We see how cautiously we must tread if we attempt to infer an identity of race from the affinity, or even the resemblance, of languages. Not only have most of the nations of Western Asia and nearly all those of Southern Europe merely adapted the speech of others to their own use, while leaving its main elements untouched; but there are also some who have taken over languages absolutely foreign to them, to which they have made no contribution whatever. The latter case is certainly rarer, and may even be regarded as an anomaly. But its mere existence is enough to make us very careful in admitting a form of proof in which such exceptions are possible. On the other hand, since they are exceptions, and are not met with so often as the opposite case, of a national tongue being preserved for centuries by even a weak nation; since we also see how a language is assimilated to the particular character of the people that has created it, and how its changes are in exact proportion to the successive modifications in the people’s blood; since the part played by a language in forming its derivatives varies with the numerical strength, in the new groups, of the race that speaks it, we may justly conclude that no nation can have a language of greater value than itself, except under special circumstances. As this point is of considerable importance, I will try to bring it out by a new line of proof.
We have already seen that the civilization of a composite people does not include all its social classes.[[131]] The racial influences that were at work in the lower strata from the first still go on; and they prevent the directing forces of the national culture from reaching the depths at all,—if they do, their action is weak and transitory. In France, about five-eighths of the total population play merely an unwilling and passive part in the development of modern European culture, and that only by fits and starts. With the exception of Great Britain, of which the insular position produces a greater unity of type, the proportion is even higher in the rest of the Continent. I will speak of France at greater length, as an instance of the exact correspondence between language and racial type; for in France we have a particular instance that strikingly confirms our main thesis.
We know little, or rather we have no real evidence at all, of the phases which Celtic and rustic Latin[[132]] passed through before they met and coalesced. Nevertheless, St. Jerome and his contemporary Sulpicius Severus tell us (the former in his “Commentaries” on St. Paul’s Epistle to the Galatians, the second in his “Dialogue on the virtues of the Eastern Monks”) that in their time at least two languages were generally spoken in Gaul. There was, first, Celtic, which was preserved on the banks of the Rhine in so pure a form, that it remained identical with the language spoken by the Galatians of Asia Minor, who had been separated from their mother country for more than six centuries.[[133]] Secondly, there was the language called “Gallic,” which according to a commentator, can only have been a form, already broken down, of Popular Latin. This fourth century dialect, while different from the Gallic of Treves, was spoken neither in the West nor in Aquitaine. It was found only in the centre and south of what is now France, and was itself probably split up into two great divisions. It is the common source of the currents, more or less Latinized, which were mingled with other elements in different proportions, and formed later the langue d’oil and the lingua romana, in the narrower sense. I will speak first of the latter.
In order to bring it into being, all that was necessary was a slight alteration in the vocabulary of Latin, and the introduction of a few syntactical notions borrowed from Celtic and other languages till then unknown in the West of Europe. The Imperial colonies had brought in a fair number of Italian, African, and Asiatic elements. The Burgundian, and especially the Gothic, invasions added another, which was marked by considerable harmony, liveliness, and sonority. Its vocabulary was further increased after the inroads of the Saracens. Thus the lingua romana became, in its rhythmic quality, quite distinct from Gallic, and soon assumed a character of its own. It is true that we do not find this in its perfection, in the “Oath of the Sons of Ludwig the Pious,” as we do later in the poems of Raimbaut de Vaqueiras or Bertran de Born.[[134]] Yet even in the “Oath” we can recognize the language for what it is; it has already acquired its main features, and its future path is clearly mapped out. It formed henceforth (in its different dialects of Limousin, Provençal, and Auvergnat) the speech of a people of as mixed an origin as any in the world. It was a refined and supple language, witty, brilliant, and satirical, but without depth or philosophy. It was of tinsel rather than gold, and had never been able to do more than pick up a few ingots on the surface of the rich mines that lay open to it. Without any serious principles, it was destined to remain an instrument of indifference, of universal scepticism and mockery. It did not fail to be used as such. The people cared for nothing but pleasure and parade. Brave to a fault, beyond measure gay, spending their passion on a dream, and their vitality on idle toys, they had an instrument that was exactly suited to their character, and which, though admired by Dante, was put to no better use in poetry than to tag satires, love-songs, and challenges, and in religion to support heresies such as that of the Albigenses, a pestilent Manicheism, without value even for literature, from which an English author, in no way Catholic in his sympathies, congratulates the Papacy on having delivered the Middle Ages.[[135]] Such was the lingua romana of old, and such do we find it even to-day. It is pretty rather than beautiful, and shows on the surface how little it is fitted to serve a great civilization.
Was the langue d’oil formed in a similar way? Obviously not. However the Celtic, Latin, and Germanic elements were fused (for we cannot be certain on this point, in the absence of records going back to the earliest period of the language[[136]]), it is at any rate clear that it rose from a strongly marked antagonism between the three tongues, and that it would thus have a character and energy quite incompatible with such compromises and adaptations as those which gave birth to the lingua romana. In one moment of its life, the langue d’oil was partly a Germanic tongue. In the written remains that have survived, we find one of the best qualities of the Aryan languages, the power of forming compounds. This power, it is true, is limited; and though still considerable, is less than in Sanscrit, Greek, and German. In the nouns, we find a system of inflexion by suffix, and, in consequence, an ease in inverting the order which modern French has lost, and which the language of the sixteenth century retained only to a slight extent, its inversions being gained at the expense of clearness. Again, the vocabulary of the langue d’oil included many words brought in by the Franks.[[137]] Thus it began by being almost as much Germanic as Gallic; Celtic elements appeared in its second stage, and perhaps fixed the melodic principles of the language. The best possible tribute to its merits is to be found in the successful experiment of Littré,[[138]] who translated the first book of the “Iliad” literally, line for line, into French of the thirteenth century. Such a tour de force would be impossible in modern French.
Such a language belonged to a people that was evidently very different from the inhabitants of Southern Gaul. It was more deeply attached to Catholicism; its politics were permeated by a lively idea of freedom, dignity, and independence, its institutions had no aim but utility. Thus the mission set before the popular literature was not to express the fancies of the mind or heart, the freakishness of a universal scepticism, but to put together the annals of the nation, and to set down what was at that time regarded as the truth. It is to this temper of the people and their language that we owe the great rhymed chronicles, especially “Garin le Loherain,” which bear witness, though it has since been denied, to the predominance of the North. Unfortunately, since the compilers of these traditions, and even their original authors, mainly aimed at preserving historical facts or satisfying their desire for positive and solid results, poetry in the true sense, the love of form and the search for beauty, does not always bulk as large as it should in their long narratives. The literature of the langue d’oil was, above all, utilitarian; and so the race, the language, and the literature were in perfect harmony.
The Germanic element in the race, however, being far less than the Gallic basis or the Roman accretions, naturally began to lose ground. The same thing took place in the language; Celtic and Latin advanced, Germanic retreated. That noble speech, which we know only at its highest stage, and which might have risen even higher, began to decline and become corrupted towards the end of the thirteenth century. In the fifteenth, it was no more than a patois, from which the Germanic elements had completely disappeared. The treasury was exhausted; and what remained was an illogical and barbarous anomaly in the midst of the progress of Celtic and Latin. Thus in the sixteenth century the revival of classical studies found the language in ruins, and tried to remodel it on the lines of Greek and Latin. This was the professed aim of the writers of this great age. They did not succeed, and the seventeenth century, wisely seeing that the irresistible march of events could in no wise be curbed by the hand of man, set itself merely to improve the language from within; for every day it was assuming more and more the forms best suited to the dominant race, the forms, in other words, into which the grammatical life of Celtic had formerly been cast.
Although both the langue d’oil and French proper are marked by a greater unity than the lingua romana (since the mixture of races and languages that gave birth to them was less complex) yet they have produced separate dialects which survive to this day. It is not doing these too much honour to call them dialects, not patois. They arose, not from the corruption of the dominant type, with which they were at least contemporary, but from the different proportions in which the Celtic, Latin, and Germanic elements, that still make up the French nationality, were mingled. To the north of the Seine, we find the dialect of Picardy; this is, in vocabulary and rhythmic quality, very near Flemish, of which the Germanic character is too obvious to be dwelt upon. Flemish, in this respect, shows the same power of choice as the langue d’oil, which could in a certain poem, without ceasing to be itself, admit forms and expressions taken bodily from the language spoken at Arras.[[139]]
As we go south of the Seine towards the Loire, the Celtic elements in the provincial dialects grow more numerous. In Burgundian, and the dialects of Vaud and Savoy, even the vocabulary has many traces of Celtic; these are not found in French, where the predominant factor is rustic Latin.[[140]]
I have shown above[[141]] how from the sixteenth century the influence of the north had given ground before the growing preponderance of the peoples beyond the Loire. The reader has merely to compare the present sections on language with my former remarks on blood to see how close is the relation between the speech of a people and its physical constitution.[[142]]