'Punic was not dead'
Within 200 years of the fall of Carthage, Punic was
replaced by Latin, according to accepted theory. Not true, according to Robert Kerr who has completed an analysis of late Punic inscriptions. Punic was a living language, probably until the arrival of Islam.
Lepcis Magna was a Phoenician colony on the coast of North Africa, in present-day Libya. The city was roughly the size of Rome, but was never inhabited after the Arabic conquest. The old Roman city has been completely excavated. 'It was an impressive experience,' says Robert Kerr. 'You can imagine how Rome must have looked in earlier times.'
Kerr, Semitic scholar and comparative linguist, travelled throughout Libya, together with his co-supervisor Karel Jongeling, making photos and sketches of Punic inscriptions. He obtained his doctorate last week - with distinction - on a systematic analysis of the 69 late-Punic inscriptions found in Libya, formerly Tripolitania. The earliest inscriptions are from Lepcis Magna, and have been dated to the first and second centuries B.C. The later inscriptions, which form the majority, come from the interior of the country, from the arid region jbordering the real desert, where the Romans had established a cordon of defensible farms. These inscriptions originate from the third and fourth centuries.
Punic is a north-west Semitic language, like the Hebrew used in the Bible. The Phoenicians, or Punics, were a trading nation; from present-day Libya, they established colonies along the north coast of Africa and in southern Spain in the eighth century B.C. The best known of these colonies is Carthage, the Mediterranean rival to Rome, which was devastated in 146 after the Third Punic War.
The Latin alphabet was used for the inscriptions. But the language, as Kerr was the first to prove systematically, was Punic. For a long time it was believed that Punic was replaced by Latin in the first two centuries after the fall of Carthage in 146 B.C. That's not the case, according to Kerr: 'The language of the inscriptions is a typical example of a functioning north-west Semitic language.'
That the inscriptions were not Latin was already known. But what was it, then? The first text in Lepcis Magna was found by Italian archaeologists in the twenties. Italian scholars realised that the language of the inscriptions contained elements of Semitic, but the date - circa 130 A.D. - means that it could not possibly be Punic since Punic was no longer spoken in Africa at that time, according to classicists. 'Then it must be Berber,' was the conclusion drawn. After the Second World War, British researchers found more texts, which were identified as Latino-Libyan; the script was Latin, the language must be Libyan, a language about which very little ls known.
Text from the wadi Umm el-Agerem
Later, when more Semitic words were found, it was thought that the language must be Punic, but in a seriously deteriorated form. Kerr: 'A number of Berber and African scholars were still prepared to believe that poor farmers continued to speak Punic, but not the elite; they spoke Latin. But it appears from the inscriptions that Punic was still spoken in the third century A.D. by the upper classes in the coastal areas.'
The type of text, the epigraph, was adopted from the Romans. Between 30,000 and 60,000 texts have been found in North Africa. The non-Roman population imitated this immensely popular and prestigious genre. The 69 Punic inscriptions from Tripolitania seem to fade into insignificance in comparison with the enormous number of Latin inscriptions from North Africa. Kerr: 'But in Tripolitania there were far more Punic inscriptions. Hardly any Latin inscriptions were found there.'
Kerr has compared he phonology of the Punic inscriptions with that of the local Latin of North Africa. He discovered that the spelling conventions of the inscriptions must have derived from this, and he suspects that the typical North African pronunciation of Latin, about which classical scholars are always so derisive, and the pronunciation of Punic were very similar. For example, unstressed vowels disappeared in both languages. Kerr believes that both Punic and the African vulgar Latin adopted the accent of another language from the regio, which he calls Berber-Libyan. 'Compare it with the similarities in pronunciation between South African Dutch and South African English, or between Irish and English as it is spoken in Ireland. The language is different, but the accent is recognisable.'
Kerr at an obelisque mausoleum between Bani Walid and Ghadames.
Who were these people who continued to speak Punic for such a long time? According to Kerr, they were a mixed population, stemming from Punic men and native Libyan women. 'Roman sources also mention such a mixed race. The Punic men were assigned to the border area by the Romans. They had been in the army and were now paid well to man the defensible border areas. They had a high degree of freedom. In Roman sources Punic-speaking men were known for being able to farm very successfully in arid regions. Archaeological finds indicate that they had a complicated but successful system for storing rainwater from the wadis to use in dry periods. It was their task to defend the borders from bands of pillagers from the Sahara. They also served as the 'last stagepost before the desert.' They received extra income from caravans in exchange for protection and supplies.'
While this system remained intact, Punic continued to be spoken, according to Kerr. 'The system of defensible farms and storing water was fragile and maintenance-intensive, and did not survive the incursions by Berber tribes from the sixth century and the Islamic conquests in the seventh century. But the whole region is now strewn with the remains of these farms. The outlines of the plots of land, olive groves and even the traces of plough tracks are still visible.'
The conclusion that Punic was a living language in the first five centuries A.D. fits well with the image portrayed in literary sources. Latin authors, for example, made frequent reference to Punic books. Kerr believes that Punic was also the native language of church father Augustine (354-430). 'It is often assumed that Augustine actually meant 'Berber' when he talked about Punic. But he was well aware that there was a difference between Punic and Libyan-Berber. He knew that this last existed, but he wasn't familiar with the language. Augustine also recognised Hebrewisms in the ancient Latin Bible translation, because he knew Punic. He did not understand Hebrew.'
Kerr's research has not only put late-Punic on the map as a fully-fledged north-west Semitic language, it also provides welcome comparative material for the study of other languages from this family, such as Hebrew. But it can also help with identifying Punic loanwords in Latin inscriptions, and gives an insight into the working of the Roman Empire in North Africa.
Kerr: 'For a long time it was thought that Punic culture came to an end when Carthage was destroyed, and 'Africa' became a province of the Roman Empire. But it was only in Tripolitania that this culture flourished. The region was fairly autonomous. Rome did not pay it much attention and the influence from Carthage came to an end after the Second Punic War when the region withdrew from the authority of Carthage. We tend to think in terms of a Rome-Carthage dichotomy, but not everyone in North Africa had posters of Hannibal as their liberator on their walls.'
R.M. Kerr (North Vancouver 1968), Latino-Punic and its Linguistic Environment
Doctorate 21 February 2007, Leiden University
Supervisor: Prof. Dr H. Gzella
Co-supervisor: Dr K. Jongeling
(27 February 2007/HP)