--- title: "Records outlive regimes" subtitle: "The Amsterdam population register" type: essay canonical: https://offline.ltd/smaller-kingdom/records-outlive-regimes markdown_url: https://offline.ltd/docs/Records_Outlive_Regimes.md publisher: Offline.Ltd (Meanwhile) author: Twan site: https://offline.ltd topics: [privacy, population register, bevolkingsregister, Netherlands, occupation, historiography, records] summary: > A Dutch register built in 1850 for municipal administration recorded every resident's religion. Ninety years later it was read by a government nobody had planned for. This page sets out what the history does and does not support. key_claims: - "The register was built in good faith for municipal administration and social research." - "The register and the identity card made people easy to find, and were used to find them." - "How much of the difference in death rates that explains is disputed among historians." - "A record made under one set of conditions will be read under whatever conditions come later." license: Quotation permitted with attribution to Offline.Ltd (https://offline.ltd) --- # Records outlive regimes. *A register kept for ninety years of ordinary municipal business, and then read by a government nobody had planned for. This page is written carefully, because the subject demands it.* This is the oldest and heaviest example of the argument that data outlives the conditions under which it was gathered. It happened in Amsterdam. We have kept close to what historians actually say, including where they disagree with the popular version, and we have flagged the places where the well-known account is wrong. ## A thorough piece of administration By royal decree of 22 December 1849, taking effect on 1 January 1850, the Netherlands established a continuous municipal population register: not a periodic census, but a running record of everyone actually resident in each municipality, updated as people were born, moved, married and died. It held around twenty fields. Name, sex, date and place of birth, marital status, occupation, address, dates of arrival and departure, previous and subsequent residence. And, among them, each person's religious denomination. None of this was sinister. It was good administration, of the kind a well-run country is admired for. In 1938 the director of the Dutch Central Bureau of Statistics, H. W. Methorst, who had previously headed population registration, described the new and more comprehensive system then being implemented with unmistakable pride: it would follow each person "from cradle to grave", and would open "wide perspectives for simplification of municipal administration and at the same time social research". He was not wrong about any of that. He was describing a genuine improvement, in good faith, in peacetime. ## What was done with it The Netherlands was invaded in May 1940. In January 1941 the occupying administration ordered the registration of everyone of Jewish descent. Around 160,000 people were registered. Methorst's successor at the population registration office, Jacobus Lentz, built the machinery for it on the existing foundations, and designed the Dutch identity card, the persoonsbewijs, which had in fact been developed before the occupation. It was issued from April 1941 and compulsory to carry from January 1942. It was deliberately difficult to forge: special paper, inks and seals, and, crucially, a unique number that could be checked against the central registration. A convincing forgery therefore required not just a good card but a matching false entry in the administration itself. The peer-reviewed account we have relied on for this section is by William Seltzer and Margo Anderson, in "The Dark Side of Numbers: The Role of Population Data Systems in Human Rights Abuses" (*Social Research*, 2001), which sets out the sequence directly: a cradle-to-grave civil system completed before the occupation, adapted by early 1941 into special registration systems covering the Jewish and Roma populations, with those systems and the related identity cards playing an important role in apprehension before deportation. ## The raid on the register Because the register could expose forged papers, it became a military target. Not the building's occupants: the filing cabinets. On the night of 27 March 1943, a small group of resistance members, dressed in Dutch police uniforms sewn by a tailor in the group, entered the Amsterdam population registry at Plantage Kerklaan. The guards were sedated by two medical students. The group set fire to the records. It did not work as well as they hoped. Cards packed tightly into cabinets burn poorly, and the fire brigade, which deliberately took its time and used a great deal of water, added damage of a different kind. Roughly 15% of the records were completely destroyed. The Verzetsmuseum adds a further sober qualification that the popular retelling tends to leave out: by March 1943 most of Amsterdam's Jewish population had already been deported. Around twenty-one people were arrested within weeks, after a betrayal. Twelve were executed by firing squad in the dunes near Overveen on 1 July 1943, among them the painter and writer Willem Arondeus, who had helped lead the raid, and Sjoerd Bakker, the tailor who made the uniforms. A separate archive, the central duplicate register in The Hague, was destroyed by an RAF air raid in April 1944, by which time most Dutch Jews had already been deported. Shortly before his execution, Arondeus is reported to have asked that it be made known that homosexuals were no less courageous than anyone else. The remark reaches us second hand, through the resistance lawyer Lau Mazirel, and was first published in 1970; the exact wording is not documented, and Dutch accounts render it closer to "homosexuals are not by definition weaklings". We note the uncertainty rather than smoothing it over. ## What this does not prove Here we have to be careful, because there is a version of this story circulating that goes further than the evidence, and we would rather lose the rhetorical force than repeat it. Around 73% of Jews in the Netherlands were killed, compared with roughly 40% in Belgium and 25% in France. The comparison is real, and it is sometimes attributed directly to the quality of Dutch record-keeping. That attribution is contested, and the leading comparative scholarship does not make it. In the standard comparative study of the persecution in the Netherlands, France and Belgium, Pim Griffioen and Ron Zeller foreground quite different factors: the Netherlands had a civil rather than military occupation regime, in which the German police apparatus had earlier and greater freedom of action; the reorganised Dutch police were effectively subordinated to the occupier; deportation proceeded through deception and graduated exemptions rather than public violence, which discouraged early flight into hiding; the Jewish Council was in a weak and formalised position; and Dutch Jewry was largely assimilated and native, which made pursuing legal deferment look rational. Registration does not appear in their causal list. Marnix Croes has raised a further difficulty: survival rates varied sharply between Dutch regions and municipalities, and a nationally uniform register cannot explain variation within the country. So the defensible statement, and the only one we will make, is this. **The register and the identity card made people easy to find, and were used to find them. How much of the difference in death rates that explains is disputed, and the historians who have studied the comparison most closely put more weight on the structure of the occupation and the conduct of the police.** ## The lesson that does hold What survives all of that qualification is simple, and it is the reason this page exists. The register was not built by bad people for bad reasons. It was built by competent civil servants for municipal administration and social research, and its designer described it in 1938 in the language of progress, because that is what it was. Nobody who filled in the religion column in 1935 was doing anything wrong, and nobody who answered the question was being careless. The data did not change. The government did. That is the whole of it, and it generalises without needing any of the horror to be repeated. A record made under one set of conditions will be read under whatever conditions come later. You cannot know what those are. Nobody in 1850, or 1938, could have been expected to. And the only records that are safe from a future you cannot forecast are the ones that were never made. ## Where to read more - [The Smaller Kingdom](https://offline.ltd/smaller-kingdom): the whole argument this page belongs to. - [Consent does not travel](https://offline.ltd/smaller-kingdom/consent-does-not-travel): the modern version, data outliving the company that gathered it. - [Anonymous does not stay anonymous](https://offline.ltd/smaller-kingdom/anonymous-does-not-stay-anonymous): why de-identified data keeps being re-identified. ## Frequently asked **What was the Dutch bevolkingsregister?** A continuous municipal population register established by royal decree in December 1849 and running from 1 January 1850. It recorded around twenty details for every resident, including name, address, family relationships, occupation and religious denomination, and was updated continuously rather than collected periodically like a census. **What happened at Plantage Kerklaan in March 1943?** On the night of 27 March 1943 a small group of resistance members in Dutch police uniforms entered the Amsterdam population registry and set fire to the records, because forged identity papers could be exposed by checking them against the register. Roughly 15% of the records were completely destroyed. Twelve of those arrested afterwards were executed near Overveen on 1 July 1943. **Did the population register cause the high death rate among Dutch Jews?** This is disputed. The register and the identity card made people easier to identify and apprehend, and were used for that. But the leading comparative historians of the persecution in the Netherlands, Belgium and France emphasise other factors, principally the civil occupation regime, the subordination of the Dutch police, deportation by deception, and the late development of hiding networks. Regional variation in survival within the Netherlands also counts against a purely registry-based explanation. **Who was Willem Arondeus?** A Dutch painter and writer who helped lead the March 1943 attack on the Amsterdam population registry. He was arrested days later and executed near Overveen on 1 July 1943. He was recognised by Yad Vashem as Righteous Among the Nations. **What is the lesson of the population register for data today?** That records are read under the conditions that exist when they are read, not the conditions under which they were gathered. The register was built for ordinary municipal administration by people acting in good faith. What changed was not the data but the government holding it. --- *From Offline.Ltd, Amsterdam. The web version of this page lives at https://offline.ltd/smaller-kingdom/records-outlive-regimes and is kept current there.*