In the massive Dictionary of Christian Archaeology and Liturgy by Fernand Cabrol and Henri Leclercq, the article “Africa (Languages Spoken in)” (1920) concludes that African Latin was not distinct enough from Latin elsewhere in the Empire to form a definable dialect.
“Africanisms” seem to be reduced to mere phenomena rather than a complete idiom. What does seem likely is that classical and official Latin hardly allowed itself to be penetrated by the various kinds of African patois, which it pushed ahead of it as far as Roman civilization extended. In the time of Apuleius, such a poor Latin was spoken at Madaura that that writer was forced to learn the language over again when he came to Rome; two centuries later the Madaurans were, on the contrary, so Latinized that Punic names sounded odd in the ears of Maximus, a grammarian of Madaura: diis hominibusque odiosa nomina, he said of them [S. Augustin, Epist., XVI, P. L., t. XXXI, col. 82.] in writing to St. Augustine, who, for his part, informs us that everyone around him speaks Latin and a child has only to hear it to learn it. [Confessions, l. I, c. XIV, P. L., t. XXXII, col. 671.] Christianity became a powerful vehicle of the Latin language. Though we lack liturgical books, which have all vanished, at least up to this time, the catechetical treatises of the Fathers have come down to us only in Latin; moreover, the inscriptions that decorate the churches, marble fronts or mosaic pavements, are found to be almost exclusively Latin. Doubtless it was necessary to make an effort to be understood; St. Augustine willingly committed grammatical faults and used incorrect words, preferring, he says, to make the wise grumble than to be misunderstood by his auditors, and the collection of Christian and pagan inscriptions of the last centuries of the empire prove that the people of the most modest condition chose a Latin epitaph for their tomb; in fact, tituli in the indigenous idiom are so rare that they are hardly worth mentioning. “If the inscriptions,” justly observes G. Boissier, “were of irreproachable correctness, we might suppose that they were composed only by professional literary men, and that below that level only the idioms of the country were understood. The improprieties of terms, the grammatical errors, the solecisms and barbarisms that are encountered on almost every line show us that we are dealing with ignorant people; that they spoke Latin badly, but at least they spoke it. We must conclude, therefore, that the Africans ended up becoming masters of a language that was at first foreign to them, since they used it to express the sentiments that meant the most to them.” [G. Boissier, L’Afrique romaine, in-12, Paris, 1901, p. 344 sq.] Finally, if “Africanisms” are found as part of the classical language, we can say as much of barbarian terms. There were not two ways of speaking Latin badly, but only one that is found in all the provinces; Africa, Spain, Gaul, Italy, and Rome itself commit similar errors, and on this point as well Africitas gives us nothing positive. “Where is the vulgar Latin that might be different from the Latin of the Church of Africa? In all the provinces of the empire, Latin was the language of the Roman people, the language of the household, of the family, of the market, of the street, of the workshop and the camps. But then why speak of an African Latin? For this reason: this language common to all the provinces first became in Africa the written language and the language of literature. In Rome, in Italy and in the other provinces, it was only spoken and had no literature. Only Africa had its Tertullians, its Cyprians, its Augustines. This is why it is permissible to speak of an African Latin and to place that name on the special character of these great writers, but not on the language itself. Even supposing that there exist expressions that cannot be found outside Africa, we should have nothing to conclude from that. Could we not perhaps find some that belonged exclusively to an Italian, to Plautus or Terence or Petronius? Would we conclude from that that there was an Italian Latin?” [J. Aymeric, Origine africaine du Codex lugdunensis, dans Les lettres chrétiennes, t. IV, p. 255.]
——Dictionnaire d’archéologie chrétienne et de liturgie: “Afrique (langues parlées en)”.