The use of violence as a political tool has been common in Nigeria both before and since President Obasanjo first came to power in 1999. The 1999 elections were also marred by violence and intimidation, as well as widespread fraud and rigging.5 Previous Nigerian governments had used political violence even more brutally and systematically, often without any real pretence of operating within a democratic system. Despite the shift from military to civilian rule, political violence has remained prevalent. It has been facilitated by the wide availability of small arms and a large population of unemployed young men who are willing to be hired and armed by politicians to intimidate their opponents. Violence became such an accepted part of political competition in some areas during the 2003 elections that politicians did not even attempt to conceal it; for example, a PDP ward chairman in the southern city of Port Harcourt told a human rights activist directly how the PDP had distributed guns in the area.6
While occasionally leading political figures fell victim to targeted killings, the overwhelming numbers of direct victims of the violence around the 2003 elections were young men whom politicians had sent out to fight each other for control of political constituencies; others were simply bystanders who found themselves caught up in the violence.
In some areas, clashes between election opponents overlapped with conflicts over other issues. For example, in the Niger delta and in Plateau State, both the 2003 general elections and the 2004 local government elections provided a focal point for longstanding inter-communal conflicts over land and resources; these conflicts were reignited as some groups used the elections as a way of venting their anger or desire for revenge in broader disputes over economic and political power.
The 2003 elections began on April 12, with voting for members of the National Assembly; elections for the presidency and governorships took place on April 19 and elections for state houses of assembly on May 3. Voters came out in overwhelming numbers on April 12. While in some areas, the process was conducted relatively smoothly, in others, there was no pretence of elections taking place at all, because of vote-rigging, violent intimidation, or both. In yet other locations, logistical problems prevented the smooth running of the elections, and some parties and candidates took advantage of these problems to falsify the results. In some locations, especially in the south, elections materials never reached polling stations, or election officials did not turn up; yet results were officially announced from these constituencies. In its final report on the elections, the Transition Monitoring Group (TMG), a national coalition of civil society groups which monitored the elections, described the activities on all three election days and the collation and declaration of results as characterized by monumental fraud.7
Many incidents of violence were recorded during the April 12 and 19 elections. By the time the state houses of assembly elections rolled around on May 3, much of the electoral violence had run its course. In several key areas, including the south and the southeast, this was at least partly because opposition parties started boycotting the polls, complaining of rigging and fraud in the earlier elections and urging their voters not to cast useless votes. Independently from these party boycotts, many voters had simply become disillusioned following the experience of the first two phases of the elections. Others were deterred by the prospect of fresh violence. Nonetheless, some serious incidents were recorded around the May 3 elections.
The highest level violence during the 2003 elections was in the south and the southeast, where PDP governors and their supporters universally succeeded in resisting opposition bids for office. These were also the areas where the greatest rigging and fraud were recorded by independent electoral observers. In these areas, the direct link between violence and election fraud was clear. More than three quarters of the incidents of violence, intimidation, harassment, ballot box stealing and stuffing and vote buying reported by TMG election observers were recorded in the south and the southeast; however, they also reported violence and disruption in other areas, including in western, central and northern states.8 The situation in some areas in the oil-producing Niger delta, in the south, was so serious that a non-governmental organization which monitored the elections stated: In parts of Rivers and Bayelsa States observed by our monitors, the elections could be characterized as a low intensity armed struggle. Weapons and firearms of various types and sophistication were freely used.9
The largest number of deaths during the elections occurred when opposing bands of political thugs, in some cases armed on both sides, fought each other for physical control of a locality, attempting to displace supporters of the opposing party. Witnesses reported numerous incidents to Human Rights Watch in which armed thugs, usually though not exclusively from the PDP, shot into the air or otherwise threatened voters with violence, created chaos, and then ran away with the ballot boxes. In some instances, these groups shot directly at individuals from opposing parties. In other cases, their threatening behavior and public display of weapons ranging from knives to firearms was sufficient to scare off their opponents, as well as ordinary voters. This type of intimidation was especially common in the south and the southeast, but also occurred in other areas. The TMGs summary of incidents recorded during the April 19 elections includes cases in which elections were violently disrupted in various locations in the northern states of Katsina and Jigawa, the central or Middle Belt states of Benue, Taraba and Adamawa, and Kogi State in the west, as well as the southern and southeastern states of Rivers, Imo, Abia and Akwa-Ibom.10
There were also cases where people who tried to intervene to prevent rigging were beaten by thugs hired by the various parties. Protests at rigging also often took a violent form, with aggrieved opposition members, as well as other frustrated voters, smashing election materials and equipment to prevent fraudulent votes from being used, and, in more serious cases, attacking PDP members or others suspected of rigging. Some election observers were threatened, and in some cases physically attacked, in order to prevent them from witnessing or reporting abuses. Some of these threats were made by state or local government officials, others by members of the security forces. At least three TMG observers in the southeast were among the victims of violent intimidation during the state house of assembly elections: in Enugu State, a TMG observer was whipped with a chain by a paramilitary mobile policeman apparently acting in collusion with PDP supporters; an observer in Ebonyi State was beaten by a candidate for the state house of assembly; and an observer in Ebonyi State was chased out of two polling stations.11
The successful portrayal of the elections as free of violence was crucial to the Nigerian governments efforts to project a positive image to its foreign partners and to consolidate its reputation for its second term. In the vast majority of violent incidents involving local government politicians, PDP officials, PDP candidates, or their supporters, no one was charged or tried. In a few cases, low-level PDP supporters were arrested or questioned, but were usually released within a short period. Many of the candidates whose supporters carried out violence and intimidation are now occupying political positions at local or state level.
In contrast, a larger number of opposition party supporters were arrested, and some were charged. The opposition parties have claimed that most of these arrests were politically-motivated. Human Rights Watch has not been able to verify the circumstances surrounding all these arrests. Many of these opposition supporters have since been released without charge, but only after the election period was over.
The role played by the security forces was mixed. Human Rights Watch documented a number of cases of human rights abuses carried out directly by members of the security forces, mostly by the police, particularly the paramilitary mobile police, acting in collusion with ruling party officials. Although such cases may not have been as widespread as anticipated, the number of abuses in which police officers were directly involved remains of serious concern, particularly in areas such as Rivers State. However, in some locations, voters and observers reported that the police helped maintain a peaceful atmosphere and were not seen to be acting in a partial way. 12
There were also incidents in which soldiers engaged in acts of violence, for example in Benue and Rivers states. Overall, however, independent observers reported that the deployment of the military in many areas of the country reduced the level of violence, although some noted that it created an intimidating atmosphere for elections.
While security forces were not partial actors in all incidents, in many cases they were unable or unwilling to protect those who came under attack. Typically, police were reported to flee the scene or to do nothing when violent clashes broke out or armed people invaded polling booths. They frequently failed to arrest the perpetrators, even in cases where violence was committed in full view of many witnesses. They also failed to respond to calls for help from some individuals who reported receiving direct threats from their political opponents.
The leadership of the various political parties also bears a heavy responsibility for the actions of their members and supporters, particularly for the failure to halt violence and intimidation during the elections. While some party leaders admitted reluctantly that their supporters may occasionally have engaged in acts of violence, most of them denied it and simply blamed their opponents.
Officials of the PDP secretariat in Abuja told Human Rights Watch that the PDP had a standard disciplinary procedure and imposed sanctions against any of its members who committed violence, ranging from verbal warnings to suspension and ultimately expulsion from the party. One official insisted that the sanctions were used, but could not name any specific case where PDP members had been expelled because of election-related violence or intimidation. Another official said that the decision not to use violence had been taken at a high level in the party, except when people were pushed to the wall. We were not committing violence unless it became very, very necessary to fight back. If at a polling station, opponents want to carry ballot boxes away, we have to resist, and this can degenerate into violence. Also if the life of agents are threatened. But sometimes police is not readily available; you have to make sure your interest is protected.13
The governments failure to bring to justice most of the perpetrators of political violence in 2003 meant that there was little to discourage candidates in the 2004 local government elections from using the same tactics. As a result, the local government elections were also marred by violence and intimidation.
Local government elections had originally been scheduled to take place in 2002, but were repeatedly postponed. The reasons given publicly for these postponements related mostly to logistical and administrative preparations. But many Nigerians speculated that the delays were more likely to have been motivated by political considerations. After the terms of local government councils expired in May 2002, state governors appointed local transition or caretaker committees to take the place of elected local government councils. A federal government announcement in June 2003 that the system of local government administration would be subjected to a wide-ranging review meant that local government elections were, for a while, indefinitely suspended; in the meantime, local governments continued to be controlled by unelected individuals, picked by state governors. When the elections were eventually announced for March 2004, many of these individuals were reluctant to give up these lucrative positions. Human Rights Watch also documented several cases in which members of these transition committees were implicated in acts of violence against their perceived opponents during the 2003 elections.
The 2003 elections had already demonstrated that the fiercest battles for political control were played out at the local level, and local disputes were the motivation behind many of the most serious incidents of violence. The 2004 local government elections confirmed this pattern. Violence broke out in many locations before, during and after polling day on March 27, leading to dozens of deaths. In addition to battles between supporters of different political parties, the period of the local government elections saw an intensification of internal fighting, in particular within the PDP, with different factions vying for control of local government positions. From February 2004 onwards, there were several incidents of apparently politically-motivated killings and attacks in different parts of the country. For example, on February 6, Aminasoari Dikibo, national vice-chairman of the PDP for the south-south zone, was shot dead in Delta State. In Kogi State, the chairman of the State Independent Electoral Commission, Philip Olorunnipa, was killed on March 7, and the PDP candidate for the chairmanship of Bassa local government, Luke Shigaba, was killed on March 3. Also on March 3, a vehicle carrying the Benue State governor, George Akume, was attacked; the governor escaped unhurt, but a friend travelling with him, Andrew Agom (who, like the governor, was a PDP member), and a police officer were both killed. There has been speculation that some of these incidents may have been caused by infighting within the PDP.
Between January and April 2004, there was also an intensification of inter-communal violence in areas such as the Niger delta and the central Plateau State. Not all these incidents were directly related to the elections, but observers concurred that the climate of heightened political tension created by the prospect of elections contributed to the increase in violence, especially in Delta State.
On the actual election day of March 27, numerous incidents of violence and intimidation and clashes between supporters of different parties and candidates were reported across the country. Human Rights Watch has not been able to verify all these incidents, but is deeply concerned at the uncontrolled proliferation of political violence. The geographical spread of locations from which electoral violence was reported during the local government elections may have been even greater than during the 2003 general elections. Whereas the worst violence in 2003 was concentrated in the south and southeast, violence around the 2004 local government elections erupted in multiple locations in the north, south, west, east and centre of the country. In an initial non-exhaustive count of incidents reported by the end of March 2004, Human Rights Watch noted at least twenty-two states (out of Nigerias thirty-six states) in which killings and other types of violent clashes were recorded by election observers, journalists and other sources.14 In very late March and early April, there were reports of further incidents, including killings, some of which may have been linked to the aftermath of the elections.15
In addition to the violence, election observers reported widespread rigging of election results. Elections did not take place at all in several locations, in some cases because state or local authorities had chosen alternative dates, and in other cases, such as in Warri in Delta State, because state governments feared that elections would aggravate ongoing violence. The cancellation or postponement of elections was in itself a source of further frustration and anger among some communities. Where elections did take place, there was a very low voter turn-out; observers speculated that people had been discouraged from voting by a mixture of apathy, fear, and disillusion with the 2003 elections.16
 See for example Annual Report 1999, Civil Liberties Organisation.
 Human Rights Watch interview, Port Harcourt, July 14, 2003.
 Do the votes count? Final report of the 2003 general elections in Nigeria, Transition Monitoring Group.
 Election monitoring report on the ongoing Nigeria federal and state general elections, April/May 2003 (executive summary), Environmental Rights Action. Reproduced in Nigeria Today, April 26, 2003.
 Summary of incidents recorded in the presidential / gubernatorial elections in 2003, Transition Monitoring Group (Appendix 8 of the TMGs final report).
 TMG preliminary report on the state houses of assembly elections held on Saturday, May 3, 2003.
 For an overview of the role of police during the elections, see The conduct of security forces by Innocent Chukwuma, in Do the votes count? Final report of the 2003 general elections in Nigeria, Transition Monitoring Group, July 2003.
 Human Rights Watch interviews with PDP secretariat officials, Abuja, July 21, 2003.
 The TMGs preliminary report on the local government elections notes incidents of violence in at least eleven states, as well as several cases where observers were threatened and harassed. Additional incidents were reported by other observers and journalists covering the elections.
 See for example Edo still boils two days after local government poll, Daily Champion, March 30, 2004; and Nigeria youths, security forces battle over disputed election results; 30 feared dead, Associated Press, April 5, 2004. Human Rights Watch has not been able to confirm these reports independently.
 For a more detailed overview of the 2004 local government elections, see Preliminary Report issued by the Transition Monitoring Group (TMG) on the Local Government Council Elections held on Saturday, March 27, 2004. The TMG, which deployed 4,144 observers and 142 monitors to observe the elections, concluded in its preliminary report: It is doubtful whether given the substantial flaws that attended the preparations for the elections in virtually all the states of the federation and the level of irregularities observed on Election Day, the elections can in anyway be considered to be reflective of the will of the people.