I have been reviewing Wei's pieces for almost four years. In the first year, I reviewed each piece before it shipped because she was still learning — that was teaching, dressed as review. In the second year, I was still teaching, mostly. In the third year, I stopped teaching and started actually reviewing — meaning, treating her pieces as finished work that another maker had completed, and asking, of each one, whether it should go. I have done that for fifty-three pieces now. This is what I have learned about doing it.
I should say first what reviewing is not. It is not editing the piece. It is not asking the maker to remake it according to my hand. It is not — and this took me longer to learn than it should have — substituting my judgment for hers about the wearer she has been listening to and I have not. The Reykjavík piece, which Wei wrote about in March, is the cleanest example of this and I will come back to it.
What reviewing actually is
What reviewing actually is, after four years, is four checks. I did not arrive at these by sitting down and writing them out. I arrived at them by noticing that I always do the same four things, in the same order, and that the work of writing this entry is mostly the work of naming the order.
The first check is the brief. I read the wearer's letter, then I look at the piece on the linen square. The question is not "is this what I would have made"; the question is "is this what the letter asked for." If the answer is yes, I move on. If the answer is uncertain, I read the letter again and look again. If after the second reading I still cannot see the connection between the letter and the piece, I write a note for Wei in the small notebook on her bench — not a request to change the piece, but a question. Three times in fifty-three reviews, I have written a question. Each time, Wei has written a paragraph back in the same notebook explaining what she heard in the letter that I had not heard. Twice the piece shipped as she made it. Once she sat with my question for two days and then re-tied two beads. All three were the right outcome. The notebook works because the question is not "do it differently"; the question is "help me see what you saw."
The second check is technical. There are things only twenty-four years of stringing teaches the hand — a knot that looks tight but will loosen across a season, a clasp seated half a millimeter off-center, a bead drilled slightly off-axis that will rotate against the silk and abrade. These are the things I look for. Wei has been making for four years; her technical work is sound, and these errors are rare. When I find one, I tell her, and she fixes it. There is no judgment in this step — it is the same step Akiko did for me for the first two years of my stringing, and I am doing it for Wei because someone has to. When I have ten more years of bench time, this check will become Wei's check on the maker after her. The technique survives by being checked.
The third check is the photograph. The discipline of the photograph is in the December 2028 entry; I will not repeat it here. The review version is: does this image show the piece the way the wearer will see it the first time she lifts it from the box? Wei takes her own photographs; she has been doing this since 2030 and is now better at it than I am — her morning patience is longer than mine, and she will wait for the light to be exactly the photograph's light. The check is brief. I look at the photograph and ask whether it is in the studio's register. It almost always is.
The fourth check is the reply. The reply is the letter that goes with the photograph. The review question is: is this reply in the register of the wearer's letter? The Marseille piece — Wei's smallest — required a twelve-word reply. The Reykjavík piece — Wei's largest — required six paragraphs. The reviewer's job is to read the wearer's letter again, then read the reply, and ask whether the latter is calibrated to the former. The seven-sentences method (Wei's name for what I had been doing) helps here as well: a reply should hit the same handful of meanings the wearer's letter was load-bearing on. If it does, the piece ships. If the reply runs longer than the letter required, I mark a sentence and Wei trims. If it runs shorter, I mark a meaning that has been left out and Wei adds. This is the only one of the four checks where I sometimes write directly in the draft.
What I have not done
I have never asked Wei to remake a piece. In fifty-three reviews, never. The closest I have come is the notebook question — three times. The piece has shipped as Wei made it in fifty of fifty-three cases. The other three, Wei adjusted because she also saw, after sitting with my question, what I had been pointing at. She was not overruled. She arrived.
I bring this up because I think a lot of brands that talk about "two-maker quality review" mean something quite different — they mean a senior maker who can stop a piece. Our review is not that. The maker ships the piece. The reviewer asks four questions and offers a notebook. The structure is built around the premise that the maker has heard the wearer most directly, and the reviewer's distance is a feature only when it lets her see something the maker, being close, might not see. Distance is not authority. It is a different angle.
The Reykjavík piece
I want to write about the Reykjavík piece honestly because Wei wrote about it honestly. When she brought the letter to me — Sigrún wanting thirteen beads — I said immediately: make her thirteen. I did not feel a hesitation about the number. I felt the description in the letter the same way Wei did, eventually: it was an eight-and-a-five letter, not a thirteen-bead request, and the piece had thirteen elements because the letter did. The number was not the question. The question was what stones would carry the eight and the five.
I left her to that. I did not look at her bench for six weeks. I do not do that with Wei's pieces; I check at the end. When she laid the finished piece on the linen square and brought me the photograph and the draft of the reply, I came over to do the four checks.
I checked the brief. The piece matched the letter — the moonstone gradient was a translation of the eight years; the named stones, in the order they appeared, were each a woman from the manuscript. The piece was what the letter had asked for. It was not what I would have asked for. If Sigrún had written that letter to me, I would have done seven moonstones and five named stones, not eight and five — a tighter gradient, a hand-friendlier weight. I think my piece would have been a good piece for Sigrún. But it would have been my piece. Wei's piece was Wei's piece, and it was correct for the letter. The check was: did the piece match the brief. It did. I moved on.
I checked the technical work. Thirteen beads is a longer strand than Wei usually makes; the knot rhythm needed to be even across all twelve gaps, and it was. Each bead was drilled straight. The clasp was seated cleanly. Nothing to mark.
I checked the photograph. The light was the morning's first light; the linen square was clean; the piece was laid in the gentle arc we use. The photograph was right.
I checked the reply. Six paragraphs. The letter Sigrún had written was 1,180 words. The reply named each stone, placed each woman in the strand, explained the moonstone gradient without explaining what she had said about her own life with it, and ran to almost exactly the length the letter warranted. I marked nothing. I signed my own name in the corner — the small signature I use only when a piece has been reviewed and approved — and walked back to my bench.
What I almost said
I will admit this here. While I was looking at the piece on the linen square, before I had checked the brief, I had a sentence prepared in my head. I was going to say: "This is a piece I could not have made." I had decided the sentence before I had even done the work of reviewing. I felt it as a kind of recognition — that Wei had produced something my own hand, by default, would not produce.
I held the sentence and did the four checks anyway. After the checks I gave the sentence to Wei out loud. She has written that she has been thinking about what I meant for four months. I will write here what I meant.
I meant: I am proud of you. That is the meaning I learned, watching my own teacher Akiko, was the most honest thing a senior maker could say to a junior maker who had just done something the senior maker could not have done. Akiko said to me once, in 2026, after I had shown her a piece I had finished without her help: "This is a piece I would not have arrived at." I understood her perfectly, and I have used the same construction since. It is not faint praise. It is the only kind of praise I think is honest in this work — because what makes a piece by another maker good is not that I would have made it the same. What makes it good is that I am surprised by it, and that, when I look, I cannot find the seam where the surprise sits. It is whole. I would not have arrived at it. I could not have made it. I am glad it exists. That is what I meant.
What a reviewer is
So this is what I have learned, in four years. A reviewer is not a senior co-maker. A reviewer is a person who has been making for longer, who looks at four things, who keeps a notebook for questions, who does not remake the piece, who recognizes when the maker has done something the reviewer could not, and says so plainly. The reviewer is a checkpoint, not a co-maker. The piece is the maker's.
The structure works only if both people understand it. Wei knows that when I sign my small signature in the corner, she has been reviewed and approved — not corrected, not graded, not over-ruled. The wearer knows that when she opens the box, she is receiving a piece made by one person, calibrated to her by that person, and then checked by another person whose only authority is duration. The brand depends on the structure being true, and the structure works because we have agreed it works.
The next entry will be Wei's. I do not know what about. I am about to start a piece for a wearer in Glasgow. Her letter was 540 words. I plan to make her six beads.