GEORGE HOLMES HOWISON Clmntan |nss $t BRACHET'S FRENCH GRAMMAR Hontiott MACMILLAN AND CO. PUBLISHERS TO THE UNIVERSITY OF Claxenfrcm IISTORICAL GRAMMAR OF THE FRENCH TONGUE BY AUGUSTE BRACKET TRANSLATED BY G. W. KITCHIN, M.A. AT THE CLARENDON PRESS MDCCCLXIX [All rights reserved'] S crn ~ ^\au\ PREFACE. THIS Historical Grammar, which proposes to study the laws of formation of the French tongue, is not meant to swell the list of those purely grammatical works whose object is to facilitate the practical acquirement of the language. For it is no longer sufficient simply to regard the study of language as an useful preparation for the study of litera- ture. It is now seen that speech, which belongs alike to all the human race, must, like all natural phenomena, follow fixed laws, and pass in its transformations through regular stages. Linguistic studies may, therefore, be an end in themselves ; for instead of pursuing them in a spirit of idle curiosity, we may investigate the manner in which the law of change, which governs all nature, is applied to languages. It is an old saying that languages are not born but trans- muted : philology investigates the law of this transmutation, using for her instruments history and comparison. Let us id|xplain : in those sciences which are based on observation, such as chemistry or natural history, it is impossible to account for any fact unless we know what fact preceded it : if we would explain how a tree is formed, we must go back from tree to young shoot, from shoot to germ or seed ; in other words, we must make out the history of the tree by careful observation of the different conditions and forms through which it has passed. We best discern that which is VI PREFACE. by knowing that which has been ; the right way to discover the causes of any phenomenon is to look at the same time at those phenomena which have preceded it. So too for philology, which is, if we may hazard the phrase, the botany of language ; here also we may best explain words or grammatical facts by the study of their history. A single example will put this in a clearer light. It is well known that before certain feminine substances, such as messe, mere, soif,faim,peur, &c., the adjective grand keeps its masculine termination, grand 'messe, grand' mere, &c. Why so ? Grammarians, who are puzzled by nothing, tell us without hesitation that grand is here put for grande, and that the apostrophe marks the suppression of the final e. But the good sense of every scholar protests against this : after having learnt in childhood that e mute is cut off before a vowel, and never before a consonant, he is told that the e is here cut off without the slightest reason in such phrases as grand' route, &c. The real explanation is in fact a very different one. In its beginning, French grammar was simply the continuation and prolongation of Latin gram- mar ; consequently the Old French adjectives followed in all points the Latin adjective ; those adjectives which had two terminations for masculine and feminine in Latin (as bonus, bona) had two in Old French, whereas those Latin adjectives which had but one (as grandis, fortis, &c.), had only one in French. In the thirteenth century men said une grand femme, grandis femina ; une dme mortel, mortalis anima ; une coutume cruel, crudelis ; une plaine vert, viridis planities, &c. In the fourteenth century the meaning of this distinction was no longer understood ; and men, deem- ing it a mere irregularity, altered the form of the second PREFACE. Vll to that of the first class of adjectives, and wrote grande, verte, forte, &c,, after the pattern of bonne, &c. A trace of the older and more correct form survives in such expres- sions as grand" mere, grand' route, grand "/aim, grand'- garde, &c., which are the debris of the older language. In the seventeenth century, Vaugelas and the grammarians of the age, in their ignorance of the historic reason of this usage, pompously decreed that the form of these words arose from an euphonic suppression of the e mute, which must be indicated by an apostrophe. Here then is a natural explanation founded on history; and even if historical grammar had no other results be- yond that of rendering ordinary grammars more logical and simple, it would still be worth much. But instead of employing this clear and fruitful method of observation, instead of studying the past to get a better understand- ing of the present, all our grammarians, from Vaugelas to M. Girault-Duvivier, have limited themselves to the study of the language in its actual form, and have tried to explain a priori (by pure reason and logic) facts which can be explained only by the history of our language and the study of its ancient state. And accordingly, for the last three centuries, they have built up systems which were both learned and puerile, intead of limiting themselves to the simple observation of facts ; they persist in treating philology as Voltaire treated geology, when he affirmed that the shells found on mountain-tops had been dropped there by pilgrims on their return from the crusades. The severe judge- ment passed by an eminent professor at the College of France 1 on French grammarians is fully justified: 'La * M. Breal, Discours d'ouverture du cours de grammaire com- paree au College de France, 1864. vill PREFACE. grammaire traditionnelle formule ses prescriptions comme les de'crets (Tune volonte' aussi impenetrable que d^cousue ; la philologie compared fait glisser dans ces t^nebres un rayon de bon sens, et au lieu d'une docilite* machinale elle demande a releve une obeissance raisonnable.' I have illustrated by one example the position that these grammatical facts must be explained by an appeal to history, and that 'the present state of an idiom is but the natural consequence of its previous state, which alone makes it intelligible/ The same is true of words : given, for ex- ample, the word dme, we will seek for its origin. Before we come to any conclusion, let us see whether the history of the word (i. e. the study of the several forms it has suc- cessively taken) can throw any light on the problem, and shew us which path to follow. The accent on the a shews that some letter has been suppressed : in thirteenth-cen- tury texts, the word is written anme; in the eleventh century it is aneme; in the tenth anime, which leads us without a moment's hesitation to anima. Thus is history the guiding- line of philology, and there is not a single broken link in the long chain which connects the French with the Latin language. When we first look at it, the distance between dme and anima, between the French of Voltaire and the peasant Latin, seems long enough; and yet it has needed only a series of infinitely small changes spread over a very long period to connect them with one another. Nature, wasteful of time, is sparing of effort ; with slow and almost imper- ceptible modifications she arrives at results far away in appearance from her starting-point 1 . 1 M. G. Paris. PREFA CE. IX To history, regarded as an instrument of philology, com- parison must be added as a precious ally. By comparison theories are proved, hypotheses verified. Thus, in the example we have already cited, the comparison of the Italian and Spanish alma with the French dme gives to the hypo- thesis we have started an invincible certainty. Armed with this double method, the historical and the comparative, an illustrious German, Frederick Diez, wrote (A.D. 1836 to 1842) a comparative grammar of the five languages which spring from Latin 1 : he shewed according to what laws they were formed from the Latin. Starting from the philological principles laid down by him, Bartsch and Matzner in Germany, and in France Littre', Guessard, P. Meyer, and G. Paris, have applied his principles to the French language in particular, and by means of many detailed investigations have thrown fresh light upon its origin 2 . 1 The Germans call these five (Italian, Spanish, French, Portuguese, Wallachian) the Romance languages; the name is clear and convenient, has been fully accepted in scientific lan- guage, and will be employed throughout this book. 2 The work of these French philologers is far from being equally good: to say nothing of the very unequal compilation published by M. Ampere, or of M. Ghevallet's bock, an ad- mirable work in its day, but now out of date, we must regard with real sorrow the success which welcomed twenty years ago M. Genin's work (Variations de la langue fran$aise), a collection of paradoxes and startling effects, performed by a juggler with words, whose business it is to astonish a dazzled audience. M. Genin was clever enough to know that his French readers would always prefer a well-turned epigram to a dry truth, and though he had never in his life read a single line of German, he was ever ready with a pleasantry rather stale perhaps, but still always applauded in France on ' the nebulous lucubrations of German brains.' He forgot that a bon mot does not do for an argument, and that in scientific matters it is no question of French or German ideas, but of right and wrong ones. X PREFACE. In. spite of these incessant efforts, the principles of French philology, scarcely recognised even by the learned, are still utterly unknown to the great majority of the literary public. My aim in this little book is to spread the knowledge of these results by freeing them from their scientific dress, and by making them accessible to a wider circle of readers. I have accordingly endeavoured to gather into a small volume the chief laws which have guided the formation of the French tongue. This is the only novelty I have to offer : for such works are not uncommon, at any rate out- side of France. In Germany and England the study of the mother-tongue has won its citizenship in colleges and schools, where it has its undisputed seat by the side of Greek and Latin 1 ; it has not as yet penetrated into French colleges, even as a branch of higher education. M. Fourtoul, who, among a number of mistakes, hit on several happy discoveries, ordered in 1853 that comparative grammar should be taught in the upper classes of the Lyceum a step towards the study of the French language which was reversed by his successor. This is much to be regretted, especially since the present ministry, which has ceased to insist on the study of Greek and Latin, and has established industrial or technical education side by side with literary training, ought all the more to have strengthened the latter by introducing the study of the three languages, 1 It will be enough to cite two elementary works, whose numerous editions prove their success: in England, Gleig's History of the English Language, in his School Series ; in Germany, Vilmar's German Historical Grammar, intended for the higher forms in the Gymnasia {Anfangsgrunde der dcutschen Grammatik, xun'dchst fur die obersten Klassen der Gymnasien. v. Dr. Vilmar. 6te Auflage, 1864). PR EFA CE. Xi Greek, Latin, and French, together with that of the three national literatures. One Frenchman, M. Monjean, Director of the Chaptal College, has ventured to introduce a course of lectures on the history of the French language in his rhetoric class, with the very best results. May his example embolden the Uni- versity of Paris to spread among the higher classes of our schools the results which have been indisputably obtained by science ! My object will have been gained if my modest manual of philology can in any way hasten this result. I cannot hope to set forth a complete historical gram- mar in two hundred pages, when three volumes would scarcely suffice. I have therefore laid aside all secondary matters and points of detail, and have thought it enough to set forth essential laws and fundamental principles, so as not to overstep the limits of space which I had im- posed on myself. Again, the subject of this book is not the grammar of Old French. The French language in its mediaeval state finds a place in it only so far as it illustrates Modern French (if I may apply to my little book what M. Littre said of his Historical Dictionary). Present usage depends on ancient usage, and can only be explained by it. Modern French without Old French is a tree without roots; Old French by itself is a tree without branches or leaves : the separation of the two is an injustice to both an injustice constantly done to them up to the present time ; and their proper combination is the only originality claimed for this book, and gives it a right to be called a Historical Grammar, The book is in three distinct parts : first, the Introduction, which sketches the history of the French language, of its Xll PREFACE. formation, and of its elements ; secondly, the Historical Grammar, which deals with the Letters (Book I), Inflexion (Book II), and the Formation of Words (Book III) ; and lastly, an Appendix containing the rules to be followed in the discovery of etymologies. Finally, I must express my gratitude to MM. Egger, Littre, and Ernest Renan, Members of the Institute, who have kindly given me the advantage of their advice and encou- ragement; to M. Emile Lemoine, formerly pupil of the Ecole Polytechnique ; last of all and most of all, to MM. Paul Meyer and G. Paris, whose friendship has strengthened me for my task. If this book has any value, it is to them that it is due. AUGUSTE BRAGHET. May 6, 1867. [The present English edition has had throughout the great benefit of the counsel and oversight of Professor Max Miiller, to whom hearty thanks are due for the interest he has taken in its welfare. There are a few Latin words in the work marked with an asterisk, as testonem*; these are late and unclassical.] CONTENTS. PREFACE INTRODUCTION. I. HISTORY OF THE FRENCH LANGUAGE I II. FORMATION OF THE FRENCH LANGUAGE . . . . 3 2 CHAPTER I. The Continuance of the Latin Accent . . 33 CHAPTER II. Suppression of the Short Vowel ... 35 CHAPTER III. Loss of the Medial Consonant . . . .37 CHAPTER IV. Conclusion 3 8 BOOK I. Phonetics, or the study of the Letters. PART I. Permutation of Letters. CHAPTER I. History of the French Alphabet .... 46 SECT. I. Origin of the French Vowels .... 46 I. Simple Vowels ...... 46 2. Compound Vowels . . . . 5 1 SECT. II. Origin of the French Consonants . . 55 i. Liquids ....... 55 2. Labials 5 8 3. Dentals 60 4. Gutturals 63 CHAPTER II. History of the Latin Alphabet . . .66 SECT. I. History of the Latin Vowels .... 67 I. Tonic Vowels 67 2. Atonic Vowels ...... 68 (1) Preceding the Tonic Syllable . . 68 (2) Following the Tonic Syllable . . 69 SECT. II. History of the Latin Consonants ... 70 i. Liquids . . . . . . 7 1 2. Dentals 73 3. Gutturals 74 4. Labials 75 xiv CONTENTS. PART II. Transposition, Addition, and Subtraction of Letters. PAGE CHAPTER I. Transposition ' 77 SECT. I. Of Consonants . . . . . 77 SECT. II. Of Vowels 77 CHAPTER II. Addition 78 SECT. I. Prosthesis ....... 78 I. Of Vowels 78 2. Of Consonants ...... 79 SECT. II. Epenthesis . 79 SECT. III. Epithesis 80 CHAPTER III. Subtraction . 80 SECT. 1. Aphaeresis 80 i. Of Vowels 80 2. Of Consonants 80 SECT. II. Syncope 80 i. Of Vowels 80 2. Of Consonants . . . . . .81 SECT. III. Apocope .82 I. Of Vowels 82 2 Of Consonants 82 PART III. Prosody. I. Tonic Accent 83 II. Grammatical Accent ....... 84 III. Oratorical Accent 85 IV. Provincial Accent ........ 85 BOOK II. Inflexion, or the study of Grammatical Forms. PART I. Declension. CHAPTER I. The Substantive 88 SECT. I. Case 88 SECT. II. Genders 96 SECT. III. Numbers 98 CHAPTER II. The Article . . 100 CONTENTS. XV PAGE CHAPTER III. The Adjective 102 SECT. I. Qualifying Adjectives ..... 102 i. Cases and Numbers 102 2. Genders . . . . . . 102 3. Adjectives used as Substantives . . -103 4. Degrees of Comparison .... 104 SECT. II. Nouns of Number 105 i. Cardinals . 105 2 Ordinals ....... 107 CHAPTER IV. Pronouns 109 SECT. I. Personal .......... 109 SECT. II. Possessive in SECT. III. Demonstrative. . . . . . . 113 SECT. IV. Relative , 114 SECT. V. Indefinite 115 PART II. Conjugation. Preliminary remarks . . . . . . . . ..118 CHAPTER I. Auxiliary Verbs 123 SECT. I. Eire . .124 SECT. II. Avoir . .127 CHAPTER II. Classification of Verbs : Conjugations . . .129 CHAPTER III. Formation of Tenses 136 CHAPTER IV. Irregular Verbs 142 CHAPTER V. Defective and Anomalous Verbs . . . .143 SECT. I. Defective 143 SECT. II. Anomalous 148 PART III. Particles. CHAPTER I. Adverbs 153 SECT. I. Of Place 154 SECT. II. Of Time 155 SECT. III. Of Manner 158 SECT. IV. Of Intensity 158 SECT. V. Of Affirmation and Negation .... 160 CHAPTER II. Prepositions 163 SECT. I. Formed from the Latin 163 SECT. II. Formed from more than one Latin Preposition 163 SECT. III. Formed from Prepositions with Adverbs, &c. . 164 XVI CONTENTS. PAGE SECT. IV. Such as are really Participles . . . .165 SECT. V. Formed from Substantives . . . .165 SECT. VI. Formed from Adjectives and Adverbs . .166 SECT. VII. Formed from an Article and a Preposition . 167 SECT. VIII. Formed from an Article and a Substantive . 167 CHAPTER III. Conjunctions 167 SECT. I. Simple 167 SECT. II. Compound 1 68 SECT. III. Conjunctival phrases 169 CHAPTER IV. Interjections 169 BOOK III. Formation of "Words. CHAPTER I. Compound Words 172 SECT. I. Of the Accent on Compound Words . . .172 SECT. II. Words compounded of Nouns . . . 1 74 SECT. III. Adjectives . . . 175 SECT. IV. Verbs . . . .176 SECT. V. Phrases . . .176 SECT. VI. Particles . . .176 I. Prepositional . . . . . .177 2. Qualitative 1 80 3- Quantitative . . . . . .181 4. Negative 181 CHAPTER II. On Suffixes or Terminations . . . .181 SECT. I. Accent of Derived Words . . . .182 SECT. II. Nominal Suffixes 183 i. Accented in Latin . . . . .183 2. Atonic in Latin ...... 187 SECT. III. Verbal Suffixes 191 i. Accented ....... 191 2. Atonic 192 SECT. IV. Diminutives 192 APPENDIX. Rules which must be followed in detecting Derivations . 195 INDEX 205 CORRIGENDA. Page ii, line 2, for the Fat, read the Simple. io, 21, omt'f n. " 5I , 9, transpose the paragraph From a primitive n . . . stare, to line 21. J 5> a / /6?r * add n ' 10^ 16, /or course, read cursus. " 5 * ;, 25 , /or dypeus buccularius, rwrf scutum bucc larium. 106, 20, for triginti, read triginta. 154, 26, for aliorsam, read aliorsum. INTRODUCTION. I. HISTORY OF THE FRENCH LANGUAGE. CAESAR tells us that he found in Gaul three races, differ- ing in speech, manners, and laws : the Belgae in the north, the Aquitani between the Garonne and the Pyrenees, and in the centre the Gallic or Celtic race. But the Belgae and the Celts really belonged to the same race, while the Aquitani were partly Iberian, and their language has per- haps survived in the Basque or Euskarian tongue. Thus, then, almost all the soil of France was occupied by the Celtic race; they were men tall and fair, eager for excitement and noise, whose ambition was to fight well and to speak well. Some six hundred years before the Christian era Mar- seilles (Massilia) was founded near the mouths of the Rhone by Phocaean refugees. This city, thanks to her relations with Rome, was destined to be the beginning of woes to the people of Gaul. She called in the Romans to defend her against the Ligurians in B.C. 153. The Romans seized the Rhone valley ; and thence, in Caesar's time, passed on to conquer the rest of the land. The Celts resisted bravely: Caesar broke their spirit only by the most cruel measures ; he massacred ten thousand women and children at Bourges ; 2 INTRODUCTION. slew the; -heads of 1 a tribe at^Vannes, and sold the rest by auction; cut off ^ his prisoners' hands at Uxellodunum. After e^gbt year^ pf'|riis ( wcrk Ga~ul was subdued, and Rome began to administer her conquest. The chief secret of Roman foreign politics lay in the perfection of her iron system of colonisation. She had two engines by which to hold down a conquered province, her military colonies set all round the frontier, so as to isolate the conquest from all external influences ; and then, secondly, an energetic ' administration ' within that circle which soon broke up all local resistance. The language and religion of the conqueror was forced on the subject: all resistance was crushed by extermination or deportation ; -the vacuum rilled up with colonists and freedmen from Rome. By this method conquerors and conquered were in a few years completely welded into one mass. Less than a cen- tury after the conquest, Latin was spoken in many parts of Gaul. But this Latin, brought in by colonists and soldiers, was very unlike the Latin of Virgil : it was distinguished from the classical or written Latin by peculiarities of vo- cabulary and of inflexion which demand our attention. It is a first law of history that all languages (just like the nations that use them), are one at first, but presently split into two parts the speech of the noble and that of the people. Every language has its epoch of division : it comes when the nation opens its eyes to arts and poetry, in a word, to culture and literature. From that time the nation may be divided into two great classes, the lettered and the unlettered. The Latin language underwent this same division at the time of the second Punic war. The separation increased as time went on. Greek art and Greek manners introduced into the literary language of Rome a crowd of purely Greek HISTORY OF THE FRENCH LANGUAGE. 3 words utterly unknown to the popular idiom 1 . These words, marks of breeding, but servile copies of the Greek, remained as strange to the common people, as the aristo- cratic French-English terms, ' turf/ ' sport/ ' steeple-chase/ &c., or the technical terms of science, ' diluvium/ ' stratifi- cation/ ' ornithology/ &c., are to the French peasantry at the present day. These borrowed words widened the breach between the literary and the popular Latin, a difference which ever increased, until the ' sermo nobilis/ the literary, aristocratic, 'classical' Latin, became in Caesar's day entirely distinct from the ' sermo plebeius, rusticus/ the ' castrense verbum/ as authors disdainfully styled it, the Latin of the people and the camp. Each had its own grammatical forms and vocabulary. For example, ' to strike ' is verberare in literary Latin ; the popular Latin said batuere : the French words, cheval, semaine, aider, doubler, ba faille, &c., were, in the classical Latin, equus, hebdomas, juvare, duplicare, pugna; in the popular, caballus, septimana, adjutare, duplare, batualia. The popular Latin was unwritten, and we might have remained ignorant of its existence had not the Roman grammarians revealed it to us by exhorting their students to avoid as low and trivial certain expressions which, they tell us, were in common vulgar use. Cassiodorus tells us that the feigned combats of gladiators and exercise-drill of the army were called batalia, ' Quae vulgo batalia dicuntur, exercitationes gladiatorum vel militum significant/ Pugna was the literary term, batalia the popular ; pugna has dis- appeared, batalia has survived in bataille. The pedants of that day could not foresee that the literary idiom, which they admired so much, would one day disappear; and that 1 As d/z(t$e'arpoi>, tTTTrdSpo/iOff, e< &C. B 2 4 INTRODUCTION. the popular Latin would reign in its room, parent of Italian, French, and Spanish, and strong enough to bear the weight of the literatures of three powerful nations. Imported into Gaul by soldiers and colonists, the popular idiom soon made itself at home, and, even in the first cen- tury of the Christian era, had supplanted the Celtic speech, except in Armorica 'and a few isolated spots *. A hundred years after the conquest, women and children used to sing Latin songs; and so universal became the use of the language, that in Strabo's time the Celt was no longer regarded as a Barbarian 2 . The lengthy sojourn of the Legions, the incessant influx of colonists, the necessity of pleading in Latin before the Roman tribunals, the con- version of the people to Christianity, and lastly, the natural vivacity and love of change 3 which distinguishes the Celt, were further causes which contributed to the adoption by the Gallic people of the language of their conquerors. But, at the same time that the people thus accepted the common Latin, the upper classes in Gaul were ambitious to adopt the literary dialect, practised rhetoric, and hoped to rise to political distinctions. From the days of Augustus, Gaul became a nursery for rhetoricians and grammarians; the schools of Autun, Bordeaux, and Lyons were renowned throughout the Empire. Pliny boasts that his works were known throughout Gaul 4 . Caesar admitted Celts to the Senate ; Claudius enabled them to undertake all public offices, on the sole condition that they knew Latin. It is easy to understand why the Celtic noble forgot his mother- tongue. 1 The Celtic lingered long after this date in Auvergne. 2 That is, the test of language (implied in the word Barbarian) placed the Gaul on the same footing with the average Roman colonist. a See Caes. B. G. 4 . 5. 4 Pliny, Ep. 9. 2. HISTORY OF THE FRENCH LANGUAGE. 5 That tongue disappeared, leaving a few faint traces as evidences that it had existed. Thus the Romans remarked that the bird they called galerita was called alauda in Gaul ; that ' beer/ in Graeco-Latin zythum, was cervisia in Gallic : they introduced the words into their own tongue, and these new Latin words, passing six centuries later into French, produced the words alouette * and cervoise. These and a few other isolated words, together with certain names of places, are all that the French language owes to the Gallic ; and indeed, if we speak more exactly, the French has borrowed nothing from it, since these words have passed through an intermediate Latin stage, and are not directly introduced into French from Gallic. But these cases are so very rare, that it may almost be affirmed that the influence of Celtic upon French has been inappreciable. Thus, whtte the French nation is really Celtic in race, its language is not so : a very remarkable fact, which shews, better than any history could do, what a strong absorbent was the Roman power. The Celtic language had scarcely accepted its defeat 2 , when the Latin, from this time forth the true mistress of 1 Alauda did not pass directly into alouette^ but into the O.Fr. aloue, of which alouette is the diminutive. 2 The Celtic language, thrust by the Romans back into Armorica, survived there for centuries, and was revived by an immigration of Kymri from Wales in the seventh century. The Bretons resisted the Frank as successfully as they had withstood the Roman ; and what is now called the Low Breton patois is the direct descendent of the Celtic language. It has a con- siderable literature of tales, songs, and plays, which, however, only date back as far as the fourteenth century. But the language, living thus for a thousand years 'in extremis,' naturally has deviated far from the primitive Celtic tongue : for beside the natural corruption and degradation of eighteen centuries, it has been forced to admit into its ranks a crowd of foreign, that is, of French, terms ; and consequently many Breton words pre- sent the singular spectacle of having two distinct forms, the one 6 INTRODUCTION. Gaul, had to enter on a fresh struggle, and to repel a new assailant. The invasion of the German tribes set in. As far back as the second century after Christ the barbarians began slowly to filter through into the Gaelic soil: they silently undermined the dykes of the Roman Empire, and prepared for the bursting of the barriers, and the terrible inundations of the fifth century. To protect northern Gaul against these German inva- sions, the Romans garrisoned their frontiers with a chain of legions or military colonies ; and when these veterans were no longer able to defend the sanctity of the Roman territory, the Romans employed an expedient which kept the great invasion at bay for a whole century, and for a few years at least gave peace to the Empire. They determined to let the barbarians settle in Northern Gaul, to attach them to the Empire, and to use them as a new and durable barrier against all further invasions. These were the Leti 1 , colonies of barbarians who recognised the nominal so- vereignty of the Emperors, and enjoyed lands granted them ancient and of Celtic origin, the other more modern, borrowed from the French, but modified by a Celtic termination. Thus in Breton we have for just egwirion or just, secretly ekuz or secretament, troubled enkrezet or troublet, anger buanegt* or coler, and so on. Here the middle column is composed of old Celtic words ; the third of corrupted French words. It would not have been necessary to insist on so elementary a truth, had not a theory been started in the eighteenth century that these Celtic impor- tations were really the origin of the French language. Le Bri- gant and the well-known La Tour d'Auvergne supported this opinion. Voltaire called this etymological folly Celto-mania : its believers amused the world by extravagant assertions that Celtic was the original speech of Paradise ; that Adam, Eve, the old serpent, all spoke Low Breton. These errors have had a still worse result ; for they have cast unmerited discredit on all Celtic studies. 1 Probably a form of the German word leute. See Du Cange. HISTORY OF THE FRENCH LANGUAGE. J under a kind of military tenure. At the same time the Emperors hired Franks, Burgundians, Alans, to fill up the blanks in their legions. The consequence was an ever-increasing introduction of German words into the common Latin ; these terms, as was natural, being chiefly connected with warfare. Vegetius, in his ' De re militari/ tells us that the Roman soldiers gave the name of burgus to a fortified work 1 . This is the German Burg. Thus, nearly a century before Clovis, Ger- man terms had got into the Latin language : it is clear that after the German invasion this influence will greatly increase in strength. But we must first note down the chief features of the Latin of the last ages of the Empire. A century after the Roman conquest Gaul was flourishing and prosperous. The Latin language in its two forms pursued a tranquil course the common dialect in cities and in the fields, the literary dialect among the aristocracy and middle classes. In the second century after Christ, the time of the highest splendour of Roman Gaul, the popular dialect was in the shade, while literary Latin shone with great brilliancy ; the Gallic schools produced lawyers and rhetoricians : and Juvenal calls Gaul ' nutricula causidicorum.' But in the fifth century, just before the German invasion, the scene is very different : the two dialects have changed places; literary Latin is dying; the popular dialect spreads widely, and this even before the invasion of A.D. 407. The institution of the l Curials' in the cities, and the extinction of the older municipal bodies, gave its deathblow to literature and the literary dialect. The better classes perished, schools were everywhere shut up, literary culture came to an imme- diate stop, and ignorance speedily recovered all the ground 1 ' Castellum parvum, quod burgum vocant.' 8 INTRODUCTION. she had lost. From this time the use of the written Latin, a dialect which only lived in books and by tradition, was confined to the Gallo-Roman nobles, a handful of men who transmitted to their children a petrified idiom, which had no life, and was destined to perish with them, when their time came. And here again popular Latin won by the losses of the literary dialect. At last the Roman Empire fell under the attacks of the barbarians. In the whirlwind, administration, justice, aris- tocracy, literature, all perished, and with them the language 1 1 M. Meyer says well that 'the invasion of the barbarians irrevocably fixed the gulf between these two idioms, between the common Latin, the mistress of Gaul, ready to be the mother of the French language, and the literary dialect, a dead language, used only by the learned, and destined to have no influence in the formation of modern languages. This dialect was kept up by Gregory of Tours, Fredegarius, the literary renaissance under Charlemagne, and by scholasticism; it was perpetuated in learned use, and in the sixteenth century experienced, after the great renaissance, a kind of artificial resurrection. Even in our own day it is the official language of the Roman Catholic Church, and, until quite lately, was the language of the learned, especially in Germany.' After the invasion under the Merovingian kings, the public personages, notaries or clergy, too ignorant to write literary Latin correctly, too proud to use the common Latin in their documents, and eager to imitate the fine style of Roman officials, wrote 'a sort of jargon, which is neither literary Latin nor popular Latin, but a strange mixture of both, with the common dialect more or less preponderant, according to the ignorance of the writer.' This jargon is what is called Low Latin. It continued to be the language of the French administration up to 1539, when Francis I ordered French to be used in all public acts. This distinction between Low Latin, a gross and barren imitation of the Roman literary dialect, and Popular Latin, the living language of the people, and parent of the French tongue, must not be forgotten. It should be added that there is, besides, a second kind of Low Latin, that of the middle ages, which reproduced French words in a servile way: as for example, missaticum produced the French message', and again message was retranslated into messagium. HISTORY OF THE FRENCH LANGUAGE. 9 which they had employed. Then the common dialect entirely supplanted the other. If proof of this were needed, we should find it in the fact that wherever the literary and the common dialect used two different words for the same thing, the French language has invariably taken the latter, and thrown aside the former: an absolute proof that the literary dialect was confined to the upper classes, and flourished and perished with them. Illustrations are innu- merable : thus LITERARY LATIN. Hebdomas Equus Verberare Pugna Oscular! Iter Verti Urbs Os Felis Duplicare Siiiere Tentamen Gulosus Jus Minae Edere Ignis Iiudus Aula POPULAR LATIN. FRENCH. septimana semaine (O.Fr. sep??iame) caballus cheval batuere battre battalia bataille basiare baiser viaticum voyage tornare tourner villa mile bucca bouche catus chat duplare doubler laxare laisser exagium essai glutonem glouton directus (drictus) droit minaciae menace manducare manger focus feu jocus jeu curtem cour, &c. These examples shew how incorrect it is to say that 10 INTR OD UCTION. French is classical Latin corrupted by an intermixture of popular forms; it is, on the contrary, the popular Latin alone. The same is true wherever the invasion of the barbarians also destroyed the literary dialect. Italian, Spanish, Portuguese, are the products of the slow de- velopment of the common Roman speech. Hence the striking family likeness often noticed between these sister- tongues ' Facies non omnibus una, Nee diversa tamen, qualem N decet esse sororum.' The German destroyed the literary dialect; but the common Latin was the gainer : eventually it succeeded in absorbing even its conquerors ; it compelled them to forget their own language, and to adopt that of their subjects. There are many causes which led to this result: first, the numerical paucity of the Franks, a few bands of men, scarcely more than twelve thousand in all, in the midst of six millions of Gallo-Romans ; next, if the Franks had not accepted the Latin, what would they have taken for their common tongue ? Each German tribe had its own dialect, Frankish, Burgundian, Gothic, &c. But, lastly, the conversion of the Franks to Christianity, which, as it were, bound them over to learn Latin, may be reckoned as the special cause which made the adoption of the Latin a necessity. So they all tried to learn Latin ; and, less than a century after the invasion, Fortunatus, Bishop of t Poitiers, con- gratulated Haribert on the great success of his efforts : ' Qualis es in propria docto sermone loquela Qui nos Romano vincis in eloquio ? ' At Strasburg in A.D. 842, Ludwig the German takes an oath in French in the presence of the army of Karl the Bald; a clear proof that the Karling soldiers no longer HISTORY OF THE FRENCH LANGUAGE. II understood German. In the next century, when Hrolf swore fealty to Karl the Fat (A.D. 911), he had scarcely begun the formula with 'Bi Got' (In the name of God) when all the company of lords burst out laughing ; so utterly was German forgotten, that it actually sounded ridiculous in their ears. Thus the Latin supplanted the German : yet a great number of German words were retained to designate those new institutions which the Franks brought in with them, such as vassal, alien, fief, &c. All terms relating to political or judicial functions, all titles in the feudal hierarchy, are of German origin. The German words mahal, bann, alod, skepeno, marahscalh, siniscalh, &c., are formed by common Latin into mallum, bannum, alodium, scabinus, mari- scallus, siniscallus, &c., whence, several centuries later, they passed into the French mall, ban, alleu, echevin, mare'chal, se'ne'chal, &c. Still more is this the case with war terms. The Franks long kept to themselves, as a privileged class, the warlike profession; and the Gallo-Romans accepted the terms which their masters employed : as halsberc, hau- bert ; helm, heaume ; heriberg, auberge ; werra, guerre, &c. There are upwards of nine hundred such words which passed from the German into Latin, and thence into French. This invasion touched the vocabulary only : there are no traces of German influence on French syntax. Common Latin was greatly affected by this sudden inroad of barbarous words : its vocabulary became less and less like that of the literary dialect; its syntax still further widened the breach. Those analytical tendencies which appear in all modern languages, and which cause the use of prepositions instead of inflected cases to mark possession and aim, soon shewed themselves in popular Latin. The literary dialect said, ' Do panem Petro^ or ' equus Petri;' but the popular Latin said, ' Do panem ad Petrumj ' caballus de Petro :' and 12 INTR OD UC TION. similarly auxiliaries were introduced in the conjugation of verbs 1 . Thus modified in its syntax, and augmented in its vocabulary, popular Latin became a really distinct language ; and the men of culture in Merovingian times called it, slightingly, ' lingua romana rustica/ Peasant- Latin. Its position as an independent language is attested early. Church writers give us the earliest proofs of it, as we should expect; for the Church, through her missionaries and her priests, first addressed the people, and in order to be un- derstood, she must use their language. Thus, as early as A.D. 660, St. Mummolinus is elected Bishop of Noyon, be- cause he can speak both German and Romance 2 . We read in the life of St. Adalhard, Abbot of Corby in A.D. 750, that he preached in the popular tongue ' with a sweet fluency ; ' and his biographer gives us clearly the distinction between the two dialects when he says, ' When St. Adalhard spoke the common, that is, the Roman tongue, you would have thought he knew no other ; if he spoke German, he was still more brilliant; but if he used the Latin, he spoke even better than in either of the others 3 / Thus in the lifetime of Karl the Great (as we see from this passage), the people understood no Latin, and the Church had taken to preaching and teaching in French. There has come to light by a fortunate chance a fragment of a glossary, called the ' Glosses of Reichenau 4 ,' and written 1 See below, p. 123. 2 * Quia praevalebat non tantum in Teutonica, sed etiam in Romana lingua.' 3 * Qui si vulgari, id est, Romana lingua, loqueretur, omnium aliarum putaretur inscius ; si vero Teutonica, enitebat perfectius ; si Latina, in nulla omnino absolutius.' Act a Sanctorum, i. 416. 4 Discovered in 1863 by M. Holtzmann in a MS. in the Library at Reichenau. HISTORY OF THE FRENCH LANGUAGE. 13 about A.D. 768, which explains many of the difficult words of the Vulgate in the French of the period. The words are written in two columns; on the left the Latin (Vulgate), on the right the French : thus TATTTV: FRENCH (of the 8th cent.) MODERN FRENCH. Minas Manatees Menaces Galea Helmo Heaume Tugurium Cabanna Cabane Singulariter Solamente Seulement Caementarii Macioni Masons Sindones Linciolo Linceul Sagma Soma Somme &c. &c. This most interesting fragment is the first written monu- ment of the French language, eleven hundred years old. The translation into modern French, in the right hand column, shews at a glance the distance between this still unformed dialect and the French of the present day. These Glosses also shew that the inhabitants of France spoke French in Karl's days; in fact, Karl himself found it necessary to learn the language of his subjects. And while Eginhard, Alcuin, Angilbert, and all the culti- vated class of that day affected to despise this half-formed patois, the Church, which had never been afraid of using this vulgar speech, quickly took in its whole importance, and instead of resisting it, and clinging to literary Latin, set herself to make a skilful use of the new movement. Hitherto she had but tolerated, or perhaps patronised, the study of this vulgar tongue by priests and missionaries ; but tow r ards the end of Karl's reign, she did more : she ordered the clergy to study it, seeing that the people no longer un- derstood Latin. In A.D. 813 the Council of Tours bid all 14 INTRODUC TION. priests expound the Holy Scriptures in the ' Romance,' and the preachers to use the same in their pulpits. Thus the Church recognised the existence of this new language, and confessed that Latin was dead and gone from among the people ; and, when once she had settled this point, she carried it out to its natural results with her habitual perseverance. After the Council of Tours, those of Rheims in A.D. 813, of Strasburg in A.D. 842, and of Aries in A.D. 851, renewed the order to preachers, until in fact the vulgar tongue was everywhere substituted for the Latin. Thus it gained ground rapidly ; so much so that five-and- twenty years after Karl's death, it was used as the language of political negociation in the famous Oaths of Strasburg which Ludwig the German took to his brother, Karl the Bald, and Karl's army took to Ludwig the German, in March, A.D. 842. Nithard, the nephew of Karl the Great, has preserved them in his l History of the Franks,' written about A.D. 843, at the command of Karl the Bald, whose intimate friendhe was. I. OATH TAKEN BY LUDWIG THE GEKMAN. Old French. Pro Deo amur, et pro Christian poblo et nostro commun salvament, d'ist di en avant, in quant Deus savir et podir me dunat, si salvarai eo cist meon fradre Karlo et in adjudha et in cadhuna cosa, si cum om per dreit son fradra salvar dist, in o quid il mi altresi fazet; et ab Ludher nul plaid numquam Modern French. Pour 1'amour de Dieu et pour le salut du peuple chre'tien et notre commun salut, de ce jour en avant, autant que Dieu me donne savoir et pouvoirje sauverai mon frere Charles et en aide et en chaque chose (ainsi qu'on doit, selon la justice, sauver son frere), a condition qu'il en fasse autant pour moi, et je HISTORy OF THE FRENCH LANGUAGE. prindrai, qui meon vol cist meon fradre Karle in damno sit. ne ferai avec Lothaire aucun accord qui, par ma volonte, porte prejudice a mon frere Charles ici present. II. OATH OF THE SOLDIEKS OF KARL THE BALD. Modern French. Si Louis garde le serment qu'il a jure a son frere Charles, et que Charles mon maitre, de son cote, ne le tienne pas, si je ne Ten puis de'tourner, ni moi, ni nul que j'en puis detourner, ne lui serai en aide contre Louis. Old French. Si Lodhuwigs sagrament, que son fradre Karlo jurat, conservat, et Karlus meos sendra de sua part non los tanit, si io returnar non Tint pois, ne io, ne neuls cui eo returnar int pois, in nulla ad- judha contra Lodhuwig nun li iv er. Next after the Reichenau Glosses, these oaths are the oldest monuments of the French language : and their value is incalculable for students of the linguistic origin of the Romance tongues ; for we here catch, as it were, the Latin language in the act of transformation into French. The importance of this will appear in the course of this book : it is sufficient to remark here that the Frankish army clearly had lost all knowledge of Latin or German ; otherwise the German Emperor, Ludwig, would never have taken oath to them in French. From this time the vulgar tongue took, once and for all, the place of the Latin which the people no longer under- stood. In common use during the last two centuries, officially acknowledged by the Church in A.D. 813, and bv the State in A.D. 842, it increased in importance, and soon broke out in poetry. In the ninth century there appears a poem in French verse, on the martyrdom of St. Eulalia ; 1 6 INTR OD UCTION. in the tenth century we find two short poems, the one on the Passion, the other on the life of St. Leger of Autun. These are the first poetic attempts of the language. These two centuries, the ninth and tenth, in which the later Karlings came to a wretched end, seem at first sight barren and desolate; but they are in reality fertile in the beginnings of French national life: with that life comes a national language, poetry, and art. All these things sprang into being from the people, not from the kings. The pre- tentious chroniclers of the time describe the last moments of the decrepit Karling dynasty ; they pass over and have not noticed how fresh a life, and what creative energy was beginning to reanimate what seemed to be the worn-out powers of society 1 . From the tenth century the French nation begins its real life : the invasions of the barbarians are over 2 . On the ruins of the Karling empire feudalism, a new form of social life half-way between ancient slavery and modern freedom, will flourish for six centuries. As the use of the French speech increased, the knowledge of Latin diminished. Hugh Capet knew no language but French : when he had an interview with Otto II, the Emperor of Germany, who spoke Latin to him, he was 1 This birth of the French language in a historical age well- known to us is of the highest importance : we learn from it how such languages as Latin and Greek (which we know only in their full age) carne first into being. And when our histories relate in full the obscure quarrels and struggles of obscure princes, and give us no details respecting this great event, we see clearly that true history has not yet found its way into the school-room. See M. Littre, Histoire de la Langue Fran$aise, i. 260, and the Revue des Deux Mondes, Feb. 15, 1867. 2 The last invasion ended with the establishment of the Northmen in north-western France. Their numbers were small : they forgot their own tongue, and adopted that of their subjects. A century after Hrolf's death Normandy was cele- brated for the excellence of her French. HISTORY OF THE FRENCH LANGUAGE. 17 obliged to get one of the bishops to act as interpreter. Even in the monasteries Latin ceased to be used after the eleventh century; there were even numbers of priests who knew nothing but French. Thus at last Latin was abandoned even by the upper classes : they had clung to it three centuries after it had died out of common use. Forthwith there sprang up, between the eleventh and the thirteenth centuries, a thoroughly original poetical literature ; graceful or brilliant lyrics, and high epics, like the l Chanson de Roland/ were written, and became exceedingly popular in other countries as well as at home. It is also worthy of notice that the French language in the thirteenth century and onwards, was well known and accepted by neighbouring nations. The Norman Conquest imposed French on England: in Germany Frederick II and his court were familiar with French poetry ; in Italy French was generally known and used ; Marco-Polo wrote his travels in it ; Brunetto Latini, Dante's master, composed his ' Tre'sor de Sapience' in it, 'because the French is the most delectable and most common tongue.' From every quarter students flocked to the University of Paris, and mediaeval Latin lines testify to the fact. ' Filii nobilium, dum sunt juniores, Mittuntur in Franciam fieri doctores.' It is time that we asked, What is this French language which Europe valued so highly in the thirteenth century ? It is a well-known fact that the first cause of phonetic changes and transformations of language lies in the structure of the vocal organs ; or, in other words, in dif- ference of pronunciation; and this again results from dif- ference of race. Thus Latin, introduced into Italy, Gaul, and Spain, and spoken by three different races, each in its own way, gradually was decomposed, as we have seen, into c 1 8 INTRODUCTION. three corresponding languages. In Gaul, popular Latin fell into the hands of two rival races, North and South, and produced two distinct idioms, that of the South, or the ' Langue d'Oc,' and that of the North, called the ' Langue d'Oil V These curious names spring from the custom, not uncommon in the middle ages, of designating languages by the sign of affirmation ; just as Dante calls Italian ' la lingua di si.' The modern French out was oil in the North, and oc in the South of France. The ' Langue d'Oil/ which prevailed in districts inhabited by populations whose characteristic differences were strongly marked (the Normans, Picards, Burgundians, &c., having their own peculiarities of pronunciation), was broken up in its turn into corresponding dialects. There was no one capital; each great feudal district was independent, with its own political and literary life, its own tongue, manners, and customs. Thus in Normandy or Picardy all official acts and literary works were in the Norman or Picard dialect; the dialect of the He de France, or French, as it then was called, was regarded in Normandy as almost a foreign language. There were in the middle ages four principal dialects of the ' Langue d'Oil' Norman, Picard, Burgundian, and 1 A line drawn from La Rochelle to Grenoble will fairly represent the frontiers of the two dialects ; north of it we have the ' Langue d'Oil,' south of it the ' Langue d'Oc.' This ' Langue d'Oc,' or, as it is now more commonly called, Proven9al, from the chief district in which it obtained, was developed alongside of the Northern dialect ; and in the twelfth century was the parent of a brilliant lyrical literature. The rivalry of North and South, which ended in the Albigensian war, and the defeat of the South, destroyed this Provencal litera- ture. In A.D. 1272 Languedoc became French, and the French dialect soon prevailed. The Provensal, Languedoc, and Gascon patois, which still remain in the South, are but the fragmentary remains of this ' Langue d'Oc,' which was so brilliant a language for two centuries. HISTORY OF THE FRENCH LANGUAGE. 19 French 1 (of the He de France) in the centre of the triangle formed by the other three. These four dialects, which were equal in power and influence, had such marked differences that even strangers were struck by it : Roger Bacon (who was in France in A.D. 1240), when seeking to shew in his ' Opus Majus' what the dialects of a language may be, chooses French as his example. ' The idioms of the same language vary in different districts, as is clearly the case in France, which has numerous varieties of idiom among the French, the Normans, the Picards, and the Burgundians ; and what is correct speech in Picardy, is looked on as a barbarism by the Burgundians, and even by the French V These differences of dialect, as in the Greek language, did not touch the syntax, but only the forms of words : thus, for example, amabam became, in the twelfth century, ameve in Burgundy, amoie in the He de France, and amoue in Normandy. This word shews us how Latin words shrank and became stiffer as they went northwards : they form a kind of sensitive thermometer, which falls as it goes farther from the South ; and this, not ' per saltum/ but by gradual change. May we not conclude that words, like plants, are modified by climate, which is one of the factors of language, as mathematicians say? In the middle ages, these four dialects (like the four Greek dialects, Ionian, Aeolian, Attic, and Dorian) produced four distinct literatures : we can immediately distinguish a Nor- man from a French or a Burgundian writer. Each of these 1 ' Frenchman,' in the middle ages, was exclusively the name of the inhabitant of the He de France. 2 l Nam et idiomata variantur "ejusdem linguae apud diversos, sicut patet de lingua Gallicana quae apud Galileos, et Normannos, et Picardos, et Burgundos multiplici variatur idiomate. Et quod proprie dicitur in idiomate Picardorum horrescit apud Bur- gundos, imo apud Gallicos viciniores.' Roger Bacon, Opus Majus, iii. 44. c 2 20 INTRODUCTION. languages had a separate and complete existence : we have now to see how the four were reduced to one, and why the dialect of the He de France was adopted as the common tongue rather than the Norman or Burgundian. The feudal system, in parcelling out the country, had secured the independence of the chief districts in politics, language, and literature ; and similarly, when the feudal system gave place to a central monarchy, the dialects also fell, and were suppressed by a central language. The dialect of the dominant province was sure to become the language of the whole people. Thus the language must depend on political movements ; and the election of Hugh Capet, Duke of France, to be king, settled the question, and made Paris the capital of France. Still, throughout the eleventh and twelfth centuries, the Capetian sovereigns, lords of little but the He de France and the Orleans territory, had no influence outside the royal domain; and the dialects retained their independent equality. But by the middle of the thirteenth century the sovereignty of the Capets grew stronger, and with its growth the French dialect also increased. The lords of the He de France are always growing stronger. In A.D. 1101 they get Berry; Picardy falls to Philip Augustus in A.D. 1203, and Touraine after it; Normandy follows in A.D. 1204; Languedoc is added in A.D. 1272, and Champagne in A.D. 1361. The French dialect followed the triumphant progress of the Duke of France, and drove out the dialects of the conquered provinces. Thus, to take Picardy as an example, French was first introduced into the official acts of the conquerors, then into literary works, and finally it was adopted by all who wished to be thought gentlemen. The people alone resisted and kept their ancient speech; and the Picard, no longer written, but only spoken by the commons, and subject to incessant alterations, fell from the rank of a dialect to that HISTORY OF THE FRENCH LANGUAGE. 21 of a patois, that is, a spoken idiom, not recognised by the French language. And so, in less than three centuries, the Norman, the Picard, and the Burgundian dialects were supplanted by that of the He de France, and became mere patois, under which attentive observation alone can discover any of the charac- teristics of those mediaeval dialects whose monuments survive in their respective literatures. But the final triumph of French over the neighbouring dialects was not won without a struggle, in which the victor received many a wound : a certain number of forms bor- rowed from the defeated dialects entered into the French language. There are words whose origin can be traced to the Norman or the Burgundian ; words which are not in com- plete harmony with the proper analogy of the French, which are, therefore, easily recognised as strangers. Thus the hard c of the Latin became ch in the He de France, and c in Picardy : campus, cantare, carta, castellum, campania, catus, cappa, cancellus, carricare, &c., became in French champ, chanter, charte, chattel, champagne, chat, chappe, chancel, charger, &c. ; but in Picardy, camp, canter, carte, castel, cam- pagne, cat, cappe, cancel, carguer, &c. Now in these instances, though modern French has generally followed the ch form, it has not done so always; thus it has taken campagne in preference to champagne. In a few cases it has adopted both forms with different senses, though they are in reality the same word : as from campus, champ and camp ; from cappa, chappe and cappe ; from cancellus, chancel and cancel; from carta, charte and carte ; from capsa, chdsse. and caisse ; from castel- lum, chateau and castel ; from carricare, charger and carguer. The same might be shewn to be true with Norman and Bur- gundian forms ; but these may serve as a sufficient example 1 . 1 Such double forms as fleurir and florir, grincer and grincher, 23 INTRODUCTION. This transformation was completed by the fourteenth century; the monarchy, previously so weak, became all- powerful, and with it rose the dialect of the He de France ; the other dialects fell into discredit, and dropped into patois, while that of the He de France became the French language. In brief, the popular Latin, transported into Gaul, pro- duced at the end of eight centuries the c Langue d'Oil/ one of whose divisions, or dialects, that of the He de France, sup- planted all the rest, and, in the fourteenth century, became the French language 2 . The same process went on in the attaquer and attacker ', ecorcher and ecorcer, laisser and Idcher, char- rier and charroyer, plier and ployer^ are also due to the dialects, and were originally the same word. Now that the history of the language has furnished us with the true explanation, it is amusing to see the grammarians decreeing that plier and player are dif- ferent in origin, and have all manner of distinctions between them. 2 Let us sum up the elements of the language. Its foundation is popular Latin with a strong German element introduced in the fifth century ; a few faint traces of Celtic may be noticed in it. When this language was fully formed, some oriental elements were thrown in about the thirteenth century ; in the sixteenth were added a number of Italian and Spanish words; in the nineteenth several expressions of English origin were accepted ; to say nothing of the scientific words drawn from the dead languages and brought in by the learned, chiefly in the six- teenth and the nineteenth centuries. The oriental elements are Hebrew and Arabic. It was a favourite theory of old etymologists that all languages are derived from Hebrew; but modern philology has proved them wrong, and has established as a law that * the elements of language answer to the elements of races.' Now the Frenchman does not belong to the same race as the Jew ; and such resemblances as may exist between their languages are accidental. When Jerome translated the Old Testament into Latin he incorporated into his version certain Hebrew words which had no Latin equivalents, as seraphim, Gehenna, pascha, &c. ; from Latin they passed at a later time into French (seraphin, gene, paque*). But they entered French from the Latin, not from the Hebrew. The same is the case with the Arabic ; its relations with French have been purely accidental. To say nothing of those words which express oriental things, such as Alcoran, bey^ cadi^ HISTORY OF THE FRENCH LANGUAGE. 23 other Latin countries : the Tuscan in Italy, the Castilian in Spain, supplanted the other dialects, and the Milanais, the Venetian, the Sicilian, or the Andalusian, and the Navarrais, fell from the dignity of written dialects into the position of patois. We will now study the constitution and forms of the French of the thirteenth century, and take note of the path followed by the popular Latin since the fall of the Empire, and of the distance which lies between this old French, and the French of to day. Every one knows that one great difference between French and Latin is that French expresses the relations of words by their position, Latin by their form, The Latin might say equally well 'canis occidit lupum/ or 'lupum occidit canis;' but in French ' le chien tua le loup' is very different from ' le loup tua le chien/ Latin, in fact, has declensions, French has none. We ask, How has this come about ? Were there always six cases in Latin? Has French never had more than one case ? Let us see what answer history will give. The tendency to simplify and reduce the number of cases appeared early in popular Latin : the rough barbarians could not grasp the more delicate shades of meaning expressed by them. They accordingly constructed a new declension to Jeryitbe, firman, janissaire, Sec. which were brought into the west by travellers, the French language received, in the middle ages, many Arabic words from another source : the Crusades, the scientific greatness of the Arabians, the study of oriental philosophies, much followed in France between the twelfth and fourteenth centuries, enriched the vocabulary of the language with many words belonging to the three sciences which the Arabians cultivated successfully : in astronomy it gave such words as azimuth, nadir, zenith; in alchemy, alcali, alcool, alambic, alchimie, elixir, strop ; in mathematics, algebre, zero, chijfre. But even so these words did not come directly from Arabic to French ; they passed through the hands of the scien- tific Latin of the middle ages. In fact, the oriental languages have had little or no popular or direct influence on French. 24 INTRODUCTION. suit their wants, far more simple, but really far less efficient, at the cost of frequent reproduction of the same form. In the fifth century there were only two cases instead of six ; the nominative to mark the subject, the accusative (chosen because of its frequent recurrence) for the object. Thence- forward the popular Latin declension was (i) subjective case, muru-s ; (2) objective case, muru-m. This afterwards became the base of French declension for the first half of the middle ages; and Old French retained these two cases in the singular and plural. Thus Old French was originally a half-synthetic language, half-way between synthetic Latin and analytic modem French. The reader is referred to the body of this book for the vicissitudes of this declension. It disappeared in the four- teenth century : from the fifteenth century onwards the modern form alone remained 1 . It would be a folly to regret the loss of this old declen- sion : we can only regard it with interest as the bridge over which the French language has passed in its journey from the ancient to the modern world. It shews us too, once more, how parallel in their movement have been the lan- guage and the political history of the country. In the fourteenth century the social edifice built by mediaeval feu- dalism begins to crumble down; after Philip the Fair comes Charles V, who strikes a fatal blow at the independence of nobles and clergy, and begins the reform of the monarchy, which is carried out by Louis XI, by Richelieu, by Louis XIV. Old French moved with the times, seeking to supply the needs of a new form of society. The movement went on throughout the fourteenth century ; the analytical or modern spirit rapidly gained ground : declension in two cases, varia- 1 The secondary modifications, consequent on the dying out of this declension, are considered below, pp. 89-93. HISTORY OF THE FRENCH LANGUAGE. 2$ tions of dialect, were abandoned; and by the end of the century Old French was gone. The fifteenth century saw the birth of modern French. With the mishaps and the shame of the House of Valois society underwent another change; the spirit of modern times began to be felt; the Renaissance dawned. The strong and expressive language of Commines is very like modern French. , By the time of the death of Louis XI, France was reorganised, and her language nearly complete. The opening of the sixteenth century brought in nothing new. The French of Calvin's famous ' Institution de la Religion Chretienne' (A.D. 1535) is completely ripe and full: it expresses with ease all shades of meaning ; and if the lan- guage had remained as it then was, it might have escaped the criticisms of Malherbe and the seventeenth -century savants; but it was damaged by an extravagant influx of foreign words, borrowed from Latin, Greek and Italian. The many expeditions of Charles VIII, Louis XII, and Francis I across the Alps, made the Italian language familiar to the French. The splendour of the Italian Renaissance in literature and art dazzled the French mind, while the re- gency of Catharine dei Medicis gave the prestige of fashion to everything Italian. This Italian influence was omnipotent at the court of Francis I and Henry II, and the courtiers handed it down to the nation. Then appeared a number of hitherto unknown words : the old military terms heaume, brand, haubert, &c., disappeared, and were replaced by Italian words, as carabine from carabina; gabion, gabbione ; escadre, scadra ; parapet, parapetto; fantassin, fantaccino; infanterie, infanteria; alerte, all'erta, &c. And not only war terms : Catharine dei Medicis introduced a number of words relating to court life courtisan from cortigiano; affide, affidato; charlatan, ciarlatano; escorte, scorta; cameriste, camerista; brave, bravo; carrosse, carozza, &c. Terms of art also entered with Primaticcio and 26 INTRODUCTION. Leonardo del Vinci ; as balcon, balcone ; costume, costume ; baldaquin, baldacchino ; cadence, cadenza ; cartouche, cartuccio, &c. ; and lastly, commercial relations between the countries left some deposits in the language, such as Ulan, bilancia ; agio, aggio ; escale, scala ; banque, banca, &c. The Italian party went further still, and tried to shoulder out French words in ordinary speech, and to substitute Italian ones : thus your man of taste would not deign to say suffire, grand revenu, la premiere fois, but baster, grosse intrade, la premiere volte. To this pernicious influence may be added another, the mania for antiquity. It was a time of great classical fervour ; and the admirers of these newly disclosed treasures despised the more homely French, and wished to bring in the majesty of expression and of thought which they found among the ancients. One of them, Joachim du Bellay, ventured to set forth a celebrated manifesto entitled ' Defense et Illustration de la Langue Frangaise' (A.D. 1548), in which he proposed a plan for the production of a more poetical and nobler lan- guage by the wholesale importation of Latin and Greek words in their natural state. He sought to ennoble the French language by borrowing largely from ancient tongues, and to enrich French poetry by introducing the literary forms employed by classical authors. One of the Duke of Orleans' pages, Pierre de Ronsard, a gentleman of Vendome, resolved to carry out Du Bellay 7 s reform. He threw aside the indigenous French poetry, and abruptly introduced Latin epic poetry and Greek tragedy. Thanks to his efforts, France for two centuries regarded these two ancient forms of narrative and dramatic poetry as the only legitimate ones in point of good taste, and as alone capable of noble inspirations. Ronsard also aspired to reform the French language, and to destroy all the traditions of the past. He threw literature into a wretched course of HISTORY OF THE FRENCH LANGUAGE. 2J imitation, which nearly proved fatal to its national character ; he recklessly seized on Greek and Latin words, and dressed up several hundred of them with French terminations : lite- rary Latin and Greek, which had given nothing to the French language before 1 , now played their parts, and, thanks to Ronsard's school, learned words, such as ocymore, enttle'chie, oligochronien, &c., passed in from every side. Ronsard's disciples 2 far outstripped their master. Not satisfied with creating handfulls of new words, they wished to reconstruct words already in being, and to bring the whole language nearer to the Latin type. Thus, for example, the Latin otiosus and vindicare had produced oiseuoc and venger ; but 1 We have already shewn this for the literary Latin. As to Greek, the two languages never came in contact with one another. Marseilles, the only Greek city which could have brought this about, was at an early date absorbed by the Ro- mans. There are indeed some few Greek words in early French, such as chere, somme, parole ; but these do not come straight from the Greek *apa, o-ay/za, TrapajSoX^, but through the Latin which first adopted them and handed them on. 2 We must distinguish between the master and his school. Ronsard was very far above his followers. He had real poetical genius, and as a reformer of language many of his ideas are happy and just. He recommended the pro j- d, z (s). soft. p,f. (q, k, c) ch. t, s (x). hard. i . Liquids : n, m, I, r, II, mm, nn, rr. This letter comes from the Latin n, m, 1. I. From a primitive n : i. Initial : nous, nos ; nez, nasus. ii. Medial : ruine, ruina ; regne, regnum ; mentir, mentiri. iii. Final: son, sonus; raison, rationem; e'tain, stagnura. II. From a primitive m : i. Initial : nappe, mappa ; nefle, mespilum ; natte, matta. ii. Medial : sente, semita ; conter, computare ; singe, simius; daine, dama; printemps, primum-tempus. iii. Final : rien, rem ; airain, aeramen ; mon, ton, son, meum, tuum, suum. 56 PERMUTATION OF LETTERS. Bk. I. Pt.i. III. From a primitive 1 : Niveau (O. Fr. m'vel), libella 1 ; poterne (O. Fr. poster ne, and very O. Fr. posterle), posterula ; marne (O. Fr. marie), margula. M. This letter comes from the Latin m, n, b : I. From a primitive m : i. Initial : mer, mare ; main, maims ; mere, mater. ii. Medial : froment, frumentum ; chambre, camera : compter, computare. iii. Final : daim, dama ; nom, nomen ; faim, fames. II. From a primitive n : nommer, nominare ; char me, carpinus. III. From a primitive b : samedi, sabbati dies. L. This letter comes from the Latin 1, r, n. I. From a primitive 1 : i. Initial : loutre, lutra ; lettre, littera ; langue^ lingua. ii. Medial: aigle, aquila; fils, filius; cercle, circulus; cdble, capulum. iii. Final : seul, solus ; poil, pilus ; sel, sal ; sourcil, supercilium. II. From a primitive r : autel, altare ; crible, cribrum ; palefroi) paraveredus, in the fifth century parafredus ; flair er, fragare. III. From a primitive n: orphelin, orphaninus * ; Paler me, Panormus ; Roussillon, Ruscinonem ; Bologne, Bononia ; Chateau- Landon, Castellum-Nantonis. 1 And compare the English, level. Ch. i. Sect. 2. FRENCH LIQUIDS. 57 R. This letter comes from the Latin r, 1, s, n. I. From a primitive r : i. Initial : regne, regnum ; deroute, derupta. ii. Medial : souris, soricem ; charme, carmen ; dro it, Low Lat. drictus for directus. iii. Final : ver, vermis ; cor, cornu ; enfer, infernum ; hivcr, hibernum. II. From a primitive 1 : i. Initial: rossignol, luseiniola*. 1 ii. Medial : orme, ulmus ; remorque, remulcum ; esclan- dre, scandalum ; chartre, cartula ; chapitre, capitulum. III. From a primitive s : Marseille, Massilia ; orfraie^ ossifraga; varlet, vassaletus*. IV. From a primitive n : ordre, ordinem ; pampre, pam- pinus ; timbre, tympanum ; diacre, diaconus ; coffre, eophi- nus ; Londres, Londinum. LL. This double consonant comes from the Latin 11, lia, lea, cl, gl, tl, chl : I. From 11 : anguille, anguilla ; bouillir, bullire ; faillir, fallere. II. From lia, lea : fille, filia ; Marseille, Massilia ; paille, palea. III. From cl, gl, tl, chl: oreille, auricula; settle, situla; vettler, vigilare ; treille, trichila ; volatile, volatilia. 1 This change of 1 into r had taken place in the late Latin texts, long before the birth of the French tongue : thus, while we find luseiniola in Plautus and Varro, we find in the Mero- vingian MSS. only the forms rusciniola, rosciniola. 5 PERMUTATION OF LETTERS. Bk. I. Ft. r. MM. This double consonant comes from the Latin mm, mn : I. From mm : flamme, flamrna ; somme, summa. II. From mn : femme, femina ; somme, somnus ; sommeil, somniculus* ; homme, hominem. EOT. This comes from the Latin run: colonne, columna; or from gn : connaitre, cognoscere. RR. This double consonant comes, from the Latin tr, dr : I. From a primitive tr: pier re, petra; verre, vitrum; larron, latronem ; pourrir, putrere ; parrain, patrinus ; marraine, matrina. II. From a primitive dr : carre, quadratum ; arriere, ad- retro ; carrefour, quadrifurcus. 2. Labials: />, b,f(ph), v, w. P. From the Latin p : i. Initial : pain, panis ; pre\ pratum. ii. Medial : couple, copula ; etouppe, stuppa; sapin, sapinus. iii. Final : loup, lupus ; champ, campus ; cep, cippus. B. This letter comes from the Latin b, p, v, m. I. From a primitive b : i. Initial : boire, bibere ; bon, bonus. ii. Medial: diable, diabolus; arbre, arbor. iii. Final: plomb, plumbum. Ch. i. Sect. 2. FRENCH LABIALS. 59 II. From a primitive p : double, duplus ; cable, capulum ; abeille, apicula. III. From a primitive v : courber, curvare ; brebis, ver- vecem ; corbeau, corvellus ; Besangon, Vesontionem ; Bazas, Vasatae. IV. From m : flambe, flamma ; marbre, marmor. P, Ph. The French language contains a great number of scien- tific and learned terms, like physique, philosophic, triomphe, in which the Greek letter $, Lat. ph, is to be met with. It would be superfluous to enumerate such elementary and obvious derivations; we will therefore limit ourselves to the remark that the French f comes from the Latin f, ph, v, p. I. From f, ph : i. Initial : faux, falx ; faisan, phasianus ; fumier, fima- rium. ii. Medial : orfraie, ossifraga ; orfevre, aurifaber ; coffre, cophinus. iii. Final : tuf, tofus. II. From a primitive v : i. Initial : fois, vice. (For the change of the Latin i into oi, see p. 52.) ii. Medial : palefroi, parafredus, form of the common Latin for paraveredus. iii. Final : vif, vivus ; suif, sevum ; nef, navis ; bceuf, bovis ; ceuf, ovum ; sauf, salvus ; serf, servus ; cerf, cervus. III. From p : chef, oaput ; nefle, mespilum ; f resale, praesaga. 60 PERMUTATION OF LETTERS. Bk. I. Pt. i. V. This letter comes from the Latin v, b, p. I. From a primitive v : i. Initial: viorne, viburnum; viande, vivenda 1 . ii. Medial : chauve, calvus ; gencive, gengiva. II. From a primitive b : five, faba ; cheval, caballus ; avoir, habere ; tivre, labrum ; souvent, subinde ; ivre, ebrius ; avant, ab-ante ; livre, libra ; niveau, libella; prouver, 'probare ; Vervins, Verbinum. III. From a primitive p : rive, ripa ; se've, sapa ; louve lupa; cheveu, capillum; chevre, capra ; savon, saponem ; savoir, sapere ; crever, crepare. 3. Dentals : /, th, d, s, 0, x,j. T. This letter comes from the Latin t, d. I. From a primitive t : i. Initial : toison, tonsionem ; faon, tabanus. ii. Medial : matiere, materia ; //#/, status ; chateau, castellum. iii. Final : huit, octo ; cuit, coctus ; fait, factus. II. From a primitive d : dont, de-unde ; vert, viridis ; souvent, subinde; Escaut, Scaldis. The Greek th is only found in technical and learned terms, such as theocratic, the'ologie, &c. 1 Originally viande signified vegetable as well as animal nutri- ment. Rabelais tells us 'les poires sont viandes tres salubres' (Pantagruel, iv. 54); and, so late as 1607, in his tragedy, Le Triompbe de la Ligue, Nereus says, speaking of God, ' II donne la viande aux jeunes passereaux' a line from which Racine drew his famous i Aux petits des oiseaux il donne la pature.' Ch. I. Sect. 2. FRENCH DENTALS. 6l D. This letter comes from the Latin d, t. I. From a primitive d : i. Initial: devoir, debere; dans, de-intus; dime, decimus. ii. Medial: tiede, t6pidus; e'meraude, smaragdus; vendre, vender e. iii. Final : sourd, surdum ; muid, modius ; froid, frigidus. II. From a primitive t : i. Initial : done, tune. ii. Medial : coude, cubitus ; Adour, Aturis ; Lodeve, Luteva. iii. Final : le'zard, lacerta ; marchand, mercantem. 3. This letter comes from the Latin s, c, t. I. From a primitive s : i. Initial : seul, solus ; serment, sacramentum ; sous, subtus. ii. Medial : cerise, cerasus; maison, mansionem; asperge, asparagus; Gascogne, Vasconia. iii. Final : mats, magis ; ours, ursus ; epars, sparsus ; sous, subtus; moms, minus. II. From t followed by the compound vowels ia, ie, io, iu : ii. Medial : poison, potionem ; raison, rationem ; oiseux, otiosus ; Venise, Venetia ; saison, sationem ; trahison, tra- ditionem; liaison^ ligationem. iii. Final: palais, palatium; tiers, tertius. III. From a soft c : i. Initial : sangle, eingulum. ii. Medial : plaisir, placere ; voisin, vicinus ; moisir t 62 PERMUTATION OF LETTERS. Bk. I. Pt. i. mueere ; oiseau (O. Fr. oisel, from the common Latin form aucellus), avicellus ; Amboise, Ambacia. Note that the double consonant ss comes from the Latin x ; as for example, essai, exagium ; essaim, examen ; laisser, laxare ; essorer, exaurare : also from an ss, as casser, quas- sare ; fosse, fossa. Z. This letter comes from the Latin s or soft c. I. From s: chez, casa; nez, nasus; rez, rasus (rez-de- chausse'e}; assez, ad-satis ; lez, latus; as in Plessis-lez-Tours Passy-lez-Paris. II. From a soft c : ttzard, lacerta ; onze, undecim ; douze, duodecina, &c. X. From the Latin x, s, c. I. From a primitive x : six, sex ; soixante, sexaginta. II. From a primitive s : deux, duos ; toux, tussis ; epoux, sponsus ; roux, russus ; oiseux, otiosus ; vineux, vinosus. III. From a primitive c: dix, decem; voix, vocem; noix, nucem ; paix, pacem ; chaux, calcem ; faux, falcem. J. From the Latin j, g, i. I. From a primitive j : i. Initial : Jean, Johannes ; je&ne, jejunium ; jeune, juvenis. ii. Medial : parjure, perjurium. II. From g : jouir, gaudere ; jumeau, gemellus ; jaune, galbinus ; Anjou, Andegavi. Ch. i. Sect. 2. FRENCH GUTTURALS. 63 III. From i : Jerusalem, Hierosolyma ; jour, diurnum ; Jerome, Hieronymus ; goujon, gobionem ; Dijon, Dibionem . For the change from i to j, see page 65. 4. Gutturals : c, q, k, ch, g, h. C. C is pronounced gutturally before a, o, and u, and is then called hard: before e, 2, and ce, it is pronounced as a dental, and is called soft. I. C hard. From the hard c of the Latins, or its equi- valent q : i. Initial : coque, concha ; coquille, conchylium ; car, quare ; casser, quassare ; coi, quietus. ii. Medial : second, secundus ; chacun (O. Fr. chascuri), quisque-unus. iii. Final : lacs, laqueus ; one, unquam ; sec, siccus. II. C soft. From the Latin c soft : ciment, caementum ; del, caelum; *cite', citatem, a common Latin form much used under the Empire for civitatem. K. This letter is employed in French terms of mensuration, as the barbarous equivalent for the Greek x, which ought properly to be rendered by ch: thus kilometre is a double barbarism for chiliomelre, ^t Q. This letter comes from the Latin c hard, qu, ch. i. Initial : quel, qualis ; queue, cauda ; queux, coquus. ii. Medial: tranquille, tranquillus; coquille, conchylium. iii. Final : cinq, quinque. 64 PERMUTATION OF- LETTERS. Bk. I. Ft. i. CH. From the Latin c hard 1 . i. Initial : chef, caput ; chose, causa ; chandelle, can- dela ; chandeleur, candelarum [festa] ; chevre, capra. ii. Medial : louche, bucca ; miche, mica ; perche, pertiea ; fourche, furea; mouche, musca; secher, siccare. iii. Final : Auch, Auscia. G hard. From the Latin g hard, c hard, q, v, n. I. From a primitive g hard: i. Initial : goujon, gobionem ; gout, gustus. ii. Medial : angoisse, angustia ; sangle, cingulum. iii. Final: long, longus; e'tang, stagnum; poing, pugnus. II. From c hard'. i. Initial : gobelet, cupelletum* ; gras, crassus ; gonfler, conflare. ii. Medial : maigre, macrum ; langouste, locusta ; vi- guier, vicarius ; cigogne, ciconia. III. From a primitive v: Gascogne, Vasconia; gui, viscum; gue, vadum ; gaine, vagina ; guepe, vespa ; sergent, ser- vientem ; Gard, Vardo ; Gapen$ais, Vappincensium ; gdter (O. Fr. gaster], vastare; guivre, vipera. IV. From a Latin n followed by a vowel : cigogne, ciconia ; Digne, Dinia ; Auvergne, Arvernia ; oignon, unionem ; Boulogne, Bononia. G soft. From the Latin g and the suffixes ia, ea. I. From a primitive g : 1 And from the Greek x i n such technical terms as chirographe , chaos (x^oy), &C. Ch. i. Sect. 2. FRENCH GUTTURALS. 65 i. Initial : gencive, gingiva ; ge'ant, gigantem ; geindre, gemere. ii. Medial : large, largus. II. From the diphthongs ia, io ea, eo. We learn from Quinctilian that the Roman i and j had originally the same sound. For a long time a great uncer- tainty existed as to the use of these two letters. Old MSS. and, after them, printed books down to the middle of the seventeenth century use i and j indifferently : it was not till the year 1750 that the French Academy recognised/ in their Dictionary as an independent letter. This is why the Latin i in some cases has become/ in French (or g soft, which is the same thing). Hierosolyma, simia, diurnus, vindemia, have passed into Jerusalem, singe, jour, vendange, proving clearly that the popular pronunciation of these words was Hjeroso- lyma, simja, djurnus, vindemja. This once granted, it is easy to see how pipionem, tibia, rabies, Dibionem, dilu- vium, eambiare*, abbreviare, &c., have respectively passed into pigeon, tige, rage, Dijon, deluge, changer, abreger, &C. 1 In these words two successive alterations have taken place : (i) from i into/, or (as the Germans call it) the ' consonni- fication' of the letter i (thus pipionem is pronounced pipjo- nem ; rabies, rabjes, Dibionem, Dibjonem, &c.) ; (2) this change of i into/ brings two consonants together, and into a sort of collision (pipionem becoming pipjonem, &c.). Now (as we will shew later on 2 ) in such cases the first of the two consonants disappears ; subjectus becomes sujet, dorsum, dos ; and similarly pipjonem,tibja, rabjes, &c., become pijonem, tija, rajes, &c., whence again come pigeon, tige, rage, &c. 1 It is hardly necessary to remind the reader that the French j is always a soft sibilant, not a soft mute, like our/'. 2 See p. 8 1. 66 PERMUTATION OF LETTERS. Bk. I. Pt. i. Similarly, ea, eo, eu, pass into je, ge, &c. In the regular Latin forms lanea, commeatus, cavea, hordeum, deusque, the e was soon replaced by z', and, long before Merovingian days, inscriptions give us as the usual forms, lania, com- miatus, cavia, hordium, diusque. These diphthongs ia, iu, next exchange their i for j after the rule just noticed ; and then lania, commiatus, cavia, hordium, diusque, having become lanja, comjatus, cavja, hordjum, djusque, passed naturally into lange, conge, cage, orge, jusgue, &c. H. From the Latin h, f : I. From a primitive h : homme, hominem ; hier, heri ; hui (in the word aujourd' hui) , hodie. II. From f : hors, foris; hormis, foris-missum 1 . CHAPTER II. HISTORY OF THE LATIN ALPHABET. The history of the French Alphabet has led us from effect to cause, from French to Latin ; and we have ascended the stream of transformation to its source. We must now follow the reverse course, in studying the history of the Latin letters, examining and describing the modifications they have under- gone before they have descended into the French Alphabet. To avoid useless repetition, we will give as few examples 1 Habler does not come directly from the Latin, but from the Spanish hablar (fabulari), and cannot be traced back beyond the sixteenth century. The Latin /folio wed by a vowel is always commuted into h in Spanish, if at the beginning of a word. Thus fabulari, facere, faba, formica, become hablar, hacer, haba, bormigua. Ch. 2. Sect. i. LATIN TONIC VOWELS. 67 as possible, and will refer our readers back to the paragraphs of the first part of this subject, where he will find a suffi- cient number of illustrations gathered together. SECTION I. HISTORY OF THE LATIN VOWELS. Every word is composed of an accented or tonic syllable, and of one or more atonic, or unaccented, syllables, which either precede or follow the tonic syllable. For example, in the word mercatus the a is the tonic vowel; e and u the atonic vowels. In writing the history of the Latin vowels we may study first the accented or tonic ones, then the unaccented or atonic. i. Accented or Tonic Vowels. Among accented vowels we may distinguish (i) the short, (2) the long, (3) those long by position (i. e. followed by two consonants). This subdivision may seem too fine and minute ; but it is in reality an important one, as will be seen by an example. Ferum, avena, ferrum, have each an ac- cented e; but their resultants in French are very different from one another : the short e becomes / latinised into acum, indicated possession. To designate the lands of Albinus or Sabinus, the Gallo- Romans fabricated the names Albini-acum, Sabini-acum. This termination in the south became ac, in the north ay, e, or y. cl : already treated of on p. 71. ct : already treated of on p. 60. \ Q. See just above, under the hard c. Q. g becomes in French g, j, t. For examples see above, pp. 62, 64. gm becomes m : pigmentum, piment; phlegrna, flemme. gn becomes ;/ : malignum, malin; benignum, be'nin. gd becomes d: smaragda, emeraude; Magdalena, Made- leine; frigidus, froid. J. See above, p. 62. H. See above, p. 66. This letter is often dropped at the beginning of words : habere, avoir ; homo, on; hora, or; hordeum, orge; hoc- illud (O. Fr. oil), out. 4. Labials : p, b, f, ph, v. P. p becomes/, b, v. For examples see above, pp. 58, 59. ps, pt, pn, as initials. This sound is unknown in French, so that the p is dropped in all these cases : ptisana, tisane ; pneuma, neume; psalmus (O. Fr. saume). Where we find Thus Sabiniacum is in the south of France Savignac; but in the north it becomes Savenay, Sevigne, or Savigny. Albini-acum similarly is Aubignac, Aubenay, Aubigne, Aubigny. Final e seems most common in the west of France ; final y in the centre ; final ay in Champagne and the east. But the distinction is not well- marked, and we must not lay too much stress on it. 7^ PERMUTATION OF LETTERS. Bk. I. these sounds reproduced in full psaume, psallette, &c., we may be sure that the words are completely modern. pt, in the middle of words, is changed into /, d\ captivus, chetif; derupta, de'route ; rupta, route; scriptus, e'er it; ad- captare*, acheter ; male-aptus 1 , malade; grupta* 2 , grotte. The words apte, captif, crypte, rupture, &c., are modern. B. b becomes b, v. For examples see above, pp. 58, 59. bt, bs, bj, bm lose their b when they pass into French, and become d, /, s, j, m : cubitus, coude ; dubitum, doute; debitum, dette; subjectum, sujet; submissum, soumis. br becomes ur : abrotonum, aurone; fabrica (O. Fr. faurge), forge. F, Ph. See above, p. 59. V. v becomes v, f, b, g. For examples see above, pp. 58, 59 6 4- 1 Aptus becomes in Old French ate, in Provencal ade. Ate or ade in the twelfth century bear the sense of being in good health, our ' well' ; malade, male aptus, is one who is in bad health. 2 Crypta became crupta in the vulgar Latin of the sixth century ; and we find this word in a Latin text of the year A.D. 887 in the form of grupta, whence the French grotte. PAST II THE TRANSPOSITION, ADDITION, AND SUBTRACTION OF LETTERS. CHAPTER I. OF TRANSPOSITION (OR METATHESIS.) When the letters of a derivative are arranged in an order different from that which they held in the word from which it is derived, we say that it has suffered metathesis (/xeratfecm), that is to say, transposition ; as when the gn of the Latin stagnum becomes ng in the French derivative e'tang. SECTION I. TRANSPOSITION OF CONSONANTS. N : etang, stagnum ; poing, pugnus ; teignant, tingentem. L : Lot, Oltis. E, : pour, pro ; trcuil, torculus ; pauvrete, paupertatem ; truffe, tuber ; troubler, turbulare* ; Durance, Druentia ; brebis, vervecem ; tremper, temperare ; fromage, *fbrmati- cum ; trombe, turbo. SECTION II. TRANSPOSITION OF VOWELS. The vowel i is often drawn towards the vowel which pre- cedes it, whence results a necessary transposition : gloire, gloria ; histoire, historia ; memoir e, memoria ; juin, junius ; muid, modius ; faisan, phasianus. 7o ADDITION OF LETTERS. Bk.I. Pt.2, CHAPTER II. OF THE ADDITION OF LETTERS. The letters added to the primitive word may be either (i) prosthetic (rrpocrOecris), that is to say, put at the beginning of a word ; (2) epenthetic (eVe'i/tfeo-ts), or put in the body of a word ; or (3) epithetic (cnidcc-is), or put at the end of a word 1 . SECTION I. ADDITION AT THE BEGINNING OF A WORD (PROSTHESIS). i. Vowels. Before the initial sounds sc, sm, sp, st (which are hard to pronounce), the French have placed an e, which renders the sound more easy by doubling the s : espace, spatium ; espece, species ; esperer, sperare ; estomac, stomachum ; esclandre, seandalum ; esprit, spiritus ; ester, stare ; escabeau, scabellum ; escient, scientem ; es clave, slavus * ; escalier, scalarium 2 . After the sixteenth century several of these words undergo a farther modification : the s goes out, and its suppression is marked by the acute accent, which is placed upon the initial e: e'tat, statum ; epice, species ; echelle, scala ; e'crin, scrinium ; e'tain, stannum ; etable, stabuluna ; etude, studium ; epais, spissus ; ecole, schola ; etroit, strictus ; 1 These technical names, borrowed from the Greek gram- marians, are here preserved, because they are in use, and are convenient in point of brevity. 2 As has often been said, the French language springs not from the literary Roman tongue, but from the popular or vulgar Latin. Now, in the fifth and sixth centuries, the vulgar Latin had ceased to say spatium, sperare, stare, &c., but ispatium, isperare, istare, as one sees by the inscriptions and diplomas of the Merovingian period. This i, thus prefixed by the people to facilitate the emission of these sounds, becomes e. in French : ispatium, e space ; istare, ester ; isperare, esperer ; &c. Ch. 2. Sect. 2. PROSTHESIS, EPENTHESIS. 79 e'poux, sponsus ; e'pine, spina ; e'pi, spica , e'toile, stella ; epe'e, spatha; Ecosse, Scotia 1 . By a false assimilation an e has been also prefixed to a number of words which, in the Latin, had no .$* : e'corce, eor- ticem; escarboucle, carbunculus, &c. 2. Consonants. 1. Padded: huit t ocfoo m , huile, oleum; haut, altus ; huitre, ostrea ; hieble, ebulum ; hache, ascia; fiuis* 1 , ostium; hurler, ullare (vulgar Latin form of ululare). 2. g added: grenotallc*, ranuncula. 3. / added : tante (O. Fr. ante*), amita. 4. / added (by the junction of the article with the word): Lille , illa-insula ; Her re, hedera ; luette, uvetta ; lors, hora ; lendemain, O. Fr. I'endemain 5 . SECTION II. ADDITIONS IN THE BODY OF THE WORD (EPENTHESIS). 1. h added: Cahors, Cadurci ; envahir, invadere ; trahir, tradere ; trahison, traditionem. The middle ages, here falling in with both the etymology and the historic reason of the words, wrote more logically envaltr, trdir, treason. 2. m added : lambruche, labrusca. 3. n added : langouste, locusta ; lanterne, laterna ; An- gouleme, Iculisma ; convoiter, cupitare* ; concombre, cucu- raerem ; jongleur, joculatorem ; peinfre, pictorem. 1 We pass over technical terms, like scandale, stomacal, stoique, &c. 2 For huts and its derivative huissier, see p. 53. 3 Grenouille in Old French is renouille, a form which does not come from the classical ranuncula, but from the vulgar Latin ranucla, a word which is often met with in MSS. of the sixth century. On the change of cl into /'/ (ranucla, renouille), see above, p. 71. 4 Cp. the English aunt. 5 Instead of saying le lendemain, le Her re, la luette, which are gross errors of the fifteenth century, the more correct forms Ven- demain, Vierre, Vuette, were in use throughout the middle ages. 80 SUBTRACTION OF LETTERS. Bk. I. Pt.2. 4. r added : fronde, funda ; perdrix, perdicem ; ires or, thesaurus. 5. For the addition of a b between the liquids mr, ml, c. (as chambre, camera, &c.), see above, p. 72. SECTION III. ADDITION AT THE END OF A WORD (EPITHESIS). s added : Us, lilium ; legs, legatum ; tandis, tam diu ; jadis, jam diu ; sans, sine ; certes, certe, &c. CHAPTER III. OF THE SUBTRACTION, OR DROPPING, OF LETTERS Letters withdrawn from the primitive words may be taken from (i) the beginning of the word (aphaeresis, a^a/p or (2) from the body of the word (syncope, orvyKoiri]) ; or (3) from the end (apocope, 071-0*07777). SECTION I. OMISSION FROM THE BEGINNING OF THE WORD (APHAERESIS). i. Of Vowels. Boutique, apotheca ; ble, ablatum ; migraine, fj^iKpavia ; leur, illorum ; riz, oryza ; diamant, adamantem ; le, ille ; Gers, Egirius ; sciatique, ischiadicus ; Natolie, Anatolia. 2. Of Consonants. Tisane, ptisana ; pdmer, spasmare * ; loir, gliris ; neume pneuma; or, hora ; orge, hordeum; on, homo; avoir, habere. SECTION II. OMISSION FROM THE BODY OF THE WORD (SYNCOPE). i. Syncope of Vowels. We have seen (above, pp. 67, 68) under what law the Latin vowels passed into the French language : the tonic vowel h.3. Sect. 2. EPITHESIS, APHAERESIS, SYNCOPE. 8 1 always remained, but the atonic vowels varied ; if short, they disappeared in two positions, ( i ) from immediately before the tonic vowel, as positura, posture; and (2) when they are penultimate, as regula, regie; but, if long, the atonic vowel always remained. 2. Syncope of Consonants. In every word the consonants can occupy two positions which differ with regard to the vowels : either (i) they are put between two vowels, as the b in tabanus, in which case they are called 'medial;' or (2) they are followed by another consonant, as b in submissum, when they are called ' non-medial/ 1. Non-medial Consonants. In the case of two consonants together, like bm in submissum, the former usually dis- appears in the French derivative : sujet, subjectum ; soumis, submissus ; deroute, derupta ; noces, nuptiae ; che'tif, eap- tivus ; peser, pensare ; avoue, advocatus ; coquille, con- chylium 1 , &c. Thus too the Latin s which had survived in most French words up to the end of the sixteenth century (cp. the O.Fr. aspre, pastre, paste, from the Lat. asper, pastor, pasta*), disappeared in the seventeenth century, and its suppression was denoted by the introduction of a circumflex accent : dpre, pdtre, pdte 2 . 2. Medial consonants. The dropping-out of these is an important element in the formation of the French language. (i) Dentals, d : cruel, crudelis ; suer, sudare ; dtmie, denudatus ; moelle, medulla ; obfir, obedire. 1 The subject of the syncope of consonants has hitherto been but little studied, and it is not yet known what exact law it follows. 2 Except in the case of the three words mouche, musca ; louche, luscus ; citerne, cisterna, in which the s disappeared much earlier. 82 SUBTRACTION OF LETTERS. Bk. I. Ch. 3. Sect. 3. t : douer, dotare; muer, mutare; rond, rotundus; saluer, salutare. (2) Gutturals, c: plier, plicare ; jouer, jocare ; vqyelle, vocalis ; delie, delicatus ; prier, precari. g : nier, negare ; ge'ant, gigantem ; nielle, nigella ; aouti augustus ; maitre, magister. (3) Labials, b : taon, tabanus ; viorne, viburnum ; ayant t habentem* v: paon, pavonem ; peur, pavorem; viande, vivenda 1 ; ctieul, aviolus*. SECTION III. CONSONANTS DROPPED AT THE END OF THE WORD (APOCOPE). i. Apocope of Vowels. On this subject see above, p. 70. 2 . Apocope of Consonants. t : gre, gratum ; aime, amatus ; aigu, acutus ; e'cu, scutum ; abbe, abbatem ; &c. n : four, furnus ; chair, camera ; cor, cornu ; htver, Mbernum ; jour, diurnum ; cahier (O. Fr. quaier), quater- num; aubour, alburnum. 1: out (O. Fr. oil), ho[c]-illud; nenni (O. Fr. nennil), non-illud. 1 See above, p. 33. PAET III, PROSODY. Prosody is that part of grammar which treats of the modifications of vowels arising from quantity and accent. Vowels can be modified in three ways, (i) In their nature: e. g. a may become o. The study of these modifications will be found under the head of the Permutation of Vowels on pp. 48-54. (2) In their length: they may be short, as in patte, or long as in pdtre. Here we have the study of quantity. There is but little to be said about it, except that it is very vague in the French language; it is never certain except in such words as mur (O. Fr. metir, Lat. maturus), which words are contractions ; or in such words as pdtre (O. Fr. pastre), in which the s has been dropped. In these two sets of words the vowel is certainly long. (3) In their elevation or accentuation. They may be tonic, as the a in celibat, or atonic, as the a in pardon. This is the study of accent- Now there are four kinds of accent, which must be kept distinct, though they are often confounded together : Tonic, Grammatical, Oratorical, and Provincial. I. Tonic Accent. In the Introduction we described f tonic accent/ or more simply ' accent/ as the incidence of the voice upon one of the syllables of a word. Thus in the word raison, the tonic accent lies on -the last syllable, but in raisonndble it is on the penultimate. G 2 84 PROSODY. Bk. 1 The accented or tonic syllable is, therefore, that on which more stress is laid than on any of the others. In Greek this elevation of the voice is called TOVOS or Trpoo-wSia, words rendered in Latin by accentus. This tonic accent gives to each word its special character, and has been rightly called ' the soul of the word.' In French the tonic accent always occupies one of two places : either (i) it is on the last vowel, when the termination is masculine, as chanteur, ai?ne'r, fimr, seigneur ; or (2) on the last vowel but one, when the termination is feminine, as sauvdge, ve'rre,porche In Latin also the accent occupies one of two places : the penultimate, when that syllable is long, as cantorem, amare, finire, seniorem; or the antepenult, when the penultimate syllable is short, as sylvaticus, porticus. If the reader will compare these French and Latin examples, he will notice at once that the Latin accent survives in the French ; that is to say, the accented syllable in Latin is also the accented syllable in French (cantorem, chanteur ; amare, aimer ; finire, finir ; seniorem, seigneur. This continuance of the Latin accent is a matter of con- siderable importance, and is, we may fairly say, the key to the formation of the French language. Its importance has been explained in the Introduction, to which (pp. 32-35) the reader is now referred. II. Grammatical Accent. In French grammar there are three accents - acute, grave, circumflex. Accent, in this sense, is a grammatical sign, which has three different functions in orthography. (i) Sometimes the accent indicates what is the proper pro- nunciation of certain vowels, as bonte, regle^pole. (2) Some- times it marks the suppression of certain letters, as pdtre, pastor; dpre, asper ; dne, asinus ; which words in Old French were pastre, aspre, asne. (3) And lastly it is used ACCENTUATION. 85 to distinguish between words otherwise spelt alike, but of different significations ; as, du and dii, des and des, la and la, tu and ///, sur and sur, &C. 1 III. Oratorical Accent. The tonic accent affects syllables within words, but ora- torical accent (otherwise styled 'phraseological') influences words within sentences. Thus oratorical accent belongs to the domain of declamation and rhetoric, and naturally has had no influence on the transformation of Latin into French words 2 . We shall therefore have no need to trouble ourselves with it in this place. IV. Provincial Accent. By provincial accent we understand the intonation pecu- liar to each province, differing from the intonation of good Parisian pronunciation, which is taken as the standard. And this is in reality what is meant by the phrase, ' He who speaks French well has no accent 7 that is, no provincial accent. The study of these characteristics of the inhabitants of certain districts does not belong to our subject, and is therefore set aside. Let us, however, say that provincial pro- nunciation limits itself to this it gives a word two accents, and lowers the value of the principal (or proper) one by subjoining to it a slight half-accent on another syllable. 1 Cp. Littre, Diet. Hist. s. v. ( Accent.' These French gram- matical accents which act as signs in writing differ widely from those of the Greek language, though borrowed from them. The acute, grave and circumflex accents in Greek simply denote the tonic syllable, and the shades of intonation on that syllable. In French, on the contrary, these accents have no connection with the tonic and etymological accent, and are purely orthographic symbols. 2 See G. Paris, Accent latin, p. 8. BOOK II. INFLEXION, OR THE STUDY OF GRAMMATICAL FORMS. BOOK II will be entirely given up to the study of inflexions; that is to say, of the modifications undergone by a noun when declined, by a verb when conjugated. Declension of substantive, article, adjective, and pronoun, and conjuga- tion of verbs, will naturally form the two divisions of this Book. To make the study of the different parts of our subject complete, we will under this division include all invariable, as well as inflected, words. PAET I. DECLENSIONS. CHAPTER I. THE SUBSTANTIVE. Let us take in order (i) case, (2) number, and (3) gender. SECTION I. % CASE. Of the six cases of Latin declension, the nominative alone indicated the subject, the other five the l government ' or relation. Now if we place Latin and French side by side we shall see that the six cases of the mother tongue are reduced to one in the daughter language. How has this come about ? Have those six cases always existed in Latin, or has the French never had more than one ? We must again turn to the history of the language; it will provide us with an answer. The tendency to simplify and reduce the number of cases was early felt in the popular Latin : the cases expressed shades of thought too delicate and subtle for the coarse mind of the Barbarian. And so, being unable to handle the learned and complicated machinery of the Latin declen- sions, he constructed a system of his own, simplifying its springs, and reducing the number of the effects at the price of frequently reproducing the same form. Thus the Roman distinguished by means of case-terminations the place where THE SUBSTANTIVE. 89 one is, from the place to which one is going : ' veniunt ad domum/ ' sunt in domo/ But the Barbarian, unable to grasp these finer shades, saw no use in this distinction, and said, in either case alike, ' sum in domum/ ' venio ad domum/ Thus, from the fifth century downwards, long before the first written records of the French language, popular Latin reduced the number of cases to t\vp : (i) the nomina- tive to mark the subject; and (2) that case which occurred most frequently in conversation 1 , the accusative, to mark the object or relation. From that time onwards the Latin declension was reduced to this: subject, muru-s; object, muru-m. The French language is the product of the slow develop- ment of popular Latin ; and French grammar, which was originally nothing but a continuation of the Latin grammar, inherited, and in fact possessed from its infancy, a com- pletely regular declension : subject, mur-fa muru-s ; object, mur, muru-m: and people said 'ce murs est haut' ; 'j'ai construit un mur*.' This declension in two cases forms the exact difference between ancient and modern French. It disappeared in the fourteenth century (as we will explain later on), not without leaving many traces in the language, which look like so many insoluble exceptions, but find their explanation and 1 The fact (which had previously been pointed out by Ray- nouard) was completely established by M. Paul Meyer in 1860, in an Essay before the ' School of Chartularies/ with proofs drawn from the study of Latin MSS. of the Merovingian era. 2 One can see at a glance the consequence of this distinction of cases ; so long as the sense of a word is given by its form (as in Latin) and not by its position (as in Modern French), inversions are possible. Consequently they are frequent in Old French. One could say equally well, ' le rois conduit le cheval ; ' or, as in Latin, ' le cheval conduit le rois (caballww conducit rex). 9 The s which marked the subject (ro/'j, rex), made ambiguity impossible. 9 o DECLENSIONS. Bk. II. Pt.i. historic justification in our knowledge of the Old French declension. This takes three forms, answering to the three Latin declensions : i. Sing. ^ Subjective rosa .. \ Objective rosa-w p, \ Subjective rosae ( Objective rosa-^ 2. ~. (Subjective muru-j.. ( Objective muru-w r>, ( Subjective mur-z' . Plur. ] __ J . ( Objective miir-w .. . ros.e . rosg . roses . roses murs .. mur mur murs ~. \ Subjective pastor .... pdtre (pastre*) \ Objective pastor-em pasteur. p. { Subjective pastor-^ .. pasteur s f Objective pastor-^ .. pasteur s. In the subjective it ran thus : ' la rose est belle ; ' ' le murs est haut;' Me pdtre est venu;' in the objective, 'j'ai vu la rose, le mur, le pasteur' &c. 2 On looking into these declensions one is struck with the facts that (i) the Latin accent is always respected; and 1 Pdtre, in Old French pastre. Pastre and pasteur were not in Old French two distinct words, but only the two cases of the same word. 2 In all these examples of Old French, we ought to have written // murs, not le murs, li being the nominative singular, and le the accusative (as may be seen below, p. 100, in the chapter on the Article) : but as we wish to pass gradually from the known to the unknown, we have for the moment sacrificed correctness to convenience. Ch. i. Sect. i. THE SUBSTANTIVE. 9! (2) that (with the exception of one case) the Old French takes s whenever the Latin has it : in other words, the French declension rests on the natural laws of derivation. Between Latin, a synthetic language, and Modern French, which is analytic, there is an intermediate, or half- synthetic, period. This transition period is marked by the Old French declension, which indicates a resting-point between synthesis and analysis 1 . But this system was still too complicated for the minds of men in the thirteenth century : though the Barbarians had reduced the six Latin cases to two, it was conceived that it would be far more regular to re- duce the three French declensions to one. Accordingly, the second declension was taken as the common form, as it was the most generally used, and its laws were applied to both the others. Now the characteristic of this second declension was an s in the subjective case of the singular ' mun-/ murus ; and accordingly, in violation of the genius of the language and of the laws of Latin deriva- tion, men took to saying ' le pastre^/ as they were wont to say ' le mun-.' The laws of derivation were broken, because the Latin pastor has no s in the nominative ; nor has it any need of that letter, since it is itself distinguished from the 1 Raynouard, who in A.D. 1811 developed the laws of French declension, gave them the general name of l the rule of the j,' by reason of the j which so commonly marks the subject. This discovery is one of the greatest services ever rendered to the study of Old French, and to the history of the language. ' With- out this key,' Littre says most truly, ' everything seemed to be an exception or a barbarism ; with it there is brought to light a system, far shorter indeed than the Latin, but still neat and regular.' Much discussion has taken place as to the usefulness and exact application of this ( rule of the j ' during the middle ages : its practical utility is doubtless restricted, and it has often been broken through ; but the existence of the rule (even more than its utility) is a fact of extreme interest, as it allows us to mark the stages of transition from Latin to French, and is, as it were, a halt in the passage from synthesis to analysis. 92 DECLENSIONS. Bk. II. Pt I. accusative pastorem by the position of the tonic accent. This addition of an s to the nominative of all such words as pastor, which has two forms in French (pastre, pasteur) seemed to simplify the inflexion of nouns; but in reality complicated it, and has in fact destroyed the whole system of French declension. For hereby the French declension, which had previously rested on the natural laws of deriva- tion, came to be founded on this suffix s, which is nothing but an arbitrary and artificial form. In its first period (ninth to twelfth century) this declension depended on etymology ; in its second (twelfth to fourteenth century) it rested on mere analogy : the former is natural, the latter artificial ; the former came from the ear, the latter from the eye. Thus then, in its first epoch, the declension was, as we have just seen, natural, based on etymology and the laws of derivation; but for that very reason it was specially fragile, 1 since its rules were only second-hand, in other words, it had relations with Latin forms and accentuation, but had no stability or guarantee in the proper junction and knitting together of its own tongue 1 / And so French declension was destined to perish forthwith, and the unlucky reform, which consisted in combining the three declensions in one by sacrificing the rarer and more individual forms to the more general ones, did not save it from ruin. Rejected from the speech of the populace, from the thirteenth cen- tury downwards, and constantly violated even by the learned, French declension was thoroughly ruined by the time it reached the fourteenth century. It disappeared, and the dis- tinction between the subjective and objective cases perished : thenceforward one case alone was used for each number. And this was the objective (or accusative) case (falconem, faucori) ; for it was usually longer and more consistent M. Littre. Ch. i. Sect. i. THE SUBSTANTIVE. 93 than the subjective (or nominative), and occurred more frequently in course of conversation. Thenceforth the sub- jective case vanished (falco, O. Fr. fauc], and modern de- clension was established. This adoption of the objective case, as the type and form of the Latin substantive had a curious result in the forma- tion of the numbers. In the older declension we had SINGULAR. PLURAL. Subject murus murs muri mur Object murum mur muros murs where the objective case was in the sing, mur, in the plur. murs. In the fourteenth century the new declension took, as we have said, the objective for its type, and con- sequently the s of the older objective case murs (muros), became the mark of the plural, while the absence of s for the objective sing, mur (murum) became the mark of the singular. But had the subjective case been taken as the type, and the objective abandoned, instead of the contrary, we should have had murs (murus) in the singular, and mur (muri) in the plural; so that the s, which now marks the plural in that case, would have distinguished the singular instead. From the moment that final s ceased to be the character- istic of the cases, and became the distinctive mark of the numbers, the French mediaeval system of declensions ceased to exist ; the fifteenth century utterly ignored it ; and when, in the time of Louis XI, Villon attempted to imitate in a ballad the language of the thirteenth century, he failed to observe the 'rule of s,' and his imitation consequently wants the distinctive mark of the middle ages. It is curious to see in the nineteenth century the detection of the mistakes committed by a writer who in the fifteenth tried to write a ballad in the manner of the thirteenth. 94 DECLENSIONS. Bk. II. Pt. i. Since declension in two cases was, as we have seen, the distinctive and fundamental characteristic of Old French, the loss of these cases immediately established between Old and Modern French a line of demarcation far more distinct than any which exists in Italy or Spain between the language of the thirteenth and that of the nineteenth centuries. There survived, however, some important traces of the Old French declension, which look to us like inexplicable anomalies explicable, in truth, only by a knowledge of the history of the language. Before entering on the study of gender let us re-state the consideration of the Old French declensions one by one, and so discover the traces they have left in modern French. i. Second Declension. Here we have suppressed the sub- jective, retained the objective case (niur from murum, serf from servum, &c.). Still, some relics of the subjective case are retained in the nine following words : fils, filius ; fonds, fundus ; lacs, laqueus ; legs, legatus ; Us, lilius ; lez, latus 1 ; putts, puteus; rets, retis; queux, coquus. In Old French these words all had also the objective case fil, filium ; fond, fundum ; lac, laqueum ; leg, legatum ; li, lilium ; //, latum ; puit, puteum ; ret, retem ; queu, co- quum. In these instances, then, the objective case has disappeared, while the subjective has survived 2 . In this way we may explain by the history of the second declension the formation of the plural in aux : mal, maux ; cheval, chevaux, &c. 1 j, x, s, regarded as orthographic signs, are equivalents in Old French ; *voix was written indifferently e voix, *vois, or voiz. A trace of this usage remains in nez, nasus ; lez, latus ; and in those plurals which end in x (cailloux, feux, maux), which used in Old French to be written with either an s or an .v. 2 It is just the same in the case of certain proper names, Charles, Carolus ; Louis, Lodovicus ; Fervins, Verbinus ; Orleans, Aurelianus ; &c. Ch. I. Sect. i. THE SUBSTANTIVE. 95 In the thirteenth century the second declension was as follows : SINGULAR. PLURAL. mals malus mal mail mal malum mals malos But the / is softened into u when it is followed by a con- sonant (as paume, palma ; aube, alba ; sauf, salvus), and so mals became maus. SINGULAR. PLURAL. maus malus mal mail mal malum maus malos Thus, when the fourteenth century abolished declension by abandoning the subjective case, and keeping only the objec- tive, they had only mal (malum) in the singular, and maus or maux (malos) in the plural. So too chevaux, travaux, &c., may be traced. 2. Third Declension. In this declension in Latin the accent is displaced in the oblique cases (pastor, pastorem) ; whence it follows, as we have seen, that the third French declension had a double form : the one pas ire (pastor) in the subjective case ; the other pasteur (pastorem) for the objective, In this declension, as in the second,*the objective case got the mastery at the same epoch, as may be seen by looking at a few instances : SUBJECTIVE. OBJECTIVE. abbas abbe abbatem abbe falco fauc falconem faucon latro lerre latronem larron serpens serpe serpentem serpent infans enfe infantem enfant Here the subjective cases have all perished, the objective cases have survived. 96 DECLENSIONS. Bk. II. Pt. j. There are a very few instances to the contrary, in which the subjective case has been retained: SUBJECTIVE. OBJECTIVE. soror sceur sororem seror pictor peintre pictorem painteur antecessor ancetre antecessorem ancesseur traditor traitre (O.Yr.trattre) traditorem traiteur In many other words the two forms have survived side by side ; but instead of continuing to be the two cases of one word, they have become two different words : as cantor chantre cantorem chanteur senior sire (O.Yr.smre) seniorem seigneur 1 SECTION II. GENDERS. The French language has adopted only the masculine and feminine genders, rejecting the third Latin gender, the neuter. The student of grammar must approve of this suppression of the neuter, for the Latin tongue had utterly lost all appreciation of the reasons which had originally made this or that object neuter rather than masculine ; and furthermore Low Latin, by uniting these two genders in one, had prepared the way for this simplication of language, which was afterwards adopted in the Romance languages. The neuter is useless except when, as in the case of English, it belongs exclusively to whatsoever is neither male nor female. 1 The Latin genitive left some traces in Old French. It is vain to quote these forms, as Modern French has rejected them all with the exception of leur, illorum, and chandeleur, cande- larum (festa). Ch. i. Sec. 2. THE SUBSTANTIVE. p; This suppression of the neuter, which dates very far back, long, indeed, before the irruptions of the Barbarians, was brought about in two ways : 1. Neuter substantives were altered to masculines. Even in Plautus we find dorsus, aevus, collus, gutturem, cubitus, &c. : in inscriptions dating back beyond the fourth century, we have brachius, monumentus, collegius. fatus, metallus, &c. : in the Salic law, animalem, retera, membrus, ves- tigius, precius, folius, palatius, tempi us, tectus, stabulus, judicius, placitus, &c. It is useless to multiply proofs of this fact, which a rhetorician of the Empire, Curius Fortu- natianus, who flourished about A.D. 450, had already ob- served, and transmitted to posterity in these words, ' Romani vernacula plurima et neutra multa masculino genere potius enunciant, ut hunc theatnim, et hunc prodigiumV 2. Neuter substantives became feminines, the neuter plural in a (pecora) having been mistaken (a strange error !) for a singular nominative of the first declension. In texts of the fifth century we find such accusatives as pecoras, per- gamenam, vestimentas, &c. We may now notice certain peculiar points which will help us to explain such anomalies as amour, orgue, hymne y deuces^ which are true grammatical irregularities. All Latin masculines ending in or became feminines in French : dolorem, douleur ; errorem, erreur ; calorem, chaleur ; amorem, amour. This feminine vexed the Latin- ists of the sixteenth century ; and as they preferred Latin to French, they tried to turn all these words into masculines, le douleur, le chaleur, &c. This attempt failed, as it de- served, except in the cases of honneur and labeur, which are masculine, and of amour , which has both genders 2 . 1 P. Meyer, fctude sur Vhistoire de la langue franqaise, pp. 31, 32 ; Littre, p. 106. 2 Littre, p. 106. 98 DECLENSIONS. Bk. II. Pt. r . Hymne was originally masculine, and the feminine use of it (in speaking of church hymnology) has nothing to justify it either in etymology or in the history of the word. Gens is properly feminine, and the idea it expresses (of men or individuals) is properly masculine ; consequently this word has both genders. But it may be said generally that these distinctions of words, sometimes masculine and sometimes feminine (as automne, gens, &c.), and of words masculine in the singular, feminine in the plural (as amour, orgue, de'lices, &c.) are mere barbarisms and idle subtleties invented by grammarians, not arising from the historical growth of the language. SECTION III. NUMBERS. French, like Latin and Aeolian Greek 1 , has two numbers, singular and plural. Of these, the latter is distinguished from the former by the addition of the letter s. And how is this? If we consider Modern French by itself, without referring back to its ' origines/ we find it impossible to understand why it has chosen this letter to indicate the plural of nouns. It certainly looks as if it were an arbitrary choice, and as if any other letter might have done as well ; and one might be tempted to see in this choice nothing but an agreement among grammarians to establish the dis- tinction between singular and plural in this particular way ; by making, in fact, a distinction which appeals to a French- man's eyes and not to his ears, as in most cases the s is mute. But in reality there is good reason for this s ; and if we pass from Modern to Old French, we shall see what 1 The Aeolian, unlike the other Greek dialects, had no dual. Ch. i. Sect. 3. THE SUBSTANTIVE. 99 it is 1 . We shall there find, it will be remembered, a declen- sion with two cases : SINGULAR. PLURAL. murs murus mur muri mur murum murs muros. We know that in the fourteenth century the subjective case was suppressed, in both numbers, and the objective retained (mur, murum ; murs, muros). Whence it came that (taking mur as the type of the singular, and murs of the plural) the letter s became the characteristic of the plural. Had the language followed the contrary course, and retained the subjective case, we should have had s as the characteristic of the singular. Certain substantives, like vitrum, glacies, &c., which had no plural in Latin, have one in French ; as verres, glaces, &c. Others which had no singular in Latin, also have both numbers in French : as menace, minaciae ; noce, nuptiae ; r clique, reliquiae ; ge'sier, gigeria ; arme, arma ; gesfe, gesta, &c. Others, again, which had both numbers in Latin, have only the plural in French : moeurs, mores ; ancetres, ante- cessor ; gens, gens. As late as the seventeenth century gens and ancetres had a singular, as we see from a couplet of Malherbe : ' Oh ! combien lors aura de veuves La gent qui porte le turban ' ; and La Fontaine has 'la gent trotte-menue/ Ancetre was employed as a singular throughout the middle ages, and even by Voltaire, Montesquieu, and Chateaubriand. The same is true of the word pleurs. Bossuet followed the seventeenth century when he wrote ' le pleur eternel/ 1 Littre, ii. 357. H 2 1 00 DECLENSIONS. Bk. II. Pt. I . CHAPTER II. THE ARTICLE. There is no article in Latin ; and, though Quinctilian pretended that the language lost nothing thereby 1 , it is certain that this was a real deficiency, and that, in order to supply it, the Romans often used the demonstrative pronoun ille, for the sake of distinctness, where the French now has Ie, la, Us. There are plenty of examples : Cicero says, ' Annus ille quo ;' ' Ille alter ;' ' Ilia rerum domina fortuna/ Apuleius has ' Quorsum ducis asinum ilium ? ' Jerome writes, ' Vae autem homini illi per quern/ &c. Though not rare in classical Latin, this usage is infinitely more common in the popular Latin, especially after the reduction of the six cases to two 2 ; a change which made the use of an article necessary. Popular Latin appropriated to this use the pronoun ille : ' Dicebant ut ille teloneus de illo mercado ad illos necuciantes 3 / This pronoun thus transformed, and also reduced to two cases, became in Old French as follows : SINGULAR. MASC. FEM. Subject ille //' ilia la Object ilium le illam la PLURAL. Subject illi li illae Its Object illos les illas les 1 He says, ( Noster sermo articulos non desiderat' (De Instit. Orat. \. 4). Of all the Indo-Germanic languages, Greek and the Teutonic languages alone have an article. Latin and Sclavonic had none ; Sanskrit only a rudimentary one. 2 About the fifth century. See above, p. 89. 3 From a chartulary of the seventh century. Ch.2. THE ARTICLE. IOI So they said, distinguishing* carsfiiUy betweeri the two cases : 1 Ille caballus fortis ' ' Li caovab e?t fcr; '' ' Ilium vidi caballum ' * ' jf'al vu le 'cheval/ And consequently, when, in due time, the subjective case disappeared, the masculine article became le, ilium, and les illos, and the feminine la, illam, and les, illas. Thus we get to the modern article 1 . Combined with the prepositions de, a, en, the masculine article in Old French gives us : SINGULAR. 1. del (de le}, which became deu 2 , and thence du, as now. 2. al (a le), au, as now. 3. enl (en le}, which has disappeared. PLURAL. 1. dels (de les}, which became des. 2. ah (a les}, aux. 3. es (en les}, which has disappeared, with the exception of a few traces, as in maiire-es-arts, docteur-es-sciences, es-mains, S. Pierre-es-liens. 1 The reader has doubtless noticed that the article is a remarkable exception to the rule of the continuance of the Latin accent in French. M. G. Paris explains this difficulty thus : ' The Latin comic writers reckon the first syllable of ille, ilia, ilium, as short ; and these words may be regarded simply as enclitics, as is shewn by the compound ellum=en ilium. Had the accent been marked, the first syllable would never have been shortened or suppressed in composition. Consequently it is not wonderful that, by a solitary exception, the French language has retained only the latter syllable of this word ; il-le = le; il-la = /a; il-li = lui ; il-los = les. 9 2 For this softening of the final / into u, see above, p. 53. 102 DECLENSIONS. Bk. II. Pt. i. CHAPTER III. THE ADJECTIVE. SECTION I. QUALIFYING ADJECTIVES. i . Case and Number. Adjectives in Old French followed the same rules of declen- sion as substantives, and had at first two distinct cases : SINGULAR. PLURAL. Subjective bon-us = bons boni = bon. Objective bonum = bon bonos = bons. They also followed the same course in the fourteenth century, abandoning the subjective case. We need not therefore reproduce the rules given above (pp. 92-95), which the student may apply for himself to the adjective. 2. Genders. We have laid it down as a general principle, that at the outset French grammar is nothing but a continuation of Latin grammar ; consequently French adjectives follow the Latin ones in every way. Those adjectives which in Latin had two different terminations for the masculine and the feminine (as bonus, bona) used also to have two in French; and those which had only one termination for these genders in Latin (as grandis) had but one in French also. Thus in the thirteenth century men said ' une grand femme, une ame mortelj &c.- The fourteenth century, not understanding the reason of this distinction, supposed it to be a mere irregularity, and accordingly, in defiance of Ch. 3. Sect. i. QUALIFYING ADJECTIVES. 103 etymology, reduced this second class of adjectives to the form of the first class, and wrote grande, cruelle, mortelle, &c., to correspond to bonne, &c. A trace of this older form remains in the expressions grand'mere, grand' route, grand' - faim, grand' garde, grand' hate, grand' chere, &c. phrases which are relics of the older language. Vaugelas and the seventeenth -century grammarians, ignorant of the historic ground for this usage, decreed with their usual pedantry and dullness, that this form came from the euphonious suppression of the e, and that the omission must be noted by an apostrophe. 3. Adjectives used as Substantives. Certain words, now substantives in French, but springing from Latin adjectives, domestique, domesticus ; sanglier, singularis ; bouclier, buccularius ; grenade, granatum ; tinge, lineus ; coursier, course, &c., were adjectives in Old French, following their Latin origin. In Old French the phrase ran thus : Un serviteur domestique, i.e. a man attached to the service of the house (domus). In Old French (with the usual regu- larity of formation) it was written domesche, so as not to neglect the Latin accent (domSsticus). Un pore sanglier, porcus singularis, a wild pig, which is of solitary habits. Un ecu bouclier, clypeus buccularius, literally an arched or bowed shield (or buckler]. Une pomme grenade, pomum grenatum, i. e. a fruit filled with pips or seeds. Une vetement linge, vestimentum lineum, i. e. a linen robe 1 . 1 For the change from lineus to linge, i. e. of -eus to -ge, see p. 66. 1 04 DECLENSIONS. Bk. II. Ft. i . Un cheval coursier, i. e. a horse kept for racing only, as opposed to carriage-horses, &c. In these expressions the epithet has in course of time ejected the substantive, and has taken its place. Then people began to say, ' un domestique] ' un sanglier' &c., just as one now speaks of ' un mort,' meaning f un homme mort/ ' un mortel,' for ' un etre mortel,' &c. 4. Degrees of Comparison. In this, as in all other parts of French declension, par- ticles have taken the place of the inflections -or, -imus, which mark in the Latin the degrees of comparison. Here, as elsewhere, we may note the analytic tendencies of the Romance tongues. I. The COMPAKATIVE is formed by the addition of the adverbs plus, moms, aussi, to the positive, in both Old and Modern French. There is one peculiarity of the Old French which must be noted : beside the form plus . . . que, it possessed, like the Italian, the form plus . . . de ' il est plus grand de moi.' It would do equally well to say, ' il est plus grand de moi/ or ' il est plus grand que moi;' just as, in Italian, we have ' piu grande del mio libro/ Some French adjectives have kept the Latin synthetic form ; as metlleur, meliorem. As the accent is displaced in the objective case (melior, meliorem), there has arisen (as we have seen) a declension with two cases, which are re- solved either into a single case, or into the retention of the two cases with different significations. The five adjectives don, mal, grand, petit, moult, have retained the old comparatives. 1. Bon : O. Fr. mieldre, melior; meilleur, meliorem, 2. Mal: pire, pejor; O. Fr. pejeur, pejorem. 3. Grand: maire, major ; majeur, majorem. 4. Petit: moindre, minor; mineur, minorem. Ch. 3. Sect. 2. NOUNS OF NUMBER. 105 5. Moult, multus; plusieurs, pluriores. The forms derived from the neuter are moms, minus ; pis, pejus; plus, plus; mieux, melius (O.Fr. muls}. We may add senior to this list ; senior has given us the O. Fr. smre 1 , and seniorem gives us seigneur. II. The SUPEKLATIVE is formed by adding le plus, or ires, to the positive. But in Old French ' moult (multum) beau 9 was as correct as tres-beau. Some Latin superlatives lingered on into Old French. In the twelfth century men said, saint-isme, sanctissimus ; alt- isme, altissimus. These vanished in the fourteenth century. The words ending in issime^, &c., which are still found in French, are technical terms, not older than the sixteenth century : like all words which do not come from the popular and spontaneous period of the language, they are very ill- formed, and break the law of accent : generalissinie, reveren- disszme, illustrissime, &c. SECTION II NOUNS OF NUMBER. i. Cardinals. Unus and duo, which are declined in Latin, passed through the same changes in Old French as did substantives 1 This word sinre has passed into sire, just as prins (Lat. pre- hensus) has become pris. - Six centuries before the birth of the French language, the superlative had already been contracted, in common Latin, to ismus from issimus, showing the growing energy and influ- ence of the Latin accent. The 'Graffiti' of Pompeii and the inscriptions of the earlier Empire give us carismo, dulcisma, felicismus, splendidismus, pientismus, vicesma, &c., for carissimo, dulcissima, felicissima, splendidissimus, pien- tissimus, vicesima, &c. 1 06 DECLENSIONS. Bk. II. Pt. i . and adjectives of quality. They had two cases down to the end of the thirteenth century. Subject uns unus dui duo Object un unum deux duos. The phrase then ran thus : ' Uns chevalr et dui boeufr moururent' (unus caballus et duo boves) : and again, 'il tua un cheval et deux bceufo' (unum caballum et duos boves). In the fourteenth century the subjective case was lost, and here, as elsewhere, the objective remained in force. There is nothing in particular to be said about the num- bers /rot's, tres ; quatre, quatuor ; cinq, quinque ; six, sex ; sept, septem; kuit, octo (O. Fr. oiP) ; neuf, novem; dix, decem. In the words onze, undecim; douze, duodecim; treize, tr6decim ; quatorze quatuordecim ; quinze, quindecim ; seize, sedecim, the position of the tonic accent has brought about the disappearance of the word decem, which gave their real force to the words undecim, duodecim, &c. 2 The words which serve to mark the decades, vingt, viginti; trente, triginti; quarante, quadraginta; cinquante, quinqua- ginta; soi Xante, sexaginta; septante, septuaginta; octante, octoginta; nonante, nonagmta, in which the Latin g has disappeared, were originally veint, treante, quareante, &c., whence came the modern contracted form vingt, trente, quarante, &c. Above one hundred, to express the even decades (120, 140, 1 60, &c.), Old French used multiples of twenty, and wrote six-vingt (120), sept-vingt (140), &c. - meaning six times, seven times, &c. twenty; just as to this day 'eighty' is expressed by quatre-vingt, (4 x 20). Traces of this ancient 1 QGto=huit; for the change of ct into it see above, p. 50. 2 See G. Paris, Accent latin, p. 61. NOUNS OF NUMBER. 1 07 usage remain even in our day, as in the hospital ' des Quinze- Vingts' (15 x 20 = 300), which was founded to support 300 blind persons ; so also Bossuet and Voltaire wrote, ' il y a six-vingts ans/ The Latin ambo ( = two together) produced in Old French the adjective ambe ; and the phrases ran, ' ambes mains/ * ambes parts/ instead of ' les deux mains/ ' les deux parts : ' and the word still survives at the gaming-table, ' j'ai gagne* un ambe a la loterie / that is to say, ' I have won a pair/ i. e. on two figures. 2. Ordinals. With the exception of premier, primarius ; second, seeun- dus, which come straight from the Latin, all the French ordi- nals are formed by the addition of the suffix -ieme, -6simus, to the corresponding cardinals : deux-z'^, trois-ieme, &c. But the system adopted in Old French for the first ten ordinals differed from that now in use. They were drawn straight from the Latin, instead of being formed from the French cardinals : thus it had tiers, t6rtius, instead of trois- ieme; quint, quintus, instead of cinquieme. These ten ordinals, prime, primus ; second, seeundus ; tiers, tertius ; quart, quar- tus ; quint, quintus ; siocte, sextus ; setme, septimus ; oitave, octavus ; none, nonus ; disme or dime, d6cimus, have had interesting fortunes of their own in the history of the French tongue : Prime, primus. This word, which has been supplanted by its diminutive premier, primarius, survives still in the phrases '/rz'^-abord/ ' /ra/^-saut/ 'parer en prime' &c. Second, seeundus, has not been suppressed by deuxieme, but has a concurrent existence. Tiers, t6rtius, remains in c AVrj-dtat/ ' tiers- parti/ and (in the feminine) in ' /z aime-a, &c . The Latins often expressed the desire of doing some- thing in the future by habeo joined with the infinitive of the 1 Except in the cases of the imperfect and perfect indicative, aimais, amabam ; aimai, amavi. 1 20 CONJ UGA TION. Bk. II. verb. Even in Cicero we have ' habeo etiam dicere' ; ' ad familiares habeo polliceri' ; ' habeo convenire' ; ' habeo ad te scribere/ * St. Augustine writes, ' venire habet ' ( = he will come). This form of the future ran side by side with the ordinary form in the writers of the Empire, and ended by supplanting it. From the sixth century downwards the forms ' partire habeo/ ' amare habeo/ ' venire habet in silvana/ became the more common, while the regular futures, amabo, partiar, veniet, seem almost forgotten. The Ro- mance (or neo -Latin) languages, as they detached them- selves from the mother-tongue, carried with them this new future ; and retaining the inverted order of the words, amare habeo became at last aimer -~ai l . At first the two elements were separable, and in certain neo-Latin languages, as the Provencal, their combination was not necessary ; and so in Provenal je vous dirai is either ' vos dir-ai] or * dir vos at.' But in French the two elements were early con- nected together, then became inseparable, and before long could not be distinguished. Lacurne de Sainte-Palaye, a scholar of the last century, was the first to notice and remark on this formation of the future ; and his discovery was confirmed by the later labours of Raynouard and Diez. The French conjugations are enriched by the conditional, a mood not known to the Latins. While the Latin con- founds faimasse and f aimer ais, under the one form amarem, the French separates these two senses and gives each its proper form. But what has been the process by which this has been arrived at ? ' The conditional indicates the future looked at from the point of view of the past, just as the 1 In Italian the Latin habeo becoming ho, the future cantare habeo became canter-6 ; in Spanish habeo = >6^, and the future is cantar-e ; in Portuguese habeo = bey, and the future becomes canter-ey. Ft. 2 . PRELIMINA RY. 121 future tense indicates a future looked at from the present. To express this shade of meaning the French language has created the conditional, under the form of an infinitive (aimer), which indicates the future, and a termination which indicates the past 1 ; and hence ai?ncr~ais ) aimer-ais, aimer -ait, &c. In one word, the conditional has been built on the lines of the future ; but the latter has the present for its material (aimer -ai> &c.), the former has the imperfect (aimer-ais, &c.). IV. PEKSONS. Both in French and in Latin the letter s is characteristic of the 2nd person singular, as amas, aimes ; amabas, airnais, &c. The ist person singular never had an s in Latin amo, credo, video, teneo ; consequently, it became in Old French j'aime, je croi, je voi, je Hen. But in the fourteenth century came in the senseless habit (senseless because not based on etymology) of adding s to the ist person singular, and of saying je vien-s, je tien-s, je voi-s. In the eighteenth century Corneille, Moliere, La Fontaine, and Racine wrote the correct formyV croi,je voi,je tien ; and Voltaire has ' La mort a respecte ces jours que je te dol' (Alzire, ii. 2.) But these forms, whose historical foundation was forgotten, appeared to be nothing but poetical licences. The letter / is the characteristic of the 3rd person sin- gular : ama-t, vide-t, legi-t, audi-t, and survived in Old French il aime-t, il voi-t, il //-/, il oui-t, &c. This etymolo- gical / disappeared from the first conjugation, but was retained in the others, // 7z-/, il voi-t, &c. It is a real gram- matical mistake and misfortune that the language has thus come to neglect the primordial characteristics of the 1 -ah, -ais, -ait, -tens, -iez, -aient, represent the Latin -abam, -abas, -abat, &c. 122 CON JUG A TION. Bk. II. Pt. 2. persons, symbols handed down to us by tradition from the highest antiquity. How clearly does the grammar of the old tongue shew its regularity when compared with the irregularities which disfigure modern grammar 1 ! V. Now that we have noted the great differences which separate French from Latin conjugation, we cannot easily begin the study of verbal inflexions in French without say- ing a few words as to the part played by the Latin accent on French conjugation. As regards their tonic accent, all Latin verbs may be divided into two great classes, strong and weak, according as the accent rests on the root (crescere) or on the termina- tion (amare) : thus, crescere, dicitis, tenui in Latin (croitre, dites, tins, in French) are strong verbs, accented on their root ; but dormire, debetis, amavi (dor mis ^ devez, aimai, in French) are weak, with the accent on their termination. This division into weak and strong verbs, or rather into weak and strong forms, for properly speaking there are no completely strong verbs (i. e. verbs which accentuate the root throughout in all tenses and persons), has thrown a strong light on the study of French conjugation, as we shall see in the next chapter. The true natural classification of French verbs should consist in their being divided into strong and weak ; that is to say, according to their form 2 ; but rather than run any risk of confusing the student, we will adopt the gram- marians' artificial classification of verbs according to their functions, and will divide them into Auxiliary, Active, Passive, Impersonal, &c. 1 Littre, Histoire de la langue franchise , i. 17. 2 But even this would not be a perfect division, seeing there are no verbs which are completely strong. Ch. i . A UXILIA R Y VERES. 123 CHAPTER I. AUXILIARY VERBS. The most important difference between the Latin and the French conjugations lies in this, that the passive and several past tenses of the active are expressed in Latin by terminations (am-averam, am-or), while in French they are expressed by the participle of the verb preceded by avoir for the active, and by etre for the passive (as favais aime\ je suis aime). This introduction of auxiliary verbs in conjugation, which seems at first sight foreign to the genius of the Latin language, was not an isolated fact, or an innovation without precedents; in germ it existed in the best ages of the Roman idiom : so Cicero said, ' De Caesare satis dictum habeo' ( = dixi); ' habeas scriptum . . nomen' ( = scripseras); ' quae habes instituta perpolies ( = instituisti). And again, Caesar, ' Vectigalia parvo pretio redempta habet ( = re- demit); 'copias quas habebat paratas'( = paraverat). Thus in the time of Augustus there sprang up beside the synthetic forms * dixi, scripseram, paravi, &c., the analytical form, habeo dictum, habebam scriptum, habeo paratum: after a time this became the form of both common Latin and of the six Romance languages; for this second form spread according as the analytical tendencies of the lan- guage developed themselves, and from the sixth century downwards Latin MSS. provide plentiful examples of it. The same is the case with the inflexions of the passive voice : common Latin substitutes for them the verb sum 1 For the difference between synthetic and analytic forms, see Egger, Grammaire compare?, p. 91. I 24 CON JUG A TION. Bk. II. Pt. 2. joined with the participle of the verb (sum ainatus instead of amor). In the collections of Merovingian diplomas we meet in every page with these new forms : ' Omnia quae ibi sunt aspecta' ( = aspectantur) ; ' Sicut a nobis praesente tempore est possessum' ( = possidetur) ; ' Hoc volo esse donatum' ( = donari) ; ' Quod ei nostra largitate est con- cessuia' ( = conceditur), &c. Just as in the declensions the new languages had aban- doned the terminations of the cases, and had substituted prepositions in their room (caball-i - du chevaT), so in the conjugations they abandoned the synthetic forms of the compound tenses, and replaced them by auxiliary verbs a natural result of that necessity which drove the Latin lan- guage from the synthetic to the analytic state. SECTION I. Eire. The Latin verb esse was defective, and borrowed six tenses (fui, fueram, fuero, fuerim, fuissem, forem) from fore and the unused fuere. In French the verb etre is composed of three different verbs: (i) Fuo, whence the preterite fus (fui), and the subjunctive fusse (fuissem) ; (2) Stare, which gives the past participle etc (O. Fr. este) from status ; (3) Esse, which gives all the rest of the tenses. I. PRESENT INFINITIVE : etre (O. Fr. estre). To such defective verbs as velle, posse, offerre, inferre, esse, which were too short to carry the usual Roman in- finitive, common Latin subjoined the termination -re, and so produced a false resemblance to verbs of the second con- jugation. Thus, from the sixth century downwards, Me- rovingian MSS. give us volere (for velle), potere (for posse), offerrere (for offerre), inferrere (for inferre), essere (for esse). Ch. i. Sect. I. AUXILIARY VERBS. 125 Essere having its accent on the first syllable (essere) became ess' re or estre, which is in fact the French infinitive. This etymology is otherwise confirmed by the form taken by the same verb in the other Romance languages ; in Italian essere, in Spanish and Portuguese ser, and in Provencal esser. And if any one doubts whether the form essere ever did exist, we may easily reply by quoting actual cases. Thus, in Gruter's collection of Roman inscriptions (No. 1062, i) may be read this epitaph found in Rome in a church of the seventh century : l Cod estis fui et cod sum essere abetis,' i. e. ' quod estis, fui, et quod sum esse habetis ( = eritis). In a series of Carolingian diplomas 1 , of the date A. D. 820, are these words: 'quod essere debuissent' ; in the year 821, 'essere de beneficio'; in A.D. 836, 'quod de ista ecclesia Vulfaldo episcopus essere debuisset.' And the same elongation by addition of -re applied to the compounds of esse (as adesse, &c.) is also to be found ; as in the Char- tulary of A.D. 818, ' quam ingenuus adessereV II. PRESENT PARTICIPLE : e'tant. This is formed from etre regularly, as mettant from mettre. III. PAST PARTICIPLE : /// (O. Fr. este), from the Latin status. 1 Perard, Recueil de pieces relatives a Vhistoire de Bourgogne (Paris, 1664), pp. 34-36. 2 Perhaps it may be thought that I have insisted too much, and with two many illustrations, on the proof that etre and essere are the same word. I have done so because I wished definitely to refute a widespread and often-repeated error, namely, that etre comes from the Latin stare. How could stare, whose accent is on the first syllable, have produced etre ? And again, how would stare go with the Provencal esser, the Italian essere, the Spanish and Portuguese ser? And lastly, we know with certainty that stare has become the French ester, and could not have produced anything else. So we have the phrase ' ester en justice, ' = stare in justitia. Ester has also survived in a few compounds, like rester, re-stare; arreter (O. Fr. arrester), ad- re-stare. 126 CON JUG A TION. Bk. II. Ft. 2. IV. PRESENT INDICATIVE. Comes from the correspond- ing Latin tense. SutSj sum (in Old French the form was sui, the more correct, as there is no final ^ in the Latin) ; es, es ; est, est ; sommes, sumus; etes (O. Fr. estes), estis ; sont, sunt. V. IMPERFECT. tat's does not come from the Latin, but has been formed straight from etre, as mettais from mettre?. Side by side with this imperfect of French origin, Old French had another drawn straight from the Latin : fere, eram ; tu eres, eras ; il ert, erat, &c. This form perished in the four- teenth century. VI. PERFECT (or definite past). From the corresponding Latin tense. Fus (O. Fr. fui\ fui ; fus (O. Fr. fuis), fuisti ; fut, fuit ; fumes, fuimus (the circumflex on this word is an error of the sixteenth century, and offends against etymological pro- priety) ; futes (O. Fr.yky/^y), fuistes ; furent, fuerunt.' VII. FUTURE AND CONDITIONAL. Serai (O. Fr. esserai). The French future is, as has been said, a compound of the infinitive of the verb and the auxiliary avoir (aimer ai = amare habeo); and thus esserai represents essere-habeo. The same is true of the conditional serais (O.Fr. twelfth century, esserais). For the formation of the conditional, see above, p. 120. VIII. PRESENT SUBJUNCTIVE. From the corresponding Latin tense. Sot's (O. Fr. sot), sim ; sot's 9 sis ; soil, sit ; soient, sint. The forms soyons, soyez, come from siamus, siatis, not from stmus, sitis (whose resultants ought to have been soins, sot'z) 2 . 1 M. Littre (Histoire de la langue f ran gals e> ii. 201), and after him G. Paris (Accent latin, pp. 79, 132), have shewn that etais or estois could not possibly come from stabam. It is surely a typo- graphical error when M. Littre, in his Dtctionnaire historique de la langue frangaise (s. v. Etre), says, ' etais vient de stabam.' 2 See the rule for the continuance of the Latin accent, above, P- 34- Ch.i. Sect 2. AUXILIARY VERBS. 127 IX. IMPERFECT SUBJUNCTIVE. From the Latin pluperfect. Fusse, fuissem ; fusses, fuisses ; ffit (O. Fr.yto/), fuisset ; fussions, fuissemus ; fussiez, fuissetis ; fussent, fuissent. X. IMPERATIVE. This tense is composed entirely of forms borrowed from the subjunctive (sots, qu'il soit, soyons, soyez, qu'ils soienf). These have been already discussed above, VIII. SECTION II. Avoir. GENERAL REMARKS. The initial h of the Latin habere, avoir, has vanished from the French conjugation, like the h of hordeum, orge ; homo, on 1 ; hora, or, &c. The Latin b has become v : habere = avoir, habebam = avazs, as in prouver, probare ; couver, cubare ; feve, faba ; cheval, caballus, &c. 2 I. PRESENT INFINITIVE. A voir (O. Fr. aver}, habere. II. PRESENT PARTICIPLE. Ay ant, for the Latin habentem (or habendo). The b has disappeared in French, as viorne, viburnum; taon, tabanus, &c. III. PAST PARTICIPLE. Eu (O. Fr. eil, au or aiif) ; in the eleventh century avut, from the Latin habitum. The old form avut shews that, at the beginning, the French language retained the Latin b. IV. PRESENT INDICATIVE. From the corresponding Latin tense. At, habeo; as, habes; a (O. Fr. at], habet the / of the Old French being etymological ; avons (O. Fr. avomes], hab6- mus ; avez, habetis ; onf, habent. V. IMPERFECT. From the corresponding Latin tense. Avais (O. Fr. avoi or avei), habebam (the Old French, always more correct, and true to etymology, had no s in the i st person sing.) ; avais, hab&bas ; avait, hab6bat ; avions 1 See above, p. 116. 2 See above, p. 60. 128 CON JUG A TION. Bk. II. Pt 2. (O. Fr. aviomes), habebamus ; aviez, habebatis ; avaient, habebant. VI. PEEFECT (or Preterite). From the corresponding Latin tense. Eus (O. Fr. eu), habui ; eus, habuisti ; eut, habuit ; eumes habuimus; eutes (O. Fr. eilsies), habuistis; eurent, habue- runt. VII. FUTURE AND CONDITIONAL. Aurai, O. Fr. avrai, twelfth century averai ; which is composed of the infinitive aver (see above, p. 119) and the auxiliary at, reproducing habere-habeo ; and is another in- stance confirming the theory of Raynouard on the formation of the future tense 1 . How useful it is to cite the Old French forms, which lie between Latin and Modern French ! They illustrate the transition and shew how the passage from the one language to the other has been effected. The conditional aurais (O. Fr. avrais), is found in the oldest texts as averais. For the formation of the conditional aver~ais, see pp. 120, 121. VIII. PRESENT SUBJUNCTIVE. From the corresponding Latin tense. Az'e, habeam; aies, habeas; atf, habeat; ayons (O. Fr. aiomes), habeamus ; qyez, habeatis ; aient, habeant. IX. IMPERFECT. From the Latin pluperfect. Eusse, habuissem; eusses, habuisses; eiit (O. Fr. eust, aiist\ habuisset ; eussions, habuissemus ; eussiez, habuisse- tis ; dissent, habuissent. Remark. We have seen (under III) that the past par- ticiple eu was originally dissyllabic eii, answering to its ety- mology. The same is true of the French imperfect. The medial b having disappeared, habuissem became ausse, which came in the twelfth century to ettsse. And the eu of 1 See above, pp. 119, 120. Ch. 2. CLASSIFICATION OF VERBS. 129 eiissions, eiissiez, eiissent, c., was both pronounced and counted in versification as two syllables. X. IMPEEATIVE. The imperative (ate, qyons, ayez) is composed of forms belonging to the subjunctive. (See above, VIII.) CHAPTER II. CLASSIFICATION OF VERBS. CONJUGATIONS. The French verbs, which are 4060 in number 1 , are arranged under four conjugations, according to the termination of the infinitive. The first, ending in -er, is the largest, embracing 3620 verbs. The second, ending in -ir, has 350 verbs. The third, which ends in -oir, counts only 30, and the fourth, in -re, has 60. Thus the first conjugation by itself embraces nine tenths of the French verbs. I. FIEST CONJUGATION (-er). The conjugation ending in -er answers to the Latin first conjugation in -are. As we have seen elsewhere 2 , a be- comes e in French, as nasus, nez; mortalis, mortel; ; whence -are = -er, portare, porter. At first this conjugation embraced only the Latin verbs ending in -are, and consequently has the weak infinitive, amare, aimer. As time went on, learned writers introduced into this conjugation verbs derived from Latin verbs in -ere, which have no true connection with the French conjugation in -er. These verbs, introduced into the French language in the fourteenth century and onwards, are of two kinds : i. Those from Latin verbs which have the weak in- 1 I have based this calculation on the Dictionnaire de r Academic, ed. 1835. 2 See above, p. 67. K J 30 CONJUGA TION. Bk. II. Ft. 2. finitive -ere, as persuad6re, exerc6re, absorbSre, reverSre ; these ought to have found their place in the French third conjugation, under the forms persuadoir ', exer$oir, absorboir, reveroir, &c., just as hab6re, deb&re, make avoir, devoir. Instead of this, which would have been the regular forma- tion, we have the mongrel verbs persuader, exercer, absorber, rtve'rer, &c. 2. Those with the strong infinitive -ere, as affligere, im- primere, t&xere. These words answer properly to the French fourth conjugation in -re (v&ndere, vendre), and ought in French to be afflire, empremdre, tistre^, not affliger, imprinter, tisser, just as pendere, vSndere, tendere, have produced pendre, vendre, tendre, not pender, vender, tender. As to verbs in -ire, there is only one such introduced into this conjugation, namely tousser, tussire ; and even this one is of modern use, for the Old French form was the correct one, tussir. Mouiller and chatouiller, which one might be tempted to put under this head, are not cases in point, as they come from the common Latin forms molliare, catul- liare, not from mollire, catulire. II. SECOND CONJUGATION (-ir). The French conjugation in -ir answers to the Latin fourth conjugation ending in -ire. It embraces words derived from Latin verbs in -ire, as fmire,jinir; in -ere, as flor6re,^#/7r; and in -ere, as colligere, cueillir. There are 350 verbs in this conjugation, which may be subdivided under two very distinct heads : i. Those which follow the Latin conjugation in all their 1 These verbs are not mere inventions ; they are to be found in the twelfth-century texts, instead of affliger, imprimer, tisser. In fact the Dictionary of the French Academy still retains empremdre and tistre. Ch. 2. CLASSIFICATION OF VERBS. 131 tenses and persons : as, for example, venir, venire ; whose present is vt'ens, venio ; imperfect, venais, veniebam ; and so on, each French part coming directly from the correspond- ing Latin inflexion. 2. Those which add -z> to the root, instead of simply fol- lowing the Latin forms : as fleurir, florire ; in the present fleur-is, imperfect fleur-iss-ais, instead offaur, floreo \fleur ais^ florebam; which \vould be formed like vtens, venais, from venio, veniebam. The question arises, What is the origin of these words thus strangely formed? by what procedure has the French language produced them? The answer is this : The Latins had such verbs as durescere, florescere, implescere, gemiscere, which marked a gradual growth (or augmentation) of the action expressed by the simple verb. (So durescere means to grow more and more hard.) These Priscian calls, for this reason, 'inchoative verbs/ Their characteristic syllable is esc, which in French became is : thus flor-esc-o became fleur-is ; fLo?-e&Q-Qb&m, fleur-is s-ais, &c. The French language seized on this syllable, and added it to those Latin verbs which, when transmuted into French, would have produced forms too short and abrupt. But while it adopted this inchoative form in iss for the (i) indica- tive present, empl-is, impl-esc-o; (2) the imperfect, empl-iss- aiSj impl-esc-ebam; (3) the present participle, empl-iss-ant, impl-esc-entem ; (4) the subjunctive, empl-iss-e, impl-esc- am; and (5) the imperative, empl-is, impl-esc-e, it refused it for (i) the infinitive (emplir comes from implere; for impliscere would have produced, not emplir but empletre, like paitre from pascere) ; and consequently (2) the future and (3) conditional tenses, formed as we have seen (p. 121) from the infinitive of the verb and the auxiliary avoir (emplir- ai), have also rejected the inchoative form. So too have (4) the perfect indicative and (5) the perfect subjunc- tive, which come direct from the Latin. K 2 J 3$ CON JUG A TION. Bk. II. Pt. 2. Thus then, to sum it up, these second-conjugation verbs are in two classes : I. The inchoatives, true irregular verbs, with five inchoative and five non-inchoative tenses, as we have just seen; and II. A small class of verbs which we may call non-inchoative (as partir, venir, &c.), which follow faithfully, and reproduce exactly, the Latin verb in all their tenses. At first sight one would say that these ought to be taken as the true types of the French second conjugation, and the inchoatives classed among the irregular verbs. But grammarians have followed the opposite course : the non- inchoative class is banished among the irregulars, and it is decided that the inchoatives are to furnish the typical form of the second conjugation and of its regularity. At any rate numbers are on their side. There are but 2 2 non-inchoatives, to set against 329 inchoatives^. III. THIKD CONJUGATION (-oir). The French conjugation ending in -oir corresponds to that of the Latins (second), which ended in -ere ; as habere, avoir ; debere, devoir. This conjugation embraces only thirty French verbs; and this number may be reduced to seventeen, as the remaining thirteen are compounds. Beside these weak infinitives in -ere, certain strong in- finitives in -ere have contributed to this conjugation : as recipere, recevoir, sapere, savoir, f allege, fattoir, concipere, concevoir, &c. 1 The following are the non-inchoatives: bouillir, courir, cou'vrir, cueillir, dormir, faillir, fulr, mentlr, mourir, qffrir, ouvrir, partir, guerlr, repentir, sentir, sortir, souffnr^ tenir, tressaillir, e venir, vetir. Several verbs, which are at the present day solely in- choative, had in Old French simple forms which they have since lost. Thus we find in Old French Us emplent, implent, in- stead of Us emplissent, implescunt ; ih gement, gemunt, instead of Us gem-iss-ent, gemescunt ; gemant, gementera, instead of gemriss-ant, gemescentem, &c. CLASSIFICATION OF VERBS. 133 IV. FOTJKTH CONJUGATION (-re). This conjugation, answering to the Latin strong (third) conjugation in -ere, includes sixty verbs. It ought properly to embrace only such as are derived from strong Latin verbs (as 16gere, lire ; defendere, defendre] ; but through a mis- placement of the accent it has come to include a number of weak verbs, as ridere, respondere, tondere, mordere, pla- cere, tacere, whose French resultants ought properly to have been ridoir^ repondoir, tondoir, &c. The accent however in these words being wrongly thrown back on the root- syllable (ridere, &c.) the resultant French verb, following the error, has become rire, repondre, tondre, mordre, plaire, laire, &c. Before beginning the study of these conjugations it will be well to point out that the conjugation in -oir differs from that in -re only in the form of the infinitive : -oir : recev-oir, recev-ant, re$~u, re$-ois, re$-us. -re : croi-re, croy-ant, cr-u, cr-ois, cr-us. Such differences as these two conjugations may happen to present arise from modifications of the root, not from changes in inflexion. It is, therefore, perfectly fair to form one conjugation out of these two; and to say that the French language has three conjugations (i) in -er, (2) in -ir, (3) in -oir, or -re. We propose to study the conjugations in detail under these three heads, and in the order here given. O H I H ^ E O U w w & , E o H 8 Q ttS H UGATION. O O *: 2 1 a f 1 1 1 g s 1 T s ::- . ? .?||.$ v?fi|| II -?! 4 T ? 11 d d * *" 3 j cO d. 03 S 3 43 S '+3 d a 3 . O O O O O O O O O Q O O O O D O O O II O 0) CO J a s s s s s S I a 1 s s S : r s. * . - "3 1 go | > " g S 3 ? 5s "? S5 5 W M S 1-1 CO t, H H o ^ g o a 1 s 5 ^ ^, ' 1^ e< . T - r - r , f , r -b ! ^ V V 'r -T V s -r 'r -T ? ? "r v *r 1 ' .2l 2 Illlll 43 d-^eL ^W^i 030303030QOQ ^5 s fl.5 -8 a 03 *- a s .a .2 .2 .2 .2 .2 ii ? -7 1 '? 'V ^ 'T 1 '? '? '? '? '? 'I '? -T -7 1 -7 1 % '? '? '? rH -H 3 c 3 5 s ^ co S 8 - -S -S 1 ^ '5 8 if J S S * 3 a? -1 -8 1 ' i ' i 1 i b cs t; ! ||| a _ |3a |||lj| 43 jj d cs cp ^ ce ^ 00)00)0)0 dddcgdd d eg C8 C3 136 CONJUGATION. Bk. II. Pt. 2. CHAPTER III. FORMATION OF TENSES. The foregoing two pages of tables of terminations are intended to make the formation of the three conjugations in (i) -er, (2) -tr, (3) -oir and -re, clearer to the eye, and set side by side all the tenses and persons of each mood. Opposite each Latin form is placed the corresponding French form, and (when necessary to mark the transition) the Old French form is put between the two, in common type. Thus, when we read under the ist plural present indicative, ' -amus, Conies, -ons,' it means that the Latin -amus be- comes in Old French -omes, and thence -ons in Modern French. Such Latin terminations as are unaccented in this table become mute in French. Remarks. I. PKESENT INDICATIVE. In the second and third conjugations the s has been wrongly added to the ist person sing., as par-s, rend-s. This letter (which violates the rules of etymology) did not exist in Old French, w r hose forms wereyV voi, je rend ; the s was properly reserved to mark the 2nd person sing,, tu rend-s, reddis; tu voi-s, vid-es. For the origin of this s, see above, p. 121. The / which marks the 3rd person sing., ama-t, vide-t, legi-t, audi-t, survived throughout in O. Fr. il aime-t l , il lit, il ou'it. But through one of those strange and inconsequent changes which often meet us in the growth of languages, 1 The -et in aimet is mute, like the -ent of aiment. Ch. 3. FORMATION OF TENSES. 337 and not uncommonly in French, this etymological / dis- appeared from the first conjugation (il aime), while it re- mained in all the others (il lit, voit, ou?f). The ist person plur. (amamus) was originally aim-omes. As time went on all the terminations in -omes were softened down into -ons, and the only relic of the form still to be found in Modern French is the word sommes (sumus), which ought to have been reduced to sons, just as aim-omes has become aim-ons. The third conjugation in Latin (legere) had the ist and 2nd persons plur. 16ginms, legitis, strong ; whence the resultants ought to have been limes, lites, not lisons, tise'z, which are weak forms. The fact is that the word came to be wrongly accented, and pronounced legimus, legitis, whence -the forms lisons, lisez, naturally followed. Dites (dicitis) and faites (facitis), which are regarded as excep- tions by grammarians, are in reality perfectly regular. In Old French the ist person plur. of these same verbs was also strong, dimes (dicinms), in place of disons, and faimes (facimus) instead of faisons. II. IMPERFECT. -abam became in French, following the dialects from south to north, -eve, -oie, -eie, -oue. Thus amabam became in Burgundy am-eve, in the He de France (or in French proper), am-oie, in Normandy am-oue^-. The dialect of the 1 Notice how near the form ameve, which retains the Latin consonant (v = b) is to the original arn-abam. And indeed it is generally true that the Romance forms, which are as clear and sonorous in the south as the Latin itself, contract and be- come dull-sounded, as one goes northward. Thus cantabam became in Spain cant aba, in Italy and Provence cantata, in Bur- gundy chantt've, in the He de France chant ois, in Normandy chan- toue. Latin words are like a very sensitive thermometer : as one goes northward it drops lower and lower; but these changes take place in continued and successive descents, not by sudden falls. ' Natura nil facit per sal turn.' 138 CONJUGATION. Bk. 2. Pt. 2. He de France having gradually supplanted all the others 1 , its imperfect -oie, -abam, prevailed, and became the type of the Modern French imperfect. In the fourteenth century an erroneous s was subjoined to the ist person sing., and hence we get the form -ois (am-ois), which prevailed up to the end of the eighteenth century, when Voltaire substi- tuted for it the now established termination in -ais (awi-ai's). A century before Voltaire, in the year 1675, an obscure lawyer, Nicolas Be'rain, had already suggested this reform. It may be further noticed that the ist and 2nd persons plur. chantions, chantiez, now dissyllabic, were trisyllabic in Old French chant-i-ons, eanta[b]-amus ; chant-i-e'z, canta- [b]atis. The older form marks the force of the Latin accent. III. PERFECT. Cantavi, cantavit, cantavimus, have resulted regularly in chantai, chanta, chantdmes. Chantas, chantdtes, chanterent, however, do not come from cantavisti, cantavistis, canta- v6runt 2 , but from the contracted forms cantasti, cantastis, cantarunt. For the same reason dormis, dormites, dormirent, come from dormisti, dormistis, dormirunt, not from dor- mivisti, dormivistis, dormiv^runt. It may also be remarked that the perfects of the first three conjugations are weak : chant- ai, cantavi ; dor mis, dormivi ; rendtSj reddidi 3 . The strong perfects, vms, veni; fis, feci, belong to the irregular verbs. IV. FUTURE AND CONDITIONAL. These tenses do not appear in the Table of Formation of Tenses, because their proper place is not there. The table 1 This fact is explained above, p. 19. 2 These longer forms, following the law of the influence of the Latin accent, would have produced in French cbanteis, chante'isteS) chanteirtnt) not chantas^ chantastes, cbanterent. 3 For perfects of the third conjugation, see the chapter on Irregular Verbs, p. 142. Ch. 3. FORMATION OF TENSES. 139 is intended to give a comparative view of those tenses which come direct from the Latin, or in other words, of the simple tenses : the future and conditional are compound tenses, made up of the infinitive of the verb and the auxiliary avoir (aimer- ai, aimer -ais). On which point see above, p. 120. V. PRESENT SUBJUNCTIVE. The t which ended the 3rd person sing, of this tense in Latin ame-t, dormia-t, redda-t, &c., though now lost in the French aime, donne, rende, &c., was present in O. Fr. aimet, dormet, rendet. It survives still in the two words ait, habeat, and soit, sit. It is now impossible to distinguish between the imperfect indicative chantions, chantiez, and the present subjunctive. But in Old French they were clearly distinguished ; for the subjunctive forms were dissyllabic, while the imperfect indica- tive was trisyllabic, following the Latin accent : Imperf. indie. : Chant-i-ons, cant-ab-amus ; chanl-i-ez, cant-ab-atis. Subjunct. pres. : Chant-ions, cant-6mus ; chant-iez, cant- etis. VI. IMPERFECT SUBJUNCTIVE. Here, as in the perfect indie. (Ill) the French form is derived from the contracted Latin form : aim-asse does not come from am-avissem, but from am-assem. VII. IMPERATIVE. The 2nd person sing, is formed from the Latin impera- tive aim-e, ama; fin-i, finis, &c. The other persons are usually borrowed from the indicative. VIII. PRESENT INFINITIVE. In addition to the details given in Section II we may here say that certain Latin infinitives in -ere (consequently strong}? have produced strong infinitives in Old French, and weak ones in Modern French. Thus currere, qua&rere, fremere, 1 40 CON JUG A TION. Bk. 2. Pt, 2. gemere, imprimere, have resulted in O. Fr. courre 1 , guerre, freindre, geindre, empreindre, but in Modern French these have become courir, que'rir, fre'mir, gemir, imprimer ; these forms arising from a misplacement of the Latin accent. IX. PRESENT PARTICIPLE. The French language has adopted the form of the objec- tive case, am-antem, aimanP ; not of the subjective, amans. X. PAST PARTICIPLE. All the past participles of what are called regular verbs are weak : aim-e, amatus ; fin-i, fin-itus, &c. There are a few strong forms among past participles ; but these belong exclusively to irregular verbs. Originally, all past participles which were strong in Latin kept the strong form in French : thus vendre, vend-ere, had vent, not vendu, as its past participle. At a later period these forms were made weak by the addition of the final u (mark of the weak participle of the third conjunction). Then the strong forms disappeared from the ranks of participles, though a considerable number of them are still in existence as substantives. Before leaving the past participle we may observe that the Romance languages, and especially French, possess the faculty of being able to form substantives out of past parti- ciples : we can say un regu, un fait t un dii words which are really the past participles of re$evoir, faire, devoir. But this is more especially the case with feminine participles as issue, vue, etouffe'e, venue, avenue, &c. The number of substantives thus added to the language is considerable; for they are formed from both classes of participles, strong and weak: 1 Still used in the phrase ' courre le cerf.' It was in use in .the eighteenth century. ' Aller courre fortune ' is a phrase em- ployed by Mme. de Sevigne, Bossuet, Voltaire, &c. 2 [Or from amando, ' une femme atmant son mari,' ' femina amando suum maritum.'] Ch. 3. FORMATION OF TENSES. 141 1. With weak, or regular, participles: chevauche'e, accouche'e, fauchee, tranche'e, avenue, battue, crue, deconvenue, entrevue, e'tendue, issue, revue, tenue, &c. 2. With strong, or irregular, participles: un dit, un joint, un reduit, un trait, &c. As we have said, these forms dis- appeared as participles, but survive as substantives ; as vente, v^ndita, which is the old form of the participle, now vendue. Subjoined is a list of these substantives 1 ' a list whose special interest lies in the illustration it affords of the history of the Latin accent, and of its influence at the time of the formation of the French language/ By the side of the old strong participle, now a substantive, and the Latin word it comes from, we will place the modern weak participle in parentheses. 1. First Conjugation: emplette, implicita (employee)', ex- ploit, explicitum (tploye). 2. Third Conjugation : meute, mota (mue), and its com- pound emeute, emota (emue) ; pointe, puncta (poindre), mean- ing to prick, pungere (this word has remained as a participle in the expression courte-pointe, O. Fr. coulte-pointe, Lat. culcita puncta) ; course, cursa (courue) ; entorse, intorta (tordue) ; trait, tractum, and its compounds portrait, retrait, &c. ; source (surgie], and its compound ressource, from the verb sourdre (surgere) ; route, rupta (rompue), and its com- pounds deroute, banqueroute (i. e. banque rompue) ; defense, defensa (de'f endue], and its kinsfolk offense, &c. ; tente, tenta (tendue), and its compounds attente, detente, entente, &c. ; rente, reddita (r endue) ; pente, pndita* (pendue), and its compounds, as soupente, susp6ndita* (suspendue] : vente, v6ndita {vendue} ; perte, perdita {perdue) ; quete, quaesita, and its compounds conquete, requete, enquete ; recette, recepta (re^ue) ; dette, d6bita (due) ; reponse, responsa (repondue)', elite, electa (elue). 1 Or rather of such of them as offer any points of interest. 1 4 2 CONJ UGA TION. Bk. II. Pt. 2 , CHAPTER IV. IRREGULAR VERBS (so CALLED). Grammarians have entitled the following verbs ' irregular/ and those treated of in Chapter III ' regular' ; but, if proper regard be paid to the place of the Latin accent, it will be seen that we are right in calling the former verbs strong and the latter weak. The terms 'regular' and ' irregular ' do but state a fact, at best ; but the distinction between strong and weak penetrates deeper, and expresses a theory. Looked at from our point of view, the old con- ception of irregularity disappears, and the word is applied solely to anomalous and defective verbs; and the strong verbs (hitherto named 'irregulars') are considered simply as another method of conjugation. ' Irregularity' presup- poses formations which, for whatever cause, have deviated from the typal form ; but, in the case of strong verbs, no such deviation has taken place : they are as regular as any others, only they obey a different law 1 . The verbs usually styled 'regular' have a weak perfect (i. e. accented on the last syllable), as amavi, aim-di ; dormivi, dorm-is ; redd-idi, rendis, &c., and all regular verbs of the strong type have their perfect strong (i.e. accented on the root), as t post-dater, post-hume, &c., are modern. 1 6. Prae, Fr. pre. Pre'cher, praedicare; prevoir, pre- server^ pretendre, &c. 17. Pro, Fr. por, pour. Pour-suivre, four-crasser, por- trait^ pro-tractus. 1 8. Be, Fr. re, re, r'. Reduire, re-ducere ; repondre, re- cueillir, re-colligere, &c. New compounds are rebuter (but), rehausser (haut), rajeunir (jeune), renverser (envers), de-re-chef, &c. 19. Betro, Fr. riere. In Old French retro made riere (like petra, pierre) ; this form remains in arriere, ad-retro, a prefix found in such compounds as arriere-ban, arriere- boutique, arriere-neveu, &c. [So too derriere, de-retro.] Retro- actif, retro-cession, &c., are modern words. 20. Se, Fr. se. Seduire, seducere ; sevrer, separare, &c. 21. Sub, Fr. se, su, sou, sous. Sour ire, sub-ridere; se- courir, suc-currere ; souvenir, sub-venire. New compound, sejourner (jour). 22. Subtus, Fr. sou, sous. Sous-traire, subtus-trahere ; sous-entendu, subtus-intendere. New compounds are sous- diacre, sous -lieutenant, souterrain. 23. Super, Fr. sur, sour. Survenir, super -venire ; sourcil, super-cilium ; surnommer, super-nominare. New com- pounds are sur-saut, sur-humain, sur -face, sur -tout. The words soubre-saut 2 , super saltum ; and subre-cargue, 1 Ii>. Old French this particle was separable. Thus par sage ( = tres sage) might be written in two parts, as ' tantpar est sage 9 ( = ' tant il est parsage'). Similarly one may still say ' G'est par tropfort.' 2 Our ' summerset.' N 2 l8o COMPOUND WORDS. Bk. III. super-carrica (the proper French forms are sursaut, and surcharge), are of Spanish origin. 24. Trans, Fr. ire, tra. Tr-averser, transversare ; traduire, trans-ducere ; &c. New compounds are trepas, trans - passus; tressaillir, trans salire 1 , &c. 25. Ultra, Fr. outre. Outre-passer ', outre-cuidance, outre- mer, &c. Such words as ultra-montain, &c., are modern. 26. Vice, Fr. vi. Vicomte, vice-comitem; vidame, vice- dominus. Modern words are vice-roi, vice-consul, &c. 2. Qualitative Particles. These are four in number : 1. Bene, Fr. bien. Bien-fait, bene-faetum ; bien-heureux, bien-venu, &c. 2. Male, Fr. mal, mau. Mal-mener, male-minare ; mal- tr alter, male-tractare ; mau-dire, male-dicere ; maussade, male-sapidus 2 ; malade, male-aptus (see above, p. 76); malsain, male-sanus. 3. Minus, Fr. mes, me. Me'dire, mefaire, meprendre, me'fier, me'sestimer*, &c. 4. Magis, Fr. mais. From this word the conjunction mais is derived, though the French plus has taken the proper sense of the Latin magis : the old use remains in the one phrase * n'en pouvoir mais' 1 Modern words : trans-cnption, trans-port, &c. 2 Sapidus, O. Fr. sade ; whence male-sapidus, maussade. 3 This prefix mes, me, does not come from the German miss, as has been thought, but from the Latin minus an etymology confirmed by the old form of the French prefix, as well as by its form in the other Romance languages. Thus the Latin minus- pretiare becomes menos-preciar in Spanish, menos-prezar in Portuguese, mens-prezar in Proven9al, and mes-priser or me-priser in French. Ch. 2. PARTICLES. l8l 3- Quantitative Particles. 1 . Bis, Fr, be, bi. Be'vue, whose proper sense is = double- vue. Learned words, compounded with bis, keep the Latin form. So biscuit, bis-coctus ; bis-aieul, bis-aviolus ; bis- cornu, bis-cornu, &c. 2. Medius, Fr. mi. Mi-di, media-die; mi-nuit, media- nocte ; mi-lieu, medio-locb ; mi-janvier, mi- car erne, &c. From dimidium we get demi ; so par mi, per medium. 4. Negative Particles. 1. Won, Fr. non. Non-pareil, non-chaloir (whose present particle exists, nonchalant a compound of chaloir, which has been discussed above, p. 147). 2. In, Fr. en. En-fant, in-fantem. The learned form is in; in-utile, in-de'cis. CHAPTER II. ON SUFFIXES OR TERMINATIONS. Suffixes, like prefixes, ought to be considered in their origin and their form. 1. As to their origin. They'may be either (i) of Latin origin, as prem-ier from prim-arius; (2) of French origin, that is, built on the lines of the Latin suffixes (as encr-ier from French encre), but having no corresponding Latin words. 2. As to form. We must carefully distinguish between suffixes formed by the learned, and those formed by the people : between such as prim-aire, secul-aire, scol-aire, which are of the former kind, and such as prem-ier, prim-arius ; secul-ier, saecul-aris ; e'col-ier, schol-aris, which are of the latter description. l82 DERIVED WORDS. Bk. III. SECTION I. OF THE ACCENTUATION OF DERIVED WORDS. Latin suffixes may be classed under two heads : the accentuated, as mort-alis, hum-anus, vulg-aris, &c. ; and the unaccented or atonies, as as-inus, port-icus, mob-ilis. The accented Latin suffixes are retained in the French, as mort-el, hum-am, vulg-aire. These suffixes (el, ain, aire) are further employed in French to produce fresh derived words, by attaching them to words which were without them in Latin : thus have been formed such words as visu-el, lomt-ain, visionn-aire, derivatives constructed for the first time by the French language. Atonic Latin suffixes, like as-inus, port-icus, jud-icem, are all shortened as they pass into the French language 1 , following therein the natural law of accent (as explained above, p. 69). So as-inus produced dm ; port-icus, pore he ; jud-icem, juge. Consequently no subsequent derivatives could be formed from these weak suffixes : it was not till a later period that the learned, ignorant of the part played by the Latin accent in forming French terminations, foolishly copied the Latin form, but gave it a false accent, displacing it from its proper syllable. Then came up such words as portique, portions; mobile ', mobilis; fragile, fragilis 2 ; words formed in opposition to the genius of the French language, bar- barous \vords, neither Latin nor French, which violate the laws of accentuation of both. 1 By the French language must be understood the collection of all words of unconscious and popular formation, as opposed to learned words introduced consciously into the language. 2 Old French, which always observed the law of the accent, said, porche, portions ; meuble, mobilis ; frele, fragilis ; instead of portique, mobile, fragile. Ch. 2. Sect. 2. NOMINAL SUFFIXES. 183 French suffixes are to be distinguished into nominal (sub- stantives and adjectives) and verbal. In each of these classes we will study successively the suffixes which are accented in Latin, and those which are not ; carefully and rigidly exclud- ing every word which has crept into the language since its proper formation. SECTION II. NOMINAL SUFFIXES. f . Suffixes accented in Latin. Alls, Fr. el, al. Mort-el, mort-alis ; chept-el, capit-ale ; hot-el^ hospit-ale; roy-al, reg-alis ; loy-al, leg-alls 1 . Amen, Fr. aim, ain, en. Air-ain, aer-amen; lev-am, lev- amen ; ess-aim, ex-amen ; li-en, lig-amen. I-men. No word with this termination has entered into French. TJ-men, Fr. on. Bet-on, bit-umen 2 . Antia, Fr. ance. Re'pugn-ance, repugn-antia. French derivatives 3 , nu-ance, se-ance, &c. Andus, endus, Fr. ande, ende, Vi-ande, viv-6nda; prov- ende, provid-enda ; leg-ende, leg-enda. French derivatives, offr-ande, reprim-ande,jur-ande, &c. Antem, Fr. ant, and ; entem, Fr. ent. March-and, merc- antem; am-ant, am-antem. Mech-ant (O. Fr. mescheant, par- ticiple of the verb mescheoir, see above, p. 148) comes 1 The learned language has kept al for this suffix ; as in hopit- al, nat-al, capit-al. 2 Learned forms are -amen, ex-amen ; -imen into -ime, reg- ime, reg-imen ; cr-ime, cr-imen ; -umen into ume ; bit-ume, bit-umen ; leg-ume, leg-umen ; vol-ume, vol-umen. 3 By ' French derivatives' are meant derivatives which are formed first-hand by the French language, and have no words corresponding to them in Latin. 184 DERIVED WORDS. Bk. III. from me's = minus (see p. 1 80), and the verb che'oir, cadere : thus mechant represents the Latin minus-cadentem ; serg- ent, servi-entem ; eche'-ant, ex-cad-entem. Anus, Fr. ain. Aub-ain, alb-anus; cert-am, cert-anus* ; rom-ain, rom-anus; hum-Gin, hum-anus. Anus becomes en, ien, after a vowel, or when the medial consonant falls out ; as chrel-ien, christ-ianus ; anc-ien, anc-ianus * ; pa'i-en, pa[g]-anus ; doy-en, de[c]-anus. French derivatives are, hautain, haut; chapelain, chapelk, &C. 1 Enus, ena, Fr. em, in, oin, ene. Ven-in, ven-enum; av- oine, av-ena; ch-aine (O.Fr. chaene), cat-ena. Ardus, Fr. ard. The German suffix -hart, Low Lat. -ardus, which indicates intensity, has furnished the French language with a very considerable number of derivatives, as pleur-ard, fuy-ard, Mt-ard, &c. Aris, arius, Fr. er, ier. Prem-ier, prim-arius : se'cul-ier, saecul-aris : gren-ier, gran-arium ; e'cuy-er, scut-arius ; riv- iere, rip-aria ; e'col-ier, schol-aris ; sangl-ier, singul-aris, sc. porcus ; fum-ier, fim-arium. New derivatives, plen-ier (plein) ; bar r -ier e (bar re], &c. 2 The suffix -ier, perhaps the most fertile in the language, has formed a number of derivatives which had no existence in Latin. It most frequently designates (i) names of trades, as boutiqu-ier, pot-ier, batel-ier, vigu-ier, c. ; (2) objects in daily use, as sabl-ier, encr-ier, fo-yer, &c. ; and (3) names of trees, as poir-ier, pomm-ier, peupl-ier, laur-ier, figu-ier, fee. Atus, Fr. e\ ata, Fr. ee. Aim-e, am-atus; avou-e, ad- 1 Learned form, -an : pl-an, pi-anus ; *veter-an, veter-anus ; &c. As to such words as courtis-an, &c., they come from the Italian (cortigiano, &c.), and date from the sixteenth century. 2 Learned form, -aire : scol-aire, schol-aris ; secul-aire, saecul- aris ; calc-aire, calcarium. Ch.2. Sect. 2. NOMINAL SUFFIXES. 185 voc-atus ; duch-e, duc-atus ; evech-e, episcop-atus ; che- vauch-ee, caballic-ata ; aim-fa, am-ata, &c. Certain derivatives in -ade, as estrap-ade, cavalc-ade, estr- ade, estac-ade, &c., come from the Italian. The French form would naturally have been -e'e, as is seen in cavalcade and chevauch-e'e ; es trade and estr-fa, strata; escapade (It. scappata), and echappee 1 . Aster, Fr. dtre. This suffix, which gives to the root the further sense of depreciation, has produced numerous French derivatives unknown to the Latin, as bell-dire, douce-dtre, gentildtre, opini-dtre, mar-dire, par-dtre, &c. Acem, Fr. at. Vr-ai, veracem ; ni-ais } nidacem, &c. The learned form is -ace : ten-ace, rap-ace, viv-ace, &c. Ela, Fr. elle. Chand-elle, cand-ela ; quer-elle, quer-ela ; tut-elle, tut-ela, &c. Elis, Fr. el, al. Cru-el, crud-elis ; fe'-al, fid-elis. Ellus, Fr. el, eau. Jum-eau, gem-ellus; b-eau, b-ellus, &c. Ensis, Fr. ois, ais, is. Such Latin derivatives as forensis, hortensis, nemorensis, have given no words to the French, which has used this termination only for words of modern formation, such as court-ois, bourg-eois, mat-ois, harn-ois, marqu-is, &c. ; or for proper names, as Orle'an-ais, Aurelian- ensis, Carthagin-ois , Carthagini-ensis, &c. Ecem, from ex, Fr. is. Breb-is, verv-ecem. Icem, Fr. is, ix, isse. Perdr-ix, perd-icem; ge'n-isse, jun-icem. Estus, Fr. ete. Honn-ete, hon-estus, &c. Ista, Fr. iste. A suffix very common in French : drogu- iste, e'be'n-iste, &c. Erna, Fr. erne. Cit-erne, cist-erna; lant-erne, lant-erna; lav-erne, tab-erna. 1 Learned form, -at : avoc-at, avoc-atus ; comul-at, consul- atus; episcop-at, &c. I 86 DERIVED WORDS. Bk. III. Etum, Fr. ay, aie. Derivatives with this termination in Latin indicated a place, or district, planted with trees. Though masc. in Latin, they became fern, in French : aun- aie, aln-6tum; orm-aie, ulm-etum ; sauss-aie, salic-etum. Hence such proper names as Chaten-ay, Casten-6tum ; Ronvr-ay, Robor-6tum ; Auln-ay, ALn-tum, &c. French derivatives are chen-aie (ch/ne); houss-aie (houx]\ chataigner- aie (chdtaignier) ; roser-aie (rosier], &c. His, Fr. il. Puer-il, gent-il, &c. The suffix -His is joined only to nouns and adverbs ; ilis only to verbs. Ignus, Fr. in, ain. Be'n-in, ben-ignus ; mal-in, mal-ignus ; de'd-ain, disd-ignum, &c. Inus, Fr. in. Dev-in, div-inus ; peler-in, peregr-inus ; vois-in, vic-inus, &c. French derivatives are mut-in, bad-in, cristall-in. Issa, Fr. esse. Abb-esse, abbat-issa; prophet-esse, prophet- issa ; venger-esse, traitr-esse, &c. Ivus, Fr. if. Chet-if, capt-ivus ; na-if, nat-ivus. French derivatives are many, pens-if, hdi-if, craint-if, ois-if, &c. Lentus, Fr. lent, lant. Vio-lent, vio-lentus ; sang-lant, &c. Mentum, Fr. ment. V$te-ment> vesti-m6ntum ; fro-ment, fru-mentum, &c. French derivatives : menage-ment, change- ment, &c. Orem, Fr. eur. Chant-eur, cant-orem ; sauv-eur, salvat- orem ; su-eur, sud-orem ; past-eur, past-orem ; pech-eur, peccat-orem, &c. Osus, Fr. eux. Epin-eux, spin-osus; pierr-eux, petr- osus ; envi-eux, invidi-osus, &c. French derivatives, heur- eux (O. Fr. heur, see p.. 175), hid-eux, hont-eux, &c. Onem, Fr. on. Charb-on, carb-onem ; pa-on, pav-6nem ; larr-on, latr-onem, &c. lonem, Fr. on. Soupf-on, suspic-ionem ; pige-on, pipi- onem ; poiss-on, L. Lat. pisc-ionem ; moiss-on, messi-onem ; mais-on, mans-ionem, &c. Ch. 2. Sect. 2. NOMINAL SUFFIXES. 187 Tionem, Fr. son. Rat-son, ra-tionem; poi-son, po-tio- nem ; venai-son, vena-tionem ; hat-son, liga-tionem ; sai- son, sa-tionem ; fa-(on, fac-tionem ; le-<;6n, lec-tionem, &c. The form -tion is of learned origin, as in the words ra-tion, po-tion, liga-tion,fac-tion, &c. Tatem > Fr. //. Ci-te, ci-tatem 1 ; sure-te, securi-tatem ; pauvre-te, pauper-tatem; &c. French derivatives: nouveau-te, opiniatre-te, &c. Icus, Fr. i ; ica, Fr. ic. Am-i, am-icus; ennem-i, inim- icus ; fourm-i, form-ica ; ort-ie, urt-ica ; vess-ie, vess-ica ; m-ie, m-ica ; p-ie, p-ica. The learned form is ique : ant-ique, pud-ique, &c. Uca, Fr. ue. Verr-ue, verr-uca ; lait-ue, lact-uca ; charr-ue, carr-uca ; fet-u, fest-uca. Orius, Fr. oir. Dort-oir, dormit-orium ; press-oir, press- orium ; dol-oire, dolat-orium, &c. French derivatives : parl- oir, abatt-oir, bruniss-oir, mdch-oire, balanf-oire. Undus, Fr. ond. Rond, O. Fr. roond, rot-undus. Unus, Fr. un. Je-un, O. Fr. jeiln, jej-unus; Verd-un, Virod-unum. TJra, Fr. ure. Mes-ure^ mens-ura; peint-ure^ pict-ura. French derivatives : froid-ure, verd-ure, &c. Urnus, Fr. our. F-our, f-urnus; j-our, di-urnus; aub- our, alb-urnum, &c. Utus, Fr. u. Corn-u, corn-utus ; chen-u, can-utus. French derivatives in abundance : barb-u,jouffl-n, ventr-u, membr-u, chevel-u, &c. 2. Suffixes which are Atonic in Latin. 'All these suffixes disappear in the French, and are consequently useless for the purpose of producing new 1 Common Latin for civi-tatem. l88 DERIVED WORDS. Bk. III. derivatives ; they have however recovered their place from the time that men utterly lost sight of the genius of the language, and became ignorant of the rule of accent 1 / Thus people began to use such words as portique, fragile, rigide, instead Q{ porche,frele, roide, from porticus, fragilis, rigidus. In considering these Latin atonic suffixes we are bound strictly to reject every word that has been introduced into the French language since the period of its natural for- mation. Eus, ins, Fr. ge, che. Etran-ge, extran-eus ; lan-ge, laii-eus ; delu-ge, diluv-ium ; lin-ge, lin-eus ; pro-che, prop- ius ; sa-ge, sap-ius ; sin-ge, sim-ius ; or-ge, hord-eum ; rou-ge, rub-ens ; au-ge, alv-ea ; son-ge, somn-ium ; Lie-ge, Leod-ium ; Maubeu-ge, Malbod-ium ; cier-ge, c6r-eus 2 . For the change of eus, ius into ge, che, see above, p. 66. Ea, Fr. ge, gne. Ca-ge, cav-ea ; gran-ge, gran-ea ; vi-gne, vin-ea ; li-gne, lin-ea ; tei-gne, tin-ea. For the change of ea into ge, see above, p. 66. la, Fr. ge, che, ce ; or it disappears altogether. Vendan-ge, vindem-ia ; angois-se, angust-ia ; cigo-gne, cicon-ia ; ti-ge^ tib-ia ; se-che, s6p-ia ; sau-ge, salv-ia ; env-ie, invid-ia ; grd-ce, grat-ia ; histoi-re, histor-ia ; Bourgo-gne, Bur- gund-ia ; France, Pranc-ia ; Gre-ce, Gra^c-ia ; Breta-gne, Britann-ia 3 . For the change of ia into ge, see above, p. 65. It-ia, Fr. esse. Just-esse, just-itia ; moll-esse, moll-itia; par-esse, pigr-itia ; trist-esse, trist-itia. French derivatives : ivr-esse, polit-esse, tendr-e'sse. 1 G. Paris, Accent latin, p. 92. 2 Learned form e, as ign-e, ign-eus. 3 Learned form ie, as chim-ie,philosoph-ie,symphon-ie, Austral-ie. But we must not confound this termination with the proper French derivatives in />, as felon-ie (felon), tromper-ie (tromper), &c., which are popular and very numerous. Ch. 2. Sect. 2. NOMINAL SUFFIXES. 189 Icem (from ex, ix, represented in French only by ce, se, ge) : her-se, h6rp-icem ; pu-ce, pul-icem ; ju-ge, jud-icem ; pou-ce, poll-icem ; pon-ce, pum-icem ; ecor-ce, cort-icem 1 . Icus, a, um, Fr. che, ge. Por-cke, port-icus ; man-che, man-ica ; ser-ge, s&r-ica ; diman-che, domin-ica ; Sainton-ge, Santon-ica ; for-ge (O. Fr. faur-ge), fabr-ica (see p. 76); per-che, prt-ica ; pie-ge, ped-ica 2 . Aticus is a suffix formed with Icus, Fr. age. Voy-age (O. Fr. viat-ge), vi-aticum ; from-age, form-aticum ; vol-age, vol-aticum; ombr-age, umbr-aticum; ram-age^ ram-aticum ; mess-age, miss-aticum ; sauv-age, silv-aticus 3 . Hence come French derivatives : mesur-age, labour -age, alli-age, arros-age, &c. It has been said that these words come from a Low Latin suffix in -agium (as message from mess-agium, horn-age from hom-agium). But though mess- agium certainly exists, it is far from being the parent of the Fr. message; on the contrary, it is nothing but the Fr. message latinised by the clergy, at a time when no one knew either the origin of the word (missaticum) or the nature of the suffix which formed it. Idus disappears in French. Pale, pall-idus ; net, nit- id us ; chaud, cal-idus (Low Lat. cal-dus) ; tilde, tep-idus ; roide, rig-idus ; sade, sap-idus ; whence maussade, male sap-idus 4 . See p. 180. 1 Learned form ice : cal-ice, cal-icem. 2 Learned form ique : port-ique, port-icus ; fabr-ique, fabr-ica ; when one rises in the morning ! But the common course was to 2 196 APPENDIX. derive from one another two words of totally different forms, and to fill up the gulf between them with fictitious inter- mediate words. Thus Manage derived the word rat from the Latin mus : " One must have first said mus, then mu- ratus, then ratus, and lastly rat /" Nay, they even went so far as to suppose that an object could take its name from a quality the very contrary of that which it possessed, on the ground that " affirmation suggests negation ;" and thus we have the famous lucus a non lucendo, on the pretext that " once in a sacred wood one has no more light 1 /' ' Finally, the illusions of etymologists became proverbial, and this branch of historical knowledge was thoroughly dis- credited. How then did a science, now established and important, emerge frctai such a mass of learned bewilder- ment ? The clue is '-" the discovery and application of the comparative method, the true method of natural sciences. 1 Comparison is the chief instrument of scientific enquiry. For science is composed of generalisations : to know is to form a group, to establisTi a law; consequently, to pick out whatever is general from among particular facts. But if we would compel facts to deliver up to us their inner meaning, we must place them side by side, explain them by one another, in a word, compare them/ ( Every one is acquainted with the discoveries of com- parative anatomy. We know how the study of the structure of animals, and the comparison of their organs (whose infinite modifications form the differentiae of class, order, and genus), have revealed, if we may so speak, the plan of nature, and have given us a firm foundation for our classifications V The same is true of the science of language : here, doubt- 1 M. Reville, Les ancetres des europeens. 2 E. Scherer, Etudes tfhistoire et de critique. APPENDIX. 197 less, as elsewhere, comparison is as old as observation ; but there are two kinds of comparison, or rather, there are two stages of comparison, through which in due succession every mind must pass. Of these the former stage is precipitate and superficial. It governed all natural sciences up to the end of the seven- teenth century, and was content to compare and class together beings or words according to their superficial like- ness. Thus the ancients put the whale and the dolphin in the class of fishes, because of their external form, their habits, and their habitat in the sea ; similarly the old etym- ologists derived the word paresse from the Greek Trapem, because, among all the languages they knew, this Greek form was most like the French word ; and so they con- cluded, without further proof, that it came from the Greek an easy way indeed of satisfying oneself! To these arbitrary processes has succeeded in our day the stage of well-considered and methodical comparison; a strict and scientific comparison, which does not stop at external resemblance or difference, but dissects objects in order to penetrate even to their very essence, and their deepest analogies. The anatomist studies the internal structure of the whale, and instantly sees that the conformation of its organs ex- cludes it from the class of fishes, and places it among the mammalia. And similarly, instead of only studying his word from the outside, the philologer dissects it, reduces it to its elements, i. e. its letters, observes their origin and the manner of their transformation. By a strict application of this new method, by letting facts lead instead of trying to lead them, modern philology has been enabled to prove that language is developed according to constant laws, and follows necessary and invariable rules in its transformations. 198 APPENDIX. We have set forth in this book the chief characteristics of this natural history of language. Especially with a view to etymology they furnish the student with unexpected as- sistance, and are, in his eyes, a precious instrument, a powerful microscope with which to observe the most delicate phenomena. Its instruments are these : Phonetics, History, Com- parison. I. PHONETICS. In the earlier part of this book 1 we divided words into their elements, that is, into their letters, and saw that the transit of the letters from Latin to French followed a regular course, each Latin letter passing into French according to fixed rules : thus e long always becomes oi in French, as me, moi ; regem, rot ; legem, lot ; te, tot ; se, sot ; tela, toile ; velum, voile, &c. The bearings of this discovery are apparent at once : for if we will but observe these laws of change as they affect each letter in succession, we shall find them a clue to guide our researches, and to keep us from straying into wrong paths ; if the etymology does not satisfy these condi- tions, it is naught. Thus then the possession in detail of the transforma- tions of the Latin letters into French 2 is the first necessity for those who would occupy themselves with etymology. If any one finds this preparatory study too minute or un- interesting, our reply is that anatomy observes and describes muscles, nerves, and vessels most minutely in detail; and draws out a catalogue of facts which may well seem dry and tiresome; but yet just as this comparative anatomy is the basis of all physiology, so is this exact knowledge of 1 See above, pp. 45-86. 2 See above, Bk. I. pp. 66-76. APPENDIX. 199 orthography the beginning of all true etymology: nothing else can give it the true character of a compact and rigorous science. We may state this new principle thus : We must reject every etymology ', which, when the rules of permutation have been laid down by orthography, does not account for letters retained, changed, or lost. By the light of this principle let us take as an example the word laitue, and seek for its origin. We have seen above, under ' Phonetics/ p. 50, that the French combination it corresponds to the Latin ct ; as fa-it from fa-ct-us ; la-it, la-ct-em ; tra-it, tra-ct-us ; fru-it, fru-ct-us ; redu-it, re- du-ct-us. Therefore the first part of the word laitue (laif) must answer to a Latin word lact-. What is the origin of the suffix -uel Now we have seen (p. 187) that this suffix is derived from the Latin suffix -uca ; as verr-ue from verr-uca ; charr-ue, carr-uca, &c. Thus we arrive at the form lact-uca, which is in fact the Latin word which expresses the idea involved in laitue, the lettuce. This enquiry into etymology is clearly analogous to the operations of chemical analysis. The chemist puts a sub- stance into his crucible and reduces it to its elements, and finds again the equivalent weight : so here too the elements are the letters, and our analysis, i. e. our etymology, is liable to suspicion so long as the elements have not been dis- covered again after the process 1 . We may sum up by saying that etymological research is subject to two rules : (i) No etymology is admissible which refuses to account for all the letters of the word it proposes to explain, without a single exception ; and (2) Every etymology which assumes a change of letters ought i M.Littre. SCO APPENDIX. to have in its favour at least one example of a change quite identical with that which it assumes ; otherwise, if no such testimony can be cited, the attempt is valueless. II. HISTORY. Every Latin word on its way into Modern French has gone through two changes ; it has passed from Latin into Old French, and thence into the French of to-day : thus festa became first feste, and then in course of time feste became fete. In finding the origin of a French word we should follow a wrong track if we speculated on it in its present state, leaping from Modern French to Latin ; we must first enquire whether there are any intermediate forms in Old French which may explain the transition and mark the path followed by the Latin on its way to the present French. And besides, these intermediate forms, by bringing us nearer to the starting-point, help us to see that point more clearly, and often guide us to the word we are seeking without any further researches. An example will best illustrate the difference in this respect between the old and the new methods of etymology. The old etymologists were much divided as to the origin of the word dme : some only thought of the sense, and there- fore declared that it came from the Latin anima, though they could not explain how the transformation had taken place; others, finding the contraction of anima into dme far too violent a change, held that it was derived from the Gothic ahma, 'breath/ The case would be still ' sub judice/ had not modern philology intervened to solve the problem in the natural way. Substituting the observation of facts for the play of imagination, modern philologers have seen that it would be absurd to talk for ever over a word in its modern form, without taking any heed to the changes it has APPENDIX. 201 undergone since the origin of the language; and so they constructed the history of the word by the study of ancient texts, and shewed that in the thirteenth century dme was written anme, in the eleventh aneme, in the tenth anime, whence we pass directly to the Latin anima. If we would obtain a secure foothold, we must move step by step over the intermediate forms; so as to be able to study in its gradations the deformation of the Latin word. But even here we must distinguish between two kinds of intermediate forms, those of the old and those of the new philological school. The former assumed at a venture some improbable word as the origin of the word under considera- tion ; and, in order to join the two ends, imagined fictitious intermediate forms to suit their purpose. Thus, Manage pretended that he had found the origin of the French haricot in faba; and to fill up the gulf between these words he added, ' They must have said first faba, then fabaricus, then fabaricotus, aricotus, and finally haricot.' Such lucubra- tions are like a bad dream ; they justify the opinion of those who have laughed at etymology, and deserve the Chevalier d'Aceilly's epigram : ( Alfana } vient d'equus sans doute, Mais il faut convenir aussi Qu'a venir de la jusqu'ici, II a bien change sur la route;' for the learned made a scientific toy of what they ought to have treated as a science. The intermediate forms, diligently sought out by modern etymology, are very different ; science does not ask what men ' must have said/ but what men did say. There are no more fanciful forms invented, as the case required 1 The name given by Ariosto to Gradasso's mare. Menage proposed to derive it from equus. 202 APPENDIX. them. French philology now limits itself to a diligent passage through old texts running back to the tenth century : then noting the birth of words and the first date of their appear- ance, it marks the changes in them century after century. Exact observation, which leaves no room for conjecture or invention, is a preliminary but essential part of all etymo- logical enquiry : before analysing a French word in its actual form, we must seek to obtain as many examples as we can of the word as it appeared in Old French. M. Littr^ has followed this course in his admirable Diction- naire historique de la langue frangaise: instead of inventing a series of arbitrary intermediate forms, he collects under each word a series of successive examples drawn from texts, running back to the very beginnings of the French language in the eighth century. These posts once firmly fixed, he goes on to build on them an etymology, which does not arise from the word in its present shape, but from it as it existed at the birthplace of the language. An attentive investigation into intermediate forms is the best help, after phonetics, that philology can have. III. COMPAEISON. While popular Latin was giving birth to the French lan- guage, it also created, as we have seen (p. 10), four sister idioms to it, formed also with astonishing regularity the per- mutation of the Latin letters into Italian, Spanish, Portuguese, that is to say, what are called the Romance languages, being as regular and unchanging as into French. Consequently, we must compare the French forms with those current in the other Romance languages ; this will be the touchstone by which to try and prove all proposed hypotheses. We have just seen (p. 199) that laitue answers letter for letter to the Latin lactuca. If this etymology is APPENDIX. m 203 correct, it will follow that the Italian lattuga, and the Spanish lechuga, whose sense is the same, spring also from the same Latin word. And this will shew us that the Italian // and the Spanish ch come from the Latin ct : thus ITALIAN no-tt-e, no-et-em ; la-H-e, la-ct-em ; o-tt-o, o-ct-o ; bisco-tt-o, bisco- ct-us ; tra-tt-o, tra-ct-us, &c. whence la-tt-uga, from la-ct- uca; SPANISH no-ch-e, no-ct-em; le-ch-e, la-ct-em; o-ch-o, o-ct-o ; bisco-ch-o, bisco-ct-us ; tre-ch-o, tra-ct-um, &c. whence le-ch-uga from la-ct-uca. Thus we see how the parallel relations of the Romance with the French languages strengthen our previous observations, and serve as verifying tests of our hypotheses. These parallels have another use ; they often shew us the road we ought to follow ; but time and space fail us, and we cannot stay to insist on the advantages that etymology can derive from careful comparison; such details would be in their right place in a ' Manual of French Ety- mology/ but are beside the mark in this short outline of the new philological method, in which we are trying to describe the great revolution which has transferred etymology from the realms of fancy to the solid ground of a historical science. CONCLUSION. By shewing that words grow and have a history, and that, like plants and animals, they pass through regular transfor- mations above all by shewing that here, as elsewhere, law reigns, and that we can lay down with certainty the rules of derivation from one language to another, modern philolo- gers have set comparative etymology on durable founda- tions, and have made a science of what seemed condemned to be confined to the regions of imagination and individual caprice. The older system of etymology tried to explain the origin of words a priori, following their apparent resemblance or 204 APPENDIX. difference ; modern etymology applies the method of the natural sciences, and holds that words ought to explain themselves, and that, instead of inventing systems, we ought to observe facts, by the help of these instruments : Pho- netics, which give us the rules of transformation from one language to another rules which we must follow implicitly, or pay the penalty of losing our way; History of words, which passes on by certain and definite stages to the original word we are looking for, or, at any rate, brings us nearly up to it ; and lastly, Comparison, which certifies and con- firms the results we have arrived at. To the fantastic imaginations of the learned of old days was due the discredit into which etymology had formerly fallen ; but by the strict application of this method and these principles, comparative etymology has risen in our time to the dignity of a science. INDEX. A. A, the French, 48 ; the Latin, 67. Accent, continuance of Latin, 33 ; grammatical, 33, 85 ; on vowels, 67 ; tonic, 84 ; oratorical, 86 ; pro- vincial, 86 ; on compounds, 172. Accusative, the Latin, retained in French, 93. Adalhard of t Corbie spoke Romance, 12. Addition of letters, 78-80. Adjectives, French, 102-108 ; used as substantives, 103; compound, 175. Adverbs, 153-163; of place, 154; time, 155; manner, 158; intensity, 158; affirmative and negative, 160. Adverbial phrases, 161. Ae, the Latin, 68. Ai, the French vowel, 52. Ailleurs, 154. Alauda, 5. A Her, 149. Alphabet, French and Latin, 46-83. Amabam, in the Langue d'Oil dia- lects, 19, 137. Amont, 155. Analytical tendencies of modern lan- guages, ii. Anglo-French aristocratic words, 4. Anomalous verbs, 148-152. Aphaeresis, 80. Apocope, 80, 82. Arabic words in French, 22, 23. Armorica, long retained Celtic speech, 4,6. Article, the French, TOO. Atonic syllables, 68. Au, the French, 53, 68. Aucun, 115. Aujourd'bui, 155. Auparavant, 158. Auxiliary Verbs, 123-129. Avaler, 155. Avantj 163. Avoir, 127-129. B. 5, French, 58; Latin, 76; omission of, 82. Bacon, Roger, on French dialects, 19. Basque tongue, i. Beaucoup, 159. Belgae, I. Benir, 150. Bercheure's translation of Livy, 39. Boire, 150. Bonheur, 175. Bordeaux, school of, 4. Boucher, originally an adjective, 103. Braire, 146. Breton language, 6, 7. Brunette Latini wrote French, 17. Burgundian French, 18. Burgus, 7. 206 INDEX. C. C, the French, 63 ; the Latin, 74 ; omitted, 82. ffl, 154- Caesar on Gaul, I. Calvin, the French of his ' Institution,' 25- Cardinal numbers, 105-107. Cases in French, reduced to two, 89- 91 ; thence to one, 93-96. Cassiodorus quoted, 3. Catharine dei Medicis her influence on the French tongue, 25. Ce, cet, celui, 113. Ceans, 155. Celtic language in Armorica, 6. Celts, i. Cervisia, 5. Ch, the French, 64. Chaloir, 147. Chanson de Roland, the, 1 7. Chaque, 115. Chez, 165. Choir, 148. Church, the, adopts the ' Romance ' tongue, 12, 13. Classicists, their influence on the French language, 26. Clore, 146. Colonisation, Roman, 2. Commines, the French of, 25. Comparison, degrees of, in French, 104 ; used to test etymologies, 202. Compounds, formation of, 172. Conjugation of French verbs, 118-152. Conjunctions, 167-169. Consonants, loss of medial, 37 ; the French, 55-66; the Latin, 70-76; transposition of, 77; addition of, 79 ; subtraction of, 8 1, 82. Convoy er, 149. Courir, 150. Curials, in Gallic cities, 7. JD, French, 6 1 ; Latin, 74; omission of, 82. Dame, 170. Dangeau, 1'Abbe, attempted a phonetic spelling, 29. Dans, 154, 164. Declensions, French, 88-117. Defective verbs, 142-148. Demonstrative Pronouns, 113. Dentals, French, 60-62 ; Latin, 73- 74- Derivation, 181. Derriere, 164. De Brosses, his primitive language, 29. Desormais, 157. Devant, 164. Devoyer, 149. Dictionnaire de V Academic, 29. Dime, 1 08. Diminutive suffixes, 192. Domestique, originally a substantive, 103. Dont, 114. Dorenavant, 157. Du Bellay's Defense et illustration de la languefran$aise, 26. Duo, 105. Durant, 165. E, the French vowel, 49 ; the Latin, 67 ; addition of, 78. Eau, French compound vowel, 53. Ecrire, 151. Ei, French compound vowel, 52. En, how derived, IIO. Encore, 156. England gave terms of civil life, &c. to France, 3 ; learnt French after the Norman Conquest, 1 7. English words imported into France, 30. Environ, 155. Envoyer, 149. Epenthesis, 78, 79. Epithesis, 78, 80. Essere, 125. Ester, 144. Etais, not from stabam, 1 26. fitre, 124-127. INDEX. 207 Etymology a new science, 195. Eu, French compound vowel, 53. Eulalia, St., French poem on, 15. Euskarian tongue, I. F. F, the French, 59; the Latin, 76. Faillir, 144. Falloir, 147. Ferir, 145. Fois, 155. Formation of tenses, 136-141 ; of words, 171-193. Fortunatianus on Latin genders, 99. Fortunatus of Poitiers, 10. Fourvoyer, 149. Franks, the, adopt the popular Latin, 10. Frederick II of Germany loved the French language, 1 7. French language owes very little to Celtic, 5 ; came from popular Latin, 9 ; in the Strasbourg Oaths, 14, 15 ; in poem on St. Eulalia, 1 5 ; divided into Langue d'Oc and Langue d'Oil, 18; He de France dialect prevailed, 20; distinguished from Picard, 21 ; in fourteenth century became the French language, 22 ; in fifteenth century, 25 ; in sixteenth, 25 ; in seventeenth, 28; in eighteenth, 29 ; in nineteenth, 29; has learned and popular words, 32 ; laws of its for- mation, 3 2 ~3^ ; popular words re- spect Latin accent, 33, 34 ; are shorter than the learned, 34, 35 ; declensions, 89-99 genders, 96- 98 ; numbers, 98, 99 ; article, loo ; pronouns, 109-117; verbs, 118- 152; particles, 153-170; adverbs, I 53~ I 63; prepositions, 163-167; conjunctions, 167-169; interjec- tions, 169, 170. Future tense, how formed, 112-138. G. G, French, 64 ; Latin, 75 ; addition of, 79 ; omission of, 82. Gallic race, I. Gallo-Roman nobles cultivate literary Latin, 8. Genders, in French, 96-98. Gens, gender of, 98 ; number of, 99. German language enters France, 6, 7 ; forgotten in France, u ; some words retained, II. Gesir, 146. Glosses of Reichenau, 12. Grammarians, philosophical, 29. Grammatical accents, 85. Greek fashionable words in Latin, 3 ; some others, through the Latin, 27; introduced by the classicists, 27. Guere, 156, 160. Gutturals, French, 63-66 ; Latin, 74, 75- H. H, French, 66 ; Latin, 75 ; addition of, 79. Haribert the Frank, congratulated on his Latin, 10. Hebrew words in French, 22. Helas, 169. Hier, 155. History of words, its value in etym- ology, 200. Hormis, 166. Hrolf the Norman, 1 1 . Hugh Capet knew no Latin, 16. Hut, 155. Hymne, gender of, 98. I. 7, the French, 49, 50 ; the Latin, 67 ; transposition of, 77. Id, 154, Idioms, aristocratic and popular, 2. le, ten, French compound vowels, 54. Hie becomes le, 100, 101. Imperative mood, 139. Imperfect indicative, 137 : subjunctive, 208 INDEX. Inchoative verbs, 131. Indefinite pronouns, 115-117. Infinitive mood, j 39. Inflexion of French words, 87. Interjections, 169. -issime, French superlatives in, 105. Issir, 145. Italians in thirteenth century valued French tongue, 1 7 ; their influence on it in fifteenth century, 25. J. J, the French, 62 ; the Latin, 75. Jadis, 155. Jamais, 157. ye, how derived, no. K. Karl the Bald, and the Strasbourg Oaths, 14. Karl the Fat, n. Karl the Great (Charlemagne), French language under, 12. L. L, the French, 56; the Latin, 71; transposed, 77 ; added, 79 ; apocope of, 83. La, 154. Labials, French, 58-60; Latin, 75- 7 6. Language and history of France con- nected, 20. Langue d'Oc, 1 8. Langue d'Oil, 18. Latin, spoken through Gaul, 2 ; in two idioms, 2,3; popular, parent of modern languages, 4, 9 ; the literary perishes, 7 ; low, 8 ; parent of French, 9 ; not known by Hugh Capet, 16; even in monasteries it died out, 1 7 ; loses all cases but two, 22, 23 ; the alphabet in, 66- 76 ; loses the neuter gender, 99. Le, 100. Leger, St., French poem on, 16. Leti, the, 6. Lez, 166. Lie, explained, 68. 'Lingua Romana Rustica,' 12. Liquids, French, 55-58; Latin, 71- 73- Literature of early French language, .17; Littre, M., on accent, 86 ; on declen- sion, 92 ; his Dictionnaire, 202. LI, the French, 57. Loin, 154. Low Latin, 8. Ludwig the German takes oath in French, jo, 14. Lyons, School of, 4. M. M, the French, 56 ; the Latin, 72 addition of, 80. Maintenant, 155. Mais, 1 68. Malherbe resisted the classicists, 28. Marco Polo wrote in French, 17. Marseilles, i. Maturus, mur, 38. Meme, 116. Mente forms the adverbial -menf, 153' 154- Meyer, M., on the two Latin idioms, 8. Mie, 162. Mm, French, 58. Moods of French verbs, 119. Monks at last abandoned Latin, 1 7. Mourir, 150. Mouvoir, 151. Moyennant, 165. Mummolinus, St., could speak German and Romance, 12. N. N, the French, 55 ; the Latin, 72 : transposed, 77 ; addition of, 80 , apocope of, 83. INDEX. 209 Nagueres, 155. Naitre, 151. -ndre, verbs ending in, 151, Neanmoins, 168. Neuter gender in Latin, when lost, 99. Nn, French, 58. Non-obstant, 165. Norman-French, 18. Nouns . of number, 105 ; compound, 174; suffixes to, 183. Numbers in French, 98, 99. O. 0, the French vowel, 50, 51 ; the Latin, 68. Oaths of Strasbourg, 14, 15. Objective case alone retained in French, 93. CEu, French compound vowel, 53. 01, French compound vowel, 52. Old French perished by the end of the fourteenth century, 25. On, 116. Or, Latin masculines in, become French feminines, 97. Or, 155, Oratorical accent, 86. Ordinals, 107. Oresme's translation of Aristotle, 39. Ou, French compound vowel, 53. Out, 1 60. Ou'ir, 145. P. P, the French, 58 ; the Latin, 75. Paitre, 147. Paris, M. G., on le, 101. Participles, 140 ; many became nouns, 141 ; also prepositions, 165. Particles, 153-170 ; words formed with, 176; qualitative, in compo- sition, 1 80 ; quantitative, 181 ; negative, 181. ,Pas, 161. "Patois distinct from dialect, 20, 21. Perfect indicative, 138. Permutations of letters, Latin to French, 46-76. Personal pronouns, 109-111. Persons of French verbs, 121. Personne, 161. Peu, 159. Pb, the French, 59. Phonetics, or letters of alphabet, 45- 86, 198-200. Picard French, 18, 21. Pliny, his works known through Gaul, t Point, 161. Popular Latin, 3, 9, 12. Position, relations of words expressed by, 23. Possessive pronouns, 113115. Prepositions, instead of inflections in common Latin, II ; the French, 163-167 ; words formed with, 177. Present indicative, 136; subjunctive, 139- Prime, 107. Pronouns, 109-117. Prosody, 84-86. Prosthesis, 78. Provincial accent, 86. Puts, 164. Q. Q, the French, 63 ; the Latin, 75. Qualifying adjectives, 102. Quant, 117. Quart, 1 08. Quelque, quelconque, quiconque, 117. Querir, 146. Qui, 114. Quint, 1 08. B. R, the French, 57 ; the Latin, 72 ; transposed, 77 ; addition of, 80. Relative pronouns, 1 14. Renaissance, the, dazzled the French 25- Rez, 1 66. 2IO INDEX. Rheims, Council of, 14. Rien, 162. Romans enter Gaul, I. Romantic school of literature in France, 29. Ronsard carries out Du Bellay's classi- cal reform, 26. Rr, the French, 58. S. S, the French, 61 ; the Latin, 74 ; addition of, 80. Sanglier, originally an adjective, 103. Savoir, 151. Seoir, 148. Sermo nobilis rusticus, 3. Setme, 1 08. Si, 158. Soudain, 138. Soudre, 147. Souloir, 147. Sourdre, 147. Souvent, 157. Spain, her influence on the French language, 28. Strabo, 4. Strasbourg, Oaths of, 10, n, 14; Council of, 14. Strong verbs, wh'tt, 122. Subjective (or nominative) case re- tained in a few words, 96. Substantives, French, declension of, 88-99. Subtraction of letters, 80. Subjunctive mood, 139. Suffixes, 181. Superlative, the French, 105. Syncope, 80, 81. T. T, the French, 60; the Latin, 73; addition of, 79 > omission of, 82 ; apocope of, 82. Tandis, 157. Tant, 159. Tel, 117. Tenses of French verbs, 119; forma- tion of, 136-141. Tiers, 107. Tisser, 144. Tonic syllable in Latin, 33 ; accent 84. Tot, 156. Tours, Council of, 13. Tout, 1 1 7. Transposition of letters, 77. Trop, 1 60. TJ. U, the French, 51 ; the Latin, 68. Ui, the French, 53. -uire, words ending in, 151. Un, 117. Unaccented or atonic vowels, 68, 69. Universal language, theory as to, 29. University of Paris, thronged, 17. Unus, 105, V. V, the French, 59; the Latin, 76; omitted, 82. Valoir, 151. Vegetius, De re militari, 7. Verb, the French, 118-152 ; auxiliary, 123-129; first conjugation, 129,* second conjugation, 130; third con- jugation, 132; fourth conjugation, 133 ; voices, 1 18 ; moods, 119; tenses, 119; persons, 1 21 ; strong and weak, 122, 142; inchoative, 131 ; irregular, 142 ; defective, 143 ; anomalous, 148; compound, 176; suffixes to, 191. Vers, 164, Villon's Old French, how detected, 93. Vingt and its compounds, 106. Vis-a-vis, 1 66. Vivre, 150. Voices of French verbs, 118. Void, voila, 166. Voir, 150, INDEX. 211 Vowels, suppression of short Latin, 35; simple French, 46-51; com- pound French, 51-55 ; Latin, ac- cented, 67 ; atonic, 68; transposed, 77; added, 78; omitted, 80-82. W. War terms introduced by the Franks, II. Weak verbs, what, 122. Words, good and bad, introduced in the present century into the French language, 29, 30 ; two forms of, popular and learned, 32 ; influx of learned, 39 ; formed from phrase? 176; with particles, 176. X. , the French, 62 ; the Latin, 74. Y. T, the pronoun, how derived, in. Z. Z, the French, 62 ; the Latin, 74. Zythum, 5. THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW AN INITIAL FINE OF 25 CENTS WILL BE ASSESSED FOR FAILURE TO RETURN THIS BOOK ON THE DATE DUE. THE PENALTY WILL INCREASE TO SO CENTS ON THE FOURTH DAY AND TO $I.OO ON THE SEVENTH DAY OVERDUE. or T^ 4rrfi /- LD 21-100m-7,'40 (6936s) YB 01395 861075 THE UNIVERSITY OF CALIFORNIA LIBRARY