AKK: Oracc Linguistic Annotation for Akkadian

This document provides an overview of language-specific annotation conventions for Akkadian used in Oracc. We focus here on the data-entry view of linguistic annotation giving only enough additional technical background to ensure that correct annotation of ATF files can be carried out. This document assumes knowledge of the introductory page on Oraccc linguistic annotation.

General principles

If you haven't done so already, start by reading some or all of the page on Oracc Linguistic Annotation page.

Lemmatisations take one of two forms:

The main thing to remember is that you are not lemmatising for other people but for the computer, and that the computer has no intuition. Inconsistencies produce duplicate entries in the glossaries, which are fiddly and time-consuming to fix.

After establishing some general principles, we describe in more detail how to manage the different components of the lemmatisation entry. At the end we show you how to deal with compound orthographic forms (crasis and logograms which straightforwardly resolve into two or more Akadian words), phrasal semanitic units (multi-word idioms), and proper nouns.

Transliteration Stylesheet

Unless your project is primarily working with legacy data, you should follow the Oracc stylesheet for Akkadian transliteration. This will save you a lot of work by enabling your project to use a related, more mature project's glossaries as the starting point for your lemmatisation, such as CAMS [http://oracc.museum.upenn.edu/cams/gkab/]. But this will only work optimally if you use these same spelling and ATF formatting conventions.

Good work habits

Check automatically entered lemmatisations for:

If you need to change either of them, use the long form of lemmatisation.

Sometimes the lemmatiser will make several suggestions for alternative lemmatisations, like this: pān[front]N|immar[see]V|innamir[appear]V|igi[reciprocal]N for the transliteration IGI. Just delete all but the relevant lemmatisation (making sure you delete all the |'s too).

When adding new lemmatisations check whether the word already exists in the online glossary for your project:

Citation Forms (CF)

We call the "dialect" of CFs "Conventional Akkadian" (CA) because it does not consistently represent any actually attested variety of the language. It is essential to follow the Concise Dictionary of Akkadian for CFs, even when it differs from CAD. This is important for inter-project compatibility and the ePAD.

However, there are some small but consistent ways in which Conventional Akkadian diverges from CDA's conventions:

Guide Words (GW) and Senses

Generally, we use the first translation in CDA as the GW. So when there are alternatives separated by commas, we use the first one, e.g., CDA's adirtu, "fear, apprehension" becomes +adirtu[fear]N.

Exceptions include:

If you want to add a local meaning (sense), follow the GW with // and then the sense, e.g. +hiāṭu[supervise//survey]V'V$hīṭma. List a sense only when there is significant variation from the GW. For a verb with a G stem GW, list the meaning of another verbal stem as a subsense only if there is significant variation in meaning, e.g., do not list an N stem passive meaning as a subsense if it is a simple passive form of the G stem active GW.

Parts of Speech (POS and EPOS)

Whenever you add a SENSE you also need to specificy a new, locally valid POS for a word, even if it is the same as that for the GW. We call this an EPOS (Effective Part of Speech) and mark it immediately after the POS, like this: +kī[like//when]PRP'SBJ$.

Here are some useful rules about POS and EPOS:

Other important points to note:

Normalisations (NORM0)

Difficult words

It's not always possible to fully lemmatise every word in a text.

Marking sentences

Insert +. in the lemmatisation line to mark sentences, not clauses - that is, wherever you judge the translation needs a full stop. A good rule of thumb is that +. follows indicative verbs that lack the -ma suffix. Its primary aim is to allow for transliteration-translation sentence-alignment in due course. It is not necessary (or perhaps is pointless to try and add it) in fragmentary texts. It is preceded and (unless at the end of a line) followed by white space.

Compound Orthographic Forms (COF)

COFs are forms that are written as one word but in fact should be understood as more than one lemma. Their meanings are transparent and they do not justify a separate entry in the glossary.

Phrasal Semantic Units (PSU)

A PSU is a multi-word idiom, whose meaning cannot be inferred from the lemmatisation of its component parts. Each PSU needs its own, manually entered glossary entry. We try to keep PSUs to a minimum.

Proper Nouns

See the page on Linguistic Annotation of Proper Nouns for more information.

POS Tags

The following lists of pronouns and function words are exemplary rather than exhaustive; the POS categorizations are based on W. von Soden, Grundriss der akkadischen Grammatik, 3rd edition (Analecta Orientalia 33). Biblical Institute Press, 1995.

BASIC WORD CLASSES
AJadjective (including statives)
AVadverb
Nnoun (including statives)
NUnumber
Vverb (including infinitives, marked with EPOS 'N and gerundive GW)

Adverbs include temporal adverbs (e.g., inanna[now], anumma[now//herewith], šattišam[yearly]), interrogative adverbs (e.g., ali[where?], ammīni[why?], kī[how?], mati[when?]), and demonstrative adverbs (e.g., kīam[thus]).

PRONOUNS
DPdemonstrative pronoun
annû[this]
ullû[that]
IPindependent/anaphoric pronoun
nom.acc./gen.dat.
anāku[I]yâti[me]yâšim[to me]
attā[you]kâta[you]kâšim[to you]
attī[you]
šū[he//it]šuāti[him//it]šuāšim[to him//to it]
šī[she//it]šiāti[her//it]šiāšim[to her//to it]
nīnu[we]niāti[us]niāšim[to us]
attunu[you]kunūti[you]kunūšim[to you]
attina[you]kināti[you]kināšim[to you]
šunu[they]šunūti[them]šunūšim[to them]
šina[they]šināti[them]šināšim[to them]
PPpossessive pronoun
yû[mine]
kû[yours]
šû[his//hers//its]
nû[ours]
kunû[yours]
šunû[theirs]
QPinterrogative pronoun
ayyû[which?]
mīnu[what?]
mannu[who?]
RPreflexive/reciprocal pronoun
ramānu[self//own]
ahāmiš[one another]
XPindefinite pronoun
ayyumma[whoever//whichever]
mamman[someone//anyone]
mimma[something//anything]
RELrelative pronoun
ša[that]
šūt[that]
šīt[that]
mala[as much as]
mimmû[all that]
DETdeterminative pronoun
ša[of]
šūt[of]
šīt[of]
OTHER PARTS OF SPEECH
CNJconjunction
lū[either]
u[and]
ū[or]
Jinterjection
anna[yes]
ai[alas]
ullu[no]
MODmodal, negative, or conditional particle
ai[not]
lā[not]
lu[either]
lū[indeed]
lū[may]
šumma[if]
ul[not]
PRPpreposition
adi[until]
ana[at//to]
aššu[about]
balu[without]
eli[on//over]
ina[in//from]
ištu[from]
itti[with]
kī[like]
kīma[like//instead of]
lāma[before]
mala[as much as]
qadu[with]
SBJsubjunction
adi[until]
ašar[where]
aššu[because]
ēma[wherever//whenever]
inūma[when]
ištu[since]
kī[when]
kīma[as]
lāma[before]
warka[after]

Roots

If you want to assign roots to Akkadian content words (nouns, adjectives, verbs and adverbs), you must do this in your project's Akkadian glossary (the akk.glo file). See the Projects and Emacs page for more information.

There is a list of Akkadian roots on a separate page.

Dialects

You must always explicitly mark the dialects of Akkadian used in your texts. You may also wish to use dialect forms as the headwords of your glossary.

23 Jul 2014 osc at oracc dot org

Eleanor Robson

Eleanor Robson, 'AKK: Oracc Linguistic Annotation for Akkadian', Oracc: The Open Richly Annotated Cuneiform Corpus, Oracc, 2014 [http://oracc.museum.upenn.edu/doc/help/languages/akkadian/]

 
Back to top ^^
 

Released under a Creative Commons Attribution Share-Alike license 3.0, 2014. [http://www.facebook.com/opencuneiform] [http://oracc.blogspot.com] [http://www.twitter.com/oracctivity]
Oracc uses cookies only to collect Google Analytics data. Read more here; see the stats here [http://www.seethestats.com/site/oracc.museum.upenn.edu]; opt out here.

http://oracc.museum.upenn.edu/doc/help/languages/akkadian/