PUBLIC ORDER POLICING

Policing is not only about protecting people from crime and responding to crime. It is also about ensuring public order and maintaining the peace when social ordering and control is insufficient. Since public order is a contested concept, capturing what it is in police doctrine is tricky.

Public order policing is normally conceived in terms of policing designed to deal with political demonstrations, protests and riots so that people and property are not endangered. According to its political nature a regime will weigh the degree of freedom of expression and assembly to be allowed, against the interests of public safety, the prevention of crime and the security of the state. It will also determine how much force it is prepared to use in any given circumstance, whether enforcement will be civilian or military, and what will be its policy should security personnel break the law while undertaking a public order role. These decisions are primarily political. Regimes of all colours use the police and the criminal law to maintain their chosen political order. Any actual or threatened breach of this order is not only impermissible, it is also punishable. ‘State power’, as Pietro Toggia observes, comes to be ‘identified as one and same with order and normalcy, even when the state itself often engages in disorderly conduct’ (Reference Toggia2008: 122).

In East Africa, as in many other parts of the continent, there are regimes that have doubts about the democratic policing advocated by ‘international policing standards’ (e.g. OHCHR 1997). Politically they argue that following a successful liberation movement's seizure of the state, as in Uganda and Ethiopia, security institutions should be coordinated with the national project directed by the party. Thus the police should be under direct political control, and are available to be an instrument to suppress opposition so as to secure national unity and political stability. Practically, the political environment is seen as requiring a police whose central mandate is intelligence gathering so as to intimidate, coerce and eliminate perceived threats to the established order. This policing paradigm has been called ‘high policing’ (Brodeur Reference Brodeur1983). It takes hold typically during or shortly after conflict when armed groups persist; when internal revolutionary and terrorist groups threaten; or in the first decades of rule by the victorious liberation movement faced with power struggles within the ruling liberation movement (Steinberg Reference Steinberg2014). This is to see political policing, not as confined to policing political activities such as protests, but as a model of policing. It is within this political context, where regimes doubt that the police should be a body independent of state control, that I examine the practice of public order policing in East Africa. I compare public order practice against political goals (rather than against ‘international policing standards’ or human rights principles).

There are different approaches to the structuring of public order policing in East Africa. Public order units may be within the police force or the public order duties may be given to separate para-military organisations. The policing model that sets the general police officer amongst the people, ‘toe-to-toe’, requires high levels of skill, expense and of public trust in the police. Thus most states have chosen some kind of specialised unit for public order.

Faced with public disorder, East African governments have largely responded in technical and legal ways. Essentially these have sought to contain protest by relying on more of the old methods. Across Africa police groups have bought anti-protester equipment (armoured vehicles, CS gas, water cannon, baton guns and electronic means of surveillance of internet and mobile phones) and new barriers (physical barriers instead of ‘human shields’), though some still regard firing live ammunition over or at protesters as the most effective form of dispersal. In the area of police training the emphasis now is on new courses (on the use of firearms and proportional force, self-protection, specialist search); new command procedures (decision protocols, mobilisation plans, intelligence gathering through social media); and new tactics (‘containment’, stop and search, overt photography, use of shields, forward intelligence teams).

There have also been revisions of legal powers (to authorise or not demonstrations, banning orders and communication intercepts). East African legislation typically says that the police may impose lawful restrictions on the exercise of rights, provided such restrictions are lawful, have a legitimate aim (such as the interests of public safety or the protection of the rights and freedoms of others) and are necessary and proportionate. In practice, governments may feel free to revoke those rights at will.

Technical and legal approaches have shortcomings. African governments are now faced, particularly within large urban areas, with single or multiple groups that are emerging as contenders for power. The cities have more young people willing to use violence against the authorities, more educated activists able to organise effectively and nationally, and every protest group has had its capacity extended by ICT. Information has long been available from the international media, but now thanks to the internet and social media information is now readily available that offers alternative conceptions of politics and justice; that can challenge the government version of events and indeed its very legitimacy; and that identifies other individuals' aspirations and critiques. It is information that can be shared within the group and that allows rapid mobilisation. Because it facilitates a new order of mass organisation, it makes conflict between elements of urban populations and the government more likely. As Charles Tilly has noted, ‘the efficiency of governmental coercion is likely to decline, at least in the short run, when the character, organization, and daily routines of the population to be controlled change rapidly’ (Tilly Reference Tilly1973: 443).

The changing social context has altered the requirements of public order policing and one would expect it to adapt in some way. As a former British chief constable argues, it is when ‘social change is taking place at a fast and, to some, disturbing pace, the role of the police may become central’; often, of course, to resist those changes (Alderson Reference Alderson1984: 19).

Using the cases studies of Ethiopia and Uganda I offer a political analysis of the police response to this changing social context in terms of its public order tactics.

ETHIOPIA

It was in 2005 that the Ethiopian government first became aware that the well-tested authoritarian approach to public order policing might have to adapt to new social conditions. An apparent election fraud shocked opposition political parties and student groups expecting that era to be passing and brought anti-regime protests to the streets of Addis Ababa and elsewhere between June and November 2005. In disregard of their own rhetoric about a new respect for human rights, the army and Ethiopian Federal Police (EFP) fired live rounds into the crowds and 193 protesters (and bystanders) were killed, 763 injured and 20,000 arrested (Abbink Reference Abbink2006; Aalen & Tronvoll Reference Aalen and Tronvoll2008; Pausewang Reference Pausewang, Ege, Aspen, Teferra and Bekele2009).

Initially the government showed no desire to either censure or to investigate the police conduct. It merely minimised the number of deaths and sought to put the blame on the victims, saying they had died when police moved in to break up roadblocks and curb looting. However, as the government took stock of the internal dissent the event fuelled and the international repercussions, an official Inquiry Commission was established, though its terms of reference sought to curtail the full story by narrowly limiting the dates and places of violence that were to be investigated.

To its credit and the government's surprise, the Commission's Report 2006 concluded that the government had concealed the extent of deaths and injuries at the hands of the police. Its chairman commented: ‘Many people were killed arbitrarily. Old men were killed while in their homes, and children were also victims of the attack while playing in the garden’ (Mariam Reference Mariam2010). The Commission also made specific factual conclusions. First, that the persons killed or wounded during the violence were unarmed protesters (contradicting the government-controlled media). Second, the shots fired by the police were intended to kill by targeting the head and chest. Third, there was no autopsy or photographic evidence that any police officers involved in the shootings were attacked or killed by the demonstrators. The Report concluded:

It called upon the government to ‘think seriously’ about changing the police response to public order and recommended specialised training. A member of the Commission reported: ‘The government was not ready for the violence that occurred … Moreover, the security forces lacked the necessary training in controlling riots’ (AFP 2006).

What the government had failed to take into account was that the expectations of many had changed. Whatever the 1995 Constitutional protection of human and political rights meant to the government, large city groups, political and student, took Article 13′s assertion that rights and freedoms will be interpreted according to the Universal Declaration of Human Rights and the International Covenant on Civil and Political Rights, to mean that the police were required to be less brutal to the public. Such expectations, coupled with a rising polling strength of the opposition, meant that the government had to revisit its public order policing policy. The Director General of the EFP admitted that he ‘regretted a lot’ the events of 2005 (Dewar Reference Dewar2008). Yet his assessment was essentially what Brodeur calls the ‘deviance approach to political policing’, namely one focused on the illegality and exception of the policing of political activity rather than recognising political policing and its repressive behaviour as a regime choice of police model or, as the Commission report it, a pervasive ‘mentality’ (Brodeur Reference Brodeur1983: 508). Hence his response was not to recommend a change in the police paradigm, but to sack 237 policemen (none of whom was ever prosecuted). The subsequent selection of the South African Police Service to ‘retrain’ the 6,000 EFP in riot control procedures was perhaps a surprising choice. It cannot be because the SAPS have a good reputation in riot control. But it may have been thought better to use a police force that worked in an African environment than a Western police force.

The recognition of individual police failure did not stop the government from proceeding to jail most of the principal opposition leaders, civic society organisers, human rights advocates and journalists on treason charges. As the theory of coercive responsiveness predicts, governments will attempt to suppress protests as quickly as possible, lest they become more threatening forms of dissent that destabilise the entire political system (Davenport Reference Davenport2007). As for the six Commission members who had recognised a political choice in the police action, all chose to go into exile rather than risk retribution for their blunt insights.

Besides adopting new training, the government commissioned a review of the EFP in 2008 (at the instigation of the so-called ‘Think Tank’ of experts with an interest in Ethiopian affairs that met at the Ethiopian Embassy in London). The review was to be written, not by a UK police officer, but by a UK retired army officer with experience in Northern Ireland.Footnote ⁴ He was asked to look at the operations of the EFP, ‘paying special attention to riot control capabilities and procedures and strategies designed to pre-empt civil unrest’ and to make ‘recommendations on the equipping and training of such a Security force’. Sadly the review focused almost entirely on the lack of equipment, as if this was the main cause of their failures, but assured: ‘Every riot control policeman is now issued with a perfectly acceptable set of personal equipment’ (Dewar Reference Dewar2008). Dewar also had an eye in his report to the international community and hoped that his review would demonstrate to them that:

For the Ethiopian government the review allowed them to minimise censure. All they had done wrong was to fail to equip the EFP properly. No one had challenged their adoption of a policing model at variance with the democratic one of international standards or for that matter at variance with the human and political rights pledged in their 1995 Constitution. Nevertheless the government did realise the need for a softer police rhetoric. In 2011 the Minister of Capacity Building said that the EFP was responsible for ‘maintain[ing] order at all times without having to resort to undue force in general and to the use of firearms in particular’ (Gebru Reference Gebru2011). The doctrine of restraint and minimal force became the norm of discourse, even as the government continued with its view of the police as an instrument to suppress political deviance. It passed the Anti-Terrorism Proclamation of 2009 which provided far-reaching powers of arrests for ‘acts of terrorism’ – so broadly defined as to include damage to property and disruption to any public service. Even a group of two or more individuals who engage in peaceful political protest can be deemed a terrorist organisation.

With economic disparity, land grabs, ethnic clashes, religious disputes, cattle raiding, and insurgencies in Oromo and Ogaden, it was predictable that the public would seize the limited space that was still available for public protest, especially when formal channels were so dominated by the ruling party. The EFP, then, would soon have to put to the test their new training and equipment. Only now, however, they began to face much more use by urban activists, both religious and ethnic based, of ICT to mobilise protest and counter government discourse. The internet that the government had introduced in 1997 for development purposes was turning political.

The next serious challenge for the EFP was in 2012 from the Muslim community. Throughout the year Muslim leaders mobilised urban protest, largely through mosques, to challenge government interference in the Supreme Council of Islamic Affairs. Believing that their limited opportunity for voice was being even more restricted, thousands took part in largely peaceful street protests in Addis. The police response reflected the government's view that the Muslim community constituted a serious potential threat politically. Witnesses speak of the police forcibly entering a mosque in Addis Ababa where an awareness raising event was taking place and of breaking doors and windows, firing tear gas inside, and violently beating those who converged on the mosque to help. Similar scenes occurred six days later as police broke up a sit-in in the mosque in response to the earlier arrest of members (HRW 2012). The government denied all charges of police violence, but accused ‘extremist elements’ of seeking to undermine the Constitution and inciting violence (Reuters Addis Ababa 2012). But that was not the end of the matter, for not only did the Muslim community give their own account in local newspapers, but also electronically to the Diaspora media, to international human rights groups and on social media sites.

In 2013 the EFP were faced with a fresh challenge. Three Ethiopians had been killed in large and violent clashes with Saudi Arabian police, following attempts by the Saudi government to repatriate 23,000 Ethiopian illegal immigrants. Once again ICT was enlisted by protesters to criticise not just the Saudi authorities, but a passive Ethiopian government. Videos posted on social media sites showing a Saudi crowd beating an Ethiopian mobilised support in Addis that quickly led to a national campaign on a social media site and an on-line petition. From this emerged a group who organised an unapproved protest outside the Saudi Embassy in Addis Ababa to call for the boycotting and closure of Saudi businesses in Ethiopia. However, the EFP broke it up using batons and arrested 100. The government justified the use of force and mass arrests by saying that the protesters did not have a permit to demonstrate, though undoubtedly sensitivities about regional reaction to what they called ‘fomenting anti-Arab sentiments’ played a part (Aljazeera 2013).

The following year the EFP faced serious violent street protest from the Oromo community. When the Addis Ababa city administration announced plans to expand into Oromia state, social media rapidly disseminated the proposals as a threat to Oromo culture. In a context of continuing government repression of Oromia opposition (Arriola Reference Arriola2013), students took to the streets to protest, burning businesses owned by the Tigrayan People's Liberation Front – the core group of the ruling party. The government was no more prepared to give space to escalating community opposition than before. At least nine (witnesses say 47) were killed and 70 injured when, according to residents, the EPF fired live ammunition into the crowds. The government once again tried minimising the casualties, downplaying police violence and blaming the protesters for the whole affair – the very same formula that had proved so ineffective in 2005. The root cause of their violent response probably lies in their accusation that ‘anti-peace’ forces were at work among the students and their leaders were trying to destabilise the country, encouraged by ‘media inside and outside the country’. In other words, this was a regime that was conscious of its fragility and whose only tool was the public order policing tactics of 2005 to prevent public displays of opposition. For the police it was another lesson in the power of the social media to cascade information, organise big events rapidly and to challenge the official accounts of the number and cause of deaths even when traditional media reports were largely blocked. One senior police officer later confided:

This year too (2015) police have forcefully disbanded an opposition party protest in Addis Ababa. A demonstration by Unity for Democracy and Justice party was said to be unauthorised since the organisers ‘failed to submit detailed information on the route of the rally, list of slogans to be displayed on the rally and other requisite information’. Social media sites reported that a dozen were beaten by the police and showed a photo of one who claimed they had suffered a serious hand injury (Horn Affairs 2015).

Ten years on from the 2005 election violence, despite criticism from the international community and a Commission of Inquiry's call for a new police mentality, the only change in police response has been to acquire new equipment. The police have not taken into account that they are facing political protest that is better resourced, including access to ICT and is able to organise protests rapidly and across many cities simultaneously; and can counter official post-event accounts or circumvent news blackouts. As one police trainer observed:

The Ethiopian government and police are, therefore, aware of the new resources available to protesters, hence their threats to those who use electronic communication to spread critical views and their closure of websites and whole phone networks.Footnote ⁵ They are aware of changing expectations about rights and scrutiny of the police, yet still the police are called to use lethal force against protesters. As the government has narrowed the political opportunities to publicly protest they have increased the likelihood of a resort to violence, yet they have failed to change their approach to public protest. They have neither gone down the democratic policing road nor turned back to full authoritarian rule where all public opposition is banned. The police are caught in ‘the tension between the ruling party's promises of democratisation and its reticence to live up to these principles’ (Hagmann & Abbink Reference Hagmann and Abbink2011: 582).

The government's hesitancy between the two political poles where they both legally allow protest and yet in practice violently attack it, may lie not in their ideology of ‘revolutionary democracy’, but in a senior leadership wedded to militarism. It may be due to the veneration by an older generation of the ideals of a professional military class, and in their valuation of military efficiency as the ideal of the state. Prime Minister Zanawi Meles saw himself as an outstanding leader of the military in war and peace. A government eulogy on his death in 2012 said that:

The ageing EPRF leadership, born of revolutionary war, views all its security forces as ‘their’ agents to suppress all challengers. The paternalism of ‘Revolutionary Democracy’ knows what is best for social development; those that resist are enemies. It is a regime whose response to challengers is a resort to military and electronic hardware alongside legal powers to maintain the current political structures. It is still ‘entirely unwilling to engage in any normal or reasonable political process’ (Clapham Reference Clapham2005). It regards violence as an effective way to intimidate anyone contemplating resistance. Its public order policing demonstrates that it cares little for research that suggests that escalating levels of repression leads to a backlash, stoking a sense of outrage that feeds further violent protest (Mason & Krane Reference Mason and Krane1989; Carey Reference Carey2006), the more so among a people informed of the political alternatives. Nor, apparently, did they or the EFP learn anything from their South African police advisers regarding the Sharpeville Massacre 1960, when the former apartheid police killed 69 unarmed black protesters; or the lessons of the Marikana Massacre 2014, when police shot dead 34 striking miners as they tried to disperse them. Massacres with impunity by a regime's security agents is a force for revolt, not a force for stabilisation. It mobilises opposition rather than quells it. In the final analysis excessive police violence against those newly resourced with access to ICT displays anocracy's fear of losing legitimacy and control.

UGANDA

Uganda in 2005, like Ethiopia, displayed repressive policing methods. President Yoweri Museveni was facing serious opposition in the pending Presidential election from the leader of the Forum for Democratic Change, Kizza Besigye. Besigye was arrested on allegations of treason and rape. Believing the charges were fabricated to reduce the opportunity of challenging Museveni, Besigye's supporters turned to public protest. Street battles took place over two days in Kampala and other urban areas between them and the anti-riot police and military police. The police responded with arrests, firing in the air and shooting gas canisters at the protesters.

Four years on nothing had changed, even though in 2007 British police officers gave training in ‘public order management’ to the Mobile Police Patrol Unit from which the anti-riot Police is drawn, ahead of the Commonwealth Heads of Government (New Vision 2007). The police in 2009 blocked the Buganda king from visiting a district within his kingdom, ostensibly due to objections from another ethnic group, but more likely because the government did not want him campaigning against their land reform proposals. The Baganda took to the streets of Kampala to protest the police action, setting light to a police station and throwing rocks at the police. This time, however, the police, backed by the military police and the military, not only fired tear gas, but live ammunition to disperse protesters. Over the next days, hundreds of young people were arrested, riot suspects beaten and indiscriminate live ammunition was fired, at times into homes or where no rioting was taking place. Hospital records suggest over 40 deaths (HRW 2010). It was not just journalists that recorded the events, but mobile phones. They provided live updates, making futile orders for the media to cease reporting on the riots, the closure of four radio stations that did broadcast news, and police seizure of journalists' cameras. If Ethiopia discovered that ICT was becoming a political threat about 2012, Uganda's government, which had introduced the internet earlier in 1993, had a wake-up call in 2009.

Though the government downplayed the numbers, maintaining that only 27 people died, largely due to ‘stray bullets’, they did not explain the use of lethal force against stone-throwing youths in the first place. Despite multiple promises from government officials, police investigations did not result in prosecutions, the police saying that investigation was difficult when it was unclear which security forces were deployed to which areas. It appears that a police officer and a soldier were eventually charged with murder, but neither had faced trial after a year. No explanation was given as to who gave them their orders. The police had fulfilled their role of frustrating Museveni's opponents by any means, with the unspoken guarantee of immunity from prosecution.

Following the riots, the Deputy Speaker of Parliament ordered the Parliamentary Committee on Defence and Internal Affairs to investigate the unrest and issue a report within two weeks. Committee members scheduled visits to affected neighbourhoods, but unlike Ethiopia, none ever took place. Committee members claimed that their efforts were ‘deliberately frustrated’ by government officials and one year later, the committee had not summoned a single witness to give evidence, and no report had been issued (HRW 2010).

Although a 2008 Constitutional Court decision nullified section 32(2) of the Police (Amendment) Act, 2006 which required written police approval for any assembly of 25 persons or more, the police continued to require advance notification and approval for public gatherings and routinely denied permission. In addition, the government was quicker than the Ethiopians to recognise the threat posed by ICT. In 2011, as the ‘Arab Spring’ erupted, Uganda Communications Commission (UCC) instructed telecom companies to block text messages that could instigate violence and unrest, and to flag if they contained keywords such as ‘Tunisia’, ‘Egypt’, ‘dictator’ ‘teargas’, ‘army/police/UPDF’, ‘people power’ (CIPESA 2014: 13). Despite the surveillance, the political opposition, inspired by the ‘Arab Spring’, initiated a ‘walk-to-work’ campaign to protest against rising prices and government corruption. ‘The movement relied heavily on social media for mobilising and publicising the brutal response of security agencies to the campaign’ (Freedom House 2012). In response, the UCC directed internet service providers to temporarily block access to Facebook and Twitter ‘to eliminate the connection and sharing of information that incites the public’ (CIPESA 2014: 13). Following the violent arrest of Besigye during the 5th ‘walk-to-work’, the protest turned to an anti-regime riot in Kampala. Footage of the arrest was broadcast on Ugandan television and YouTube showing plain-clothes police officers beating up Besigye's supporters, smashing the window of his car and spraying the inside with pepper spray, before taking him away. Later he was released on bail following a charge of ‘inciting violence’, and allowed to fly to Kenya for medical help.

The next day the demonstrators carried posters praising Besigye and challenging the brutality of the police. Despite being aware of the power of ICT to scrutinise their actions and subvert their accounts, police tactics did not change. Clearly under orders, the military police reacted with disproportionate violence. They beat members of the crowd and fired tear gas, rubber bullets and live ammunition at street barricades and adjacent residential areas. At least two people were killed and more than 100 wounded. 360 were arrested (Guardian 2011). The police action caused rioting to spread to several suburbs and towns outside the capital.

Why did the police resort to unrestrained and unnecessary violence despite the likelihood of it fuelling further conflict? The government, like Ethiopia, was caught in the uncertain middle ground of anocracy. On the one hand they admitted a democratic right to protest, on the other they held the theory of coercive responsiveness that protests must be suppressed as quickly as possible to prevent escalation to a level of regime threat. Museveni, while conceding the right to protest, also asserted a right to determine what was a ‘legal’ protest; and by fiat denounced as ‘illegal’ the ‘walk-to-work’ campaign. Meanwhile his Information Minister claimed the right to determine what was a ‘worthwhile’ demonstration and with it the right to ban all others: ‘We're not going to encourage worthless demonstrations which don't have clear and genuine objectives to continue’ (BBC 2011). And defending police deployment in most public spaces to block pro-opposition assemblies, the junior minister for Internal Affairs, said, ‘Those demonstrating have a right to do so, but as police it is our mandate to make sure that they [demonstrators] enjoy their constitutional right without infringing on the rights of others who are not part of the demonstration’ (Observer 2014). Finally, the Internal Affairs Minister provided the justification for police abuse that they had done nothing that was not legal. Those injured had mainly been hurt by ‘stones, sticks and bottles’ from the rioters themselves, not by police action; and whatever the police force did, it was ‘within its constitutional mandate to restore law and order’ by removing road blocks and ‘disengaging crowds’ (Guardian 2011). In an anocracy the police are expected to carry out the bidding of the ruling party, yet only ever act within the law.

Though Museveni and his National Resistance Movement (NRM) had won the 2011 Presidential and Parliamentary elections, he was left with Kampala as an opposition stronghold and one that had elected Erias Lukwago, a former opposition Member of Parliament, as the first Lord Mayor of Kampala, following the enactment of the Kampala Capital City authority. Upon Lukwago taking office, Museveni hurriedly created the post of City's Executive Director and personally appointed Jennifer Musisi. As intended, the dual and overlapping roles caused conflict. Then, no doubt at the instigation of higher authorities, 19 councillors petitioned the Minister in charge of Kampala to censor the Lord Mayor for failure to execute his obligations. The Minister constituted a judicial Commission of Inquiry to investigate the alleged misuse of office, which looked like little more than a pretext for unseating the Mayor. Upon being summoned to appear before the Inquiry, Lukwago began to hold public rallies to defend himself. These caused street battles between crowds and the police. On 20 June 2013, Besigye joined him at a meeting at Kiseka Market, despite a police ban on such rallies on the grounds of intent to incite the crowd against NRM councillors. Heavily armed anti-riot police, assisted by the Military Police, arrested both politicians ‘for trying to incite people to commit violence’. Such a politically motivated arrest led to Kiseka's traders stoning the police. Police responded by firing teargas and live bullets. One person died, allegedly hit by a police bullet (Red Pepper 2013). The episode was little more than the harnessing of police coercion to maintain the political status quo. Given it fuelled the opposition cause in Kampala, it was bad politics as well as bad policing.

Though Museveni did not choose, as his Ethiopian counterparts, to have independent reviews or commissions of inquiry that might have censured his response to the challenges of urban groups, he was given public advice from an unexpected quarter. His Minister of Justice and Constitutional Affairs criticised the police tactics of ‘preventive arrest’ that barred Lukwago and Besigye from leaving their homes or escorted them wherever they went. ‘I really wonder why the police is hell bent on making the lives of these two difficult. Why don't they let them walk?’ He accused the police of not only infringing on the politicians' rights, but of suffocating the rights of the people who elected the Kampala mayor. ‘Why do you arrest a leader in such a forceful manner yet he has done nothing illegal?’ He advised Besigye and Lukwago to challenge the police's actions in court (Observer 2013).

Museveni ignored the advice and conceived his own reforms regarding police strategy and public order legislation. Up until 2014 Uganda had relied on the regular police to handle public order, but had frequently found that to cope they needed the assistance of the military police and the military. In 2014, therefore, Museveni formed a new unit within the police, the Field Force Unit (FFU) to handle demonstrations. The Inspector General of Police (IGP) said the FFU would help deal with paramilitary threats, including riots, communal violence, as well as terrorism and insurgencies, cattle rustling, and other forms of extreme violence. However, initially it would be responsible for policing Kampala, with a reserve to respond to emergencies. Clearly Museveni is determined to crack down on the opposition stronghold of Kampala with the use of the specialised paramilitary force. Far from learning that policing that is partisan, brutal and immune from prosecution has strong political disadvantages, or that new tactics are required to meet a networked opposition, Museveni is trusting that old-fashioned repressive methods will maintain power and subdue unrest. Hence at a 2014 East African Police Chiefs Cooperation Organisation meeting, when speaking about ‘the recent wave of violence and instability in North Africa and its implications to our region’, he boasted of his own resources to handle such protests. Pointing to the Ugandan IGP, he said: ‘he has enough tear gas … Once the demonstrators take enough smoke, they go home’ (BBC 2014).

Museveni's new legal powers to limit the rights of assembly and free speech were introduced through The Public Order Management Act 2013. This prohibits protests of more than three people without prior police authorisation and requires event organisers to list the names of all participants. Its vague requirement for public statements not to contravene any laws is a provision which allows censoring of a wide range of criticism. Regarding police powers, it authorises them to use guns when policing public events, but says nothing about ensuring that this use is to be the last resort after exhausting all other possible means, in compliance with the UN Basic Principles on the Use of Force and Firearms by the police. Thus Museveni has criminalised ‘unauthorised’ public assemblies and given the police a free hand in using weapons to deter or kill political opposition.

Meanwhile the political use of social media continues to expand. In 2011 the social media may have been the preserve of the middle class and it is not thought they had a large impact on those protests. Since then, however, they have been widely adopted by young people. As one Ugandan journalist records: ‘more recently, young people are beginning to use social media to discuss political issues – drawing parallels with social media use during the Arab Spring … there is a lot of youth this time around who have finished school and are jobless and who are now users of social media’ (BBC 2014). The IGP has already cautioned about social media: ‘it is so quick in terms of disseminating information … if it's good information that is nice, but if it's dangerous information … and somebody tells lies like the Kayunga riots, then you know how much damage it can do’ (BBC 2014).

Why has there been this failure by the regime to change their approach to public protest and riots despite their awareness of the increasing access by activists to information via ICT? The answer surely has parallels to that of Ethiopia. Once again we find a belief that though protest cannot be criminalised in toto, it must in practice be suppressed quickly and firmly to prevent its spread. And we find, too, an elderly militarism in the leadership that having led a successful armed revolt reveres a coercive solution. Museveni has never hidden his pride in his army values of discipline and efficiency.Footnote ⁶ Like the EPRF, the NRM's own ageing leadership views the security forces as their personal agents to suppress urban challengers. Once again the paternalism of a ‘progressive’ liberation movement believes it knows what is best for the people. It, too, is a regime that understands what military hardware can achieve. However, although it is aware of the increasing significance of the software that powers the social media of its youth, like the Ethiopian regime, it has until recently underestimated it and failed to link its understanding with police public order tactics.

The failure to lessen the violence of the anti-riot police also suggests an underlying insecurity; the more the regime's grip on power is felt to be slipping, the more it lashes out at those groups that would throw away its perceived gains. Thus, in preserving the old approach to public order policing, Uganda disregards the changes that have taken place in Ugandan urban groups over the last decade. It is a recipe for further unrest and ultimately unsustainable in the new order of public opinion and public resources for mobilisation. The 2014 annual budget performance report shows expenditure on handling of 57 cases of public order, 49 cases of incitement to violence, six election-related cases, and one case of promoting sectarianism, among others. According to the 2014/15 Internal Affairs ministerial policy statement, Shs 14bn (£3·1 m) will be spent on procuring riot gear, shields, jackets, boots and other equipment necessary to put down riots and other unlawful assemblies. Another Shs 1·8bn will be used by the police to meet the cost of public order management and riot control (Observer 2014). Current Ugandan policy on public order policing will create not stability but instability.