All right. Dr. Speed.
Terence Speed, called as a witness by the Defendant, was sworn and testified as follows:
Raise your right hand, please. You do solemnly swear that the testimony you may give in the cause now pending before this court shall be the truth, the whole truth, and nothing but the truth, so help you God?
Please have a seat on the witness stand and state and spell your first and last names for the record.
Thank you, your Honor. Good afternoon, ladies and gentlemen.
THE JURY: Good afternoon.
DIRECT EXAMINATION BY MR. NEUFELD
Okay. And could you please tell the ladies and gentlemen of the jury what your occupation is and what positions you hold?
And could you, please, for the ladies and gentlemen of the jury summarize your education?
I received a bachelor's degree in science majoring in mathematics and statistics at a University in Melbourne, Australia, and later a Ph.D. majoring in mathematics at another university, Monish University also in Australia.
And could you also summarize your professional appointments since the time you received your Ph.D.?
Okay. After I got my Ph.D., I went to the United Kingdom, worked for about four years at the University of Sheffield. And I went back to Australia, worked for nine years as a professor in mathematics at the University of Western Australia. Then I left academia for a few weeks and had a job in a government laboratory as chief of a group of mathematicians and statisticians, and that was about eight years ago. I left there and went to the University of California, Berkeley, where I've been ever since.
Now--one moment. Let's back up a second. You said that you taught at the University of Western Australia?
And while you were at the University of Western Australia, did you have a particular title at some point during that time?
Yes. Towards the end of my period there, I was chair of the department of mathematics.
And you also mentioned, sir, that--excuse me--Professor, that you took some time off and you worked for something called the division of mathematics and statistics?
That's a division of an organization with a very long title which I'll give, commonwealth scientific and industrial research organization, which is a very large multi-disciplinary organization spread all around Australia, doing research in those areas, and I had a division of mathematics and statistics with about a hundred mathematicians and statisticians again located all around Australia, and I was the chief of that group.
You weren't something one of the hundred in that department. You were actually the chief of it?
Okay. And you mentioned that currently you are a Professor at the University of California at Berkeley. Is that referred to as Cal or Berkeley? What's the easiest way for me to refer to it for the future so I can defer? Is it Cal?
Okay. All right. During the time that you were at the University of California at Berkeley in the statistics department, were you ever made the chairman of that department?
Yeah. I was chair for four years, 1989 to `93, and then I went on a sabbatical and gave up the chair.
Could you tell the jury and myself and other counsel in the court, what is statistics?
Well, a one-line definition might be something like this. That is the use of numerical information or it's the art and science of using numerical information to answer questions of interest to people.
Yeah. One example might be, why does the casino always win at roulette? Another one might be, how do I design an experiment to test whether a new drug or a new medical treatment is better or not better than an existing one? A third one might be, how do we design sampling procedures so we can predict the outcome of something like an election with a very, very small number of people, just a few thousand out of a country the size of the United States.
Can you give us some--I'm sorry. Are you aware of any examples of where good statistics are used to overthrow the--some accepted, but nevertheless erroneous view?
Certainly. There's lots in the medical area where new medical techniques and in former days, new drugs, became available, and without proper testing, were found by the proponents to be effective. Yet, when well-designed statistically valid studies were conducted, they were found to be ineffective. So it's now quite commonly used to screen drugs, not quite as commonly used to screen surgical procedures and other medical inventions.
All right. Are you a member of any professional organizations within the field of statistics?
Well, I'm a member of statistical societies, the major ones in the United States, which are the American Statistical Association and the Institute of Mathematical Statistics, remain active in the Australian Statistical Society and one in the United Kingdom and another one, the Biometric Society.
Well, that's the professional organization of people with a somewhat more theoretical vent, mathematical statistics, rather than just statistics.
And you also mentioned that you were a member of the Biometric Society. What is the Biometric Society, Professor?
Well, that's the society of people who are interested in applying statistics to biology and medicine.
Yeah. I've had the presidency of one of the regions. The--that's a worldwide society which is divided up I think into about 10 regions, and for a period a few years ago, I was the president of one of the regions.
And by the way, the Biometric Society which you said is involved in the interface of statistics and biology and medicine, does it also involve the application of statistics to genetics?
And by genetics, I mean, does it also involve the application of statistics to DNA typing?
Yeah. I think genetics these days includes molecular genetics, which is pretty much things related to DNA.
Now, are you a currently--are you currently a member of any advisory body which focuses on human genetics research?
Yeah. I'm a member of the--what's called the human geno study group formed by the national institutes of health.
Well, as people may know, there is a thing going on these days called the human geno project, which is a very detailed study of DNA in a typical human, and there's a whole institute of the national institutes of health divided to research and promoting the broad study of that area; and they give away many millions of dollars each year to researchers around the world, but mainly in the United States, and the greatest--or the proposals that come in are assessed by a study group of which I am a member.
Now, other than your appointments in these various organizations, Professor, have you served on the editorial boards of any peer review journals in your field?
Yeah. I've served on the editorial board of most of the major journals in my field, although right now I'm only--I've got off most of them recently and I'm on only one at the moment.
And aside from your editorial appointments, Professor, have you authored any technical papers in the field of statistics?
I think it's, you know, about a hundred, 120. I'd have to just have a glance. Yeah. I think up to about 120.
Could you please tell us a little bit what these papers are about and how they've evolved over the years?
Well, shortly after I did my Ph.D., most of them were somewhat theoretical in nature, but over the last decade or so, I've got increasingly involved in the application of statistics to a variety of disciplines, but primarily biology and more recently genetics.
And even before you became involved in the field of probabilities in genetics and in DNA typing, were you involved--did you--I'm sorry--did you write a paper on nuclear reactor accidents?
And did that paper involve the likelihood that certain errors can occur which can cause a nuclear mishap?
Yes. That was--it was a critique of a big study conducted by what is now called the nuclear regulatory commission to try to find the accident--the probability of risks of an accident from a commercial nuclear reactor.
And in your opinion, was your work on that paper at all related to any of the issues that you're going to be testifying to today?
Well, in the general sense that that paper was a critique of the statistics that was used in that study. I was very critical of the study because in my view, it misused statistics quite badly, and to some extent, there was an element of that in the use of statistics in DNA profiling. To that extension, there's a similarity.
Over the last four or five years, approximately how many articles have you authored on the application of statistics to genetics?
Do you believe you need to be a forensic scientist before you can give an expert opinion on the application of statistics to forensic DNA profiling?
I certainly hope not, because if that were the case, you would have to be professional in any discipline to which statistics was applied before you could come forward, and that's not the way things work.
Well, if a statistician is going to make comments on the use of statistics in another discipline, then obviously they need to be familiar to some extent with that other discipline and, in particular, to be able to appreciate the way statistics is being used. They certainly don't have to be professionals in that discipline. If they were, they wouldn't be statisticians.
Now, how did you become interested in the application of statistics to forensic DNA profiling?
Well, the interest in statistics and genetics is long-standing, and I think it was around 1990, there was a one-day symposium on actually that topic, statistics in DNA identification done here in L.A., and I and a few students came down and attended it and we got our first taste of the subject.
As a result of that first taste on the subject five years ago, did you continue to do reading and thinking and participating in scholarly discussion about that subject?
Have you been asked at any time during the last five years to actually become involved in criminal cases involving the application of statistics to forensic DNA profiling?
And approximately how many times have you become involved in cases of that nature other than this case?
And in those two cases where you became involved, did you consult for the Defense or the Prosecution?
Okay. And in the course of working on those cases, did you have an opportunity to review database autorads produced by a DNA laboratory?
Well, we made a visit to Quantico, which is where the FBI research, DNA research operation is based.
Now, in addition to those two cases that you were involved in, have you remained interested in the issue?
Have you organized and/or attended any statistics--statistics conference sessions devoted to the forensic issues?
By the way, have you ever been approached by law enforcement to become involved in this issue to assist the Prosecution?
I think it was about spring 1993, I got a call from the people at the Berkeley lab, which is referred to here as DOJ.
Well, I was just invited to come and look around the lab and chat with people because they had heard, presumably through my involvement with these cases, that I'm interested in the subject. So we--I looked around and we exchanged views.
And at that time with that meeting at the Department of Justice laboratory, did you share your thoughts and criticisms of the statistical applications being used at the laboratory?
Now, aside from that experience you had at the DOJ laboratory when they invited you there to view their laboratory work and then to discuss these issues with them, have you ever on any other occasion assisted law enforcement?
Uh, I think it was early 1993, I was approached by people from--well, they were based in the Alameda County District Attorney's office, but one of the people was a federal Prosecutor I believe.
And to your knowledge by the way, Professor Speed, is Mr. Harmon here a member of the Alameda County District Attorney's office?
Well, they were working on a case involving the possibility of the sale of contaminated meat in a supermarket chain, that being quite a number of tests proved to be positive for contamination, and I wanted to organize a sampling plan to see if this was systematic and not just occasional contamination.
And so did law enforcement at that time use your expertise in the field of statistics to help them design the appropriate study for the Prosecution?
By the way, when you did that work for the Alameda County District Attorney's office back in 1993, did you--were you--did you personally receive any compensation for that work?
Would I be correct in saying, Professor, that this is actually though your first time testifying in a court of law on the forensic application of DNA?
Are you a little bit nervous? No? Okay. Now, you mentioned that in the two times where you assisted the Defense, that you didn't charge any compensation and you mentioned at the time you assisted the Prosecution in their case, that you didn't charge any fee as well. In this case, are you accepting any fee, any money at all in exchange for your time and your expertise and your testimony?
Now, Professor, are you familiar with Dr. Bruce Weir, who testified in this case as a Prosecution witness?
Has Dr. Weir ever sought your input on his writing on statistical applications of forensic DNA issues?
Now, Dr. Weir testified that he had no background in forensic science and no training in forensic science. Do you have any training at all in that area?
After you became involved in this case, Professor, did you review any of the testimony of different witnesses?
I've reviewed Dr. Weir's, I've reviewed Mr. Sims', I've reviewed Dr. Cotton's and I've seen some of the testimony from the LAPD of Mr. Yamauchi. So--
Have you also--are you also familiar with the testimony that Dr. Gerdes gave during the last several days?
Have you reviewed any of the open proficiency test data that these laboratories talked about?
Now, other than reviewing the materials that you described, have you met with the lawyers to discuss these issues?
And what, if anything else, have you done to prepare yourself for your testimony today?
As an expert in the field of statistics, Professor, is it the case that sometimes people misinterpret the meaning of a statistical statement?
Well, I think when people make statements, particularly statements about probabilities or frequencies, it's often unclear just what is the event, the precise event, whose probability or whose frequency is being spoken of. It's also often clear what is the assumptions under which this calculation is being made, and how the calculation is being made is also relevant, and that's very rarely something people know.
Professor, have you seen small Xerox copies of the various evidence boards that have been prepared by the Prosecution? And I'll just use this as--for illustrative purposes, the--
Okay. And have you also seen reduced Xerox copies of the other boards for Rockingham DNA profiling, Bundy DNA profiling and the glove DNA profiling?
Now, let me ask you this apropos of what you just said, sir, about the meaning of statistical statements. For instance, are the frequencies reflected on the last column on these evidence boards, are they the probability of Mr. Simpson's guilt or innocence?
Are the frequencies on these evidence boards the probability that someone other than Mr. Simpson is the source of any particular stain?
What do these numbers tell the jury about the probability of a false or misleading match being reported due to errors in the collection and handling of samples?
What do these numbers tell us about the probability of a false or misleading--I'm sorry. What do these numbers tell us about the probability of a false or misleading match being reported due to evidence tampering?
Are these numbers on the boards relevant if the matches are determined to be due to cross-contamination?
Now, are you familiar, sir, with Dr. Robin Cotton's testimony about her reporting of how Cellmark did on the California Association open proficiency testing? Do you recall that?
And do you recall when she testified that with respect to one of the batches of 50 samples that they received, that--there was a match that she gave as a--gave a frequency for of 1 in 1.8 billion? Do you remember that?
And do you remember her testimony when she said that later on, the submitting agency told her that they had a false positive? Do you recall that, sir?
And do you recall her then saying that having learned that it was a false positive, that the statistic of 1 in 1.8 billion was irrelevant? Do you remember that?
Now, what, if anything, do these frequencies tell us about the probability that swatches were mixed up at the Los Angeles Police Department?
Now, Professor, Dr. Cotton, Mr. Sims and Dr. Weir have each given their own opinion as to how rare particular DNA profiles are in this case. Are you familiar with their testimony?
In order to assess the evidentiary significance of these DNA typings, are there other frequencies that are also important?
And I think you said that you're also familiar with the testimony of Dr. Gerdes regarding the various opportunities to err in this case?
Now, when you're thinking about the other frequencies which would also be important to consider, is it also too important--I'm sorry--is it also important from a statistical standpoint to consider the chance that the reported matches are the result of error in crime scene collection and packaging?
Is it important to consider from a statistical standpoint the chance that the reported matches are the result of error when the evidence is unpacked at the Los Angeles Police Department?
Is it important to consider from a statistical standpoint the chance that the reported matches are the result of error when the evidence is put into bindles on the next morning?
Is it important from a statistical standpoint to consider the chance that the reported matches are the result of error when the Bundy drops and Mr. Simpson's reference sample are processed by Mr. Yamauchi?
What, sir, would be the importance--well, is there any importance at all to--from a statistical standpoint to assessing the chance that reported matches are the result of error when the Bundy drops and Mr. Simpson's reference sample are processed by Mr. Yamauchi at the LAPD?
Now, again, from a statistical standpoint, is it important to consider the chance that the reported matches are the result of error in the DNA testing laboratory, for instance?
Is it also important to consider the chance that a reported match is a true match, but that the source is someone other than Mr. Simpson, but someone who has the same DNA profile of him at the loci that had been studied?
Okay. From a statistical standpoint, is it important to consider the chance that a reported match is a true match?
Okay. And is it important to consider that--from a statistical standpoint, that the source of that match is someone other than Mr. Simpson, but that someone is simply someone who has the same DNA profile as Mr. Simpson?
And in fact, Professor, in response to that last question, that the Prosecution experts listed those frequencies in the final column of those various boards?
Yes. As far as I can see, that's the only question that those frequencies speak to.
From a statistical standpoint, Professor, is it important to consider the chance that it may really be Mr. Simpson's blood, but that the drops were left by Mr. Simpson at some earlier date?
Okay. From a statistical standpoint, if you were to consider simply for the purpose of this question, the Bundy--the Bundy drops on the walkway, sir, would it be important to consider the chance that it may really be Mr. Simpson's blood on that walkway, but that those drops were left by Mr. Simpson at an earlier date?
From a statistical standpoint, is it important to consider the chance of various alternative hypotheses to explain certain data?
Okay. Are you familiar with Dr. Weir's testimony where he said that all of his calculations assume the validity of all the DNA testing done in this case?
From a statistical standpoint, is Dr. Weir's assumption scientifically appropriate in your opinion?
Well, I believe that the possibility of errors can never be ruled out and, therefore, there is an incompleteness in any evaluation that is based on the assumption that no errors could have happened.
Now, Professor, are these views criticizing Dr. Weir's position just yours or have you considered the opinions of other experts in reaching your conclusions on this point?
All right. In reaching your opinion that you just gave about Dr. Weir's conclusion, did you consider the report issued by the national research council of the national academy of science?
By the way, Professor, are you aware of the views of other experts in your field on this particular issue that I just addressed?
And have you considered the views of other experts in helping to reach your opinion that you're providing this jury with today on the witness stand?
And I believe you said that one source that you have considered in giving your testimony today are the conclusions of the national research council, the national academy of science; is that correct?
And could you please tell us which portions of this book, of this report that you considered in helping to frame your position on the witness stand today?
Well, could you please tell us the first--can you give us a page reference, please?
Well, perhaps I'll start with the very first one, which is a page that doesn't actually have a number. That is opposite the contents page. It's a discussion there of errors.
Mr. Neufeld, let's do this by question and answer rather than having the doctor read it from the book.
All right. Professor, could you turn to page 160 first. In reaching your opinions as an expert on these issues involving the statistical applications to forensic DNA profiling, have you considered the following quote? Reading from the bottom of page 160. Quote--
Well, your Honor, I don't think it's appropriate. If he says no, then the content of that's not--
All right. Would you please look at the paragraph beginning at the bottom of page 160 and ending at the top of 161.
Is that one of the quotes that you have considered and taken into account in helping to develop your opinion that you're giving on the witness stand today?
And I'm quoting now from the NRC report at the bottom of page 160. Quote: "Expectations regarding the parallel of DNA typing can lead to overlooking or ignoring sources of error or mistakes in applying the technology. For example, jurors focusing on the probability of correctly identifying a perpetrator might lead them to discount the possibility of laboratory error, whether it stems from incompetence or carelessness of personnel, malfunctioning equipment or unavoidable mistakes." Did I read that accurately?
Well, I think it says that you should be conscious not just of the frequencies of the particular multilocus genotypes that are quoted, but also be aware of the possibility of errors in the--of the kinds that have been stated and the frequencies with which these errors might occur and contribute to a wrongful result.
Now, referring you specifically to different sections of this report, have you considered--I'm sorry. You began to say before that you considered the--a portion of the preface or prefatory language?
Now, I call your attention, sir, to the page directly preceding the table of contents.
In the third paragraph, did you consider--in helping to develop your own opinion on these matters, consider that first sentence in the third paragraph?
Now, the report says that, quote: "We regard the accreditation and proficiency testing of DNA typing laboratories as essential to the scientific accuracy, reliability and acceptability of DNA typing evidence in the future." By the way, do you know when the--this report was published?
Well, that the authors of the report think it important that agencies that do DNA testing get accredited and participate in proficiency testing. Otherwise, the scientific accuracy, reliability and acceptability might be compromised.
And the specific portion that I'm referring to--one moment--beginning with the second paragraph on the page.
And is that one of the portions of this report that you have considered in reaching your opinions today on the witness stand?
Okay. Now, I'm going to ask you if you agree with the following quote from the NRC report. "Interpretation of DNA typing results depends not only on population genetics, but also on laboratory error. Two samples might show the same DNA pattern for two reasons: Two persons have the same genotype at the loci studied or the laboratory has made an error in sample handling procedure or interpretation. Coincidental identity and laboratory error are different phenomena. So the two cannot and should not be combined in a single estimate. However, both should be considered." What are the authors saying at this point, Professor?
Well, it states very strongly that in addition to the frequencies that are associated with coincidental match, there should be some estimate of frequency of errors, the different kinds.
Because both can contribute to an incorrect conclusion about the source of a DNA profile or the source of a stain.
Now, sir, in addition to that paragraph, paragraph on page 88, did you also--can I call your attention to page 89? And I call your attention to the first paragraph. Is that a paragraph or portion of this report that you have considered in the development of your own opinions that you're giving on the witness stand today?
I'm quoting from page 89: "Especially for a technology with high--" I'm sorry. "Especially for a technology with high discriminatory power such as DNA typing, laboratory error rates must be continually estimated in blind proficiency testing and must be disclosed to juries. For example, suppose the spans of a match due to two persons having the same pattern were 1 in 1 million, but the laboratory had made one error in 500 tests. The jury should be told both results. Both facts are relevant to a jury's determination." In your opinion, sir, are both facts relevant to a jury's determination?
Because, as I said earlier, there are two different ways in which incorrect conclusions can be reached. One is a chance coincidental match and the other is that the result could be as a result of laboratory error.
And, sir, again, in helping to develop your opinions of this particular matter, did you also consider page 15?
Your Honor, before we go any further, I have been reminded by Mr. Cochran that I did not ask for a number for this for the next in order. It's my recollection that this exhibit was shown to someone else.
May we approach, your Honor, because I think Mr. Neufeld has spoken some interesting words. May we approach, your Honor?
Well, I'd like to discuss what he just said and what he told me before we started, your Honor.
Referring you now to page 15, and I quote from the section. It says: "Laboratory error rates should be measured with appropriate proficiency tests and should play a role in the interpretation of results of forensic DNA typing." Do you agree with that statement?
It means that there should be appropriate proficiency tests where appropriate remains to be defined for establishing laboratory error rates which then should be available for interpreting the results of forensic DNA testing.
Referring you now to the third paragraph on the page, have you considered that portion of the report also of the national research counsel in helping you develop your own opinion on these matters?
And again, Professor, I'm quoting from the national research council's report, page 89, where it states, quote: "Reported error rates should be based on proficiency tests that are truly representative of case materials with respect to sample quality, accompanying description, et cetera." What does that--first of all, do you agree with that?
Well, it means that the reported or the error rates that should be reported should be based on proficiency tests which are in as many ways as possible similar to actual cases.
And also of Collin Yamauchi with respect to proficiency testing that they've had at their three laboratories?
When they describe the type of proficiency tests that they had, did they state that they were using samples that were representative of casework?
Are you familiar with the testimony of Dr. Cotton, Mr. Sims and Mr. Yamauchi that they did not have degraded samples in their proficiency tests?
Does the NRC report declare elsewhere the necessity of relying on external blind proficiency testing?
Do you agree, sir, with the statements expressed by the authors of the NRC report contained at the bottom of page 106 and the top of 107?
And in those statements, sir, does it talk about the necessity of external blind proficiency testing?
Oh, by the way, from a statistical standpoint, Professor, why is it necessary that laboratory error rates be calculated on the basis of external blind proficiency tests administered on a continuing basis?
Well, let's start at the beginning. Why is it--I'm sorry. Why is it essential that they be external proficiency tests?
Well, I think that's so that the tests will have credibility, so that there will be uniformity in their application, so that the results can be widely disseminated and also, in fact, so that they can be fully blind. It's rather hard to imagine blind in the sense that we're talking about, tests which are internal.
Okay. And by the way, does the NRC report define what they mean by "Blind external proficiency testing"?
Do you agree with the definition given by the national research council in their report?
Well, "External" means, of course, that the agency doing the proficiency testing is separate from the body that's being tested. "Blind" means that the people in the organization being tested are not aware that the samples they are processing, the test samples, are any different from any other samples that they are processing in their normal casework.
Are we about to wind this up? I mean, this is the fifth witness who's told us about the difference between blind and open proficiency testing.
KEY QUOTEMoving on from blind and external, why is it essential that the testing be done on a continual basis?
Well, for a start, if people make errors, then they should obviously take steps to correct them so such errors don't occur in the future. So that you need a continuing program of testing so that you see that the corrective measures you took are actually working. And also, of course, your error rate may be going down if things are getting better. So you want the most readable, up to date estimate of your error rate.
Can you give examples of external blind proficiency testing to calculate laboratory error rates in other applications?
All right. Are you familiar with Dr. Gerdes' testimony regarding the national marrow donor program in which his laboratory participates?
Are you familiar with the use of external blinds in urine testing for the presence of illegal drugs by any governmental agency in this country?
Well, the federal highway administration does it and I think a number of others do. They certainly do.
Well, the urine samples are tested for illegal drugs by accredited testing organizations, and the samples should be sent to them by the people, you know, by the employers of the drivers, and they're instructed how many samples that they should include that will be blind to the testing agency, which will either have no drugs or there will be some that are spiked to have specific drugs; and then the test will be whether the testing agency actually detects the ones that have no blood--no drugs and finds the appropriate drugs in the ones that have been spiked with drugs.
By the way, even outside the fields of chemical analysis, are there other examples that you can think of where external blind proficiency testing is utilized in a similar way?
Yes. I've made it in the mining area where if you go out and do some prospecting and you have some samples that you think contain valuable ore, you're going to send them off to a chemical laboratory for analysis to say, for example, gold or silver or copper, then you're very interested in whether you got accurate determinations of the concentrations of these samples. So it's very, very common in that area to include samples with known concentrations out of a whole variety of levels so that you can have a check with your unknown samples on the laboratory doing the work. And the laboratory, of course, doesn't know which are the new samples and which are the known samples. This is quite routine in the mining industry.
In this particular case, are you aware that Dr. Weir testified that he rejects the notion of an error rate, page 33--
Did you review that portion of Dr. Weir's testimony where he gave his--I'm sorry--where he gave his opinion on the relevance of error rates in forensic DNA profiling?
This is not as I precisely recall, but I believe she said, "I made errors in the past and I don't make them anymore."
KEY QUOTEYour Honor, I object. That calls for speculation. That's not what she said. That misstates the testimony.
Overruled. The jury can compare the two and see if there's a difference that's significant.
Are you familiar with the opinion that Dr. Cotton gave on the role of laboratory error rates in evaluating the weight of DNA evidence?
Now, in your opinion, sir, does the NRC report also disagree with the opinions of Dr. Weir and Dr. Cotton on that point?
Well, as an expert who is asked to interpret those portions of the report that deal with statistical issues, what is your opinion as to whether the NRC report rejects or disagrees with the opinions of Bruce Weir and Robin Cotton?
Well, does the NRC report endorse the use of error rates based on external blind proficiency testing as a necessity for interpreting forensic DNA profiling?
Are there other experts besides the NRC report whose opinions you have considered as well?
Okay. In the field of statistics as opposed to--I'm sorry--withdrawn. In the field of statistics as applied to genetics, who would those other experts be whose opinions on this matter you have listened to?
Objection. The names, your Honor--it's clear where we're going. This calls for hearsay, no foundation.
Well, a number of professors of statistics that I know who are interested in or knowledgeable about genetics.
And could you please tell us the names of those people whose opinions you considered?
Would you please tell us the names of the experts in the field of statistics as applied to genetics whose opinions you have considered on this particular question?
Dr. David Balding, B-A-L-D-I-N-G, Professor Peter Donnelly of the University of Chicago, D-O-N-N-E-L-L-Y, Professor Elizabeth Thompson of the University of Washington in Seattle.
Nothing.
I believe he stated that he rejects the notion of an error rate.
This is not as I precisely recall, but I believe she said, 'I made errors in the past and I don't make them anymore.'
Especially for a technology with high discriminatory power such as DNA typing, laboratory error rates must be continually estimated in blind proficiency testing and must be disclosed to juries.
Are we about to wind this up? I mean, this is the fifth witness who's told us about the difference between blind and open proficiency testing.