Reprinted in part from Pelast's book by workingforchange.com.

'This series is part of the WorkingForChange campaign, in cooperation with Martin Luther King III of the Southern Christian Leadership Conference, to prevent the theft of the presidential election of 2004. There is a link included to sign onto the WorkingForChange/King petition.

McKinney Nails the Confession

There's Thomas Cooper who's crime is still in the future.

And then Johnny Jackson Jr., on the purge list because his name partially matched that of a man convicted in Texas, John Fitzgerald Jackson (no "Jr."). Johnny Jr.'s never been to Texas, and his mama swears to me he never had the middle name Fitzgerald. Neither is there any evidence that John Fitzgerald Jackson, the felon, ever left Texas -- or ever left his jail cell. There are 638 John and Johnny Jacksons (and permutations thereof) in the Florida phone book. How did the state know they had the right Johnny? They didn't; and it looks like they didn't want to know. Using the address history database, as the state was promised, would have saved Jackson, a Black man, his right to vote.

Then there's Wallace McDonald, age sixty-four. Wallace tells me how in 1959 he fell asleep on a bus-stop bench and was busted. Even for a Black man in then-segregated Florida, that was a misdemeanor, not a felony. He never lost his right to vote; and the state agrees he was wrongly "scrubbed." Had DBT checked the databases, as promised, they would not have named Wallace.

Willie Dixon is on the list, too. The Reverend Dixon was convicted decades ago, and has received full executive clemency. That would have been an easy one to catch if the state had checked and verified the clemency records as per the contract.

Read down the list and mismatches jump out at you. Note they have taken voting rights away from Randall Higginbotham, age forty-one, because of the crimes of Sean Higginbotham, age thirty. The list is lousy with suspicious matches: pairing voter David Russell Butler Jr. of Florida to convict David Butler of Ohio. No question why David R. registered with his full name and appended the Junior. There are sixty-six other David Butlers listed in the Florida phone book and they must get one another's mail all the time. It is disturbingly improbable that they purged the right Butler. That should have been a no-brainer to correct.

The wrong Butlers, Smiths and Jacksons remained on the list because of DBT's "matching logic" and "matching criteria." Credit card companies can require thirty-five matches for verification before they will issue you plastic. The State of Florida was content with a partial match of four: names (the first four letters were good enough), date of birth, gender and race. Not even the address or state mattered in the mad dash to maximize the number of citizens stripped of their civil rights.

Rather than add matching criteria to verify the list, the state told DBT to remove criteria. For example, Messrs. Butler and Jackson so carefully added "Jr." to their official names to avoid such confusion. Tough luck. I found an internal mail in Roberts's office, dated June 14, 2000, in which clerks fretted about what they called "tweaked" data, allowing "matches" between Edward and Edwin (and Edwina!); deliberately ignoring middle names and initials; and skipping the "Jr." and "Sr." suffixes.

I met with a Willie D. Whiting of Tallahassee. The Reverend Whiting confessed he had a speeding ticket a decade ago, but doubted that should cost him his right to vote. But there he was: on the purge list, matched with Willie J. Whiting -- no "Jr." -- whose birthday was two days different from Willie D.'s. Our experts looked at the paltry number of match criteria and were horrified. One, Mark Hull, told me the state and ChoicePoint could have chosen criteria that would have brought down the number of "false positives" to less than a fragment of 1 percent. He said it made him ill to learn what the company had agreed to do. These revelations were especially upsetting to him; he had been the senior programmer for CDB Infotek, a Choice-Point company.

Vulture Opportunism versus Conspiracy

I now began to understand the brilliant deviltry of the Black voter purge game. It did not matter if, on Day One of the purge process, Republicans had some grand plan, some elaborate conspiracy, to eliminate the vote of African-American innocents. Rather, document after document suggested that, once the operatives saw the demographics of the raw lists -- tens of thousands of names of mostly Democratic voters -- they moved heaven and earth to prevent its reduction. A list of 57,000 voters, mostly Black, erased with the flick of a switch was just fine with Ms. Harris and crew.

Make verification phone calls? Have statisticians check the findings? Correct the methods? Why, that would only cut the list . . . by 90 percent at least. Why should a Republican administration pay for that?

It's not "conspiracy," but opportunism. The Department of Elections Republicans began to act like a bank customer who accidentally receives a million-dollar deposit that is not theirs: To fail to correct the error, to actively conceal the error, is theft in any court. Only here the crime was far bigger: the theft of our democracy. Opportunism does not require planning and conspiracy; it does require a cover-up.

In any investigation, I try to imagine myself in the perps' shoes. If I had a magic list falsely accusing my opponents' voters of crime, how would I prevent the discovery that it is bogus? First, don't dare verify the list; not one phone call. Second, don't correct the methodology: Ignore every warning about crap inputs, crap methods, crap results. And third, for God's sake, don't allow any independent statistician near it.

The Case of the Missing Statistician

Florida's contract with DBT states:

DBT shall consult a professional statistician. . . . Upon the return of the processed data, DBT shall supply the formulas and mathematical calculations and identify the professional statistician used during the verification process.

The 8,000-name Texas list had a 100 percent error rate -- which seemed a wee bit high to me. What kind of "academically based formula" was used to verify the accuracy of these data? Who was the consulting "professional statistician"? Inscrutably silent on whether he or she exists, ChoicePoint DBT referred me back to Clay Roberts. His minions could not name this Man of Mystery either, although the contract requires DBT to provide evidence of the statistician's hiring and analysis. Neither the name nor the calculations were filed as required.

Eventually, I found this: a letter dated March 22, 1999, from DBT to the state. "Our" statistician, said the one-page note, "certified" their list as 99.9 percent "accurate"! I can imagine why "our" statistician would remain nameless: 99.9 percent accurate but almost every name an eligible voter. No backup. Nada. How convenient. No independent technician, no expert to see things go rotten, no one to blow the whistle.

Evidence of Innocence: "Don't Need"

I turned back to the question of Florida hiring DBT for $2.3 million, booting the company charging $5,700. When questioned, George Bruder, ChoicePoint DBT's senior vice president, said, "a little birdie" told him to enter that astonishing bid. What else did the little birdie tell him?

What happened to the 1,200 databases, the millions and millions of records that DBT used in its Carl Saganesque sales pitch to the state? In fact, the state paid for this vital cross-check -- or at least DBT's bid said that for their two-million-dollar fee, they would use artificial intelligence for "cross-referencing linked databases . . . simultaneously searching hundred of data sources, conducting millions of data comparisons, compiling related data for matching and integration."

In all, they had four billion records to check against. Under "Offer and Bid" it read:

"DBT will process total combined records from:

"8,250,000 Criminal Conviction Records
69,000,000 Florida Property Records
62,000,000 National Change of Address Records
12,590,470 Florida Driver License Records. . . ."

And so on. The phone calls, the massive data crunching, it all justified the big payoff to DBT and scared away competitors who could not match DBT's database firepower. DBT's offer promised "273,318,667 total records to be processed." But they didn't do it.

Once the contract was nailed, it seems a little birdie in the state told DBT not to bother with all that expensive computing work. In the state files, on the DBT bid, I found a handwritten notation, "don't need," next to the listing of verification databases (the 62 million address histories, etc.), though this work was included in the price.

Each pass would have cut the list by thousands, thereby letting thousands more Democrats vote. So when the state said, "Don't need," the underlying motive was, "Don't want."

"Wanted More Names Than We Can Verify . . ."

DBT's "expertise" in obtaining data justified their hiring. But it was a con. Janet Mudrow, the state's liaison with DBT, confessed to me that DBT merely downloaded lists from eleven states that make the data available publicly, such as Texas. Any high school kid with a Mac and a credit card could have grabbed the names off the Internet. And that was okay with Florida, even though eight of those states do not take away an ex-felon's voting rights, and therefore should not have been used at all.

DBT's negligence in handing Florida the bogus Texas list cost Florida and its counties a pretty penny when they tried to reverse that error. Yet Mudrow, in Harris's office, says the state neither demanded reimbursement nor sought any penalty as permitted under the contract. In fact, the state awarded DBT another contract renewal, bringing total fees to over $4 million.

Why didn't the state complain, sue, or withhold payment? Following my first reports, when the stats hit the fan, ChoicePoint DBT agreed to a one-year extension of their contract without charge. But why didn't the watchdog bark?

One can only conclude that Harris's office paid an awful lot of money for either (a) failed, incomplete, incompetent, costly, disastrous work that stripped innocent citizens of their rights, or (b) services performed exactly as planned.

Was DBT paid to get it wrong? Every single failure -- to verify by phone, to sample and test, to cross-check against other databases -- worked in one direction: to increase the number of falsely accused voters, half of them Black.

How could ChoicePoint, such an expert outfit, do such a horrendous job, without complaint from their client? You'd think their client, the state, ordered them to get it wrong. They did. Just before we went on air in February 2000, ChoicePoint vice president James Lee called us at the BBC's London studios with the first hint that the state of Florida instructed the company to give them the names of innocents. The state, he said, "wanted there to be more names than were actually verified as being a convicted felon." What an extraordinary statement. When ChoicePoint saw the story with their own words -- "more names than were actually verified" -- printed across the screen, the company went ballistic. They demanded in writing to my network chiefs that we retract it all. The BBC wouldn't back down an inch.

McKinney Nails the Confession

Following the February 15, 2001, broadcast, only one member of the U.S. Congress called BBC to ask for our evidence: Congresswoman Cynthia McKinney. This lady is trouble, the kind of trouble I like. A Black single mom and doctoral candidate at Princeton's School of Diplomacy, she is always asking questions. And in the world of politics, that makes her dangerous -- "radioactive," as a staffer from the Democratic National Committee describes her. Unusual for a member of Congress, she reads the detailed memos and evidence herself, not delegating the research to underlings. She knows her stuff. McKinney represented Atlanta, the location of ChoicePoint's headquarters. She demanded their executives appear before a special hearing. As usual, she had some questions she wanted answered, in public. So I handed McKinney --and ChoicePoint -- the evidence. ChoicePoint was shoveling a lot of nonsense my way, but I figured the company might hesitate about shucking and jiving a member of the U.S. Congress.

On April 17 ChoicePoint VP James Lee opened his testimony before the McKinney panel with notice that, despite its prior boast, the firm was getting out of the voter purge business. Then the company man, in highly technical, guarded language, effectively confessed to the whole game. Lee fingered the state. Lee said that, for example, the state had given DBT the truly insane directive to add to the purge list people who matched 90 percent of a last name -- if Anderson committed a crime, Andersen lost his vote. DBT objected, knowing this would sweep in a huge number of innocents. The state then went further and ordered DBT to shift to an 80 percent match. It was programmed-in inaccuracy. Names were reversedófelon Thomas Clarence could knock out the vote of Clarence Thomas. He confirmed that middle initials were skipped, "Jr." and "Sr." suffixes dropped. Then, nicknames and aliases were added to puff up the list.

"DBT told state officials," testified Lee, "that the rules for creating the [purge] list would mean a significant number of people who were not deceased, not registered in more than one county or not a felon would be included on the list. Likewise, DBT made suggestions to reduce the numbers of eligible voters included on the list."

Correct the list? Remove those "not a felon"? The state, says DBT, told the company, Forget about it.

Hunting the Black Voter -- the June 9 Letter

Florida was hunting for innocents and, it seems, the Blacker the better. To swing an election, there would be no point in knocking off thousands of legitimate voters if they were caught randomly -- that would not affect the election's outcome. The key was color. And here's where the computer game got intensely sophisticated. How could it be that some 54 percent of the list were Black? There is no denying that half of America's felons are African Americans, but how could it be that the innocent people on the list were mostly Blacks as well?

In November, ChoicePoint's PR men jumped up and down insisting in calls to me that "race was not part of the search criteria." The company repeated this denial in press releases after they were sued by the NAACP for participating in a racist conspiracy against citizens' civil rights. DBT complained to my producers and to federal investigators: Race was not a search criterion, period!

Then, I obtained a letter dated June 9, 2000, signed by Choice-Point DBT's Vice President Bruder written to all county elections supervisors explaining their method: "The information used for the matching process included first, middle, and last name; date of birth; race; and gender; but not Social Security Number."

They had not lied to me. Read closely. They used race as a match criterion, not a search criterion. The company used this confusion between "match" and "search" criteria to try to pull the BBC off the track. They tried to slide the race question by the U.S. Civil Rights Commission. However, on the morning of February 16, the day after our broadcast, I faxed to the commission the June 9 letter. Later that day, the commission questioned Bruder.

COMMISSION: Was race or party affiliation matching criterion in compiling that list?
BRUDER: [under oath] No. . . .
COMMISSION: [June 9 letter read into record.] Did you write this letter? It has your signature on it.
BRUDER: Can I see it, please?
COMMISSION: So, you misinformed the Florida supervisors of elections that race would be used as a matching criterion?
BRUDER: Yes.

Wise answer, Mr. Bruder. Misleading elections officials is not a crime; perjury would be. He pleaded confusion. So if race was not a match criterion, how did Black people get matched to felons? I was perplexed by this until I looked again at the decoded scrub sheets: There were columns for felon race and voter race. How could DBT deny that? [Click here to see the scrub list.] However, DBT had simply identified race for every real felon, and the secretary of state provided the race of the voters. It was left to the county supervisors to finish the Jim Crow operation: They would accept racial matches as "proof" that the right person was named. Therefore, a Black felon named Willie Whiting wiped out the registration of an innocent Willie Whiting (Black) but not the rights of an innocent Will Whiting (white).

Fair Use Notice: This site contains copyrighted material the use of which has not always been specifically authorized by the copyright owner. We are making such material available in our efforts to advance understanding of environmental, political, economic, democratic, domestic and international issues, etc. We believe this constitutes a 'fair use' of any such copyrighted material as provided for in section 107 of the US Copyright Law. In accordance with Title 17 U.S.C. Section 107, the material on this site is distributed without profit to those who have expressed a prior interest in receiving the included information for research and educational purposes. For more information go to: http://www.law.cornell.edu/uscode/17/107.shtml. If you wish to use copyrighted material from this site for purposes of your own that go beyond 'fair use', you must obtain permission from the copyright owner.