11/21/00 4:45 p.m.
The War of the Dimples
The more dire the numbers look for Gore, the more infinitesimal the marks argued over will be.

By Robert Alt, adjunct fellow, John M. Ashbrook Center for Public Affairs

ll eyes are currently focused on the Florida supreme court, but in all probability the decision they render will not decide the election. While it is possible that they will issue an order that effectively settles the matter one way or the other, the serious betting is that they will split the baby — allowing some but not unlimited time for counting, and offering nebulous and inconclusive guidance as to what should count as a vote. If this happens, the real question — the question which may well decide the election — will remain unresolved: whether a dimple counts as a vote.

The hand recounts are not going as well as Gore had hoped. With 103 of Palm Beach County's 531 precincts reporting, Gore has picked up a total of 3 votes. Gore had hoped to pick up hundreds of votes in Palm Beach. He therefore has filed an emergency petition for a Wednesday hearing before Palm Beach County Judge Labarga, in which he is seeking to require the election board to count the dimpled chads. Judge Labarga had previously issued an order finding that a per se rule excluding partially punched chads was impermissible where evidence of voter intent was clear, but he did not require the dimpled chads to count.

Unlike the rapid revisions and re-revisions in "standards" that the canvassing boards have implemented in this election, the Palm Beach County board had, prior to the current dispute, maintained a uniform rule since 1990: "A chad that is fully attached, bearing only an indentation, should not be counted as a vote." This standard was not arbitrary, or intended 10 years before the fact to disenfranchise Gore voters, but was used because "an indentation is not evidence of intent to cast a valid vote." Given the risk of fraud and self-dealing after an election, it is important to maintain pre-election standards unless they are clearly unlawful, which this one is not.

A quick look at the elusive dimpled chad demonstrates why the old standard is correct, and why a dimpled chad does not demonstrate the voter's intent and therefore should not be counted. In Palm Beach County, the Democrats have disputed 276 dimpled chads which the board said did not evidence intent of a vote. In some cases, there were dimples on more than one candidate. These clearly should not have been counted because they would not have been counted if the dimples were completed perforations. In others cases, all the votes on the ballot in question were perforated, but the presidential vote was merely dimpled. While the board did not suggest why the presidential vote in the last example would be dimpled but not perforated, the argument is relatively clear: buyer's remorse. Standing in the voting booth, a reluctant voter may set the stylus to a name, but retract it before making the perforation. Seeing a tiny bump, but reading on the ballot that the chad must be removed, the voter could turn in the ballot without intending to vote for either of the candidates. Gore supporters will doubtlessly discount this explanation, but it is more probable than the theory that every mark, no matter how slight, evidences the true intent of the voter. The buyer's remorse theory also better explains why a voter would perfectly perforate every vote except for the presidential slot. While other theories will abound, given the examples of ballots in Palm Beach and the alternative explanations of what they may mean, dimples simply do not provide a sound basis for determining a voter's intent.

The problem of the dimpled chads is illustrative of why Bush has resisted the hand counts: With a lack of defined standards, interested parties will push the criteria as far as they can until they get the result for which they are looking. If canvassing boards or the courts require some form of dimpled chads to count, you can be assured that there will be skirmishes to the bitter end about degree: The more dire the numbers look for Gore, the more infinitesimal the marks argued over will be. A better solution would be to rely on the 1990 no dimpled chad standard: Its pre-election status means that it is less likely the product of bias, and its clear boundary means that it is less subject to manipulation.