A | B | C | D | E | F | G | H | CH | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
 | This article needs additional citations for verification. (September 2022) |
| Geʽez | |
|---|---|
| ግዕዝ Gəʽ(ə)z | |
 | |
| Pronunciation | [ˈɡɨʕ(ɨ)z] |
| Native to | Eritrea, Ethiopia |
| Extinct | Before 10th century to 14th century[1][2] Remains in use as a liturgical language.[3] |
| Geʽez script | |
| Official status | |
Official language in | Liturgical language of the Eritrean Orthodox Tewahedo Church, Ethiopian Orthodox Tewahedo Church, Eritrean Catholic Church,[3] Ethiopian Catholic Church, and Beta Israel[4] |
| Language codes | |
| ISO 639-2 | gez |
| ISO 639-3 | gez |
| Glottolog | geez1241 |
Geʽez (/ˈɡiːɛz/[5][6] or /ɡiːˈɛz/;[7][8] ግዕዝ Gəʽ(ə)z[9][10][11][12] IPA: [ˈɡɨʕ(ɨ)z] ⓘ, and sometimes referred to in scholarly literature as Classical Ethiopic) is an ancient South Semitic language. The language originates from what is now northern Ethiopia and Eritrea.
Today, Geʽez is used as the main liturgical language of the Eritrean Orthodox Tewahedo Church, the Ethiopian Orthodox Tewahedo Church, Ethiopian Catholic Church, Eritrean Catholic Church, and the Beta Israel Jewish community.
Hawulti Obelisk is an ancient pre-Aksumite Obelisk located in Matara, Eritrea. The monument dates to the early Aksumite period and bears the oldest known example of the ancient Geʽez script.
In one study, Tigre was found to have a 71% lexical similarity to Ge'ez while Tigrinya had a 68% lexical similarity to Geʽez followed by Amharic at 62%.[13] Most linguists believe that Geʽez does not constitute a common ancestor of modern Ethio-Semitic languages but became a separate language early on from another hypothetical unattested common language.[14][15][16]
Phonology
Vowels
| Front | Central | Back | |
|---|---|---|---|
| Close | /i/ i | /ɨ/ ə | /u/ u |
| Mid | /e/ e | /o/ o | |
| Near-open | /æ ~ ɐ/[a] a | ||
| Open | /a ~ ɑ/[b] ā | ||
Historically, /ɨ/ has a basic correspondence with Proto-Semitic short *i and *u, /æ ~ ɐ/ with short *a, the vowels /i, u, a/ with Proto-Semitic long *ī, *ū, *ā respectively, and /e, o/ with the Proto-Semitic diphthongs *ay and *aw.[20][21] In Geʽez there still exist many alternations between /o/ and /aw/, less so between /e/ and /aj/, e.g. ተሎኩ taloku ~ ተለውኩ talawku ("I followed").[22]
In the transcription employed by the Encyclopaedia Aethiopica, which is widely employed in academia, the contrast here represented as a/ā is represented as ä/a.
Consonants
Transliteration
Geʽez is transliterated according to the following system (see the phoneme table below for IPA values):
| translit. | h | l | ḥ | m | ś | r | s | q | b | t | ḫ | n | ʼ |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Geʽez | ሀ | ለ | ሐ | መ | ሠ | ረ | ሰ | ቀ | በ | ተ | ኀ | ነ | አ |
| translit. | k | w | ʽ | z | y | d | g | ṭ | p̣ | ṣ | ḍ | f | p |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Geʽez | ከ | ወ | ዐ | ዘ | የ | ደ | ገ | ጠ | ጰ | ጸ | ፀ | ፈ | ፐ |
Because Geʽez is no longer spoken in daily life by large communities, the early pronunciation of some consonants is not completely certain. Gragg writes that "he consonants corresponding to the graphemes ś (Geʽez ሠ) and ḍ (Geʽez ፀ) have merged with ሰ and ጸ respectively in the phonological system represented by the traditional pronunciation—and indeed in all modern Ethiopian Semitic. ... There is, however, no evidence either in the tradition or in Ethiopian Semitic what value these consonants may have had in Geʽez."[23]
A similar problem is found for the consonant transliterated ḫ. Gragg notes that it corresponds in etymology to velar or uvular fricatives in other Semitic languages, but it is pronounced exactly the same as ḥ in the traditional pronunciation. Though the use of a different letter shows that it must originally have had some other pronunciation, what that pronunciation was is not certain.[24]
The chart below lists /ɬ/ and /t͡ɬʼ/ as possible values for ś (ሠ) and ḍ (ፀ) respectively. It also lists /χ/ as a possible value for ḫ (ኀ). These values are tentative, but based on the reconstructed Proto-Semitic consonants that they are descended from.
Phonemes of Geʽez
The following table presents the consonants of the Geʽez language. The reconstructed phonetic value of a phoneme is given in IPA transcription, followed by its representation in the Geʽez script and scholarly transliteration.
| Labial | Alveolar | Palatal | Velar | Pharyngeal | Glottal | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| central | lateral | plain | labialized | |||||||
| Nasal | /m/ መ m | /n/ ነ n | ||||||||
| Stop/ Affricate |
voiceless | /p/ ፐ p | /t/ ተ t | /k/ ከ k | /kʷ/ ኰ kʷ | /ʔ/ አ ʼ | ||||
| voiced | /b/ በ b | /d/ ደ d | /g/ ገ g | /gʷ/ ጐ gʷ | ||||||
| emphatic[a] | /pʼ/ ጰ p̣ | /t’/ ጠ ṭ | /t͡sʼ/ ጸ ṣ | /t͡ɬʼ/ ፀ ḍ | /k’/ ቀ q | /kʷ’/ ቈ qʷ | ||||
| Fricative | voiceless | /f/ ፈ f | /s/ ሰ s | /ɬ/ ሠ ś | /x/ ኀ ḫ | /xʷ/ ኈ ḫʷ | /ħ/ ሐ ḥ | /h/ ሀ h | ||
| voiced | /z/ ዘ z | /ʕ/ ዐ ʽ | ||||||||
| Approximant | /r/ ረ r | /l/ ለ l | /j/ የ y | /w/ ወ w | ||||||
- ^ The emphatic consonants of Geʽez were likely realized as ejectives, as in the modern Ethiopian Semitic languages.
Geʽez consonants in relation to Proto-Semitic
Geʽez consonants have a triple opposition between voiceless, voiced, and ejective (or emphatic) obstruents. The Proto-Semitic "emphasis" in Geʽez has been generalized to include emphatic p̣ /pʼ/. Geʽez has phonologized labiovelars, descending from Proto-Semitic biphonemes. Geʽez ś ሠ Sawt (in Amharic, also called śe-nigūś, i.e. the se letter used for spelling the word nigūś "king") is reconstructed as descended from a Proto-Semitic voiceless lateral fricative . Like Arabic, Geʽez merged Proto-Semitic š and s in ሰ (also called se-isat: the se letter used for spelling the word isāt "fire"). Apart from this, Geʽez phonology is comparably conservative; the only other Proto-Semitic phonological contrasts lost may be the interdental fricatives and ghayn.
Stress
There is no evidence within the script of stress rules in the ancient period, but stress patterns exist within the liturgical tradition(s). Accounts of these patterns are, however, contradictory. One early 20th-century account[26] may be broadly summarized as follows:
- primary stress only falls on the ultima (the last syllable) or the penult (the second-to-last syllable)
- in finite verbs (including the imperative), stress falls on the penult: ቀተለት qatálat ("she killed"), ንግር nə́gər ("speak!", masculine singular), with the important exception of the 2nd-person feminine plural suffix ክን -kə́n
- in nouns and adjectives (in citation form), and most adverbs, stress falls on the ultima: ንጉሥ nəgúś ("king"), ሀገር hagár ("city"), ግዕዝ Gə́ʽz ("Geʽez"), ጠቢብ ṭabíb ("wise"), ህየ həyyá ("there"); an exception among adverbs is ዝየ zə́ya ("here")
- the suffix -a, marking the construct state or the accusative case (or both), is not stressed: ንጉሠ nəgúśa, ሀገረ hagára, ግዕዘ Gə́ʽza, ጠቢበ ṭabíba
- cardinal numbers are stressed on the ultima, even in the accusative, e.g. ሠለስቱ śalastú accusative ሠለስተ śalastá ("three")
- pronouns have rather unpredictable stress, so stress is learned for each form
- enclitic particles (such as ሰ -(ə)ssá) are stressed
- various grammatical words (short prepositions, conjunctions) and short nouns in the construct state are unstressed
As one example of a discrepancy, a different late 19th-century account[27] says the masculine singular imperative is stressed on the ultima (e.g. ንግር nəgə́r, "speak!"), and that, in some patterns, words can be stressed on the third-, fourth- or even fifth-to-last syllable (e.g. በረከተ bárakata).
Due to the high predictability of stress location in most words, textbooks, dictionaries and grammars generally do not mark it. Minimal pairs do exist, however, such as yənaggərā́ ("he speaks to her", with the pronoun suffix -(h)ā́ "her") vs. yənaggə́rā ("they speak", feminine plural), both written ይነግራ.[21]
Morphology
Nouns
Geʽez distinguishes two genders, masculine and feminine, the latter of which is sometimes marked with the suffix ት -t, e.g. እኅት ʼəxt ("sister"). These are less strongly distinguished than in other Semitic languages, as many nouns not denoting humans can be used in either gender: in translated Christian texts there is even a tendency for nouns to follow the gender of the noun with a corresponding meaning in Greek.[28]
There are two numbers, singular and plural. The plural can be constructed either by suffixing ኣት -āt to a word (regardless of gender, but often ኣን -ān if it is a male human noun), or by using an internal plural.[29]
- Plural using suffix: ዓመት ʿāmat ("year") plural ዓመታት ʿāmatāt, ገዳም gadām ("wilderness, uninhabited area") plural ገዳማት gadāmāt, ሊቅ liq ("elder, chief") plural ሊቃን liqān, ጳጳስ p̣āp̣p̣ās ("(arch)bishop") plural ጳጳሳት p̣āp̣p̣āsāt.
- Internal plural: ቤት bet ("house") plural አብያት ʾabyāt, ቅርንብ qərnəb ("eyelid") plural ቀራንብት qarānəbt.
Nouns also have two cases: the nominative, which is not marked, and the accusative, which is marked with final -a. As in other Semitic languages, there are at least two "states", absolute (unmarked) and construct (marked with -a as well).
| Singular | Plural | |||
|---|---|---|---|---|
| Absolute state |
Construct state |
Absolute state |
Construct state | |
| Nominative | ሊቅ liq | ሊቀ liqa | ሊቃን liqān | ሊቃነ liqāna |
| Accusative | ሊቀ liqa | ሊቀ liqa | ሊቃነ liqāna | ሊቃነ liqāna |
As in Classical/Standard Arabic, singular and plural nouns often take the same final inflectional affixes for case and state, as number morphology is achieved via attaching a suffix to the stem and/or an internal change in the stem.
There is some morphological interaction between consonant-final nouns and a pronoun suffix (see the table of suffix pronouns below). For example, when followed by የ -ya ("my"), in both nominative and accusative the resulting form is ሊቅየ liqə́ya (i.e. the accusative is not *ሊቀየ *liqáya), but with ከ -ka ("your", masculine singular) there's a distinction between nominative ሊቅከ liqə́ka and accusative ሊቀከ liqáka, and similarly with -hu ("his") between nominative ሊቁ liqú (< *liq-ə-hu) and accusative ሊቆ liqó (< *liqa-hu).[30][31]
Internal plural
Internal plurals follow certain patterns. Triconsonantal nouns follow one of the following patterns.
| Pattern | Singular | Meaning | Plural |
|---|---|---|---|
| ʾaCCāC | ልብስ ləbs | 'garment' | አልባስ ʾalbās |
| ፈረስ faras | 'horse' | አፍራስ ʾafrās | |
| ቤት bet | 'house' | አብያት ʾabyāt | |
| ጾም ṣom | 'fast' | አጽዋም ʾaṣwām | |
| ስም səm | 'name' | አስማት ʾasmāt | |
| ʾaCCuC | ሀገር hagar | 'country' | አህጉር ʾahgur |
| አድግ ʾadg | 'ass' | አእዱግ ʾaʾdug | |
| ʾaCCəC(t) | በትር batr | 'rod' | አብትር ʾabtər |
| ርእስ rə's | 'head' | አርእስት ʾarʾəst | |
| ገብር gabr | 'servant, slave' | አግብርት ʾagbərt | |
| ʾaCāCəC(t) | በግዕ bagʽ | 'sheep' | አባግዕ ’abāgəʽ |
| ጋንን gānən | 'devil' | አጋንንት ’agānənt | |
| CVCaC | እዝን ’əzn | 'ear' | እዘን ’əzan |
| እግር ’əgr | 'foot' | እገር ’əgar | |
| CVCaw | እድ ’əd | 'hand' | እደው ’ədaw |