A trustworthy voting system should reliably record the voter’s choices and provide a dependable method of recounting the votes. However, South Carolina’s voting system provides neither.

All voting machines used in South Carolina are iVotronics. Mathematical formulas and computer instructions are used to match the position of the voter’s finger on a touch-screen with a candidate’s name and then to store the vote in the machine’s memory. After the selections are verified on the screen, the contents of the touch-screen are erased, leaving no trace of the selection on the screen. The voter must trust that his/her selection has been correctly recorded in the memory of the machine because there is no way to compare the selections made by the voters with the results reported by this voting machine.

The State Elections Commission claims that these machines always work correctly, but there is no way to prove or disprove that the iVotronics have always worked correctly in South Carolina. Since there are no paper ballots to compare against the machine totals, nobody knows whether the machine totals are correct or not.

Experience elsewhere strongly suggests that iVotronics machines do not always report votes correctly. In addition to a number of reports in other states of unlikely results in elections that seem to suggest machine error, there is also a well-documented failure of the iVotronics in Faulkner County, Arkansas, in 2008. Two voting machines allocated votes cast in one race to an entirely different race that wasn’t even on the electronic ballot. The voting machines reported votes for candidates in the race for constable as votes for a different candidate on a non-existent ballot for a State House seat. How this happened remains a mystery, but, this event provides indisputable evidence that iVotronics machines can make mistakes.

In Faulkner County the iVotronics machines each had a voter-verifiable paper tape recording each vote as it was cast. Examination of these tapes revealed the correct vote, but did not explain why the complicated computer instructions inside the iVortonics failed to record the vote for the correct candidate in their memories. The iVotronics used in South Carolina do not have a voter-verifiable paper tape, so that method of verification is not available. If results from a South Carolina election are implausible, there is no recourse, no way to check accuracy, because there are no original ballots to compare with the reported results.

If the results for two candidates are sufficiently close, South Carolina law requires that the votes be recounted. But, with the iVotronics machines, no such recount of the votes is possible because there are no ballots to recount. In a recent primary race in Florence, SC, totals for the two leading candidates differed by a single vote. However, there were no ballots to recount. Instead, the totals from the precincts were added together a second time and, since the totals were correctly added both times, the same “winner” prevailed.

The voting machines used in South Carolina provide neither a reliable method of recording the voter’s choices nor a method of recounting the votes. South Carolina voters deserve better. The League of Women Voters supports a permanent ballot (such as a paper ballot), marked by the voter and retained for use in a possible recount. The ballots can be tallied by an optical scanner like one already in use in most counties to count absentee ballots. Such a system would provide both a reliable method of recording the voter’s choices and a dependable method of recounting the votes.

Eleanor Hare is an Associate Professor Emerita of Computer Science at Clemson University and a member of the Board of Directors of the League of Women Voters of South Carolina