TRANSACTIONS OF SECTION H. 893 



accounted for Ly the settlement of tbe Bretons who followed William the Con- 

 queror, or of the Welsh who came into England when the penal laws against them 

 were repealed by Henry YIII. But the advocates of the theory of extermination 

 had always one argument which seemed to them unanswerable, and which indeed 

 was the origin of their theory. The language of the Anglo-Saxons contains scarcely 

 any words borrowed from Keltic. Such a fact was held to be inexplicable except 

 on the hypothesis that the speakers of the Keltic dialects were all exterminated 

 before any intercourse was possible between them and the invading Teuton. 



But I think I can show that the fact admits of quite another explanation. 

 Roman Britain was in the condition of Roman Gaul ; it was a Roman province, 

 so thoroughly Romanised indeed that before the end of the first century, according 

 to Tacitus ('Agric' 18-21), even the inhabitants of North Wales had adopted the 

 Roman dress and the Roman habits of luxury. After four centuries of Roman 

 domination it is not likely under these circumstances that the dialects of the British 

 tribes would have resisted the encroachment of the Latin language any more than 

 did the dialects of Gaul. The language, not only of government and law, but also 

 of trade and military service, was Latin, while the slaves and servants who culti- 

 vated the soil were bound to understand the language of their masters. Moreover, 

 Britain was a military colony ; the natives were drafted into the army, and there 

 perforce had to speak Latin. If Latm had not been the language of the country 

 at the time the Romans left it, the fact would have been little short of a 

 miracle. 



That it was so is certified by more than one piece of evidence. The inscriptions 

 which have survived from the period of the Roman occupation are numerous; with 

 the exception of three or four Greek ones, they are all in Latin. Of a Keltic 

 language or dialect there is no trace. When the Romans had departed, and the 

 inhabitants of Wales and Cornwall had been cut off from intercourse with the 

 civilised world, Latin was still the ordinary language of the mortuary texts. It 

 is only gradually that Keltic oghams take their place by the side of the Roman 

 characters. When St. Patrick wi-ites a letter to the Welsh prince of Cardiganshire, 

 addressed not only to him but to his people as well, it is in the Latin language ; 

 when St. Germanus crosses into Britain to settle a theological controvert, and' 

 leads the people to victory against the Saxon invader, he has no difficulty in being 

 understood ; and the proper names of the British leaders continue to be Roman 

 long after the departure of the Roman legions. What clinches the matter, however, 

 is the positive statement of Gildas, the British writer, the solitary witness who has' 

 survived to us from the dark period of heatlien invasion. He asserts that the ships 

 called 'keels' by the Saxons were called longce naves 'in our language' ('nostra 

 lingua ').' In the middle of the sixth century, therefore, Latin was still the language 

 of the Kelt south of the Roman Wall. Such being the case, it is not Keltic but 

 Latin words that we must expect to have been borrowed by Anglo-Saxon, if the 

 British population, instead of being exterminated, lived under and by the side of 

 their Teutonic invaders. Now these borrowed Latin words exist in plenty. They 

 have come not only from the speech of the towns^ but also from the speech of the 

 countiy, proving that the country population must have used Latin like the in- 

 habitants of the towns. In an interesting little book by Professor Earle on the 

 Anglo-Saxon names of plants a list is given of the names of trees and vegetables 

 that have been taken from a Latin source. Where the tree or the vegetable was 

 one with which the invaders had not been acquainted in their original home, the 

 name they gave to it was a Latin one, like the cherry or cerasus, the box or biixus, 

 the fennel or feniculum, the niallotv or malva, the poiypy or impaver, the mdish or 

 radix. Such names they could have heard only from the serfs who tilled the 

 ground for their new lords, not from the traders and soldiers of the cities. It is 

 much the same when we turn to the names of agricultural implements which imply 

 a higher order of culture than the simple plough or mattock, the name of which 

 last, however, is itself of Keltic origin. Thus the coulter is the Latin culter, the 

 sickle is the Latin secula. That other agricultural implements bore Teutonic names 



' Hist. 23. 



