The Berber Latin alphabet (Berber languages: script=Latn|Agemmay Amaziɣ Alatin) is the version of the Latin alphabet used to write the Berber languages. It was adopted in the 19th century, using varieties of letters.
The Berber languages were originally written using the ancient Libyco-Berber script and then centuries later by the Tuareg Tifinagh script in Tuareg language areas, of which the Neo-Tifinagh alphabet/abjad is the modern development.
The use of a Latin script for Berber has its roots in European (French and Italian) colonial expeditions to North Africa. Dictionaries and glossaries written with Latin letters, ordered alphabetically and following European orthography (mainly French) began to appear in print in the 19th century, they were intended to the colonial administration, traders and military officers. With the arrival of linguists specialized in Semitic languages there emerged a system based on Semitic romanization conventions: diacritics were used, and dictionary entries were now ordered by root. This system has since become the most common way of Berber transcription in scientific documents and literature.
Various writing standards were used since the 19th century, some are phonetically oriented, other phonologically oriented. While the Tuareg languages use a phonetically oriented transcription, the northern Berber languages use on the other hand a mixed transcription, the latter is recommended by the French institute of languages, INALCO and has been adopted by the HCA in Algeria and IRCAM in Morocco (although in Neo-Tifinagh).
The Berber Latin alphabet of Northern-Berber usually consists of 34 letters:
| The 34-Letter Alphabet of Northern-Berber | ||||||||||||||||||||||||||||||||||
| A | B | C | Č | D | Ḍ | E | Ɛ | F | G | Ǧ | Ɣ | H | Ḥ | I | J | K | L | M | N | Q | R | Ř | Ṛ | S | Ṣ | T | Ṭ | U | W | X | Y | Z | Ẓ | |
| Lower case | ||||||||||||||||||||||||||||||||||
| a | b | c | č | d | ḍ | e | ɛ | f | g | ǧ | ɣ | h | ḥ | i | j | k | l | m | n | q | r | ř | ṛ | s | ṣ | t | ṭ | u | w | x | y | z | ẓ | |
In Northern-Berber texts, foreign words and names are written in their original form even if they contain the letters: O, P, V, or any other non-Berber letter (like Ñ). According to SIL, the letter P is used in Kabyle.
The following table shows the Northern-Berber Latin alphabet with its Neo-Tifinagh[1] and Arabic equivalents:
| Berber-Latin | IRCAM's Tifinagh equivalent | Arabic equivalent | IPA equivalent | Similar sound in other languages | ||
|---|---|---|---|---|---|---|
| 1 | A a | Berber languages: ⴰ | أ / ا / َ | pronounced as /link/ | By default like English a in "map". When there is an emphatic Berber consonant then the Berber "a" is pronounced like the English a in "car". | |
| 2 | B b | Berber languages: ⴱ | ب | pronounced as /link/ or pronounced as /link/ | English b or a soft Spanish b / v | |
| 3 | C c | Berber languages: ⵛ | ش | pronounced as /link/ | English sh in "ship" | |
| 4 | Č č (tc) | Berber languages: ⵞ | تش | pronounced as /link/ | English ch in "China" | |
| 5 | D d | Berber languages: ⴷ | د / ذ | pronounced as /link/ or pronounced as /link/ | English d as well as English th in "this" | |
| 6 | Ḍ ḍ | Berber languages: ⴹ | ض / ظ | pronounced as /link/ | Thick English d in "dick" or "door". Emphatic d | |
| 7 | E e | Berber languages: ⴻ | none | pronounced as /link/ | English unstressed a in "attack". | |
| 8 | Ɛ ɛ | Berber languages: ⵄ | ع | pronounced as /link/ | ‘ayn (voiced equivalent of ḥ, similar to English onomatopoeia for retching) | |
| 9 | F f | Berber languages: ⴼ | ف | pronounced as /link/ | English f | |
| 10 | G g | Berber languages: ⴳ | (گ) | pronounced as /link/ | English g in "gate" or "gold" | |
| 11 | Ǧ ǧ (dj) | Berber languages: ⴵ | ج | pronounced as /link/ | English j in "joke" or English g in "George" | |
| 12 | Ɣ ɣ (gh) | Berber languages: ⵖ | غ | pronounced as /link/~pronounced as /link/ | like French / German r or between this sound and Spanish intervocalic g | |
| 13 | H h | Berber languages: ⵀ | هـ | pronounced as /link/ | Strong English h in "hello" or "hold" | |
| 14 | Ḥ ḥ | Berber languages: ⵃ | ح | pronounced as /link/ | Arabic ḥ in Muḥammad (stronger than h, similar to English onomatopoeia for being cold) | |
| 15 | I i | Berber languages: ⵉ | ي / ِ | pronounced as /link/ | English ee like in "sheet" or English i in "hidden". | |
| 16 | J j | Berber languages: ⵊ | (ج) | pronounced as /link/ | English s in "measure" or "television", or English j in "déjà-vu". | |
| 17 | K k | Berber languages: ⴽ | (ک) | pronounced as /link/ | English k | |
| 18 | L l | Berber languages: ⵍ | ل | pronounced as /link/ or pronounced as /link/ | usually British English clear L in "light" (as in French, Spanish, German) | |
| 19 | M m | Berber languages: ⵎ | م | pronounced as /link/ | English m | |
| 20 | N n | Berber languages: ⵏ | ن | pronounced as /link/ | English n | |
| 21 | Q q | Berber languages: ⵇ | ق | pronounced as /link/, pronounced as /link/ or pronounced as /link/ | like k, but deeper in the throat | |
| 22 | R r | Berber languages: ⵔ | ر | pronounced as /link/, pronounced as /link/ | Spanish or Italian r | |
| 23 | Ř ř | Berber languages: ⵔ | ر | pronounced as /link/ | Soft Italian/Spanish r, but even softer, almost l | |
| 24 | Ṛ ṛ | Berber languages: ⵕ | ڕ | pronounced as /link/ | Thick emphatic Spanish r | |
| 25 | S s | Berber languages: ⵙ | س | pronounced as /link/ | English s in "seed" | |
| 26 | Ṣ ṣ | Berber languages: ⵚ | ص | pronounced as /link/ | Thick English s as in "sold". Emphatic s | |
| 27 | T t | Berber languages: ⵜ | ت / ث | pronounced as /link/ or pronounced as /link/ | English t in "tea", as well as English th in "thought" | |
| 28 | Ṭ ṭ | Berber languages: ⵟ | ط | pronounced as /link/ | Thick English "t" in "toll". Emphatic t | |
| 29 | U u | Berber languages: ⵓ | و / ُ | pronounced as /link/ | English u in "put" or "rule" | |
| 30 | W w | Berber languages: ⵡ | وْ | pronounced as /link/ | English w | |
| 31 | X x | Berber languages: ⵅ | خ | pronounced as /link/~pronounced as /link/ | German / Dutch ch in "Nacht", European Spanish j | |
| 32 | Y y | Berber languages: ⵢ | يْ | pronounced as /link/ | English y in "yes" or "yard" | |
| 33 | Z z | Berber languages: ⵣ | ز | pronounced as /link/ | English z in "zoo" | |
| 34 | Ẓ ẓ | Berber languages: ⵥ | (ژ) | pronounced as /link/ | Thick English z in "Zorro". Emphatic z | |
The letter "O" does occur often in Tuareg-Berber orthography and sometimes in Northern Berber. In Northern-Berber orthography it usually corresponds to the letter "U".
In the interest of pan-dialectal legibility, the Berber Latin alphabet omits the partly phonemic contrasts found in some Berber language varieties (notably the Kabyle language and Riffian Berber) between stops and fricatives.[2]
Phonemic labiovelarization of consonants is widespread in Berber varieties, but there are rarely minimal pairs and it is unstable (e.g. ameqqʷran "large", in the Ainsi dialect of Kabyle, is pronounced ameqqran in At Yanni Kabyle-Berber, only a few kilometers away).[3] The INALCO standard uses the diacritic (ᵒ) for labiovelarization only when needed to distinguish words, e.g. ireggel vs. ireggᵒel.
| Tifinagh equivalent | IPA equivalent | ||
| Bʷ bʷ / Bᵒ bᵒ | Berber languages: ⴱⵯ | pronounced as /bʷ/ | |
| Gʷ gʷ / Gᵒ gᵒ | Berber languages: ⴳⵯ | pronounced as /ɡʷ/ | |
| Ɣʷ ɣʷ / Ɣᵒ ɣᵒ | Berber languages: ⵖⵯ | pronounced as /ɣʷ/ | |
| Kʷ kʷ / Kᵒ kᵒ | Berber languages: ⴽⵯ | pronounced as /kʷ/ | |
| Qʷ qʷ / Qᵒ qᵒ | Berber languages: ⵇⵯ | pronounced as /qʷ/ | |
| Xʷ xʷ / Xᵒ xᵒ | Berber languages: ⵅⵯ | pronounced as /xʷ/ |
The letter (ṛ) is used for pronounced as /[rˤ]/ only when it contrasts with (r) (e.g. ṛwiɣ "I am satisfied" vs. rwiɣ "I am moved"). In all other cases (r) is used, e.g. tarakna "carpet" (pronounced taṛakna). This is because pronounced as /[rˤ]/ is often an allophone of pronounced as //r// in the environment of other emphatics, and it rarely contrasts with pronounced as //r// otherwise.[4] Exceptional cases of other emphatics, e.g. pronounced as /[ʊʃˤːæj]/ "hound", are ignored (i.e. written as uccay).
In most Riffian areas (northern Morocco), the letter "L" in the word alɣem is pronounced [ařɣem]. "Ř" is pronounced as something between "L" and "R".pronounced as /needed/[5]
Riffian Berbers pronounce the "LL" (in a word like yelli, "my daughter") like "dj" or "ǧǧ" (yedji). Depending on the author's whim, this might be represented in writing as "ll", "dj", a single "ǧ", or "ǧǧ".
| Riffian letter | Riffian word | The word in other Berber dialects | meaning in English |
|---|---|---|---|
| Ř ř | uř | ul | heart |
| aɣyuř | aɣyul | donkey | |
| awař | awal | speech / talk | |
| Ǧ ǧ | azeǧif | azellif | head |
| yeǧa | yella | (he) is / (he) exists | |
| ajeǧiđ | agellid | king | |
| Č č | wučma | weltma | my sister |
| t̲acemřač | tacemlalt | white | |
| t̲aɣyuč | taɣyult | female donkey (jenny) |
In Souss (mid-southern Morocco), Berber writers rarely use the neutral vowel "e", because the unphonemic schwa is rarer in Tachelhit due to a different stress system than its sister languages.[6]
In Kabyle-Berber (northeastern Algeria), the affricates pronounced as //t͡s, d͡z// have traditionally been notated as (ţ, z̧) for over thirty years. However these affricates are uncommon in other dialects (except in Riffian) and they are morphologically conditioned, so for the sake of pan-dialectal legibility the INALCO standard omits them.[7] In Kabyle the affricate pronounced as /link/ may derive from underlying pronounced as //tt// or pronounced as //ss//. In the former case the INALCO standard uses (tt), and in the second it uses (ss) (e.g. yettawi vs. ifessi deriving from the verb fsi).
| INALCO equivalent | IRCAM Tifinagh equivalent | IPA equivalent | Pronunciation | ||
| Ţ ţ | Tt tt | Berber languages: ⵜⵙ | pronounced as /link/ | ts like in "Tsetse fly" | |
| Ss ss | |||||
| Z̧ z̧ | Zz zz | Berber languages: ⴷⵣ | pronounced as /link/ | dz / the English "ds" in words |
Labiovelarization is indicated with the superscript letter (ʷ) (examples: kʷ, gʷ), or with the "degree sign": "ᵒ" (examples: kᵒ, gᵒ), or simply by using the letter (w). (ḇ ḏ ǥ ḵ ṯ) may represent spirantization.
On the internet, it is common to replace the Latinized Greek epsilon and gamma, (Ɛɛ) and (Ɣɣ), with actual Greek letters:[8]
Among non-Kabyle Berber writers a number of alternative letters are used:
| INALCO equivalent | ||
| Â â | Ɛ ɛ | |
| Ġ ġ | Ɣ ɣ | |
| Gh gh | ||
| Dj | Ǧ ǧ |
There has been a long and fierce debate on whether to use the Latin, Tifinagh, or Arabic alphabets for Berber in Algeria and Morocco, between Berber activists and anti-Berber establishments, mainly those with an Arab-Islamic orientation. Berber activists overwhelmingly favor the use of the Latin alphabet in order to ensure a quick development and proliferation of the Berber language (Tamazight) in schools, in public institutions, and on the internet.[9] A small number of them prefer the Neo-Tifinagh alphabet. The states of Morocco and Algeria usually distance themselves from Latin-based Berber writing, fearing that it would strengthen the position of Berber against Arabic and French, and thus leading to a stronger Berber political activism. The Arab-Islamic establishments and political parties often reject the Latin alphabet as a Berber alphabet for the same reasons, and they usually brand it as a tool to westernize and Christianize Berbers.[10]
In 2003, Mohammed VI of Morocco approved the Royal Institute of the Amazigh Culture (IRCAM) Berber Institute's decision of using Neo-Tifinagh as the sole official alphabet for the Berber language in Morocco. The IRCAM's decision was met with much disapproval among independent Berber activists and they saw it as a way of neutralizing Berber and preventing it from quick flourishing and development.
The Southern-Berber (Tuareg) Latin alphabet is made of 36 letters. They are mostly Latin letters with one IPA character and one Greek letter incorporated.
The vowel O is used in the Latin alphabet of Southern Berber (Tuareg), but is also used in some (but not all) Northern Berber languages. The vowel "O" in Tuareg words mostly corresponds to "U" in Northern Berber words.
| A | Ă | B | Ḅ | D | Ḍ | E | Ǝ | F | G | Ɣ | H | Ḥ | I | J | K | L | Ḷ | M | N | Ŋ | O | Q | R | S | Ṣ | Š | T | Ṭ | U | W | X | Y | Z | Ž | Ẓ | Γ |
| a | ă | b | ḅ | d | ḍ | e | ǝ | f | g | ɣ | h | ḥ | i | j | k | l | ḷ | m | n | ŋ | o | q | r | s | ṣ | š | t | ṭ | u | w | x | y | z | ž | ẓ | ʕ |
| 37-Letter Latin alphabet for Tuareg-Berber (Tamahaq), official in Niger since 1999 | ||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A | Ă | Ǝ | B | C | D | Ḍ | E | F | G | Ǧ | H | I | J | J̌ | Ɣ | K | L | Ḷ | M | N | Ŋ | O | P | Q | R | S | Ṣ | Š | T | Ṭ | U | W | X | Y | Z | Ẓ |
| a | ă | ǝ | b | c | d | ḍ | e | f | g | ǧ | h | i | j | ǰ | ɣ | k | l | ḷ | m | n | ŋ | o | p | q | r | s | ṣ | š | t | ṭ | u | w | x | y | z | ẓ |
The Malian national literacy program DNAFLA has proposed a standard for the Latin alphabet, which is used with modifications in Karl G. Prasse's Tuareg French Dictionary and the government literacy program in Burkina. In Niger a slightly different system was used. There is also some variation in Tifinagh and in the Arabic script.
The DNAFLA system is a somewhat morphophonemic orthography, not indicating initial vowel shortening, always writing the directional particle as (dd), and not indication all assimilations (e.g. (Tămašăɣt) for tămašăq.
In Burkina Faso the emphatics are denoted by "hooked" letters, as in Fula, e.g. (ɗ ƭ).