The approximately 250 Afroasiatic languages, spoken by about 340 million ethnically and racially different people, occupy today the major part of the Middle East, all of North Africa, much of North-East Africa and a considerable area in what may roughly be defined as the northwestern corner of Central Africa. Though the distribution and spread of the specific languages was substantially different, about the same area was covered by Afroasiatic languages in antiquity. In the Middle Ages, Sicily and the southern half of Spain were also conquered by those who were to become the largest Afroasiatic-speaking people, the Arabs. Today, only Maltese represents this family as a native language in Europe.
The term ‘Semitic’ was proposed in 1781 for a group of related tongues, taken from the Bible (Genesis 10-11) where Noah's son Shem is said to be the ancestor of the speakers of these languages - showing, incidentally, awareness of linguistic relationships at this time. When it was realised that some other languages were further related to this group, the term ‘Hamitic’, based on the name of Shem's younger brother Ham (Cham), the biblical ancestor of Egypt and Kush, was coined for the entire family. Later the composite term Hamito-Semitic (sometimes Semito-Hamitic) was introduced. However, this created the wrong impression that there exists a ‘Hamitic’ branch opposed to Semitic. Of all the other terms proposed (Erythraic, Lisramic, Lamekhite), ‘Afroasiatic’ has been gaining ground. Even this name has the inconvenience of being misinterpreted as a group including all the languages of Africa and Asia. To dispel this, a further contraction, Afrasian, has also been used.
Afroasiatic is composed of several branches. Various proposals have been made concerning the internal relationship between the branches, but none of these subdivisions are convincing enough to be adopted. The main branches are the following.
Egyptian is the extinct language of one of the major civilisations of antiquity, that of Pharaonic Egypt (in today's Egypt, Arabic is spoken). This language can boast the longest continuous history. Its earliest documentations are from 3000 BC. From ad 300 on, the term ‘Coptic’ is used for the Egyptian idiom of mono-physite Christians. It was spoken till the sixteenth century, perhaps even later; it is still used as a liturgical language.
Semitic (see Chapter 32).
Cushitic consists of about 40 languages, spoken by perhaps 40 million people in Ethiopia, Somalia, northwestern Kenya and adjacent areas. Beja (of eastern Sudan and northern Ethiopia), with about a million Muslim speakers, has been classified as North Cushitic, but there is some likelihood that it constitutes a separate branch of Afroasiatic. Central Cushitic or Agaw used to be the major language of Ethiopia before the Semitic conquest. It has split into a number of languages and is still spoken, by few, in scattered enclaves. Rift Valley (or Highland East) Cushitic is spoken by nearly two million people around the Ethiopian Great Rift Valley. Its best known representative is Sidamo. Lowland (East) Cushitic is numerically the most important group. Among others, it comprises Afar-Saho (Dankali) along the Red Sea, Oromo (formerly Galla), spoken by 20 million or more people, Somali, the official language of the Republic of Somalia and the vehicle of about 12 million Muslims, the Dullay languages, etc. The status of South Cushitic is debated; many consider it a separate main branch, but it may also be a southern offshoot of Lowland Cushitic. The oldest Cushitic texts are from the eighteenth century. Note that the term ‘Cush’ was originally applied to an unrelated country and civilisation: Meroë.
Omotic is the name of a group of about 40 languages in the Omo Valley of southern Ethiopia, with about 1,500,000 speakers. It used to be classified as West Cushitic. Yet the great divergences led scholars to list it as a separate branch. On the other hand, since the divergences mainly consist of absence of some typical Cushitic features, Omotic may also be a simplified, pidginised offshoot of some branch of Cushitic.
Berber is a cluster of closely related yet not always mutually intelligible dialects. Once the major language of all of North Africa west of Egypt, it still has between 14 and 20 million (or even more) speakers, with the heaviest concentration in Morocco. The earliest documentation is provided by the Lybian inscriptions (the only one dated is from 139 bc). The major dialects are Tuareg, Tamazight, Tshalhit, Tirifie, Kabyle, Chawiya and Zenaga. An old consonantal alphabet, the tifinagh, has survived among the Tuareg. The extinct language of the Canary Islands, Guanche, may have also been a Berber tongue.
Chadic (see Chapter 36).
The assertion that certain languages are related means that it is assumed that they are descended from a single common ancestor. Naturally, this is not necessarily true of the speakers themselves. It often happens that the same sedentary population switches language, adopting, with a certain degree of modification, the type of speech that has been imported by a relatively small, yet dominant group of newcomers. Thus, it could be just the language that wanders, whereas the people remain stationary and only change linguistic allegiance. This explains why so many anthropological types are found in this family: the brown-skinned Mediterranean Semites, the white-skinned Berber, the black-skinned, yet in many ways still different, Cushites and Chadic speakers.
Since Semitic, a linguistically fairly homogeneous group, seems to have had its major branches already established at least 5,000 years ago, and further, taking into consideration the great internal heterogeneity of Cushitic and Chadic, the period when the putative ancestral common Afroasiatic language was spoken must be placed at a much earlier period than the previously assumed sixth millennium bc. The location of this hypothetical tongue has been assumed to have been in North Africa, perhaps in the area which is now the Sahara Desert, and the various branches must have diffused from there.
Theories have been advanced about further relationships of Afroasiatic with other languages, especially with Indo-European within a wide superfamily, Nostratic, also including Uralic, Altaic, Kartvelian, Dravidian, etc. In view of the enormous time-depth that has to be accounted for, it is extremely hard to form any critical opinion of the reconstructions proposed to support this or other such proposals.
In view of the great diversity among the branches of Afroasiatic, one should not expect many features in common that are to be found everywhere. Some such features do exist, such as gender distinction with t as a mark of the feminine, an element k as a mark of the second person, some vocabulary items such as the root *mut ‘die’. Otherwise, we have to content ourselves with features that are found in several, but not all, branches, yielding an intertwined system that ultimately makes the unity of the family quite obvious. Thus, the root *šim ‘name’ is found everywhere but in Egyptian, the prefix conjugation is attested in Semitic, Cushitic and Berber, the stative suffix conjugation in Semitic, Egyptian, Berber and possibly Cushitic, etc. Naturally, for comparative purposes, it is sufficient for an item to be attested in at least one language of a branch to be used as an isogloss, e.g. the suffix conjugation only in Kabyle within Berber, the root *mut clearly only in Rendille within Cushitic.
Because of the fact that Semitic exhibits such a great deal of regularity and also because of its being the best known branch, some of the reconstructions have been strongly inspired by phenomena of Semitic. The opposite attitude, rejecting Semitic phenomena in reconstruction in order to avoid bias, has also been seen. Other disturbing factors are: lack of knowledge of Egyptian vowels (only Coptic provides clues about them), quite recent attestation and no ancient documents of most Cushitic, Omotic and Chadic languages, contrasting with millennia-old Semitic and Egyptian data. Nevertheless, one should not dogmatically believe that older data necessarily reflect a more archaic situation. Some phenomena found in recently discovered languages may be direct survivals from the oldest times.
The following is a brief listing of linguistic features that may be original Afroasiatic.
All branches except Egyptian exhibit a special set of consonants, besides voiced and voiceless pairs, the ‘emphatic’ series, realised as pharyngealised (velarised) in Arabic and Berber, glottalised (ejective, explosive) in South Arabian, Ethiopian and Cushitic and glottalised (explosive or implosive) in Chadic; Egyptian, incidentally, also lacked voiced consonants (d stands for /t/, t for /th/, in the standard transliteration). There is evidence for several lateral consonants in Proto-Semitic; they are still used in modern South Arabian, South Cushitic and some Chadic languages (e.g. balsam ultimately comes from the Semitic root bśm where ś must have been a lateral fricative). Laryngeal sounds ‘, ḥ and x are found in Egyptian, Cushitic, Berber and Semitic. A prenasalised phoneme *mb has also been reconstructed.
The original vowel system is assumed to be long and short a, i, u, as still in Classical Semitic. Cushitic, Omotic and Chadic have tonal systems, e.g. Awngi (Cushitic, Agaw) aqá ‘(turn) into a man’, aqâ ‘I have been’ and áqâ ‘I have known’; a represents mid tone, á high tone, à low tone and â falling tone.
In the pronominal system, *an for ‘I’ in Semitic and Cushitic vs *ana:ku ‘I’ with a further velar in Egyptian and marginally in Semitic (perhaps also in the Berber suffix -γ), or ka for masculine ‘thee, thy’ in Semitic and Chadic vs ku in Cushitic and marginally in Semitic (unclear for Egyptian) with different vowels, may represent original dialectal variations in Afroasiatic. The opposition u/i for masculine/feminine, especially in third person singular pronouns, seems to be original as well: Akkadian (Semitic) šu:/ ši:, Somali (Cushitic) -uu/-ay ‘he/she’, Omotic: -o/-e gender markers in Kafa, parts of the third person singular masculine/feminine verb endings in Dizi, noun gender markers in Mubi (Chadic) (e.g. mùndúró/mìndíré ‘boy/girl’) and perhaps Egyptian -f/-s ‘his/her’ (from *hw/hy?).
In the demonstrative system the following gender-and-number markers are found: m. sg./f. sg./pl. n/t/n (Semitic, Chadic, traces in Berber), ku/ti/hu (Cushitic, also Chadic: Mubi g-/d-/h-), p/t/n (Egyptian) and for m./f. w/θ (in Berber). It is possible that both p and w come from *ku.
Two verbal conjugation systems are found in more than one branch. One, found in Semitic, Cushitic and Berber, operates with the prefixes: ?- or a- for first person singular, n- for first person plural, t- for second person and for third person singular feminine and y- for the other third persons. Further suffixes added to the second and third person plurals and, in Semitic and Beja, to the second person singular feminine make up the full conjugations. Note the homonymy of second person singular masculine and third person singular feminine. The Cushitic languages have all switched to suffix conjugations by means of prefix-conjugated postposed auxiliaries, though a few of them have maintained the original conjugation for a limited number of verbs. This suffix conjugation is not to be confused with the original Afroasiatic suffix conjugation which can be reconstructed for predicates expressing a state, rather than an action, and is attested in Semitic (with the original value in Akkadian), Egyptian, Kabyle (Berber, for predicative adjectives) and probably in Cushitic.
In spite of its absence from Egyptian, Omotic and Chadic, it is likely that the prefix conjugation harks back to Proto-Afroasiatic.
Internal inflection, i.e. internal vocalic changes within a consonantal root to express tense, mood and other categories (the root-and-pattern system) is an operative principle in Semitic (Akkadian i-prus ‘he divided’, i-parras ‘he divides’, root p-r-s), less systematically in Berber (-θ-lal ‘she (will) be born’, θ-lula ‘she was born’), in traces in Cushitic (Beja ?adanbíil ‘I collect’, ?adbìl ‘I collected’, root d-b-l). In Chadic, where the person of the subject is expressed by means of preposed particles which are very similar in shape to the oblique pronouns of other branches and where other categories like tense, mood, etc., are expressed either by elements attached to these particles or, in part at least, by the stem form of the verb, alternations like Mubi nítúwà ‘I (will) n ‘I ate’ have been considered traces of the Afroasiatic internal inflection by some scholars, while others have attributed them to independent developments. It is likely that an internal a is to be posited to mark the non-past in Afroasiatic. Internal a/u for non-past/ past is attested in Semitic, Berber and Cushitic.
The verbal derivation system plays an important part in Afroasiatic vocabulary. Verbal roots are subject to modification; new verbs are created by the addition of derivative affixes. The element s produces a causative, the addition of t or n makes the verb intransitive (passive or reflexive). Repetition of the root or part of it or mere consonantal gemination expresses repeated action. Berber: aγәm ‘to get water’, ss-iγәm ‘cause to get water’, ttuy-uγәm ‘(water) be drawn’; Beja tam ‘eat’, tamtam ‘gobble’.
Classical Semitic and Egyptian used to have a dual in their nominal system, e.g. Egyptian sn ‘brother’, sn.wy ‘two brothers’, sn.w ‘brothers’. For plural marking, several devices are found. The endings -u:/-w and -n seem to be attested all over. Repetition of the last consonant is found in Cushitic (Somali miis/miisas ‘table/tables’) and Chadic (Mubi lísí/lésas ‘tongue’). In Cushitic and Chadic, one finds singulative systems where the basic form is a collective and the addition of a suffix makes it singular, e.g. Mubi (Chadic) mándàr ‘boy(s) (in general)‘/mùndúrò ‘boy’. Yet the most interesting plural formation is what has been called the broken plural, based on internal inflection, sinn-/asna:n- in Arabic (Semitic), sini/san in Logone (Chadic) for ‘tooth/ teeth’, Xamta (Agaw, Cushitic) gezéŋ/agzéŋ ‘dog/dogs’, Berber ikәrri/akrarәn (with a further -n) ‘ram/rams’. Though the basic principle seems to be the infixation of an a, the broken plural forms cannot be predicted automatically from the singular. This is also an argument in favour of their archaic character. Thus, some form of internal inflection must have existed indeed in Afroasiatic. The Afroasiatic noun also distinguished between the genders masculine and feminine. The latter is used not only for female animates, but often also for derivatives such as diminutives, e.g. Berber axam ‘tent’ - θaxamθ ‘small tent’. Furthermore, Semitic and Cushitic have traces of polarity whereby a noun changing number may also change its gender, e.g. Sidamo (Cushitic) ko beetti ‘this boy’/te ooso ‘these boys’ vs te seemo ‘this girl’/ko seenne ‘these girls’ (m. ko, f. te).
In nominal derivation, the prefix ma- plays an important role to form agent, locative or instrumental nouns.
Classical Semitic, Egyptian and Berber are VSO languages, Cushitic is almost all SOV, while Chadic is mainly SVO. The reconstruction of Proto-Afroasiatic word order is open to speculation.
Diakonoff (1965) is a short yet highly informative comparative presentation, the best so far in the field. Hodge (1971) is a collection of chapters from the Current Trends in Linguistics series, somewhat uneven, partly inevitably obsolete, but still an important research tool. Cohen (1947) is a pioneering work in comparative Afroasiatic, but restricted to vocabulary and with only few references to Chadic. Bender (1976) provides concise yet comprehensive sketches of the structure of the major Cushitic languages, with state-of-the-art introductions, and is a ground-breaking publication. Greenberg (1963) includes a valuable discussion of Afroasiatic. Reference should be made to Hayward (2000) for an up-to-date current appraisal.
Related Credo Articles
Theodor Benfy demonstrated the relationship between Egyptian and the Semitic languages in 1844, and named the family Semito Hamitic in 1869....
(ca. 1890) : any of various groupings of non-Semitic Afro-Asiatic languages (as Berber, Egyptian, and Cushitic) that were formerly thought to make u
The Omotic group are AFROASIATIC LANGUAGES spoken in Ethiopia. They differ strongly from the Semitic languages like Amharic and the Cushitic...