PDGS - Seminar Governance, Security and Military Institutions in Democracies - The Military Role in Internal Defense and Security: Some Problems

THE MILITARY ROLE IN INTERNAL DEFENSE AND SECURITY: SOME PROBLEMS

Occasional Paper #6

María José Moyano Rasmussen
CCMR

October 1999

Draft - Not for citation or attribution without the author's permission

THE MILITARY ROLE IN INTERNAL DEFENSE AND SECURITY: SOME PROBLEMS

María José Moyano Rasmussen

A key issue in the post-Cold War world is whether military forces should be deployed in domestic law enforcement missions. Advocates of military involvement in domestic law enforcement see it as a useful strategy for avoiding immediate force reductions. In many new democracies this practice is not even questioned. Due to the traditional involvement of the military in law enforcement missions, when violence breaks out policy-makers and society-at-large assume that the military should be brought in.

This paper will analyze some of the difficulties that emerge when the military is employed in domestic law enforcement. It discusses the American military's involvement in riot control in Los Angeles in 1992, the British military's thirty-year experience of riot control and counter-terrorism in Northern Ireland, and some lessons we derive from these two cases. In Los Angeles and Britain, public officials deployed the military in law enforcement missions as an act of desperation, without giving much thought to the impact that these decisions might have on military organizations, democratic practices, and the orderly functioning of civilian law enforcement agencies.

This paper is organized into four sections. Section I describes the conditions that led to military involvement in Los Angeles and analyzes the performance of the military during that crisis. Section II does the same for Northern Ireland. Section III draws common lessons from the performance of the military in both cases, and concludes that the military were not ideally suited to these missions, which should have been performed by a well-trained police force. Finally, Section IV draws some lessons on the behavior of civilians during these crises, and provides some options for those interested in establishing a sound and democratic public order policy.

I. Military Involvement in Riot Control: The Los Angeles Riots, 1992

The Los Angeles riots of 1992, or the "Rodney King" riots, were catalyzed by the acquittal of four white Los Angeles Police Department (LAPD) officers, who were on trial for savagely beating King, an African-American. A private citizen videotaped the incident, which took place on March 3, 1991. An edited version of the videotape was constantly broadcast on national television during the twelve months that elapsed between the beating and the beginning of the officers' trial. The televised version omitted footage that showed King resisting arrest and attacking one of the police officers. The jury at the officers' trial viewed the complete videotape and found it exculpatory.(1) On the basis of the televised version, the citizens of Los Angeles, the American public and public officials including President George Bush, expected convictions.

The verdict in the "Rodney King trial" was announced on Wednesday, April 29, 1992, at 3.15 PM. One hour later, the rioting began. Even though the verdict precipitated the riots, there were two other factors that contributed to the violence: the killing of Latasha Harlins, and social conditions in South Central Los Angeles. Harlins, a black adolescent, was shot in the back of the head by Soon Ja Du; a Korean shop owner who thought Harlins wanted to steal a carton of orange juice. In October 1991, Du was convicted of voluntary manslaughter. Even though she faced a maximum of 11 years in prison and an additional five for using a gun, the judge freed her on a suspended sentence. The disposition of the Harlins case convinced many blacks in Los Angeles that they could not obtain justice through the legal system, and this view became reinforced by the verdict in the King trial. The hostility between blacks and Koreans, palpable during the Harlins trial, erupted in the riots: 1,867 of the 3,000 businesses destroyed or looted in the riots were Korean.(2)

Latasha Harlins was killed in South Central, the Los Angeles district that became the epicenter of the riots. South Central was an area "awash in poverty, drugs, and violence."(3) The district presented the usual characteristics of inner cities: unemployment, family breakdown, exodus of the middle class, increase in violent crime. What distinguished South Central and other inner city districts of Los Angeles county was the presence of a large number of heavily armed gangs. The "hard policing" tactics of the LAPD, while somewhat effective against the criminal fraternity, had alienated the law-abiding population of South Central.(4)

The city leadership and the LAPD leadership were totally unprepared for the riots. The Mayor of Los Angeles, Tom Bradley, and the LAPD Chief, Daryl Gates, expected guilty verdicts (as did everyone else), so they saw no reason to do any contingency planning. Bradley and Gates were also engaged in a personal feud. By the time of the riots, they had not talked to each other in 13 months.(5) This was a serious problem, since the LAPD is ultimately accountable to the mayor. When the verdicts were announced, Mayor Bradley left his post to attend a rally at a black church, intended to launch "Operation Cool Response." This was a well-intended initiative, the brainchild of the Reverend Cecil Murray, designed to appeal to blacks to channel their frustration through conventional political means.(6) However, it came too late, since the rioters were already taking to the streets. "Cool Response" should have been launched three or four days earlier. The jury in the King trial had been deliberating for six days (since April 23), which in itself was a sign that guilty verdicts were not a certainty. Chief Gates' conduct was even more inexcusable. When the riots broke out, the Chief left his post to attend a political fund-raiser.(7)

The LAPD was in no position to face the rioters. Most of the senior officers (captains) in the different divisions were away from their desks, on a three-day course outside Los Angeles.(8) Shifts in many divisions changed at 3 PM, so when the verdicts were announced, the LAPD had a total of 838 officers on duty.(9) These officers had no civil disturbance equipment, no reserve ammunition, and no riot control plan. The only LAPD unit that had undertaken minimal riot control training was the elite Metropolitan Division. However, "Metro had 233 officers, less than 3 percent of the LAPD's sworn personnel."(10) In addition, on the day the riots began, Metro officers were told to report to duty at 6 PM, by which time the riots were in full swing.(11) The riots started as two separate incidents of vandalism. Individuals involved in both incidents converged on the intersection of Florence and Normandie avenues, in South Central. Thirty police officers in 18 patrol cars responded to the disturbance. At Florence and Normandie, these 30 officers confronted a crowd of some 200 people throwing bricks, pieces of concrete and other improvised missiles. A decisive show of force might have stopped the riots at this juncture.(12) Standard riot control doctrine, in the LAPD and elsewhere, calls for the cordoning off of the intersection with patrol cars. Properly equipped with riot gear, and in sufficient strength, police then lock arms and advance. However, at 5.45 PM the LAPD abandoned the intersection. While the riots spread and the local television stations showed live images shot from helicopters, LAPD officers assembled at an emergency command post that had no televisions, no computers, and few radios or telephones.(13)

Given the LAPD's initial paralysis, standard procedure called for the implementation of the Mutual Aid system. "Mutual aid in California is a well-designed system of law enforcement officers helping neighboring law enforcement jurisdictions when situations escalate beyond local capabilities."(14) Mutual aid agreements between law enforcement agencies and other emergency services dictate that if a particular sheriff or police department or fire department is overtaken by events, the state Office of Emergency Services coordinates the provision of reinforcements from other jurisdictions. However, having failed to take any precautions against rioting, civilian officials panicked and overreacted. Mayor Bradley and the governor of California, Pete Wilson, circumvented the Mutual Aid system and called in military forces. Two thousand California National Guard soldiers were mobilized that evening, and an additional 2,000 were mobilized on the second day of rioting, April 30.(15)

The National Guard is a uniquely American organization. Guard units are funded at the federal level and are considered part of the military reserves. But they work for the state governors. Also, a small portion of the Guard officer corps has full-time military jobs. The vast majority of Guard personnel lead civilian lives and perform military duties and training during short periods each year.

As this narrative will argue, the Los Angeles riots should have been met by a well-trained police force. Soldiers did not necessarily represent the best response to this crisis. However, if soldiers were going to be employed at all, it can be said that guardsmen were the soldiers best suited to this task, for a number of reasons. First, the initial 2,000 mobilized lived in the neighborhoods affected by the riots, and were familiar with the terrain. Second, many of these citizen soldiers held civilian jobs in law enforcement or related areas. Third, the California National Guard has vast institutional experience responding to emergencies, such as fires or floods, always in aid of civilian law enforcement. Finally, many of the older guardsmen had been deployed during the Watts riots of 1965, also in Los Angeles.(16) However, guardsmen were not given a chance to do their job. On the third day of rioting, May 1, President Bush authorized the deployment of Regular Army troops and Marines, and federalized the California National Guard. By May 2, 10,465 guardsmen, 2,023 Army soldiers and 1,508 marines were in Los Angeles, in addition to 1,717 other federal law enforcement officers.(17)

This massive deployment of troops was unnecessary because the rioting was over after the first night. During the second day (Thursday, April 30) the rioting gave way to massive looting. There was only sporadic looting and violence during the third day
(Friday, May 1).(18) By the time federal forces arrived in Los Angeles on Saturday, May 2, the problem was not the restoration of order but the maintenance of order. Thousands of troops stood by waiting to be assigned missions. This is not meant to suggest that the human and financial costs of the riots were not enormous, or that the police officers, firemen and soldiers did not face great dangers. The statistics in Table 1 are clear. Rather, the argument here is that troop commitment was excessive, particularly in view of the fact that the Mutual Aid system had hardly been invoked.

Table 1 - The Toll of the Los Angeles Riots, 1992

Date	Deaths	Injuries	Fire Responses
Wednesday April 29	10	n/a	601
Thursday April 30	21	1,000 (cumulative Apr.29-30)	3,244
Friday May 1	12	257	1,172
Saturday May 2	4	859	517
Sunday May 3	4	267	5,579
Monday May 4	3	n/a	n/a

Source: James D. Delk, Fires and Furies. The L.A. Riots (Palm Springs, CA: ETC Publications, 1995), pp.46, 95, 138, 180, 222.

In addition, all of the missions assigned to military troops were strictly law enforcement missions - i.e., tasks normally performed by police officers (and which the LAPD should have been able to perform had it been properly equipped and trained.) Guardsmen, soldiers and marines patrolled neighborhoods not yet affected by the violence, to prevent the spread of the riots, and protected shopping malls under threat of arson or looting. They stopped looting in progress at some stores, and provided protection against further looting. They protected the post office when postal workers handed out welfare checks, and escorted firefighters who went out on missions. Guardsmen also directed traffic and accompanied police detectives as they recovered some of the loot.(19)

Military troops performed all these missions with enormous restraint, and earned the recognition of the citizens and public officials of Los Angeles.(20) However, these were not missions for which the military were adequately trained or equipped. This was the assessment of Major General Marvin L. Covault, commander of all military forces in Los Angeles, who sought to extricate his troops from the city almost from the moment he got there.(21)

There were several reasons why the military were not ideally suited to these missions. Unlike the LAPD or the Los Angeles County Sheriff's Department, soldiers, marines and guardsmen from Northern California were unfamiliar with Los Angeles neighborhoods and were not even equipped with adequate city maps.(22) Their military radios were not suited to a city environment.(23) Their intelligence estimates were flawed because police intelligence, which relies heavily on informers and community contacts, was alien to them.(24) More important, the troops did not know "how to recognize 'the enemy.'"(25) They were uncomfortable in the role of mediators between Korean shop owners, armed to the teeth and threatening to shoot anyone who approached their businesses, and the crowds. The troops had little knowledge of gangs and found it difficult to distinguish between gang members and simple adolescents with an attitude.(26)

Above all, the troops were unfamiliar with the inner city environment and its crime problem. On average, there are four murders per day in Los Angeles county, "where death by gunfire per 100,000 inhabitants is more than triple the national average."(27) On numerous occasions while they were on patrol between May 2 and May 7, troops suffering verbal abuse by gang members or hearing shots in the distance thought they were in imminent danger, only to be told by police officers that crime and violence levels were much lower than normal.(28) On another occasion, marines accompanied police officers responding to a domestic dispute. On receiving a request to "cover me," the marines fired over 200 bullets into a house where there were children.(29) The problem on this occasion was one of vocabulary as well as training: in police parlance, "cover me" means "have your weapon ready and prepare to fire only if I am in danger;" whereas in marine parlance, "cover me" means "open fire at once." This points to the most important difference between military troops and police officers. Police officers are trained to respond to crime and violence with the minimum force necessary to accomplish the task. This principle of minimum force is alien to a soldier.

Every guardsman, soldier and marine deployed in Los Angeles had to sign a copy of very restrictive Rules of Engagement, in order to acknowledge his/her understanding and acceptance of these rules. The Rules of Engagement dictate, for any military operation, the conditions under which force will be used. During the Los Angeles riots, the Rules of Engagement stated that:
The use of deadly force is authorized only where all three of the following circumstances are present:

(30)

To indicate that deadly force can be used after "all other means have been exhausted" implies that non-lethal weapons were available to military troops, but this was not the case. Some, but not all troops were equipped with riot batons and face shields. No other standard, non-lethal riot and crowd control means (water cannon, tear gas, rubber bullets or dogs) were available.(31) Nor had military troops (with the exception of some Guard units) trained in the use of these non-lethal weapons. In practice, the soldiers, marines and guardsmen had few options other than the use of deadly force. Table 2 shows the Arming Orders, which dictate what equipment and/or weaponry will be used in a given military operation.

Table 2 - Arming Orders during the Los Angeles Riots

Arming Order	Rifle	Bayonet	Pistol	Baton	Magazine/Chamber	Control (see below)
AO-1	Sling	Scabbard	Holstered	Belt	In pouch/Empty	OIC/NCO
AO-2	Port	Scabbard	Holstered	Belt	In pouch/ Empty	OIC/NCO
AO-3	Sling	Fixed	Holstered	Hand	In pouch/ Empty	OIC/NCO
AO-4	Port	Fixed	Holstered	Hand	In pouch/ Empty	OIC/NCO
AO-5	Port	Fixed	Holstered	Hand	In weapon/ Empty	OIC/NCO
AO-6	Port	Fixed	In Hand	Belt	In weapon/ Locked	OIC

Control:
OIC= Officer in charge
NCO= Noncommissioned officer in charge
Source: James D. Delk, Fires and Furies. The L.A. Riots (Palm Springs, CA: ETC Publications, 1995), p.343.

Between Thursday (April 30) and Saturday (May 2), the California National Guard patrolled the streets fully armed, at Arming Order 5. This changed with the arrival of federal forces, when General Covault ordered all troops to change to Arming Order 1, which made it more difficult for the soldiers to use lethal force. This decision has been heavily criticized, on the grounds that it put soldiers at risk and impacted morale.(32) However, General Covault made a wise and courageous decision. His decision was courageous because it put his troops at risk, and wise because it likely averted bloodshed. His soldiers confronted heavily armed gang members. But they also confronted crowds armed only with bottles or rocks; adolescents and children; looters who lived below the poverty line and who were committing a crime out of desperation; and irresponsible citizens out for thrills who wanted to photograph the action. What is the adequate response? How does one chose between the prevention of crime, the preservation of order and the preservation of life? Police officers are trained to make these choices on a daily basis, and are also trained to react to street conditions in ways that soldiers are not. This was recognized by National Guard commander General James Delk, who told his troops to "listen to your [police] counterpart, they know the territory and the threat."(33) The military intervention in the Los Angeles riots could have resulted in a bloodbath. Deploying soldiers in the streets armed with rifles and bayonets does not allow for a gradual escalation of the state's response. What is a soldier to do but shoot after he has issued his warning to looters? That the military intervention in Los Angeles did not result in the death of many civilians is due to Covault's decision to expose his soldiers, as well as to the restraint shown by those men.

It must be pointed out that decisions on the Rules of Engagement were left to all intents and purposes in the hands of the military, and the enforcement of those Rules of Engagement was at best haphazard. The Rules of Engagement and Arming Orders were devised by Generals Delk and Covault. Both senior officers based themselves on existing Department of Defense operational plans and regulations. Since civilian officials have at some point approved these plans and regulations, it could be said that there was some civilian input, albeit indirect. But there is no evidence that civilians in California (the mayor or the governor) provided any input or evinced any interest in the issue of the Rules of Engagement or how these were enforced. There is also substantial evidence that during the riots different units operated at different arming levels, in spite of Covault's orders, and that this was widely known at the time. Considering that the likely outcome was the civilian loss of life, this issue constitutes a serious indictment of the civilian leadership of Los Angeles.

In fact, one constant in the military intervention in Los Angeles was that the military operated by and large devoid of any civilian control. This was not the military's fault. The military were obliged to make decisions that should have been made by civilian law enforcement or political leaders, because of the lack of any civilian coordination, or of an established law enforcement chain of command.

In 1980, city authorities in Los Angeles passed an ordinance that created an Emergency Operations Plan. According to this plan, an Emergency Operations Organization (EOO), headed by the mayor and with representatives from every major city agency, was supposed to plan for contingencies such as riots before they occurred. The EOO was also expected to assume command of operations once an emergency developed. Before and during the 1992 riots, the EOO "should have been the nerve center of the City's emergency response. In this case, however, the EOO was dysfunctional."(34) Many city officials were unfamiliar with the emergency legislation, and therefore unaware that they had a role to play. When, in agreement with the Emergency Operations Plan, the mayor opened an Emergency Operations Center in the basement of City Hall, many city officials who should have been involved were unaware of this development. In addition, since the Emergency Operations Center did not have adequate communication with the outside world, it could hardly issue orders or respond to events as they unfolded. Therefore, the military were left with no choice but to respond to ad hoc requests from law enforcement officers in the field, and to exercise their own initiative.

Civilian officials in California called in the military because they perceived the LAPD was incapable of providing a response, and then completely abdicated their responsibility and left the military to fend for themselves. To some extent, this was a problem of leadership. But the LAPD's shortcomings were also the result of long-term trends for which officials and citizens of California bear responsibility. In 1978, California voters approved Proposition 13, a measure that froze property taxes and drastically reduced funds available for public services. Over the next fifteen years, the LAPD appealed to city officials and to the voters, but though everyone wanted improved police services, nobody was willing to pay for them. As a consequence, by the time of the riots the LAPD had less than two officers per 1,000 inhabitants, which is under half the ratio in other large cities. LAPD training facilities were in disrepair, training standards had collapsed, and the force suffered equipment shortages and malfunctions.(35) It should also be said that the LAPD, devoid of any meaningful civilian control, had lost all legitimacy among the population of South Central.(36) Resorting to the military in order to quell disorders has one added disadvantage. The military can become a crutch that absolves civilian officials from the responsibility of providing for and overseeing the work of the police and its relations with the community it is expected to serve and protect. Some of these issues will be developed further in the concluding sections.

II. Military Involvement in Counter-terrorism: The British Army in Northern Ireland, 1969-2000

The Irish have been fighting British domination for 300 years. In 1921, this led to the partition of the island and eventually to the creation of an independent Republic of Ireland in the south. The six counties of Northern Ireland (also known as Ulster) remain part of the United Kingdom to this day. For fifty years following partition, Ulster voters elected representatives to a local parliament known as Stormont, as well as to the British parliament in London. Catholics, who constituted one third of the population of Northern Ireland in the 1960s, labored under severe discrimination. "Demands for reform centred particularly on local government where gerrymandering had ensured Unionist control even in boroughs like the city of Derry with a Catholic majority. Catholics were sharply discriminated against in the allocation of local authority jobs and housing, and even the local franchise ... gave businessmen (who were usually Protestant) additional votes according to the value of their property."(37) (Political language in Northern Ireland differentiates between Protestants, unionists and loyalists. A Protestant is someone raised in that religion. Most, but not all, Protestants are unionists, which means they favor the union with the United Kingdom. Loyalists are those unionists who advocate or practice violence in order to further their cause. A similar distinction exists between Catholics, nationalists and republicans. Catholics are individuals raised in that religion. Most Catholics are nationalists, which means they want the reunification of Ireland, north and south. Republicans are those nationalists who advocate or practice violence in order to achieve this end.)

In 1967, Catholics formed the Northern Ireland Civil Rights Association (NICRA) in order to campaign against discrimination. Encouraged by lengthy coverage in the international media, NICRA thrived. Protestants, who dominated the Stormont parliament, tenaciously clung to their privileges, and interpreted the mildest demand for reform as an assault on the constitutional order. Street violence escalated rapidly in 1968-1969, between Catholics and Protestants and Catholics and the police, mainly drawn from the Protestant community. Demonstrations, counter-demonstrations, petrol bombs, and barricades became a daily occurrence. It was felt that the local police, the Royal Ulster Constabulary (RUC), could not cope with the disorders. Two factors gave rise to misgivings. First it was recognized that the RUC was severely under-staffed. Second, there were doubts about the RUC's political neutrality. Television crews had captured images of the RUC siding with Protestant demonstrators against the Catholics.

After one night of rioting (August 14-15, 1969) left 10 civilians dead and another 145 wounded,(38) the British government decided to send in the army: "The General Officer Commanding Northern Ireland [the GOC] has been instructed to take all necessary steps, acting impartially between citizen and citizen, to restore law and order. Troops will be withdrawn as soon as this is accomplished."(39) Thirty years later, the troops are still there, though their duties changed dramatically - from public order to counter-terrorism - after the first three years.

In August 1969 the British army took to the streets of Belfast and Derry to separate the two warring tribes, and to protect the Catholics from the ferocity of the Protestants. The army also engaged in a number of community activities. It organized sporting events for young people, ran discos, even helped the elderly with household chores.(40) Through patient negotiation with community leaders, the army managed to "talk down" the barricades that Catholics and Protestants had erected to protect their neighborhoods from the opposing side. But as soon as a barricade went down, another one came up. The barricades "created areas of virtual self-government."(41) The existence of these barricades allowed the terrorist organizations, which had stagnated for years, to reorganize and rearm. The barricades came finally down in 1972 as a result of Operation Motorman, the largest British military operation since the Korean War. By this time, an effective program of reform had appeased NICRA. Riots have occasionally erupted over the years, but they have never again been the daily occurrence they were in 1969-1972. Therefore, by 1972 the army's main concern was no longer public order but terrorism, primarily by the Irish Republican Army (IRA), and to a lesser extent by the loyalist Ulster Defense Association (UDA) and smaller groups.(42)

During this period, the army's relationship with the Catholics also changed dramatically. "When the British Army arrived on the streets of Northern Ireland in 1969 in response to pleas from Catholic politicians, in traditional Republican areas soldiers were welcomed much like the troops who arrived to liberate Paris in 1944."(43) Photographs and television news coverage of British soldiers on patrol in Belfast and Derry in 1969 depict an endless stream of Catholic housewives bringing them tea and muffins. This relationship was soured in July 1970, when the army decided to impose a three-day curfew in the Catholic enclave of Falls Road in Belfast, in response to a tip-off that there were arms hidden there. The whole area was cordoned off. Citizens were confined to their homes without access to basic staples such as bread and milk while the army looked for weapons. The house searches uncovered a not insignificant number of guns and explosives, but in street clashes with demonstrators, the army killed four civilians and injured 68. Whether the operation could be considered a military success, in terms of the number of weapons seized, is open to question. In political terms, it was a disaster, as well as a turning point in the army's relationship with the Catholics. As one soldier remembers, "[t]he week before the curfew, I'd actually marched my troops unarmed into the Falls to a Catholic church to go to church to show that we had confidence in [the Catholics]. We were unarmed and we marched down the road and into the church. A week later, it was a reverse situation. I don't believe the Army would have marched down since that day, unarmed down the road."(44)

A second event that damaged the army's relationship with the Catholic population was the introduction of internment without trial (indefinite detention without charges) in August 1971. The decision to introduce internment was made by the Stormont and British governments, but it was the army that executed it poorly. Internment was a failure for several reasons. Intelligence gathering by the RUC was deficient. Rumors about internment had been floating around Belfast for months. Lieutenant General Harry Tuzo, GOC Northern Ireland, had even discussed the pros and cons of internment in an interview with the Belfast Telegraph. Therefore, when internment came, the real gunmen were in hiding. Internment was also seen as biased, since there was no Protestant among the 346 persons lifted initially. Soon there were rumors of the army's ill-treatment of internees during interrogation.(45) The army made its third serious mistake on January 30, 1972, when soldiers of the Parachute Regiment killed 13 unarmed civilians during a demonstration against internment in Derry, an incident which became known as "Bloody Sunday" and which has never been properly investigated or explained.(46)

The combined effect of the Falls curfew, internment and "Bloody Sunday" was to drive Catholic moderates into the arms of the IRA. In fact, it could be said that the army confronted after 1972 an enemy of its own making. In 1969, when the army was sent to Northern Ireland, the Catholics were fighting to end discrimination, not to gain independence from Britain. It was to a large extent the behavior of the army, described in the preceding paragraphs, which nurtured a climate favorable to the rebirth of the IRA. Thus, the army lost the battle for hearts and minds at the start of its counter-terrorist campaign. It should also be pointed out that the army was sent into Northern Ireland without much direction. The British government had only one goal: to avoid having to govern the province directly. London simply continued to pour more troops into the province (see Table 3) to bolster the Stormont government, in the hope that the soldiers would find a military solution to what was (and remains) a political problem, and that order would be finally and somewhat miraculously restored. The violence that accompanied the introduction of internment (see Table 4) and the debacle of "Bloody Sunday" forced the British government to abolish the Stormont parliament and assume direct control, which it maintains to this day.(47)

The ad hoc nature of political and military responses was influenced by the fact that all the participants thought that the conflict would end soon. The IRA, boosted by its recruitment after the Falls curfew, believed victory was at hand. Britain had recently abandoned Aden as a consequence of a terrorist campaign that resulted in the deaths of 36 soldiers. The IRA believed that if it killed the same number of soldiers, the British government would pull out.(48) The British government and sectors of the army though that the soldiers could win a military victory over the IRA.(49) By the end of the 1970s, the participants had a clearer understanding of the situation. The IRA settled into a "long war" strategy in 1977, and the following year a classified army intelligence document (which the IRA obtained and published) stated that the IRA's "campaign of violence is likely to continue while the British remain in Northern Ireland.... We see little prospect of political development of a kind which would seriously undermine the [IRA's] position."(50)

The IRA has been a formidable adversary. The organization moved from car bombs to letter bombs to sophisticated devices. In the early 1970s, while it still kept Ulster as the focus of its activity, the IRA began to stage operations on the British mainland. From the early 1970s, and until all terrorist organizations (Catholic and Protestant) declared a cease-fire in 1997, the IRA periodically targeted pubs, department stores, parks, and the financial district of London, as well as British army barracks on the European continent. The IRA routinely attacked members of the security forces in Northern Ireland, and occasionally killed a high profile target, such as the queen's uncle, Earl Mountbatten. The group almost succeeded in killing Prime Minister Margaret Thatcher at the Conservative Party annual conference in 1984, and during the Gulf war (1991), it carried out a mortar attack against 10 Downing Street while Prime Minister John Major presided over a cabinet meeting.

As far as the British government and the security forces are concerned, it is the IRA that drives security policy. Loyalist terrorists are not perceived as constituting the same type of threat, for a variety of reasons. Loyalist violence is reactive. It responds to IRA violence. Loyalist have killed and bombed less frequently than the IRA (see Table 4). More important, Loyalist terrorists confine themselves to killing Catholics, and refrain from attacking the security forces.(51)

The ad hoc nature of political and military responses to violence extended to the legal and practical definitions of the soldiers' role. In 1969, the army marched into Northern Ireland as "military aid to the civil power" - in other words, as support to the police. In practice, the military took over the responsibility for public order and counter-terrorism, and the police became subordinate to the army. The RUC resented and resisted its de facto subordinate role. As a consequence, army-police relations were strained during most of the 1970s, and the counter-terrorist effort suffered. During this period, the police was relegated to fighting what the Irish humorously term "ordinary decent criminals" in Protestant areas.(52) In Catholic areas, the army was the law. The army was also in control of every aspect of counter-terrorism. It operated a system of saturation patrolling, manned rural and urban observation posts and vehicle checkpoints, monitored public gatherings, carried out house searches, ran a network of informers, and reorganized intelligence gathering.(53)

In 1977, the government decided to inaugurate the "era of police primacy," not out of a desire to democratize law enforcement. What motivated the government was the realization that the pervasive military presence in the streets gave credence to the "army of occupation" argument and allowed the IRA detainees to refer to themselves as prisoners of war. The army resisted the advent of police primacy forcefully, yet for almost a decade the change was merely cosmetic. It was only in the mid-1980s that the police took over duties such as manning vehicle checkpoints, house searches, or the policing of demonstrations, marches and political funerals.(54) The army continues to have responsibility for the policing of rural areas, and of republican enclaves in cities. The army will also intervene, at the request of the RUC, in the policing of major public gatherings.(55) On these and all other occasions, the army retains operational independence. More important, since the mid-1980s, the army has been increasingly devoted to undercover operations in plainclothes, and to surveillance of civilians.

Under normal circumstances, the advent of police primacy in security policy should be considered a democratic advancement, since it will lead to a "civilianization" of law enforcement, and to the adoption of more flexible and open public order responses. However, according to observers of the Ulster conflict, the advent of police primacy resulted in the "militarization" of the police, in the sense that the police has adopted tactics, practices and equipment borrowed from the military, and has also benefited from the increasing secrecy clouding security operations in the province. The only real change that the era of police primacy seems to have brought about, is that there is now regular consultation and cooperation between the General Officer Commanding and the Chief Constable, extending down both chains of command.

It is easy to conclude that the term "police primacy" is a little ambiguous, and that "the criteria for military intervention remain open to fairly arbitrary change with little opportunity for proper parliamentary scrutiny, and the legal relationships between soldiers, police and other civil authorities are far from clear."(56) Scholars and journalists argue that the ambiguity in police-military relations, and the substantial autonomy with which the police and military operate, suit the civilian officials, who can dissociate themselves from mistakes, particularly when these result in loss of civilian life. Whether or not this is true, there is evidence that many security decisions are made without much civilian input, either from the Northern Ireland Office (London's executive arm in the province) or from the civilian body appointed to oversee the RUC.

Has the army been effective? What are the lessons that can be derived from the army's 30-year experience in counter-terrorism? There are two ways to answer these questions. The first is to look at some statistics. Tables 3 - 5 provide figures on force structure and on incidents of violence in Northern Ireland.

Table 3 - Security Forces Personnel in Northern Ireland, 1969-1999

Year	British Military	UDR/RIR	RUC	RUCR	Total RUC	Total
1969	3,000 (a)	------	3,000 (c)	-----	3,000	6,000
1970	11,243 (a)	4,008 (b)
1971	7,742 (a)	6,786 (b)
1972	14,218 January (a) 21,688 July (a)	9,074 (b)	4,200 (d)	2,500 (d)	6,700	29,992 Jan. 37,462 July
1973	16,854 (a)	7,982 (b)	4,500 (c)	2,500 (c)	7,000	31,836
1974	15,702 January (a) 17,000 May (a) 14,217 November (a)	7,795 (b)
1975	13,913 (b)	7,861 (b)
1976	13,672 (b)	7,769 (b)
1977	13,632 (b)	7,843 (b)
1978	13,600 (b)	7,900 (b)
1979	12,986 (c)
1980	12,141 (c)	8,000 (c)	7,000 (c)	4,500 (c)	11,500	31,641
1981	11,098 (c)	7,500 (f)	7,300 (f)	4,800 (f)	12,100	30,698
1982	10,538 (c)	7,500 (e)	8,000 (e)	2,500 (e)	10,500	28,538
1983	9,882 (c)	7,000 (e)
1984	9,090 (c)		8,000 (e)	4,405 (e)	12,405
1985		6,700 (e)	8,259 (h)	4,508 (h)	12,767
1986	10,086 (c)	6,530 (c)	8,234 (h)	4,414 (h)	12,648	29,264
1987		6,531 (h)	8,236 (h)	4,650 (h)	12,886
1988		6,393 (h)	8,231 (h)	4,649 (h)	12,880
1989	10,500 (i)	6,230 (h)	8,259 (h)	4,623 (h)	12,882	29,612
1990	10,500 (m)	6,043 (h)	8,231 (h)	4,544 (h)	12,775	29,318
1991		6,276 (h)	8,217 (h)	4,560 (h)	12,777
1992	11,000 (k)	5,417 (h)	8,478 (h)	4,593 (h)	13,071	29,488
1993		5,412 (h)	8,464 (h)	4,572 (h)	13,036
1994	13,500 (h)	5,241 (h)	8,493 (h)	4,690 (h)	13,183	31,924
1995	12,000 (h)
1996
1997		5,000 (j)
1998	12,000 (g)	5,000 (g)
1999		5,500 (h)	8,489 (l)	4,967 (l)	13,456

Note: The Ulster Defence Regiment and the Royal Irish Rangers merged in 1992, and became the Royal Irish Regiment. Technically the RIR is part of the Regular Army, but only serves in Northern Ireland.
Source:
(a) Michael Dewar, The British Army in Northern Ireland (London: Arms and Armour Press, 1997), p.105.
(b) David Butler, British political facts, 1900-1979 (London and Basingstoke: The Macmillan Press, 1980), p.435.
(c) Tom F. Baldy, Battle for Ulster. A Study of Internal Security (Washington, DC: National Defense University Press, 1987), pp.11, 106.
(d) Christopher Hewitt, The Effectiveness of Anti-Terrorist Policies (Lanham, MD: University Press of America, 1984), p.79.
(e) Desmond Hamill, Pig in the Middle. The Army in Northern Ireland 1969-1985 (London: Methuen, 1986), pp.265, 267, 271, 273, 299.
(f) Northern Ireland: Problems and Perspectives (London: Institute for the Study of Conflict, Conflict Studies # 135, 1982), pp.20, 21.
(g) Ministry of Defence, "The British Army in Northern Ireland - The Armed Forces in Northern Ireland." Available at http://www.army.mod.uk/army/world/nireland/n_armf.htm (August 13, 1999).
(h) CAIN Web Service, "Background Information on Northern Ireland Society - Security." Available at http://cain.ulst.ac.uk/ni/security.htm (July 30, 1999).
(i) Glenn Frankel, "An Anniversary That's No Cause for Celebration," Washington Post National Weekly Edition, August 21-27, 1989, p.15.
(j) Ministry of Defence, "The British Army in Northern Ireland - The Royal Irish Regiment." Available at http://www.army.mod.uk/army/world/nireland/n_roir.htm (July 30, 1999).
(k) John Strawson, "The Thirty Years Peace," in David Chandler and Ian Beckett (eds.), The Oxford History of the British Army (Oxford: Oxford University Press, 1996), p.357.
(l) CAIN Web Service, "Background Information on Northern Ireland Society - Law and Order." Available at http://cain.ulst.ac.uk/ni/order.htm (August 5, 1999).
(m) Ronald Weitzer, Policing under Fire. Ethnic Conflict and Police-Community Relations in Northern Ireland (Albany, NY: State University of New York Press, 1995), p.74.

Table 4 - Deaths in the Northern Ireland Conflict, 1969-1993

Year	Killed by Republican Terrorists	Killed by Loyalist Terrorists	Killed by the Security Forces	Other Killings	Total
1969	3	3	10	0	16
1970	15	2	5	2	24
1971	96	20	45	9	170
1972	261	111	83	17	472
1973	131	86	32	3	252
1974	144	125	19	6	294
1975	120	121	7	9	257
1976	155	114	16	10	295
1977	75	25	8	2	110
1978	56	10	10	1	77
1979	101	17	2	0	120
1980	51	14	9	6	80
1981	68	12	17	13	110
1982	83	14	12	1	110
1983	61	10	12	2	85
1984	47	7	12	3	69
1985	47	4	5	0	56
1986	39	15	5	1	60
1987	70	15	9	3	97
1988	70	23	10	1	104
1989	54	18	3	0	75
1990	52	19	10	0	81
1991	49	40	6	0	95
1992	40	38	10	1	89
1993	38	48	0	1	87
Total	1,926 (58.63%)	911 (27.73%)	357 (10.87%)	91 (2.77%)	3,285 (100%)

Source: Malcolm Sutton, Bear in mind these dead ... An Index of Deaths from the Conflict in Ireland 1969-1993 (Belfast: Beyond the Pale Publications, 1994), p.206.

Table 5 - Security Incidents in Northern Ireland, 1969-1994

Year	Shootings	Explosions	Devices Neutralized	Firearms Found	Explosives Found (kg)
1969	73	9	1	14	102
1970	213	153	17	324	305
1971	1,756	1,022	493	716	1,246
1972	10,631	1,382	471	1,259	18,819
1973	5,019	978	542	1,313	17,426
1974	3,208	685	428	1,236	11,848
1975	1,803	399	236	820	4,996
1976	1,908	766	426	736	9,849
1977	1,081	366	169	563	1,728
1978	755	455	178	393	956
1979	728	422	142	300	905
1980	642	280	120	203	821
1981	1,142	398	131	357	3,419
1982	547	219	113	288	2,298
1983	424	266	101	166	1,706
1984	334	193	55	187	3,871
1985	238	148	67	173	3,344
1986	392	172	82	174	2,443
1987	674	236	148	206	5,885
1988	538	253	205	489	4,728
1989	566	224	196	246	1,377
1990	557	166	120	179	1,969
1991	499	231	137	164	4,167
1992	506	222	149	194	2,197
1993	476	206	83	196	3,944
1994	348	123	99	178	1,285

Source: CAIN Web Service, "Background Information on Northern Ireland Society - Security." Available at http://cain.ulst.ac.uk/ni/security.htm (August 18, 1999), p.3.

As Table 3 shows, over the years the conflict has involved a significant troop commitment. There seems to have been no conscious decision, on the part of the British government, to adjust force structure to the level of actual violence. There is little variation in troop levels between 1972, when there were 472 deaths, and 1993, when there were 87. For the British government, more is always better, irrespective of the irritant factor that the troops' presence represents. Advocates of the use of the military in domestic counter-terrorism should pay close attention to the figures in Tables 4-5, which put the military's effectiveness into question. Clearly, the highest levels of violence coincide with the period of military primacy in counter-terrorism, and violence decreased significantly in the era of police primacy. More important, the figures above demonstrate that there can be no military solution to a problem that is essentially political.

Advocates of the use of the military in domestic counter-terrorism should also consider that the British army, viewed by many as the world's most expert counter-terrorist force, cannot do more than contain the problem. This is not necessarily what the soldiers in Ulster believe. According to one sergeant, "[w]e've all basically got the same idea of how we could end this, but world opinion wouldn't stand for it." A senior officer believes that "[a] majority of officers are now strongly backing a shoot-to-kill policy as the principal method of defeating the IRA. Otherwise we shall be marching toward an eternity of duty in Northern Ireland."(57) These statements run counter to key findings of terrorism research. Through studies on the life histories of former terrorists we have learnt that repression will only strengthen the individual terrorist's resolve, increase the group's cohesiveness, and even encourage the terrorist organization to stage further acts of violence.(58) In fact, there are only two cases, Argentina and Peru, in which military repression succeeded in putting an end to a terrorist campaign. In both cases, this involved massive human rights violations. In a democratic regime, these "dirty wars" are not an option. Therefore, unless repression is conducted on such a scale that leads to the extermination of the terrorist organization (which in a democracy cannot happen), it will be counterproductive, since it will only stimulate more violence.

The issue of the effectiveness of a military response to terrorism can be approached from a different angle, by looking at the effects on military organizations and on democratic practices. Colonel Michael Dewar, a veteran of several Northern Ireland tours, refers to it as "a distraction" from the British army's main purpose, the defense of Europe through NATO.(59) The Ulster campaign has involved an enormous expenditure in specialized equipment and requires a complex system of relief that disrupts army life. The army brass initially assumed that Northern Ireland would provide useful training in infantry tactics. However, "[n]ot only was the infantry involved; armoured, artillery, engineer regiments, and the other corps all took their share.... We should not underestimate the profound influence of Ulster on soldiers' lives. This seemingly endless commitment calls much of the tune. It conditions deployment, training, movement and logistics; it affects morale; it shapes the soldier's life."(60) Therefore, from the perspective of a proper and efficient use of military resources, it would seem that the British army's involvement in Ulster has had mostly adverse effects.

The commitment of troops to Ulster has also negatively affected British law and British democracy. Over the years, the British government has felt it necessary to cloud in secrecy different aspects of the security situation in Northern Ireland. The Northern Ireland Office has refused to make available basic data, such as the number of deaths attributable to the security forces and to the terrorists. The Director of Public Prosecutions has refused to make public a list of killings by the security forces that have resulted in criminal proceedings.(61) The army's doctrine for internal security is "unavailable for public scrutiny."(62) Even the "Yellow Card," the one-page instruction which tells the soldier when to use lethal force, is classified.(63) The government routinely turns down requests for this type of information on the grounds that it would not be in the public interest to make this information available, while in reality the very opposite is true. Disclosure of this type of information is not only in the public interest; it is also a vital means to achieve democratic civilian oversight of internal security policy. The British government's desire for secrecy has merely stimulated investigative journalists and scholars to obtain data and documents by other means. In turn investigative journalists have been threatened with onerous libel laws and with prosecution under the Official Secrets Act, a draconian and, according to legal scholars, very vague piece of legislation seeking to punish the disclosure of information pertaining to national security.(64)

The government has also felt it necessary to introduce a number of legal reforms in order to protect soldiers from criminal prosecution. Between 1969 and 1993 the security forces killed 348 civilians and 9 soldiers engaged in undercover work. Of those civilians killed, 154 were members of terrorist organizations and 194 were totally unconnected with terrorist activities. These 194 individuals died because they were mistaken for terrorists, or they were in the vicinity of a gun battle, or they were shot during street disturbances. The RUC killed 51 civilians. The other 297 (85%) were killed by the military.(65) A very high proportion of these killings have occurred in disputed circumstances, "but criminal prosecutions of those responsible have been brought in only 28 cases, only two of which have to date resulted in a conviction."(66)

In 1969, the army hierarchy wanted a guarantee that no soldier posted to Northern Ireland would face criminal charges. This the government could not promise. Instead, what has evolved is an implicit agreement between the government, the army brass and the soldiers, that the government will try to prevent those prosecutions. It is in this context that a number of legal reforms were introduced. Unlike the police, after an incident that results in deaths the soldiers abandon the scene. They only submit written statements about the incident, after extensive consultation with the Army Legal Service. Mark Urban examined a variety of statements issued after incidents between the late 1970s and the late 1980s, and concluded that since the statements were all so similar, they were probably not entirely truthful. In mainland Britain, whenever there is a death in disputed circumstances, a coroner is required to call an inquest. Supported by a jury, the coroner issues a verdict as to the cause of death, and recommends further legal action. The relatives of the deceased and their legal counsel are entitled to question all witnesses to the incident. However, in Northern Ireland, soldiers who have killed civilians while on duty are exempted from testifying at inquests. Coroners' inquests in Northern Ireland are not empowered to issue verdicts as to cause of death, or make any recommendations as to legal action, and the juries are selected by the police. It is easy to conclude as many have done in Ulster, that the law affords soldiers protections not available to ordinary citizens.(67) This de facto legal immunity enjoyed by the soldiers creates two different classes of citizens - soldiers and civilians - and is a violation of the most basic democratic principle, equality before the law. And, while the trend in democratic societies is for an increasing transparency of government actions, and a concomitant increasing civilian empowerment, what the citizens of Ulster have suffered is an increasing militarization of politics.

III. Summary: Characteristics of the Military Performance in Los Angeles and Northern Ireland

The analysis in the preceding two sections indicates that the military participation in law enforcement in Los Angeles and Northern Ireland shared some common characteristics and dynamics. This section reviews those common traits.

1. All the missions undertaken by the military were traditional police missions.
In Los Angeles and Ulster, the military spent most of the time directing traffic, conducting street patrols, manning vehicle checkpoints, conducting house searches, and protecting commercial establishments. All of these missions are normally undertaken by police officers. The military did not contribute any specific capabilities, nor was there anything in the nature and dynamics of these crises that required the use of military force.

2. The military were neither trained nor equipped to perform constabulary work.
In the case of Los Angeles, the military were unfamiliar with the terrain - not only the geography of their area of responsibility, but also the street conditions, and the public's attitudes and reactions. Some military equipment, such as radios, did not work well in a city environment. In Northern Ireland, the army has been forced to purchase equipment that has no military use beyond the Ulster theater. The army has also spent considerable resources developing training programs to teach soldiers how to behave like policemen. And, because the army and civilian officials implicitly acknowledge that the constabulary mission detracts from the army's warfighting skills, an elaborate system of rotation has been established, so that duty in Northern Ireland becomes a determining influence in the functioning of the entire British army. This begs the question: if soldiers have had to be re-trained to become policemen, why not train policemen in the first place?

3. Military-police interoperability was problematic.
In Los Angeles, individual policemen and soldiers, demonstrating considerable good will towards each other, developed good relations as they operated side by side. But incidents such as the one described earlier, when the police requested cover and the marines opened fire, reveal that it was difficult for these organizations, with different procedures and equipment, to act harmoniously, particularly in the absence of any previous joint training. In Northern Ireland, the police and army battled for control of security policy well into the 1980s. This enmity and rivalry over which of the two organizations would dictate the orders hampered the entire security effort.

4. The police are trained on the principle of minimum force, whereas the military operate on the principle of maximum force.
Protesters, demonstrators and rioters can and sometimes do engage in behavior that the penal codes term criminal, such as the destruction of property. But the criminal codes in democratic societies never punish these actions with the death penalty. Therefore, it is important to guarantee that, in responding to protest and riot, the security forces will not apply the death penalty in extra-judicial fashion. Policemen routinely employ minimum force, while soldiers are trained to employ maximum force against an identified enemy. It is not customary in war to ask for explanations of individual deaths or woundings. In constabulary work, it is. And while in military terms the impact of one bullet may be negligible, the political impact of one bullet is immense. This is why policemen are, or should be, equipped and trained in the use of non-lethal means of crowd control. Soldiers hardly ever are. In Los Angeles, soldiers were only armed with lethal weapons, and their only option, if persuasion failed, was to use lethal force. In Northern Ireland the soldiers are equipped and trained in the use of lethal and non-lethal weapons. However, the high number of fatalities in disputed circumstances, and the failure to prosecute those responsible, has left a legacy of bitterness that will be difficult to overcome.

5. Military intervention in Los Angeles and Ulster became necessary because the police could not cope with the situation, but constant recourse to the military hampers the development of an adequate police force.
The city of Los Angeles seems to be involved in a vicious cycle. In 1965, after the Watts riots, a commission of inquiry argued that the riots had two causes: the social deprivation in inner city areas, and the bigotry displayed by the police in these areas. Also, in 1965, the lack of legitimacy and riot control training of the LAPD necessitated the involvement of the California National Guard. The commission of inquiry made a series of recommendations for change in social policy and in the behavior of the LAPD, but little was achieved. In 1992, the same conditions that had led to the Watts riots led to the Rodney King riots, and once more, the National Guard had to be called in. A new commission of inquiry in 1992 made fresh recommendations on social policy and on radical changes to the culture and processes within the LAPD. Almost a decade later, nothing seems to have changed. In fact, over the last year (1999-2000), the media has regaled the American public with stories about rampant corruption and excessive use of force (with racial overtones) by the LAPD. It is easy to conclude that if another riot were to break out in Los Angeles today, the LAPD would be paralyzed once more, and the National Guard would have to step in.

To some extent, the same arguments about a vicious cycle can be made in the case of Ulster. In 1969, the army was sent into the streets of Belfast because the police, which was only 13% Catholic, lacked any legitimacy with the minority, and was overtaken by events. Today, the RUC is one of the best-equipped police forces in the world. However, Catholic membership in the force has plummeted to 7%, and a recent (1999) commission of inquiry argues that the RUC is neither legitimate nor responsive to the needs of the Catholic minority. Therefore, on the basis of the experience in Los Angeles and Ulster, it can be said that the military may provide short-term relief to a public order crisis. But such crises are normally the result of complex political and social problems, and inadequate policing. Unless those problems are addressed, the military might have to be redeployed, and becomes less of a deterrent. In addition, in Los Angeles and Ulster, continuous reliance on the military has prevented the police from facing up to its responsibilities.

IV. Conclusion: Lessons and Recommendations
"No army, however well it conducts itself, is suitable for police work."(68) This is a conclusion that Colonel Michael Dewar draws from his tours of duty in Northern Ireland. It is also one of the threads that runs through this monograph. Democratic public order is a job for well-trained policemen, but it also requires a broad consensus within civil society. In conclusion, this section draws some political lessons from the two cases described, and proposes some remedies.

Lesson # 1: The time to think about public order emergencies is not when the rioters are on the streets, but well before that. Crises are not inevitable; they can be caused by government action, or inaction.
Civilian officials in Los Angeles in 1992 and in London in 1969 believed that they were faced with an emergency so severe that it warranted the deployment of military forces. This may have been the case. But it is also true that in Los Angeles and Ulster, problems became emergencies through the actions or inaction of those in authority. In Los Angeles, politicians had allowed social conditions to fester, in spite of the warning signs of the Watts riots in 1965, and the verdict of the commission of inquiry, which argued that what explained Watts was the deprivation of inner city minorities. Politicians and voters in Los Angeles had failed to provide the funds that would have given the LAPD adequate equipment, training, and personnel. In addition, the LAPD operated without meaningful oversight, and the force had alienated the community of South Central. In Northern Ireland, decades of British neglect had allowed the Protestants to turn the Catholics into second class citizens, and the RUC into a force identified with one party to the conflict. In both cases, the emergency might have been avoided if social and political problems had been addressed in timely fashion. To compound the problem, civilian officials in both cases gave little though to what should happen if the violence escalated. Confronted with widespread rioting, they resorted to the military, provided little guidance, and expected the military to provide a solution.

Recommendations:
Just as other social policies, such as health or education, are widely debated in democracies, the question "what kind of domestic security policy do we want?" is one that should be permanently asked and debated in government, in the press, and by the citizenry. The answer to this question will change depending on national circumstances and resources. But the question must be addressed, and a consensus reached, particularly in view of the potential human costs. That consensus must find expression in laws that regulate both the behavior of the security forces and the oversight mechanisms for those forces. In addition, through political parties and other social organizations, citizens must demand that their government meet its basic responsibility. Governments have a responsibility to provide solutions to pressing social problems and to provide a public order policy that is compatible not only with democratic values but also with the needs of individual communities.

Lesson # 2: The police must be trained and equipped to deal with disorders in a manner that does not exacerbate the situation, and is compatible with democratic practices.
Protest is a basic right and an inescapable fact of democratic life. Protest need not turn violent, and develop into a riot. Experience shows that the role of the police is decisive. The police can be prepared to adopt a gradual response, and thereby diffuse a crisis. Conversely, the police can, by their actions, provoke increased violence.

Recommendations:
Police forces in several countries have adopted practices worth emulating. These practices have successfully minimized social conflicts and guaranteed a balance between the right to free expression and the right to order.
a) Adequate force structure and budgets: In Los Angeles and Ulster, one of the problems at the start was that the police were not available in sufficient numbers, and lacked riot control training and equipment. Therefore, an important first step is to make sure that a police force has adequate personnel, available in sufficient numbers to deal not only with the population's everyday needs but also with the incidents of public protest and mass disorder. However, numbers are not enough. Studies of police forces indicate that as soon as they face financial constraints, senior policemen make cuts in the area of training, particularly riot and human rights training. Therefore, in addition to a proper force structure, the police must have adequate budgets, to support adequate training and equipment.
b) Policing protests as a labor-intensive activity: This monograph has consistently distinguished between lethal and non-lethal means of crowd control, emphasized that only policemen are trained in the use of non-lethal weapons, and stressed that such non-lethal weapons are a more humane method to deal with crowds. However, some European police forces resist the use of such technical aids, on two grounds. First, there is a big debate over the use of plastic and rubber bullets, which are only effective at a range between 20 and 45 meters. Plastic and rubber bullets can kill at short distances, and are ineffective at long distances. In addition, police forces in some European countries argue that other non-lethal means of crowd control, such as water cannon, horses, dogs, or large shields, are too provocative. Therefore, the preference in these countries is for the deployment of a very large number of police officers, armed only with batons. It is assumed that those police officers will be able to negotiate with the crowd, and that those officers armed with technical aids should stay in the background, ready to be employed only if persuasion fails.
c) Innovations in public order management: Some European police forces negotiate with protestors and demonstrators prior to an event.(69) The police meet with demonstration organizers prior to an event and assist the organizers with preparations for the march, discussing issues such as transportation, first aid, alternative routes. The police also present their views, and explain the nature of their jobs, to the potential demonstrators. In Germany, police forces have occasionally distributed flowers to the protesters, and have sent policemen in uniform to demonstrate alongside the protesters. In the Netherlands and Germany, the police do not limit these contacts to the period immediately preceding a protest event. The police maintain frequent contact with those groups such as ecology or peace groups, which they expect to have to police at some point in the future. In the Netherlands, these groups are routinely invited to the training sessions of the Dutch riot police.

Lesson # 3: The police are the most visible arm of the state. The attitudes and behavior of the police will influence what the citizens think of the social order under which they live. Therefore, the police must at all times be held accountable for their actions.
The innovations described above are only possible if the police enjoy a relatively harmonious relationship with the community it serves. This was not the case in either Los Angeles or Northern Ireland, where the police had lost all legitimacy, primarily in the eyes of the minority populations. Civilian oversight of the police is vital in guaranteeing that this situation does not develop. Civilian oversight can also ensure that the police comply with the laws and regulations on their behavior.

Recommendations:
Democracy is compatible with a variety of mechanisms of police control. Particular institutions or processes have not proved more effective than others. Rather, experience seems to indicate that the co-existence of a variety of governmental and non-governmental controls works best. What follows is a list of different mechanisms for control. Readers should keep in mind that what is important is not so much the institution or process in place, but whether oversight is effectively exercised. For example, in Los Angeles the mayor is supposed to exercise ultimate control over the police, but we know that by the time of the riots, the mayor had not spoken to the LAPD chief for over a year. Therefore, the mechanism was in place, but it was not employed effectively.
a) Legislatures: parliaments have the power to define the structure of the police force, its rules of conduct, its internal disciplinary codes, its patterns of deployment, its equipment. Parliaments also have the power of the purse. It is not enough for parliaments to legislate on these matters. Ideally, parliamentary commissions should also monitor that the law is being followed.
b) The locus of everyday control: in addition to the law, police forces are accountable to civilians on a daily basis. There is enormous variation across countries, dictated by tradition and cultural preferences. In most of the United States, the police is accountable to an elected official, the mayor or state governor. In continental Europe, the police is accountable to the minister of the interior or the minister of justice (sometimes, to both), who is assisted by civil servants. Britain (including Northern Ireland), Japan and some large American cities subscribe to the philosophy of "police independence." This is the idea that the police should be accountable directly to the law, and not to public officials identified with the political party in power. This is a lofty ideal, but in practice, in these cases the police have been placed under the control of civilian review boards without clearly defined powers and without budgetary resources. The result has been an absence of meaningful control.
c) Judicial review of the use of force and police misconduct: in a variety of countries, the police are criminally liable for misconduct and for excessive use of force. This is a positive development that enforces the notion that the police are at all times servants of the law. The courts also allow members of the public to obtain redress. However, in itself this mechanism is not enough. In the two years prior to the Rodney King incident, LAPD officers had been taken to court in record numbers, on charges of excessive use of force, and the LAPD had paid astronomical sums in compensatory damages. However, this did not prompt any police re-evaluation of its conduct.
d) The media and non-governmental organizations: the media and NGOs have a vital role to play. The media can and must expose and discuss problems in the provision of law enforcement services. NGOs are uniquely placed to conduct research, lobby political power, disseminate information, even act as mediators between the community and the police.
e) Internal police controls: some police forces have written codes of conduct, and disciplinary boards designed to enforce those codes. Some scholars emphasize that this is the best way to ensure the accountability of the police. However, the internal review of police conduct will not do much to increase the public's confidence in the police.

Lesson # 4: Police organizations must develop procedures and techniques that make them legitimate in the eyes of all sectors of society, including the ethnic minorities.
We have seen that one of the key issues in Los Angeles and Northern Ireland has been the persistent failure of the police to earn the confidence of the minorities they are supposed to serve. Unless this problem is addressed, a repetition of periodic outbursts of violence is likely.

Recommendations:
Two complimentary issues must be addressed if a police force is to enjoy harmonious relationships with a multi-ethnic and multi-cultural community:
a) The ethnic composition of the police force: in order for the police to gain acceptance in a multi-ethnic and multi-cultural community, the composition of the police force must reflect the ethnic composition of the community. Efforts must be made to reduce imbalances between the social composition of the community and that of the police. The police must develop recruitment strategies that increase minority membership, and must also conduct cultural awareness training of its personnel, so that retention of minority police officers does not become an issue.
b) Community policing: Communities plagued by ethnic conflict have experimented successfully with a new policing stylea (70), known as community policing, alternative policing, or neighborhood policing. Community policing is also the norm in some countries that have not suffered major ethnic conflict.(71) In community policing, the constabulary is expected to be a constant presence in the community, and to engage that community in creative problem solving partnerships. Community policing requires a significant investment in time and funds, since it relies on foot patrols. Police officers "on the beat" become a constant presence in the local schools, parks, homeowners' or merchants' associations. Local police stations also formalize contacts with the media, the NGOs, and citizens' committees. Community policing has proved effective in fighting crime, but more important, it has provided a constant channel of communication with the community.

Lesson # 5: If widespread disorder and rioting overwhelm the local police and determine a need for reinforcements (police or military), this must occur according to well established procedures and societally approved criteria.
In Los Angeles and Ulster, civilians called in military troops because they perceived that the police were not available in sufficient numbers to quell the violence. This situation might arise in other places, given that it may not be financially possible to maintain a large police force.

Recommendations:
There are several options that do not necessarily involve military troops. Austria, Spain and France maintain a separate riot control police. The advantage is that this force can be deployed wherever disorder breaks out. However, critics argue that a national riot police will apply the same techniques and the same response, irrespective of local conditions. Germany, Portugal, Sweden and the Netherlands apply a second option. They maintain a national riot police, and deploy this police in support of local forces. A third option, if protest and riot policing is left in the hands of the regular police, is to deploy reinforcements from other jurisdictions. This can happen in several ways, depending on the organizational structure of the police. In countries with a single police directed from the national capital, this mobilization of reinforcements can happen quickly. To some extent, the same happens if a country has a regional police organization, directed from a state or provincial capital. But in countries with an atomized police structure like the United States, where there are approximately 25,000 local police forces, it will be necessary to have pre-existing agreements, such as the Mutual Aid plan which was bypassed in Los Angeles. Finally, there is the option of military forces.

Whichever option is selected, a plan of action must be in place well before disorders break out. This plan should be simple and well understood by those entrusted with carrying it out, unlike the Los Angeles emergency regulations. This plan should also be sanctioned by community involvement. In addition, the more forces involved, the more difficult it will be to coordinate action. Therefore, if an emergency plan contemplates the participation of different police forces, or of the police and the military, there should be annual or bi-annual joint exercises. Planning and training should emphasize a clear chain of command, and if the military are to be involved, military subordination to the police. Planning and training, especially if the military are to be involved, should also emphasize discussion of the appropriate use of force, and training should also emphasize compliance with the rules of engagement, and penalties for non-compliance.

PDGS - Seminar Governance, Security and Military Institutions in Democracies