Forward Attribution: Rebuilding Recognition When the Record Is Gone
Almost every system we use to recognise what a person can do looks backward. It verifies a record of what already happened — the degree earned, the licence issued, the loan repaid, the passport stamped. When the record is lost, the machinery has nothing to work with, and the person is treated as if their capability went with the paper. There is another direction, and I learned it from a group of artisans who had been written off.
What modern societies actually run on
Strip away the language of merit and you find that modern institutions run on attribution. Degrees, passports, licences, employment histories, references, credit files — these are the mechanisms that let strangers trust claims about one another. I have never met you, but your diploma lets me believe you can do the work; your repayment record lets a bank believe you will pay again. Attribution is the infrastructure of recognition, and most of the time it is invisible, the way plumbing is invisible until it fails.
It fails completely for a refugee. The tragedy of displacement is usually told as a loss of resources, but that is the smaller part. The larger part is the collapse of the verification architecture that connected a person's capabilities to society's ability to recognise them. The papers are gone. The position is gone. The references, the curriculum, the institutional affiliations — gone, and almost impossible to retrieve. The person is exactly who they were. What disappeared is everyone else's ability to see it.
Notice the shape of the problem. Identity and attribution are not the same thing. A refugee has not lost their identity. They have lost their markers. And because our entire apparatus of recognition is built to read markers, the loss of the marker is treated as the loss of the capability. The surgeon without papers is a labourer. The teacher without a transcript is unemployed. The system cannot see past the missing record, so it concludes there is nothing behind it.
Backward attribution, and the half it cannot reach
Every attribution system I can think of works in one direction: backward. It looks at a record of something that already happened and certifies it. This is not a flaw — it is what verification is for. A degree certifies that you completed a course of study in the past. A credit score certifies a history of repayment. A licence certifies that you once passed an examination. The whole machinery is retrospective by design.
This is also where a great deal of energy is now being poured, and where technologies like blockchain make a real and narrow contribution. If the problem is that records get lost or revoked by whichever office happens to hold the file, then a tamper-proof, portable record that lives on a person's own device is genuinely useful. A credential a camp bureaucracy cannot misplace is worth something. But be precise about what that solves. It solves the carrying of recognition. It moves the record somewhere safer and makes it harder to destroy. It does nothing at all about generating recognition where the record is already gone, or never existed. Crypto is a better filing cabinet. It is not a way of seeing a person.
And the carrying problem, however well solved, leaves the harder half untouched. Because the refugee's deepest difficulty is not that their old diploma is hard to retrieve. It is that the system has no way to recognise what they can do now — today, in a new context, with none of the old markers — and no way to recognise what they and the people around them could do together. That recognition has to be built forward, from present capability rather than past record. I have started calling it forward attribution, because naming it as the opposite of the thing we already have makes the gap obvious.
The artisans who had been written off
I learned the difference between these two directions in South Africa, years ago, almost by accident. I was staying at the house of someone who ran a charity. She heard what I do, and on something close to a whim asked me to come and facilitate a group of artisans her organisation was supporting.
There were two groups, each with a backer. One of them had just secured roughly a million euros to build the artisans a house. The engineer attached to the project was delighted — he had his commission and his funding lined up, a clean year or two of work ahead of him. I walked into a room of about twenty-five people, ages eighteen to sixty-five, and found them completely despondent. Present but checked out. There, more or less. They did not understand what the project was for, and they were not connected to any sense of purpose in it. They were the beneficiaries of a million euros and they looked like people waiting at a bus stop in the rain.
So we did the only thing I know how to do in a room like that. We drew. We painted. We talked. We split into groups they chose themselves — some took marketing, some took construction, some took community. And over a few hours something turned. They began to re-see themselves. They were already entrepreneurs. They were already on the street selling what they made. They already knew how to build. The value was not waiting to be conferred on them by a donor; it was in what they did, and it was already there.
Once they could see that, they redefined the entire problem. They worked out that they did not need a million-euro house that would take two years to deliver. With traditional materials, community labour, and the free time of people who were unemployed anyway, they could build for under a hundred thousand. The group defined a catalytic investment of a hundred thousand euros that they would control — and in doing so they stopped waiting, powerless, on someone else's million-euro decision.
The inversion
No ledger would have changed that room. There was no record to retrieve, no credential to verify, no filing cabinet whose contents would have made the slightest difference. What changed was that the people in the room came to recognise their own capability, and the agency moved from outside the group to inside it.
That is the part worth sitting with. The donor model had placed both the money and the decision outside the group. The artisans were positioned as recipients of an outcome chosen for them, and they responded exactly as people respond when the important choices have already been made elsewhere — they disengaged. The facilitation did not give them anything. It rearranged who held the decision. And the moment the decision sat inside the group, the group did something no external planner had managed: it produced a smaller, faster, cheaper, better-fitted answer than the million-euro plan, because the people defining it were the people who would live in it and build it.
This is forward attribution in its rawest form. Recognition was not retrieved from a record; it was regenerated from what the group could demonstrably do, and crucially from what they could do together. Individual capability mattered, but the thing that actually shifted was collective — the group's sense of what it was and what it could make. You could feel the coherence of the room change in an afternoon.
I want to be careful with that word, change, because the instinct everywhere now is to speed things up. That is not what happened, and it is not what should happen. The group's coherence was not accelerated. It was enhanced — deepened, made richer and more legible to itself. Those are different operations. Acceleration flattens; it collapses a complex situation into whatever metric is fastest to read, which in a development context is almost always short-term survival. Enhancement does the opposite. It holds the space open long enough for a group to find a fuller answer than the obvious one. The artisans were never only trying to survive. The assumption that they were is precisely what the million-euro house was built on.
Why this matters well beyond refugees
It is tempting to file all of this under humanitarian work and move on. That would be a mistake, because the same pattern is running, quietly, through every organisation now putting AI into the core of its work.
When a company automates a function and lets the people go, it does not only shed cost. It sheds the recognition system that was holding those people's capability legible — the informal knowledge of who was good at what, who to ask, whose judgment to trust in an edge case. That recognition lived in the relationships, not in the records, and it leaves with the people. What remains is a backward-looking attribution system — org charts, performance files, role descriptions — that can certify what was true last year and is blind to what the remaining team can actually do under new conditions. The organisation, like the refugee, is left holding markers that no longer match the territory.
The answer is the same in both cases, and it is not a better filing cabinet. It is forward attribution: a deliberate, slower practice of rebuilding recognition from demonstrated present capability, individually and collectively, rather than from the record of a world that has already changed. It is the work of enhancing a group's coherence so it can see what it is now, instead of accelerating it past the point where anyone can tell.
The artisans understood their own value within an afternoon, once someone held the space for them to see it. They had not lost it. They had lost the system that recognised it. Most of the people and organisations struggling through this technological transition are in the same position, and the move that frees them is the same move: stop trying to retrieve the old record, and start rebuilding recognition forward, from what they can demonstrably do today.
A note on the idea
Forward attribution is a working concept — the counterpart to the backward-looking recognition systems we have built our institutions on. It connects to the wider question of coherence between fast and slow clocks that we have been developing on this blog: backward attribution serves the fast clock of transactions, while forward attribution is patient, slow-clock work. The South Africa anecdote is drawn from the author's own facilitation experience.