Apsû — the one true spelling, and the forms that fail it
The internet spells Apsû more ways than one, and only one of them is the name the sources wrote. This is the register's ruling — canonical, fallback, contested, and false, each in its proper place.
At a Glance
- Canonical form: Apsû
- Original script: 𒀊𒍪
- Pantheon: Mesopotamian
- Convention: Assyriological transliteration — š, ṣ, ṭ, ḫ as the sign lists require
- False forms on record: 1
- Contested forms: 1
- Series: The Canonical Register, No. 27 of 287
The Canonical Form
The attested form is 𒀊𒍪, from the tradition's own record. The canonical transliteration is Apsû.
Every mark is evidence, and every piece of evidence is checkable:
- u → û — Circumflex: a visible question mark — the length of Apsu's final vowel is discussable, not certain: the vowel length, as the sources give it.
Where the Flattened Form Stands
The plain form apsu is not false — it is a fallback, and the register is exact about what that means. In systems that cannot carry diacritics (legacy databases, older tooling, parts of the address-bar pipeline) the flattened form is the accepted vehicle, and the rulebook names it explicitly as the final rung of the fallback hierarchy: full restoration first, circumflex where attested, macron-only by the academic standard, an attested variant, and only then plain ASCII.
What the fallback status forbids is promotion: apsu may carry the name where marks are impossible, but it may not represent the name where marks are possible. Every mark it drops — u → û — is a fact the sources recorded. A fallback is a vehicle, never the primary.
The Register of False Forms
False form 1: Apsu
Origin: the publisher compromise — older books and hurried websites that print the "English equivalent" as if it were the name itself.
Violation: every substitution here (û→u) is a romanization of convenience, not an attested spelling — it teaches the reader to guess, not to read.
The Contested Forms
Not every non-canonical form is false. These forms carry real scholarly support — they are recorded in the lexicon as variants, with their sources, and the register presents them honestly rather than pretending unanimity:
- Abzu — Sumerian form of the primordial freshwater abyss. Support: ETCSL, Black-Green, Enuma Elish.
The canonical form stands above the contested ones here by weight of evidence, not by decree. If the scholarship shifts, the register shifts with it — that is what "canonical" means when it is honest.
The Convention's Own History
The oldest writing system on earth encodes its gods in wedge marks on clay, and its consonant system outruns the Latin alphabet: š is a hushing sibilant, ṣ and ṭ are emphatics produced deep in the mouth, ḫ is a rasp English lost centuries ago. When nineteenth-century decipherers built the transliteration conventions Assyriology still uses, they chose diacritics precisely so that none of those distinctions would be flattened. This register honors the same choice for the same reason: the tablets distinguish; so do we.
What the Address Bar Allows
IDNA 2008 permits the letters these restorations use — macrons, acutes, underdots, thorn, ogonek, the lot — and the DNS carries them as punycode (`xn--` labels). Two honest constraints remain. First, registrar inventory: the protocol permits a character; the registrar's stock decides whether the exact domain can be bought today. Second, combined marks: forms like Apṓllōn (acute stacked on a macron) are philologically ideal yet effectively untypeable on phones and unsupported across much of the registration system.
That is why the rulebook keeps a fallback hierarchy — full restoration first, then the circumflex when the source carries one, then the macron-only academic standard, then an attested variant, and only as a last resort the plain ASCII form. Every step down the ladder is still defensible; nothing on the ladder is a wrong mark. A fallback is a choice among attested forms. A false form is not on the ladder at all.
Where the Name Stood First
Archaeological evidence for the abzu centers on temple architecture and textual deposits. The abzu shrine of Enki at Eridu — among the earliest sacred structures in southern Mesopotamia — was rebuilt repeatedly from the Ubaid period onward as the cosmic freshwater source. At Babylon, Esagila's cella of Ea and its apsû-basin embodied the god's dwelling upon the primordial deep. Cuneiform copies of the Enuma Elish from Nineveh and Sippar preserve the myth of Apsû's defeat. The temple on the web is the newest of these addresses — the first one that fits in a pocket.
The Verdict
The register rules for Apsû, and the ruling is checkable: every mark sourced, every violation stated, every borderline form given its honest standing. apsu is the sanctioned fallback for systems without marks — a vehicle, never the scholarly primary. Anything else in circulation has its origin and its failure recorded above. That is the whole of the law on this name — and the temple at /sites/apsu/ enforces it.
The docket for this ruling: 1 false form ruled against, Apsu first on the docket; 1 preserved mark verified: u → û; 1 contested form recorded with its support; the attestation rests on 8 cited sources — Enuma Elish, CAD (Chicago Assyrian Dictionary), ETCSL (Electronic Text Corpus of Sumerian Literature).
The Register Continues
This is one of 287 entries in the Canonical Register — the authoritative reference for every flagship temple. The temple's founding dispatch tells the name's story; its Restoration File details the marks; its Resonance File reads the archetype into the industries. The blog index holds the whole archive, and the Rulebook states the convention this register enforces.
Continue the register — previous entry: Apóllōn (read it) · next entry: Arachnē (read it).
The Canonical Register is written from the same canonical record as the lexicon itself. If a false form is missing from this page, it is because the evidence does not support calling it false — or because we have not seen it yet. Show us, with a citation, and the register will answer.