/ 14 August 1998

Fair or fraudulent result?

Roger Southall

A Second Look

Your correspondent William Boot is being incautious in suggesting that Lesotho’s May election was rigged by the ruling Lesotho Congress of Democrats (LCD) (“Lesotho’s election farce”, August 7 to 13). A more careful look at the election is required.

Preparations for the election began under the Independent Electoral Commission (IEC) in September 1997. A constituency delimitation commission had already delimited the 80 constituencies now required to replace the 65 into which the country was divided in 1993. Registration began in January 1997.

The preparations were not without problems. An IEC vehicle carrying registration papers was stolen and the papers dumped; the director of elections was suspended following allegations of involvment in actions inconsistent with his role; and on May 14, the leaders of the main opposition parties complained that the provisional electoral lists were not being freely provided to them for scrutiny. This complaint was later referred to the high court in a bid to have the election postponed, but it was ruled that although there had been some irregularities, these were insufficient to justify a postponement.

The official result recorded the LCD – which had broken away under Prime Minister Ntsu Mokhehle from the then ruling Basotholand Congress Party (BCP) in 1997 – as having won 78 out of the 79 constituencies contested on general election day. The LCD later also won a postponed election in the Moyeni constituency.

A preliminary comparison between the 1993 results (won by the BCP by 65 seats to nil) and data for 77 out of the 80 constituencies in 1998 is illustrated in the table above.

The election was observed by 400 observers from NGOs in Lesotho and 150 observers from other countries and international organisations. Their joint preliminary declaration noted that “the election was conducted in a manner … acceptable according to international standards”.

Even so, the major opposition parties began to challenge the validity of the results. In 1993, the Basotho National Party (BNP) had argued that a “neo- robotic” outcome had elucidated a conspiracy to fraudulently rig the election. This charge was sustained neither by the high court nor by post- electoral academic analysis. When faced with a similar resounding loss in 1998, the party again claimed the election was rigged, but this time it was joined by the BCP and the Maramatlou Freedom Party in launching a sustained campaign against the result.

The parties declared they had appointed “experts” to audit the results. Their statement which released the result of this audit was based on an examination of 31 out of the 79 constituencies, obstruction by the IEC having allegedly necessitated a “truncated analysis”. The burden of the audit is as follows:

(i) Comparison between the number of votes cast, as recorded by the IEC, and the number of voters’ names crossed off the voters’ lists for the 31 constituencies covered by the audit, reveals a substantial difference. In essence, the allegation is that the IEC invented “ghost” voters to swell LCD totals.

(ii) To quote the Mail & Guardian (August 7 to 13): there were also “some seemingly inexplicable demographic peculiarities … In every constituency, analysis of the roll revealed a vastly higher number of births … on January 1 than on any other day of the month. The same was true in the ascending sequence: February 2, March 3 … a larger number of people had birthdays in the first half of the month than the second.”

A case based upon the above complaints was brought to the high court on July 22. The judge ruled that the evidence was insufficient to prove systematic fraud, but ordered the full records of the IEC be made available for scrutiny.

The M&G (August 7 to 13) reports that a subsequent unofficial recount of votes cast in half the constituencies indicates there was systematic rigging, and that the LCD should have won only 12 out of these 40 seats. Only time will prove the validity or otherwise of the opposition case. However, initial comment must be that the claim of election fraud must be treated with caution:

(i) Terminology used in the press release reporting the results of the audit is less than impressive: the total votes cast as recorded by the IEC is termed an “exit poll”, which is something wholly different. No attempt is made to suggest how the IEC misled or received covert co-operation from the host of independent observers to boost LCD majorities.

(ii) Voters’ lists could well concentrate birth dates in the first half of the month. Many people in Lesotho do not know their actual birth date. Computer default in the registration process may have allocated such people to the first day of the first month, etcetera.

(iii) If, for argument’s sake, the fairness of the electoral process is assumed, then the LCD’s overwhelming victory recorded by the IEC is wholly consistent with the tendency of the British-style plurality electoral system to over-represent the winners, just as it did in 1993. The failure of the opposition to win more than one seat is not inconsistent with the reported capture of 60% of the vote by the LCD. In contrast, if the LCD/BCP vote had split down the middle, so rather than amassing 60% and 10,5% of the vote respectively, they had attained some 35% of the vote each, then it might be expected that the seats they obtained would tend towards equality, with the split of votes allowing the BNP to emerge as winner in seats where it has historically enjoyed strong support.

However, according to the IEC this did not happen. Indeed, if the IEC figures are accepted, they demonstrate a remarkable voter consistency in the face of confusion within the party that won the last election. In short, the proportion of votes going to the smaller parties plus the BNP remained roughly the same, as did the total vote going to the LCD and BCP combined. But, in the face of allowing a BNP comeback, the vast majority of those who voted for Mokhehle and the BCP in 1993 would appear to have voted for Mokhehle and the LCD in 1998. The simplest explanation of the election outcome is that this is what happened.

Demands for the king to intervene to annul the elections amount to an attempt to precipitate a coup. (Contrary to Boot’s article, the king has no “constitutional right” to do so.) Yet whatever the outcome, it is apparent segments of Lesotho’s political class will refuse to acknowledge the validity of the election – even if independent analysis were to confirm the result.

The problem stems from the huge extent of the LCD victory, which, as in 1993, demonstrates the inappropriateness of the continued use of a first-past-the- post electoral system. Were the LCD to have won a 60% popular vote in a single- list proportional representation system as operates presently in South Africa, the opposition would have 40% representation in the country’s Parliament. In contrast, 40% of the electorate presently have no effective voice within the National Assembly. Under proportional representation, opposition leaders would have something of a stake in the political system.

The way forward would seem to require:

l the early holding of a neutrally chaired constitutional conference to consider whether the electoral system should be reformed in the direction of proportional representation;

l scrutiny and appropriate revision of the voter’s list by a reconstituted IEC, headed by a director of elections appointed from a neutral body from outside the country as in 1993;

l commitment by the government to the holding of a further election following completion of the above tasks.

Roger Southall is professor of political studies at Rhodes University, and co-editor (with T Petlane) of Democratisation and Demilitarisation in Lesotho: The General Election of 1993 and Its Aftermath.