The Ruling Postponing the California Recall Election:
Why It Is Questionable, and Why It Was Not Compelled by Bush v. Gore

On September 15, the U.S. Court of Appeals for the Ninth Circuit ruled that the gubernatorial recall election must be pushed forward from its October 7 date to a date that could be as late as next March. During the interim, antiquated punch-card machines currently in use in some populous counties are to be replaced by more modern and ostensibly more accurate machinery. The opinion was issued by a three-judge panel, per curiam - that is, without a named author.

As of this posting, that ruling is in legal limbo as the entire Ninth Circuit decides whether to hear the case en banc - that is, by a panel of eleven of its judges, rather than just three. But whatever happens from here on out, an analysis of the reasoning of the three-judge panel's opinion is certainly warranted.

The Three-Judge Panel's Opinion: Admirable In Some Respects

Moreover, the three-judge panel was rather persuasive in its treatment of some of the non-constitutional questions, such as whether the punch-card challengers were barred by an earlier lawsuit or were to be faulted for not having sued earlier. I haven't heard much critical commentary, from academics or others, on these points.

But what about the merits of the panel's key holding that equal protection bars the October 7 election? That holding, in turn, was based largely on the Supreme Court's Bush v. Gore decision nearly three years ago. What are we to make of the Ninth Circuit's reliance on Bush v. Gore?

Understanding Bush v. Gore Itself

For starters, it's important to realize that Bush v. Gore is itself a hard decision to understand. Even defenders of the Supreme Court's 2000 action would have to admit that the Court's opinion, due to the tremendous time pressure under which it was written or for other reasons entirely, was no model of judicial clarity or craftsmanship.

We do know that a majority of the Court thought that the manual recount proceeding that was unfolding in Florida - whereby individual vote talliers in each county were using different criteria from those being used in other counties to determine how to read a ballot - violated the Equal Protection Clause of the Fourteenth Amendment by treating voters in some counties differently than voters in other counties. The Court said that a statewide uniform standard for hand-counting the votes was constitutionally required.

But the Court also said that the question before it was "not whether local entities. . .may develop different systems for implementing elections." So some diversity in voting mechanisms and systems across counties is permitted. But how much? And exactly what were the key "present circumstances" existing in Florida to which the Supreme Court's ruling was explicitly "limited?"

If the Ninth Circuit had trouble figuring out exactly what the animating principle of Bush v. Gore was, such difficulty is much more the fault of the Supreme Court than anyone else.

The Ninth Circuit Ruling's Questionable Analysis of Bush v. Gore

Having said that, however, I must quickly add that the Ninth Circuit ruling makes some analytic moves that are certainly open to question, if not criticism.

One is that the errors in counting votes in Florida were being made by people, not machines. Why might that matter? Because people who count votes by hand may make mistakes in counting the votes not just because they may be defective counters, but also because they may be politically motivated.

It is hard to completely understand Bush v. Gore without appreciating the concern shared by many people in the country that Democratic and Republican county elections officials might try to take advantage of the unfettered discretion state law gave them to hand count votes to get to a particular result. Machine mistakes, even ones that are predictably skewed in favor of or against particular groups or parties, may be different than mistakes made by individual persons who are motivated to make those mistakes on purpose.

Purposeful intent is often treated differently in the constitutional law of equal protection than is predictable disparate impact. And while many people alleged intentional manipulation in the Florida vote count nightmare, no one today alleges that the old-fashioned punch cards are used in urban populous counties in California because of a desire to disadvantage the causes or candidates who are most popular in those counties.

The Florida Recount Versus the California Recall Election

Nor is it just that the votes in Florida were being hand-counted (as opposed to machine-counted in California.) They were being recounted in Florida. That is crucial, because by the time of a recount - when the Supreme Court stepped in allegedly to promote fairness - we all knew that every miscounted vote (whether miscounted by design or by negligence) was a big deal, because we all knew the margin of victory was going to be a few hundred votes one way or another.

So the problem was not just manipulable standards that various counties were using; it was that manipulators knew exactly how many votes they needed to steal to tip the balance.

Things are very different in California. Let us assume the Ninth Circuit is correct when it says that use of punch-card machines will lead to as many as 40,000 votes going uncounted in the affected counties. Even so, there would be some uncounted votes in these counties using any vote counting machines, even the best available technology. So the real key number is not 40,000, but rather 40,000 minus the 15,000 or so inevitable uncounted votes.

Of that difference - of 25,000 - let us assume that 15,000 uncounted votes would have been counted for Governor Gray Davis, and 10,000 against him. (That would be pretty strong Davis country, inasmuch as he is running about even statewide.) All of this means that the punch-card machines are likely to cost Davis at most about 5,000 votes.

Was the Ninth Circuit's Chosen Remedy Truly Justified by the Law?

And yet the Ninth Circuit opinion never really addresses why its remedy was justified to avoid a constitutional violation that in all likelihood was going to be what we call in the law "harmless error." Indeed, imagine that Secretary of State Kevin Shelley had told the Ninth Circuit that he would like to run the election as planned, so as to not waste the money already spent, and that he would stipulate that he would not object to any after-the-fact remedy should the election turn out to be an extremely close one and should the punch card machines be proven to have caused an undercount that could have affected the outcome. If the Ninth Circuit is acting completely honestly, such a concession should cure its concerns.

Now perhaps the Ninth Circuit panel thinks that that we should simply not run any risk of a disenfranchisement that might affect the outcome of the election, because moving the election back from October to March simply doesn't cost society very much at all. This is where the Ninth Circuit's reasoning is most inflammatory. The court says that the State's interest in sticking to the October date (on which everyone had been counting) is "weak" because the choice between October and March is "arbitrary."

It is arbitrary, says the court, because if the signatures that triggered the recall hadn't been collected or counted until August (rather than July, when they were in fact certified), then state law would itself dictate a March recall date. (If there is a regularly scheduled election within 180 days of the certification of signatures, then state law directs the elections authorities to save resources by consolidating the recall election with the regularly scheduled election.)

The Court Ignored the Value of Sticking to a Pre-Set Schedule

According to the Ninth Circuit, extending that 180-day outer limit to 225 or so days (the time from the July certification until the March election, the earliest time punch card machines could be replaced) is no big deal, because the choice between 180 and 225 days is "arbitrary."

If so, then so is the choice to have a Presidential election every 4 years, rather than every 5 years. And yet, could we imagine a court pushing back a Presidential election because a year here or there never hurt anyone?

Or consider the choice to give a defendant 20 days within which to respond to a complaint, rather than 30 days. At some level, that rule is arbitrary, but only in the sense that every rule, at some level, is arbitrary.

But what is not arbitrary is the value of sticking to a pre-agreed upon schedule, around which everyone has arranged their financial and emotional commitments. Just imagine a lawyer saying to a judge: "Your honor, I realize I didn't file my response within the required 20 days, but that 20-day time frame itself is rather arbitrary, so if you don't excuse my 3-days tardiness, you are acting arbitrarily and thus unconstitutionally." That the Ninth Circuit didn't see the weakness of its "arbitrariness" analysis is rather surprising.