Editor’s Note: This article is the fourth in a series from thinkers at the Center for a New American Security that explores the U.S. military’s phasing construct and the line between war and peace. Read the first, second, and third articles at War on the Rocks.
The difficulties of applying the U.S. military’s phasing construct to the realities of conflict are rarely more evident than when examining the American experience in Iraq. Though U.S. involvement in Iraq has traditionally been divided into two distinct periods of conflict, the 1990–1991 Gulf War and the 2003–2011 Iraq War, the reality is that the U.S. military has been nearly continuously engaged in Iraq for the past 26 years. The United States has conducted special operations raids into, launched cruise missiles at, imposed no-fly zones over, and outright invaded Iraq. The United States also provided humanitarian aid, financially supported local actors, and even governed the country. The six-phase planning construct does a poor job of accurately representing the range of activities over a quarter-century of U.S. foreign policy and military strategy in Iraq.
The phasing construct is optimized for traditional conflict, and the continuous conflict in Iraq has been anything but. Operation Desert Storm is straightforward: Phase II (Seize Initiative) closely followed by Phase III (Dominate). The same was the case for the opening days of Operation Iraqi Freedom, when the United States toppled Saddam’s regime in just 21 days. Unsurprisingly, a phasing construct conceptually designed for sharp periods of conventional state-on-state conflict easily mapped onto those scenarios. It is much more difficult, however, to categorize other periods of the Iraq experience.
One of these harder-to-define parts of U.S. involvement in Iraq is the period from 1991 until the invasion in March 2003. Soon after Desert Storm ended, the U.S. Air Force started dropping humanitarian relief supplies to Iraqi Kurdish refugees. This rather quickly evolved to include Air Force fighters enforcing a no-fly zone. Occasionally during the 1990s, these operations involved kinetic combat operations.
In late 1992 and early 1993, American F-16s shot down two Iraqi MiGs. Later in 1993, the U.S. launched cruise missiles from Navy warships to destroy a suspected nuclear facility. At several times throughout 1993, U.S. aircraft targeted Iraqi air defense and radar capabilities when they fired at or targeted coalition planes enforcing the no-fly zones in the north and south. In 1996, in reaction to Iraqi actions during the Kurdish Civil War, the United States conducted Operation Desert Strike, a one-day cruise missile campaign against Iraqi air defenses in the south, and U.S. forces expanded the southern no-fly zone closer to Baghdad. Over time, the military dimensions of Operation Provide Comfort eclipsed the original humanitarian tasks, Provide Comfort was replaced by Operation Northern Watch in 1997 to match its southern counterpart. Over four days in 1998, the United States and United Kingdom carried out Operation Desert Fox, a major bombing campaign against Iraqi targets believed to contribute to the development of weapons of mass destruction.
For a period generally not viewed as a time of war, the level of military involvement during those years was fairly high. Enforcing the no-fly zone was resource-intensive, and keeping Saddam in a box required semi-regular kinetic action.
It is precisely these spaces short of full-scale war that the phasing construct fails to capture very well. Was the period from 1991 to 2003 Phase I (Deter)? In a sense, the goal was to deter the Iraqi state from aggression against the Kurds in the north and the Shias in the south. However, the definition of Phase I in Joint Publication 5-0 never refers to offensive operations, only to shows of force to demonstrate capability and resolve. Some operations during this period, such as launching hundreds of missiles and aircraft sorties in Operation Desert Fox, go beyond mere demonstrations. Some kinetic actions may have been aimed at restoring deterrence, but others were clearly intended to actually degrade Iraq’s capabilities. Does that qualify these operations as Phase III (Dominate)? While the U.S. military controlled the operational environment, it did not involve a full employment of joint force capabilities and did not conclude with decisive operations, which is how JP 5-0 characterizes Phase III.
Did those years instead qualify as Phase II (Seize Initiative)? This seems to fit the bill, as Phase II is defined by offensive operations that set the conditions for decisive operations in Phase III. However, the phasing construct is designed so that the phases are, as JP 5-0 says, “planned in support of each other and should represent a natural progression.” Classifying Operations Provide Comfort, Southern Watch, and Northern Watch as Phase II would imply that the goal of the no-fly zone was not to punish Iraqi aggression against the Kurds, but rather to set conditions for a large-scale conflict. We know now that that is exactly what happened when the U.S. military went into Iraq in 2003, but that is clear only in retrospect.
Should most of this period be described as Phase I operations, with some punctuated Phase II or Phase III activities in Operations Desert Strike (1996) and Desert Fox (1998)? Again, the problem is that the phasing construct explicitly states a linear progression from one phase to the next, with phases occurring “sequentially” and the ending conditions of one phase being the start of another.
JP 5-0 acknowledges that there may be some overlap between phases, but doesn’t acknowledge the possibility that the force may not need to proceed to the next phase because political objectives were accomplished. In the world of the phasing construct, the progression is forever forward. The only time when a conflict might move backwards in phase is if the enemy regains the initiative: “Occasionally operations may revert to a previous phase in an area where a resurgent or new enemy reengages friendly forces,” JP 5-0 says. The period between the Gulf War and Operation Iraqi Freedom still fits awkwardly in the phasing construct.
Another glaring disconnect between the phasing construct and the U.S. military’s experience in Iraq is the temporal dimension. On the phasing chart, Phase III (Dominate) is the most significant phase in time and level of effort. However, at least in Iraq and Afghanistan, the Dominate phase was actually the shortest and most straightforward. In Afghanistan, the Taliban were displaced without even a major force deployment by the United States. The other phases take up less space on the phasing chart, but more pages in the history books, and the level of effort required to execute them was also greater. Stabilizing Iraq required more troops and cost more casualties than the swift defeat of Saddam’s conventional forces. In Afghanistan, the difference in the level of effort required to seize the country and stabilize it was even greater. At best, the phasing construct’s emphasis on Phase III is not justified by recent U.S. experiences.
If defining the period from March 1991 to March 2003 is challenging, reaching agreement on how to define the lengthy U.S. occupation and stabilization campaign from April 2003 to December 2011 is almost impossible. The initial invasion can conceivably be labeled as Phase III and everything afterward as Phase IV (Stabilize). Yet that would just go to show how inadequate the planning construct is in visualizing the relative level of effort in recent military engagements. The United States spent not even a month in Phase III and almost a decade in Phase IV.
In an informal focus group we assembled composed of a half-dozen field-grade military officers, the only thing they all could agree on was that the Operation Desert Storm and the initial March 2003 invasion were Phase III, a period of time that covers less than two months of the quarter-century the U.S. has spent in Iraq. Everything else was a hodgepodge of every other phase, each with valid and convincing justifications.
One view is that the U.S. never left Phase III operations in much of Iraq from 2003–2011. In the Kurdish-controlled areas in the country’s north, the U.S. may have been in Phases IV and V. By contrast, in areas such as Anbar, Baghdad, and Diyala, the U.S. never truly achieved the conditions for ending Phase III and beginning Phase IV. Again, this is only clear in hindsight.
A perhaps more accurate view is that the U.S. thought it finished Phase III in April 2003 and tried to move into Phase IV operations with the help of the United Nations, the Coalition Provisional Authority, and later the Iraqi Interim Government and the Iraqi Transitional Government. The 2007 troop surge was another attempt to bolster Phase III operations so that the starting conditions for Phase IV and eventually Phase V (Enable Civil Authority) could be realized. This is clear in President George W. Bush’s 2007 address to the nation explaining the need for the troop surge, in which he stated that security and holding territory was the “most urgent priority for success in Iraq.” The President expounded that the troop surge would over time allow Iraqi troops to take charge, provide security (Phase IV), and allow the Iraqi government the “breathing space it needs to make progress in other critical areas” (Phase V).
The fact that multiple activities were underway at any point in time is not a failing of the phasing construct. In fact, the phasing construct explicitly captures the fact that there may be overlap between phases and multiple different kinds of activity underway simultaneously. JP 5-0 states:
Phases are designed to be conducted sequentially, but some activities from a phase may begin in a previous phase and continue into subsequent phases. … A joint campaign or operation may be conducted in multiple phases simultaneously if the OA has widely varying conditions. For instance, the commander may transition to the stabilize phase in some areas while remaining in the dominate phase in those areas where the enemy has not yet capitulated.
The phasing diagram in JP 5-0 similarly shows that more than one type of activity can occur during a phase. Phases are defined by the focus of activity, not necessarily the absence of activities related to other phases. Still, it remains difficult to clearly define which phase the United States was in for large swathes of the past 26 years of Iraq operations. Similar to the no-fly zone operations that followed the Gulf War, the long and messy occupation of Iraq blurs the lines of the phasing construct, something that became abundantly clear when the United States thought it could move onto new types of operations only to revert back repeatedly.
Whether this disconnect between JP 5-0’s notional phases of a campaign and reality is a problem depends largely on one’s point of view. JP 5-0 states, “the actual phases used will vary (compressed, expanded, or omitted entirely) with the joint campaign or operation and be determined by the [Joint Force Commander].” Thus, the six phases (Shape, Deter, Seize Initiative, Dominate, Stabilize, Enable Civil Authority) may not even apply for any given campaign! This clause operates as a sort of escape hatch for the phasing construct, a sensible precaution to planners to use their best judgment when crafting a campaign. This is a prudent warning. Planners should apply knowledge of the specific mission and operating environment to a campaign plan. The phases as crafted in JP 5-0 may not be a perfect template.
Even if specific phases might require adjustment, there are still areas where the concept of phasing matches poorly with the Iraq experience. One area in which the phasing construct misses the mark is the presumption that distinctions between phases are clear, a condition that certainly was not the case in Iraq from 2003-2011. JP 5-0 states:
Phases are distinct in time, space, and/or purpose from one another, but must be planned in support of each other and should represent a natural progression and subdivision of the campaign or operation…. Each phase should have a set of starting conditions (that define the start of the phase) and ending conditions (that define the end of the phase). The ending conditions of one phase are the starting conditions for the next phase.
The ending conditions for the initial invasion were clear, but milestones for progress in the stabilization campaign that followed were fuzzier. What were the right metrics for transitioning from Phase IV to V and then ending Phase V? The 2007-2008 troop surge succeeded in tamping down much of the violence in Iraq. Did that mark the end of Phase IV and the beginning of Phase V? U.S. forces eventually left, but only because the United States did not reach a Status of Forces Agreement with the Iraqi government. Was Phase V accomplished? Did it fail? Clear transitions depend on clear conditions on the ground. This type of planning becomes significantly more difficult when those conditions cannot be sufficiently defogged. Insurgents cannot be counted like members of a conventional military, and counterinsurgency success cannot be assessed by battles won and combatants killed. The operational environment cannot be controlled if possible battlefields include everything under the sun, and the enemy’s will cannot be measured if the composition of the enemy is fluctuating and changing constantly. Again, this disconnect could be interpreted as a failure of the phasing construct or simply the reality that any actual campaign will diverge significantly from a notional template. (Alternatively, one could interpret the lack of clarity on ending conditions as a failure on the part of U.S. leadership to clearly define the metrics for success.)
The six-phase planning construct is a square peg, and a square peg is easy to fit into a square hole. In reality, the vast majority of conflicts the U.S. has been involved in have taken all shapes and sizes. Iraq is a prime example of one such conflict (ironically so, given that the Gulf War was the template for the original phasing construct). Much of the logic of campaign planning in JP 5-0 remains sound. It notes that different efforts should be planned in support of each other, that transitions in strategy should be based on conditions and not on time, and that there should be flexibility built in while actually conducting operations. These constitute helpful guidance, but the phasing construct does not reflect reality in other areas.
Then again, perhaps it is too much to expect that reality neatly lines up with the phases of a notional campaign. Planners will always have to use their best judgment and understand the particular characteristics of a conflict. The question is not whether the phasing construct serves a perfect template for real-world operations; it isn’t intended to. The question is whether any alternative models would be better. The phasing construct has evolved considerably since its inception, with multiple phases added and other recharacterized over time. Future articles in this series will explore ways it might continue to evolve in the future.
Kevin Shi is a researcher in the Defense Strategies and Assessments Program at the Center for a New American Security. Paul Scharre is a Senior Fellow and director of the Future of Warfare Initiative at CNAS.
Image: U.S. Air Force photo by Master Sgt. Cecilio Ricardo