Dear successor, whoever sits beneath the clock I have finally learned to distrust,
There is a blue numbered ticket in the drawer of this desk. It says 041. I have kept it for eleven months, although it belongs to a day our reports celebrated as a success. If you ask the clerk what happened that afternoon, she will tell you we reduced the average waiting time by a quarter. If you ask the woman who held the ticket, she will tell you that her father’s heating assistance vanished between two counters. Both answers are accurate. That is why I am writing to you.
The city in this account is imaginary. I have given it an old municipal hall, a tram depot, a steep northern ward, and offices with windows painted too high for a seated visitor to see through. My characters and figures are invented to test a familiar moral problem, not to assert what happened in a particular government. I want you to recognize the problem without confusing invention with evidence: an institution may improve every number it has chosen to collect and still become harder for its people to use.
When I took the reformer’s post, the registry had three public counters. At the first, a resident established identity. At the second, a clerk verified household circumstances. At the third, a supervisor decided whether the application qualified for assistance. Each desk demanded its own queue. A person could spend half a day crossing a room without leaving it. Everyone knew the arrangement was absurd. I arrived carrying a diagram that promised to remove the absurdity with one tidy arrow.
So I proposed a single service window. One clerk would scan an applicant’s papers, enter the details, and issue a numbered ticket. A shared system would route the request to whichever office could process it fastest. Residents would no longer need to explain their lives three separate times. The council approved the plan before the month ended. The newspaper called it a small victory for reason. I put the clipping above my desk like a certificate of virtue.
At first, the plan genuinely helped. A carpenter renewing a trade license left in fifteen minutes instead of three hours. A young parent registering a birth no longer had to climb a staircase with a sleeping child. Two clerks who had spent their careers copying records began resolving errors on the same day they were discovered. Their relief mattered. Reforms are rarely pure mistakes. The most dangerous ones contain enough real improvement to make criticism sound ungrateful.
Then the blue ticket entered my life. It belonged to Nessa, who had come from the northern ward with her father’s medical papers and a request for winter fuel assistance. Nessa worked mornings in a laundry and could afford to miss only one shift. She arrived before the doors opened, stood through a cold drizzle, and reached the scanner with the correct documents. The clerk fed them into the machine. It produced ticket 041 and told her she would receive a message.
The message never came. A week later she returned and was advised to submit a new request because no active case appeared under her name. The second request was rejected as a duplicate. At the appeals desk, which now operated only by appointment, she was told there was nothing to appeal because no decision had been recorded. She had moved from an unreasonable journey through three visible counters into a perfect circle that did not appear on any office floor.
I met her by accident. She was speaking to the doorman after closing time, with the exhausted politeness of someone who has discovered that anger can cost more than silence. I heard her say, “I am not asking anyone to approve it. I am asking someone to admit that I gave it to them.” When I introduced myself as the person responsible for the new system, she looked less relieved than frightened. I had expected my position to promise help. She seemed to think it increased the number of things she might say incorrectly.
I took her ticket and searched the system myself. A scanning error had left her father’s surname without its final letter. An automatic match had attached the application to a closed record. Our software labeled the transaction complete because the incoming file had been received and routed. No measure distinguished successful movement of data from successful access to service. In our dashboard, Nessa’s visit had counted as a victory before her father had even been considered.
The following morning I studied our celebrated report. The charts showed fewer face-to-face visits, shorter lines, and a higher percentage of requests resolved within two days. They did not show the people who stopped returning, the residents unable to receive messages, or the time spent finding a path through an error. I remembered an argument about efficiency that excludes, and understood that a process can become faster by ceasing to count those who cannot finish it.
A council member asked whether Nessa was an unusual case. I said I did not know. He wanted reassurance, not an honest answer, and I was tempted to give it. Instead I went to the waiting room with a notebook. I asked people where they had been before reaching us, what they expected would happen next, and whom they would contact if nothing did. Several answered the last question by pointing vaguely toward the ceiling. They had learned the name of a department but not the name of anyone authorized to help.
The clearest lesson came from Tovan, a cleaner whose son translated his letters. Our new instructions were printed in small type, with a code leading to an online portal. I asked why he had not used the portal. He showed me a phone with a fractured screen and said he bought its data in small portions. Then he asked why the letters called him an applicant instead of a person. It was not a philosophical demand. He could recognize his own name but not the category we had assigned him.
I had once described a public counter as a bottleneck. Now I wondered whether, for some people, it had also been the only place where an unanticipated difficulty could become visible. A bad counter humiliates. A good counter interprets. By abolishing both, I had mistaken the elimination of a barrier for the elimination of a relationship. I thought of the ways bureaucratic gates can hide inside ordinary paperwork, and the way my cleaner interface had concealed rather than removed them.
The clerks had warned me, though I had not understood their warnings. Before the change, Mara from counter two kept a handwritten list of cases that needed follow-up. She knew that the forms could not describe every household and that some residents would not know what to ask. I called her list an undocumented workaround. I ordered it discontinued because it could not be audited consistently. I did not ask what need had made an experienced worker create it.
When I returned to Mara, she did not say she had been right. She placed a box of returned envelopes on the desk. Each had been marked undeliverable. She said the system counted a letter as sent, not received, and she could no longer contact its intended reader without opening a new case. Mara had done this job for twenty-eight years. She had never considered an envelope successful because it had left her hand. That distinction seemed embarrassingly obvious once she spoke it.
It would be convenient to blame the machine. I designed the measures and defended them in public. I eliminated the discretion that would have allowed a clerk to correct a bad match without making a claimant begin again. I asked for a simpler system but did not specify whose work ought to become simpler. The burden had not vanished. It had moved from paid staff to residents with little time, uncertain language, and no reliable route back to a human decision.
We began with a different principle: every resident must be able to find a responsible person without proving that the system has already failed. We posted a named duty officer in the waiting room each day. We allowed written, spoken, and assisted applications to enter the same record. We gave clerks authority to correct obvious mistakes while preserving a record of what changed and why. These measures were less elegant than my first diagram. They were also more honest about the variety of lives entering our building.
The appeals procedure required harder work. A resident could previously challenge a denial but not a missing decision. We created a traceable complaint for any application that had disappeared, stalled, or been closed without explanation. The person filing it received a copy that could be understood without the portal. The officer handling it had to give a name and a next step. I had begun to see how accountability disappears when every worker controls only a fragment of a person’s journey.
Not everyone welcomed the change. Some clerks feared that direct accountability would expose them to anger for mistakes made elsewhere. They were right to worry. We needed enough staff, authority, and protective procedures to prevent a duty officer from becoming a convenient target for the whole department. I had to learn that responsibility assigned without tools is only blame wearing an official badge. A reform that makes frontline staff absorb unmanageable pressure merely moves the institution’s failure closer to the public.
Nessa returned after we found her missing file. Her father’s assistance was approved, but winter had already arrived, and she had borrowed money for fuel. I offered an apology in the language of service failure. She listened and asked whether the laundry would compensate her for the shifts she had missed. I said no. She nodded, as if that answer was the only part of our conversation she had expected to be simple.
I asked whether the new procedure felt better. She said it might help someone who had not yet lost time. Then she asked me to preserve her original ticket as evidence that a person could be present and still be absent from the record. I promised I would. The blue paper is worn along one edge because I have taken it from this drawer so many times. I wanted to remember that repairing a process does not restore every cost inflicted while it was broken.
At the next council meeting I replaced my favorite chart. Instead of showing only how quickly requests moved through our system, it showed how many were returned for correction, how long people waited for a real decision, and how often a resident needed to contact us again. We began measuring unresolved cases alongside completed ones. I could no longer claim that a shorter queue automatically meant a better service. Some council members disliked seeing more problems on paper. I told them the problems had existed before the paper acknowledged them.
I still believe in improvement. I do not long for the old three-counter maze, and I will not pretend every veteran clerk was wise. Familiarity can conceal careless habits as readily as it preserves useful judgment. The task is to keep the benefits of coordination without forcing residents to become experts in our structure. In that sense governance by default is not a reason to surrender to existing procedures. It is a reminder that every default is a choice someone can examine and revise.
If you inherit my office, resist the temptation to promise a complete solution. Ask the people waiting outside which part of the process requires them to tell the same story twice. Ask those who stopped coming why they stopped. Ask clerks which unofficial repairs they perform to make the official procedure usable. Those answers will not fit neatly into a single arrow. Your work is to make room for them before you remove the old furniture.
The clock above this desk still runs six minutes fast. The maintenance department has offered to fix it, but I have left it for you with the blue ticket. Visitors can see the clock from the waiting room. It reminds them how long they have been there. For years I saw it only from behind my desk, where it reminded me how much I hoped to finish. We occupied the same room and watched different kinds of time pass.
If you fix the clock, please also stand outside the counter when it strikes the hour. Listen for the person explaining that a letter never arrived, that a name was entered incorrectly, or that a rule does not recognize the shape of their household. Your diagrams will be necessary. Your listening will tell you what they have left out. The measure of our reform is not whether the institution can describe itself more simply. It is whether someone frightened and unfamiliar can find a person who will answer.
With responsibility for the mistakes I once called improvements,
The reformer who kept ticket 041



