Chapter I (2)

27. It may render this more clear, if we mention a few of the growing corruptions, which have in fact transformed the Latin into French and the sister tongues.--The prepositions were used with no regard to the proper inflexions of nouns and verbs. These were known so inaccurately, and so constantly put one for another, that it was necessary to have recourse to prepositions instead of them. Thus _de_ and _ad_ were made to express the genitive and dative cases, which is common in charters from the sixth to the tenth century. It is a real fault in the Latin language, that it wants both the definite and indefinite article; _ille_ and _unus_, especially the former, were called in to help this deficiency. In the forms of Marculfus, published towards the end of the seventh century, _ille_ continually occurs as an article; and it appears to have been sometimes used in the sixth. This of course, by an easy abbreviation, furnished the articles in French and Italian. The people came soon to establish more uniformity of case in the noun, either by rejecting inflexions, or by diminishing their number.--Raynouard gives a long list of old French nouns formed from the Latin accusative by suppressing _em_ or _am_.[44] The active auxiliary verb, than which nothing is more distinctive of the modern languages from the Latin, came in from the same cause, the disuse, through ignorance, of several inflexions of the tenses; to which we must add, that here also the Latin language is singularly deficient, possessing no means of distinguishing the second perfect from the first, or ‘I have seen’ from ‘I saw.’ The auxiliary verb was early applied, in France and Italy, to supply this defect; and some have produced what they think occasional instances of its employment even in the best classical authors.

[44] See a passage of Quintilian, l. 9, c. 4, quoted in Hallam’s Middle
Ages, iii. 316.

In the grammar of Cassiodorus, a mere compilation from old writers,
and in this instance from one Cornutus, we find another remarkable
passage, which I do not remember to have seen quoted, though
doubtless it has been so, on the pronunciation of the letter
_M_. To utter this final consonant, he says, before a word
beginning with a vowel, is wrong, durum ac barbarum sonat; but it
is an equal fault to omit it before one beginning with a consonant;
par enim atque idem est vitium, ita cum vocali sicut cum consonanti
_M_ literam, exprimere. Cassiodorus, De orthographia, cap. 1.
Thus we perceive that there was a nicety as to the pronunciation of
this letter, which uneducated persons would naturally not regard.
Hence in the inscriptions of a low age, we frequently find this
letter omitted; as in one quoted by Muratori, Ego L. Contius me
bibo [vivo] archa [archam] feci, and it is very easy to multiply
instances. Thus the neuter and the accusative terminations were
lost.

|Continuance of Latin in seventh century.|

28. It seems impossible to determine the progress of these changes, the degrees of variation between the polite and popular, the written and spoken Latin, in the best ages of Rome, in the decline of the empire, and in the kingdoms founded upon its ruins; or finally, the exact epoch when the grammatical language ceased to be generally intelligible. There remains, therefore, some room still for hypothesis and difference of opinion. The clergy preached in Latin early in the seventh century, and we have a popular song of the same age on the victory obtained by Clotaire II. in 622 over the Saxons.[45] This has been surmised by some to be a translation, merely because the Latin is better than they suppose to have been spoken. But, though the words are probably not given quite correctly, they seem reducible, with a little emendation, to short verses of an usual rythmical cadence.[46]

[45] Le Bœuf, in Mém. de l’Acad. des Inscript. vol. xvii.

[46] Turner, in Archæologia, vol. xiv. 173. Hallam’s Middle Ages, iii.
326. Bouterwek, Gesch. der Französen Poesie, p. 18, observes,
that there are many fragments of popular Latin songs preserved. I
have not found any quoted, except one, which he gives from La
Revaillère, which is simple and rather pretty; but I know not
whence it is taken. It seems the song of a female slave, and is
perhaps nearly as old as the destruction of the empire.

At quid jubes, pusiole,
Quare mandas, filiole,
Carmen dulce me cantare
Cum sim longe exul valde
Intra mare,
O cur jubes canere?

Intra seems put for trans. The metre is rhymed trochaic; but that
is consistent with antiquity. It is, however, more pleasing than
most of the Latin verse of this period, and is more in the tone of
the modern languages. As it is not at all a hackneyed passage, I
have thought it worthy of quotation.

|It is changed to a new language in eighth and ninth.|

29. But in the middle of the eighth century, we find the rustic language mentioned as distinct from Latin;[47] and in the council of Tours held in 813 it is ordered that homilies shall be explained to the people in their own tongue, whether rustic Roman or Frankish. In 842 we find the earliest written evidence of its existence, in the celebrated oaths taken by Louis of Germany and his brother Charles the Bald, as well as by their vassals, the former in Frankish or early German, the latter in their own current dialect. This, though with somewhat of a closer resemblance to Latin, is accounted by the best judges a specimen of the language spoken south of the Loire; afterwards variously called the Langue d’oc, Provençal, or Limousin, and essentially the same with the dialects of Catalonia and Valencia.[48] It is decidedly the opinion of M. Raynouard, as it was of earlier inquirers, that the general language of France in the ninth century was the southern dialect, rather than that of the north, to which we now give the exclusive name of French, and which they conceive to have deviated from it afterwards.[49] And he has employed great labour to prove, that, both in Spain and Italy, this language was generally spoken with hardly as much difference from that of France, as constitutes even a variation of dialect; the articles, pronouns, and auxiliaries being nearly identical; most probably not with so much difference as would render the native of one country by any means unintelligible in another.[50]

[47] Acad. des. Inscript. xvii. 713.

[48] Du Cange, p. 35. Raynouard, passim. M. de la Rue has called it,
“un Latin expirant.” Recherches sur les Bardes d’Armorique. Between
this and “un Français naissant” there may be only a verbal
distinction; but, in accuracy of definition, I should think M.
Raynouard much more correct. The language of this oath cannot be
called Latin without a violent stretch of words: no Latin scholar,
as such, would understand it, except by conjecture. On the other
hand, most of the words, as we learn from M. R., are Provençal of
the twelfth century. The passage has been often printed, and
sometimes incorrectly. M. Roquefort, in the preface to his
Glossaire de la Langue Romane, has given a tracing from an ancient
manuscript of Nitard, the historian of the ninth century, to whom
we owe this important record of language.

[49] The chief difference was in orthography; the Northerns wrote
Latin words with an _e_ where the South retained _a_; as charitet,
caritat: veritet, veritat; appelet, apelat. Si l’on rétablissait
dans les plus anciens textes Français les _a_ primitifs en place
des _e_, on aurait identiquement la langue des troubadours.
Raynouard, Observations sur le Roman du Rou, 1829, p. 5.

[50] The proofs of this similarity occupy most part of the first and
sixth volumes in M. Raynouard’s excellent work.

It is a common error to suppose that French and Italian had a
double source, barbaric as well as Latin; and that the northern
nations, in conquering those regions, brought in a large share of
their own language. This is like the opinion, that the Norman
Conquest infused the French we now find in our own tongue. There
are certainly Teutonic words, both in French and Italian, but not
sufficient to affect the proposition that these languages are
merely Latin in their origin. These words in many instances express
what Latin could not; thus _guerra_ was by no means synonymous
with _bellum_. Yet even Roquefort talks of “un jargon composé
de mots Tudesques et Romains.” Discours Preliminaire, p. 19;
forgetting which, he more justly remarks afterwards, on the oath of
Charles the Bald, that it shows “la langue Romane est entièrement
composée de Latin.” A long list could, no doubt, be made of French
and Italian words that cannot easily be traced to any Latin with
which we are acquainted; but we may be surprised that it is not
still longer.

|Early specimens of French.|

|Poem on Boethius.|

30. Thus, in the eighth and ninth centuries, if not before, France had acquired a language unquestionably nothing else than a corruption of Latin, (for the Celtic or Teutonic words that entered into it were by no means numerous, and did not influence its structure), but become so distinct from its parent, through modes of pronunciation as well as grammatical changes, that it requires some degree of practice to trace the derivation of words in many instances. It might be expected that we should be able to adduce, or at least prove to have existed, a series of monuments in this new form of speech. It might naturally appear that poetry, the voice of the soul, would have been heard wherever the joys and sufferings, the hopes and cares of humanity, wherever the countenance of nature, or the manners of social life, supplied their boundless treasures to its choice; and among untutored nations it has been rarely silent. Of the existence of verse, however, in this early period of the new languages, we find scarce any testimony, a doubtful passage in a Latin poem of the ninth century excepted,[51] till we come to a production on the captivity of Boethius, versified chiefly from passages in his Consolation, which M. Raynouard, though somewhat wishing to assign a higher date, places about the year 1000. This is printed by him from a manuscript formerly in the famous abbey of Fleury, or St. Benoit-sur-Loire, and now in the public library of Orleans. It is a fragment of 250 lines, written in stanzas of six, seven, or a greater number of verses of ten syllables, sometimes deviating to eleven or twelve; and all the lines in each stanza rhyming masculinely with each other. It is certainly by much the earliest specimen of French verse;[52] even if it should only belong, as Le Bœuf thought, to the eleventh century.

[51] In a Latin eclogue quoted by Paschasius Radbert (ob. 865) in the
life of St. Adalhard, abbot of Corbie (ob. 826), the romance poets
are called upon to join the Latins in the following lines: “Rustica
concelebret Romana Latinaque lingua, Saxo, qui, pariter plangens,
pro carmine dicat; Vertite huc cuncti, cecinit quam maximus ille,
Et tumulum facite, et tumulo superaddite carmen.”

Raynouard, Choix des Poésies, vol. ii. p. cxxxv. These lines are
scarcely intelligible; but the quotation from Virgil, in the ninth
century, perhaps deserves remark, though, in one of Charlemagne’s
monasteries, it is not by any means astonishing. Nennius, a Welsh
monk of the same age, who can hardly write Latin at all, has quoted
another line; “Purpurea intexti tollant aulæa a Britanni;” which is
more extraordinary, and almost leads us to suspect an
interpolation, unless he took it from Bede. Gale, xv. Scriptores,
iii. 102.

[52] Raynouard, vol. ii. pp. 5, 6, and preface, p. cxxvii.

|Provençal grammar.|

31. M. Raynouard has asserted what will hardly bear dispute, that “there has never been composed any considerable work in any language, till it has acquired determinate forms of expressing the modifications of ideas according to time, number, and person,” or, in other words, the elements of grammar.[53] But whether the Provençal or Romance language were in its infancy so defective, he does not say; nor does the grammar he has given lead us to that inference. This grammar, indeed, is necessarily framed, in great measure, out of more recent materials. It may be suspected, perhaps, that a language formed by mutilating the words of another, could not for many ages be rich or flexible enough for the variety of poetic expression. And the more ancient forms would long retain their prerogative in writing: or, perhaps, we can only say, that the absence of poetry was the effect, as well as the evidence, of that intellectual barrenness, more characteristic of the dark ages than their ignorance.

[53] Observations philogiques et grammaticales, sur le Roman de Rou
(1829), p. 26. Two ancient Provençal grammars, one by Raymond Vidal
in the twelfth century, are in existence. The language therefore
must have had its determinate rules before that time.

M. Raynouard has shown, with a prodigality of evidence, the
regularity of the French or Romance language in the twelfth
century, and its retention of Latin forms, in cases when it had not
been suspected. Thus it is a fundamental rule, that, in nouns
masculine, the nominative ends in s in the singular, but wants it
in the plural; while the oblique cases lose it in the singular, but
retain it in the plural. This is evidently derived from the second
declension in Latin. As, for example--

Sing. Li princes est venus, et a este sacrez rois.
Plur. Li evesque et li plus noble baron se sont assemble.

Thus also the possessive pronoun is always _mes_, _tes_,
_ses_, (meus, tuus, suus) in the nominative singular;
_mon_, _ton_, _son_, (meum, &c.), in the oblique
regimen. It has been through ignorance of such rules that the old
French poetry has seemed capricious, and destitute of strict
grammar; and, in a philosophical sense, the simplicity and
extensiveness of M. Raynouard’s discovery entitle it to the
appellation of beautiful.

|Latin retained in use longer in Italy.|

32. In Italy, where we may conceive the corruption of language to have been less extensive, and where the spoken patois had never acquired a distinctive name, like _lingua Romana_ in France, we find two remarkable proofs, as they seem, that Latin was not wholly unintelligible in the ninth and tenth centuries, and which therefore modify M. Raynouard’s hypothesis as to the simultaneous origin of the Romance tongue. The one is a popular song of the soldiers, on their march to rescue the Emperor Louis II. in 881, from the violent detention in which he had been placed by the duke of Benevento; the other, a similar exhortation to the defenders of Modena in 924, when that city was in danger of siege from the Hungarians. Both of these were published by Muratori, in his fortieth dissertation on Italian Antiquities; and both have been borrowed from him by M. Sismondi, in his Littérature du Midi.[54] The former of these poems is in a loose trochaic measure, totally destitute of regard to grammatical inflections. Yet some of the leading peculiarities of Italian, the article and the auxiliary verb, do not appear. The latter is in accentual iambics, with a sort of monotonous termination in the nature of rhyme; and in very much superior Latinity, probably the work of an ecclesiastic.[55] It is difficult to account for either of these, especially the former, which is merely a military song, except on the supposition that the Latin language was not grown wholly out of popular use.

[54] Vol. i. pp. 23, 27.

[55] I am at a loss to know what Muratori means by saying, “Son versi
di dodici sillabe, ma computata la ragione de’ tempi, vengono ad
essere uguali a gli endecasillabi.” p. 551. He could not have
understood the metre, which is perfectly regular, and even
harmonious, on the condition only, that no “ragione de’ tempi”
except such as accentual pronunciation observes, shall be demanded.
The first two lines will serve as a specimen:--

“O tu, qui servas armis ista mænia,
Noli dormire, moneo, sed vigila.”

This is like another strange observation of Muratori in the same
dissertation, that, in the well-known lines of the emperor Adrian
to his soul, “Animula vagula, blandula,” which could perplex no
schoolboy, he cannot discover “un’esatta norma di metro;” and
therefore takes them to be merely rhythmical.

|French of eleventh century.|

33. In the eleventh century, France still affords us but few extant writings. Several, indeed, can be shown to have once existed. The Romance language, comprehending the two divisions of Provençal and Northern French, by this time distinctly separate from each other, was now, say the authors of the Histoire Littéraire de la France, employed in poetry, romances, translations, and original works in different kinds of literature; sermons were preached in it, and the code, called the Assizes de Jerusalem, was drawn up under Godfrey of Bouillon in 1100.[56] Some part of this is doubtful, and especially the age of these laws. They do not mention those of William the Conqueror, recorded in French by Ingulfus. Doubts have been cast by a distinguished living critic on the age of this French code, and upon the authenticity of the History of Ingulfus itself; which he conceives, upon very plausible grounds, to be a forgery of Richard II.’s time: the language of the laws indeed appears to be very ancient, but not probably distinguishable at this day from the French of the twelfth century. It may be said, in general, that, except one or two translations from books of Scripture, very little now extant has been clearly referred to an earlier period.[57] Yet it is impossible to doubt that the language was much employed in poetry, and had been gradually ramifying itself by the shoots of invention and sentiment; since, at the close of this age, or in the next, we find a constellation of gay and brilliant versifiers, the Troubadours of southern France, and a corresponding class to the north of the Loire.

[56] Vol. vii. p. 107.

[57] Roquefort, Glossaire de la Langue Romane, p. 25, and État de la
Poésie Française, p. 42, and 206, mentions several religious works
in the royal library, and also a metrical romance in the British
Museum, lately published in France on the fabulous voyage of
Charlemagne to Constantinople. Raynouard has collected a few
fragments in Provençal. But I must dissent from this excellent
writer in referring the famous poem of the Vaudois, La Nobla
Leyczon, to the year 1100. Choix des Poésies des Troubadours, vol.
ii. p. cxxxvii. I have already observed, that the two lines which
contain what he calls la date de l’an 1100, are so loosely
expressed, as to include the whole ensuing century. (Hallam’s
Middle Ages, iii. 467.) And I am now convinced that the poem is not
much older than 1200. It seems probable that they reckoned 1100
years, on a loose computation, not from the Christian era, but from
the time when the passage of Scripture to which these lines allude
was written. The allusion may be to 1 Pet. i. 20. But it is clear
that, at the time of the composition of this poem, not only the
name of _Vaudois_ had been imposed on those sectaries, but they had
become subject to persecution. We know nothing of this till near
the end of the century. This poem was probably written in the south
of France, and carried afterwards to the Alpine valleys of
Piedmont, from which it was brought to Geneva and England in the
seventeenth century. La Nobla Leyczon is published at length by
Raynouard. It consists of 479 lines, which seem to be rhythmical or
aberrant Alexandrines; the rhymes uncertain in number, chiefly
masculine. The poem censures the corruptions of the church, but
contains little that would be considered heretical; which agrees
with what contemporary historians relate of the original Waldenses.
Any doubts as to the authenticity of this poem are totally
unreasonable. M. Raynouard, an indisputably competent judge,
observes, “Les personnes qui l’examineront avec attention jugeront
que le manuscrit n’a pas été interpolé,” p. cxliii.

I will here reprint more accurately than before the two lines
supposed to give the poem the date of 1100:--

“Ben ha mil et cent ancz compli entièrement,
Que fo scripta l’ora car sen al derier temps.”

Can M. Raynouard, or any one else, be warranted by this in saying,
_La date de l’an 1100_, qu’on lit dans ce poème, merite toute
confiance?

|Metres of modern languages.|

34. These early poets in the modern languages chiefly borrowed their forms of versification from the Latin. It is unnecessary to say, that metrical composition in that language, as in Greek, was an arrangement of verses corresponding by equal or equivalent feet; all syllables being presumed to fall under a known division of long and short, the former passing for strictly the double of the latter in quantity of time. By this law of pronunciation all verse was measured; and to this not only actors, who were assisted by an accompaniment, but the orators also endeavoured to conform. But the accented, or, if we choose rather to call them so, emphatic syllables, being regulated by a very different though uniform law, the uninstructed people, especially in the decline of Latinity, pronounced, as we now do, with little or no regard to the metrical quantity of syllables, but according to their accentual value. And this gave rise to the popular or rhythmical poetry of the lower empire; traces of which may be found in the second century, and even much earlier, but of which we have abundant proofs after the age of Constantine.[58] All metre, as Augustin says, was rhythm, but all rhythm was not metre: in rhythmical verse, neither the quantity of syllables, that is, the time allotted to each by metrical rule, nor even, in some degree, their number, was regarded, so long as a cadence was retained in which the ear could recognise a certain approach to uniformity. Much popular poetry, both religious and profane, and the public hymns of the church, were written in this manner; the distinction of long and short syllables, even while Latin remained a living tongue, was lost in speech, and required study to attain it. The accent or emphasis, both of which are probably, to a certain extent, connected with quantity and with each other, supplied its place; the accented syllable being, perhaps, generally lengthened in ordinary speech; though this is not the sole cause of length, for no want of emphasis or lowness of tone can render a syllable of many letters short. Thus we find two species of Latin verse: one metrical, which Prudentius, Fortunatus, and others aspired to write; the other rhythmical, somewhat licentious in number of syllables, and wholly accentual in its pronunciation. But this kind was founded on the former, and imitated the ancient syllabic arrangements. Thus the trochaic, or line, in which the stress falls on the uneven syllables, commonly alternating by eight and seven, a very popular metre from its spirited flow, was adopted in military songs, such as that already mentioned of the Italian soldiers in the ninth century. It was also common in religious chants. The line of eight syllables, or dimeter iambic, in which the cadence falls on the even places, was still more frequent in ecclesiastical verse. But these are the most ordinary forms of versification in the early French or Provençal, Spanish, and Italian languages. The line of eleven syllables, which became in time still more usual than the former, is nothing else than the ancient hendecasyllable; from which the French, in what they call masculine rhymes, and ourselves more generally, from a still greater deficiency of final vowels, have been forced to retrench the last syllable. The Alexandrine of twelve syllables might seem to be the trimeter iambic of the ancients. But Sanchez has very plausibly referred its origin to a form more usual in the dark ages, the pentameter; and shown it in some early Spanish poetry.[59] The Alexandrine, in the southern languages, had generally a feminine termination, that is, in a short vowel, thus becoming of thirteen syllables, the stress falling on the penultimate, as is the usual case in a Latin pentameter verse, accentually read in our present mode. The variation of syllables in these Alexandrines, which run from twelve to fourteen, is accounted for by the similar numerical variety in the pentameter.

[58] The well-known lines of Adrian to Florus, and his reply, “Ego
nolo Florus esse,” &c., are accentual trochaics, but not wholly so;
for the last line, Scythicas pati pruinas, requires the word pati
to be sounded as an iambic. They are not the earliest instance
extant of disregard to quantity, for Suetonius quotes some
satirical lines on Julius Cæsar.

[59] The break in the middle of the Alexandrine, it will occur to
every competent judge, has nothing analogous to it in the trimeter
iambic, but exactly corresponds to the invariable law of the
pentameter.

|Origin of rhyme in Latin.|

35. I have dwelt, perhaps tediously, on this subject, because vague notions of a derivation of modern metrical arrangements, even in the languages of Latin origin, from the Arabs or Scandinavians, have sometimes gained credit.[60] It has been imagined also that the peculiar characteristic of the new poetry, rhyme, was borrowed from the Saracens of Spain.[61] But the Latin language abounds so much in consonances, that those who have been accustomed to write verses in it well know the difficulty of avoiding them, as much as an ear formed on classical models demands; and as this gingle is certainly pleasing in itself, it is not wonderful that the less fastidious vulgar should adopt it in their rhythmical songs. It has been proved by Muratori, Gray, and Turner, beyond the possibility of doubt, that rhymed Latin verse was in use from the end of the fourth century.[62]

[60] Roquefort, Essai sur la Poésie Française dans le 12me et 13me
siècles, p. 66. Galvani, Osservazioni sulla poesia de’ Trovatori.
(Modena, 1829) Sanchez, Poesias Castellanas anteriores al 15mo
siglo, vol. i. p. 122.

Tyrwhitt had already observed, “The metres which the Normans used,
and which we seem to have borrowed from them, were plainly copied
from the Latin rhythmical verses, which, in the declension of that
language, were current in various forms among those who either did
not understand, or did not regard, the true quantity of syllables;
and the practice of rhyming is probably to be deduced from the same
original.” Essay on the Language and Versification of Chaucer, p.
51.

[61] Andrès, with a partiality to the Saracens of Spain, whom, by an
odd blunder, he takes for his countrymen, manifested in almost
every page, does not fail to urge this. It had been said long
before by Huet, and others who lived before these subjects had been
thoroughly investigated. Origine e Progresso, &c., ii. 194. He has
been copied by Ginguéné and Sismondi.

[62] Muratori, Antichità Italiane dissert., 40. Turner, in Archæologia,
vol. xiv., and Hist. of England, vol. iv. pp. 328, 653. Gray has
gone as deeply as any one into this subject; and, though writing at
what may be called an early period of metrical criticism, he has
fallen into a few errors, and been too easy of credence,
unanswerably proves the Latin origin of rhyme. Gray’s Works by
Mathias, vol. ii. p. 30-54.

|Provençal and French poetry.|

36. Thus, about the time of the first crusade, we find two dialects of the same language, differing by that time not inconsiderably from each other, the Provençal and French, possessing a regular grammar, established forms of versification (and the early troubadours added several to those borrowed from the Latin[63]), and a flexibility which gave free scope to the graceful turns of poetry. William, duke of Guienne, has the glory of leading the van of surviving Provençal songsters. He was born in 1070, and may possibly have composed some of his little poems before he joined the crusaders in 1096. If these are genuine, and no doubt of them seems to be entertained, they denote a considerable degree of previous refinement in the language.[64] We do not, I believe, meet with any other troubadour till after the middle of the twelfth century. From that time till about the close of the thirteenth, they were numerous almost as the gay insects of spring; names of illustrious birth are mingled in the list with those whom genius has saved from obscurity; they were the delight of a luxurious nobility, the pride of southern France, while the great fiefs of Toulouse and Guienne were in their splendour. Their style soon extended itself to the northern dialect. Abelard was the first of recorded name, who taught the banks of the Seine to resound a tale of love; and it was of Eloise that he sung.[65] “You composed,” says that gifted and noble-spirited woman, in one of her letters to him, “many verses in amorous measure, so sweet both in their language and their melody, that your name was incessantly in the mouths of all, and even the most illiterate could not be forgetful of you. This it was chiefly that made women admire you. And as most of these songs were on me and my love, they made me known in many countries, and caused many women to envy me. Every tongue spoke of your Eloise; every street, every house resounded with my name.”[66] These poems of Abelard are lost; but in the Norman, or northern French language, we have an immense number of poets belonging to the twelfth, and the two following centuries. One hundred and twenty-seven are known by name in the twelfth alone.[67] Thibault, king of Navarre and count of Champagne, about the middle of the next, is accounted the best, as well as noblest of French poets.

[63] See Raynouard, Roquefort, and Galvini, for the Provençal and
French metres, which are very complicated.

[64] Raynouard, Choix des Poésies des Troubadours, vol. ii. Auguis,
Recueil des Anciens Poètes Français, vol. i.

[65] Bouterwek, on the authority of La Ravaillere, seems to doubt
whether these poems of Abelard were in French or Latin. Gesch. der
Französen Poesie, p. 18. I believe this would be thought quite
paradoxical by any critic at present.

[66] Duo autem, fateor, tibi specialiter inerant, quibus feminarum
quarumlibet animos statim allicere poteras, dictandi videlicet et
cantandi gratia; quæ cæteros minimè philosophos assecutos esse
novimus. Quibus quidem quasi ludo quodam laborem exercitii recreans
philosophici pleraque amatorio metro vel rithmo composita
reliquisti carmina, quæ præ nimiâ suavitate tam dictaminis quam
cantus sæpius frequentata tuum in ore omnium nomen incessanter
tenebant, ut etiam illiteratos melodiæ dulcedo tui non sineret
immemores esse. Atque hinc maxime in amorem tui feminæ suspirabant.
Et cum horum pars maxima carminum nostros decantaret amores, multis
me regionibus brevi tempore nunciavit, et multarum in me feminarum
accendit invidiam. And in another place: Frequenti carmine tuam in
ore omnium Heloissam ponebas: me plateæ omnes, me domus singulæ
resonabant. Epist. Abælardi et Heloissæ. These epistles of Abelard
and Eloisa, especially those of the latter, are, as far as I know,
the first book that gives any pleasure in reading which had been
produced in Europe for 600 years, since the Consolation of
Boethius, But I do not press my negative judgment. We may at least
say that the writers of the dark ages, if they have left anything
intrinsically very good, have been ill-treated by the learned, who
have failed to extract it. Pope, it may be here observed, has done
great injustice to Eloisa in his unrivalled Epistle, by putting the
sentiments of a coarse and abandoned woman into her mouth. Her
refusal to marry Abelard arose not from an abstract predilection
for the name of mistress above that of wife, but from her
disinterested affection, which would not deprive him of the
prospect of ecclesiastical dignities, to which his genius and
renown might lead him. She judged very unwisely, as it turned out,
but from an unbounded generosity of character. He was, in fact,
unworthy of her affection, which she expresses in the tenderest
language. Deum testem invoco, si me Augustus universo præsidens
mundo matrimonii honore dignaretur, totumque mihi orbem confirmaret
in perpetuum præsidendum, charius mihi et dignius videretur tua
dici meretrix quam illius imperatrix.

[67] Auguis, Discours Préliminaire, p. 2. Roquefort, Etat de la Poésie
Française aux 12me et 13me siècles.

37. In this French and Provençal poetry, if we come to the consideration of it historically, descending from an earlier period, we are at once struck by the vast preponderance of amorous ditties. The Greek and Roman muses, especially the latter, seem frigid as their own fountain in comparison. Satires on the great, and especially, on the clergy, exhortations to the crusade, and religious odes, are intermingled in the productions of the troubadours; but love is the prevailing theme. This tone they could hardly have borrowed from the rhythmical Latin verses, of which all that remain are without passion or energy. They could as little have been indebted to their predecessors for a peculiar gracefulness, an indescribable charm of gaiety and ease, which many of their lighter poems display. This can only be ascribed to the polish of chivalrous manners, and to the influence of feminine delicacy on public taste. The well-known dialogue, for example, of Horace and Lydia, is justly praised; nothing extant of this amœbean character, from Greece or Rome, is nearly so good. But such alternate stanzas, between speakers of different sexes, are very common in the early French poets; and it would be easy to find some quite equal to Horace in grace and spirit. They had even a generic name, _tensons_, contentions; that is, dialogues of lively repartee, such as we are surprised to find in the twelfth century, an age accounted by many almost barbarous. None of these are prettier than what are called _pastourelles_, in which the poet is feigned to meet a shepherdess, whose love he solicits, and by whom he is repelled, (not always finally,) in alternate stanzas.[68] Some of these may be read in Roquefort, Etat de la Poésie Française, dans le 12me et 13me siècles; others in Raynouard, Choix des Poésies des Troubadours; in Auguis, Recueil des Anciens Poètes Français; or in Galvani, Osservazioni sulla Poesia de’ Trovatori.

[68] These have, as Galvani has observed, an ancient prototype in the
twenty-seventh pastoral of Theocritus, which Dryden has translated
with no diminution of its freedom. Some of the Pastourelles are
also rather licentious; but that is not the case with the greater
part. M. Raynouard, in an article of the Journal des Savans for
1824, p. 613, remarks the superior decency of the southern poets,
scarcely four or five transgressing in that respect; while many of
the fabliaux in the collections of Barbazan and Méon are of the
most coarse and stupid ribaldry; and such that even the object of
exhibiting ancient manners and language scarcely warranted their
publication in so large a number.

38. In all these light compositions which gallantry or gaiety inspired, we perceive the characteristic excellencies of French poetry, as distinctly as in the best vaudeville of the age of Louis XV. We can really sometimes find little difference, except an obsoleteness of language, which gives them a kind of poignancy. And this style, as I have observed, seems to have been quite original in France, though it was imitated by other nations.[69] The French poetry, on the other hand, was deficient in strength and ardour. It was also too much filled with monotonous commonplaces; among which the tedious descriptions of spring, and the everlasting nightingale, are eminently to be reckoned. These, perhaps, are less frequent in the early poems, most of which are short, than they became in the prolix expansion adopted by the allegorical school in the fourteenth century. They prevail, as is well known, in Chaucer, Dunbar, and several other of our own poets.

[69] Andrès, as usual, derives the Provençal style of poetry from the
Arabians; and this has been countenanced, in some measure, by
Ginguéné and Sismondi. Some of the peculiarities of the Trobadours,
their tensons, or contentions, and the envoi, or termination of a
poem, by an address to the poem itself or the reader, are said to
be of Arabian origin. In assuming that rhyme was introduced by the
same channel, these writers are probably mistaken. But I have seen
too little of oriental, and, especially, of Hispano-Saracenic
poetry, to form any opinion how far the more essential
characteristics of Provençal verse may have been derived from it.
One seems to find more of oriental hyperbole in the Castilian
poetry.

|Metrical romances. Havelok the Dane.|

39. The metrical romances, far from common in Provençal,[70] but forming a large portion of what was written in the northern dialect, though occasionally picturesque, graceful, or animated, are seldom free from tedious or prosaic details. The earliest of these extant seems to be that of Havelok the Dane, of which an abridgment was made by Geoffrey Gaimar, before the middle of the twelfth century. The story is certainly a popular legend from the Danish part of England, which the French versifier has called, according to the fashion of romances, “a Breton lay.” If this word meant anything more than relating to Britain, it is a plain falsehood; and upon either hypothesis, it may lead us to doubt, as many other reasons may also, what has been so much asserted of late years, as to the Armorican origin of romantic fictions; since the word Breton, which some critics refer to Armorica, is here applied to a story of mere English birth.[71] It cannot, however, be doubted, from the absurd introduction of Arthur’s name in this romance of Havelok, that it was written after the publication of the splendid fables of Geoffrey.[72]

[70] It has been denied that there are any metrical romances in
Provençal. But one called the Philomena, on the fabulous history of
Charlemagne, is written after 1173, but not much later than 1200.
Journal des Savans, 1824.

[71] The Recherches sur les Bardes d’Armorique, by that respectable
veteran, M. de la Rue, are very unsatisfactory. It does not appear
that the Bretons have so much as a national tradition of any
romantic poetry; nor any writings in their language older than
1450. The authority of Warton, Leyden, Ellis, Turner, and Price
have rendered this hypothesis of early Armorican romance popular;
but I cannot believe that so baseless a fabric will endure much
longer. Is it credible that tales of aristocratic splendour and
courtesy sprung up in so poor and uncivilised a country as
Bretagne? Traditional stories they might, no doubt, possess, and
some of these may be found in the lais de Marie, and other early
poems; but not romances of chivalry. I do not recollect, though
speaking without confidence, that any proof has been given of
Armorican traditions about Arthur, earlier than the history of
Geoffrey: for it seems too much to interpret the word _Britones_ of
them rather than of the Welsh. Mr. Turner, I observe, without
absolutely recanting, has much receded from his opinion of the
Armorican prototype of Geoffrey of Monmouth.

[72] The romance of Havelok was printed by Sir Frederick Madden in
1829; but not for sale. His Introduction is of considerable value.
The story of Havelok is that of Curan and Argentile, in Warner’s
Albion’s England, upon which Mason founded a drama. Sir F. Madden
refers the English translation to some time between 1270 and 1290.
The manuscript is in the Bodleian Library. The French original has
since been reprinted in France, as I learn from Brunet’s Supplement
au Manuel du Libraire. Both this and its abridgment, by Geoffrey
Gaimar, are in the British Museum.

|Diffusion of French language.|

40. Two more celebrated poems are by Wace, a native of Jersey; one, a free version of the history lately published by Geoffrey of Monmouth; the other, a narrative of the Battle of Hastings and Conquest of England. Many other romances followed. Much has been disputed for some years concerning them, and the lays and fabliaux of the northern trouveurs; it is sufficient here to observe, that they afforded a copious source of amusement and interest to those who read or listened, as far as the French language was diffused; and this was far beyond the boundaries of France. Not only was it the common spoken tongue of what is called the court, or generally of the superior ranks, in England, but in Italy and in Germany, at least throughout the thirteenth century. Brunetto Latini wrote his philosophical compilation, called Le Tresor, in French, “because,” as he says, “the language was more agreeable and usual than any other.” Italian, in fact, was hardly employed in prose at that time. But for those whose education had not gone so far, the romances and tales of France began to be rendered into German, as early as the latter part of the twelfth century, as they were long afterwards into English, becoming the basis of those popular songs, which illustrate the period of the Swabian emperors, the great house of Hohenstauffen, Frederic Barbarossa, Henry VI., and Frederic II.

|German poetry of Swabian period.|

41. The poets of Germany, during this period of extraordinary fertility in versification, were not less numerous than those of France and Provence.[73] From Henry of Veldek to the last of the lyric poets, soon after the beginning of the fourteenth century, not less than two hundred are known by name. A collection made in that age by Rudiger von Manasse of Zurich contains the productions of one hundred and forty; and modern editors have much enlarged the list.[74] Henry of Veldek is placed by Eichhorn about 1170, and by Bouterwek twenty years later; so that at the utmost we cannot reckon the period of their duration more than a century and a half. But the great difference perceptible between the poetry of Henry and that of the old German songs proves him not to have been the earliest of the Swabian school: he is as polished in language and versification as any of his successors; and though a northern, he wrote in the dialect of the house of Hohenstauffen. Wolfram von Eschenbach, in the first years of the next century, is, perhaps, the most eminent name of the Minne-singers, as the lyric poets were denominated, and is also the translator of several romances. The golden age of German poetry was before the fall of the Swabian dynasty, at the death of Conrad IV., in 1254. Love, as the word denotes, was the peculiar theme of the Minne-singers; but it was chiefly from the northern or southern dialects of France, especially the latter, that they borrowed their amorous strains.[75] In the latter part of the thirteenth century, we find less of feeling and invention, but a more didactic and moral tone, sometimes veiled in Æsopic fables, sometimes openly satirical. Conrad of Wurtzburg is the chief of the latter school; but he had to lament the decline of taste and manners in his own age.

[73] Bouterwek, p. 95.

[74] Id. p. 98. This collection was published in 1758, by Bodmer.

[75] Herder, Zerstreute Blätter, vol. v. p. 206. Eichhorn, Allg.
Geschichte der Cultur. vol. i. p. 226. Heinsius, Teut, oder
Lehrbuch der Deutschen. Sprachwissenschaft, vol. iv. pp. 32-80.
Weber’s Illustrations of Northern Antiquities, 1814. This work
contains the earliest analysis, I believe, of the Nibelungen Lied.
But above all, I have been indebted to the excellent account of
German poetry by Bouterwek, in the ninth volume of his great work,
the History of Poetry and Eloquence since the thirteenth century.
In this volume the mediæval poetry of Germany occupies nearly four
hundred closely printed pages. I have since met with a pleasing
little volume, on the Lays of the Minne-singers, by Mr. Edgar
Taylor. It contains an account of the chief of those poets, with
translations, perhaps in too modern a style, though it may be true
that no other would suit our modern taste.

A species of love song, peculiar, according to Weber (p. 9), to the
Minne-singers, are called Watchmen’s Songs. These consist in a
dialogue between a lover and the sentinel who guards his mistress.
The latter is persuaded to imitate “Sir Pandarus of Troy;” and when
morning breaks, summons the lover to quit his lady; who, in her
turn, maintains that “it is the nightingale, and not the lark,”
with almost the pertinacity of Juliet.

Mr. Taylor remarks, that the German poets do not go so far in their
idolatry of the fair as the Provençals, p. 127. I do not concur
altogether in his reasons; but as the Minne-singers imitated the
Provençals, this deviation is remarkable. I should rather ascribe
it to the hyperbolical tone which the Troubadours had borrowed from
the Arabians, or to the susceptibility of their temperament.

42. No poetry, however, of the Swabian period is so national as the epic romances, which drew their subjects from the highest antiquity, if they did not even adopt the language of primæval bards, which, perhaps, though it has been surmised, is not compatible with their style. In the two most celebrated productions of this kind, the Helden Buch, or Book of Heroes, and the Nibelungen Lied, the Lay of the Nibelungen, a fabulous people, we find the recollections of an heroic age, wherein the names of Attila and Theodoric stand out as witnesses of traditional history, clouded by error and coloured by fancy. The Nibelungen Lied, in its present form, is by an uncertain author, perhaps, about the year 1200;[76] but it comes, and as far as we can judge, with little or no interpolation of circumstances, from an age anterior to Christianity, to civilisation, and to the more refined forms of chivalry. We cannot well think the stories later than the sixth or seventh centuries. The German critics admire the rude grandeur of this old epic: and its fables, marked with a character of barbarous simplicity wholly unlike that of later romance, are become, in some degree, familiar to ourselves.

[76] Weber says,--“I have no doubt whatever that the romance itself
is of very high antiquity, at least of the eleventh century,
though, certainly, the present copy has been considerably
modernised.” Illustrations of Northern Romances, p. 26. But
Bouterwek does not seem to think it of so ancient a date; and I
believe it is commonly referred to about the year 1200. Schlegel
ascribes it to Henry von Offerdingen. Heinsius, iv. 52.

It is highly probable that the “babara et antiquissima carmina,”
which, according to Eginhard, Charlemagne caused to be reduced to
writing, were no other than the legends of the Nibelungen Lied, and
similar traditions of the Gothic and Burgundian time. Weber, p. 6.
I will here mention, as I believe it is little known in England, a
curious Latin epic poem on the wars of Attila, published by Fischer
in 1780. He conceives it to be of the sixth century; but others
have referred it to the eighth. The heroes are Franks; but the
whole is fabulous, except the name of Attila and his Huns. I do not
know whether this has any connection with a French poem on Attila,
by a writer named Casola, existing in manuscript at Modena. A
translation into Italian was published by Rossi at Ferrara in 1568:
it is one of the scarcest books in the world. Weber’s
Illustrations, p. 23. Eichhorn, Allg. Gesch. ii. 178. Galvani,
Osservazioni sulla poesia de’ trovatori, p. 16.

The Nibelungen Lied seems to have been less popular in the middle
ages than other romances; evidently because it relates to a
different state of manners. Bouterwek, p. 141. Heinsius observes
that we must consider this poem as the most valuable record of
German antiquity, but that to overrate its merit, as some have been
inclined to do, can be of no advantage.

|Decline of German poetry.|

43. The loss of some accomplished princes, and of a near intercourse with the south of France and with Italy, the augmented independence of the German nobility, to be maintained by unceasing warfare, rendered their manners, from the latter part of the thirteenth century, more rude than before. They ceased to cultivate poetry, or to think it honourable in their rank. Meantime a new race of poets, chiefly burghers of towns, sprung up about the reign of Rodolph of Hapsburgh, before the lays of the Minne-singers had yet ceased to resound. These prudent, though not inspired, votaries of the muse, chose the didactic and moral style as more salutary than the love songs, and more reasonable than the romances. They became known in the fourteenth century, by the name of meister-singers, but are traced to the institutions of the twelfth century, called Singing-schools, for the promotion of popular music, the favourite recreation of Germany. What they may have done for music I am unable to say: it was in an evil hour for the art of poetry that they extended their jurisdiction over her. They regulated verse by the most pedantic and minute laws, such as a society with no idea of excellence but conformity to rule would be sure to adopt; though nobler institutions have often done the same, and the Master-burghers were but prototypes of the Italian academicians. The poetry was always moral and serious, but flat. These meister-singers are said to have originated at Mentz, from which they spread to Augsburg, Strasburg, and other cities, and in none were more renowned than Nuremberg. Charles IV., in 1378, incorporated them by the name of Meistergenoss-schaft, with armorial bearings and peculiar privileges. They became, however, more conspicuous in the sixteenth century; scarce any names of meister-singers before that age are recorded; nor does it seem that much of their earlier poetry is extant.[77]

[77] Bouterwek, ix. 271-291. Heinsius, iv. 85-98. See also the
Biographie Universelle, art. Folez; and a good article in the
Retrospective Review, vol. x. p. 113.

|Poetry of France and Spain.|

44. The French versifiers had by this time, perhaps, become less numerous, though several names in the same style of amatory song do some credit to their age. But the romances of chivalry began now to be written in prose; while a very celebrated poem, the Roman de la Rose, had introduced an unfortunate taste for allegory into verse, from which France did not extricate herself for several generations. Meanwhile, the Provençal poets, who, down to the close of the thirteenth century, had flourished in the south, and whose language many Lombards adopted, came to an end; after the reunion of the fief of Toulouse to the crown, and the possession of Provence by a northern line of princes, their ancient and renowned tongue passed for a dialect, a patois of the people. It had never been much employed in prose, save in the kingdom of Aragon, where, under the name of Valencian, it continued for two centuries to be a legitimate language, till political circumstances of the same kind reduced it, as in southern France, to a provincial dialect. The Castilian language, which, though it has been traced higher in written fragments, may be considered to have begun, in a literary sense, with the poem of the Cid, not later than the middle of the twelfth century, was employed by a few extant poets in the next two ages, and in the fourteenth was as much the established vehicle of many kinds of literature in Spain as the French was on the other side of the mountains.[78] The names of Portuguese poets not less early than any in Castile are recorded; fragments are mentioned by Bouterwek as old as the twelfth century, and there exists a collection of lyric poetry in the style of the Troubadours, which is referred to no late part of the next age.[79] Nothing has been published in the Castilian language of this amatory style older than 1400.

[78] Sanchez, Collection de poesias Castellanas anteriores al siglo
15mo. Velasquez, Historia della poesia Español; which I only know
by the German translation of Dieze, (Göttingen, 1769,) who has
added many notes. Andrès, Origine d’ogni litteratura, ii. 158.
Bouterwek’s History of Spanish and Portuguese Literature. I shall
quote the English translation of this work, which, I am sorry to
say, is sold by the booksellers at scarce a third of its original
price. It is a strange thing, that while we multiply encyclopædias
and indifferent compilations of our own, there is no demand for
translations from the most learned productions of Germany that will
indemnify a publisher.