The Case Against Police Militarization

Eliav Lieblich,Adam Shinar June, 2018

We usually think there is a difference between the police and the military. Recently, however, the police have become increasingly militarized – a process which is likely to intensify in coming years. Unsurprisingly, many find this process alarming and call for its reversal. However, while most of the objections to police militarization are framed as instrumental arguments, these arguments are unable to capture the core problem with militarization.

This Article remedies this shortcoming by developing a novel and principled argument against police militarization. Contrary to arguments that are preoccupied with the consequences of militarization, the real problem with police militarization is not that it brings about more violence or abuse of authority – though that may very well happen – but that it is based on a presumption of the citizen as a threat, while the liberal order is based on precisely the opposite presumption. A presumption of threat, we argue, assumes that citizens, usually from marginalized communities, pose a threat of such caliber that might require the use of extreme violence.

This presumption, communicated symbolically through the deployment of militarized police, marks the policed community as an enemy, and thereby excludes it from the body politic. Crucially, the pervasiveness of police militarization has led to its normalization, thus exacerbating its exclusionary effect. Indeed, whereas the domestic deployment of militaries has always been reserved for exceptional times, the process of police militarization has normalized what was once exceptional.

INTRODUCTION

In late December 2014, Ferguson protesters posted an “eviction notice” on the walls of the St. Louis Police Department, declaring that it was “to be reclaimed by its citizens,” due to the transformation of the police into an “occupying force.”¹ In early December 2016, 2,000 veterans gathered at the Standing Rock Sioux Reservation in North Dakota to serve as “human shields” for protesters against a plan to construct an oil pipeline through the Reservation.² The choice of language was striking: the terms “occupation” and “human shields,” belong to the world of war rather than to that of law enforcement and civil protest.³

This specific language, in both cases, does not necessarily reflect the protesters’ belligerent attitude, nor is it coincidental. Rather, it responds to and mirrors the long process of police militarization: a process whereby police forces come to look and operate like military forces. Indeed, in recent years, police forces across the globe—and particularly in North America—have increasingly adopted military models.⁴

The United States is leading this process. In the last few decades, U.S. police forces have acquired military weapons and equipment; set up paramilitary units, often with the assistance of the military or former soldiers; and embraced overall a militaristic mode of operation far removed from the antiquated police officer “walking the beat.” This process is likely to intensify in the near future. While some constraints on police militarization were put in place during the Obama presidency, the Trump administration has repealed these restrictions, limited as they were.

That the police should not be an occupying force is intuitive, and many object to police that are reminiscent of armed forces. However, this intuition remains surprisingly unexplored. Indeed, various instrumental arguments against police militarization have been advanced in recent years. Specifically, critics have argued that police militarization is ineffective in fighting crime, constitutes an inefficient allocation of resources, and incentivizes police brutality at the expense of constitutional rights.⁵ However, let us imagine well-regulated, well-trained, and well-supervised militarized police that do not cause these adverse consequences. Is the problem with militarized police only rooted in such consequences? Our answer is no. Extant arguments against police militarization, correct as they may be, are contingent and do not fully grasp the true nature of the problem.

Police militarization is part of a wider trend in which exceptional measures—traditionally reserved for extreme emergencies—are normalized.⁶ Originally established in the 1960s to ostensibly counter extreme threats, militarization was empowered in recent decades through the “war” discourse. From the “war on drugs” to the “war on terror,” militarization saw a process of normalization that culminated in the camouflaged police-troops and armored personnel carriers that deployed in the streets of Ferguson in 2014 and in the fields of Standing Rock in 2016.

Accordingly, this Article identifies the key problem of police militarization in its normalization . But invoking normalization is not enough. What exactly is being normalized? It is in this context that the Article makes a novel contribution: suggesting that what is normalized is a presumption that the policed community is threatening . In other words, militarization is wrong because it is based on a presumption of threat , while the liberal order, if it is to have any substantive justification, must be based on precisely the opposite. The assumption that others are threatening, in essence, reduces the liberal order to a Hobbesian state of nature. The real case against police militarization, then, does not lie with its consequences, but rather its implicit reversal of one foundational aspect of the liberal order.

Police militarization implies a presumption of threat because it exhibits two salient characteristics. First, it is primarily preventive rather than strictly reactive: deployment of militarized police reflects the anticipation of extreme violence of the type that could require a forcible response. Second, it is collective: since it is not (always) aimed at specific individuals, it tends to rely on collective assumptions of the potential for violence. When militarization becomes normalized, the presumption of threat becomes normalized as well. This capacity for normalization is enhanced when it is specifically police that becomes militarized, since police— contra the military—are elements of the normal, rather than the exceptional, legal order.

Importantly, we do not argue that militarized forces actually fight the civilian population, like militaries fight. Rather, we claim that the essential effect of the presumption of threat—manifested in militarization—is symbolic or expressive . Actual combat is not needed: the mere deployment of militarized police carries the symbolic power to exclude the policed community from the political collective. This is because what is perceived as military is generally understood to operate outside the state. We demonstrate this excluding effect by invoking Carl Schmitt’s famous distinction that possibility of combat— which militarization implies—creates a friend/ enemy distinction that constitutes (and delineates) the political collective.

The Article proceeds as follows. Part I offers a definition of police militarization as a process in which police adopt the appearance and behavior that symbolizes military in a given political culture. It then distinguishes between exceptional and normalizing militarization, the latter reflecting the process in which the exception becomes normalized.

Part II examines the process of police militarization, demonstrating how it became normalized, from the “war on drugs,” through the “war on terror,” and until this day. It ends by surveying the federal backlash against militarization during the late Obama presidency, but notes that the attempts to curtail militarization were neither principled nor sufficient, and indeed have been largely reversed by the Trump Administration, which is amenable to further militarization.

Part III discusses the prevalent instrumental arguments against police militarization, namely that militarization generates more violence. While we are sympathetic to these arguments, they suffer from a shortcoming common to consequentialist arguments: it is possible to counter them with slight adjustments of our factual assumptions. We therefore argue that the question of police militarization must be discussed on the non-instrumental, principled level.

Part IV presents the core of our case against police militarization. It first lays down the argument that police militarization correlates with a presumption of threat by virtue of the fact that it is based on preventive and collective reasoning. It thereafter claims that the liberal order must be based on the opposite presumption of non-threat: when a state assumes that its citizens are threatening, it undermines its own authority and in fact moves closer to totalitarianism. We then move to demonstrate that the presumption of threat, manifested in the militarization of police, implies the exclusion of the policed community from the body politic. Merely by virtue of its symbolic power, militarization generates an excluding friend/ enemy distinction. Significantly, when the distinction is carried out by police forces, the exclusion is more forcibly normalized, since the police are part of the normal, rather than exceptional, order, which is the domain of the military.

Part V addresses several possible challenges to our theory. In general, these objections suggest that some of militarization’s ostensible benefits might justify the exclusion that it creates. For instance, militarization might contribute to officers’ protection; it might deter unlawful acts and increase the public’s sense of security. However, we claim that these challenges are unconvincing, both empirically and on a principled level.

I. Defining Militarization A. Militarization as a Symbolic Process

Militarization is the process by which an organization adopts the operation mode of a military or embraces military values and culture.⁷ Accordingly, its primary problem-solving tools are borrowed from the military (for example, its exercise of military technology and power, its equipment, and its organizational and operational style). According to Peter Kraska, “[t]o militarize means adopting and applying the central elements of the military model to an organization or particular situation. Police militarization, therefore, is simply the process whereby civilian police increasingly draw from, and pattern themselves around, the military model.”⁸

The word process is key. While the police, since their inception, have always exhibited some militaristic tendencies, the phenomenon we wish to address is their increasing militarization . We focus on the process in which otherwise civilian law enforcement agencies come to resemble militarized forces and the symbolic effect ushered by this transition. It is through this process that the exceptional gradually becomes normalized.

Police militarization is usually viewed through the lens of four parameters: material, cultural, organizational, and operational.⁹ The material lens focuses on the types of weapons, uniform, technology, and equipment police use. The cultural lens examines the type of language, style, appearance, and values used by the police. The organizational lens views the way the police choose to organize themselves in terms of hierarchy, special units, and forces. Finally, the operational lens looks at the patterns of police action in the various areas of its operation.

To be clear, we do not claim that police militarization amounts to a wholesale importation of the military paradigm to law enforcement. To the best of our knowledge, no police department has altered its rules of engagement to reflect those of fighting military units. To us, however, militarization rests not on the actual adoption of military-style shoot-to-kill rules of engagement, but rather on the cultural message it projects. From the point of view of the citizen, what matters is not only what militarized forces actually do, but also what they symbolize. This is why our definition of militarization is not contingent upon the actual adoption of military operational approaches across the board.

For this reason, when we discuss militarization, we generally refrain from suggesting a checklist for defining the exact point at which a police force becomes militarized. It is possible to discuss whether, for instance, a blue-clad policewoman carrying a handgun would not be considered militarized, while the same policewoman wearing black and carrying an assault weapon would. But to us, this would be overly technical and contingent. We therefore opt to treat militarization as a cultural phenomenon in which certain attributes are generally perceived by the public as reflecting militarization. In sum, we address militarization as a symbolic process .

This is why, when we consider the four parameters that characterize militarization, we focus on their observable external dimensions, which can be viewed as an aggregate of characteristics that are perceived as militarized in a specific culture. The paradigmatic example, of course, is the prevalence of the now ubiquitous Special Weapons and Tactics (“SWAT”) units in the United States. Like military units, SWAT teams are equipped with separate uniforms (urban tactical gear), full body armor, and Kevlar helmets, and armed with automatic weapons, commonly in use in the military,¹⁰ as well as other technologies such as sound suppressors, laser sights, and semi-automatic and automatic shotguns. Some police units have acquired armored personnel carriers.¹¹ All of these were once exclusive to the military, but are now routinely part of police equipment. SWAT teams adopted military style governance and discipline, and their tactics, language, and training differ from the ordinary police officer.¹² Their operation looks more like, and indeed is, a paramilitary nature.¹³

B. Exceptional and Normalizing Militarization We draw an essential distinction between the exception and the

norm. We might accept that in exceptional and well-defined situations of emergency, some special police units are needed (for instance, the taking of hostages during a terrorist attack, or when a grave and imminent threat to life requires decisive action beyond the capabilities of simple police). When militarized forces are utilized only in such situations, this is exceptional militarization . Exceptional militarization traces the classic distinction in political theory between the normal state of affairs and the exceptional emergency—the state of exception—in which extreme measures are seemingly required.¹⁴ While exceptional militarization can be abused— indeed, as we later discuss, the mere availability of such capabilities and the incentive to acquire more of them results in a tendency to use them even when not actually required—in well-defined emergencies, it could be agreed upon that such forces might be required.

A different type of militarization is normalizing militarization . As a factual phenomenon, this occurs when militarized forces are increasingly deployed in settings that are by no means exceptional, as the term is commonly understood. They can be used in day-to-day police activities, such as serving search and arrest warrants, in drug raids, and in patrolling the streets, often wearing full battle dress uniforms and carrying automatic weapons.¹⁵ These forces might also be deployed for “public order” pur-poses in political demonstrations.¹⁶

While exceptional militarization raises a host of difficult questions, our central focus is normalizing militarization . This is because a central claim of this Article is that the key problem of militarization lies in its power to exclude the policed community from the political collective. If this is true, then it becomes clear that when militarization is normalized, so is the exclusion.

In this regard, it is necessary to situate normalizing militarization within the wider phenomenon identified by leading theorists as one in which “states of exception,” or emergencies, are becoming the rule.¹⁷ If a clean-cut distinction between the exceptional and the normal was once thought to be possible, nowadays it has become muddled.¹⁸ Terrorism, never-ending wars,¹⁹ wide-ranging national emergency measures,²⁰ and extraordinary international reactions to counter ongoing refugee influx²¹ are only some current aspects of this phenomenon. Indeed, normalization of exceptional measures is dangerous because we lose sight of the fact that it existed as an exception to begin with: measures that were once extreme become the new normal. New and more extreme “exceptions” arise as our standards become lower and lower.²²

Police militarization squarely belongs within this general process. The relationship between militarization and the “exception” is revealed when we consider militarization’s potential for extreme physical violence. When we normally think about police, we intuitively think about the maintenance of law and order. We usually do not think about police in terms of their capacity to kill. Giorgio Agamben famously connected the increasing capacity (or power to decide) to kill and the erosion of the distinction between the normal and the exceptional, focusing on those who are protected by the legal order and those excluded from it. In his penetrating analysis, states of exception were once distinct points in which sovereignty moved from the regulation of law and politics into the regulation of mere physical existence, or “bare life.”²³ To Agamben, as the borders between the normal and exceptional become blurred, the regulation of bare life (and death) moves from the margins of the political realm to the fore.²⁴ In our context, while traditional police could (ideally) be seen as regulators of human behavior, normalizing militarization—by virtue of its capability to employ extreme deadly force—is an instance in which policing moves to regulate, at least in potential, bare life itself.²⁵

Moreover, normalizing militarization cannot be separated from the normalization of “war” in the general political discourse. War, of course, is the quintessential military concept, packing almost mythical powers as the ultimate state of exception.²⁶ Once problems are reframed as “wars,” hardline policy initiatives can garner more support.²⁷ Once a certain situation is described as “war”—and therefore exceptional—the executive’s margin of discretion grows.²⁸ Accordingly, this elevation of threat could partially account for the erosion of judicial review of militarized police action, for instance.²⁹

The metaphor of “war” proved key to the process of police militarization. With the winding down of the Cold War, policymakers set their sights on domestic problems, namely drugs.³⁰ Soon enough, no longer was there a drug “problem” or a crime “problem.” Those problems became the “war on drugs” and the “war on crime.”³¹

The rhetorical tool of “war” in these contexts facilitated, indeed incentivized, police militarization. When the term is used by higher political echelons, it permeates street-level discourse as well. Indeed, one of the hallmarks of normalizing militarization is the infiltration of military language into the day-to-day:³² sometimes police go as far as to describe city streets as a “war zone,” borrowing directly from military language.³³ In terms of political discourse, the war metaphor served to elevate certain categories of crime to existential national security threats, certain types of criminals to enemies, and thus certain types of situations to exceptional ones. However, since these exceptional situations are not well defined in time or space, they are perpetuated and normalized.³⁴

Last, the interrelationship among the war discourse, militarization, and normalization cannot be viewed separately from the “war against terror,” in which the war discourse serves to justify continuous and global forcible actions against ambiguous terrorist networks.³⁵ Such actions—implying a general erosion of boundaries—affect the way states approach external and internal threats.³⁶ Police militarization should therefore be viewed as part of a global process in which war becomes closer to law enforcement,³⁷ and law enforcement becomes closer to war, creating a significant overlap between the two.

We turn now to demonstrate how normalizing militarization came about.

II. The Processof Police Militarization

A. The War on Drugs and the Shift from Exceptional to Normalizing Militarization

There was a time when there was crime without a war on crime; a drug problem without a war on drugs. The war metaphor came about in the wake of World War II, when leaders sought to rally public support to solve social problems. The “war on crime” was coined by President John-son in 1966, whereas the “war on drugs” began during the Nixon administration. Nixon called drugs “public enemy number one,” equating them with “foreign troops on our shores.”³⁸ President Reagan took it to another level, declaring drugs to be a national security threat, a sentiment echoed by Presidents Bush and Clinton.³⁹

The war on drugs not only prioritized drug offenses, but also led to a proliferation in police paramilitary units—namely SWAT units—and a change in their character. SWAT teams are a relatively new creation. First established in Philadelphia in 1964, they spread as a result of the Watts Riots in 1965 and were perfected by would-be LAPD chief, Daryl Gates.⁴⁰ Whether in response to the police’s inadequate performance during the riots, or as a product of racially charged moral panic, Gates became convinced that the police force must familiarize itself with Guerilla warfare. He asked the military, the Marines in particular, for help, and they obliged.⁴¹ The connection between SWAT and the military was tight from the beginning. Not only were SWAT teams trained by the military, but they also adopted military behavior and norms.⁴² At first, even the name SWAT had a military orientation. Initially, Gates wanted SWAT to stand for “Special Weapons Attack Teams.”⁴³ However, the idea was rebuffed by his supervisor, precisely because of the military connotation.⁴⁴ Police do not attack; they relate to their community.⁴⁵ They are not meant to use preventive violence. The word “attack” was therefore replaced with “and tactics.”⁴⁶

SWAT teams were quickly established in most American cities.⁴⁷ The creation of SWAT teams was a product of various material incentives. Federal funds incentivized police to purchase military hardware, contributing to skewed police incentives. For example, provisions which allowed the police to keep the proceeds of assets forfeited in anti-drug activities led them to focus on drug offenses at the expense of other crimes.⁴⁸ The money obtained through this activity was often used to expand paramilitary units.⁴⁹ Significantly, the Military Cooperation with Law Enforcement Act of 1981 gave the military a foothold in domestic affairs in the war on drugs, by authorizing it to make available to the police equipment, information, research facilities, and access to military bases.⁵⁰

Once a SWAT unit is in place, it will be used. Initially, SWAT teams were limited to what we have described as exceptional militarization— meaning, the occasional hostage situation, barricaded suspect, or the rare act of terrorism. Over time, however, SWAT teams have expanded their scope of activity. Now, whenever there is “high risk” police activity, a term which itself has been expanded, these units are summoned.⁵¹ Since the mid-1980s, militarized units are used in a wide spectrum of day-to-day police activity. Most SWAT “call outs” in this period had nothing to do with terrorism or hostages, but rather with serving warrants or drug raids, which were redefined as high-risk activities.⁵² As Peter Kraska observed, twenty percent of paramilitary units were used for routine patrol work.⁵³ As a 2014 ACLU report states, 79 percent of police paramilitary activity is now devoted to serving warrants, one of the more routine types of police deployment.⁵⁴ The availability of military hardware led police to use helicopters, camouflage, and battering rams, even when the suspicion amounted to little more than growing marijuana.⁵⁵ Two recent appellate decisions have criticized police departments that used SWAT teams even for regulatory inspections of barbershops and nightclubs.⁵⁶ The legacy of the war on drugs can thus be described as a key moment in which the shift from exceptional to normalizing militarized police gained ground.

B. Normalizing Militarization and Physical Transformation: Personnel and Equipment

The proliferation of militarization was accompanied and enhanced by the increasing interaction between police and the military in training, personnel, and equipment. Military training and the increasing number of exmilitary personnel in the ranks of the police augment the process of normalizing militarization, while the acquisition and deployment of military hardware affects the symbolic representation of policing.

A useful point of departure is, again, the establishment of SWAT in Los Angeles. As aforementioned, the initial SWAT was the result of cooperation with the Marines and was inspired by guerrilla warfare in the Vietnam War.⁵⁷ Soon thereafter, elite military units such as SEALS and Rangers began to train police officers throughout the country.⁵⁸ This training normalizes militarization not only directly through SWAT units, but also indirectly because “ordinary” police units are trained by their SWAT colleagues.⁵⁹

Not only does the military train the police, many veterans have joined the police after their discharge, partly due to government incentives. The Community Oriented Policing Services program (COPS) in the Department of Justice, a program “advancing the practice of community policing,”⁶⁰ allocates millions of dollars every year ($14 billion since 1994) in grants that are used to defray the costs of police officers.⁶¹ In 2012, the program restricted the hiring of non-military personnel.⁶² Although the restriction was repealed a year later, it indicates an underlying sentiment that the police can benefit from a military mindset by having veterans serve among its ranks, thus infusing the police with a military culture.⁶³ “Community policing,” for which veterans are hired by the police, is a vague term.⁶⁴ Thus, it is perhaps unsurprising that much of the COPS money went into establishing SWAT teams.⁶⁵ Sustaining militarization through ex-military personnel and military training that eventually permeates to “ordinary” police units, has a strong, systemic, normalizing function.

Normalizing militarization led to the physical transformation of police forces throughout the United States. As aforementioned, after the Cold War, the military began selling (or donating) weaponry, surplus, and technology to the police (for example armored personnel carriers, M16 rifles, and night vision technology).⁶⁶

While the requirement to pay the military usually discouraged police departments from spending exorbitant sums on equipment they probably did not need,⁶⁷ two federal programs operated by the DOD, program 1122⁶⁸ and program 1033,⁶⁹ made such acquisitions easier. Program 1122 allows local law enforcement agencies to purchase military equipment at subsidized rates,⁷⁰ including weapons and surveillance equipment.⁷¹ Program 1033 is the more ambitious of the two. Passed by Congress in 1997, it is a program run by the U.S. Defense Logistics Agency Disposition Ser-vices.⁷² The program allows for surplus weapons and gear to be transferred to local law enforcement agencies, free of charge.⁷³ Although this made things easier for the police, the maintenance costs still made the transaction cost for some agencies prohibitive.⁷⁴

Still, in its first three years, Program 1033 handled 3.4 million orders for military gear from 11,000 police agencies. In 2005, the number of police agencies participating in the program climbed to 17,000.⁷⁵ Since its inception, the program has disbursed over $5 billion worth of military equipment.⁷⁶ As the United States began to withdraw from Iraq and Afghanistan, transfers to local police increased significantly.⁷⁷ In 2014 alone, local law enforcement agencies received nearly $1 billion worth of military equipment.⁷⁸ Importantly, most of the applications come from small and medium sized police agencies, which are not as well funded as larger agencies, which presumably already possess such equipment.⁷⁹

The possibility of free military equipment created an arms race, so to speak, among police agencies. Small agencies saw that larger agencies were arming themselves and did not want to be left behind. Militarization thus became a matter of prestige.⁸⁰ Indeed, the change in the physical appearance of police necessarily led to a symbolic change: from a policing force involved in law enforcement to a militarized force standing ready for the possibility of combat.

C. The “War on Terror” and Crowd Control as the Epitome of Normalization

The terrorist attacks on September 11, 2001, brought both continuity and change: continuity, because in many ways militarization increased and, in certain respects, became more normalized;⁸¹ change, because in recent years there is also a realization, at least among some, that there is something deeply troubling about this process.

As part of the “Global War on Terror,” virtually every law enforcement agency was enlisted in the war effort. The establishment of the Department of Homeland Security (DHS) was one major consequence of the terrorist attacks. The DHS, with its $66 billion budget for the fiscal year 2017,⁸² spends much of that money in the form of grants to local law enforcement agencies that in turn serve to procure military equipment.⁸³

DHS grants dwarf other funding programs. Through various federal programs, billions of dollars are transferred from the federal government to the local police level, mostly by the DHS.⁸⁴ Between 2001 and 2011, the DHS gave $34 billion in anti-terror grants, many of which went to unlikely places such as Fargo, North Dakota, or Canyon County, Idaho, hardly prime terrorist targets.⁸⁵ Many of these places, unsurprisingly, chose to buy military equipment.⁸⁶ It did not take long for militarized police, ostensibly meant to counter grave security threats, to pop up in a quintessentially democratic setting: political demonstrations. Those were now subject to militarized crowd control, perhaps the epitome of normalizing militarization.

Even before 9/11, militarized police, together with the National Guard, were sent to quell the demonstrations in Seattle surrounding the 1999 WTO ministerial conference as part of the anti-globalization campaign, sparking what became infamously known as the “Battle in Seattle.”⁸⁷ The police changed to full riot gear with helmets, arming themselves with pepper spray, tear gas, stun grenades, and even rubber bullets.⁸⁸ An internal investigation by Seattle police commended its performance, referring, inter alia , to police officers as operating on the “front line.”⁸⁹ Others, however, thought differently. Seattle Chief of Police Norm Stamper resigned over the handling of the demonstrations, referring to his performance as “disastrous.”⁹⁰

Smaller scale demonstrations happened in the 2009 Pittsburgh G-20 Summit. In Pittsburgh, like Seattle, police used tear gas (even firing it into dorm rooms), sound canons, bean bags, and rubber bullets.⁹¹ Moreover, police in Pittsburgh wore paramilitary garb, camouflage, and combat boots.⁹² This is hardly the gear required for urban settings, where camouflage uniforms stand out more than conceal.⁹³ This gives rise to the suspicion that such gear is meant to signify rather than to serve an operational function.

Overall, when facing large events with the possibility of protests, militarized police became the norm.⁹⁴ To us, most important is that the appearance of militarized police for crowd control signifies the normalization of militarization. Originally conceived to operate in states of exception, militarized forces now routinely appear in the most fundamental of democratic activities.

D. The Occupy-Ferguson Backlash and the Trump Administration The public outcry following the heavily militarized police reaction to

the 2011 Occupy protests and the 2014 Ferguson demonstrations provoked the official acknowledgement of the problems of unchecked militarization. However, as we argue below, the preliminary steps taken were unlikely to adequately address the problem and indeed have been reversed by the Trump administration.

Militarized police were on full display during the Occupy movement. Reflecting on police militarization during Occupy Wall Street, former Seattle Chief Stamper wrote:

[T]he police response to the Occupy movement, most disturbingly visible in Oakland—where scenes resembled a war zone. . . brings into sharp relief the acute and chronic problems of American law enforcement. Seattle might have served as a cautionary tale, but instead, US police forces have become increasingly militarized, and it’s showing in cities everywhere . . . Such agencies inevitably view protesters as the enemy. And young people, poor people and people of color will forever experience the institution as an abusive, militaristic force—not just during demonstrations but every day, in neighborhoods across the country.⁹⁵

Stamper was not alone. Others have equated police response to protestors as militarized or paramilitary, tracing it to the war on drugs and funding programs like COPS and Program 1033.⁹⁶

Recently, the images from Ferguson and later Standing Rock brought police militarization to every home.⁹⁷ Battle-ready police in combat uniforms, with weapons pointed at citizens and armored personnel carriers alarmed many observers.⁹⁸ While disconcerting, Ferguson also signaled a potential for change. After Ferguson, both state and federal govern-ments seem to have started to recognize the potential harms of the process.

A 2015 report submitted by the government-appointed Ferguson Commission⁹⁹ criticized the police response and gear as “military-style,” recommending that “St. Louis law enforcement agencies should aim to have a more human, non-militarized, proportional response to future protest or demonstration activity in the region.”¹⁰⁰ It further issued a call to “[d]irect the state to cease providing, and local departments to cease using, militarized weaponry that does not align with a use of force continuum that authorizes only the minimal amount of force necessary.”¹⁰¹

The Ferguson Commission Report was preceded by a Department of Justice Investigation into the police response in Ferguson.¹⁰² Essentially calling to halt normalizing militarization, the DOJ report found that the

Use of military weapons and sniper deployment atop military vehicles was inappropriate, inflamed tensions, and created fear among demonstrators. Agencies possessing military-type equipment or weaponry should restrict its deployment to limited situations in which the use of the equipment or weapons is clearly justified . . . The equipment and weapons should be kept out of sight and not be used routinely or in the absence of special circumstances.¹⁰³

The Report stated that “[f]or persons who would have seen the SWAT teams arrive, the action would have resembled that of ‘soldiers arriving at a war zone.’”¹⁰⁴ For example, whereas the police use an armored personnel carrier, ostensibly to protect police officers, members of the community often referred to them as tanks, thus contributing to the perception of the police as military.¹⁰⁵ The importance of these observations, as we note later,¹⁰⁶ is precisely in its recognition of the effect of militarization on the perception of police by the policed community.

The DOJ recommendations did not go unheeded. On January 16, 2015, President Obama issued Executive Order 13688 titled “Federal Support for Local Law Enforcement Equipment Acquisition.”¹⁰⁷ Acknowledging the contribution of federal funds to keeping “the American people safe,” the Order notes that the government “must ensure that careful attention is paid to standardizing procedures” regarding the provision of military equipment and that more must be done to ensure that law enforcement agencies are aware of the civil rights concerns resulting from militarization.¹⁰⁸ To that end, the Order established an inter-agency working group tasked with, among others, compiling a list of controlled equipment that could be given to the police, but with additional oversight on the allocation, use, and training of agencies that receive controlled equipment.

Pursuant to the Executive Order, the Law Enforcement Equipment Working Group submitted its recommendations in May 2015.¹⁰⁹ It recommended, among other things, that law enforcement agencies be prohibited from acquiring certain equipment, chosen both because of its propensity for overuse or misuse, and because its militaristic nature undermines community trust in the police.¹¹⁰ The acquisition of other equipment will be controlled, meaning that agencies may acquire them, subject to further oversight, assurances, and certification. Moreover, agencies must employ protocols on the use, supervision, evaluation, accountability, transparency, and operation of the equipment.¹¹¹

While the Order limits militarization to an extent, police can still purchase the prohibited equipment or bypass the extra oversight provided for controlled equipment if the acquisition is not made through federal programs or federal funds. Moreover, the Executive Order has sparked criticism from both the House of Representatives and police sheriffs.¹¹² This is significant, since Executive Orders can be changed relatively easily.¹¹³

And indeed, during his campaign, President Trump pledged to reverse the post-Ferguson restrictions on militarization.¹¹⁴ Unsurprisingly, the National Fraternal Order of the Police—the largest police union in the United States—expected him to deliver: their first “potential action” for Trump’s first 100 days was to “rescind Executive Order 13688.”¹¹⁵ Upon assuming office, President Trump quickly delivered. On August 28, 2017, he signed Executive Order 13809, revoking both Executive Order 13688 and the recommendations issued pursuant to Executive Order 13688,¹¹⁶ thus restoring Program 1033 to its scope prior to Ferguson.

III. Key Instrumental Argumentsagainst

Police Militarization Until now we have described the process of police militarization in

the United States, demonstrating how police embarked on a militarization spree since the 1960s, ushering a move from exceptional to normal militarization. Still, it is possible to claim that militarization, on its own, should not bother us too much if it brings about good results. Consequentialists might argue that if militarized police are supervised and disciplined not to use excessive force, perform efficiently overall, prevent crime, and keep communities safe, this is all that matters.

As discussed later, we argue that even if militarized police could achieve these goals, militarization is inherently problematic from a nonconsequentialist perspective. Nonetheless, in this Part we detail numerous instrumental arguments that have been made against police militarization. To be clear, we do not believe that these arguments fail. In fact, we (generally) agree with them, and some have also been empirically substantiated.¹¹⁷ However, like all instrumental objections, they are factually contingent, as it is possible to imagine a world where the concerns are met satisfactorily. Therefore, in this Section we elaborate on the instrumental arguments against militarization not only to complete the picture, but also to underscore, in general, why a non-consequentialist, normative argument is required.

A. Militarization Undermines Trust and Does Not Reduce Violence The main instrumental argument against police militarization is that

it is inefficient since it only increases violence or undermines police legitimacy. When citizens believe the police will respond with violence, there could be an escalation. A display of force could catalyze both sides and increase the level of violence. For instance, the “war on drugs” ultimately exemplifies how normalized militarization is in its essence a cycle of violence. A heavy handed approach to narcotics crimes led to more turf wars among drug dealers, which led to more violence, leading to public pressure on police to get even tougher, leading to police wanting more authority and equipment, and consequently to use even more force.¹¹⁸ At the same time, the war on drugs showed poor results, with scholars, judges, police officers, and policymakers hailing it a failure, given the billions of dollars spent, the lives lost, the devastating effects on minorities, and the rise in incarceration.¹¹⁹

Militarization can thus create conflict instead of defusing it, contributing to loss of trust in the police.¹²⁰ Trust can be replaced by fear or hatred,¹²¹ which in turn decreases the legitimacy of police.¹²² For instance, the DOJ report stated that militarization often brings about “unintended consequences” that galvanize a negative reaction and aggravate community concerns about the police and the justice system.¹²³ While a voluminous literature examines police legitimacy and its relationship with compliance,¹²⁴ scant attention is devoted, if at all, to how militarization generates fear among the population, which in turn affects the legitimacy of the police. As Rachel Harmon argues, fear of the police has substantial costs that are rarely, if ever, factored into the cost-benefit equation.¹²⁵ When the police patrol the streets with armored personnel carriers, some of which are mine-resistant; when the police use military weapons, helicopters, bayonets, or disperse riots with various grenades; when SWAT teams cover themselves with masks to conceal their identity, they create a shock and awe effect that is designed to create a menacing presence.¹²⁶

A possible reply is that the use of military equipment and military weapons could decrease the overall level of force exercised by police in that the mere display of power will discourage resistance. We address this specific argument for deterrence below.¹²⁷ For now, it suffices to point out that there will be times when the presence of the equipment and weapons will increase the severity of force.¹²⁸ For example, military weapons will be used for operations that in the past relied on less powerful weapons. An example is the increasing reliance on SWAT teams to execute warrants that in the past were executed by ordinary units, or the use of armored military vehicles where none were used in the past.¹²⁹

Militarization creep also contributes to these adverse effects. Once one starts using militarized police for one thing, for example drugs, one may start using it for other things, for example counter-terrorism and ultimately for day-to-day policing.¹³⁰ Once military means are available, more initiatives using those means will be undertaken.¹³¹ We pointed this out in the context of call out rates.¹³² Over time, paramilitary police units have been increasingly dispatched, whereas the rise in violent crime could not, in itself, explain the increased deployments.¹³³ Thus the rise of police militarization cannot be explained as a “rational” response to increased crime rates. Once the equipment is there, police will want to use it. This is of course inefficient economically, and also contributes to heightened tensions.

A further problem, which might also generate more violence, relates to the changed self-perception of the militarized police officer. As Camp-bell and Campbell point out, the roles of an infantryman and a patrol officer are different. When the police officer’s job is framed in military terms, discretion about arrests and the appropriate level of force give way to aggressive intimidation rather than to the delicate problem solving required from the police officer.¹³⁴

Similarly, as Benjamin Beede argued, the military is usually illequipped for tasks other than war, such as civilian law enforcement.¹³⁵ When law enforcement is militarized, it might become clumsy and inefficient because of vastly different training. Consequently, some argue that soldiers are generally less flexible than civilian police and are prone to over-reaction when dealing with the public.¹³⁶ Soldiers are perceived, also by themselves, as “warriors,” whereas police officers are in the business of law enforcement.¹³⁷ Conflating the two thus invites a heightened risk of police violence where police officers now behave in a military or quasi-military fashion.

To sum up, the increased likelihood of using force, the tendency to frame social problems as wars, the addition of manpower through federal budgets that goes toward increased militarization, and the transfer of military equipment has led observers to argue that militarization has caused police to become more aggressive and violent, often with dramatic consequences to policed communities.¹³⁸ These consequences are enhanced in marginalized communities, which see the most militarized forces.¹³⁹ If policing depends on receiving the cooperation of the policed community and maintaining good relations with the community, then in communities that feel targeted by militarization, police will naturally be approached with suspicion or not approached at all, thus undermining their effectiveness.

B. The Contingency of Instrumental Arguments Taking these arguments into account sheds a different light on the

security-rights tradeoff. Militarization is often promoted by asserting that although rights are important, personal and community safety eradicating crime can and should override the rights of criminals.¹⁴⁰ But if militarization makes us less safe because it tramples on the individual rights of many, because it generates fear and alienation between the community and the police, because it leads citizens to distrust their police, and because in some cases it is ineffectual in reducing crime, then the security-rights tradeoff is false.

Nevertheless, as compelling as these arguments are, they do not offer a principled objection. This is because theoretically, instrumental arguments can be met with counterarguments that invoke the option of tightening secondary norms. Responding to instrumentalist concerns, the problem of militarization can presumably be solved with better regulation and supervision. If we just make slight adjustments, the argument goes, the problems might be fixed. Non-reduction of violence and undermining of trust can be dealt with through more training, discipline, and enhanced police-public relations. Self-perception of police can be controlled through education and is not a salient issue since soldiers also engage in policing in various operational settings.¹⁴¹ If our objection was purely instrumental, we would have to concede that there would be nothing wrong with such militarized police forces. Yet, there still remains something disturbing in the vision of good-mannered, disciplined, and approachable police rolling in tanks through our neighborhoods. This is the predicament of instrumentalist arguments: they are virtually always contingent .

There seems, then, to be something else, not-strictly-instrumental, that underlies the genuine objection to police militarization.¹⁴² This is why we need a principled, non-consequentialist argument to advance the discussion.

IV. The Principled Caseagainst Police Militarization

In this Part, we propose a principled objection to police militarization. First , militarization entails a presumption that the policed community and individuals are threatening. A liberal order, conversely, must be based on precisely the opposite presumption: that of non-threat. Indeed, the prevalence of the presumption of non-threat is a key difference between liberal states and totalitarian ones.

Second , the presumption of threat, manifested in militarization, trig-gers the exclusion of the policed community, through the symbolic power of police. Importantly, this exclusion emanates from the cultural perception of militarization rather than from a particular instance of use of force by militarized police.

Third , when this distinction is created by militarized police —rather than by locally deployed military forces during emergencies—the potential for exclusion is more significant due to the normalizing effect of police.

A. Militarization as a Presumption of Threat Imagine that you participate in a political demonstration in your city.

The police—entrusted with preserving order—appear militarized: an armed personnel carrier is deployed and personnel wear armor, helmets, and carry assault rifles. The message must be that the state views you, and other participants, as posing a level of threat of the type that can only be properly countered by deploying, in advance, forces capable of using extreme violence. Militarization thus entails a presumption that the policed community poses a certain level of threat, so intense that it requires the availability of combat-ready forces.¹⁴³ We refer to this state of mind and its effects as the presumption of threat.¹⁴⁴

The presumption of threat is a product of two interrelated characteristics of militarization. First, when militarization becomes normalized, it also becomes collective, in the sense that its reference point is not specifically threatening individuals in specific circumstances, but rather an unspecified group of people—the policed community.¹⁴⁵ Second, militarization is preventive, in the sense that it is not based upon an actual, imminent and grave threat, but rather on the assumption that such threat is present, whether emanating from society at large or (perhaps more often) from certain groups within society, or from certain individuals inter alia because of their belonging to specific groups.

To clarify how militarization implies a presumption of threat, it is helpful to draw from the distinction made in international law between law enforcement operations undertaken in peacetime and hostilities con-ducted during armed conflicts.¹⁴⁶ In absence of an armed conflict, law enforcement operations are conducted under the presumption that unless circumstances absolutely prove otherwise, individuals do not pose a threat of the scope and level that must be confronted with severe, possibly lethal violence. This view is behind the use-of-force continuums required by international law in law enforcement operations,¹⁴⁷ and also in the practice of many police departments.¹⁴⁸ The upshot is of course that threat must be individually determined and must be imminent.

Traditional military operations, conversely, operate under the exact opposite presumption: certain categories of individuals are a priori threats, and therefore can be collectively neutralized—including by premeditated, proactive use of lethal force.¹⁴⁹ The hostilities paradigm is thus collective, since a person becomes targetable on account of formal status. It is also preventive in the sense that status implies threat, even if the individual combatant was not threatening, in the strict sense, at the time of attack.¹⁵⁰ In short, armed conflict entails the reversal of the presumption of non-threat precisely on these two levels. Due to this reversal, international law is clear that the switch between the law enforcement paradigm and that of hostilities requires (at least) the factual existence of organized armed violence.¹⁵¹ These understandings of military operations are not only abstract legal concepts but reflect deeply entrenched societal conventions on the nature of war.¹⁵²

An example of how militarization can reflect a presumption of threat can be found in the infamous use by the LAPD, during the 1980s “war on drugs,” of battering rams attached to armored vehicles for the purpose of rapidly tearing down doors and walls and forcibly breaking into suspected drug houses.¹⁵³ This tactic exhibited both elements of militarization: a complete use-of-force continuum is not envisioned due to a predetermination of severe threat;¹⁵⁴ and the threat is attached to categories of persons rather than individuals—in this example, residents in neighborhoods predominantly inhabited by minorities, albeit suspected of drug offenses.¹⁵⁵ Indeed, the choice of theaters in which such units are deployed implies that, like in military operations, levels of threat are at least partially determined in relation to categories of individuals or their geographical location. In other words, militarization of police—just like “proper” military operations—inherently assumes a heightened level of threat, usually from a group of people, at a certain place, rather than from individuals. It thus transforms law enforcement activity from an individual-based action to one which assumes some collective threat.

This reversal seems to permeate the discourse of militarized police personnel. Consider, for instance, SWAT units. Recall that, initially, LAPD chief Daryl Gates envisioned SWAT as “Special Weapons Attack Teams.”¹⁵⁶ As aforementioned, the word “attack,” which connotes proactive acts of violence against an enemy¹⁵⁷ was replaced by “tactics” in order to better place SWAT units within the ethos of policing.¹⁵⁸ Police chiefs have also reported that outfitting police in battle uniforms creates a mind-set that “you’re a soldier at war.”¹⁵⁹ In a sense, the presumption of threat is a necessary byproduct of military training of police.¹⁶⁰

However, a possible counterargument is that during hostilities, militaries actually target individuals in line with the presumption of threat. Conversely, militarized police, although adopting certain military attributes, do not. Thus, it could be argued that since police in practice still operate within the traditional law enforcement notions of individual threat and strictly reactive lethal force, they do not really operate under a presumption of threat. Our claim, however, is more nuanced. To us, the mere deployment of combat-ready police conveys to the community that it must be threatening; otherwise such forces would have not been needed to begin with. Thus, the point of assessment of the action is already in time of deployment, which must be understood as collective preventative action. Indeed, as we detail later, once this possibility of combat is performed and asserted, the presumption of threat is already established.

A presumption of threat is a factual mode of operation. However, as we argue, a foundational principle of the liberal order must require just the opposite: that the State does not act upon the presumption that its citizens are threatening.

B. The Presumption of Non-Threat as a Principle of the Liberal Order

The liberal legitimization of the State is traditionally grounded on its fundamentality for the achievement of the triumvirate of “life, liberty and the pursuit of happiness.”¹⁶¹ Other, more progressive versions, advance justice, fairness, and welfare based justifications.¹⁶² Common to all these visions is that the individual must be the end of all political action¹⁶³ and that freedom can only be restricted when its use harms others.¹⁶⁴ As we have shown, instrumental arguments against police militarization claim that it adversely affects these basic rights and principles.

Yet we believe that a meaningful liberal order must be based on another foundational principle, one that captures the salient, principled problem of militarized police. To an extent, this foundational principle is derived from the Hobbesian notion of security as a basic end of the commonwealth.¹⁶⁵ However, it differs from the basic Hobbesian idea since it does not focus on the individual as a potential victim in the absence of security.¹⁶⁶ Nor does our account afford absolute primacy to the concept of security, as Hobbes suggested.¹⁶⁷ Rather, we focus on the exclusionary effect that materializes when the state perceives the individual as a potential threat. As we argue, the liberal order cannot presume that citizens pose a threat absent a clear, individually determined factual basis, without unraveling. Therefore, the liberal order cannot be preserved unless operating under a presumption of non-threat.

The upshot of the presumption of non-threat is that threat must always be determined individually, and accordingly that violent or coercive action can be only reactive or preemptive. When threat is presumed, state action becomes collective and preventative.¹⁶⁸ A presumption of threat is behind policies such as mass preventative internment;¹⁶⁹ counter-terrorism policies that hinge on group belonging;¹⁷⁰ “early warning” systems;¹⁷¹ racial profiling;¹⁷² barriers to citizenship based on ethnic grounds for ostensible “security” reasons;¹⁷³ or mass surveillance, in itself a concept centered on preventative law enforcement.¹⁷⁴ More generally, the institution of a presumption of threat is a key element of an exception becoming normalized.

Conversely, the presumption of non-threat complements, and is entangled with, traditional justifications for liberal rights. For instance, the presumption of innocence, as well as due process rights in general, are usually justified as necessary for the protection of freedom and autonomy.¹⁷⁵ However, they can just as well be understood as protecting a freedom from being considered a threat to the State and others.

Essentially, one key difference between liberal and totalitarian states is that in the former citizens are presumed non-threatening, while in the latter, the entire State apparatus is structured in anticipation of internal threats.¹⁷⁶ Indeed, even the hardnosed realist Carl Schmitt conceded that “[n]o democracy exists without the presupposition that the people are good” and thus their will is valid,¹⁷⁷ while absolutism invokes the axiom that humans are naturally evil, and therefore strong authority is justified.¹⁷⁸ In terms of political theory, the threat/non-threat dialectic can be traced to the differing perceptions in classic thought of human beings in pre-State society (the “state of nature”). In this context, social contractarians frequently point out that anarchy breeds a security dilemma.¹⁷⁹ While theo-rists differ on the morality of violence under anarchy, most agree that an organized political community—a commonwealth—is a solution to insecurity under the state of nature.¹⁸⁰ Yet, these theories largely follow two archetypal models—for our purposes, the Hobbesian and Lockean— which differ precisely on the scope of power in the hands of the sovereign, and consequently, on the extent to which the sovereign can assume the citizen/subject is threatening. Simply put, while the Hobbesian (absolutist) model leads to a sovereign that can act in accordance with a presumption of threat, the (liberal) Lockean model does not.¹⁸¹

To Hobbes, the state of nature is an unavoidable condition of war because of the natural inclinations of individuals, their possession of “the right to everything,” and their equal powers to harm one another.¹⁸² Under such conditions, individuals are always mutually threatening, and therefore preventative action is always necessary and justified.¹⁸³ In other words, Hobbes’s state of nature is defined precisely in that individuals are presumed to be mutually threatening . It is this condition that drives individuals to enter into a reciprocal agreement with others for mutual security.¹⁸⁴ War is terminated when persons, through this agreement, divest themselves from the right of all things—including the right to act preventatively—and leave that power only to the sovereign.¹⁸⁵ In essence, therefore, individuals abdicate their right to presume that others are threatening while retaining the right to act in self-defense only when actually assaulted.¹⁸⁶

Hobbes’s idea of the commonwealth famously envisioned an absolute sovereign that imposes peace through fear.¹⁸⁷ A Hobbesian sovereign could indeed presume that its subjects are threatening by nature. A “visible power to keep them [the subjects] in awe,” Hobbes writes, is needed to keep their natural passions at bay.¹⁸⁸ Of course, such a sovereign could deploy militarized forces to instill fear, to counter this presumed threat. It is exactly on this point that Hobbesian social-contractarianism differs from the ideas undergirding (ideal) liberal-democratic states. While we might agree (or not) that in a hypothetical state of nature individuals can presume that others threaten them, we can hardly retain a liberal view of the State if we argue that such powers remain in the hand of the sovereign. Even under the Hobbesian view of the state of nature, the establishment of political society must at least entail a shift in the perception of individuals from threatening to non-threatening. Otherwise, the State simply recreates the state of nature it is meant to avert.¹⁸⁹

Locke shared with Hobbes the position that the purpose of the State is security.¹⁹⁰ In his state of nature, too, persons enjoy “perfect freedom” and equality.¹⁹¹ However, contrary to Hobbes, equality does not provide for preventive violence but precisely the opposite: a prohibition on harm-ing one another, unless in imminent self-defense or as punishment.¹⁹² For Locke, a state of nature is not necessarily a state of war, but rather a condition in which war is both more likely and potentially more destructive. It is this that leads individuals to form a political society.¹⁹³

Thus, for Locke, even in the state of nature individuals cannot act on the basis that others are threatening. Since the Lockean sovereign derives its power from the transfer of “executive power” from individuals to the public,¹⁹⁴ this a priori means that it cannot assume a presumption of threat that is unavailable to individuals in the state of nature.¹⁹⁵ Put differently, individuals cannot give to the State something they themselves do not possess. It follows that when acting under a presumption of threat, we are precisely following the absolutist Hobbesian logic concerning the powers of government.

Indeed, the notion of security is central both to Hobbes’s absolutist view and to Locke’s balanced approach, in which security is but one value within the ends of “peace, safety, and public good of the people.”¹⁹⁶ While it is beyond this Article to develop a theory of a liberal right to security—a contested issue in political theory¹⁹⁷—it is sufficient, for our purposes, to argue that any proper understanding of “security” cannot be convincingly reconciled with the presumption of threat. In a sense, a normalized presumption of threat is tantamount to a normalized state of emergency, in which the Hobbesian logic of the state of nature as constant threat prevails. An individual thus cannot enjoy security when she is presumed to be a threat, since being perceived as such implies exclusion from the same order ostensibly charged with maintaining security and protecting her: it implies that she is in a Hobbesian state of nature in relation to the State, even as she is situated within the State.¹⁹⁸

Until now we only implied a connection between the presumption of threat and the notion of exclusion. In the next Section, we expand on the concept of militarization and exclusion by demonstrating that militarized police possess the symbolic power to mark the policed community as the enemy, to exclude it from the political collective, and to ultimately normalize this exclusion.

C. The Symbolic Power of Police, Exclusion, and the Normalization of the Exclusion

1. The Symbolic and Exclusionary Power of Militarization

Both police and military are organs of State coercion. Functionally, one can argue that they are interchangeable if necessity dictates.¹⁹⁹ If we adopt a simplistic perception of State power as violence, the police, in descriptive terms, are simply an agency that applies coercive force for the purpose of imposing a legally justified end within a given society. In Weberian terms, it is merely one arm through which the State’s monopoly on force is carried out.²⁰⁰ All other conditions being equal, if the police do not bear an additional—even symbolic—meaning, it shouldn’t matter whether coercion is conducted by police or military forces, and consequently, by traditional or militarized police forces.

However, when looking at the constitutional practice of States, it is clear that deploying military forces within State boundaries packs explosive political meaning. For instance, it was famously cited in the U.S. Declaration of Independence as part of the “long train of abuses and usurpations” that justified secession of from Great Britain.²⁰¹ Similarly, in modern constitutional law of many nations, the role of armed forces is generally restricted to the defense of the State, while their internal deployment is permitted only during emergencies or when special national security con-cerns require so.²⁰² While these restrictions could be partly explained by the historically abusive tendencies of standing armies (and specifically the practice of quartering),²⁰³ the objection to standing armies is, at bottom, a principled one, relating to their military character rather than their actual abuses.²⁰⁴

Indeed, the distinction between different forms of State coercion is as old as the State itself. For instance, although the most important precursor to the modern administrative state—the Roman Republic and then Empire—did not possess a specialized police force,²⁰⁵ there was some distinction between the military and other agents of coercion. Roman magistrates were always accompanied by lictors carrying fasces, but axblades would be fixed only when leaving city limits.²⁰⁶ Since magistrates held dual civil and military roles, the presence of lictors —rather than soldiers—symbolized their civilian function. For this reason, when Mark Antony was accompanied by soldiers rather by lictors, he was condemned for acting as a tyrant.²⁰⁷

Why, then, do societies generally distinguish between different agents of coercion? The answer is all about drawing boundaries. In ancient Rome, a fundamental constitutional principle distinguished between the city as a “pacified sphere” ( pomerium ), surrounded by a “sacred boundary,” from which military power was excluded.²⁰⁸ As discussed above, this idea can also be found in liberal social contractarian thought, in which the State represents a realm of peace, distinguished from the state of nature—and possibly war—lurking outside.²⁰⁹

Militaries are thus perceived as forces of external coercion, supposedly operating in the anarchic state of nature, and acting under a presumption of threat emanating from the enemy. They are meant to protect the social contract from outside threats, where the state of nature supposedly reigns. Police, on the other hand, are an agent of the internal social contract. Phrased in this manner, we can already understand the exclusionary power of militaries or militarized forces. Police are perceived as a coercive power acting within the community, within and part of the social contract; militaries act against those outside.²¹⁰ Police act as trustees, militaries act as adversaries.²¹¹ When the military or militarized police are deployed internally, they exclude the targeted communities from the body politic.

In our view, then, the root of the police/military distinction cannot be captured only by referring to the historical problem of standing armies, or by functional or instrumental grounds such as the fear of excessive force.²¹² Rather, the root lies in the symbolism of the perception of “police” and “military” in a given society; and specifically, in their respective perceptions as inclusionary versus exclusionary forces.

To understand how militarization excludes we must first understand the power of symbols. Symbols are spoken or visual forms that convey a collective meaning, whether directly or implicitly.²¹³ They can serve as powerful political signifiers, possessing the capacity to “preserve or to transform objective principles of union and separation. . .of association and dissociation.”²¹⁴ Those who possess symbolic power—for instance, those who act upon state authority—²¹⁵ have the “power to make groups” and to make “visible and explicit social divisions.”²¹⁶ In the context of police, theorists describe its “symbolic power” as a source of “powerful, efficacious collective representations about community, order, the distinction between good and evil, and about security and protection.”²¹⁷ Those who are served by the police are thus part of a de facto community.

However, when symbols build on representations that tap into cultural associations that inspire (and normalize) fear and subservience—even without exercising direct violence—symbolic power turns into symbolic violence.²¹⁸ This is precisely the transformation that occurs when police are militarized. A militarized police force— qua police—retains its symbolic power to “make groups” and to make collective representations about community. This power turns exclusionary as the symbolism becomes increasingly violent. As we demonstrate below, the potential for an especially intense type of violence—that amounting to the possibility of combat—is a strong signifier of exclusion.

The power of symbols has not been overlooked by legal scholars. Expressive theorists²¹⁹ of law argue, much in line with the idea of symbolic power, that “[t]he expressive dimension of governmental action plays a central role. . .in constitutional law,”²²⁰ and that governments cannot act in ways that convey that certain “social impressions” override all competing values, since these cause expressive harms.²²¹ For instance, in their work on voting rights and district redistribution, Pildes and Niemi argued that redistribution (gerrymandering) can be unconstitutional even without causing material harm (such as by vote dilution), precisely because it conveys the social impression that “race consciousness” is an overriding consideration.²²² Militarization and the presumption of threat that underlies it pack symbolic power that can generate the expressive harm of exclusion.

When this process is pervasive, as is the case with normalized militarization, it can solidify and reify this exclusion in the public eye. Indeed, some scholars argue that the expressive power of law (in the broad sense) is reflected, inter alia , in its ability to convey information.²²³ Law in general, police action included, signals attitudes and information that affect behavior and public beliefs.²²⁴ Arguably, the practice of police militarization— especially in an era in which media coverage is ubiquitous—²²⁵ can affect the general perception of the policed community as particularly threatening.²²⁶ As images of militarized police marching in the streets of Ferguson were transmitted into every screen, a particular image of the local residents was also transmitted.

As discussed above, the deployment of militarized police in a given arena is laden with exclusionary potential. However, does this remain true even if these forces do not actually engage in combat, but are only present in this or that community? In the next Section, we turn to explore the notion that the mere deployment of militarized forces indicates the possibility of combat, and that this in itself results in exclusion from the political community.

2. Deployment of Militarized Forces as a Friend/Enemy Distinction

To better understand the exclusion generated by the symbolic power of militarization, a helpful point of departure is Carl Schmitt’s famous view of “the political”—and the concept of the State itself—as based on the friend/enemy distinction.²²⁷ Particularly relevant is the central meaning Schmitt attributes to the act of distinction, marked by the performance of the possibility of combat.

To Schmitt, a people’s collective political existence hinges on a distinction between friend and enemy.²²⁸ Now, not every disagreement results in such a distinction, but only one which exhibits “the utmost degree of intensity of a union or separation, of an association or dissociation.”²²⁹ Meaning, the distinction results in the inclusion in or exclusion of the enemy—the perennial “other”—from the political collective.²³⁰

Importantly, the degree of intensity, which implies political dissociation, is achieved only when the possibility of violent conflict is recognized.²³¹ To Schmitt, “[a]n enemy exists only when, at least potentially, one fighting collectivity of people confronts a similar collectivity.”²³² The enemy concept thus hinges on “the ever present possibility of combat.”²³³ Combat here is not a metaphor: rather, the potential of real war and killing must exist. Otherwise, at hand is only a local, non-excluding, dispute within members of the community. Now, it is important to stress that for a friend/enemy distinction to be made, combat does not have to actually take place, but must be present as a possibility .²³⁴ The political, in sum, is the “mode of behavior” which follows the possibility of combat.²³⁵

How does this reflect on police militarization? As the possibility of physical killing must be present for a political distinction, the concept of the weapon—and more so, the type of weapon culturally perceived as fit for combat—is central.²³⁶ The appearance of such weapons may be the ultimate signifier of a friend/enemy distinction, and correspondingly, of the exclusion of the subject from the political collective. The same can be said concerning military-style battle dress fatigues or combat helmets, as representations with clear cultural connotations: these can effectively symbolize the declaration of a “domestic enemy.”²³⁷

In sum, when the police, whether due to their weapons, equipment, or uniform, take the form of a militarized force—meaning, that through their cultural representations they are perceived as equipped for the possibility of combat—essentially a new friend/enemy distinction emerges, whereby the policed community is excluded from the political collective. The fact that these weapons are not always used, but rather are displayed, does not alter this effect, since it is the possibility of combat, coupled with the power to decide that lethal force should be used,²³⁸ that generates the exclusionary power of militarized police.²³⁹

We are now in a better position to understand the complaint by Ferguson demonstrators of being under “occupation.”²⁴⁰ This labeling, in fact, captures a remarkably precise distinction, reflecting exactly the exclusionary power of militarization. While a State might conceivably impose martial law within its territory in extraordinary circumstances and subject to constitutional mechanisms,²⁴¹ it by definition cannot “occupy” its own sovereign territory. Occupation, as commonly understood, is a factual situation in which a State gains control over external , hostile territory, territory over which the it has no sovereign title and is not indebted politically to the local population.²⁴² Importantly, occupation implies both spatial and personal exclusion. When protesters complain of “occupation,” they are noting the effect of being marked as an enemy and thus placed in a zone of exclusion within their own State, where “outside and inside” become intermingled.²⁴³

This effect is all the more significant when the policed community is comprised primarily of minorities. Indeed, the exclusionary result of the friend/enemy distinction is amplified when aimed at historically marginalized communities, as it could reify already prevailing notions of exclusion and alienation.²⁴⁴ Experience shows that militarized police forces have been deployed more in African American and other minority communities.²⁴⁵ As suggested in the literature, one reason for this is precisely a collective presumption of threat.²⁴⁶ This realization highlights a further troubling aspect of police militarization: its potential not only to exclude, but also to normalize the exclusion in a way that disparately impacts those that are already wielders of inferior social power.

3. Militarized Police and the Normalization of the Exclusion Above, we claimed that police possess the symbolic power to exclude, when it bears the cultural hallmarks of militarization. It is arguable, however, that as long as State coercion is conducted by forces formally belonging to the police, the symbolic distinction between police and military remains. If this were true, the mere fact that militarized police maintain their formal definition as “police” and retain some symbolic distinction, such as badges or insignia, would alleviate our concerns. If you will, the inclusionary power of police would “cancel out,” the exclusionary power of combat-ready appearances. Perhaps for this reason the U.S. Posse Comitatus Act prohibits the “use” of “any part of the Army or Air Force” for policing, but the Secretary of Defense may authorize the use of military equipment by the police.²⁴⁷ However, the usual procedure is that military markings are removed or covered in such cases.²⁴⁸

Here, however, we must point out a key paradox: the same inclusionary symbolic power of police also breeds its especially harmful potential. Since police are an organ of the “normal” legal order, constantly present and unrestricted to extreme emergencies, its transformation can lead to a normalization of a certain type of distinction.²⁴⁹ In other words, since police symbolizes inclusion within a political system, it can be a vehicle through which exclusion can become internalized as part of the norm, as part of the system.²⁵⁰ When it is the police—in contrast, say, to armed forces deployed in “true” extreme emergencies such as civil wars—that carries out the friend/enemy distinction, the message is all the more powerful: the enemy status becomes part of the regular political life of the community.

Critical thinkers have long noted the potential of the police as a normalizer of state violence. Hannah Arendt, for instance, pointed out that totalitarian regimes tend to disproportionately strengthen the police, since it is more efficient as a force of oppression than the military. To Arendt, the protective military ethos results in that “even under totalitarian conditions they [members of the military] find it difficult to regard their own people with the eyes of a foreign occupier.”²⁵¹ Indeed, this rings true, at least in States where the military is by and large representative of a relatively homogenous society.²⁵² Already in the nineteenth century, Britons complained, at times of social unrest, that the military forces they encountered “know how to behave themselves, but the police don’t.”²⁵³ More recently, during the 2011 revolutions in Egypt and Tunisia, the military refused to act against demonstrators, as opposed to the police and internal security apparatuses.²⁵⁴

Walter Benjamin, too, pointed out this nature of police violence, manifested precisely in its norm-setting potential. Benjamin characterized all “legal” (contra “pure”) violence as serving either a lawmaking or law preserving function.²⁵⁵ To Benjamin, the military serves a “lawmaking” function vis-à-vis a vanquished enemy,²⁵⁶ while serving a “law preserving” function vis-à-vis the state’s citizenry through general conscription.²⁵⁷ The police, conversely, combine these two functions. In obvious cases, the police employ coercion to preserve law; however, since the police operate in the vanishing point of written law’s coercive power, it necessarily operates in ambiguous situations, thus effectively creating new law.²⁵⁸ In other words, the creation of new law by police action is a creation of the new normal.

The normalizing aspect of police also clarifies the impact of the friend/enemy distinction described above. For this distinction to hold, the possibility of combat must be normalized : it must “remain a real possibility for as long as the concept of the enemy remains valid.”²⁵⁹ It is obvious, therefore, that as much as militarization becomes normalized and open-ended, and as more traditional police units exhibit militarization, their excluding effect becomes more entrenched. Conversely, deploying special units for specific tasks, for a limited time in exceptional conditions, might result only in a temporary effect.

In sum, the key difference between police and military can be described by reference to the inclusionary versus exclusionary symbolic power of these two organs of State coercion. The military symbolism implies a friend/enemy distinction; when militarization specifically affects the police, the danger of normalization of the distinction is amplified. It is in this sense that militarization of police normalizes the exception.

V. Possible Objections

In this Part, we address several possible objections to our theoretical case against militarization. The first objection is factual and comparative: it points out that, in fact, militarized police forces of various types have been acceptable in some of the world’s most well-established democracies. We then consider three instrumental objections, which make the claim that our argument is trumped by the benefits reaped by militarization. First, that militarization protects police officers. Second, that militarization deters unlawful acts. And third, that militarization increases the sense of security of the public at large. Finally, the last objection we attend to concerns our idea of militarization as a symbolic process and argues that if so, normalization will eventually neutralize the exclusionary effect of militarization. We believe these objections ultimately fail.

A. Established Democracies in fact have “Hybrid” Militarized Forces

A possible challenge to our theory emanates from the existence of two types of “hybrid” forces.²⁶⁰ The first is the (mainly) Continental European phenomenon of units formally belonging to the military but charged with wide policing duties. The most famous examples of such units are the French National Gendarmerie ²⁶¹ and the Italian Royal Carabinieri ,²⁶² or the Spanish Guardia Civil .²⁶³ Such forces act as law enforcement agencies, but can also be deployed externally, in some form, during armed conflict.²⁶⁴ They are therefore doubly affiliated with ministries of defense and of interior.²⁶⁵ If established democracies use such forces for public order, does our theory of militarization and exclusion still stand?

Four reasons lead us to answer in the affirmative. First, historically, these forces are remnants of old authoritarian regimes—both in form and in attitudes to policing—and therefore cannot be celebrated as particularly democratic institutions.²⁶⁶ Second, and more importantly, both the Gendarmerie and the Carabinieri for instance, predate the modern French and Italian states—the former established in 1720 and the latter in 1814.²⁶⁷ As troubling as that may be in theoretical terms, in our view, the focus should be less on the historical existence of grandfathered paramilitary police forces in various countries, but rather on the process of militarization of previously “civilian” forces.²⁶⁸ Third, it should be noted that in any case, even formal military police, in actuality, don more “traditional” police appearance when acting for public order.²⁶⁹ Fourth, it seems that there is some tendency in Continental Europe to (at least formally) civilianize paramilitary police.²⁷⁰ Thus, Germany, for instance, gradually civilianized its cold-war era paramilitary Federal Border Guard (now Federal Police).²⁷¹ Greece disbanded the Hellenic Gendarmerie after the fall of the dictatorship.²⁷² Belgium demilitarized its gendarmerie in 1992,²⁷³ and Austria did so in 2005.²⁷⁴

The second hybrid force is “border police” or border guards, which are forces charged with national border control.²⁷⁵ Sometimes these agencies formally belong to the police or other civilian agency.²⁷⁶ Especially in non-peaceful borders, these forces are characterized by their heightened militarization.²⁷⁷ Arguably, the prevalence of such forces challenges our distinction between military and police action and symbolism. However, upon close look, militarized border police actually strengthen our argument. This is because they are acting on the border: their “target” is external to begin with – meaning their militarization signifies the exclusion of those beyond the border.²⁷⁸

B. Self-Protection of Police Officers

A second objection argues that militarization makes police safer, as they are better able to protect themselves.²⁷⁹ Indeed, advocates of militarization might claim that military equipment, such as armored personnel carriers, is only used to defend and not attack; and that despite the potential to inflict violence, police act with the utmost discretion to minimize the use of force.

Two responses are available. First, at bottom this is an empirical argument. Whether in general militarization makes police safer is a matter of dispute. For instance, while militarization allows police to respond more forcefully to violence, it is just as likely that militarization might provoke graver violent resistance to begin with. As we pointed out earlier, this “arms race,” if you will, characterized the “war on drugs:”²⁸⁰ the deployment of militarized police might incentivize criminals to organize and arm themselves in a manner that will allow them to counter militarized forces. Thus, the instrumental argument does not seem determinative.

A second response is principled. Even if militarization did make police safer (which is disputed), one can argue that as a point of departure, a police force can only make itself safer to a certain extent. Police assume a special role in society, which requires them to assume some form of risk. Indeed, if police were not required to take any risk, they would have not been required to use force only as a last resort but could act preventively. As David Luban points out, “[s]ome professions [such as police] include risk taking in their vocational core.”²⁸¹ This role-based requirement for risk-taking ascends from their consent to become police officers but also descends as an inherent part of their professionalism.²⁸² This is precisely the substance of the common police motto, “to protect and to serve.” Protection inherently means assuming risk for the sake of another, or for a greater good. It follows that police cannot follow “zero-risk” policy, but rather, the question is one of proper balance.²⁸³ We can therefore legitimately ask whether militarization, to the extent that it reduces some risk to police, can be justifiable in light of the considerations detailed above.

C. Militarized Police Deters Unlawful Acts

A further argument for militarization concerns deterrence. Indeed, some proponents of militarization may argue that militarized police forces may deter criminal and terrorist activity.²⁸⁴ However, this claim is also unconvincing, both on the instrumental and principled levels. First, it is empirically unclear whether militarization deters crime. Indeed, it seems that criminals that would be willing to use the type of violence that militarization wishes to deter—for instance terrorists—are unlikely to be deterred from such forces to begin with.

Moreover, the problem with arguments for deterrence is that they are inherently vague, are impossible to validate, and as such can be used to justify almost everything.²⁸⁵ The same forces that might deter criminals can easily be used in practice to deter legitimate dissent through pretext.²⁸⁶ Even if such abuse does not take place, a “chilling effect” might occur, as “[t]he very essence of a chilling effect is an act of deterrence.”²⁸⁷ Therefore, the mere presence of intimidating militarized forces, originally meant to deter criminals, might effectively deter individuals that seek to engage in legitimate activities.

However, to us, the key problem with the argument from deterrence in the context of militarization is a principled one. We should ask those who invoke deterrence, what is the “or what?” that underlies the presence of militarized forces? In other words, what is the stick with which the potential perpetrator is threatened by such forces? In the context of war, it is clear that the threat is death and destruction.²⁸⁸ In such cases, demonstration of military capacity might be understandable.²⁸⁹ However, in the context of law enforcement, the stick of deterrence must be in the form of justice, meaning arrest, trial, and punishment in line with due process.²⁹⁰ Deterrence by militarized police conflates the two: it applies to law enforcement deterrence the logic of war deterrence. It conveys to the policed community not the message that if you commit a crime, justice will be meted out; but rather, that if you transgress, you might encounter extreme, possibly deadly violence. It becomes clear that such perception of deterrence is unacceptable in a rule of law society.

D. Increasing the Sense of Security A further objection could be that even if all else fails, militarization

enhances the “sense of security” of the public. The public sees militarized forces and is reassured that the State is strong and able to protect it, and thus the system’s legitimacy is strengthened.

This argument fails for several reasons. Key among them is the fallacy found at the basis of broad policies which aim to achieve a subjective sense of security through preventive actions. Indeed, when a state adopts policies based on the presumption that its civilians are threatening, it actually defeats itself. Peter Ramsay convincingly identified an ongoing process whereby States attempt to ensure not only the physical security of their citizenry, but also their subjective sense of security. The problem, according to Ramsay, is that when doing so, the State admits its inability to provide a sense of security normally; thus, paradoxically, the State becomes an insecurity State.²⁹¹ In Ramsay’s words, such actions are nothing but “an authoritative statement of law’s lack of authority.”²⁹² In our context, deploying militarized forces because of a presumption of threat amounts to an admission by the State of law’s general inefficacy, of the general anxiety of the citizenry, and the its own authority deficit.²⁹³ When militarizing police, the State essentially concedes that its citizens – and even, perhaps, its police – are insecure within their own polity. This of course only generates more subjective insecurity:²⁹⁴ when the presumption of threat constructs some citizens as threatening-if-not-proven-otherwise, it simultaneously constructs others as potential victims.²⁹⁵

Even if we would assume militarization could instill some sense of security among certain (privileged) segments of society, this would hardly change our conclusion. As we established earlier, normalizing militarization assumes threat, implies a friend/enemy distinction, and therefore excludes the policed community. Now, assuming that militarized police are not deployed everywhere but selectively,²⁹⁶ the “sense of security” of community X must be achieved at the expense of the exclusion of community Y. The only way to counter this problem would be to deploy militarized police always and everywhere. In such an attempt to uphold formal equality, the state would complete the merger between democracy and totalitarianism, norm and exception.

E. If Militarization Becomes Normalized, It Would Cease to be Excluding

A related counterargument claims the following: if we are correct that militarization is a symbolic process, then can’t it be said that once normalized, the excluding power of the symbol is diminished? In other words, once militarization becomes normal, it loses its power to exclude since there will no longer be an alternative cultural backdrop to which to compare it. So, to speak, if it “excludes” everyone, it excludes no one.

While this argument makes sense theoretically, it ultimately fails to convince. This is because the alternative cultural backdrop would disappear if and only if all police forces, wherever they are deployed, become militarized. As long as in certain contexts “civilian” police are still deployed, the relative difference will continue to generate the exclusionary effect concerning those communities which face militarized units.²⁹⁷ Since, as we highlighted above, selective deployment is part and parcel of the phenomenon of police militarization, it is highly unlikely that in the foreseeable future this relative difference will disappear, and so its power to exclude will remain.

Conclusion

Like other situations in which measures that were once designed for the exceptional case become normalized, the discussion of police militarization is preoccupied with instrumental arguments. When engaged in such arguments, we frequently lose sight of the larger principled issue at hand.

As this Article demonstrated, the principled problem of police militarization is not found in its actual use of military-style violence, but rather in the presumption of threat that it implies. The presumption of threat, manifested symbolically through the tacit acknowledgment that the possibility of combat is constantly present, carves out a domestic enemy, one which is excluded from the political order. Since this distinction is carried out by the police, which is an organ of the normal, rather than the exceptional, legal regime, the distinction has an especially powerful potential to normalize and solidify this exclusion. This is what underlies the strong objections to such forces among policed communities.