I couldn't agree more -- with both groups. The Votomatic punch-card machines discussed in the court's ruling are a national travesty; they've been widely condemned for decades as inaccurate, insecure devices seemingly designed to facilitate errors, manipulation, and fraud. But the court's cure -- postponing the recall until they are replaced with touch-screen voting terminals -- would disenfranchise far more voters than it might aid. Specialists in elections technology who've studied the aftermath of the 2000 Florida vote count learned a clear lesson that didn't get much publicity: The best way to disenfranchise voters is to switch balloting technology immediately before a major election.
"It's very clear that if you change your voting system, you cause a huge spike in the voter error rate," said Douglas Jones, a University of Iowa political science professor specializing in the history of voting technology. "You don't want to change voting systems very frequently. You also don't want to change them before a high-profile election. You want to try it first during an off-cycle election, something like an election for library district, something where 2 percent of the voters show up. You can use that as a dry run, then do it in a major election."
If any switch in voting technology causes problems, the touch-screen machines slated to replace California's Votomatic lemons are likely to be double vote-count trouble. The new touch-screen systems have already exhibited security problems, and it doesn't look like these problems will be solved by March. Of particular concern is the lack, at present, of any paper record of votes cast.
In other words, if problems arise, the new, supposedly improved systems -- whether they are first used in October or in California's March primary, expected to be a major battleground for Democratic presidential candidates -- would not allow a manual recount.
"I want our counties to be able to recover from the kinds of snafus and glitches we see in every election," said Kim Alexander, president of the California Voter Federation, a Sacramento group advocating for voter rights. "If something goes wrong with the software, when those problems and glitches occur, we may not have any way to recover from them."
Buried under the partisan bickering surrounding the Ninth Circuit's initial recall decision (which a larger panel of the Ninth Circuit is reconsidering as I write this column) is a scandalous history of American voting technology, one that may continue to be ignored once the current round of arguing dies down. Despite a quarter-century of widespread, highly publicized criticism, this country's slipshod system of administering elections has drifted from scandal to scandal, unperturbed.
An apparent alignment of interests -- actually, a lack of interest in change -- extends from voters to poll workers to candidates to party bosses to vote-machine entrepreneurs, allowing the system to persist.
Punch-card machines -- essentially slightly reworked surplus IBM computer punch-card readers -- have been cheap, durable, and easy to understand. They have, therefore, been favorites of poll workers. Candidates have been as likely as not to benefit from inaccurate poll results, so rarely will even half the electorate extend much outrage. Bad polling technology has been seen as affecting minority and other disenfranchised voters, and therefore has not been an issue regularly championed by Republicans. And though nobody likes to say it, voting error and fraud have a natural constituency: whichever side happens to benefit.
For decades, small, secretive, for-profit firms have provided poorly designed, failure-prone voting machines to local poll workers, producing an intermittent stream of mangled results. Hanging and dimpled chads represent but a small portion of the problems associated with the Votomatic punch card machines discussed in the Ninth Circuit's decision to postpone the recall. Perhaps more important is the way the cards are counted; experts who've seen the closed-source software used in the counting process say it is easily subject to vote-tampering.
Battles over Votomatic's wide-ranging electoral deficiencies erupt with regularity following close or suspicious-seeming votes, usually in tiny local elections, sometimes in regional contests, once in a presidential election. In these debates, the machines become a symbol used to allude to everything one side hates about their political foes. Typically, though, the ballot machines are sidelined in the conversations.
I'll admit, for example, to not having read Bush v. Gore until last week; I'd considered the U.S. Supreme Court decision a judicial coup d'état and didn't think there was much more I needed to know. Likewise, the voluminous history of the Votomatic punch-card machine comprising the bulk of the 66-page Ninth Circuit decision in Southwest Voter Registration Education Project et al. v. Kevin Shelley has not informed the gist of most public discussion of the case.
For example, as I waited in the chow line at a Republican fund-raiser last week, the woman in front of me said, "They're saying minorities can't figure out how to use ballot cards; that's insulting." The decision actually says nothing about the technical abilities of minority-group members. Rather, it says error-prone punch-card balloting machines are concentrated in minority-heavy voting districts.
A columnist wrote in last week's Chronicle that she had interviewed experts on both ends of the political spectrum, and all of them said the Ninth Circuit panel decision was political retaliation for the U.S. Supreme Court's Bush v. Gore decision. Actually, though, when people of different political persuasions blindly speculate about the motives behind a court decision and reach the same conclusion, it doesn't mean they have engaged in anything more meaningful than blind speculation.