The phrase “criminal justice system” is everywhere, but no one working in—or writing about—the system day-to-day spends much time debating what sort of “system” this system might actually be.
It is what it is.
But a recent New York Times account of a startling amicus brief filed by former Justice Department dignitaries (including a former Attorney General and a former Solicitor General) urging the Supreme Court to reverse a government victory in a lower court provides a fascinating look at the clashing visions of the justice system that co-exist even among the people who administer it.
The case, United States v. Georgiou, a securities fraud prosecution, presents a simple issue.
The long-standing rule of Brady v. Maryland requires that exculpatory evidence be given to the defense on request.
Do prosecutors violate Brady’s due process guarantees when they fail to disclose evidence, even though diligent defense lawyers should have found the evidence themselves?
In Georgiou, the defense asked for any damaging information that the prosecutors possessed about the mental health history of a key prosecution witness, Georgiou’s former partner. The prosecutors turned over some scraps, but withheld others. (A Bail Report and a sentencing transcript.) The defense didn’t learn about the withheld evidence until after Georgiou was convicted.
The current DOJ prosecutors argue that since the defenders would have found the material themselves if they had looked harder, the due process clause was not violated. The United States Court of Appeals for the Third Circuit agreed.
The amicus brief filed by the chorus of former DOJ leaders criticizes this result as elevating winning over doing justice.
But in fact it does more than that. The Third Circuit’s opinion lifts the lid covering the inner workings of the criminal justice system. It shows why the design of those workings matters.
The Third Circuit (unintentionally) debunks any claim that the criminal justice system is designed for safety.
Our most common conception of the criminal system envisions a sequence of increasingly fine screens. We picture a potential error (such as a wrongful conviction) moving in a straight line from left to right until it is blocked by protections at the police level, or the prosecution level, or the grand jury level, or by the defense screen.
Backstopping all of those devices is the elaborate inspection of their output by a formal adversary trial screen. Even the product of that daunting screen is subjected to further inspection by appellate review. We think that the safety margin these screens provide is cumulative.
Errors can only take effect when every screen has a hole, and all of those holes happen to line up. In this folk version of safety expert James Reason’s “Swiss Cheese” diagram (Reason’s model is actually much more sophisticated) the screens and their defects are static, and they are independent of each other. It looks like a model of defense-in-depth: a belt-and-suspenders operation.
The first thing that Georgiou shows is that from the perspective of the current DOJ prosecutors (and the Third Circuit Court of Appeals that agreed with them) the safety features of the system are not redundant or cumulative and, what’s more, they shouldn’t be.
The challenge of discovering the mental frailties of the prosecution’s witnesses is exclusively the defense “screen’s” problem. If there is a hole in the defense screen because the defender is lethargic, or unimaginative, or unlucky, or overwhelmed by a gigantic caseload, the prosecutors feel no need to patch that hole. They’ve done their job; they see no further individual responsibility for the collective outcome.
Georgiou also shows that the screens aren’t really independent; the screens deform each other.
When the Georgiou prosecutors turned over a fraction of the mental health information, it implied that they had turned over all of the existing mental health information that answered the defense request. The predictable response of the defense lawyers was to stop looking. The prosecutors’ partial disclosure not only exploited an existing hole in the defense screen; it helped to create one.
The trial jury is assigned the role of ultimate safety inspector, but once the defenders were deprived of the mental health evidence, the jury couldn’t learn the nature of the witness’s mental health issues, and it was in no position to do meticulous quality control.
The decision not to disclose made a hole in that screen too.
Besides, Georgiou shows that the destructive potential of screens acting on screens works in both directions. It is not just a matter of upstream (for example, prosecution) errors (like withholding evidence) crippling downstream (such as jury) protections.
What the Third Circuit’s new rule does to damage future safety of the criminal justice system is more interesting than what it says about the system’s current state.
Withholding inconvenient evidence was always attractive to some prosecutors. It is more attractive now.
If prosecutors keep doubtful evidence to themselves they can convince the intimidated defendant to plead guilty early, and there will be no unpredictable trial inspection to worry about.
If a trial has to happen, the prosecution’s presentation can be cleaner and more compelling. Even when the embarrassing evidence turns up after the defendant is convicted, the question will be ineffectiveness of defense counsel, not prosecutorial misconduct.
Now, prosecutors operating under political and caseload pressure only have to decide whether the defender should find the evidence, not whether it is exculpatory.
But which defender sets this limit? The ideal defender? Or the defender actually involved in the case: the one who has 200 open cases and no investigator?
The path of least resistance for prosecutors is to take the most flattering view of defense capabilities: there will be next to nothing the imaginary Super Defender couldn’t find. This hurts the innocent—who can’t know what really happened—more than the guilty, who might.
If you know anything about how systems operate, you have to fear that the prosecutors’ vision of what their generic “reasonable defender” should find and the reality of what any actual defender can and will find will diverge further and further. Less and less will be turned over. The holes in the safety screens will grow wider and wider.
It won’t be a race to the bottom exactly; it will be more like slow, inexorable ooze.
That’s how things go wrong in systems: they drift toward failure.
As safety expert Sidney Dekker observes:
“There is a long and steady progression of small, incremental steps that unwittingly take an operation toward its boundaries. Each step away from the original norm that meets with empirical success (and no obvious sacrifice of safety) is used as the next basis from which to depart just that little bit more.”
That’s how Space Shuttles get launched in lower and lower temperatures until, finally, the O-rings fail, and the Challenger explodes. That’s how many wrongful convictions happen too.
Wait until this new constitutional rule of the Third Circuit’s is turned loose in busy urban courts, where the cases number in the thousands.
Then watch—you may have to wait a little while, but not long—for the inevitable explosions.
James Doyle, a Boston defense lawyer and author, was a 2011-2014 Visiting Fellow at the National Institute of Justice. The opinions expressed here as his own. He welcomes comment from readers.