Where all of this comes from.

The objects

The 41 carpets here were traced from plates in “The Azerbaijani Carpets”, published by the Heydar Aliyev Foundation. 41 of the 41 are attributed to a specific physical object — a plate, a collection and an inventory number that all point at one carpet — and 41 carry a date in the volume’s own wording.

The objects themselves are held by: The Latif Karimov Azerbaijan State Carpet and Applied Arts Museum.

Note: the museum has been renamed twice since. The name the source volume uses — the Latif Karimov Azerbaijan State Carpet and Applied Arts Museum — was in use from 1993 to 2014; in 2014 it became the Azerbaijan Carpet Museum, and since 2019 the Azerbaijan National Carpet Museum. The records here keep the volume’s name, because a record should say what it says.

What was actually made

Each carpet was redrawn as a lattice — one cell for every knot, in the dyes the weaver used — which is why it can be zoomed to the single knot and taken apart by name. Across the collection that is 14,815,696 cells and 203 named parts. The figure counts cells in our tracings, at the resolution each carpet was traced. The source records no dimensions and no knot density for these objects, so it is not a measurement of how many knots they physically carry.

Every named part is verified against the carpet it claims to come from: a run taken from the middle of each motif has to be found, character for character, in that carpet’s own grid. A motif that cannot be found in its own object fails the build.

What you are looking at

No photograph on this site is a photograph. Every carpet image is a render of our own traced grid at one pixel per knot. The museum’s plates are copyright the Heydar Aliyev Foundation and the collection that holds the objects; they were used as reference to trace from and are not republished here, in whole or in part.

This is a real constraint and it shaped the design. A museum site usually leans on macro textile photography for its sense of material. This one has no photography at all — so its material is the weave itself, which is the one thing it has that a photograph could not give it.

What this site does not claim

• The classification is one tradition’s. The eight schools, and the school given for each object, come from the Azerbaijani volume this collection is drawn from. Scholars of Caucasian weaving note that the region’s traditions were intermixed, and that assigning a village carpet to a single national tradition is often contested — Karabakh most of all. This site reports what its source records; it does not adjudicate.
• The motif readings are ours. Every object record is checked against the volume it comes from, but the volume does not discuss individual motifs. What each of the 203 parts is called here, and what its description says it does, is this project’s reading of the object. Colour claims are checked against the carpet’s own dyes on every build; where a description touches symbolism the source is largely the rug literature — trade knowledge rather than scholarship — and the text says “read as” rather than “means”.
• It is not a history. Six of the eight weaving centres here have no sourced descriptive text in this project, so the region pages carry measurements rather than accounts. Where you see a number it was counted from the objects; where you would expect a paragraph of history, there is deliberately none.
• It is not a chronology. 32 of the objects are dated to the early twentieth century. The collection has no chronological spread to draw a timeline from, and drawing one anyway would be an invention.
• Attribution is stated, not assumed. Where the volume records a carpet without a composition name, the name on the page is ours and the object’s own page says so.
• The weavers are not here. These carpets were made by people, almost all of them women, and this project holds no record of any of them. That absence is in the sources, and repeating it is better than filling it with invented figures.

Heritage status

Azerbaijani carpet weaving was inscribed on the UNESCO Representative List of the Intangible Cultural Heritage of Humanity in 2010.