The Volokh Conspiracy

Canadian legal blogger Leonid Sirota has posted an interesting response to my post advocating extension of the franchise to politically knowledgeable children. He argues that it is unjust to require children to pass a knowledge test for voting, but not adults. He would therefore prefer to lower the voting age to 16 instead:

Pour ma part, je pense que l’option du vote à 16 ans est préférable à celle d’un test. Au-delà problèmes d’administrabilité évoqués par prof. Somin, ce sont arguments qu’il apporte lui-même qui semblent militer contre l’instauration de tests pour les mineurs. S’il n’y a pas de bonne raison de traiter les jeunes différemment des adultes, et les arguments de prof. Somin pour dire qu’il n’y en a pas sont très convaincants, alors il est sûrement injuste d’instaurer un test pour les premiers mais pas pour les seconds. Si les connaissances du système politique devraient être un critère pour pouvoir voter, il n’y a pas de raison pour ne pas appliquer ce critère aux adultes.

Sirota’s post is in French, which I understand, but most of our US readers probably don’t. I would translate the key passage roughly as follows: “If there is no good reason to treat the young differently from adults, and Prof. Somin’s arguments that there isn’t one are very convincing, it is surely unjust to institute a test for the former, but not the latter. If knowledge of the political system should be a prerequisite to getting the right to vote, there is no reason not to apply that criterion to adults.” If I have gotten the translation wrong, I hope Sirota or one of our French-speaking readers will correct me.

I am not convinced by Sirota’s objection. It’s true that my proposal doesn’t eliminate all unequal treatment of children and adults. Children would have to pass a test to be eligible to vote, whereas adults would not. But the same is true of both the status quo voting age of 18 and Sirota’s proposal to cut it to 16. In both cases, people younger than the voting age are categorically denied the franchise solely because they are minors. My proposal doesn’t completely eliminate the unequal treatment. But it does reduce it by at least giving children some opportunity to vote.

I would prefer to minimize government policies that discriminate on the basis of age. But there are some cases where age is so closely correlated with some important ability that it’s difficult to avoid. In this case, it’s undeniably true that children, on average, have much lower political knowledge than adults. And political ignorance is a serious problem that we don’t want to exacerbate by, say, allowing completely ignorant six year olds to vote. Giving the vote to knowledgeable children both diminishes the extent of age discrimination in voting policy and actually increases the average knowledge level of the electorate. It’s a win-win situation all round.

Why not, then, require adults to pass a knowledge test as well? As Sirota says later in his post, this is not a completely ridiculous idea. But it does have two serious flaws. First, as I discuss in an earlier post, there would be a serious danger of potential partisan bias in the design of the test. I’m not certain we can overcome them even with a test confined to children. But the danger is much greater if the test applies to all voters rather than just a small subset.

Second, there is a big difference between using a test to expand the franchise to a group that has always been barred from it, and using it to take away voting rights from millions of people who have them now. The latter is both politically infeasible and quite likely unjust as well. In general, I’m not a big fan of Burkean conservatism. But this is one of those areas where Burkean suspicion of drastic change is probably justified. Allowing knowledgeable children to vote isn’t subject to the same objection, since it is likely that it would alter the composition of the electorate only modestly. Probably only a small percentage of children would both want to take the test and be able pass it.

My proposal is far from perfect. It wouldn’t completely eliminate age discrimination in voting rights, and it certainly isn’t a complete solution to the problem of political ignorance. But it could well be a genuine improvement over the status quo on both fronts. I think that’s reason enough to at least give it serious consideration.

UPDATE: Sirota has put up a response to this post:

I think that what drives the disagreement between us is that we have different views on what justifies the denial of a legal right to vote (note that I am only talking about the legal situation―prof. Somin makes the case, in a separate post, that there is something like a moral duty to abstain from voting on issues or candidates about which one is ignorant, and I agree with him on that). In prof. Somin’s view the key factor is political knowledge, and lack thereof. I think that the real issue is not so much knowledge as maturity and capacity for judgment.

It is true that minors are generally less knowledgeable about politics (and other things) than adults. But they are also, on average, less mature and less capable of responsible judgment, and the law recognizes this diminished capacity by making them, depending on their age, less criminally liable, incapable of entering into (certain kinds of) contracts, etc. At least in criminal cases, it is quite clear that the reason for the distinction made between minors and adults is not knowledge of the relevant facts, but capacity for judgment. I think that it is the reason for the other distinctions too. Note, too, that the one category of adults to whom we uncontroversially deny the franchise are those too mentally ill, too lacking in judgment and decision-making ability to be responsible for their own decisions; such people have guardians, just like children do. They need not be ignorant―but their judgment faculty is severely impaired.

Now, in those areas where―I think―the law makes distinctions on the basis of age that are grounded in maturity and capacity for judgment, it usually does so by drawing bright lines.

Obviously, some degree of judgment is necessary for almost any decision. But in my view, there is a big difference between a situation where we are making choices in our own personal lives (e.g. – signing contracts, to use one of Sirota’s examples), and decisions where we are choosing rulers for an entire society. The former kind of decision usually requires only a modest amount of knowledge, and it is reasonable to assume that anyone with reasonably good judgment could acquire it. By contrast, good judgment of the sort that enables us to make effective personal decisions is not the most important factor in making voting decisions. There, we need knowledge of how policy affects people in very different circumstances from our own. The key ingredient here is not judgment of the kind that helps us make personal decisions, but knowledge of how large-scale political and economic systems work. Personal experience and judgment is only of limited relevance. A person can easily be an above-average quality voter while displaying poor judgment in their personal life, and vice versa. I make that case in more detail here. My guess is that any child who has more political knowledge than the average adult is also likely to have enough judgment to analyze that knowledge at least as well as the average adult eligible to vote (which may, of course, not be a very high level of competence in an absolute sense). On net, therefore, my proposal would still improve the overall quality of the electorate, even if perhaps only slightly.

Even with respect to criminal law and contracts, I think that knowledge plays a bigger role in the law than Sirota suggests. One big reason why we distinguish between minors and adults in these areas is that the latter are less likely to understand the true consequences of certain contracts, or of criminal acts. They may also be less likely to realize that a given act is wrong. That’s also a major factor in the differential legal treatment of the mentally insane. Indeed, the traditional legal definition of insanity actually defines it in terms of lack of knowledge: the defendant gets off if, as a result of his mental disease or defect, he (i) did not know that his act would be wrong; or (ii) could not understand the nature and quality of his actions.

It’s also worth noting that Sirota’s proposal of reducing the voting age to 16 without imposing tests of either knowledge or judgment would result in a reduction of the quality of the electorate on both dimensions (assuming, of course, that 16 to 18 year olds on average have worse judgment than adults). And, as I previously noted, it would continue to engage in more blatant age discrimination than my idea, since those under 16 would be forbidden to vote regardless of their level of knowledge or how good their judgment might be.

Finally, contra Sirota’s description of my view, I don’t necessarily regard it as a “mistake” that we have extended the franchise to include all or nearly all adults. As discussed above, I’m skeptical that we can trust government enough to come up with a knowledge test that would apply to the entire electorate. Creating a knowledge test that applies only to children poses fewer risks of abuse.

In response to my post arguing that we should grant voting rights to politically knowledgeable children, Paul Horwitz of Prawfsblawg makes the following suggestion:

What spurred this post is Ilya Somin’s argument on the VC yesterday that knowledgeable children ought to be allowed to vote. He addresses some standard objections in his post, but a number of his commenters wrote to argue that such a rule, if enforced by knowledge or literacy tests, would end up privileging some groups and disadvantaging others (as, indeed, previous tests have done in the United States). Indeed, given massive educational inequality in this country, it’s hard not to see how this proposal wouldn’t give much more electoral power to the wealthy, well-educated, mostly white elite. Unless....perhaps Ilya would welcome a trade-off: knowledgeable children get the vote, in exchange for guarantees of massive public/private efforts to assure meaningful educational and welfare rights to ensure that the opportunity to be a knowledgeable child voter is fairly and widely distributed among the entire population rather than limiting that vote to enclaves with better resources. I’m just going to go ahead and consider this Ilya’s very subtle case for overruling San Antonio School District v. Rodriguez [the 1973 decision that ruled that there is no constitutional right to equal education spending].

Paul is, of course, entitled to interpret my argument however he wants. But I have no desire to overrule Rodriguez. Setting aside the legal merits of the case, extensive evidence compiled by economists Eric Hanushek and Alfred Lindseth shows that increasing education spending in public schools does little or nothing to increase educational achievement. On the other hand, I would be happy if my child-voting proposal were paired with increased school choice, which does have a demonstrated record of increasing educational achievement among poor minority children.

That said, I think the child-voting proposal could potentially be implemented even in the absence of increasing school choice. Most studies of racial attitudes suggest that racism is correlated with age (the young are much less likely to be racist than their elders) and inversely correlated with knowledge and education (the knowledgeable and educated are less likely to be racist). Therefore, even if knowledgeable children are disproportionately likely to be white and/or wealthy, the resulting electorate is not likely to be more hostile to minorities than the one we have now, and may well be less so. Much evidence also suggests that there is little correlation between political attitudes and narrow self-interest (after controlling for other variables that affect political opinion). So it’s unlikely that knowledgeable children (or knowledgeable adults) would seek to promote policies that benefit the rich or some narrow interest group at the expense of the general public or the poor. Indeed, at this point in history, the young, including the well-educated young, are disproportionately liberal Democrats.

Finally, it’s worth noting that the unequal historical impact of literacy and knowledge tests was in considerable part the result of biased application. Unbiased test administration would have a much less skewed effect. And, obviously, racial or ethnic bias in test administration would today (unlike in the Jim Crow era) be grounds for lawsuits under the Voting Rights Act and the Fourteenth and Fifteenth Amendments.

That said, as I noted in my original post, I do take seriously the danger that incumbent political leaders would try to skew the test to favor their political supporters and disfavor opponents. I don’t at this point have anything approaching a complete solution to this problem. But I’m not convinced it’s completely intractable either. One possible approach would be to leave the design of the test to a politically balanced, nonpartisan commission. Some election scholars believe that this approach has helped solve the comparable problem of gerrymandering of political districts in favor of the incumbent party in places like Nebraska. But I admit that for my proposal to actually work in practice, this issue would have to studied much more carefully.

UPDATE: I should note that I’m aware that the Voting Rights Act of 1965 suspended the use of tests for voting in much of the country because they had been used to exclude African-Americans. It’s likely that the Act would have to be modified to permit knowledge tests for minors in order for my proposal to be allowed in states covered by that section of the Act. I think there are obvious and important differences between using knowledge tests to exclude adults, and using them to include children who otherwise would be categorically barred from voting in any case. In any event, allowing states to use knowledge tests for children is still compatible with also giving both individuals and the federal government to sue states that attempt to use the test for purposes of racial or ethnic discrimination, whether by designing it in a biased way, or through unequal enforcement.

On this election day, as on most others, we will hear a lot about the need to increase turnout and the dangers of voter suppression. But few will even consider questioning the systematic exclusion of a huge part of our population from the franchise: children under the age of 18. We allow even the most ignorant and irresponsible adults to vote, but exclude even the most knowledgeable and insightful children. And to add insult to injury, we saddle them with a mediocre education system and trillions of dollars in public debt that they will someday have to repay.

For reasons I outlined during the last presidential election, this is both unjust and counterproductive. We should at least consider allowing children to vote if they are more knowledgeable than the average adult voter:

The main objection to giving children the vote is that they lack the knowledge to make informed choices. Of course the same is true of most of the adult electorate, who are rationally ignorant about politics and public policy, and often don’t know even very basic facts. Nonetheless, it’s probably true that the average child knows a lot less about politics than the average adult, and that may be a good reason to deny most children the franchise. But why deny it to all of them? If a minor can pass a test of basic political knowledge (say, the political knowledge equivalent of the citizenship test administered to immigrants seeking naturalization), why shouldn’t he or she have the right to vote? Such a precocious child-voter would probably be more knowledgeable than the majority of the adult population. Giving her the right to vote would actually increase the average knowledge level of the electorate and thereby slightly improve the quality of political decision-making. I’ve met twelve-year-olds with far higher levels of political knowledge than that of the average adult. You probably have too.

Once the knowledge objection is off the table, all the arguments for giving adults the right to vote also apply to sufficiently knowledgeable children. Like... adults, children have a claim to the franchise because government policies affect them too, because otherwise their interests might be undervalued in the political process, because it affirms their status as citizens with equal rights, and so on.

Obviously, there might be some difficult administrative issues. For example, we might not trust the government to put together an adequate knowledge test. But I don’t see any principled reason to deny the franchise to children whose political knowledge is greater than that of most adult voters.

Other standard objections to letting knowledgeable children vote also don’t hold much water, and in some cases resemble long-discredited justifications for excluding women from the franchise:

Some people might worry that even knowledgeable child-voters will be “unduly” influenced by their parents’ preferences. Given the existence of the secret ballot, I doubt that this would be a major problem. Moreover, children who are knowledgeable enough to pass the test and interested enough to take it will probably have at least some political ideas of their own that aren’t easily susceptible to parental suasion. In any event, I’m not sure that the possibility of parental persuasion would necessarily be a bad thing. The objection is in fact similar to one of the arguments once raised against giving women the right to vote – that they would be unduly influenced by their husbands or fathers. Husbands will often influence the views of their wives (and vice versa); similarly, parents will influence those of their children. That doesn’t by itself justify denying either married people or children the right to vote....

[C]hildren might lack maturity or life experience, as well as knowledge.... I’m just not convinced that either is tremendously useful for voting. Most voting decisions have to do with complex, large-scale policy issues that can’t easily be weighed based on personal experience. Realistically, even most adults have little life experience that is directly useful in assessing difficult policy issues... At the very least, it seems to me that superior knowledge might well outweigh inferior maturity and life experience. And I’m only advocating giving the franchise to children who can demonstrate knowledge levels superior to those of the average adult voter...[Moreover, we don't exclude even the most immature adults from the franchise, even if they are highly ignorant to boot].

[Some cite] the value for voting of such “adult” experiences as holding a job, paying taxes, owning property, and so on.... I’m skeptical that these experiences greatly improve the quality of voting decisions. Even more to the point, however, we don’t exclude from the franchise the many adults who lack some or all of these experiences – even if they are also ignorant of even the most basic political knowledge. If lack of life experience is not enough to justify exclusion of even the most ignorant adults from the franchise, I don’t see why it should be considered sufficient to exclude vastly more knowledgeable minors.

The key conclusion is this: There is no plausible justification for excluding knowledgeable children from the franchise that doesn’t also apply to large numbers of adults. We could easily exclude adults who don’t have a job, don’t own property, or lack whatever other life experience supposedly makes you a qualified voter. But virtually everyone agrees that we shouldn’t.

One can argue that the exclusion of children is more permissible than that of comparable adults because it is “only” temporary. But every election leads to policy decisions that have permanent long-term effects. Knowledgeable children who were denied the vote in 2004, 2008, and this year, are going to be massively affected by the decisions made by the winners of these elections. And, of course, the exclusion of adults who don’t have jobs or other relevant life experience might also be temporary, lasting only until they manage to get that experience.

Argentina and the German state of Bremen recently extended the franchise to people over the age of 16, irrespective of their knowledge levels. We should consider doing the same for at least those children who know as much about politics as the average adult. I don’t underestimate the practical difficulties of implementing this idea. For example, it may be very hard to come up with an unbiased knowledge test for aspiring child voters. But the issue at least deserves serious consideration. We should not continue to exclude millions of knowledgeable potential voters from the franchise, unless there really is no good way to avoid it.

I am not the first to advocate giving at least some children the right to vote. Harvard political scientist Paul Peterson has been making that case for a long time (albeit on somewhat different grounds). And columnist Michael Kinsley defended a similar idea in 2011. But Kinsley and Peterson are arguing that parents should be allowed to cast an extra vote for each of their children (though Peterson would give parents the option of letting the children make their own choices). This is a little like giving husbands the right to cast an extra vote for their wives, instead of letting married women vote for themselves. I say let knowledgeable children cast their own ballots.

Finally, it’s worth noting the commonality this post and my last one, in which I urged adult voters to consider not voting on issues they know little or nothing about. Knowledge, not age, should be the main qualification for exercising political power at the ballot box. We may understandably shy away from giving government the power to use knowledge tests to narrow the franchise. But it’s much tougher to argue against using them to expand it.