Too Narrow and Too Broad

This post is inspired from reading the beginning of another amicus brief from Perry, this one by several law professors, just in response to a certain little point they make. In certain legal contexts but most notably First Amendment jurisprudence, there is a concept of "overbreadth." This is the idea that, even if there is a sufficiently strong justification for some apparently speech-suppressing law or government action, that law/action might still be unconstitutional if it has effects beyond its justification. To take a comical example, suppose that anti-black hate speech had been held to be unprotected speech, i.e. that preventing hate speech against people of African ancestry had been held to be a sufficient justification for the suppression of speech. A law prohibiting all white people from ever speaking, on the grounds that nearly all anti-black hate speech comes from white people, would be overbroad. It would accomplish its goal, yes, and the goal is an acceptable one (in this scenario, though not in the real U.S. courts), but it would then also accomplish a whole lot of other stuff not included in the acceptable goal, so it would be obviously invalid.

There is also the concept of a law's being underbroad, although I don't know if that word is ever used. The idea here is that, sometimes, a law in pursuit of a sufficient justifying interest will be invalid if it fails actually to accomplish that interest. This only occurs when the law is acting in an area where strong justification is required, and might be most at home in Equal Protection jurisprudence. In a way the principle I discussed in my last post was an example of this idea: in an insult-without-injury scenario we can be reasonably sure that, whatever the justification proffered for the policy might be, it is in fact under-broad to accomplish that policy, since it accomplishes no material effect at all. The principle basically is, even if your aims are sufficiently noble, if your proposed policy both fails to actually attain those goals and involves significant injury to some constitutional right or value we will find it invalid. For a good metaphor, suppose that there is some dam, or really any wall of a large container of water, in which there is a large hole through which water is pouring. Someone proposes some means to stop up the hole, which would have a great cost to implement. If that method will not actually plug the hole, if the plug is too narrow to fill the hole, and we can know that for sure, then we should not pay that cost for no actual benefit, even though the proposal resembles an action to fix a genuine problem.

Note that this is distinct from the principle, which is false and which the Court has repudiated in so many words, under which the government would be prohibited from only solving part of a problem. It may do that, but as the means it chooses becomes more and more constitutionally problematic we become more and more insistent that it at least come close to solving the problem entirely, and if it will not actually solve the problem at all we should not hesitate long to say "stop!"

So a law can be either too broad or not broad enough. Can it be both? Definitely. Lots and lots of laws (or, hopefully, lots and lots of hypothetical laws) are both too narrow and too broad. The brief I'm reading describes gay marriage bans as one example, relative to their purported justifications in terms of protecting traditional marriage. Too broad because they involve substantial violations unrelated to that goal; too narrow because they don't actually do anything to achieve that goal. But, hang on... isn't there a shorter way of saying "too narrow and too broad"? Why yes, yes there is. Here are some possibilities: "wrong," "bad," "terrible." If you're doing a coloring-book and you're supposed to color in a certain area red, and you draw a big red patch on some completely different part of the page, your red patch is too broad because it covers a lot of area outside the lines, but also too narrow because it misses a lot of area inside the lines. It is, in other words, in the wrong place. (Not that I endorse prescriptivist coloring-book norms, mind you.) One way to think about this is in terms of "type I" and "type II" errors in the statistical context. A too-broad-and-too-narrow policy is one with both very high "false positive" errors, applications to places where it shouldn't be applied, and very high "false negative" errors, non-applications to places where it should be applied. Normally we think of these two errors as being balanced against each other; if you want less of one, you might need to accept more of the other. But these policies don't bother having less of either! They just have lots of errors, lots of falseness. They get it wrong, and apply themselves in all the wrong places. They suck!

Of course, those words only apply to an analysis of the policy relative to its purported justification. There's something at which anti-gay marriage policies are excellent. Perhaps lots of things. Expressing the sense of the community that gay people are inferior, for instance. They do a great job of that. But they do a terrible job of "protecting traditional marriage," if you don't just interpret that phrase as a cipher for "keeping the gays out because they're icky and we don't like them." In fact, any time you see a law that appears to be both too broad and too narrow you should be tipped off that the purported justifications are not the real ones. Probably the people enacting it were not idiots, which means there probably is some other motivation behind the law, in light of which it won't look so bad. That inference gets stronger, of course, as the relative degree of the two-way miss gets higher. All laws are slightly over-broad and over-narrow, because life is imperfect. There might even be pretty substantial errors on both sides and yet have the law be a good-faith effort to address the problem it claims to address, either because that problem is hard or because the legislators or executive officers just made good-faith mistakes or miscalculations about the effects of their policy. (That last explanation might not be enough to avert a constitutional challenge if we're in a problematic area.) But when a policy has substantial effects unrelated to its purported justification and does virtually nothing to advance that purported justification, we will be justified in inferring that there was a hidden agenda behind it, probably aimed at doing whatever it is the law actually does.