Kim Davis and the Nature of the Fundamental Right to Marry
by Michael Dorf
In a useful update on the Kim Davis case on Balkinization, Marty Lederman raises the question whether the improvised forms that are now being used to issue marriage licenses in Rowan County, Kentucky--insofar as they differ substantially from the forms used in other counties--violate the equal protection rights of applicants for marriage licenses in Rowan County, even if the forms suffice to make the ensuing marriages legal under state law. Marty asks (in his question 5):
Palmer, recall, is the case in which Jackson, Mississippi closed all of its public swimming pools rather than integrate them. The Supreme Court, in a 5-4 ruling, held that the city did not thereby violate the Equal Protection Clause. Jackson, after all, was under no obligation to establish or operate public swimming pools, and so by closing them for people of all races it did not discriminate on the basis of race.
While Marty thinks about Palmer, let me suggest that there is a potentially crucial difference between swimming pools and marriage licenses--namely, there is, under the SCOTUS cases, a fundamental right to marry but no fundamental right to swim in a government-operated pool. The cases recognize that even when the government does not draw any distinctions based on any illicit criteria, discriminatory classifications with regard to fundamental rights trigger strict scrutiny.
The relevant case law includes two categories of fundamental rights: (1) those that are recognized as fundamental for substantive due process purposes are, ipso facto, also fundamental for equal protection purposes, so that distinctions drawn regarding their exercise must satisfy strict scrutiny; and (2) rights that are fundamental only for equal protection purposes, such as the right to vote. A state need not have an elected attorney general at all--and thus there is no substantive due process right to vote for the office of attorney general--but if the state does have an elected attorney general, then inequalities in the distribution of the franchise with regard to attorney general elections are strictly scrutinized.
One fair objection to Obergefell v. Hodges--and indeed to the entire line of the Supreme Court's right-to-marry cases--is that the right to marry ought to be the second kind of fundamental right. Under this approach, a state would not be obligated to create a civil institution of marriage at all, but if it does, then inequalities with respect to marriage are strictly scrutinized. In this view, Obergefell is rightly decided as an equal protection fundamental right case, because all of the states did have an institution of marriage and their reasons for denying marriage to same-sex couples were barely rational, much less compelling; but if marriage is the second kind of fundamental right, then Justice Kennedy and the rest of the Obergefell majority oughtn't to have relied on the Due Process Clause. Marriage would be a fundamental right "in" equal protection, just as voting is.
One obvious advantage of this approach would be that it would answer an objection raised by the dissenters: The Constitution, they said, generally protects negative rights against state interference, not affirmative rights to state recognition or assistance; yet the marriage right looks like the latter; when the government denies recognition to a same-sex couple's marriage (or to an opposite-sex couple's marriage), it does not do anything to them; it fails to do something for them.
That is a sound objection in principle but it is not fairly lodged against the Obergefell majority. The Court in Loving v. Virginia quite clearly rested its holding on the alternative ground (endorsed by eight justices) that, in addition to the equal protection violation occasioned by the use of a race-based classification, the Viriginia laws at issue were invalid because they "deprive[d] the Lovings of liberty without due process of law in violation of the Due Process Clause of the Fourteenth Amendment." Zablocki v. Redhail also relied on a substantive due process right to marry as well as equal protection. (The other main right-to-marry case, Turner v. Safley, arose in the prison context, so the standard for evaluation was somewhat different, although even then, the Court found the infringement on the right to marry unconstitutional.)
Although it would be an overstatement to characterize the key portions of the cases recognizing a substantive due process right to marry as mere dicta, it is true that in each of the cases equal protection would suffice for the result. Thus, critics who think that there is something highly problematic about a substantive constitutional right to government recognition (as opposed to non-interference) could argue in some future case that the right to marry ought to be reconceptualized as simply a type-2 fundamental right: one that, like voting for state AG, cannot be distributed unequally if the state recognizes it, but that the state need not recognize at all.
Despite efforts in Oklahoma and elswhere to eliminate civil marriage, it seems unlikely that a test case will actually arise in which the Supreme Court will be asked to consider "demoting" marriage from a fundamental right for substantive due process purposes to a fundamental right only for equal protection purposes. But even if that were to happen--and even if one were to think that marriage is already best understood as fundamental only for equal protection purposes--it wouldn't make a difference in the Rowan County case. Whichever kind of fundamental right marriage is, state laws that treat some people differently from others with respect to marriage trigger strict scrutiny.
All such laws? Well, no. As the Court said in Zablocki:
Certainly there are a whole range of unimportant variations. If some counties issue licenses on 8x11 paper while other issue licenses on 8x14 paper, no one would--and certainly no one should--care. Likewise, if in some counties the license is signed by the clerk while in others it is signed by the deputy clerk, it is hard to see any harm. And even if there is some small symbolic harm, if the reason for the variation is to accommodate the religious beliefs of the clerks, that would seem to suffice.
At the other end of the spectrum, suppose that the clerk takes the view that her religion requires her to ensure that any same-sex licenses issued from her office bear some mark of disapproval from her (even if they do not contain her name), and that she chooses to implement this policy by confiscating the regular license forms and substituting for them license forms that are scrawled in green crayon on soiled paper towels. Even assuming such licenses would be valid under state law, and even assuming that both same-sex couples and opposite-sex couples receive these ugly licenses, it strikes me that any couple issued such a license would have a plausible constitutional claim. The official disapproval of marriage in the county could be counted as significant interference with the decision to enter the marital relationship.
Thus I conclude with a question: How different are the hypothetical crayon-on-soiled-paper-towel licenses from the irregular license forms that Kim Davis has provided to her deputy clerk in Rowan County?
In a useful update on the Kim Davis case on Balkinization, Marty Lederman raises the question whether the improvised forms that are now being used to issue marriage licenses in Rowan County, Kentucky--insofar as they differ substantially from the forms used in other counties--violate the equal protection rights of applicants for marriage licenses in Rowan County, even if the forms suffice to make the ensuing marriages legal under state law. Marty asks (in his question 5):
Do the Davis-amended licenses violate the 14th Amendment, as plaintiffs suggest, even if they do not affect the legality of the resultant marriages under Kentucky law, and even if same-sex and opposite-sex couples are treated equally within Rowan County? The theory here presumably would be that Rowan County is symbolically disfavoring same-sex marriage by virtue of appending an effective asterisk, or "issued under protest" message, to all licenses in Rowan County (same-sex and opposite-sex alike).Marty then provides a cautious non-answer: "I'll need to think about this further if and when plaintiffs offer the argument. Cf. Palmer v. Thompson."
Palmer, recall, is the case in which Jackson, Mississippi closed all of its public swimming pools rather than integrate them. The Supreme Court, in a 5-4 ruling, held that the city did not thereby violate the Equal Protection Clause. Jackson, after all, was under no obligation to establish or operate public swimming pools, and so by closing them for people of all races it did not discriminate on the basis of race.
While Marty thinks about Palmer, let me suggest that there is a potentially crucial difference between swimming pools and marriage licenses--namely, there is, under the SCOTUS cases, a fundamental right to marry but no fundamental right to swim in a government-operated pool. The cases recognize that even when the government does not draw any distinctions based on any illicit criteria, discriminatory classifications with regard to fundamental rights trigger strict scrutiny.
The relevant case law includes two categories of fundamental rights: (1) those that are recognized as fundamental for substantive due process purposes are, ipso facto, also fundamental for equal protection purposes, so that distinctions drawn regarding their exercise must satisfy strict scrutiny; and (2) rights that are fundamental only for equal protection purposes, such as the right to vote. A state need not have an elected attorney general at all--and thus there is no substantive due process right to vote for the office of attorney general--but if the state does have an elected attorney general, then inequalities in the distribution of the franchise with regard to attorney general elections are strictly scrutinized.
One fair objection to Obergefell v. Hodges--and indeed to the entire line of the Supreme Court's right-to-marry cases--is that the right to marry ought to be the second kind of fundamental right. Under this approach, a state would not be obligated to create a civil institution of marriage at all, but if it does, then inequalities with respect to marriage are strictly scrutinized. In this view, Obergefell is rightly decided as an equal protection fundamental right case, because all of the states did have an institution of marriage and their reasons for denying marriage to same-sex couples were barely rational, much less compelling; but if marriage is the second kind of fundamental right, then Justice Kennedy and the rest of the Obergefell majority oughtn't to have relied on the Due Process Clause. Marriage would be a fundamental right "in" equal protection, just as voting is.
One obvious advantage of this approach would be that it would answer an objection raised by the dissenters: The Constitution, they said, generally protects negative rights against state interference, not affirmative rights to state recognition or assistance; yet the marriage right looks like the latter; when the government denies recognition to a same-sex couple's marriage (or to an opposite-sex couple's marriage), it does not do anything to them; it fails to do something for them.
That is a sound objection in principle but it is not fairly lodged against the Obergefell majority. The Court in Loving v. Virginia quite clearly rested its holding on the alternative ground (endorsed by eight justices) that, in addition to the equal protection violation occasioned by the use of a race-based classification, the Viriginia laws at issue were invalid because they "deprive[d] the Lovings of liberty without due process of law in violation of the Due Process Clause of the Fourteenth Amendment." Zablocki v. Redhail also relied on a substantive due process right to marry as well as equal protection. (The other main right-to-marry case, Turner v. Safley, arose in the prison context, so the standard for evaluation was somewhat different, although even then, the Court found the infringement on the right to marry unconstitutional.)
Although it would be an overstatement to characterize the key portions of the cases recognizing a substantive due process right to marry as mere dicta, it is true that in each of the cases equal protection would suffice for the result. Thus, critics who think that there is something highly problematic about a substantive constitutional right to government recognition (as opposed to non-interference) could argue in some future case that the right to marry ought to be reconceptualized as simply a type-2 fundamental right: one that, like voting for state AG, cannot be distributed unequally if the state recognizes it, but that the state need not recognize at all.
Despite efforts in Oklahoma and elswhere to eliminate civil marriage, it seems unlikely that a test case will actually arise in which the Supreme Court will be asked to consider "demoting" marriage from a fundamental right for substantive due process purposes to a fundamental right only for equal protection purposes. But even if that were to happen--and even if one were to think that marriage is already best understood as fundamental only for equal protection purposes--it wouldn't make a difference in the Rowan County case. Whichever kind of fundamental right marriage is, state laws that treat some people differently from others with respect to marriage trigger strict scrutiny.
All such laws? Well, no. As the Court said in Zablocki:
By reaffirming the fundamental character of the right to marry, we do not mean to suggest that every state regulation which relates in any way to the incidents of or prerequisites for marriage must be subjected to rigorous scrutiny. To the contrary, reasonable regulations that do not significantly interfere with decisions to enter into the marital relationship may legitimately be imposed.So, what kinds of inter-county variations in the form of a marriage license either don't trigger any heightened judicial scrutiny or, if they do trigger such scrutiny, readily satisfy it?
Certainly there are a whole range of unimportant variations. If some counties issue licenses on 8x11 paper while other issue licenses on 8x14 paper, no one would--and certainly no one should--care. Likewise, if in some counties the license is signed by the clerk while in others it is signed by the deputy clerk, it is hard to see any harm. And even if there is some small symbolic harm, if the reason for the variation is to accommodate the religious beliefs of the clerks, that would seem to suffice.
At the other end of the spectrum, suppose that the clerk takes the view that her religion requires her to ensure that any same-sex licenses issued from her office bear some mark of disapproval from her (even if they do not contain her name), and that she chooses to implement this policy by confiscating the regular license forms and substituting for them license forms that are scrawled in green crayon on soiled paper towels. Even assuming such licenses would be valid under state law, and even assuming that both same-sex couples and opposite-sex couples receive these ugly licenses, it strikes me that any couple issued such a license would have a plausible constitutional claim. The official disapproval of marriage in the county could be counted as significant interference with the decision to enter the marital relationship.
Thus I conclude with a question: How different are the hypothetical crayon-on-soiled-paper-towel licenses from the irregular license forms that Kim Davis has provided to her deputy clerk in Rowan County?