There is a principle in criminal procedure that most people take for granted without understanding what it actually protects.
The accused has the right to know the accusation. Not at the verdict. Not during deliberation. Before the hearing begins. They have the right to examine the evidence, to call their own witnesses, to confront the opposing testimony, to respond to each element of the case against them.
If you strip this principle away, you do not get a faster trial. You get a ceremony that looks like a trial and produces a conviction every time.
I think about this when I walk into an organization and discover a decision that has already been taken about a technical situation where only one version of events has been heard.
The one-sided hearing
The most common configuration I encounter when I enter a technical mission in crisis is this: the executive team has built an interpretation of what is going wrong, they have a candidate explanation, they have a candidate responsible party, and they are now looking for the independent expert who will confirm it.
They do not describe it this way. They describe it as wanting a diagnosis. But the scope of the diagnosis has already been narrowed. The questions that would open alternative explanations are not on the agenda. The people who would provide the opposing testimony are not in the meeting.
I have sat through executive briefings where the CEO described the situation, the COO confirmed the COO version, the CFO added a financial angle that supported the same reading, and the entire room agreed on what needed to happen next, while the technical team responsible for the execution of the contested decision was not in the building, had not been consulted, and would only learn about the verdict when the implementation order arrived.
This is not a conspiracy. It is the default configuration of organizational decision-making in the absence of procedural discipline. The people with the most access to executive attention get to tell the story. The people with the least access to executive attention have the story told about them.
The problem is not that the story being told is wrong. Sometimes it is exactly right. The problem is that there is no procedure for finding out whether it is right before a decision based on it is executed.
Charge of proof, inverted
In a proper trial, the burden of proof sits on the accusation. The prosecution must demonstrate guilt beyond reasonable doubt. The defense is not required to prove innocence; the absence of proof against them is sufficient.
Inside organizations, this principle is almost universally inverted.
A technical team produces a system that is performing below expectations. A decision is contemplated to replace the vendor, terminate the partnership, restructure the team, or change the technical approach. The burden of proof, in practice, sits on the technical team to prove that the current situation is acceptable, that the expectations were unrealistic, that the external pressures were unmanageable.
They are not asked to defend a specific accusation. They are asked to justify their continued existence against a diffuse sense that things should be going better.
This is a structurally losing position. You cannot prove the counterfactual. You cannot demonstrate what would have happened if a different decision had been made two years ago. You cannot adequately contest an accusation that has never been formally articulated.
The accused technical team does what anyone in that position does. They produce defensive documentation. They explain context that nobody asked to hear. They preemptively justify choices that may or may not have been the actual source of the problem. And they lose, because they are fighting against a verdict that was reached before they understood they were on trial.
The auditor I encountered by accident
Some years ago, I was leading a technical team inside an organization going through a governance crisis. A minority shareholder had commissioned an independent audit to understand what was actually happening in the business.
The audit was conducted professionally. The auditor interviewed the executive committee, mapped the formal decision-making structure, reviewed the financial data, and built a picture of the organization from the vantage point of its governance documents and its leadership team.
My team was not mentioned in those interviews. We were in a different building, working on a different floor, isolated from the political center for reasons that had more to do with internal positioning than with technical necessity. The auditor had no reason to know we existed, and nobody in the governance reviews had any incentive to surface our existence.
One afternoon, the auditor came to the building for a coffee break. Someone happened to mention my team. He came to find me. We spoke for fifteen minutes.
At the end of that fifteen-minute conversation, he called his assistant from my office, and canceled every meeting he had scheduled for the rest of the afternoon.
The audit’s conclusions changed. Not because my testimony was privileged or definitive. Because it was a version of events that no one at the governance level had any interest in surfacing, and the auditor recognized immediately that the picture he had been building was structurally incomplete.
The most useful information about what is actually happening inside an organization is almost never where the formal inquiry is looking for it. It is in the rooms nobody thought to check, with the people nobody thought to call.
When the process is perfect and the outcome is unjust
There is a specific kind of failure mode in organizational decision-making that looks like success from every procedural angle.
The process was followed. The committee was consulted. The documents were produced. The decision was ratified by the appropriate governing body. Nothing was skipped. Nothing was expedited. Nothing was irregular.
And someone walks out of the outcome with a completely legitimate sense that they have been treated unjustly.
I have seen this repeatedly in technical organizations. A procurement process that selected the lowest bidder over the incumbent partner who knew the system, because the procurement rules rewarded price over capability transfer. A reorganization that respected every applicable labor protection while eliminating the person who carried two critical systems in their head, because the formal criteria did not value tacit knowledge. An investment committee that approved a project because it ticked every box, while ignoring the weak signals from the operational team that the assumptions were wrong.
In each case, the rigidity of the process produced the illusion of neutrality. The decision was “not personal” because the process had been followed. The outcome was “not anyone’s fault” because every step was documented.
The procedural integrity of the decision was used, implicitly, to avoid the substantive judgment that someone in authority needed to make and did not want to make.
This is how organizations avoid accountability for decisions that require it. The process absorbs the responsibility. Nobody has to say: I chose this, and here is why, and if it is wrong it is wrong on me.
The designated fuse
In electrical engineering, a fuse is a deliberately weak point in a circuit. It is designed to fail first so that the more expensive, harder-to-replace components behind it survive.
When a fuse blows, the electrician’s first question is not “why did this fuse fail?” The question is “what upstream condition forced this component to sacrifice itself?”
Organizations use people as fuses.
A project goes badly. The explanation converges on a single individual or a single team, usually the one that arrived most recently, or that had the least political cover, or that was in the room last when the failure became visible. That person absorbs the organizational verdict. A reorganization follows. The problem is declared solved.
Six months later, a similar failure occurs. A different fuse blows. The cycle repeats.
The reason it repeats is that the fuse is not the cause of the failure. The cause of the failure is further upstream, in the governance configuration, in the decision-making protocols, in the political alignment that forced the fuse into a role it was never equipped to succeed in. The fuse fails because something else, structurally, could not hold.
When the fuse is replaced, the underlying condition remains. The next fuse fails the same way.
Recognizing this pattern is difficult from inside the organization, because everyone upstream of the fuse has a direct interest in the fuse being the explanation. The explanation that blames the fuse protects the circuit behind it. The explanation that audits the circuit threatens the positions of everyone who designed it.
This is not a moral failure. It is a structural one. It requires, almost by definition, someone from outside the circuit to see what is actually happening.
What the right to confrontation actually does
The right to confront one’s accuser, one of the oldest principles in adversarial legal systems, is not primarily about fairness to the accused. It is about the quality of the evidence that reaches the decision-maker.
Testimony that cannot be cross-examined is structurally less reliable than testimony that can. Not because the witness is lying. Because the full picture of what the witness knows, and does not know, and how the information came to them, is only visible when the testimony is contested.
In corporate decision-making, the equivalent principle is almost never applied. The technical team’s version of events is heard, if at all, in a separate meeting from the executive team’s version. The consultant’s report is received. The vendor’s pushback is dismissed as self-interested. Each source of testimony is evaluated in isolation, by people with their own prior commitments, without the discipline of contradictory examination.
The decision that emerges from this process is not better than the best testimony. It is an average of the available testimony, weighted by the political influence of each source.
In a courtroom, this would be recognized immediately as a miscarriage of procedure. In a boardroom, it is the normal operating mode.
Who convenes the absent
When I enter a technical situation in crisis, the first thing I look for is not who is being accused. It is who has not been heard.
The auditor who found my team by accident is the template for how this work actually gets done. You cannot trust the formal inquiry to surface the most important testimony, because the formal inquiry is structured by the same political dynamics that produced the crisis in the first place. The people with the best view of the actual mechanics of the failure are almost always the people the formal inquiry has the least incentive to interview.
You have to go find them. Not as an afterthought, not as a completeness check, but as the primary activity of the diagnosis. The verdict that reaches the executive committee should reflect every material version of events, examined under contradiction, with the procedural discipline that is mandatory in courts and optional in corporations.
This is not a neutral activity. It changes the outcome. In many of the situations where I have done this work, the conclusion that the executive team initially favored was not the conclusion that held up under full examination. A vendor was not replaced. A team was not dissolved. A reorganization was halted. The actual cause was further upstream than anyone wanted to look.
The cost of not doing this work is not only that innocent parties take the fall. It is that the organization learns nothing. The circuit continues to blow fuses. Each verdict confirms the previous bias. The institutional memory of what actually goes wrong degrades over time, because every autopsy is conducted by the same people with the same interests.
The value of the adversarial principle is not that it is fair. It is that it produces decisions grounded in the strongest available evidence. That value does not disappear when you move from the courtroom to the boardroom. The requirement for it does not diminish because the defendant is a technical team instead of a citizen.
It just becomes optional. And in its absence, the verdict is written before the hearing begins.
The pattern of stories told about teams rather than with them is the inversion of Looking Inside the Walls. The external vantage point required to surface unheard testimony is the argument of The Pit Wall and the Cockpit.
