Method
About & method
This site exists because the antiquity of the Amazigh language is a settled question in scholarship and an unsettled one in public argument. The gap is not evidence. The gap is that the evidence is scattered across journals, museum records and constitutional texts that nobody is going to read.
What this is, and what it is not
It is a reference. It presents established evidence with its sources attached and stops there. It is not a campaign, it does not argue with anyone, and it has no position to defend beyond accuracy.
It is maintained independently by one person. It is not affiliated with any government, political party, NGO or university, it carries no advertising, and it takes no funding. There is nothing here to sell.
How sources are graded
A dated inscription and a nineteenth-century colonial historian are both sources. They are not the same kind of evidence, and a reference that presents them identically is hiding the thing you most need to know. Every entry in the bibliography therefore carries a tier:
- Primary source (6 entries). An artifact, an inscription, or an official legal text. The evidence itself, not a description of it.
- Peer-reviewed (14 entries). Research published in a journal with external review, cited to volume, page and DOI so you can check it.
- Reference work (3 entries). A standard scholarly survey or classification. Authoritative for the field's consensus, not a report of new findings.
- Secondary (2 entries). Historical, colonial-era or otherwise framed material, used only where the framing itself is the point, and labelled as such.
Citations are enforced by the build, not by discipline
Every inline citation in the prose is a reference to an entry id in the source collection. If an id does not resolve to a real entry, the site fails to build. It is not possible to publish a page here with a citation that points at nothing.
That guarantees a citation exists. It does not guarantee the source says what the sentence says, which is a different and harder problem, and the only answer to it is checking each claim against the original. That work is done in verification passes, and what has and has not been checked is recorded in each source entry.
Where the scholarship is unsettled, the page says so
Some of the relevant research is genuinely contested, and population genetics especially so. Presenting a live scholarly debate as a closed question would be the same failure as denying the evidence, just pointed the other way. Where readings compete, the page names the competing reading and cites it.
Genetics in particular cannot establish anything about a language. Languages are learned, not inherited. The population record answers a different question, and the genetics pillar is explicit about which one.
Corrections
Corrections are welcome and wanted, and the address for them is being set up. Rather than print one that bounces, this page will carry it as soon as it works. A reference that cannot be corrected is not much of a reference, so this is not a detail that will be left sitting.
Corrections are logged in the relevant source entry rather than quietly overwritten, so the record of what was wrong stays visible. A reference that silently edits its own mistakes is asking for trust it has not earned.
Reuse
Everything here is licensed CC BY-SA 4.0. Mirror it, translate it, quote it, build on it. Attribution and the same licence onward is the only condition. Translations into French, Arabic, Tifinagh and Darija are planned, and anyone who wants to work on one is welcome to get in touch.