Lexicon
The lexicon is the dictionary of the corpus: one entry per lexeme, with its citation form, its lexical description, and the occurrences that have been assigned to it. It is open to every reader.
An entry is a citation form, not an attested one. i-ša-am is not in the lexicon;
šâmu is, and i-ša-am is one of its attestations. The link between the two is an
editorial statement — see Headword or named entity?.
Lexicon
1. Reading an entry
Each entry shows:
- the citation form of the lemma, and beside it a link to the corresponding entry in the electronic Babylonian Library — the reference dictionary the project aligns to;
- the guideword: the short gloss that identifies the meaning, and that distinguishes homonymous lemmata from one another;
- language and morphology (noun, verb, adjective, preposition, and so on), and the root where one is recorded;
- the logograms by which the lexeme is written, as a row of tags: this is where the Sumerographic spellings of an Akkadian word are recorded;
- the number of attestations, and the bibliographic references attached to the entry.
The attestations open beneath the entry and lead to the texts. The list of occurrences respects your account's perimeter, so the count is the number of attestations you can see.
2. Finding an entry
Two ways in, and they combine.
The alphabet bar runs on the Assyriological alphabet, not the Latin one: ʾ and ʿ come first, Ḫ is distinct from H, Š from S, Ṣ and Ṭ have their own positions. Clicking a letter restricts the list to the lemmata that begin with it.
The filter column holds, from top to bottom:
| Filter | What it does |
|---|---|
| Search lemma | matches the citation form |
| Search logograms | matches the Sumerographic spellings recorded on the entry |
| Find word in results | highlights a string inside the entries currently listed — it does not filter |
| Lemma language | Akkadian, Hurrian, Hittite, Sumerian, West Semitic… |
| Tablet language | the language of the texts in which the lemma is attested |
| Genre | grouped into Primary (legal, administrative, letters) and Secondary (divinatory, lexical, literary, ritual) |
| Classification | words with no class, numerals, urban topography (TOP), professions and roles (WN) |
| Site | the archive the attestations come from |
Filters are cumulative and the list refreshes without reloading the page. Note the distinction between lemma language and tablet language: the first is a property of the word, the second of the document — a West Semitic lexeme attested in Akkadian texts is found by combining them.
The Classification filter is the lexical side of the class system
Only classes that belong to the lexical plane appear here — TOP and WN alongside
unclassed words and numerals. Proper names are not in this register at all: they are in
Named entities.
3. Where the entries come from
3.1 The lexicon is not maintained here
DAPCA does not author its dictionary. The entries are created and maintained by two Assyriological projects, and this platform hooks into them to retrieve what it needs:
| Project | Status |
|---|---|
| eBL — electronic Babylonian Library | the current source of the entries |
| LAD — Leipzig Akkadian Dictionary | forthcoming |
You can see the provenance in every entry. Each one keeps the identifier of its source record — the same identifier that opens the corresponding article in the eBL dictionary — together with the imported description: guide word, senses, attested forms, derivations, logograms, ORACC correspondences.
A wrong or missing dictionary entry is reported upstream, not fixed here
If a headword is wrong, a sense is missing, or a lexeme is absent altogether, the place to raise it is the source project. Correcting it locally would fix this copy and nothing else.
What belongs to DAPCA is not the entry but the link between a token and an entry — the lemmatisation, and the attestations that follow from it. That is the editorial work this section documents, and it is entirely ours.
3.2 The local record, and why editing it is discouraged
The platform does hold a full editable record for each entry, organised in four panels plus a raw view. Read it as a description of what arrives from the source, not as a form to be filled in: the panels reproduce the structure of the source record.
| Panel | What it holds |
|---|---|
| Lemma | the identity of the entry: citation form, language, morphology, root, guideword |
| Forms | the attested forms — the paradigm as documented |
| Meanings | the blocks of meaning, so that a polysemous lexeme is described sense by sense |
| Derived | derived and related lexemes, grouped |
| JSON | the underlying record as it was imported |
At present, editing any of this locally is strongly discouraged, and the reason is not pedantry:
- the change does not travel back. It reaches neither eBL nor LAD, it is invisible to everyone consulting the source, and it diverges the moment the data is realigned;
- maintaining an Akkadian dictionary is not within the scope of this project.
Three things are legitimately local, and are worth keeping apart from the rest:
- the corpus class —
TOP(urban topography) andWN(profession, role) — which is a routing decision of this platform, not a lexicographic one (why the two planes never meet); - internal notes;
- the attestations, which no other project holds.
Creating a new entry: possible, and deliberately not set up
The administrative interface offers a create button, but the record requires the source identifier, and the form does not ask for it. An entry created that way is left without one: its link to the dictionary leads nowhere, and it stands outside the alignment with eBL.
In other words the mechanism is not prepared for local authorship, which matches the policy. If a lexeme you need is genuinely absent, check it in eBL first — and if it is missing there too, that is where it should be added.
3.3 Linking occurrences — the part that is ours
The entry comes from outside; the link is yours to make. Occurrences are attached to a lemma in the Global lemmatizer, Lemmata tab, where they arrive grouped by written form and are assigned in bulk.
Two consequences:
- a lemma with no occurrences attached is simply an unused dictionary entry: it does not surface in ordinary browsing of the register until the corpus attests it;
- a lemma is never "wrong" because of its attestations. If a form has been attached mistakenly, the correction is made in the lemmatizer — with include already lemmatized occurrences enabled — and never by editing the lexical record.