Datasets:
ShikharLLM
/
llm2

Dataset card Data Studio Files Community
text
stringlengths
0
51.6k
Greenberg relied on his own method of mass comparison of vocabulary items rather than the comparative method of demonstrating regular sound correspondences to establish the family. An alternative classification, based on the pronominal and conjugation systems, was proposed by A.N. Tucker in 1967. As of 2023, widely accepted sound correspondences between the different branches have not yet been firmly established. Nevertheless, morphological traits attributable to the proto-language and the establishment of cognates throughout the family have confirmed its genetic validity.
Origin
Timeline
There is no consensus as to when Proto-Afroasiatic was spoken. The absolute latest date for when Proto-Afroasiatic could have been extant is , after which Egyptian and the Semitic languages are firmly attested. However, in all likelihood these languages began to diverge well before this hard boundary. The estimations offered by scholars as to when Proto-Afroasiatic was spoken vary widely, ranging from 18,000BCE to 8,000BCE. An estimate at the youngest end of this range still makes Afroasiatic the oldest proven language family. Contrasting proposals of an early emergence, Tom Güldemann has argued that less time may have been required for the divergence than is usually assumed, as it is possible for a language to rapidly restructure due to areal contact, with the evolution of Chadic (and likely also Omotic) serving as pertinent examples.
Location
Likewise, no consensus exists as to where proto-Afroasiatic originated. Scholars have proposed locations for the Afroasiatic homeland across Africa and West Asia. Roger Blench writes that the debate possesses "a strong ideological flavor", with associations between an Asian origin and "high civilization". An additional complicating factor is the lack of agreement on the subgroupings of Afroasiatic (see Further subdivisions) – this makes associating archaeological evidence with the spread of Afroasiatic particularly difficult. Nevertheless, there is a long-accepted link between the speakers of Proto-Southern Cushitic languages and the East African Savanna Pastoral Neolithic (5000 years ago), and archaeological evidence associates the Proto-Cushitic speakers with economic transformations in the Sahara dating c. 8,500 ago, as well as the speakers of the Proto-Zenati variety of the Berber languages with an expansion across the Maghreb in the 5th century CE.
An origin somewhere on the African continent has broad scholarly support, and is seen as being well-supported by the linguistic data. Most scholars more narrowly place the homeland near the geographic center of its present distribution, "in the southeastern Sahara or adjacent Horn of Africa." The Afroasiatic languages spoken in Africa are not more closely related to each other than they are to Semitic, as one would expect if only Semitic had remained in an West Asian homeland while all other branches had spread from there. Likewise, all Semitic languages are fairly similar to each other, whereas the African branches of Afroasiatic are very diverse; this suggests the rapid spread of Semitic out of Africa. Proponents of an origin of Afroasiatic within Africa assume the proto-language to have been spoken by pre-Neolithic hunter-gatherers, arguing that there is no evidence of words in Proto-Afroasiatic related to agriculture or animal husbandry. Christopher Ehret, S.O. Y. Keita, and Paul Newman also argue that archaeology does not indicate a spread of migrating farmers into Africa, but rather a gradual incorporation of animal husbandry into indigenous foraging cultures. Ehret, in a separate publication, argued that the two principles in linguistic approaches for determining the origin of languages which are the principles of fewest moves and greatest diversity had put “beyond reasonable doubt” that the language family “had originated in the Horn of Africa”.
A significant minority of scholars supports an Asian origin of Afroasiatic, most of whom are specialists in Semitic or Egyptian studies. The main proponent of an Asian origin is the linguist Alexander Militarev, who argues that Proto-Afroasiatic was spoken by early agriculturalists in the Levant and subsequently spread to Africa. Militarev associates the speakers of Proto-Afroasiatic with the Levantine Post-Natufian Culture, arguing that the reconstructed lexicon of flora and fauna, as well as farming and pastoralist vocabulary indicates that Proto-AA must have been spoken in this area. Scholar Jared Diamond and archaeologist Peter Bellwood have taken up Militarev's arguments as part of their general argument that the spread of linguistic macrofamilies (such as Indo-European, Bantu, and Austro-Asiatic) can be associated with the development of agriculture; they argue that there is clear archaeological support for farming spreading from the Levant into Africa via the Nile valley.
Phonological characteristics
Afroasiatic languages share a number of phonetic and phonological features.
Syllable structure
Egyptian, Cushitic, Berber, Omotic, and most languages in the Semitic branch all require a syllable to begin with a consonant (with the exception of some grammatical prefixes). Igor Diakonoff argues that this constraint goes back to Proto-Afroasiatic. Some Chadic languages allow a syllable to begin with a vowel; however, in many Chadic languages verbs must begin with a consonant. In Cushitic and Chadic languages, a glottal stop or glottal fricative may be inserted to prevent a word from beginning with a vowel. Typically, syllables only begin with a single consonant.
With the exception of some Chadic languages, all Afroasiatic languages allow both closed and open syllables; many Chadic languages do not allow a syllable to end in a consonant. Most words end in a vowel in Omotic and Cushitic, making syllable-final consonant clusters rare. Diakonoff argues that proto-Afroasiatic syllables disallowed consonant clusters or vowels at the end of a syllable.
Syllable weight plays an important role in AA, especially in Chadic; it can affect the form of affixes attached to a word.
Consonant systems
Several Afroasiatic languages have large consonant inventories, and it is likely that this is inherited from proto-Afroasiatic. All Afroasiatic languages contain stops and fricatives; some branches have additional types of consonants such as affricates and lateral consonants. AA languages tend to have pharyngeal fricative consonants, with Egyptian, Semitic, Berber, and Cushitic sharing and . In all AA languages, consonants can be bilabial, alveolar, velar, and glottal, with additional places of articulation found in some branches or languages. Additionally, the glottal stop usually exists as a phoneme, and there tends to be no phonemic contrast between and or and . In Cushitic, the Ethiopian Semitic language Tigrinya, and some Chadic languages, there is no underlying phoneme at all.
Most, if not all, branches of Afroasiatic distinguish between voiceless, voiced, and "emphatic" consonants. The emphatic consonants are typically formed deeper in the throat than the others; they can be realized variously as glottalized, pharyngealized, uvularized, ejective, and/or implosive consonants in the different branches. It is generally agreed that only the obstruents had a contrast between voiceless and voiced forms in Proto-Afroasiatic, whereas continuants were voiceless.
A form of long-distance consonant assimilation known as consonant harmony is attested in Berber, Chadic, Cushitic, and Semitic: it usually affects features such as pharyngealization, palatalization, and labialization. Several Omotic languages have "sibilant harmony", meaning that all sibilants (s, sh, z, ts, etc.) in a word must match.
Consonant incompatibility
Restrictions against the co-occurrence of certain, usually similar, consonants in verbal roots can be found in all Afroasiatic branches, though they are only weakly attested in Chadic and Omotic. The most widespread constraint is against two different labial consonants (other than w) occurring together in a root, a constraint which can be found in all branches but Omotic. Another widespread constraint is against two non-identical lateral obstruents, which can be found in Egyptian, Chadic, Semitic, and probably Cushitic. Such rules do not always apply for nouns, numerals, or denominal verbs, and do not affect prefixes or suffixes added to the root. Roots that may have contained sequences that were possible in Proto-Afroasiatic but are disallowed in the daughter languages are assumed to have undergone consonant dissimilation or assimilation.
A set of constraints, developed originally by Joseph Greenberg on the basis of Arabic, has been claimed to be typical for Afroasiatic languages. Greenberg divided Semitic consonants into four types: "back consonants" (glottal, pharyngeal, uvular, laryngeal, and velar consonants), "front consonants" (dental or alveolar consonants), liquid consonants, and labial consonants. He showed that, generally, any consonant from one of these groups could combine with consonants from any other group, but could not be used together with consonants from the same group. Additionally, he showed that Proto-Semitic restricted a sequence of two identical consonants in the first and second position of the triliteral root. These rules also have a number of exceptions:
velar consonants can occur with pharyngeals or laryngeals;
dental consonants can co-occur with sibilants; However, there are no Proto-Semitic verbal roots with ḍ and a sibilant, and roots with d and a sibilant are uncommon. In all attested cases of a dental and a sibilant, the sibilant occurs in first position and the dental in second.
Similar exceptions can be demonstrated for the other AA branches that have these restrictions to their root formation. James P. Allen has demonstrated that slightly different rules apply to Egyptian: for instance, Egyptian allows two identical consonants in some roots, and disallows velars from occurring with pharyngeals.
Vowel systems
There is a large variety of vocalic systems in AA, and attempts to reconstruct the vocalic system of Proto-Afroasiatic vary considerably.
All branches of Afroasiatic have a limited number of underlying vowels (between two and seven), but the number of phonetic vowels can be much larger. The quality of the underlying vowels varies considerably by language; the most common vowel throughout AA is schwa. In the different languages, central vowels are often inserted to break up consonant clusters (a form of epenthesis). Various Semitic, Cushitic, Berber, and Chadic languages, including Arabic, Amharic, Berber, Somali, and East Dangla, also exhibit various types of vowel harmony.
Tones
The majority of AA languages are tonal languages: phonemic tonality is found in Omotic, Chadic, and Cushitic languages, but absent in Berber and Semitic. There is no information on whether Egyptian had tones. In contemporary Omotic, Chadic, and Cushitic languages, tone is primarily a grammatical feature: it encodes various grammatical functions, only differentiating lexical roots in a few cases. In some Chadic and some Omotic languages every syllable has to have a tone, whereas in most Cushitic languages this is not the case. Some scholars postulate that Proto-Afroasiatic may have had tone, while others believe it arose later from a pitch accent.
Similarities in grammar, syntax, and morphology
At present, there is no generally accepted reconstruction of Proto-Afroasiatic grammar, syntax, or morphology, nor one for any of the sub-branches besides Egyptian. This means that it is difficult to know which features in Afroasiatic languages are retentions, and which are innovations. Moreover, all Afroasiatic languages have long been in contact with other language families and with each other, leading to the possibility of widespread borrowing both within Afroasiatic and from unrelated languages. There are nevertheless a number of commonly observed features in Afroasiatic morphology and derivation, including, the use of suffixes, infixes, vowel lengthening and shortening as a morphological change, as well as the use of tone changes to indicate morphology. Further commonalities and differences are explored in more detail below.
General features
Consonantal root structures
A widely attested feature in AA languages is a consonantal structure into which various vocalic "templates" are placed. This structure is particularly visible in the verbs, and is particularly noticeable in Semitic. Besides for Semitic, vocalic templates are well attested for Cushitic and Berber, where, along with Chadic, it is less productive; it is absent in Omotic. For Egyptian, evidence for the root-and-template structure exists from Coptic. In Semitic, Egyptian, Berber, verbs have no inherent vowels at all; the vowels found in a given stem are dependent on the vocalic template. In Chadic, verb stems can include an inherent vowel as well.
Most Semitic verbs are triliteral (have three consonants), whereas most Chadic, Omotic, and Cushitic verbs are biliteral (having two consonants). The degree to which the Proto-AA verbal root was triliteral is debated. It may have originally been mostly biconsonantal, to which various affixes (such as verbal extensions) were then added and lexicalized. Although any root could theoretically be used to create a noun or a verb, there is evidence for the existence of distinct noun and verb roots, which behave in different ways.
As part of these templates, the alternation (apophony) between high vowels (e.g. i, u) and a low vowel (a) in verbal forms is usually described as one of the main characteristics of AA languages: this change codes a variety of different functions. It is unclear whether this system is a common AA trait; the Chadic examples, for instance, show signs of originally deriving from affixes, which could explain the origins of the alterations in other languages as well.
Word order
It remains unclear what word order Proto-Afroasiatic had. Berber, Egyptian, and most Semitic languages are verb-initial languages, whereas Cushitic, Omotic and some Semitic subgroups are verb-final languages. Proto-Chadic is reconstructed as having verb-initial word order, but most Chadic languages have subject-verb-object word order.
Reduplication and gemination
Afroasiatic Languages use the processes of reduplication and gemination (which often overlap in meaning) to derive nouns, verbs, adjectives, and adverbs throughout the AA language family. Gemination in particular is one of the typical features of AA. Full or partial reduplication of the verb is often used to derive forms showing repeated action (pluractionality), though it is unclear if this is an inherited feature or has been widely borrowed.
Nouns
Grammatical gender and number
The assignment of nouns and pronouns to either masculine or feminine gender is present in all branches – but not all languages – of the Afroasiatic family. This sex-based gender system is widely agreed to derive from Proto-Afroasiatic. In most branches, gender is an inherent property of nouns. Additionally, even when nouns are not cognates, they tend to have the same gender throughout Afroasiatic ("gender stability"). In Egyptian, Semitic, and Berber, a feminine suffix -t is attested to mark feminine nouns; in some Cushitic and Chadic languages, a feminine -t suffix or prefix (lexicalized from a demonstrative) is used to mark definiteness. In addition to these uses, -t also functions as a diminutive, pejorative, and/or singulative marker in some languages.
Afroasiatic languages have a variety of ways of marking plurals; in some branches, nouns change gender from singular to plural (gender polarity), while in others, plural forms are ungendered. In addition to marking plurals via a number of affixes (with the suffixes -*uu/-*w and -*n(a) widely attested), several AA languages make use of internal vowel change (apophony) and/or insertion (epenthesis). These so-called "internal a" or "broken" plurals are securely attested in Semitic, Berber, Cushitic, and Chadic, although it is unclear if the Chadic examples are an independent development. Another common method of forming plurals is reduplication.
Noun cases and states
Nouns cases are found in the Semitic, Berber, Cushitic, and Omotic. They are not found in Chadic languages, and there is no evidence for cases in Egyptian. A common pattern in AA languages with case is for the nominative to be marked by -u or -i, and the accusative to be marked by -a. However, the number and types of cases varies across AA and also within the individual branches. Some languages in AA have a marked nominative alignment, a feature which may date back to Proto-Afroasiatic. Zygmont Frajzyngier states that a general characteristic of case marking in AA languages is that it tends to mark roles such as genitive, dative, locative, etc. rather than the subject and object.
A second category, which partially overlaps with case, is the AA linguistic category of "state." Linguists use the term "state" to refer to different things in different languages. In Cushitic and Semitic, nouns exist in the "free state" or the "construct state". The construct state is a special, usually reduced form of a noun, which is used when the noun is possessed by another noun (Semitic) or is modified by an adjective or relative clause (Cushitic). Edward Lipiński refers to Semitic nouns as having four states: absolute (free/indeterminate), construct, determinate, and predicate. Coptic and Egyptian grammar also refers to nouns having a "free" (absolute) state, a "construct state," and a "pronominal state." The construct state is used when a noun becomes unstressed as the first element of a compound, whereas the pronominal state is used when the noun has a suffixed possessive pronoun. Berber instead contrasts between the "free state" and the "annexed state," the latter of which is used for a variety of purposes, including for subjects placed after a verb and after certain prepositions.
Modifiers and agreement
There is no strict distinction between adjectives, nouns, and adverbs in Afroasaiatic. All branches of Afroasiatic have a lexical category of adjectives except for Chadic; some Chadic languages do have adjectives, however. In Berber languages, adjectives are rare and are mostly replaced by nouns of quality and stative verbs. In different languages, adjectives (and other modifiers) must either precede or follow the noun. In most AA languages, numerals precede the noun.
In those languages that have adjectives, they can take gender and number markings, which, in some cases, agree with the gender and number of the noun they are modifying. However, in Omotic, adjectives do not agree with nouns: sometimes, they only but take gender and number marking only when they are used as nouns, in other cases, they take gender and number marking only when they follow the noun (the noun then receives no marking).
A widespread pattern of gender and number marking in Afroasiatic, found on demonstratives, articles, adjectives, and relative markers, is a consonant N for masculine, T for feminine, and N for plural. This can be found in Semitic, Egyptian, Beja, Berber, and Chadic. A system K (masculine), T (feminine), and H (plural) can be found in Cushitic, Chadic, with masculine K also appearing in Omotic. The feminine marker T is one of the most consistent aspects across the different branches of AA.
Verb forms
Tenses, aspects, and moods (TAMs)
There is no agreement about which tenses, aspects, or moods (TAMs) Proto-Afroasiatic might have had. Most grammars of AA posit a distinction between perfective and imperfective verbal aspects, which can be found in Cushitic, Berber, Semitic, most Chadic languages, and some Omotic languages. The Egyptian verbal system diverges greatly from that found in the other branches. Additionally, it is common in Afroasiatic languages for the present/imperfective form to be a derived (marked) form of the verb, whereas in most other languages and language families the present tense is the default form of the verb. Another common trait across the family is the use of a suppletive imperative for verbs of motion.
"Prefix conjugation"
Conjugation of verbs using prefixes that mark person, number, and gender can be found
in Semitic, Berber, and in Cushitic, where it is only found on a small set of frequent verbs. These prefixes are clearly cognate across the branches, although their use within the verbal systems of the individual languages varies. There is a general pattern in which n- is used for the first person plural, whereas t- is used for all forms of the second person regardless of plurality or gender, as well as feminine singular. Prefixes of ʔ- (glottal stop) for the first person singular and y- for the third person masculine can also be reconstructed. As there is no evidence for the "prefix conjugation" in Omotic, Chadic, or Egyptian, it is unclear whether this was a Proto-Afroasiatic feature that has been lost in those branches or is a shared innovation among Semitic, Berber, and Cushitic.
"Suffix conjugation"
Some AA branches have what is called a "suffix conjugation", formed by adding pronominal suffixes to indicate person, gender, and number to a verbal adjective. In Akkadian, Egyptian, Berber, and Cushitic this forms a "stative conjugation", used to express the state or result of an action; the same endings as in Akkadian and Egyptian are also present in the West Semitic perfective verb form. In Akkadian and Egyptian, the suffixes appear to be reduced forms of the independent pronouns (see Pronouns); the obvious correspondence between the endings in the two branches has been argued to show that Egyptian and Semitic are closely related. While some scholars posit an AA origin for this form, it is possible that the Berber and Cushitic forms are independent developments, as they show significant differences from the Egyptian and Semitic forms. The Cushitic forms in particular may be derived from morphology found in subordinate clauses.