet in office. Once in public office, they would be in a position of trust that does entail a strong duty of that sort. But it might be argued, this is the case only for officeholders; it does not apply to those aspiring to office.⁴⁴ True, the scope of the (moral) duty of due diligence of the officeholder is of course much broader than that of the office-seeker. Still, it would be wrong to think that office-seekers have no such duty whatsoever. After all, those seeking such positions of trust are openly and publicly requesting voters’ consent to occupy office; and they are asking voters to consent and to entrust them with office on the basis of information they provide during the campaign. ‘Office-seeking’ is a public role just as much as ‘office-holding’. The duty of due diligence attached to it may be much more limited, but it is hardly non-existent. It requires office-seekers not to make false statements, either intentionally or negligently, when seeking voters’ electoral consent.

In virtue of candidates’ duty to be trustworthy communicators on those matters, it is reasonable for voters to expect that of candidates. Some might query whether it is reasonable to expect politicians to be trustworthy simply because they have a duty to be so. After all, just because morality calls on people to do something does not mean that they will do it. Shouldn’t the reasonableness of expectations, they may ask, be based instead on the likelihood that people will actually do what they should do?

This however equivocates between descriptive and prescriptive expectations, expectations about what ‘will be’ and expectations about what ‘should be’. Descriptively, it might be unreasonable to expect politicians not to be untruthful when asking for our consent to occupy elected office; it might be unreasonable to lay a bet on that. But that does not mean that it would be prescriptively unreasonable to expect them to be so. Our expectations of someone are reasonable in the moral sense when they are based on what his moral duties require him to do (however likely or unlikely he is to do as morally demanded).

Most importantly in the present context, that moralized account of ‘reasonable expectation’ is the one that is appropriate to an analysis of the moral validity of electoral consent. Frequencies have no moral significance and should not influence what we judge to be morally valid or invalid consent. The fact that a certain fraud is widespread and thus expected does not mean that a victim of such fraud ‘offers morally valid consent to a con’.⁴⁵ Similarly, just because culpable ignorance and political deceit are widespread among the political class does not make citizens’ ‘consent’ morally valid when that ‘consent’ was induced through culpable negligent misinformation. The frequency of others’ bad behaviour (lying or negligently misinforming) does not make that behaviour morally permissible or consent based on it morally valid.

To recap: I argued, first, that even if there was no intention to deceive, voters may rightfully consider themselves ‘deceived’ and not bound by their ‘consent’ when politicians’ unawareness that their facts were wrong was itself a result of their culpable ignorance. I argued, second, that politicians’ ignorance is culpable where they could and should have known the facts about matters meant to influence voters’ choices at the election. I argued, third, that voters can reasonably expect candidates not to utter falsehoods when seeking their electoral consent in the election in virtue of the latter’s duty not to do so.

V. PRACTICAL CONSEQUENCES

Sometimes we can be confident (on the basis of a priori or/and a posteriori evidence) that political misinformation was substantially more likely than not to have been a difference-maker in an election. Then collective authorization of an elected official to occupy office would be morally void. Morally, that official would have no permission to occupy that elected office. But what should politically and legally follow from that moral fact?

Sometimes maybe nothing should follow. There are some obvious benefits to a stable democracy where election results are not easily contested or overturned. Political and legal stability, with their economic corollaries, argue against removing people from office willy-nilly. Still, there should clearly be an institutional mechanism by which the collective can, as appropriate, eject people from office when collective authorization to their occupying elected office has been deemed morally invalid.

One mechanism for ejecting incumbents from office is a recall election.⁴⁶ That is one way of removing a representative from office before their term ends. Currently, recall elections are available only in a limited number of jurisdictions.⁴⁷ There has been no provision for recall of national officials in the US since 1789.⁴⁸ Neither is there any provision for recall of national officials in Canada or Germany or even Switzerland, although in all those countries (as well as in the US) there are recall provisions in some of their subnational units.⁴⁹

While details of electoral law vary across polities, two facts pertaining to the present argument bear emphasizing. First, many polities currently lack any provision for recall elections. And second, even where they are available, recall elections are not currently permitted in specific response to impactful misinformation having invalidated consent at the previous election.⁵⁰

My proposal would be for recall elections to be available in all jurisdictions as a mechanism for removing elected representatives from office in specific response to political misinformation having undermined collective consent at the previous election. I propose that those be initiated by a petition signed by a requisite (but not unduly excessive) number of registered voters. I further propose that the recall petition should explicitly refer to the reason for recalling the targeted official: the fact that they spread misinformation that was impactful at the previous election in a way that morally invalidated consent at that election. I propose that the person targeted by the recall be permitted to contest the ground for a recall in court, where the judge(s) would be permitted to issue an opinion (which would be purely advisory for voters) on whether the targeted person does indeed have a case to answer at the recall election.⁵¹

Thus, it would be up to the voters to petition for a recall election. They might not exercise this option every time deception has rendered electoral consent morally invalid and the collective authorization to hold office morally void. Collective consent to hold office may already be morally void owing to deception. But from a political and legal perspective that consent is only voidable until and unless a successful recall election has been held to remove the person from office.⁵²

There is a particular advantage to making recall elections the means by which incumbents elected under false pretences are expelled from office. This additional step would further confirm that misinformation was most likely impactful in the previous election. If the recall succeeds, that shows that a substantial portion of the collective had previously mistakenly and unintentionally signaled consent due to misinformation. Allowing courts to advise voters on the merits of the case for recall helps prevent the process from being groundlessly manipulated purely for political advantage. Holding recall elections too often would obviously undermine governability. But we could fine-tune that trade-off between governability and morality by adjusting the number of signatures required for a successful recall petition.⁵³ When putting electoral consent on firmer moral grounds by finding ways to correct for the effects of impactful deception, the challenge is to maintain governability by balancing moral and practical concerns. In this article I have suggested one such way.

ACKNOWLEDGEMENTS

I am grateful to the journal’s editor, Jeff Howard, and its referees for their feedback. A version of this paper was presented at the 2020 American Political Science Association Annual Conference, the 2021 Midwest Political Science Association Annual Conference, and a 2020 International Ethics Research Group Work-in-progress Seminar. For questions/comments, thanks to Claire Benn, Ned Dobos, John Dryzek, Bob Goodin, Luke Glanville, Toni Erskine, Paul Formosa, Mianna Lotz, Cian O’Driscoll, Jonathan Pickering, Wendy Rogers, Peter Stone, and Albert Weale. This research was funded through a Discovery Early Career Researcher Award from the Australian Research Council (DE210101068). The views expressed are my own and do not represent the opinion of the institutions with which I am affiliated.

COMPETING INTERESTS

The author declares that she has no competing interests.