Enforcing (but Not Defending) Unconstitutional Laws

When should the executive decline to defend in court a federal law it has determined to be unconstitutional, yet still enforce that same statute against third parties? The question is prompted by the Obama administration\u27s decision to enforce, but not defend in federal court, Section 3 of the Defense of Marriage Act ( DOMA ). But the DOMA Section 3 decision is not the first time the executive has bifurcated the enforcement of a statute from its defense before the bench. The practice of enforcement-litigation gaps dates back at least to World War I. Commentators tend to judge the practice by focusing on the merits of each enforcement-litigation gap but remain inattentive to its systemic effects. This Article sidesteps debate on specific cases, such as the DOMA Section 3 decision. It instead develops a default rule as a guide for executive branch practice. To that end, it analyzes the question whether a conscientious executive branch lawyer should view enforcement-litigation gaps as presumptively acceptable (and hence available for use) or presumptively disfavored (such that an especially compelling argument must be made to justify its use in each case). As a threshold matter, conventional wisdom views enforcement-litigation gaps as a kind of departmentalism, and either condemns or endorses the practice on that basis. But the Constitution does not categorically allow or prohibit independent executive branch judgment on matters of fundamental law. Working within this zone of executive branch discretion, this Article analyzes enforcement-litigation gaps in terms of their effect on constitutional values such as legality, accountability, and public confidence in the Constitution. This analysis suggests that the desirability of enforcement-litigation gaps turns on what sort of constitutional question is at stake. The practice is presumptively justified when the executive defends an Article H value, but rests on weaker ground when the constitutional rights of individual third parties are in play. The government therefore ought to be presumptively willing to use the practice with respect to structural issues, and presumptively unwilling to use it in individual rights cases