Areal and dialectal divisionsThe of Old French in contemporary terms corresponded to the northern parts of the (including and , which in the 12th century were ruled by the ), and the . The Norman dialect was also spread to and , and during the , Old French was also spoken in the , and in the and the in the . As part of the emerging dialect continuum, the were contrasted with the (the emerging group, at the time also called Provençal), adjacent to the Old French area in the south-west, and with the group to the south-east. The group developed in Upper Burgundy, sharing features with both French and Provençal; it may have begun to diverge from the as early as the 9th century and is attested as a distinct Gallo-Romance variety by the 12th century. Dialects or variants of Old French include: * Burgundian in , then an independent whose capital was at ; * Picard of and , with , and as some of the more prominent cities. It was said that the Picard language began at the east door of , so far-reaching was its influence. It would also spread northwards in the area of that had a strong presence of and ; * , in , whose principal cities were and . The of brought many Norman-speaking aristocrats into the British Isles. Most of the older Norman (sometimes called "French") words in reflects its influence, which became a conduit for the introduction into the Anglo-Norman realm, as did Anglo-Norman control of Anjou and Gascony and other continental possessions. was a language that reflected a shared culture on both sides of the . Ultimately, the language declined and fell, becoming , a jargon spoken by lawyers that was used in until the reign of ; however, the , still survives in Normandy and the , as a regional language; * Wallon, around , now in , ; * Gallo of the ; * of the . Some modern languages are derived from Old French dialects other than Classical French, which is based on the dialect. They include , , Bourguignon-Morvandiau, , Franc-Comtois, Gallo, , , Picard, Poitevin, and .
Evolution and separation from Vulgar LatinBeginning with ' time (254–184 ), one can see phonological changes between and what is called , the common spoken language of the . Vulgar Latin differed from Classical Latin in and as well as exhibiting lexical differences; however, they were mutually intelligible until the 7th century when Classical Latin 'died' as a daily spoken language, and had to be learned as a second language (though it was long thought of as the formal version of the spoken language). Vulgar Latin was the ancestor of the , including Old French. By the late 8th century, when the began, native speakers of Romance idioms continued to use Romance rules while speaking and reading Latin. When the most prominent scholar of Western Europe at the time, British deacon , was tasked by with improving the standards of Latin writing in France, being not a native Romance speaker himself, he prescribed a pronunciation based on a fairly literal interpretation of Latin spelling. For example, in a radical break from the traditional system, a word such as ⟨viridiarium⟩ 'orchard' now had to be read aloud precisely as it was spelled rather than */verdʒjær/ (later spelled as ). Such a radical change had the effect of rendering Latin completely unintelligible to the general romance-speaking public, which prompted officials a few years later, at the , to instruct priests to read sermons aloud in the old way, in or 'plain Roman speech'. As there was now no unambiguous way to indicate whether a given text was to be read aloud as Latin or Romance, various attempts were made in France to devise a new orthography for the latter; among the earliest examples are parts of the and the (see below).
GaulishSome words influenced Vulgar Latin and, through this, other Romance languages. For example, classical Latin was uniformly replaced in Vulgar Latin by 'nag, work horse', derived from Gaulish (cf. , ), yielding , Occitan (), Catalan , Spanish , Portuguese , Italian , Romanian , and, by extension, English '' '' and '' '' (both via different forms of French">ldFrench). An estimated 200 words of Gaulish etymology survive in modern French, for example , 'oak tree', and , 'plough'. Within historical phonology and studies of language contact, various phonological changes have been posited as caused by a Gaulish substrate, although there is some debate. One of these is considered certain, because this fact is clearly attested in the Gaulish-language epigraphy on the pottery found at la Graufesenque ( 1st century). There, the Greek word (written in Latin) appears as . The consonant clusters /ps/ and /pt/ shifted to /xs/ and /xt/, e.g. > ''*kaxsa'' > ''caisse'' ( Italian ) or ''captīvus'' > ''*kaxtivus'' > (mod. ''chétif''; cf. Irish ''cacht'' 'servant'; ≠ Italian , Portuguese , Spanish ). This phonetic evolution is common in its later stages with the shift of the Latin cluster /kt/ in Old French ( > ''fait'', ≠ Italian , Portuguese , Spanish ; or ''lactem''* > ''lait'', ≠ Italian , Portuguese , Spanish ). This means that both /pt/ and /kt/ must have first merged into /kt/ in the history of Old French, after which this /kt/ shifted to /xt/. In parallel, /ps/ and /ks/ merged into /ks/ before shifting to /xs/, apparently under Gaulish influence. The Celtic languages, Celtic Gaulish language is thought to have survived into the 6th century in France, despite considerable cultural Romanization. Coexisting with Latin, Gaulish helped shape the dialects that developed into French, with effects including loanwords and calques (including , the word for "yes"), sound changes shaped by Gaulish influence, and influences in conjugation and word order. Recent computational studies suggest that early gender shifts may have been motivated by the gender of the corresponding word in Gaulish.
FrankishThe pronunciation, vocabulary, and syntax of the Vulgar Latin spoken in Roman Gaul in Late Antiquity was modified by the Old Frankish language, spoken by the Franks who settled in Gaul from the 5th century and conquered the future Old French-speaking area by the 530s. The name ''français'' itself is derived from the name the Franks. The Old Frankish language had a definitive influence on the development of Old French, which partly explains why the earliest attested Old French documents are older than the earliest attestations in other Romance languages (e.g. Strasbourg Oaths, ). It is the result of an earlier gap created between Classical Latin and its evolved forms, which slowly reduced and eventually severed the intercomprehensibility between the two. The Old Dutch, Old Low Franconian influence is also believed to be responsible for the differences between the ''langue d'oïl'' and the Occitan language, ''langue d'oc'' (Occitan), being that various parts of Northern France remained bilingual between Latin and Germanic for some time, and these areas correspond precisely to where the first documents in Old French were written. This Germanic language shaped the popular Latin spoken here and gave it a very distinctive identity compared to the other future Romance languages. The very first noticeable influence is the substitution of the Latin melodic accent by a Germanic stress and its result was diphthongization, differentiation between long and short vowels, the fall of the unaccented syllable and of the final vowels: * ''decimus'', ''-a'' 'tenth' > ''disme'' > French ''dîme'' 'tithe' (> English ''dime''; Italian ''decimo'', Spanish ''diezmo'') * ''dignitate'' > ''deintié'' (> English ''dainty''; Italian ''dignità'', Romanian ''demnitate'') * ''catena'' > ''chaeine'' (> English ''chain''; Italian ''catena'', Cast./Occitan ''cadena'', Portuguese ''cadeia'') Additionally, two phonemes that had long since died out in Vulgar Latin were reintroduced: and (> ''g(u)-'', ONF ''w-'' cf. Picard ''w-''): * ''altu'' > ''halt'' 'high' (influenced by Old Low Frankish  ''*hōh'' ; ≠ Italian, Portuguese ''alto'', Catalan ''alt'', Old Occitan ''aut'') * ''vespa'' > French ''guêpe'', Picard ''wèpe'', Wallon ''wèsse'', all 'wasp' (influenced by ''*wapsa''; ≠ Occitan ''vèspa'', Italian ''vespa'', Spanish ''avispa'') * ''viscus'' > French ''gui'' 'mistletoe' (influenced by ''*wīhsila'' 'morello' with analogous fruits, when they are not ripe; ≠ Occitan ''vesc'', Italian ''vischio'') * ''vulpiculu'' 'fox kit' (from L ''vulpes'' 'fox') > ''golpilz'', Picard ''woupil'' 'fox' (influenced by ''*wulf'' 'wolf'; ≠ Occitan ''volpìlh'', Old Italian ''volpiglio'', Spanish ''vulpeja'' 'vixen') In contrast, the Italian, Portuguese and Spanish words of Germanic origin borrowed from French or directly from Germanic retain ~ , e.g. It, Sp. ''guerra'' 'war', alongside in French ''guerre''). These examples show a clear consequence of bilingualism, that sometimes even changed the first syllable of the Latin words. One example of a Latin word influencing an loan is ''framboise'' 'raspberry', from ''frambeise'', from ''*brāmbesi'' 'blackberry' (cf. Dutch ''braambes'', ''braambezie''; akin to German ''Brombeere'', English dial. ''bramberry'') blended with LL ''fraga'' or ''fraie'' 'strawberry', which explains the replacement > and in turn the final ''-se'' of ''framboise'' added to ''fraie'' to make ''freise'', modern ''fraise'' (≠ Wallon ''frève'', Occitan ''fraga'', Romanian ''fragă'', Italian ''fragola'', ''fravola'' 'strawberry'). Mildred Pope (1934) estimated that perhaps still 15% of the vocabulary of modern French derives from Germanic sources (while the proportion was larger in Old French, because the Middle-French language borrowed heavily from Latin and Italian).
Earliest written Old FrenchThe earliest documents said to be written in the Gallo-Romance that prefigures French – after the Reichenau Glosses, Reichenau and Kassel glosses (8th and 9th centuries) – are the (treaties and charters into which King Charles the Bald entered in 842): The second-oldest document in Old French is the Eulalia sequence, which is important for linguistic reconstruction of Old French pronunciation due to its consistent spelling. The royal House of Capet, founded by Hugh Capet in 987, inaugurated the development of northern French culture in and around , which slowly but firmly asserted its ascendency over the more southerly areas of Aquitaine and Tolosa (Toulouse); however, the Capetian dynasty, Capetians' ''langue d'oïl'', the forerunner of modern standard French, did not begin to become the common speech of all of France until after the French Revolution.
Transition to Middle FrenchIn the Late Middle Ages, the Old French dialects diverged into a number of distinct ''langues d'oïl'', among which proper was the dialect of the region. During the early modern France, Early Modern period, French was established as the official language of the Kingdom of France throughout the realm, including the ''langue d'oc''-speaking territories in the south. It was only in the 17th to 18th centuries – with the development especially of popular literature of the ''Bibliothèque bleue'' – that a standardized Classical French spread throughout France alongside the regional dialects.
LiteratureThe material and cultural conditions in territorial evolution of France, France and associated territories around the year 1100 triggered what Charles Homer Haskins termed the "Renaissance of the 12th century", resulting in a profusion of creative works in a variety of genres. Old French gave way to in the mid-14th century, paving the way for early French Renaissance literature of the 15th century. The earliest extant French literary texts date from the ninth century, but very few texts before the 11th century have survived. The first literary works written in Old French were hagiography, saints' lives. The ''Sequence of Saint Eulalia, Canticle of Saint Eulalie'', written in the second half of the 9th century, is generally accepted as the first such text. At the beginning of the 13th century, Jean Bodel, in his ''Chanson de Saisnes'', divided medieval French narrative literature into three subject areas: the Matter of France or Matter of ; the Matter of Rome (Romance (heroic literature), romances in an ancient setting); and the Matter of Britain (Arthurian romances and Breton lais). The first of these is the subject area of the ''Chanson de geste, chansons de geste'' ("songs of exploits" or "songs of (heroic) deeds"), epic poems typically composed in decasyllable, ten-syllable assonanced (occasionally rhymed) ''laisses''. More than one hundred ''chansons de geste'' have survived in around three hundred manuscripts. The oldest and most celebrated of the ''chansons de geste'' is ''The Song of Roland'' (earliest version composed in the late 11th century). Bertrand de Bar-sur-Aube in his ''Girart de Vienne'' set out a grouping of the ''chansons de geste'' into three literature cycle, cycles: the ''Geste du roi'' centering on Charlemagne, the ''Geste de Garin de Monglane'' (whose central character was William of Gellone, William of Orange), and the ''Geste de Doon de Mayence'' or the "rebel vassal cycle", the most famous characters of which were Renaud de Montauban and Girart de Roussillon. A fourth grouping, not listed by Bertrand, is the ''Crusade cycle'', dealing with the First Crusade and its immediate aftermath. Jean Bodel's other two categories—the "Matter of Rome" and the "Matter of Britain"—concern the French Romance (heroic literature), romance or ''roman''. Around a hundred verse romances survive from the period 1150–1220. From around 1200 on, the tendency was increasingly to write the romances in prose (many of the earlier verse romances were adapted into prose versions), although new verse romances continued to be written to the end of the 14th century. The most important romance of the 13th century is the ''Romance of the Rose'', which breaks considerably from the conventions of the chivalric adventure story. Medieval French lyric poetry was indebted to the poetic and cultural traditions in Southern France and Provence—including Toulouse, Poitiers, and the Aquitaine region—where ''langue d'oc'' was spoken (Occitan language); in their turn, the Provençal poets were greatly influenced by poetic traditions from the Muslim Spain, Hispano-Arab world. Lyric poets in Old French are called ''trouvères'' – etymologically the same word as the ''troubadours'' of Old Occitan, Provençal or langue d'oc (from the verb ''trobar'' "to find, to invent"). By the late 13th century, the poetic tradition in France had begun to develop in ways that differed significantly from the troubadour poets, both in content and in the use of certain fixed forms. The new poetic (as well as musical: some of the earliest medieval music has lyrics composed in Old French by the earliest composers known by name) tendencies are apparent in the ''Roman de Fauvel'' in 1310 and 1314, a satire on abuses in the medieval church, filled with medieval motets, Lai (poetic form), lais, Rondeau (music), rondeaux and other new secular forms of poetry and music (mostly anonymous, but with several pieces by Philippe de Vitry, who would coin the expression ''ars nova'' to distinguish the new musical practice from the music of the immediately preceding age). The best-known poet and composer of ''ars nova'' secular music and chansons of the incipient period was Guillaume de Machaut. Discussions about the origins of non-religious theater (''théâtre profane'') – both drama and farce—in the Middle Ages remain controversial, but the idea of a continuous popular tradition stemming from Latin comedy and tragedy to the 9th century seems unlikely. Most historians place the origin of medieval ''drama'' in the church's liturgical dialogues and "tropes". Mystery plays were eventually transferred from the monastery church to the chapter house or refectory hall and finally to the open air, and the vernacular was substituted for Latin. In the 12th century one finds the earliest extant passages in French appearing as refrains inserted into liturgical dramas in Latin, such as a Saint Nicholas (patron saint of the student clercs) play and a Saint Stephen play. An early French dramatic play is ''Le Jeu d'Adam'' (c. 1150) written in octosyllabic rhymed couplets with Latin stage directions (implying that it was written by Latin-speaking clerics for a lay public). A large body of fables survive in Old French; these include (mostly anonymous) literature dealing with the recurring trickster character of Reynard the Fox. Marie de France was also active in this genre, producing the ''Ysopet'' (Little Aesop) series of fables in verse. Related to the fable was the more bawdy ''fabliau'', which covered topics such as cuckolding and corrupt clergy. These ''fabliaux'' would be an important source for Chaucer and for the Renaissance short story (''conte'' or ''nouvelle''). Among the earliest works of rhetoric and logic to appear in Old French were the translations of ''Rhetorica ad Herennium'' and Boethius' ''De topicis differentiis'' by John of Antioch (translator), John of Antioch in 1282.
PhonologyOld French was constantly changing and evolving; however, the form in the late 12th century, as attested in a great deal of mostly poetic writings, can be considered standard. The writing system at this time was more phonetic than that used in most subsequent centuries. In particular, all written consonants (including final ones) were pronounced, except for ''s'' preceding non-stop consonants and ''t'' in ''et'', and final ''e'' was pronounced . The phonological system can be summarised as follows:
ConsonantsNotes: * All obstruents (plosives, fricatives and affricates) were subject to word-final devoicing, which was usually indicated in the orthography. * The affricates , , , became fricatives (, , , ) in . ** had three spellings – ''c'' before ''e'' or ''i'', ''ç'' before other vowels, or ''z'' at the end of a word – as seen in ''cent'', ''chançon'', ''priz'' ("a hundred, song, price"). ** was written as ''z'', as in ''doze'' "twelve", and only occurred in the middle of the word. * (''l mouillé''), as in ''conseil'', ''travaillier'' ("advice, to work"), became in . * appeared not only in the middle of a word, but also at the end, as in ''poing'' "fist". At the end of a word, was later lost, leaving a nasalized vowel. * was found only in Germanic loanwords or words influenced by Germanic (cf. ''haut, hurler''). It was later lost as a consonant, though it was transphonologization, transphonologized as the so-called aspirated h that blocks liaison (French), liaison. In native Latin words, had been lost early on, as in ''om'', ''uem'', from ''homō''. * Intervocalic from both Latin and was Lenition, lenited to in the early period (cf. contemporary Spanish: ''amado'' ). At the end of words it was also devoiced to . In some texts it was sometimes written as ''dh'' or ''th'' (''aiudha, cadhuna, Ludher, vithe''). By 1100 it disappeared altogether.
VowelsIn Old French, the nasal vowels were not separate phonemes but only allophones of the oral vowels before a nasal consonant. The nasal consonant was fully pronounced; ''bon'' was pronounced ( ). Nasal vowels were present even in open syllables before nasals where Modern French has oral vowels, as in ''bone'' ( ''bonne'' ).
MonophthongsNotes: * had formerly existed but then closed to ; the original Western Romance having previously been fronted to across most of what is now France and northern Italy. ** would later appear again when monophthongized and also when closed in certain positions (such as when it was followed by original or but not by , which later became ). ** may have similarly become closed to , in at least in some dialects, since it was borrowed into Middle English as > ( ''computāre'' > ''conter'' > English ''count''; ''rotundum'' > ''ront'' > English ''round''; ''bonitātem'' > ''bonté'' > English ''bounty''). In any case, traces of such a change were erased in later stages of French, when the close nasal vowels were opened to become . * may have existed in the unstressed third-person plural verb ending ''-ent'', but it may have already passed to , which is known to have happened no later than the Middle French period.
Diphthongs and triphthongsNotes: *In Early Old French (up to about the mid-12th century), the spelling represented a diphthong instead of the later monophthong , and represented the diphthong , which merged with in Late Old French (except when it was nasalized). *In Early Old French, the diphthongs described above as "rising" may have been falling diphthongs (, , ). In earlier works with vowel assonance, the diphthong written did not assonate with any pure vowels, which suggests that it cannot have simply been . *The pronunciation of the vowels written and is debated. In the first records of Early Old French, they represented and were written as , and by , they had both merged as , but the transitional pronunciations are unclear. *Early Old French had additional triphthongs and (equivalent to diphthongs followed by ); these soon merged into and respectively. *The diphthong was rare and had merged into by Middle French ( ''tiule'' > ''tuile'' 'tile'; ''siure'' > Late ''suire'' > ''suivre'' 'follow').
HiatusIn addition to diphthongs, Old French had many instances of hiatus (linguistics), hiatus between adjacent vowels because of the loss of an intervening consonant. Manuscripts generally do not distinguish hiatus from true diphthongs, but modern scholarly transcription indicates it with a diaeresis (diacritic), diaeresis, as in Modern French: * ''audīre'' > ' 'hear' ( ''ouïr'') * *''vidūta'' > ' 'seen' ( ''vue'') * ''rēgīnam'' > ', 'queen' ( ''reine'') * ''pāgēnsem'' > ' 'country' ( ''pays'') * ''augustum'' > ' 'August' ( ''août'') * ''patellam'' > ' 'pan' ( ''poêle'') * ''quaternum'' > ' 'booklet, quire' ( ''cahier'') * ''aetāticum'' > ', ' 'age' ( ''âge'')
Sample textPresented below is the first laisse of ''The Song of Roland'' along with a broad transcription reflecting reconstructed pronunciation circa 1050 C.E.
NounsOld French maintained a two-case system, with a nominative case and an oblique case, for longer than some other Romance languages as Spanish language, Spanish and Italian language, Italian did. Case distinctions, at least in the masculine grammatical gender, gender, were marked on both the definite article and the noun itself. Thus, the masculine noun ''li veisins'' 'the neighbour' was declined as follows: In later Old French, the distinctions had become moribund. As in most other Romance languages, it was the oblique case form that usually survived to become the Modern French form: ''l'enfant'' "the child" represents the old oblique (Latin accusative ''īnfāntem''); the nominative was ''li enfes'' ( ''īnfāns''). There are some cases with significant differences between nominative and oblique forms (derived from Latin nouns with a stress shift between the nominative and other cases) in which either it is the nominative form that survives or both forms survive with different meanings: * Both ''li sire'', ''le sieur'' ( ''seiior'', ''seiiōrem'') and ''le seignor'' (nom. †''sendre''; ''senior'', ''seniōrem'') survive in the vocabulary of later French (''sire'', ''sieur'', ''seigneur'') as different ways to refer to a feudal lord. * ''wikt:sœur, sœur'' "sister" is the nominative form ( ''wikt:suer, suer'' < Latin nominative ''soror''); the oblique form ''seror'' (< Latin accusative ''sorōrem'') no longer survives. * ''wikt:prêtre, prêtre'' "priest" is the nominative form ( ''wikt:prestre, prestre'' < ''presbyter''); the oblique form ''wikt:prevoire, prevoire'', later ''wikt:provoire, provoire'' (< ''presbyterem'') survives only in the Paris street name ''Rue des Prouvaires''. * indefinite pronoun ''wikt:on, on'' "one" continues Old French nominative ''wikt:hom, hom'' "man" (< ''homō''); ''wikt:homme, homme'' "man" continues the oblique form ( ''wikt:home, home'' < ''hominem''). In a few cases in which the only distinction between forms was the nominative ''-s'' ending, the ''-s'' was preserved. An example is ''wikt:fils, fils'' "son" (< Latin nominative ''fīlius''). The fact that the ''-s'' in the word is still pronounced today is irregular, but has to do with the later developments, namely the and Early Modern French system of pausa, pausal pronunciations. As in Spanish and Italian, the neuter gender was eliminated, and most old neuter nouns became masculine. Some Latin neuter plurals (which ended in ''-a'') were reanalysed as feminine singulars: ''gaudium'' was more widely used in the plural form ''gaudia'', which was taken for a singular in Vulgar Latin and ultimately led to ''la joie'', "joy" (feminine singular). Nouns were declined in the following declensions: Class I is derived from the Latin first declension. Class Ia mostly comes from Latin feminine nouns in the third declension. Class II is derived from the Latin second declension. Class IIa generally stems from second-declension nouns ending in ''-er'' and from third-declension masculine nouns; in both cases, the Latin nominative singular did not end in ''-s'', which is preserved in Old French. The classes show various analogical developments: Class I nominative plural ''-es'' from the accusative instead of ''-∅'' (''-e'' after a consonant cluster) in Class I nominative plural ( ''-ae'', although there is evidence to suggest Romance plurals#Origin of plural -s, this analogy had already occurred in VL), ''li pere'' instead of ''*li peres'' ( ''illi patres'') in Class IIa nominative plural, modelled on Class II, etc. Class III nouns show a separate stem in the nominative singular that does not occur in any of the other forms: * IIIa nouns are agent nouns which ended in ''-ātor'', ''-ātōrem'' in Latin and preserve the stress shift. * IIIb nouns also had a stress shift, from ''-ō'' to ''-ōnem'' (although several IIIb nouns actually continue Frankish language, Frankish Weak inflection#Nouns, weak nouns with a similar inflection: Frankish wikt:Reconstruction:Proto-West_Germanic/barō, ''*barō'' ~ ''*baran'' becomes ''ber'' ~ ''baron''). * IIIc nouns are an Old French creation and have no clear Latin antecedent. * IIId nouns represent various other third-declension Latin nouns with stress shift or a change of consonant (''soror'', ''sorōrem; īnfāns, īnfāntem; presbyter, presbyterem; seiior, seiiōrem; comes, comitem''). Regular feminine forms of masculine nouns are formed by adding an ''-e'' to the masculine stem (unless the masculine stem already ends in ''-e''). For example, ''wikt:bergier, bergier'' (shepherd) becomes ''wikt:bergiere, bergiere'' ( ''wikt:berger, berger'' and ''wikt:bergère, bergère'').
AdjectivesAdjectives agree in terms of Grammatical number, number, gender and case with the noun that they are qualifying. Thus, a feminine plural noun in the nominative case requires any qualifying adjectives to be feminine, plural and nominative. For example, in ''femes riches'', ''wikt:riche#Old French, riche'' has to be in the feminine plural form. Adjectives can be divided into three declensional classes:Moignet (1988, p. 26–31), Zink (1992, p. 39–48), de La Chaussée (1977, p. 39–44) *Class I corresponding roughly to Latin declension#First and second declension adjectives, Latin 1st- and 2nd-declension adjectives *Class II corresponding roughly to Latin declension#Third declension adjectives, Latin 3rd-declension adjectives *Class III containing primarily the descendants of Latin declension#Comparative and superlative forms of adjectives, Latin synthetic comparative forms in ''-ior'', ''-iōrem''. Class I adjectives have a feminine singular form (nominative and oblique) ending in ''-e''. They can be further subdivided into two subclasses, based on the masculine nominative singular form. Class Ia adjectives have a masculine nominative singular ending in ''-s'': :''wikt:bon#Old French, bon'' "good" (< ''wikt:bonus#Latin, bonus'', > ''wikt:bon#French, bon'') : For Class Ib adjectives, the masculine nominative singular ends in ''-e'', like the feminine. There are descendants of Latin second- and third-declension adjectives ending in ''-er'' in the nominative singular: :''aspre'' "harsh" (< ''wikt:asper#Latin, asper'', > ''wikt:âpre#French, âpre'') : For Class II adjectives, the feminine singular is not marked by the ending ''-e'': :''wikt:granz#Old French, granz'' "big, great" (< ''wikt:grandis#Latin, grandis'', > ''wikt:grand#French, grand'') : An important subgroup of Class II adjectives is the present participial forms in ''-ant''. Class III adjectives have a stem alternation, resulting from stress shift in the Latin third declension and a distinct neuter form: :''wikt:meillor#Old French, mieudre'' "better" (< ''wikt:melior#Latin, melior'', > ''wikt:meilleur#French, meilleur'') : In later Old French, Classes II and III tended to be moved across to Class I, which was complete by Middle French. Modern French thus has only a single adjective declension, unlike most other Romance languages, which have two or more.
VerbsVerbs in Old French show the same extreme phonological deformations as other Old French words; however, morphologically, Old French verbs are extremely conservative in preserving intact most of the Latin alternations and irregularities that had been inherited in Proto-Romance. Old French has much less analogical reformation than Modern French has and significantly less than the oldest stages of other languages (such as Old Spanish) despite the fact that the various phonological developments in Gallo-Romance and Proto-French led to complex alternations in the majority of commonly-used verbs. For example, the verb ''laver'' "to wash" ( ''lavāre'') is conjugated ''je lef'', ''tu leves'', ''il leve'' in the present indicative and ''je lef'', ''tu les'', ''il let'' in the present subjunctive, in both cases regular phonological developments from Latin indicative ''lavō'', ''lavās'', ''lavat'' and subjunctive ''lavem'', ''lavēs'', ''lavet''. The following paradigm is typical in showing the phonologically regular but morphologically irregular alternations of most paradigms: *The alternation ''je lef'' ~ ''tu leves'' is a regular result of the final devoicing triggered by loss of final /o/ but not /a/. *The alternation ''laver'' ~ ''tu leves'' is a regular result of the diphthongization of a stressed open syllable /a/ into /ae/ > > . *The alternation ''je lef'' ~ ''tu les'' ~ ''il let'' in the subjunctive is a regular result of the simplification of the final clusters /fs/ and /ft/, resulting from loss of /e/ in final syllables. Modern French, on the other hand, has ''je lave'', ''tu laves'', ''il lave'' in both indicative and subjunctive, reflecting significant analogical developments: analogical borrowing of unstressed vowel /a/, analogical ''-e'' in the first singular (from verbs like ''j'entre'', with a regular ''-e'' ) and wholesale replacement of the subjunctive with forms modelled on ''-ir''/''-oir''/''-re'' verbs. All serve to eliminate the various alternations in the verb paradigm. Even modern "irregular" verbs are not immune from analogy: For example, ''je vif'', ''tu vis'', ''il vit'' (''vivre'' "to live") has yielded to modern ''je vis'', ''tu vis'', ''il vit'', eliminating the unpredictable ''-f'' in the first-person singular. The simple past also shows extensive analogical reformation and simplification in Modern French, as compared with Old French. The Latin pluperfect was preserved in very early Old French as a past tense with a value similar to a preterite or imperfect. For example, the (878 AD) has past-tense forms such as ''avret'' (< ''habuerat''), ''voldret'' (< ''voluerat''), alternating with past-tense forms from the Latin perfect (continued as the modern "simple past"). Old Occitan also preserved this tense, with a conditional mood, conditional value; Spanish language, Spanish still preserves this tense (the ''-ra'' imperfect subjunctive), as does Portuguese language, Portuguese (in its original value as a pluperfect indicative).
Verb alternationsIn Latin, stress was determined automatically by the number of syllables in a word and the syllable weight, weight (length) of the syllables. That resulted in certain automatic stress shifts between related forms in a paradigm, depending on the nature of the suffixes added. For example, in ''pensō'' "I think", the first syllable was stressed, but in ''pensāmus'' "we think", the second syllable was stressed. In many Romance languages, vowels diphthongized in stressed syllables under certain circumstances but not in unstressed syllables, resulting in alternations in verb paradigms: Spanish language, Spanish ''pienso'' "I think" vs. ''pensamos'' "we think" (''pensar'' "to think"), or ''cuento'' "I tell" vs. ''contamos'' "we tell" (''contar'' "to tell"). In the development of French, at least five vowels diphthongized in stressed, open syllables. Combined with other stress-dependent developments, that yielded 15 or so types of alternations in so-called Irregular verb, strong verbs in Old French. For example, diphthongized to before nasal stops in stressed, open syllables but not in unstressed syllables, yielding ''aim'' "I love" ( ''amō'') but ''amons'' "we love" ( ''amāmus''). The different types are as follows: In Modern French, the verbs in the ''-er'' class have been systematically levelled. Generally, the "weak" (unstressed) form predominates, but there are some exceptions (such as modern ''aimer''/''nous aimons''). The only remaining alternations are in verbs like ''acheter''/''j'achète'' and ''jeter''/''je jette'', with unstressed alternating with stressed and in (largely-learned) verbs like ''adhérer/j'adhère'', with unstressed alternating with stressed . Many of the non-''er'' verbs have become obsolete, and many of the remaining verbs have been levelled; however, a few alternations remain in what are now known as irregular verbs, such as ''je tiens'', ''nous tenons''; ''je dois'', ''nous devons'' and ''je meurs'', ''nous mourons''. Some verbs had a more irregular alternation between different-length stems, with a longer, stressed stem alternating with a shorter, unstressed stem. That was a regular development stemming from the loss of unstressed Romance languages#Intertonic vowels, intertonic vowels, which remained when they were stressed: *''j'aiu''/''aidier'' "help" < ''adiūtō'', ''adiūtāre'' *''j'araison''/''araisnier'' "speak to" < ''adratiōnō'', ''adratiōnāre'' *''je deraison''/''deraisnier'' "argue" < ''dēratiōnō'', ''dēratiōnāre'' *''je desjun''/''disner'' "dine" < ''disiēiūnō'', ''disiēiūnāre'' *''je manju''/''mangier'' "eat" < ''mandūcō'', ''mandūcāre'' *''je parol''/''parler'' "speak" < ''*paraulō'', ''*paraulāre'' < ''parabolō'', ''parabolāre'' The alternation of ''je desjun'', ''disner'' is particularly complicated; it appears that: Both stems have become full verbs in Modern French: ''déjeuner'' "to have lunch" and ''dîner'' "to dine". Furthermore, ''déjeuner'' does not derive directly from ''je desjun'' (< ''*disi(ēi)ūnō'', with total loss of unstressed ''-ēi-''). Instead, it comes from ''desjeüner'', based on the alternative form ''je desjeün'' (< ''*disiē(i)ūnō'', with loss of only ''-i-'', likely influenced by ''jeûner'' "to fast" < ''jeüner'' < ''je jeün'' "I fast" < ''iē(i)ūnō'': ''iē-'' is an initial rather than intertonic so the vowel ''-ē-'' does not disappear).
Example of regular ''-er'' verb: ''durer'' (to last)Non-finite forms: * Infinitive: wikt:durer#Old French, ''durer'' * Present participle: ''durant'' * Past Participle: ''duré'' Auxiliary verb: ''avoir''
Example of regular ''-ir'' verb: ''fenir'' (to end)Non-finite forms: * Infinitive: wikt:fenir#Old French, ''fenir'' * Present participle: ''fenissant'' * Past participle: ''feni(t)'' Auxiliary verb: ''avoir''
Example of regular ''-re'' verb: ''corre'' (to run)Non-finite forms: * Infinitive: wikt:corre#Old French, ''corre'' * Present participle: ''corant'' * Past participle: ''coru(t)'' Auxiliary verb: ''estre''
Examples of auxiliary verbs
=''avoir'' (to have)= Non-finite forms: * Infinitive: wikt:avoir, ''avoir'' (earlier ''wikt:aveir, aveir'') * Present participle: ''aiant'' * Past participle: ''eü(t)'' Auxiliary verb: ''avoir''
=''estre'' (to be)= Non-finite forms: * Infinitive: wikt:estre, ''estre'' * Present participle: ''estant'' * Past participle: ''esté(t)'' Auxiliary verb: ''avoir''
Other parts of speechAdverbs, prepositions, conjunctions and interjections are generally invariable. Pronouns are usually declinable.
See also* Anglo-Norman literature * Bartsch's law * History of French * History of the English language * Influence of French on English * Languages of France
General sources* * * * * * * Grandgent, Charles Hall (1907). ''An introduction to Vulgar Latin''. Boston: D.C. Heath & Co. * Hall, Robert Anderson (October 1946). "Old French phonemes and orthography". ''Studies in Philology''. Vol. 43, No. 4. 575–585. . * * * Laborderie, Noëlle (2009). ''Précis de Phonétique Historique''. Paris: Armand Colin. * * * * * * Rickard, Peter (1989). ''A history of the French language''. London: Unwin Hyman. * *