Another “Crisis” in Civil-Military Relations?
What should be the role of Defense Department civilians below the secretary of defense in policymaking? During James Mattis’s tenure as the secretary of defense, senior civilians reported that they felt bypassed in the decision-making process and that their responsibilities were taken over by senior military officers. In November 2018, a Foreign Policy article quoted one former official that civilian control of the military “was already weakening in the last administration, and I think it basically fell off a cliff.” A congressionally mandated commission and commentary by former Defense Department officials echoed these concerns.
To what extent are these reported changes really a problem and what to do about them? We assess that the reports of the military’s behavior in dealing with top civilians are — arguably — compatible with current regulations, and largely on par with past policy processes. Civilian control over military actions, at least as legislated by Congress, does not appear at risk. Nevertheless, declining civilian input is a concern for good policymaking. Changes in legislation or regulation may be needed to bolster civilian input, although before such steps are taken, the defense community needs to understand better the evolution of defense civilian input into policy processes since the passage of the Goldwater-Nichols Act.
How Is It Supposed to Work?
By civilian control of the military, we mean that a specified, politically accountable civilian authority has the final say on national security and defense policy. Civil-military relations refers to the wider set of interactions between civilians and military personnel. Relevant academic literature highlights four issues in civil-military relations: 1) curbing the political power of the military; 2) ensuring that the military acts to protect rather than endanger the state; 3) ensuring civilians do not use the military for partisan political goals; and, 4) solving the puzzle of how civilians can control the military and ensure military effectiveness even as they lack the specific knowledge and expertise of military officers. The current debate centers on the fourth issue: Amidst the understanding that civilians and military have a shared responsibility for the security of the country, the tension centers on interpreting rules and procedures in a way that is respectful of civilian and military roles and achieves military effectiveness while preserving civilian control.
As a component of the U.S. government, the Department of Defense is authorized and constrained by law. Personnel within the department draw on law and regulation, or at least their understanding of it, to determine their roles and responsibilities. While the academic literature is useful in evaluating performance and proposing remedies, we use the legal and regulatory framework as a baseline, especially on the narrow question of civilian control. The degree to which observed behavior is within or outside the legal and regulatory framework highlights where changes may be needed.
The basic structure and functioning of the department is specified by Department of Defense Directive 5100.01, Functions of the Department of Defense and Its Major Components, issued by the secretary of defense and based on the last major defense reform legislation, the Goldwater-Nichols Department of Defense Reform Act of 1986. The directive describes the components of the department. The Office of the Secretary of Defense, led by the deputy secretary of defense, acts as the “principal staff element of the Secretary of Defense.” The Joint Chiefs of Staff is led by the chairman of the Joint Chiefs of Staff, and supported by the Joint Staff. The military services are responsible for recruiting, training, and equipping military forces. The combatant commands are regional and functional commands who exercise “authoritative direction over all aspects of military operations, joint training, and logistics.”
Some of the structure and functions of the department are specified in Goldwater-Nichols, although the legislation leaves flexibility, such as by empowering the secretary of defense to define the chairman’s role, as is done in the directive on functions. The current version of the directive dates back to 2010 and was issued by Robert Gates, but the basic organization articulated in the directive has not changed greatly since 1987.
As specified in Goldwater-Nichols, the directive on functions delineates that the operational chain of command goes from the president to the secretary of defense to the combatant commands. The secretary therefore exercises civilian control over the military in the sense of being the only civilian other than the president with the authority to issue orders to military personnel. The directive stipulates that all communication between president or the secretary of defense and the combatant commands, including military orders, is normally required to be transmitted through the chairman. Indeed, the chairman of the Joint Chiefs of Staff is responsible for overseeing the activities of the combatant commands, and acting as their spokesperson, “especially on the operational requirements.” Furthermore, the directive states that communication between other organizations and the combatant commands “normally shall be coordinated with” the chairman.
Another statutory role for the Joint Chiefs of Staff is to provide their advice to both the president and the secretary of defense, first as members of the Joint Chiefs of Staffs and second as representatives of their respective services. While recent commentary calls out the military commanders for “often preempt[ing] the advice and analysis of civilian staff by sending their proposals straight to the secretary of defense,” the functions directive effectively permits the chairman to bypass civilians in the Office of the Secretary of Defense in providing military advice. In principle, these civilians could also present their advice to the secretary without inputs from the chairman, although standard procedures generally require that they at least attempt to coordinate with the Joint Staff.
To look deeper at the issue of the alleged declining role of the civilians within the Office of the Secretary of Defense, we focus in particular on the role of the Office of the Under Secretary of Defense for Policy (hereafter referred to as “Policy”). Other civilian organizations in the department are no less important, but may have different dynamics. Policy has specified responsibilities and functions, such as issuing guidance and reviewing campaign plans and contingency plans as well as representing the department in interagency meetings and international defense negotiations. But direct oversight of the management of the use of force, setting requirements, and developing plans is executed by the military chain of command.
Indeed, while the directive on functions calls for the Office of the Secretary of Defense and the Joint Chiefs of Staff to “function in full coordination and cooperation,” it also enables the Joint Staff to act as gatekeepers between Policy and the combatant commands. In practice, officials in Policy may have routine informal discussions with their counterparts on the Joint Staff and combatant commands, including asking questions of military commanders and gaining visibility on military advice as it is being formulated. But in conditions of more antagonistic relations between Policy and the top military leadership and/or in cases where the top military leadership resorts to hardball bureaucratic politics, Policy civilians may not see military advice from the combatant commands until it has been approved by the chairman, by which point it has sufficient gravitas to make it difficult to question. As Alice Hunt Friend argues, the less information available to civilian leaders, the more that civilian preferences tend to align with those of the military.
It is no accident that legislation and Department of Defense regulations authorize military officers to offer their advice with little input from civilians below the secretary of defense. Senator Goldwater, one of the key architects of the 1986 reorganization, shared a common view that the U.S. failure in Vietnam stemmed from civilian interference in professional military decisions. The strengthening of the chairman of the Joint Chiefs of Staff and the combatant commands ensured that civilian officials do not prevent military leaders from providing their advice to the secretary or the president. The real question is whether the strengthening of the military role in the policy process in practice has evolved too far in the 35 years since the passage of Goldwater-Nichols.
How Has It Worked?
Since Goldwater-Nichols, the influence of civilians appears to have ebbed and flowed, although an accounting is challenging with available data. Ultimately, the details of these relationships are adjudicated in non-public internal Defense Department processes: who attended which meeting and what was discussed? Who did or did not see or coordinate on memoranda at a particular time? Accounts of the recent civil-military relations in the department rely primarily on journalistic narratives, heavy on unnamed sources, and offering broad generalizations of staffing processes, making it impossible to reach an objective assessment.
One example where there is a relatively detailed account is the 2009 debate about troop levels in Afghanistan. The events of this account are telling because the key protagonists — Secretary Gates and Under Secretary of Defense for Policy Michèle Flournoy — are identified as good practitioners of civil-military relations.
During the process that Gates describes, Policy officials, especially Flournoy, played an important role. But the military advice about troop numbers — the heart of the future policy in Afghanistan — was carried up through the military chain of command, with apparently limited debate by civilians until it reached Gates and the president. For example, in 2009, Gen. Stanley McChrystal was appointed as the new NATO commander in Afghanistan. Following a request for increased force levels above and beyond what President Barack Obama had reluctantly approved, Gates directed McChrystal to conduct a 60-day review of the situation in Afghanistan. McChrystal’s assessment, once complete, initially was provided only to his military superiors — Gen. David Petraeus and admirals Michael Mullen and James Stavridis. Petraeus endorsed McChrystal’s recommendations and then Gates describes how Flournoy became involved by discussing the process for considering the assessment within the National Security Council.
When Obama received McChrystal’s assessment, it had already been endorsed by McChrystal’s higher headquarters. Gates provided his opinion only in an “eyes only” memo to the president, rather than for internal defense department discussion. In the midst of a series of interagency meetings about McChrystal’s assessment, a copy was leaked to the press, and McChrystal made public comments that the White House interpreted as limiting the president’s freedom of action. Effectively, the military advice of the commanders set the stage for the president’s decision, with only the secretary of defense, rather than subordinate civilians, providing the primary feedback, and the key debate about the policy taking place above the level of the Defense Department, within the National Security Council.
Few participants in the policy process on Afghanistan are likely to cite it as a high point in U.S. civil-military relations. Even so, the process Gates describes provided for a greater role for civilians below the secretary, including through Flournoy’s participation and a greater deliberative process, than what is described to have transpired under Trump. The episode nevertheless indicates that limited defense civilian inputs to military recommendations are not a new phenomenon.
Contrary to the idea that an apparent increase in military roles inside the Department of Defense represents a break with prior patterns or a crisis in civil-military relations, an independent military role that sometimes excludes input from civilian officials below the secretary seems consistent with the letter if not the spirit of existing regulations and not far outside of past patterns of behavior.
This is not to say that recent observations of an expanded military role are of no concern. In the 2009 debate, questioning by civilians in the National Security Council might have improved the policy outcome on Afghanistan, and defense policymaking would probably be better served by incorporating civilian opinions earlier in the process.
While personalities are no doubt important, if a greater role for Policy or other civilians is desired over the long term, a change in law or regulations is required. The current secretary of defense, Lloyd Austin, faced questions about the decline in civilian inputs during his confirmation, and pledged to strengthen civilians’ roles. He has emphasized a return to “professionalism and normal order and good process.” While Austin’s tenure could lead to a shift back to greater civilian input, it could be context-dependent and it may not persist unless it is institutionalized with changes in regulations.
One potential change could be to amend the defense department directive on the functions of its major components. A revised version could require that information copies of military advice provided by the combatant commands to the Joint Chiefs of Staff also be provided to Policy. Another change could be to permit Policy to communicate with the combatant commands without first coordinating with the chairman or Joint Staff, so long as the Joint Staff is aware and the communication is appropriately caveated. Offices in Policy may already regularly communicate with the combatant commands with the awareness of their counterparts in the Joint Staff. In those cases, a change to the formal rules toward more information-sharing would pose little challenge and it would prevent civilians from being cut out in the future.
But caution is in order, as any change in the department’s organization could have unintended consequences. Despite all of the writings on U.S. civil-military relations, there is little broad assessment of the typical interactions between senior officers and civilians. Judgements in the current debate seem to focus on a few incidents based on partial views from anecdotes and memoirs. There are also additional factors that are worthy of further study prior to any reform.
First, there is a need to understand to what degree Policy civilians have weighed in on different issues over time. Ideally, this would involve a review of documents, supplemented with interviews with a wide range of participants in these processes. Important questions include the frequency with which military advice is developed and presented to the secretary without inputs from Policy, and the extent to which there is a collaborative relationship between Policy and the combatant commands. Such an analysis would need to go hand-in-hand with an evaluation of why past practices evolved. Current practices are likely the result of changes in the functions directive and other formal rules, as well as shifts in informal, widely understood practices.
Second, there is a highly normative question of what should be the proper role of Policy civilians vis-à-vis top military leadership. Karlin and Schulman observe that civilian input can help to provide “collaborative friction” — “Departmental debate is healthy, and if one portion of the building stovepipes their advice on the way to Secretary, such debate is stifled.” Karlin also writes that “Without a capable and informed staff, no individual has the wherewithal to do the job decently, much less effectively.” The question is, to what degree is civilian control or policymaking by the secretary weakened by an unempowered civilian staff if the secretary still has an empowered military staff.
Answering this question must consider the unique contributions of civilians in the Department of Defense. Political appointees play a role in communicating the policy of the elected administration and ensuring that military actions are linked with and commensurate to political goals. Civilians in the civil service do not rotate as frequently as their military counterparts and consequently provide institutional memory. Defense civilians also may have advantages in engagements with civilian organizations, including Congress, State Department, and civilian components of the intelligence community. Finally, civilians are not obligated by the military chain of command. Once a senior officer in the chain of command has decided on a recommendation, other officers may be reluctant to openly question such decisions. Civilians may have an easier time asking hard questions to the combatant commanders or other senior officers.
Recognizing the potential for civilian contributions, a third issue is how to improve the quality of Policy’s inputs. Given greater power, civilians have a greater responsibility to provide good advice. As with any organization, a key challenge is staffing. If Policy is to present a civilian perspective, one criticism of current procedures is that a preference for hiring veterans may interfere with the ability of Policy to recruit qualified civilians. Improving the diversity of Policy may also strengthen its impact.
A fourth consideration is tradeoffs on secrecy. One potential justification for excluding civilians during high-profile deliberations is concern about leaks. However, staff within military organizations could leak information as well — Gates relays an account that McChrystal’s assessment was leaked by his staff. Limiting the number of individuals involved in sensitive deliberations may be necessary. At the same time, senior leaders within Policy and some of their staff members must be read into sensitive deliberations to enable them to fully participate.
With these inputs, and ideally an ability to link possible changes to military effectiveness, more persuasive arguments about reforming Department of Defense procedures can be made. Decision rights theory provides one way to proceed. Applied to the defense department, the theory recognizes that having the right to participate in all steps of the policy process does not mean that civilians will actually choose to participate in all the steps. For reasons of efficiency, they may delegate many aspects of policymaking — especially the initiation and implementation of policies — to the military, although they would retain the rights of ratifying policies and then monitoring their implementation. Some senior military officers may be skeptical of the need for the civilians to have a greater role, but understanding how civilian input can lead to better policy outcomes, as well as socialization of a new cohort of military leaders, may change their perspective.
Andrew Radin is a political scientist at the RAND Corporation. From December 2018 to December 2020, he was detailed from RAND to serve as a country director for Afghanistan in the Office of the Under Secretary of Defense for Policy. He is the author of Institution Building in Weak States: The Primacy of Local Politics, published by Georgetown University Press.
Thomas Szayna is a senior political scientist at RAND. His research has focused on strategic planning for the U.S. armed forces, the future security environment, and coalition interoperability. He is the lead author of the RAND report The Civil-military Gap in the United States: Does It Exist, Why, and Does it Matter?