9. The first multi-party elections

THE PARLIAMENTARY elections of April 27, 1993, although delayed five months beyond their appointed date, were nevertheless a milestone in Yemen’s political development: the first popular expression of the new pluralism. Beyond that, the fact that Yemen became the first state in the peninsula to hold competitive elections on the basis of universal suffrage [1] was, for all the flaws and problems, no mean achievement. More subtly, the elections re-affirmed the constitutional axiom that “the people of Yemen are the possessors and source of power” [2] and helped to implant the idea that ordinary citizens can hold their leaders accountable to public opinion, even if the leaders themselves do not always heed it. This is not to suggest that everyone in Yemen valued the exercise or even approved of it. Attitudes among the public ranged from excitement to apathy, while expectations varied between those who hoped for dramatic improvements and those who saw the process as little more than window-dressing.

In reality the elections were not an occasion for choosing between alternative governments. Even though all the main parties had apparently accepted the principle of peaceful alternation of power, there was never much prospect of them being required to apply it. This was an election to choose a new parliament, not a party of government: it had been made clear beforehand that in the interests of national unity there would be another coalition, though the balance of parties in the new parliament would obviously affect the composition of the future government. It was also expected that the government would draw its members from the GPC and YSP, though perhaps not exclusively. The contest, therefore, was to establish the relative strengths of the parties, and in broad terms the outcome was not difficult to predict. Although some 21 parties were contesting the election, only three – the GPC, YSP and Islah – had the kind of organisation and support that would enable them to win large numbers of seats.

In some respects the competitive element was less important than the fact that the elections took place at all, since the mere holding of them fulfilled a number of needs. Constitutionally, together with the earlier referendum, they were an opportunity for popular endorsement of the new political system and provided the main parties with a legitimacy that had hitherto been lacking. As with many emerging democracies, attention focused on the electoral process as much as the outcome: the perception of a reasonably fair contest was important, not just for the benefit of relations between government and governed, but in terms of Yemen’s external relations and its worthiness to receive foreign aid and investment. For this reason, attendance of international observers was an essential part of the process.

An estimated 6,283,000 Yemenis were entitled to vote in the election [3], of whom 2,688,000 actually registered to do so (see table). While this figure would be considered low in a more established democracy, many Yemenis saw it as a positive achievement. Certainly the total represented a huge increase on the 1,890,000 voters who registered in the 1991 constitutional referendum. Nevertheless the electoral system contained an in-built disincentive to both registration and voting, which some critics thought was deliberate, but which could also be explained by administrative and geographical factors. In the case of the largest unregistered group – women – there were additional cultural factors but again, the fact that around 16%-20% of the eligible women did register was viewed as an achievement rather than a failure. A report by the Washington-based National Democratic Institute [4] observed:

Many women maintained that the greatest achievement of the election process was the formal incorporation of women into the election law, even while practical problems remained to inhibit full participation. The election system provided women with a legal guarantee of equal status that they rarely enjoy in other fields of endeavour. Before the 1993 elections, women in North Yemen were permitted to vote but not to be candidates. (Though one woman did register herself as a candidate for the General Assembly in 1988 as a form of protest.) In South Yemen, women were allowed to run for seats in the legislature, as well as to vote, although voter preference and meaningful choice played no role in these Soviet-style one-party elections. [5]

The mechanics of registration and polling placed heavy demands on an administrative system which, at the local level, was often minimal. It was therefore necessary to strike a balance between meeting the convenience of voters and avoiding the creation of a structure so large than proper supervision became impossible. The eventual arrangement was to provide five polling stations for each of the 301 constituencies, staffed and managed by a total of 13,000 people. While this was adequate in the towns, in rural areas it was not; according to one estimate, voters had to travel between four and five kilometres on average to their polling station, and in remote areas considerably more [6}. They also had to make this journey twice: once to register and once to vote. Proof of registration had to be presented on polling day.

The registration process involved queuing to fill in a form and have a photograph taken. In some areas shaykhs attended to verify that people were from the right place. Voters were allowed to register in one of three places: the constituency where they were born, lived or worked. This gave the two government parties a theoretical advantage in that nothing prevented them from moving military units into marginal constituencies in the hope that the soldiers’ votes would influence the outcome. In Ma’rib province an extraordinarily large number of men – 88% of those eligible – registered to vote. This was 18 percentage points above the national average for male voters, and eight points above the next highest province [7]. The suggestion was that military votes were being used to manipulate the outcome, though there were only three parliamentary seats at stake in Ma’rib (and, in the event, each of the main parties won one). Despite the suspicions raised by opposition parties and independent observers during the registration period, on election day there was no evidence of large-scale abuse of the military vote, apart from some minor irregularities [8]. One electoral officer was dismissed for allowing soldiers to vote who were not registered in his constituency [9] and at a polling station in Sana’a six soldiers found different names against their electoral registration numbers – all of which were marked as having already voted. The six were allowed to vote because they had no ink on their fingers and had valid registration papers. There were also claims (not fully substantiated) that Islah, and the two government parties to a lesser extent, had bought thousands of identity cards from civil servants which were then used to register voters who were either from another district or under the legal voting age [10].

The level of female registration varied widely throughout the country, ranging from 2% of eligible women in al-Jawf province to 37% in Abyan (see table). Within provinces there were also huge variations; in parts of Ta’izz province, for example, female registration reached 30% but only 5% in other districts [11]. In general the highest levels were found in the southern provinces (attributable to the relative emancipation of women there) and the lowest in rural parts of the north. However, registration was quite high in northern cities. Surprisingly perhaps, in view of its traditionalist attitudes, Islah was particularly active in encouraging women to register, though it opposed the idea of women as candidates. The election law required “all appropriate measures to encourage women to exercise their voting rights” [12] and, at the instigation of Islah, this was strengthened to require separate registration committees and polling sites for women (on the assumption that women would be reluctant to stand in the same queues as men for registration and voting, or to have their identity photographs stored alongside those of men) [13]. Unfortunately the effect was to remove one obstacle and replace it with another: delays in establishing the separate registration centres for women meant that many of them did not open until half-way through the four-week registration period, while some were open only for the last three days [14].

Another difficulty relating to women concerned the photograph needed for the registration card – a sensitive issue in strict Muslim societies. The election law specified one photograph, and instant cameras were provided for the purpose at registration centres. But because of the shortage of time, it was not possible to issue the permanent cards intended for use in all future elections. The body responsible for organising the elections, the Supreme Elections Committee (SEC), therefore decided that two photographs should be taken: one for a temporary card to be used on April 27 and the other for issuing permanent cards later. This gave rise to complaints that the second photograph was illegal and aroused suspicions about its possible use by the authorities. Although the SEC pointed out that men do not normally mind their wives being photographed for passports, it appears that some parties, having mobilised their own women supporters to register, spread suspicion in the hope of discouraging other women from registering [15].

Several of the factors that discouraged voter registration in general applied to women disproportionately. The inaccessibility of some polling and registration centres (a frequent complaint among both sexes) tended to affect women more, given the cultural restraints imposed upon women’s mobility outside the home [16]. The higher level of illiteracy among women (85% compared with 46% among men) was also a strong disincentive [17]. However, this was probably less significant than the traditionalists’ reluctance to treat voting as an appropriate activity for women – especially without a husband’s consent. The chairman of an SEC supervisory committee in Ibb explained: “There is a perception that women are involved in domestic work and do not have time to think about political activities. The husband’s responsibility is to participate in the political process for both of them.” Another government official said the most common reason given by women for not registering was: “My husband is away. I cannot register without his permission. He would divorce me.” [18]

In contrast to the discouragements faced by potential voters, standing as a candidate was extremely easy. There was no deposit to be paid and a plan requiring nominees to obtain 300 signatures from voters registered in their constituency was eventually dropped. This meant that virtually any adult who was literate, of good moral standing, religiously observant and without a criminal record could put his or her name forward. As a result, 4,602 did so – an average of more than 15 candidates for each constituency – and of these, 3,246 were independents [19]. Withdrawals before polling day brought the final total down to 3,181, of whom 1,956 were independents [20].

The reasons for this large number of independents were many and varied. Firstly, there was the novelty of the country’s first free elections and while many candidates were sincere in their aims, there were certainly others with an inflated view of their own importance – including a number of disgruntled party members who had sought and failed to win official nomination. Others stood as independents because they were civil servants or members of the armed forces who could not be formally affiliated to a party. There were also tactical candidates who stood with the intention of withdrawing later (and being rewarded financially for doing so). The large parties’ attitude towards independents varied according to the local situation: in some areas they sought to eliminate them and in others encouraged them in the hope of splitting the opposition. Another factor encouraging independent candidates was that the electoral arithmetic made victory look deceptively easy: on average each constituency had about 20,000 eligible voters, of whom no more than 10,000 were likely to vote. This meant that in a straight contest the winner would need little more than 5,000 votes – and considerably fewer where there were multiple candidates [21]. In the event even this proved an over-estimate, since the average number of votes cast in each constituency was only about 7,500, but the optimism of many independents was flawed because it failed to take account of the large parties’ superior organisation.

There were 50 women candidates [22], though they faced additional difficulties. Almost all were independents, because the main parties were reluctant to nominate women (the GPC nominated two, the YSP four and Islah none) [23]. The GPC reportedly asked several of its prominent female members to withdraw their candidacies on the grounds that “the competition would be too much trouble for them” [24]. Similarly, the YSP seemed prepared to nominate women only in areas where they had a good chance of success, so as not to subject them to “the humiliation of failure” [25]. This was a delicate way of saying that female candidates could be an electoral liability – a not altogether unreasonable conclusion, given Islah’s declaration that to vote for a woman would be un-Islamic [26]. The account of an election meeting held by a woman doctor, contesting a seat in Sana’a as an independent, shows the effort female candidates were obliged to expend on justifying their candidacy rather than discussing their policies:

People slowly move towards the chairs under the tent, and the rows fill up as the speech begins … Invoking popular and religious elements, she traces the origins of Yemeni democracy back to the pre-Islamic period, and emphasises the role women have played in Yemeni history, from Bilqis to Queen Arwa … She explains that religious leaders have assured her that men may vote for a woman, and that a woman may take a seat in parliament. [27]

There had been 10 women in the transitional parliament created by unification – all elected by the south under its non-competitive single-party system. In 1993, competitive elections reduced the number to two (again, both from the south), though Yemen was not unusual in this: it is a common phenomenon when countries move from one-party to multi-party systems [28].

The selection of candidates to represent the main parties revealed a variety of criteria. In the case of the GPC, political views and even party affiliation were often secondary to strong links with the local community; lacking the party machine of the Socialists, the GPC was obliged to use traditional social networks – especially outside the cities – as a means for mobilising support. Typically, the candidates included tribal leaders, businessmen and senior officials. In the south, supporters of Ali Nasser Muhammad, the PDRY president ousted in 1986, provided a source of candidates for the GPC. The YSP, with its tradition of grass-roots activism and more efficient organisation, was less concerned with fielding candidates who had local roots, hence the large number of central committee members and government ministers on its list. In the north, its electoral vehicle was the old National Democratic Front, which had since been absorbed into the YSP. The Islah candidates fell mainly into two groups: the notables (rural shaykhs linked to Shaykh Abdullah al-Ahmar), and university-educated Islamists in the cities. Al-Haqq fielded 67 candidates from the most distinguished sayyid families, while the League of the Sons of Yemen was represented by prestigious families from Shabwa and Hadramawt. In all, the GPC fielded 279 candidates, the YSP 217, Islah 196 and the Ba’ath 160. Apart from the three main parties, the League and the Ba’ath were the only ones that contested the election nationwide, with candidates in almost all provinces [29]. 

The elimination of well over 1,000 candidates before election day indicates the extent of pre-election bargaining. In the north, particularly, there were agreements between the GPC and the Islah to withdraw a candidate from one party in favour of another who was better placed. Similar agreements, though probably smaller in number, were arranged between the GPC and YSP [30]. Although this led to complaints that the large parties had carved up the election among themselves, such agreements were less significant than many Yemenis imagined. The GPC refrained from contesting 22 of the 301 seats. If one assumes that all 22 were the result of inter-party agreements and that Islah or the YSP had made a reciprocal number of withdrawals elsewhere, the maximum number of seats affected would be 44 – just under 15% of the total. 

There were, however, other ways in which the smaller parties and independents were squeezed. Probably the main one was the lack of co-ordination among the minority candidates themselves: where they failed to agree on withdrawals the opposition vote was split, with the result that government parties won. “Yemen,” as one observer put it, “is a country full of prima donnas” [31]. At the same time, independents who were expected to perform well faced pressure to withdraw, sometimes accompanied by threats or inducements. Others were invited to join one of the big parties – especially the GPC which was less concerned about ideology than the others. At least one independent received a personal phone call from Abd al-Aziz al-Ghani (the most senior figure in the GPC after the president) and an invitation to discussions in Sana’a [32]. Several independents who sympathised with the YSP claimed the party had broken its promises not to stand against them [33]. In one of the Ta’izz constituencies, the Socialists and Nasserists failed to agree on who should step down, with the result that Islah won. In the southern city of Mukalla, on the other hand, the Islah candidate was deprived of almost certain victory by the withdrawal of the YSP candidate in favour of the Ali Nasser Muhammad faction of the GPC [34].

Judged by the election manifestos, there was a substantial degree of unanimity on policy, not only among the large parties but many of the small ones, too. The differences tended to be those of emphasis and personalities. The GPC presented itself as democratic and reformist while upholding Islamic tradition. The YSP, having abandoned most of its Marxist ideas, adopted a social-democratic stance, championing democracy, modernisation and order. Islah focused on “restoring” Islam to a central position in Yemeni society but presented its case with a subtlety that appeared progressive rather than reactionary or revolutionary. In terms of actual policies, particularly in the fields of economics, foreign policy, health, education, decentralisation of government and freedom of speech, there was a remarkable measure of agreement (see comparison of policies). That was not entirely surprising since the GPC and YSP were supposedly co-ordinating their campaigns and preparing to merge, but it was also partly a result of the big parties’ attempts to claim the middle ground by stealing each other’s political clothing. Thus the YSP, conscious of accusations that it was secular or atheistic, proposed the creation of an Islamic university, while Islah made a point of including proposals on women’s rights – an area where it was often considered weak or even hostile. In other areas Islah used religious teaching to justify remarkably liberal policies, declaring, for example, that freedom of expression was “part of the freedoms guaranteed by sharia”. 

There were, however, a few important areas of difference. One was in the area of security, corruption and judicial reform, where the YSP’s proposals were couched in especially strong terms that could be read as an implicit criticism of the GPC. The other concerned Islah’s attitude to the constitution, and democracy in particular. Although committed to the republican system, Islah campaigned under the slogan: “The Qur’an and the Sunna supersede the constitution and the law.” This was potentially more flexible than the slogan of Egyptian Islamists: “The Qur’an is our constitution”, but it remained problematic. While the manifesto reassuringly declared support for “peaceful alternation of power through the ballot box”, there was no mention of the multi-party system and Islah generally preferred the word shurawiyya (“consultationalism”) to dimukratiyya (democracy). Despite this, the party sought to strengthen the role of parliament so as to make it “an effective instrument”. Support for parliament was consistent with either shurawiyya or dimukratiyya; though cynics might argue that Islah saw parliament as the most suitable vehicle for promoting its ideas, increasing its influence and, in the longer term, achieving power [35].

The most serious flaw in the voting process – and one which significantly compromised secrecy – was that ballot papers did not carry the candidates’ names. Because the papers were to be printed abroad as a protection against forgery, the time between finalising the list of candidates and polling day was not sufficient to include the names. The result was that instead of making a mark on the paper, voters had to write the name of their chosen candidate. This caused particular difficulties among illiterate voters, who were allowed to ask a friend, a member of the committee or another person to write the name for them [36]. To reduce the possibility of abuse, no individual was permitted to fill in more than 10 ballot papers on behalf of others, and assisted votes had to be checked to ensure that the name had been written in accordance with the voter’s instructions. The checking system appears to have worked quite efficiently, at least in Sana’a, according to an American observer [37]. However, another witness reported that a group of women had become impatient with waiting at one polling station and had forced themselves inside where they cast a collective vote, completing a series of ballot papers for their illiterate friends in the presence of passive officials. Elsewhere soldiers at the booths filled in ballot papers for others that no one bothered to check [38]. Another difficulty with the write-in voting system was that voters had to know the name, rather than party, of their preferred candidate. Also, prominent candidates were usually referred to by a “popular name” and voters were not necessarily aware of their correct name, which of course had to be used on the ballot paper. It was not disclosed how many votes were disallowed because of incorrect names. Together, these factors meant that voting was secret only for those who (a) were able to write and (b) knew the correct name of their preferred candidate.

There was some evidence of multiple voting, despite efforts to prevent it. A journalist in Sana’a reported seeing one man vote three times in the space of two hours, while at another polling station he saw a man vote twice [39]. After voting, electors were required to place their thumb in ink, to prevent them voting a second time. The effectiveness of this depended largely on the diligence of individual election officials in checking that the procedure was observed – often difficult if the polling station became crowded. Some voters managed to avoid putting their thumb in the ink and a few (including a cabinet minister) flatly refused to do so [40]. The inking system is widely used for voting in various parts of the world and according to one experienced observer of elections, tends to discourage casual abuse but not determined attempts at double-voting. The ink used in the Yemeni elections was specially imported from Germany and supposed to be indelible, although – as often happens – there were claims later that it could be washed off without much difficulty. The same observer reported several cases of ink being used in attempts to prevent people from voting by daubing it on hand-rails and doors where people might touch it accidentally on the way to the polling station [41].

Even if multiple voting was not entirely preventable, it was, at least in principle, detectable. One ballot box was issued for every 500 voters, which meant that the number of ballot papers in each box should not exceed that total. To prevent the premature opening or substitution of boxes, a three-person committee of candidates’ representatives was assigned to accompany each box at all times. 

The elections were monitored by more than 50 accredited international observers, plus foreign diplomats, journalists and interested individuals. The largest number came from the United States: the International Republican Institute (IRI), which has links with the Republican party, provided 19, the International Foundation for Electoral Systems (IFES) three, and the National Democratic Institute for International Affairs (NDI), linked with the Democratic party, sent three officially-registered observers together with a number of others. The British Electoral Reform Society (ERS) sent three observers. Other groups included representatives of French and Bulgarian organisations. The observers themselves came from a variety of backgrounds, including professional electoral officers and politicians, ranging from highly professional observers with experience of numerous elections worldwide, to enthusiastic amateurs. 

Invitations to observers have become routine practice in developing democracies and are aimed primarily at securing international acceptance of the validity of elections (an important criterion when seeking aid from the International Monetary Fund, the World Bank and most western donor countries). One of the reasons for having foreign observers is that by virtue of their nationality they are disinterested in the outcome of the election. Another is that a well-balanced international team will bring a wide experience of electoral practices in a variety of countries – both fair and unfair. They will therefore be able to distinguish between procedures which are generally sound and those which are more open to abuse. On the debit side, the number of observers is necessarily small and they are unable to observe the entire election in person; they may not understand the local language; and they may have little or no experience of the country they are observing. Some of the gaps may be filled by careful balancing of the observer team, background reading, briefings from diplomats, etc, but detailed local knowledge will usually be lacking. 

Normally, observer teams make a preliminary visit to see the preparations and interview the supervisory body – in Yemen’s case, the Supreme Elections Committee (SEC). At this early stage they begin to form an opinion as to whether or not the elections are suitably structured and intended to produce a fair result. In Yemen, although there were some complaints that preparations had started too late, first impressions were generally favourable. One British observer commented: “I was very impressed by the commission [the SEC]. There was a Marxist-Leninist doctor who looked after fair play. He was very meticulous and required the media to give fair representation to all sides.” By polling day much of the observers’ work has already been done: they have scrutinised the electoral machinery, identified weaknesses and possibly recommended improvements. If they have been satisfied by the electoral preparations, their task on polling day is mainly a matter of establishing how closely the established procedures are followed. 

Inevitably, the presence of international observers makes them participants to some extent. Although impartial in terms of the country’s party politics, most observers readily admit to taking a personal interest in the furtherance of democracy which at times places them in a dilemma: is their duty merely to observe and report, or does it also include offering guidance towards a successful outcome? The position adopted by the British Electoral Reform Society was to do both, but keep the two functions as separate as possible. Thus one of their team attended at the planning stage to give formal advice, and later, during the election, as an observer. In the view of one ERS representative intervention should not be totally avoided while acting as an observer, but should always be circumspect. If a problem occurs at a polling station, he said, “you have to realise that you are in a situation of transition to democracy, and give information such that the presiding officer will be directed. You don’t tell him what to do but you point out relevant bits of the instructions to help him make his decisions.” It is, however, a sensitive area where foreign observers must be tactful to avoid compromising their impartiality in the eyes of the local people. The American NDI, for example, undertook a study of the role of women in the election which, although a worthwhile project in itself, at times appeared like a naive and insensitive attempt to promote feminism in Yemen.

As far as monitoring by Yemenis was concerned, the election law provided for committees representing all the candidates, on the principle that their suspicion of each other would ensure a fair fight. This offered an reasonable level of security for the ballot boxes and the count, both in theory and, for the most part, in practice. However, conflict arose when a non-governmental Yemeni body, the National Committee for Free Elections (NCFE), sought the right to conduct its own monitoring on equal terms with the foreign observers. The NCFE had been established in January 1993 by the Yemeni Organisation for the Defence of Rights and Liberties with the aim of “guaranteeing free and clean elections”. Its chairman was Mustapha Nu’man, a human rights activist and former diplomat whose father had twice been prime minister of north Yemen and its committee was drawn mainly from the professions: lawyers, a civil engineer, businessmen, two journalists, a teacher and a doctor. In addition to analysing the electoral process as a whole, the committee aimed to recruit and train at least one volunteer to monitor each ballot box [42]. The American NDI embarked on a plan to train the Yemeni volunteers as observers and then to work alongside them. This arrangement, which was intended to combine the experience and guaranteed independence of the American observers with the local knowledge and greater geographical coverage of the Yemenis, had previously been tried by the NDI in other countries, with some success:

International delegates benefit from the knowledge and experience of the local observers ... Cultural subtleties and relevant history are more likely to be appreciated by the domestic monitors who can interpret events or circumstances for their international counterparts. In cases where observer deployments are co-ordinated, the international team is able to enlarge its geographical reach by deploying mixed teams to more areas. [43]

In Yemen the plan proved little short of disastrous and threatened to compromise the NDI’s independence when the organisation became embroiled in political intrigue. While there was nothing to prevent anyone from observing and reporting on the elections, in order to carry out detailed monitoring NCFE volunteers would require the right to enter polling stations and counting centres. The law made no provision for this, and the decision on whether or not to grant it appeared to be a matter of discretion for the Supreme Elections Committee. Despite this uncertainty, at the beginning of February, the Americans committed themselves to working with the NCFE and helping to train its volunteer observers [44]. That in turn meant the Americans would have to persuade the SEC to grant special status to NCFE observers. According to the Americans, the plan was formally approved by the Supreme Elections Committee in February 1993 but abruptly rescinded a month later by its chairman, under pressure from the GPC [45]. It is doubtful, however, whether the committee ever firmly approved the plan: published documents suggest that the February decision amounted to nothing more than tentative approval which was hedged with conditions [46]. Abd al-Karim al-Iryani, the Yemeni foreign minister (and a member of the GPC) explained the official reasons for rejecting the plan in a letter to an American congressman:

This organisation [the NCFE] and the other domestic organisations that have been formed with the same objective are absolutely free and encouraged to train local persons to act as election observers ... However our election law requires that election monitors be appointed by candidates, and does not permit self-appointment. Therefore, the only restriction on the activities of the NCFE and the other domestic organisations working on election monitoring is that the individual monitor be appointed by a candidate ... this is consistent with practices in numerous countries like ours in the early stages of democratic development ... The outcome of such an experiment would most likely be the creation of controversy and confusion. With over 40 parties participating in our elections, the SEC felt that adequate monitoring capability would be provided by the candidates’ representatives. [47]

Meanwhile, according to the NDI, there were various efforts to undermine the NCFE. Articles in GPC newspapers accused it of “well-known” but unspecified partisan connections. (Elements in the GPC, and to some extent Islah as well, apparently regarded the organisation as a YSP-sponsored project to denounce the election if the GPC won a large majority. However, the NDI later stated in its report that no evidence of a pro-YSP bias was ever produced and it continued to believe the organisation was truly independent.) [48] The GPC and Islah also attempted pack the NCFE’s advisory council with their own supporters. When that failed, a rival monitoring organisation, the Yemeni Committee for Free and Democratic Elections, came into being less than a month before the elections. The NDI had little doubt that this had been created at the instigation of the GPC with the express purpose of causing confusion [49]. The rival organisation then demanded help from the NDI at the same level as that offered to the NCFE and, when it was not forthcoming, accused the NDI of being biased in favour of the NCFE.

Despite these pressures, the NCFE went ahead with a revised monitoring plan. Volunteers were instructed to wear their special badge and present themselves to vote at the station where they were registered. Once inside the station they should remain there to observe for as long as possible. If asked to leave they should interview voters outside [50]. This was partly successful: about 2,000 of the original 4,200 volunteers eventually filed reports and about 800 of these were allowed to remain in polling stations and conduct thorough observations. Elsewhere, observers were harassed by security officers or party agents and in Sana’a two were arrested [51].

The results, announced a few days after the election, gave the GPC a substantial lead over Islah, its nearest rival, and five more seats than Islah and the YSP combined:

However, with 47 independents and 12 representatives of smaller parties, no party had a clear mandate to govern alone. Indeed, because of the questionable allegiances of some of the newly-elected members, the scale of the GPC’s success was less than it appeared on paper. It was known, for example, that a number of GPC representatives were Islah sympathisers. One estimate put the figure as high as 30, plus three “independents”, and if this were correct it would potentially have given Islah 95 votes in parliament against only 92 for the GPC.52 Similarly, the YSP claimed that in addition to its 57 official members, 13 “independent socialists” had won with its support and a further 17 “independents” shared its “vision of the future” [53].

In the case of the GPC and YSP, the division of spoils was almost entirely geographical (see results by province). The GPC won seats in all the provinces of the north but only three seats in the south. The YSP secured an overwhelming victory in the south, where the party’s organisational apparatus proved its enduring effectiveness. Even in Abyan, the power base of the ousted socialist leader, Ali Nasser Muhammad, the YSP won seven of the eight seats. In the north, the YSP performed best in areas where there was opposition to the traditional Hashid establishment, winning two seats in the territory of the Sufyan tribe, and one among the bedouin of Ma’rib. While this was the first firm indication that the Socialists were able to make common cause with some of the northern tribes, the alliance appeared to be tactical rather than ideological. Predictions that Ta’izz, the northern province with the strongest links to the south, would swing to the YSP were unfulfilled – mainly because of the Socialists’ failure to agree a common strategy with the Nasserists, resulting in a split vote and victories for Islah and the GPC.

The successful Islah candidates came mainly from its conservative, tribal wing rather than the radical Islamist elements [54]. The party won no seats in the south, though it performed well in several provinces, having campaigned especially hard in Hadramawt, which has a strong religious tradition and close ties with Saudi Arabia. As a mark of personal respect for Shaykh Abdullah al-Ahmar, the Islah leader, neither the GPC nor YSP opposed his candidacy in Khamir, the Hashid capital north of Sana’a, and he secured an easy victory over the Haqq party. The victory of one of his sons, in Hajja province, was marred by several deaths and the destruction of the YSP headquarters by a rocket.

Among the smaller parties the Ba’ath was most successful, winning seven seats; al-Haqq won two in Sa’da province, the historical Zaydi stronghold, and the three Nasserist parties won one seat each (see results by party). The League of the Sons of Yemen, which campaigned in 87 constituencies, won none; on average, its candidates polled a derisory 186 votes each. None of the other parties won a seat. Where the smaller parties succeeded, their success was probably due more to the personalities involved than to policies or ideology. The leader of the Ba’ath party, for instance, Shaykh Mujahid abu Shawarib, is a brother-in-law of the Islah leader and was also at the time deputy prime minister in charge of tribal affairs.

Among the many independent candidates, 47 won seats. Critics lost no time in pointing out that the total votes cast in favour of independents was almost the same as the total obtained by the GPC – and yet because of the vagaries of the electoral system independents won only 15% of the parliamentary seats while the GPC won 41%.

Shortly after the polls closed, the American embassy in Sana’a issued a statement congratulating the people and government of Yemen on “the success of their first multi-party elections”, adding that the US looked forward to working with the new government [55]. In a similar vein, the official Saba news agency reported that international observers had “commended the measures enforced during the secret, free and direct balloting as well as the procedures carried out afterwards, including the sorting process. They stated that the Yemeni democratic experience constituted a big cultural leap forward in the region.” [56] With the counting still incomplete, such statements were seen by some as prejudging the question of whether the elections had been free and fair. They were not, however, without foundation: the international observers had formed an opinion based mainly on the electoral mechanisms and the safeguards they contained. Even before polling day, some observers had said they expected the process to be largely fair. For example, a representative of the American-based International Foundation for Electoral Systems said: “There is a system of checks and balances … the process is time-consuming and laborious but it will make allegations of rigging difficult.” [57]

More critically, a statement by the Yemeni NCFE complained that “numerous actions inconsistent with the spirit and letter of the law undermined the proper conduct of the elections.” Some of these, it said, “can be attributed to SEC officials at the local level, who seemed to have received uneven or inconsistent training. More often they can be attributed to the actions of agents of political parties.” It added: “The climate of fear and suspicion in many quarters, the minimal participation by women, and the lack of public education about democratic politics all rendered these elections less meaningful than they could have been and should have been.” However, it concluded: “Based on the information we now have, it is not clear that the outcome of any particular constituency election was altered by the election day voting or counting irregularities. Therefore, at present, our conclusion is that the announced results generally reflect the intentions of the voting population on April 27.” [58]

Elections, as a central feature of any democracy, offer a means for measuring its performance both in absolute terms and in comparison with other democracies. In new democracies such as Yemen they also provide a measure of the extent to which democratic principles have taken root more generally. In assessments of this kind the absolute, or ideal, standards should always be kept in mind, though in practice their usefulness is limited. In any electoral system there are elements which someone, somewhere, considers unfair, and it is quite usual for parties to seek advantage through tactics which are not based purely on the merits of their argument. Ultimately, therefore, the issue is a relative one: did the elections provide a reasonable test of opinion in the circumstances? 

After the elections the effectiveness of the international observers was questioned by some Yemenis, amid accusations that they had been ill-informed, easily duped and over-hasty in concluding that the elections were generally fair. This arose partly from a misunderstanding about the nature of their observations. Observers are not particularly looking for individual instances of cheating which can occur in almost any national election but tend to have no effect on the outcome; their main concern is the possibility of systematic fraud. The number of ways an election can be rigged is actually quite small and there are also well-established ways of structuring the electoral procedures to prevent systematic fraud or at least to ensure that if it occurs it will be noticed. A country which refuses to incorporate the normal safeguards into its system will immediately arouse suspicion.

Subsequent interviews with several of the international observers indicate that they had been aware of all the types of malpractice complained about by Yemeni citizens; the disagreement lay in what significance to attach to them. Experience of numerous electoral systems gives the seasoned observer a perspective on malpractice which is generally more sanguine than that of a voter or candidate experiencing it for the first time. For example, one observer said: “No election is perfect, and the Yemenis were starting from an imperfect situation.” He added that some of the more critical foreigners had been “looking for things they don’t even observe themselves in elections at home”. In general, then, the level of abuse was considered not sufficient to invalidate the whole election, especially since no party appeared to have benefited from it disproportionately.

Many of the complaints came from western-educated Yemenis with unrealistically high expectations or exacting standards. This is not to say that they were wrong to complain; insofar as their complaints might lead to future improvements they were valuable, but there is a distinction between that and what is immediately feasible. In the words of one seasoned British observer: “You have to ask: what is perfection, what is realistic? No election is perfect.” The problems, he said, can come from three sources:

1. Those who wish to control the outcome;
2. Those who wish to prove it unfair;
3. Those who fear the outcome or the process itself.

Without excusing the instances of malpractice or abuse that did occur, it is essential to keep them in perspective. The Yemeni experience illustrates the difficulty of holding elections in a poor country with large, sparsely populated areas, marginally effective central administration and a high illiteracy rate. Virtually all the general criticisms of electoral practice, as opposed to the localised cases of abuse, are directly attributable to one or more of these factors, or to the lack of adequate preparation. The manipulation, such as it was, occurred mainly before the election in the haggling over constituency boundaries and inter-party pacts whereby independent and minor-party candidates were squeezed out. Though the losers might consider that unfair, it was a legitimate part of the rough-and-tumble of politics. It was certainly not the kind of election sometimes seen in the Middle East (and elsewhere) where the government candidates routinely attract more than 90% of the votes. The problems faced by the small parties and independents were those that are found in any electoral system which does not set out deliberately to encourage them (for example, through proportional representation).

The British observer commented: “There were areas of influence and some fiddles, but it is difficult to swing a whole election. There was certainly an identification problem [in the polling stations]. Some women were getting the vote for the first time and their husbands were trying to decide for them. And the Saudis were disrupting, paying people and buying off tribes in the north. The election in Yemen was not as bad as I feared, not as good as I hoped. It was reasonable, and not atypical.”

© Copyright Brian Whitaker 2009