All Content Related to Sub-Saharan Africa

Ethiopia

evan — Wed, 09 May 2007 17:09:58 -0400

Background

Ethiopia’s record on human rights and political openness took a turn for the worse after the legislative elections of May 2005. Though originally hailed by the U.S. State Department as “a milestone in creating a new, more competitive multi-party political system in one of Africa’s largest and most important countries,”1the elections were quickly followed by protests and riots by opposition parties alleging voter intimidation and rigging by the Ethiopian People’s Revolutionary Democratic Front (EPRDF).2 Ethiopian police in turn arrested more than 10,000 people in Addis Ababa during the protests. According to Human Rights Watch, most were released within a month, but hundreds remained locked up. The government recently released 400 prisoners in March of last year, but it is unknown how many remain.3 The EPRDF continued its crackdown on opposition by arresting seventy-six “politicians, journalists, and civil society activists”4 and charging them with “‘treason’, ‘conspiracy’ to overthrow the government and ‘genocide.’”5 In foreign affairs, Ethiopia is involved in a border dispute with Eritrea, the subject of a war between the two states from 1998 to 2000.6 Ethiopia, on behalf of the U.N.-recognized transitional government of Somalia, has also entered into conflict with Union of Islamic Courts (UIC). The UIC is an Islamist group vying for control of Somalia.7 Because of the rise of Islamist extremism in Somalia, the United States views Ethiopia as an important ally in the global war on terror. As a result, Ethiopia still receives the largest amount of U.S. aid in sub-Saharan Africa despite U.S. disapproval of the repression following the elections of May 2005.8 Self-censorship in the media is driven by the fear of government reprisal. Foreign journalists have difficulties acquiring authorization to work in Ethiopia, and an Associated Press reporter was sent out of the country early this year after “tarnish[ing] the image of the country.”9 The government has reportedly entered into censorship of blogs and opposition Web sites, though it officially denies doing so.10

Internet in Ethiopia

Ethiopia lags behind much of Africa in Internet availability and is currently attempting a broad expansion of access throughout the country, though efforts have been hampered by the largely rural makeup of the Ethiopian population and the government’s refusal to permit any privatization of the telecommunications market. Only 113,000 people had Internet access in 2005, for a penetration rate of 0.2 percent, one of the lowest in Africa.11 The state-owned Ethiopian Telecommunications Corporation is the sole Internet service provider (ISP) in the country. Internet cafés are a major source of access in urban areas, and an active community of bloggers and online journalists now plays an important role in offering alternative news sources and venues for political dialogue. However, three-quarters of the country’s Internet cafés are in the capital city, and even there access is often slow and unreliable.12

In 2005, Ethiopia announced plans to spend hundreds of millions of dollars over the next three years to connect all of the country’s schools, hospitals, and government offices—and most of its rural population—to broadband Internet via satellite or fiber-optic cable.13 Currently satellite Internet is available to some large corporations, but individuals are not permitted to have private satellite connections. The ETC also bans the use of Voice-over Internet Protocol (VoIP) in Internet cafés and by the general population.14

Legal and regulatory frameworks

The Ethiopian government maintains strict controls over access to the Internet and online media, despite constitutional guarantees of freedom of the press and free access to information.

The state-owned Ethiopian Telecommunications Corporation (ETC) and the Ethiopian Telecommunication Agency (ETA) have exclusive control of Internet access throughout the country. The ETA is not an independent regulatory body and its staff and telecommunications policies are controlled by the national government.15 It grants the ETC a monopoly license as Ethiopia’s sole ISP and seller of domain names under the country code top-level domain, “.et”. Internet cafés and other resellers of Internet services must be licensed by the ETA and purchase their access through the ETC.16 Individual purchasers must also apply for Internet connections through the ETC. Though Ethiopia has considered some limited privatization of the telecommunications market, these plans are on hold indefinitely despite acknowledgments that the ETC has not been an effective service provider.17

In the face of political turmoil over the last two years, the ruling party in Ethiopia has become an increasingly active censor. In mid-2006 the government cut off access within the country to online publications run by political dissidents and to all blogs sponsored by blogspot.com (the ETC claimed that the blockage was a technical glitch but offered no further explanation).18 The government has also banned reporters for the state-run news agency from using the Internet at all and now frequently jails journalists, including online journalists, for charges including treason; most private news outlets have now been shut down.19 The Committee to Protect Journalists named Ethiopia one of the top four jailers of reporters in the world in 2006.20

In late December 2006, the ETA began requiring Internet cafés to log the names and addresses of individual customers, apparently as part of an effort to track users who engaged in illegal activities online. The lists are to be turned over to the police, and Internet café owners who fail to register users face prison.21 Bloggers believe that their communications are being monitored.22 And the state maintains the right to cut off Internet access to resellers or customers who do not comply with security guidelines. In practice, it has shut down Internet cafés in the past for offering VoIP services and other policy violations.23

ONI testing results

ONI conducted testing on Ethiopia’s sole ISP, the ETC. The ETC blocking effort appears to focus on independent media, blogs, and political reform and human rights sites, though the filtering is not very thorough and many prominent sites that are critical of the Ethiopian government remain available within the country.

The prime target of Ethiopia’s filtering is political bloggers, many of whom oppose the current regime. Ethiopia blocks all the blogs hosted at blogspot.com and at nazret.com, a site that aggregates Ethiopian news and has space for blogs and forums. Though many of the filtered nazret blogs are critical of the government, the scope of the filtering is wide: one blocked blogger wrote solely about the 2006 World Cup. The blogspot-hosted sites that are blocked include Ethiopian and international commentators on politics and culture, including popular blogs EthioPundit and Enset.

The Web sites of opposition political parties appeared to be a priority for blocking (www.kinijit.org, www.hebret.com, and others), as did pages for groups that represent ethnic minorities within Ethiopia (www.anaukjustice.org, www.oromia.org). Although women’s rights groups in general were not filtered, the ETC did block one Web site aimed at connecting women involved in politics in Asia (www.onlinewomeninpolitics.org).

Many independent news sites covering Ethiopian politics or compiling international and local coverage were blocked, including CyberEthiopia, the Tensai-Ethiopia radio site, EthioMedia, EthioX, and EthioIndex. But some media sites carrying news and editorials that are unfavorable to the Ethiopian government remained available, including Addis Voice and Ethiopian Review, which had been blocked as part of the ETC’s initial filtering of blogs and media sites in 2006.24 International news sites such as CNN and Voice of America radio were not blocked.

Some human rights sites focusing specifically on Ethiopia were filtered. The Ethiopian Democratic Action League, which advocates for political prisoners, was blocked, as was a page calling for the freedom of jailed opposition leader Yacob Haile-Mariam (www.freeyacob.com) and a site about the imprisonment of human rights activist Mesfin Woldemariam (www.mesfinwoldemariam.org). However, information about these and other imprisoned dissidents is available via a number of human rights pages that are not blocked, including Human Rights Watch, Amnesty International, and various Ethiopian-focused rights groups. Reporters Without Borders, which has chronicled Ethiopian Internet filtering on its Web site (www.rsf.org), is not banned.

ONI testing found that search engines, including Google, Yahoo, MSN, and others, were available in Ethiopia, and no e-mail sites have been blocked. Though VoIP has been banned within the country, sites offering that service, such as Skype, were not filtered. The ETC did not block censorship circumvention tools such as www.anonymizer.com, and Internet users within Ethiopia appear to have found alternative means of accessing banned sites.25

Conclusion

Ethiopia’s current approach to filtering can be somewhat spotty, with the exception of the blanket block on two major blog hosts. Much of the banned political and human rights–related content is available at sites that are not blocked. The authors of the blocked blogs have in many cases continued to write to an international audience, apparently without sanction. But Ethiopia is increasingly jailing journalists and the government has shown an increasing propensity toward repressive behavior online; it seems likely that the trend will be more extensive censorship as Internet access expands across the country. When the ETC becomes more sophisticated as an ISP, its filtering regime may become broader and more comprehensive, particularly if the Ethiopian political situation remains unstable.

NOTES

1. Sean McCormack, “Ethiopian Elections,” U.S. State Department, http://www.state.gov/r/pa/prs/ps/2005/53355.htm.
2. ABC News, “One student killed in Ethiopian election protest,” June 2005, http://www.abc.net.au/news/newsitems/200506/s1385945.htm.
3. Human Rights Watch, World Report 2007: Ethiopia, 2007, http://hrw.org/englishwr2k7/docs/2007/01/11/ethiop14704.htm.
4. Ibid.
5. Reporters Without Borders, Ethiopia: Annual Report 2007, 2007, http://www.rsf.org/article.php3?id_article=20755&Valider=OK.
6. BBC News, “Country Profile: Eritrea,” December 2006, http://news.bbc.co.uk/2/hi/africa/country_profiles/1070813.stm.
7. BBC News, “Ethiopian army faces Somali test,” December 2006, http://news.bbc.co.uk/2/hi/africa/6208759.stm.
8. Human Rights Watch, World Report 2007: Ethiopia, 2007, http://hrw.org/englishwr2k7/docs/2007/01/11/ethiop14704.htm.
9. Reporters Without Borders, Ethiopia: Annual Report 2007, 2007, http://www.rsf.org/article.php3?id_article=20755&Valider=OK.
10. Ibid.
11. Internet World Stats, Internet Usage Statistics for Africa, http://www.internetworldstats.com/stats1.htm.
12. Lynn Hartley and Michael Murphree, “Influences on the partial liberalization of Internet service provision in Ethiopia,” Critique, Fall 2006, http://lilt.ilstu.edu/critique/fall2006docs/Influences on the Partial Liberalization of Internet Service.pdf.
13. Ian Limbach, “Waking up to a laptop revolution: From grand infrastructure projects to small grassroots initiatives in education, technology is bringing about change in the developing world,” Financial Times, March 29, 2006; Xinhua News Agency, “Ethiopian firm launches standard virtual internet service,” February 6, 2006.
14. Economic Commission for Africa, Consultative workshop held on Ethiopia’s new broadband initiative, March 29, 2005, http://www.uneca.org/eca_resources/news/032905disd_dna.htm ; Ethiopian Telecommunication Agency, Telecommunication (Amendment) Proclamation No. 281/2002, July 2, 2002, http://www.eta.gov.et/Button/Scan/Telecom Proc 281_2002 (amendment) NG.pdf.
15. Lynn Hartley and Michael Murphree, “Influences on the partial liberalization of Internet service provision in Ethiopia,” Critique, Fall 2006, http://lilt.ilstu.edu/critique/fall2006docs/Influences on the Partial Liberalization of Internet Service.pdf.
16. Ethiopian Telecommunication Agency, About ETA, http://www.eta.gov.et/Button/About%20ETA.html ; Ethiopian Telecommunication Agency, Licenses, http://www.eta.gov.et/Button/Licenses.htm.
17. Balancing Act News Update, “Ethiopia puts off privatising ETC but rolls out nationwide fibre network,” Issue 272, September 19, 2005, http://www.balancingact-africa.com/news/back/balancing-act_272.html.
18. Reporters Without Borders, “Three more sites unaccessible, government denies being involved,” May 29, 2006, http://www.rsf.org/article.php3?id_article=17783.
19. Julia Crawford, “‘Poison,’ politics, and the press: In Ethiopia’s toxic political climate, Zenawi’s government sweeps up journalists and shuts down newspapers,” Committee to Protect Journalists, April 28, 2006, http://www.cpj.org/Briefings/2006/DA_spring_06/ethiopia/ethiopia_DA_spri....
20. Committee to Protect Journalists, “Internet fuels rise in number of jailed journalists,” December 7, 2006, http://www.cpj.org/Briefings/2006/imprisoned_06/imprisoned_06.html.
21. Groum Abate, “Ethiopia Internet cafes start registering users,” The Capital, December 27, 2006, http://nazret.com/blog/index.php?title=ethiopia_internet_cafes_start_reg....
22. Weichegud! ET Politics, “Clarification,” December 13, 2006, http://weichegud.blogspot.com/2006_12_01_weichegud_archive.html.
23. Groum Abate, “Ethiopia Internet cafes start registering users,” The Capital, December 27, 2006, http://nazret.com/blog/index.php?title=ethiopia_internet_cafes_start_reg....
24. See, for example, CyberEthiopia, “Internet repression in Ethiopia,” September 2006, http://www.cyberethiopia.com/net/docs/internet_repression_in_ethiopia.ht....
25. Carpe Diem Ethiopia, “Blogging in the valley of the shadow of death,” June 2, 2006, http://carpediemethiopia.blogspot.com/2006/06/blogging-in-valley-of-shad... ; Meskel Square, “The magnificent eight,” June 1, 2006, http://www.meskelsquare.com/archives/2006/06/the_magnificent_1.html.

Zimbabwe

miriam — Tue, 08 May 2007 17:52:56 -0400

Background

Zimbabwe’s government is tightly controlled by President Robert Mugabe and the ruling Zimbabwe African National Union-Patriotic Front (ZANU-PF). They have dominated the political landscape since the country’s independence from Great Britain in 1980 and have manipulated political structures to ensure that they stay in control. 1 The ZANU-PF–controlled government is known for its brutal repression and continuing violations of human rights. The best example of this is 2005’s “Operation Murambatsvina,” or “Operation Tsunami,” as it is called locally. 2 Officially described as an effort to eliminate illegal housing and commerce, the “mass evictions and demolitions” were, 3as reported by the U.N., “carried out in an indiscriminant and unjustified manner, with indifference to human suffering, and, in repeated cases, with disregard to several provisions of national and international legal frameworks.” 4Though the actual motivations are unknown, one theory is that the operation was meant to be retribution toward regions in which voters for opposition parties lived.5 Free assembly is dramatically curtailed as the government often violently breaks up peaceful protests under the Public Order and Security Act. 6There have been allegations of police abuse and the torture of detainees.7 A severe press law passed in 2002 allows the Media and Information Commission to crack down on dissent within the media by controlling the licensing of journalists.8 And, finally, the government jams a number of radio stations critical of the government, such as Voice of America, Voice of the People, and SW Radio Africa.9

Internet in Zimbabwe

The number of Internet users in 2005 was reportedly 1,000,000, or approximately 8 percent of Zimbabwe’s population.10 The number of Internet service providers (ISPs) has risen from six in 2003 to twenty-seven in 2004, due to growing demand.11 The Business Mirror in 2005 performed a survey showing that Harare has over thirty Internet cafés, up from about twenty, two years prior.12 The Internet is a less expensive means of communication than the telephone service in Zimbabwe, fueling its growth. In 2004, electronic messaging cost between ZIM $200 (USD0.04 by 2004 exchange rate) and ZIM $250 (USD0.05 by 2004 exchange rate) per minute, while international telephone calls cost between ZIM $3,800 (USD0.72 by 2004 exchange rate) and ZIM $5,800 (USD1.10 by 2004 exchange rate) per minute.13 However, because of limited awareness its capabilities, Internet use is mostly limited to e-mail.14 The low level of Internet penetration overall is likely the result of the increasingly rapid decline of the economy and quality of life in the country over the past seven years. In January 2007, inflation rates reached a staggering 1,593.6 percent.15 The government is bankrupt, eight in ten Zimbabweans are destitute, and workers in Harare see their bus fares to and from work take up their entire salaries.16 In such an environment, demand for luxury goods such as computers and Internet use is low. In September 2006, a large majority of the Internet went offline when the international satellite communications provider, Intelsat, cut service to the country, following the failure of government-owned telecommunications company, TelOne, to pay its debts to the company. Service was restored after the reserve bank paid the outstanding debt.17

Legal and regulatory frameworks

Zimbabwe’s government mainly focuses its regulation of Internet use on e-mail.18 The Post and Telecommunications Act of 2000 allows the government to monitor e-mail usage and requires ISPs to supply information to government officials when requested.19 The Supreme Court, however, ruled in 2004 that the sections of the law that permit monitoring violated the constitution.20

The government struck back with an initiative in 2004 that requires ISPs to renew contracts with TelOne, the government-owned telecommunications company, with the stipulation that they report any e-mail with “offensive or dangerous” content.21 In essence, this requires ISPs to do what the Supreme Court has ruled is unconstitutional. As of yet, no ISPs have signed new agreements.22

The government responded again with the Interception of Communications Bill of 2006. Under its provisions, the government would establish a telecommunications agency called the Monitoring and Interception of Communications Center to monitor, among other things, all telecommunications.23 The government withdrew the bill in November 2006 over constitutionality objections from the Parliamentary Legal Committee and plans to revise it.24 Even without explicit powers, the authorities appear to be pursuing a crackdown on e-mail dissent unabated. In 2005, for example, authorities arrested forty people in a raid on a local Internet café because an e-mail insulting President Robert Mugabe allegedly was sent from the location.25

ONI testing results

ONI testing of two Zimbabwean ISPs, Econet and YoAfrica, revealed no evidence of a filtration regime in the country. Though the ZANU-PF regime is very repressive, this is not an unexpected finding. Internet use in Zimbabwe is extremely low (10 percent) and, as mentioned earlier, is generally limited to e-mail rather than Web-browsing. As a result, Zimbabwe’s main efforts toward control of the Internet are e-mail focused. A large-scale Internet filtration system in all likelihood does not hold much value to the Zimbabwean government relative to the price of its implementation.

Conclusion

Zimbabwe is a highly repressive country with a failing economy and a poverty-stricken population. Internet penetration is extremely low and the Internet is mainly used for e-mail. As a result, the government restricts its efforts toward Internet control to e-mail monitoring and censorship. Though its legal authority to pursue such measures is contested, the government appears to be following through on its wishes to crack down on dissent via e-mail. If Internet usage were to rapidly expand and increasingly spill over to Web browsing, it is likely, given its history, that Zimbabwe would move to Web site filtration. Given the state of the country, however, this does not appear imminent.

NOTES

1. Freedom House, Zimbabwe: Country Report, 2005, http://www.freedomhouse.org/template.cfm?page=22&year=2005&country=6866.
2. U.N. Special Envoy on Human Settlements Issues in Zimbabwe, Report of the Fact Finding Mission to Zimbabwe to Assess the Scope and Impact of Operation Murambatsvina, 2005, http://www.zimbabwesituation.com/zimbabwe_rpt.pdf.
3. Human Rights Watch, World Report 2007: Zimbabwe, 2007, http://hrw.org/englishwr2k7/docs/2007/01/11/zimbab14720.htm.
4. U.N. Special Envoy on Human Settlements Issues in Zimbabwe, Report of the Fact Finding Mission to Zimbabwe to Assess the Scope and Impact of Operation Murambatsvina, 2005, http://www.zimbabwesituation.com/zimbabwe_rpt.pdf.
5. Ibid.
6. Human Rights Watch, World Report 2007: Zimbabwe, 2007, http://hrw.org/englishwr2k7/docs/2007/01/11/zimbab14720.htm.
7. Ibid.
8. Reporters Without Borders, Zimbabwe: Annual Report 2007, 2007, http://www.rsf.org/article.php3?id_article=20744&Valider=OK.
9. Ibid.
10. International Telecommunication Union, World Telecommunication Indicators 2006.
11. Zimbabwe Internet Usage and Marketing Report, 2005, http://www.internetworldstats.com/af/zw.htm.
12. Ibid.
13. Ibid.
14. Ibid.
15. Reuters, “Zimbabwe inflation surges to new record, February 12, 2007, http://www.zimbabwesituation.com/feb13_2007.html#Z1.
16. Michael Wines, “Economic free fall in Zimbabwe,” February 6, 2007, http://www.iht.com/articles/2007/02/06/news/zim.php?page=1.
17. BBC News, “Zimbabwe internet link restored,” September 26, 2006, http://news.bbc.co.uk/2/hi/africa/5382518.stm.
18. Freedom House, Zimbabwe: Country Report, 2006, http://www.freedomhouse.org/template.cfm?page=22&year=2006&country=7092.
19. Reporters Without Borders, “Zimbabwe,” 2004, http://www.rsf.org/article.php3?id_article=10710.
20. Ibid.
21. Andrew Meldrum, “Mugabe introduces new curbs on Internet,” June 2004, http://www.guardian.co.uk/international/story/0,3604,1230096,00.html.
22. ZimObserver News, “IPS(s) block e-mails with political content,” June 2006. http://www.zimobserver.com/index.php?mod=article&cat=Technology&article=....
23. Clemence Manyukwe, “Zimbabwe: Government drops snooping bill,” November 3, 2006, http://www.allafrica.com/stories/200611030016.html.
24. Ibid.
25. APC Blogs, Expression under Repression: WSIS and the Net, 2005, http://blog.apc.org/en/index.shtml?x=2503009.

Kenya's Civil Servants Forbidden to Access Gov Whistleblowing Site?

sally — Fri, 30 May 2008 12:04:00 -0400

Typically, when a government wants to keep people from seeing a web site, they ban or block it from everybody. An article posted on Wikileaks describes another scenario: blocking from one government department to another. The report states that several government officials in Kenya have forbidden those in their departments to access the Kenya Anti-Corruption Commission’s (KACC’s) anonymous reporting service. In addition, “Some ‘senior officials’ have instructed IT personnel to monitor civil servants visits to “forbidden web pages” such as those of the KACC corruption reporting system.”

This is not technical blocking, but appears to be a targeted formal ban on accessing the site, plus a request for some level of surveillance. However, it is unclear how much monitoring could be or is actually taking place.

Like many places, Kenya protects whistleblowers against retaliation, but has no formal requirement that access to whistleblowing sites or hotlines be provided. However, Kenya, along with a number of other countries, committed to combating corruption when it ratified the United Nations Convention against Corruption in 2003. The Convention entered into force at the end of 2005, and meetings of the signing countries occur yearly, to discuss implementation. The Convention includes obligations to aid and protect whistleblowing:

Article 8
Codes of conduct for public officials
4. Each State Party shall also consider, in accordance with the fundamental principles of its domestic law, establishing measures and systems to facilitate the reporting by public officials of acts of corruption to appropriate authorities, when such acts come to their notice in the performance of their functions.

Article 13
Participation of society
2. Each State Party shall take appropriate measures to ensure that the relevant anti-corruption bodies referred to in this Convention are known to the public and shall provide access to such bodies, where appropriate, for the reporting, including anonymously, of any incidents that may be considered to constitute an offence established in accordance with this Convention.

Article 33
Protection of reporting persons
Each State Party shall consider incorporating into its domestic legal system appropriate measures to provide protection against any unjustified treatment for any person who reports in good faith and on reasonable grounds to the competent authorities any facts concerning offences established in accordance with this Convention.

Prohibiting government employees from accessing the KACC's anonymous reporting service, and monitoring the activities of those who visit the KACC's site, runs counter to these obligations, no matter how soft they may be. However, it does not actually stop anyone from reporting. The ban extends only to those directly ordered, so any individual could access the site from anywhere—and could do so without being monitored if they access it from a non-government location.

Blocking a subset of a population from accessing certain Web sites is rare. It happens in the private sector, when corporations block productivity-draining sites, such as personal e-mail services, games, and social tools, from their office networks. However, forbidding another department’s Web site is a new twist.

Internet Watch Report: The 2007 Presidential Election in Nigeria

sally — Thu, 15 Nov 2007 10:37:44 -0500

To view this report as a pdf, with full charts and visuals, click here.

Executive Summary

Despite widespread charges of fraud and disenfranchisement, Nigeria's recent elections were not marked by Internet tampering. While certain sensitive political sites were inaccessible around the time of the elections, these blockages were not caused by intentional tampering but rather by structural problems in Nigeria's faulty telecommunications network. The results of the technical monitoring included no evidence of attempts to block or disable Web sites critical of the current regime, either during or directly preceding the elections. These conclusions were reached through the analysis of tests carried out by the OpenNet Initiative, a partnership between research institutes at the universities of Cambridge, Toronto, Harvard, and Oxford.

The ONI team conducted two types of tests during the election period, which were carried out by a field team in Lagos and researchers in Cambridge, UK. The fi rst testing program, developed by ONI researchers, was run from a standard personal computer. Researchers on the ground in Nigeria used the program to attempt to make contact with a list of politically sensitive sites using several Internet service providers (ISPs) and then sent the results of those tests back to the Citizen Lab at the University of Toronto for analysis.

The second series of tests were run through a specialized computer designed to interact with networks, which was controlled remotely by technical researchers at the University of Cambridge. Testing began a week before the local elections and continued through the national elections. The ONI has been developing technical methods of monitoring for evidence of just-in-time Internet fi ltering or other tampering with Internet access during election periods.

The ONI has conducted Internet-related election monitoring in Kyrgyzstan and Belarus prior to these tests in Nigeria. The ONI ran the election monitoring project in Nigeria because of widespread concerns that the elections would not be free and fair. Observers have claimed that both the local elections of April 14, as well as the national elections on April 21, were marred by blatant and widespread violence, fraud, and disenfranchisement. In reference to the elections on April 14, Peter Takirambudde, Africa Director of Human Rights Watch, stated that “the Nigerian government failed completely in its conduct of a free and fair election” in several key states. Commenting on the presidential elections from the northern town of Kaduna, Max van den Berg, head of the European Union's Observer Mission, noted, “for now the assessment is outspokenly negative ... I'm very concerned.” In addition, the National Democratic Institute went so far as to say that the elections represent “a step backward in the conduct of elections in Nigeria.” Nevertheless, the elections appear to have been free from Internet-related attacks and Web site blocking.

Introduction

The Nigerian Context

Africa’s progress towards stable democratic systems of government since the end of the colonial era has been slow. According to one measure championed by the US-based nongovernmental organization (NGO) Freedom House, in 1976 just three countries in Africa were said to be “free,” while twenty-five were “not free.” Thirty years later, the number of “free” countries has grown to seven while those deemed “not free” has dropped to eighteen.(1) In addition, all but five countries on the continent have held elections within the last five years (though of varying degrees of legitimacy).(2)

Africa has also been slow to enter the digital age, and lags significantly behind other regions of the world in this respect. That said, the past decade has seen a narrowing of the digital divide. Cell phone ownership has risen exponentially in many African countries, with wireless lines often outnumbering land lines. Similarly, access to the Internet is also rising, albeit at a slower pace due to the limitations imposed by forces such as high levels of poverty, illiteracy and lack of basic infrastructure.

The graphs below show this growth for five representative African countries. Although the two graphs have similar growth patterns, note that the final totals are quite different. The highest national figure for cell phones is 20.9 million subscribers, while the highest figure for Internet is only 3.5 million users, both in South Africa.

Nigeria is good example of trends in telecommunications development in Africa. Nigeria has experienced one of the most dramatic increases in cell phone ownership of any country in the world, jumping from 13,000 subscribers in 1995 to over 9 million in 2004. In the same period the number of Internet users grew from under 10,000 to 1.7 million.(3) Strong growth has continued; the number of Internet users in Nigeria in December 2006 and January 2007 were 32,322,202 and 33,603,761 respectively.(4)

Nigeria is also an example of a country progressing towards democratic governance. In 1999 and 2003 the country experienced its first and second presidential elections since the end of military rule, and there is great hope that the age of coups and military dictatorships is firmly in the past.

The intersection of technological and political change in countries such as Nigeria is of interest to ONI researchers. While the Internet offers easier access to information and new means of personal expression, it poses a challenge to traditional political interests, and makes it possible for state authorities to implement new forms of political information control and censorship.

Monitoring the Internet during Elections: the ONI Experience

ONI has shown through its past research that countries that do not filter Internet content on a regular basis may nonetheless control Internet communications during election periods, and may use methods subtler than outright filtering or blocking. For this reason, ONI has begun to undertake investigations of the Internet during elections, with Nigeria as the fourth such effort.

ONI began its monitoring of the Internet during elections with the 2004 US Presidential elections. Testing showed that while President Bush's official Web site was blocked to Internet users outside the country, users within the United States were able to access the site.(5)

ONI election monitoring efforts followed with a larger research project in Kyrgyzstan during the 2005 elections. ONI researchers were able to track two Denial-of-Service (DoS)(6) attacks on ISPs hosting opposition newspapers back to a group of Ukrainian hackers-for-hire. However, ONI was unable to connect this group to the Kyrgyz government. Both the government and the opposition newspapers were able to spin the Web site malfunctions to their own advantage.(7)

In 2006, ONI mounted a third election monitoring project, in Belarus. This project encountered similar challenges to the Kyrgyz project. Several opposition Web sites were inaccessible via the state-owned Beltelecom ISP both on and directly after election day. However, ONI was unable to conclusively prove that DoS attacks had taken place because researchers did not have access to server logs. The fact that these sites were blocked on a network controlled by the regime indicates that official manipulation was the likely explanation. However, ONI found no conclusive evidence of systematic and comprehensive interference by state authorities or their agents with the Internet in Belarus, although it was clear that tampering did occur. The report recommended that election monitoring become a special focus of ONI attention, and that the issue of Internet openness during elections should be raised as an important emerging issue among those groups mandated with measuring the freedom and fairness of elections.(8)

Structure of this Report

In this Internet Watch, we report on ONI's efforts to monitor the April 2007 presidential election in Nigeria. This report is presented in four parts.

Part 1 details why Nigeria was a leading candidate for ONI election monitoring. Nigeria has a long history of electoral manipulation, as well as state control and manipulation of the media. Recently, the government of President Obasanjo initiated legislative and institutional changes to increase the government's ability to control the Internet, with the ostensible goal of curbing cybercrime. In addition, Nigeria's position as Africa's most populous and second-richest country, combined with its active foreign policy, make it an important standard-bearer for African reform—or stagnation.

Part 2 reports on the testing process and findings of the 2007 ONI monitoring carried out during the national and state elections. The testing found no evidence of blockage or interference. While there were several instances where politically sensitive Web sites were unavailable, ONI found that these failures were caused by flaws in Nigeria's weak telecommunications infrastructure rather than deliberate meddling.

Part 3 builds out on the findings and considers a prognosis for Internet freedom in Nigeria. The government's tradition of muzzling dissent along with the new Internet policy environment that is unfriendly towards Internet freedom combine to make filtering seem a possible next step. On the positive side, Nigeria's extremely decentralized telecommunications infrastructure makes the Internet much more difficult to control than in other countries, as the Nigerian Internet does not have fixed ‘choke points’ through which outside content could be blocked or filtered.

Part 4 provides a short summary of the overall findings of ONI testing in Nigeria.

Part 1. Why Test in Nigeria?

The ONI considered Nigeria an important test case for monitoring the Internet during elections for four reasons: 1) a history of electoral manipulation; 2) the critical nature of the election as the nation's first transfer of power between civilian governments; 3) Nigeria's importance as a bellwether of reform for Africa; and 4) the government's growing interest in regulating the Internet, with the stated goal of curbing cybercrime.

A History of Electoral Manipulation

Following the death of Nigeria's military dictator Sani Abacha in 1998, his successor, General Abdusalami Abubakar, began the process of returning the country to democratic governance. Abubakar set up the Independent National Electoral Commission (INEC) to conduct elections for local and national positions and legalized political parties, which had been forbidden under Abacha. Several new political parties were created, of which the All People's Party (APP), Alliance for Democracy (AD), and People's Democratic Party (PDP) were the most prominent. INEC successfully organized elections in late 1998 and early 1999. Olusegun Obasanjo, a former military leader who had been imprisoned by Abacha, was elected president; his party PDP continues to hold power to this day. Although there were instances of ballot falsification and ballot box stuffing, especially in the perennially troubled Niger Delta oil region, the theme for the day was hope and progress as Nigeria moved out from under the weight of thirty-three years of military rule.

In 2003, Obasanjo ran successfully for re-election, though in this second outing the press and international observers were decidedly more critical of the election proceedings. The National Democratic Institute (NDI) witnessed “vandalised, stolen and stuffed ballot boxes”(9) while the leader of the European Union observer delegation, Max van den Berg, said that he was “very concerned” about the voting (he was to repeat the same phrase during the 2007 election).(10) In addition, Human Rights Watch issued a report which argued that at least one hundred people had been killed as a result of political violence.(11)

It appeared that, after only four years in power, the PDP was consolidating their influence, leading one BBC journalist to title his analysis of the 2003 elections “Nigeria's One-Party Creep.” In 2005, Obasanjo began an unsuccessful attempt to do away with the Constitution's term limit clause which prevented him from seeking a third term. After failing in this effort, he turned his sights on assuring the 2007 victory of his chosen successor and PDP presidential candidate, Umaru Yar’Adua, the governor of Katsina State.

Obasanjo and the PDP used the courts to keep hundreds of opposition candidates off the ballot, including Atiku Abubakar, the Vice President and Obasanjo's former ally (Abubakar’s candidacy was later reinstated by the Supreme Court). In the words of The Economist, “the lengths to which Mr. Obasanjo's ruling People's Democratic Party (PDP) has gone to cling to power have undermined and discredited so many of Nigeria's institutions and office-holders that the country now seems more a prisoner of its bleak past than a beacon for the future.”(12) The nation's vast oil riches (valued in 2005 at USD 50 billion per year), combined with limited checks on the executive branch and minimal accountability in the use of public funds, makes Nigeria a tempting prize for any political party. Given these circumstances, the PDP, not surprisingly, seemed unwilling to cede power.

Hopes for a Legitimate Transfer of Power

This 2007 election was to be historic, the first time in the nation's forty-seven year history since independence that power would pass from one ruler to another based upon the results of an election rather at the barrel of a gun. While the election of 1999 left many with hope for Nigeria's future, it was not a contest between an incumbent and a challenger, but rather a contest to decide Abacha's successor. The election in 2003 was also an inconclusive test of the health of Nigeria's democracy. Even in established democracies like the United States, incumbent candidates regularly have an advantage over challengers when running for re-election because they can marshal some of the powers of the state behind their campaigns.(13) Obasanjo's re-election in 2003 did not conclusively point to a dysfunctional electoral system as the president's electoral success could be attributed simply to incumbent advantage.

It was for this reason that the 2007 election was so important. It was a measure of whether Nigeria was really on the path to a stable democratic system of government or, as many feared, elections were being used to legitimize a strongman political system in which he who has the oil makes the rules.

With Repercussions Beyond Nigeria’s Borders

Why does Nigeria's political situation matter? In many ways, Nigeria is a bellwether for reform in Africa. It is the continent's most populous country; remarkably, one in five Africans is Nigerian. Its vast oil wealth also makes it Africa's second richest nation after South Africa. In fact, Nigeria's GDP makes up a mammoth 55% of the economic output of the West African region. The United States alone imports more than 1 million barrels of crude oil from Nigeria every day and has pegged part of its strategy of reduced dependence on Middle Eastern oil to imports from Africa.(14)

Nigeria's most significant international influence is in Africa. It helped found the Organization for African Unity (now the African Union) in 1963 and President Obasanjo served as the organization's chairman from 2004 to 2005. Nigeria contributed significant numbers of troops and financial resources to regional and UN-sanctioned peacekeeping forces that intervened in Liberia's civil conflict in the 1990s. “When there are crises, the countries have looked upon Nigeria to be an arbitrator,” said Ngozi Okonjo-Iweala, Nigeria’s former Minister of Finance and Foreign Affairs.(15) Nigeria has the economic power, regional influence, and visibility to mark a path to improved governance on the continent—or to excuse other smaller countries from backsliding into dictatorship and political inertia.

A New Internet Policy

In addition to the political situation, the country's changing relationship to the Internet made this election a significant one to monitor. In 2004, the Nigerian Cybercrime Working Group was established by President Obasanjo. The group is charged with “the responsibility to create a legal and institutional framework for securing computer systems and networks in Nigeria and protecting critical information infrastructures in the country.”(16)

In 2005, the “Computer Security and Critical Information Infrastructure Protection Bill” was introduced into the National Assembly. The bill is concerned with cybercrimes such as hacking and identity theft, and provides law enforcement with the power to collect digital evidence.(17) If passed, it would make unlawful many undesirable Internet activities, such as spamming, data forgery, computer fraud, cyber-terrorism, and unlawful interception, and it would require that ISPs keep records of all traffic and subscriber information for a period of time to be determined by the president.(18) While it has received a good deal of attention, the Bill appears to be stuck in Assembly and has not yet been voted on.(19)

Even though the Bill is focused on providing measures for combating cybercrime, it leaves open the possibility of abuse. At the time of the Bill’s introduction, local press noted several provisions in the Bill that appear overbroad, in comparison to similar Bills in other jurisdictions. “[T]here are no checks and balances provisions whatsoever in the Bill. There are no mandatory reporting procedures to either the Nigerian parliament or the Nigerian judiciary on the activities of law enforcement agencies in carrying out these wiretapping or lawful interception activities.”(20) There appears to be no recourse for individuals whose civil liberties are compromised. In addition, it was noted that there is, “in the matter of obtaining a warrant [f]or release of information for legitimate reasons, a recurring reference to either a Court of Law or ‘any other lawful authority’ for obtaining either the warrant or the information. There is no enumeration or definition anywhere in the Bill, as has been done in other jurisdictions, of whom or what constitutes ‘lawful authority.’”(21) A number of other worries have been listed, including concern over the security of personal data storage and transmission at ISPs; the ability of the president to make additional regulations pertaining to provisions in the Bill; and a provision that allows for data to be released without a warrant in extreme circumstances without an accompanying provision that requires a warrant to be sought with similar urgency.(22)

In 2007, the Directorate for Cybersecurity (DfC) was founded to “respond to security issues associated with growing usage of Internet and other information and communication technologies (ICTs) in the country.”(23) In its first year of existence the new agency was given a hefty budget of USD 9.3 million to accomplish its goal.

These recent actions demonstrate that the Nigerian government would like to control online activities more effectively within the country. While there have been no allegations of Internet blocking within Nigeria so far, the elections served as an excellent opportunity to test Internet freedom there.

Within the Context of Limited Freedom of Expression

These threats to Internet speech arise within a national environment of limited freedom of expression. By one measure, the US government-backed NGO Freedom House currently classifies Nigeria as “partly free” (an electoral, but not liberal, democracy) and gives it a freedom rating of 4 out of 7, where 1 represents the most open societies and 7 the least.

Some newspapers in Nigeria have been controlled by political interests. Two governors, Orji Uzo Kalu and James Ibori, whose terms expired with the spring 2007 elections, are reported to own newspapers The Sun and The Daily Independent, respectively.(24) They both now face charges related to corruption while in office, and Orji Uzo Kalu has fallen out with his former party, PDP.(25) The Guardian, which is widely considered to be an open and objective newspaper and is one of the country's most widespread papers, is also owned by the Ibru family,(26) who are also close to Obasanjo.(27)

In addition, journalists who write about the misdoings of politicians have been harshly punished. In 2005, Owei Kobina Sikpi, publisher of the Weekly Star, was secretly detained for nearly a week and then charged with seven counts of false information after his newspaper published an article alleging that a state governor was involved in money laundering.(28) That same year, members of the State Security Service (SSS) raided the offices of the Lagos-based weekly The Exclusive and confiscated more than 200 copies of the tabloid in an effort to censor coverage of ethnic Igbo nationalist groups.(29) Given the level of corruption in Nigeria, however, it is noteworthy that there is a certain level of press freedom. There are a number of smaller newspapers not owned by politicians, and many print articles that include negative speech about the government.

Because many Nigerians do not or cannot read, radio is an important source of information.(30) In a positive 2005 development, the Nigerian Broadcasting Commission (NBC) decided to auction broadcast network licenses to private operators. (Radio had previously been the unique domain of the government.) This change in the radio market primarily benefited large broadcasting companies. At the same time, the NBC opened a path for community broadcasting licenses for nonprofit stations run by members of a particular locality, though the license regime for community broadcasting is largely considered prohibitive. Despite these recent moves towards greater openness, the Nigerian government maintains control over what is broadcast by refusing or rescinding licenses. The vulnerable position of regime critics makes the Internet a likely next frontier for political control.

Box 1. The Nigerian Elections in the Blogosphere

Nigeria’s bloggers are overwhelmingly affluent and urban, and are thus hardly a representative population. Nevertheless, they offer important first-hand accounts of the election. These bloggers, acting as citizen journalists, noted a salient point overlooked by the mainstream media: while many voters were eager to participate in the democratic process, voter apathy was a considerable problem. Ore of “Ore’s Notes” described her experience voting in the local elections on April 14, 2007. “I chided myself for not getting there earlier, but I heard from people who had been there at 8AM, as instructed, that the INEC [election commission] officials had not yet arrived at that time. Thank goodness the line moved (albeit very slowly) and there were interesting conversations going on around me to participate in and listen to.”(A1) Funmi of “Funmi Iyanda’s Blog” had a similar experience on the 14th. “I … stood in line in the sun for two hours, hat and sunglasses firmly on, large bottle of water in hand as resolutely determined as most of my fellow countrymen to cast my vote,” she wrote.(A2) However, during the presidential elections a week later, her experience was quite different. “This weekend, the polling booth was a ghost town, my people had lost hope. I voted and left.... I had observed proceeding from the Alimosho area [of Lagos], through Agege, Ikeja, to Maryland and the apathy was palpable. The streets were empty as boys took to the highway playing football.”(A3) Since many voter tallies were falsified, the actual rates of voter turn-out were often not recorded.(A4) Blogger accounts offer important qualitative evidence of true participation rates.

Box 2. Obasanjo Hot and Cold

The outgoing president, Matthew Olusegun Aremu Obasanjo, is a contradictory fi gure. His admirers note that he is Nigeria’s first democratically elected leader since the 1960’s. He also wears the badge of a dissident; he was imprisoned by the military dictator Sani Abacha for criticizing his regime. In addition, Obasanjo has a reputation as a corruption fi ghter. His economic reforms inspired the Paris Club to forgive $18 billion of Nigeria’s debt in 2006(B1) and he has served on the advisory council of Transparency International. Yet recent actions have seriously damaged his reputation. In 2005 he pushed the National Assembly to alter the constitution to allow him to run for a third term,(B2) allegedly offering bribes to lawmakers willing to support him.(B3) Once this effort failed, he set about the task of ensuring that power passed to a member of his own party. According to the Economist magazine, “Mr. Obasanjo has pursued a highly partisan campaign by manipulating and abusing the very institutions, such as the elections commission and the anti-corruption agency... that were touted as paragons of his reforms, so undermining their hard-won credibility in the eyes of many Nigerians.”(B4) We now doubt whether history will look fondly upon Obasanjo or his political legacy.

Part 2. Monitoring the Internet During the Nigerian Elections

ONI monitored the Nigerian Internet for two weeks, beginning a week before the April 14, 2007 local elections and ending two days after the April 21, 2007 presidential elections. No filtering of Internet content, deliberate or accidental, was detected by ONI research.

Methodology: How we tested

Much of the testing took place from within Nigeria. ONI employed two methods to monitor the status of the Nigerian Internet and accessibility of Web sites. First, a field team based in Lagos employed an effective ONI tool to test a comprehensive list of potentially vulnerable Nigerian Web sites. These included political and independent media sites, as well as other sites critical of government or key Nigerian officials. Tests were carried out across five different Nigerian Internet service providers (see Table 1). Test results were then cross-checked against results of the same test run from ONI’s control locations outside of Nigeria to determine whether failed connections were due to internal blocking, which would result in an error message only in Nigeria, or a problem originating at the Web site itself (or more general failures in the Internet), which would result in an error message occurring regardless of location.

Second, the ONI field team deployed a dedicated testing machine designed to measure the status of the Nigerian Internet throughout the testing period, recording local network failures and anomalies. This machine was controlled remotely by operators at the University of Cambridge, providing a platform for ONI researchers to perform testing metrics designed to eliminate technical failure as a possible cause of the inaccessibility of websites.

How we chose which ISPs to monitor

In selecting which ISPs to test the team was guided by three criteria: 1) likelihood of being filtered, 2) infrastructural diversity, and 3) availability of dial-up service.

According to the first criterion, Nitel was the most significant ISP to monitor. It is the historic national operator and was owned by the Nigerian government until it was privatized in 2006.(31) Despite being under new ownership, Nitel operates as if it were still government-owned. (The field team had to submit passport-sized photos in order to sign up for an account, making the process feel more like a visa application than buying Internet service.) In addition, one ONI field researcher recalled that a director at Nitel's main facility in Lagos asked the field team to write a letter to Abuja to request permission to place ONI equipment in the building, despite the fact that the Nigerian government should technically no longer have any control over Nitel.

Table 1: ISPs Tested
Name	Type of Internet gateway
Hyperia	Highway Africa satellite (Nairobi)
Linkserve	SAT-3 submarine fiber-optic cable (Lagos)
Cyberspace	SAT-3 submarine fiber-optic cable (Lagos)
Nitel	SAT-3 submarine fiber-optic cable (Lagos)
Tara	SAT-3 submarine fiber-optic cable (Lagos)

The second criterion dictated that ONI test a variety of Internet gateways. Because all the Web sites on the test list were hosted outside of Nigeria, it was important to test the integrity of a variety of different paths to the external Internet. There are several reasons why ONI only tested sites located outside of Nigeria. In testing for election-oriented filtering, ONI wished to test opposition sites, which are usually hosted outside of Nigeria. For Africa in general, few sites are hosted on the continent; they are more often hosted by services in other parts of the world.(32) Filtering is also more likely to be found on sites hosted outside the country, as filtering is one of the only ways that a group can limit access to a rival’s Web site if it is located elsewhere (rather than simply pulling the plug on servers located in-country).

In Nigeria, the most common types of Internet connection are VSAT (very small aperture terminal) and fiber-optic cable. VSAT is a satellite connection designed for easy deployment which uses a satellite dish of less than 3 meters in diameter. Most large businesses, particularly banks, use VSAT for their Internet connection because it allows them to bypass Nigeria's inadequate telecommunications infrastructure. These businesses use their VSAT dishes to connect to privately-owned telecommunications satellites which route their Internet traffic. The VSAT-based ISP we used, Hyperia, accesses the Internet through a satellite owned by Highway Africa, a news agency based in Nairobi, Kenya.

The second common way to connect to the external Internet is by fiber-optic cable. There is only one fiber-optic cable that connects to Nigeria, the SAT-3/WASC (South Atlantic 3/West Africa Submarine Cable). SAT-3, as it is called, runs from South Africa to Portugal and makes landfall in nine countries along Africa's west coast. SAT-3 is controlled by a closed consortium of historical telecommunications companies, one for each country it connects to.(33) Nitel is the consortium member in Nigeria, and as such controls all Nigerian access to the SAT-3 cable. Other ISPs who wish to use SAT-3 to connect their clients to the Internet must pay Nitel for the privilege.

Monopoly over the SAT-3 cable gives Nitel an excellent opportunity to filter the Internet traffic that passes through that gateway. For this reason, ONI decided to test not only Nitel, but also three ISPs that rent bandwidth from Nitel: Linkserve, Cyberspace, and Tara.

What we found

There were several sites that were inaccessible at varying times during the testing period, among them the sites belonging to the BBC, Amnesty International, the Human Rights Library at the University of Minnesota, and the Independent National Electoral Commission (INEC). Among these sites, the inaccessibility of INEC was of the most concern because it was inaccessible for the longest period of time, from April 12 through 14, 2007.

However, ONI researchers determined that all the sites, including INEC, were inaccessible due to technical errors in the Nigerian network. In the case of INEC, for example, the site was most likely inaccessible due to an incorrect domain name system (DNS) configuration. DNS connects a Web site's name (www.inecnigeria.org) with its IP address (66.226.64.30). If the DNS is improperly configured it will fail to connect the user to the server on which the Web site is stored and an error message will appear.

This analysis is consistent with what Nigerian experts told the field team about the technical weaknesses of the Nigerian Internet. Both the director of the Nigerian Internet Exchange and the CFO of Tara Systems told members of the field team that they were likely to come across errors due to malconfigured DNS.

This technical diagnosis also makes sense from a political perspective. It is highly unlikely that the regime would block a government agency, especially given that INEC, despite its name, is far from independent. Throughout the election process, Obasanjo has used INEC as a tool in his effort to rig the election in favor of his own party.(34) This included a ruling in which INEC disqualified Atiku Abubakar, one of Obasanjo's most bitter enemies, from running for president. The Nigerian Supreme Court later reinstated Abubakar's candidacy, stating that INEC did not have the right to disqualify candidates.(35) Given its reputation as a staunch ally of the ruling regime, it is highly unlikely that the INEC site was blocked by the government, furthering supporting the technical failure hypothesis.

Part 3: And so, is the Internet Under Threat in Nigeria?

Although no blocking or filtering was discovered by ONI research during the April 2007 elections, there appear to be reasons to maintain vigilance.

As mentioned in Part 1, there have been several legislative and structural changes in the Nigerian government's Internet policy. Specifically, the recent creation of the Directorate for Cybersecurity could serve as a new locus for increasing constraints on free expression over the Internet. Moreover, the Computer Security and Critical Information Infrastructure Protection Bill of 2005 could grant new Internet surveillance powers to the central government with a low level of external oversight.

The Will to Censor but Perhaps Not the Means

In Nigeria, significant obstacles to Internet blocking remain. Ironically, the Nigerian government may be foiled by their own refusal to build and maintain telecommunications infrastructure. This lack of reliable infrastructure has led those who can afford it to build their own. Most major businesses, and even many private individuals, own diesel-powered electric generators. The “do it yourself” trend applies to Internet infrastructure as well. Those who can afford the expense are opting for private satellite connections via VSAT (see Part 2) instead of Nigeria's own fiber optic and telephone infrastructure. As a result, a significant percentage of Nigeria's Internet traffic travels through infrastructure that the Nigerian government does not control, namely private communication satellites operated by independent companies in foreign countries.

The Nigerian government could theoretically filter international Internet traffic through the SAT-3 fiber optic gateway and domestic Internet traffic through the NIXP Internet exchange. Both pieces of infrastructure are located in the same building in Lagos, which is owned by Nitel.

However, because of the existence of private satellite connections, Nigeria's filtering could never be as thorough as that found in, for example, China, Iran, or Tunisia. Those who could afford satellite connections would be able to bypass filtering mechanisms on Nigeria's land-based infrastructure.

In order to effectively filter the Internet, the Nigerian government would have to enlist the private ISPs, mandating that they filter their clients’ Internet requests. ISPs that are private companies are less reliable allies in censorship than a government agency, as their prime concern is the bottom line, not political control. This effort to engage ISPs in filtering has not been made, though the Critical Information Infrastructure Protection Bill is making movements in this direction.

In sum, while filtering is possible along Nigeria's phone and fiber optic infrastructure, the presence of private satellite connections to the Internet presents a hurdle to comprehensive filtering.

The Importance of Monitoring the Internet around Elections

The combination of possible threats to the openness of Nigeria's Internet and a history of corruption make monitoring all the more important. Projects such as the ONI may help to highlight the importance of ensuring Internet openness and could contribute to making such openness a criterion for judging the freedom and fairness of elections. While ONI research did not find evidence of filtering during the April 2007 elections, conducting this type of testing is important in establishing norms that can act against future limitations on freedom of expression on the Internet.

Part 4: Summary

Despite legitimate concerns about the openness of the Nigerian Internet, ONI research found no evidence of filtering during the April 2007 election. Although some sites were inaccessible during the testing period, these errors were ascribed to malconfigured DNS which failed to connect Web site names with their IP addresses.

Nevertheless, the Nigerian Internet may be under threat in the future. The fraud and violence perpetuated by the Nigerian government and documented by international observers, which led to voters being disenfranchised during the elections, suggests that the current leadership may be willing to go to some lengths to ensure a desired political outcome.(36) Moreover, the creation of the Directorate for Cybersecurity, along with cybercrime legislation in 2005, indicates that the Nigerian government may be interested in increasing its capacity to control the Internet.

Note: all photos are licensed under a Creative Commons Attribution – Non-Commercial – Share-Alike License, credited to Mary Joyce

Notes

(1) Lydia Polgreen, “Africa’s Crisis of Democracy,” The New York Times, 23 April 2007, online edition (www.nytimes.com); Map of Freedom 2006 at www.freedomhouse.org

(2) The countries which have not recently held elections are Angola, Cote d'Ivoire, Eritrea, Libya, and Somalia. In this context, “recent” means within the last five years.

(3) International Telecommunication Union Yearbook of Statistics: Telecommunication Services Chronological Time Series 1995-2004, (Geneva, Switzerland: ITU, 2006), 126.

(4) Nigeria Communications Commission subscriber data 2001-Jan. 2007, www.ncc.gov.ng/subscriberdata.htm.

(5) OpenNet Initiative, “Geolocation filtering: www.georgewbush.com blocked during run-up to election,” (OpenNet Initiative Bulletin 007, October 27, 2004), http://opennet.net/bulletins/007/.

(6) In a Denial-of-Service (DoS) attack a hacker or group of hackers, usually with the help of automated “bots,” make multiple simultaneous requests to a Web site they wish to disable. The goal of the attack is to overwhelm the server on which the Web site is hosted so that it can no longer respond to all the page requests and displays an error message when legitimate Internet users try to access the site. This can be a preferred method of Internet censorship because it is difficult to prove that the Web site is inaccessible as a result of a DoS attack rather than because of an overwhelming number of hits by legitimate users.

(7) OpenNet Initiative, “Election Monitoring in Kyrgyzstan,” (OpenNet Initiative Special Report, February 2005), http://opennet.net/special/kg/.

(8) OpenNet Initiative, “The Internet and Elections: The 2006 Presidential Election in Belarus (and its implications),” (OpenNet Initiative Internet Watch Report, April 2006), http://www.opennetinitiative.net/studies/belarus/ONI_Belarus_Country_Stu....

(9) Joseph Winters, “Analysis: Nigeria's One-Party Creep” BBC News, 21 April 2003, http://news.bbc.co.uk/2/hi/africa/2964759.stm

(10) ?Ibid.

(11) ?Nigeria’s 2003 Elections: The Unacknowledged Violence (Human Rights Watch, 2004), 1.

(12) “How to Steal Yet Another Election” The Economist, April 21, 2007, Vol. 383 Issue 8525, p51-52, 2p, 1c (Hereafter referred to as “How to Steal”).

(13) Some of these advantages that US elected officials hold are laid out in “Why are Sitting Members of Congress Almost Always Re-elected?” Citizens for U.S. Direct Initiatives, 2003, http://www.cusdi.org/reelection.htm. The authors analyze some of the numbers at play in the 1998 mid-term Congressional elections.

(14) Jad Mouawad, “Growing Unrest Posing a Threat to Nigerian Oil” The New York Times, April, 21 2007, online edition (www.nytimes.com).

(15) “Regional Giant Nigeria Looms over West Africa” The Online Newshour, April 5, 2007, online edition, http://www.pbs.org.

(16) Nigerian Cybercrime Working Group homepage (http://www.cybercrime.gov.ng).

(17) Computer Security and Critical Information Infrastructure Protection Bill 2005, http://www.cybercrime.gov.ng/site/index.php?option=com_content&task=view...

(18) Ibid., § 11.

(19) ?Supra note 16.

(20) ?Eijeagbon Ohicheoya, “Nigeria: New Wire Tapping, Cyber Crimes Bill in Nigeria,” This Day (Lagos), Oct. 18, 2006, available at http://www.infosecnews.org/hypermail/0610/12262/html.

(21) ?Ibid.

(22) ?Ibid.

(23) ?Shina Badaru, “FG Okays N1.2b for Cybersecurity Directorate” This Day (Lagos), April 6, 2007, available on allAfrica, http://allafrica.com.

(24) Interview, in-country researcher with Mary Joyce, OpenNet Initiative, April 2007.

(25) ?Email from Dan Smith, Brown University, to Sally Walkerman, OpenNet Initiative, Oct 10, 2007.

(26) Supra note 24.

(27) Supra note 24.

(28) ?“Nigeria,” Freedom in the World (Freedom House, 2006).

(29) ?Ibid.

(30) ?As of 2004, literacy was estimated at 68% (“Nigeria,” The World Factbook, CIA, http://www.umsl.edu/services/govdocs/wofact2004/geos/ni.html). Radio, especially call-in or SMS-in radio, has increasingly been a mechanism for information exchange in Africa (Ethan Zuckerman, “What is Citizen Media,” Panel Presentation, September 21, 2007). NGO Global Knowledge Partnership has found that radio is a good method of communication for locals implementing development projects (See “Virtual Consultations: Build on Local Resources”, http://www.globalknowledge.org/gkps_portal/index.cfm?menuid=184&parentid...).

(31) The current owner of Nitel is the Nigerian investment company Transcorp.

(32) ?Infrastructure, security, and ease of set-up are factors that Web site owners consider when determining where to host a site. For a general discussion, see Ronald Deibert et. al., eds., Access Denied (forthcoming 2008). Regarding security, hosting a site outside the country can help a Web site owner maintain anonymity. See Ethan Zuckerman, “How to Blog Anonymously,” Handbook for Bloggers and Cyber-dissidents, Reporters Without Borders, Sep. 2005, http://www.rsf.org/IMG/pdf/handbook_bloggers_cyberdissidents-GB.pdf.

(33) ?This arrangement, which gives every consortium member monopoly rights to the cable in their country, is very worrying to advocates of an open telecommunications market.

(34) ?“Abuse of State Power,” Human Rights Watch, http://hrw.org/backgrounder/africa/nigeria0407/6.htm

(35) ?“INEC to Allow Atiku Abubakar to Contest at Court’s Behest,” INEC Bulletin (March 14, 2007), http://www.inecnigeria.org/uploaddocs/Bulletin%20of%2014%20March%202007%20(P1).pdf.

(36) ?For example, see “Election or Selection?: Human Rights Abuse and Threats to Free and Fair Elections in Nigeria” Human Rights Watch, http://hrw.org/backgrounder/africa/nigeria0407/index.htm; “Election Monitoring Report,” Network of Mobile Election Monitors (NMEM), eds., http://www.kiwanja.net/miscellaneous/NMEM_Election_Report.pdf. This report found that, while in some areas voting went more smoothly than in previous elections, the overall level of documented fraud indicates that there is still significant need for improvement.

Box Footnotes

(A1) ?Ore’s Notes, “Voting in Progress,” April 14, 2007, http://orenotes.blogspot.com/2007_04_01_archive.html.

(A2) Funmi Iyanda’s Blog, “Power from the People,” April 23, 2007, http://fiyanda.blogspot.com/2007_04_01_archive.html.

(A3) Ibid.

(A4) Ben Rawlence and Chris Albin-Lackey, “Briefi ng: Nigeria’s 2007 general elections Democracy in retreat,” 106 African Affairs 497.

(B1) ?“Nigeria settles Paris Club debt,” BBC News, April 21 2006, http://news.bbc.co.uk/2/hi/business/4926966.stm.

(B2) ?Toye Olori, “Uproar over Obasanjo’s third term campaign,” News from Africa (December 2, 2005), http://www.newsfromafrica.org/newsfromafrica/articles/art_10550.html.

(B3) ?Craig Timberg, “Nigerian Senate Blocks Bid for 3rd Presidential Term,” Washington Post (May 17, 2006), http://www.washingtonpost.com/wp-dyn/content/article/2006/05/16/AR200605....

(B4) ?“How to Steal,” The Economist.

Internet Watch Report: The 2007 Presidential Elections in Nigeria

sally — Thu, 15 Nov 2007 09:47:18 -0500

ONI is pleased to release our latest Internet Watch Report, "The 2007 Presidential Election in Nigeria" (here in HTML). The report documents the ONI's efforts monitoring the Internet in Nigeria during elections held in April 2007. No filtering was found, either during the elections or in the two weeks preceding, though several sites were down at times because of faulty infrastructure. This Internet Watch Report also discusses some of the recent changes in the legal environment that governs the Internet in Nigeria and identifies some potential emerging points of control.

Nigerian Internet Remains Free of Tampering During Tumultuous Elections

Rob — Mon, 30 Apr 2007 16:17:23 -0400

Lagos, Nigeria: Nigeria’s local and national elections of April 2007 were marked by widespread charges of fraud and disenfranchisement. A team of election monitors from the OpenNet Initiative (ONI) traveled to Nigeria during the elections to test for evidence of Internet tampering. Certain sensitive political sites were inaccessible around the time of the elections, but these blockages appear to be caused solely by structural problems in Nigeria’s faulty telecommunications network. ONI’s technical monitoring did not reveal evidence of deliberate attempts to block or disable websites critical of the current regime, either during or directly preceding the elections. These conclusions were reached through the analysis of tests carried out by the ONI, a partnership among research institutes at the universities of Cambridge, Toronto, Harvard, and Oxford.

“Nigeria was an important test case given its size and importance to the African continent," said Rafal Rohozinski, an ONI principal investigator and team leader for the ONI’s election monitoring work. “This result may be atypical given the reported increase in what can be termed ‘just in time filtering’ of time sensitive Internet and communications services such as SMS that we have observed around other elections, including in several other African countries."

"The Internet is increasingly an important venue for access to information and discussion among citizens during election periods around the world,” said John Palfrey, one of the ONI’s principal investigators and executive director of the Berkman Center for Internet & Society at Harvard Law School. “The Nigerian elections of 2007 were hardly a case study in freedom and fairness, but our research offers a bit of good news: there was no evidence of anyone tampering with the ability of Nigerians to access information online during the key period of the campaigns."

The ONI team conducted two types of tests during the election period, both of which were carried out by a field team in Lagos. The first testing protocol monitored activity related to a set of politically sensitive sites accessed through a number of Nigerian Internet service providers (ISPs). The results of these tests were analyzed by researchers at the Citizen Lab at the University of Toronto. A second series of tests was run by researchers from the University of Cambridge scanning for anomalies in overall network traffic. Testing began a week before the local elections and continued through the national elections under the coordination of a team of researchers from the Berkman Center for Internet and Society at Harvard University.

The ONI has been developing methods of monitoring for evidence of event-based Internet filtering or other tampering with Internet access during election periods. In 2004, the ONI mounted Internet-related election monitoring efforts in Kyrgyzstan and Belarus in cooperation with local partners.

ONI ran the Internet election monitoring project in Nigeria because of widespread concerns that the elections would not be free and fair. Observers have claimed that both the local elections of 14 April, as well as the national elections on 21 April, were marred by blatant and widespread violence, fraud, and disenfranchisement. In reference to the elections on 14 April, Peter Takirambudde, Africa Director of Human Rights Watch, stated that "the Nigerian government failed completely in its conduct of a free and fair election" in several key states. Commenting on the presidential elections from the northern town of Kaduna, Max van den Berg, head of the European Union’s observer mission, noted "for now the assessment is outspokenly negative ... I’m very concerned." In addition, the National Democratic Institute (NDI) went so far as to say that the elections represent "a step backward in the conduct of elections in Nigeria."

Nevertheless, the elections appear to have been free from Internet-related attacks and website blocking. According to local statistics from 2006, only 1.1% of Nigerians have access to the Internet. As the Internet becomes a more important source of political information, many states around the world have begun to filter the Internet more extensively. The absence of tampering with the Internet during this election cycle does not mean that the network in Nigeria will remain free of censorship.

"When very few people have Internet access, the state has few reasons to worry about monitoring or censorship. It’s when the Internet begins to serve some of the roles of mass media that governments can perceive and respond to a threat," said Jonathan Zittrain, an ONI principal investigator and professor of internet governance at Oxford University.