KABUL: One of the two Afghans on a UN-backed commission looking into vote fraud in the August presidential election resigned on Monday, citing interference by foreigners. Officials acknowledged that errors and miscommunication have plagued the investigation.
The panel’s rulings on how many votes to throw out will determine whether the vote goes to a runoff between President Hamid Karzai and former Foreign Minister Abdullah Abdullah. Preliminary results show Karzai winning with about 54 percent of the vote, but if enough votes are voided, he could dip below the 50 percent needed to avoid a second round.
In announcing his resignation from the commission, Maulavi Mustafa Barakzai alleged that the three foreigners on the panel -- one American, one Canadian and one Dutch national -- were “making all decisions on their own” without consultation.
The complaints commission rejected the allegation.
Barakzai was “an integral part of the commission and took part equally in all commissioner meetings,” the group said in a statement. It said the resignation “will not distract” it from investigations. There will be no interruption in the group’s work, said Nellika Little, a spokeswoman for the commission.
Barakzai was appointed by the Afghan Supreme Court. It was not immediately clear if the court will be able to appoint a replacement.
Meanwhile, the head of the commission, Canadian Grant Kippen, said a number of errors of interpretation have slowed or complicated the investigation and partial recount. The process started a week ago.
Officials have finished examining a sampling of suspect ballot boxes, but rulings are likely to take days.Kippen acknowledged that the commission misinterpreted the statistical analysis that would be used to decide what percentage of votes to void for each candidate. He told reporters last week that each candidate would lose votes in proportion to the number of fraudulent ballots cast for them in a sampling of suspect boxes.
Instead, each candidate will lose the same percentages of suspect votes, based on the number of fraudulent ballots found in the sample, Kippen said on Monday.
He said the actual process has not changed and that it is statistically sound, but that confusion stemmed from miscommunication between statisticians who designed the mathematical procedure and commissioners whose role is to determine whether the individual boxes are fraudulent.
He said the mistake was caught and explained to candidate representatives before the commissioners started on their decisions. “It hasn’t affected the process,” he said. “It has probably affected people’s perception of the process.”
Carlos Valenzuela, who helped design the sampling method, said he only learned that the commissioners had misinterpreted his system after they talked to reporters. “We actually saw this because of reports in the press,” he said.
The system appears to penalize all candidates equally, regardless of whose camp committed the fraud. Valenzuela said this type of “simple sample” was the only option that would provide accurate results within a limited time.
