The Yale Law Journal


Toward Internal Separation of Powers

26 Oct 2006

The standard American conception of separation of powers presumes three branches of government, each replete with ambition to maximize its power. But due to a complicated interplay of party dynamics and executive branch assertiveness, Congress has often been content to stay at the sidelines of regulating the executive. Particularly when it comes to legal issues in the aftermath of the horrible attacks of September 11, 2001, Congress has passed open-ended legislation that fails to check the executive or has passed no legislation at all. The result is an executive that subsumes much of the tripartite structure of government.

Many commentators have bemoaned the current state of affairs. My essay—and the larger project of which it is a part—begins where others left off. If major decisions are going to be made by the President, or delegated to him via open-ended legislation, then perhaps we should begin seriously thinking about implementing the “separation of powers” principle within the executive branch.

I argue that the first-best concept of “legislature v. executive” checks and balances must be updated to place more emphasis on second-best “executive v. executive” divisions. And, instead of doing away with the unitary executive, I propose institutional designs that impose internal checks but permit temporary departuresfrom themwhen the need is great.

We can start by trying to resuscitate a dirty word in American politics today: bureaucracy. The structural advantage of bureaucracy is its empowerment of a group of long-term players who accrue expertise in particular subjects. However, for bureaucracy to perform the type of checking function I envision, at least two things need to happen.

First, bureaucracies should be set up in partial competition with one another, with overlapping jurisdictional mandates. By placing individuals charged with similar duties in separate executive structures, different viewpoints emerge. These different viewpoints, in turn, allow agencies to function more like laboratories, devising new solutions to new problems. When the State and Defense Departments have to convince each other of why their views are right, for example, better decision-making results. Similarly, the existing rivalry between the Department of Justice’s Antitrust Division and the Federal Trade Commission can promote smarter enforcement of our antitrust laws.

Second, we need to transform bureaucratic service into a more attractive career option. While top graduates of France’s premier educational institutions compete for civil service positions in large numbers, America’s top graduates more frequently vie for political appointments to supervise agencies, not jobs in the agencies themselves. At least one American institution, however, works differently: the Foreign Service. The Foreign Service rewards employees, discourages spoils, and creates extensive opportunities for promotion through its “up or out” system, whereby management must either be promoted or exit the Service after a term of years.Perhaps most impressively, the State Department has a “Dissent Channel” through which employees anywhere in the world can send a cable to the highest reaches of the State Department if they disagree with a policy decision. Rather than suffering retaliation, individuals who use the channel are often promoted. The result of all of this is, comparatively speaking, a Foreign Service with skilled employees who are motivated to do their jobs.

Unfortunately, the civil service today cannot make the same claims. While civil service reform has been a longstanding goal since the 1883 Pendleton Act, the truth is that the civil service today is dispirited. The growth of “Schedule C” political appointees, combined with executive branch policies and academic theories that emphasize the agencies’ political accountability rather than their expertise, have contributed to the demise of bureaucracy’s checking function. That checking function also requires sufficient job protection so that civil servants do not fear retaliation when they blow the whistle and expose wrongdoing.

Overlapping bureaucratic mandates and career protection for civil servants can only accomplish so much. With bureaucratic friction comes the need for an overarching decision-maker. With respect to legal debates, the Department of Justice’s Office of Legal Counsel (“OLC”) is supposed to act as the neutral decision-maker. But the OLC’s two roles as adviser to the President and adjudicator of legal issues are in fundamental tension. After all, no one wants an advisor who has ruled against him as an adjudicator. The result is an OLC that is too tempted to tell the President what he wants to hear.

Over the last five years, the OLC’s legal judgment has been atrocious. In a November 2001 opinion that still has not been released to the public or Congress, the OLC stated that the President’s proposed use of military commissions to try alleged terrorists was legal—but the Supreme Court strongly disagreed. In a different memo, the OLC claimed that federal courts lacked habeas corpus jurisdiction over detainees held at Guantanamo Bay—yet the Supreme Court said that was wrong, too. The OLC also issued a legal opinion stating that the Geneva Conventions’ most rudimentary requirements of the laws of war, embodied in “Common Article 3,” did not apply to the war on terror—but the Supreme Court rejected that implausible claim as well. The OLC evidently said that American citizens could be detained indefinitely without due process—but again, the Supreme Court found that assertion wrong. The OLC has also poorly analyzed the Constitution’s restrictions on the executive’s power to conduct domestic electronic surveillance of Americans and its power to torture detainees. It might be said that the only post-September 11 legal issues on which the OLC has not been proven wrong are those it has so far managed to keep out of the courts.

It is simply implausible to assert that the repeated repudiation of these OLC positions is due to an “activist” Supreme Court. Because the Court includes seven Justices appointed by Republican Presidents, that’s a tough claim to sustain. And, curiously, Bush Administration officials have been complaining for years about the activism of the federal judiciary—a claim that, if they believed it, should have led the OLC to be far more cautious about reaching its legal positions. After all, if judges are activist, then they are more likely to scrutinize the government’s claims, and that should induce a modicum of caution in the OLC’s legal decision-making. Instead, however, the Administration decided to adopt wild-eyed theories that bore little relationship to the traditions of this country or even to the academic theories—such as the unitary executive—that they claimed supported their positions.

On legal questions such as those posed by the aftermath of the horrible September 11 attacks, agencies are bound to have differing views. Secretary of State Colin Powell, for example, was tremendously opposed to the OLC’s interpretation of the Geneva Convention, whereas other members of the Administration supported it. Such disagreement is normal and healthy, for one function of agencies is to push the law forward and to avoid ossification, particularly in national-security matters. But when that push is made, the agencies need somewhere to go to voice their opinions and seek adjudication. And that adjudication has to come from a neutral decision-maker, instead of a pep squad for one side masquerading as a quasi-judge.

A crucial function of the OLC is to resolve inter-agency legal disputes. In performing this and other functions, the OLC prides itself on its independent judgment and expertise. When its high-ranking officials become advocates for one side, as they did in many of the recent relevant decisions, the system breaks down. The decisions of that office begin to look suspect, resembling a courtroom guided by political influence rather than the law. Yet the political pressure on OLC officials is unavoidably immense. They are, after all, political appointees themselves—the head of the office and all its deputies are politically appointed. They are expected to advise the President, rather than merely adjudicate disputes, and are regularly present at White House meetings. In this climate, there is simply no way that the OLC’s aspirational purpose of acting as a neutral decision-maker can play out in practice the way one would hope. Simply put, the OLC is composed of lawyers with a client to serve.

For these reasons, I believe that the OLC should be stripped of its adjudicatory role and permitted to function only as an adviser to the government. The adjudication function would be transferred to a separate official, a Director of Adjudication, who would resolve inter-agency disputes. The Director’s decisions would be subject to presidential override.

Several other changes could create structural incentives that would force deliberation by the executive and reduce the current concentration of power that has been deposited in that branch: reporting requirements on agency action, hearings held by the minority party in Congress, and judicial doctrines that prod agency action. Reporting requirements create an information flow to Congress that promotes external checks. Similarly, during periods of single-party government, minority-party hearings would produce modest constraints on domineering executives and ensure better information flow. Courts can also jump-start the process of creating internal checks and balances by choosing not to defer to presidential interpretations of international law, such as the Geneva Conventions, unless those interpretations are the product of internally checked processes.

My point is to highlight the ways in which separation of powers can become laced into the executive branch itself, rather than hoping that it will rematerialize in Congress and the courts. Each of those actors has limitations, the former due to the rise of the party system, the latter due to well-designed institutional limits on jurisdiction. The pendulum today has begun to swing so far toward executive branch vigor that one must fear that the principles of divided government that our Founders embraced are no longer working. By giving force to traditions that are already part of our subtle constitutional landscape—bureaucratic overlap, civil-service protections, internal adjudication, and reporting requirements—some of this movement can be pulled back toward equilibrium.

Neal Katyal is Professor of Law, Georgetown University, and served as lead counsel in Hamdan v. Rumsfeld.

Preferred Citation: Neal Kumar Katyal, Toward Internal Separation of Powers, 116 Yale L.J. Pocket Part 106 (2006),