BFFs: Byforms

BFFs are the Oracc glossary mechanism for word byforms of various kinds. The general mechanism allows the individual byforms to be treated under their own entries within the glossary and during lemmatization, but then be kept separate or grouped together when the glossaries are rendered.


BFFs are implemented in glossaries using the @bff tag; this should be read as 'ByForm-oF', since BFFs are only given under the entries of byforms, not of the main form. The choice of main form depends on the language and the kind of byform.

The @bff tag's syntax consists of two required and two optional arguments. The required arguments are:

A single word which indicates which kind of byform this is; the names of the classes are language-specific.
A glossary link which must resolve to an @entry in the glossary. Note that the @bff linkage points only from the byforms to the main form; it is an error for a word named as an OWNER in a @bff tag to have its own @bff tags. The link has the same form as other glossary entry links: <CF [GW] POS>. So, to link to ki aŋ, love, the OWNER argument would be: <ki aŋ [love] V/t>.

The optional arguments are:

A human-readable label which may be rendered in the glossary and is intended to convey information about the byform to readers. This may contain spaces and must be given in double quotes.
A machine-readable code which is stored in the XML version of the glossary and is intended for use in machine-processing; may not contain spaces.

The required order of arguments (with optional arguments indicated with a question-mark after them) is:


Sumerian BFFs

Main Forms

The main forms in Sumerian BFFs depend on the CLASS of the BFF and are described under Classes below.


Suppletive byform: used for suppletive verbs. The main form is the entry that is used for Perfective/Singular/Subject/Human.

Compound byform: used for collapsed compounds of the type mu-un-si-sa₂ instead of expected si mu-un-sa₂. With this schema, there is an entry for the byform sisa which refers back to the main form si sa:

  @entry sisa [set straight] V/t
  @bff COMP <si sa [make straight] V/t>
  @bases si-sa₂
  @form mu-un-si-sa₂ /si-sa₂ #mu.n:~
  @sense V/t set straight
  @end entry

The main form is the uncollapsed version of the compound.

Grammatical byform: used primarily for frozen verbal forms which exhibit variation in morphological markers. The main form is a matter of choice rather than rule, but should be a widely attested or earliest-attested form.


Labels are not used for Sumerian; they are constructed from the CODE argument


Sumerian codes are used only for suppletive verbs. They are constructed of hyphen-delimited quadruples giving the following information, in the following sequence:

P (perfective) or I (imperfective)
Sg (singular) or Pl (plural)
S (subject) or O (object)
H (human) or N (nonhuman)

If the selection criterion is not relevant to the suppletion, an asterisk (*) is used. Thus, typical suppletion codes include:



18 Dec 2019 osc at oracc dot org

Steve Tinney

Steve Tinney, 'BFFs: Byforms', Oracc: The Open Richly Annotated Cuneiform Corpus, Oracc, 2019 []

Back to top ^^

Released under a Creative Commons Attribution Share-Alike license 3.0, 2014. [] [] []
Oracc uses cookies only to collect Google Analytics data. Read more here; see the stats here []; opt out here.