Old Babylonian finite verb analyser v2.3

Aleksi Sahala 2012 (updated 29.09.2014)
University of Helsinki

Short feature description

Program supports all, but doubly weak quadriradicals and doubly weak triradicals verbs. All affixes are supported. Input can be given in transcription uparris or transliteration u2-par-ris. Accents are not supported *ú-par-ris. Please note that you must ignore distinctive vowel lenghts when parsing from transliteration. To do this, press F2 or write !ilv in the console.

Some symbols must be substituted

    Long vowels:         ā  ē  ī  ū    ->     aa  ee  ii  uu  
    Contracted vowels:   â  ê  î  û    ->     A   E   I   U
    Semitic consonants:  ţ  š  ş  '    ->     T   c   S   x

    Spaces are written normally: i nillik

For example, verb ušta''il must be entered as uctaxxil and ultēšibšinātima as ulteecibcinaatima.

Version history

v2.3 (24.09.2014)
- Support for some irregular verbs (izuzzum, alāku, mēlulu etc.)
- Primitive support for transliteration.
- Temporarily supports only 64-bit Windows (will be fixed).

v2.2 (1.11.2013)
- Verb variants with root vowel /e/ are now recognized properly
- Option to hide unattested analyses
- Statives added
- Precative and negatives are now recognized with statives

Lexicon update (27.11.2012)
- Download the latest lexicon: lexicon20.zip. To install it, copy and unzip it into /data folder in your Babyparser directory.

v2.1.1 (30.09.2012)
- Program is now compatible with 32-bit Windows.

v2.1 (23.09.2012)
- Lexicon finished, see its contents (here).
- Program now gives a tentative deep structure for parsed verb, e.g. /i+ndin+ki/ for iddikki

- Expanded lexicon (80% complete). Added Assyrian roots.
- Fixed formatting issues
- Fixed a bug that prevented vowel classes (?-?) from analysing correctly.
- F1 shows general help

- Expanded lexicon (60% complete).

- Entirely new method for parsing and radical recognition.
- External patterns (stems.xml)
- Support for Xt and Xtn stems in every verb class.

Missing features

- No dummy generation or vowel class ignoring (means that only roots included in lexicon20.xml are parsed).
- Hiding verbs with unattested vowel classes if input is not in D- or Š-stem.

Known bugs

- II infirmae D-stems are occasionally analysed as G-stems due to prefix allomorphy validation bug.
- Some representations of underlying morphology are inconsistent and may contain odd symbols and excess alephs.

Installing and running the analyser v2.0

To install, create a folder and unzip the package there, start the program by running analyser.bat.

Version 2.3 (64-bit Windows): BabylonianParser23.zip (4688kb)

Version 2.1 (64/32-bit Windows): BabylonianParser2.zip (4838kb)

Manual (EN): Babylonian Parser Manual (EN)

Suomenkieliset ohjeet: Manuaali (FI)

Back to index