About this site
This is a history of the southern Koup: the ground below the Swartberg around Klaarstroom and Prince Albert, taken up as loan farms from about 1777 and fixed as property by the grants of 15 April 1841 — and of the families who settled it, their history and the ground they held afterwards, and the roads by which they came there.
The Koup is bigger than that. The basin runs from the Swartberg north to the Nuweveldberge, with Beaufort West as its principal town and Merweville in its west. This site covers its southern rim, and grows northwards only as far as the records reach.
I have been researching my family for many years. For most of that time the material sat in separate places: a tree here, a pile of deeds there, scans in one folder and half-remembered stories in another. I always thought it would be worth mining it deeper, not just for names and dates, but for the connections between them.
That is now possible in a way it was not before. Working with AI, it has become practical to read across everything at once: to join a story to a place, a place to a deed, a deed to a person, and a person to the people standing beside them at a baptism. Some of the most interesting things on this site came out of witness lists, the social connections nobody writes down deliberately, but which are sitting in the registers waiting to be read.
How this site is put together
Everything here is one model, not a set of pages. A fact is recorded once, in one place, and every page that needs it reaches it by an id. That is why a name in a story is the same person as a name on a deed, and why the map, the timeline and the family lists cannot disagree with each other, they are three views of one set of rows.
What the site holds
| the things | rows | what one row is |
|---|---|---|
| people | 1646 | one person, with the names they were actually called as well as their baptismal name |
| parcels | 269 | one piece of ground: a farm, or a portion cut out of one. 233 are drawn on the map |
| documents | 284 | one sheet: a diagram, a deed, a death notice, an inventory |
| sources | 426 | where a document or a claim came from, with the archive reference in full |
| events | 290 | something dated that happened, from a grant to a baptism to a battle |
| places | 80 | a town, a district, a river, a field cornetcy |
| stories | 38 | a longform piece. 76 parcels have one set on or about them |
| inventories | 29 | an estate taken after a death, 218 individual goods listed |
| open questions | 194 | a research lead: what we expect, where to look, what would settle it |
And how they are joined
The joins are the point of the whole exercise. Names and dates were never the hard part; what could not be done before was reading across them.
| the joins | rows | what it connects |
|---|---|---|
| parentage | 1040 | a child to a father and a mother |
| unions | 303 | two people who married |
| ownership | 105 | a person to a piece of ground, between two dates |
| residences | 33 | a person living on ground they did not necessarily own |
| graves | 1450 | a person buried on a named piece of ground |
| witnesses | 314 | who stood at whose baptism. The quietest evidence on this site of which families were close to which |
| neighbours | 0 | a farm to the farms named on its boundary by the surveyor |
| parcel links | 130 | a portion to the farm it was cut from, or the pieces a farm was consolidated out of |
| story links | 1111 | a story to every person, farm, document and event it rests on |
| timeline | 425 | every dated row above, gathered into one chronology |
Reading a page
The ids are permanent. PER- a person, PAR- a
parcel, DOC- a document, SRC- a source, EVT- an
event, STO- a story, PLC- a place, LEAD- an open
question. They are never renumbered or reused, so a reference written down today still
resolves in ten years. Gaps in the numbering are normal.
Dates say only what the record says. 1841-04-15 is a day.
1883-10 is a month. 1821~ means about 1821, not 1821.
../1968 means before 1968. 1899/1902 is a range. Where
a source hedges, the site hedges, because flattening a hedge into a fact is the one thing
a history site must not do.
Tags mark how good a claim is. [TO VERIFY] is not
established. [FAMILY] is memory. [TREE] is the genealogy.
[FS] is FamilySearch. [DIAGRAM] is read off a survey sheet.
An untagged claim has a source cited beside it.
Every farm says why it is here. The panel traces the ground back to the three families the project is anchored on, over recorded parentage and marriage, and names every step so you can check it. Where there is no such trace it says that instead, which is the honest description of ground that came in through the land rather than through the family.
What is deliberately not shown
- Living people. Anyone who cannot be shown to have died is left off entirely. Where a transaction names them, the transaction is shown and the name is not. This is why the site cannot honestly tell you that a name or a line "died out", and it says so where the question comes up.
- Recent deeds searches. Anything after 1980 is cited but not reproduced, so current owners are not published by this site.
- Scans are gated by date. A PDF's text cannot be read by the checks that guard everything else, so only documents dated before 1936 are served as images. What has been transcribed from a withheld sheet is on its page.
The questions are part of it
A gap is not a failure to be hidden. What the record is missing lists what this model cannot answer yet, found by reading it against itself, and every research lead carries a prediction and the archive to go and look in. If you can close one, the contact form below is the place.
Please use it, and please credit the people who did the work
Anyone is welcome to use what is here. History does not belong to anyone.
But please recognise the people who did the hard work, as I try to. Almost nothing on this site is mine. It rests on the Surveyor-General's diagrams, on church registers kept by clerks whose names we do not know, on the Deeds Office, on FamilySearch and the Genealogiese Instituut, on local historians like Rose Willis, Doreen Atkinson, Helena Marincowitz and Judy Maguire, and on family members who wrote things down. Every page here lists what it rests on and links to the original wherever one exists online.
If your work is on this site and the credit is wrong, thin, or missing, tell me and I will fix it. If you would rather it were not here at all, tell me that and I will take it down.
What this is not
It is biased towards my own family, and it is not extensive. It is one family's records, not a census of the district. Where a page says something ended, or was the first, or was the only, read it as as far as the records gathered here go. There are farms in this district with no page, families with no mention, and whole decades with nothing in them. That is a gap in the collection, not in the history.
I intend to grow it. If you have something, a photograph, a deed, a family history, a correction, it is genuinely welcome.
The clearest example: the coloured community has lived in this place longer than the Boers have, and is almost absent from these pages. Part of that is what this collection is, and part of it is the paper it runs on, which is title deeds, diagrams and registers, generated by and for the people who held title. Where a document does name people they are here, and there are about six: the 1838 letter from Kouka names Piet Stuurman, a second man of the same name, Kobus Stuurman, Oranje and Dikkop, and Jan Koevies picked up the nugget that started the 1891 rush. I would welcome collaboration to make that fuller, and if it is your family, please get in touch.
Living people
People who are or may still be living are deliberately not published here. That is a rule the site enforces mechanically rather than by memory, and it fails closed: where there is no evidence that somebody is beyond privacy, they are left off.
Send me a message
Vul dit in en jou eie e-posprogram maak oop met die boodskap reg om te stuur. Niks word deur 'n derde party gestuur nie en niks word hier gestoor nie.