It does not bode well for the rule of law when the standards for determining whether a statute is constitutional change from case to case. Justices Kennedy, Souter, and O’Connor of the US Supreme Court once wrote, “Liberty finds no refuge in a jurisprudence of doubt.” (This line opened a plurality opinion in which they refused to overrule Roe v. Wade because, they said, for 19 years women had organized their lives around Roe. Why the babies that were aborted and would in the future be aborted could not be given a chance to organize their lives around the opposite outcome was not addressed.)
For over 50 years now, American federal courts have, in fits and starts, subjected statutes enacted by legislatures to a jurisprudence of doubt by moving the goal line that a government must cross in order to validate a statute when someone claims it violates the US Constitution. Not all judges have done so, but enough have to make it impractical to predict whether statutes affecting certain “hot button” issues are constitutional or not.
The May 31 opinion of the US Court of Appeals for the First Circuit that invalidated one section of the federal Defense of Marriage Act (DOMA) illustrates this process at work. DOMA was enacted by Congress in 1996 in response to efforts in certain states to redefine marriage to include two persons of the same sex. Both houses of Congress voted overwhelmingly for it. It is quite short as federal statutes go, just three sections, contained on one printed page. Section 1 names it. Section 2 says that no state must recognize the validity of any same-sex marriage that has been contracted in another state. Section 3 defines the words “marriage” and “spouse” for the purposes of federal benefits as including only marriages of one man and one woman.
The case that reached the First Circuit began when private plaintiffs challenged the constitutionality of Section 3 of DOMA under the Equal Protection Clause. They claimed that federal social security, veterans, and government employee health benefits were denied to them as same-sex spouses, notwithstanding that they were legal spouses as marriage is defined in Massachusetts. Because they were as equally married in Massachusetts as other married couples, they claimed that the federal government was violating their equal rights. (Massachusetts also filed a lawsuit over Medicaid and same-sex marriage, but the First Circuit did not distinguish between its claim and those of the private plaintiffs, and no separate discussion of the Commonwealth’s case is needed here.)
The federal district court ruled in favor of plaintiffs, and the government appealed. After filing its brief, the Obama Administration decided to abandon the defense of DOMA, an event that considerably delayed the appeal while the House of Representatives obtained attorneys to defend the law. The Administration did not merely bow out of the cases, however; it actually switched sides on the Equal Protection issue and filed a new brief agreeing with the plaintiffs.
Three standards for unconstitutional discrimination
The First Circuit Court of Appeals affirmed the result reached by the trial court, but it felt constrained to invent a new standard for measuring violations of Equal Protection in order to do so. In Equal Protection jurisprudence, there are three different goal lines for statutes to cross in order to be upheld when plaintiffs assert unconstitutional discrimination, depending on the nature of the case.
One, “strict scrutiny,” is applied to laws that discriminate by suspect classes (eg, race, religion, citizenship, and ancestry) or discriminate in regard to fundamental rights, like the right to free speech. The 14th Amendment was intended to preclude such discrimination. Under the test, the government must show a compelling need for the law and also that the statute was carefully crafted to attain only the needed remedy. This goal line is very distant, and few statutes survive to cross it when tested by “strict scrutiny.”
A shorter goal line is presented to statutes that do not create “suspect classes” or impair “fundamental rights.” This standard is called the “rational basis test.” Under this test, a law is presumed to be constitutional unless no rational purpose can be advanced for the classification it creates or the classification created by the law has no rational relationship to achieving a legitimate governmental purpose (eg, a law requiring all persons named “Smith” to pay a sales tax.) Most laws can satisfy the “rational basis” test.
In regard to laws that classify by sex, however, the federal courts of the 1970s did not like either of these alternatives. They made up a middle tier of measurement called “intermediate scrutiny.” The goal line is set not so close as in the normal “rational basis” case, but not so distant and unreachable as in “strict scrutiny.” The idea was that sex could not always be a “suspect class,” because there were obviously some types of classification by sex that inherently made sense (eg, strength and endurance requirements for physically-intensive jobs). Thus “strict scrutiny” was not justifiable. Yet the Supreme Court felt that women had been held down by the law and needed extra help to overcome that discrimination. They could not help being born as women. Hence “intermediate scrutiny.”
To cross that goal line, the courts held that the government must prove that there is an important governmental interest at stake and that the means employed have a substantial relationship to the goal. Many law review articles have been written on whether these concepts actually have any meaningful content.
Intermediate scrutiny would have probably been used by the First Circuit in the DOMA case, but for an inconvenient precedent. The First Circuit had already ruled against using that standard in a case challenging the so-called “don’t ask don’t tell” policy toward homosexuals in the armed services. Under the Court’s own procedures, a three-judge panel like the one hearing this appeal could not overrule a precedent already adopted in another case.
A new standard: “heightened scrutiny”
If intermediate scrutiny could not apply, even more was strict scrutiny out of the question. The only alternative left was the rational basis standard, but the Court could not bring itself to accept that, for as it candidly admitted, under that standard the private plaintiffs could not succeed. Homosexuals have been historically discriminated against for too long to allow that to happen, it appears. Therefore, the Court created a new measure, that of “heightened scrutiny.”
What “heightened scrutiny” means in practice is unclear, as its proponents seem to prefer. The Court said it did not want to create an “algorithm” — that is, a discernible formula for use in the future. Rather, it described its task as “to require a closer than usual review based in part on discrepant impact among married couples and in part on the importance of state interests in regulating marriage.” To exercise “closer than usual review” is so loosely defined as not to make it application of a rule of law, but an exercise in rhetoric. In this case, it masked a policy choice, not a legal judgment.
As precedent for such a loose, unprincipled, standard, the Court plucked three Supreme Court cases from history that it interpreted as involving an “historically disadvantaged or unpopular” protesting group and a justification for the law that “seemed thin, unsupported or impermissible.”
One case involved a federal law on eligibility for food stamps that disqualified from the definition of “households” groups of unrelated individuals that lived together. US Dept. Agriculture v. Moreno, 413 US 528 (1973). The First Circuit quoted the case as saying the law over-reached by disqualifying many eligible people and was the product of a “bare Congressional desire to harm a politically-unpopular group.”
The second case presented a municipal ordinance that operated to deny a permit to operate a group home for mentally retarded people within a certain neighborhood. City of Cleburne v. Cleburne Living Center, 473 US 482 (1985). The First Circuit again quoted the Supreme Court as finding that the ordinance was based on “mere negative attitudes, or fear, unsubstantiated by factors which are more properly cognizable in a zoning proceeding.”
Finally, the Court invoked Romer v. Evans, 517 US 620 (1996), which addressed an amendment to the Colorado constitution that forbade making homosexuals into a group that could invoke special protections of state and local civil rights laws. The US Supreme Court found that the amendment was an unprecedented “disqualification of a class of persons from the right to seek specific protection from the law.” Such a law, the Court thundered through Justice Kennedy, could be nothing other than the product of irrational hatred of homosexuals.
This is pretty thin gruel for creating a new level of “heightened scrutiny” in Equal Protection cases. All of these cases were decided under the rational basis test, and the conclusions focused on whether there was a rational connection with a legitimate state interest. (Whether or not they were decided correctly is beyond the scope of this article.)
It is worth noting that the First Circuit cited only cases in which the Supreme Court also characterized the motives of the government officials as biased and prejudiced. Although the First Circuit’s opinion stated on its surface that the cases reflected “the case-specific nature of the discrepant treatment, the burden imposed, and the infirmities of the justifications offered,” the quotations that the First Circuit pulled from these cases focused not on these factors, but on the bias and irrational prejudice that the Supreme Court saw in them.
In so doing, the First Circuit departed from expounding law and entered the realm of rhetorical persuasion. The portion of its opinion that cited and quoted from Supreme Court cases appears to have been intended to associate such motives with DOMA. Perhaps the First Circuit aimed at framing the issues in the mind of Justice Kennedy, the author of Romer and widely considered the Court’s “swing vote” on contentious issues.
The First Circuit judges attempt to preclude any such conclusion from being drawn. Near the end, the opinion specifically disclaimed finding that Congress had been actuated by irrational, base motives when it enacted DOMA: “In reaching our judgment, we do not rely upon the charge that DOMA’s hidden but dominant purpose was hostility to homosexuality.” It remains open to question, therefore, why the only quotations from the three key Supreme Court cases relied upon for the “heightened scrutiny” given to DOMA were quotations focusing on irrational bias against the groups involved.
The federalism argument
The First Circuit also stressed federalism as a basis justifying “heightened scrutiny.” It said DOMA interfered with the rights of the State in domestic relations law by impairing federal benefits married folks should receive. Here, the First Circuit has the federalism argument exactly backward. What constitutional necessity can require the federal government to let state law dictate the eligibility of people for federal benefits? That is a gross impairment of federal sovereignty. Federalism cannot mean that states determine disbursements from the federal treasury if the federal government does not consent or that state laws force the federal government to give greater justifications for its monetary allocations than otherwise. McCullough v. Maryland, decided early in the history of the Republic, ought to rule out this type of federalism.
Moreover, federalism concerns are not as unique to Equal Protection cases as the First Circuit described. The Supreme Court pointed out in San Antonio Independent School District v. Rodriguez that any time a state law comes before the federal courts on an Equal Protection challenge, federalism concerns are raised. It would seem less important to take account of such concerns when federal statutes involving federal money are involved than when state statutes are involved, for state sovereignty is not impinged upon nearly as much by DOMA as by federal court judgments invalidating state statutes. Again, the First Circuit has federalism backwards.
After creating a new standard of “heightened scrutiny” for this case, the First Circuit found the justifications offered for section 3 of DOMA to fall short. Saving money was found to be of little weight, for an historically disadvantaged group like homosexuals, the Court claimed, cannot protect itself in the political process from making all the sacrifices that result in the savings.
Bolstering the institution of marriage as a justification was rejected on the ground that the record did not show that DOMA would increase any governmental benefits to heterosexual married couples. This is a non sequitur. The Court failed to acknowledge that an official endorsement of same-sex marriage by government will be perceived by the public as an official rejection of the traditional view of marriage. It is in precluding such an official rejection that heterosexual marriage is defended. Finally, the connection between rearing children in a traditional marriage and DOMA was found to be nonexistent, as none of the federal benefits at issue were connected with rearing children.
The result that the Court reached was an ad hoc rationalization of a policy preference, not application of the rule of law. The Court moved the goalposts, as it were, so that the government could not win, as it would have under the applicable rational basis standard of review. Federal courts should do better than that.
James S. Cole graduated from Harvard Law School and practices law in St Louis, Missouri.