UNITED STATES COURT OF APPEALS

FOR THE NINTH CIRCUIT

 

 

 

 

SUSAN MARIE WEBER ,

Plaintiff-Appellant,

 

v.

 

KEVIN SHELLEY,* in his official

capacity as California Secretary of  

State; MISCHELLE TOWNSEND, in her

official capacity as Riverside

County Registrar of Voters,

Defendants-Appellees

 

 

*Kevin Shelley is substituted for his predecessor, Bill Jones, as California Secretary of State. Fed. R. App. P. 43(c)(2)

 

 

   No. 02-56726

 

D.C. No.

CV-01-11159-SVW

 

Plaintiff-Appellant's

Petition for

Re-Hearing En Banc

 

 

 

On Appeal from the United States District Court

for the Central District of California

Stephen V. Wilson, District Judge, Presiding

___________________________________ 

This is an appeal from a decision by a 3-judge panel of this court, seeking re-hearing by that panel, or review by the court sitting en banc.

Introduction

This is a case of exceptional national importance. The heart of this case involves the near-unanimous opinion of computer science professionals which was misapprehended by the three-judge panel and the District Court under review.

The eyes of the entire nation were on this court, sitting en banc, as it reviewed a long chain of cases evaluating “punch-card” voting systems and determining the fate of a recall election.

The constitutional significance of the punch-card cases is vastly exceeded by this case. Those cases involved claims of single-digit percentages of the electorate having their right to vote infringed. This case challenges the new “touchscreen” systems, in which there is virtually no limit to the disenfranchisement possible, and the legitimacy of representative government is at stake. A firestorm of controversy is erupting nationwide, with state and local election authorities in need of guidance from this court. This lawsuit was filed over two years ago after the publication of a warning by the computer experts at MIT and CalTech which vastly exceeds the threat by punch-card systems:

A Provocative Scenario

A programmer at SlickVotingMachines Corp. adds malicious code to a DRE (Direct Recording Electronic device) machine for the California 2004 Presidential election, so that every fiftieth vote for a Republican candidate is changed to a vote for the corresponding Democratic candidate. This only happens when the machine is in “real” mode as opposed to “test” mode, so the election officials never discover the fraud during their testing. The electronic audit trail made by the DRE machine is also affected, so “recounts” never discover anything amiss.
( http://web.mit.edu/newsoffice/nr/2001/VTP_report2.pdf  )

 

Spurred by federal funds, touchscreen systems are being sold by voting machine corporations whose salespersons are working day and night. This court needs to take control of this issue, removing control from the hands of voting machine corporations and giving guidance to legislatures and election officials.

There are three reasons why both the District Court and the 3-judge panel erred in sustaining the constitutionality of such a dangerous and unreasonable voting system:

1)  They both completely confounded the material issue of  “verifiability” and the immaterial issue of “accuracy.”

2) This confusion resulted in the misapplication of the requirements of the Constitution as spelled out in Burdick v. Takushi

3)  The judges shirked their constitutional responsibility to give judicial guidance to the other branches of government.

1. Verifiability vs. “Accuracy”

Although the right to vote is the most fundamental of all our rights, it is not explicitly spelled out in the Constitution or the Bill of Rights. The issue in this case is more fundamental than whether a small percentage of the electorate is unable to completely punch out chads. This case goes to the bedrock of the right of the People to create a legitimate representative government.

The three-judge panel called the infringement alleged by the Plaintiff merely “hypothetical”:

Weber has raised at most a hypothetical concern about the ability to audit and verify election results. . . . [T]he impact on her right to vote is minimal.

Slip op at 15429

 

The infringement in this case is not “hypothetical,” but it is so staggering that it defies statistical quantification. It is fundamental. This analogy illustrates how fundamental the issue in this case is.

Imagine that in 1765 King George III concluded that public support for the Stamp Act needed to be buttressed. The king determined to hold an election on the issue. Not leaving anything to chance, the king determined that voters would tell British soldiers whether they supported taxation without representation or opposed it, and the British soldiers would mark a ballot for the voter. The voters would not be permitted to verify how their ballots had been marked by the British soldiers. British soldiers would then count -- in secret -- the ballots they created for the voters, and then announce the results of the “election.”

Both the District Court and the 3-judge panel of this court ruled in favor of this constitutional monstrosity on the grounds that no evidence was presented showing that the abilities of British soldiers to count the “ballot images” they created were “inherently less accurate” than those of other vote-counters (slip op. at 15433).

This issue of “accuracy” is a complete red herring. The problem with a total reliance on computers is that voters are not allowed to cast their own ballot, but their “ballot images” are created by proprietary software, and there is no way voters or voter authorities can verify that the voting machine corporation recorded the voter’s preference for Candidate A, or whether the voting machine corporation created a “ballot image” for Candidate B, negating the intent of the voter.

The State correctly and consistently calls these “ballot images.” They are not voter-created or voter-verified ballots. It does not matter how “accurately” “ballot images” are counted if they do not faithfully represent voter intent.

It is logically and constitutionally irrelevant what kind of security measures the King employs to protect the “ballot images” created by the British soldiers. The 3-judge panel mentions “redundant data paths” and “instantaneous systems checks,” echoing the sales pitch of voting machine vendors (slip op. at 15431), but this would not have prevented the Revolutionary War from breaking out a decade earlier. The “accuracy” with which these software-created “ballot images” are counted, stored, or re-counted is completely irrelevant.

The Constitution guarantees us a Republican form of government. If the President, Legislature and Judiciary were replaced by a Führer, it would be shocking to hear someone say that our suspicions were merely “hypothetical.” This form of government is our right, regardless of whether any harm has occurred or can be “proven.” Similarly, the right to vote is the most fundamental of all rights, and when votes are created by a corporation or the government itself, and when the overthrow of an election becomes undetectable because verification has been lost, the right to vote has been infringed.

When auditing a citizen, the IRS would never let the taxpayer off the hook if the taxpayer simply hit the “refresh” button on the spreadsheet software and printed out another copy of the tax return. If Enron boasted of storing all its (cooked) entries in “redundant memory” and claimed its computers made calculations with great speed and accuracy, no sensible auditor would be satisfied without seeing the original source documents. The government election officials, the District Court, and the 3-judge panel of this court were deceived by flashy claims by voting machine vendors. Computer scientists have not been fooled. They demand a genuine verifiable audit of election returns. In time, a majority of voters will too.

The issue is verifiability, not “accuracy,” and beneath this is the issue of trust. The 3-judge panel wrote:

Weber makes a number of interrelated arguments, which boil down to distrust of a system that records a vote without a paper trail. 

Slip op at 15431-2

 

The right of the voters to distrust the government is the heart of the American experiment in self-government. Madison observed in Federalist No. 55,

[T]here is a degree of depravity in mankind which requires a certain degree of circumspection and distrust. . . .

 

We have “checks and balances” and a “separation of powers” because Americans recognize that human beings are easily tempted and cannot be trusted with unmonitored power. It is un-American to require voters to trust the government with a computer voting system which computer experts almost unanimously condemn.

We are also being asked to trust voting machine corporations and their “proprietary” software.

Courts have recognized that implicit in the right to vote is the right to have our votes counted. Now this court is called upon to declare that the People have the right to cast their own ballot and count their own votes.[1]

 

2. The Burdick[2] Standard

Infringement of the Right to Vote is examined under strict scrutiny. But merely minor infringements will not be examined under strict scrutiny if they are incidental to an otherwise “reasonable” voting system. The three-judge panel quoted Burdick:

A court considering a challenge to a state election law must weigh the character and magnitude of the asserted injury to the rights protected by the . . . Fourteenth Amendment[ ] that the plaintiff seeks to vindicate against the precise interests put forward by the State as justifications for the burden imposed by its rule, taking into consideration the extent to which those interests make it necessary to burden the plaintiff’s rights.

Slip op at 15434

 

 The magnitude of the injury described by the computer experts – such as the CalTech/MIT scenario above – is nothing less than the destruction of legitimate government. Unverifiable elections under such a voting system are not perceived to be legitimate. Elections become the creation of the voting machine corporation. The magnitude of this injury is staggering.

This injury has to be weighed against the state's interest in NOT adding a voter-verified audit trail to the touchscreen machines.

What is that interest? What is the state protecting by refusing to allow voter-verified elections?

Law vs. Facts.     The District Court and the 3-judge panel analyzed the Burdick Standard based on a factual conclusion that conducting an election in complete reliance upon the proprietary software of a computer system without a voter-verified audit trail was “reasonable.” But who determines “reasonable?” It would be virtually impossible to find a professionally-trained computer science expert – such as the over 1,000 computer science professionals who have signed Stanford University Professor David Dill’s Internet petition [http://www.verifiedvoting.org ] -- who would agree that it is “reasonable” to conduct an election in complete reliance upon the proprietary software of a computer system without a voter-verified audit trail. None of the experts who signed Prof. Dill’s petition or who were offered to the District Court by this plaintiff were consulted by Riverside County or California state voter authorities before certifying or adopting the Touchscreen system.

The 3-judge panel summarized Burdick:

the question is whether using a system that brings about numerous positive changes (increasing voter turnout, having greater accuracy than traditional systems, being user-friendly, decreasing the number of mismarked ballots, saving money, etc.), but lacks a voter-verified paper ballot, constitutes a “severe” restriction on the right to vote.

Slip op at 15435

 

None of the “positive changes” are lost by attaching a voter-verified audit trail to the touchscreen system. Is forcing voters to allow the government to create the voters' ballots -- without voters being permitted to verify the ballots -- a "severe" restriction on our right to vote?  Would King George III's system of Redcoat-marked ballots be considered by Jefferson to be a "severe" restriction of our rights?

The lower court should be directed to examine the warnings of computer science experts and ask whether the Defendants in this case have a compelling state interest in NOT modifying computer voting systems to permit voter verification. And just as the lower court accelerated the timetable of decertification of punch-card voting systems, the lower court should accelerate the timetable of decertification of computer systems which lack a voter-verified audit trail.

 

3. This is a Judicial Responsibility, Not a Legislative or Executive One.

At the end of their opinion, the 3-judge panel simply shifted responsibility to other branches of government:

[I]t is the job of democratically-elected representatives to weigh the pros and cons of various balloting systems. So long as their choice is reasonable and neutral, it is free from judicial second-guessing

Slip op. at 15436

 

Whether true or false, this statement requires a reversal of the District Court’s summary judgment. It is false in the sense that it is courts, not the legislature, that must apply the Burdick standard. Punch-card voting systems had been certified by the Secretary of State, but were declared unconstitutional by the same District Court from which this appeal was taken. In Common Cause, ACLU,  et al vs. Secretary of State, 01-03470SVW (RZx), the District Court did not accept the idea that either the Legislative or Executive branches had a monopoly on determining the constitutionality of voting systems. That court should have listened to the computer science professionals in this case who declare with a nearly unanimous voice that computer voting without voter-verification is systemically “unreasonable.” Voting systems which are “unreasonable” are unconstitutional when examination under strict scrutiny reveals massive infringement of the right to vote.

It is virtually the unanimous opinion of computer science experts across the country that complete reliance upon computers to create “ballot images” -- with no voter-verified audit trail -- is dangerous. They would use the term used by the Supreme Court in Burdick: “unreasonable.” The reasonableness of a computer voting system without a voter-verified audit trail is the material heart of this case. This material issue is hotly disputed, and the District Court erred in granting summary judgment and not declaring such an unreasonable system to be unconstitutional.

If the statement of the 3-judge panel is true, then our “democratically-elected representatives” still need the guidance of this court. Neither the California Legislature nor the Riverside County Board of Supervisors weighed the evidence presented by the Plaintiff. The “cons” presented by the computer science experts were not weighed against the “pros” of vendors representing voting machine corporations. No computer experts were involved in the process in Riverside County, and a newly-elected Secretary of State did not form a Task Force to examine the problems inherent in touchscreen systems until after this Plaintiff recommended such in the course of this litigation.

It is the constitutional duty of the judicial branch to erect constitutional boundaries within which the legislature must operate. Such boundaries have not yet been handed down, and the entire landscape of election science is now being defined by voting machine vendors with a bloodhound’s sense of the scent of federal funds.

It is notable that in oral argument before the 3-judge panel, counsel for the state conceded that Plaintiff’s arguments had been weighed and the scales had been tipped in Plaintiff’s favor. She is now “preaching to choir,” state’s counsel told the court, explaining that the Secretary of State had begun establishing timetables for implementing voter-verified audit trails. But at this point the State believes such modifications of unverifiable election systems are discretionary rather than constitutionally mandated.

This court must take control of this issue. The District Court must be ordered to examine the conclusions of computer science and public policy experts that computer voting systems without voter-verified audit trails are unreasonable. Such an unreasonable system must be found unconstitutional under Burdick and modifications ordered implemented as soon as possible. This court cannot allow voting machine corporations to mislead election authorities anxious to increase voter turnout with “new and improved” but unconstitutional voting systems.

 

Future prospects

It must be realized that a majority of Americans no longer vote. Those who do vote tend to be more informed than those who do not. They tend to be computer-savvy, and increasingly get their news from the Internet, which is now awash in warnings about unverifiable “black box” computer voting systems. Already in Riverside County nearly 40% of those who vote choose to vote using paper absentee ballots. If the right of the people to create and cast their own ballots is not affirmed by this court, there is no doubt that in 2004 there will be a flight from touchscreen machines, and more than 50% of the voters will seek refuge in paper-based absentee voting.

There is presently exponential growth of voters who are computer-savvy enough to see the difference between “verifiability” and “accuracy.” Fewer and fewer voters will care if a computer “accurately” counts “ballot images” created by proprietary software which do not faithfully reflect voter intent.

 

Conclusion

1)  The 3-judge panel followed the District Court’s error of confusing the issues of “accuracy” and “verifiability.”

2) The 3-judge panel followed the District Court’s error of misconstruing the requirements of the Constitution as spelled out in Burdick

3)  Both the 3-judge panel and the District Court shirked their constitutional responsibility to give judicial guidance to the other branches of government.

Therefore the decision of the 3-judge panel should be reviewed by this court sitting en banc, and the District Court’s summary judgment reversed.



[1] The California Election Code contains some protection for voter verification. EC 15272 states the voting public must be allowed to observe the count; EC 15360 states that they will be reassured by 1% manual count.  Now this is all done internally by a computer program that may or may not be correct or corrupt.  Now these codes are rendered absurd. 

[2] Burdick v. Takushi 504 U.S. 428, 112 S. Ct. 2059 (1992)