This year the Supreme Court prohibited the Trump Administration from undoing the Obama Administration’s “DACA” policy without first providing a more thorough justification; a year earlier it prohibited the Trump Administration from adding the citizenship question without first providing a more candid justification. The common line connecting the cases is clear, and it suggests a trend of slight reinvigoration of judicial review of agency action. But there is another connection worth drawing, with a case that might seem very different: this year’s House subpoenas case. 

The DACA and Census cases are by now well known. In Department of Homeland Security v. Regents of the University of California, the Court did not prohibit the Trump Administration from rolling back DACA per se; it simply held that to do so would require—as in any action subject to the Administrative Procedure Act—a reasoned explanation for doing so. But the Administration had simply failed to provide any reasoned analysis of some of the most salient issues implicated by the agency’s action.

And in Department of Commerce v. New York, the Court did not prohibit the Trump Administration from restoring the citizenship question per se; it held that to do so would require, again, a reasoned explanation for doing so. But while arbitrary-and-capricious review is normally deferential, the Court “cannot ignore the disconnect between the decision made and the explanation given.” The Trump Administration’s “rationale—the sole stated reason—seems to have been contrived.” And “[a]ccepting contrived reasons would defeat the purpose of the enterprise. If judicial review is to be more than an empty ritual, it must demand something better than the explanation offered for the action taken in this case.”

This week’s House subpoena case, Trump v. Mazars USA, is also well known, but its prominent holding against the President—namely, that the president does not enjoy categorical privilege against congressional subpoenas for his files—overshadows the Court’s equally blunt assessment of the House’s own explanation and motives. In a case where both sides invoked sweeping powers not actually granted explicitly by the Constitution, both sides were denied outright victory. And the Court’s rejection of the House’s sweeping claims of power is noteworthy.

In rejecting the House’s categorical assertions of subpoena power, the Court called upon the House to offer a much more thorough justification for its subpoenas, a la the DACA case. As the Court observed, any exercise of Congress’s subpoena power requires a demonstrable connection to “a legitimate task of Congress”—a phrase that reminds us that, in this context, actions by Congress are not self-evidently “legitimate.” Rather, the task’s legitimacy has to be justified—and, in the context of subpoenas for presidential papers, the Court framed justification in terms of burdens imposed on the president and the particular interests asserted by the Congress. The Court contrasted legitimate purposes with “impermissible” purposes—and, with reference to the subject at hand, observed bluntly that a president’s personal papers have a “less evident connection to a legislative task” sufficient to justify the subpoena. (That part of the Court’s analysis stands in particularly stark contrast to the D.C. Circuit’s much less demanding analysis; it will be interesting to see how the D.C. Circuit handles those matters on remand.)

In describing the more detailed and weighty justifications that the House would need to provide to justify its subpoenas, the Court stopped far short of the colorful language with which it characterized the Administration’s own justifications in the Commerce Census case. But while the Court didn’t go so far as to describe the House’s rationale as “pretextual,” it did state candidly its assessment that at least one part of the House’s rationale strained credulity: to the House’s assertion that these subpoenas do not actually implicate presidential interests, the Court quipped that “[w]e would have to be ‘blind’ not to see what ‘[a]ll others can see and understand.’” (It might as well have added, “this wolf comes as a wolf.”)

* * *

We are accustomed (to say the least) to seeing courts demand reasoned and candid explanations for agencies’ actions. We are far less accustomed to seeing courts demand such explanations for legislatures’ actions. As the Court observes from time to time, Congress is not required to articulate the reasons that motivated its enactment of a given statute.

But here, of course, the legislative action is not the enactment of a statute but the enforcement of a subpoena, and therein lies the difference. To the extent that Congress attempts to exercise power not through the Constitution’s expressly granted legislative powers, but rather through “oversight” activities, Congress cedes the discretion and power that it otherwise would enjoy as the first branch of constitutional government.

This is not to say that Congress should not conduct oversight of the rest of government; far from it. (As I observed in December, one of the major problems of governance facing us today is that the Senate, for example, seems far too little interested in its promoting steady administration, a role for which the Senate was originally created.)

But it is to say that “oversight” responsibilities should be seen for what they are: an implicit adjunct to Congress’s specific powers: to legislate substantive laws and wield Congress’s “power of the purse“; to grant or withhold the Senate’s “advice and consent” to officer appointments; to impeach; and so on. “Oversight” came to be seen not an adjunct to Congress’s express constitutional powers but as Congress’s foremost raison d’être; and at the same time, Congress grew much less interested in the work of regular legislation, appropriations, and executive-branch appointments—that is, the legislative powers that Congress itself could leverage to make its oversight more effective. Instead, modern “oversight” is enforced not with the threat of legislative sanction, but judicial sanction. In short, Congress made its own power largely dependent upon either executive goodwill or judicial intervention.

Hence the result in Mazars. By calling upon federal judges to settle its oversight dispute with the President, the House necessarily ceded to the courts the power to sit in judgment of the House’s own justifications; the House could not simply rely on categorical assertions of power or dubious rationalizations. You take the bitter with the sweet.

While the executive branch is already well familiar with the duties of reasoned explanation and candor that the courts impose pursuant to the Administrative Procedure Act, we would do well to step back, from time to time, to consider how at odds such duties of explanation seem for the part of government that was originally described in terms of “decision, activity, secrecy, and despatch.” When the work of the executive branch came to resemble legislation or adjudications more than classical execution, then the executive branch found itself saddled with procedural and substantive rules and norms befitting a legislature or judiciary. (The APA vindicates many values, but it is hard to count “activity, secrecy, and despatch” among them.)

In Trump v. Mazars, and other litigation surrounding congressional oversight, we see something similar happening to Congress. The modern Congress is far less interested in legislating or exercising other expressly enumerated powers, and far more interested in the nebulous work “oversight” ever more removed from the real work of legislating, appropriating, and granting (or withholding) Senate advice-and-consent. The more that congressmen dedicate their energy to playing roles more akin to prosecutors or investigators than legislators, the more that their work will be judged accordingly.

Adam J. White is a resident scholar at the American Enterprise Institute, and an assistant professor at George Mason University’s Antonin Scalia Law School, where he directs the C. Boyden Gray Center for the Study of the Administrative State.