Introduction to the problem of African Latin
Béla Adamik (University ELTE Budapest, Hungary)
LLDB = Computerized Historical Linguistic Database of the Latin Inscriptions of the Imperial Age
(http://lldb.elte.hu/)
1
II. Phonological changes in African Latin as evidenced in inscriptions
1. “Labdacismi”
LLDB-71554: l > LL, SVSTVLLIT = sustulit, ILAlg 1, 3310, 14 = CIL 8, 1966, 14,
Africa proconsularis, Theveste , undated
LLDB-51058: l > LL, AVRELLII = Aurelii, ILTun 1385, 2 = CIL 8, 26457, 2, Africa
proconsularis, Thugga , 222-235 AD
LLDB-54401: ll > L, BELA = bella, MAD 1559, 8 = CIL 8, 26450a, 8, Africa
proconsularis, Thugga , 151-300 AD
LLDB-51382: ll > L, GALIENI = Gallieni, ILTun 1732, 8 = CIL 8, 22051, 8, Africa
proconsularis, by Tunes, 253-260 AD
Bibliography:
Acquati, A. (1971). Il vocalismo latino-volgare nelle iscrizioni africane, Annali della Facoltà di Lettere e
Filosofia dell’Università degli studi di Milano 24, 155–84.
Adams, J. N. (2007). The Regional Diversification of Latin 200 BC–AD 600. Cambridge.
Omeltchenko, S. W. (1977). A Quantitativ and Comparative Study of the Vocalism of the Latin Inscriptions of
North-Africa, Britain, Dalmatia and the Balkans, Chapel Hill.
Petersmann, H. (1998). Gab es ein afrikanisches Latein? Neue Sichten eines alten Problems der lateinischen
Sprachwissenschaft, in B. García-Hernández (ed.), Estudios de lingüística latina: Actas del IX Coloquio
internacional de lingüística latina 1. Madrid, 125–36.
Introduction to the problem of African Latin
Béla Adamik (University ELTE Budapest, Hungary)
LLDB = Computerized Historical Linguistic Database of the Latin Inscriptions of the Imperial Age
(http://lldb.elte.hu/)
2
2. Vowel lengthening and shortening
LLDB-52670: productio vocalium cum accentu, ERIGERE MAN = erigere manum,
ILAlg 1, 276, 4 = CIL 8, 5352, 4, Africa proconsularis, Calama, AD 539.
clausula hexametri, ērīgere mānum pro ērigere manum
LLDB-64387: productio vocalium sine accentu in arsi, ROGATIANE TIBI = Rogatiane,
tibi, ILAlg 2, 829, 3-4 = CIL 8, 7228, 3-4, Numidia, Cirta, undated.
Initio hexametri, Roga:tiá-ne > Ro:gatiá:ne
LLDB-53796: correptio vocalium sine accentu, IN GALLIA MORTE = in Gallia morte,
ILTun 582, 1 = CIL 8, 702, 1, Africa proconsularis, Chusira, 201-300 AD,
clausula hexametri, Gallia pro Gallia:
3. Vowel mergers - letter confusions
cl. Latin > vulgar Latin: ḗ í > ẹ, é > ę; ṓ ú > ọ, ó > ǫ; ē e i > ẹ, ō o u > ọ → E ~ I / O ~ V
 LLDB-46796: é: > I, FICI = fecit, ILTun 905, 7
 LLDB-48692: e: > I, FIDILIS = fidelis, IFCCarth 2, 182, 3,
 LLDB-37408: e > I, SINATO[R] = senator, ILAlg 1, 26, 1
 (LLDB-66050: é > I , INGINVA| = Ingenua, ILAlg 2, 2567, 3)
 (LLDB-12163: í: > E, PEREGRENO = peregrino, InscrAqu 3, 3103, 6)
 LLDB-70975: í > E, DOMETIVS = Domitius, ILAlg 2, 3735, 1
 (LLDB-43681: i: > E, ESIDORVS = Isidorus, IRT 201, 2)
 LLDB-47819: i > E, FEDELIS = fidelis, ILTun 1152, 1
 LLDB-46207: ó: > V, MENSVR = Mensori, ILTun 766, 1
 LLDB-58250: o: > V, NEPVS = nepos, ILAlg 1, 2261, 3,
 LLDB-59279: o > V, FECERON[T] | = fecerunt, ILAlg 1, 2881, 7
 (LLDB-71392: ó > V, LVNGOS | = longos, ILAlg 1, 3149, 5)
 (LLDB-58849: ú: > O, IOL = Iulius, ILAlg 1, 2500, 2)
 LLDB-73663: ú > O, OXOR = uxor, ILAlg 1, 2872, 1
 (LLDB-72796: u: > O, FOFIDIVS | = Fufidius, ILAlg 1, 3755, 2)
 LLDB-51365: u > O, ISTERCOLVS = Sterculus, ILTun 1710/74, 1
Introduction to the problem of African Latin
Béla Adamik (University ELTE Budapest, Hungary)
LLDB = Computerized Historical Linguistic Database of the Latin Inscriptions of the Imperial Age
(http://lldb.elte.hu/)
3
1. 100% = 1687 C = 77% V = 23% E ~ I + O ~ U = 5% E ~ I / O ~ U = 1,5
2. 100% = 226 C = 77% V = 23% E ~ I + O ~ U = 6% E ~ I / O ~ U = 1
3. 100% = 87 C = 62% V = 38% E ~ I + O ~ U = 14% E ~ I / O ~ U = 13
4. 100% = 2453 C = 59% V = 41% E ~ I + O ~ U = 15% E ~ I / O ~ U = 2
The first exploration of the vowel system of later African Latin based on a selection of Christian inscriptions (ILCV)
was carried out by Omeltchenko 1977, who concluded that the Latin of Africa was characterized by the extreme
conservatism of its vocalism (i.e. by the extreme rarity of E/I and O/U confusions).
Introduction to the problem of African Latin
Béla Adamik (University ELTE Budapest, Hungary)
LLDB = Computerized Historical Linguistic Database of the Latin Inscriptions of the Imperial Age
(http://lldb.elte.hu/)
4
5. (6.) 100% = 170 C = 68% V = 32% E ~ I + O ~ U = 26% E ~ I / O ~ U = 0,9
6. (5.) 100% = 1047 C = 57% V = 43% E ~ I + O ~ U = 22% E ~ I / O ~ U = 2,7
7. 100% = 478 C = 43% V = 57% E ~ I + O ~ U = 32% E ~ I / O ~ U = 2,2
8. 100% = 1614 C = 24% V = 76% E ~ I + O ~ U = 64% E ~ I / O ~ U = 2
The results of Omeltchenko were supported by Adams 2007, who analysed the spelling errors of the Bu
Njem ostraca and the Albertini tablets (both from Africa) not only regarding vowels but also consonants,
with special attention to the b-w merger reflected in B/V confusions.
Introduction to the problem of African Latin
Béla Adamik (University ELTE Budapest, Hungary)
LLDB = Computerized Historical Linguistic Database of the Latin Inscriptions of the Imperial Age
(http://lldb.elte.hu/)
5
1. 100% = 2007 C = 70% V = 30% E ~ I + O ~ U = 3% E ~ I / O ~ U = 2
2. 100% = 1047 C = 67% V = 33% E ~ I + O ~ U = 7% E ~ I / O ~ U = 2,5
3. 100% = 806 C = 60% V = 40% E ~ I + O ~ U = 9% E ~ I / O ~ U = 2
4. 100% = 919 C = 57% V = 43% E ~ I + O ~ U = 12% E ~ I / O ~ U = 5
Nevertheless, the substantial investigations of Omeltchenko and Adams could not present a full picture since they did
not use extensive epigraphic corpora but narrow selections or small corpora of inscriptions, while almost entirely
overlooked the huge Pre-Christian epigraphic material.
Introduction to the problem of African Latin
Béla Adamik (University ELTE Budapest, Hungary)
LLDB = Computerized Historical Linguistic Database of the Latin Inscriptions of the Imperial Age
(http://lldb.elte.hu/)
6
5. 100% = 2552 C = 55% V = 45% E ~ I + O ~ U = 8% E ~ I / O ~ U = 3
6. 100% = 114 C = 71% V = 29% E ~ I + O ~ U = 5% E ~ I / O ~ U = 4
7. 100% = 79 C = 78% V = 22% E ~ I + O ~ U = 0% E ~ I / O ~ U = 0
(8. 100% = 23 C = 57% V = 43% E ~ I + O ~ U = 17% E ~ I / O ~ U = 0,3)
The preliminary results of the investigation suggest that while in the later Christian period the African vocalism was
indeed very close to the Sardinian one and quite far from the Gallic one, in the Pre-Christian period the situation was
quite the opposite, and African vocalism resembled the Hispanic and Gallic ones instead of the Sardinian one.