Learning From the Karen Read Case
It's the Ordinariness That Makes It Valuable
There were atypical facts in the Karen Read homicide prosecution that caught the public’s eye. The dead man found in the snow was a police officer. The people in the house were police too. The defendant was a middle-class, suburban, white woman—these factors tip the balance toward the man-bites-dog class of news. Read’s defense team embraced the media and social media rather than hiding from them, and the reported facts left enough mystery to intrigue an audience—to encourage viewers to settle around the television for nightly debates.
But the real value in the Karen Read case’s history lies in its essential ordinariness, not in its incidental anomalies. An unexplained death triggers another investigation; that investigation triggers one more prosecution. There are hundreds of iterations every year. The piquant features of the Read case matter mostly because they catapulted our ordinary criminal processes out into a bright light—we can learn a lot about them if we know how to look.
Within moments of the jury’s acquitting Karen Read of all charges except Operating Under the Influence, an avalanche of commentary cataloged the “lessons” taught by the Read saga. These lessons constituted performance reviews of assorted players at the sharp end of the criminal system. The prosecutors, for example, overcharged the case. The police conducted a slovenly investigation which failed to “rule out” alternative scenarios. The crime scene work was shambolic: blood was collected in Red Solo cups. The lead State Police investigator was a misogynist bozo, who assumed Read’s guilt and texted demeaning messages.
This accounting of individuals’ personal and professional missteps only gets us so far. Yes, we know that these people zigged when they should have zagged. What we don’t know—maybe because we haven’t asked—is why?
When sociologist Diane Vaughan painstakingly anatomized the space shuttle Challenger launch decision, she demolished the conventional narrative which held that amoral managers violated safety rules in order to meet political and career goals. Vaughan concluded that: “No extraordinary actions by individuals explain what happened: no intentional managerial wrongdoing, no rule violations, no conspiracy. These are mistakes embedded in the banality of organizational life and facilitated by environments of scarcity and competition, uncertain technology, incrementalism, patters of information, routinization and organizational structures.”
Vaughan showed that it was conformity with what they believed was expected within their local culture, not willful deviation from any rules set down in writing, that had driven the managers’ decisions.
Like the NASA administrators, the players in the Read prosecution displayed all of the characteristics of workers in a poorly resourced system under chronic production pressure. Faced with a crime, their job is to produce a conviction; they have to find a way to do that with the tools at hand. There will be more crimes tomorrow. And the weaknesses and incentives in the system will still be there tomorrow too.
It is perfectly fair to describe the decision to charge Read with second degree murder as “over-charging.” In Massachusetts, second degree murder requires proof that Read intended to do “an act which, in the circumstances known to the defendant, a reasonable person would have known created a plain and strong likelihood that death would result.” There was never much hope that the Commonwealth could prove those elements beyond a reasonable doubt.
Still, in a system, all of the sources of danger can’t be seen within its individual components (here, in the prosecution); the dangers also live in the interactions of the components.
The prosecutors did not charge second degree murder despite the weakness of the evidence; it was charged because of the weakness of the evidence. The over-charge was not the product of an incompetent person; it was the emergent adaptive product of people-in-systems.
Charging Read with second degree murder conformed to a normalized strategy for confronting a chronic reality of shoddy investigations and dubious forensics, a strategy designed to generate plea bargaining leverage—to threaten the defendant with the possibility (however small) of a mandatory life sentence if she lost at trial. It was a step taken because the prosecutors recognized that (as usual) the evidence generated by the State Police might not withstand trial scrutiny. The over-charge was a standard “work-around” or “informal work rule.”
The police, for their part, had learned over the years that a “good enough” investigation would suffice to get the case off their desks and onto the prosecutors’, and in all likelihood, generate a plea bargain when the pressure on the defendant was turned up. A more careful investigation just generates complications and exculpatory hypotheses and slows things down. A process of “practical drift” normalizes shortcuts, and if there is no immediate catastrophe today’s new shortcut provides the departure point for tomorrow’s further step. Tunnel vision comes to be seen as a virtue—a sign of valuable focus, not of vulnerability.
Look at the Read story through the lens that aviation or medicine would use to understand a disaster and you see a system failure, an “organizational accident” that can be properly studied as a “near miss” and that supplies crucial safety lessons.
In the Read verdict we can also see that crucial system resiliency is provided by adequately funded defense counsel who can present a case to a jury. Counsel and juries, the two elements that made the Read case a near miss are under persistent attack. The Massachusetts Committee for Public Counsel Services maintains a “murder list” of experienced lawyers qualified to handle homicide cases, but even those lawyers have to battle for dwindling expert forensic resources. In routine cases private “Bar Advocates” are paid at below subsistence rates.
The cost of Read’s defense team is, of course, her private business, but it would be a public service if her team disclosed it so that it could be contrasted with the funding of the appointed counsel supplied in the vast majority of cases.
Meanwhile, we should also note that Read’s probationary sentence after the OUI conviction indicates that Judge Beverly Cannone foreswore the common judicial practice of signaling that there would be a “trial tax” imposed if the defendant dared to reject a plea bargain and insist on a jury’s scrutiny.
The Read case illustrates how the provision of adequate counsel and the inspection by a citizen jury—two features of the system that the system itself has drifted into treating as expensive luxuries (if not nuisances)—are actually crucial sources of resilience.
These is nothing wrong with firing a dysfunctional investigator like the egregious State Police detective in the Read case. Slapping his supervisor’s wrist (with a loss of five vacation days) sent some sort of message too. But “near miss” cases call on system leadership to hold itself “accountable for learning” how the system they control contributed to the dangers. After the “bad apples” are gone, the conditions and influences that motivated them are still present.
These episodes require more than performance reviews of individuals; they call for full event reviews, that uncover their deeper, abiding system origins. As Dr. Donald Berwick, a pioneering leader in medicine’s patient safety movement, observed, every one of these events can be treated as “a treasure”—as a source of learning.
Recent explorations of “Sentinel Event Reviews” in criminal justice, derived from medicine’s experience, supply a model to emulate. They involve all stakeholders, at all ranks, include the community, and aim at learning, not at blame. They provide an opportunity to allow the stakeholders to assess what they can and should expect from each other, and what resources they should demand.
And they provide all of them with the currently missing opportunity to state a shared goal: “We don’t want this to happen again.”


I appreciate your exegesis.