In the order it was done, newest first 219
Two generations out of one parish register, and the census the row sent me to could never have answered it Documented
did Work list row 128 had asked since 16 September for the parents of JOHN HAWKINS b.1820 and EMMA b.1822 of THE RED LION COTTAGES, ALDENHAM — Derrick's great-great-grandparents through their daughter Emma, who married Robert Wear at Southwark in 1881. It nominated the 1841 and 1851 censuses. *** THE 1841 CENSUS COULD NEVER HAVE ANSWERED IT. *** All twenty Hawkins entries for the parish were read and at twenty John WAS ALREADY BOARDING AWAY FROM HOME, at Aldenham Checkers with Katherine Nicholls, 70, and William Kimpton, 20. The row's instrument was the wrong one. SO THE PARISH REGISTERS WERE READ INSTEAD — 141 Hawkins baptisms, 44 Hawkins marriages, 30 Larkin marriages, every page. *** JOHN HAWKINS, BAPTISED ALDENHAM 1819, SON OF WILLIAM HAWKINS AND MARY APPLEYARD. *** The only John the register offers in the window, the rest excluded on age, his whole sibling set present from 1801. *** AND EMMA IS EMMA LARKIN, DAUGHTER OF ROBERT LARKIN AND MARY BURCH *** — banns at Aldenham 18 September 1842, both resident in the parish, the only John-and-Emma pair among the 44 and falling exactly where the baptism of their eldest in 1843 said it must; then her own baptism, ALDENHAM 1822, father Robert, mother Mary, the birth year matching the 1871 census to the year. APPLEYARD and BURCH are surnames this archive has never held. *** AND THE NINE CHILDREN CAME OUT WHOLE WHERE THERE WERE SIX: *** Robert 1843, Betsey 1849, Richard 1851, Jane 1853, John 1856, Emma 1859, Mary 1861, Henry 1863, Alfred 1865 — the 1871 household holds six and the three missing are the three eldest after Robert, which is what a household looks like when the older children have gone. *** HELD AS PROBABLE AND NOT DOCUMENTED, AND THE REASON IS THE SAME FOR ALL THREE MARRIAGES: EVERY ONE IS A BANNS ENTRY WITH AN EMPTY MARRIAGE-YEAR FIELD. *** Banns are the announcement on three Sundays, not the wedding; what supports each is a baptism to that couple in the same parish soon after. And each identification is BEST-CANDIDATE — one pair of the right names in the right window — which is weaker than a document naming the person. HALF A WARNING DISSOLVED ON THE WAY: this archive has carried a caution about TWO EMMA HAWKINSES, mother and daughter, after merging them once and publishing it. THE MOTHER WAS NEVER AN EMMA HAWKINS AT ALL — that was her married name. What is left is the sharper half: she appears as Emma Hawkins in every census of her married life, which is exactly how a woman gets filed under a name she was not born with. AND TWO PROPERTIES OF THE INSTRUMENT, MEASURED: this set INDEXES EACH BANNS CALLING SEPARATELY so one wedding appears two or three times under different ids — counting entries is not counting marriages — and on Findmypast `page=` IS NOT THE PAGINATION PARAMETER and does not fail loudly: it returns “0 results found” because an unknown parameter is treated as a filter nothing matches. It is `_page=`. That is the same shape as the year-filter trap already on the books: a zero that is an artefact of the query and looks exactly like an absence.
The one page the subscription had not given up, four days before it lapses Documented
did The probate CALENDAR page for George William Castle the younger — Martha Maud's elder brother — had been recorded on 20 September as the thing that got away: IIIF at 2864x4361 behind Findmypast's image gateway, SESSION-LOCKED, a request carrying the visible cookies answering 403. David signed the session in on 12 October and it was read the same hour, in four overlapping native-resolution regions. *** IT IS WORTH FOUR NEW FACTS AND IT CORRECTS ONE OF OURS. *** The entry: “CASTLE George William of 85 JACKMANS-PLACE LETCHWORTH Hertfordshire died 23 August 1939 at ST. BARTHOLOMEWS HOSPITAL LONDON E.C.1 ADMINISTRATION London 2 November to MARGARET CASTLE WIDOW and CYRIL GEORGE CASTLE PRINTER. Effects £1166 13s. 2d.” *** HE DID NOT DIE AT LETCHWORTH, WHICH IS WHAT THIS ARCHIVE SAID. *** Letchworth is his RESIDENCE — the only thing the index gave — and the archive read it as the place of death. He died in a London hospital forty miles from Jackmans-place. The thirteen-day reading is unharmed and slightly sharpened: his sister sailed FROM LONDON on 10 August and he died IN LONDON on the 23rd. *** AND THE WORD ON THE PAGE IS ADMINISTRATION, NOT PROBATE *** — so there was NO WILL, and he died intestate at about fifty-four; the index column had called it probate and this archive repeated it. *** IT ALSO NAMES A WIFE NOBODY HAD. *** MARGARET CASTLE, WIDOW, taking the administration jointly with CYRIL GEORGE CASTLE, PRINTER — who now has an occupation and sits in the position a son ordinarily occupies, WHICH IS STILL NOT A PROVEN RELATIONSHIP and is not claimed as one. *** AND A FIGURE THAT A SMALLER VIEW GOT WRONG. *** Effects £1166 13s. 2d. A downscaled read of the same line gave “£1100 15s. 2d.” and would have been written down; the numerals are the first thing to go, which is the second time in two days that a figure has had to be taken at native resolution rather than from a page view. *** THE TRAP IS RECORDED TOO: THERE ARE TWO GEORGE WILLIAM CASTLES IN THE 1939 CALENDAR, TWENTY LINES APART *** — the other of 3 Cissbury-road WORTHING, probate not administration, 4 August, £2466 12s. 3d. Only the residence town separates them. THE IMAGE IS NOT REPUBLISHED HERE: the LABSA and ZLB crops on this site are from public archives and this one is Findmypast's licensed scan, so it is transcribed and the record id given instead.
A sentence that was holding up an identification turns out to have had nothing under it, and now does Documented
did Row 242 found Joachim Peter Leibholz's Berlin birth on 11 October by scoping the search with one line: “Joachimsthaler Straße lies in the district of Standesamt Charlottenburg I”. *** IT WAS RIGHT AND NOTHING IN THE 1903 VOLUME HAD BEEN CITED FOR IT. *** The Lutherstraße chain of 8 October had just established that a Charlottenburg street belongs to a STADTBEZIRK and the Bezirk to an office, and that THE HOUSE NUMBER DECIDES — Lutherstr. 12–43 is one Bezirk and the rest of the street is not. So the step was re-run at the image. *** JOACHIMSTHALERSTRASSE 7.8 IS IN STADTBEZIRK “AM KURFÜRSTENDAMM 1”, AND THE BRACKET IS ON THE SAME PAGE. *** Printed page 20 of the 1903 Charlottenburg part: “1. Fasanenstr. 23–29 und 99–104, … JOACHIMSTHALERSTR. 5–9, … Knesebeckstr. 66–81, Kurfürstendamm 18–39”, with Bezirk 2 immediately below taking Joachimsthalerstr. 10–21. Seven and eight fall inside five to nine, bounded above. THE OCR IS USELESS HERE — it renders the range “5 -’i” — and the numerals were read at the page image, which is the same lesson the Lutherstraße house number taught when the text layer said 31 and the page said 34. *** AND THE ONE LINK THE 1903 VOLUME CANNOT SUPPLY IS NOW NAMED RATHER THAN GLOSSED. *** It does not define its offices by district at all: the administration pages give “Standesamt 1. Rankestr. 10” and “Standesamt 2. Kirchstr. 10” with their registrars and nothing more, so CHARLOTTENBURG HAD TWO REGISTRY OFFICES IN 1903 AND NOT THE THREE OF 1913, and no 1903 page says which Bezirke either one served. Standesamt I is at Rankestraße 10 in both years and in 1913 covers Am Kurfürstendamm 1–6 — a bridge, and labelled as one. NOTHING ABOUT THE FINDING MOVES: Nr. 875 was found in Charlottenburg I's own register index and read at the image, which settles the office without any directory reasoning at all. What changes is that a load-bearing sentence has a citation behind it instead of an assertion, and the hole in it is written down. *** THE RULE, EARNED TWICE IN FOUR DAYS: IN CHARLOTTENBURG A STREET NAME IS NOT AN ADDRESS. *** The house number selects the Stadtbezirk, and a reading that names the street without the number has not finished.
East Berlin was an argument and is now two thirds a reading, and every 1990 volume in it is blind Documented
did THE LAST UNCLOSED READING ON AGNES WAS THAT SHE DIED IN THE EAST. The 8 October sweep left East Berlin on the argument that “a West Berliner listed in a 1979 telephone book did not move to the DDR” — right, and now with a date under it: UNTIL 9 NOVEMBER 1989 A WEST BERLIN PENSIONER COULD NOT RELOCATE EAST AT ALL. So the hypothesis lives only from 1990, when she was 101, to 1994, when the index closes and she would have been 105. *** SCOPING TO THOSE YEARS TURNED 605 VOLUMES AND 41.9 GB INTO 47 VOLUMES AND ONE GIGABYTE *** — the whole of East Berlin, eleven offices, affordable in an afternoon. All 47 were downloaded and scored. THIRTY-THREE LEGIBLE AND READ, FOURTEEN BLIND AND NOT READ. No Leibholz, no Leipholz, no Leipsholz, no Luwinski and NO KERGER in any of the thirty-three — the first time her maiden name has been put to any East Berlin office. *** AND EVERY SINGLE 1990 VOLUME IN EAST BERLIN IS BLIND, IN ELEVEN SEPARATE OFFICES — EVERY OFFICE IN EAST BERLIN — ALL AT 0.000, *** while the 1991 volumes of the same offices score 1.4 to 2.5. THAT IS A SCANNING BOUNDARY AT 1990/91 and it is the THIRD this archive has measured — Kreuzberg's text layer begins at 1963, Schöneberg's at 1984, East Berlin's at 1991. A date where a series changes scan generation is a property of the digitisation and not of the records; it differs per office and has to be measured every time. Prenzlauer Berg proves the per-office part: it is blind for 1990, 1991, 1992 AND 1993, and only 1994 reads. TWO OF THE ELEVEN BOROUGHS STOP AT 1990 REGARDLESS — Treptow and Hohenschönhausen have no 1991-94 index in LABSA at all, which is a holding limit and cannot be closed by reading. *** THE FOURTEEN BLIND VOLUMES ARE NAMED AND ARE NOT BEING READ, AND THAT IS A JUDGEMENT RECORDED AS ONE. *** 1,759 pages, perhaps fourteen L-block reads, against a prior that requires a woman of 100 to leave the borough she had lived in for decades, cross a border that had been shut her whole adult life and open for weeks, and die in the east within five years — and an L-block reading would carry exactly the narrowness this archive spent the same morning writing a caution about. Scoped and named so anyone who disagrees can do it in an afternoon. ONE KERGER CAME OUT and she is not Agnes: “Kerger, Karoline Henriette Lucie”, Köpenick 1992, three forenames and no married name. AND KÖPENICK CROSS-INDEXES ITS WOMEN BOTH WAYS — Leibner also as Wilk geb. Leibner, Leipner also as Kirbach geb. Leipner — which is the behaviour SPANDAU DID NOT SHOW for Else Hedwig Herfert the same day. CROSS-INDEXING IS AN OFFICE'S HABIT AND NOT A RULE, now observed in two directions rather than assumed in one. SO AGNES REDUCES TO TWO READINGS AND NEITHER IS REACHABLE TODAY: she died outside Berlin, with no address to start from, or after 1994 and behind the thirty-year closure.
The sweep that counted boroughs and not years, and the three volumes that were Agnes's own Documented
did THIS ARCHIVE HAS SAID SINCE 8 OCTOBER THAT ALL TWELVE WEST BERLIN BOROUGHS WERE SWEPT FOR THE SURNAME. They were — BUT NOT ALL FROM THE SAME YEAR, and nobody had laid the twelve coverages side by side. Schöneberg 1938-1994 · Kreuzberg 1938-1994 · Reinickendorf 1974-1994 · Neukölln 1967-1992, the whole holding · Charlottenburg, Steglitz, Tempelhof, Tiergarten, Wedding, Wilmersdorf and Zehlendorf 1967-1994 — *** AND SPANDAU I 1983-1994 ONLY. *** AGNES LEIBHOLZ NÉE KERGER LEAVES THE TELEPHONE BOOK BETWEEN THE 1979-80 AND 1980-81 VOLUMES, SO THE THREE UNREAD SPANDAU YEARS WERE HER THREE YEARS. That is not a near miss; it is the shape of a sweep that counted boroughs and not years. *** ALL THREE ARE NOW READ AND THERE IS NO KERGER IN ANY OF THEM. *** P Rep. 480 Nr. 1238, 1239 and 1240, all LEGIBLE at 1.861, 1.725 and 1.691 control names a page, and no Leibholz in 1980 or 1981 either. AND THE FORENAME IS ITS OWN CONTROL: “Agnes” appears 90, 70 and 78 times across the three volumes — two hundred and thirty-eight other women — so the instrument plainly reads that word. SO THE SENTENCE CAN BE SAID: AGNES DOES NOT APPEAR IN ANY WEST BERLIN DEATH REGISTER INDEX THAT EXISTS FOR HER WINDOW. She died outside Berlin, or after 1994 when she would be 105 and the thirty-year closure still hides it, or in East Berlin — which is now bounded rather than shrugged at, because UNTIL NOVEMBER 1989 A WEST BERLIN PENSIONER COULD NOT RELOCATE EAST AT ALL, so that reading is live only for 1990-94, when she would be 101 to 105. *** AND THE ONE VOLUME 7 OCTOBER LEFT “NEITHER READ NOR CLOSED” IS CLOSED BY STRUCTURE. *** P Rep. 480 Nr. 2458 is catalogued “Namensverzeichnis zum Sterberegister” AND IT IS NOT A NAME INDEX AT ALL: read at the image it is a chronological running list, two columns a page of REGISTER NUMBER · FORENAME SURNAME with date headings breaking the run. It duplicates in date order the deaths the single-year volumes index by name, and single-year volumes exist for every one of 1982-86. No need to read 202 faint pages; the catalogue description is simply wrong. *** A PERSON OF THE EIGHTH SPELLING CAME OUT OF 1980: *** “HERFERT verh. LEIPHOLZ · ELSE HEDWIG · 24.11. · 3483 · ev. · verh.”, read at the page image. Born Herfert, married a Leipholz, and the column says still married — so there was a living Mr Leipholz in Spandau in 1980, in the same borough and the same window as the ERNA LEIPHOLZ this archive already had at Streitstraße 39 across six telephone books. Nothing joins them but that. *** AND IT PUT A LIMIT ON FOUR READINGS MADE THE SAME MORNING. *** Else Hedwig is filed under H, with the family's spelling only as the second element of her line, and that office did NOT cross-index her under L — while the SAME office DID cross-index its 1982 Leibholz woman under both L and S. Kreuzberg 1959, 1960, 1961 and 1962 were read BY L BLOCK because their text layers are blind, so a woman who married into this surname and whom the registrar did not cross-index would not be in the band that was read. THE FOUR NEGATIVES STAND AS EXACTLY WHAT THEY SAY and that is a narrower sentence than the one it is tempting to write. Recorded as a limit the same day rather than discovered later as a fault.
The six rows that unblocked, worked the same afternoon — and MOTA turns out to be an African surname Documented
did THE SERVICE CAME BACK IN THE MORNING AND SIX ROWS HAD BEEN WAITING ON IT. *** ROW 137 CALLED MOTA “THE MOST IMPORTANT UNTESTED NAME”, *** because MOTA FIGUEIREDO is the fullest form anybody ever wrote of Ivone's maiden name and 288 records was past the cap used on every other spelling. ALL 278 ARE NOW READ, AND THE NAME IS A FALSE LEAD FOR A REASON NOBODY HAD GUESSED: *** MOTA IN THE SOUTH AFRICAN ARCHIVES IS OVERWHELMINGLY A SOTHO, TSWANA AND NGUNI SURNAME. *** The bulk of the 278 are NATIVE ESTATE files and magistrates' case files from the Free State and the Witwatersrand — Lehlohonolo, Litaba, Mohasoa, Mbuti Meshack, Maphehello Agnes, Monamoholo Johane. EIGHTEEN ARE PORTUGUESE AND NOT ONE IS HERS, and the two-keyword test settles it in one line: MOTA and FIGUEIREDO together return EXACTLY ONE RECORD across all twenty repositories, and it is her own first divorce, held here since September. *** ROW 175 WANTED A NUMBER AND NAMED ITS OWN TRAP WITHOUT SEEING IT. *** It asked for “the WLD number for Kurt versus Frances, 1957” and said to search LUWINSKI, LUWINSKY and LUWENSKI. The file is TAB WLD 1878/1956 — THE YEAR IS 1956, the year he married her — and it came up under HOSTOWSKY, her maiden name, which none of those three queries would return. And the row's last line points at “the neighbour already in the table, WLD 2542/1957, Standard Bank versus K. Luwinski, in the same window”. THAT NEIGHBOUR IS WHERE THE 1957 CAME FROM. Correction 9500919. *** ROW 138: ALL SIX SALKOW FILES CONFIRMED AND A SEVENTH FOUND *** — TAB WLD 2462/1966, Isaac Lewis Benjamin versus ROSALINE EVE, the first record to give her a second forename, and her fourth marriage ending. AND IT CONTRADICTS THE 1960 FILE ABOUT THE ORDER OF HER MARRIAGES: one lists them “formerly Spence-Wilson, formerly Salkow”, the other numbers them “(1) Burns (2) Spencer-Wilson (3) Salkow”, and they cannot both be right. The estate is indexed SALKOW, ISAAC MORRIS — forenames reversed from every court file. *** ROW 154: *** seventy-five Rappoport records and exactly two are Pauline's. No fourth husband. *** ROW 140 IS DONE AND THE WAY IT HAD TO BE DONE IS THE FINDING: A REFERENCE CODE IS NOT SEARCHABLE ON THIS SERVICE AT ALL. *** “61416” returns ZERO while the record carrying it sits in the index, and “534/1948” returns NINE THOUSAND FOUR HUNDRED AND SIXTY-NINE, because the keyword box tokenises on the slash and matches 534 OR 1948 anywhere in a title. The portal searches TITLE TEXT. So all eighteen citations were re-verified BY SURNAME and all eighteen came back with the reference this archive already held. NOT ONE CITATION HERE IS STALE. *** AND A NEW SPELLING, OFF THE ARCHIVE'S OWN MOST-CITED FILE. *** TPD 534/1948 is Ivone's first divorce and this archive has cited it since September. Its index title reads “JOHANNES ANTONIE OELOFSE DE WITT VERSUS YVONNE (BORN DE FIGUEITEDO)” — A T WHERE THE R BELONGS, and nobody here had recorded it, because the file was known by its number and the title was never read. SEVENTEEN FORMS OF HER MAIDEN NAME ARE NOW ON THE NAMES PAGE AND THE NEWEST WAS ON A DOCUMENT THE ARCHIVE ALREADY OWNED. A CITATION IS NOT A READING — which is the same fault as this morning's, twice in one day.
NAAIRS was not down, it had moved again — and the three negatives it was blocking now hold under seventeen spellings Documented
did THE WORK LIST HAD FIVE ROWS OWNED BY “NAAIRS IS DOWN” AND A SIXTH BLOCKED BEHIND THEM. The endpoint was re-probed to put a date on the block and it turned out not to be a block. The old Search Manager address still answers 503 and naairs.htm is still 404; THE SERVICE CAME BACK SOMEWHERE ELSE UNDER A DIFFERENT NAME — nationalarchives.gov.za/search/, the NATIONAL ARCHIVAL INFORMATION SEARCH SYSTEM, free, no account, all twenty repositories, with a JSON API underneath. That is the second time in a month a dead URL has been mistaken here for a dead service; the first was negative #21, withdrawn on exactly that ground. *** AND THE BETTER HALF IS NOT THE ADDRESS. *** NARSSA still publishes its two-database warning — 8.3 million entries migrating, the “fulltext search” admittedly incomplete — AND IT NAMES THIS PORTAL AS THE COMPLETE ONE: “To search the complete database please click on: Search NAAIRS Database”, linking exactly there. On 27 September this archive added a bound to negative #18 saying it could not establish which of the two databases had answered, and that the balance of evidence favoured the complete one. FOR TODAY'S SEARCHES THAT IS NO LONGER A BALANCE OF EVIDENCE; IT IS THE ARCHIVE'S OWN SENTENCE. *** CONTROLS FIRST: SMITH 47,936, SCHMIDT 3,418. *** Then LEIBHOLZ 0, LEIPHOLZ 0, LEIBHOLTZ 0, LEIPHOLTZ 0, LAIBHOLZ 0, LEIBHOLS 0 — the negative that was called “the exact shape of the 1975 error” is widened from one spelling to six AND IT SURVIVES, and the eighth spelling has at last been put to a South African archive. LUWINSKI 7, LUWINSKY 1, LEWINSKI 5, LOWINSKY 7 and seven more spellings zero, with no naturalisation file and no immigration file under any of them; LUWINSKY, the spelling this archive knew the South African record used and had never searched, returns exactly the 1948 divorce file already held. A RE-RUN THAT CONFIRMS THE ORIGINAL IS THE USUAL OUTCOME AND IS ALMOST NEVER WRITTEN DOWN. *** AND TWO COURT FILES FELL OUT THAT THE ARCHIVE DID NOT HAVE. *** TAB WLD 1878/1956, RESTITUTION OF CONJUGAL RIGHTS, KURT LUWINSKI VERSUS FRANCES NÉE HOSTOWSKY — a seventh case where the litigation table held six. He married her on 29 FEBRUARY 1956 and sued her IN THE SAME YEAR, and in South African practice a restitution order is the first step of a divorce: THE MARRIAGE LASTED MONTHS. And TAB WLD 126/1959, Frances Eisenhardt v Rolf, judicial separation, within a year of her 23 May 1958 remarriage. BOTH ARE INDEXED UNDER HER MAIDEN SURNAME SPELT ONE LETTER APART — HOSTOWSKY AND HOSTOWSKI — AND NEITHER QUERY RETURNS THE OTHER. Three family surnames also went to a South African archive for the first time: KERGER 3, two of them immigration papers; SCHMUL 5; POLLNOW 1. None connected to anyone here, all recorded as tried and empty.
The house Jacob was deported from, and its owner disappearing out of it in print Documented
did HE IS ON THE 30TH OSTTRANSPORT OF 26 FEBRUARY 1943 AT TREUCHTLINGERSTRASSE 4, and the relief card adds “b/SPIER” — lodging with a family called Spier. The 1939 census has him at Meraner Straße 2 with Auguste and Siegfried, and nobody knew what happened in between. *** AND THE FIRST THING TO SAY IS THAT I REDISCOVERED HALF OF IT. *** The Leibholz page has carried the 1939 Teil IV entry for this house for a fortnight, with the owner read as GERTRUD EISENBERG NÉE HAAKE and Spier H. identified as HANS SPIER, born 2 December 1894. The work list row still described the question as open and nobody had read the row against the page that answered it. AND THAT ADMISSION WAS ITSELF WRONG WITHIN THE HOUR: the 1941 entry is on the same page too, quoted in full since 21 September, owner's SARA and all. The first check stopped at the panel that matched the search term and the answer was in the panel after it. *** WHAT IS GENUINELY NEW IS THREE VOLUMES — 1940, 1942 AND 1943 — AND TWO OF THEM ANSWER A QUESTION THE PAGE POSED IN WRITING. *** The 21 September panel ends: “the 1942 and 1943 volumes would close it further, and the text layer will not give them up … THOSE TWO PAGES HAVE TO BE FOUND BY EYE.” They have been. The house is 2 and 4 together on a corner, cross-referenced to Barbarossastraße 36a, so the transport list's number and the directory's are one door. *** AND THE OWNER GOES FROM A NAME, TO A NAME PLUS THE NAME THE STATE IMPOSED, TO NOBODY. *** 1939: “E Eisenberg, G., Ww.” 1940 and 1941: “E EISENBERG, G. SARA, Ww.” 1942 and 1943: “E UNGENANNT.” The compulsory middle name dates her as Jewish on the state's own record — the decree took effect on 1 January 1939, which is exactly why the 1939 volume, compiled from 1938 returns, does not carry it and the 1940 one does. *** WHAT “UNGENANNT” MEANS IS NOT PROVED AND THIS ARCHIVE DOES NOT PRETEND. *** It is consistent with a forced sale, with administration, with her death and with an editorial gap. What is recorded is that the house Jacob was deported from had a named Jewish owner in the 1941 volume and no owner at all in the 1942 one, and that one tenant — Kephan, H. Israel, Einrichter — leaves at the same boundary. SPIER CARRIES NO “ISRAEL” IN ANY OF THE FOUR YEARS while the same page marks two others and the same volume marks four men in one house round the corner, so the absence is evidence and not silence. NOT CLAIMED: that Spier hid him or sheltered him. Forced relocation into designated rooms was ordinary administration in Berlin by 1942 and “b/Spier” records an address, not a kindness. *** AND A NEGATIVE FROM YESTERDAY IS EXPLAINED RATHER THAN OVERTURNED: *** a full-text sweep had found Spier and Treuchtlinger co-occurring only at page level, which was right — THE STREET PAGE DOES NOT CONTAIN THE STRING “TREUCHTLINGER” AT ALL, because the heading OCRs as “Tremhtliagrr Str.” A search for both terms could never have found it. What works is walking the running heads, which survive the OCR far better than the body.
A birth found first try, a marriage that is not the one we wanted, and a hole in an instrument used for three weeks Documented
did THE AUSTRALIAN DEATH INDEX NAMES JOACHIM PETER LEIBHOLZ'S FATHER AS WILLIAM AND HIS MOTHER AS NANETTE NÉE NETTER, and the work list proposed hunting that marriage through the Berlin indexes by eye. THE SCOPING COST ONE PAGE OF A DIRECTORY. The 1903 Berliner Adressbuch: “WILLIAM, Fabrkbes., CHARLOTTENBURG, JOACHIMSTHALERSTR. 7. 8 II.” — a factory owner, AND THE ONLY LEIBHOLZ IN CHARLOTTENBURG in that volume, the other twelve all in Berlin proper. Joachimsthaler Straße is in Standesamt Charlottenburg I's district and the boy was born on 20 August 1903, so the office was fixed without guessing. *** AND THE INDEX HAS HIM: “LEIBHOLZ · JOACHIM PETER · 20.8. · 875” — STANDESAMT CHARLOTTENBURG I, GEBURTSREGISTER 1903, NR. 875, *** read at the page image. The man who sailed on the ANCHISES in 1939 and was naturalised at South Yarra in 1944 with previous nationality STATELESS now has a German registration behind an Australian index. *** THEN THE MARRIAGE, AND IT IS NOT A NETTER. *** Charlottenburg I 1898-1903, both volumes' L blocks read at the image, hold EXACTLY ONE LEIBHOLZ: “LEIBHOLZ | FRIEDLAENDER | 42”, 1902 Nr. 42. Charlottenburg II 1900-1903 holds none. The shape is right and William is the only Leibholz in the borough — BUT THE INDEX GIVES SURNAMES ONLY, with no forenames and no indication of which party is the groom; the same page carries “Leistikow geb. Marschalk | Heimann”, a woman indexed under L. A MISS LEIBHOLZ MARRYING A MR FRIEDLAENDER WOULD LOOK IDENTICAL, SO NOTHING IS CLAIMED. The weakest document in the chain is the Australian index, which records what an informant said about a man's parents decades later in another country. *** AND SEPARATELY, BERLIN WAS TRIED FOR SALLY LEIBHOLZ AND IS EMPTY. *** No Sally in any Berlin directory from 1903 to 1920 — eighteen years of the Leibholz run in volumes that print full forenames, the 1916 one read at the image and agreeing with its text layer entry for entry. A man who is not a householder is not in a directory, so this is not proof he never came; it is the only cheap instrument, run across eighteen years, returning nothing. *** AND THE INSTRUMENT HAD A HOLE NOBODY HAD NOTICED. *** The ZLB anchor this archive has queried for three weeks covers 1903 TO 1943 AND NOT 1799 TO 1943 — 1902 and earlier return zero pages AND ZERO SCHMIDT, so it is the anchor and not the surname. The earlier volumes are a different record, 1873 to 1896, and 1897 TO 1902 ARE DIGITISED UNDER NEITHER. A note here had said “1799 to 1970” since September. THREE THINGS CAME OUT OF THE 1916 PAGE ON THE WAY: Samuel's hat business is at Jerusalemer Straße in 1916, SEVENTEEN YEARS EARLIER than this archive's first record of it; William owned Kurfürstendamm 169 AND 170, the “E.” being Eigentümer; and there is a SECOND Leibholz household in Chodowieckistraße, Else née Kabelitz at number 8, twenty-three doors from Georg at 31.
Isidor and Cecillie have a second child, and Isidor has a Hebrew name Documented
did A WORK LIST ROW HAD BEEN SITTING ON ONE SENTENCE SINCE 21 SEPTEMBER: Yad Vashem indexes SUBMITTERS as well as victims, so searching the survivor who wrote Jenny Leipholz's Page of Testimony would give every page he filed. It was never run. It has now been run and it was right. A names search returns THREE TABS and the second — “Results Regarding Data Submitters” — lists every record whose SUBMITTER matched, so SEARCHING A SURVIVOR'S NAME RETURNS THE DEAD HE WROTE DOWN. *** GEORG ROSENBERG, BORN SCHLOCHAU 3 DECEMBER 1888, KAUFMANN, OF BERLIN, MURDERED, AGE 54 — father ISIDOR, mother CECILLIE, submitted BY HIS SON. *** He is Jenny's brother, four years and two months older, out of the same house. A SECOND PAGE FOR THE SAME MAN BY THE SAME SON GIVES THE PARENTS AS YITZKHAK AND SETZILIA — the German and Hebrew names of one couple, which is the thing that would matter if a Jewish community register at Schlochau is ever reached, because it will not be written in German. The two pages also DISAGREE ABOUT HOW HIS FATHER DIED, Auschwitz undated against Theresienstadt 1942, and neither is published. A second family came with it: JOSEPH EHRLICH of Reetz in Pomerania, of Wardenburgstraße 11 Berlin, deported to Theresienstadt on 17 March 1943 and dead there on 1 February 1944, filed by the same man as his GRANDSON — so Ehrlich is Georg Rosenberg's father-in-law and names the family Georg married into. *** AND THE METHOD LESSON APPLIES TO EVERY PAGE OF TESTIMONY THIS ARCHIVE HOLDS: *** the Ehrlich Page disagrees with the German federal memorial list on his birth DAY and his death YEAR and the lists win both times. A PAGE OF TESTIMONY IS A RELATIONSHIP DOCUMENT, NOT A DATE DOCUMENT. What it gives that no list gives is the single word in the relationship field: son, nephew, grandson. STILL NOT KIN TO ANYONE HERE on any evidence, and that has not moved; what has moved is that the row opened with two forenames and no dates and now has four named people of Schlochau with dates — which is Sally Leibholz's own Kreis, and the reason the row exists. THE SAME DAY, GENETEKA WAS PUT TO THE SAME GROUND AND THE CONTROLS ARE THE FINDING: pomorskie births give SCHMIDT 1,134 and KOWALSKI 1,023 against COHN 1, LEVY 0, LEVIN 0 and ISRAEL 1. THE INDEX HOLDS NO JEWISH RECORDS, so a Jewish surname's zero there is the index speaking. Człuchów, which is Schlochau, is not indexed at all, and Czarne, which is Hammerstein, is not either.
Kreuzberg is closed, a row sat blocked for nineteen days on work finished before it was blocked, and a DNA reading reversed Withdrawn
did *** KREUZBERG DEATHS 1938 TO 1994 ARE NOW CLOSED FOR THIS SURNAME IN EVERY SPELLING. *** The last two blind volumes, 1961 and 1962, were read page by page at the image — no Leibholz, no Leipholz, no Leipsholz. THEY PROVE NOTHING ABOUT KLARA, who died on 4 February 1960 and was in the ground before 1961 opened; they were read so the sentence is about fifty-seven years instead of forty-nine. Forty-five volumes scanned by text layer, the four blind years that mattered read by eye, and the office yields exactly two of our people in fifty-seven years. *** AND THE ADRESSBUCH ROW TURNED OUT TO HAVE BEEN WAITING FOR NOTHING. *** A note of 22 September said the borough matching needed the page images for 1939, 1941, 1942 and 1943. THOSE FOUR VOLUMES HAD BEEN READ AT THE IMAGE ON 20 SEPTEMBER, TWO DAYS BEFORE THE NOTE THAT BLOCKED THE ROW WAS WRITTEN. All four were re-read anyway, independently, which is the only reason it was found; eleven, nine, six and four, agreeing entry for entry. The matching itself: FORTY-SEVEN Berlin Gedenkbuch people, not the forty-four the row claimed, and of them the method attaches FOUR to one printed household — Else née Seelig, Kurt, Lieselotte and Walter to “A Israel, Fleischermeister, NW87 Levetzowstraße 13”, who is “A Israel, ARBEITER” at the same address a year later — corroborates three already attached, confirms one already known, and puts a direction on one unordered pair of boroughs. IT DOES NOT TOUCH THE OTHER THIRTY-EIGHT, and thirty-eight is the number that matters. *** AND THE Y-DNA ROW REVERSED ITSELF. *** It had told the next session to go and read the two GEOGRAPHIC projects on Derrick's branch, judging them worth more than the surname ones. Nobody had put a base rate under any of it. Against 21 of 23,730 in the master R1b project — 0.0885 per cent — ENGLAND GB GROUPS EIJ is 1.1 TIMES BASE and GERMANY-YDNA is 0.8. THE BRANCH SITS IN A LARGE ENGLISH PROJECT AND A LARGE GERMAN PROJECT AT CHANCE. The five surname projects the row ranked below them are enriched 53 to 292 times. Which corrects the row's own standing sentence: “no surname has surfaced and none is predicted to” is TRUE IN ITS MEANING — nothing names Derrick's biological father's line — and MISLEADING IN ITS WORDS, because five surnames sit on this branch far above chance AND NOT ONE OF THEM IS WEAR, WEIR OR WARE. The countries flatten the same way: the United States 2.3 times base, England 1.5, GERMANY 0.95 — exactly chance — and 167 of 242 branch members state no origin at all.
Fifty-one volumes of Berlin death indexes, four people of the name, and an instrument that had to be taught to doubt itself Documented
did THE FOUR OTHER REGISTRY OFFICES ARE READ. Kreuzberg 1983-1994, Reinickendorf 1982-1994, Spandau I 1983-1994 and ALL OF NEUKÖLLN 1967-1992 — fifty-one volumes, free, no account, nobody to write to. *** FOUR PEOPLE OF THE FAMILY’S LETTERS CAME OUT, AND THE TIDIEST READING OF THEM IS WRONG. *** FRITZ LEIPHOLZ, Sterberegister SPANDAU 1985 NR. 2926, died 24 November 1985, ev., VERWITWET. Fritz is the ordinary short form of Friedrich; FRIEDRICH LEIPHOLZ, Pensionär of Schwabstraße 4, leaves the Berlin telephone book during 1974; and Reinickendorf’s own registers do not hold him. A CANDIDATE, NOT AN IDENTIFICATION — no birth date, no address, no wife’s name — and it is written down as a candidate precisely because it fits too well. MARIE LEIPHOLZ, Sterberegister REINICKENDORF 1988 NR. 2306, died 5 November 1988, ev. SHE IS NOT HIS WIDOW: he is entered verwitwet in 1985, so his wife was already dead, and Marie died three years after him. GERTRUD ELLA MEIRITZ née LEIPHOLZ, SPANDAU 1993 NR. 2047 — a Leipholz-born woman who married out, which is exactly the case the method was built to catch. And LEOKADIA LOWINSKI née BARANIAK, NEUKÖLLN NR. 815, KATHOLISCH, who is almost certainly the POLISH CATHOLIC strand the family tree once confused with Derrick’s line, and is recorded for that reason. *** AND THE SWEEP ALMOST PUBLISHED TWELVE WORTHLESS NEGATIVES ON ITS FIRST PASS. *** Every one of these volumes is labelled “Durchsuchbares PDF durch OCR-Texterkennung” and the label is not a promise about the NAMES. KREUZBERG 1983 is handwritten and its text layer is blind — 6 Schmidt, 4 Müller and 0 Becker across 211 pages, 0.095 common surnames a page, against 1.272 for 1984. Row 253 had measured that by hand on 22 September AND THE FIRST PASS OF THIS SWEEP WALKED STRAIGHT INTO IT, returning a clean bracket built on a layer already proved unable to read names. EVERY VOLUME NOW SCORES ITS OWN EYESIGHT against six common surnames before its silence is allowed to count, and the two classes separate without argument: the legible run 0.93 to 1.96, the blind 0.00 to 0.095, nothing in between. *** WHICH CLOSED A ROW AND CANCELLED THE WORK IT WAS SAVING UP FOR. *** Row 253 planned to read eleven Kreuzberg volumes at the image, “roughly 220 reads”, on the premise that the office stayed handwritten to 1994. IT CHANGES TO A TYPED INDEX AT 1984. AGNES LEIBHOLZ NÉE KERGER DID NOT DIE IN KREUZBERG BETWEEN 1980 AND 1994, and the 220 reads are not needed. REINICKENDORF IS NOW UNBROKEN FROM 1974 TO 1994, twenty-one years, and FRIEDRICH IS IN NONE OF IT. *** THREE VOLUMES HAD NO TEXT LAYER AT ALL AND WERE READ AT THE IMAGE INSTEAD. *** NEUKÖLLN 1967-68 — 570 pages, not one carrying text — taught the sharpest lesson of the day: IT IS NOT ALPHABETICAL INSIDE THE LETTER. Page 263 opens L under “Januar 1967” with Lienau, Lorenz, Liebster, Lorenz, Liesegang, Liedtke, Lawischka. The entries run BY MONTH AND DATE, so a nearest-name-either-side bracket proves nothing and any method that narrows to part of the letter walks past the name. Pages 263 to 289 were read entire, bounded by the end of K and the opening of M. NEUKÖLLN 1985-87 was stranger still: EVERY PAGE SCANNED MIRRORED, three separate alphabets, and the pages OUT OF SEQUENCE — one L block reads 252, 251, 254, 253, 256, 255, 258, 257, and the only reason it counts as a complete reading is that the surnames chain unbroken across all eight, Labrenz to Luxens. All three of its L blocks were found and read. *** SO NEUKÖLLN IS CLOSED, 1967 TO 1992, AND CHARLOTTE OF THE HASENHEIDE IS NOT IN IT. *** The same caution as always, and this archive owns the control case: Heinrich Leibholz vanished from the Berlin directory after 1942 and did not die until 1948. LEAVING A DIRECTORY IS NOT DYING. ONE VOLUME OF THE FIFTY-ONE IS NEITHER READ NOR CLOSED — Spandau P Rep. 480 Nr. 2458, blind at 0.03 — and it is recorded as unread rather than quietly counted as a negative.
Derrick had Kurt’s own papers in a drawer, and one of them puts a date of ours in doubt Documented
did DERRICK PHOTOGRAPHED KURT’S DOCUMENTS AND SENT THEM. A BERLIN DRIVING LICENCE AND ITS INTERNATIONAL PERMIT, 1934, AND A GERMAN PASSPORT ISSUED AT JOHANNESBURG IN 1939 — all eight pages now published with a line-by-line German transcription and an English translation beside each. *** THE LICENCE CONTRADICTS A YEAR THIS ARCHIVE HAS PRINTED EVERYWHERE. *** “Ausgestellt für Herrn KURT LUWINSKI, geboren am 11.5.1912 zu Berlin, wohnhaft in Bln.-Schöneberg, MERANER STRASSE 2 … BERLIN, den 30. JANUAR 1934”, class 3, four Reichsmarks of Prussian fee stamps on it. THE FAMILY TREE SAYS HE EMIGRATED IN 1933 AND THIS ARCHIVE REPEATED IT AS FACT ON THE HOME PAGE, THE ATLAS, THE DNA PAGE, DERRICK’S PAGE AND HIS OWN. A man who had emigrated in 1933 did not take out a Berlin licence from a Schöneberg address in January 1934. EITHER THE TREE’S YEAR IS WRONG OR HE LEFT AND CAME BACK; the paper cannot say which and the archive is not choosing. The withdrawal is registered and check-superseded found ELEVEN PLACES still asserting it, every one corrected. AND IT EXPOSED SOMETHING WORSE AND QUIETER: THE EARLIEST DOCUMENT THIS ARCHIVE HOLDS PLACING KURT IN SOUTH AFRICA IS THE 1939 PASSPORT. Everything before that was the tree. *** THE LICENCE ALSO MOVES THE FAMILY INTO MERANER STRASSE FIVE YEARS EARLY *** — the census of 17 May 1939 was the earliest the archive could put Jacob, Auguste and Siegfried in that flat, and the son was living in it in January 1934. *** AND THE PASSPORT ANSWERS AN OPEN QUESTION OUTRIGHT. *** Kurt’s page has asked for weeks where “ISRAEL” came from, reasoning that he left before the decree and as a Mischling was outside it anyway. THE GERMAN CONSULATE IN JOHANNESBURG APPLIED IT TO HIM ON 26 JULY 1939 AND HE SIGNED IT HIMSELF: “ISRAEL KURT LUWINSKI”, in his own hand, under a consular stamp with an eagle and a swastika. What is NOT answered is why a consulate treated him as a Jew rather than a Mischling; the question is not closed, it is SHARPER, because it is now about an act by a named office on a known day. *** ONE MORE THING THE PASSPORT CLOSES: *** Beruf, HUTMACHER. The archive had Auguste a MILLINER, Samuel selling HÜTE EN GROS, and Kurt buying the PICCADILLY HAT COMPANY in 1941 — but Kurt’s own stated trade was “dress designer” in 1933 and 1956, so the hats looked like something he picked up in Johannesburg. HE WAS A HATTER IN 1939, TWO YEARS BEFORE HE BOUGHT THE SHOP. And the Ehefrau and Kinder columns are struck through, which agrees with the 1948 register calling him a bachelor. *** AND THE THING THAT IS NOT A FACT AT ALL: THREE PHOTOGRAPHS OF HIM. *** Twenty-one and smiling in profile; the same man in sepia; and twenty-seven at a consular counter in 1939, the year his father had four left. This archive has written about him for weeks out of register numbers and gazette notices. DERRICK HAS HAD HIS FACE ALL ALONG.
The home page had been contradicting a correction for nine days, and a reader found it Withdrawn
did David looked at the GRANDPARENTS block on the front page and asked whether it was right. IT WAS NOT, IN TWO PLACES, AND ONE OF THEM WAS THIS ARCHIVE’S OWN PUBLISHED CORRECTION BEING IGNORED BY ITS OWN SPINE. *** GEORG WAS LISTED UNDER “JACOB’S BROTHERS”. *** The Leibholz page has said since 20 SEPTEMBER that he cannot be one: Jacob born at SCHUBIN on 8 August 1884, Georg at HAMMERSTEIN on 7 June 1884, SIXTY-TWO DAYS APART AND ONE MOTHER, both dates from the Bundesarchiv. Cousin is the strong reading. The spine went on naming him a brother, and counting “all four brothers” where there are three, for nine days. THE CORRECTION EXISTED; NOTHING HAD EVER READ THE SPINE AGAINST IT. *** AND THE SECOND ERROR WAS MADE WORSE BY YESTERDAY’S OWN FINDING. *** The block said Auguste “SURVIVED THE WAR IN BERLIN”. Survival is documented — she died at Kraichtal in 1975. BERLIN IS NOT: the support is a census card of 17 May 1939 and a 1940 directory entry this archive itself says may not be her, the directory sweep records her ABSENT from 1941, and her second husband died at SOMMERFELD in January 1946. The framing rested on her being ‘the Catholic wife of a deported Jewish husband’ — AND THE DIVORCE OF 21 MARCH 1940, PUBLISHED HERE YESTERDAY, MEANS SHE WAS HIS EX-WIFE BY NEARLY THREE YEARS. *** WHICH BROKE A LOAD-BEARING EXPLANATION ELSEWHERE. *** mappingthelives-leibholz.tsv reasons from the 1939 census card that the Mischehe ‘is the standard explanation for why Jacob was not deported until 26 February 1943’. A PROTECTION THAT ENDED IN MARCH 1940 CANNOT EXPLAIN A SURVIVAL UNTIL FEBRUARY 1943. That row was saved by its own last sentence — ‘the archive states the category, which is on the record, and not the inference that follows from it’ — which is exactly why it was right not to have made the inference. WHAT SURVIVES THE NEW DATE IS A DIFFERENT READING: 26 February 1943 is the EVE OF THE FABRIKAKTION, which seized Jewish FORCED LABOURERS, and Gertrud — registered as an Arbeiterin — died on the morning of the 27th. Forced labour rather than marriage. LABELLED A HYPOTHESIS, because no document says where Jacob worked. *** AND MY WITHDRAWAL STRING WAS TOO GREEDY FOR THE THIRD TIME TODAY: *** a bare ‘four brothers’ matched ‘TWENTY-four brothers’ on the register page, about a different family and true. Narrowed. The lesson is now three for three and belongs in the tool’s own docstring, not just in mine.
An archivist spent a morning explaining something that was sitting unsent in our drafts folder Documented
did THE LANDESARCHIV BERLIN ANSWERED, AND THE ANSWER IS THREE DOCUMENTS AND A LESSON. Arabella Oberle sent, free of charge, GERTRUD LEIBHOLZ’S DEATH ENTRY (Standesamt Wedding 1943 Nr. 1291, P Rep. 830 Nr. 114) and BOTH SHEETS OF THE 1924 LEIBHOLZ–LUWINSKI MARRIAGE (Schöneberg II Nr. 6, P Rep. 161 Nr. 171 — note the INDEX this archive had been citing is Nr. 271, a different volume). All three are now published with a line-by-line German transcription and an English translation beside them. *** WHAT THE MARRIAGE ENTRY SETTLES: *** the divorce — Landgericht Berlin 251. R. 80. 40, final 21 MARCH 1940 — which is the third of three possibilities this archive listed for the 1940 directory’s “Auguste Leibholz, WITWE” and refused to choose between. THE CONTRADICTION IS GONE AND THE WORD IS NOT: a divorced woman is not a widow, and the volume was compiled before the decree was final. It also gives AUGUSTE’S BIRTH REGISTER CITATION, GROß-BÖSSAU 1893 NR. 88, after weeks of sweeping the Ermland on a village name; SAMUEL LEIBHOLZ standing as witness, which is the first document putting him beside Jacob; and a floor of 21 March 1940 under the Fricke remarriage. *** AND IT BREAKS A PREDICTION THIS ARCHIVE PUBLISHED IN ADVANCE. *** On 20 September, holding only the index, it wrote that the entry would name BOTH SETS OF PARENTS. IT NAMES NO PARENTS AT ALL — the 1920s Prussian form cites each party’s birth register instead. SECOND TIME IN ONE DAY that a row here was planned around a field the relevant form revision does not have; the other was BT 27 in a different century and a different country. *** THE GATE BUILT FOR THIS CAUGHT TWO MORE PLACES AND I MADE IT CRY WOLF TWICE. *** check-superseded found the claim still standing in twenty-alone.tsv and berlin-namensverzeichnisse.tsv — both real, both corrected. It ALSO flagged two statements that are TRUE, about the 1874 Schubin entry and a 1940 Johannesburg certificate, BECAUSE THE STRING I WITHDREW WAS THE SENTENCE PATTERN RATHER THAN THE CLAIM. The 1874 form does name both sets of parents and this archive proved it. A WITHDRAWN VALUE MUST BE SPECIFIC TO THE CLAIM OR THE GATE STARTS REFUSING CORRECT WORK. Narrowed twice before it passed. *** AND THE WORST THING IN THE LETTER IS OURS. *** She could not find Kurt’s birth: the 1912 registers are not digitised and she needs the responsible Standesamt — “die bloße Angabe Berlin reicht nicht”. THIS ARCHIVE ESTABLISHED IT ON 20 SEPTEMBER: STANDESAMT BERLIN VIII, GEBURTSREGISTER 1912, NR. 963, LUWINSKI · KURT MYSZISLAUS. A reply saying exactly that has been sitting in a drafts folder for nine days, because a direct send was refused by permission mode — correctly, and it was not worked around — and then NOBODY TOLD DAVID IT WAS WAITING. An archivist did unpaid work teaching a method this archive had already used, on the same day the draft was written. *** AND SHE GAVE US SOMETHING WE DID NOT KNOW EXISTED: *** the SAMMELAKTE survives and holds the 1923 AUFGEBOT PAPERS, which normally contain both parties’ birth certificates — so the four parent names the marriage entry failed to give may be in a box in Berlin. Not digitised, orderable for a fee, AND THAT IS A PURCHASE AND DAVID’S TO MAKE.
A question from the shared map paid for a reading nobody had planned Documented
did Record Atlas asked whether the BT 27 finding holds for other year ranges. CHECKING IT WAS HOUSEKEEPING AND IT TURNED INTO EVIDENCE. Three eras sampled: 1907 has NO last-UK-address column at all and buckets ages by nationality; 1937 AND 1939 ARE THE SAME FORM, P.M. 21, columns (1)-(9); a 1951 piece is a typed US military transport manifest filed inside BT 27 with none of the Board of Trade form left. The series is not homogeneous and the era has to be checked before a row is planned around a field. *** THE 1937 HALF OF THAT ANSWER WAS THE ONE THAT MATTERED HERE. *** It meant Leonard's own Windsor Castle sheet of 12 March 1937 carries column (6) — so it was opened, and it reads “WEAR, MR L A R · 20 · 18, RIDGE AVENUE, LETCHWORTH, HERTS. · ENGINEER”. THREE DEPARTURES ACROSS TWENTY-NINE MONTHS — Leonard in March 1937, his father in April 1939, his mother and the four youngest in August — ALL GIVING THE SAME HOUSE. This morning’s correction rested on two entries four months apart; it now rests on three across more than two years, and “the family’s own house” is no longer an inference from two departures but a fact about where they lived. IT ALSO SETTLES A WORD. Both 1939 entries abbreviate to “18 Ridge Ave” with a flourish and this archive published the reading with a query against it; Leonard’s clerk wrote RIDGE AVENUE out in full, with the county. AND IT DOES NOT DO THE OTHER THING. The move out of 72 Pix Road is still bracketed 1921 to October 1935 from the Pix Road side. What is new is the far end: settled at Ridge Avenue by 12 March 1937, so the gap is at most seventeen months, AND THE 1936 AND 1937 ELECTORAL REGISTERS WOULD CLOSE IT. The lesson is not about Letchworth. A HOLDINGS QUESTION FROM SOMEBODY ELSE’S MAP, ANSWERED PROPERLY INSTEAD OF GUESSED AT, HANDED BACK A RECORD THIS ARCHIVE HAD OWNED SINCE 16 SEPTEMBER AND NEVER OPENED.
A word the archive made up, and the record that took it back Withdrawn
did Leonard's page said Janette was “a cinema cashier IN LODGINGS on Ridge Avenue, half a mile away”, and the worklist row was titled “the 1939 Register finds her in Letchworth lodgings”. THE 1939 REGISTER SAYS NO SUCH THING. It gives an address — 18 Ridge Avenue — and an occupation, and nothing about tenure. The archive knew the family house as 72 Pix Road, inferred that a different address must be a room she had taken, and WROTE THE INFERENCE DOWN AS A FACT. *** THE PASSENGER LISTS TOOK IT BACK, AND THEY HAD BEEN IN THE ARCHIVE SINCE 16 SEPTEMBER. *** BT 27 column (6), LAST ADDRESS IN THE UNITED KINGDOM, is NOT IN THE FINDMYPAST TRANSCRIPT and exists only on the page image. It reads 18 RIDGE AVE, LETCHWORTH for Henry Robert on the Durban Castle of 5 April 1939, and again — one brace over all five — for Martha Maud and the four youngest on the Dunvegan Castle of 10 August. IT WAS THE FAMILY’S OWN HOUSE. She had not moved out; she was alone in the house her father left in April and her mother left in August. THE TRANSCRIPT HAD BEEN READ AND THE IMAGE HAD NOT, which is the same fault as reading one line of importer output and not opening the file it described — twice in one day. *** AND THE ROW’S OWN QUESTION TURNED OUT TO BE UNANSWERABLE BY THIS RECORD SET. *** The row wanted a Letchworth neighbour aboard, to explain why South Africa. BT 27 in 1939 stops at “country of intended future permanent residence” — A COUNTRY AND NOTHING FINER, no destination address at all — so no amount of reading these lists says where in South Africa anyone was going. Two pages were read in full and TWENTY-FOUR WERE NOT, and the row says so instead of implying a voyage-wide negative: no Hertfordshire address on Durban Castle p6, one on Dunvegan Castle p7 (Roundwood Lane, Harpenden) sitting immediately below the Wear brace. A passenger list is written in booking order. ADJACENCY IS NOT ACQUAINTANCE, and it is recorded as a fact and not as a connection. ALSO SETTLED FROM THE IMAGE: they entered S AFRICA as intended permanent residence, so they sailed as emigrants rather than on a passage out and back; and his occupation is MACHINE MINDER in a clear hand where the index has “Machine minister”. ONE METHOD NOTE FOR WHOEVER IS NEXT: the Findmypast viewer’s image endpoint speaks IIIF, so a single column can be cropped and read at native resolution. REGION CROPS ARE REFUSED ON ANY PAGE THE VIEWER HAS NOT OPENED. That is an access control, it was not worked around, and it is why twenty-four pages stayed shut.
Harvested the source Documented
did The MyHeritage tree read in full — 164 people, 648 facts with their citations. Five duplicate records found and collapsed; about a hundred people identified as a surname index rather than descent.
Checked it against the Bundesarchiv Documented
did All four Leibholz men found in the Gedenkbuch. Jacob was deported on 26 February 1943, not killed in 1945 — and Georg, born at Hammerstein two months after Jacob was born at Schubin, cannot be his brother.
Read the DNA Inferred
did Derrick carries 0.0% Ashkenazi Jewish where a quarter is required, and 1.4% Germanic where a quarter is required. The paternity of record does not survive it.
Found her name Documented
did The 1948 divorce file names her in full: Ivone, born MOTA FIGUEIREDO. "Mota" was never a nickname. Her father was a Figueiredo and her mother a Mota.
Surveyed Mozambique Documented
did FamilySearch has never filmed anything genealogical for Mozambique. The Arquivo Histórico de Moçambique holds the civil registers for 1865–1934 and takes requests online — and Ivone was born in 1924.
Worked the passport registers Documented
did Porto closed as a dead end: its series begins in 1951. Aveiro gave one candidate, Augusto Pereira de Figueiredo, bound for Lourenço Marques in October 1904.
Read the 1883 baptism book Documented
did Fifty-one images, no name index, entry Nº 35. Augusto born 12 August 1883, four grandparents named — and a marginal note giving his death at Amadora in 1954, with the registration number that will decide whether he is Ivone’s father.
Worked the British DNA cluster Inferred
did Five matches who are first cousins to each other are all second cousins to Derrick. One sits a generation closer — her grandparent was a sibling of his biological father, and her own father was born in Johannesburg. Surnames: Halsey, Parr, Fenton.
Yad Vashem Documented
did Samuel appears in the List of Jews murdered in Auschwitz as Samuel ISRAEL Leibholz — which is where the compulsory Nazi middle name on this family’s records comes from. Oskar confirmed on the Theresienstadt list, Georg on the Łódź list.
Opened the photographs Documented
did Twenty-two people in the tree carry attached images, and several of the “photographs” are documents nobody had read: Julius Leibke’s 1942 Berlin death register, naming a whole new generation — Abraham Leibke and Christine née Berger — and Henriette Katz’s 1926 newspaper death notice, in which the family itself pairs Trude with Oskar.
Read the 1956 marriage register Documented
did The largest of those images, 3,918 × 5,245 px, turned out to be the duplicate original register of Kurt’s second marriage — entry No. 169 of 1956. It records his status as DIVORCEE with an order filed, which confirms the 1951 decree from outside the court; and its parents’ columns, which this form has and the 1948 form did not, are ruled through and empty.
Wrote to three archives Documented
did Letters sent 15 September 2026, in Portuguese, to the Arquivo da Defesa Nacional at Paço d'Arcos, the Arquivo Histórico Militar in Lisbon and the Arquivo Histórico de Moçambique in Maputo, asking whether Henrique was recenseado at Lourenço Marques in 1961 and what his military registration says. The Maputo letter is a different letter: this archive already knew their civil registers stop at 1934, so it does not ask for a birth certificate — it asks whether the recruitment-district records stayed in Mozambique after 1975. Each archive is told what the others have been asked.
WITHDRAWN: the estate notice that was not Kurt's Withdrawn
did On 14 September this archive announced that it had found Kurt's estate notice in Government Gazette 2525 of 26 September 1969. On 15 September the gazette was downloaded and read in full. The word Luwinski does not appear in it once. The estate number 7584/69 on that page belongs to KOTZE, Pieter Johannes of Bethal, and the Sanlam Building address quoted with it belongs to the notice immediately above, ROBBERTZE, Alida Christina. Two adjacent entries in a two-column list, and the archive took the number from one and the address from the other. The notice is not found. Kurt's estate is known from the National Archives index and nothing else.
Found Henrique's estate Documented
did The same free gazette archive that found the hat shop also found the deceased estate of Ivone's first son — number 12294/2023, gazetted three times in 2023 as DA ROCHA, HENRIQUE FIGUEIREDO PINTO. A South African estate file contains form J294, which asks for the father's full names. The man the family tree can only call "Unknown da Rocha" is named on a printed form in a government office, and the file can be ordered by number.
Found the shop Documented
did A page of the Union Government Gazette for 30 October 1942, supplied by the family, names Kurt in a partnership notice. Chasing it found Gazettes.Africa — free, full-text, 54,313 South African issues — and eight Luwinski entries. He bought the Piccadilly Hat Company at 81 Eloff Street on 1 November 1941, ran it with Walter Oppenheimer for exactly a year, took it over alone on 31 October 1942, and was still trading there in 1952. Fifteen blank years between his arrival and his marriage are no longer blank.
Swept all 5,464 matches by surname Documented
did The archive's oldest outstanding item, closed. The search box on the DNA Matches page takes a name OR an ancestral surname and searches the whole list. LEIBHOLZ: no matches. LUWINSKI: no matches. MyHeritage indexes both as Derrick's own ancestral surnames, so the search had real terms and still returned nothing across 5,464 people. The German and Jewish absence is now an exhaustive search rather than a sample of six.
Withdrew the Portuguese half of it, hours later Withdrawn
did The same sweep returned 27 for FIGUEIREDO and 13 for MOTA, and this archive called it the best lead on Ivone's parentage it had ever had — before opening a single match. Opened, the Figueiredo tag turns out to pair her with a woman born before 1713, and MyHeritage states that the Mota matches share no ancestral surname with Derrick at all. A surname tag on a match card is a string comparison between two trees, not a finding. Withdrawn in public, because that is what the corrections layer is for.
Tested the Y-branch surnames Inferred
did FamilyTreeDNA Discover names the surname projects on R-S11477 — Ballew, Stafford, Blakely, Byrd, York. All five were searched against Derrick's match list. Everything came back between 10 and 28 cM, fourth-to-fifth cousin, the noise floor — exactly what a common ancestor in 200 BCE predicts. A negative worth recording, because it says which tool to reach for next and it is the autosomal one.
The Y-DNA came back Documented
did R-S11477 — R1b, branch U106/Z381, the North Sea lineage that peaks in the Netherlands and England, and not one characteristic of Ashkenazi paternal descent. The archive had named this as the test that would break the paternity finding or confirm it. It confirmed it, from an independent direction, and the label stays Inferred because a conclusion does not become Documented by surviving a test.
The 1939 census Documented
did Addresses at last. Jacob, Auguste and Siegfried at Meraner Straße 2, Schöneberg. And two corrections: Gertrud Leibke was Oskar’s wife, not Georg’s — and Georg’s wife, Lydia née Perl, died on the same day he did.
Read two words in column four Documented
did The 1952 Johannesburg marriage register was opened at full resolution to look for the parents of Mervin Halsey. The 1952 form has no parents' columns — but it has column (4), Country of Birth, and it says Mervin was born in the CAPE PROVINCE and his bride Cynthia Wear in ENGLAND. A week spent on the surname Halsey had been spent on the wrong half of the couple. Cynthia's GRO birth entry (Q2 1931, Hitchin, 3A/1345) names her mother as a CASTLE; the only Wear-Castle marriage in the entire English index is Hitchin, Q4 1914; the 1921 census gives her as Martha Maud Wear, born Baldock, Hertfordshire; and the 1901 and 1911 censuses put the whole Castle household at Baldock — with two children, Hilda b.1902 and Ernest b.1904, that the match's own tree never recorded. Two of Derrick's closest matches descend from that one couple through two different daughters, 297.5 and 292.7 cM. The Castles were never Norfolk; only the Browns were.
Cross-matched the cluster, then read the segments Documented
did Centimorgan totals could not separate two readings of where Derrick sat. The shared SEGMENTS could, because they answer a different question — not how much he shares with each match but which parts of his own genome. Of the 297.5 cM he shares with the closest match, only about 44 fall in regions he also shares with any Castle cousin: one block on chromosome 2. The other 253 lie nowhere near them. Two people who share a single ancestral couple cannot do that. He and she descend from Henry Robert Wear AND Martha Maud Castle together — a couple, not a line — so his father was their son.
Found the man Inferred
did LEONARD AUBREY ROBERT WEAR, birth registered Hitchin 1917, on the 1921 census at 72 Pix Road, Letchworth, aged four — and then absent from every English record there is. He was in Johannesburg: married Phyllis Agnes Robertson at Rosebank on 7 September 1940, divorced by her in 1951, married again that October, divorced again in 1962. 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.
Closed the sibship Documented
did The GRO birth index carries the mother’s maiden name from 1911, so the whole family can be listed rather than guessed at. Surname WEAR, mother’s maiden name CASTLE, all of England and Wales: six children, every one at Hitchin — 1917, 1919, 1921, 1931, 1933, 1936 — of whom two were sons. In June 1948 one was thirty-one and the other was eleven.
Then the whole family turned up in Johannesburg Documented
did Henry Robert Wear died there on 11 September 1969, aged 80, and is buried at West Park; Martha Maud died there on 16 March 1986, aged 90; Bruce Wallace married at Kuruman in 1959. The household this archive had reconstructed in a Hertfordshire market town had picked itself up and moved to the city Derrick was born in.
Somebody else had joined him to a Castle in 2018 Documented
did A seven-person MyHeritage tree, kept by a South African family who know nothing of Baldock or Letchworth, gives Leonard Wear b.1917 UK d.1973 a mother called “Muad Wear, born Castle”. Muad is Maud. No record anywhere else gives him parents — FamilySearch’s own profile has none to this day. And its profile page labels him “Father (of site manager)”, which identifies the masked daughter as the woman who built it.
Swept a source that indexes women by maiden name Documented
did gendatabase.com, given by the family. It answered in four seconds a question that had taken a day: EDWINA ANN WEAR, the candidate half-sister, was born EDWINA ANN GREEN and married Bruce Wallace Wear — Leonard’s younger brother — in 1980. A Wear by marriage, no blood relation at all. Every dead end around her had been the answer telling us so. It also produced three Luwinski court files nobody had seen: WLD 1950/1948, the divorce Kurt filed against Ivone within months of marrying her, and WLD 497/1951 and 2368/1951 behind it.
Rebuilt four cemetery pages a tile at a time, and answered the 1986 question with a no Inferred
did The West Park deep-zoom images were reassembled from their tiles — 377 for the 1986 spread alone — and magnified to five times. The result: the register DOES NOT NAME who authorised reopening grave 56045 in 1986. Martha Maud’s Remarks column is empty, and the index side of the private-graves folio holds exactly one Wear, her husband. A second interment into an existing private grave generated no grave transaction, so nobody signed for it. The record was never made, which is a different answer from not having found it, and the row closes with the reason attached.
CORRECTED: the grave folio is W 8, not W58 Withdrawn
did Published on 15 September as folio W58. The pencilled head-margin number, re-cropped and read on its own at 4.3 times, is a looped W and a single 8 — the flourish on the W had been read as a 5. Three records agree: the purchaser page’s own margin, the index side of the same folio, and the burial register’s cross-reference “P. Fol. W.8.” Nothing else in the entry changes — not the purchaser, not the address, not the sixty pounds. But a folio number is how the next person orders the page, and published wrong it sends them to the wrong sheet.
Got the parents’ own address, and demoted a guess Documented
did Burial register folio W 56 has an address column, and row 4 is Henry Robert Wear: died 11 September 1969 of bronchopneumonia, aged 80, at the General Hospital, Johannesburg; usual residence 5 LIVINGSTONE CT., 4TH ST., PARKHURST. The first address ever documented for either parent in South Africa. It also bracketed an inference this archive had been leaning on: 64 Greenfield Road, Greenside was never documented as their house — it was reasoned from an unmarried daughter of twenty-one. The parents are now documented in Parkhurst in 1969 and the widow still in Parkhurst in 1986. Greenside survives as a guess, and now says so.
Read a dead hostname as a dead service, and was wrong Withdrawn
did On 16 September the National Archives search system stopped resolving and this archive published that NAAIRS was down. The hostname really had died. THE SERVICE HAD NOT — it had moved to www.national.archsrch.gov.za under a new path, and the portal page the family opened gave the new address in one click. A paragraph telling the Reading Room supervisor that her own system was down was cut out of a letter minutes before it was sent. A URL that does not resolve is a fact about a URL.
Published two zeroes that were not zeroes Withdrawn
did The spelling sweep reported LAWINSKI and LUBINSKI as returning nothing in the archives index. Both return three records each. The page now carries the correction in place rather than a silent edit, because NOTHING FOUND and FOUND, AND NOT OURS are different claims and only one of them is a negative.
Claimed DNA above George William Castle, and withdrew it inside the hour Withdrawn
did Two match holders carry the surname GIDDINS, 43.36 and 62.89 cM, both in cluster 12 — and James Giddins of Norton End had just been proved as Derrick's three-times-great-grandfather. The arithmetic fitted a fourth cousin exactly. Three paragraphs were written calling it the first DNA to touch the Castle line above George William. THEN THE TREE WAS OPENED. Its George Giddins was born on 17 July 1846 at WITHYBROOK, WARWICKSHIRE, father Richard — a different family in a different county. The matches are real and the surname is not the reason for them. The arithmetic was sound and the premise was never checked, and checking it took four minutes.
Settled WALLIS against WALLS, six to nothing Documented
did This archive had used both spellings and asserted neither, which is a contradiction to leave lying on a page for weeks. The GRO birth index can be searched on the MOTHER'S maiden name: Castle births at Hitchin return SIX children under WALLIS — 1884, 1887, 1889, 1892, 1901, 1904 — and none at all under Walls. That is a registrar writing her maiden name down six times across twenty years, against one census in which an enumerator heard her widowed father as Joseph Walls. Walls is withdrawn.
Reversed the emigration and had it backwards in public first Withdrawn
did The published account had the father go out first and the son follow. LEONARD SAILED ON 12 MARCH 1937, two years before either parent — a man of twenty going alone, not a family sending for its children. Henry Robert followed on 5 April 1939 and Martha Maud on 10 August with the four youngest, three weeks before the declaration of war.
A line of Fraktur, and at the end of it a photograph Documented
did The 1940 Berlin address book has “Wilhelm, FLEISCHERMEISTER, Neukölln, LENAUSTRASSE 25” — an entry this archive nearly missed, because only the first name of a run spells the surname and the rest hang off a dash. Mapping the Lives has ERNA NÉE HAMMER and a son, HEINZ PAUL HUGO, born 1922, at that same address, both dead before the war ended. And Yad Vashem has his Page of Testimony: “He was a BUTCHER. Prior to WWII he lived in Berlin.” Married, spouse Erna. PLACE OF DEATH: AUSCHWITZ, 1943 — which is why the 1943 volume has no entry for him, and which no directory could ever have said. SUBMITTED BY EVA KHAVA HAMMER OLSHER, NIECE, who also wrote Erna’s. She is a Hammer: the survivor who recorded this household came through the wife’s side. His record lists four documents and TWO OF THEM ARE PHOTOGRAPHS. None of it connects him to Kurt’s family — Jacob, Samuel and Oskar were born at Schubin and nothing says where Willi was born beyond “Germany”.
Raised as new, for the third time in a day, something the archive already had Withdrawn
did A row was written this morning saying the Chevrah Kadisha was “the one cemetery this archive ruled out” and that one enquiry would give Kurt a plot and a date. IT HAS ALREADY BEEN CONSULTED. It is the source of the FIRST correction in corrections.json: “Kurt Israel Luwinski, died — was Dec 15 1969, now 15 DECEMBER 1968. Johannesburg Chevrah Kadisha burial register: died 15.12.1968, buried 16.12.1968, WESTPARK BLOCK L GRAVE 211, Yahrzeit 24 Kislev 5729, which is itself 15 December 1968.” The date, the plot and a Hebrew cross-check, all held. The discrepancy the row raised was settled before it was raised. AND THE PATTERN IS WORTH NAMING: three times today I have presented as new something already in this archive — because I have been reading the work list and the page text and NOT corrections.json, which is where the hardest-won facts live.
Kurt lies in the same ground as the Wears Inferred
did Kept from the struck row because it deserves not to be lost with it. The Chevrah Kadisha register puts Kurt at WEST PARK, BLOCK L, GRAVE 211, in the Jewish section, buried 16 December 1968. Henry Robert Wear is at West Park in GRAVE 56045, SECTION E.C., bought by his son on 13 September 1969, and Martha Maud was laid in the same grave in 1986. The man who married Derrick’s mother and the parents of the man who fathered him are in one cemetery, a few hundred yards apart, and neither family ever knew the other existed.
Added the de Witt correction, and did not collapse the duplicate Documented
did Ivone’s first husband now has a date of death in the register — 2 April 1963 at Sasolburg — carried as a was/now correction with its reasoning attached, because the source is the tree’s own duplicate and not an outside record. The duplicate itself gets a second correction saying it is not a second person. WHAT IT DID NOT GET IS COLLAPSING. The DUPLICATES map in build-register.py takes records only where the pair shares AN IDENTICAL BIRTH AND DEATH DATE, and says in its own comment that those are not guesses. This pair does not meet that bar — Johannesburg against Pretoria, one birth “circa”, only one death recorded. Widening a rule quietly to fit a case is how the two Emma Hawkinses happened. Both records stay visible with the correction on them.
Audited every citation in the tree, and the first pass of the audit was wrong Withdrawn
did Fifty-nine cited facts across three hosts. AROLSEN: seven citations, ALL SEVEN POINTING AT ONE DOCUMENT, and only one of the seven is supported by it. FAMILYSEARCH: forty-six, and only TWO are wrong. NAAIRS: six, none verifiable — they are URLs with session tokens and the service has not answered since 16 September. THE FIRST PASS SAID THIRTY-THREE FAMILYSEARCH CITATIONS WERE WRONG. It was counting parish registers: the record is the BAPTISM or the BURIAL and the tree correctly holds the birth or the death — Stanilaus born 27 September 1810 and baptised 14 October, Valentin dead on 12 July 1878 and buried on the 15th. Twenty-five of the thirty-three were that. The two that are real are one error made twice: Ivone’s and Johannes de Witt’s marriage of 2 FEBRUARY 1948 both cite the record of KURT and Ivone’s marriage of 19 JUNE 1948. The right record was already in hand from three days ago.
A man the tree has twice, and the duplicate was carrying his death Documented
did The same audit checked 164 people by arithmetic alone and found three impossibilities. Two are Martha Schuenke, dead in 1930 with sons born 1933 and 1940. THE THIRD PAYS FOR THE WHOLE EXERCISE. JOHANNES ANTONIE OELOFSE DE WITT, born December 1923 at Johannesburg, has a SON OF THE SAME NAME born CIRCA 1923 AT PRETORIA — a father aged nought — and both men die at SASOLBURG. They are one man. And the duplicate carries a fact the original lacks: DEATH 2 APRIL 1963. So IVONE’S FIRST HUSBAND DIED AT SASOLBURG ON 2 APRIL 1963, AGED THIRTY-NINE, and the date was sitting inside a duplicate nobody had looked at. The split also invented a phantom wife and a spurious half-sister.
A citation that cannot be falsified from its own text is not much of a citation Inferred
did The reason the FamilySearch citations are 44 out of 46 sound and the Arolsen ones are 1 out of 7 is not that one archive is better than the other. It is that a FamilySearch citation carries the RECORD’S OWN DESCRIPTION in its text — “Entry for X and Y, 19 Jun 1948” — so a wrong one announces itself without anything being opened. An Arolsen citation is a bare URL. There is nothing in it to check against, which is exactly why six wrong ones survived in a tree that is otherwise careful. The audit is now a tool, tools/audit-tree.py, so it can be re-run rather than remembered.
An Arolsen index entry is a pointer, not a fact Withdrawn
did Four documents were opened with the consent and TWO OF THEM SHOW THE TREE HAS MISATTACHED A RECORD, both in the same shape. Alfred, born 9.9.1917, is made a father of children born in 1933 and 1940; his card says Polish, Catholic, farm labourer, SINGLE, Sachsenhausen, dead in July 1942 at twenty-four. Felix, born 19.2.1923, is given a birthplace of GRAFENAU, GERMANY; his document gives no birthplace at all — it is a Form 11 return of the LANDRAT GRAFENAU dated 19 AUGUST 1946, an Allied listing of foreigners in the American Zone, and its nationality field reads RUSSLAND. The place of registration has become the place of birth. THE RULE FOLLOWS, AND THIS ARCHIVE SHOULD APPLY IT TO ITSELF FIRST: an index entry gives a name, a date of birth and a DocID, and everything else is on the document. The 191 Leibholz index entries recorded here without opening anything are pointers, and the file now says so.
The man the tree calls a father died at Sachsenhausen, single, at twenty-four Withdrawn
did The tree makes ALFRED LUWINSKI, born 9 September 1917, the father of Manfred and Joachim — and cites an Arolsen record as its source. That record was opened. It is a registration card of the Berlin Office for the Registration of War Victims, and every field contradicts the tree: Geburtsort a POLISH place, Beruf LANDWIRT, religion KATH., Nationalität POLN., Familienstand LEDIG — SINGLE — inhaftiert SACHSENHAUSEN, and “† 11.7.42 15:20, Herz- und Kreislaufschwäche”, which is the standard camp formula and is not a cause of death. The reverse gives his home: PRZASNYSZ, PIŁSUDSKIEGO 117. A Polish Catholic farm labourer who never married and died at twenty-four. HE IS NOT THE FATHER OF ANYBODY BORN IN 1933. The tree cites this exact card for Manfred’s 1933 birth and for Martha Schuenke’s 1930 death, and it contains neither fact and neither person. Manfred and Joachim’s father is unknown; the only candidate left is the Alfred Luwinski who was a BÜGLER and a head of household at Putbusser Straße 24 in 1933.
A branch of this family flew from Berlin to Australia in 1965 Documented
did The consent declaration was ticked and the emigration file opened. It is an INTERGOVERNMENTAL COMMITTEE FOR EUROPEAN MIGRATION NOMINAL ROLL, headed MELBOURNE, marked “AUSTRALIAN GOVERNMENT FINANCIALLY ASSISTED”, stamped SEPTEMBER 1965, filed under “Emigrations — transport by airplane”. Four lines of it are one family: MANFRED LUWINSKI, born 14.11.33, aged 32, of BERLIN, occupation MODELL FITTER; HELGA, aged 31; and two children aged ten and three, who are living and are not named here. Forwarding address: c/o Mr H. MARQUARDT, Canberra ACT — and the tree carries Helga as HELGA ERNA MARQUARDT. Her family was already in Australia and brought them out. THE BERLIN DIRECTORY LINES UP EXACTLY: Manfred is a SCHLOSSER at Egellsstraße 16, Tegel, in 1957, 1959 and 1961, and vanishes after 1961. Schlosser and model fitter are the same trade. He vanishes because he left the country. And the tree already knew where he ended up without knowing how: MANFRED JOACHIM LUWINSKI DIED AT BRISBANE IN 2012.
A generation in the tree that cannot be true as written Withdrawn
did Reading the tree against the documents turned up a fault in the tree. It makes ALFRED LUWINSKI, born 9 September 1917, the father of Manfred born 1933 and Joachim born 1940 — and gives his wife, Martha Schuenke, a death in 1930. A father of sixteen, and a mother three years dead at the first birth. The 1933 Berlin directory meets the same problem from the other side: its Alfred Luwinski is a BÜGLER AND A HEAD OF HOUSEHOLD, which fits a man old enough to father a child that year and not a boy of sixteen. Three readings and none chosen. WHAT WOULD SETTLE IT IS IN QUEENSLAND: Manfred died at Brisbane in 2012, and an Australian death certificate names the deceased’s parents.
Ruled out the one cemetery Kurt is actually in Withdrawn
did When this archive swept the Johannesburg municipal cemetery registers it stated its bounds carefully, including that the sweep did not cover the Chevrah Kadisha, “a separate database — a Wear would have no business in a Jewish cemetery, but the exclusion is stated rather than assumed”. That is right about the Wears and WRONG ABOUT KURT. The family tree gives his burial as CHEVRA KADISHA, JOHANNESBURG. He was registered as a Jew in Germany and carried the imposed name out of it. The burial society keeps its own database, one enquiry would give a plot and usually whoever arranged the burial — and it may settle a discrepancy this archive is carrying, because the work list says he died on 15 December 1968 and the tree says 15 December 1969.
Arolsen holds Kurt himself, and six more of his family Documented
did Searching the Arolsen Archives for the whole surname returns NINE records, and seven of them match this archive’s own register by exact date of birth. KURT IS ONE OF THEM — “LUWINSKI, Kurt, 05/11/1912”, DocID 11214358, “1 Incarceration Documents / 1.2.1 DEPORTATIONS AND TRANSPORTS / Part 24: Folder MFN01335”. That is the same DocID already held for Jacob and Auguste, so the three of them are in one folder of five hundred documents, and his index date of birth matches the register to the day. The archive knew the DocID; it did not know Kurt was indexed under his own name inside it. The others sit in four different classes: FELIX in the Allied registration of foreigners and persecutees; MANFRED AND HELGA SHARING A SINGLE DocID in “3.1.3 Emigrations — correspondence and nominal rolls”; ALFRED in post-war evaluations. Two of the nine are people flagged living in the family tree and are not named here. EVERYTHING ABOVE IS THE INDEX ENTRY: every document behind it needs a consent declaration ticked, and this archive does not tick declarations on anybody’s behalf.
Two leads closed by absence, and only one of the absences means anything Inferred
did OSKAR LUWINSKI, the master tailor of Steglitzer Straße 7, is in neither Arolsen nor Mapping the Lives. He exists in ONE LINE of ONE directory and in no record of Nazi persecution at all — consistent with a man who was not persecuted, and one absence in two archives proves nothing. ANTON LUWINSKI of Deuthen is not in Arolsen either, AND THAT ADDS NOTHING: his own 1939 card already reads “race NNNN, basis of persecution unknown”, so a persecution archive should not hold him, and it does not. The two sources agree and neither takes him further. He stops being a separate lead and becomes part of the Ermland case — a Luwinski born twenty miles from Lizzy’s village, waiting on the same undigitised films.
Named the film this archive has been reading for a week without naming it Inferred
did Twice now a West Park register page has been reached by opening images by hand, and neither time was the film written down — so every grave question started from nothing. Behind the indexed collection sits catalogue record olib:4148464, TWO HUNDRED AND TEN FILMS of Johannesburg cemetery registers, twelve of them West Park private graves by year range. Leonard bought grave 56045 on 13 September 1969, so folio W 8 is in FILM 107638608, “WPC Private Graves Registers 1963–1970”, 488 images, and nowhere else. Opened, it reads alphabetically by purchaser surname — image 401 is the T section, Theunissen, Tyson, Terblanche, Taljaard, Taylor — which puts the F folios around 180 to 200. Folio F 3/8 itself was NOT read: the viewer kept reverting to its thumbnail grid and the hunt was abandoned rather than padded out. The film number is worth more than the folio, because the folio may not even be Wear’s.
Found Lizzy’s parish, and found that eight of its nine films are not online Inferred
did A village means a parish and a parish means a register. The FamilySearch catalogue names it in its own words: “Parish register of baptisms and deaths for GROSS BÖSSAU, Ostpreußen, now Biesowo, Olsztyn, Poland. Also includes Kunzkeim, SAUERBAUM, Willims and Kleisak (Kreis Rössel)” — 1690 to 1897, on nine films. EIGHT OF THE NINE ARE NOT DIGITISED. The two that matter both answer “film number not found” while signed in: 7948905, TAUFEN 1793–1890, which would hold Joachim Luwinski’s 1865 baptism and NAME HIS PARENTS; and 7769256, GEBURTEN (KOPIE) 1881–1958, which would hold Lizzy’s birth. The one film that does open was checked to the frame: at image 591 the camera photographed the title board, “GROSS-BÖSSAU / KRS.RÖSSEL / 1798-1847 / †”. Right parish, wrong years.
Seventy-one Luwinskis in the world and not one of them East Prussian Withdrawn
did With the sign-in restored, the surname was matched exactly across all of FamilySearch: 71 records. Twenty-two West Prussia, thirteen Posen, two Silesia, and the rest American immigration and the two South African marriages. NONE IN EAST PRUSSIA — where Lizzy was born. AND THAT IS A FACT ABOUT THE INDEX, NOT THE FAMILY: the Ermland registers are held as catalogued film and no name index reaches them, and the full-text instrument that reads unindexed images returns, for Luwinski, exactly what it returned three days ago — Milwaukee and Chicago naturalisations, all American. One more label to distrust: the viewer files this Catholic parish under “Germany, LUTHERAN Baptisms, Marriages and Burials”. Anyone trusting it would conclude these were the wrong registers and go home.
Lizzy’s own census card, and it answers three questions at once Documented
did AUGUSTE LEIBHOLZ NÉE LUWINSKI, born 28.08.1893 in SAUERBAUM / RÖSSEL / OSTPREUSSEN — and this archive already knew that village, because Johanna Thimm OF SAUERBAUM is Kurt’s great-grandmother and was used as the East Prussian control when counting Berlin directory pages. Her card records the ancestry of all four grandparents and reads “RACE”: NNNN — four times nicht-jüdisch. THE GERMAN STATE CLASSIFIED HER AS NOT JEWISH, ON ITS OWN FORM, IN 1939. And her basis of persecution is given as COLLECTIVE, the term for someone taken as part of a household rather than for their own recorded ancestry — a mixed marriage, which is the standard reason Jacob was not seized until the Fabrikaktion of 26 February 1943. The same record’s family block gives Jacob a place of death the Gedenkbuch leaves blank: AUSCHWITZ.
Fixed the wording, and the page had already said it one paragraph further down Withdrawn
did Correction 5 called Georg “a different Leibholz family” and gave as its reason that Hammerstein is “two hundred kilometres” from Schubin. Both are now withdrawn: SALLY LEIBHOLZ, JACOB’S OWN FATHER, WAS BORN AT HAMMERSTEIN on 7 July 1847, so the distance is not evidence of anything and the surname there is not a stranger’s. WHAT RULES OUT BROTHERHOOD IS THE SIXTY-TWO DAYS, not the two hundred kilometres, and the correction now rests on that alone. The Leibholz page carried the same fault — “These are two Leibholz families. Ours is the Schubin one” — ONE PARAGRAPH ABOVE A PANEL THAT SAYS THE OPPOSITE: “the family had people in both towns… right surname, RIGHT FAMILY, wrong generation”. The page now says in as many words what it used to claim and that it was wrong.
And the panel I had not read gave the letter its best line Withdrawn
did The same panel carries the marriage date: SZUBIN, 24 NOVEMBER 1874. Prussian civil registration began on 1 OCTOBER 1874, so the marriage is SEVEN WEEKS INSIDE IT and must be an ordinary Standesamt entry rather than a Jewish community register. The Bydgoszcz letter drafted an hour earlier had raised that boundary as an open question and asked the archive to check both. It no longer speculates — it gives the date, says which book the record should be in, and keeps the community register only as a fallback. That is the fourth time today something was raised as new that this archive already held, and the cause has not changed: I read the work list and the top of a page, and the answer was further down it.
Hammerstein is not a stranger’s town, and a published correction leans on the idea that it is Withdrawn
did Drafting the Bydgoszcz letter turned up something that cuts against this archive. CORRECTION 5 disputes Georg Leibholz’s brotherhood on the ground that he was born at HAMMERSTEIN, West Prussia, two months and two hundred kilometres from Jacob at SCHUBIN in Posen. The argument treats Hammerstein as a place this family has nothing to do with. IT IS NOT. The front-page chart says the great-grandparents were “Hammerstein · married Szubin 1874”, and the letter sent to Koszalin on 15 September asked for “Standesamt Hammerstein 1884 — Georg; AND SALLY, 1847”. SALLY LEIBHOLZ, JACOB’S OWN FATHER, WAS BORN AT HAMMERSTEIN. The arithmetic still rules out brotherhood — two sons of one mother cannot be born two months apart — but the geography is gone from the argument. A Leibholz born at Hammerstein in 1884 is not a stranger sharing a surname; he is a Leibholz born in the town Jacob’s father came from, WHICH IS WHAT A COUSIN LOOKS LIKE.
No East Prussian index exists, which is why the Thimms cannot be found Inferred
did David asked what could be discovered about the Thimms. All 2,500 FamilySearch collections were enumerated. For the Prussian provinces it has indexed church records for POSEN — 11.6 million — SILESIA, POMERANIA, BRANDENBURG and WEST PRUSSIA. FOR EAST PRUSSIA IT HAS TWO: both Königsberg funeral sermons, both image-only with zero indexed records, ending in 1794 and 1900. There is no East Prussian church index at all. That is also the retrospective explanation for something odd found on Friday — seventy-one Luwinskis in the whole world and not one East Prussian. IT IS THE INDEX THAT STOPS AT THE PROVINCIAL BORDER, NOT THE FAMILY. And a join the archive already had the pieces for: JOHANNA THIMM died at ZEUTERN in 1955, AUGUSTE died at KRAICHTAL in 1975, AND ZEUTERN IS A DISTRICT OF KRAICHTAL. The page lists them as two places. Mother and daughter died in the same small town twenty years apart.
Friederike Schmul is not in the church books, and she would not be Documented
did The Posen collection holds thirty-three people with that exact surname and all thirty-three were read. No Friederike, no Szubin 1874. What IS there is Jewish — Wolf Schmul and Beila Lepotke, Isidor Schmul and Liebchen Katz — including a RICKE SCHMUL, which is a diminutive of Friederike, born in January 1832 and therefore FIFTY-TWO when Jacob was born in 1884. Right name, right province, wrong generation; written down so nobody else chases her. She is absent because those are CHURCH books and a Jewish marriage is not in one. Prussian civil registration began on 1 OCTOBER 1874, so a Szubin marriage that year is a Judenregister or a STANDESAMT entry — and either way it is at Bydgoszcz, where this archive already has an acknowledged letter sitting. A Prussian marriage certificate of 1874 names BOTH SETS OF PARENTS, a generation above the top of the chart.
A Geneteka negative that the control destroyed Withdrawn
did The Polish Genealogical Society’s index was searched for Leibholz marriages in Kujawsko-Pomorskie, 1860 to 1890. Nothing. Then the same query for KOWALSKI — the commonest Polish surname there is — and that returned nothing too. The URL parameters are wrong or the search needs a posted form. THE LEIBHOLZ ZERO MEANS NOTHING, is not recorded as a negative, and Geneteka remains untested.
The file the consent does not reach, and the one it does Inferred
did T/D 52.971, the tracing case for OSKAR LEIBHOLZ, serves METADATA AND NO IMAGES even with the declaration ticked — no document list, no DocIDs, no viewer — and says why: “parts of the file are not displayed for data protection reasons. In such cases, please contact the Arolsen Archives.” The consent opens deportation lists and registration cards and STOPS AT THE CORRESPONDENCE, which is what a T/D file is. One thing the metadata does give: CREATION DATE 1949 TO 2002 — somebody asked about Oskar in 1949 and the file was still being added to fifty-three years later. A letter is drafted and not sent. BUT A DIFFERENT SERIES IS SERVED WITHOUT RESTRICTION: 6.3.1.1, Tracing Inquiries 1945–1946, which returns forty-three topics for this surname.
A woman in Palestine wrote to Jerusalem in May 1946 asking what had become of them Documented
did JEWISH AGENCY FOR PALESTINE, SEARCH BUREAU FOR MISSING RELATIVES, Questionnaire No. 2, “Location of Relatives Abroad”, dated 23 MAY 1946, file 50764. THE PERSON SOUGHT: LEIBHOLZ, WILLI, aged 55, BORN AT STOLP IN POMMERN, last information MARCH 1943 FROM BERLIN, last address BERLIN-NEUKÖLLN, LENAUSTRASSE 25, occupation BUTCHER. With him: ERNA, 48, born HAMMER, wife; HEINZ, 23, son; AND RUTH, 22, DAUGHTER — a child this archive did not have. THE PERSON WHO ASKED: TABORI-HAMMER born NACHUM, RUTH-TRUDE, of BENEI-BRAQ, Palestine, born 9 September 1899 at Hamburg, formerly of Berlin-Moabit, Turmstraße 59, relationship “BROTHER IN LAW AND HIS FAMILY”, signed in her own hand. The Tracing Service answered her through the Jewish Agency on 8 December 1947 and closed the case a fortnight later.
And it settles where Willi was born, which this archive left open this morning Documented
did The Page of Testimony said only “Germany”, and the note written this morning said so in as many words. The 1946 form says STOLP IN POMMERN. Jacob, Samuel and Oskar were all born at SCHUBIN IN POSEN — two hundred and fifty miles and a different province — SO THE LENAUSTRASSE LEIBHOLZES ARE DOCUMENTED AS NOT KURT’S FAMILY rather than merely unproven, and this morning’s caution turns out to have been the right one. Two Hammers survived and both went looking: Ruth-Trude in 1946, and Eva Khava Hammer Olsher, who wrote Willi’s and Erna’s Pages of Testimony later — BOTH THROUGH ERNA’S FAMILY. The honest disappointment is that the row wanted a survivor on KURT’S side, and the file that would give one is the closed one.
Read corrections.json properly, and it caught two more things Inferred
did After raising the Chevrah Kadisha as new when it was already the first entry in the file, the whole of corrections.json was read against the eighty-four open rows. TWO HITS. First, a row plans to have MORRIS ISAAC SALKOW’S CAPE TOWN ESTATE “ride along” with a Cape Town request — and correction 3 says he died and was buried in JOHANNESBURG, Westpark block U grave 2469, on 8 November 1991. A better candidate for the Cape file surfaced in Friday’s FamilySearch sweep: ISAAC MORRIS SALKOW, born 1897 in Lithuania, DIED IN THE CAPE PROVINCE ON 10 OCTOBER 1938, probate 1938–39. The forenames are reversed and the deaths are fifty-three years apart. Nobody should order that file as Ivone’s husband’s until the reference is checked against a date. Second, the row blocked on the Arolsen consent is now openable, and it named its own next step: a TRACING CASE FOR OSKAR, T/D 52.971, fourteen documents — and a tracing file names the person who asked.
Called Jacob a stepfather, and the archive’s own record calls him the father Withdrawn
did This morning’s entry reasoned that because Kurt was born in 1912 and Lizzy married Jacob in 1924, Kurt must be “her son by another man” and Jacob his stepfather. THAT IS AN ASSUMPTION AND THE SOURCE SAYS OTHERWISE: the family tree records JACOB LEIBHOLZ AS KURT’S FATHER, and work-list row 152 states the archive’s own position in the same terms. A CHILD BORN TWELVE YEARS BEFORE HIS PARENTS MARRY IS ILLEGITIMATE, NOT FATHERLESS — and that is precisely why he carried his mother’s surname while his younger brother was registered a Leibholz. Auguste was eighteen in May 1912 and Jacob twenty-seven. THE POINT SURVIVES AND IMPROVES: if Lizzy was NNNN then Kurt’s classification as a Jew came from his father, and the father has a name. Under the 1935 laws a child of one Jewish and one non-Jewish parent was a MISCHLING and did not carry the imposed name; somebody counted as a Jew did. Kurt was counted a Jew — a question with a documentable answer rather than a loose end.
Called it a persecution database, and the card said otherwise Withdrawn
did An hour after writing that Anton Luwinski was “in a persecution database”, his record was opened. “Race”: NNNN. “Basis of persecution: unknown.” Mapping the Lives indexes surviving 1939 census supplementary cards, and being in it is not evidence of persecution. What the record does give is a birthplace: DEUTHEN, KREIS ALLENSTEIN — in the Ermland, about twenty miles from Sauerbaum where Lizzy was born. Two Luwinskis from two villages in the same small Catholic corner of East Prussia. A better reason to look than the one that was claimed, and still not a relationship.
Laid the directory against the memorial database, household by household Documented
did The Berlin address book gives Leibholz HOUSEHOLDERS with trades and street numbers. Mapping the Lives, built on the 1939 census card, gives PEOPLE with dates of birth and death at addresses — 114 of them under this surname. Matched on STREET AND HOUSE NUMBER rather than on name, they line up. “A. Israel, Fleischermeister, Levetzowstraße 13” is ALFRED LEIBHOLZ, born 04.12.1887, DIED 8 AUGUST 1942 — which is why the 1942 volume has no entry for him. And he is not alone at that address: ELSE née SEELIG, died August 1943; KURT, born 1927; LIESELOTTE, born 1923, died December 1943; MARIANNE, born 1921. A master butcher, his wife and three children in one back-building flat. The Weinstraße traveller who became a building labourer had a wife, JENNY née ROSENBERG, dead before the war ended. The Lenaustraße butcher lost ERNA and a son of twenty.
And the control that stops the whole reading running away with itself Withdrawn
did HEINRICH LEIBHOLZ, Verwaltungsbeamter, Leinestraße 32, Neukölln, is in the directory every year to 1942 and gone from 1943. HE DIED ON 9 AUGUST 1948. He survived the war. So a Leibholz vanishing from the Berliner Adressbuch does not mean a death, and the series read off the page images today — seventeen, eleven, ten, nine, six, four — cannot be read as a casualty count. It never claimed to be. Now there is a named man proving why it must not be. One more caution from the same table: ARTHUR LEIBHOLZ, Sanitätsrat, is printed in the 1938 volume and had died in NOVEMBER 1937. These books are compiled a year ahead and carry the dead for a year.
Two Luwinskis in the whole memorial database, and one of them is new Inferred
did Searching it for LUWINSKI returns exactly two people. One is Lizzy — “Leibholz, Auguste née Luwinski, *28.08.1893, Meraner Str. 2”. The other is ANTON LUWINSKI, BORN 16 OCTOBER 1919, at SCHINDEKOPSTRASSE 7, LEHRWERKSTATT OSTPREUSSISCHES BAUGEWERBE, KÖNIGSBERG — the training workshop of the East Prussian building trade. This archive’s East Prussia page is built on nineteenth-century parish entries; Anton is the first twentieth-century Luwinski found in the province, and he is in a persecution database. Neither Oskar the master tailor nor Alfred the presser is there at all. Nothing connects Anton to Lizzy but a surname and a province, which is a reason to look and not a relationship.
A search that turns Luwinski into Löwinsky, caught by opening the page Withdrawn
did Sweeping the Berlin householder section for LUWINSKI across 1900 to 1943 returned the name in two years: 1933 and 1909. A 1909 householder would be Lizzy’s father’s generation, which this archive has never had — and a work-list row was written about him. THEN THE PAGE WAS OPENED. It runs “LÖWENTHAL — 1639 — LÖWISOHN” and there is no Luwinski on it at all: it is Semu Lowinski, Dr Julius Lowinskis, Willy Lowinsky Ingenieur G.m.b.H., R. Lowinsky Nachfolger Dampf-Waschanstalt. The search fuzzes LU to LO and Ö. Four minutes to check, and it would have taken a day to unpick.
Turning that doubt on a finding already published, and finding a master tailor Documented
did If the 1909 hit was a mirage, the 1933 one had to be re-verified too. IT HOLDS — page 1671 runs LUTZ to LUX, the right alphabetical place, and the line reads “Luwinski Alfred Bügler N31 Putbusser Str 24”. But directly beneath it, on the dash, is a name the text layer had hidden: “—OSKAR SCHNEIDERMSTR W35 STEGLITZER STR 7”. OSKAR LUWINSKI, A MASTER TAILOR, with his own business. Lost to the same convention that hid four Leibholz households this morning, twice in one day. There were TWO Luwinski households in Berlin in the only year the name appears at all — A PRESSER AND A MASTER TAILOR, BOTH IN THE GARMENT TRADE. Lizzy Luwinski was a dressmaker and milliner.
Kurt’s mother, perhaps, in Charlottenburg in 1957 Inferred
did The post-war Berlin volumes are not in the series this archive had been querying — that one stops at 1943 — but they exist under other identifiers, and they are set in Antiqua, so the text simply reads. The 1957 Namenteil, page 518: “Fricke Auguste Rentn Chlb 1 Kais-Friedr-Str 79”. AUGUSTE FRICKE, A PENSIONER, KAISER-FRIEDRICH-STRASSE 79, CHARLOTTENBURG. Lizzy was born in August 1893 and died in June 1975, so she was sixty-four in 1957 and “Rentnerin” is exactly right; she is the only Auguste Fricke in the Berlin name sections of 1955 to 1980. A CANDIDATE AND NOT A FINDING: Fricke is a common North German surname and one name in one directory is how the Bloom mariner went wrong. Her death registration of 22 June 1975 would carry an address to compare.
Separated the two Kurts at last Documented
did Every build prints the same complaint: “KURT LUWINSKI matches more than one register entry and is deliberately NOT LINKED” — Kurt Israel of Johannesburg who died in 1968, and a Kurt who died in 1995. THE SECOND KURT IS NOW PLACED. The 1964 Berlin name section gives “Luwinski Kurt Postfacharb -21 Stephan[str]” — A POST OFFICE WORKER IN MOABIT — and the 1970–71 telephone book gives him a number. Two more of the family are there with him: MANFRED, a fitter, at Egellsstraße 16 in Tegel across 1957, 1959 and 1961, and JOACHIM, a stove-fitter, in Ramlerstraße in 1967. The register already holds Manfred Joachim Luwinski born 1933 and Joachim Luwinski born 1940. The two Kurts are now separated by city, trade and decade instead of by a death date alone.
Published that Kurt lived at the hat shop, and it was the same fact twice Withdrawn
did This morning’s entry said the wholesale hatter’s address matched one of KURT’S four Berlin addresses, and that a second matched Oskar’s. IT IS CIRCULAR AND IT IS WITHDRAWN. Those four addresses are not Kurt’s. They are the four Leibholz households in the CENSUS OF 17 MAY 1939 — Jacob, Auguste and Siegfried at Meraner Straße 2; Samuel at Kommandantenstr. 29; Oskar and Gertrud at Waghäuseler Str. 9/10; Georg and Lydia at Michaelkirchstr. 26 — carried on the atlas under a loosely worded pin. Saying the directory agreed with them was saying the same thing twice, and it went into a commit message before it was checked. Third correction of the day.
What the directory actually adds: four men, four trades Documented
did The census gave four addresses and no occupations. The 1938 and 1939 pages give the occupations. JACOB traded as JAQUES — “Jaques Kaufm Schönebg Meraner Str 2” — and printed directly beneath him is “Jaques Leibholz & Co, KONFEKTION, W8 Krausenstraße 59.60”, so the ready-made clothing firm this archive has been watching since 1924 IS HIS. SAMUEL sold hats wholesale. OSKAR was a Kaufmann. And GEORG RAN A LAUNDRY at Michaelkirchstraße 26. None of that disturbs the argument that Georg was not their brother — he was born at Hammerstein, they at Schubin — it just gives four men their working lives.
Read the whole surname off the page images, year by year, and found the law doing its work Documented
did The text layer could not read these pages and this archive said so. The IIIF page images can. Leibholz householders in the Berlin directory: SEVENTEEN in 1938, ELEVEN in 1939, TEN in 1940, NINE in 1941, SIX in 1942, FOUR in 1943. But the number is not the finding. THE NAMES ARE. In 1940 Samuel is no longer Samuel — the entry reads simply “Israel, Hüte, Kommandantenstr 29”. In 1941 two more men carry it, and both have lost their trade in a single year: A. ISRAEL LEIBHOLZ of Levetzowstraße 13 from FLEISCHERMEISTER to ARBEITER, WILHELM ISRAEL LEIBHOLZ of Weinstraße 6 from VERTRETER to BAUARBEITER. IN 1942 EVERY NAME CARRYING “ISRAEL” IS GONE. The four who survive to 1943 — a police official, a nurse, a clerk, a bank official — are exactly the four who never carried it. The directory marks them, and then loses them.
Jaques Leibholz, Meraner Straße 2, last seen in the 1939 volume Inferred
did Item 7 of the Berlin list asked for Jacob at Meraner Straße 2 year by year, on the reasoning that “the year he stops appearing is the year the record of his life in that city stops”. He is in the 1938 volume and the 1939 volume and NO VOLUME AFTER. The 1936 street section lists the tenants of Meraner Straße 1 to 5 and no Leibholz is among them, so the household arrived between 1936 and 1938. He was deported from Berlin on 26 February 1943; the directory loses him three volumes before that. One entry away from him, in 1940 only, is “Auguste, WITWE, Schöneberg, Badensche Straße 9” — and the archive will not join those two up, because Jacob was alive until 1943 and a widow in 1940 cannot be his wife unless something happened that nobody here knows about.
Measured the wrong thing twice in one day, on one page Withdrawn
did The morning’s correction was that a page count had mixed four sections of the book. The afternoon’s is worse and on the same page: IN A GERMAN DIRECTORY ONLY THE FIRST ENTRY OF A RUN SPELLS THE SURNAME, and every entry after it begins with a dash. Searching the text for “Leibholz” finds one line in ten. The file published at lunchtime said that from 1939 nothing was left under the name but a printer’s cross-reference. FOLLOWED PROPERLY, THE 1940 VOLUME HOLDS TEN LEIBHOLZ HOUSEHOLDERS (this entry said ELEVEN until 23 September; the eleven belongs to the 1939 volume, and the enumeration here has always listed ten) — a master butcher, a police official, a widow, a pensioner, a clerk, a bank official, the hat wholesaler and two Wilhelms. What saved the published version was that it had refused to state the disappearance as a finding. The refusal was right; the description underneath it was wrong.
One page number, four grooms, and two brick walls on the same certificate Inferred
did SARAH BLOOM, marriage registered Q3 1842, STEPNEY, VOLUME 2 PAGE 430 — and the other names printed on that page are WILLIAM WEARE, Thomas Moyes, William George Smith and James Wicks. Before 1912 the index does not name the spouse, so this is one-in-four on a page and not a marriage. But the certificate behind it would name BOTH FATHERS: William Wear’s father, who is the brick wall at the top of this line — born about 1823 at Newington, before civil registration — and Sarah Bloom’s father, which is the whole point of the Yarmouth enquiry. Two of the oldest open questions in this archive now hang on one order.
A year filter that answers zero instead of filtering Withdrawn
did The row above had been recorded as a failure: “not found, and the failure is recorded”. It was not the family, it was the search box. On Findmypast’s England & Wales Marriages, PUTTING A YEAR ON A SEARCH SILENTLY RETURNS NOTHING. The control that proves it: James Shardalow Mort married at Yarmouth in 1885, the row was read on screen — then searched again for that same surname and place with the year set to 1885 give or take two, and it returned NOTHING AT ALL. Both parameter spellings do it. The casualty here was measured: this archive had recorded three Sarah Bloom marriages in 1841–1853, and unfiltered the search returns SIXTY-SEVEN. Every year-filtered marriage search run here has to be redone.
Counted four sections of a book and called it one family Withdrawn
did The Berlin address-book sweep reported LEIBHOLZ page hits falling from 24 in 1933 to 2 in 1943 and read it as a surname leaving a city. A Berlin volume has FOUR SECTIONS: householders and firms, the trades index, institutions, and the street directory. In 1933, twenty-four pages carried the name AND EXACTLY ONE OF THEM WAS THE HOUSEHOLDER SECTION. The other twenty-three were the street and trade indexes, where a single household generates a dozen hits. The count measured the size of the book as much as the size of the family, and the Thimm control did not catch it because the control had the same flaw. The name section has now been read on its own, volume by volume, 1920 to 1943.
Read a garment presser off a page and told two Alfreds apart Documented
did The 1933 householder section, page 1671: “LuwinSki Alfred Bügler N31 Putbusser Str 24” — ALFRED LUWINSKI, A GARMENT PRESSER, A HEAD OF HOUSEHOLD in Wedding. The Arolsen index holds an Alfred Luwinski born 09.09.1917, and a boy of sixteen is neither a householder nor a qualified presser. THEY ARE NOT THE SAME MAN. Father and son is the likeliest reading and the archive does not assert even that. Luwinski reaches the householder section in ONE YEAR of the entire run, which is how rare the name is in Berlin.
The family was in the rag trade on both sides of the equator Inferred
did The Leibholz entries in the Berlin householder section are not scattered families. They are BUSINESSES, and they are all clothing: Jaques Leibholz & Co., KONFEKTION, W8 Krausenstraße; M. Leibholz, TRIKOTAGEN und WOLLWAREN, C2 Rosenstraße 17; and SAMUEL LEIBHOLZ, “HÜTE EN GROS” — HATS, WHOLESALE. Kurt Luwinski kept a hat shop, the Piccadilly Hat, at 81 Eloff Street, Johannesburg. Lizzy was a dressmaker and milliner. The trade did not begin in Johannesburg. Whether the hat wholesaler IS Samuel Israel Leibholz of the Auschwitz death register is a candidate and not a finding — same uncommon surname, same city, same decade, and two addresses that have not yet been compared.
Refused to publish a disappearance that the page could not support Withdrawn
did From the 1939 volume onward the only thing the text layer returns under LEIBHOLZ is the printer’s cross-reference “see also Leipholz” and a LEIBHOLD Wilhelm, who is a different surname. Written as a finding that would have read beautifully: the name leaving the register of Berlin householders in 1939. IT DOES NOT HOLD. Page 1719 of 1939 comes back as ONE UNBROKEN BLOCK of seventeen thousand characters with the columns interleaved, and its running head still says Leibholz. Whether an entry sits beneath it is not legible. Counting works; reading a named line needs the page image, which was written down as a risk before the run and is now a measured fact.
Martha Giddings did not die in childbirth Documented
did Three General Register Office entries, all Hitchin: the marriage of George Castle to a Giddings in Q3 1862; the birth of GEORGE WILLIAM CASTLE in Q4 1862, within three months of the wedding; and the death of MARTHA CASTLE in Q3 1863, volume 3A page 188. She died nine to twelve months after her son was born, not at his birth. That also corrects a sentence in an unsent letter — “she cannot have known her son at all” — which is wrong by about nine months. The bound is stated rather than buried: the marriage index does not print the bride’s forename and the pre-1866 death index carries no age, so two certificates are still needed to close it.
Lost a woman to the surname Smith, and measured how completely Inferred
did Ada May Castle married Sidney Smith, and that is where she leaves the record. The England and Wales death index gives THREE HUNDRED AND SIXTY-ONE Ada May Smiths born 1885–1889. Narrowed to Hertfordshire with a wildcard: seventeen Ada Smiths and not one carries the middle name MAY. The 1939 Register, which prints exact dates of birth and would have keyed the death index in a single step, holds two Ada Smiths in the whole county and neither is she — and her husband is not at Hitchin in 1939 either. They left the county. A national search on that surname and that forename is not a search.
Counted an index instead of trusting its title, twice in one morning Withdrawn
did FamilySearch names a collection “South Africa, Civil Marriage Records, 1801-1974”. Counted year by year it holds 178,037 marriages for 1960 and ONE HUNDRED AND TWENTY-FOUR for 1961. It is a name index to 1960. The death collection called 1953-1967 is a name index to 1966: 59,263 entries for that year, 178 for the next. THIS MATTERS BECAUSE OF WHO FALLS ON THE FAR SIDE OF IT — Kurt died in 1968, Henry Robert in September 1969, Leonard in 1973 or 1979. All three are missing from an index that stops before they died, and a fortnight of searching had been reading that absence as a fact about them. It is a fact about the index.
Read every South African record FamilySearch holds for this family Documented
did The work list said eleven record sets. All 3,502 FamilySearch collections were enumerated and filtered: THERE ARE THIRTY-TWO. Matching the surname exactly rather than loosely drops 16,156 rows to 111, and all 111 were read. TWO NEW RECORDS. PHYLLIS AGNES WEAR MARRIED ROBERT SELWYN NORQUOY ON 14 JUNE 1952 at Malvern — which is where the defendant’s surname in WLD 790/1957 comes from, a name this archive had from a court index and could not explain. And KATHLEEN EDITH WEAR AND JAMES CECIL FOOT, 26 May 1945, in an ANGLICAN parish register: the marriage was known from a transcription, the record was not, and it is the first documentary evidence of this family’s denomination in South Africa.
Wrote up three discoveries that were already on this site Withdrawn
did The same sweep produced what looked like five new records. Three of them are published here already, and one of them far better than an index can put it — Ivone’s first marriage at GERMISTON on 2 February 1948 has been READ, entry 27 before magistrate A. J. Rossouw, and Frances’s remarriage to Rolf Eisenhardt on 23 May 1958 has been on Kurt’s page for days. The error cost nothing because the site was checked before the page was written, but it is recorded rather than quietly fixed: AN INDEX AGREEING WITH A DOCUMENT YOU HAVE ALREADY READ IS CORROBORATION, NOT A FINDING, and a tool that answers in two seconds makes it very easy to forget which of the two you are looking at.
Kurt’s third marriage was not the end of his story Inferred
did FRANCES HOSTOWSKY, BERNER, SEFTON, LUWINSKI — the surname stack is four marriages — REMARRIED ON 23 MAY 1958, to Rolf Eisenhardt. She had married Kurt on 29 February 1956. That marriage lasted two years and three months, and Kurt lived another TEN YEARS after it ended. The archive has been closing his life at the 1956 register entry. There is a divorce between those two dates, in the same court as the five Luwinski files already held, and nobody has looked for it.
Found the one document that would answer everything, and it is not online yet Inferred
did A South African DEATH NOTICE names the dead man’s parents. FamilySearch holds “South Africa, Transvaal, Estate Files, 1950-1993” — 191,580 images of exactly those notices, from the National Archives in Pretoria — and its title brackets Kurt in 1968, Henry Robert in 1969, both of Leonard’s candidate years and Martha Maud in 1986. ITS WAYPOINTS SAY IT IS PUBLISHED FOR 1955 TO 1960 ONLY. Seven of them. The Orange Free State equivalent has fifty-seven, running to 2006, so the gap is a publishing schedule and not a destroyed record. Checking whether that has changed costs one request and returns a number.
Leaked three living match holders into a build, and found out from a guard that was not running Withdrawn
did Two named living people reached the published pages and a third reached the work list. The check that exists precisely to stop this — it compares every published page against the private match list — HAD NEVER BEEN WIRED INTO THE BUILD CHAIN. It is now, and it runs on every build: 111 match holders checked against every page. All three names were stripped. A guard that is written and not called is not a guard.
CORRECTED: Martha Giddings had her son for something under a year Withdrawn
did This archive wrote, and put in a letter to a living second cousin, that if Martha Giddings died in 1863 she could not have known her son at all. SHE DIED IN THE THIRD QUARTER of 1863; GEORGE WILLIAM WAS BAPTISED ON 4 JANUARY of it. Two numbers, both already on file, and the quarters were never compared. It changes the reading of the 1871 census as well: the widower is IN the Giddins house at Norton End, enumerated as son-in-law with his boy as grandson, so he did not hand the child over — he went to his dead wife's parents and stayed eight years. A correction has been drafted to the cousin who was told the wrong thing.
Found Caroline where nobody could follow her, because she changed her name Documented
did Every tree loses Caroline Wallis after her husband died in 1916. She remarried in Q3 1920 at Hitchin — CAROLINE CASTLE to JAMES JAMES — and was CAROLINE JAMES for the last thirty-seven years of her life, which is why her death in 1957 could not be found under either of the surnames people were searching. The 1939 Register has her at 38 Norton Road, Baldock, widowed a second time, and gives her exact date of birth: 28 FEBRUARY 1863, not the 1862 every secondary source repeats.
The transport has a number, and the memorial book was not the one we thought Documented
did Yad Vashem had been used here for exactly one list. Searched properly, LEIBHOLZ returns 88 victim records across four collections — and Jacob's record carries a field this archive never had: "Details of Transport: TRANSPORT 30 from Berlin to Auschwitz Birkenau, Extermination Camp, Poland on 26/02/1943." The date has been on these pages for a fortnight. THE TRAIN HAD NOT. It also gives his citizenship as GERMANY, which matters to the statelessness argument two pages away. OSKAR'S RECORD closes the distance between his two deportations: Berlin to Theresienstadt 17 March 1943, Theresienstadt to Auschwitz 16 October 1944, and DATE OF DEATH 16 OCTOBER 1944 — the same day the second transport left. AND THE SOURCE BEHIND THE TRANSPORT NUMBER IS A BOOK THIS ARCHIVE DID NOT KNOW IT WAS STANDING ON: not the Bundesarchiv Gedenkbuch cited on every page here, but the GEDENKBUCH BERLINS DER JÜDISCHEN OPFER DES NATIONALSOZIALISMUS, Freie Universität Berlin, Edition Hentrich, 1995 — Berlin's own. It has never been consulted directly and it should be. One discrepancy settles against Arolsen and in the Bundesarchiv's favour: Yad Vashem has GEORG born 1884, not 1885. And one decoy is written down before it costs anybody a day — "KURT LEIBHOLZ, 1927, Berlin, murdered" IS NOT OUR KURT, who was a Luwinski, born 1912, and died in Johannesburg in 1968.
Three of them are on one document, and one of the three had been gone ten years Documented
did The Arolsen Archives had sat on the sources page under "identified, not yet worked" since 14 September. Worked on the 17th. The index needs no consent from anybody and it holds 213 Leibholz entries. It confirms JAKOB 08.08.1884 SCHUBIN and OSKAR 13.05.1887 SCHUBIN from records with no connection to the Bundesarchiv — and only those two Leibholzes in the whole index are recorded as born at Schubin. It carries two of the brothers under the compulsory name, LEIBHOLZ ISRAEL OSKAR and LEIBHOLZ GEORG ISRAEL, which puts the 1938 decree published here this morning onto the family's own paper. It holds a post-war tracing case for Oskar, T/D 52.971, fourteen documents, which exists because somebody asked. AND THEN A SEARCH FOR AUGUSTE RETURNED SOMEBODY THIS ARCHIVE DID NOT KNOW WAS IN THE PERSECUTION RECORDS AT ALL: "LEIBHOLZ, Auguste, maiden name LUWWINSKY, born 28.08.1893". That is Lizzy, carried here for a fortnight as 1893-1975 with no date — and the maiden name is spelt with a DOUBLED W, a seventh form of the surname and the first that is German rather than South African. THEN THE DOCUMENT IDENTIFIERS WERE COMPARED. Jacob, Auguste and Kurt ALL RESOLVE TO DocID 11214358, folder MFN01335, under Deportations and Transports. IT IS ONE DOCUMENT. The father who was deported, the mother who was not, and the son who had been in Johannesburg for ten years, all three named on a single sheet. This morning it was written here that Kurt was probably "named inside a document about somebody else"; by the afternoon the somebody else had a name and it was his father. WHAT STAYS LABELLED: the shared DocID is a hard fact from the index; what kind of document it is, is not, because nobody here has read it — opening one requires ticking a declaration of consent about the use of sensitive personal data, and this archive does not tick that on anybody's behalf.
A surname leaving a city, and a control that proves it is not the scanner Documented
did David asked what more could be done on the Leibholz and Luwinski side. THE WHOLE LINE LIVED IN BERLIN AND THIS ARCHIVE HAD NEVER LOOKED IN THE BERLIN DIRECTORY. The Zentral- und Landesbibliothek has digitised the Berliner Adressbücher from 1799 to 1970 — one volume a year, six to seven thousand pages each, householders by NAME, TRADE AND STREET — free, no account, answering a JSON query. 1,047 page hits for LEIBHOLZ, 23 for LUWINSKI. THE FIRST THING OUT OF IT: Leibholz page hits per annual volume run 24 in 1933 and hold between 15 and 24 through the thirties — then 12 in 1940, FIVE in 1941, six in 1942, TWO IN 1943. AND THE CONTROL IS NOT AN IDLE ONE. THIMM — Johanna Thimm of Sauerbaum is Kurt's own great-grandmother, East Prussian and not Jewish — runs 66, 53, 47, 65, 72, 68, 45, 75, 69, 63 AND 75 IN 1943: as common in the last volume as in the first. The scanner did not get worse, the Fraktur did not get harder, the book did not get smaller. THE LIMITS GO IN WITH IT AND THEY MATTER MORE THAN THE NUMBER: a page hit is a page and not a person, twenty-four to two is a small base, and DEPORTATION, EMIGRATION AND ADMINISTRATIVE EXCLUSION FROM DIRECTORIES ALL LOOK IDENTICAL IN A COUNT — Jews were being struck out of German commercial listings from 1938 quite apart from being deported, and a count cannot tell those apart. AND A LEAD: the Luwinskis of Berlin are in the garment trade. ALFRED LUWINSKI, BÜGLER — a presser — at Putbusser Straße 24 in Wedding in 1933 and 1936, a lorry driver by 1943, while Lizzy was a dressmaker and milliner in the same city. A brother is the obvious guess and it is a guess.
Two hundred and nineteen commits citing Kew, and the catalogue had never been opened Documented
did David asked whether this archive had ever looked at The National Archives. IT HAD NOT. The pages carry RG10, RG11, RG12, RG13, RG14, RG15, BT 27, WO 329, WO 372 and WO 392 — hundreds of references to records held at Kew — and every one came second-hand through a commercial index's transcript. Discovery is free, needs no account and answers JSON. The first hour produced four things. EVERY CITATION VERIFIES, and two verify BETTER than they were published: WO 392/15 is specifically "Section 5: South Africa (Union Defence Forces)", which is exactly the claim this archive makes about Leonard Wear, and WO 329/1120 adds the page range 312-622. WO 372/4/50790 IS THE EXACT ITEM REFERENCE for Fred James Castle's medal card, in Kew's own words — "Scottish Rifles 10061 Serjeant. Scottish Rifles 7815432 Serjeant. Machine Gun Corps 195568 Serjeant." Findmypast gave only the piece and PRINTED THE DISCOVERY ID ON ITS OWN TRANSCRIPT; this archive had it for hours and never followed it. DP/12/1/14 AND DP/12/1/15 ARE THE BALDOCK BAPTISM REGISTERS, 1880-1895 and 1895-1910, at Hertfordshire Archives — the register this archive had been gesturing at all day, now an orderable reference. And LUWINSKI RETURNS ZERO across 2,500+ British archives, filed with the reason a catalogue is not a name index. A reusable tool went in with it, tools/tna.py, because the failure here was not ignorance of the source but never having reached for it.
Found him: Sergeant 10061, Cameronians (Scottish Rifles), and a career soldier into the 1920s Inferred
did FREDERICK JAMES CASTLE is the only Castle male line this archive has and he had been missing from every English record since 1901. He is found, and the chain runs: the 1901 census has him at eleven at NORTON END, BALDOCK, son of George W and Caroline; the 1911 census has a Frederick Castle, 21, single, born Herts Baldock, in MEEANEE BARRACKS, COLCHESTER — and the ORIGINAL IMAGE, read at 299 per cent, gives column 5, UNIT OR ARM OF SERVICE, as SCOTTISH RIFLES, which the transcript leaves blank; and the campaign medal roll, TNA WO 329/1120, gives FRED JAMES CASTLE, SERGEANT, 10061, CAMERONIANS (SCOTTISH RIFLES), British War Medal and Victory Medal. THIS MORNING THE IDENTIFICATION RESTED ON ONE CENSUS AGE. It now rests on two matches arrived at from opposite ends — the forenames FRED JAMES, exactly as the General Register Office registered him in 1889, and the regiment, exactly as an enumerator wrote it over his line in 1911. AND HIS MEDAL INDEX CARD, WO 372/4, EXPLAINS EVERY ABSENCE THE ROW WAS BUILT ON: Sergeant throughout, THREE service numbers — 10061 Scottish Rifles, 195568 MACHINE GUN CORPS, and 7815432 Scottish Rifles. An eight-digit number is a post-1920 issue. HE WAS STILL A SERGEANT AFTER THE WAR, which is why he is in no 1921 household, no 1939 Register and no English death index. Not proved, and the file says so: the medal roll carries no birth date and no birthplace, so the link is forenames plus regiment and not a document stating identity. A service record would state it, because it names next of kin.
Nine born, eight living, one died — and all nine are now named Documented
did The 1911 census asks a married woman how many children she has borne. Caroline answered NINE. This archive could name seven, and the baptism register turned out to stop in 1879. It closed anyway, from the household. THE 1901 CENSUS HAS THE FAMILY WHOLE — RG13/1298 folio 49 page 26, schedule 192, NORTON END, NORTON, BALDOCK: George W Castle 36, Caroline 36 born WESTON, and five children including FREDERICK, 11, and MARTHA, 5. THEY HAD GONE BACK TO NORTON END, the hamlet where George William was himself raised by his GIDDINS grandparents after his mother died; he brought his own family home to it. THE NINTH CHILD IS SARAH ANN WALLIS, born 19 August 1882 at Weston, TWO MONTHS BEFORE HER PARENTS MARRIED AND REGISTERED UNDER HER MOTHER'S MAIDEN NAME — not a Castle birth in the index at all, which is why no search for Castle births could have found her. AND THE ONE WHO DIED IS NAMED: SIDNEY CHARLES CASTLE, born Q2 1898 at Hitchin, DIED Q4 1900 AT HITCHIN, AGED TWO. One entry in each index, no competitors, and he was gone five months before the 1901 census, which is why no census ever showed him. The arithmetic closes exactly. AND THE SIX-INSTEAD-OF-NINE WAS THREE SEPARATE INDEXING FAILURES: Sarah Ann filed under Wallis, Martha Maud with a blank mother's maiden name, Sidney Charles missing too. A filter on a field that is sometimes empty is a filter on the indexing, not on the family.
Frederick Castle was a soldier, which is why nobody could find him at home Inferred
did He is in the 1901 census after all, aged 11 at Norton End — the earlier search that came back wrong had found the OTHER Norton End Castle family, three schedules away on the same night. And in 1911 there is a BALDOCK-BORN FREDERICK CASTLE, 21, SINGLE, AT MEEANEE BARRACKS, COLCHESTER: RG14 piece 10304, schedule 9999, Military Road and Mersea Road. A regular soldier is not in anybody's household. The age is the discriminator and it is thin but real — 21 on 2 April 1911 means born between April 1889 and April 1890, which fits ours and excludes both of the other Baldock Freds. IT IS NOT PROOF and it is labelled probable: the entry says Frederick, not Frederick James, and it rests on one age and a birthplace. What it changes is the shape of the question. The 1921 census holds ONE Baldock-born Frederick Castle in all England and Wales and it is the other family's; ours is out of the country by 1921, is not in Soldiers Died, and is in no death index. A regular soldier who vanishes like that went abroad and stayed, or died in a service Soldiers Died does not cover.
Part of this morning's withdrawal comes back, in a different record set Documented
did The electoral registers were withdrawn today because the set tested — 1832 to 1932 — returns page images and indexes no elector at all. There is a SECOND AND DIFFERENT SET, ENGLAND & WALES ELECTORAL REGISTERS 1910-1932, AND IT CARRIES NAMES: a search for Castle in Hertfordshire returns Arthur Frederick Castle in 1929 and 1930, Frederick Castle in 1919 and 1920. It cannot do what the withdrawn row wanted, because it starts in 1910 and the question was Bermondsey 1885 to 1900. It can place the Baldock Castles year by year from 1919 — the decade in which Caroline became Caroline James and Frederick disappeared. The withdrawal stands; what was wrong was the claim about one set, not the idea.
Koszalin answered, and the record that would settle Georg Leibholz does not exist Documented
did The first of eleven institutional letters to come back. Reference O-S.6342.406.2026, signed by the Head of the Slupsk Branch: the archive holds NO registers of the German civil registry office at CZARNE — Standesamt HAMMERSTEIN, Kreis Schlochau — for 1874 to 1899, and NO books of the Jewish community there at all, and NO information about what became of either or where they might now be. That last clause is what makes it a good letter rather than a brush-off. Hammerstein is where GEORG LEIBHOLZ was born on 7 June 1884, and an 1884 birth entry is the one document that would name his parents and settle whether he belongs to this family. This archive holds that he does not — Jacob was born at SCHUBIN in Posen on 8 August 1884, two months later and in another province — and that finding now rests on the Gedenkbuch alone, because the record that could overturn it is gone. SALLY LEIBHOLZ, born 1847, predates civil registration entirely, so only the synagogue books could have held him, and those are the ones that do not exist. Acknowledged the same day, in Polish, in their own thread.
The baptism register ends five years before the family starts Documented
did The 1911 census has Caroline answering NINE CHILDREN BORN, EIGHT LIVING, ONE DIED, and this archive can name seven. The baptism register names both parents on every line, so it should have named all nine. ALL 823 CASTLE BAPTISMS IN THE HERTFORDSHIRE SET WERE SORTED BY MOTHER'S FIRST NAME AND THE CAROLINE BLOCK READ: between the Barbara block and the Catharine block there is EXACTLY ONE ROW. The reason is a coverage bound and not a search failure — sorted by year, the latest of 195 Castle baptisms at Baldock is 1879, and the children were born 1884 to 1904. Martha Maud's own Baldock baptism of September 1895 is not in the set either. THE INSTRUMENT ENDS FIVE YEARS BEFORE THE FAMILY STARTS, the two unnamed children stay unnamed, and the answer is in the parish register at Hertfordshire Archives rather than in anything searchable from here. THE ONE ROW IS WORTH HAVING: ADA MAY CASTLE, BAPTISED 10 APRIL 1887 AT WESTON, father GEORGE WILLIAM CASTLE, LABOURER, mother CAROLINE. The first baptism read for any of Caroline's children, and it puts the family at Weston — her own birthplace, and where Sarah Ann Wallis was born two months before the marriage.
Ada May did not die a Castle, and a second woman goes missing the same way Withdrawn
did The harvested match-holder's tree records "Ada May Castle, died Jul 1969, Hitchin". SHE MARRIED IN Q4 1909 AT HITCHIN, volume 3A page 1171, aged twenty-two. And the death index has no such entry: Ada-anything Castle born 1884 to 1890, all England and Wales, returns sixteen rows, NOT ONE at Hitchin and NOT ONE in 1969. A woman who married in 1909 did not die a Castle sixty years later. This is the CAROLINE JAMES trap in miniature, and it is the second time in two days the same mechanism has hidden a woman from this archive — a tree recording a death under a maiden name its owner had long stopped using. Her married surname is NOT guessed here: MarriageFinder offers two page-mates from the index and page-mates are not a marriage. The general lesson is worth more than Ada May — every married daughter in that tree with a death recorded under CASTLE is now suspect.
Called a result count a source, for the third time in a week Withdrawn
did Yesterday this archive filed a row saying the England and Wales Electoral Registers had returned NINE ENTRIES FOR WILLIAM WEAR, Surrey and London, 1885 to 1900, and called them the best unworked source for the Bermondsey Wears — an ANNUAL address record covering the years that bracket both brothers' deaths. TODAY ONE WAS OPENED. Every row in that record set says "NO NAME ON RECORD" and carries a PDF icon: they are page images of volumes, not entries about people. Searched without a place filter, WILLIAM WEAR returns 3,478 of them in DEVON and DURHAM — Weare in Devon, the River Wear in Durham. JOHN SMITH returns the same nameless rows in Lancashire. LUWINSKI returns zero, and so does a nonsense string, so the fields are not inert; they simply match a place. The "Surrey and London" came from a place filter typed by the same hand that then reported it as a finding. A RESULT COUNT IS A PROPERTY OF A SEARCH BOX. This archive has now made that mistake over two Giddins DNA matches, over two archives-index zeroes that were three records each, and here — and it wrote the warning after the first one. WHAT SURVIVES: the Bermondsey volumes are digitised and can be READ page by page for 1885 to 1900. It is an afternoon of images, not a search, and the row now says so.
TWENTY-ONE EVIDENCE FILES AND NOT ONE BUILD SCRIPT READ ANY OF THEM Withdrawn
did Asked to check for a fault the sister archive at Mazza had just been caught by — a per-person builder that never opened its own evidence files, so twelve documented people were told on their own pages that nothing had been read about them. This archive has no per-person pages, so it could not fail in that exact shape. IT FAILED IN A WORSE ONE. Every file in data/ was audited against every build script: build-register.py reads the MyHeritage export and corrections.json, the two living-persons guards read the export and a hash file, and THAT IS THE WHOLE LIST. All twenty-one .tsv evidence files — the Gedenkbuch, six censuses, the 1939 Register, the passenger lists, the court files, the Bermondsey households — were read by NOTHING. Their contents reached the site only where a human had retyped them into a paragraph. The clearest casualty: the register printed Samuel Israel Leibholz's birth as "June 28 1877?", the source tree's question mark intact, while data/leibholz-gedenkbuch.tsv held Bundesarchiv entry en909753 giving 28 June 1877 at Schubin and his death at Auschwitz on 10 April 1943. The archive had the citation and the page showed the doubt. Eleven people — three Blooms of Yarmouth, four Hawkinses of Watford, two Castles, Gertrude Wear, a Letchworth lodger — were documented in the repository and appeared NOWHERE on the published site at all.
Built the evidence in, and then a gate so it cannot fall out again Documented
did tools/build-evidence.py now reads every evidence file through an EXPLICIT MANIFEST and writes what each record says about each person; /evidence/ publishes all 135 records about 100 people; and the register carries a RECORD READ column. tools/check-evidence.py runs on every build and refuses it three ways: a file in data/ the manifest does not declare, a stale build, or a person named in an evidence file reaching the published site with neither a record nor a correction. All four failure modes were reintroduced deliberately and all four exit 1. AND IT CHANGES NO DATE. The register and the charts still draw the export as given — the question mark on Samuel's birth is still there, because that is what the source says. The record now sits beside it instead of in a file nobody reads. Where a named source genuinely overrides the family record that is a correction, carried as was and now, and it is a different mechanism on purpose.
The new builder hung three divorce actions on the wrong Kurt, and the fix is that it now refuses to choose Withdrawn
did Written in the same hour as the entry above, and caught by reading its own output. The register holds KURT LUWINSKI, died 1995, a surname-index entry, and KURT ISRAEL LUWINSKI, 1912 to 1968, who is the man this archive is about. The linker tested exact names first and only fell back to first-plus-last, so the index man matched exactly and won while the real one never got a vote — and WLD 1950/1948, WLD 497/1951 and WLD 2368/1951 were attached to a stranger who outlived Kurt by twenty-seven years. It now pools every candidate before deciding and LINKS NEITHER when more than one person answers to a name, naming both on the page so a human resolves it. A name match is not a person match. This archive has now said that about a census index, a surname-tagged DNA match, three Frederick Castles, eighteen Bermondsey Wears and two Emma Hawkinses — and then written a script that did it anyway. The same check also turned up a DUPLICATE the register's own collapse list does not cover: Johannes Antonie Oelofse De Witt is in the source twice, once with a death date and once without.
And the second guard caught the first one leaking Documented
did The first build of the new evidence page failed check-matches: a free-text note in the match-derived Castle file identifies a living DNA match holder as somebody's grandson, and publishing the file published him. The builder now carries a per-file redaction list and that column never reaches the page. The guard written eight days ago, after three living people were leaked into a build, caught the tool written to fix a different problem on its first run.
Followed the Pix Road household to the end of all four lives Documented
did The Pix Road find opened one question worth the day: Louisa Shenton was Henry Robert Wear's sister, so a descendant of hers would carry the WEAR half of Derrick's ancestry and not the Castle half — the independent test this archive has wanted since it read the segments. THE ANSWER IS NO. Births, surname SHENTON, mother's maiden name WEAR: one result, STANLEY W SHENTON, Q1 1924, Hitchin. Deaths, surname Shenton, born 1924: THE SAME CHILD, Q1 1924, Hitchin. Born and died in the same quarter, to a mother of forty-one. And the trap that had caught this archive over Martha Maud was checked for: the district was searched with NO mother's name at all, Shenton at Hitchin across twenty-one years, and there is exactly one Shenton birth in it. WILLIAM SHENTON married a Wear at Hitchin in Q3 1918 and died there in Q3 1953. LOUISE S A SHENTON died at Hitchin in Q4 1967, aged 84 — and the initials close it, because her birth is registered as LOUISA SARAH A WEAR, Q1 1883, at KINGSTON, SURREY, which is a place neither of her parents had ever been put before. AND EMMA HULL DIED IN Q2 1944 AT HITCHIN, AGED EIGHTY-FIVE, four years after the grandson she had known as a boy was taken prisoner. Eighty-five years from Aldenham to Letchworth, two husbands buried, and thirty miles.
Wrote a grandmother over her own mother, by matching on a name Withdrawn
did The Pix Road finding was applied to the tree by a script that selected people BY NAME. Emma Hawkins the daughter, born 7 February 1859, and Emma Hawkins the mother, born about 1822, are two entries two generations apart with identical names — and the code did not look at the generation. For part of 17 September this archive published that a woman born about 1822 was blind at Letchworth in 1939. The same run appended a sentence to the Giddins household at Norton End saying that THEIR daughter Emma was born on 7 February 1859; their Emma is EMMA GIDDINS, born about 1855 at Norton, a different woman in a different parish who shares only a forename. Both are withdrawn on the pages themselves. A match on a name is not a match on a person — which is the thing this archive says about every census index it touches, said while doing it.
Caroline James stops being an index line and becomes a household Documented
did The remarriage had rested on a GRO index entry. The 1921 census settles it from both directions. There is NO Caroline Castle anywhere in Hertfordshire in 1921 — she leaves the surname exactly when the index says she changed it. And there is one Caroline James: RG15/7079, schedule 125, 9 BOSCOMBE PLACE, LETCHWORTH, wife of JAMES JAMES, born at WESTON, Hertfordshire — with ERNEST ALBERT CYRIL CASTLE and FLORENCE MARGARITE HILDA CASTLE in the house, enumerated as his STEP-CHILDREN. That is an enumerator writing down a remarriage. Her age, 56 years 4 months, puts her birthday in February and agrees with the 28th; the year is two out and she is the one who gave it. She had also moved to Letchworth, where her daughter was raising the Wears.
Nine children born, eight living, and only six have been found Documented
did The 1911 census at St Mary's Cottages, Baldock — RG14/7572, schedule 173 — asks a married woman how many children she has borne. Caroline answers NINE BORN, EIGHT LIVING, ONE DIED. This archive has names for six from the GRO index and a seventh in Martha Maud. So at least two Castle children have never been identified and one of the nine was already dead by 1911. Her own father, JOSEPH WALLIS, is in the house as well, spelt Wallis — a seventh independent instance of the spelling that was in dispute a day ago.
Excluded the wrong Frederick, with an exact date this time Inferred
did Frederick James Castle, born 1889 at Hitchin to a mother née Wallis, is the only Castle male line that could ever carry a Y test. The Frederick James Castle who died at Hitchin in 1975 is NOT him: the 1939 Register has that man at 17 ORCHARD ROAD, BALDOCK, born 10 APRIL 1887, a vertical driller on aircraft work, married to Ella. And Baldock supplies the warning as well as the problem — a search of the 1921 census returns a SECOND unrelated Castle household at 5 Cemetery Road with nine children and its own Frederick, born about April 1888. Three men of one name in one small town within two years. Our Frederick is in no English census after 1901 and in no death index at all, which is now the open question rather than a solved one.
Sixty-three and seventy-two Pix Road Documented
did The 1939 Register was searched for Leonard's aunt Louisa Shenton and returned his grandmother. Schedule 305, 63 PIX ROAD, LETCHWORTH: EMMA HULL, born 7 February 1859, widowed, "Incapacitated - Blind"; Louisa Shenton, born 29 November 1882; William Shenton, incapacitated. Emma Hull is EMMA HAWKINS — born at Aldenham, widowed at thirty-one in 1890, remarried to John Hull and widowed again. Henry Robert and Martha Maud had raised their six at 72 PIX ROAD, the same street. The register was taken on 29 September 1939, seven weeks after the last ship sailed, and it finds an eighty-year-old blind woman and the daughter keeping her still in the road the family had left. The record says who was where and nothing about what anybody felt. But it is no longer true that the family simply left.
Learned that the paywall was not one Documented
did The estate references looked to be behind a subscription. They are not: every South African estate over the threshold must be advertised as a Form J193 notice in the Government Gazette, which is free and full-text to 2026. The only trick is the name order — modern notices read surname first. Four estates in twenty minutes, all four women born Wear, with exact birth dates, identity numbers and addresses.
Oskar left Theresienstadt on a card that names the transport at both ends Documented
did His four Arolsen person records were on the work list as “all openable and none read”. Read, three of them lock together. The Gestapo list of the “4th Large Transport of the Elderly”, 17 March 1943, page 161 line 54: Leibholz, Oskar Israel, 13.5.87, Schubin, W.50 Marburgerstr. 13, number 014577. An AJDC index card repeats it as “49/30999 II — 4.gr.Alterstrsp. 17.3.43” and adds his birthplace in full, SCHUBIN/BROMBERG, and his wife, GERTRUD geb. LEIBKE, 13.7.94 Berlin. And in Terezín a Czech clerk wrote out a card of the central transport file: DEPORTACE NA VÝCHOD, Číslo Er 606, dne 16.X.1944, with the original transport number 11025-I/90 — which is the Berlin transport’s own designation. The two ends of the journey name each other. The card does not say where he was sent; it says east.
WITHDRAWN: the document that was not a deportation form Withdrawn
did On 17 September this archive found Jacob, Auguste and Kurt sharing one Arolsen DocID in a series called Deportations and Transports, reasoned that it was “a deportation form that records a man’s household and his next of kin”, and called that reading “evidence instead of the charitable guess”. David ticked the consent and the document was opened. IT IS A BERLIN RESIDENTS’ REGISTRATION CARD, and the DocID covers two separate images — the family’s card and a sheet of other households. KURT’S NAME IS ON THE SECOND SHEET, not on his family’s card. The series name was read as the document. What the card does give is worth the correction: AUGUSTE’S BIRTHPLACE IN HER OWN HOUSEHOLD’S HAND — SAUERBAUM, the East Prussian village her mother Johanna Thimm came from, which until today rested on a single census card.
Tested a restriction instead of assuming it Documented
did T/D 755.334 for Heinz Leibholz was listed as “may be restricted like Oskar’s — BUT THAT HAS NOT BEEN TESTED”. Tested: eleven documents, creation date 1959–1988, and the page serves no DocID, no image and no document list. The consent declaration does not reach the T/D repository at all. The same sweep listed every Leibholz topic in the archive — forty-three of them — and the list is itself a finding: ten consecutive case numbers, 90.465 to 90.474, are one enquiry about one household, and they match the Levetzowstraße 13 family from the 1939 census to the day. One of those ten is “LEIBHOLZ, KURT”, and it is Alfred’s son born in 1927, not Derrick’s father. Written down so nobody finds it later and draws the obvious wrong conclusion.
An emigration card with the word emigrating crossed out Documented
did Row 197 was written to ask whether Oskar’s wife was on the transport that took him to Theresienstadt, because she is not on the page that carries him. It was answered the same afternoon and the answer is no. The Kultusvereinigung Berlin — the community organisation the state had turned into an instrument of its own administration — kept Zählkarten, counting cards, and hers survives: “Leibholz geb. Leibke · Gertrud · 13.7.94, Berlin · verh.”, household address Marburgerstr. 13, written on 8 March 1943. The card is headed “Zählkarte für AUSWANDERNDE Personen” with AUSWANDERNDE struck through in ink, and in the column headed ZIELLAND — country of destination — somebody has written VERSTORBEN 27/2.43. She died eighteen days before the transport left. Oskar’s line on the list carries no Ehepaar bracket because by then there was no pair. The date itself was already published here from the Memorial Book, which is a compilation; it is now on a card filled in nine days later by the office that knew her.
Two Leibholz families, and they are finally separable by place Documented
did The 43 tracing topics threw up “T/D 161.094 for MEYER, GERTRUD born 09.03.1899”, indexed under the maiden name LEIBHOLZ. A Leibholz daughter born in 1899 would have fitted the Schubin family, which has no sister in it at all, and a row was written to chase her. Her second person record answers it in one word: PLACE OF BIRTH GLOWITZ — Główczyce, KREIS STOLP, in Pomerania. That is Willi’s district, not Jacob’s. She is a Pomeranian Leibholz and she is not anybody’s sister here. THE NEGATIVE IS WORTH MORE THAN THE HYPOTHESIS WAS: for weeks the two Leibholz families could only be told apart by argument, and now they can be told apart by place — SCHUBIN IN POSEN against STOLP IN POMMERN, with two documented birthplaces on each side. Her own fate is on the record: Wave 48, the 34th Osttransport to Auschwitz, 4 March 1943 — thirteen days before Oskar went to Theresienstadt, on a different kind of list altogether.
Searched the names database by who WROTE the page, and found somebody alive Documented
did A Page of Testimony names the person who filled it in, and so does a Relico questionnaire, so searching the SUBMITTER field asks a different question: not who died but WHO SURVIVED TO WRITE IT DOWN. Submitter last name LEIBHOLZ returns three records and one woman — SOPHIE LEIBHOLZ, filling in forms for the World Jewish Congress relief committee IN SHANGHAI IN 1946. She was looking for her brother SALLY BINDER, a tailor of Motzstrasse 13 in Neukölln, deported to Auschwitz on 1 March 1943; for his wife Mathilde; and for her brother-in-law HEINRICH LEIBHOLZ, born Berlin 16 October 1890, a tax agent of Hornstrasse 19, wife Agnes. Heinrich is this archive’s control case — the man used to prove that vanishing from the Berlin directory after 1942 does not mean death, because he died in 1948. He now has a birth date, a wife, a street and a trade, and somebody was looking for him from Shanghai two years before he died, not knowing he had lived. Nothing ties any of them to Kurt’s family, and it is the first Leibholz survivor this archive has a name for.
WITHDRAWN: the niece who was a niece by marriage Withdrawn
did The work list has argued for three days that IRMGARD SHTERN was the best lead in the archive on the oldest open question — whether Georg Leibholz’s line touches Sally Leibholz and Friederike Schmul of Schubin. The argument was clean: she called herself Georg’s NIECE, so her parent was Georg’s sibling, so her other Pages of Testimony would name GEORG’S PARENTS. She filed seventeen of them, and three of them end it. Lidia Laibholz b.1884 gives her father as ELIUS and her mother as MINA; Klara Lewinzon b.1906 gives YULIUS and MINA; Ema Mendelzon b.1892 gives YULIUS and MEINE. Elius and Yulius are one name through the Hebrew and so are Mina and Meine. THEY ARE THREE SISTERS, AND IRMGARD IS NIECE TO ALL THREE — so Georg is the uncle BY MARRIAGE, her blood is Lydia’s family, and her Pages could never have given Georg’s parents. Georg’s own Page has no father and no mother on it and gives his birthplace as the single word Germany. WHAT THE SEVENTEEN DID GIVE: Lydia’s parents and two sisters, which nobody had — and GIDEON SHTRUZIK, BORN 1940, marital status “child”, of Berlin, murdered, father Harry. Harry Struzik was Elli Leibholz’s husband, so Gideon is almost certainly Elli’s son and Georg’s grandson, although his mother is not named on the page.
Followed a survivor to Shanghai and found her husband’s death certificate in a list Documented
did This morning’s find was a name and a city: SOPHIE LEIBHOLZ, filling in relief questionnaires in Shanghai in 1946, the first Leibholz survivor this archive had a name for. Who she was married to was unknown. Mapping the Lives has an EMIGRATION PLACE field nobody here had used — surname LEIBHOLZ, emigration place SHANGHAI, two results, and one of them is ALBERT LEIBHOLZ born 17 May 1886, who is the “Albert, Antiquitäten, Halensee” of the 1938 Berlin directory. Arolsen then has him twice. A CENTRAL LOCATION INDEX card, L 337909: “Leibholz · Albert · died 8.18.44 · 5.17.86 · Berlin · dealer · last known address SHANGHAI” — and under OTHER MEMBERS OF THE FAMILY ABROAD, “SOPHIE · W · 6.19.87 · Berlin”. W is wife. The card also carries the crossing — “arr. 6.27.39 from Berlin” — and a third person looking for them, Curt Sterngast of Rehavia, Jerusalem. And the H.I.A.S. list of Jewish nationals who died and were buried in Shanghai 1939–1948, page 24: “LEIBHOLZ, ALBERT · 5/17/86 · 8/18/44 · paralisis progressitis”. A Berlin antiques dealer, out on 27 June 1939 with ten weeks to spare, dead in Shanghai in 1944, his widow still there in December 1945. It also separates the two Sophie Leibholzes for good — née Binder, born Berlin 19 June 1887, alive; née Rothenberg, born Schlochau 9 January 1887, murdered. None of it touches Kurt’s family, and it is the first Leibholz household this archive has followed all the way out.
Searched the census by address instead of by name, and found our own Kreis in Charlottenburg Documented
did Every earlier pass at the 1939 census database started from a surname in the Berlin directory. Starting from a STREET AND HOUSE NUMBER returns the whole flat, and it produced a correction and a household. The correction is this archive’s own: Levetzowstraße 13 holds SIX Leibholz, not five — WALTER, born 20.09.1922, was simply missed, and he is the T/D case 90.473 that looked like a name the census lacked. The household is Danckelmannstraße 44, Charlottenburg, which this archive has listed for a week as a Leibholz Stolperstein “belonging to another family”. HANS LEIBHOLZ, born 25 January 1912 AT PREUSSISCH FRIEDLAND, KREIS SCHLOCHAU, deported from Berlin on 12 January 1943 and dead at Auschwitz on 22 February; his mother ELSE NÉE ROTHENBERG, born at SCHLOCHAU; his sister LISELOTT, born at Preussisch Friedland; and a second Rothenberg woman of 1884, LAURA, in the same flat. KREIS SCHLOCHAU IS HAMMERSTEIN’S DISTRICT, and Hammerstein is where SALLY LEIBHOLZ — Jacob’s father, Derrick’s great-grandfather — was born on 7 July 1847. Every other Berlin household placed so far is Schubin, Stolp or Berlin; this is the first from our own Kreis. A district is not a family and this archive has been burnt twice by good fits, so it is not claimed as kinship. What it does is name the place to write to, and nobody here has ever asked Poland for a Hammerstein record.
Proposed writing to Poland for a letter that was answered six days ago Withdrawn
did The Charlottenburg household led to Kreis Schlochau, and Kreis Schlochau is where Sally Leibholz was born, so a row was written saying nobody here had ever asked Poland for a Hammerstein record and it was time to. SOMEBODY HAD. A letter went to Archiwum Państwowe w Koszalinie on 14 SEPTEMBER asking for exactly those two records — Sally’s birth of 1847 and Georg’s of 1884 — and the synagogue books. It was answered on 16 SEPTEMBER, reference O-S.6342.406.2026, signed by the Head of the Słupsk Branch: no Standesamt Hammerstein registers for 1874 to 1899, no synagogue-community books for Czarne at all, and no information about where either went. It was acknowledged on the 17th. What the morning’s work adds is one line of corroboration: the public finding aid holds fond 27/364/0, Standesamt Hammerstein [Czarne], at Koszalin’s Słupsk branch, 127 digitised units, and its own date facet reads 1900+ : 128, 1800–1899 : 0. The branch’s written negative is confirmed by its own catalogue. That is worth a line and it is not worth a row. SIX TIMES IN ONE DAY NOW, and the cause has not changed: a row or a paragraph written off the back of a fresh find without first reading the file that already answers it. The rule that follows is one command — grep letters.json and requests/ for the place name before proposing to write to anybody.
The surname that cannot be searched in a newspaper Inferred
did Priority three of the Findmypast subscription is a newspaper sweep of the Hitchin and Letchworth titles for Wear and Castle notices. It was tried, and the instrument will not do it. WEAR IS ONE OF THE COMMONEST WORDS IN BRITISH NEWSPAPER ADVERTISING, and unlike the census sets the newspaper name box is an OCR string match rather than an index of transcribed names. Wear with the keyword Letchworth returns 9,442 articles headed “MEN’S WEAR AT PUGH’S, Letchworth”; “Martha Maud Wear” returns 285 headed by a knitting joke and a Yorkshire sock advertisement. Two further properties of the tool go down beside the rule that its marriage index silently returns zero when year-filtered: the EXACT-KEYWORD switch does not enforce a phrase, so “Pix Road” with it ticked returns 85,719 articles — and the results URL ignores the county and date parameters, so this one set cannot be swept by constructing URLs the way every other set can. A query with all five filters returned twenty million, more than the unfiltered one. The sweep is re-scoped rather than abandoned: it has to be driven by strings that are not ordinary English — Shenton, Giddins, Wallis with Baldock — and by the ship names, Windsor Castle in March 1937 and Dunvegan Castle in August 1939, looking for a departure notice. The surname on its own is not worth trying again.
The broken year filter is not only the marriages Withdrawn
did On 16 September this archive proved that Findmypast’s marriage index SILENTLY RETURNS ZERO when year-filtered, using a positive control — James Shardalow Mort, married at Yarmouth in 1885, who disappears when the search is narrowed to 1885 give or take two. The rule written then named the marriage index. Looking for the six Hitchin Wear children in HERTFORDSHIRE BAPTISMS found the same defect in a second set. Lastname Smith, no year: 15,122 baptisms. The same search narrowed to 1880 give or take five: ZERO. Fifteen thousand Hertfordshire Smiths and not one of them baptised between 1875 and 1885. The 1920 window also returns zero, and without the 1880 control that would have been read as “the set does not reach the twentieth century” and written down as a coverage bound. THE RULE IS RESTATED WIDER: never year-filter a Findmypast search in any set, and if one is run, run a positive control in the same set in the same breath. The failure mode is the dangerous one — a clean, plausible zero instead of an error. The children themselves are not there: eleven Wear baptisms in the whole county set, the latest 1891, the only Hitchin one 1786 — and a whole-category search for Cynthia Y E Wear, born Hitchin 1931, returns her GRO birth entry and no parish baptism anywhere.
A user tree’s date, confirmed to the day by a government index Documented
did Priority seven of the Findmypast subscription is probate, and it paid. GEORGE WILLIAM CASTLE the younger — Martha Maud’s elder brother — has been carried here on a date taken from a match holder’s tree and nothing else: died 23 August 1939. The England & Wales Government Probate Death Index has him: DEATH DATE 23 AUG 1939, RESIDENCE TOWN LETCHWORTH, probate granted at the London registry on 2 November 1939, linked person CYRIL GEORGE CASTLE. Two things come with it. He was in LETCHWORTH — the Wears’ town, and the youngest Castle brother was at 9 Boscombe Place, Letchworth in the 1921 census; the Castles had moved from Baldock and the Wears went to the same place. And the dates sit thirteen days apart: Martha Maud sailed from London on the Dunvegan Castle on 10 August 1939, and her brother died at Letchworth on the 23rd. She had been gone a fortnight. Both halves are documented — BT27 on one side, the probate index on the other. CYRIL GEORGE CASTLE is a new name and is not the ninth child, Ernest Albert Cyril Castle of Baldock; likeliest a son of the deceased, and not asserted. The original calendar page would give the street, the occupation, the executors and the estate value, and it could not be read: Findmypast’s image gateway is session-locked and returns 403 to anything outside the signed-in browser.
Line 1039, and he was not taken from the family flat Documented
did This archive has had Jacob Leibholz’s deportation date since the Gedenkbuch was checked on day two. What it never had was the paperwork, and the paperwork moves him to a different street. Wave 44, the 30th Osttransport to Auschwitz of 26 February 1943, page 153, line 1039: “Leibholz · Jakob Israel · 8.8.84 · Schubin · Beruf OHNE · verh. ja · Alter 58 · arbeitsfähig ja · Wohnung TREUCHTLINGERSTR. 4 · Kennkarte A492552 · Kennzeichen 31822.” An AJDC index card repeats the address and adds one word: b/SPIER — bei Spier, care of a family called Spier. The same page of the list carries “49 b/Grindel”, “b/Gottfeld” and “b/Neumann” against other names, so the convention is not a guess: he was lodging in somebody else’s flat, as many of the 1,100 were. THE 1939 CENSUS HAS HIM AT MERANER STRASSE 2 WITH AUGUSTE AND SIEGFRIED. Between May 1939 and February 1943 he was moved out of his own home, and nothing read here says when or by whom; the word Judenhaus is on neither document and is not used. One column is worth reading twice. In the 1938 directory he is Jaques, Kaufmann, with a clothing firm at Krausenstraße 59–60. On the list of February 1943 the occupation column says OHNE. And Samuel, re-searched with the consent now ticked, is still not in the Arolsen index at all — the second test of an absence this archive first recorded on 17 September.
Two letters on a card, and a house already full Documented
did The AJDC card for Jacob Leibholz ends “Treuchtlingerstr. 4 b/Spier”. Searching the 1939 census by street and house number instead of by surname turns the two letters into three people: SPIER, MAX, born 12 April 1864, died 26 March 1941; SPIER, HANS, born 2 December 1894; SPIER, ALICE née WOLLSTEIN, born 3 November 1899, the last two both dead before 8 May 1945. And it says what kind of building it was. Treuchtlinger Straße 4 held SEVENTEEN persecuted people from ELEVEN family names on census night — Brück, Cohn, Eisenberg, Eisenberger, Kaphan, Kochanowska, Löwenheim, Podbielski, Rosenstock, Sundheimer, Spier. Jacob was moved into a house that was already full, three and a half years before the transport. THE SAME SEARCH ON MERANER STRASSE 2 SOLVED SOMETHING ELSE. Fourteen people at the family’s own address, and one of them is ANNA PUFF, BORN 1 JUNE 1895 — who is one of the eleven names indexed on DocID 11214358, the registration card whose second image nobody here could account for. So that second image is the HOUSE LIST FOR MERANER STRASSE 2, which is exactly what the stamp on the first image says: “Lt. Hausliste 1945 n. erfaßt”. One exact match out of seven unplaced names, and the other six would not be in a database of persecuted people if they were not Jewish — a strong reading, not a proof. It also gives Kurt’s name on that sheet a reason at last: an absent son written against his parents’ flat. That is the THIRD account this archive has given of why he is on that DocID. It is offered as the best one and it is flagged as the third.
Samuel had a wife, and the census had her all along Documented
did The address search was then run across every Leibholz address the Berlin directory had given, and the best of them is Kommandantenstraße 29 — Samuel the hat wholesaler’s. It returns two people on consecutive census cards: SAMUEL LEIBHOLZ, born 28.06.1877 in Szubin, “Race” JJJJ, deported to Auschwitz, died 10 April 1943, card VZ153419 — and FRIDA LEIBBHOLZ NÉE SPREE, born 12 May 1888 in Berlin, “Race” NNNN, no date of death, card VZ153420. SO BOTH BROTHERS WERE IN MIXED MARRIAGES. Jacob married Auguste née Luwinski, NNNN, persecuted “collectively”; Samuel married Frida née Spree, NNNN. Two men from Schubin, two non-Jewish German wives, and both men deported anyway. Neither wife has a death date. A Mischehe delayed the two brothers and it did not save them. Samuel is not in the Arolsen index at all, searched twice, the second time with the consent ticked — so the gap there is Arolsen’s. The clerk spelt her surname Leibbholz, with a doubled B. The same sweep separated the two Wilhelms at last: the Weinstraße 6 Vertreter is born 30.12.1887 and Yad Vashem’s deported Wilhelm 28.12.1890, so they are not one man. AND IT PRODUCED A COINCIDENCE THAT WAS CHECKED AND DID NOT SURVIVE. At Weinstraße 6, in the same building as a Leibholz household, lives HENRIETTE ROSENFELD NÉE SCHMUL — and Schmul is Friederike’s surname, Jacob’s mother, married to Sally Leibholz at Szubin in 1874. Her record was opened inside a minute: born at SWINEMÜNDE in Pomerania, not Schubin, not Posen. The coincidence is real and the connection is not evidenced, and it is written down so the next reader does not spend an hour on it.
Opened the half of the address books nobody here had opened Documented
did Every earlier pass at the Berliner Adressbücher read TEIL I, householders by name. There is a TEIL IV that lists the same people BY STREET, house by house, with the owner first and the tenants beneath — and it answers the one question Teil I cannot: who else lived in the building. The 1939 volume, printed page 1629, gives Treuchtlinger Straße 4 entire: “4 E EISENBERG, G., vw. Handelsgerichtsrat · Bayer, Oberstleutnant · Eberhardt, Kaufmann · Jakob, Maurer · Mannowsky · Petras · Rosenstock · SPIER, H., Handelsvertreter · — M., Vertreter · Sundheimer · Wendorf”. The E is Eigentümer: the house belonged to GERTRUD EISENBERG NÉE HAAKE, born 9 June 1879, a commercial court councillor’s widow, who also lived in it. Spier H. is HANS, born 2 December 1894; the dash line under him is MAX, born 12 April 1864, dead by March 1941 — the same dash convention that cost this archive four Leibholz households in a day last week. AND THE INSTRUMENT IS THE BIGGER FIND. digital.zlb.de runs a Goobi viewer with a Solr query API: a POST to /viewer/api/v1/index/query/ returns the IMAGE NUMBER of every page in a volume containing a word, and /pages/{n}/text/ returns that page’s OCR. Three limits were measured and written down. Fraktur OCR mangles words — “Treuchtlinger” returns nothing in Teil IV and the fuzzy form “Treuchtlinger~1” returns exactly the two right pages. The Teil IV OCR is column-interleaved and reads as nonsense, so it finds a page and cannot read one. And the table of contents gives PRINTED page numbers, not image numbers: printed 1629 is image 5933, an offset of about 4,304 that differs per volume. The whole lookup took six requests.
Went looking for corroboration, got one match and six misses, and kept the reading anyway Inferred
did Teil IV was then turned on the family’s own address. Meraner Straße 2, 1939, printed page 1598: the owner is SALOMONSKI, M., living away at Tauentzienstraße in W 50, and the householders under him are Binder, Briliant, Dübe, Dunje, Gadischle & Kaulmann, Hesler, Helm, Kamnitzer, v. Koch, *** LEIBHOLZ, J., KFM. ***, Neumann, Pehser, Putzar-Ehrenberg, Schade twice, Wesse. Jacob is listed at his own address as a KAUFMANN — the trade the transport list of 1943 would record as OHNE. AND IT WAS A TEST, NOT A VICTORY LAP. This morning this archive decided that the unexplained second image of DocID 11214358 is the HOUSE LIST for Meraner Straße 2, on the strength of Anna Puff appearing both on it and in the 1939 census at that address, plus the stamp “Lt. Hausliste 1945”. The directory was checked against the sheet’s other six names — Schmidt, Heinicke, Herzog, Biere, Koch, Nicksch — AND NOT ONE OF THEM IS A HOUSEHOLDER THERE IN 1939. The reading survives for two ordinary reasons and they are written down as an argument rather than a fact: the directory lists householders only, so lodgers never appear in it — Anna Puff herself does not — and a list stamped 1945 in a city where the bombed-out were rehoused constantly should not match a 1939 householder list. It is now a reading with a test behind it instead of a reading with one match behind it.
Three Luwinskis in post-war Berlin, with trades, addresses and a telephone number Documented
did The Solr API found this morning searches forty volumes in one request, so the POST-WAR Berlin books were swept by surname for the first time — 27 Namenteil volumes 1949–1970 and 13 telephone books 1969–1980, all set in Antiqua, so the text layer reads cleanly and no page images are needed. LUWINSKI returns ten pages in eight volumes and three men, every one already in the family tree and every one dead. MANFRED, “Schlosser” and then MODELLSCHLOSSER, Tegel, Egellsstraße 16, from 1957 to 1970 — and Modellschlosser is MODEL FITTER, the exact trade on the 1965 ICEM emigration roll that flew him to Melbourne care of a Mr H. Marquardt, and the tree holds Helga Erna Marquardt née Luwinski. Three sources, one man. JOACHIM, OFENSETZER — a stove-fitter — of Ramlerstraße 15, Wedding, in 1967 and 1970: a trade and an address the tree does not have for him at all. And KURT, a post-office worker at Stephanstraße 7 in 1964 and Peter-Vischer-Straße 25 in 1967, who by the 1970–71 telephone book is “Luwinski Kurt · 4 12 25 12 · Beamter · 52 Waldstr. 60”. THAT KURT IS NOT DERRICK’S FATHER, who left Germany in 1933 and died in Johannesburg in 1968 and cannot be in a 1970 Berlin telephone book. It is the tree’s other Kurt, Leo’s son, recorded as dying in 1995 with no birth date, no place and no occupation. A surname, a city and an era: a correspondence, not an identification, and written down as one. OSKAR LUWINSKI, the master tailor of Steglitzer Straße in 1933, is in none of the forty volumes.
A candidate in one year, and in no other Inferred
did The same sweep did the other half of row 179. “Fricke Auguste” across all forty post-war volumes returns THREE pages and they are TWO WOMEN. The 1957 Namenteil has “Fricke Auguste Rentn Chlb 1 Kais-Friedr-Str 79”, a pensioner in Charlottenburg — the candidate for Kurt’s mother under her last married name. The 1963 and 1970 street sections have “Fricke Auguste Putzfrau” in the same building both years, among Fröhlich, Kaldenbach, Luckow and Stinka. A working cleaner in 1963 is not the seventy-year-old pensioner of 1957. AND THE PENSIONER APPEARS IN NO OTHER VOLUME OF THE RUN, AND IN NO TELEPHONE BOOK. Four Fricke blocks were opened by hand as well — 1952, 1954, 1963, 1970 — and none holds an Auguste under her own name or on the dash. If she is Lizzy, she was in Berlin in 1957 and gone by 1959 — and she was still in Berlin TWO YEARS AFTER HER MOTHER DIED AT ZEUTERN in October 1955. She did not go to Baden to nurse her mother. She went afterwards, and died at Kraichtal, the same town, in 1975. The identification is still unproved and nothing here changes that: one line, no age, no birth date, no maiden name. What the sweep buys is the difference between “there is a candidate in 1957” and “there is a candidate in 1957 and in no other year”.
Asked the opposite question of the Berlin directories, and Heinrich’s widow answered it Documented
did Every pass at these books has tracked the surname DISAPPEARING — seventeen Leibholz householders in 1938, six in 1942, every name carrying Israel gone. Nobody here had asked WHO WAS STILL THERE AFTERWARDS. The post-war run is forty volumes to 1980 and set in Antiqua, so it searches outright. LEIBHOLZ returns nineteen pages in seventeen volumes and six names across thirty years. AGNES IS THE ONE THAT CLOSES A LOOP. “Leibholz Agnes, Rentn., SW 61 Mehringdamm 109” in 1957 and 1959, and still at Mehringdamm 109 in the 1975–76 telephone book — nineteen years at one address, with a line of her own by the end. Heinrich Leibholz is this archive’s control case, the man who proved that vanishing from the directory after 1942 does not mean death, because he died on 9 August 1948; the census gives his wife as AGNES NÉE KERGER, born 1889. And SOPHIE LEIBHOLZ, filling in a relief questionnaire in SHANGHAI IN 1946, named him with “wife Agnes” and asked where he was. He had survived; he died two years later; his widow lived on at Mehringdamm for another quarter-century. The others are Ella the nurse at Dominicusstraße and then Willmanndamm, Charlotte at Hasenheide, Gerhard a doctor of economics in Steglitz — not the constitutional lawyer of the same name, who died at Göttingen — Felix at Wiesbadener Straße 57 and then Hilde in the same flat, and Erna alone in Spandau in 1980. NOT ONE IS DOCUMENTED AS BELONGING TO THE SCHUBIN FAMILY, and that is the point: the survivors of this surname in Berlin are the OTHER Leibholz families. Kurt’s people were in Johannesburg, in Potsdam, and in a grave at Auschwitz. One absence was looked for on purpose — Frida Leibbholz née Spree, Samuel’s non-Jewish widow with no death date, is in none of the forty volumes. Telephone numbers were deliberately not transcribed.
The Geneteka zero was a one-character bug, and the control had been right all along Withdrawn
did On 20 September this archive searched Geneteka, the Polish Genealogical Society’s index, for Leibholz marriages in kujawsko-pomorskie and got nothing — then ran KOWALSKI as a control, got nothing again, and correctly refused to publish the negative. The cause is now found and it is one character. GENETEKA’S PROVINCE CODES ARE POSITIONAL: 01ds dolnośląskie, 02kp KUJAWSKO-POMORSKIE, 03lb lubelskie, 04ls LUBUSKIE. The query used w=04kp, which does not exist, so the select fell back silently to dolnośląskie and every search ran against the wrong province. There is a second trap in the same site: the results table is filled by AJAX after the page loads, so a plain fetch of the search URL returns the shell with an empty table, which reads exactly like a zero. WITH w=02kp THE CONTROL PASSES — Kowalski returns thousands from 1621 on. And the real answers arrive. LEIBHOLZ: zero in kujawsko-pomorskie and zero in wielkopolskie, births and marriages alike. SCHMUL: eighteen marriages in wielkopolskie, fuzzy across Szmul, Szmił, Schmul, Szmiel and Szmielewski — Lwówek, Przedecz, Pyzdry, Turek, Dąbie, Krotoszyn, Gniezno, Pniewy, Koło — and NOT ONE AT SZUBIN. AND THERE IS A STRUCTURAL REASON NONE COULD BE. Szubin is in Geneteka — parish 13066, indexed from 1715 — but it is the CATHOLIC parish. Geneteka marks Jewish registers “(mojż.)”, and kujawsko-pomorskie contains NO (mojż.) parish at all, not one across 358 entries. So Geneteka cannot answer the Szubin question and never could. Sally Leibholz’s marriage to Friederike Schmul on 24 November 1874 needs the letter to Bydgoszcz, which is drafted and unsent — and that is now a tested statement instead of an assumption.
The register this archive has been writing letters about is online Documented
did Geneteka having been proved structurally incapable of holding a Jewish Szubin record, the question became where the 1874 register actually is. It is in three places and one of them needs no letter at all. ARCHIWUM PAŃSTWOWE W BYDGOSZCZY holds the Szubin civil registers in three fonds — 6/1792/0 SZUBIN MIASTO, 1874–1935, with marriages from 1874 and, crucially, SKOROWIDZE, alphabetical indexes, FOR 1874–1878; 6/1795/0 Wieś I; 6/1796/0 Wieś II. AND FAMILYSEARCH FILMED THEM FROM BYDGOSZCZ. Catalogue koha:402152, Standesamt I, film note “Geburten, Heiraten, Tote 1874–1876”, DIGITAL FILM 007998644, 1,121 images. Catalogue koha:402191, Standesamt II, 1874–1875, DIGITAL FILM 008016003, 1,160 images. Both open in the viewer with no restriction and no affiliate-library wall. Neither is in the full-text index — the service’s own groupNumber lookup returns an empty id list for both — so they cannot be searched by surname and have to be browsed. That is tractable rather than hopeless: Prussian civil registration began on 1 OCTOBER 1874, so the 1874 volumes cover three months, the marriage book should hold a few dozen entries, and the skorowidz should be a single page. A marriage register of 1874 names THE PARENTS OF BOTH SPOUSES. That is the generation above Sally Leibholz and Friederike Schmul, and nobody in this family has ever had it. Request 30 to Bydgoszcz has been updated to cite the three fond references, so a busy office does not have to find them — and if the browse succeeds the letter is only needed for a certified copy, if at all.
WITHDRAWN: an image count is not an image Withdrawn
did An hour ago this archive published that the Szubin civil registers of 1874 are filmed and that “both are online and neither is restricted — no permission wall, no affiliate-library restriction, no login beyond the ordinary free account”. THAT IS WRONG AND IT IS WITHDRAWN. The claim was made from the viewer reporting a page count — 1,121 images for film 007998644, 1,160 for 008016003 — WITHOUT LOOKING AT THE PAGE, because the browser pane was hidden and only the text layer was being read. With the pane displayed, EVERY THUMBNAIL IN THE FILMSTRIP CARRIES A PADLOCK, and opening one returns FamilySearch’s own notice: “IMAGE RESTRICTED. Image access is typically determined by local laws or the custodian who has the original document. To view these images do one of the following: ACCESS THE SITE AT A FAMILYSEARCH CENTER.” THE LESSON IS GENERAL AND GOES BESIDE THE OTHERS. FamilySearch serves a film’s LENGTH to everybody, restricted or not. The padlocks exist only in the rendered page. A film’s availability cannot be read from its metadata, and reading a count as an answer is the same mistake as reading a dead hostname as a dead service. WHAT SURVIVES IS MOST OF IT: the registers exist, they are at Archiwum Państwowe w Bydgoszczy in three named fonds, the town fond has SKOROWIDZE — alphabetical indexes — for 1874–1878, and the films are 007998644 and 008016003. Only the viewing is blocked, and by the CUSTODIAN rather than by FamilySearch. AND IT LEAVES A FREE ROUTE THIS ARCHIVE CANNOT TAKE ITSELF. The restriction lifts inside a FamilySearch Centre or affiliate library, of which there are several in Brisbane, and access there costs nothing. Film 007998644, the 1874 skorowidz, the marriage of Sally Leibholz and Friederike Schmul on 24 November 1874 — an afternoon at a terminal. Request 30 to Bydgoszcz is therefore necessary after all, and it is a better letter for citing the three fond references.
Checked the Polish side too, and the register is unreachable from any desk Inferred
did Having got the FamilySearch answer wrong once, the rest was checked properly. Film 008016003, Standesamt II, is padlocked as well — so the restriction is on the Szubin material, not on one reel. And the Polish State Archives’ own portal has no scans of it either: szukajwarchiwach returns the three USC Szubin fonds AT FOND LEVEL ONLY, with no unit rows, no inventory and no scan counts, across four differently-worded searches. THE CONTROL IS WHAT MAKES THAT A FINDING RATHER THAN A SHRUG. The same search box returns 127 UNIT-LEVEL ROWS, each with a “Number of scans” line, for “Standesamt Hammerstein” at Koszalin’s Słupsk branch. The portal does show units and does show scans when they exist. It has neither for Szubin. So the position is now fully mapped: THE REGISTERS EXIST AND NOBODY CAN SEE THEM FROM A DESK. Not digitised in Poland; filmed by the Mormons and locked to a FamilySearch Centre. Two routes remain and both need a person in a room — a free terminal at a FamilySearch Centre in Brisbane, or the letter to Bydgoszcz, drafted and unsent.
The town’s paper stops twenty years before the family left Inferred
did The newspaper sweep was re-run the way the re-scoping said — drive the FILTER PANEL, and search strings that are not ordinary English. First, why the panel works now: THE BROWSER PANE HAD BEEN HIDDEN, and a click cannot reach a pane that is not compositing frames, so every earlier attempt to open a facet failed with the panel apparently inert. Displaying it fixed the problem instantly — and the same hidden pane is what produced the Szubin padlock error an hour later. Twice in one day, one cause. THE SHIP WAS THE BEST IDEA AND IT CAME BACK EMPTY. Keyword “Dunvegan Castle”, 47,783 articles, faceted to Hertfordshire: eleven. 1890, 1897 twice, 1898, 1900, 1901 twice, 1902 — then 1982 twice and 1986. Mrs Louis Botha landing at Southampton, and coach tours to Skye. NOT ONE IN 1939. AND THEN THE COVERAGE WAS CHECKED, WHICH SHOULD HAVE COME FIRST. LETCHWORTH HAS EXACTLY ONE DIGITISED TITLE: Citizen (Letchworth), 1906–1917, 587 issues. The Wears lived at 72 Pix Road and left in 1937–39. The town’s only digitised paper stops twenty years before they went. Hitchin has exactly one, the Hertfordshire Express, 1859~1956 — and 1,403 issues across ninety-seven years is fourteen a year for a weekly, a quarter-sampled run at best. So the priority closes rather than being re-scoped again. A departure notice for this family is not findable here, and the reason is neither the search nor the surname: the paper was never digitised for the years that mattered. That is a coverage bound and it is the end of the route.
Sauerbaum is Zerbuń, and one spelling is not a search Documented
did Two of the twenty were done at once, and the second was rescued by David. FIRST: SAUERBAUM, KREIS RÖSSEL, IS ZERBUŃ — today in Gmina Jeziorany, powiat olsztyński. Its parish and civil district were BÖSSAU: GROß BÖSSAU is BIESOWO, and the Amtsbezirk Bössau was formed on 9 July 1874 with its seat there, so the STANDESAMT BÖSSAU is where Auguste’s birth of 28 August 1893 must be registered. Every Polish archive and portal is organised on the current name, and this archive has spent weeks searching a German one. SECOND: Geneteka’s East Prussian province, never searched until today. The first pass ran ONE spelling, found one woman in one village, and wrote “nothing else in the entire province”. David then said to try every spelling. THE INSTRUCTION CHANGED THE RESULT. LOWINSKI returns THREE BIRTHS AT STANDESAMT RÖSSEL — Franz 1876, Albert 1881, Bernhard 1884 — and Kreis Rössel is the district Sauerbaum is in, where Johanna Thimm was born in 1871 and Auguste in 1893. LUWENSKI returns the best lead of the day: entry ELEVEN of the new Allenstein civil registry, 1874 — MARTHA LUWENSKI, FATHER JOSEPH, MOTHER ELISABETH BILITEWSKI. Joachim Luwinski was born at Allenstein in 1865, and the family tree already holds a Joseph Luwinski with no dates, inside the hundred-name Polish block this archive has always called a surname index rather than a descent. IT IS THE FIRST TIME THAT BLOCK HAS TOUCHED A RECORD. The same registry spells one woman LOWINSKI in four years and LUWINSKI in two, which is the whole argument in one line. Neither find is claimed as kinship. Both name exactly which entries Olsztyń should be asked for. AND THE RULE GOES DOWN: ONE SPELLING IS NOT A SEARCH. Every surname sweep from here runs the archive’s whole documented list of variants, one at a time, exact.
The other half of the emigration roll, in Canberra Documented
did The National Archives of Australia is free, name-indexed and had never been searched. It holds NAA: A2559, 1965/86/85-88 — the same roll number and the same four line numbers as the Arolsen sheet, BGO 65/135 lines 85 to 88, filed on opposite sides of the world by two archives that do not know of one another. Nine folios, paper, Canberra, not digitised.
Published that they flew, and the Australian title says they sailed Withdrawn
did This archive announced this morning that Manfred Luwinski flew from Berlin to Melbourne in 1965. The NAA catalogue entry for the same document reads “arrived per ship FLAVIA ex Bremerhaven/Germany 17 September 1965”. The error was reading the mode of travel off the Arolsen FOLDER heading, “TRANSPORT BY AIRPLANE”, rather than off the sheet. A folder heading describes the folder. And the ship changes the story: Bremerhaven is a four-week sailing from Melbourne, so they left Germany in August.
A Berlin-born Leibholz who reached Melbourne ten weeks before the war Documented
did Joachim Peter Leibholz, born Berlin 20 August 1903, landed from the ANCHISES on 12 June 1939 with his wife and a child the catalogue does not name, and was naturalised at South Yarra on 27 July 1944 with his previous nationality recorded as STATELESS. Six files. One of them — A1478, EPCP63 — is “German property taken over by the Polish Government”, and ours are from Schubin in Posen. Not kin on any evidence held, and kept as a lane of its own.
LAIBHOL* found a file that LEIBHOL* does not Documented
did Eight surname wildcards were run at the NAA rather than one. A1444, EPC164 is “German currency bonds - J P LAIBHOLZ” — the same man, spelt differently by his own government on a different form, and invisible to any search that assumes a surname has one shape. The LOWINSK* sweep also turned up Helga Lowinsky, born 14 April 1931, who sailed to Australia from Britain in 1964: the right forename, the wrong year, the wrong country, and exactly the kind of near-miss a careless search would have taken.
Wrote up Paula Luwinski as a discovery, and the Kurt page already had her — with the correction Withdrawn
did A gazettes.africa sweep returned “Paula Luwinski, a commercial traveller, Mark Hotel” and this archive wrote it up: Kurt’s widow, on the road in 1972, her own shop by 1975. Both notices were already on the Kurt page. So was the identification. AND SO WAS A PUBLISHED CORRECTION OF THAT EXACT MISREADING — the commercial traveller at the Mark Hotel is Phillip Crystal, the man being sequestrated, and Pauline Luwinski is the creditor who brought the application. The archive repeated its own corrected error, from the same snippet, four days later. The standing rule has been widened accordingly: before writing up any find, grep the site for the name.
What the gazette sweep did add: a second family, and a clean absence Documented
did Running LOWINSKI, which had never been run, returns eight documents and a separate South African family — Thos. H. Lowinsky in 1922, Mark Lowinsky in 1944 and 1961, Rupert Esmond Lowinsky of Maclear whose estate 671/55 was advertised twice in 1955. They are in print eleven years before Kurt landed, so they are not ours, and they are recorded so that nobody rules them out twice. And LEIBHOLZ, LAIBHOLZ, LEIBHOLTZ and LEIBBHOLZ return NOTHING in any African gazette, in any jurisdiction. Kurt went to Johannesburg alone.
The compensation files moved on 1 May 2026, and the plan had the wrong addressee Documented
did The twenty-item list called the Entschädigungsakten the largest unopened seam and put them at the Landesarchiv Berlin in B Rep. 025. They were not there. Until this year they belonged to the ENTSCHÄDIGUNGSBEHÖRDE, a department of LABO — a live authority, not an archive. On 1 May 2026 the Landesarchiv took the whole of it: about 280,000 proceedings and FIFTEEN KILOMETRES OF PAPER, stem files and subsidiary files together, under the federal project Transformation der Wiedergutmachung. The question is four and a half months old, and the archive's stated minimum to ask is a name and life dates — which this archive has for Kurt, for Auguste and for Siegfried.
Ten files a year, scanned free, and a status this archive will not claim Inferred
did The Landesarchiv's own page says relatives of Nazi-persecuted people, or those they authorise, may have up to ten files a year digitised at no charge, and that the quota covers Wiedergutmachungs- and Haftakten too. The letter mentions it and does not assert it. Derrick is Kurt's son on every document there is, and this archive has published at length the DNA result that puts that paternity in doubt. The German text says so plainly and leaves the decision to them. An archive that has questioned a relationship in public does not quietly rely on it in private.
A national card index nobody had heard of Documented
did Named on the Landesarchiv's own page: the BUNDESZENTRALKARTEI at the Bezirksregierung Düsseldorf, a central index of compensation claims across the whole Federal Republic. It matters because § 185 BEG gave each western Bundesland its own authority, and this family scattered across three jurisdictions — Kurt to Johannesburg, Auguste to Baden-Württemberg, Siegfried to Potsdam and the far side of the wall. A Berlin-only enquiry can come back empty while the file sits in Stuttgart. The BZK answers the prior question — did they ever file, and where — for all four at once.
Trove does not work here, and that is a measurement rather than a result Withdrawn
did Tried again an hour after the first failure. The application loaded; the search box was clicked, the surname typed a character at a time, Return pressed, and the submit button clicked twice — once by coordinate, once by element reference. THE RESULTS AREA NEVER POPULATED AND NO SEARCH REQUEST WAS EVER SENT: the network log holds the configuration calls and the fonts and nothing else, and the console is clean. Trove has not been searched. Not one query has run, and anything said here about what it holds would be invention. The Ryerson Index answered “Server Overloaded” to six attempts in the same sitting.
Asked Olsztyn for a marriage, not a birth Documented
did The letter to the Warmia archive is drafted, and it asks first for the marriage of Joachim Luwinski and Johanna Thimm — undated, but bracketed between her birth in 1871 and their daughter's in 1893, so somewhere about 1888 to 1893 at Biesowo or Reszel. A marriage record names the parents of both parties. Johanna's are known; JOACHIM'S ARE KNOWN TO NOBODY, and they are the oldest generation reachable on this side. It also names the four indexed entries the spelling sweep found — Allenstein 1874 no. 11, Reszel 1876 no. 68, 1881 no. 31, 1884 no. 9 — so an archivist can go straight to the book. And it gets the 1874 line right: civil registration began on 1 October that year, so the 1865 and 1871 births are church records, and the letter asks whether those books sit at Olsztyn or at the archdiocesan archive.
Berlin publishes its registry name indexes free, and nobody here had looked Documented
did The Landesarchiv's Standesamtsabfrage lists the alphabetical name indexes to the Berlin birth, marriage and death registers as free PDFs, office by office. A hundred and forty-two offices, harvested from its own autocomplete and all three register types queried for each. THE RESULT IS A BOUNDED JOB: Kurt's 1912 birth entry number is in Schöneberg II's index for 1911-1913, a single 33 MB file, or in one of twenty single-year 1912 volumes for the old city. And the 1924 marriage — which would name both sets of parents and carry a marginal note if Kurt was legitimised — is in one of two Schöneberg volumes. The PDFs render blank in this browser, so reading one means downloading it, and that is a question for David rather than a decision for the archive.
Four negatives with one letter in common Withdrawn
did This was about to be published: “Schöneberg, this family's own district, has no online name index at all.” Three queries returned nothing and a sweep of all 142 offices put Schöneberg, Neukölln, Köpenick and Weißensee in the empty column. EVERY NAME WITH AN UMLAUT RETURNED ZERO AND EVERY NAME WITHOUT ONE RETURNED ROWS. The site stores its office names double-encoded — the stored string is literally “Standesamt Köpenick” — and matches on the whole string, so the correct spelling finds nothing and the mangled one finds seventeen volumes. Schöneberg has births from 1874 to 1913 and deaths year by year to 1994. When a set of negatives shares a character, suspect the character.
Read the 1912 L section at Schöneberg II, and Kurt is not in it Documented
did Four pages, entries 14 to 1119, every surname from Laband to Lüttke. NO LUWINSKI. Schöneberg I's birth indexes stop at 1910 and Schöneberg II's volume covers 1911 to 1913, so between them 1912 is covered exactly once and he is not in it. HE WAS NOT BORN IN SCHÖNEBERG. The pages were read by loading the archive's own PDF into a renderer inside the browser rather than downloading it — 192 images, only the cover carries a text layer, and the rest is a clerk's handwriting photographed from a worn book.
A Luwinski at number 32 and a Leibholz at number 31 Documented
did This archive held one line — 1913, “LuwinSki, I., Invalide” — with no street, no house and no district, and had never opened it at the page. The page gives CHODOWIECKISTRASSE 32, on the corner of Winsstraße, Berlin N 55, Prenzlauer Berg; the initial is J and not I; and THE HOUSE NEXT DOOR, NUMBER 31, HAS “LEIBHOLZ, G., KFM.” The volume was compiled in the autumn of 1912 and Kurt was born on 11 May 1912. Eleven years later Auguste Luwinski married Jacob Leibholz. Nothing proves the two households are those families and both surnames are borne by other Berliners — but that marriage has never had a geography before, and Kurt's birth has never had a district.
Proved the street instead of reading it off the running head Inferred
did A running head names the first and last street on a page, not the street a column belongs to — and taking a number from one entry and a name from its neighbour is exactly the error that produced the false estate notice in September. So the heading was hunted down on the previous page: “NO. 55 · Chodowieckistr. · A.G. Berl. Mitte — L.G. I”, with the numbers running unbroken from 1 to 40 before the street ends at an arrow marked Greifswalder Str. and a new heading begins. The corner confirms it: 32 is marked “s. a. Winsstr. 42” and 33 “s. a. Winsstr. 33”, consecutive numbers facing each other across a junction. A TEXT-LAYER HIT IS NOT A READ ENTRY: the OCR gave the name and swallowed the street, the number, the neighbour and the initial.
Standesamt Berlin VIII, 1912, number 963 Documented
did *** KURT'S BIRTH ENTRY, FOUND. *** “Luwinski · Kurt Myszislaus. · 963”, the fourth line of the May block in the L section of P Rep. 523 Nr. 1395, the Landesarchiv's name index to the birth register of Standesamt Berlin VIII for 1912. Twenty-one candidate volumes became one in three steps: Schöneberg excluded by reading its entire 1912 L section; the 1913 directory giving Chodowieckistraße 32; and the Berliner Adressbuch's own street index, whose FIRST COLUMN IS HEADED STANDESAMT, giving houses 23 to 32 as office 8. Four things then agree — surname, forename, month, and a district arrived at independently — and the arithmetic checks: entry 963 of a register that passes 2561 by December sits 37.6 per cent through it, and 11 May is 36 per cent through the year. Every L page of 1912 was read; there is exactly one Luwinski in the whole year.
A second forename nobody has ever seen Documented
did MYSZISLAUS. Not the family tree, not the Arolsen index, not the Berlin registration card, not one South African document has ever carried a second forename for Kurt. It is a German clerk's spelling of the Polish Mieczysław — and it is the first thing this archive has learned about him that came from him rather than about him.
Schöneberg II, 1924, number six — and the clerk wrote down the Mischehe Documented
did *** THE MARRIAGE ENTRY, FOUND. *** “Leibholz · mos. · verheiratet mit Luwinski · kath. · Nr. 6”, and the same entry again under her name, on page 221 of the marriage index of Standesamt Schöneberg II. The tree gives the wedding as 5 January 1924; a register restarts at 1 each January and the entries above are 1, 2 and 4, so the sixth marriage of the year is a marriage in the first week of it — date and number agreeing without either being derived from the other. AND THE REGISTRAR WROTE THE RELIGIONS: mos. against Jacob, kath. against Auguste. Her Catholicism had been inferred from the Ermland, from Sauerbaum and from a census card. Here it is written down at her own wedding, in the same line as her husband's Judaism.
Two offices ruled out, and the ruling out dates a move Documented
did A Berlin marriage is registered where a party lives, and no 1924 address was known for either of them. Berlin VIII was tried first because the Luwinskis were at Chodowieckistraße in 1913 — its whole January 1924 L block is Lehmann, Liedtke, Langer, Levy, Lewandowski, Löffer, Langhans, Lange, Lauterbach, Lenz, and nothing else. Schöneberg I next: Liefländer 4, Lehmann 11, Lucke 15, Leu 19. Nothing. Then Schöneberg II, number 6. WHICH IS A FACT ABOUT THE FAMILY AND NOT JUST ABOUT THE SEARCH: they were in eastern Schöneberg by January 1924, nineteen months before Siegfried was born at Martin-Luther-Straße 60, which had been the earliest Schöneberg document there was.
The neighbour is Georg, and Jaques is four lines below him Documented
did The 1913 directory's name section spells out the street entry: “— Georg, Kaufm., NO 55, Chodowieckistr. 31 II.” — the Leibholz next door to the Luwinskis. This archive already holds a Georg Leibholz born at Hammerstein who cannot be Jacob's brother; whether he is this man is not established, though the volume lists exactly one Georg Leibholz in Berlin. AND FOUR LINES BELOW HIM: “— Jaques, Kaufm., W 62, Lutherstr. 34 II.” The man Auguste married eleven years later, on the same page, twenty-five years earlier than any address previously held for him. The directory also prints its own eighth spelling of the name: “Leipholz, s. a. Leibholz”, which has never been searched anywhere.
A Neubau in 1912, gone by 1914 — and it qualifies this morning's claim Withdrawn
did The volumes either side of 1913 were opened. In 1912 Chodowieckistraße 32 is “Neubau. E. Berl. Terrain-Gesellschaft” — a new building still owned by the developer, with no tenants at all — and number 31 has no Leibholz. In 1914 number 32's alphabetical list runs Landow, Michaelis with nothing between, while 31 still carries Leibholz, G., Kfm. SO THE LUWINSKI TENANCY IS ONE VOLUME WIDE: into a brand-new house during 1912, out again before the autumn of 1913, and Georg Leibholz arriving next door in the same year. AND IT QUALIFIES THE BIRTH-ENTRY FINDING: this archive counted the district as one of four agreeing facts, but that address dates from the autumn and the birth was in May. Three agree outright and one partly. The entry is untouched — the register index was opened and read — but the district was the reason for opening the volume, not part of the proof.
The eighth spelling produced a person Documented
did LEIPHOLZ run at Mapping the Lives, which this archive had only ever searched as LEIBHOLZ. Exactly one result, surfaced through the maiden-name field: ELISABETH WEILER NÉE LEIPHOLZ, BORN 1 JANUARY 1891 AT BERLIN — the generation of Jacob, Samuel and Oskar, and in none of this archive's files under any spelling. The control was run on the same screen and returns the hundred-capped LEIBHOLZ list the archive already holds, so the single result is a result rather than a failure to search. NOT KIN ON ANY EVIDENCE HELD. It incidentally confirmed something else: Agnes Leibholz née Kerger, born 10 March 1889, at Leinestraße 32 III rechts in Neukölln — the address this archive had for Heinrich Leibholz, with a maiden name and a birth date attached for the first time.
Fifty in the memorial book, and only three born at Schubin Documented
did The Gedenkbuch was opened only to control a search for LEIPHOLZ, which returns nothing. The control returned FIFTY Leibholz, and the fifty say what the four already held could not. THREE WERE BORN AT SCHUBIN — Jacob, Samuel and Oskar — AND NO OTHER LEIBHOLZ IN THE BOOK WAS. The other forty-seven cluster in Pomerania and West Prussia within about sixty kilometres of Schlochau: Glowitz 6, Hammerstein 5, Karwen 5, Schlochau 3, Bärwalde 2, Preußisch Friedland 2, Tempelburg 2. The German centre of gravity of this surname is Kreis Schlochau, not Posen — and Sally Leibholz, Jacob's father, was born at Hammerstein in 1847 and married at Szubin in 1874. The archive held both facts for days without drawing the line. The Schubin branch is a Hammerstein family that moved east.
Georg's wife was born eighteen kilometres from Auguste's village Inferred
did The same fifty gave a birthplace the archive did not hold: LYDIA LEIBHOLZ, born 16 January 1885 at BISCHOFSTEIN — Lydia née Perl, wife of Georg of the Michaelkirchstraße laundry, dead with him at Chełmno. Bischofstein is in the Ermland, about eighteen kilometres from Sauerbaum, where Auguste Luwinski was born. AND GEORG LIVED NEXT DOOR TO THE LUWINSKIS at Chodowieckistraße in 1913 and 1914. It proves nothing — two Ermländer households side by side in Prenzlauer Berg is how chain migration works. But it is the first mechanism this archive has ever had for how a Catholic dressmaker from Kreis Rößel came to marry a Jewish merchant from Posen.
The same couple, spelt with a B in Berlin and a P in Jerusalem Documented
did Yad Vashem returns two genuine LEIPHOLZ: WILLI, born 1887 at Berlin, and JENNY, born 1893 — both murdered, both with a Page of Testimony. The Bundesarchiv's fifty, read an hour earlier, contain “Leibholz, Wilhelm Willi, born 30.12.1887 in Berlin” and “Leibholz, Jenny, born 11.02.1893 in Loosen”. SAME COUPLE. A German federal archive wrote the name with a B and a Jerusalem memorial wrote it with a P. The Berliner Adressbuch said the two spellings were one name in 1913; this shows it on two named people in two databases compiled by different institutions from different sources — and it supersedes the weaker version of the same claim, the policeman Adolf of Friedenstraße, which rested on OCR and stays unverified. EVERY NEGATIVE THIS ARCHIVE HAS RECORDED UNDER ONE SPELLING IS PROVISIONAL UNTIL IT IS RUN UNDER THE OTHER.
The eighth spelling reaches our own province, and brings a warning with it Inferred
did Geneteka, run properly after the empty-parameter trap caught this archive a second time — both spellings first returned an empty table with the province and record type left blank, and the Kowalski control with them filled in returned rows at once. AN EMPTY PARAMETER IS NOT “ALL PROVINCES”; IT IS NO RESULT. Re-run: in KUJAWSKO-POMORSKIE, the province Szubin is in, LEIBHOLZ returns “Brak danych” and LEIPHOLZ returns Gottlieba Leipholz, mother of a child baptised at Runowo Krajeńskie in 1905. AND THE WARNING: warmińsko-mazurskie returns an eighteenth-century PROTESTANT Leipholz family at Orzysz-Arys in Masuria — Gotlib, Daniel and Maria, children of Gotlib Leipholz and Maria Kliss. A Lutheran family in Masuria is not a Jewish family from Posen by any reading. Leipholz is also a surname in its own right. THE RULE WIDENS THE SEARCH; IT DOES NOT WIDEN THE FAMILY.
Nine instruments, five new people, and three proofs of the same thing Documented
did The eighth-spelling sweep finished at Arolsen, which returns BODIN, ELSE, maiden name LEIPHOLZ, born HAMMERSTEIN 12.08.1884 — and the Gedenkbuch's fifty contain “Bodin, Else (* 12.08.1884 in Hammerstein)” under LEIBHOLZ. A THIRD INSTITUTION, A THIRD PERSON, THE SAME TWO SPELLINGS, after Willi and Jenny across the Gedenkbuch and Yad Vashem. And this one is Hammerstein-born, which is where Sally Leibholz was born and where five of the fifty were born: the spelling question and the geography question turn out to be one question. Nine instruments swept in an evening — the Berlin directories, the NAA, Gazettes.Africa, Mapping the Lives, the Gedenkbuch, Yad Vashem, Queensland, Geneteka and Arolsen — producing five people this archive did not have, one Masurian Protestant family that has to be kept out, and a rule that is no longer a hunch.
The living-people gate watched the pages and not the data Documented
did A peer session found that these archives' check:living walked rendered HTML only, and widened the shared kit's copy. THIS ARCHIVE RUNS ITS OWN COPY, so the fix did not reach it and had to be made here — the one thing a peer cannot do for you is fix a tool you do not use. This site serves two JSON files a browser fetches directly: the atlas data behind the map and the search index, and a living person could have reached a reader through either without the build saying a word. MEASURED BEFORE CHANGING AND THE LEAK WAS ZERO: a first crude search looked like three hits and every one was a published dead namesake — Joachim Luwinski born 1865, Sally Leibholz, Gertrud Leibke. The blind spot was real and the leak was not, and those two have to be established in that order. Now walks html and json, 27 pages become 29 files. AND PROVED RATHER THAN ASSUMED: a scratch build with a guarded living person in a served atlas-data.json and nothing in the HTML — the old gate answered “ok” and exited zero, the new one named the file, the person and the matched text and failed the build.
The Weinstraße household was at Yad Vashem all along, under the eighth spelling Documented
did This archive's register of lines said the Weinstraße Wilhelm was NOT the Wilhelm Yad Vashem has. That was written off a LEIBHOLZ search and was true of that search. BOTH WILHELMS ARE THERE — the 1890 one under LEIBHOLZ and the Weinstraße one under LEIPHOLZ. Page of Testimony 5369884 matches the household exactly: born 30.12.1887, wife Jenny née Rosenberg, address WEINSTR 6. And it says what a census card cannot — place of death, AUSCHWITZ, where Mapping the Lives could only give “died before 08.05.1945”. Jenny's own page gives her parents, ISIDOR and CECILLIE ROSENBERG, and her birthplace: SCHLOCHAU, the district Sally Leibholz was born in and the district the Gedenkbuch's fifty cluster in. The Gedenkbuch says Loosen instead; both are the same corner of Pomerania and the archive is not choosing between two sources on no evidence.
Somebody sat down and wrote it out Inferred
did Both pages were filled in by KHANAN ROSENBERG VARADI — “Family” on Willi's, NEPHEW on Jenny's. A Rosenberg, a child of one of Jenny's brothers or sisters, who survived and who at some point set out what had happened to his aunt and his uncle. Everything else in this archive is a clerk's form, a register line, a transport list, a card index. A PAGE OF TESTIMONY IS THE ONLY KIND OF DOCUMENT HERE THAT SOMEBODY CHOSE TO MAKE. It is not evidence of kinship and is not treated as any; no approach has been made to him or to anyone connected with him, and whether one ever should is not this archive's decision.
Checked a distance, and found the map fifty kilometres out Withdrawn
did The claim under test was small: that Bischofstein, where Georg Leibholz's wife was born, is about fifteen kilometres from Sauerbaum, where Auguste Luwinski was born. The Kreis checks out — German Wikipedia has Bischofstein assigned to Kreis Rößel in 1815 and still Landkreis Rössel in 1945. The distance came out at 33 kilometres, which was wrong twice over: the real figure between the villages is 17.8, and THE ATLAS PIN FOR SAUERBAUM WAS AT 54.35/21.15, FIFTY KILOMETRES FROM ZERBUŃ. It had been placed before the village was identified at all, and the published map has been showing Auguste's birthplace in the wrong place ever since. Pin corrected, distance corrected to eighteen everywhere it appears, correction 9500903. The arithmetic that tests a claim also tests everything the claim is measured against, which is the cheapest reason to do it.
Lutherstraße belonged to Charlottenburg, and the reason given last night was wrong Withdrawn
did The street section's own heading, printed page 534 of the 1913 volume, reads “W. 62 Lutherstr. · A.G. Charlottb. — L.G. III. · (geh. z. Charlottenburg)”. Last night this archive published that W 62 was Berlin proper and reasoned from it. IT BELONGED TO CHARLOTTENBURG, an independent city until 1920 exactly as Schöneberg was. The conclusion survives — Jacob was not in Schöneberg in 1913 — and the reasoning does not. It also explains a puzzle: the Berlin Straßenverzeichnis, the index whose first column gave Kurt's birth office, has no Lutherstraße at all; its L column runs Lustgarten, Lutherbrücke, Lütticher with nothing between, because that index lists Berlin's own streets and this was not one. So Jacob's 1913 office is one of Standesamt Charlottenburg I, II or III (this entry said I to IV until 23 September, when the Landesarchiv's holdings were enumerated rather than guessed at: CHARLOTTENBURG IV DID NOT EXIST IN 1913, its earliest volume of any kind being 1922, and the unnumbered office closes in 1897). And the geography deserves saying beside the birth entry: Kurt was registered at Berlin VIII in Prenzlauer Berg, where the Luwinskis lived, and Jacob was nine kilometres west in another municipality.
He stayed, and Victoria's free index names his parents Documented
did Joachim Peter Leibholz, the Berlin-born man who reached Melbourne on the ANCHISES ten weeks before the war, died at TOORAK in 1971 aged sixty-eight — reg. 26252/1971, two suburbs from where he naturalised in 1944 and consistent to the year with a birth on 20 August 1903. AND THE INDEX CARRIES HIS PARENTS: father LEIBHOLZ WILLIAM, mother NANETTE née NETTER. This archive has already met a William Leibholz of Berlin — Generaldirektor and Stadtrat in the 1920 directory, and “William, Kaufm., W15 Kurfürstendamm 169” in 1913, the only William on that roll. Strong and not proof. It does not connect him to this family: ours are Konfektion merchants from Schubin and William is a managing director on the Kurfürstendamm, and the Melbourne lane stays a lane. The wildcard L*HOLZ across births, deaths and marriages 1900 to 1995 returns exactly this one record, with the honest limit that Victorian birth indexes close about 1922 and marriages about 1942.
Spier has a trade, and the house has a landlord with a forced name Documented
did Jacob's transport line reads “Treuchtlingerstr. 4 b/Spier” and the archive could not say who Spier was. The 1941 street directory gives the house: owner EISENBERG, G. SARA, WW. — a widow carrying the name the 1938 decree forced on her — with STEPHAN, H. ISRAEL, ROSENSTOCK, L., and SPIER, H., HANDELSVERTRETER among the tenants. A Jewish-owned house filling with Jewish tenants in 1940 is what a Judenhaus was, and the directory shows it without using the word. AND JACOB IS NOT IN IT: that volume was compiled in autumn 1940, so the forced move is now bounded at both ends — Meraner Straße 2 on 17 May 1939, absent from Treuchtlinger in autumn 1940, Treuchtlinger 4 bei Spier on 26 February 1943. The 1942 and 1943 pages would close it further and the OCR will not give them up.
Szubin said yes, and the registers had been online since 2015 Documented
did Two letters went to Polish archives on 14 September. KOSZALIN answered that the Hammerstein books are gone and nobody knows where. BYDGOSZCZ answered on 22 September by pointing at genealogiawarchiwach.pl, where the registers of USC Szubin Miasto had been published since 30 September 2015. Both documents were read at the image within the hour — Jacob Leibholz's birth entry of 8 August 1884, and the marriage of 24 November 1874. THE ARCHIVE SPENT EIGHT DAYS WAITING FOR A DOCUMENT IT COULD HAVE OPENED ON THE FIRST. The rule that earns: before writing to a Polish state archive, search the portal for the registry office by name. The letter still earns its place — it produced the exact link — but it is the second move, not the first.
Hammerstein, from a document at last — and a generation above Documented
did TWO THINGS IN THIS ROW WERE CORRECTED ON 27 SEPTEMBER AND IT IS LEFT STANDING SO THE MISTAKE IS VISIBLE: the marriage is 24 DECEMBER 1874, not November, and the mother's given name is NOT WILHELMINE — it is unreadable past its initial H. Standesamt Schubin, marriage akt nr 9 of 24 November 1874. DER KAUFMANN SALLY LEIBHOLZ, siebenundzwanzig Jahre alt, GEBOREN ZU HAMMERSTEIN, WOHNHAFT ZU HAMMERSTEIN, son of the Kaufmann JACOB LEIBHOLZ and WILHELMINE GEB. POLLNOW, married FRIEDERICA SCHMUL, born and living at Schubin. The Schubin branch is a Hammerstein family that moved east, and that is no longer inferred from the Gedenkbuch — a registrar wrote it down. Twenty-seven in November 1874 puts his birth in 1847, the year this archive has always carried and which no surviving register can confirm. THE MARRIAGE ENTRY DID WHAT THE DESTROYED BIRTH REGISTER WOULD HAVE DONE, and it named two people who did not exist here in any form that morning.
Neither Wilhelmine nor Philipine — the name read against the clerk's own capitals Withdrawn
did Two transcribers disagreed about Sally Leibholz's mother: this archive read WILHELMINE at the image on 22 September, the portal's volunteer indexer read PHILIPINE. Settled on 27 September WITHOUT PREFERRING EITHER, by asking what this registrar's letters look like where the answer is already known. HIS CAPITAL P — four words along the same line, in POLLNOW — IS A SINGLE ROUND BOWL, and the disputed capital has no bowl at all; Philipine is out on the same stroke of the same nib. HIS CAPITAL H — in HAMMERSTEIN, written twice on the page as the groom's birthplace and residence — has a long flat baseline lead-in, two steep diagonals and A LOW CROSSBAR joining them, AND THE DISPUTED CAPITAL HAS ALL THREE. So the name begins with H, the rest is x-height minims at the scan's limit, and she is published as “H—— geb. Pollnow”. HENRIETTE IS THE NAME THAT WOULD OCCUR TO ANYONE READING A JEWISH PRUSSIAN REGISTER OF 1874, WHICH IS EXACTLY WHY IT IS NOT WRITTEN DOWN. Stated weakness: Kurrent W and H are alike and there is no W in this hand anywhere on the spread, so W is excluded on the positive H match rather than on a W exemplar. AND THE EARLIER REASONING IS NAMED: a pass this same week read Philipine and defended it by claiming a W “has no ascender loop” — false, and worse, A CLAIM ABOUT THE ALPHABET IN GENERAL WITH AN EXEMPLAR SITTING FOUR WORDS AWAY.
A catalogue is not an archive — Bydgoszcz sent the index this site said did not exist Withdrawn
did On 22 September this archive recorded, with a control it was rather pleased with, that SZUBIN DEATHS HAVE NO ALPHABETICAL INDEX: filtering the portal to skorowidze zgonów returns zero while the identical query on akta zgonów returns 190 units, so the zero is a real absence and not a broken filter. Row 264 then costed Sally Leibholz's death at 190 blind volumes and sent the reader to Berlin instead. THE REASONING WAS SOUND AND THE CONCLUSION WAS WRONG. On 25 September the Archiwum Państwowe w Bydgoszczy answered a letter about something else entirely by sending USC SZUBIN MIASTO SYGN. 246 — “Alphabetisches Verzeichnis zu dem Sterbe-Register über Standes-Amt Schubin von 1894 bis 1908”, 103 scans, shared from the archive's own drive by Sandra Dończyk and uploaded by the case officer. THE ZERO WAS A FACT ABOUT A WEBSITE AND THIS ARCHIVE TURNED IT INTO A FACT ABOUT A BUILDING. A portal shows what has been digitised; an archive holds what it holds; the only instrument that reads the difference is a letter to an archivist. AND THE VOLUME ANSWERS IN FOUR PAGES WHAT WAS COSTED AT 190. It is not alphabetical in the usual sense — within each letter the entries run BY YEAR, so you read the whole letter, which is the slot method as a physical object. K ends on scan 44, L runs 45 to 48, M begins on 49, every year from 1894 to 1911 present: Lehmann, Lis, Luckstaedt, Lewandowska, Liesske, Lurkiewicz, Lesner, Luczny, Leonski, Lawrenz, Lesniewski, Lüttke, Lichy, Loewe, Langer, Loeffler, Leu, Leiser, Łowiński, Lange — NO LEIBHOLZ. AND THE SLOT CARRIES ITS OWN CONTROL, which is worth more than the negative: the first entry in it is LEHMANN MOSES, Schneider, and 1909 opens with LEISER ADOLF, with Samuel, Silberberg, Salzbrunn and Schobelt elsewhere in the book. THE INDEX LISTS JEWS, so the absence is real and not an artefact of who got recorded. Schmul was read the same sitting, scans 80-88, and is absent too. WHAT IS NOT COVERED IS THE LIKELIER WINDOW: the volume begins in 1894, Sally was alive in August 1884 and would have been 46 in 1894, so 1884-1893 is untouched, as are the two Szubin Wieś offices. Row 264 is narrowed, not closed.
The newspaper archive was paid for on 16 September and opened on the 27th Documented
did Findmypast's British Newspaper Archive is included in the subscription David took, and the work-list row recorded it as “UNTOUCHED, AND THE SUBSCRIPTION IS PAID FOR”. It paid on the first search. THE 1914 MARRIAGE HAD BEEN AN INDEX LINE — Castle-Wear, Hitchin, Q4 1914, vol 3A p.1552, a three-month window. TWO PAPERS REPORTED THE DAY: the Citizen (Letchworth) of Friday 6 November 1914 and the Hertfordshire Express of Saturday 7 November under a BALDOCK heading. One says “last Saturday”, the other “on Saturday”, and BOTH RESOLVE TO 31 OCTOBER 1914. The index could not give the church (the CONGREGATIONAL CHAPEL at Baldock, the Rev. T. Smith pastor — Nonconformist, which is a fact about the family), her street (NORTON STREET, BALDOCK, the first address ever held for Martha Maud before her marriage), or the circumstance: he was “in full uniform”, ON FURLOUGH FROM BURY ST EDMUNDS where the 1st Herts Territorials were headquartered, the wedding “took place quietly”, he rejoined, and THE 1/1ST HERTFORDSHIRES LANDED IN FRANCE ON 6 NOVEMBER — the day the Citizen printed its notice. HE MARRIED ON ABOUT A WEEK’S LEAVE. The regiment itself was NOT news, and that is the point: his WO 363 record already gave it, and a service file and a village paper agree without either knowing the other. ALSO: the Express calls him HARRY WARE, so a WEAR search misses him — Ivone’s six spellings have an English counterpart. AND A NEGATIVE THE ROW WANTED MORE: no death notice for the 1863 Martha Castle in any Hertfordshire paper, on a search string the row itself had wrong — she died under her MARRIED name, not as Martha Giddings. Coverage control holds, the absence is real and weak, and it does not separate the 1863 death from the 1865 one.
I read one line of output and concluded the wrong thing about my own work Withdrawn
did Reporting this archive's searches into the shared Record Atlas, the importer printed “Luwinski Family: 8 rows, 0 MATCHED TO AN ATLAS VOLUME”. I took that to mean nothing had landed, and said so — to the user and to a peer session — concluding that an archive working non-FamilySearch sources “complies perfectly and still contributes nothing the map can colour”. THAT WAS WRONG AND A PEER CHECKED IT RATHER THAN AGREEING. The per-archive line counts ONLY volume marks, which are keyed on a FamilySearch waypoint. SOURCE marks are counted once in a separate total and never broken out per archive. TWO OF MY EIGHT ROWS HAD LANDED AS SOURCE MARKS AND WILL COLOUR THE MAP — verified in the generated file, both carrying this archive's name and the right states. I read the one line that existed and did not look at the artefact it described. *** AND CHECKING THEIR CORRECTION FOUND A REAL BUG NEITHER OF US HAD SEEN. *** They reported Findmypast as an unregistered provider and guessed my dates would have updated the shared marks anyway. Neither holds. FINDMYPAST IS REGISTERED (id findmypast, url findmypast.co.uk) and the mark exists — but it carries ANOTHER ARCHIVE'S note and ANOTHER ARCHIVE'S DATE, 9 September, where my rows are the 27th and 28th. Same for Trove. THE MECHANISM IS ONE LINE: import-research-sessions.py does «m["since"] = max(m.get("since",""), when)» — a STRING max over dates written in two different formats. max("9 Sept 2026", "2026-09-28") is "9 Sept 2026", because ASCII 9 beats ASCII 2. SO ANY HUMAN-FORMAT DATE BEGINNING 3 TO 9 BEATS EVERY ISO DATE, PERMANENTLY. Across the estate: 1,072 rows human-format against 156 ISO. This archive is 8 of 8 ISO because the skill’s own template shows ISO — so following the instruction put us in the losing format, and the map will say these sources were last worked on 9 September when they were worked on the 28th. Reported with the line numbers rather than patched: it is a shared script, another session was committing in that repo minutes earlier, and a silent fix to somebody else’s tool is how the estate got the problem it spent today naming. *** RESOLVED THE SAME DAY, AND BY THE SAME METHOD. *** The peer ran max("9 Sept 2026", "2026-09-28") themselves rather than take my word, got "9 Sept 2026", and read the live mark: source:findmypast still says 9 September with another archive’s note, so the 27-28 September work is STUCK, not merely uncounted. They retracted “not a registered provider” plainly. THEY ALSO DECLINED TO PATCH IT, for the reason I declined to, and passed the line numbers and the 1,072-to-156 count to Record Atlas whose script it is — and declined to switch the skill’s template to human format, which would have hidden the defect behind a permanent dependency on it. NOBODY WHO FOUND THIS BUG FIXED IT, AND THAT IS THE CORRECT OUTCOME: two sessions each checked the other and neither edited a tool it did not own.
A page argued with itself for a fortnight, and the gate built for it could not see this kind Withdrawn
did Derrick's summary panel said his father was “Kurt Israel Luwinski, of record — DISPUTED BY DNA”. The exclusion had been published for a fortnight: 0.0% Ashkenazi where a quarter would be required, a Y haplogroup that is North Sea rather than Ashkenazi, and the tree page grading him RULED OUT and saying in as many words that this is not a level on the ladder but a claim tested and failed. THE EXCLUSION IS NOT A DISPUTE AND IT IS NOT OPEN. What is open is WHO the biological father was. A PEER SESSION THEN COPIED THAT PHRASE OFF THE PANEL INTO A CHART NOTE, so the page contradicted itself within one screen — the panel saying disputed, the chart beneath it saying excluded and settled — and the peer found it when I corrected their note without correcting the sentence they had taken it from. IT WAS TRUE WHEN WRITTEN AND WAS NEVER READ AGAINST THE FINDING THAT REPLACED IT. *** AND IT EXPOSES A BLIND SPOT IN THE GATE THIS ARCHIVE BUILT YESTERDAY FOR EXACTLY THIS FAULT. *** check-superseded refuses a build that still asserts a withdrawn VALUE — a date, a name — because a value is declared and a string can be matched. THIS WAS NOT A STALE VALUE BUT A STALE CHARACTERISATION: nobody had thought to declare “disputed” as a withdrawn way of describing a settled exclusion, so there was nothing for the gate to match. Declared now, as the third fact in data/superseded.json, which is the only fix available — THE GATE CANNOT FIND WHAT NOBODY DECLARES, and its coverage is exactly as wide as somebody’s memory. The peer reports the same shape twice more in the estate this week: an /about page saying the living were “absent from the build” while six were named in its own living.json, and a /register saying “not one is living” in an archive holding a living woman. THREE INSTANCES, ONE CLASS: a true sentence left standing beside the data that replaced it.
A withdrawn fact was asserted in twelve files, and the hand sweep missed five Withdrawn
did Correcting the 1874 marriage date meant changing it in twelve places. Every occurrence was listed with grep, read in context, sorted into must-change and must-not-change — the withdrawn claim and the akt nr 7 near-miss genuinely are 24 November and had to stay — and corrected. THE SWEEP WAS DECLARED COMPLETE AND IT HAD MISSED FOUR, all asserting the old date as live fact, all matching deep inside long strings where the grep output had been read at a glance: data/corrections.json, the register generated from it, negatives.json, and mappingthelives-leibholz.tsv. A FIFTH WAS WORSE — negatives.json carried “JACOB LEIBHOLZ AND WILHELMINE GEB. POLLNOW” in the SAME SENTENCE where the date had just been fixed and the name had not. SO THE WITHDRAWN VALUES ARE NOW DECLARED INSTEAD OF REMEMBERED. data/superseded.json names each withdrawn fact, what it used to be published as, and every path where the old value may still legitimately stand — which means the file IS the correction, or it quotes the claim in order to withdraw it, or it is a dated record superseded in place. check-superseded.py fails the build on anything else, and IT FOUND TWO MORE WITHIN SECONDS OF FIRST BEING RUN, after the hand sweep had finished. Controlled by planting a violation: FAIL and exit 1, ok when removed. Twelfth gate.
The W was one scan back, and the archive had already been standing on it Documented
did A finding was published with its weakness stated: the mother's given name in the 1874 entry reads as beginning with H, but KURRENT W AND H ARE ALIKE and there was no W in this clerk's hand anywhere on the spread, so W was excluded on a positive H match rather than on a W exemplar. Worklist row 271 was filed to go and find one. IT WAS CLOSED WITHIN THE HOUR, AND THE EXEMPLAR WAS SOMEWHERE THIS ARCHIVE HAD ALREADY BEEN. Akt nr 8 of the same volume marries HEINRICH GUSTAV JULIUS WENGATZ — the very near-miss the reader landed on in September while hunting for akt nr 9, written up in the archive's own notes as “the right date and the wrong people”. HIS CAPITAL W IS TWO ROUNDED LOOPING ARCHES, no straight strokes, no crossbar; the disputed capital is two steep straight diagonals joined low down. Not the same letter and not close. AND THE SAME PAGE THREW IN A SECOND CONTROL NOBODY ASKED FOR: akt nr 8 IS ALSO DATED DECEMBER, in the same hand, same construction — which orders the volume, nr 7 on 24 November, nr 8 in December, nr 9 on 24 December, and shows what was wrong with the November reading, which had nr 7 and nr 9 sharing a date with nr 8 stranded between them. THE CONTROL COST ONE CLICK IN A UNIT ALREADY OPEN AND SAT THERE UNUSED FOR FIVE DAYS. The rule: when you write down what would close a finding, check first whether you have already walked past it.
The wedding was never in November, and the archive had been holding the contradiction Withdrawn
did The same re-read moved the marriage by a month. The entry opens “Schubin am 24 vier und zwanzigsten DECEMBER tausend acht hundert siebenzig und vier” — day written twice and both give the 24th, month written once and it is December, a single looped capital D where this clerk's N is two tall strokes. THIS SITE PUBLISHED 24 NOVEMBER 1874 AND WROTE THAT IT WAS “the date the family record has always given, NOW STANDING ON THE REGISTER ITSELF”. It does not stand on the register. Everything around it agreed with December and never with November: PRUSSIAN CIVIL REGISTRATION BEGAN 1 OCTOBER 1874, so the volume covers three months and holds nine marriages, and akt nr 9 is the last wedding of the year. AND THE ARCHIVE'S OWN NOTE ALREADY SAID AKT NR 7 WAS DATED 24 NOVEMBER — “the right date and the wrong people”. Two entries cannot both be it. HOW IT HAPPENED IS THE WHOLE POINT: the reader went to the document knowing the answer, was shown a genuine 24 November on the way in, and then read the right entry with the date already confirmed in mind. Sally's age is unaffected — twenty-seven on 24 December 1874 still contains the family's 7 July 1847.
The maiden name put to its own alternatives Documented
did POLLNOW was published with a warning that P, T and Z confuse in Kurrent. It was then tested against a list of German-Jewish surnames. In the Gedenkbuch POLLNOW returns EIGHT; TOLLNOW, ZOLLNOW and SOLLNOW return NONE; controls either side, Leibholz 50 and a nonsense string zero. Of the four ways that capital could be read, only one is a name anybody bore. AND THE DISTRIBUTION DOES NOT FIT THE STORY, WHICH IS WORTH MORE THAN IF IT HAD: the eight Pollnows are East Prussian — Mehlaucken, Königsberg, Darkehmen, Lyck — and not one is Pomeranian. The archive's own guess that the name might be toponymic from Pollnow in Kreis Schlawe is withdrawn.
Rosebank answered, and closed a route this archive was relying on Withdrawn
did The Church of the Immaculate Conception replied in a week to all five questions. The 1940 parish marriage register SURVIVES and CONTAINS the entry; it DOES NOT NAME THE PARENTS — “Unfortunately not, as attached”. And there is no second book to ask for: “the Catholic Church keeps one Marriage Register, the banns are included in the attached”. THE PAGE THEY SENT IS THE PARISH REGISTER. This site had argued that a Catholic sacramental register “ordinarily names the father and mother of both parties”; the reasoning was sound and the premise was wrong for this parish. Route 2 to Leonard Wear's parentage is closed by the only person who could close it, and the SANDF attestation form is now the whole of it.
Three wrong answers about one negative, in one evening Withdrawn
did The 1973 probate negative was re-examined three times before it was right. FIRST this archive read a zero from the half of Find a Will that structurally cannot hold a 1973 grant, and published it as an answer about 1973. SECOND it corrected that using a null control that returned 0 and 0 — a control run down a path where the year never reached the form, published without being repeated. THIRD it corrected THAT to “the image half runs no search at all”, on the strength of the page printing “41 document pages returned” for every surname, for nonsense, and for both years. ALL THREE CAME FROM READING THE RENDERED PAGE INSTEAD OF THE REQUEST UNDERNEATH IT. Called directly, the endpoint returns 41 for WEAR, 469 for SMITH and 0 for nonsense: a working search behind a broken display. THE RULE: where a page and its own API disagree, the API is the instrument and the page is a report about it.
Wilhelmine was in the data, in the manifest, past every gate — and not on the site Withdrawn
did The most important new person of the day never reached a page. Her row key was written “*** Wilhelmine Leibholz née Pollnow ***”, and person_name() refuses a cell that starts with an asterisk, so she was silently dropped. Jacob Senior likewise — “the elder” trips the prose rule. EVERY REJECTED KEY CELL HAS ALWAYS BEEN PRINTED, and that was not enough: the list is forty legitimate row labels long and the two that mattered were invisible in it. A warning you cannot find your finding in is not a warning. build-evidence now re-tests each rejected cell with its decoration stripped and reports separately, and loudly, any that would then pass the rule UNCHANGED. Caught only because David asked whether the front page was up to date.
Six times in one day, the archive treated what it already held as a discovery Withdrawn
did The count is the useful part. The Rosebank scan was written up as new when every field on it had been on the DNA page for five days, rebuilt from FamilySearch tiles. A row was raised saying the SANDF request had been turned from a guess into a document, when it was drafted with that evidence already in it. A follow-up letter was drafted asking the parish for two books its own email had already ruled out — because only the PDF attachment had been opened and not the message. THE PATTERN IS CONSISTENT: the new artefact gets read and the context around it does not — the email body, the existing page, the rejection list. Five of the six were caught by a control or by David rather than by the rule that exists to catch them.
A bound is not a date, and four of them were being printed as dates Documented
did year() takes the first four digits of a date string and throws the qualifier away. So “Before 1955” reached the bloodline as 1955 and the timeline PLOTTED A POINT AT IT; “After May 13 1942” likewise. Four people were affected. Fixed in the builder rather than the data: every record now carries byq/dyq beside by/dy, the correction loop sets them so a correction cannot re-introduce the fault, and both pages print the word. One of the four was then replaced outright — Elli Leibholz's “After May 13 1942” is the DEPORTATION DATE, and Gedenkbuch entry en1169257 has her dying at Kulmhof in May 1942, birth date matching to the day.
The name search was the wrong instrument, and a healthy control hid it for a day Withdrawn
did Row 251 asked for one cheap thing: bracket Friedrich Leipholz’s disappearance from the Berlin telephone books the way Erna’s was. A full-text sweep of all twelve volumes answered it immediately — present in four, absent in eight — and the answer was false. READING THE SLOT INSTEAD OF SEARCHING THE NAME turned eight zeros into six positives and two verified absences. The OCR prints this surname LEIPHOTZ in 1972-73, LELPHOLZ in 1976-77 and 1978-79, LEIPKSLZ in 1977-78 and LEIBHOLZ in 1980-81: FIVE OF THE ELEVEN READABLE VOLUMES MANGLE IT, and a search for the correct spelling misses every one. What licensed the error was a control that looked healthy — SCHMIDT, 133 to 191 hits a volume, in every book that has a text layer at all. IT CANNOT CLEAR A RARE-NAME NEGATIVE: a volume can be richly indexed and still lose one line to a broken glyph. The fix is the instrument row 252 already uses on the death registers — LEIPHOLZ sorts between LEIPERT and LEIPIC, so print the whole slot and read it, and treat a missing side as inconclusive rather than negative. It gained a third Friedrich volume, made Erna’s run continuous across all six, and moved the bracket a year: he is in 1973-74 and the slot is complete and empty in 1974-75. AND THE BRACKET THEN DISAGREED WITH THE REGISTERS — Reinickendorf 1974-1981 is already read in full and he is in none of it, so the tidy reading, that he left the book because he died, is the one thing this cannot say.
The firm was in the wrong street, and the proof was a letter that is not there Withdrawn
did Reading the Adressbuch householder runs for row 260 — as runs, not as a name search — turned up two published facts that did not survive. THE 1940 VOLUME HOLDS TEN LEIBHOLZ HOUSEHOLDERS AND THREE FILES SAID ELEVEN, with the correct enumeration of ten printed directly beneath the claim in one of them; the eleven belongs to 1939. And JAQUES LEIBHOLZ & CO., JACOB’S OWN FIRM, WAS AT KRAUSENSTRASSE 59-60 AND THIS SITE HAD IT AT KRONENSTRASSE on the Leibholz page, in the deportation file, in the directory file and in two entries of this log. Four readings in three volumes and not one says Kronen: 1925 prints it cleanly, 1934 and 1938 give it twice more, and 1938 again in a different section of the same volume. THE DECIDING EVIDENCE IS THE OCR NOISE RATHER THAN THE SPELLING — the f of “Kraufen” and the j of “Kraujeu” are both this scanner’s long s, and KRONENSTRASSE HAS NO S IN THAT POSITION. A misreading of Kronen could not produce those letters. That is worth more than four agreeing transcriptions, because it explains the noise instead of voting on it. The wrong street survived four days because both are real, both are in W8, and both are in the Konfektion quarter around Hausvogteiplatz; there was nothing implausible to catch. IT WAS CAUGHT BY READING THE SAME LINE IN A SECOND VOLUME, which is the only thing that would have.
Catalogued 5,383 volumes, then found the catalogue had a hole shaped like an umlaut Withdrawn
did Four open rows all reduced to one unanswered question — which register volumes exist and what do they cover — so the Landesarchiv’s whole free holding was enumerated: 5,383 NAME-INDEX VOLUMES ACROSS 139 OFFICES, 2,845 death, 1,595 marriage, 943 birth. The office names had to be got from the form’s own autocomplete first, because LABSA matches them exactly. THE FIRST PASS REPORTED 4,838 VOLUMES AND SAID TWENTY-NINE OFFICES HELD NOTHING, AND ALL TWENTY-NINE HAVE AN UMLAUT OR AN ESZETT IN THE NAME — Schöneberg, Neukölln, Köpenick, Weißensee. The autocomplete hands names back DOUBLE-ENCODED and the search needs them that way, but the result table prints them correctly encoded, so the loop queried in one form and filtered in the other. Every query succeeded; every row was then thrown away. A quiet, confident zero, twenty-nine times, worth 545 volumes. IT WAS CAUGHT BY A CONTROL AND BY NOTHING ELSE: Schöneberg has sixty-one death volumes, a figure this archive already rests row 252 on, and the catalogue said zero. The fault was not in the archive’s instrument but in the code reading it, which is the harder kind to see, because nothing errored and 4,838 looked plausible. Two things fell out of the corrected list. Yesterday’s Charlottenburg inventory was A THIRD SHORT — 207 volumes in six designations, not 143 in five — because the offices had been guessed at rather than enumerated. And THE MARRIAGE INDEXES STOP AT 1944 EVERYWHERE, which means the “free test before any purchase” that row 254 is built on does not exist if that marriage happened after the war, and for a Jewish woman widowed in Berlin by 1982 a late-1940s marriage is at least as likely as a pre-war one.
Read a borough out, and the first pass had bracketed the wrong name Documented
did Row 251 had thirteen Reinickendorf death-index volumes outstanding, 1982 to 1994, and they are now read — 807 pages of 2,120 through pdf.js over range requests, nothing downloaded. THE FIRST PASS WAS DISCARDED AND THE VOLUMES READ AGAIN. It tested whether each volume’s alphabet was open on both sides of LEIBHOLZ, and then reported thirteen negatives covering LEIPHOLZ as well. THOSE ARE TWO DIFFERENT SLOTS: Leibholz falls between LEH- and LEIC-, Leipholz between LEIN- and LEIS-. In the 1990 volume the nearest name above Leibholz is LEIPE, which sorts BELOW Leipholz, so the Leipholz slot had not been shown to be read at all. The conclusion would have been true and unearned, which is the same fault as trusting a zero from an instrument that was not looking. Two volumes, 1992 and 1993, failed mid-read with network errors and were NOT counted as absences; a failed fetch is not an absence. Both read clean on the retry. THE RESULT: twenty-one consecutive years of Friedrich Leipholz’s own borough, 1974 to 1994, are now read, and his death is not registered there. He leaves the telephone book during 1974 and Reinickendorf never buries him. AND THE SWEEP TURNED UP A PERSON THIS ARCHIVE DID NOT HAVE — LEIPHOLZ MARIE, Sterberegister 1988 Nr. 2306, the only Leipholz in all twenty-one years, dying in the borough he lived in. Not Erna Frida Else, who died in 1975, and not the Erna of Streitstrasse. A wife, a sister, or no relation at all; nothing joins them, and what would is a certificate nobody here can buy.
Went to the image for one line, and it moved a date four months and added a word Withdrawn
did Marie Leipholz was published the same morning as a TEXT-LAYER reading, with her death date given as “@507 — most likely 5 July, but not legible enough to publish as a date”. Rendering page 76 of the volume and cropping the row settles it: THE ROW IS TYPEWRITTEN AND PLAIN — “Leipholz | Marie | 05.11. | 2306 | ev.” THE DATE IS THE FIFTH OF NOVEMBER, and the guess was wrong by four months. It did no harm only because this archive’s rule made it say the layer was too poor to trust rather than round the guess into a fact. AND THE IMAGE CARRIED A COLUMN THE TEXT LAYER NEVER SHOWED AT ALL: she is EV., evangelisch, Protestant. The same column reads mosaisch for the Jewish woman in the Spandau 1982 index, and reads mosaisch for Leinkram Gertrud three lines above Marie on her own page — so the distinction is being made on that very sheet. A Protestant Leipholz belongs with the Masurian Leipholz family of Pianki and Arys, whom this archive already holds and already calls certainly not the same people, and not with Jacob Leibholz of Schubin. One word, no proof, and for once it points AWAY from the family rather than toward it — which is worth as much.
The family record was right and this archive was wrong Withdrawn
did Derrick sent photographs of his mother’s papers, and they close what this site itself called “the largest single hole in this archive”. Her death certificate names her parents — JOAQUIM JÚLIO DE FIGUEIREDO and LEÓNIA DOS SANTOS MOTA — and her birth registration goes a generation further, naming all four grandparents and both her parents’ home parishes in mainland Portugal. She died at home at Gulpilhares, Vila Nova de Gaia, on 14 March 2010, aged 88. BUT THE MARGINAL NOTES ARE WHAT MATTER. IVONE MARRIED FREDERICO PINTO DA ROCHA ON 18 NOVEMBER 1939 AND WAS DIVORCED FROM HIM BY A SENTENCE OF 6 DECEMBER 1945. This archive had struck that marriage out of the family tree as correction 9500002, on one word in one column — the Germiston register of February 1948 entering her as a SPINSTER — and then built three panels on the strike: “She had never been married”, “it did not happen”, “there is no marriage record to find, because there was no marriage”. ALL OF IT IS WITHDRAWN. The same register gives her age as 22 when she was 25, and a second South African register four months later gives 24 — two declarations, two years apart, neither true. A divorcée remarrying abroad gave an office the version of herself she wanted; a registrar at Lourenço Marques did not invent a husband, a date, an office and a court sentence. THE COST WAS NOT ONLY THE MARRIAGE. Henrique, born 11 March 1941, was born INSIDE it — sixteen months after the wedding — so his surname Pinto da Rocha needs no theory of acknowledgement, and his father has a forename. Her birth year was wrong by two years, which is where the page’s own contradiction came from: it called her “an unmarried mother of sixteen” in one panel and “seventeen” in another, and she was nineteen. THE RULE THIS ARCHIVE KEEPS RELEARNING: a single field on a single form is not a finding, and the instinct to correct a family’s own record deserves a higher bar than the instinct to believe it.
Two candidate dates, and the document already excluded one of them Withdrawn
did The death certificate’s boxed day-of-month is smudged, and this archive published “the 16th or the 17th” and declined to choose. THE SEVENTEENTH WAS NEVER AVAILABLE. The same certificate carries the certifying physician’s signature dated 16 March, and a doctor cannot certify a death that has not happened yet — so one of the two readings was impossible on the page’s own evidence. Worse, the bound was written into the very paragraph that offered the pair, as the thing that “bounds it”, and was then not applied to it. The family gives the date as 14 MARCH 2010, which fits everything else: the certificate written and the whole bundle faxed two days later. THE RULE: a document’s legible dates constrain its illegible ones, and the constraint has to be applied before the candidates are published, not printed alongside them. Refusing to guess is the right instinct and it is not a substitute for reading the rest of the page.
A recollection, a paragraph built on it, and a withdrawal inside the hour Withdrawn
did Asked who travelled to Portugal when Ivone died, the family first said HENRIQUE — known as Eric — and that he passed the papers to Derrick afterwards. The declaration in the bundle names “o Sr. DERIKE”. This archive recorded the disagreement rather than resolving it, which was right, BUT ALSO WROTE A PARAGRAPH ON WHAT WOULD FOLLOW IF THE RECOLLECTION HELD — that a Henrique at Vila Nova de Gaia in 2010 would mean his estrangement from his mother had not lasted, and that the half-brothers were in contact late in life. The family then confirmed it was Derrick, and the paragraph is withdrawn. NOTHING PLACES HENRIQUE IN PORTUGAL, and the archive still knows nothing of him between Johannesburg in 1962 and his death in 2022. The argument that settled it was in the document the whole time: a clerk taking a name by ear does not turn HENRIQUE, a name he writes every week, into DERIKE — he does turn DERRICK into it. THE LESSON IS NOT “do not record family recollection”, which would be absurd in an archive built on it. It is that an inference drawn from an unresolved claim inherits the claim’s uncertainty, and this one was written in the confident voice the rest of the page uses for documents. ERIC survives as a name variant not previously held, which is all the detour left behind.
Nothing of that kind.