Cirque de Landis: Can anyone speak French?

Following the first day's swing from enthusiastic and dramatic opening statements to the more mellow...

News feature: Floyd Landis' arbitration hearing, May 16, 2007

Cirque de Landis: Can anyone speak French?

Following the first day's swing from enthusiastic and dramatic opening statements to the more mellow (but equally dramatic) testimony by two expert witnesses, the second day began with the conclusion of cross-examination of one of the expert witnesses, Dr. J. Thomas Brenna, by Landis' counsel Maurice Suh.

Again, the examination centered around the measurement of uncertainty with the Isotope Ratio Mass Spectrometry (IRMS) test for synthetic testosterone. Dr. Brenna was asked about the differences in the IRMS data returns between the manual, or quality control step, and the auto analysis by the machine.

Q: Should [the difference] take into account measurement of uncertainty?
A: [Dr. Brenna] I think it should take it into account.

Q: Do you see any values being the same?
A: No.

Q: If you saw this process done resulting in this total variance, in your own lab, would you have been concerned?
A: Yes.

The final parts of the examination of Dr. Brenna focused on the retesting of the samples done at LNDD. The re-running of the samples produced a log sheet - which Mr. Suh commented that USADA had worked to prevent the Landis team from accessing. This log sheet indicated a re-rerunning of the test within ten minutes of the first analysis. The data from the first analysis was not saved, but overwritten by the second analysis, and multiple times on the same day.

The USADA side rebutted Mr. Suh's questions about the differences in the auto versus the manual steps of the test. As well, in regards to the overwritten log files, the USADA counsel asked if there was an opportunity to review the logs given to the Landis team. "[Dr. Davis, Landis' expert present at the testing] was given an opportunity to re-analyse the data on the system but there was confusion as to why he would want to," said Dr. Brenna. "The next day he came back and said he did not need to."

Lost in translation

The biggest drama of the day involved the French interpreter, Pierre Debboudt, of the National Court Reporters. Mr. Debboudt made noticeable errors with specifics of the testimony from the actual LNDD tester of the samples, Cynthia Mongongu. Mr. Suh finally objected at the point when the translator misinterpreted the quantification of "a day-and-a-half" to be "an hour-and-a-half." Mr. Debboudt was dismissed from his duties and the hearing was adjourned for 45 minutes while a replacement for Mr. Debboudt could be located.

More than an hour later the replacement, Martitia Palmer of the Judicial Council of California, resumed the translation duties. However, she was distressed by the fact that she was not able to familiarize herself with the specifics of the case and the terminology of the procedures. Nonetheless, the hearing pressed on. When the answer in question came up again, and Ms. Palmer correctly interpreted it, applause from the press room ensued.

The examination of Ms. Mongongu commenced with questions regarding the specific process of testing the samples. The questions were describing the specific steps of the procedures and then verifying that these steps were the ones followed, to the letter, regarding the testing of the Landis samples. The process consisted of three steps, including a quality control, an instrument verification process, then injection of the actual samples into the IRMS machine for measurement.

Again, one of the areas of concentration by the USADA counsel was that of the auto versus the manual verification of the results.

Q: For what purpose [do you perform manual adjustments]?
A: [Cynthia Mongongu] First I control to see if the software correctly checked the background noise, because sometimes it will go to peaks that are not background noise; therefore I correct the error.

Q: Are there other manual adjustments you make?
A: On the level of the peak integration.

Q: Why do you find it necessary?
A: I have to look at the background noise because sometimes the software will add background noise which falsifies the background noise. As far as the integration is concerned, you have the possibility of seeing the trace of a ratio to delineate the actual real value of the chromatography peak.

Q: Was the affect of those adjustments to enhance the quality of the results?
A: Yes.

Q: If you had not made those adjustments, those results would not have been reliable?
A: Absolutely.

Following this, USADA counsel turned to the previous arguments made by the Landis team about the software log which indicated that multiple processing of the sample had occurred.

Q: Was there a time when you performed twenty functions [of reprocessing]?
A: [Cynthia Mongongu] No, I certainly did it twenty times.

Q: So what Mr. Suh stated was a mistake?
A: Yes.

Q: When did you first see this log?
A: I saw it at the time of the reprocessing of the data.

Q: Can you explain what you were doing and why there are two entries in the log?
A: If I look here at the 17th of April, there was no acquisition because the instrument had stopped. There was no data reported. Then at 11:49 I injected a MixCal IRMS and at 12:16 I injected a second. That was to charge the insert [liner] of the instrument.

Q: So you weren't erasing previous data?
A: No.

Q: You simply had run a function to prime the machine?
A: Yes.

The USADA counsel asked Ms. Mongongu to fully explain why there were multiple entries. "When we reprocessed the results I was asked to bring out data to do three reports on the data," Ms. Mongongu said. "One which dealt with the automatic reprocessing of the data, the verification I had originally performed and the third was the subtraction of the background noise. Then after that we inserted or replaced the electronic data into the MastLink software. Then I reprocessed the results using that other software."

Much more to come

The Landis side was not able to begin its cross-examination of Ms. Mongongu today, and will likely lead-off Wednesday with it. Also, the witness list that was disclosed yesterday indicates some potential further drama, with Greg Lemond, Joe Papp and Floyd Landis himself listed as witnesses. And the nearly four hours of delay means that the hearing could spill over into the weekend to make up the lost time.

As was the case yesterday, nobody involved with the proceedings, including Landis, made any comment as all are prohibited from doing so per the rules of the arbitration.


For a thumbnail gallery of these images, click here

Images by AFP

Cyclingnews' coverage of the Floyd Landis case

September 28, 2008 - Landis takes case to US federal court
September 10, 2008 - Landis signing with current Health Net-Maxxis team for 2009
July 1, 2008 - CAS delivers final blow to Landis legal challenge
June 30, 2008 - Landis loses final appeal
June 28, 2008 - Landis decision due Monday
March 12, 2008 - Landis' judgment day nears
October 21, 2007 - Landis files appeal with CAS
October 18, 2007 - AFLD takes another look at Landis case
Thursday, October 11 - Landis continues fight, appeals to CAS
Saturday, September 22 - UCI officially names Pereiro 2006 Tour champion, Landis case raises issues
Friday, September 21 - Landis' appeal denied, two year suspension levied

Cyclingnews' complete coverage of the Floyd Landis case

Back to top