Status

And a structural gap, now named The cause of today’s duplicate

Two archives share one mailbox. This one and the Lerena/Booyzen family have separate sites, separate notes and separate questions — but one email account, one National Archives, one Reading Room contact, and now one private researcher. Neither project’s notes knew what the other had sent.

A shared file now sits in both projects listing every contact, its live state, the addresses known to be dead, and the constraints that apply to both. And a rule at the top of it: before writing to any institution, search the sent mail. One query. It would have caught all three of today’s problems at once — the dead domain, the bound-files constraint, and two people already written to.

Checked rather than assumed. The free profile says so itself: “dates show the year only and fields marked Members only are hidden.” Membership runs R110 / R245 / R400 — roughly £4.60, £10, £17 — and a single verified report is R20, about 85p.

A paid report gives full names and maiden name, full date of birth, identity number, citizenship, last marital status, living status, and sometimes address and telephone history.

Corrected within the hour. That was written from the marketing copy and it was too flat. Tested on a profile whose estate number is already known from the gazette, the hidden fields are: identity number, old identity number, marriage details, name changes, the 1984 and 2004 voter rolls, place and cause of death — and Estate Information.

Estate information is the one that matters. It does not give parents. It gives the estate reference — and the death notice inside an estate file names the father and the mother. gendatabase → estate number → order the file → parents. That is exactly the chain this archive needs, one step removed. Worth the R110.

The lookups that would pay for it on the first day: Gillian Agnes Louisa Nel (1941–2024) — is Leonard her father; Leonard himself — his parents, his heirs, and whether he died in 1973 or 1979; the three other Wear-born daughters of 1943, 1946 and 1949; and Mervin Richard Halsey, which would close a question this archive has carried for a fortnight.

And then the paywall turned out not to matter Settled, free

Every South African estate over the threshold must be advertised as a Form J193 notice in the Government Gazette, and that corpus is free, full-text, and runs to 2026. The estate numbers were never behind a paywall at all. The only trick is the name order — modern notices read surname first, so “Gillian Agnes Louisa Nel” returns nothing and “Gillian Agnes Louisa” finds her immediately.

Gillian’s notice adds her exact birth date, her identity number, her address at Hermanus, and a blank where a surviving spouse would go. Two of these four retired to the Cape and died under the same Master. Derrick was born between the second and the third.

So: do not buy the subscription for this. It would still reach people the gazettes cannot — chiefly Leonard himself, who died in 1973 or 1979 where the OCR is poor — and it would add identity numbers and the 1984 and 2004 voter rolls. That is worth R110 to somebody. It is not needed here. The next step is a letter, not a payment: the Master of the High Court at Cape Town, for 10761/2024 and 018561/2021. Each file holds a death notice with a field for the father’s name.

And a condition this archive has to meet Unresolved

Every gendatabase profile carries this: “Information found on GenDatabase may not be shared online or in print without prior written permission.” This archive is published on the open web.

The National Archives references — the 1948 and 1951 court files, the estate numbers — are not affected: NAAIRS is a free public index and this archive queried it directly. The affected list turned out shorter than it looked. Gillian’s and Patricia’s dates, identity numbers, addresses and estate numbers are all in the Government Gazette, read from the PDF — not theirs. What is genuinely theirs is six facts: Edwina’s maiden name and 1980 marriage, Leonard’s sisters by their married names, his 1979 death, and the maiden name Wear for three more women.

The permission request was submitted on 15 September, naming the six facts, the one site they would appear on, and an undertaking to publish no identity numbers, addresses, telephone numbers, causes of death or living persons’ details from their records. They review each request and email the decision, with a certificate attached if it is granted.

If it is refused, those six facts come off the public pages into the working notes, and the site will say a source asked for them to go rather than deleting them quietly. It costs the argument nothing: the Wear reconstruction rests on English birth, census and marriage indexes and on the Government Gazette, none of which is theirs.

Item 8 answered itself, 15 September Closed

The letter asking the closest match who her grandfather’s parents were went out in the morning. By the afternoon the question was moot: a column on the 1952 marriage register shows her grandfather was born in the Cape Province and her grandmother in England, and the English side is the one that matters. Her grandmother’s birth entry names the mother as a Castle, of Baldock, Hertfordshire — the second of Derrick’s top matches descends from the same couple through a different daughter. The chain →

And a sentence in both of them is already overstated Noted, not resent

Both messages say Derrick’s father was most likely a son of George and Caroline. An hour later that was walked back — the generation is not settled. No third message has been sent. The question actually asked of them — what became of the Castle sons, and did any go to South Africa — is the right question under either reading, because under the further one Derrick’s father is a son of one of those sons. Only the confidence of one sentence was wrong, and three messages to the same two people in one morning would cost more than it corrected. It will be put right in the reply, if a reply comes.

Three things, and none of them is research.

• Reply to the match who offered his whole tree in June 2025 and was never answered. He is on Ancestry, not MyHeritage, which is why his tree here holds three people. Fifteen months late, but the offer may stand.
• All four letters to living matches have now gone, the last of them a cold approach to a stranger about her own father. The concern that Derrick should be told before a stranger is was raised twice and overruled twice; that is the family’s call and the file says so.
• Nothing, for the first time in this archive’s life. Request 11 went to the Master at Cape Town on 15 September; every letter and message is out. The next movement comes from somebody else’s filing cabinet.
• Big Y-700. A purchase, and the one test that would place Derrick on a branch young enough for a surname to mean something.
• The two ante-nuptial contracts — now asked about rather than stuck. They sit in a Deeds Registry, which is a different institution from the archives, and the researcher already working in Pretoria has been asked whether she does that work or can name someone who does.

Everything else is either in the post or finished. The next movement in this archive comes from somebody else’s filing cabinet, and Bydgoszcz has already warned that a complicated request can take more than a month.