PROGRESS OF ROMANCE PHILOLOGY 239 



langues de I'Europe latine dans leurs rapports avec la langue des 

 troubadours. But the very title of this book shows that the work was 

 intended to be the demonstration of a preconceived idea; all its 

 conclusions were necessarily vitiated because they were subordinated 

 to an erroneous theory. In fact, Raynouard's leading idea was that 

 between Latin and the various Neo-Latin tongues there had existed 

 an intermediate stage; this he called, using the term in a special 

 sense, the langue romane. According to his theory, this language had 

 developed in close succession to Latin all over Latin Europe; but, 

 while it was preserved by a miraculous exception in the south of 

 France, everywhere else it had undergone the special modifications 

 which led to the formation of French, Italian, Spanish, etc. 



The point of departure for this conception, which after all is not 

 so radically false as it may seem, is an unfortunate interpretation 

 of the expression Romana lingua, which, in Latin writings of the 

 ninth and tenth centuries and even later, is used to designate a 

 language quite different from the Classic Latin, but one whose 

 relationship to the ancient Roman idiom was clearly understood at 

 the time. By lingua romana, or lingua rustica romana, people under- 

 stood in every Romance country nothing more nor less than the 

 vulgar speech as opposed to literary or grammatical Latin. The use 

 of the same expression in different countries did not in the least 

 imply that the lingua romana was everywhere the same. People 

 of that time cared little for such a question. Raynouard, finding 

 that the poets of the south of France often gave the name romanz, 

 or lenga romana, to the language they employed, argued that the 

 name being the same, the language must be the same, and persuaded 

 himself that during a period of some length the people of the Latin 

 West had spoken the language of Provence, his native region. 



Herein lay his error. On the other hand, we cannot doubt that 

 there was a stage between the Classic Latin and the Romance lan- 

 guages: this is a fact long since recognized. But the intermediate 

 stage, generally designated as Vulgar Latin, had no closer connection 

 with the Romance of South France than with that of other regions. 

 A good many years were to elapse before the study of Vulgar Latin 

 I mean of that small part of it that we can ever really know 

 was undertaken in a methodical way. 



Nevertheless, Raynouard's work, in spite of its fundamental error 

 and in spite of a thousand mistakes and confusions in matters of 

 detail, was by no means useless, for its author may be termed in a 

 certain sense the precursor of Diez, and to Diez belongs incontestably 

 the honor of having founded the comparative grammar of the 

 Romance languages. 



Diez, as one may conclude from his writings, and as he appeared to 

 me forty years ago when I visited him in Bonn, was a modest and 



