One final instance of the survival of Celtic must be mentioned for the controversy which it has evoked. Jerome says that the Galatians in his day had practically the same native language as the Treveri.[112] Was this language Celtic? Freeman[113] thinks that Jerome’s word may be doubted, as he was not a philologist. This would seem to rule out all the witnesses, for philology is an entirely modern development; and it would hardly have been indispensable for forming so simple a judgement. Jerome, moreover, is a considerable authority, the most learned of the Fathers. Lavisse[114] has recently accepted his statement. He mentions the objection of Perrot,[115] who maintains that Celtic had long vanished from Asia Minor, and of Fustel de Coulanges,[116] who says that the language of the Treveri was German, and answers: (1) that Celtic survived in the speech if not in the documents of Asia Minor, and (2) that Coulanges is wrong; the names of the Treveri are Celtic. This being granted, it would seem that Celtic survived well into the fifth century, and this conclusion appears to be reinforced in extent and significance by Freeman’s statement that there was a survival of the Celtic language and sentiment in Brittany during the fifth century.[117] But this statement is misleading. It is generally admitted that Gallic had entirely gone out of use in Armorica, when the fugitives from Great Britain settled in the country and introduced their insular speech from the fifth century onwards. And Breton is more closely allied to Welsh and Old Cornish than to Gallic. On the whole we must say that the evidence of modern philology points to a less considerable influence of the Celtic element than we should expect. Celtic was overshadowed by German, and especially, of course, Latin. Hitherto modern philology has found traces of Celtic loan-words in the following spheres: agriculture, carriage-building, the names of animals, trees, and plants, the parts of the body, items of clothing, weapons, and geographical terms.[118] The inscriptions in Gaul show words like _cantalon_,[119] a kind of building, and _cantuna_ (canteen) which the philologists pronounce to be of Celtic origin.[120] We must conclude, therefore, that while sporadic traces of Celtic are undoubtedly found in the fifth century (as in the case of the Treviri, who in their secluded valley would naturally retain the ancient language longer than the people around them), the language had, in the main, disappeared by that time.