I now approach, not unconscious of its difficulties, the etymology of the word. In the lax Latin of the middle ages, we find Desvres spoken of as Divernia Bononiensis. There is the epitaph of a churchman, born in the place, which says on his behalf:
"Me Molinet peperit Divernia Bononiensis."
The local historian, Baron d'Ordre, speaks of the place as "Désurène, Divernia, aujourd'hui Desvres." [Footnote 134] The name Desvres itself evidently has undergone strange, yet traceable, variations and modifications. [Footnote 135] Its first appearance as a French word is "Desurennes," and this is derived from Desvres sur Enna, or Desvres upon the Enna or Liane, which, as I have said, flows past the place, giving its name to a little village near the forest. By this derivation, however, only the first two letters of the original word Desvres are left. How do they disappear, why do they reappear in the modern form of the word, and what is its original derivation?
[Footnote 134: "Notice historique sur la ville de Désurène Divernia, aujourd'hui Desvres." Par M. d'Ordre. Boulogne, 1811.]
[Footnote 135: "II n'y pas 50 ans que le nom de Desvres a prevalu sur celul de Desurenne que cette ville avait tonjours porté auparavant."—M. L. Cousin "Mémoires de la Société des Antiquarires de la Morinie," vol. iv., p. 239. M. Cousin's papers on Monthulin and Tingry, in the Transactions of this society, are in general accord with what I have said of the ancient military importance of the whole district of Desvres.]
It is a very curious fact, that in England the Roman camps seem to have been always known as "Castra," while in Gaul the Tabernae is the name which generally adhered to them. Lanigan says, and correctly, so far as I have been able to discover, that there is no trace of a Roman station called Tabernae in England, while the affix chester is the most common in its topography. In England, it may be said the Romans encamped; in France, the Tabernae meant a more settled and familiar residence, as familiar as the Caserne of the empire. It would be interesting to inquire whether as many cities in France do not derive their origin from these military stations as England has of Chesters. But the student who attempts this task will be sure to find the Latin word almost defaced beyond power of recognition by the etymological maltreatment which it has sustained in that conflict of consonants which has resulted in the present high polish of Academic French. I may mention one or two instances to show how little violence I do to French philology in identifying the Divernia Bononiensis of the middle ages with the Tabenae of Boulogne. Saveme in Lorraine is well known to be the Tabenae Triborocrum. It was known in a semi-Germanic form as Elsas Tabern. Gradually the sibilant ss of the first word invaded the second; and it has long settled down into one word in the form of Saveme. The Tabernae Rhenanae, on the other hand, retained the hard b instead of converting it into v, as inevitably happened in the south, and instead changed the T into Z Rhein-Zabren. In ages which had no hesitation in changing the pure dental T into the sibilant dentals S or Z, it will not be considered surprising that it was sometimes changed into D—the only other pure dental sound. Indeed, of all the transmutations of letters, those of d and t and those of v and b, are notoriously the most common. "The Irish d," says O'Donovan, "never has such a hard sound as the English d. " Again, "In ancient writings, t is frequently substituted for d. " Again, "It should be remarked that in ancient Irish MSS. consonants of the same organ are very frequently substituted for each other, and that where the ancients usually wrote p, c, t, the moderns write b, q, d." [Footnote 136] Decline the Irish word Tâd, father. It becomes Ei dâd, his father; Ei thâd, her father; by nhâd my father. We carry the tendency into English. The mistake is one from which certain parts of Ireland as well as certain parts of France are not exempt even to the present day; and in Munster one may still [{755}] hear, as in the times when the ballad of "Lillibullero" was written, the letter d occasionally used where the tongue intended t or th. Nor is this vagary of speech confined to the Irish. Why do the Welsh say Tafyd for David? It is the most frequently recurring of that systematic permutation of consonants which is one of the chief difficulties of the Cymbric tongue. The Welsh d and t turn about and wheel about in their mysterious alphabet without the slightest scruple. In Germany the convertibility of the same letters is also very marked. The German says das for that, Dank for thanks, Durst for thirst; and again Teufel for devil, Tanz for dance, Theil for dial. As to the same abuse in France, the dictionary of the Academy and that of Bescherelle [Footnote 137] lay down the principle very plainly: "Le t est une lettre à la fois linguale et dentale, comme le d son correlatif, plus faible, plus doux, avec lequel il est fréquemment confondu, nonseulement dans les langues germaniques, mais dans la plupart des langues. En latin, cette lettre so permute fréquemment avec le d: attulit pour adtulit. On écrivit primitivement set, aput, quot, haut, au lieu de sed, apud, quod, hand."
[Footnote 136: O'Donovan, John, LL.D., "A Grammar of the Irish Language." Dublin, 1845.]
[Footnote 137: "Dictionnaire de l'Académie Française," Bescherelle, "Dictionnaire National. " Paris, 1857.]
So far as to the permutation of T and D. I will not waste the time of the reader in order to show that the conversion of v into b is even more common. We find a familiar illustration of it in the old Latin name of Ireland, which, as every one knows, is variously written Ibernia, Ivernia, Hibernia, Juvernia, and Iernia. But the English word tavern, which is exactly derived from the Latin Taberniae, is a still more apposite illustration in the present case. In this word, finally, the intermediate vowel swayed in sound with the consonants which inclosed it. As the primary Latin T changed into the softer and feebler D, and the b into v, the intermediate a lost its full force. The mediaeval Latin melts into i in Divernia. The modern French form, Desvres, brings it half-way back toward its place at the head of the alphabet. It does not run the whole gamut of the vowels, as from Ibernia to Juvernia.
This Divernia Bononiensis, then, I claim to identify with the Taberniae Bononienses, Tournehem with Nemtur or Emtor, Enna with Enon. If it were necessary even to push the proof a step further, there is the district called Le Wicquet, which M. Jean Scoti, who was lieutenant particulier de la Sennechaussée de Boulogne, tells us is undoubtedly derived from the Latin Vicus, and which might naturally be the vico Bonaven Taberniae of which the "Confession" speaks; but the historian of Desvres, Baron d'Ordre, whom I have already cited, disputes this derivation, and says the word is Celtic, and comes from Wic, Celtic for wood, like our word wicket. Both may be right, for Vicus may be a Latin form of the same word. [Footnote 138] But the point is not material.