After Judge Leon declared a mistrial in the second Africa Sting trial last week (see here for the prior post), I was contacted by an individual who identified himself/herself as the jury foreman in the case. After taking certain steps to reasonably assure myself of the individual’s identity and role in the Africa Sting case, I am pleased to publish this guest post from the jury foreman.
“As foreperson of the jury that acquitted Messers. Caldwell and Godsey of FCPA charges, and as a non-practicing attorney, I thought it might be useful to those who practice and study the FCPA to pull back the veil a bit on jury deliberations. Professor Koehler has kindly provided this forum. The following view and recollection of the trial and deliberations is mine alone. In the interest of full disclosure, I voted not guilty in the final vote for all defendants and on all charges, convinced that the government’s evidence did not establish guilt beyond a reasonable doubt on any of the charges.
Jury Composition and Working Relationship
The jury was composed of four African-American men, three African-American women, one Asian-American woman, and four Caucasian men. General conversation over the course of the trial established that at least half of the jurors had at least some college education. Jury interaction was collegial at all times, excepting only a few confrontational moments during the last three days of deliberations.
Approach to Deliberations
We began deliberations with a secret ballot straw vote on all defendants. At this vote we did not vote specifically on each of the charges, and jurors were encouraged to vote “Leaning Guilty,” “No Leaning,” or “Leaning Not Guilty.” The results of the vote clearly indicated that it would be easiest to reach unanimity on Mr. Caldwell as ten jurors indicated a leaning toward his innocence and two indicated no leaning whatsoever. The straw vote also indicated that Messers. Godsey and Morales would be next most likely to receive a verdict and that the charges against the Mushriquis would be the greatest challenge. The initial straw vote on the Mushriquis returned approximately six “Not Guilty” votes each with the balance evenly split between “Guilty” and undecided. Over the course of the deliberations, no defendant ever received more than five guilty votes, though I did once incorrectly report a straw vote of 7-5 as 6-6 to the jury.
My initial recommendation that we walk through the elements of the offense was quickly overwhelmed by an unstoppable flood of opinion on process and evidence. It was not disorderly, but indicated the challenge of focusing a jury on a single element as it relates to a single defendant after nearly three and half months of being unable to discuss the case. We settled into a pattern for the first week of reviewing chronologically the evidence against a particular defendant.
In our first two days of deliberations we were able to reach unanimity on Mr. Caldwell and were only one vote from a “Not Guilty” verdict for Mr. Godsey. Some jurors were persuaded, based on the character evidence offered in Mr. Caldwell’s defense, that he could not have committed the offense. Others had reasonable doubt that he was able to hear the terms of the deal or that, based on his relative inexperience in the industry, he understood the terms of the deal. Some jurors gave significant weight to what they perceived to be Mr. Caldwell’s deference to Mr. Giordanella during the meeting in which the deal was pitched.
The initial straw poll did include two or three “Guilty” votes against Mr. Godsey. However, in the review of the evidence against Mr. Godsey, it became clear to the jury that there was very little indication in the evidence of his state of mind. The evidence included video of him in a meeting in which the terms of the deal were described, but the terms of the deal were described less fully than in the meetings with other defendants. In addition, Mr. Godsey made very few statements or even passive affirmations (“yeah,” “ok,” “uh huh”) near those portions of the meeting that contained the statements about part of the agent’s commission going to the Gabonese Minister of Defense. The last “Guilty” vote against Mr. Godsey fell on the morning of our third day of deliberations. In our review of the evidence against Mr. Morales, the holdout developed reasonable doubt of Mr. Godsey’s guilt and switched his or her vote.
We continued to deliberate for a couple more days in the same manner – walking chronologically through the evidence against Mr. Morales and then the Mushriquis. We then sent in our first note indicating that we were deadlocked and Judge Leon instructed us to try again, as we knew he would. At this point it became clear that we were not near agreement on these defendants and that we needed to revert to the original suggestion of walking through the elements of the offense for each of the three remaining defendants.
Adjustment to Deliberations
Upon our switch to an element-based deliberation, we quickly obtained agreement that each of the defendants was a “domestic concern” (as Judge Leon had instructed us) and that each of the defendants had made a payment. These were the first and third of six elements of the offense in our instructions.
The remaining elements of the offense were that the defendant had: 2) acted corruptly and willfully; 4) known that at least part of the payment was going to a foreign official; 5) intended the payment for at least one of four disallowable purposes (most notably, to influence the foreign official to do something or to gain an unfair advantage); 6) made the payment to obtain or retain business.
Deliberation following this approach was an arduous task because of the difficulty of keeping a single defendant and a single element in focus at a time. In addition, the idiosyncratic views of a few of the jurors came to light during this period. For example, one juror would not have convicted any of the defendants because he or she was not persuaded it would have been illegal for the Gabonese Minister of Defense to accept such a payment. “What,” she or he asked, “if that is part of his compensation? I don’t know. And the prosecution didn’t offer any evidence.” In this juror’s view, the fifth element of the offense could not be met without such evidence.
Another juror had difficulty parsing the language of the instructions. For example, he or she repeatedly pointed to a gloss of one of the terms of art (“corruptly,” I believe) included after the elements of the offense. The gloss used a brief phrase (something like “for some illegal purpose”) to broadly describe the four sub-elements of the fifth element of the offense. This juror would have voted “Not Guilty” on any of several grounds and so the extended detour to address this misreading of the instructions only muddied the water for others.
After a week of deliberation along these lines, we realized that there was no more to be done. For Mr. Morales, one member of the jury simply had insufficient doubt about his guilt to vote “Not Guilty.” This juror leaned heavily on his or her perception that Mr. Morales’s was more engaged in the meeting where the terms of the deal were described than was Mr. Godsey. This juror also relied heavily on a recorded phone call that corroborated Mr. Bistrong’s testimony that Mr. Morales had been made aware of the concerns another subject of the sting had about the legality of the deal.
Counsel for the Mushriquis had suggested during the trial that they had suspicions that Gabon was not the end destination for the goods in the first phase of the transaction. The jury agreed that at some point they were aware that Gabon was the destination of the goods. So we started from the end of the transaction and walked that thread back through the evidence until all jurors agreed that they probably understood Gabon to be the end user before the date of a specific meeting. This helped us determine, so far as we were able, their states of mind on the dates they were alleged to have violated the FCPA.
Even with this agreement, a significant majority of the jurors continued to vote “Not Guilty” on the Mushriquis because of reasonable doubt as to whether they believed at least a portion of the commission payment was going to a foreign official. And, of course, without agreement on that point, we could not even get to the question of their intent in that payment. In the end, only three jurors concluded that the evidence demonstrated beyond a reasonable doubt that the Mushriquis believed the goods were going to Gabon and that a portion of the commission they remitted in the first phase of the deal was going to a foreign official. All jurors admitted there was language in an email from Ms. Mushriqui and on a phone call from Mr. Mushriqui that could be reasonably construed as skepticism, but we could not secure unanimous agreement that the skepticism related specifically to the allegedly illegal payment.
Credibility of Witnesses
It may be clear from the description above that very little of the jury’s deliberations took into account the testimony of prosecution witnesses. In part, that is because the jury with near unanimity found nearly all of the prosecution witnesses to be evasive and combative. The very low view of their credibility was also based on the concerns of many jurors related to the nature of the sting operation. Though, in the end, I am not sure the credibility concerns were an important aspect of this case because the jury had the most difficult time ascertaining the state of mind and intent of the defendants. And very little testimony would have been useful in reaching those determinations.
Nature of the Sting Operation
As noted above, a number of jurors were troubled by the nature of the FBI sting operation. Specifically, some seemed unwilling to convict on the basis of vague language (e.g., “commission” instead of “bribe”) and where the defendants had not sought out the deal. These jurors were largely not participatory in the deliberations and when specifically called upon for their views would typically voice agreement with views expressed by some other juror voting “Not Guilty.” But enough small comments through the course of deliberations lead me to believe that their underlying view was that the defendants had acted in good faith and the FBI/DOJ in bad faith. Along the same lines, more than one juror voiced concern that it would be unjust for the defendants in this case to be convicted when the government relied so heavily on Mr. Bistrong who freely admitted on the stand more illegal acts than the entire group of defendants was accused of, yet was able to plead to only one count of conspiracy to violate the FCPA.
A similar issue was raised specific to Messers. Morales and Godsey. In their case, the goods they were selling would not have received State Department clearance to be exported to Gabon. Evidence indicated that nine of the subjects of the sting would be lost if State failed to grant clearance and the FBI advocated such clearance on the grounds that the goods would not actually be exported to Gabon. Some jurors expressed concern that this was overreaching by the government and that it was impossible to determine what the intent of these defendants would have been at the point in time they were alleged to have violated the FCPA because they would never have been engaged in the deal at that point absent government interference. This “impossibility” defense was not raised in court – it originated in the deliberation room and may have carried special weight for some jurors due to concerns expressed by Mr. Morales (upon first being told of the potential deal) that treaties might not permit the sale and export.
Complications Related to the Size and Length of the Trial
In my opinion the larger and longer trial accomplished several things contrary to the goals of the prosecution. First, it gave the jury multiple iterations of the deal invitation language. When comparing the less clear language used in some pitches with the clearer language used in others, it provided a mechanism to conclude that there was relatively less likelihood of culpability for those who heard less clear language or who were less participatory in their pitch meetings.
Second, the decision of the government to prosecute at least one defendant who seemed to most of the jurors to be clearly innocent leant greater weight to concerns that the government was conducting the sting or prosecutions in bad faith.
Third, facing the defendants nearly every day for more than three months seemed to me to humanize the defendants in the eyes of many jurors. At least one juror expressed absolute confidence that none of the defendants would intentionally have participated in a crime and this seemed to me to be rooted in empathy. Additionally, jurors noticed that defendants were being burdened with a much longer trial than was necessary as very little of the testimony was applicable to more than two defendants at a time.
Fourth, it gave the jurors a very long time to develop their own theories of the case. Although the judge appropriately warned us at least once a day to not form conclusions, I suspect that it is largely, if unintentionally, violated in a trial of this length.
The government has the option to try Mr. Morales and Mr. and Ms. Mushriqui again. As a taxpayer, I sincerely hope they will instead dismiss the charges. The evidence simply does not exist, even if they get their witnesses to behave better under cross, to convict. This is a case that makes one wish that a supermajority was sufficient to acquit. Prolonging this prosecution is a waste of government resources.
At some point in the deliberations, I described this sting and prosecution as a quarterback sneak. Although I came to regret that analogy for the frequency with which it was recalled in the jury room, I think it apt. The FBI and DOJ designed a play to get the ball just across the goal line. Unfortunately, in the ensuing pileup, no camera angle shows the ball with clarity and it is anyone’s guess as to whether they scored.”