The Problem That Belongs to No One

There is a kind of problem that has no home.

It always starts somewhere specific. A product team ships a feature that turns out to carry a licensing question nobody priced. A distribution partner goes quiet for reasons that look commercial and turn out to be personal. A regulator asks a question the company can answer accurately or answer well, but not both in the same sentence. Each begins as a discrete issue with an obvious owner. Then it spreads sideways, collecting dependencies in legal, finance, communications and engineering, until the honest description is that it belongs to everyone, and therefore to no one in particular.

I have spent most of my working life inside problems like this. For a long time I thought of them as a professional category, a type of matter that came across my desk. I now think of them as something closer to a human constant. They are what happens wherever people organize themselves into groups with boundaries, which is to say everywhere. And the way a society, a company or a family handles them says a great deal about how it creates value and how it loses it.

Everyone behaves reasonably, and the problem still grows

Organizations find the orphan problem hard for structural reasons, not because they lack talented people.

Ronald Coase asked a deceptively simple question in 1937: if markets are so efficient, why do firms exist at all? His answer was that some coordination is cheaper to do by direction inside a boundary than by contract across one. The firm is a machine for reducing the cost of working together. But every boundary that makes coordination cheaper on one side makes it more expensive on the other. Departments are small firms inside the firm. Each is built to be accountable for outcomes it can control and each is rewarded for staying inside its lines.

Confronted with a problem that crosses four of those lines, each function will correctly identify the part it owns, handle that part competently and leave the connective tissue alone. Legal manages the legal exposure. Finance models the financial one. Communications drafts a statement in case it is needed. Engineering fixes the bug. Everyone behaves reasonably. The problem compounds anyway, because what it needed was a decision about the order of things across all four and the order of things sits on nobody’s scorecard.

Max Weber admired bureaucracy for its precision, its reliability and its impersonality. He also saw its cost. A system designed so that every task has an assigned office will struggle with the task that has not been assigned yet. The gaps between offices are not empty. They are where the new, the unexpected and the dangerous tend to arrive.

Companies have been answering this for years, mostly by inventing titles: the chief of staff, the head of special projects, the director of strategic initiatives, the vice president of business operations. In many organizations the same role exists under no title at all, as a trusted senior person who is pointed at whatever matters most this quarter. The proliferation of names is itself a signal. Organizations have recognized something valuable before they have found the language for it.

Every culture invents the go-between

I was trained as a cultural anthropologist before I was trained as anything else, and anthropology has a useful habit of making the familiar strange. Seen from that distance, the corporate chief of staff is a recent costume on a very old figure.

Every society with boundaries invents people whose job is to cross them. The Ottoman court relied on dragomans, interpreters who did far more than translate words. They carried meaning, tone and intention between powers that did not trust each other, and they were valued precisely because they belonged fully to neither side. Marriage brokers, village elders who settle disputes between families, the trusted intermediary in a negotiation between clans: each exists because two groups need something from each other and cannot quite say it directly.

The Greeks gave this figure a god. Hermes was the deity of boundaries and roads, of messengers, travelers and merchants, and, tellingly, of thieves. Stone markers called herms stood at crossroads and property lines. The Greeks understood something we sometimes forget. The person who moves between worlds is useful and slightly dangerous, because the skill that lets him carry a message faithfully would also let him carry it falsely. The value of the go-between and the trust placed in him cannot be separated.

The anthropologist Victor Turner wrote about liminality, the condition of being “betwixt and between,” the threshold state in a ritual when a person has left one status and not yet entered the next. Turner noticed that liminal spaces are where social structure loosens and where new arrangements become possible. Orphan problems live in organizational liminality. They sit between categories, which is exactly why they are both dangerous and generative. A problem that fits no existing box is also a place where a better box might be designed.

Inherited work

People who handle orphan problems well often come to them from careers that trained them under real pressure. Mine was the law. I spent roughly twenty-four years as a litigator and counselor, much of it on technology, cybersecurity, entertainment and media matters, and a good deal of it in situations where the law had not yet caught up with the technology. I no longer practice. What carried over was not doctrine. It was a way of working.

High-stakes litigation is inherited work. The matter is already in motion when it reaches you, and its shape was set by people you may never meet. The facts are incomplete and some of what you have been told will turn out to be wrong in ways nobody intended. There is a capable opponent and a calendar set by a court rather than by your preferences. Certainty never arrives on schedule. So you build a theory of the case from what you have, commit to it, act on it, and revise as the picture fills in. You plan for the revision from the start.

Before law I worked as a quality assurance engineer and then a product manager at a software company that made educational games and desktop publishing tools. The first thing that job taught me is that you cannot fix a bug you cannot reproduce. Engineers, artists and testers would sit together trying to make a failure happen on purpose, because until it could be made to happen it could not really be understood. I have carried that lesson into every room since. Before anyone solves an orphan problem someone has to be able to describe exactly how it happens in terms every affected group accepts.

Making the problem legible

A string quartet has no conductor. Four people listen so closely that leadership passes among them without anyone announcing it.

That description is the first contribution and often the most valuable one. Before anything can be solved across four functions, someone has to produce an account of the situation that all four recognize as true. In my experience that single deliverable, an honest and shared description of what is actually going on, is frequently worth more than the plan that follows it. Once it exists, the plan tends to write itself. Without it, everyone is quietly solving a slightly different problem, and the effort of capable people cancels out.

From there the work comes down to three things that do not photograph well.

Translation. Most friction between groups is not a disagreement about facts. It is two groups using the same word to mean different things. Legal says risk and means liability. Product says risk and means schedule. Finance says risk and means variance. Someone has to sit in the middle and convert, without treating any of the three as wrong.

Mary Parker Follett, a management thinker writing in the 1920s who deserves to be far better known, put this better than anyone. She distinguished three ways of handling conflict: domination, where one side wins; compromise, where both sides give something up; and integration, where the parties find a solution in which neither has to sacrifice what actually matters to them. Her illustration was small and domestic. Two people in a library disagree about a window. One wants it open; the other wants it shut. The compromise is to open it halfway, which satisfies no one. The integration is to ask what each person actually wants. One wants fresh air; the other does not want a draft blowing on him. Open a window in the next room. Nothing was given up, because the desires underneath the stated positions were never in conflict.

Translation is the work of finding the next room. It requires taking each group’s position seriously enough to ask what it is protecting.

Sequencing. When a problem touches five functions, the live question is rarely what to do. It is what to do first and what can safely wait. The order often carries more value than any single decision inside it. A statement issued before the facts are fixed can create a liability that did not exist. A technical fix shipped before the partner is told can turn an annoyance into a betrayal. Litigation trains you to think in sequences because every step forecloses some options and opens others, and because the other side is watching the order as closely as you are.

Subtraction. This is the least visible of the three and, I have come to think, the most important. Organizations under pressure generate activity and a good deal of it is motion that feels productive without reducing the exposure. Meetings multiply. Workstreams are launched to demonstrate seriousness. Deciding what the group will not do frees the capacity to do the rest well. It also takes a certain kind of standing, because nobody gets credit for the fire drill that was cancelled.

Four people, no conductor

For more than fifteen years I worked with a string quartet, as a board member and advisor, helping with the business of bookings, touring, fundraising and contracts. I spent many evenings listening to them perform and rehearse, often in rooms small enough that you could see every glance.

A string quartet is the best model I know for how capable people solve problems that belong to all of them. There is no conductor. Nobody stands in front with a baton assigning ownership. Four people listen so closely that leadership passes among them, bar by bar, without anyone announcing it. The first violin leads a phrase; the cello takes the next; the viola, so often treated as the quiet middle voice, turns out to be holding the harmony together. Each player is responsible for a part and each is also responsible for the whole. When something goes wrong in performance there is no time to ask whose problem it is. Someone adjusts and the others adjust to the adjustment.

What makes this possible is not a better organizational chart. It is listening, practiced as a technical skill. It is trust built over years. And it is a shared understanding of the piece deep enough that any of the four can see, at any moment, what the music needs next.

Most organizations will never function like a string quartet, nor should they try. Scale requires structure. But the quartet shows what the structure is ultimately for and what it cannot replace. Process can assign the predictable work. Only people who are listening to the whole will catch what falls between the parts.

When to build a process, and when to send a person

There is a fair question underneath all of this. If a company keeps running into orphan problems, does it simply need a better operating model?

Sometimes, yes. Where the same category of issue crosses the same three functions every quarter, the answer is process. Build it once, define ownership clearly, write it down and the recurring version of the problem stops requiring a senior generalist to direct traffic. That path is cheaper and more humane and it is usually the right one. A good generalist’s first instinct with a recurring problem should be to make himself unnecessary.

The calculus changes with the genuinely novel. Some situations arrive once, matter enormously, and will not recur in any form a process would recognize. Standing up permanent structure for those is expensive and slow and the structure tends to be designed for the last crisis rather than the next one. What serves better is a person who can be pointed at the situation with enough standing to make calls across boundaries and enough range to understand what each boundary exists to protect. The two approaches complement each other. The organizations that get the most from the second have usually invested properly in the first.

What hiring tends to miss

Recruiting rewards demonstrable depth, for the sensible reason that depth can be verified. A credential can be checked, years can be counted, a specialty can be confirmed by references who share it.

The harder attribute to evaluate is whether someone has seen enough varieties of situation to recognize the shape of a new one early, while the options are still open and inexpensive. Isaiah Berlin borrowed a line from the Greek poet Archilochus: the fox knows many things, but the hedgehog knows one big thing. Our institutions are mostly built by and for hedgehogs and they need them. But orphan problems are fox territory. They reward someone who has been a translator in several languages, who has watched deals fail and products break and partnerships sour in a dozen different industries, and who therefore sees the outline of trouble before it has a name.

This does not show up neatly on a résumé. That is part of why the role keeps being reinvented under new titles by organizations that can feel the need before they can describe it.

The space between people

I want to end somewhere a little larger than the org chart, because I think the orphan problem tells us something about where value actually comes from.

We usually locate value inside things and inside people: in the product, the asset, the skill, the expert. But a remarkable share of what any group creates is generated in the space between its members. Trust is relational. So are reputation, coordination, timing and the shared understanding that lets four musicians breathe together before a downbeat. None of these belong to any single person or function. They exist only between people and they are maintained only by people who pay attention to the between.

That space is also where value leaks away. Most of the damage I have seen in my career did not come from one party doing something obviously wrong. It came from the seams. Each side did its part and nobody owned the joint.

This matters more now, not less. We are building machines that are extraordinarily good at executing within a defined lane. Give a modern AI system a well-specified task inside a clear boundary and it will do in seconds what used to take a team a week. As execution gets cheaper, the scarce thing shifts. It moves toward the work that has no clean specification: noticing that a problem exists before anyone has assigned it, describing it in terms that different groups accept, deciding the order of things, choosing what not to do and carrying meaning faithfully between people who do not fully trust each other. That is go-between work. It is Hermes at the crossroads. It is also, I suspect, among the most durably human things we do.

The value of the person who takes on orphan problems was never that he arrives knowing the answer. It is that the group stops spending its scarcest resource, which is time, arguing about whose problem this is and starts spending it on the problem.

The honest answer to “whose problem is this?” is almost always “ours.” Someone just has to be willing to say it first.