Mightier Than the Sword
Civilian Control of the Military and the Revitalization of Democracy
Alice Hunt Friend



Why Study Civilians?

THE NEW PRESIDENT WANTED A general to be his secretary of defense. The nomination was unusual. By law, the country’s senior-most military officers were not eligible to serve as secretary of defense unless they had been retired from active duty for at least seven years.1 According to Title 10, the section of the US Code that regulates the armed forces, the leader of DoD had to be “appointed from civilian life.” When president-elect Donald Trump named retired Marine Corps general James Mattis secretary of defense, no president had asked Congress to make an exception to that law for six and a half decades.

If the nomination was unusual, the circumstances were unprecedented. Donald Trump had not served in the military himself, nor had he ever held public office. Not only that, but during the presidential campaign he made a number of statements that many foreign policy professionals deemed reckless. Jim Mattis, who served in the Marines for forty-one years, seemed like an “adult” who would steady the neophyte president. Military historian Eliot Cohen, appearing at a hearing before the Senate Armed Services Committee (SASC) on the question of issuing an exception to law so Mattis could be confirmed, said he hoped that “a Secretary Mattis would be a stabilizing and moderating force” in the Trump administration (SASC 2017).

In any case, not many people understood why nominating a recently retired military general to run the military was a problem. Civil-military relations scholars stepped into the public square to explain. “In healthy democracies,” political scientist Peter Feaver wrote in the New York Times, “the command authority is civilian.” He added that as a retired four-star general, Mattis would “never become fully civilian” because his values, assumptions, and personal relationships were too steeped in the Marine Corps (Feaver 2016). Meanwhile, the Congressional Research Service explained that the legislators who drafted the provision mandating a break between officership and political service believed it would “preserve the principle of civilian control of the military” (McInnis 2021, 8). Nevertheless, scholars agreed the circumstances warranted making an exception to the rule. Feaver himself endorsed the nomination because of the extraordinary nature of the untested president. So long as the Senate did not make such an exception for any other recently retired general officer for another seventy years, Cohen and others argued, secretaries of defense would continue to exercise sound civilian control over the military.

Four years later, president-elect Joseph Biden nominated retired Army general Lloyd Austin III to run DoD. That Biden picked another general so soon after Mattis’s nomination sparked controversy on Capitol Hill. “I supported a one-time waiver in the case of Secretary James Mattis with the belief that the circumstances at the time warranted a rare exception, not the establishment of a new precedent,” Senator Susan Collins (R-ME) told reporters (Steinhauer, Schmitt, and Broadwater 2020). Protests on the Hill and in the press were substantial enough that Biden made the rare move of publishing a written defense of Austin’s nomination in The Atlantic magazine: “Why I Chose Lloyd Austin as Secretary of Defense.” In the piece, Biden argued that although Austin’s primary experience was as a military leader, that same role had tested his diplomatic skills. Moreover, Austin had “served as a statesman” when he led the withdrawal operations from Iraq, proving he was “a true and tested soldier and leader” (Biden 2020).

The problem with Biden’s argument was that it didn’t acknowledge that being drawn from civilian life is normally a prerequisite for the role. Ignoring this fundamental gap in Austin’s professional experience blunted the argument that his personal qualities and time as a military leader were sufficient.2 Although Biden demonstrated his personal confidence in Austin, he did not grapple with the central questions before lawmakers: Had a career in the military become the best preparation for running the DoD?3 Were Mattis and Austin proving that the requirement for the secretary of defense to have a primarily civilian identity was no longer necessary? In the end, the Senate confirmed both men, but confusion lingered about the value of civilians in the civil-military context.

Americans weren’t always so fuzzy about the unique value of a civilian background for managing military affairs. The history of Americans’ approach to civilians roles in national defense reveals that generations of government leaders thought civilian control over military affairs was essential to self-rule. Yet the emphasis on civilian dominance of military policy has atrophied over time. Why?

Historical Trends for Civilians in Civil-Military Relations

Wariness of military power was one of the ideas that helped found the United States. In the prerevolutionary era, seventeenth-century English pamphleteers persuaded American colonists that large, permanent militaries posed a danger to society. One influential pamphlet authored by British parliamentarian John Trenchard warned that “unhappy Nations have lost that precious Jewel Liberty” because “their Necessities or Indiscretion have permitted a standing Army to be kept among them” (Trenchard 1697, 4).4 In other words, there was a correlation between a permanent army and government oppression. Under King George III, the colonists began to see this correlation for themselves. “The keeping of a standing army in several of these colonies,” the Continental Congress declared in 1774, “without the consent of the legislature of that colony in which such army is kept, is against the law.”5 The Declaration of Independence protested the king “affected to render the Military independent of and superior to the Civil power.” The king’s policies, according to military historian Richard Kohn, “made hatred of the standing army axiomatic in American politics” (Kohn 1991, 82).

Importantly, Americans did not blame the military itself for their oppression. It was the king who was threatening their liberty. The army was a mere instrument; the despot in possession of the army was the real problem (Bailyn 2017). “The time may come, when we may have to contend with the designs of the crown, and of a mighty kingdom . . . backed by a STANDING army,” read another colonial pamphlet (Dickinson 1768, 45).

This keen recognition of the source of danger was essential to the eventual structure of the American federal government. Because it was really overly powerful rulers that put freedom at risk, the American founders worried mainly about controlling civilians who commanded the military. If every politician or functionary could act like George III or Oliver Cromwell, then no civilian could ever be allowed to amass too much power, let alone the personalist loyalty of a full-time army (Bailyn 2017). “It is of great importance in a republic,” wrote James Madison in Federalist no. 51, “not only to guard the society against the oppression of its rulers; but to guard one part of the society against the injustice of the other part.”6 Early Americans were not just worried about guarding the guardians, but also about guarding the guardians’ guards. Civilian control of civilians was the paramount necessity. And those civilians could not have too large a military force at their disposal.

Centralized control over the military thus became a problem for the nascent American democracy. To constrain power-hungry civilians, the Founders divided control over military affairs into different civilian roles. Under Article I of the Constitution, Congress has the authority to “raise and support armies” and “provide for a Navy,” regulate the military forces it creates, collect taxes and allocate funds for national defense activities, declare war, define violations of international law, and “call forth the militia” and set standards for its training and equipping. Congress itself, being divided into two chambers, also cannot wield power as a unified group. Meanwhile, Article II makes the president “commander-in-chief” of the Army and Navy and of the militia when Congress activates it for national service. The division of labor was deliberate: one civilian body prepares for defense in peacetime and another leads in wartime, yet neither achieves its goals for long without the other. Without Congress, the president has no military with which to use force; without the president, Congress cannot actually launch whatever war it declares—at least not effectively. Civilian control of civilians would prevent a tyrant from total power over the military in the United States, but civilians would remain in overall control of the military instrument.7

The division of peacetime and wartime labor worked more or less as the Founders intended for a century and a half. In war, the size of the army ballooned rapidly as an emergency measure, and then shrank back to a small standing force in peacetime. Civilian control of civilians was most imperiled during these wartime episodes—most notably during the Civil War, when President Abraham Lincoln invoked controversial if effective military measures. Generally, however, the US maintained its constitutional system through major wars and kept to its “axiomatic” bias against large standing armies.

But in the years after World War II the United States changed its mind about a large, permanent military force. That period began a slow fading of the distinction between peacetime and wartime, and with it a steady shift in primary responsibility over controlling the military to the executive branch. To many influential Americans, the speed and destructiveness of modern warfare had obliterated the notion that it was safe to dismantle military capacity (Herring 1941). Long-range flight, missiles, and nuclear weapons collapsed the time envisioned in the Constitution for observation and political deliberation before military action. There had to be a permanent vigilance, a belief that drove an urgent need for structures able to ready forces, call upon industrial resources, and generally organize policy with speed and intelligence. The lessons of the war were also fresh in terms of military organization. Competition between the War (Army) and Navy departments had generated grave inefficiencies. Despite the modern sense that America’s triumph in WWII was inevitable, it felt much more like a near-miss at the time, and policymakers pushed for military unification into a single department to avoid further hair raising.

Regardless, Americans were as adamant as ever about the importance of civilian control over the military, but the introduction of a standing army shifted the balance of their anxiety away from the sense that civilians themselves needed to be controlled. For the first time, Americans had a military that truly had the capacity to upend democratic governance on its own (Hogan 1998). Americans in the 1940s worried about making civilians powerful enough to constrain this unprecedented military organization. For this reason, Presidents Truman and Eisenhower both insisted on centralized civilian control over the nascent DoD. When Congress passed the National Security Act of 1947—the law that unified the military services into a single department, among other changes—civilian preeminence over the entire institution was an explicit priority. In his letter to Congress supporting the law, Truman best summarized the era’s faith in centralized civilian control. Acknowledging the concern “that the concentration of so much military power would lead to militarism,” he assured Congress, “there is no basis for such fear as long as the traditional policy of the United States is followed that a civilian, subject to the President, the Congress and the will of the people, be placed at the head of this Department” (Cole et al. 1978, 13). As assistant secretary of war for air (and later secretary of defense), Robert Lovett asserted at one of the House hearings on military unification, it was “not just the theatre that needs unity of command, but the country as a whole needs it—at the top.”8 The effect of the debate was to minimize concerns about excess in a single civilian’s control over the military and to maximize anxiety about military subordination to civilians.

But crafting the secretary’s role to give the position full authority over military activities proved challenging. A series of amendments to the original 1947 law progressively amplified and centralized the secretary’s power over the military departments. In 1949 Congress expanded the secretary’s bureaucratic capacity by laying the foundation for a civilian staff, providing for a deputy secretary and three assistant secretaries, including a comptroller to oversee service budgeting (Stuart 2008). The 1949 amendment also clarified the all-encompassing nature of the secretary’s jurisdiction, stipulating that the secretary exercises “direction, authority, and control over the Department of Defense” (Cole et al. 1978). This latter phrase, the House Armed Services Committee (HASC) Chair Carl Vinson explained, was “the heart of this legislation.”

Direction means the act of governing, management, superintend[ing]. Authority means legal power; a right to command; the right and power of a public officer to require obedience to his order lawfully issued in the scope of his public duties. Control means power or authority to manage, to direct, superintend, regulate, direct[sic], govern, administer, or oversee. So under this law the Secretary of Defense is to have clearcut [sic] authority to run the Department of Defense. (Cole et al. 1978, 145–46)

Even this was not satisfying to President Dwight Eisenhower—himself a retired Army general who understood the difficulty of controlling military affairs. In 1958, Eisenhower wrote to Congress asking them to remove statutory authority for military operational command from “any official other than the Secretary of Defense” (Cole et al. 1978, 180). Eisenhower also informed Congress of his plans to create unified command structures, strengthen the secretary’s power over research and development, give the chairman of the Joint Chiefs of Staff (CJCS) and the secretary more control over military three- and four-star flag officer appointments, and give the secretary more budgetary “flexibility” (Cole et al. 1978, 177). In the final DoD reform of the period in 1958, the service secretaries were told to “cooperate fully with personnel of the Office of the Secretary of Defense” to ensure the secretary of defense’s authority would be carried out.9 That law also expanded the Office of the Secretary of Defense (OSD) once again, to include a new director of defense research and engineering (DDR&E) and assistant secretaries for legislative affairs and public affairs—all three moves designed to shift power from the services to OSD. By 1959, the secretary of defense had ten assistant secretaries, a general counsel, and the DDR&E.

To align itself with the new organizations in the executive branch, Congress also redesigned its own committee structure, combining the committees that had overseen the War and Navy departments into unified armed services committees. But in their determination to strengthen immediate civilian control over the newly sprawling defense enterprise, national leadership duplicated many congressional responsibilities in OSD, shifting the initiative for defense budgeting and much of the regulation of the armed forces from Congress to the DoD itself. Some members of Congress at the time expressed alarm about the rebalance of power. Senators Mike Mansfield and Paul Douglas worried that the law “clear[ed] the way for a major transfer of constitutional legislative powers and duties to the Executive Branch” (Stuart 2008, 226). Even Carl Vinson fought the 1958 reorganization because he feared it would cede too much authority to the president. In the HASC’s report on the law, he protested that “the organization of our national defense system must . . . at all costs, be the creature of our form of government—with the responsibility for national defense placed equally upon the President, as commander in chief, and the Congress.”10 But it was too late. In the span of just under twenty years, Congress ceded much of its predominance over military policy and budgeting to the DoD. Civilian control over civilians hadn’t been eliminated, but it had been handicapped.

Instead, much as Eisenhower had worried, the appropriate balance between civilian and military influence over decision making became a major point of contention. Once the DoD was established, the center of gravity for civilian control of the military increasingly fell between the secretary and the uniformed leaders of the military services—the Joint Chiefs of Staff. Civilians at DoD and the White House wrestled with how to include military leaders in the policymaking process without being overwhelmed by service parochialism. “The problem of the proper set-up of the Joint Chiefs of Staff,” Secretary of Defense Robert Lovett had written to President Truman in 1952, “involves the striking of a proper balance between civilian and military control.” Although civilian control of the military was “fundamental to our form of government,” he continued, “civilian judgment must be based on adequate military advice given by professional military men in an atmosphere as free as possible from service rivalries and service maneuvering.”11

At the time of the 1947 law and its amendments, lawmakers prevented unified military advice because they feared that the United States would duplicate the German “general staff” model, which they viewed as the engine of German militarism that sparked World War II. But the failed war in Vietnam made many in Congress conclude that the Joint Chiefs of Staff needed a stronger voice in the policymaking process. In 1980, the disastrous failed attempt to rescue American hostages in Iran and subsequent military operations in Grenada and Panama also spurred new thinking about military organization and the role of military advice in policymaking.

To reform the military advisory system, the Ronald Reagan administration’s Blue Ribbon Commission on Defense Management recommended that the CJCS become “the principal uniformed military advisor to the President, the National Security Council, and the Secretary of Defense” (Cole et al. 1978). The Goldwater Nichols Department of Defense Reorganization Act of 1986 implemented the recommendation, putting the CJCS in charge of spearheading strategic, contingency, and logistical planning, and developing joint doctrine and training for the services.

In contrast to the reforms of the 1950s, civilians—the secretary and OSD—received no new powers. It was not that the law’s proponents felt civilian control of the military at DoD was unimportant. It was simply that they did not think it needed reinforcing. As SASC staff wrote in their summary of the history of American civil-military relations, “the military has never posed a serious threat to civilian control in the United States.” Consequently, “fears that the U.S. military might threaten American political democracy are misplaced.”12 This declaration of victory for civilian control marked a turning point in practitioners’ perspective on civilians in American civil-military relations: In contrast to the post-WWII generation, government officials in the last decade of the Cold War weren’t concerned about reinforcing civilian preeminence. They had faith that civilians were suitably ensconced at the top of the military hierarchy (Locher 2002, 439).

More recently, Congress has moved to reduce the civilian voice at DoD. In 2015, Senator John McCain, chairman of the Senate Armed Services Committee, launched a review of DoD’s structures and practices and found that failures in the wars in Iraq and Afghanistan and bureaucratic waste could be attributed to top-heavy DoD headquarters. Skeptical of the size of OSD and the service secretary staffs, McCain asked rhetorically, “Is the quality of civilian oversight and control of the military better?”13 The results of the study shaped the 2017 National Defense Authorization Act (NDAA), which required the secretary of defense to “streamline the organizational structure and processes of the Office of the Secretary of Defense” no later than eighteen months from the enactment of the bill and downsized the number of civilian Senior Executive Service members by 25 percent.14

Yet the same law expanded the role of the CJCS by making it “responsible for” the “strategic direction of the armed forces” including “developing strategic frameworks and preparing strategic plans, as required, to guide the use and employment of military force and related activities across all geographic regions and military functions and domains.”15 According to the SASC summary of the bill, the changes were intended “to improve military advice to civilian leaders” and “strengthen the Chairman’s ability to assist the Secretary with the global integration of military operations.”16 But it ceded initiative to the chairman, the “global integrator” role echoing the global force management authority of the secretary of defense. Although the law stipulated that the chairman had to act consistently with the national defense strategy that the secretary wrote, and the secretary was free to revoke those delegated authorities at any time, the law obscured the distinction between the secretary’s strategic functions and the chairman’s.

Over American history, as the military grew and Congress ceded influence over budgeting, force structure, and war policy to DoD, civilians’ distinct professional and institutional contributions to military policymaking were subsumed in a more collaborative civil-military model. The permanent wartime of the Cold War, reinforced by the “forever war” on terrorism, atrophied best practices for civilian control over the military and elevated the status of military expertise relative to political expertise. Given the ground civilians have lost legally and bureaucratically, doubts about the unique value of a civilian background for positions of control over the military—such as the secretary of defense—are easy to understand. This book aims to revive that lost sense of civilians’ value to military policymaking.

Civilians in Scholarly and Contemporary Context

In the United States, civilians have distinct status, functions, and expertise that revolve around the exercise of political control. It is through political processes that national values are articulated, national priorities set, and national resources apportioned. Civilians’ ability to integrate military policy into national politics makes democratic control of the armed forces possible. Civilians decide how large or small the force should be, manage the division of labor among the military services, and decide whether and how those forces will be used. Civilians author national defense strategies, make decisions about global military basing, approve contingency plans, and buy weapons. They provide for military justice frameworks, nominate and confirm general and flag officers, and provide guidance on readiness, recruitment, retention, and promotion for military personnel.

Yet civilians are underappreciated, understudied, and underestimated. Neither practitioners nor the public pay much attention to the roles civilians play in military affairs, and scholars have built little systematic knowledge about who civilians are or how they fulfill their responsibilities. Political scientists tend to treat civilian elites’ supply of money, strategy, and oversight as givens (Betts 1991; Feaver 2003). Historical treatments contain rich descriptions of civilian wartime and defense policy leadership, but are often limited to an individual, a role, or an era rather than providing systematic patterns and findings (e.g., Halberstam 2001; McPherson 2008; Moten 2014).

In contrast to our limited knowledge about civilians, we have a detailed sense of what it means to be “military.” A rich library of military studies examines military professionalism, military institutions, military coups, military influence over policymaking, and the militarization of foreign policy (e.g., Huntington 1957; Janowitz 1960; Abrahamsson 1972; Dubik 2017; Lupton 2017). Civilians are not absent from these analyses—even those studies that are not explicitly couched in the civil-military relations subfield often examine the relationship between militaries and regimes. But these studies tend to simplify the roles civilians play and minimize civilians’ influence on military affairs. For example, the military innovation literature often argues that civilians are irrelevant to advances in military technologies or doctrine (Rosen 1991; Avant 1994). Ideas about military professionalism often focus on the “unique expertise” the armed forces provide in the policymaking context (Dempsey 2012, 4). Because there is no commensurate concept of civilian expertise there is little sense of what value civilians bring to military affairs.

There are practical reasons that academics and practitioners alike neglect studying civilians as intensively as they study the military. Scholars see the military as a tractable unit of analysis. Civilians are not institutionally organized or culturally homogenous, making them harder to aggregate into meaningful categories. “The military ethic,” wrote political scientist Samuel P. Huntington, “is concrete, permanent, and universal. The term ‘civilian’ on the other hand, merely refers to what is nonmilitary. No dichotomy exists between the ‘military mind’ and the ‘civilian mind’ because there is no single ‘civilian mind’” (Huntington 1957, 89). There is the military, and there is everyone else. This perspective has allowed analysts to lower civilians’ profiles so that even civil-military relations scholars often treat civilians like supporting characters: there to react to military initiatives or restrain military impulses, but not to shape the military’s actions.

But if civilian control is a prerequisite for democratic order, we need to know as much about civilian superiority as we know about military obedience. To build such knowledge, we must specify which civilian actions precede and complement military subordination, and why civilians choose to take such actions in the first place.

The fixation on the military and the definition of military professionalism as apolitical has also obscured the effects of politics on civilian control of the military. Samuel Huntington’s edict that civilians ought to isolate the military from domestic political disputes and that politicians ought not to meddle in operational matters generally has inhibited much of the subfield from incorporating civil-military studies into relevant political contexts (Huntington 1957; Brooks 2020). More recent scholarship has tried to call attention to this discrepancy. Drawing on Prussian military theorist Carl von Clausewitz, Eliot Cohen argues that war is a political activity and so politicians’ interventions in military strategies are consistent with the logic of war itself (Cohen 2002a). And Risa Brooks shows that equating the military profession with an apolitical identity blinds the military and those around it to the political drivers of military actions (Brooks 2020). Because civilians engage in politics to manage military affairs, methods of civilian control that emphasize insulating the military from politics just marginalize civilians from civil-military analysis more.

Scholars and practitioners alike need a revived concept of “civilian” in civil-military relations, not just to understand the role civilians play in American military policymaking, but to reinvigorate and perhaps even redeem it.


1. Congress amended Title 10 in 2021 to require that general and flag officers be retired for ten years prior to serving as secretary of defense. For officers at the rank of colonel and below, the law left the waiting period as seven years.

2. The Austin nomination also gave senators the opportunity to confirm the first Black head of DoD in the wake of renewed civil rights and racial justice unrest over the summer of 2020. See Bishop Garrison, “Representation at the Top: The Importance of Race in the Austin Nomination Debate,” Just Security, December 11, 2020,; Meg K. Guliford, “What Lloyd Austin’s Nomination Really Reveals, and What It Really Means for Me,” Inkstick Media, December 23, 2020,

3. For an exploration of the qualifications that career military officers bring to the secretary of defense role, see Alice Hunt Friend, “A Military Litmus Test? Evaluating the Argument that Civilian Defense Leaders Need Military Experience,” Just Security, August 19, 2020,

4. The historian of American revolutionary thought Bernard Bailyn references John Trenchard as an influence on American colonists.

5. Extracts from the Votes and Proceedings of the American Continental Congress Held at Philadelphia on the 5th of September 1774 (Philadelphia: William and Thomas Bradford, 1774), 8.

6. Federalist, no. 51, 317. The Federalists were highly partisan, and I use the Federalist Papers here not because they reflect general sentiment at the time, but because they represent the strain of thought that came to dominate and justify the content of the Constitution.

7. Of course, the civilians included in government as reflected in the Constitution of the late eighteenth century was a category limited to free Americans eligible to vote, run for office, and wield economic power. Congress, the presidency, and the judiciary represented a generally free, White, male, and monied subset of the American population. Only those with enough social power were invited to participate in the balancing of interests in government. The idea that both the militia and the government controlling it represented “the people” elided these qualifiers placed on representation.

8. House of Representatives, Hearing before the Select Committee on Post-War Military Policy, “Proposal to Establish a Single Department of Armed Forces,” 78th Congress, 2nd Session, April 26, 1944.

9. Department of Defense Reorganization Act of 1958, Section 202, (c)(7).

10. Representative Carl Vinson, chairman, House Armed Services Committee, Report Accompanying the Department of Defense Reorganization Act of 1958, May 22, 1958, 85th Congress, 2nd Session. Emphasis mine.

11. Cole et al. 1978, Secretary Lovett’s Letter—18 November 1952, 119.

12. SASC. “Defense Organization: The Need for Change: Staff Report to the Committee on Armed Services.” 99th Congress, First Session, October 16, 1985, page 40 and 45. Of course, any statistician will point out that the probability of an event is not determined by the frequency of previous occurrences of such an event.

13. The Future of Defense Reform: Hearing before the Committee on Armed Services of the United States Senate, October 21, 2015, 3.

14. Public Law 114–328: National Defense Authorization Act for Fiscal Year 2017, December 23, 2017. Hostility toward large staffs in OSD and the military departments persisted into the Trump administration. Secretaries Mattis and Esper both abhorred large headquarters, with the latter taking action to cut the size of leader support staffs. See Seamus P. Daniels, “Understanding DoD’s Defense-Wide Zero-Based Review.” Center for Strategic and International Studies, September 4, 2019,

15. Public Law 114–328, Section 921.

16. US Senate Armed Services Committee. National Defense Authorization Act for Fiscal Year 2017, 2.