The Luwinski Archive
An evidence-first archive for the family of Lynn Derrick Luwinski — Berlin, Schubin, Sauerbaum, Lourenço Marques, Johannesburg, Porto.
Four of this family's men were deported from Berlin between 1941 and 1944 and did not come back. One young man got out and made hats in Johannesburg. A woman born in Portuguese East Africa married four times in that city and was buried in Porto. This archive sets down what can be documented about all of them — and one finding that unsettles how they are joined together.
Two spines
Lynn Derrick Luwinski is living, and this archive names him and says nothing else about him. Behind him run two lines, and they are not the same line. One is the family of record — carried on a name, documented from the 1870s, and as of 22 September 2026 reaching one generation further back through a marriage entry at Schubin. The other is the family his DNA keeps pointing at, reconstructed from 5,464 matches. As of 15 September 2026 that second lane has names on it all the way down, and the lean has become steep: the shared segments place him inside a Hertfordshire household called Wear, and a man of that family was living in Johannesburg in 1948. Both are still set down side by side, because no document yet joins the second one to him — only the arithmetic does.
Paper, from Berlin and the Posen–West Prussia borderland. The family Derrick was born into, named for, and belongs to.
Everyone here is documented — English registers, South African registers, a court index. What is not documented is the last joint, the one to Derrick.
grandparents
Found 22 September 2026, in their son’s marriage entry at Schubin: Sohn des Kaufmann Jacob Leibholz und der H—— geb. Pollnow. He was a Kaufmann like the son and the grandson after him. Neither has a date here yet — the entry names them and says nothing else. The grandson born ten years later was named Jacob for him.
Her given name stood here as “Wilhelmine” until 27 September, when the entry was re-read against the clerk’s own capitals. It is not Wilhelmine, and it is not the Philipine a volunteer indexer read either. It begins with H and the rest of the word is unreadable, so an initial is what this chart can honestly carry.
The oldest names on this lane are George Castle and Caroline Wallis, both born 1863. Their parents have not been identified, and Henry Robert Wear’s have not been found either. This lane is a generation shorter than the one beside it.
grandparents
Married at Szubin on 24 December 1874, akt nr 9 — the ninth and last marriage of that year. The entry gives him as geboren zu Hammerstein, wohnhaft zu Hammerstein, twenty-seven years old, and her as born and living at Schubin. The move east happens at this wedding. A Jewish family of the borderland.
Joachim Luwinski & Johanna ThimmAllenstein and Sauerbaum, East Prussia. Catholic farming people of the Ermland.
Married 1883. Farm labourer, later a labourer in the Baldock brewery. Hertfordshire — not Norfolk, an error this archive carried for weeks.
The Bermondsey sideHenry Robert Wear, born Bermondsey, London, 1889 — a printer’s machine minder at J. M. Dent & Sons, Letchworth. His own parents are not yet found.
Deported from Berlin to Auschwitz, 26 February 1943. Did not come back.
Auguste “Lizzy” Luwinski 1893–1975Dressmaker and milliner. Catholic, and recorded by the German state in 1939 as not Jewish. They divorced on 21 March 1940, nearly three years before he was deported. She outlived him by thirty-two years and died at Kraichtal.
Married on Saturday 31 October 1914 at the Congregational Church, Baldock — the only Wear–Castle marriage in the whole English index, and registered at Hitchin in the last quarter of that year. By 1921 at 72 Pix Road, Letchworth, with four children and a lodger.
The date, the church and her street come from two local papers read on 27 September 2026, which had never been opened. He was “in full uniform”, home on furlough from Bury St Edmunds where the 1st Herts Territorials had their headquarters; she was Maude Castle of Norton Street, Baldock; and the wedding “took place quietly”. He rejoined, and the battalion landed in France on 6 November. He married on a week's leave and went to the Front.
And they went too. He died in Johannesburg on 11 September 1969, aged 80, and is buried at West Park; she died there on 16 March 1986, aged 90. The whole household crossed.
The segments name them as a couple, not a line. Of the 297.5 cM Derrick shares with his closest match, only ~44 falls in Castle territory; ~253 cM lies outside every region he shares with any Castle cousin. Two ancestors in common, not one.
Out of Germany and in Johannesburg by 1939. The tree says he left in 1933; his own Berlin driving licence is dated 30 January 1934. Dress designer; partner in the Piccadilly Hat Company, 81 Eloff Street. Married Ivone in Johannesburg, June 1948.
Very probably not the biological father: 0.0% Ashkenazi where a quarter is required, and a Y-DNA haplogroup — R-S11477, U106 — that is North Sea, not Ashkenazi.
Born at Hitchin, on the 1921 census at Letchworth aged four — and then absent from every English record there is. He was in Johannesburg, where he married Phyllis Agnes Robertson at Rosebank on 7 September 1940, was divorced by her in 1951, and married again that October.
Thirty-one in June 1948, in the right city, married to somebody else. And he is why his sister Cynthia was in Johannesburg to marry there in 1952 — her brother had been there twelve years.
And there were only two sons. The birth index, searched on the mother’s maiden name, gives the couple six children — all at Hitchin, 1917 to 1936 — of whom two were boys. In June 1948 one was thirty-one and the other was eleven.
Not in dispute on either reading. Three marriages that happened — de Witt, Luwinski, Salkow — and a fourth in the family tree that did not. Whichever man fathered Derrick, she was his mother, which is also why the Portuguese half of his DNA is the half that behaves.
And she is no longer only a mother on this chart. Her own birth registration reached this archive on 27 September 2026, from Derrick, and it names a generation nobody here had: her father Joaquim Júlio de Figueiredo of Vimieiro, her mother Leonia dos Santos Mata of Negozela — two parishes of the same concelho, Santa Comba Dão in the Beira Alta — and all four of her grandparents. She was born in the freguesia of Nossa Senhora da Conceição at Lourenço Marques and died at home at Gulpilhares, Vila Nova de Gaia. Her page carries the documents.
The surname on this chart is not settled. Her birth entry spells it MATA twice, including a legal rectification of her full name; her 2010 death certificate spells it MOTA twice, typed. This archive has published Mota. The two documents disagree and it is not being decided quietly.
No relation to Derrick whatsoever, and not his mother by any reading. She is here because Gillian’s mother is not Ivone — Gillian was born eleven months into this marriage and carries Agnes in her own name. If Gillian is Leonard’s daughter, this is whose daughter she is, and Derrick would share a father with her and nothing else.
Born sixteen months into his parents’ marriage. Ivone married Frederico Pinto da Rocha at Lourenço Marques on 18 November 1939 and was divorced from him by a sentence of 6 December 1945 — both recorded in the margin of her own birth registration. He carries the da Rocha name because his parents were married to each other, and his father has a forename. Derrick’s brother on either reading, because the dispute has only ever been about his father. The marriage entry of 18 November 1939 should name Frederico’s own parents.
Living. Named here, and described nowhere.
Nobody on the Luwinski side. Kurt married three times and one child is recorded across all three. Both of the people either side of him here reach him through a different parent.
Born eleven months after Leonard married Phyllis Agnes Robertson, and carrying her mother’s name. Married Willem Adrian Nel in 1969; he predeceased her. She died at 14 Fynbos Park, Church Street, Hermanus — estate 10761/2024, identity number 4108080012088, no surviving spouse, wound up by an attorney rather than by family.
She built the tree. Its own page labels Leonard “Father (of site manager)”, which is how the <Private> daughter is identified — and it is the only source anywhere that gives Leonard a mother called Castle. She died in March 2024, so the half-sibling test that would have ended this in a single reading is gone.
Three more women born Wear inside that marriage, all dead, all with estates now found free in the gazette: Doris Dawn Joubert 1943, Patricia Mary Maud Grant 1946 (018561/2021, and she too died in the Cape), Heather Jill Hunter 1949. Her sisters, or another family altogether. Two death notices were requested from the Master at Cape Town on 15 September — each has a field for the father’s name.
Everything dated, on one axis
The two spines say who stands where. This says when. Every life the archive can date, drawn to the same scale — the Berlin line above, the Hertfordshire line below, and the two people both readings share between them. Derrick is not on it, because he is living and this archive gives him no dates at all.
Every dot on the chart is clickable. Forty-odd events — marriages, divorces, deportations, departures — each with what the archive actually holds on it.
marriage divorce or action filed left the country deported, or died abroad the pivot a dotted tail is a life this archive cannot see the end of
Three things the chart says that the prose does not. The Hertfordshire family and the Berlin family are contemporaries — Henry Robert Wear and Jacob Leibholz were born five years apart, and their sons were in the same city by 1940. Everything converges on 1948: Ivone marries twice in four months, Kurt files against her within the year, and Leonard is thirty-one and married to somebody else. And the right-hand lane keeps going — Gillian dies in March 2024, five months before this archive began looking for her.
The finding this archive cannot avoid
Kurt Israel Luwinski is very probably not Derrick's biological father Inferred
If he were, Derrick's DNA would carry about a quarter Ashkenazi Jewish from Jacob Leibholz and about a quarter East Prussian German from Auguste. It carries 0.0% and 1.4%. His nearest matches are British and Portuguese, and there is no German or Jewish cluster anywhere near the top of a list 5,464 long.
And the Y-DNA now agrees. Derrick's paternal haplogroup, read on 14 September 2026, is R-S11477 — R1b, branch U106/Z381, the North Sea lineage that peaks in the Netherlands and England. It is not a lineage characteristic of Ashkenazi paternal descent. Two independent systems, the same answer, and they agree on direction as well: the autosomal cluster is British, and the haplogroup's largest European population is England.
The evidence, the arithmetic, what the Y can and cannot prove →
That does not make the Berlin history untrue, and it does not make it less his. Jacob Leibholz died at Auschwitz; his brothers Samuel and Oskar died there too, and Georg at Chełmno. Kurt escaped in 1933 and carried the name. It is the family Derrick was born into and named for. Whether it is also the family he descends from is a different question, and this archive keeps the two apart rather than pretending they are one.
Where to start
Four men, four entries in the German Federal Archives' Memorial Book. Two corrections to the family tree fall out of them.
Born at Lourenço Marques on 4 March 1922, daughter of Joaquim Júlio de Figueiredo and Leonia dos Santos Mata of Santa Comba Dão. Married a Pinto da Rocha, a de Witt, a Luwinski and a Salkow. Died at Vila Nova de Gaia on 14 March 2010.
Two marriages four and a half months apart, and a divorce filed before the child was born.
The Catholic half: a farmer, a milliner, and a son born twelve years before his parents married.
Every person in the source, labelled — including the hundred who are a surname index and not ancestry.
What “documented”, “inferred” and “disputed” mean here, and why living people are absent.