When discriminatory US laws are constitutional

It occurs to me that some of y'all may be as unclear as I've been on why the plaintiffs in Perry v. Schwarzenegger (the Prop 8 trial) have been making some of the points they've been making. [Added later: especially all the stuff about whether gays and lesbians are politically powerless or not.]

So here's an attempt at a backgrounder for one aspect of the case. Note that most of what I think I know about this comes from Wikipedia; if I'm wrong about anything here, I hope someone will correct me.

The core issue underlying many questions of whether a given discriminatory state law in the US is constitutional is the Equal Protection Clause of the Fourteenth Amendment to the US Constitution. It says that "no state shall . . . deny to any person within its jurisdiction the equal protection of the laws."

So when a state enacts a discriminatory law, people affected by that law can go to Federal court to try to get that law overturned under the Equal Protection Clause.

[I originally had a discussion here of the criteria that determine whether a group is a suspect class; I got that information from Wikipedia. ~Elliot pointed out that some of the information was dubious. After doing some further research, I ended up rewriting that section of the Wikipedia entry. I'm therefore reluctant to let my misleading quote of the old material stand. So although I hate to make significant changes to an entry after publication, I'm doing so here: about five hours after my original post, I'm removing some material that was here and replacing it with, I hope, more accurate info.]

The court then decides what category the group of affected people falls into, using a combination of various criteria. In particular, the court determines whether the people affected are a "suspect classification" (often shortened to "suspect class") deserving of special protection. Here's Wikipedia's discussion of those criteria (I wrote most of this discussion, based on a book, some Supreme Court decisions, a few websites, and bits of an earlier Wikipedia list of criteria):

Some of the criteria that have been cited include:

  • The group has historically been discriminated against, and/or have been subject to prejudice, hostility, and/or stigma, perhaps due, at least in part, to stereotypes.
  • The group is a "discrete" or "insular" minority.
  • They possess an immutable and/or highly visible trait.
  • They are powerless to protect themselves via the political process.

[End of edited material.]

If a group doesn't quite qualify as a suspect class, the court may determine that they're a quasi-suspect class, or that they're not a suspect class at all.

Depending on which category the group falls into, there are three different levels of scrutiny that the court may choose to apply to the discriminatory law:

Discriminated class Level of scrutiny Description of scrutiny
Non-suspect classification rational basis review Considers whether the law is "rationally related" to a "legitimate interest" of the government. In most cases where this level of scrutiny is applied, the law is allowed to stand.
Quasi-suspect classification intermediate scrutiny Considers whether "the law or policy being challenged furthers an important government interest in a way that is substantially related to that interest." (Wikipedia)
Suspect classification strict scrutiny Considers whether there's a compelling governmental interest, whether the law is narrowly defined, and whether it's the least restrictive means of serving the interest.

So that's one part of what's going on in the Prop 8 trial. The plaintiffs are attempting to show that GLB people in the US satisfy those suspect-class criteria. If the Federal court finds that GLB people are indeed a suspect class (as the California Supreme Court found a while back), then Prop 8 must be subject to strict scrutiny, and can stand only if it's for a compelling governmental interest and is narrowly defined. (Of course, a court may find that GLB people are a suspect class, but still find that Prop 8 withstands strict scrutiny. Wikipedia says that "laws survive strict scrutiny over thirty percent of the time.")

(Note that Federal standards are the minimum standards that states must apply; state courts can also apply stricter standards than the Federal ones.)

In particular, this is why the plaintiffs have been trying to show that GLB people are "powerless" in the political process, and why the defendants have been trying to suggest that GLB people (especially in California) have vast amounts of political power, and powerful protection from straight politicians.

(Reminder: you can read a quasi-transcript of the whole proceeding so far at firedoglake.)

The wacky part of all this, to me, is that it feels like it comes down to the Prop 8 proponents saying "Gays aren't widely discriminated against, so it's okay to pass a law that discriminates against them."

Which sounds silly to me when I put it that way, but I think that sort of argument is kind of built in to the "suspect class" system. Wikipedia says that (for example) age, disability, political preference, and political affiliation are not considered (by Federal standards) to be suspect classes, so laws that discriminate on the basis of those things are subject only to rational-basis scrutiny. In other words, if the Federal court can think of some potential legitimate government interest the state might have in passing such a law, then the law is allowed to stand.

[Added later: Does that mean that a law barring elderly people from marrying would be subject only to rational-basis scrutiny? I'm not sure.]

I can see the underlying idea here: if a group is subject to widespread systemic prejudice, then the US government has a stronger-than-usual interest in making sure that states don't pass laws discriminating against that group, even if the state claims to have some vaguely legitimate interest in such a law. (And, on the flip side, the US government doesn't want to be interfering in state law except in really important situations, such as those in which a law discriminates against a group that's already subject to widespread systemic prejudice.)

But it seems weird to me that it's possible to argue that since the discrimination against a group isn't really all that bad, it's okay for a state to discriminate against them.

P.S.: The New York Times's Bay Area blog has a video of the press conference that came after the close of arguments the other day. Fifteen minutes of awesomeness by plaintiff lawyers David Boies and Theodore Boutrous, followed by fifteen minutes of smug repellence from defense lawyer Andy Pugno. (Might I be biased? Just a teensy bit?) Worth watching the first half; I watched the second half out of a misplaced sense of duty to hear the opposition, but there's no real need to do so. Pugno's argument boils down to "The plaintiffs made a lot of good arguments, but the points they made are irrelevant to the legal issue at hand, so we didn't bother wasting time rebutting them."

8 Responses to “When discriminatory US laws are constitutional”

  1. ~Elliot

    I’m highly dubious about Wikipedia’s list of criteria. The stuff about discreteness and insularity comes from the famous “footnote four” of United States v. Caroline Products Co., and some of the other criteria come from San Antonio Independent School District v. Rodriguez, which remarked that heightened protection is merited to those “saddled with such disabilities, or subjected to such a history of purposeful unequal treatment, or relegated to such a position of political powerlessness as to command extraordinary protection from the majoritarian political process.” Whether this is the real test is highly dubious, as it doesn’t much resemble the actual analysis in the Supreme Court’s cases finding suspect or quasi-suspect classifications. (My personal view is that the Supreme Court’s decisions on this issue have been almost entirely arbitrary.) But following the Supreme Court’s officially declared criteria is the best a lower court can do.

    I don’t know where Wikipedia got the immutability criterion. I’m aware of no federal case that’s said immutability is required for suspect classification status. Certainly the Supreme Court has never said any such thing. Alienage is a suspect classification, but is very much not immutable. I think Wikipedia is just wrong about this one.

    You’re right that the plaintiffs’ equal protection argument has mostly been directed at the “history of purposeful unequal treatment” and “powerlessness” criteria, which come from the San Antonio Independent School District test I mentioned above. Conversely, the defendants have emphasized that there’s a lot less anti-gay discrimination than there used to be and they’ve argued that gays and lesbians have a lot of political power.

    The conventional legal view on “it’s OK to discriminate if discrimination isn’t all that bad” is to say that all laws discriminate. Unemployment insurance discriminates in favor of the unemployed, insider-trading laws discriminate against corporate insiders, the drinking age discriminates against people under 21, homicide laws discriminate against killers. So what the law is trying to do with this suspect-classification stuff is to distinguish between what kinds of discrimination should be prohibited (unless there’s a really good reason) and which should generally be allowed.

    One more point: you talk about GLB people, but this case is really just about GL’s. The plaintiffs are claiming that Prop 8 is anti-gay discrimination, not that it’s anti-bisexual discrimination.

    (There’s an entirely different legal argument going on in parallel, based on Romer v. Evans, but that’s an entirely different legal bag of worms.)

    • Jed

      Okay, I’ve now edited the Wikipedia entry (which, I agree with you, appears to have been just plain wrong in some regards); and because I hated the idea of perpetuating incorrect info in my entry, I’ve also replaced my list of criteria that I had quoted from Wikipedia with the new list that I’ve just added to Wikipedia.

      Thanks again for calling attention to these problems!

      Also thanks for the “all laws discriminate” explanation; I think that makes more sense to me now.

      Re GLB vs GL: interesting; hadn’t occurred to me that the plaintiffs are leaving bisexuals out of it. But now that you say that, I suppose I’m not surprised.

      Some musings off the top of my head, which may be way offbase: On the one hand, the bisexual case magnifies the ridiculousness of the law (if I fall in love with a woman, I can marry her, but not if I fall in love with a man); but on the other hand, it muddies the waters, and leaves the door open for the argument that I can get married, it’s just that the state is putting a restriction on who I can marry (just as it puts other limitations and restrictions on who people can marry).

      So I guess it’s probably just as well that they’re leaving out bisexuals.

  2. Jed

    Thanks much for all this info!

    I ought to have run this entry by an actual legal professional before posting it. I’m sorry not to have done so.

    I’m having a hard time finding any sources online that even attempt to outline what the criteria are (or whether there are any) for determining whether a class is a suspect class. Most of the sites I’m looking at say things like “a suspect class is one subject to strict scrutiny.” And then they list some suspect classes. So maybe they’re agreeing with you that there aren’t really any such criteria, that the Supreme Court has simply declared a few particular classes to be suspect?

    It appears that a couple of the cases where suspect classifications have been discussed are Korematsu v. United States (1944) and Graham v. Richardson (1971); I’ll look those up soon (along with Loving v. Virginia (1967), of course). Looks like intermediate scrutiny is discussed in Plyler v. Doe (1982).

    The USLegal page on suspect classification gives a list of four criteria fairly similar to Wikipedia’s, but slightly different; for example, they say “highly visible” instead of “immutable.” Unfortunately, their page gives no references/citations either.

    Aha! Here we go: Sexual Orientation and Human Rights, by Robert Wintemute (1995), gives a list of seven criteria (one of which is immutability), but adds: “In its various suspect classification decisions, the Supreme Court has referred to different combinations of these requirements, but has never provided a coherent theory explaining their purpose and relative importance.”

    Interestingly, political powerlessness (which was really the main point of my entry) is not one of the seven criteria, though the “discrete and insular” criterion does add “whose political participation has been seriously curtailed because of prejudice.”

    Of course, I’m not sure how reliable that source is. But it’s the only source I’ve found so far that even attempts to give or discuss criteria. And each one of the seven listed criteria is heavily footnoted; for example, the word “immutable” appears to derive from Lyng v. Castillo (1986), if I’m understanding right.

    • Jed

      Aha again! From the decision in Lyng v. Castillo (1986):

      The District Court erred in judging the constitutionality of the statutory distinction under “heightened scrutiny.” The disadvantaged class is that comprised by parents, children, and siblings. Close relatives are not a “suspect” or “quasisuspect” class. As a historical matter, they have not been subjected to discrimination; they do not exhibit obvious, immutable, or distinguishing characteristics that define them as a discrete group; and they are not a minority or politically powerless.

  3. ~Elliot

    Interesting, I had not been aware that there were cases where the Supreme Court used that rhetoric about immutability in some of those cases. It still makes no sense given the status of alienage as a suspect classification, but the Supreme Court has often failed to make sense.

    • Jed

      Well, I think the idea is that immutability is one of several possible factors to consider, none of which are necessary or sufficient by themselves.

      The Wikipedia article originally made it sound like all the listed criteria had to be satisfied; as far as I can tell, that was just wrong. So I’ve changed it to being a list of relevant factors, which seems to be more in line with what the Wintemute book says about it.

  4. Benjamin Rosenbaum

    The point you are neglecting to observe here, I think, is the full range of the meaning of the word “discriminate”. It means discerning a difference between two things, and acting according to that difference. When you pick “age, disability, political preference, and political affiliation”, you are pointing out things that really are suspect classes — subject to persistent patterns of discrimination — even if the courts don’t think they are. But those things are just a tiny fraction of the ways we “discriminate”.

    For a bar to offer free drinks to those who wear green on St. Patrick’s Day requires the bar to legally discriminate between people based on what color they are wearing. For Nick Mamatas to ban people from submitting to Clarkesworld because they argued with his rejection letters requires him to be able to legally discriminate on that basis. Or, to look at actions of the state, for the state to set an age for legal driving or drinking requires it to discriminate on the basis of age; for it to require you to pass a swim test to get a boating license requires it to be able to legally discriminate between people who can swim and those who can’t.

    Now, you’ve done more research here, so it may be that I am confused about the legal meaning of the word “discriminate”. But this is my impression.

    What I’m saying is that I don’t think the notion of “suspect classes” is at all weird — certainly not in the case of the state restricting the ability of businesses to discriminate. I think discriminating against someone for being gay is profoundly different from discriminating against them because they swim poorly.

    • Jed

      I don’t know anything about the legal meaning of “discriminate,” but what you’re saying sounds basically reasonable/plausible to me.

      I agree with you that “discriminating” by allowing some kinds of people to do something and not allowing other kinds of people to do that thing is legitimate, and it seems like common sense that the criteria by which such discriminations are made should be somehow relevant to the situation. This is basically what ~Elliot was saying near the end of his first comment above.

      And there are plenty of situations in which there could be discrimination using criteria that are irrelevant to the situation but that nonetheless don’t constitute a suspect or quasi-suspect class. So in those situations, the rational-basis test is used, which also seems to make sense in general.

      The part that I’m hung up on (though this is really just a minor side note to my entry here) is what happens when there’s a class of people who are discriminated against but who don’t quite meet the criteria for being a suspect or quasi-suspect class. In those cases, apparently, the rational-basis test (which I gather is generally an easy test to pass) applies, which could result in the gov’t adding to the burden of prejudice on such a class of people.

      You said that age, disability, etc “really are suspect classes … even if the courts don’t think they are,” but that’s really my point in this aside: a suspect class is by definition whatever the courts say it is, so if the courts don’t consider age (for example) to be a suspect class, then laws that discriminate against people because of their age are subject only to rational-basis scrutiny. And I’m not convinced that that’s sufficient. If I’m understanding it right, it would be okay for a state to pass a law against allowing (for example) wheelchair users to marry, as long as someone could claim that that law was “rationally related” to a “legitimate interest” of the government.

      In other words, I would like more things to count as suspect classes than the law currently recognizes; it sounds like you agree.

      It may well be that this is almost never an issue in real life. But the situation at hand is exactly such an issue: there’s a class of people who are widely discriminated against; the state passed a law that furthers that discrimination; the people defending that law argued that the class is not discriminated against enough to deserve protection from further discrimination.


Join the Conversation