The Future of War Is Unilateral but Small

Recent moves and countermoves by the U.S. and Iran in the Persian Gulf over the past few months have increased speculation about the prospect of war in the region. Some members of Congress, including a few Republicans, have stated that the president cannot use military force against the Islamic Republic without the approval of the legislature. The executive branch, meanwhile, is purported to have at one point put effort into arguing such an intervention would fall under the 2001 Authorization for Use of Military Force (AUMF), and in any case it can always claim that such a use of force would fall under the president’s own Article II authority. Most recently, attacks on American military assets and oil tankers may have provided a plausible justification for the president to use force against Iran regardless of Congress’s own preferences. A military response by the United States could come in many forms, ranging from further nonkinetic cyberattacks (in addition to those already undertaken) to limited airstrikes to full-fledged invasion. Concerningly, while recent Office of Legal Counsel (OLC) opinions have suggested that there is a limit to the scale of conflict that the executive can wage absent congressional approval, there is nothing to stop the attorney general or president from superseding these opinions and adopting a broader view of the president’s power. Even with a casus belli, however, historical patterns driven by a logic of political consequences suggest that the president will likely not use major military force absent congressional authorization—regardless of the creative legal arguments that would likely be put forward by executive branch lawyers.

Any time the use of military force is considered or realized, a debate ensues as to the exact authority under which the president may undertake such action. While some commentators put great effort into legitimizing or delegitimizing a use of force based on its legality, others see any such exercise as futile because of the rise of what Arthur Schlesinger has memorably described as an “imperial presidency,” in which the president acts regardless of Congress’s will. Indeed, scholars and pundits frequently point out the plethora of uses of force presidents have undertaken absent congressional approval as evidence that the legal status of an operation—contemplated or realized—has no effect on presidential decision-making. In this line of thought, the existence of a standing military is more than sufficient to allow a president to act unilaterally.

While it is true that presidents have repeatedly used force absent congressional approval, attention needs to be focused not only on the simple binary choice of using force or not, but also on the size and seriousness of the operation contemplated or undertaken. When considering past unilateral uses of force, it should be kept in mind that not all uses of force are equal in scale: The category includes both massive wars such as the Second World War and short engagements lacking casualties. Historical practice shows that presidents consistently have been quite comfortable using force on a small scale without Congress’s approval but even today refrain from undertaking major uses of force without the public backing of the legislature.

One way to visualize this is to sort different military operations based on the number of American casualties suffered.

The figure above plots U.S. uses of force over time by the number of U.S. combat deaths suffered in each engagement as coded by the Correlates of War project. To be sure, the dataset is not perfect as it includes only interstate disputes and extends between 1815 and 2010. It is, however, the best dataset available at this time for looking at the scale of conflicts over such a long time range and has already been used by preeminent legal scholars. Red data points represent those uses of force undertaken with explicit congressional approval (OIF refers to Operation Iraqi Freedom, or the Iraq War, and OEF refers to Operation Enduring Freedom, or the Afghanistan War), while the blue points represent engagements lacking the legislature’s formal backing.

Interestingly, while there is a popular belief that presidents have been increasingly willing to act alone, this only seems to be reflected in the frequency of uses of force but not in the scale of uses of force. Indeed, there does not seem to be any trend in the data suggesting a correlation between year and the scale of the operations presidents order on their own authority. There is—if anything—remarkable stability over time. One conspicuous outlier is the Korean War—and it is possible that, writing in the 1970s, an observer like Schlesinger would, without the advantage of forty more years of data, assume that the Korean War was part of a general trend in presidents increasingly undertaking large-scale operations unilaterally. In reality, the Korean War proved to be fully unique in its status as the only U.S. major war fought without formal congressional approval. The Korean War is better interpreted not as the end of congressional influence over the initiation of armed conflict but, rather, as a lonely outlier that later presidents consciously sought not to follow.

If anything, the Korean conflict may serve as the exception that proves the rule that presidents do not enter major wars without congressional backing. While Truman was persuaded by Secretary of State Dean Acheson not to ask Congress for authorization to use force in Korea (the argument being that “the President was entitled to use armed forces in protection of the foreign policy represented by the [U.N.] Charter,” and that the U.N. Security Council had given its approval for the use of force), congressional support for the intervention was strong and formal authorization would likely have been forthcoming had Truman asked for it. Thus, even the Korean War had strong congressional backing even if formal approval was never acquired.

The real precedential value of the conflict, however, was not that presidents could take the nation into a major war without congressional authorization but, rather, that, in failing to have members of Congress formally commit themselves to a use of force at the outset, presidents left themselves dangerously exposed to opportunistic criticism by political opponents as the conflict endured. For instance, as the Korean War wore on, Republican members of Congress were able to freely criticize Truman both for being too hawkish and, at the same time, for not being hawkish enough. Republicans simultaneously denounced “Truman’s War” and pointed to the “wooden boxes” in which American soldiers were coming home, while also arguing for the possible expansion of the war into China and the use of nuclear weapons. These seemingly incoherent attacks so depressed Truman’s poll numbers that he chose not to run for reelection in 1952. Truman’s immediate successor, Dwight Eisenhower, explicitly sought—and received—congressional blessing for using force both around Taiwan and in the Middle East after seeing the political trouble Truman ran into during the Korean War. Lyndon Johnson likewise specifically contemplated using force in Vietnam without congressional approval but wisely realized the political consequences of Truman’s folly the decade prior. Instead, he sought to “seal the lips of Congress against future criticism” by having them “tied, bound and delivered” in support of the war. Subsequent presidents have apparently agreed with Eisenhower’s and Johnson’s logic: The United States has not entered into a major conflict without Congress’s formal approval in the seven decades since 1950.

The conflicts in the bottom right corner of the plot represent other points of interest. Included among these are the Mayaguez Incident in 1975; the Tanker War of the 1980s; and the invasions of Grenada and Panama under Presidents Reagan and George H. W. Bush, respectively. The two invasions in the Caribbean basin were significant events, but the number of U.S. casualties sustained and the temporal length of the operations were both several orders of magnitude less than other major uses of force.

But perhaps focusing on fatalities alone is an improper measure of the seriousness of an operation because of the enormous “capitalization” of the U.S. military over time.  Undoubtedly, one cause of decreasing casualty figures over time is an improvement in technology—both in better protecting American soldiers (e.g., body armor) and in replacing humans in some missions altogether (e.g., drones). Put another way, it is possible that the U.S. spends less blood but more treasure today than in the past—a potential bias that suggests fatalities figures are not a perfect measure of the scale of an operation.

Considering the escalation level reached in each encounter can help visualize the scope of individual uses of force while avoiding the possible biasing of fatality figures over time. Below, I have used the “highest action” variable from the Correlates of War’s Militarized Interstate Dispute dataset. The escalation level of the dispute is rated along a 22-point scale, ranging from mere threats to mobilizations to occupations to interstate warfare. Here, because only actual uses of force are considered, the scale ranges from blockades to full interstate warfare.

While reasonable minds can differ over exact categorizations, the figure above shows a similar trend to the fatalities figure: The most serious uses of force have generally been undertaken only with congressional approval, while minor uses of force are often executed by the president alone. Again, the blue observations and red observations are relatively separated from one another—with the familiar exception of the Korean War. The other blue observation of an “interstate war” is the Boxer Rebellion around the turn of the 20th century—a conflict in which American involvement was rather small and that involved the loss of 53 American lives.

Perhaps of interest are the “clashes” taking place after 1950, since they appear to be similar to the OEF deployment that was authorized explicitly. These “unauthorized” observations consist almost exclusively of border clashes involving U.S. forces in a war clearly authorized by Congress—clashes with Syrians and Iranians during the Iraq War, encounters with Pakistani military units by American forces in Afghanistan, encounters with Chinese forces during the Vietnam War, and so on. Note that the Lebanon deployment and subsequent barracks bombing—the red datapoint between the Vietnam War and the Gulf War on both plots—actually had congressional authorization, although this was pushed for by Congress and resisted by Ronald Reagan. The other possible instances of unauthorized conflict also still seem to be closely related to otherwise congressionally approved activity—low-level disputes with North Korean forces in the 1960s after the U.S. Senate ratified a mutual defense agreement with the Republic of Korea and the establishment of “no-fly” zones in Iraq after Congress authorized the president in 1991 to use force in fulfillment of U.N. Security Council resolutions in relation to Iraq. Aside from the conspicuous exception of the Korean War, there is no apparent increasing propensity for the president to use force in more escalated conflicts over time.

Thus, after accounting for the scale of the use of force—measured in either U.S. fatalities or in the escalation level involved—there is actually little evidence for a full cession of the war-making authority by Congress in the 1950s. Instead, we see that consistently over time large-scale uses of force have had congressional authorization while small-scale uses of force have lacked it. The table below shows the postwar interventions that many critics have used to justify the belief that the president has fully usurped the war-making power from Congress. Note the strong relationship between U.S. combat fatalities and congressional authorization.

As can be seen, aside from one exception—the Korean War—every use of force had either a small number of casualties or explicit congressional approval. The invasion of Cambodia in 1970 is the only case aside from the Korean War in which more than 30 Americans died and arguably no congressional authorization was given (Schlesinger specifically discusses the Cambodian invasion as well). Even this case, however, has problems for proponents of the imperial presidency thesis as it relates to the scale of unilateral uses of force. First, the invasion was inextricably linked to the Vietnam War—indeed the Congressional Research Service itself does not recognize this as a separate conflict—as the targets of the operation were North Vietnamese Army and Viet Cong forces. Second, the Gulf of Tonkin Resolution itself had specified Southeast Asia—and not Vietnam exclusively—as its jurisdiction. Indeed, future Chief Justice William Rehnquist cited to the Gulf of Tonkin Resolution as the source of authority in his 1970 OLC opinion on the incursion.

In other words, while the Cambodian campaign may have been undertaken under controversial legal authority, it was also not a clear case of a lack of authority. And, despite all of this, even if one were to consider this a conflict wholly separate from the Vietnam War, it is hard to make the case that the Cambodian campaign was a clear counter-example to the general rule of presidents not undertaking major wars absent congressional approval.

What explains these trends? Perhaps presidents have consistently been unwilling to enter into “big wars” without knowing Congress is in on it as well, while being inclined to use force in “small wars” alone given that the cost in lives and money will be minimal. While there is a huge potential payoff for the president in getting Congress’s public commitment to a large conflict—which implies years of future funding and public support of the effort—going to Congress is also costly in three ways: first, the transaction costs in precious time and energy devoted to lobbying the Hill for support; second, the very real risk that Congress will embarrassingly reject the authorization sought; and third, the precedent created by affirmatively asking for congressional permission. Because of these costs, presidents will only seek authorization for uses of force they anticipate to be highly costly in blood and treasure, in which the added benefit of receiving congressional authorization is extremely high. Otherwise, for small, quick uses of force, the added benefit of congressional support is minimal and outweighed by the costs in time and effort needed to achieve approval, as well as the real risk that the authorization will be humiliatingly rejected publicly. 

It is this logic of consequences and politics—not any logic of appropriateness, constitutionality or norms—that drives the observed lack of serious war absent congressional authorization over time. Indeed, even Schlesinger himself noted this logic behind Johnson’s request regarding the Gulf of Tonkin Resolution, writing that LBJ “had come to believe that Truman made a mistake in not asking for a congressional resolution” in the Korean War but that this was a “political, not constitutional, error.”

Bringing this forward to the present simmering crisis with Iran, evidence from history suggests the following: The president will likely not bother asking for congressional approval if a planned use of force is relatively minor but will also not enter into a major operation absent the legislature’s blessing. Consider the rumored airstrikes against Iran called off by the president in June: The executive did not seek congressional approval for this engagement. This forecast tracks well with the executive branch’s own view of the president’s power to use military force: The president can use force in pursuit of “important national interests” so long as the “‘anticipated nature, scope and duration’ of the conflict [does not] rise to the level of a war under the Constitution.” But this view is itself heavily and expressly informed by past historical practice. If, as I argue, this observed pattern is driven by political considerations—and not the suasive force of legal opinions—even a president with little concern for respecting constitutional norms and willing to supercede the current test put forth by the OLC would nonetheless be compelled de facto to conform with this paradigm or face the political consequences of “going alone.” If this practice holds, the absence of—and likely impossibility of obtaining—an AUMF against Iran may not prevent the Trump administration from using force per se. But it will likely make any use of force undertaken unilaterally relatively minor.