Windsor: Encouraging Constitutional Change, Not (Just) Clearing the Channels of Political Change

Neil Siegel

In recent posts describing a new article, Heather Gerken offers an account of the U.S. Supreme Court’s decision in United States v. Windsor that draws from John Hart Ely’s theory of judicial review. Gerken contrasts her “internalist” account of Windsor with “psychoanalytic” ones offered by scholars such as Rick Pildes, Michael Klarman, Mary Dudziak, and myself. Gerken describes her work as reflecting a distinctive focus on what Justice Kennedy’s majority opinion in Windsor “actually says,” as opposed to divining what it may portend doctrinally.

There is a more accurate way to characterize our disagreement. Gerken offers a process account of the majority opinion in Windsor, according to which the Court “dislodge[ed] an outdated consensus at the national level” about the status of gay people, thereby “ensur[ing] that the interlocking gears of our democracy—rights and structure—were free to move without committing to them moving in a particular direction.” In contrast, I view the Windsor Court as accomplishing something more substantive than just clearing the channels of political change. I view the Windsor Court as encouraging—but not yet requiring—citizens and courts to secure marriage equality.



In Winsdor and Hollingsworth v. Perry, the Court invalidated federal law’s limitation of marriage to opposite-sex couples and declined to decide the constitutionality of California’s ban on same-sex marriage. In the wake of those June 2013 decisions, I suggested in an article that the Windsor opinion leaned in the direction of marriage equality but ultimately resisted domestication by disciplined legal analysis. It was neither the full-throated “equality” opinion that Justice Scalia feared (with the implication that state bans on same-sex marriage were doomed), nor the straight-up “federalism” opinion that Chief Justice Roberts hoped (with the implication that state bans were unaffected).

In seeking to understand why Justice Kennedy’s opinion seemed to preserve for itself a Delphic obscurity, I suggested that the opinion is best read as an exemplar of what judicial opinions may look like in transition periods, when the Court may seek to invite, not to end, a national conversation, and to nudge it in a certain direction—“persuading before it attempts to coerce,” as Alexander Bickel put it in 1962. In such times, federalism reasoning and rhetoric—like declining to announce the level of scrutiny in Windsor and appearing to misapply the justiciability doctrines in Hollingsworth—may be used as a way station toward a particular later resolution.

In American politics, federalism may serve a temporizing function—it may be invoked by politicians who do not want to take a decisive national stand on a divisive issue for the time being. In the years before the Civil War, for example, Senator Stephen Douglas pushed a “federalism” solution to the explosive constitutional problem of federal regulation of slavery in the territories. (His “popular sovereignty” approach sought to stop the spread of slavery in what he thought would be the least politically controversial way: defeat at the polls.) More recently, President Obama initially framed debates over same-sex marriage in terms of respect for local politics, only to eventually describe the issue as one of national citizenship and equal rights.

So, too, in Windsor, the Court seemed to be using the analytical and rhetorical resources of federalism both as a Bickelian passive virtue and as an enabling device—as a way station toward an eventual declaration of marriage equality. Such an interpretation allows one to take seriously the federalism language in the majority opinion without embracing the Chief Justice’s “federalism” reading.

According to the Chief Justice’s interpretation, the Court inferred animus against gay people primarily or exclusively from Congress’s unusual interference with the traditional authority of the states to regulate marriage. From the internal point of view of the faithful participant in the practice of constitutional adjudication, there are several problems with this interpretation.

First, such a federalism analysis was logically unnecessary. Equal protection and/or due process reasoning sufficed to support the Court’s holding that Section 3 of DOMA violated the Fifth Amendment.

Second, and more importantly, such a federalism analysis is unpersuasive on its own terms because it lacks substance neutrality. The Court would not have inferred animus from Congress’s unusual interference with the traditional authority of states to regulate marriage if Congress has defined “marriage” to include same-sex couples. (Imagine a federal Defense of Marriage Equality Act, or DOMEA.)

Third, and equally importantly, such a federalism analysis is unlikely to decide any future cases. It is hard to think of another example in which federalism concerns render unconstitutional an otherwise permissible law that adversely affects equality or liberty interests.

Those legal considerations lead to my article’s conclusion that Windsor’s federalism analysis and language were doing work other than providing stable federalism doctrine going forward. Like declining to announce the level of scrutiny in Windsor and declining to reach the merits in Hollingsworth, the federalism reasoning and rhetoric in Windsor allowed the Court to avoid requiring lower courts to invalidate all state bans on same-sex marriage at that time.

By leavening its opinion with federalism, the Windsor majority avoided coercing lower courts to strike down state bans even as it allowed and even encouraged them to do so. The Court encouraged lower courts to interpret Windsor broadly by emphasizing—repeatedly—that the federal law’s discriminatory definition of marriage had the purpose, effect, and dominant social meaning of re-enforcing the inequality of same-sex couples and their children by excluding them from an institution that is not inherently unsuited to their inclusion. It does not take much imagination to apply the same reasoning to state laws.

The Court also encouraged courts to invalidate state bans by celebrating the then-distinct minority of states that allowed same-sex couples to marry while ignoring the large majority of states that prohibited same-sex marriage. For example, the Court commended voters in New York—but not, say, voters in North Carolina—for acting on “the community’s considered perspective on the historical roots of the institution of marriage and its evolving understanding of the meaning of equality.”

In her Foreword in the Harvard Law Review, Reva Siegel (no relation) also has emphasized the equality-infused portions of Kennedy’s opinion. So has Russell Robinson in his forthcoming article in the Stanford Law Review. Gerken does not adequately account for those features of Kennedy’s opinion, which also are part of what the opinion “actually says.” Instead she writes that a “process account [explains] why Kennedy was willing to invalidate a federal law without saying anything, either way, about the constitutional status of similar state laws.”

But the Court’s emphasis on equality strongly suggests that it was not merely clearing the channels of political change, as Gerken asserts. Rather, the Court was encouraging citizens and courts to make a particular kind of constitutional change, as the Court’s equal protection and/or due process holding likely will confirm later this year.

Gerken, like Ely before her, offers a process account that misses the substance of process.
Posted 2:05 PM by Neil Siegel [link]