Apologies for the delay in posting the fourth part of my five-part series on character-based voting. Here are parts one, two, and three, which are probably necessary as background to part four. Earlier in the series, I make reference to what I call a “Murad-type meeting,” referring to Donald Trump’s behavior at a recent meeting with Nobel Peace Prize winner Nadia Murad.
The first part introduced the topic of character’s ambiguous relation to “policy.” The second part focuses on character’s instrumental relation to policy. The third part considers the possibility that expressions of character might be constitutive of “governance.” This part considers the possibility that expressions of character might have normative significance out of relation to policy or governance, at least on conventional construals of those terms.
(3) Third grade: bad character past a threshold disqualifies the candidate for office as such
Consider the possibility that Murad-type meetings are neither relevant to policy nor constitutive of governance in the sense described earlier in this series, but are still relevant to a candidate’s fitness for office, hence a permissible basis for voting. The relevant idea here is primarily that of disqualification for elected office by sufficiently bad character: that is, certain behaviors (and by implication, the states of character that express them) are so morally egregious that having passed a certain threshold, they disqualify someone from holding office simply by virtue of crossing the threshold. These threshold-crossings may, and in most cases will, have adverse consequences on policy and governance, but they need not do so to qualify as a justifiable rationale for character-based voting.
Put another way, the adverse policy- and governance-consequences of egregiously bad character need not be epistemically accessible to a voter in cases like this, even when the consequences obtain (or would obtain, or are likely to obtain, or are warrantedly thought to be likely to obtain). The badness of very bad character might, at least in certain contexts, be sufficient to rule the worst candidate out, and by default, justify a vote for his less-bad rival.
Brennan doesn’t explicitly deal with a disqualification-oriented rationale for character-based voting, deriding what strikes me as a less plausible and more simpleminded idea:
The office of the presidency is not an honorific meant to show that we respect that person’s character. Giving someone the presidency is not bestowing a medal or a certificate of commendation but giving him (some) control of the state, an institution that makes rules, and forces innocent people to comply with these rules using violence and threats of violence. We need to be sure he will do a good job controlling it. (Ethics of Voting, p. 84).
This set of claims seems part strawman and part ignoratio elenchi: a reader could grant the crux of Brennan’s claims but still doubt that those claims rebut character-based voting conceived as disqualification.
Let’s grant, with Brennan, that elected office is not an honorific meant merely to show respect for someone’s character, especially if “character” is understood in generic, apolitical terms. Let’s also grant that voting for someone is not like bestowing a medal or certificate of commendation for character thus conceived. And let’s accept Brennan’s claim that voters need to ensure that elected officials don’t abuse the power we give them, an assurance not secured simply by determining that a given candidate is, in some politically uninformative or irrelevant way, a nice guy. It doesn’t follow from any of these claims, whether singly or jointly, that every part–or even every significant part–of every elected office is so tightly connected to policy and governance that no other, broader moral considerations are ever relevant to judging a candidate’s fitness for office.
Brennan overlooks the possibility that a candidate’s moral character may be relevant to fitness for office without determining policy or governance in office. In particular, he overlooks the fact that we have an interest in ensuring that candidates deserve the offices they seek, where part of deserving “the office” is deserving receipt of the (often gigantic) perquisites that come with it.
It’s worth noting that a politician’s entirely private corruption (whether covert or overt) bears no obvious connection either to policy or to governance as so far explicated (or at least no clear, determinate, ex ante accessible connection to the voter). A politician might, as a doxastic matter track the truth on policy, and as a political matter be effective at promulgating truth-tracking policies, and yet still (separately from both facts) be sufficiently corrupt as not to deserve office, hence be disqualified from it. If there is such a thing as fitness or worthiness for office, where fitness or worthiness is not conceived by the voter as a proxy for policy or governance, character-based voting may well track the candidate’s sheer fitness or worthiness for office rather than the causal/proxy relation of character to policy or governance.
Consider a thought-experiment to this end. Imagine that Nadia Murad visits the White House. After the photo-op, the president drags her to a room somewhere in the White House, then rapes and murders her. Suppose that the rape and murder are discovered, and that the president is demonstrably implicated in them. Imagine that the crime has no bearing on policy, and doesn’t affect the president’s capacity for governance, either. Once the news comes out, the president and his administration simply carry on after the fact as though nothing had happened. I’m no expert on how such a case would actually be handled as a matter of law, but for present purposes, imagine that the ban on indictments of sitting presidents is taken to apply to the case at hand, so that the president can be impeached but not indicted even for violent felonies like rape and murder. Suppose, finally, that impeachment/removal is (for whatever reason) not in the cards. Now suppose that the president runs for re-election.
It seems plausible to think that character-based voting is justified in this context on some fairly likely or plausible assumptions about the president’s challenger. Consider three cases.
- The president faces a policy-equivalent but morally decent challenger, where policy-equivalence is a known, demonstrable fact. In this case, character breaks the tie between moral indecency and moral decency; we ought to vote for the morally decent candidate.
- The president faces a morally decent challenger where the policy issues are unclear or indeterminate, but the evidence doesn’t clearly indicate that the challenger is wildly suboptimal on policy. Again, it seems plausible enough to think we ought to vote for the morally decent candidate.
- The president faces a morally decent challenger whose policy views, though sub-optimal (both absolutely and relative to the president), do not clearly cross a threshold of policy disastrousness. This is a controversial or difficult case involving a variety of considerations, but it is not obvious that policy trumps character. It could be that we ought, at least in cases like this, to prefer the morally decent but somewhat policy sub-optimal candidate to the egregiously immoral candidate who is only somewhat better on policy and governance.
The general principle here is that when it comes to voting, even if policy is weighted higher than character, character might still have a certain weight of its own. If a candidate’s character is sufficiently bad, then past a certain threshold, that fact disqualifies or tends to disqualify the candidate outright.
There are at least two ways of thinking about this. One is to conceive of disqualification somewhat mechanically as varying inversely with policy/governance considerations (on a weighted conception of their normative significance), so that the worse a person’s character, the more that considerations of character come to outweigh or override matters of policy. We might imagine (very simplistically) awarding two points for rightness on policy, and subtracting one point for badness on character. In a case where two candidates are sufficiently close on policy, but one is much worse than the other on character, character will function as a deciding factor in voting.
The “tricky” issue here is epistemic: how does the voter capable of tracking the difference on character manage to track the similarity on policy without having a detailed knowledge of the policy issues, and without character’s itself tracking them? It’s not clear how, but it’s also not clear that the phenomenon in question is impossible. As I’ve been at pains to insist, there’s a lot here that’s not clear.
Another approach is to treat disqualification as a threshold, conceiving this threshold as a side-constraint on the pursuit of policy, where policy considerations are themselves subject to side-constraints. In other words, certain character-types are so bad that they’re regarded as off the table, as are certain (probable) policy outcomes (understood as probable conditional on the election of the candidate). The implication is that morally egregious candidates can be vetoed independently of policy outcomes unless the policy outcomes are so bad that those outcomes are themselves vetoed on grounds of their disaster-conduciveness.
In a case where one vote leads to the election of a morally indecent candidate (e.g., a rapist murderer) and the competing vote leads to an egregiously unacceptable policy outcome (e.g., a war that wouldn’t otherwise have taken place), the only option is not to vote. But in cases where one vote leads to the election of a morally indecent candidate and the competing vote leads to a mixed-bag policy outcome, it doesn’t obviously seem irrational to choose the latter over the former. Nothing of Brennan’s that I’ve read explains why it would be. Indeed, it’s Brennan himself who assures us that advanced democracies like the United States are sufficiently stable as to be immune to wild perturbations in policy. Forget elections: as far as Brennan is concerned, you can assassinate a Kennedy or elect a Trump and still be left with business as usual.*
A critic of my view might still legitimately ask: if moral character is ex hypothesi irrelevant to policy or governance, why is it nonetheless relevant to elected office, hence relevant to voting for elected office? Surely we need some way of avoiding Brennan’s legitimate complaint: as he correctly says, an elected office can’t simply function as an honorific for moral character generically conceived. Let me just sketch a few relevant considerations.
Take a case like that in the thought-experiment where one candidate’s behavior (and by implication character) is not just morally indecent in some generic sense, but involves criminal misconduct that creates a victim. Different conceptions of criminal justice will have different things to say about what justice requires with respect to such acts. But no plausible account will allow a criminal either to hold on to the very perquisites that facilitated his crime, or to be rewarded with those perquisites after the fact, or to keep the ill-gotten gains of the crime as though no crime had been committed. Certain wrongs demand a response, and past a threshold, the response they demand is incompatible with conferring the kinds of rewards that political office yields up.
Given the size of the reward involved in electing someone to public office—to say nothing of electing someone to the presidency of the United States—there is simply a mismatch (to put it mildly) between what the perpetrator deserves for having perpetrated a crime and the reward he gets as a result of being elected to office. If we conceive the mismatch involved as either a miscarriage or obstruction of justice, then contributing to it is itself a miscarriage or obstruction of justice. Since voting is such a contribution, it seems to follow that voting for a criminally indecent candidate is a case of wrongful voting—specifically, of wrongful voting on grounds of character. That leaves decent-but-not-wildly-suboptimal candidates as justifiable options-by-default.
There are probably other relevant considerations here, related to but not necessarily identical to or reducible to considerations of moral desert.
One is standing to govern.** The President is the nation’s chief law enforcement officer. Arguably, a law enforcement officer has to have the moral standing to execute the laws; there’s something morally wrong with a situation in which an egregious law-breaker is put in charge of law enforcement. The point is not simply that such a law-breaker is apt to undermine or subvert the rule of law once in office (though that’s probably true). It’s that morally speaking, a law breaker is not an appropriate choice for the role of law enforcement officer because he lacks the standing to do the job. A murderer-rapist president lacks the standing to execute the laws. So it seems wrong to vote for him. A morally decent candidate likely has better standing. To that extent, at least, it seems justified to vote for her.
Another consideration is deterrence. Even apart from considerations of policy or governance, we may not want to send the message that elected officials can murder and rape with impunity. One way of sending that message, admittedly an understated one, is to refuse to elect or re-elect such officials to office. Weak as this deterrent may be, it seems justified enough on grounds of deterrence. Deterrence is, after all, a matter of cumulative effectiveness, and not voting for a rapist-murderer may be one tool in the toolbox of deterrents we have at our disposal. Hopefully, we have more than this one tool, but it seems implausible to think that we’re debarred from using one tool because we’re obliged to use the more powerful ones we have to the same end.
It might be argued that the argument I’ve presented is an artifact of the extreme (and implausible) nature of the thought-experiment it’s based on. Yes, we get the character-based conclusion I’ve been discussing here if we imagine a murderer-rapist president, but that’s a pretty outlandish scenario. Does the argument really work on less fanciful situations?
It’s a judgment call, but I think it does. The argument works (if it works) with the murderer-rapist example because murder and rape are egregiously immoral acts indicative of an egregiously defective character, because they demand a rectificatory response by the wrongdoer, and because the response demanded is wildly incompatible with electing him to office. On top of this, the wrongdoing makes the candidate unfit to govern by depriving him of the standing to govern, and by generating worries that he might repeat a destructive behavior in office (using the office to shield himself from scrutiny). These features seem exemplifiable in a wide range of cases, not just in the particular case I’ve conjured up. If so, character-based voting will have application wherever such cases arise. How often they arise and where is an empirical matter. My point is simply that they’re possible in a sense that implies that character-based voting has more plausibility than Brennan’s discussion suggests.
In the fifth and final part of this series, I’ll respond to a Brennan-friendly objection mentioned in correspondence by my friend Ben Rossi (along with any other miscellaneous issues I have time for).
*A couple of provisos about things I say in this paragraph.
- I don’t have the time right now to hunt down the citations for the textual claims I’m making about Brennan in this paragraph, but will do so if anyone insists on it.
- Separate point: someone might argue that Brennan concedes the points I’m making a few pages before his discussion of character-based voting (compare Ethics of Voting p. 76 with pp. 84-85). I see the earlier concession as inconsistent with the later discussion.
- My “Kennedy/Trump” claim is meant more rhetorically than literally: I’m provisionally accepting the conventional (but highly disputable) view that Kennedy was a wonderful president and Trump an awful one, mostly because I can’t think of a recent U.S. president whom I regard as both wonderful and assassinated, not because I actually admire JFK. Lincoln, Garfield, McKinley, Ford, and Reagan don’t really work, either.
**I think of this by analogy with the more familiar topic, standing to blame. It’s not clear to me whether standing to govern is reducible to moral desert, or is a separate, non-reducible consideration. I could use some feedback on that. (Do I deserve it, as well? Probably.)