TEMCO'S OBSERVATION: A CRITICAL EVALUATION

TEMCO was founded in April 1994 by twenty-two member organisations, including the University of Dar es Salaam.Footnote ¹ This number increased to sixty non-governmental organisations (NGOs) in 2000. Unlike other observers, TEMCO has a national character. It normally covers elections countrywide and for the longest period ranging from registration, nomination, campaigns, voting and counting to declaration of results. In 1995, TEMCO started observation on 6 August and continued until 31 October. It deployed 136 monitors in the 180 constituencies in mainland Tanzania. Likewise, in 2000, observation commenced on 17 August and continued until 31 October. In that election, TEMCO had 129 constituency observers. It is unfortunate that in its report for the 2005 general elections, TEMCO did not state the coverage of its observation. However, it notes the starting time of observation, November 2004, when preparation of the Permanent Voters' Register (PNVR) was underway. Methodologically, TEMCO employs a wide range of techniques to solicit data such as observation, interviews with key stakeholders in elections (NEC, political parties, voters, etc.), and reports from other observers, particularly the media.

In executing its activities, TEMCO depends on donor money. In the 1995 elections, for example, TEMCO received funds from six donors: Denmark, Finland, Norway, Sweden, the Netherlands, and the Commission of the European Union. TEMCO submits: ‘they made us a generous and timely grant, without any conditions attached, except that money be spent as planned and be fully accounted for’ (TEMCO 1997: 6). Interestingly, in its reports for the 2000 and 2005 elections, TEMCO did not spell out whether it received funds free of conditionalities from the donor community. In its report for the 2000 elections TEMCO (2001: x) noted: ‘we are also grateful to the Basket Group of Donors led by the Royal Danish Embassy for their financial support’. In its report for the 2005 elections it remarked: ‘we would like to give special thanks to the Donors Basket (coordinated by UNDP) whose financial assistance made it possible to carry out the election observation work’ (TEMCO 2006: xii). The silence of TEMCO in its reports for the 2000 and 2005 elections leaves a lot to be desired. In a spirit of transparency, TEMCO ought to disclose its financial independence for the purpose of raising its legitimacy and credibility. I shall tackle the issue of budget in due course.

In each election, TEMCO had specific objectives. However, such objectives at some point considerably changed. In the 1995 and 2000 elections, TEMCO had almost the same objectives: to monitor and observe the whole electoral process; to observe and assess the fairness of the administration of the elections in general, especially the impartiality of the legal infrastructure (electoral laws, regulations, and directives), the electoral machinery, and the state instruments of law and order; to assess the fairness in the allocation and use of relevant state resources by political parties, especially state subsidy and public media; to assess the extent to which the election results could be said to have been free and fair; and to recommend improvements in electoral efficacy and fairness for future politics of the country (TEMCO 1997: 1; 2001: ix). One objective was added in the 2000 elections, concerned with provision of training to various groups that were involved in election monitoring. These objectives are relevant to election observation, since they address fundamental issues of impartiality of the legal and institutional framework that guides the electoral processes, as well as the administration of such rules to elections. Besides, they are clear and directly linked to election observation.

However, a significant departure of TEMCO's objectives in terms of content appeared in the 2005 elections. TEMCO had four objectives which are quoted herewith: to provide feedback to the major stakeholders (NEC, government, political parties, candidates, voters, the media, etc.) and other interested observers; to make suggestions for significant improvements in the electoral system and process; to provide indications for the extent of democratic consolidation achieved; and to be able to say whether, all things considered, the elections were free and fair or not (TEMCO 2006: xi, 161). It was at this juncture that TEMCO's objectives became flawed. If read critically, they are vague and do not suggest anything worth election observation. As can be seen, even the words ‘to observe’, ‘to monitor’ and ‘to assess’, which were key features of TEMCO's objectives in the 1995 and 2000 elections, were omitted. Furthermore, to state the extent of democratic consolidation based on elections is reductionism of analysis since it means equating democracy to elections. This is what scholars view as the fallacy of electoralism (Brownlee Reference Brownlee2007). Democratic consolidation transcends elections. It includes, but is not limited to, the following preconditions: free and fair elections, rule of law (constitutionalism), respect for human rights, civilian control over military/armed forces, independent legislatures and courts, viable political parties, viable and autonomous civil society organisations, a free press and peaceful settlement of conflicts resulting from competition (Bakari & Mushi Reference Bakari and Mushi2005: 37). The next section examines TEMCO's reports for the 1995, 2000 and 2005 general elections. It shows how TEMCO observed these elections, with a particular focus on four major issues: legal framework, electoral management body, state institutions and personnel, and certification.

LEGAL FRAMEWORK

In 1992, just three years before the first multiparty elections took place, TEMCO (1997: 18) noted that the ruling party had unilaterally defined all the rules of the game in its favour, observing that ‘the government sent a bill to the parliament (which was “100%” CCM) when it wished and the opposition parties were kept guessing, not knowing what would happen next. This strategy enabled the ruling party to monopolize the process of defining new rules of the game, including changes in election laws.’ Baregu (Reference Baregu2003: 23) aptly calls this transition ‘a CCM controlled transition’. It was the view of TEMCO that the new pluralist system inherited ‘a wrong constitution suited to a monolithic system’, and that ‘the writing of an entirely new constitution rather than patching up the one-party constitution as the CCM government has been doing’ was needed to reinvigorate democracy (TEMCO 1997: 12). The simple interpretation is that the inherited monolithic constitution has a significant impact on the entire electoral system.

Despite having the same objectives as in 1995, TEMCO shied away from mentioning anything with regard to the new constitution in the 2000 elections. However, it pointed out that the project of delinking the ruling party from the state which started in 1992 was yet to be completed (TEMCO 2001: 195–6). For the 2005 elections, there was a complete shift in TEMCO's objectives as previously discussed. The reports for the 1995 and 2000 elections each had a chapter that dealt with the legal framework at length. In 2005, TEMCO did not include a specific chapter on legal framework. The questions that went unanswered include: did Tanzanians write a new constitution founded on the pillars of multiparty democracy? To what extent is the ruling party delinked from the state? The silence of the 2000 and 2005 election reports on the legal framework would imply that these factors had already been resolved. On the first question, the response is definitely ‘no’. Opposition parties are still struggling for a completely new constitution.Footnote ² On the second question, the reforms effected since 1992 are by and large cosmetic, and work to favour the ruling party. In 2003, the United States Agency for International Development (USAID) in Tanzania commissioned a study (ARD 2003: 2),Footnote ³ which stated in its report:

Other scholars have pointed out some of the problematic amendments that were ushered in before and after the 2000 elections. They included a simple majority electoral formula, empowering the president to appoint up to ten members of parliament, introduction of 5 million shillings deposit for petitioning an electoral case,Footnote ⁴ the enactment of a law that legalised ‘treating’Footnote ⁵ which is corruption, the law that prevents independent candidates,Footnote ⁶ and the Election Expenses Act no. 6 of 2010 (Biddle et al. Reference Biddle, Cassidy and Mukandala2002; Liviga Reference Liviga2009; Makulilo Reference Makulilo2008; Mmuya Reference Mmuya2003; Raphael Reference Raphael2010; Whitehead Reference Whitehead2009). It is against this reality that Mukandala et al. (Reference Mukandala, Mushi, Barkan and Njema2005) posit that although CCM is formally detached from the governmental structures, the old ties between the party and the administration still exist, both formally and informally. Mmuya (Reference Mmuya2003: 63) submits that ‘it is obvious that … constitutional and legal reforms have only consolidated the dominant position of the party in power. In this regard, democratic consolidation becomes elusive.’ All these concerns about constitutional and legal reforms leave a lot to be desired and would therefore deserve a critical assessment in the context of free and fair elections. This is one of the major omissions of TEMCO's reports, particularly those for the 2000 and 2005 elections.

ELECTION MANAGEMENT BODY

Effective management of electoral systems requires institutions that are inclusive, sustainable, just and independent. It includes, among other things, electoral management bodies (EMBs) that have the legitimacy to enforce rules and assure fairness with the cooperation of political parties and citizens (Elklit & Reynolds Reference Elklit and Reynolds2005; UNDP 2000). As main referees of elections, EMBs should be impartial and autonomous from any interested parties in a given election. The African Conference on Elections, Democracy and Governance (ACEDG) set the benchmarks for assessing independence of EMBs in Africa (ACEDG 2003). These include: that independence of EMBs should be secured constitutionally, and their budgets should be voted directly by the legislative bodies; that selection and appointment procedures for commissioners should be inclusive; that EMBs should conduct themselves impartially; that EMBs should independently appoint their secretariat; and that EMBs should be subjected to public scrutiny.

In Tanzania, general elections are solely managed by the NEC. The commission is vested with constitutional powers to register voters, to nominate candidates, to supervise campaigns, to manage voting and counting, and finally to declare results. The commissioners are unilaterally appointed by the president of the United Republic, and can be dismissed by the same at his or her discretion. The president has always been the chair of the ruling party, and on some occasions a presidential candidate during elections. In its report for the 1995 general elections, TEMCO (1997: 137) posed a question: ‘Is NEC independent?’ It responded to this question by arguing that ‘ideally, Article 74(7) and 74(11) of the Union Constitution purports to accord an independent status to NEC at least at national level … Practically, however, the NEC does not pass the basic tests of an independent institution.’ TEMCO advanced four reasons to support its position. Firstly, the NEC commissioners are appointed by the incumbent president, who is also the chairman of the party in power; secondly, appointees to the NEC have no guaranteed security of tenure since the president can revoke their appointments at will; thirdly, neither the constitution nor the Election Act secures funds for use by the NEC; and fourthly, NEC has no staff of its own at the regional and constituency levels. It relies on local government staff, who in most cases are CCM cadres. TEMCO (1997: 193) concludes: ‘how could the National Electoral Commission (NEC) delink itself from CCM given its composition, manner of its appointment, reliance on CCM government discretionary funding, and even more compromising, reliance on borrowed government personnel, most of whom were believed to be (or to have been in the immediate past) CCM members?’

In its report for the 2000 elections, while questioning the independence of NEC, TEMCO slightly changed the tone of its wording. It ceased to be authoritative and left it open for readers to judge the independence of NEC. TEMCO (2001: 19) writes: ‘in assessing … independence we need to delineate factors which support … independence and those which prejudice it at the national, regional, district and constituency levels’. TEMCO cites the same articles from the constitution as in its report for the 1995 elections. While in 1995 TEMCO posited that ‘ideally’ these laws ‘purport’ to accord independence to the NEC, in its report for the 2000 elections it calls the same laws ‘factors supporting NEC's independence’, indicating that TEMCO then believed that these laws actually ‘support’ and ‘not purport’ the independence of NEC. Likewise, for TEMCO's position that ‘practically, however, the NEC does not pass the basic tests of an independent institution’ was substituted ‘factors prejudicing NEC's independence’ (TEMCO 2001: 20).

TEMCO finally washed its hands of the independence debate in the 2005 report. To the question, ‘Is NEC independent?’, TEMCO (2006: 108) starts by positing ‘as it was the case for the 2000 elections, the composition of NEC was guided by the same framework during the 2005 elections’. This is quite right. I can quickly add that this was also exactly the case for the 1995 elections. How did TEMCO then distance itself from its earlier position on NEC? It tactfully initiated a debate, stating: ‘the issue of the independence of NEC has been perennial debate. There are some of the players and actors who discredit NEC as a stooge of its appointing authority – the President. They argue that …’ (TEMCO 2006: 120–1). These arguments are the very same as are found in TEMCO's reports for the 1995 and 2000 general elections when the debate had not yet begun. The other side of the debate, claiming NEC to be independent, is not presented. So, on which side was TEMCO in the 1995, 2000 and 2005 elections?

In its report on the 1995 elections, TEMCO clearly argued that NEC was not independent. Implicitly in 2000 it held a similar position, since the same factors held as in 1995, and it recommended that NEC should be reconstituted before the 2005 elections, in order to win the confidence and trust of all political parties and voters. For the 2005 elections there is confusion. First, it noted as stated above that the composition of NEC was guided by the same framework as before. This alone is sufficient to place TEMCO among those actors who argued that NEC was not independent. Additionally, the factors advanced by TEMCO in 1995 and 2000 were the same in 2005. Third, the failure to present the other side of the debate is a serious omission. Nonetheless, TEMCO (2006: 165) held that ‘all constituency reports by TEMCO observers affirmed that NEC election officials performed their roles independently and impartially. They consulted all participating parties in preparing campaign timetables and listened to their complaints and suggestions.’ Here comes a serious methodological problem. How are independence and impartiality measured? There are various ways. However, in 1995 and 2000 TEMCO advanced the following criteria: appointment procedures of NEC's commissioners, tenure of office by commissioners, independent budget deliberated by the parliament, and that TEMCO should have its own staff nationwide. In contrast, the independence of NEC in 2005 elections derived from ‘compiling timetables’ and ‘solving complaints’. Based on these criteria, I would argue that TEMCO stood on the side that viewed NEC as an independent organ. Applying TEMCO's criteria for assessing the independence of NEC as was the case with the 1995 and 2000 elections, Makulilo (Reference Makulilo2009) finds that NEC is yet to be independent since such factors have remained unchanged.

STATE INSTITUTIONS AND PERSONNEL

In a democratic state, institutions and personnel should act impartially towards all contending parties during elections. The reason for this is that they are funded by taxpayers' money, and so should ensure a level playing field for all actors. TEMCO has consistently observed the police force, regional and district commissioners, division secretaries, ward executive officers and village executive officers. These are directly involved in the management of elections.

CERTIFICATION

In the case of TEMCO, certificates are issued on the basis of ‘freeness and fairness’ of an election. On the one hand, elections are said to be ‘free’ if no person(s) are prevented from participating or forced to participate in elections. On the other hand, ‘fairness’ concerns the treatment of candidates and their parties with regard to competition. It is about the impartiality of the laws, personnel and institutions that manage elections. From that backdrop, TEMCO presents one of six certificates, depending on the nature of elections: ‘clean, free and fair’ means that nothing was wrong with the elections or that the observed shortcomings were insignificant and could not alter the results; ‘qualified, free and fair’ means that election was generally free and fair, but significant irregularities affected some or all participants negatively; ‘free but not fair’ means that there was pervasive non-compliance with electoral rules, often taking the form of government and/or election officials' actions that militate against the fair participation of some parties, candidates or voters; ‘unfree and unfair’ suggests that there was significant bias in key aspects of the electoral system, such as laws and rules, and management of the elections and/or pervasive intimidation; ‘totally mismanaged elections’ indicates a complete failure to properly manage the elections due to incompetence and poor planning, making it difficult to even decide on an assessment; and ‘aborted elections’ indicate that there is deliberate and severe widespread violation of electoral rules and procedures, causing a substantial number of voters, candidates and/or parties to withdraw from the race, making it impossible for elections to come to a legal finality.

TEMCO certified the 1995 and 2000 general elections as ‘free but not fair’. The basis for this verdict in the 1995 elections was that ‘they were free because no person or category of persons was prevented from participating or forced to participate in the elections under the existing rules and circumstances. All were free – at least legally speaking – to participate or not to participate. The elections were unfair because of many problems, irregularities and managerial inadequacies pointed out’ (TEMCO 1997: 252). In the 2000 elections, TEMCO's ground was that ‘the unfairness comes from the big state bias in favour of the ruling party, the heavy-handedness of the police in campaign rallies of opposition parties, and the incomplete separation of state resources from the ruling party’ (TEMCO 2001: 198–9). It appears that while Tanzania's electoral system is open to participation by voters, it constrains candidates and their respective parties. Independent candidates are strictly not allowed by law. The main source of unfairness in elections is the incomplete delinking of the ruling party from the state, creating a playground tilted in its favour. As the ruling party identifies opportunities for the opposition parties, new laws are enacted to make sure that they suffocate the political space in its favour. This is normally effected through the CCM-dominated national assembly. Arguably, there is a correlation between the contents of the 1995 and 2000 reports and their respective certificates. However, one serious omission on these certificates is that TEMCO does not state the extent to which the identified shortcomings would have altered the results or otherwise.

The big problem, however, lies with the TEMCO's certificate for the 2005 elections, that is, ‘clean, free and fair’. The ground for this certificate was, in TEMCO's words: ‘taking into account the balance between the positive and negative things that occurred in the management of all stages of the election, TEMCO awards a clean certificate of free and fair elections. The observed shortcomings were insignificant and could not alter the results.’ This was despite the fact that all the factors that militated against fairness in the 1995 and 2000 elections were still pervasive in 2005. It was the same constitution of the single-party era that TEMCO was against in 1995; the same NEC that was not independent; and the same partisan state media, police and personnel. If these factors had an impact in rendering the 1995 and 2000 elections unfair, why not for 2005 too? For ease of reference, I recap TEMCO's statements for the 2005 elections. In its report TEMCO vehemently stated that the RCs, DCs, DS, WEOs and VEOs are public servants appointed by the president (who at the same time is the chairman of the ruling party and in some cases a presidential candidate during general elections) on the basis of their loyalty to the ruling party, and therefore cannot be impartial during elections. TEMCO notes that ‘they mobilize voters on behalf of the ruling party and in many different ways facilitate campaigns of candidates of the ruling party using state resources (vehicles, security personnel, etc.). This area was controversial in 1995 and 2000, and remained unchanged in 2005. This is a systemic problem, and it will be difficult to have a level political playfield in Tanzania without finding a way of making these powerful people in the regions, districts and divisions act impartially’ (TEMCO 2006: 168). Can one argue that this shortcoming was insignificant and could not alter the results? If TEMCO believes that this problem alone severely militates against ‘a level political playfield in Tanzania’, what else is needed to certify the 2005 elections as unfair? In the case of The Attorney-General v. Aman Walid Kabourou, [1996] T.L.R 156 following the Kigoma by-election of 1994, the Court of Appeal of Tanzania (Supreme Court) ruled that the use of government property and government employees by CCM during the elections is prohibited in accordance with the principle of fairness. Such acts influenced the results in favour of the ruling party and its candidate, and provided grounds for the nullification of the by-election result. Since the state personnel mentioned by TEMCO are scattered all over the country from the national to grass-roots levels, their impact is severe, irreparable and systemic. TEMCO thus failed to certify the 2005 elections in accordance with the data available to it. Probably, this could partly be explained by the drastic shift in its objectives in 2005, which indicated reluctance to ‘observe’, ‘monitor’ or ‘assess’ the election.

EXPLAINING TEMCO'S 2005 SHIFTS

As shown in the preceding sections, TEMCO certified the 1995 and 2000 elections as free but not fair. Paradoxically, despite the same authoritarian realities it found in those elections, TEMCO gave a clean, free and fair verdict on the 2005 elections. Yet, unlike in its reports for the 1995 and 2000 elections, TEMCO's objectives changed significantly in the 2005 report. The following four factors may help to explain TEMCO's shifts in its objectives and its verdict for the 2005 election report.

The first explanation is anchored on the issue of professionalism, which undermines the organization's sense of impartiality. To substantiate this point, let me consider the recent disputed opinion polls of 2010 conducted by the Research for Education and Democracy in Tanzania (REDET). I hasten to emphasise here that REDET is TEMCO and vice versa: REDET is the lead agency of TEMCO, and TEMCO's core management team is constituted by the same members as REDET's team (TEMCO 2010a: 1; 2010b: 4). In its March 2010 opinion poll for the October 2010 elections, REDET found that CCM and its candidates were in the lead by a wide margin. This was also the case with the September 2010 poll. I wish to point out one methodological deficit of these polls. In its March poll, REDET's sampling method considered population ratio between urban and rural residents. Though rural residents are the majority compared with urban ones, REDET did not give estimated figures for these people. However, it proceeded to sample at the ratio of ‘60%’ rural residents and ‘40%’ urban residents (REDET 2010a: 4). This had been the practice by REDET in its previous polls. In contrast, in its September poll of 2010, REDET, while considering the same rural–urban population ratios, sampled ‘70%’ rural residents and ‘30%’ urban residents (REDET 2010b: 4). It is a matter of fact that rural regions are the strongholds for the ruling party. Opposition parties are localised in some urban areas (Chaligha Reference Chaligha2005; Kaya Reference Kaya2004; Kiragu & Mukandala Reference Kiragu and Mukandala2005; Mmuya Reference Mmuya1998). The September Poll thus inflated results in favour of CCM. It is surprising for REDET to change the urban–rural population ratio so drastically in a period of just six months. This would suggest that the urban–rural migration in Tanzania is extremely high, which is not the case. A further weakness is that REDET, unlike in its past polls (e.g. Polls no. 11 and no. 12 of 2006) where it analysed results by political party affiliation, ignored this important parameter in both the March and September 2010 polls, making it difficult to ascertain the results on a partisan basis (REDET 2006a: 12; 2006b: 9).

The 2010 REDET polls were highly disputed by academics, politicians, civil society, and the general public. For instance, Professor Peter Maina of the School of Law, University of Dar es Salaam said that the polls were questionable and unrealistic (This Day 12.10.2010). Similarly, opposition parties dismissed the polls claiming that they were strategically designed to favour CCM. The Tanzania Labour Party (TLP) deputy secretary-general, Mr Hamad Tao, said his party did not agree with the way the institution conducted its surveys, and described its report as ‘debatable’. He furthered that while it could be true that Mr Kikwete was the front-runner ahead of the elections, the margin of his lead in REDET's report was ‘outrageous’. Mr Tao contended: ‘my worry is that this misleading survey can change voters’ perception … people might start thinking that opposition parties are too weak to win any seat, and thus decide to vote for CCM' (Citizen 9.10.2010). Yet TEMCO found itself in professional crisis when it observed the 31 July 2010 referendum in Zanzibar. One unfortunate statement it made was about the under-age registration of voters. In its newsletters, TEMCO stated that the problem of under-age registration was acute in several registration centres throughout the registration process (TEMCO 2009: 2; 2010c: 5). Surprisingly, in its August 2010 interim statement on the referendum, TEMCO stated that this problem was not observed at all; this was repeated in its Special Issue Newsletter on the Referendum (TEMCO 2010d: 2; 2010e: 6). As usual, it certified the referendum as ‘transparent, free and fair’, a certificate which is not even found in its list: TEMCO's highest certificate is ‘clean, free and fair’. It is worth noting that the quality of TEMCO/REDET's reports and books has been questioned (Van Donge 2000: 542–4). In the USAID study (ARD 2010),Footnote ⁷ it is well stated that in Tanzania elections can relatively be free but not fair.

The second explanation is serious allegations of strategic relations between TEMCO/REDET and individual leaders of CCM. Following the September 2010 poll, Professor Mwesiga Baregu, one of the founders of REDET, urged REDET's leadership to demarcate between professionalism and partisanship in order to safeguard the reputation of the University of Dar es Salaam. He added that some of REDET's stakeholders were friends of the CCM presidential candidate, Mr Jakaya Kikwete, and some of them have been transferred to State House (Habari Leo 10.10.2010). One of REDET's members and former co-chairman was appointed by President Jakaya Kikwete as his political advisor. This may raise doubt on the impartiality of REDET/TEMCO. In line with this view, the former UDSM Vice-Chancellor, Professor Mathew Luhanga, remarked: ‘Kwa uzoefu nilioupata hapa Chuo kikuu cha Dar es Salaam, nimeona kuwa wasomi wana hofu kupoteza maslahi yao … badala ya kusema kweli wanaamua kujikomba kwa viongozi’Footnote ⁸ (Mwananchi 15.10.2010). This statement from the highest level of the former university management speaks for itself. The Tanzania Media Women's Association (TAMWA) denounced REDET's September 2010 poll as cooked and in favour of CCM (Tanzania Daima 13.10.2010). Since there has been an increasing outcry on opinion polls, not only by REDET/TEMCO but also SYNOVATE and others, further study is needed to understand the relationships between research organisations and the ruling party in Tanzania, and how such relationships shape the behaviour and impartiality of these organisations.

The third factor is fear. It is the habit of the ruling party to react negatively to opinions, facts or actions that seem contrary to its beliefs. For example, in 1995 Baraza la Wanawake wa Tanzania Footnote ⁹ (BAWATA), which provided voters' education to women, was deregistered on allegations by the ruling party and its government that it was campaigning against CCM. BAWATA filed a case in the High Court, which ruled in favour of BAWATA.Footnote ¹⁰ Similarly, HAKIELIMU, a civil society organisation which is critical of government educational policies, has persistently found itself attacked by the ruling party and its government (Tanzania Daima 4.2.2007). In relation to election observers, Kamata (Reference Kamata2002: 45–6) reminds us that politicians and parties in power view election monitors and observers with suspicion. He contends that while this was not explicit in the 1995 general elections in Tanzania, it was very obvious in the October 2000 general elections. President Benjamin Mkapa said in a speech in Harare that election observers are disappointed when election results return incumbents in power. Kamata proceeds: ‘the President remarked that elections are free and fair only when the incumbent is defeated and the opposition takes power’. As one of the observers, TEMCO cannot shield itself from such threats.

The fourth explanation concerns the budget. TEMCO's budget is ‘100%’ funded by donors. Usually, donors' money is attached with conditionalities (Plattner Reference Plattner, Zoltan and Moser2009: 1–12). In its report on the 1995 elections, TEMCO noted that such conditionalities were not imposed. However, the organisation was required to account for how it spent the money according to the budget proposal it submitted to the donors. In contrast, TEMCO was silent as to whether its budgets for the 2000 and 2005 general elections had such conditionalities. This leaves a lot in a black box. While donors are in most cases highly sensitive to the disclosure of their contracts, it is not known why TEMCO was able to do that in 1995. Admittedly, there is no adequate information to hold a strong position on this factor. I suggest this as an issue of interest for future researches in order to uncover the motives of actors who sponsor election observers, so as to explain whether such motives go beyond the promotion of democracy. During the 1991 elections in Zambia, the former president Kenneth Kaunda accused election observers of facilitating his downfall (2009 int.), whereas former president Frederick Chiluba, winner of the 1991 elections, applauded that observers were instrumental in containing vote rigging by Kaunda's regime (2009 int.).

* * *

Five remarks are made on TEMCO's experience of observing elections in Tanzania. First, TEMCO overemphasises the administration of elections on the voting day and leaves aside the legal and institutional framework within which such elections take place. Second, while the reports are well detailed with election irregularities, the certification is at times repugnant to the content. Third, TEMCO has not been able to concretely establish the extent to which election irregularities impact on the electoral outcome: that is ‘whether the result would have been substantially different if this maladministration had not been the case’ (Van Donge 2000: 544). The conclusion is merely drawn that by considering all that has been said the election was ‘free but not fair’ or ‘clean, free and fair’. Fourth, given similar facts across elections, TEMCO has failed to issue consistent verdicts. Fifth, for reliability and boldness in the issuance of its certificates, TEMCO should use previous studies, reports and case laws to aid its decisions.