Americans like to believe that every vote counts, that we govern ourselves by majority rule and that the principle "One Man, One Vote" ensures equality.
So much idealism is attached to the act of voting that its mechanics are usually overlooked. But you don't have to look far for evidence that the slogans are wrong:
· Every vote doesn't count, because not all votes are counted. In the 2000 presidential election, nearly 3,000 votes in Hamilton County were never counted.
· The majority doesn't always rule, because laws sometimes prevent it. Al Gore received more popular votes in 2000 than George W. Bush, but Bush occupies the White House.
· "One Man, One Vote" is a principle whose application depends on an unscientific process called reapportionment, with sometimes scandalous results.
The right to vote is central to the integrity of a republic. Some citizens — notably women and African Americans — got access to the polls only after heroic struggle. But a combination of mechanics, regulations and political manipulation affect the value of our votes — in ways that until recently received little attention.
Hang the chads
Chad is a man's name, a country's name and the name of the tiny perforated rectangle that, when punched out of a card, indicates a voter's preference on an issue or among a slate of candidates.
The chad is the operative portion of a punch-card ballot.
Sometimes a voter fails to push hard enough to separate a chad from a ballot. Vote-counting machines tabulate the number of holes left by chads that have been pushed out, so a chad that is not completely removed can result in a vote going uncounted.
Chads were at the heart of the dispute over the 2000 election in Florida, which allegedly gave Bush a victory in the Electoral College.
But chads are not a problem in Florida alone. Hamilton County also uses punch cards, and chads have caused problems here, too.
In October 2001 The Columbus Dispatch quoted Ohio Secretary of State J. Kenneth Blackwell on the voting system: "I think the punch-card system should go tomorrow."
Tomorrow has come and gone, but we're still punching chads.
Chads given too light a touch result in "pregnant" chads or "hanging" chads, with a voter's intent not quite clear. The debacle of the 2000 presidential election in Florida — when the word "chad" came into the nation's vocabulary — caused Ohio to develop guidelines for determining voters' intent.
Ohio law now requires the removal of hanging chads, according to Tim Burke, chair of the Hamilton County Board of Elections. Normally, a hanging chad has two or three corners broken out. Such a chad is removed, Burke says, "because there's a clear indication of the fact that the voter intended to vote that punch. It just didn't break out cleanly the way it does 99.9 percent of the time."
Hanging chads are only one problem with punch-card ballots. Over-voting, in which a voter punches holes for two candidates for the same office, is another.
Getting your vote to count depends in large part on your ability to vote correctly — not over-voting or under-voting the ballot. Say, for example, you vote in the upcoming race for governor and you punch holes for both Tim Hagan and Bob Taft. This would be an over-vote, and your vote in that particular race would not count.
In Hamilton County, 384,336 people cast ballots in the November 2000 presidential race. But 2,916 were over-votes — and therefore were not counted. In November 2001, in the city of Cincinnati, there were 113 over-votes in the mayor's race. The city council race had 2,279 over-votes among the 89,079 ballots cast.
The punch cards used by Hamilton County and many other parts of the country — a pre-perforated variety — are especially susceptible to problems, according to Howard Strauss, information technology manager at Princeton University. He says it's the worst system.
"It's very difficult to secure," Strauss says.
Every time someone tries to hand-count the ballots, some of the perforations get knocked out, he says.
"It will soon look like you voted for everybody," Strauss says. "Even doing the right thing with them ... can cause some of the chads to pop out."
'Taking people's votes away'
The number of over-votes in the city council election — 2.56 percent of the total ballots — is a cause for concern, Strauss says.
"It may not be malicious, but at least your ballots are confusing," he says.
Making ballots too complicated for voters is a problem, according to Strauss.
"In a real sense, though, what you're doing is taking peoples' votes away," he says.
Hamilton County Commissioner Todd Portune says those numbers point to a serious problem.
"Hamilton County loses a sizable amount of votes on average every year," he says. "There's no doubt that the kind of punch card, butterfly ballot system that we use in this county is a problem."
Hand recounts of elections tabulated by punch-card machines "tend to confirm the general accuracy of a count," Burke says. But he, too, believes the system should be replaced.
After examining election data, Burke says the highest incidence of over-votes occur in poorer, generally African-American communities.
Burke is not alone in his assessment. The American Civil Liberties Union (ACLU) of Ohio filed suit Oct. 11 in federal court in Akron, saying the voting systems in Hamilton, Summit, Montgomery and Sandusky counties violate the rights of minority voters.
More than 94,000 Ohio voters had their ballots rejected in the 2000 presidential election, according to the ACLU.
"The ACLU of Ohio has conducted statewide research on this problem, which reveals a strong relationship between the racial composition of a precinct and the percentage of discarded ballots," according to a statement by the organization.
It's time to change the system in Hamilton County, according to Burke.
"I do believe we need to be moving toward the elimination of the punch card," he says. "Each year we have thousands of people who are casting a vote ... and for one reason or another they cast more votes than they're entitled to."
Julie Stautberg, director of the board of elections, says the butterfly ballot used in Hamilton County is not the same kind used in Florida, which has only one column for two pages.
"Ours has always been the two open faces but a column for each page," she says. "This year we added a blank separation down the middle."
She says this helps in cases such as the election for Cincinnati City Council, in which voters are allowed to pick up to nine candidates. If a large number of candidates are on the ballot — there were 26 in 2001 — all the names wouldn't fit on the same page. Voters might choose all their candidates from the first page, not realizing there are more on the next page. Or voters might try to vote for an additional slate of candidates on the second page after having chosen nine on the first page.
"We present the names to the voter so that they're looking at all the candidates at one time, rather than trying to flip back and forth," Stautberg says.
Hamilton County Commissioner John Dowlin has served on voting-system panels with the County Commissioners Association of Ohio and the National Association of Counties. In Hamilton County, the punch-card system works, he says.
In February 2001 Dowlin The Columbus Dispatch quoted Dowlin saying, "My question is, if it ain't broke, why fix it? In my county, people aren't complaining about the ballot system."
"If John doesn't think losing thousands of votes each year in Hamilton County is a problem, well then I guess that's his opinion, but I would disagree," Portune says.
Bryan Flannery, a Democrat running for Ohio Secretary of State, also believes the punch-card system is flawed.
"It's an outdated system that goes back 30 years," he says. "The reason it's inherently flawed is because it disallows votes to be counted."
There is an easy way to eliminate under-votes and over-votes. It would mean, however, punching the punch cards to the curb. Blackwell says technology to protect against over-votes can be found in the second-chance optical scan or an electronic touch-pad system that five counties in Ohio already use.
Franklin County uses an electronic touch pad system, which operates like an automated teller machine.
"You vote your ballot with your finger," Blackwell says.
In Franklin County, the vote is collected by electronic impulse, without a paper trail. But the new electronic touch pad systems leave a paper trail, Blackwell says.
"When you vote, instead of getting an 'I Voted Today' sticker, you get an immediate receipt that you voted," he says.
The system also has an internal paper trail, he says.
Blackwell predicts that in three to four years, punch cards will be decertified and new technology will be mandatory. He says cost is the reason Ohio hasn't yet overhauled its voting system. But Flannery says the state can't afford to wait.
Tests of electronic systems have been run in many counties around Ohio, according to Flannery.
"They have found that people that use that — they're very satisfied with it, because it's technology that they're familiar with," he says.
The electronic system alerts a voter if he under-votes or over-votes and allows him to correct it, Flannery says. This, he believes, would help ensure that all ballots are always counted.
"That's what democracy is based on," Flannery says. "What differs us from other forms of government is our vote, and when you have Third World countries that have a more advanced voting system, that's a slap in the face."
'No way to know'
Some, however, believe technology is not the solution, but rather part of the problem. With fully computerized voting, there's no way of doing a recount, according to Rebecca Mercuri, an expert on electronic voting and a professor at Bryn Mawr College.
"The recount is just the same data that got sucked in to begin with," she says.
Without any paper trail of your vote, there's no way to verify that it was accurately recorded.
"There's absolutely no way to know ... once it gets into the machine whether or not it was tallied correctly," Mercuri says.
Chads aren't the only problem in counting votes. Questions about the use of computers in elections surfaced 17 years ago in Cincinnati. In 1985 Hamilton County Common Pleas Judge Richard Niehaus ruled, "There is no adequate and proper safeguard against the computers being programmed to distort the election results."
That was three years before Cincinnati even heard of Leonard Gates, a former Cincinnati Bell installer who admitted widespread illegal wiretapping for the Cincinnati Police Division and the FBI.
In 1988 Gates told a federal grand jury he wiretapped the Hamilton County Regional Computer Center, where votes were counted, on election nights in 1977, 1979 and 1981. The purpose, he said, was to manipulate vote tabulations in Cincinnati City Council races.
Votes in Hamilton County are no longer counted at the Regional Computer Center. Today that work is done at the office of the board of elections. The system is only available to certain employees and is protected by a password, according to Dennis Predmore, elections specialist for the board.
"The modem is not even on the system," Predmore says. "It's only hooked up if and when we would need assistance from our vendor — and that's not on Election Day."
The computers used on election night are in the vote counting room, where they can be watched by the staff. Ballots are watched by representatives from both parties, according to Tony Reissig, administrator of operations at the board of elections.
"The law stipulates that there be a Democrat and a Republican that witness every action where ballots are concerned," he says.
But even though the system is not hooked up to a modem on election night, the fact that a modem is even available means something could happen, according to Mercuri.
Using a modem, an off-site person could influence the data or programming inside the machine, she says.
Without inspecting how the system works, Mercuri says, it's impossible to know whether a person could use a modem prior to an election to change the firmware (the software inside the machine). But it's theoretically possible, she says.
"I don't have any way of knowing without inspection of the machines whether it could be programmed or not from the outside," she says.
Even if the board of elections staff programmed the computers, it doesn't write the code for the system down to the firmware level.
"If they're completely legit, it still could be being rigged by some other facility," Mercuri says. "It could be triggered to go off at some other point. There's absolutely no way to know what's going on within the machine."
Robert Strunk, a computer programmer who gave an affidavit in the 1985 case heard by Niehaus, says deceptive programs can be written.
"Once you're inside that black box, you can spin the wheels and make it sing 'Happy Birthday' if you want," he says. "The computer is only like a fourth-grader. It can only add, subtract, multiply and divide. The computer can do absolutely nothing without the programmer. It takes a program to make a computer boot."
You can watch, but you can't see
Since August, visitors are able to watch votes being counted at the board of elections. The computer has been moved to a third-floor office where it's visible through windows. Visitors can witness stacks of ballots brought in and placed on a table, then fed into the machine for counting. Stautberg says this gives people interested in the election process another opportunity to see it at work.
"It's more of us trying to make the process more understandable to people," she says.
But what real value does the public display provide? After all, the actual counting occurs inside the computer, a function of lines of code decipherable only by a programmer. A program to handle elections might have 20,000 lines of code, Strunk says.
"Reading a program to tell what's going on — it is really a difficult job," he says.
After watching vote counts when they were at the Regional Computer Center, Strunk went back to observe again when the procedure was moved to the board of elections.
"I'd say that nothing really changed," he says. "You could have 1,000 people watch this program run and they wouldn't be able to say, 'Yeah, I certified this and it did the right thing.' "
Seventeen years ago Strunk testified, "It is very easy to make a computer program seemingly read the data on a reel of tape, then completely ignore that data and print out some pre-arranged answers."
Today, he still believes so. He says it would be a "piece of cake."
Mercuri says most problems in computer programming are the result of mistakes, not necessarily done maliciously.
"I'm more concerned about general stupidity or somebody just did something silly or they didn't test for the right thing and now it's going haywire on you," she says.
But not all the problems would have to occur within a machine. In 1985 the Cincinnatus Party hired James and Kenneth Collier — brothers who spent much of their lives investigating voting fraud — to videotape and evaluate Hamilton County's election system.
In their 1992 book Votescam, the Colliers wrote they had observed women using tweezers to pluck chads out of punch-card ballots.
"The women pointed out that the vote card was blistered on the back," the book says. "Ken focused the unlit camera on a card which showed about seven little 'pocks' on the back. They were in the exact same spot on every card. The women were tweezing these pocks off each card because the blister prevented the card from passing through the counting machine. It appeared that the Port-O-Punch had been used to quickly punch a slate of seven candidates."
In Miami, Fla., the brothers reported "hole-punching parties," using a machine that could punch identical holes in a pad of 50 ballots.
You gotta have faith
Stautberg says computers at the Hamilton County Board of Elections are checked for problems.
"We run tests before and afterwards, logic and accuracy tests," she says.
In public dry runs of the system, the board demonstrates how the equipment operates, putting test decks through the system.
A white paper by Election Systems and Software, the vendor that maintains Hamilton County punch-card system, assures many safeguards are in place to prevent tampering in elections.
"It should be noted that despite periodic allegation of 'vote fixing' and 'equipment tampering' in the news, to date, there has not been solid evidence that provides proof of computer voting equipment manipulation," the company says. "Contrary to the public's common misconceptions pertaining to proprietary computer programs, manufacturers and vendors that develop and provide election-related software are often required to escrow these software source codes with a third party depository. The claim of protection of proprietary intellectual property rights is not valid in this case as data redundancy, accessibility and the public trust is involved."
But Mercuri says there is no way of tracing what that escrow is and whether it relates to what is actually in the voting machine. The source code does not live inside the machine; rather, a translation of the source code does.
Mercuri also scoffs at the claim of data redundancy as proof of anything, because the machine could just spit out the same redundant bogus data over and over again.
The last paragraph of the white paper by Election Systems and Software reminds us that in order to believe our votes are counted, we have to trust the people who do the counting.
"Election security does require the public to 'trust but verify,' with vigilant, informed voters and responsible election administrators," the paper says. "Any voting system is only as secure and dependable as the people who administer and operate the equipment."
But the verification the company is suggesting is not possible, Mercuri says.
"You can't," she says. "There's no way unless you do a hand recount of the punch cards."
Stautberg says Republican and Democratic employees of the board of elections program the vote counting machines. The Ohio Secretary of State requires county boards to do their own programming, she says.
"Here in Hamilton County we do all our own programming," Stautberg says. "We do all of that in-house."
But Strauss, the information technology manager at Princeton University, is doubtful.
"I don't think they have written the program themselves," he says.
The program that counts the holes in a card is normally purchased from a manufacturer, according to Strauss.
He's right. The programming by election officials only tells the computer which candidate or issue corresponds to which hole in the punch card.
"When you punch things in a card, you're not voting for Sue Smith or John Jones," Strauss says.
Instead, you're voting for column 6 or 9 or 12 — and the computer has to be told what that means.
Predmore says the machines used by the county have a margin of error of less than one-tenth of 1 percent.
But Mercuri says this is the margin of error when counting pristine ballots, in which everything is perfect on a ballot, with no hanging or pregnant chads. In real-life circumstances, she says, the margin is higher.
"Actually the margin of error is 3 to 5 percent," she says.
Even if the margin of error were only one-tenth of 1 percent, Mercuri points out this isn't small. It means that one in 1,000 votes are not correctly counted. In the 2000 presidential race, this alone would mean that 384 votes in Hamilton County were inaccurately counted.
In deciding elections, everything depends on the machine. A complete hand count isn't done to verify results.
All ballots are checked by hand for hanging chads, Stautberg says. But that's not the same as a hand count.
"We do not do a hand count for the official count," Stautberg says. "It's all done by a machine."
In fact, even in a recount of a close election, only 3 percent of ballots are actually counted by hand. The tally of that 3 percent sample is recounted by the machine to ensure the numbers match. The machine then recounts all the ballots in that race.
Stautberg says that since she started working at the board of elections in 2000, recounts have turned up the same results as the first count. Hand counts aren't done, she says, because there is trust in the machines and the systems chosen by the Ohio Secretary of State.
There is also the issue of efficiency.
"I wouldn't even want to guess how long it would take to count all of the ballots cast in Hamilton County by hand," Stautberg says.
Burke says hand counts are impractical.
"It would take forever," he says.
People who were involved in elections in the 1950s say it used to take a week to count the results in a city council election, according to Burke.
"Today, for better or worse, everybody is into instant gratification," he says.
"Folks don't have the cultural orientation to wait the days that it would take to count those ballots," he says. "To revert back to hand counting ballots would be like perhaps saying it would be more environmentally friendly for us to ride horses. Why not just go back to riding horses? People just aren't going to go backwards that way."
Besides, Burke points out, there is still such a thing as human error.
"Even a hand count is certainly not going to be perfect," he says.
But counting ballots by hand was not nearly as slow as Burke claims. Election Day in 1957 was Nov. 5. The Nov. 7 morning edition of The Cincinnati Enquirer reported the unofficial count of the city council election. The results had become available the night of Nov. 6, one day after the elections. The total votes cast in that election were 140,584 — about 50,000 more than the most recent election. Yet the count only took about 24 hours.
The November 1963 election was the last to be counted by hand in Hamilton County, according to The Cincinnati Post-Times Star.
'Strange stuff floating around'
Assuring the integrity of the vote has two layers, according to Strauss. One is the actual voting system. The other is the procedure used to run it and ensure its integrity.
"You have to have good procedures," he says. Tampering with a voting system, he says, requires attacking either the procedures or the system.
The people involved in the system are also capable of making mistakes.
"Humans are humans," Strauss says. "Humans get tired."
Humans can also be bribed.
"Give any programmer lots of Coke and pizza — they'll do anything you want," he says.
Strauss points out that most voters wouldn't notice if a hole had been punched out for a candidate before they vote. If a person wanted to tamper with an election, he could punch out a hole for his favorite candidate before it reached the voter.
Strauss also points out that when cards are being moved to a location to be counted, it would be fairly easy to replace voted cards with fake cards.
Strauss says he's seen vote counting machines used between elections to generate water bills. The elections officials explained that it was OK because they cleared the data prior to the election and the printout on the machine read zero.
This concerns Strauss.
"I could get a computer to print all zeros and there could be strange stuff floating around in there," he says.
Strauss says there are hundreds of ways to attack the system.
Blackwell says no one can satisfy every "conspiracy theorist," but he believes the systems are safe. He believes it helps that Ohio has a decentralized system, with standards that have to be met statewide but no requirement that a county pick a certain vendor or use a particular technology.
"That way it would take thousands of people collaborating to mess with a statewide election," he says.
The Hamilton County Board of Elections has approximately 45 employees. About 15 more people are hired for Election Day, Stautberg says.
In addition, 4,100 poll workers staff the 1,025 precincts throughout the county. Two Democrats and two Republicans work at each precinct.
In some cases, extra workers help by examining ballots as they arrive at the board of elections, according to Burke. A Republican and a Democrat look to see that there are no hanging chads and the ballots are not damaged.
The extra employees don't work in the secure counting room, however.
The number of ballots used and the number of ballots collected are counted by poll workers at the close of a polling station and also at the board of elections.
Most of the time the count ends up the same, according to Stautberg. When the numbers don't match, efforts are made to find the problem. Occasionally poll workers say a certain number of people voted, but they don't have that many ballots, Stautberg says. The poll workers go through their notes and find that someone took a ballot but didn't turn it in.
'Never heard boo' from Blackwell
In order to have your vote counted, you have to be able to vote in the first place. Across Ohio in November 2000, this was sometimes easier said than done.
With the mandatory redistricting that follows every U.S. Census, the boundaries for state and congressional representatives change. So do precinct lines.
When precincts were redrawn following the 2000 Census, confusion developed about where to vote.
Flannery, a state representative from the 17th district, says people still call him, thinking he's their representative — and he has to tell them he's not.
"People were going to their old polling places to vote and they were being told, 'You're not registered here. Go home and we'll call you and tell you where to vote,' " Flannery says.
Portune says he voted for years at a polling place in the basement of a neighbor's home. After redistricting, the people who live on the other side of the street still vote there. His precinct, however, changed. Now he has to vote a mile and a half from his home.
But even worse than that kind of confusion is the redistricting controversy at the heart of a lawsuit against the Republican-controlled Ohio Apportionment Board. The suit accuses state officials of redrawing district lines to dilute the voting power of African Americans.
In April, WCPO (Channel 9) reported the transcript of a phone call between Brett Buerck, chief of staff for Ohio House Speaker Larry Householder, and Doug Mink, a fellow Republican. Mink was planning to run in the primary election against Jim Raussen.
Redistricting is supposed to fairly reapportion boundaries so that districts are roughly equal in population. The goal is "one person, one vote." Otherwise the vote of a person in a district with five people would be far more powerful than the vote of a person in a district of 5,000 people.
But in the transcript obtained by WCPO, Buerck points to a different reason for the new district lines.
"Because of the changes that we've made in redistricting to help Jim, we essentially took 13,000 African Americans out of the Raussen district and put 14,000 Republicans in," Buerck said, according to the transcript. "We sat down with Senate President Dick Finan and asked Dick Finan to allow us to take his hometown of Sharonville (Evendale) and move it out of the Michelle Schneider district into the Raussen district. It was not a very pleasant conversation in order to get President Finan to do that. But he likes Raussen, he understood the commitment that the Speaker had and he was willing to make a personal sacrifice in order to make this happen."
A lawsuit in U.S. District Court for the Southern District of Ohio claims the 2001 Ohio Reapportionment Plan violates the Voting Rights Act of 1965 and the 14th and 15th amendments to the U.S. Constitution. The plan makes it harder for African Americans to elect representatives of their choice, according to the plaintiffs, who include State Sen. Mark Mallory (D-Cincinnati) and Tyrone Yates, a candidate in the 33rd House District.
"The 2001 plan 'packs' some African Americans into a single or few districts and 'fragments' others into many districts," the lawsuit says. "In both situations, the effect is to dilute the voting influence of African Americans and place them in a segregated political environment."
As secretary of state, Blackwell sits on the Ohio Apportionment Board. Flannery says Blackwell did what his fellow Republicans wanted with the redistricting, in the process disenfranchising many African-American voters.
"You never heard boo from him," Flannery says.
Blackwell says the law was followed. Districts were made compact and reasonable, minority districts were preserved and "opportunity districts" were created, he says.
"The fact of the matter is I was in constant conversation with the Legislative Black Caucus," Blackwell says.
But Shelley Cooper, executive director of the Legislative Black Caucus, says she never talked with Blackwell about redistricting and he never called her office.
State Sen. C.J. Prentiss (D-Cleveland), president of the caucus, says Blackwell didn't talk to her about redistricting either.
"There has not been a discussion between me and Mr. Blackwell on this issue," she says.
Prentiss says neither Blackwell nor his staff have attended the caucus' meetings to discuss the issue.
"We have Wednesday meetings and Ken Blackwell never attended one of those meetings," she says.
Portune says the reapportionment process raises serious questions.
"If there is intent to cut people out of a district to not empower them to elect a representative of their choice, then yeah, I think those motives are questionable," he says.
In order to elect a representative of your choice, the choice has to first be available on the ballot. This won't be the case for Marilyn Hyland's supporters next month.
Hyland thought she had collected enough signatures on candidacy petitions to run as an independent for the Hamilton County Board of Commissioners. But the board of elections decided she didn't have enough valid signatures to make the cut.
"The board voted to reject the signatures of 127 residents of Hamilton County because they were printed, rather than cursive," Hyland says.
That ruling effectively disenfranchises people who wanted the chance to vote for her, according to Hyland.
"Signing a petition is a part of exercising your voting rights," she says.
At a hearing before the board of elections, Hyland pointed out that Burke, the president of the board, printed his signature, rather than use cursive writing.
"If that's the case, is he a legally registered voter in the State of Ohio?" Hyland asks. "If he's saying that in his judgement the Ohio Supreme Court wants a signature in cursive, according to his logic, is his signature even valid?"
Burke isn't alone. Hyland says Hamilton County Auditor Dusty Rhodes printed his name on the candidacy petition for Dr. Jean Siebenaler, who's running for the county commission.
But Burke says the problem with Hyland's petitions is that signatures on her petition didn't match signatures on voter-registration cards at the board of elections.
He says the job of the board of elections is to determine whether a signature is valid. Petition signatures have to match signatures on file, as directed by the Ohio Secretary of State, Burke says.
"The majority of these folks that were not accepted — the printing doesn't match anything at the board," he says.
Burke says his printed signature, like Rhodes', is on file with the board of elections.
Hyland says that kind of pickiness is a disservice to voters.
"I think the citizens of Hamilton County have been treated unfairly," she says. "We need to ensure that people are dealt in, not thrown out."
But Hyland's complaint is more than a matter of quibbling over signature styles. An additional 150 people who signed her petitions were not counted as registered voters — even though, as was later determined, they actually are registered.
Hyland says that because 47 people had written zip codes on the petition, instead of the date, the board invalidated their signatures.
But even if all of Hyland's petitions — those with non-matching signatures, those with zip codes for dates — were counted, the law acts to make it harder for independent candidates to get on the ballot.
As an independent, Hyland needed 2,830 signatures to get on the ballot. If she'd been running with a party endorsement, she would have needed only 50.
"We need, as a democracy in the United States in 2002, to be dealing people in," Hyland says. "This just is a rude awakening as to how fundamentally tight-fisted the power's being used at the leadership of the board of elections." ©