01
The Languages of Sheba
Old South Arabian is the name given to four closely related languages written in the kingdoms of ancient Yemen: Sabaic, of Saba (the biblical Sheba) with its capital at Marib; Minaic, of the kingdom of Maʿin in the Jawf valley; Qatabanic, of Qataban; and Hadramitic, of Hadramawt. They are known from thousands of inscriptions on stone and bronze, from the early first millennium BCE to the 6th century CE. The kingdoms grew rich on the incense trade, and Minaean merchants left inscriptions far to the north, at Dedan in the Hejaz, and as far away as Egypt and the Aegean island of Delos.
Sabaic is by far the best attested and lasted longest. Minaic, Qatabanic and Hadramitic stopped being written as their kingdoms were absorbed by their neighbours, and by about 300 CE, when Himyar had united the south, Sabaic was the only written language left. The last dated inscriptions come from the mid-6th century, not long before Arabic replaced it.
The name suggests an ancestor of the Modern South Arabian languages, but it is not one: those languages are a separate branch. How Old South Arabian itself fits into West Semitic is uncertain, partly because its script does not show vowels.
02
The Musnad Script
The inscriptions use an alphabet of 29 letters, called musnad by later Arab writers. It descends from the same early alphabetic invention as the Levantine scripts, and it is the ancestor of the Ge'ez script of Ethiopia. Its letters keep apart all the consonants reconstructed for Proto-Semitic, including three sibilants usually transcribed s₁, s₂ and s₃, whose exact pronunciation is debated. Like other early Semitic scripts it writes consonants only; short vowels and doubled consonants are never marked, so much of the grammar has to be inferred.
What can be seen shows real differences between the four languages. Sabaic forms causative verbs with a prefix h- and uses h- in its third-person pronouns, while Minaic, Qatabanic and Hadramitic use s- in both places, the older pattern also found in Akkadian (š-). For this reason some scholars treat Sabaic as one group and the other three as another.
The monumental inscriptions are formal and formulaic. Since the 1970s, several thousand palm-leaf stalks and wooden sticks inscribed in a cursive form of the script have come to light, carrying letters, contracts and writing exercises, and they give a rare view of everyday Sabaic.
Where it’s spoken
- Eritrea
- Ethiopia
- Oman
- Saudi Arabia
- Yemen
Turning points
- WritingContested
The first Sabaic inscriptions
The earliest Sabaic inscriptions are carved in the musnad alphabet. Their dates depend on a long-running dispute between a 'short' chronology, which placed the beginnings around the 5th century BCE, and a 'long' chronology, which radiocarbon dating and links with Assyrian records now generally support.
Sources (2)
- ContactContested
Sabaic across the Red Sea
Inscriptions in Sabaic, with South Arabian-style temples, appear in the Ethiopian and Eritrean highlands, at Yeha and other sites of the kingdom of Daʿmat. Whether they reflect settlers from Saba or a local elite adopting South Arabian culture is debated.
Sources (2)
- ExtinctionContested
Sabaic stands alone
Minaic, Qatabanic and then Hadramitic cease to be written as their kingdoms fall to their neighbours. By about 300 CE, when Himyar controls the whole south, Sabaic is the only written language of Yemen. The exact dates are uncertain.
Sources (2)
- ExtinctionContested
The last inscriptions
The written record ends in the 550s CE. Among the last long texts is the inscription of the Aksumite ruler Abraha recording the repair of the Marib dam (548 CE). With Islam, Arabic replaces Sabaic; how long it survived in speech is unknown.
Sources (2)
Contested classifications
- Widely accepted
South Semitic
Evidence for
The traditional classification, found in most handbooks, rests on geography, the close link between the South Arabian and Ethiopian scripts, and features shared with Ethiosemitic and Modern South Arabian such as the -k endings of the first-person singular and second-person past tense (Sabaic -k).
Evidence against
The script hides vowels and doubled consonants, so it cannot show whether Old South Arabian kept the old doubled imperfective of Ethiosemitic or had the Central Semitic form. Huehnergard and Rubin and others note that it shares some features with Central Semitic, such as a definite article, and treat it as a separate branch of West Semitic whose exact position is unknown. Some scholars also doubt that its four languages form a single group.
Sources (3)