ad info

Editions | myCNN | Video | Audio | Headline News Brief | Feedback  





Bush signs order opening 'faith-based' charity office for business

Rescues continue 4 days after devastating India earthquake

DaimlerChrysler employees join rapidly swelling ranks of laid-off U.S. workers

Disney's is a goner


4:30pm ET, 4/16









CNN Websites
Networks image

Special Event

Florida Hearing: 0900-1212

Aired December 5, 2000 - 9:00 a.m. ET


BILL HEMMER, CNN ANCHOR: Again, Judge Sanders Sauls, back on the bench as we convene now day number two of the contest hearing in Leon County Circuit Court.

Now the judge.

JUDGE N. SANDERS SAULS: Good morning to you all. Madam Clerk, are we ready to proceed?

MADAM CLERK: Yes, sir.

SAULS: Madam Reporter?


SAULS: All right. Has everyone returned that we know? Is anybody missing? All right, then at this time the defendants will continue with their case. For what purpose does the gentleman arise?

UNIDENTIFIED MALE: Your Honor, (UNINTELLIGIBLE) the Vice President and Senator Lieberman.

Judge, we've been furnished today one-two-three-four-five-six- seven witnesses from the -- from Governor Bush, and there's also appears to be some more coming in a category from Broward County. We would like to ask how -- the plaintiff's case, Your Honor, we basically done our case subject to some potential rebuttal, which wouldn't be extensive. How much time are we going to give the defendants to put on this case? Is there any constraint on that?

SAULS: (UNINTELLIGIBLE) hopefully for their proper presentation but not over presentation. Do you wish me to look at the witness list? I haven't seen them, and how many did you indicate that -- who has called who? I don't know, we have various defendants, and they all have separate status and identities, and I don't know -- who are we referring to at this time? Which particular defendants? Is it the defendants of Bush?

UNIDENTIFIED MALE: Judge, I have -- we were given this, and I'm sorry, I can't answer any further ...

SAULS: Will you approach and let me just view that?

UNIDENTIFIED MALE: Yes, sir. SAULS: All right. Let you - make your way up here. Let's just see what that says. Let's see, one-two-three-four and then the second category, are those defendants -- are those witnesses of the defendant Miami Canvassing Board?

PHIL BECK, BUSH CAMPAIGN ATTORNEY: No, Your Honor, what that list is - is it's my handwritten version this morning of what we gave them yesterday except with fewer names on it, as we this evening -- or last evening trimmed down our list.

And, the first three or four, I think, I've got underneath my name, those are some of our experts and people who will be testifying about dimpled ballots and then the next, I think, three, with a question mark after the fourth, are witnesses that my colleague Irv Terrell (ph) will put on concerning what happened in Miami-Dade, and then we're also going to have, I believe, three, maybe four, witnesses on -- or we may, we haven't made a final decision, we're going to do that over lunch -- on whether we present testimony concerning what happened in Broward County, and that would be -- those witnesses would be presented by my partner Fred Bartlit, so we ...

SAULS: We won't discuss that over lunch, we want to try to complete your case, if we can, by this morning. Let's do our best.

BECK: Then we'll discuss that over the coffee break.

SAULS: Let me back up, apparently the second category here perhaps would be witnesses that are - perhaps in common, or would have factual testimony concerning the defendant Miami - or the Dade County Canvassing Board, I assume, but let's look at the first four now; the experts. We've had two experts presented. We've had, I suppose, one expert in ...

BECK: ... rubber ...

SAULS: ... machines, and what have you, and the functioning of machines. We've had one statistician; I don't purpose to have multiple statisticians ...

BECK: ... I could tell you exactly what each one is ...

SAULS: Do so.

UNIDENTIFIED MALE: The first one, I believe, on the list is Dr. Laurentius Marais. He is our statistician, and he's our only statistician.

SAULS: All right.

BECK: The second one on the list is John Ahmann (ph). Mr. Ahmann is -- has several patents on the Votomatic, and he is our machine person to counteract what their political scientist said. The third one on our list is a Mr. Rohloff (ph), he's not an expert, he's an individual who put the stylus in the hole over Al Gore's name, thought about whether he really wanted to carry through with that and vote for Mr. Gore, and decided he didn't. SAULS: All right. All right.

BECK: And then the forth is Mr. Burkhardt (ph). Mr. Burkhardt was an observer at Broward County during the voting, and he will testify about a kind of a narrow issue, what's called "over votes", and that is that while the Broward people were counting dimples as votes, if somebody had punched out a chad for one of the nominees, and then there was a dimple or an impression for another one of the candidates, they were not counting that second dimple as a (UNINTELLIGIBLE) ...

SAULS: (UNINTELLIGIBLE) the last two are relatively ...

BECK: ... very short ...

SAULS: ... limited witnesses and then we have basically, I suppose, two experts. It looks satisfactory at this point. I'm not going to have them using multiple experts, but take that back, see if you can read it -- kind of hard to read, but at least ...


SAULS: ... for the clarification, so let's proceed.

Call the next witness.

BECK: Governor Bush and Dick Cheney call as their next witness, Dr. Laurentius Marais.

SAULS: Ask him to come forward and be sworn.

Are you going to designate Ms. Joyce (ph) to swear this witness in?


SAULS: Ms. Joyce is going to swear the witness? (UNINTELLIGIBLE). You want to swear the witness?

UNIDENTIFIED MALE: Raise your right hand please. Do you solemnly swear that the evidence that you will give on this issue should be the truth, the whole truth, and nothing but the truth, so help you God?


SAULS: Be seated, sir.

MARAIS: Thank you.

SAULS: You may proceed.

BECK: Dr. Marais, would you please state your name for the record.

SAULS: And spell your last name for the reporter, sir. MARAIS: Laurentius Marais, m-a-r-a-i-s.

UNIDENTIFIED MALE: You may want to spell your first name for the court reporter as well.

MARAIS: L-a-u-r-e-n-t-i-u-s.

BECK: Dr. Marais, what do you do for a living?

MARAIS: I'm a professional statistician and applied mathematician in a consulting practice.

BECK: What is the name of the firm that you work with?

MARAIS: William E. Wecker (ph) Associates, Incorporated.

BECK: Have you been asked by us to come here for this hearing and testify about some of the statistical presentations that have been made by the plaintiffs and their implications in this case?

MARAIS: Yes, I have.

BECK: Before we get to those, please describe briefly for the court, since we are trying to compress this hearing as much as we can, your educational and professional background that enables you to give this testimony.

MARAIS: I have a Bachelor's degree in mathematics, applied mathematics and computer science. I have Master's degrees; one on mathematics, one in statistics from Stanford University in California. I have a Doctorate in business administration and mathematics in computer-intensive statistical methods from Stanford University in California. I've served on the faculties of the University of Chicago and also on the faculty of Stanford University in California.

BECK: Dr. Marais, are you -- in your review of the materials in this case, did you look at the complaint that was filed by the Gore legal team?

MARAIS: I did.

BECK: And, are you aware sir, that in the complaint that they have filed, they have an allegation that, based on the results of the partial hand count in Miami, where 20 percent of the precincts approximately were counted, that if you project the result of that partial hand count over the entire county, then Vice President Gore could pick up approximately 600 votes.

MARAIS: I am aware of that.

BECK: From a statistical point of view, is the allegation made by the Gore legal team a valid one?

UNIDENTIFIED MALE: Objection Your Honor; foundation.

SAULS: Well, I believe you can develop questions of foundation and the basis for the opinion on cross, so in Florida we are permitted to express an opinion and support it and be cross-examined on it.

Go ahead.

BECK: Is the allegation that the Gore legal team made about how you can take the results from that partial hand count in 20 percent of the precincts and then just do the multiplication -- you know, multiply it by five -- to get the votes they claim they'd pick up for the entire county, is that a valid statistical approach?

UNIDENTIFIED MALE: Objection Your Honor. That's not what the intention is of the complaint.

SAULS: An (UNINTELLIGIBLE) form of the question. Restate the question.

BECK: Let me get a copy of the complaint and that (UNINTELLIGIBLE).


UNIDENTIFIED MALE: Greta, we are off to another start here. Quickly, tell us what we're going through right now with the statistician on the stand.

GRETA VAN SUSTEREN: We are. What was happening here is that the lawyers are being held to the strict words of the complaint, and the Gore people are objecting because the Bush lawyer has strained a little bit from the theory that the Gore team claims it presented in it's complaint. So the Bush lawyer's going back to the words in the complaint, which starts the case, to make sure he does exactly as the theory presented in the Gore complaint.

UNIDENTIFIED MALE: And you heard the first question from Dino Kitchen (ph), the Gore attorney, asking about the length of the Bush witness list; a concern, again, about time in this case.

Back to circuit court now, here is Phil Beck again, the Bush attorney.

BECK: I've got to apologize to all concerned that the only copy I have has some underlining that I'd ask you to ignore, Dr. Marais.

SAULS: Why don't you just read from the complaint (UNINTELLIGIBLE).

BECK: Sure.

"The election canvassing board" -- this is paragraph three -- "certified" -- it's talking now about Miami-Dade -- "certified 2,912,790 votes for George Bush and Richard Cheney and 2,912,253 votes for Al Gore and Joe Lieberman, a difference of 537 votes. The difference was entirely the result of -- and then it starts going through and we get to paragraph D, which -- I'm sorry the earlier numbers was for state wide -- and then we get to paragraph 'D and it said, "not counting approximately 9,000 ballots in Miami-Dade County that have not been recorded as a result - as a vote for any presidential candidate, and which were never counted manually, because the Miami-Dade Canvassing Board prematurely ceased its manual count, with only approximately 20 percent of the precincts counted."

And now this sentence, Dr. Marais, "If these approximately 9,000 uncounted ballots result in the same proportional increase in net votes as the ballots that were counted by the board before it stopped counting, those -- or these ballots would result in approximately 600 net additional votes for Gore/Lieberman."

So, I'm focusing on that last sentence about if they used the same proportions that were used in the partial recount, they claim they'd get another 600 votes. OK?


BECK: Now, is that approach that is laid out in the Gore/Lieberman complaint a valid one from a statistical point of view?

MARAIS: It is not, and it makes for an unreliable and inaccurate projection, because it is based on a false premise.

BECK: What is the false premise?

MARAIS: The false premise is stated in the portion of the text that you read, that is if the proportion of net votes gained by the Gore side were the same in the remaining approximately 9,000, then a certain result would follow, but to interpret that and assess it, one needs to know that the portion of the precincts that were recounted by hand were heavily Democratic, in fact, those are precincts in which Gore won over Bush by a margin of greater than 75 percent to 25 percent.

That is very different from the remaining precincts where, in fact, in the underlying machine recount totals Gore -- Bush beats Gore by a margin of about 52 to 48 percent. So, in other words, where the recount -- the manual recount was done, was in heavily Democratic precincts with a very large preponderance for Gore, and it is -- there is no foundation -- there is no basis for projecting from those precincts into the remaining precincts where, in fact, Bush wins.

BECK: Thank you.

One of the charts that was shown, I don't know if they're still here, but I've got one from the handouts that they gave us yesterday.

Oh great. Maybe I could put up one of these ...

SAULS: Just don't put in front of (UNINTELLIGIBLE).


UNIDENTIFIED MALE: I'll put it up next to (UNINTELLIGIBLE). Work together with (UNINTELLIGIBLE).

UNIDENTIFIED MALE: Clearly, the Bush attorney's at this time trying to undercut the Gore argument about that sample recount that was done in Miami-Dade. A recount that did show the Vice President gaining votes, however, clearly, at this time the witness on the stand is making the case that it was only done in heavily Democratic precincts, and one cannot project over an entire county that that number would play out the way the Gore team alleges.

Back inside again, Phil Beck now with a chart.

BECK: And you see here the point they made was that in Broward, when they did the manual recount, they turned 26 percent of the non- votes into votes. You understand that from this chart?


BECK: And in Miami-Dade, as to the Democratic portions of the county where they did the recount, they turned 22 percent of the non- votes into votes. Do you see that?


BECK: And in Palm Beach when they did the recount -- and were you here yesterday when the witness from Palm Beach described the standards they applied and all the procedures they went through in their recount?

MARAIS: (UNINTELLIGIBLE) yes, I was here, sir.

BECK: And when they did that recount in Palm Beach, they only turned eight percent of the non-votes into votes. Is that how you understand this chart?


BECK: Now, I ask you from a statistical point of view, is there any valid -- any validity to the contention that just because Broward and Miami turned more non-votes into votes that they must have done a better job, or applied better standards in the recounts that they did, versus the recount that was done in Palm Beach?

MARAIS: Absolutely non that I have seen or heard in the testimony for which I was present yesterday.

BECK: And I take it, as a statistician, it's not within your realm of expertise to pass on which county did the best job in applying the standards in the recount, is that right?

MARAIS: No, sir. That is correct.

BECK: But from this chart, you can't draw any conclusions just because they turn more votes -- more non-votes into votes that somehow they did a better job or applied a better standard.

MARAIS: Not if you're assessing it from the perspective of a professional statistician.

BECK: The -- you were here yesterday when Dr. Hengartner (ph) talked about how with Votomatic machines, according to his numbers, there is higher undervotes than with Optical Scanning voting machines.


BECK: And he used the word "association". That there is an association between Votomatic machines and a higher number of undervotes. Do you remember that?


BECK: And I fumbled around with him trying to ask some questions about the difference between association and causation, let me take another try at it with you. Would you please describe to someone in statistics, what's the difference between saying that two things are associated, and saying that one thing causes the other.

MARAIS: Best way for me to do that is by example, if I may.

BECK: Yes, please do.

MARAIS: An example that occurs to me is ...

BECK: Please use the stork example, if you would, so that people know what I was talking about yesterday. (UNINTELLIGIBLE).


MARAIS: I will use the stork example briefly, if I may. The stork example refers to a commonly used textbook example illustration of counts of the number of stork's nests unobserved (ph) and counted on the roofs of houses in a particular region of Europe. The region varies from textbook to textbook. I think yesterday we heard it was Poland, I've also seen it as being Germany ...

BECK: But never London.

MARAIS: Never London, no, not London.

The other element of information in that illustration is the number of births in each of those years, and what the textbook illustrations like to do is to create a graph that a statistician would call a "scatter plot."

On which, in the vertical axis, that's the, what is sometimes called the Y axis, they plot the number of births and on the horizontal axis, sometimes called the X axis, they plot the number of stork's nests. So, in a given year, if 55 storks nests were counted and 75 births, there would be one point -- one dot in that graph at a position corresponding to 55 stork's nests and 75 births, and then you do that over again for some other years, and you notice that, amazingly or interestingly, those points tend to fall on a line. It looks like in years where there are more stork's nests, there are also more births.

Now, that is an association, and if one were to apply the kind of statistical test that Professor Hengartner applied and talked about yesterday to those data, you would discover that there is what would be called a highly statistically significant association between those two variables. If you applied a test like that and from that you could make a statement like the one that Professor Hengartner made, that this could not happen by pure chance anymore frequently than one would win the lottery, or anymore frequently than one person would be struck by lightning five times, I think, were the two illustrations.

But that does not prove that stork's nests or storks cause or bring babies to town -- cause births or bring babies to town, in that case, the explanation for both of those numbers is a slow increase in population, an increase in the population of the town accounting for a greater number of births with the passage of time over a 10 or 20 year period, and an increase in the number of storks in town for reasons that I don't know, but that is coincidence with the increase in births. So that is the storks and babies example.

BECK: Do you have any shorter examples used to illustrate this same point?


MARAIS: Yes, my shorter example has to do with the seatbelt sign in the -- in an airliner, like the one on which I came from California to here.

People who fly may have observed that frequently after the seatbelt sign comes on in flight, there is some turbulence. And if one were to do a statistical test, it would appear to be a highly statistically significant association between those two variables, but it is not the case that the seatbelt sign turning on causes the turbulence outside the airplane, even though there is a strong and statistically significant association that is very unlikely to be explained by pure chance.

BECK: I want to share with you here what Dr. Hengartner said about whether the information that he displayed about this association whether in fact it proves that the -- that there was something wrong with the Votomatic machines and people were trying to vote, but the Votomatic machines were failing to record their vote, and then I'm going to ask you to comment on the analysis.

Excuse me for one minute, Your Honor.

HEMMER: David Cardwell here in Tallahassee, certainly the testimony that they're trying to get at right now is the whole issue of voter intent.

DAVID CARDWELL, CNN ELECTION LAW ANALYST: Right. What the Bush campaign is doing here is not disputing the numbers that were presented yesterday by the Gore statistician, but rather the conclusions that he reached stemming from those numbers.

HEMMER: The interpretation.

CARDWELL: The interpretation, right.

HEMMER: Thank you. BECK: Now, that's a little faded back there.

Can you see on the screen where we are here on this transcript? This was page 46 of the transcript of Dr. Hengartner yesterday.


BECK: And he begins -- actually I begin at line two asking the questions. "Let me -- we're going around and around on this -- I take it you have no opinion that you're here expressing today about why there is this variance between the undervote and the Votomatic machines versus the undervote and Optical Scanning machines." And he answered, "I have many opinions, but I have no proof."

And then I ask, "Well, I want to make sure that I didn't miss something. Did you express any of those opinions so far?" Answer, "No." Question, "So, if you have any opinions on why that happens, you're keeping it to yourself for today's hearing?" Answer, "Because I don't have the data to support that."

Now, Dr. Marais, do you agree with the statistician that was brought in by the Gore legal team that there just is not enough information to draw any conclusions about why there is this association between higher undervotes and the use of the Votomatic machine?

MARAIS: I do agree. There is not enough information that I have seen or heard in this courtroom to draw any such conclusion.

BECK: I'd like to now explore with you what a statistician would like to look at if he had -- how long ago were you contacted to get involved in this proceeding? MARAIS: About ...


MARAIS: Approximately nine days.

BECK: OK. Now, I'd like to have you share with the court the kind of analysis that would be necessary from a statistical point of view before anyone could draw any valid conclusions about why there seem to be more undervotes with Votomatic machines versus Optical Scanning machines, and first thing I want to ask is - did you take a look at the different counties that use Votomatic machines and look at whether they all have pretty much the same level of undervotes?

MARAIS: I did.

BECK: And the only statistics we saw yesterday from the fellow brought in by the Gore legal team was this kind of average for all the Votomatic counties, what did you find when you looked at the individual counties?

MARAIS: Well, I found that the individual counties have a lot of -- that among the Votomatic counties, which were represented in the exhibit shown to the court yesterday by a single bar showing the average undervote percent in those counties ...

BECK: Let me just get that, so the judge knows what chart we're talking about. I'll put it through, and you tell me when to stop when we get to the one that you're talking about.

MARAIS: That looks like the one.

BECK: OK. I'll put this one up on the easel.

OK, what did -- now continue on, what was your observation when you looked at the information for each one of the counties rather than this average.

MARAIS: I could do a better job at this if I could get up and point ...

SAULS: You may.

MARAIS: Thank you, Your Honor.


MARAIS: My observation, Mr. Beck, is this that this drop kind of creates the impression that this is what is going on in punchcard counties as they are labeled here - "punchcard", and that this is what is going on in the optical ballot counties, but .3 percent is not what is happening in each of the separate optical ballot counties, it is an average, and the numbers are being averaged there vary quite widely.

Moreover, 1.5 percent is not what is happening in each of the punchcard counties, that is an average, and the numbers that are being averaged vary quite widely. In fact, the variation around this average and the variation around this average is so great that there is a substantial overlap between the optical and the punchcard counties, in fact, there are about a dozen of the 67 counties in Florida that fall in that overlap range.

Moreover, looking at the punchcard counties on their own, using this same kind of statistical test that Professor Hengartner used to conclude that there is a statistically significant difference between the punchcard counties and the optical counties that applying that test instead to the punchcard counties alone to answer the question of whether they all really have this same rate of undervote, or whether there is difference -- whether there are differences among the punchcard counties, there is a highly statistically significant difference among the punchcard counties themselves, and, in fact, it is just about as highly statistically significant as the difference that Professor Hengartner points to between the punchcard counties and the optical counties.

In other words, there are other factors accounting for differences among counties, because not all punchcard counties have the same rate of undervotes.

BECK: So there's something going on in the different counties, something - or more than one thing, other than the fact that they've got punchcard ballots. Is that right?


BECK: Let's just look up there. They've emphasized that Palm Beach has 2.2 percent for the undervote for this last Presidential election, and did they present any data at all about Palm Beach's undervotes in prior elections to see whether there's something wrong with the machines that keep showing up year after year in Palm Beach, or whether there might have been something peculiar going on in Palm Beach this year that caused the undervotes for President?

MARAIS: Yes. They presented, or Professor Hengartner had an analysis for Palm Beach results of column one versus column two ...

BECK: Well, we'll come back to that, but did they go back in historical data and look at Palm Beach versus the other punchcard counties? Did they present anything like that yesterday?

MARAIS: Not when I -- while I was in the room, sir.

BECK: Now, from a statistician's point of view, if you wanted to do a meaningful analysis of why Palm Beach may have been -- I suppose that there were other counties, actually, with punchcards that were up in this same range. Is that ...

MARAIS: Yes that is correct.

BECK: OK, but they're at least (ph) above the average, right, from the number of undervotes. If you wanted to do a statistically sound analysis, would you just accept the hypothesis that there's -- that in Palm Beach the rubber on the left side of the ballot gets hard, and they're not able to punch the chads through. Is that the way that a statistician would go about the analysis?

MARAIS: No, sir.

BECK: Would a statistician trying to analyze this look to see whether there might be anything else unusual in Palm Beach during the Presidential election that could account for the fact that people just didn't vote as often in the Presidential election as some other people did elsewhere in the county?

MARAIS: Certainly, yes.

BECK: Now, I know that you haven't had time to do a thorough investigation, but I think you learned just by reading the newspaper and watching TV over the last several weeks that there was quite a controversy in Palm Beach concerning the use of what's called a "butterfly ballot".

MARAIS: I am aware of that.

BECK: And if a statistician were trying to do a thorough analysis of this problem, would you look at the fact that in that same Presidential ballot where there appear to have been people who didn't vote at all, there were also a whole bunch of people who claimed that they were so confused by the ballot that they ended up voting for two people?

MARAIS: It's one of the things one would want to consider.

BECK: And would you also want to consider, if you were going to do a thorough investigation, that there were other people who claimed they were by the ballot, and they ended up voting for the wrong guy?

MARAIS: That also would suggest another explanation to be investigated, yes, sir.

BECK: And so would you want to investigate whether there is a possibility that with that same butterfly ballot that people complained about, some people might have just thrown their hands up in the air and said I can't figure this out, I'm not going to vote for any of these guys.

MARAIS: It would be a facet to consider.

BECK: The -- let's just stand on this a little bit. I want to focus now on how a statistician doing a thorough analysis would differentiate between the association and the question of causation. What causes this phenomenon where people using the punchcard system might have more undervotes than people using the Optical Scanner? Is there a area of academic study that looks at whether the meanings by which you perform a task affects the way in which you perform that task?


BECK: I'll just take this chart down since we're done with it. What is the name of that field of study where people look at what I was talking about?

MARAIS: The type of study that you are talk -- that you are referring to goes by a number of names, depending on the particular area of study. In the behavioral theory of decision-making, it is sometimes referred to as "framing", and it's known that the way a question is framed or a task is framed influences, even if objectively speaking the task is otherwise identical, does influence the way people respond. In the Political Science arena, it has been studied using statistical methods; there have been statistical studies of the way in which different methods of voting influence voter behavior, including voter participation.

BECK: Before we get to that, just back on a more general subject, since we're in a lawsuit involving political elections, sometimes we hear that the way you ask a question in a public opinion poll can influence the way that the answers come out. In other words, you could have the same basic idea, but the way that you frame the question and the place that you put the question in your list of questions ends up causing a big difference in the number of responses. Is that the sort of thing that you're talking about?

MARAIS: It's exactly the sort of thing I'm talking about. Yes, sir. BECK: And you're saying that people have actually studied whether the type of method for voting, whether it's, you know, checking a box or filling a little whole or punching a card, that influences voter behavior?


BECK: And does it influence how many people show up to the polls and vote?


BECK: And does it also influence how many people, once they get inside the polling booth, actually vote for every office, rather than skipping an occasional office?

MARAIS: Yes, there is a well-known phenomenon referring to that effect called "ballot roll-off", and I believe I heard about it from Mr. Brace (ph) in this courtroom yesterday.

BECK: Mr. Brace referred to it, I think, in terms and suggested that it's only the folks at the end of the ballot that might get left off. In fact, might people be left off at the front of the ballot and in the middle of the ballot as well?


BECK: What are a couple of the academic studies that have looked at this?

MARAIS: One that I have reviewed quite recently was published from the University of Iowa, I believe by the Iowa University Press, in 1986, and I know it by the name of "The Iowa Voter Study," although I'm not sure that that is its formal title, by a scholar named Mather, m-a-t-h-e-r.

BECK: Let me just see if I can get these two studies.

BILL HEMMER: By now, so many of our viewers are quite familiar with the terms we're hearing in court today; butterfly ballots and undervotes, and even the Votomatic came up again today after so much testimony yesterday.

BECK: May I approach, Your Honor?

SAULS: You may.

BILL HEMMER: Phil Beck handing off documents now to David Boies, the lead attorney for the Gore team.

BECK: (UNINTELLIGIBLE) the study that you just referred to, back on the right (ph)?


BECK: And what is the title of this study? MARAIS: "Lost votes: effects of methods of voting on voter participation in Iowa 1920 through 1984."

BECK: And, since we are a bit pressed for time, I'm not going to ask you to walk through page by page in the study, but when you prepared yourself to testify here, were there any, sort of, general conclusions that you drew when you reviewed this academic piece?

MARAIS: This academic piece has 110 pages, but one of the general conclusions was that it documents that the method of voting affects voter participation or influences voter participation, both at the level of how many voters go to the polls and among those voters who go to the polls, the extent to vote to which they vote all the offices and initiatives on the ballot.

BECK: I'm going to hand you another study, I don't know if we've assigned a number to it yet, but we will be. This'll be exhibit 59.

SAULS: What was the prior one (ph)?

BECK: Sorry?

SAULS: What was the prior one (UNINTELLIGIBLE)?

BECK: The prior one, Your Honor, was Defendants exhibit 36.

SAULS: All right (ph).

BECK: This is Defendant's exhibit 59, and we'll put a mark on that and provide one for the clerk later, Your Honor.

What is Defendant's exhibit 59?

MARAIS: It is a journal article from the spring of 1998 by an author named Nichols (ph), and it is entitled "State Referendum Voting, Ballot Roll-Off and the Effect of New Electoral Technology."

BECK: And again, without going through page by page, what was helpful to you in reviewing this study?

MARAIS: A statistical documentation of the effect of electoral technology. In other words, voting method on the participation of voters in an election, including voter choices to vote all or not all of the offices on the ballot.

BECK: Now, you've reviewed these two statistical -- or the one statistical study and the other one, the Iowa study, and if someone was going to do an actual analysis of what accounts for the difference in the undervotes, would you want to look at issues such as - with Optical Scanners you're filling out a ballot that has the names next to the holes, and you fill it out with a number two pencil, that kind of thing?

MARAIS: Right. That would be how the kinds of issues you asked about earlier, framing issues, would show up in this context. There are different levels of user friendliness of different methods of voting. Some of them involve procedures that are quite foreign and strange to the kind of things we do in everyday life.

BECK: For example, in everyday life, we don't use little styluses with plastic handles that have little chains attached to the table that we're writing on, correct?

MARAIS: I do not in my everyday life.

BECK: But lots of us use pencils, and that's a more familiar, user friendly ...

MARAIS: That is a familiar thing to do.

BECK: And when we, in our everyday life, when we're writing something, we generally have the entire page in front of us, and we can see what we've done, right?

MARAIS: Exactly, so you get some visual feedback from the page in front of you, and you see all of what you've done at once in one frame, as it were.

BECK: And the things (ph) like the Cognitive Psychologist do they tell us that if there's something left blank on that page and you've filled in the rest of the dots, it kind of pulls you in and says to you psychologically, you know, let's fill in all the blanks here.

MARAIS: It would not surprise me if they said that, but I'm not a Cognitive Psychologist.

BECK: And you have looked at the punchcard machines and the ballots, do you get the same kind of visual feedback that you get when you have a Optical Scanning ballot that you fill out.

MARAIS: You do not.

BECK: And, in fact, if you -- if somebody was undecided about who they wanted to vote for for President, one of those people that make up their minds at the last minute, and they decided to vote in the rest of the races first, while they kind let (ph) this make their last final decision, if you did that with the Optical Scanning ballot, if you glanced down on it, would you get a reminder that you still had to come back and vote for President?

UNIDENTIFIED MALE: Objection, Your Honor, he's already said he's not an expert in this area.

BECK: I'm talking about the ballots that he's examined here, Your Honor.

UNIDENTIFIED MALE: He still has no expertise in cognitive psychology.

SAULS: Sustain the objection.

BECK: Well, we'll just leave that, and the lawyers can talk about it later. We talked about the question of user friendliness. You know, you're not an expert in cognitive psychology, but, as a statistician, before you drew any conclusions about why there may be more undervotes with one kind of a voting method versus another, would you want to make sure that someone who did know what they were talking about in cognitive psychology looked at those two methods to see if one was just more user friendly than the other and could explain the fact that people just don't vote for all the offices in one, more than they don't vote for all the offices in the other? Would you want to look at that?

MARAIS: Absolutely. And it could even be studied by a properly designed experiment.

BECK: Oh, it could be studied by a properly designed statistical experiment?


BECK: If you were designing such a statistical experiment, would you want to investigate whether one explanation for undervotes for President in the punchcard ballot could be that people -- it's possible for people to walk in and open that ballot and just flip it open by accident to page two and start voting in the Senators and never even open the page to one?

MARAIS: It's a factor that you would want to consider in the design of such an experiment.

BECK: And would you want to design into the experiment the difference between a ballot like that where you might not ever even notice that you didn't vote for President and one where it's staring you in the face, because it's a punchcard, and you can see that you didn't fill in -- I'm sorry it's a thing that you fill in with the pencils, so it's staring you in the face, and you can see that you didn't fill in one of those ovals with the number two pencil?

MARAIS: It is certainly one of the differences between the situations, all which you would want to control if you were going to attempt to draw a valid inference.

BECK: Now, let's move from the user friendliness to the possibility that I keep saying the Gore legal team has raised, but I'll just state it as a hypothetical to you. That we have this difference between the undervotes in Votomatics and Optical Scanning, and I want you to assume that someone is claiming that the explanation for that difference is that there's something wrong with the Votomatic machines, especially the left hand column where the rubber's gotten hard, or the plastic's gotten scratched, or there's chads that have built up into little mountains underneath there, and people are trying to level that to vote for Al Gore, but they just can't push the stylus all the way through and, instead, they're leaving these little dimples.

Now, if that were a hypothesis for the difference in the undervotes, as a statistician, would you want to subject that hypothesis to scientific testing to see whether there's anything to that? MARAIS: Yes.

BECK: Before you drew an conclusions based on a hypothesis like that, would you want to hear from the people who were proposing that explanation from somebody who understands rubber and can tell you whether rubber gets harder the more times it's impacted with a stylus?

MARAIS: I would want to do that.

BECK: Would you want to hear from somebody who actually works with the machines and designs the machines rather than a Political Scientist about whether little mounds of chad can build up and stop somebody from puncturing the ballot and voting?

MARAIS: I would want to learn that, because there is -- it serves no useful purpose to try to learn from statistics alone about a situation where there a lot of subject-matter-specific information that can be brought to bear on the question that is being examined.

BECK: And, as a statistician, would you want to look -- and we referred to this earlier in your examination, and I asked you to defer your discussion of it -- but would you want to look to see whether there was a pattern from prior elections where people in the left hand column voted fewer times than people in the right hand column, and then the -- than they did in the right hand column, and then the -- and in the middle of the ballot?

MARAIS: Yes. Particularly if I had encountered anecdotal evidence that something like that could be discerned.

BECK: And of course, you've read the affidavit and the proper proof concerning Professor Hengartner, have you not?

MARAIS: I have.

BECK: And you read the paragraphs in both those documents concerning the 1998 election in Florida, right?

MARAIS: I did.

BECK: And were you here yesterday while he testified about that?


BECK: And, without going through all the details, do you generally recall that in his sworn affidavit and in the proper proof (ph) that was submitted by the Gore legal team, they claimed that the ballot in 1998 had the Senators on the far left column and then had the Governors race in the second?


BECK: And we saw yesterday that that was wrong, that both those were in the left hand column. Remember that?

MARAIS: Yes. BECK: Now, did you take a look at the number of votes that were cast for the different races that were in the left hand column on that 1998 ballot and the number of votes that were cast in the races that were in the number two column to the right ...

MARAIS: I did.

BECK: ... in the 1998 election. And that is our exhibit 60, which I'll put up on the screen.

Now, and I'll represent to the court -- we can put the sample ballot up if somebody wants, but I'll represent to the court that the sample ballot actually had in column one the race for Senator, and then the race for Governor, the race for Secretary of State, and the race for Attorney General. Have you checked the sample ballot? Do you know that to be the case?

MARAIS: Yes, I have scrutinized it closely.

BECK: So you actually did have a close observation of the sample ballot.

MARAIS: Yes, I did.

BECK: OK. And then in column two, there were races for State Controller, State Treasurer, State Commissioner of Education and State Commissioner of Agriculture, right?


BECK: Now, let's take a look here. Here in column one for Senator, how many votes total were cast for the two candidates?

MARAIS: Three hundred thousand four hundred and twenty-six.

BECK: And over here in column two on the top race, how many votes were cast for the Controller?

MARAIS: Three hundred two thousand five hundred and fifty-seven.

BECK: Now, can you draw any conclusion on that that said, gee whiz, people were trying to vote over here in column one, but they couldn't do it, because they couldn't punch the chads through?

MARAIS: That's not a conclusion I would draw.

BECK: Well, let's look down at the next race. For Governor, what were the total numbers?

MARAIS: Three hundred ten thousand nine hundred and twenty-five.

BECK: And then over on the second race in column two, what was the total number?

MARAIS: Three hundred nine thousand one hundred and four. BECK: So there were, for the first one, a little bit more votes were cast in column two, and for the second set of races a little bit more votes were cast in column one, right?

MARAIS: That is correct.

BECK: Let's move to the third race. Secretary of State in column one, how many votes were cast?

MARAIS: Three hundred and four thousand seven hundred and eighty-six.

BECK: And then, correspondingly, the third race in column two, State Commissioner of Education?

MARAIS: Three hundred and seven thousand and seventy-two.

BECK: So, this time column two's got a little bit more votes for the third spot, right?

MARAIS: Column two wins.

BECK: Then let's go to the fourth race, Attorney General, how many votes were cast?

MARAIS: Three hundred eight thousand five hundred and twenty- three.

BECK: And then State Commissioner of Agriculture, the fourth race in column two.

MARAIS: Three hundred two thousand seven hundred and forty-two.

BECK: So recognizing, Dr. Marais, that looking at just one prior election may not be the greatest analysis in the world, would you agree with that?

MARAIS: I would agree with that.

BECK: But it is the analysis that they did. Right?

UNIDENTIFIED MALE: Objection Your Honor. It is not the analysis that we did.

SAULS: You may cross examine.

BECK: It's the analysis that they offered to this court and they submitted other (UNINTELLIGIBLE) in the sworn affidavit from the statistician; let me put it that way.

UNIDENTIFIED MALE: Objection, Your Honor. Again, that misstates it, and Counsel is continuing to lead -- I've been patient on that, but I do object his leading the witness.

SAULS: Well, he's an expert, and I don't believe he's going to confuse the expert. You have some latitude to lead your experts. Overrule the objection. Go ahead. Rephrase your question (UNINTELLIGIBLE).

BECK: Doing the analysis that Professor Hengartner put forth in his sworn affidavit, in which the Gore legal team included in a (UNINTELLIGIBLE) that was submitted to this court and the Florida Supreme Court, when you look at the ballot as it actually existed, and you look at the vote totals as they actually existed for the eight races in column one and column two, is there any statistical support whatsoever for the claim that there's something wrong in column one, and people are trying to vote to their level (ph) best, but they're unable to punch the chads through?

MARAIS: On the basis of what we're looking at here, absolutely not.

BECK: Now, the -- another possibility -- is another possibility that a statistician would want to consider for this difference between the undervotes with the Votomatics versus the undervotes with the Optical Scanners that there might be demographic differences in the people who are using the two different kinds of machines?

MARAIS: It is certainly another possibility that a statistician would want to consider and analyze.

BECK: And, in fact, in the last couple of weeks, have you seen Vice President Gore going on television saying that it's a shame because the Vote-o-matic machines are older machines and they're in poorer areas and the optical scanning machines are newer and more expensive machines and they are in wealthier areas.

MARAIS: I have seen that, yes.

BECK: And I don't know whether Vice President Gore's right or wrong about that, but if he's right, if what he says is true, then the fact that one type of machine might be used with people with one level of education, one level of wealth, one level of voting past participation and another kind of machine would be used by people with a different level of education, different demographics. Is that the kind of thing statisticians need to look at before they draw any conclusions?

MARAIS: That's the kind of thing a statistician would need to look at before concluding that it is the machines that create the effect that is observed.

BECK: And did you look at the individual counties to see whether in fact there were some demographic differences that might be significant to a statistician before you leaped to the conclusion that it's hard rubber and chad build up that's causing this differential?

MARAIS: I did.

BECK: And what did you see when you did that kind of analysis?

MARAIS: Well, you understand, of course, that I've been on this project for nine days.

BECK: Yes.

MARAIS: And so this is a very preliminary look. But there are demographic difference between the counties. And the Vote-o-matic machines are not, for example, used predominantly, from that look, in poorer counties.

BECK: You say they are not?

MARAIS: They are not.

BECK: So Vice President Gore is just wrong about that?

MARAIS: Yes. The vice president was wrong about that.


MARAIS: And for that matter Professor Hengartner does not, I think, from his testimony disagree with that.

BECK: Well, let me read it and ask you whether you agree with what Professor Hengartner said on this. This is at page 32 of his transcript, where he says, "I said if I look at voter demographics like racial profile of the counties, which is the appropriate aggregation in which I have the data, then they can't explain all that difference," referring to the difference between Vote-o-matics and an optical scanning.

Do you agree with Professor Hengartner that, based on your preliminary look at the demographic data, they cannot explain all of the difference between the two levels of under voting?

MARAIS: Based on my preliminary look, I do not have an independent opinion as to whether they can explain all of the under vote difference. But based on my preliminary look at the data there are demographic differences.

Since I've not seen all of Professor Hengartner's analysis, I cannot comment on his analysis in any detail, but I understand from his testimony that demographic differences in his analysis do explain some of the difference in the under vote rates.

BECK: And if somebody wanted to do a thorough analysis that you could draw statistically valid conclusions from about the reasons for these differences, would you want to do a more thorough look at the demographic data to see how much of a difference is just a accounted for by the kind of demographic data that you and Professor Hengartner accounts for some of this difference?

MARAIS: You certainly would. As Professor Hangartner testified yesterday, the available data at the county level instead of at the level of individual voters is not a very precise kind of data for studying that matter. And so one would not expect to learn the full extent to which other factors could explain a -- the difference in under vote rates from the kind of preliminary look that I and apparently Professor Hengartner have been able to take. BECK: Last question, I hope, Dr. Marais. From a statistical point of view, is there any valid basis for drawing the conclusion that people were in the Vote-o-matic voting booth trying to vote for Al Gore but they simply weren't able to push the stylus through the chad?

MARAIS: Absolutely none.

BECK: Thank you, sir.

SAULS: They need a minute to take that down.

BOIES: Maybe we could put that back up the picture that you had on the screen before.

BECK: You'll have to tell me which picture.

BOIES: I think it was the last one with the Palm Beach County charts.

SAULS: Sure.

HEMMER: At issue here is that under vote that is critical to the Gore team -- the proving why the ballot should be recounted again by hand. David Boies now with the cross-examination. The witness will stay there, Lawrencious Marais, a doctor, after just being questioned by Phil Beck.

BECK: No, no, no. I just have to find in my notes where it was so I what to plug in there.

HEMMER: Throughout the testimony we just heard, the witness on several occasions trying to make a point which would explain for an under vote. And clearly throughout this matter, Phil Beck is trying to make the point that voters don't necessarily vote for every office and every candidate when they go to the polls.

BOIES: Good morning.

MARAIS: Good morning.

BOIES: Did you prepare this chart?

MARAIS: Yes, sir.

BOIES: And what was the purpose that you meant to convey to the court about this chart? What were you trying to use this chart to prove?

MARAIS: I was using this chart to explore for my own understanding that the analysis that Professor Hangartner had described in his affidavit in this matter and in his proffer in this matter in which he drew certain inferences from a comparison of column one and column two in which he was misinformed apparently.

BOIES: Did this have anything to do with the testimony that Professor Hengartner gave in this courtroom yesterday? This chart? Or did it only relate to that affidavit and the proffer?

MARAIS: I heard, I believe, Professor Hengartner testify about column one and column two in this courtroom yesterday. Yes, sir.

BOIES: And what did you hear him testify to about column one and column two that you think this related to?

MARAIS: I believe I heard him testify about his proposition in his affidavit and in the proffer that there was a conclusion or an inference to be drawn from comparing the ratio of votes in column two to the votes in column one by the ballot represented by this chart.

BOIES: You were here in court when Professor Hengartner testified?

MARAIS: I was.

BOIES: And you know that what you are now talking about is when on cross-examination, Governor Bush's lawyer read him from a proffer and the affidavit. Correct, sir?

MARAIS: If I -- I do not have a clear recollection of the order of events. I remember Professor Hengartner testifying on that point.

BOIES: Do you remember whether it was on direct examination or whether it was simply when Governor Bush's lawyer was reading him something?

MARAIS: No, sir. That's the distinction that matters to lawyers but not so much to statisticians.

BOIES: Well, sir, in terms of whether you're a statistician or a professional witness, you testify a lot and deal with lawyers a lot, do you not, sir?

MARAIS: I have met a number of lawyers and, yes, I work with a number of lawyers.

BOIES: And you give testimony. For example you go into court and you testify. And that's basically your business -- consulting -- right?

MARAIS: Well, there are two propositions in your sentence. My business -- my line of work -- is statistics and applied mathematics in a consulting context and on occasion that calls for me to testify.

BOIES: Yes. And for example, you'll go into court and you will testify that a statistical analysis that some other statistician has put into evidence or that some other organization has published does not meet your standards as to a thorough scientific analysis. Correct?

MARAIS: If a matter involves such an analysis that I have assessed and that I have found does not meet a particular standard, that would be the subject of a testimony but not necessarily. There could be analyses that I would support. BOIES: Well, for example you testify that certain statistical analyses that link lead paint with injuries to children didn't meet your standard for statistical scientific analysis. Correct?

MARAIS: No, sir.

BOIES: You didn't do that?


BOIES: Well, sir. What did you testify?

PHIL BECK, BUSH CAMPAIGN ATTORNEY: Your honor, I'm going to object to the relevancy of this in any event. The time is short. Why is he quizzing him about other cases?

SAULS: Does it matter in some other case?

BOIES: Well, your honor, I think it goes to his credibility and I think it goes to way he performs statistical analyses.

BECK: Your honor, I think he's being -- the counsel's trying to tire because he testified in a case that Mr. Boies thinks would be unpopular with the public. And I don't think it has any relevance here.

SAULS: Sustain the objection. Irrelevant. Irrelevant -- what the testimony was in some other case.

BOIES: Let me -- let me inquire, your honor, whether and I . . .

SAULS: You asked him what did he testify to in some other case and it has absolutely no materiality or relevance in this case.

BOIES: Would the court permit a comparison of the way he approaches what is a scientific statistical analysis in other cases with the way he has done it here?

SAULS: Not a detailed so that we get involved in other cases. No, sir.

BACK: And, your honor, I'm going to ask that the court admonish Mr. Boies about his obligation not to go down that route unless he has a legitimate, held belief that there is a difference in methodologies because otherwise all he's doing is grandstanding and saying that this man testified in cases that Mr. Boies thinks will be unpopular with the public and that's not right.

BOIES: You honor, I'm not grandstanding.

SAULS: I understand. Go ahead and let's see where you're trying to go. But I'm not going to get off into other cases.

BOIES: Your honor, I will abandon the line of inquiry. I think I understand the court's rule. Let's go back to this exhibit here. Did you mean to imply by this exhibit that you could tell something from these statistics as to whether voters were more likely to vote in column one or column two?

MARAIS: No, sir. I meant by this exhibit to test a proposition that I understood was being advanced by Professor Hengartner in the present matter and to see whether he was right or whether he had got it wrong. It appears that he got it wrong.

BOIES: Would you agree with me that you cannot tell anything at all from this chart as to whether or not voters are more likely -- all other things being equal -- to vote in column one or column two.

MARAIS: It appears that...

BOIES: Could I have a yes or no than an explanation, your honor?

SAULS: Yes, even though he's an expert, he may certainly in detail explain his answer.

MARAIS: No, because in the context in which the chart is being offered there is already testimony from another expert in this matter that one can tell something from these data about likelihood of voting in column one and column two.

BOIES: And do you believe that this statistical chart that's in front of you here or behind you that's evidence that is contrary to that proposition.

MARAIS: Oh, but of course, sir. It is evidence that the expert that advanced that proposition couldn't even get column one and column two straight.

BOIES: And, sir, did you get them straight?

MARAIS: I think the ballot in Palm Beach got them straight.

BOIES: How many candidates were there in column two on the Palm Beach ballot?

SAULS: Which ballot are we referring to? This one or . . .

BOIES: November 3, 1998 Palm Beach County general election ballot.

MARAIS: I don't know that with precision. It's my recollection as I sit here that I have shown all of the candidates from column two but I'm not sure of it whether that is so or not and it doesn't matter a wit to the point that I make from this exhibit.

BOIES: Well, sir, did you look at the ballot yourself?

MARAIS: I did.

BOIES: And did you notice how many people were in column two?

MARAIS: I looked at the ballot and I must have noticed and I recall I think that I actually held up that ballot next to this exhibit to make sure that I got the columns right and to make sure that I got the offices right.

BOIES: And was this a paper ballot or punch card ballot or both.

MARAIS: This was a -- the -- the -- actually what I was looking at, I believe, was the sample ballot and so it's not a punchable voting medium itself. It is a list of candidates and offices with arrows pointing to presumably where the punch holes would be.

BOIES: And did you look at the punch card ballot at all?


BOIES: Did you compare the ratio of votes in column one and column two in Palm Beach? With the ratio of votes for the same offices in optical ballot counties.

MARAIS: No, sir.

BOIES: Is that something that you would want to do if you were doing what you called a thorough statistical analysis?

MARAIS: Not if I were responding to Professor Hengartner, who did no such thing.

BOIES: So, you would only do what the other person did? You wouldn't do what you think is a thorough statistical analysis?

MARAIS: That would depend on what I had been asked to do and in this matter, as I testified at the beginning of my testimony this morning, I had been asked to respond to statistical propositions raised by the Gore legal team.

BOIES: So you weren't asked to do a thorough statistical analysis?

MARAIS: I was asked to respond to statistical propositions raised by the Gore legal team thoroughly.

BOIES: Were you asked to do what you define as a thorough statistical analysis?

MARAIS: I was not asked to do a...

BOIES: May I get a yes or no answer and then an explanation, please?

SAULS: He just said he was not asked to. Go ahead and finish your answer.

MARAIS: Thank you, your honor. I was not asked to do an -- a statistical analysis of any kind independent of what I was responding to in propositions advanced by the Gore legal team.

BOIES: Let me see if I can understand what you're saying. You were asked to respond to Professor Hengartner, but you were not asked to do anything else. Is that what you are saying? MARAIS: No, sir.

BOIES: OK. Were you asked, either in an attempt to respond to Professor Hengartner or otherwise, were you asked as part of your work to do what you would consider to be a thorough statistical analysis?

MARAIS: When I am asked to do anything statistical such as responding...

BOIES: Your honor, may I get a yes or no and then an explanation? That's been the rule with other witnesses.

SAULS: Could you answer that with a yes or no?

MARAIS: I could, your honor, if I could hear that question again.

SAULS: State the question again. I'm not sure if it was phrased the way he could, but let's see.

MARAIS: At least I will try, your honor.

SAULS: All right. Well, if he asks so you can then try to do so.

MARAIS: I'll give it my best shot, sir.

UNIDENTIFIED FEMALE: The question is, "Were you asked either in an attempt to respond to Professor Hengartner or otherwise, were you asked as part of your work to do what you would consider to be a thorough statistical analysis?"

MARAIS: I would have to answer yes, and explain that what I mean by that yes is that not that anybody said, "Please do a thorough statistical analysis" but just that whenever I am asked to do a statistical analysis I take it that people are asking me to do that work thoroughly. I do not interpret a request for statistical analysis to be something slip shod and slap dash.

BOIES: Well, sir, you used the term "thorough statistical analysis" several times in your direct examination.

MARAIS: I don't recall it specifically but it's possible.

BOIES: And you said that in order to do a thorough statistical analysis, you would want to investigate other causes for the statistical association that you might see. Do you recall that?


BOIES: Did you do that, sir? Did you investigate these other causes?

MARAIS: I have not investigated these other causes because that answer was in the context of, "What would a statistician want to do in a full statistical analysis of this matter?" I've not been asked to do any such analysis and I've not attempted any such analysis.

BOIES: Do you distinguish between what you talked about on direct examination which was a thorough statistical analysis with what you now talk about as a full statistical analysis or are those the same thing?

MARAIS: No, sir. These things have to be understood in the context of what is being talked about. And the context of my work in this matter is a response to the statistical claims raised by the Gore legal team not an independent and elaborate study of voting patterns.

BOIES: Did you look at any election statistics for the November 3, 1998 elections outside of Palm Beach County?

MARAIS: I must have because I believe when consulting the source of the data used here, what one sees on the screen is all of the election results for 1998. But I've no specific recollection and nothing that I have said here this morning depends specifically on any other kind of data.

BOIES: My question just is whether you analyzed or looked at it?

MARAIS: I understood your question to be whether I looked at and when an item appears on my computer screen I see it.

BOIES: Other than sort of seeing it because it was on your computer screen, did you consider it at all?

MARAIS: Not that I recall.

BOIES: Did you look at on your computer screen and do so sufficiently to remember what the statistics were in the November 3, 1998 election in optical ballot counties in Florida?

MARAIS: No, sir. I don't recall that difficulty.

BOIES: Do you know whether the undervote in the November 3, 1998 election in Florida, there was the same disparity in terms of undervotes comparing punch card counties with optical ballot counties?

MARAIS: No, sir. Like Professor Hengartner I don't know that either.

BOIES: Is that something that you had available to you on your screen if you had looked at it?

MARAIS: No, sir.

BOIES: Did you ever ask anybody for that information?

MARAIS: No, sir.

BOIES: Let me turn to the election that we're involved in. And if I can find where the charts were put back I'm going to show you some of the charts that you were asked about. Plaintiff's exhibit seven. Now, I'm correct, am I not, that as a matter of arithmetic you don't have any disagreement with the statistics that are shown here.

MARAIS: I believe the division was done correctly. Yes.

BOIES: Now, am I also correct that you agree that the difference in the percentage of undervotes for optical ballot counties compared to punch card counties is very, very unlikely to have occurred purely as a matter of chance?

MARAIS: Yes, sir. That would be true. Just like the correlation between the seat belt sign and the turbulence in the airplane. It doesn't happen by chance.

BOIES: Exactly. That happens because the seat belt sign is turned on by the pilot, when the pilot thinks, based on information that the pilot has, that there's likely to be turbulence. Correct, sir?

MARAIS: That is correct. There is a mechanism in-between there. It is not the seat belt that causes the turbulence to occur.

BOIES: Is there a mechanism like that that applies to the difference between optical ballot counties and punch card ballot counties?

MARAIS: Several such potential mechanisms came up in my previous testimony, in fact, and in Professor Hengartner's testimony. Yes.

BOIES: Well, my question was not whether potential ones came up but whether you've identified any such cause?

MARAIS: I have not identified and studied in detail in the past nine days such causes. No.

BOIES: Have you reached a conclusion that there is any such cause that explains that difference?

MARAIS: I do not have a conclusion about what it is that causes those differences. My conclusions are about the absence of evidence for certain causes that have been advanced.

BOIES: OK. Now, would your answer be the same for the other charts that Mr. Hengartner put in? That is you don't have an independent conclusion as to what causes statistical differences that are shown.

MARAIS: I do not have independent conclusions about the true causes of the differences that appear in Professor Hengartner's charts. No.

BOIES: OK. Let me just try to clear up a few additional items and then I think I'm through. You mentioned ballot roll off and you mentioned a particular article by Steven Nichols (ph). Do you recall that?

MARAIS: I do. BOIES: And ballot roll off refers to the fact that people, when there are a lot of names on the ballot will tend to vote less for the less contested, high visibility races. Correct?

MARAIS: That is one interpretation of that term. Yes.

BOIES: That is in fact interpretation that is given in this article that you rely on. Correct?

MARAIS: I would have to review the article to make sure that was the and the only interpretation and that would take more time than we have.

BOIES: Well, let me just ask to you to read one sentence that I've highlighted and see if you agree with that.

BECK: What page are you on, please?

BOIES: It's on the second page of the -- could you read that into the record, please?

MARAIS: I do. I could. This is on page 107 in the second column and it reads -- this is a portion taken from the middle of the paragraph and it reads: "Also known as voter fatigue, ballot roll off is the long-recognized decline in the number of votes cast in lower ballot contests as compared to the tally in the higher profile partisan contests that attract most voters to the polls."

BOIES: Do you agree with that definition of ballot roll off?

MARAIS: I doubt that there is a single truly precise and universally recognized definition, but this is a fair characterization.

BOIES: Now in the election that just occurred this last November, did you vote in that election?


BOIES: And as a voter, did you have a view as to what the highest profile race was in that election?

MARAIS: I thought it would probably be the presidential part of the election. Yes.

BOIES: Yes. Let me ask you just a couple questions about the Iowa study that you referred to. Did that Iowa study deal with punch card ballots?


BOIES: Did it deal with optical ballots?


BOIES: Did it deal with a comparison of punch card ballots with optical ballots at all?

MARAIS: No, sir. And that's why I didn't bring it up in that context. It clearly has nothing to do with the vote or it does not reference the actual machines used in Florida. Its relevance is the effect of differences in voting method on voter behavior.

BOIES: And the causes of the lost votes that are referred to in this article. MARAIS: Yes, sir.

BOIES: Are what, sir?

MARAIS: There are numbers of potential causes including the appearance and the organization of the ballot, the transition from paper ballots to voting machines itself is cited in some portions of the report as a cause. But again, just to be clear in case I was unclear before, the relevance of the report is not because it refers to Votomatic machines and optical machines and systems because it does not.

The relevance of the report is that it is a documentation over a 64-year period in Iowa of the effect of changes in voting method on voter behavior establishing the proposition that voting method influences voter behavior.

BOIES: And the voting behavior -- the main voting behavior that was analyzed here -- was the behavior as to whether or not voters came to the polls. Correct, sir?

MARAIS: Various voting behaviors were analyzed including voter choice to vote offices once at the polls.

BOIES: Well, sir -- may I approach, your honor?

SAULS: Go ahead. I think you have another report up here somewhere, too.



BOIES: You do. Would you turn to page 36?

MARAIS: Thirty-six of the 110 approximate pages. This would be chapter four?

BOIES: Yes. And what is the title of that?

MARAIS: Causes of Lost Votes.

BOIES: OK. And would you turn to the last page of that chapter?

MARAIS: That would be page 40?

BOIES: Yes. MARAIS: I have it.

BOIES: And would you read into the record the last four paragraphs beginning, "What we seem to have here is not one but two voting machine effects. Two kinds of undervoting." Do you see that?

MARAIS: OK. So this would be the last four paragraphs of chapter four beginning in the middle of page 40. "What we seem to have here is not one but two voting machine effects, two kinds of under voting. One, a failure on the part of some voters to operate the machines properly to record their votes. This effect was transitory and may have been solved by improvements in the voting machine mechanism.

"Two, a failure on the part of some voters to come to the polls at all. They don't vote because they don't try because they don't want to be embarrassed or for some other reason. This effect which represents about 5 percent of the potential voters, is pervasive and persistent. These considerations should guide our efforts to educate citizens for effective participation."

BOIES: What is your understanding of the word "transitory" in that paragraph when it talks about the first effect being transitory?

MARAIS: Since this is a summary paragraph, sir, the specific use of transitory here clearly refers back to analysis in this 110 page document which I do not have freshly in my mind right now. So, I can give you my understanding from the dictionary and as a term from plain English of "transitory" but I cannot give you a precise characterization of what "transitory" means in this context in this 110 page report.

BOIES: All right, sir. The only reason I ask is because this was a report that was introduced under your direct. If you can't, you can't.

MARAIS: I can -- if it's fair to interpret your question is what does "transitory" mean as a word in English, I can do that for you.

BOIES: What I really meant was what was your understanding of what it meant in this article that you used on your direct examination and that you referenced to me when I took your deposition?

MARAIS: I can characterize that generally for you again with the caveat that this is a report based on numerous statistical analyses, not all of which that I have in my head. But the general sense of that term here is that certain of the influences on voter behavior that were observed in this analysis appeared most strongly shortly after the introduction of a new voting method and became attenuated there after. In other words, there was an initial spike and then things were not quite so strong or that they were somehow different. But I can't give you any more detail than that.

BOIES: OK. Mr. Beck asked you some questions about the butterfly ballot. Do you recall that?

MARAIS: I recall that he mentioned it. Yes.

BOIES: And are you aware of any suggestion that the butterfly ballot lead to under votes?

MARAIS: I don't recall a -- that suggestion specifically. I recall numerous suggestions in the news media, which are really my only source of information on butterfly ballots that butterfly ballots influenced voter behavior in a variety of ways.

BOIES: Do you recall that it was influencing voter behavior that maybe would lead some people to vote for two presidential candidates?

MARAIS: I believe I do remember that.

BOIES: And that would not result in what we refer to as an undervote or a non-vote, would it?

MARAIS: Voting for two presidential candidates would not be what we are calling an undervote here.


MARAIS: Although confusion about how to vote for the single candidate of one's choice might conceivably -- conceivably cause one to skip voting for either one. But this is not something that I know as a statistician. It's an issue that one could study. I have not studied it.

BOIES: And indeed you haven't seen any assertion even in the press that you read that the butterfly ballot caused people not to vote at all as opposed to vote for two people, have you?

MARAIS: As I testified a minute ago, sir, I recall a variety of propositions about the butterfly ballot and I have no clear recollection one way or the other of whether I've seen that one.

BOIES: OK. Have you made any attempt to determine, with respect to the Miami-Dade vote that you mentioned -- the Miami-Dade precincts that have not been recounted. Have you made any effort to project or estimate how many of those uncounted votes would be votes for Governor Bush and how many would be votes for Vice President Gore?

MARAIS: I am not entirely sure that I have. I have done it for the net effect of those two, and I guess implicitly that must mean the additional votes for Gore and Bush but only for the -- but what I had in mind was the net effect because I understood the claim of potentially 600 net Gore votes to be about the net effect and not about the individual totals.

BOIES: OK. So you focused on the net effect, and that's fine?


BOIES: But what I'm asking you is with the respect to the 9,000 ballots that have not been counted -- and you understand it's about 9,000 ballots, do you? MARAIS: Yes I do, roughly. I believe that's the approximation.

BOIES: Approximately. The approximately 9,000 Miami-Dade ballots that have not been manually counted, what would be the net effect on terms of presidential votes of counting those 9,000 ballots in your opinion?

KLOCK: Your honor, the secretary and the commissioner objects to any sort of statistical presentation of what the votes would be for one candidate or another. We think that it is well under the court's statutes. The only standards would be for the manual count which we will maintain must be on a state-wide basis. So having this witness testify as to what he thinks one person would or would not have done in the county is irrelevant under Florida law. We object.

SAULS: I understood the witness to say that he had not conducted a thorough statistical analysis of this and I don't recall him having testified concerning this on direct. But if he...

KLOCK: That is correct, your honor.

SAULS: But if he did then I would suppose that it is a subject matter -- what's correct or not.

BECK: You're correct, Your Honor, that he did not testify about a statistical analysis to come up with such a projection. This is beyond the scope of direct examination.

BOIES The very first thing that I wrote on my notes, your honor, was his testifying about how there were heavily Democratic precincts that were counted and then there were Republican precincts that weren't counted. And he talked about 48 percent to 52 percent and 75 percent to 25 percent. And I think that related to Miami Dade.

SAULS: Well, he was testifying that - with respect to whether or not that was a valid approach to project the 600 vote increase that they - I understood this witness to say that in his opinion, no, because it was based upon a false premise. And that is if - if the - that it would be dependent upon a proportion of the net were the same and pointed out that in the three counties that have been selected - I believe - that those were - the vote there was - were greater than 75 percent to Gore versus roughly 25 percent to Bush versus county wide the vote was 52 percent to 48 the other way. Therefore there was no basis for any projections from those precincts. Did I miss understand what you said?

MARAIS: No, your honor.

SAULS: All right.

MARAIS: That was it exactly and I wish I had said it that way.

SAULS: But did you testify that the statistical analysis, sir, so that you could project?

MARAIS: Yes, sir. SAULS: All right. That was my recollection.

BOIES And I just want to follow up on the last question. Could I just have that question answered back to because I wasn't sure that I heard it?


UNIDENTIFIED MALE: Did you say to the court that you had not done your own analysis of what those votes would be?

MARAIS: No, sir.

UNIDENTIFIED MALE: Your honor, he said that he wasn't asked to comment direct examination. That was the object beyond the scope of direct examination.

MARAIS: I understood your honor's question to be, "Did I testify to a statistical projection?" and I answered, "No."

BOIES Your honor, my I ask him whether he has done such an analysis?

BECK: I would like to reiterate what I said in the objection. I don't mind being over ruled, your honor, but I do not believe it is the least bit relevant that a statistician should sit up there and testify as to what (UNINTELLIGIBLE) in one county. That is not the issue before the court under the statute to have to demonstrate that there would be a difference on the statewide basis. And we would need, your honor, that the only basis under Florida statutes and Florida law is a manual recount of the entire state if you could do such a thing which we also question.

SAULS: Well it's a matter of argument but I don't know at this point. We're still trying to take evidence. And whether or not and whether or not there is a legal basis is a fact of determination. I frankly don't know. It was my understanding that the witness said he had not under taken and done a thorough statistical analysis of this. And if that is the case I don't know how he can express an opinion. Frankly, it's outside the scope of cross. But your question is, you want to ask him if he has done a thorough statistical analysis of this?

BOIES Let me start with that.

SAULS: Even though we're not going to get into that?

BOIES Yes. Let me start with that if I may, your honor?

SAULS: Yes, sir.

BOIES Have you done what you consider to be a thorough statistical analysis as to what the net votes would be in the presidential election if the precincts in Miami Dade that have not been manually counted were manually counted?

SAULS: Can we specify that we're talking about in that county alone or are we counting - are we talking about statewide?

BOIES I'm talking about the votes in that county alone. Although obviously the votes in that county alone if they were enough would effect the statewide election.

BECK: I would object again because (UNINTELLIGIBLE) statewide election.

SAULS: That would be a matter of argument. I understand that would be his position to argue. But we're trying to figure out the facts.

BECK: It is not appropriate for someone to try to make someone else look like an expert. And it's clearly beyond the scope of direct. And Mr. Boies's people are very interested in moving this along. This is not moving along.

SAULS: All right. Let me just ask the question. Have done a thorough statistical analysis of this particular matter relating to this particular county?

MARAIS: I have not done a thorough analysis . . .

SAULS: Thank you very much. OK. All right.

BOIES Now, can I ask him whether he has done a preliminary analysis of that?

SAULS: I think it's all irrelevant whether he's done a preliminary or not. If he hasn't done a thorough one, he has nothing of any appropriate value for that so . . .

BOIES All right. Would you agree that the only way to tell in fact what those ballots represent is to look at the ballots themselves. You cannot tell that purely from a statistical analysis.

BECK: Your honor, I'm going to object. We're into argument now. This is not cross-examination of a statistician. We want to move it along.

SAULS: It's all in the scope of the direct.

BECK: Right.

SAULS: Let me raise that.

BOIES I thought it was fairly raised by his testimony, your honor, but I will accept the court's guidance if you think it's beyond the scope.

SAULS: I believe it is.


SAULS: If you want to rephrase it, I'll try it one more time. BOIES OK. OK. Do you believe that the only way to tell for sure what the votes are in the 9,000 uncounted Miami Dade votes, who those votes were cast for if anyone is to look at the actual ballots.

MARAIS: So that's outside my expertise as a statistician but as a citizen watching television I can't say I saw any process that I would characterize as telling for whom those votes were cast for sure.

BOIES I was meaning to talk about the court reviewing those ballots. And I take it that you would have confidence that the court was able to look at the ballots and make a determination as to whether as to they should be counted.

BECK: Your honor . . .

SAULS: I believe he was brought in as an expert now he's testifying - your asking him questions about as an individual citizen so I think we need to move back into his testimony as an expert.

BOIES I'll move things along, your honor. Thank you.

SAULS: Correct.

BECK: Does that mean you're done?

BOIES It does.

BECK: Because I had a couple of points. First, when Mr. Boies was asking you whether you've done this thorough statistical analysis that I was also asking you about I think you indicated that you were first contacted by me or one of the members of my team about nine days ago?

MARAIS: That's about right.

BECK: And the kind of thorough analysis that we were discussing that would have to be undertaken in order to really find out what the cause is of these under votes, is that humanly possible to do in nine days?

MARAIS: No, sir.

BECK: If one were to undertake that kind of analysis, would it still be ongoing when the Electoral College met and decided who was going to be the next President of the United States?

SAULS: It is an objection to answer him.

BOIES I'm going to pass.

BECK: Patrick could you put the chart up on there? I need for you to flip me over to - there we go, thanks. Mr. Boies was asking you whether you remembered how many races were in each column, do you recall that?

MARAIS: I do. BECK: On the sample ballot that you looked at. And just you've got four races in each column on your chart, right.

MARAIS: That's right. Although I should mention that this chart is not to scale so if one were to look, as I recall, at the ballot - the races - there are four races shown here but they don't sit on the ballot book exactly next to each other in the way they are shown here. This is the right order.

BECK: And, of course, the ballot book doesn't say how many votes one person got. What you - you've got a chart summarizing the results of the vote, right?

MARAIS: Yes, that's correct.

BECK: Let's look over - please switch us to the ballot. How many races were in column one?

MARAIS: One, two, three four.

BECK: And how many races were in column two?

MARAIS: One, two, three, four.

BECK: So is that the same numbers that you showed on your chart?


BECK: Thank you. That's all I have, your honor.

SAULS: Thank you very much. May the witness stand down without objection. All right, at this time would it be appropriate to take a - well, we've got so many people - is 10 minutes satisfactory? I like to stay on the hours if we can. All right. Let's stand in recess until 11:00.

HEMMER: That's a 10 minute break on my watch. Judge Sauls recessing for a short break. I'm not sure if either side really has moved the ball in any direction just yet today. But certainly it is fascinating to watch these high profile attorneys at work. We'll talk with our attorneys, David Cardwell of Greta, Van Susteren and Ken Gross when we come back here. Our coverage of the hearing from Leon County Circuit Court will continue right after a quick time out.


HEMMER: Once again, welcome back to Tallahassee again as we sit here in this 10 minute break in Leon County Circuit Court. Let's bring in our panel for some quick thoughts. Two hours in, first of all to Greta Van Susteren in Washington.

Greta, your thoughts thus far on today's first and only witness to date on this Sunday?

VAN SUSTEREN: Well, the interesting part of the Bush statistician is that he said that the statistician that Gore's people used was flawed in his analysis since he only considered the heavy Democratic precincts in trying to determine whether those 9,000 votes in Miami Dade that the Gore campaign says have not been counted whether if they were counted 100 percent whether they would come out to be in favor of Gore or Bush. So that's an important point for the Bush team to make.

Now, the big question is whether or not this hearing is going to be over or not today. But even more than that earlier in the day Representative Gephardt from Missouri said this about whether about whether the entire matter will be over ever, "You know it would be terrible a month from now after this is all over if Gore is not the winner - not declared the winner. Somebody's going to go back and recount these ballots and look - under the Freedom of Information Act - and look and see who actually won. Would we want to wake up a month from now and have the national popular winner and the actual winner in Florida not be the President of the United States?"

Bill, the other interesting thing about the testimony this morning in terms of when it will be over when it will not be over is these - the underlying statistical analysis and one question that the Bush expert was asked by David Boies if it happened to the Gore campaign and if it was not answered, the Judge would not let him answer, is whether he had done a statistical analysis - a partial one - of those 9,000 votes. He was asked whether he had done a thorough analysis and he said he had not so he didn't have to answer but the answer we didn't get is that he do something less than a thorough analysis in order to give us some hint as to those 9,000 votes. Didn't answer that. Back to you, Bill.

HEMMER: OK. Greta, thank you. David Cardwell here in Tallahassee, jump on in.

CARDWELL: Well, we have the dueling statisticians now.

HEMMER: Yeah, we do. A lot of numbers out there.

CARDWELL: A lot of numbers are thrown out. We're getting some more terms that we're hearing now that we haven't heard before as we're getting into now some of the statistical analysis. But what was interesting I thought about the witness today was that he was really trying to refute the conclusions that were reached by the Gore statistician. They're really going off in opposite directions and keep in mind that the Gore campaign has to rather conclusively show that these statistics that they've produced are showing that there is a need for a recount. All that the Bush campaign needs to do is cast some doubt on that.

HEMMER: Our two reporters also at the courthouse, Mark Potter's outside, Gary Tuchman is inside. Quickly to Gary by telephone. Hello?

TUCHMAN: Hey, Bill. Well, this courtroom has been crowded and very hot but an icy wind blew through when we just watched that exchange between David Boies and between Lawrencious Marais, the statistician. One thing that the unflappable Boies tried to bring up was that fact that Marais was apparently involved in other cases as a statistician involving lead paint. The Judge did not allow that testimony. But the point that he was trying to bring up perhaps as a statistician is not credible if he is hired by certain people to testify in cases. One thing that was brought up by the last statistician, Nicholas Hengartner, from Yale University, the professor, and the Gore team made a point to bring this up is that Hengartner has not been paid to testify. He is enjoying the publicity basically for Yale University and for himself. And so that's something we see in a lot of cases. Attorneys try to point out to a Judge or a jury that someone is getting paid to testify. Of course the Judge can consider that. It doesn't necessarily mean he is less credible. But that's something that a savvy lawyer likes to bring up.

The other point that we want to say is that in was interesting this morning when court started the Judge asked the Bush team how many witnesses they would call. And they were talking about - the Bush team said to the Judge, "We'll over lunch discuss if need a couple of other witnesses." And the Judge made a quick point and said, "Well, I don't know if we're going to be having lunch. We want to finish this this morning." So the Judge still wants that quick pace although everyone seems to be taking their time and it's not so clear whether that quick pace will occur. Bill?

HEMMER: Indeed. And certainly time is certainly a factor in this particular contest. Just about five minutes left before this break and court will reconvene. Let's get a quick time out here. Back with more in about two minutes time.


HEMMER: Again, welcome back on a Sunday morning here in Tallahassee waiting for Judge Sander Sauls to come back into his courtroom. While we still have time let's check in with Mark Potter outside of Circuit Court here in Leon County. Mark, hello to you. Are we moving the ball any just yet?

POTTER: Well, this is - this is like watching paint dry. This is going very slowly, and we went nine hours yesterday, got through four witnesses, and now we're -- what -- two hours today, getting through another. And we have many more witnesses to go.

We're told that we're going to hear from a man who was involved in building the Vote-o-matic, designing the Vote-o-matic machine, an individual voter who put a stylus on the ballot but then changed his mind, creating a dimple, but that's not what he meant to do. He meant to vote for somebody else, and the point there is obvious.

Also, a vote count observer, we're told, are going to -- we're going to have a vote count observer, who talks about how the dimpled ballots were counted, and then we'll have officials from Miami-Dade and Broward Counties, perhaps Nassau, to talk about the vote count procedures. They're all -- these people are being brought in to counter the points that have been made by the attorneys for Vice President Gore.

Now we have Tucker Eskew with us here from the Bush side. Tell us first about the statistician that was saw today. What was the point of having him testify?

TUCKER ESKEW, GORE CAMPAIGN SPOKESMAN: It makes clear with this scientific witness that the case the Gore Campaign has to make, which really is a very high burden -- heavy burden to carry, was not carried by the Gore Campaign. We've been able to prove that the kind of statistical so-called aberrations they cite simply are matter of fact, and they occur in elections.

POTTER: And you have some concerns also about some of the other cases that we have here. It's hard to believe, but there are other cases...

ESKEW: There are.

POTTER: ... besides this big one involving Seminole County and Martin County, some absentee concerns -- a ballot -- actually, the call is to throw out absentee ballots -- there are 25,000 -- and you have some thoughts on that.

ESKEW: Well, there's news this morning, Mark. First of all, Secretary Christopher, on behalf of Al Gore, this morning embraced these lawsuits, which would have the effect of throwing out and disenfranchising as many as 20,000 voters in the state of Florida on some hyper-technicality, a minor mistake. They say they want every vote to count, and yet they embrace this strategy.

Meanwhile, the Orlando Sentinel reports that the man who filed suit in Seminole has a lot of direct ties to Al Gore, including a $50,000 donation to him this year.

POTTER: Well, thank you very much. We'll be talking to you later in the day.

ESKEW: Thank you.

POTTER: I appreciate your time.

So that's the situation here. We're going to be going back into court shortly. I think we have a long day ahead of us. There was talk -- the judge wanted to end all this within 12 hours total. We're only an hour away from that, and, obviously, we're not going to meet that deadline.

Bill, back to you.

HEMMER: All right, Mark.

Mark Potter again. We were told at the outset today we had a witness list for the Bush team that extended about 10 different witnesses. We have gotten through one thus far, and, again, based on what Mark was saying from yesterday, the judge tried to wrap it up in 12 hours. They quit after nine hours, and, again, it's anybody's guess as to whether this will wrap up to day.

There are many who do believe, though, that it's quite possible we could go into Monday, or even beyond Monday. So, again, no conclusive word on that just yet.

And, again, before Judge Sauls comes back into the courtroom, again, it's important to point out that he is the sole arbiter in this case. There is no jury. Judge Sauls' decision and his opinion and his verdict is the verdict that these attorneys are fighting for right now.

Back to Mark Potter for equal time. He's now with Doug Hattaway -- Mark.

POTTER: Yeah, just to prove that we are fair, we're going right down the middle, I now have next to me a minute later Doug Hattaway, spokesman for the Gore Campaign. Now, your thought on the statistician that they brought in there. He seemed to really be chipping away at the points made by the statistician brought in by the Gore Campaign. Has the argument -- has the Gore side been seriously hurt by his testimony?

DOUG HATTAWAY, GORE CAMPAIGN SPOKESMAN: Not at all. I think we made the case pretty clearly that thousands of people showed up to vote, and their votes weren't counted. He admitted he didn't do a thorough statistical analysis of Miami-Dade. They're asking us to believe that 10,000 people showed up at the polls, voted halfway, then changed their mind. That just defies common sense, much less any statistical analysis.

POTTER: Now, there is a big threshold that must be met to get -- to convince this judge to order yet another manual count. Can that be met? Only two witnesses were put on. Do you think that that was met in the presentation so far?

HATTAWAY: Sure. Certainly there is a high legal threshold, but the case is so strong that thousands of people's votes haven't been counted here in Florida, and they ought to be counted by a human being. I think the evidence showed yesterday -- we had two strong witnesses making a very simple case. The machines spit out ballots without counting them. The law provides they ought to be counted by a human being. That's all we're asking for here.

POTTER: Well, thank you for your time, and we'll be talking to you later in the day, I'm sure.


POTTER: Thank you very much.

So the case proceeds, and shortly, we'll be going back to court. But, first, back to Bill.

HEMMER: All right, Mark.

We've been fairly punctual throughout this matter. Judge Sauls has kept a pretty tight courtroom, but, at this point, extending now -- that 10 minute break has now run 15 again.

But the issue at hand now for the Gore team is the undervote again. They believe that there are votes from two different counties, Miami-Dade and Palm Beach, which show that their man, the vice president, Al Gore, has enough votes to win the White House.

Republicans counter that the undervote is strictly just that. Nobody voted in that column for a president in this Election 2000.

As we await the attorneys and the judge, we'll have a quick time- out here in Tallahassee. Back with more after this.


HEMMER: Again, here in Tallahassee, we are in a recess, the first one of the day today. Back inside the courtroom, as the attorneys mill about here, once again is Gary Tuchman on the telephone inside the courtroom there -- Gary.

GARY TUCHMAN, CNN CORRESPONDENT: Bill, one thing I think that's notable -- I've been covering court cases for almost 20 years, and this case -- I've never seen a court that has such loose rules when the court is not in session. The reason I bring that up is because, as we speak right now from the courtroom, you have lawyers from both sides, actively spinning reporters, reporters asking questions, doing interviews in the courtroom.

Out in the hallway outside the courtroom, cameras are rolling, tape recorders are rolling. Lawyers from both sides are out there doing interviews during every single break, trying to put their spin in.

A lot of times in major court cases -- and this is certainly a major court case -- there are very strict rules about what should be done -- the conduct inside a courthouse, inside a courtroom. No such rules apply here.

Normally, lawyers don't like to talk. They don't want to anger the judge. There's no such fear here. Everyone is talking, and reporters and photographers are inside the courtroom with their cell phones, with their computers, with their cameras rolling -- not allowed to use your cell phone while court's going on, but in between all these breaks, there's plenty of reporting and spinning going on, and that is very, very unusual. No problem from the judge or his staff with doing that, and that's why everyone's doing that right now.


HEMMER: Gary, it was my impression from being over there earlier in the week that it was virtually an all-access pass. You could come and go as you pleased throughout the entire hearing.

But up to Greta Van Susteren in Washington -- more thoughts on that, Greta, or do you want to take this in a different direction?

GRETA VAN SUSTEREN, CNN LEGAL ANALYST: Well, let me take it in a little different direction, Bill. Let's talk about what happened this morning. What the Gore people have to prove in order to get the hand count done is that the hand count matters. They called a statistician yesterday to make that point.

Well, today, we could see an opposite view. We've got a statistician on behalf of the Bush people, who says a hand count won't matter, and that the work done yesterday by the Gore Campaign -- the Gore people's statistician was flawed.

So you have dueling statisticians -- not an unimportant point. They must prove in order to get -- to prevail in this matter to get the hand count is that, in the event there were a hand count, that it would make a difference, and this is a race with 537 votes separating the two. So that's why you saw such hammering on both witnesses on the issue about their reports.

HEMMER: And the gentleman there in the red tie with the receding hairline there is Irv Terrell out of Houston, Texas. We do anticipate hearing from him as we get underway again, once again inside the courtroom.

But before we go back inside and while we wait, let's check in on the campaigns themselves. First, Tony Clark is live in Austin with the Bush Campaign, and from Washington, CNN's Eileen O'Connor traveling with the Gore Campaign and watching the steps of the vice president there.

First, though, to Texas. Here is Tony. Tony, good morning to you.

TONY CLARK: Good morning, Bill. The governor is at his ranch about 90 miles from here. He has no public events planned today, no public exposure planned to day.

But the campaign itself is moving ahead, trying to plan for a Bush Administration, if there is one. In fact, they've set up a transition web site that has an application on it so that people can apply for the thousands of jobs that become available with a change of administration. The web sites says the Bush-Cheney Campaign is looking for men and women of the greatest ability and highest ethical and professional integrity.

But it goes on to warn that government service isn't for everyone. The hours are long, there is an intense pace, and the applicants have to lay bare their financial situation, their education, their health, all sorts of facts about themselves, and it says those facts would likely become public at some point. And so that's the sort of thing that they're doing in preparation for an administration.



Tony Clark from Austin. Do we have time for Eileen O'Connor? Is court about to resume?

All right. Let's go back inside. Circuit court here in Leon County. UNIDENTIFIED MALE: (UNINTELLIGIBLE)

SAULS: Mr. Kitchen?

DINO KITCHEN: Judge, quickly, we, yesterday, expressly asked you about the opportunity to go through Plaintiff's Exhibits that have been offered, admitted, and make sure everyone's on the same page so that the record will be clear.


KITCHEN: The 388 ballots that were undervotes that were actually counted in Dade County we've now had separately marked as Plaintiff's 59 for identification. Out of an abundance of caution, we want to move 59 in. It may already be in, but -- so -- but that won't be...

SAULS: Without objection, so ordered.


SAULS: Yes, sir.

UNIDENTIFIED MALE: I'm not sure (UNINTELLIGIBLE) I'm sorry to address, Your Honor, but is the exhibit that the plaintiffs have offered actual ballots that are undervotes? I couldn't hear.


SAULS: Yes. And I assume they want to put in an exhibit number, but as far as I know, those ballots are still...

KITCHEN (?): Yes, yes.

SAULS: ... where they're supposed to be and apparently are separated and kept in a different place. So they're available in the event they're needed.

UNIDENTIFIED MALE: My point is I do object on behalf of Governor Bush to any admission of Miami-Dade ballots for reasons that'll become clear in the testimony of my witnesses. At some point...

SAULS: All right. At this time, then, there is an objection, so we'll -- you will have them -- they've been identified with an exhibit number, and then we'll see about whether or not there's a subsequent admission. Just make a note of it. All right.

UNIDENTIFIED MALE: And, Your Honor, we have objections as well. We have the right to just hand out and distribute (UNINTELLIGIBLE) exhibits on behalf of the secretary (UNINTELLIGIBLE)...

SAULS: All right. I believe that what we said is that we would give them preliminary admission subject to a determination of any specific objections.

UNIDENTIFIED MALE: We'll just hand them (UNINTELLIGIBLE), Your Honor. Thank you. SAULS: All right.

Mr. Beck?

PHIL BECK: George W. Bush and Dick Cheney call John Ahmann as their next witness.

SAULS: Ask him to come forward and be sworn. Would you allow the court to swear you in?

UNIDENTIFIED MALE: Do you solemnly swear that the evidence you shall give in this issue shall be the truth, the whole truth, and nothing but the truth, so help you God?


UNIDENTIFIED MALE: Please be seated.

SAULS: Be seated, sir.

AHMANN: Thank you, sir.

SAULS: You may proceed.

BECK: Sir, would you please state your full name and spell your last name for the record?

AHMANN: My name is John Ahmann, A-h-m-a-n-n.

BECK: And, Mr. Ahmann, what do you do for a living?

AHMANN: At present, I'm a rancher in Oregon and California, and I also have a small elections supplies company in California.

BECK: Have you had experience over the years with punch-card ballot machines?

AHMANN: Since 1966, yes, sir.

BECK: And have you been asked by me to come here and offer testimony concerning the design and operation of punch-card ballot machines?

AHMANN: Yes, sir.

BECK: And, particularly, have you been asked to come and give testimony about how indentations and dimples can find their way onto ballots used in punch-card machines?

AHMANN: Yes, sir.

BECK: You have to speak up just a little bit. Before we get to that testimony, I'd like you to spend a few minutes describing your background in the industry, and try to be as brief as you can while still giving the court a good grounding in your experience, so -- starting out with your first job that was in this area, and then kind of run through chronologically as briefly as you can.

AHMANN: I joined IBM in February, as I recall, 1966, and I worked in the tab card plant in Campbell, California. On or about May-June of 1966, I was asked to assist in the rework of the Vote-o- matics that IBM had built and were used in the primary elections in 1966.

We did that, and we continued to work that summer on the Vote-o- matic. I installed -- I was assisting in the installation and rebuilding of all the equipment that went into San Bernandino County. I was in charge of the work -- rework operation.

After that, I made about 85 suggestions as to how they could improve the product. For about two months, I was asked to work on the crimped hinge, which was an idea they had which they thought would speed up the assembly of the ballot pages. In doing that, I designed the crimped hinge, patented -- I didn't patent, but I worked on the mask punch and the crimper and made a single unit out of that crimper puncher.

I then was asked to go back to the engineering department in Dayton, New Jersey, for IBM and become the responsible engineer for IBM to assist in developing and debugging the IBM Vote-o-matic. I did so. A year later, they sold Los Angeles County, and they sent me back to California, and I installed Los Angeles County.

BECK: You say you installed Los Angeles County. You mean you implemented the changeover to the Vote-o-matics in Los Angeles?

AHMANN: Yes, sir. That is correct.

BECK: And what year are we in now?

AHMANN: We're in 1968.


AHMANN: And at the end of 1968, IBM exited the voting equipment field and decided to concentrate on computers. Three of the salesmen and the chief programmer for IBM, Ken Hazelet (ph), decided to form what was then known as Computer Election Systems, CES.

BECK: So CES -- and this was the end of 1968, beginning of '79.

AHMANN: Sixty-nine, yes.

BECK: I'm sorry, '68-'69.

AHMANN: Right.

BECK: And did you go with CES?

AHMANN: Yes. They asked me to come with them and join them, and they did make me a minor partner in the company. I was the engineer that actually designed, developed, and built all of their equipment for seven years. BECK: And so that brings us up to what, the mid '70s?

AHMANN: That is correct. During that time, we developed the Model 3 concept, which is what most of the counties here in Florida are using, and the 3-P was developed the year after I left, after we installed Cook County and precinct ballot counters, which I also helped design and develop. And that is the unit that you have down in Dade County, the 3-P.

After that, I worked two years for another company, Consolidated Foods Corporation, as a plant manager, and approximately June, 1978, my wife and I decided to start our own little supplies business and go into business. We called our company Election Supplies Limited.

Within two years, I re-invented the Model 3, basically, and we call it the Model 5. It was a streamlined, more compact version of the Model 3. We put a full five-year warranty on it. I also patented the new stylus, which was stronger, had a more flexible tip, had virtually no breakage on the tips of that stylus, and many of them were sold and distributed throughout the state of Florida.

We also sold from our company at that time in the mid '80s approximately 6,000 new punch frames to Dade County. So we also had our labels -- a very large instruction label which you saw here in the back of the booth yesterday. That was from our company. That was our design and development. So that was to aid in making sure that the voters turned their card over and checked the back side of the ballot for a hanging chad, and it was thought to be a very good improvement, and we've sold a lot of those -- had sold a lot of those.

We continued with Election Supplies Limited, until Business Records started to buy up all the small voting -- election machine companies. We continued as ESL until 1986, at which time we did merge with Business Records.

My wife and I had a five-year contract to continue managing our business. We manufactured all the voting equipment, punch-card voting equipment, and booths, developed a new Optech (ph) ballot box for the Optech (ph) -- which is still used with the Eagle, I understand, in some locations, as well as a new electronic voting machine, which I'm listed as a co-patent holder on.

I have 11 patents, actually implemented over 50 design changes and new pieces of equipment and gear for the Vote-o-matic system, including the semi-automatic crimper, the pneumatic crimpers, the new auto-inchfeed crimper, and so forth. This is the auxiliary equipment which I have serviced here in Florida just this last March in about five counties.

We sold the Business Records, as I mentioned, in '86. At the end of 1991, they felt that they could go ahead and manufacture all of our equipment on their own. They didn't take any people. They transferred it to Texas.

A year and a half later, I received a phone call that said, John, I don't know how you did it, but our hat's off to you. By the end of the week, everyone in their department had been fired, and they decided to send the equipment and all the manufacturing back to Berkeley, which is where most of it began. And...

BECK: So did you continue, then, working in the field in the '90s?

AHMANN: Actually, in the early '90s, I had a golden parachute with Business Records not to compete, a non-compete clause which went for about three and a half years. In 1996, I was asked by Ed O'Day (ph), one of our patent salesmen that I had known -- we used to play guitar together and sing at the county clerks meetings and the supervisor elections meetings here in Florida.

And Ed (ph) says, John, they're trying to kill punch-card voting. You need to get back in the business, and you need to start making the crimped hinges and the masks, and while you're at it, can you help me with my voting booth, and I did, and we have, and the first year and a half, we sold -- I think it was five million crimped hinges and about half a million masks.

BECK: Are you still active in the business today? I think you said that, primarily, you're a rancher at this stage.

AHMANN: Primarily a rancher. We have one full time employee in Napa, California, who basically builds the lamp assemblies for both ES&S (ph) and for Ed O'Day (ph) in his voting booths, and we do have the crimped hinges manufactured at the company that's been making them for the last 20 years, and we do our mask blanking in our plant, so it's a small involvement.

BECK: OK. Now, with all your years in the -- in the industry, are you familiar with which kind of voting machines are in Dade, Broward, and Palm Beach Counties?

AHMANN: I think I am, yes. I have seen them.

BECK: Which models are in Dade County?

AHMANN: Dade County has the 3-P voting booth.

BECK: Is that the one with the plastic case?

AHMANN: Yes, that's why it's called P, for plastic.

BECK: Is that the one that we saw up there yesterday?


BECK: Which one was that?

AHMANN: That was a -- actually, that was a late model 3-A. I recognized it. That was my manufacture out of our plant in Napa.

BECK: And the 3-P was made by -- was that CES who made the 3-P?

AHMANN: Yes. As it happens, when anyone leaves an organization, other people have other ideas, and they say let's make it cheaper, and cost reduction was to make it out of a plastic case, and so they did.

BECK: OK. And then in Broward -- what kind of models are in Broward?

AHMANN: Broward has predominantly Model 3-As.

BECK: And that's the aluminum?

AHMANN: Yes, that was the 1981 -- or 1970-71 patent, I believe -- '72, yes.

BECK: And was that the kind of machine that was up there on the stand yesterday?

AHMANN: That type, yes.

BECK: Are there any other kind of models that you know of in Broward?

AHMANN: They have various -- they may -- they may have some Model 5s. I believe they do. I think I remember seeing those. They're some of our late model units they picked up used. I don't believe they purchased them new.

BECK: How about Palm Beach? What kind of models do they have?

AHMANN: Palm Beach has the 3-As also, and I believe they have some Model 5s, and I know they have the Pollstar (ph), but I don't believe they have any 3-Ps...

BECK: And even though you don't -- you've not been involved in manufacturing the Pollstar units, are you familiar with the Pollstar unit through your years of work in the industry?

AHMANN: I am familiar with the new one somewhat. Some of the previous ones, I really didn't get a chance to look at.

BECK: Do you understand that the Pollstar units that are in Palm Beach in use were purchased in approximately 1995?

AHMANN: That's what I understand, yes.

BECK: And are you familiar with that unit?

AHMANN: I have seen it. I have voted on it -- test voted on it, and that's about the extent of it.

BECK: I'd like to put up on the screen an illustration that we worked on, which is our Exhibit 513, and you can either turn back to the screen, or you've got it in front of you. Would you please describe for the court what the various things are that are shown here in this illustration?

AHMANN: May I get up and...

SAULS: You may, sir. AHMANN: ... point to them? We have here the -- actually, it's an oversized stylus, because this tip is only fifty-thousandths diameter.

BECK: The tip is only what?

AHMANN: Fifty-thousandths in diameter.

BECK: Fifty-thousandths of what?

AHMANN: Around. That's the diameter. It's stainless steel. This is a rubber body, size SBR (ph) rubber, very flexible. These are your T-strips sitting -- pardon me, templates sitting on top here.

BECK: The gray things show a cross section of plastic templates?

AHMANN: A cross section, usually with hash marks if you show them on a drawing. But, anyway, that's your template and your (UNINTELLIGIBLE). Our guide section is not nearly that vertical, or I should say angled, you know, a very shallow guide, and the vertical portion of the template hole is very much straight up and down. That's the...

BECK: John, can I ask you to step to the other side so the judge can see the diagram?

AHMANN: You bet. Sorry. Anyway, that's your template, and then you have the ballot card underneath with your little breaks in here. Those are your chads being outlined. These are your T-strips, and we called them T-strips but they actually are molded -- compounded rubber which is molded...



UNIDENTIFIED MALE: I have to object to any testimony regarding any rubber aspects for the T-strips and give the court a copy of his deposition. On page 63, he was specifically asked about these issues, and he said, and I quote from line eight, "Well, we have quite a few rubber parts to the Vote-o-matic, as you know, the flexible body stylus, the T-strips, but I don't consider myself an expert. I have to rely on chemists and also people such as 3M who do have great engineering staff." I assume that's why they had a chemical engineer here yesterday.

Any testimony about the qualities of the rubber, the T-strip, which he said he was not an expert about, we would object to, both for the fact that there is no foundation and he's acknowledged his lack of expertise, and also it'll be (UNINTELLIGIBLE).

SAULS: Overruled.

BECK: Go ahead, please.

SAULS: I understand he said he's the man that -- you designed -- this was your design?

AHMANN: Yes, sir.

SAULS: And he's going to testify about his design, and I don't believe that I've heard any questions concerning his being asked about the properties of the rubber itself. He's merely said that these are rubber T-strips, and I assume we're going to find out at some point where he got them and what his specifications were for.

UNIDENTIFIED MALE: I very much appreciate the (UNINTELLIGIBLE). The only thing I'm talking about is the rubber aspects of any of the parts of the machine. We understand that he's going to testify about the mechanics (UNINTELLIGIBLE).

SAULS: At the proper time, you may have an objection.

UNIDENTIFIED MALE: Thank you, Your Honor.

SAULS: Go ahead.

AHMANN: These are the T-strips, as I was saying. They're T- shaped rubber molded parts, and this is the -- simulating the plastic part of the punch frame.

You can see there's quite a distance between the T-strips and how far they go down here on the plastic. There is a crossbar at the bottom which joins these together. Then there is an open space below that, which gives you a substantial accumulation area for those T- strips -- or for the chads to accumulate beneath it.

In fact, in the Model 3-A, there's a substantial area beneath those, if you don't mind my saying so. In fact, it's -- if you calculated it, it's 90 cubic inches plus of storage space in a Model 3-A or 3-P. If you calculate the width of the chad -- by length, by the width, by the thickness, you have approximately one and a half million chad -- pieces of chad that can be accumulated in a Model 3 vote recorder.

BECK: OK. Now, let's just -- with my illustration here, we've got here -- take a look at the styluses going to -- let's see -- comes down, you can see there. What's happening now in the machine that you designed?

AHMANN: This is -- this is the -- the basic design is that the rubber actually pulls the chad while the corner points are being broken. OK. If there is a weakening of the rubber or wear in the rubber...

BECK: I just want to -- we can get to this later. I want to just now talk about what happens in a normal voting process.

AHMANN: In a normal voting process, this is what happens. The chad is broken loose, because the rubber has held it there, and then as the voter pushes it in all the way down, the chad then goes between the rubber T-strips -- and you can see the rubber T-strips riding on the barrel of the stylus -- and the chad then falls down to the bottom of the voting device, which can be almost an inch distance down below.

BECK: OK. You can -- you can resume your seat now, please. In your work at IBM and in your work over the years in this industry, do you work with the actual ballots that are used in elections?


BECK: And do you have to become familiar with the characteristics of the ballots in order to design the equipment that's going to be used when people cast their votes on those ballots?

AHMANN: At IBM, yes, you do, and at IBM, I actually did some of the design work on the ballot card itself and the location of it, et cetera -- changed the specs.

BECK: Are the ballots susceptible to having imperfections appear on them through various ways?

AHMANN: There can be flaws in the paper. There can be nicks or dings in the manufacturing, if that's what you're referring to.

BECK: Well, once you get beyond the manufacturing, even if you assume that a ballot comes in perfectly pristine condition with no nicks or dings anywhere in it, is the -- is the ballot itself of such a nature that it's easy or hard to make impressions and further nicks and dings in it?

AHMANN: There are basically two types of -- yeah -- yes.


AHMANN: There are basically two types of ballots. We have the demonstrator ballot, and we have the official ballots. And I think the testimony has been already that the demonstrators are made with the tools that get worn and gives you a higher burst test, and the official ballots have a lighter punching -- punch-out requirement, because the tools are new and the blades are sharper. And so it is possible in rough handling -- if they're not handled properly, you can get nicks and dings or fingerprints in the ballots, yes.

BECK: What kind of things now -- and I -- and I don't want to limit you to dimples made by a stylus, but rather could you explain to the court what kind of things could happen in the normal handling of ballots by court personnel -- not court -- election personnel and the voter herself, and then -- and then further handling of the ballots? What kinds of things could come into contact with the ballots that could cause indentations and could cause a chad, for example, to be pushed out a little bit but not completely dislodged?

AHMANN: Well, as you know, the ballots are first received from the manufacturer in a cardboard container, and it says right on those containers they must be stored in a -- in a type of environment which is cool, dry, and you don't want to get them wet because they can expand and curl and warp -- so they're stored properly. When you go to transfer them to the precincts, you need a transfer carrier or case if they're taken out of that carton, which usually they are, and put into a transfer case.

That transfer case that you design has to have edges on it and padding inside of it so that the card is not damaged by the follower (ph) blocks, which is a typical type steel slide -- steel tray slide which is put into those type of transfer cases. And then when it gets to the polling place, depending on whether -- what type of ballots you have -- you either have a fold-over or you have the short stub ballot with a ballot envelope. Normally, that ballot is placed inside the envelope so it can't get handling damage by the voter or by the precinct officers.

BECK: Do you know which kind are used in Florida in these three counties?

AHMANN: They are using the fold-over ballot, which the fold-over ballot -- the top stub folds over after it's used to hide the...

HEMMER: You're listening to the contest hearing in Leon County Circuit Court in Tallahassee, Florida. This in the courtroom of Judge Sanders Sauls, and, at this time, we welcome our international viewers watching from all over the world here in Tallahassee, Florida. Testimony continues now with John Ahmann on the stand.

BECK: ... the voter puts the ballot in the machine, whether it originally comes in an envelope or not, is the -- is the voter then handling the ballot?

AHMANN: Yes, definitely, the voter is handling the ballot.

BECK: And I suppose you've seen on television, as we all have over the last few weeks, these illustrations of different kinds of imperfections that could appear on chads in a ballot, where sometimes they could -- they're sort of like pregnant or pushed out, but they're not detached in the corners, and the various variety of imperfections. Could those kinds of things be caused by handling of ballots?

AHMANN: Yes, they can. Some can.

BECK: How would that happen?

AHMANN: Well, as an example, in some precincts, over the years, when voting gets very heavy, they will issue the ballots to the voters, who then, in turn, stand in line possibly for as much as, well, 30 minutes or so while they're waiting to get into the voting booth while another person votes. And while you're waiting in line, you -- the cards actually pick up the moisture off your hands, for one thing, and if they can pick up the moisture off your hands, by handling them with your fingernails, your ring, or whatever, you can cause dents or dings in the ballot. This can happen.

BECK: How about after the votes have been cast and the ballots have been collected, and then they are further handled by the -- by the voting personnel in order to be put into the machines? Can further dings and nicks and imperfections occur then?

AHMANN: If -- are you referring to the ballot inspection boards or in the precinct where they actually sit there and count them? Depending on how they're handled in the precinct -- you know, they tear off the stub, lay them on -- if they don't tear off the stub properly, you end up with a big wad of card still sitting on the ballot card.

In fact, it's thought that -- I won't say it's thought -- it's been -- I've seen it happen before where people would tear off the stub and actually remove part of the card, or in tearing off the stub, you're pushing down with your finger -- this is in the precinct, when they are doing -- you're tearing off the stub, you can move or get your finger on the chad when you're doing that. So it has to be -- it has to be handled carefully so you do not disrupt the ballot -- the chad in the ballot when you are tearing off the stub. And this is something that has to be done before they can be processed through the card readers.

BECK: And once they get into the card readers, the -- we've all heard about how the chads get dislodged, but, sometimes, do chads only get dislodged part-way, like -- so that they're those hanging chads, where there's -- two corners are still there, or even one corner, but the chad hasn't quite come loose?


BECK: Do the -- do the folks who work for the election boards -- before they put the ballots into the machine to be counted, do they do anything to take off those hanging chads, or whatever the term is for these chads that are hanging by one or two corners?


BECK: And when they are doing that, when they're -- when they're manipulating the ballot, say, in row two, and pulling off the loose chads so they won't gum up the machine, is there the possibility that in that process they might touch a chad somewhere else that could leave an imperfection?

AHMANN: Yes, that is possible.

BECK: Now, does it sometimes happen that when the ballots are run through the counting machine, that, notwithstanding all the work that the people have done, there are still some loose chads that sort of get in between ballots?

AHMANN: That can happen as well, yes.

BECK: And if the machine -- if they happen to sit behind another chad in between the ballots, what effect, if any, would that have on the other ballots that they're next to?

AHMANN: If it's going through the card reader -- which, of course, that's the way you normally count them -- and you have a hanging chad on one card, and you have open holes in another card next to it, and it's being roller fed and air separated, hopefully -- they sometimes don't always air separate so well because of the fact if there's too many chad hanging in this card, it may go -- be introduced into the hole of the card next to it, you can get damage, or you can actually have the chad from this card come off in the hole of the next card if it happens to be punched the same way. So...

BECK: Could it also -- if you -- if you have a hanging chad, and you've got the ballots next to one another, and they're pressed against one another, could it -- could the hanging chad from one that's kind of folded on top of that ballot cause an indentation in a chad on the -- on the ballot right next to it?

AHMANN: If you have a thousand or so cards up in the hopper, and they're all coming down -- and that would put a lot of weight on them -- there's a possibility that that could happen.

BECK: Now, of course, is it also possible that somebody can make a dimple, for example, in a chad with the stylus that is used in the voting machine?


BECK: Now, in your years of working with this -- with these cards, the actual ballots, and the voting machines, have you come to any conclusions about how likely it is if someone is actually following the instructions and putting the ballot in like they're supposed to and attempting to vote would instead end up leaving a dimple instead of punching the chad through?

UNIDENTIFIED MALE: Objection, Your Honor. This is outside his expertise. He was designated as a mechanical engineer. He is now talking about psycho -- neurological, psychological issues, clearly outside the scope of what he's been submitted to this court for.

BECK: I can ask this question...

SAULS: Overruled.

BECK: Go ahead. Answer my questions.

AHMANN: At IBM, in 1966, we conducted extensive studies on human factors and how the voter rolls in. If he's right-handed, he tends to roll in, and because of the way the hand works, you tend to pull towards yourself, and, as a result, templates get worn.

They can -- you know, because you're rolling in. And, sometimes, people actually come over and they will sit the stylus down basically in the hole, and if you -- what the voter doesn't know sometimes is that the chad breakpoint is about -- can go as low as 150 grams. That's about six ounces.

If he goes in and taps it, it's quite possible you can dimple the ballot and have no intention of actually voting. Conversely, the upper limit is 350 grams on your chad burst strength, and you have to punch.

It takes about -- almost three quarters of a pound to punch down, which, if you take the weight of your hand, it punches real easy. But if you just touch it, you haven't voted, and -- does that answer your question?

BECK: Yeah. Now, if somebody, for example, were to put the stylus through the hole next to Al Gore's name, and then they're making up their mind whether they want to carry through with that and actually vote for him, and then decide you know what -- I'm a Democratic, but I just don't like this guy -- and they pull the stylus out, could that leave the kind of indentation you're talking about?

AHMANN: Depending on -- depending on if he comes back and touches it hard enough, he may unknowingly have dimpled it. It's that -- it -- one or two ounces -- it's that finite, a small amount of force that's required. Yes.

BECK: Now, we have these various ways that chads can have imperfections that you talked about, including that the dimples could come from a stylus. If someone were so inclined, could -- are there means by which you could figure out whether a particular imperfection in a particular chad came from a stylus versus one of -- all these other causes that you've talked about?

AHMANN: Definitely.

BECK: How would somebody determine that an imperfection in a particular chad came from a stylus versus one of these other possible causes?

AHMANN: Well, it's rather simple. If you're using a needlepoint stylus, you will actually see a needlepoint in the center of the chad, and if you look at the chad, it will look like the old fashioned two- man tent. It actually creases right in the center, and you'll see the needlepoint going right through the center of the chad.

BECK: And can you look at that with your naked eye, or do -- would you have to do something else?

AHMANN: Preferably, it'd be much easier, especially for older people, to use a magnifying glass, a very strong magnifying glass, or a low powered microscope.

BECK: And would you be looking not just at whether the -- there was some form of impression or the chad had been pushed out a little bit, but actually looking for a witness mark that you could match up with the stylus tip?

AHMANN: Yes, that does show, especially on the -- well, on the serrated (ph) stylus, you don't have the needlepoint, but you will see -- because the rubber is black, and you push on it with this fifty- thousandths -- forty-thousandths diameter tip at the bottom of the serrations (ph) -- when it makes contact with the rubber, it does leave a slight blackening right there and you can see this dimple in the chad without breaking the chad loose.

And I've observed those and done tests with it, and it's also true of the absentee voting punch as well that is used. If you push on it, you can actually see the little impression on the chad. BECK: Now, from your experience in the -- in the industry -- we've been talking about how it's -- how it's possible to leave an indentation with a stylus. From your experience in the industry, the fact that there is, let's say, a particular chad that we're confident the imperfection did come from the stylus -- does that mean that the person intended to vote but was unable to push that chad through?

AHMANN: I seriously doubt that the voter would be unable to push the chad through on a normal voting device.

BECK: Now, let's talk about how it might be possible for somebody to attempt to vote and just not push the chad through. Were you here the other day when one of the witnesses for Vice President Gore talked about how sometimes people will not follow the directions that your company prints up and they put on the back of the machine and instead they'll put their ballot on the -- on the top of the unit? Does that sometime happen in your experience?

AHMANN: If I may go back to my experience, in 1968 -- I don't know how many hundreds of precincts I actually observed in the Robert Kennedy election, but in Los Angeles, out of a whole precinct, we would have maybe two cards, and, invariably, every other precinct, there'd be two cards that someone -- two voters had voted upon, but they had all hanging chad, and...

BECK: Now, let me just stop you there. You say they had all hanging chad.

AHMANN: That's correct.

BECK: If somebody doesn't follow the directions and puts the ballot either on the top of the -- of the ballot folder or off to the side, is there a pattern that shows up of chads that are not fully dislodged?

AHMANN: Very definitely. Normally, if the voter doesn't use the vote recorder, what normally happens is it will -- the pattern will be very definite that they have either hung most of the chad, or they've either got dimples, or they hang it, and so the whole card will be that way. And this is how they were in 1968 with those cards we had there.

BECK: In your work with different voting officials in the different states, have you become familiar over the years with the regulations that are in effect -- that are in place in order to evaluate whether an imperfection in a chad should be treated as a vote?

AHMANN: I am familiar with those, yes, not in every state.

BECK: Are you familiar with the California...

AHMANN: Yes, sir.

BECK: ... regulations?

AHMANN: Very definitely.

BECK: I'm going to put up on the screen our Exhibit 52...

UNIDENTIFIED MALE: Your Honor, I presume he's giving legal testimony of what the standards in California -- we are OK if he wants to talk about Texas as well.

UNIDENTIFIED MALE: They can put up...

SAULS: Do you have an objection?


SAULS: What is it?

UNIDENTIFIED MALE: The objection is it calls for a legal statement from the witness.

SAULS: Overruled.


SAULS: He's testifying about his experience and what his knowledge is of the standards.

BECK: The -- is this the front page of the California standards?

AHMANN: It appears to be.

BECK: I'm going to skip over to the 29th page, and does this page contain information about the standards employed in California to ascertain voter intent according to these imperfect chads?

AHMANN: Yes, it does.

BECK: Where -- just tell me where on the screen, and then I'll blow it up so everybody can see it.

AHMANN: Well -- when voter marks his ballot by indentations, impressions for all races and measures...

BECK: So this is one?

AHMANN: It's number two, yeah -- number one, yeah.

BECK: OK. We'll do it here. And what does this say here?

AHMANN: Voter marks his or her ballot by indentations, impressions for all races and measures, and there are no overvotes, it shall be counted.

BECK: And then I'll see if we can focus over under the -- under the notes column, where that -- where that same indication...

AHMANN: It says voter intent can be determined.

BECK: OK. So...

AHMANN: Duplicate the ballot, and then it goes in with the ballots to be counted by machine.

BECK: So this situation -- do you understand that this situation deals with that hypothetical we were talking about before, where somebody doesn't put the ballot in the machine, and they -- so they never really fully dislodge any of the chads.

AHMANN: That's correct.

BECK: Or, I suppose, it's possible for someone to be so feeble that they consistently apply just barely enough pressure to form a dimple but not enough to knock the chads out.


BECK: Then what -- then we move down to the next item here, and what does this say under the -- under the column, situation? What's the next situation that's dealt with in the California regulations?

AHMANN: Voter marks his or her ballot by indentation, impression on selected races and measures, however, cleanly punched holes on other races, measures.

BECK: So that is a ballot where there's -- less than all of the races are these imperfectly punched chads?

AHMANN: Yes. It says voter intent is unclear with an indentation, impression on parts of the ballot card. Count properly cast votes only.

BECK: So in that situation, at least in California, the X indicates that you're not supposed to count it. Is that right?

AHMANN: Yes, in that column.

SAULS: What did you say, in that column?

AHMANN: Yes. There's two columns on the chart here. The first one is to count, and then the other column is not to count.

BECK: I'll put it back up on the screen here, Your Honor.

AHMANN: See, the don't count is the top of the...

BECK: Up on the screen, Judge, under one, where all they leave are indentations for all the races...

SAULS: I understood that.

BECK: ... there's an X underneath the column, count.

SAULS: I understand.

BECK: And then under two, where they have some of them that are -- that are fully dislodged, then there's an X under the column that says don't count, and then there's an explanation that you don't count because the people figured out how to vote for some of the races, and, therefore, it's not right to say that their indentations somehow were intentions to vote in the other races.

SAULS: I understood that. That could be anywhere clear on -- punched through on any other -- in some other column on the ballot, and I asked -- I understood the witness to say it was limited to a column by column basis. I was just trying to understand what he said.

BECK: Oh, I'm sorry. I thought you were talking about the document, so we had some confusion. Sir, would you explain to the judge how this is implemented in practice? Does that mean one column on the ballot or...

AHMANN: No, it's the entire ballot card. It's the entire ballot card you're looking at.

BECK: Are you also familiar with standards that were in place before this recount commenced, or maybe partly into the recount, in Palm Beach in Florida?

AHMANN: I am familiar with -- I have read it, and as far as I know, it was very similar to the California. I don't recall whether or not the dimpled ballots were counted here, but it does say that the -- it had to be -- the chad had to be loosened on three sides, which is a two-point hanger, as we call it, and those need to be counted.

BECK: Let's take a look at the -- at our Exhibit 46, which was a document from, I think, 1990 in Palm Beach. The date was November 2, 1990. And, first, on this document, they quote the instructions that are in the voting machines that say to vote, hold the punch straight up and punch down through the card next to the preferred candidate's name or issue presented. And is that the -- that the big label that you yourself manufactured and put in the back of the machines?

AHMANN: I believe it's the vote recorder label. It's on the instructions to vote on the left-hand side, which is also enlarged and put in the back of the booth as well, yes.

BECK: And then the document continues -- the guidelines assume that these directions have been understood and followed, therefore, a chad that is hanging or partially punched may be counted as a vote since it is possible to punch through the card and still not totally dislodge the chad. I'm going to just stop here for a minute.

Do you agree, sir, that it is possible, when using the Vote-o- matic, to punch through and the chad won't actually break off all four corners, but will just partially loosen on two or three corners?


BECK: Then it says: but a chad that is fully attached, bearing only an indentation, should not be counted as a vote. An indentation may result from a voter placing the stylus in the position but not punching through -- let me continue on here. So here we were -- we're now up at the top -- a chad that is -- but a chad that is fully attached, bearing only an indentation, should not be counted as a vote. An indentation may result from a voter placing the stylus in the position but not punching through, thus an indentation is not evidence of intent to cast a valid vote.

Now, sir, from your years of experience in the industry, do you agree with the -- with the approach that was adopted on November 6, 1990, by the canvassing board of Palm Beach County here in Florida?

AHMANN: As far -- yes, I do, as far as it goes.

BECK: OK. I want to turn now to -- back to your days at IBM, and in the 1960s, did you participate in a test at IBM concerning the effect on the T-strips of extended use by -- extended impacts by a stylus?

AHMANN: Yes, sir. I did.

BECK: Would you please describe for the court the test that you...

UNIDENTIFIED MALE: Your Honor, this is where our objection -- this (UNINTELLIGIBLE) -- this is where our objection that we raised earlier, where he said he could not testify because he wasn't an expert. (UNINTELLIGIBLE) if I could hand up a copy of the deposition to the court.

BECK: Your Honor, I may be able to short circuit this. I'm not going to ask him about chemical formulas. I'm not going to ask him anything about the areas that were covered yesterday by our chemist. What I'm going to ask him, as a mechanical engineer and a person who conducted the tests, what were they testing for and what did they see?

SAULS: That'll be permissible.

BECK: OK. So now please describe for the court what you were doing in this test at IBM. Tell us when it was and tell us about the test.

AHMANN: I'm trying to think of the actual year. It was probably 1967, probably the spring of 1967. We had just come up with a new Hazelet (ph) format, which is the 228 format. We, at that time, named it after Ken Hazelet (ph), the programmer who came up with it.

We wanted to do a life test on the new vote recorder that we had, as we had improved it with the props on the bottom and the pads on the side to properly locate the ballot card and align it with the T- strips. We wanted to do a test that would determine not only how much chad the vote recorder could take before having to clean, but we wanted to really determine how many years this device would work by giving it a functional test and having all 228 positions of the template punched for the equivalent of 15 years or 10 years of use.

BECK: How many times -- well, did you actually have somebody sitting there punching ballots over and over again so that the stylus was hitting the rubber? AHMANN: We had what you would call, I think, a double blind test. We had two men punching out the cards. Every card was punched, all 228 positions. We punched out 15,000 cards per voting device. We wore out many templates. We broke many styluses, but we proved that you did not get a degradation of the T-strips until it was beyond 10,000 punches in every position, and it was still punching reliably up to 15,000 punches.

BECK: And when you did get to 10,000 punches on the T-strips, when there was the degradation, was it that the rubber became harder, or was it that something else happened?

AHMANN: We experienced, as I have witnessed in many vote recorders that have been used since 1966 ...

UNIDENTIFIED MALE: I'm just talking about this test now.


UNIDENTIFIED MALE: Was it -- did the rubber become harder, or did something else happen?

AHMANN: Well, what I was stating was that I have seen other devices that looked similar to those since then, and what happens is the -- you get a witness mark of the chad on the rubber. You can actually see that. And the rubber starts to deteriorate at the edge where the stylus goes through, which in turn makes it easier to punch. And the rubber does not support the chad.

So the incidence of hanging chad increases the longer you have the device, the more the T strips wear.

UNIDENTIFIED MALE: How about, in this 15,000 times -- how many holes are in the ...

AHMANN: Two hundred twenty-eight in that format.

UNIDENTIFIED MALE: OK. So you got 228 holes that you are punching through, and you got all these T strips underneath, and you punch all 228 15,000 times, right?

AHMANN: That's correct.

UNIDENTIFIED MALE: And someone could do the math, and I haven't, but that would be 15,000 times 228. In all those times, when someone was putting the stylus through, was there ever occasion where they made a dimple instead of either a fully dislodged chad or the occasional hanging chad?

AHMANN: Not that I recall.

UNIDENTIFIED MALE: The last subject I want to deal with is this supposed phenomenon called chad buildup. Were you here in court the other day when the political scientist called by Vice President Gore's legal team talked about his hypothesis that piles of chad build up underneath a ballot, and that might keep somebody from casting an effective vote?

Were you here for that testimony?

AHMANN: Yes, sir.

UNIDENTIFIED MALE: Now, your experience over the years, designing and implementing these voting systems, is that even remotely likely to happen?

AHMANN: Definitely it can happen.

UNIDENTIFIED MALE: Objection, your honor. No proper foundation. (UNINTELLIGIBLE) talk about the specific machines. There are different kinds of machines (UNINTELLIGIBLE).

SAULS: He designed the machines.

UNIDENTIFIED MALE: Well, there are different machines in the different counties, your honor.

SAULS: All right. (UNINTELLIGIBLE) talk about the specific machine.

UNIDENTIFIED MALE: The model 3A and model 3P you actually designed, right?

AHMANN: Yes. The vote recorder, yes, sir, I did.

UNIDENTIFIED MALE: And one or the other of those models, the difference is that -- aluminum case versus a plastic case, but that area underneath where the chad fall in, is it the same size in either model.

AHMANN: It's -- the same blue base is used in both -- same mold.

UNIDENTIFIED MALE: So, for those machines, is it even -- and you say those are what were in place in Broward and in Miami-Dade, right?

AHMANN: Yes, and for most part, in West Palm Beach as well.

UNIDENTIFIED MALE: OK. So first let's just talk about all of Miami, all of Broward, and for most of Palm Beach, for those machines, is it even remotely possible that little mountains of chad are going to build up underneath Al Gore's name, and somebody is going to try to vote for them, and they're going to be thwarted and they're going to leave a dimple instead of punching through that chad?

AHMANN: Well, first I would like to say, no, I do not believe it's possible. But let me qualify why, if I may.


AHMANN: The booths -- first of all, the booths are allocated to precincts, approximately one booth per 125 voters. So if you are looking at the first column, the punching (ph), then you're are looking at possibly 125 votes being cast in that column, in that particular booth per election.

UNIDENTIFIED MALE: I want you to assume that it was a really crowded election, and everybody went in same booth, and so there were maybe 1,000 people. Even assuming there was 1,000 people per machine, would it be remotely possible that the chads could build up and stop somebody from voting?

AHMANN: In the model 3A or 3P, I have to say no.

UNIDENTIFIED MALE: OK, And explain why that would not be possible.

AHMANN: OK. Anyone familiar with the punch press, or punch press principles, will tell you, as I will tell you, that as you are punching downward, the material that is being punched, especially when it is not in a captive channel, will tend to go outward, and they will form a pile upward (ph).

And we are dealing with a chad which is only .125 in length or width, .170 roughly in height, and seven thousandths thick. You have three-quarter of an inch clearance underneath that, seven thousandths doesn't take that much room.

And if they scatter out a little bit, which they do, because they just don't stack up and stack (ph) and go up there, you end up with an area that is approximately 16 times or more the length -- assuming that they don't push out even further; I'm just assuming they are going to go in a pyramid type pile -- you are going only to be approximately an eighth of an inch high.

If you had -- you know, figure 16 pieces or 20 pieces are going to lay on the bottom, that's seven thousandths. Then you go up a little higher, you might (UNINTELLIGIBLE) have a 32nd of an inch behind 100 of them.

UNIDENTIFIED MALE: You'd only get up a 32nd of an inch if you had 100 chad down?

AHMANN: Approximately, yeah.

UNIDENTIFIED MALE: All right. And that assumes that they don't get dispersed out as the stylus keeps going down.

AHMANN: That's is correct. It's an open area underneath. You have an area that's 14 inches by 9 inches, roughly, that's totally open. And the beauty of it is, with the model 3 concept, when the case is closed up at the end of election...

UNIDENTIFIED MALE: Well, let me stop you there.

Just with the size of the -- and we don't have the machine up here, so I can't actually reenact the -- yesterday's demonstration. But with the size of the opening down there, during the voting day, at a typical precinct, would it actually be necessary for somebody to grab that machine, like I think Mr. Kitchen (ph) did yesterday, and shake it over his head in order to make room to vote underneath Al Gore's name?

AHMANN: Not at all.


AHMANN: Not at all, if you only had the normal number of voters. If you got up in the thousands of voters, then you might need to, for a particular machine. That is correct.

UNIDENTIFIED MALE: Now, what about the demonstration that I reenacted, of carrying it around with a suitcase? I think it was Mr. Zak (ph) who pointed out that these things aren't cleaned out after every election. Is that your understanding?

AHMANN: Indeed they are not cleaned out after every election.

UNIDENTIFIED MALE: Let's say that somebody didn't clean it out even -- any more often than 10 years. Would that be a problem in terms of chad buildup underneath Al Gore's name versus everybody else on the ballot? .

AHMANN: I would first say no, but then that would also depend on whether you had an election every week or if you only had an average of two or three elections a year.

UNIDENTIFIED MALE: Or maybe even five a year.


SAULS: (UNINTELLIGIBLE) testified -- what -- three million pieces of chad can be accumulated in once (UNINTELLIGIBLE) how ever (ph) many elections does it take for 1.3 million...

AHMANN: In order to totally fill it up, you could probably 50 years, but...

SAULS: All right.

AHMANN: ... 8 to 10 years, it might not cause you any problem, depending on how many elections, how many votes have been cast on that unit.

UNIDENTIFIED MALE: And I think you probably were here yesterday when they actually brought the machines in.

AHMANN: Yes, sir.

UNIDENTIFIED MALE: And do you remember that, when they brought these machines in, they were holding onto handle just like a suitcase?

AHMANN: It's designed to be handled that way and transported. In fact, that's the way it's handled from the truck to the precinct, the precinct back to truck, then gets on the road and goes back to the warehouse where they take it and pick it up and they handle it again by the handle and put it into a stack. UNIDENTIFIED MALE: Now, when -- let's say that there's a mound of chad that's built up under the number 5 hole, which happens to be where Al Gore is in a particular ballot, during another election, and then the fellows pack up the machine and carry it in a suitcase and put it in a truck and drive over and carry about it in warehouse.

And then six months later they take it back on the truck and drive it back, and they bring it back in the thing and they open it up, and they haven't cleaned it out in years, is there going to be a little pile underneath the five hole, or what's going to have happen to those chad?

AHMANN: I can, with a great deal of assurance, guarantee there is not going to be a big pile underneath the five hole.

UNIDENTIFIED MALE: Is there any reason to believe that, after all this handling of machines, and they install them in Palm Beach and Broward and Miami, that somehow or another this phenomenon that's been hypothesized of chad buildup is going to stop people from voting for Al Gore in the five hole, but is going to allow them to vote for everybody else in rest of the holes in the ballot?

AHMANN: I know of no way that that could happen.

UNIDENTIFIED MALE: Now finally, sir, you said you're familiar with the Vote-o-matic, the machines that you made. Are you as familiar with the portable units, the Pollstar (ph) units?

AHMANN: They are a compact unit. I know the area that's beneath them as far as accumulation of chad and how you get them open, and I have looked at the T strips and things, and I know how they function. They're very similar to the Vote-o-matic, yes.

UNIDENTIFIED MALE: Do you feel that you are capable of expressing an opinion about the likelihood of this chad buildup hypothesis in the Pollstar units?


UNIDENTIFIED MALE: And what is your opinion about the likelihood that this chad buildup theory might operate so as to deprive Vice President Gore of votes in the Pollstar unit?

AHMANN: The Pollstar unit needs to be cleaned out more frequently, obviously, than the model 3, because it has no vast reservoir around it. And also, when you go to transport it, all of the chad doesn't drop down behind the top of the punch frame and go into the other half of the cavity.

So all of your chad is contained in this opening. There is a lower portion which, if it stood on end, could go down. That's where the cleanout hole is.

UNIDENTIFIED MALE: In fact, the other day, when probably Mr. Zak (ph) had one of those little Pollstar units and (UNINTELLIGIBLE) spilling chad all over the judge and then showed that there is a pile of chad at the bottom, is that the cleanout unit?

AHMANN: Yes, it is.

UNIDENTIFIED MALE: OK. So, if there is a little pile of chad in the bottom of the unit -- cleanout unit, is that going to stop anybody from voting for Al Gore?

Assuming that it has been cleaned out within, you know, one election cycle, or three or four elections, and it hasn't been overly used, more than 125 or 150 voters per election, there should not be a problem.

UNIDENTIFIED MALE: And we don't know those things, but if we assume the worst, that it hasn't been cleaned out a long time, and that for some reason having (ph) elections every week, is there any reason to believe that there is going to be chad buildup under Al Gore's hole and not chad buildup under any of other holes?

AHMANN: With only one punch position row 1 and a number of punch positions over in the center of the card, it would stand to reason you're going have to a buildup in the center of that unit and that's going to stop you from voting there a lot sooner than it does over on the left-hand column.

UNIDENTIFIED MALE: Is that pretty much what I went through with fellow the other day about there is a lot more offices (ph) in middle of the ballot than there is on the left.

AHMANN: That is correct. That is correct.

UNIDENTIFIED MALE: Thank you, sir.

HEMMER: For the past hour, the Bush team has clearly been trying to show how ballots and chad can be damaged when handled, and then for the past several minutes, the key witness on the stand now for Bush team, John Ahmann, trying to shoot down the theory about chad buildup.

Now we're going to leave this hearing momentarily, but we'll return in a couple of minutes now. But first of all, some of the key players on both sides, the Bush team and the Gore team making, the rounds today on this Sunday morning.



Back to the top  © 2001 Cable News Network. All Rights Reserved.
Terms under which this service is provided to you.
Read our privacy guidelines.