On Friday, a Denver jury convicted Willie Clark in the killing of Denver Bronco Darrent Williams during a drive-by shooting. Much of the case, although not all, was based on the testimony of heavily rewarded criminal informants. Stories here and here. One witness in particular, Daniel “Ponytail” Harris, admitted to being in the car from which the bullets came, and testified that he saw Clark, and only Clark, shoot out the window at the limousine in which Williams was riding. Harris was facing a life sentence for an unrelated federal drug charge, but in exchange for his testimony, he will see that sentence cut down to five years. He will also avoid being prosecuted for the shooting himself. Another witness, gang member Vernone Edwards, will get a decade shaved off his crack-cocaine trafficking sentence. This sort of heavily compensated, self-serving testimony is one of the prime reasons that informant testimony has become such a problematic source of error. Three alternate and released jurors who spoke to reporters after the case was over said they did not believe Harris. One of the lead prosecutors in Harris’s drug case candidly explained that prosecutors can only do their best to determine whether such witnesses are telling the truth.
It used to be that informant unreliability issues were litigated, if at all, on habeas, or by volunteer attorneys at innocence projects long after the case was over. Those days are coming to an end. With heightened public and media awareness of the problem, I predict that we will see more cases in which the problem of informant reliability is addressed early on in the process, at trial or on appeal, and not, as so often has happened in the past, as an afterthought or not at all.