The status of French in Medieval England: Evidence from the use of object pronoun syntax 1 Cet article se penche sur la syntaxe des pronoms personnels compléments d’objet direct en Anglo-Normand. À partir d’un examen de textes allant du XII e au XIV e siècles, il est demontré que l’Anglo-Norman a étroitement suivi l’évolution du français en ce qui concerne la place des pronoms clitiques dans les propositions infinitives. A également été respectée la place du pronom c. o. d. dans les propositions à auxiliaire fléchi, ainsi que dans les impératives. L’influence de la syntaxe du moyen anglais ne s’est fait aucunement sentir dans ces domaines. Cette analyse rejoint une perspective selon laquelle l’Anglo-Normand faisait partie d’un continuum dialectal francophone, au lieu d’en être coupé par son statut de langue seconde. Le problème est posé de la transmission de ces compétences linguistiques, soit par l’enseignement soit par d’autres moyens. The special status of Anglo-Norman (AN) in relation to continental French has recently been subject to renewed debate, one perspective emphasising its sui generis status as a second language in England (Kibbee 2000), and another that it was part of the medieval French dialect continuum (Trotter 2003a, b). This study presents evidence that syntactically AN was not an isolated variety, but closely reflected ongoing internal systemic change within continental French in the syntax of object pronouns, which is known to be vulnerable in the acquisition of French as a second language. An extensive sample of Anglo-Norman texts datable between the mid 12 th and the mid 14 th centuries is examined. Object pronoun use in four contexts is analysed: finite clauses with auxiliaries, non-finite clauses, V1 imperative clauses, and V2 imperative clauses. In all these contexts Anglo-Norman usage with respect to object pronoun morphosyntax very closely parallelled continental French, and did not reflect different object pronoun positions in Middle English. In finite clauses with auxiliaries, object pronouns continued to precede a tensed auxiliary. In nonfinite clauses, clitic pronouns underwent a switch towards preceding an infinitive after about 1320, as in continental French. Pronouns preceded verbs in V2 imperatives but not in ordinary V1 imperatives. English offered no model for such accurate positional discrimination across these differing contexts. It is concluded that, for this outcome to be observed, AN must have retained close ties with continental French linguistic developments. The issue is raised of how such linguistic competence was transmitted, whether by instruction or by other means. 1 I should like to thank Andres Kristol for the thought-provoking observations that encouraged me to undertake this investigation,Anne Curry and Gwilym Dodd for helpful advice on AN parliamentary petitions, and David Birdsong for his insightful and challenging comments on a later stage of this enquiry. All factual errors and misinterpretations here remain entirely my own, however. The assistance of the Anglo-Norman Hub in searching two of the texts is also acknowledged. Vox Romanica 65 (2006): 86-107 The status of French in Medieval England 1. Rationale for the study The extent and nature of the use of French in medieval England has provoked lively debate over many years (Suggett 1946, Legge 1950, 1980, Berndt 1972, Rothwell 1976, Richter 1979, Clanchy 1993, Kibbee 1991, 2000), focusing on how long it continued to be a spoken vernacular after the Norman Conquest. According to Suggett 1946: 79 «The French used in England was a true vernacular whose roots had penetrated deeply into all classes of English society who could read and write», and Legge 1980 contended that Anglo-Norman was a spoken variety of French, «absorbed naturally» in a communicative setting. However, Rothwell 1976: 44 considered that by the 13 th century «French was no more a vernacular than was Latin», while Kibbee 1996: 7-11 emphasised the «essential difference» between Anglo-Norman (henceforth AN) and continental French in this respect, drawing attention to AN gender errors and «special syntactic constructions that reflect English rather than French». Contemporary awareness of these differences is documented in Brereton 1939, who showed that continental French scribes corrected AN manuscripts on grammatical points such as the use of a les for aux, de les for des, and que for qui as a subject relative pronoun. Nevertheless, it is also quite common to find scholars such as Rothwell 1996 drawing attention to the often high standard of accuracy and fluency with which many later Anglo-Norman writers used French. The notion sometimes entertained that in the later medieval period French in England was no more than a «half-understood jargon» must clearly be rejected. Trotter 2003a: 239 drew attention to the extensive use of AN by English merchants for international correspondence in the 14 th century, describing it as «a perfectly acceptable variety of the Middle Ages’ second international language, after Latin». He presented grounds for viewing AN as one part of a medieval Francophone dialect continuum, which included Gascony, Flanders and other North-Eastern regions of what is now France. Völker’s 2003 analysis of 13 th century texts from the county of Luxembourg showed that the use of French in this region showed grammatical variation comparable with that of more central regions of the medieval Francophonie. Little work on grammatical variation has been been done for AN: its syntax has been particularly poorly explored (Wilshere 1993), which is regrettable, for at least two reasons. Certain aspects of ordinary clausal syntax are now fairly wellstudied with respect to diachronic developments in Old French (see e. g. Vance 1997, Labelle/ Hirschbühler 2005), and therefore allow fruitful comparisons to be made between the development of AN and continental usage to a much greater extent than previously. Secondly, formal syntactic contrasts, especially those lacking a clear semantic rationale, are known to be particularly vulnerable in second language learning (see e. g. van Boxtel et al. 2005). If, as claimed by advocates of the «essential difference» position, AN was «clearly a language learned at school» (Berndt 1972: 354), we might expect evidence of some difficulty in acquiring formal syntactic properties, taking the form of errors influenced by English, as indeed 87 Richard Ingham 88 Kibbee 1996 and others have claimed. Since we do not have direct access to how AN was pronounced (pace Pope 1934), we cannot be sure how far the phonology of AN diverged from that of Continental French, but the AN textual record provides ample opportunity to assess how far the formal systemic level of syntax did so. We seek to take this issue further by investigating the performance of AN writers in a domain of French grammar which is known to be vulnerable in second language learners: the morphosyntax of object pronouns, in particular their status as clitics rather than syntactically independent items. The difficulty posed to contemporary learners of French 2 by clitic pronouns has been well documented, e. g. by Gundel/ Tarone (1983) 3 . Given that Old French object pronoun morphosyntax (see section 2) was in key respects more complex than the modern system, non-nativelike performance might be expected at higher levels of proficiency than those at which errors with modern French are found. Such an outcome, if obtained, would surely argue in favour of the «essential difference» position as regards the status of Anglo-Norman. Indeed, Kibbee 1996 has already drawn attention to the tonic versus clitic object pronoun distinction as a point of difficulty mentioned in a late 14 th century didactic treatise produced for use in England, the French version of the Orthographia Gallica (Johnston 1987) 4 . In order to see what light can be shed on the psycholinguistic status of AN by this particular syntactic variable, we first analyse some of the major AN sources from the 12 th century, and then turn our attention to the later 13 th and 14 th centuries. We follow Pope 1934: 424 who distinguished an earlier period in which Anglo-Norman was «a living local form of speech, handed down from generation to generation» and a later «period of degeneracy in which insular French was cut off from its base and . . . gradually became a dead language that . . . always had to be taught.» If it is true that the status of French in England became that of a language taught in a classroom (see also Berndt 1972), evidence of increasing difficulty may be forthcoming as regards the abstract grammatical rules acquired by native speakers fairly early in childhood, such as those relating to verb finiteness and clitic versus non-clitic pronoun behaviour mentioned above. In a second language learning scenario, it would be normal to suppose that L1 English could have influenced the patterns of object pronoun use, since the languages differed in the morphosyntax of object pronouns, in ways presented in the next section. 2 Clitic object pronouns are also particularly vulnerable in French Specific Language Impairment (Paradis et al. 2005/ 06). 3 However, White’s 1996 study of child L2 learners of French, who were Anglophone Canadians placed in an immersion context at around age 5, showed that they acquired clitic object pronouns after two years or so from initial exposure to the language. 4 The rule - shown as F42 in Johnston 1987: 35 - was presented by the author of the Frenchversion Orthographia Gallica as a matter of using the clitic series as the object of a verb and the tonic series as the object of a preposition, which clearly does not suffice as a formulation of the Old French patterns with verbs: see section 2. The status of French in Medieval England 2. Object pronoun forms in Old French and Middle English Old French object pronouns were governed by rules that depended on abstract syntactic factors involving verb finiteness and structural position. There were two series of direct object pronouns 5 , tonic forms and clitic forms, as shown in Table 1: Table 1: Morphology of Old French direct object pronouns Persons: 1 st sing. 2 nd sing. 3 rd sing. 3 rd pl. clitic me te le, la les tonic mei tei lui, li eus The 1 st and 2 nd plural forms nos et vos, which figured in both series, were designated «pronoms indifférents» by Buridant 2000 and as such were not considered in the present study. When an object pronoun stood adjacent to a finite verb in a declarative clause, use of the clitic form was normally obligatory; a clitic pronoun had to precede the finite verb, even in an auxiliated clause. When an object pronoun appeared in a non-finite clause, a member of either series could be used, with the proviso that the weak form could follow, but not precede, the non-finite verb, e. g.: (1a) . . . por doner le au chien DitsSQ P, 108 (Buridant 2000: 443) (1b) Je avois appris a veoir les sovent Graal 17, 10-12 (Foulet 1930: §182). The respective distributional possibilities in declarative clauses were thus as follows (adapted from Buridant 2000: 447, spelling modernised): object pronoun in clause with finite auxiliary (2a) Il le peut voir (2c) *Il lui peut voir (2b) *Il peut le voir (2d) *Il peut lui voir object pronoun in non-finite clause (3a) *Il vint por le voir (3c) Il vint pour lui voir (3b) Il vint por voir le (3d) Il vint pour voir lui The system of Old French object pronoun morpho-syntax was clearly a matter of considerable complexity: in finite declarative clauses, even those with an auxiliary, an object pronoun had to stand before the finite verb, as in (2a), but in non-finite 89 5 In this study we have not considered indirect object pronouns, which raise similar but not identical issues of comparability with continental French. An additional complication here is that an indirect object pronoun could be replaced by a prepositional phrase, e. g. a eus, a li, cf.: (i) . . . et encore les voulent constraindre a rendre a euls ou nom du roy tous les profiz (Lettre d’une dame de Cassel, Pir. 1900: 199) Richard Ingham 90 clauses there was considerable, though not complete, flexibility, as in (3b-d) In imperative clauses, complexity took a different form. In V1 imperatives, i. e. imperative clauses without an initial sentence constituent, 1 st and 2 nd person objects took the tonic form, while 3 rd person objects took the clitic form. In V2 imperatives the object pronoun had to stand before the imperative verb, and be in the clitic form in the singular and 3 rd plural. These differences are illustrated in the following examples: (4a) Souviegne toi. Bible française du XIII e (ed. M. Quereuil, Genève 1988) siècle, Genèse, p. 93 (4b) Pursiu les Li quatre livre des reis 58, 8 (Hirschbühler/ Labelle 2001) (4c) Si me fai boire. Jeu de S. Nicholas, v. 693 (ed. A. Henry, Droz, 1981) (4d) Car m’eslisez un Roland, 275 (Hirschbühler/ Labelle 2001) barun de ma marche. The rather intricate distributional possibilities needing to be acquired by a learner of Old French in declarative and imperative clauses are summarized in Table 2: Table 2: Syntax of Old French object pronouns declaratives finite verb/ auxiliary infinitive in non-finite clause clitic pron. precedes √ X clitic pron. follows X √ tonic pron. precedes X √ tonic pron. follows X √ imperatives V1 imperative V2 imperative 1 st person sing. clitic pronoun precedes X √ clitic pronoun follows X X tonic pronoun precedes X X tonic pronoun follows √ X 2 nd person sing. clitic pronoun precedes X √ clitic pronoun follows X X tonic pronoun precedes X X tonic pronoun follows √ X 3 rd person sing. & plur. clitic pronoun precedes X √ clitic pronoun follows √ X tonic pronoun precedes X X tonic pronoun follows X X Since formal alternatives within the system of Old French object pronoun syntax would have had relatively low perceptual salience and limited communicative value, they would have been vulnerable to imperfect second language acquisition, The status of French in Medieval England thus giving rise to outcomes in individual learners that might be incomplete or divergent (Sorace 2003) with respect to the assumed target grammar. Furthermore, the system was vulnerable to errors produced by influence from the L1 English system. In early Middle English (c. 1150-1300) declarative clauses, object pronouns could stand before a finite verb/ auxiliary (5), before a non-finite verb (6a-b), or after a non-finite verb (7a-b): (5) & te lundenissce folc hire wolde tæcen PC 58, 21 (1140) and the London-ish people her wanted take-INF ‘and the people of London wanted to take her’ (6a) . . . pet na gastlich cunfort ne mei hire gleadien AW f48a (c. 1225) that no spiritual comfort NEG may her console-INF ‘that no spiritual comfort may console her’ (6b) . . . pet ping se feble as flesch is . . . schal him ouerstihen Hali Meid 204 (c. 1225) that thing so weak as flesh is . . . shall him surpass-INF ‘. . . that a thing so weak as flesh is . . shall surpass him’ (7a) . . . dat he uuolde iiuen heom up Wincestre PC 58, 26 (1140) that he would give-INF them up Winchester ‘. . . that he would surrender Winchester to them’ (7b) . . . al dat he cuthe axen him PC 58, 1 (1140) all that he could ask-INF him ‘all that he could ask of him’ In imperative clauses, however, Middle English (ME) positioned the verb uniformly before object pronouns in both V1 and V2 imperatives: (8) Wasshed ow hwar se neod is AW f115a (c. 1225) wash-IMP you where the need is ‘Wash yourselves as necessary’ (9) Wid peawfule talen schurted ow to gederes AW f115a (c. 1225) with instructive tales divert-IMP you together ‘Entertain yourselves together with instructive tales’ 6 Thus English influence, if present, might show up in AN texts as a tendency to give either or both of the French pronominal series the free distribution in declaratives 91 6 ME texts of different periods and regional origins display considerable variation in terms of positional preferences, but object pronouns in declaratives seem not to have followed fixed rules governing the distribution of two sets of pronoun forms, as they did in Old French. Richard Ingham 92 and the fixed position in imperatives as English object pronouns had. If so, errors would be expected along the lines illustrated by the following (spelling modernised): Declarative (10a) *Il lui doit voir (10b) *Il veut le voir (10c) *Il vint por le voir Imperative (10d) *Vois lui! (10e) *Or vois le! (10f) *Vois me (10g) *Or moi vois In this study we enquire whether distributional patterns of object pronouns in AN texts diverged from the continental norm towards Middle English, displaying some or all of the above potential errors in declarative and imperative clauses. First, however, we note that during the period covered by this study Continental French usage itself underwent a change with respect to the syntax of object pronouns in non-finite clauses which must be taken into account. 3. The evolution of object pronoun use in Continental French Marchello-Nizia’s (1997: 248-49) study of Middle French pronoun morphosyntax observed a shift towards preposing weak forms before the non-finite verb that made itself felt from the mid-14 th century onwards. The process took some time to complete, with the strong forms moy, toy and soy surviving longer in preverbal position than lui, li/ elle and euls. Further examination of continental French that we have conducted using non-literary texts concords with her analysis. Around 1300, texts still generally show the older patterns. The tonic forms were used before a non-finite verb, e. g.: (11a) Que elle le lairoit aler Plainte contre le bailli Recueil des Hist et soi pourchacier de Vermandois de Fr XXIV/ 2, 699, L. (11b) (. . . comme il les vorront faire F-B: 84-86, N4 (1302) deviser des dittes cozes a entendre) et eaus tenir sans damage (11c) . . . et de lui allier au roy, ensi F-B: 282-83 N1 (1298) comme ses peres i fu (11d) Et ont recue la davant dite Hauuy Monfrin, Vosges, p. 143 en serour de lor ordre et por li encevelir a la-mort Where clitic pronoun forms occurred in non-finite clauses, they followed nonfinite verbs, e. g.: (12) Et voir le nous convenra autrement que de paroles Pir. 163 (1324) The status of French in Medieval England By the mid-14 th century, as in the Correspondance of Charles V, clitic object pronouns regularly appear preverbally in a non-finite dependent clause, e. g.: (13) Qui y auront este a le mener en noz dis greniers Charles V, p. 14 (1364) The later Middle French preference for placing object pronouns in non-finite clauses before the non-finite verb probably began in late Old French. In the later 13 th century Coutumes du Beauvoisis, of Philippe de Beaumanoir, it makes an occasional appearance, e. g.: (14) . . . qui sont mis es garnisons . . . pur les garder B 57 Mostly however, Beaumanoir still displayed the Old French patterns, e. g.: (15a) S’il n’a especial comandement de son seigneur de fere le. B 31 (15b) . . . grant peine metre en li maintenir sagement et loialment en l’office la u il est B 57 (15c) Il se doit presenter . . . et soi ofrir contre cex a qui . . . B 61 It seems that the later 13 th and early 14 th centuries were a period of transition between Old and Middle French, in this as in other syntactic respects (see e. g. Vance 1997). A consequence of this development was that the target for Anglo-Norman writers seeking to follow continental practice with clitic pronouns had changed somewhat by early Middle French. In non-finite clauses a clitic pronoun could now appear before the infinitive, though it could still not do so in finite auxiliated clauses. 4. Anglo-Norman data sources Most surviving texts from the earlier AN period are in verse, whereas in the later period French became used for a much wider range of administrative and other purposes. The grammar of the earlier period appears to show little or no divergence from continental Old French, but thereafter the standard of French, in terms of how well AN observed continental grammatical norms, is supposed to have declined, according to e. g. Kibbee 1991. To the best of our knowledge, however, very few methodical investigations have been carried out to demonstrate in what respects this decline can be assessed as regards grammatical accuracy 7 . Assuming 93 7 Chapple 1938: 146 briefly mentioned object pronouns in the mostly 14 th century Correspondence of the City of London, but appears from the following statement to have misunderstood the situation as regards certain targetlike forms in Old French: «The strong forms of the pronouns are used occasionally where the weak are expected: . . . pur lui esbaignier, . . . et lui justicier.» These strong forms are, on the contrary, entirely to be expected if AN writers were following Old French non-finite clause norms. Richard Ingham 94 such a development took place, one would expect it to take the form of a decline in the extent to which French was acquired in a nativelike fashion. Abstract syntactic rules such as those for Old French object pronouns presented in section 2 would be prime targets for imperfect learning, and thus constitute linguistic variables allowing us to address the hypothesis that the quality of French in England declined as a result of its having become an imperfectly learned second language. We know this was to be the outcome in later Law French, where by the later 15 th century we find usage that clearly diverges from continental norms, such as: (16a) . . . issint quil nauera vnques eux [sc. a hen or a capon] arrere SCEC 124 (c. 1456) 8 (16b) Le chaunceler adiourne eux en leschekere Chambyr SCEC 147 (c. 1458) (16c) Ieo ne barre luy pur ceo mater SCEC 151 (c. 1458) (16d) . . . sicome le Roy graunt a moy toutz lez finez et amercimentz SCEC 146 (c. 1458) deinz vne certeine lieu coment que le Roy voille pardoner eux Here, non-contrastive uses of strong form pronouns are placed after the finite verb as in (16a-c), and after the non-finite verb as in (16d). These are the positions in which pronouns were regularly found in the corresponding English sentences by the later 15 th century. In this study, Anglo-Norman sources of the 13 th and 14 th centuries are examined for evidence that this trend indicative of L1 influence was already underway at that time. We sought to compare object pronoun use in texts from Pope’s period I, which lasted roughly into the early decades of the 13 th century, and from period II, which ran from the mid-13 th to the later 14 th centuries. 1362 was chosen as a later cut-off point, since in that year court hearings were officially required to be held in English, which amounts to formal recognition that some people of importance did not speak French 9 . 1250 was designated as a fairly arbitrary starting point for period II, but one which corresponds roughly to the time from which it was claimed by Rothwell 1976 that French was being acquired in conditions of second language instruction. Since the study undertook a diachronic analysis of object pronoun syntax in AN, as compared with continental French, it was important to use only texts where the date of composition is reasonably well-established, and where the manuscript evidence is not compromised by a later re-working of an earlier text. The bigger the gap between date of composition and the date of the manuscript, the more opportunity there is for this to have occurred. Unfortunately, many AN works are difficult to date, particularly those of a literary nature. We therefore preferred for this purpose to opt for non-literary data sources, which tend to be easier to date, 8 The dates in examples (16)a-d refer to those of the law cases reported in the text. The mss used by Hemmant 1933 to edit the SCEC are apparently late 15 th century. 9 This decree was not implemented as regards the writtten record of court cases, though various oral procedures within the judicial process may already have been taking place in English (Ormrod 2003). The status of French in Medieval England notably correspondence, which is most often internally dated. For the 1250-1352 period, these criteria were met, but for the earlier time-period, this was less straightforward. However, seven verse texts that can be dated to the 12 th century or early 13 th were identified: Clemence of Barking’s The Life of St. Catherine, Hue de Rotelande’s Ipomedon and Protheselaus, Jordan Fantosme’s Chronicle, Le Mystère d’Adam, Le Petit Plet, and La Seinte Resureccion. Ideally speaking, it would have been preferable to use the same genres of text for both earlier and later periods, but this was simply not feasible in terms of the data sources available. As already mentioned, most pre-1200 AN material is in verse form, whereas AN was much more rarely used in verse in the C14. To exemplify period II we examined several sets of AN correspondence 10 (Litterae Cantuarienses, Royal correspondence of Henry III, Peckham’s Register, letters of Walter of Wenlok, Correspondence of the City of London, for the period 1320-62), as well as administrative texts (Oak Book of Southampton, c. 1300, responses of the King’s council to parliamentary petitions 1314-62), late 13 th century treatises on husbandry (Walter of Henley, etc.) and law (Britton, Placita Corone), a later 13 th century prose chronicle (Livere de reis de Engleterre), three religious works (Jerarchie, Holkham Bible Picture Book, and Henry of Lancaster’s Le Livre de Seyntz Medicines), and early 14 th century moral fables (Bozon). 5. Data analysis 5.1 Imperative clauses 5.1.1 Period I: 1150-1230 A total of 35 object pronouns in imperatives were identified in our Period I data sources. They showed almost complete accordance with continental models, cf. Table 2: Table 2: Object pronouns in V1 and V2 imperatives, Anglo-Norman C12 verse Pre-V Post-V V1 imperative 0 18 V2 imperative 16 1 (tonic) V1 imperatives always placed the object pronoun after the verb; 3 rd person pronouns were always clitic form, as required in Old French: 95 10 However, letters from and to royal personages themselves were excluded, on the basis that the court is known to have been French-speaking throughout the period under study, and indeed kings of England appear to have acquired French as a native language. For us, the question was whether French was used in other circles in England in a way that approximated continental French. Richard Ingham 96 (17a) Manje le Ad. 165 (17b) Guard les, Sire, d’aversitez CB 2575 (17c) Apelez les SR, ms. P. 84 (17d) Mandez les sempres ca a nus Prot. 535 (17e) Maintenez-le Pet. Pl. 1661 (17f) Guardum la JF 1230 1 st person pronouns were likewise targetlike in using the tonic form in V1 imperatives: (18a) Di mei que li oisel ferunt CB 1343 (18b) Baillez mei ca cel uinnement SR ms. P. 251 (18c) Mes dites mei Pet. Pl. 464 In V2 imperatives, the object pronouns were nearly always placed preverbally, e. g.: (19a) A Adam le done Ad. 263 (19b) Pois la me faites amener CB 2039 (19c) Mais or les alum assaillir Prot. 1234 (19d) Sun cors me donez pur enterer SR ms. C. 90 (19e) Si me cuntez de cel effroi Pet. Pl. 92 (19f) Tel kunseil me dunez JF 489 The single case of syntactic divergence from Old French norms shown in Table 2 occurred in Jordan Fantosme’s Chronicle, and placed the object pronoun postverbally: (20) Or veez mei ci JF 334 The Petit Plet twice uses a tonic form in a V2 imperative, which is untargetlike - cf. (10g) above - but still places the pronoun pre-verbally, as required: (21a) Ore mei metez en en bon espeir Pet. Pl. 616 (21b) Ore mei dites Pet. Pl. 1541 This is the only indication in the period I data of the use in AN of tonic forms in clitic contexts in the singular pronoun paradigm. 5.1.2 Period II: 1250-1362 In our period II data, imperative clauses continued to follow the Old French rule according to which an object pronoun stood after the verb in a V1 imperative, but before it in a V2 imperative: (22a) Maunde le moy LC II 108 (22b) E alowet les donc Walter c. 62 (22c) Dite lur ke . . . Rules xxiii (22d) Veiez le la Plac. Cor. 25 The status of French in Medieval England 97 (23a) Donk les lessez en pees LC I 418 (23b) En yver les sustenez Walter c. 94 (23c) Al pie senestre le attachez Boz 101 (23d) Vistement le fetes prendre Plac. Cor. 25 In period II, no cases of postverbal pronouns in V2 imperatives analogous to the English model in (8)-(9) occurred, such as *Donc laissez-les en paix. The single untargetlike example (20) noted in period I thus does not herald the collapse in Anglo-Norman of the Old French positional distinction. Where AN writers permitted themselves variation in the position of pronouns with V1 imperatives, it was just in those cases where continental Old French itself showed variation (Labelle/ Hirschbühler 2005), first where an imperative clause was modified by an initial subordinate proposition, as in (24 a-b), and secondly after the connector et, as in (24c-d): (24a) E si vos devez estor acheter, le achetez entre la Walter c. 77 Paske e la Pentecoste (24b) Si vus trovez nul qe ne seient mye seyn, les remuez Walter c. 93 (24c) Et le merciez moult de sa lettre LC I 188 (24d) Et les fetez arder Boz 114 On the assumption that AN was an instructed second language, it is interesting to note the absence of analogical errors in AN that might have been produced as a result of second language learners’ confusion arising from surface syntactic similarity across two different constructions, et with an imperative clause, and et introducing a subjectless indicative clause. When et was followed by an indicative verb governing an object pronoun in Old French the pronoun always preceded the verb. The AN data observed this restriction without exception: (25a) Et la remaunderoms a vous LC I 188 (25b) Vynt un venour . . . od un mot des chienz e les descoupla Boz 134 al gopil (25c) qui . . . trop poyz des cheveux ad a la teste e les veot Boz 113 ennoyter Pope (1934: 482) discussed the analogical generalisation in AN of various verbal morphological forms in the 12 th century and later. On the basis of examples such as those in (25) as regards object pronoun use, it seems that analogy played no role in the syntactic constructions studied here. The rules governing the placement of object pronouns were faithfully respected by AN writers, and were not extended to inappropriate contexts. Richard Ingham 98 5.2 Object pronouns in non-finite clauses 5.2.1 Period I All object pronouns 11 preceding the infinitive in non-finite clauses were tonic in period I, e. g.: (26a) Por ceo ne sevent nule mesure de mei empeirer Pet. Pl. 774/ 5 (26b) Od vos aler lui despendre SR ms. P. 204 (26c) Pur els maintenir JF 1111 (26d) Il vindrent pur li escharnir CB 1219 Of 24 examples of object pronouns depending on non-finite verbs, 22 took the tonic form and stood before the infinitive. Two stood after the infinitive, one tonic in form, the other clitic; all of these possibilities were grammatical in Old French. It is not clear why the data showed such a strong preference for the position preceding the infinitive, but it may be that metrical factors lie behind this, given that the period I sample consisted entirely of verse. 5.2.2 Period II To anticipate, the period II data shows the maintenance of the Old French object pronoun system in non-finite clauses very strongly until about 1320. For purposes of presentation, we have accordingly divided our discussion of the data into two sub-periods, before and after that date. In the sub-period up to 1320 object pronouns preceding the non-finite verb were in almost every case strong form, e. g.: (27a) Il ad un angle [sc. ange] ki entent a li garder Jerarchie 87, 27 (c. 1285) (27b) Nus sumes pourveu de cydre pur eus emplir Wenlok 294, 10 (c. 1299) (27c) Repentaunce . . . de sey amaunder et lesser ses folies Peckham II 489 (1282) (27d) . . . qe aucun des jurours est procuré a ly dampner Britton 5, 32 (c. 1290) In this subperiod, a single example of a preposed clitic pronoun form occurred: (28) Il iert tenuz de les rendre . . . Britton II, 2,4 (c. 1290) In other words, when clitic object pronouns were encountered, they almost always followed the non-finite verb, e. g.: (29a) . . . chasteaus le rei garder et sustenir les Burt. 479, 6 (1259) (29b) Metez peine d’estorer la Walter c. 31 (c. 1285) 11 In this and subsequent analyses we assumed that a pronoun such as me/ moy etc. governed by a verb was a direct object if that verb’s normal valency in continental Old French was transitive. For this reason, examples with forms of Mod. Fr. aider were excluded, since 13 th century Old French texts show some uses of aider as a verb taking the preposition à + Noun Phrase. The status of French in Medieval England 99 (29c) . . . pour eslire les draps et carier les a Cantorberis LC I 40 (1318) (29d) . . . ou W. a aver la si cum il la demande Brev. Plac. 43 (c. 1260) Altogether, out of 25 clitic object pronouns in non-finite clauses identified in texts dating between 1250-1319, 24 obeyed the Old French pronoun rules described in section 2 and stood after the non-finite verb, the sole exception being (28) above. A further twenty items were strong form, of which all except three stood before the non-finite verb. In the second subperiod, 1320-62, we find frequent examples of the modern pattern, e. g.: (30a) Et de le returner Rot. Parl. 436, 27 (1325) (30b) Et la mettre en bone fyn CCL p. 305 (1327) (30c) De les enjoyndre de par nostre seigneur le roi NR p. 407 (1359) Out of a total of 55 clitic object pronouns in this sub-period, 32 were now placed before the non-finite verb. Almost all were 3 rd singular or plural personal pronouns, conforming to the findings of Marchello-Nizia that in continental Old French change took place first with these forms: only a single case of se and none of me preceding non-finite verbs were noted. Strong object pronoun forms continued to be used both before (N: 39) and after (N: 5) non-finite verbs. The trend as compared with the first sub-period seems clear, and very closely follows continental developments, in which the late Old French/ Middle French innovation was the ability of clitic pronouns, especially third person personal pronouns, to precede a non-finite verb. In this respect, AN behaved as would be expected of a francophone dialect in touch with mainstream developments. 5.3 Finite clauses with auxiliaries We turn next to the syntax of object pronouns in finite clauses with auxiliaries. In Old French, the object pronoun preceded the tensed auxiliary accompanying the infinitive of which it was the logical object, as in (2a) above. In later Middle French, as already mentioned in section 3, this began to give way to the modern pattern, but as of the early 14 th century this had not yet happened, so the placement of object pronouns in other positions in auxiliated clauses would have been untargetlike, and potentially a sign of English influence, cf. the examples in (6)-(7). We searched the period II texts 12 in order to see whether the syntax of object pronouns 12 Corresponding period I data was not analysed, as the almost total uniformity of the Period II results made it highly unlikely that those from Period I would yield an interesting outcome indicating a departure from the Old French norms. Richard Ingham 100 in finite auxiliated clauses underwent change. If it did, this would clearly constitute a major divergence from continental French, which saw no change here during the period of enquiry. In the texts analysed, the syntactic position of object pronouns in AN finite auxiliated clauses was almost entirely targetlike in standing before the finite auxiliary, e. g.: (31a) . . . qe vous me voudrietz prendre en vostre protection LC I, 444 (1332) (31b) Il les deust quere meismes LC I, 488 (1332) Clitic pronouns never stood immediately before the non-finite verb in such clauses, despite the potential L1 model for this structure afforded by English, e. g. (6ab) above 13 . In a scattering of cases in each sub-period, a tonic pronoun form was used instead of a clitic before a finite auxiliary. The use of a tonic pronoun preceding the finite element is well-known as a later AN idiosyncrasy (Pope 1958), and a couple of cases were noted above with imperatives in le Petit Plet. The nature of such departures from the CF norm is neverthless worthy of comment. As shown in Table 3 below, they never involved the replacement of les by eux, or of la by elle, cases where the tonic and clitic forms were phonologically very distinct. They were found with soi for se and lui for le, where the greater phonological similarity of the forms might have been responsible for the confusion. Table 3: Object pronouns in clauses with finite auxiliaries 1250-1362 Sub-period 1, 1250-1319 tonic clitic total me/ moi 0 2 2 le/ lui/ li 3 30 33 la/ ele (li) 0 6 6 les/ eux 0 28 28 se/ soi 2 32 34 5 98 (95.2 %) 103 13 A single non-targetlike case was found where an atonic pronoun followed the non-finite verb: (ii) Eissi ke les seignurages mes ne pussent destriendre les par tel achesun (ita quod domini postea non possint distringere propter illam occasionem) Burt. 472, IV, 9 (1259) This was from a text consisting of a series of regulations translated from Latin, though it is not clear whether that fact had any role in the occurrence of this example. The absence of an object pronoun in the Latin source means that the syntax of the source text cannot have influenced the Anglo-Norman in this respect. The status of French in Medieval England 101 Table 3: Continued Sub-period 2, 1320-62 tonic clitic total me/ moi 0 10 10 le/ lui 3 55 58 la/ elle(li) 0 7 7 se/ soi 1 26 27 les/ eux 0 31 31 4 129 (97.0 %) 133 The relative rarity in both sub-periods of strong form pronouns preceding a finite auxiliary verb is interesting. The use of strong form pronouns preceding finite verbs has often been commented on, as a feature of later AN texts, yet it seems to have been a very uncommon lapse from continental norms, and did not grow in frequency up to the mid-14 th century, on the evidence presented here 14 , where AN writers continued to follow CF morphosyntax over 95 % of the time. Whether this particular syntactic context, or these particular texts, were in some way unrepresentative may be a matter for further enquiry. 6. General discussion The overall generalisation is that in each context studied, and in each time period studied, object pronoun use in AN very closely conformed to the syntactic patterns of continental French. Clear «interference errors», i. e. non-targetlike performance caused by the influence of L1 English on L2 French, were conspicuous by their absence. In Period I (later 12 th to early 13 th centuries), the distribution of the clitic object pronoun almost entirely respected the positional possibilities of Old French; in Period II (1250-1362), it underwent the same change as took place in Continental French, at first preserving the Old French pattern, until around 1320, and then shifting towards the Middle French pattern in just the same parts of the personal pronoun paradigm as on the continent. In short, with respect to this variable at least, AN was evolving very much as we should expect a dialect of French to do 15 . The dialect continuum position of Trotter (2003) thus receives support from this finding. In line with his conclusion that AN lexis was mostly simply French, rather than being peculiarly insular, we would argue that in the period studied AN grammar, in this respect at least, was essentially the grammar of French: it closely observed the 14 It should, however, be noted that about 45 % of the datapoints from subperiod 2 in Table 3 were contributed by Lanc., and all the strong forms occurred in other sources. 15 Note that later AN syntax, if this particular syntactic variable is anything to go by, differed from phonology, where, according to Pope (1934: 429) «in the later period . . . the influence [of] contemporary Continental changes was superficial only». Richard Ingham 102 distinctions that continental French made, and evolved in much the same way that continental French evolved. Because this syntactic variable, in common with much else, appears to have been little studied in continental French dialects other than Central French, it is unfortunately impossible to state in the present state of our research just how similar AN was to them in the nature and timing of this evolution. Nevertheless, it seems fair to say that if the AN texts we have been using as sources had been produced on the other side of the Channel, their object pronoun syntax would hardly provoke comment. This strongly contrasts with the grossly untargetlike pronoun usage in later Law French illustrated by (16) above, and suggests that the divergence of insular French grammar from the mainstream of continental varieties should be seen principally as a fifteenth century development. An often-noted issue in AN studies is the heterogeneity of the source material available: how far is the pattern of usage in the texts analysed here general and typical of insular French? It might be thought that a respect for changing continental usage would have affected chiefly official users of AN, who - perhaps by their position at the seat of government, with its institutional links with France, or perhaps by special training - might have been in closer contact with current linguistic developments on the European mainland than other users. Now it is true that clitic forms preceding the infinitive in non-finite clauses, the new development in the 1320-62 subperiod, were found chiefly in administrative texts, such as the responses to parliamentary petitions. However, a shift from the postverbal to the preverbal pattern in non-finite clauses can also be seen in AN correspondence sent from religious houses, and in other authors, as illustrated by the contrast between (32a-d) and (33a-d): (32a) Metez peine d’estorer la Walter c. 31 (c. 1285) (32b) . . . as chasteaus le rei garder et sustenir les Burt. 479, 6 (1259) (32c) Len a mester de prendre le Sen c. 41 (c. 1285) (32d) . . . pour eslire les draps et carier les a Cantorberis LC I 40 (1318) (33a) . . . de les retenire LC II 182 (1338) (33b) Nous fumes la prest . . . de les recevire LC II 228 (1340) (33c) . . . li meisme de la tenier LC II 274 (1344) (33d) . . . de ensi le faire Lanc. p. 29 (c. 1354) The new pattern illustrated by (33) was thus not restricted to the work of government clerks, and achieved some diffusion within AN. That said, it undoubtedly remains desirable, in future work on later AN, especially post-1320, to establish a much richer evidential database than we have been able to do in this initial study, and one representing central government and regional users of the language in a more balanced way 16 . 16 Only the Council’s responses to the pleas were used for this investigation, since it was not known whether the pleas as enrolled were written in the petitioner’s locality, or re-written by central government scribes. The status of French in Medieval England 103 We have so far emphasised the targetlike performance of AN writers analysed, yet some small incidence of tonic pronouns used in pre-finite position did occur (see Table 3). This is well-known as a later AN feature, and was identified as nontargetlike by contemporary authorities. Here, the morphosyntax of AN was undoubtedly divergent from the continental norm; the question is how to interpret this low level of non-targetlike performance. It is not clear that the source of the error was confusion over the grammatical distinction between tonic and clitic forms per se. Where the tonic and clitic forms were phonologically distinct, as with les/ eux, no such divergent uses of the tonic form ever occurred. The source of the errors may thus have been that spoken Anglo-Norman was tending to merge the tonic and clitic forms with e. g. le and luy, me and moy (Pope 1934), so that the writer’s task in such cases amounted to maintaining a syntactic distinction between different spelling forms, rather than preserving a true morphological distinction. Even measured in this fashion, AN writers displayed a very high degree of success: The (graphemic) clitic form remained overwhelmingly the majority variant in finite auxiliated clauses (Table 3). Even the low level of errors noted above was entirely absent in an area where grammatical confusion could have arisen, i. e. between the syntactic position of the object pronouns in auxiliated clauses and in non-finite clauses, both of which featured a non-finite verb form. Yet, albeit that usage was tending towards the Obj Pro - V inf order in nonfinite clauses, no overgeneralisation of this pattern to finite auxiliated clauses was observed, along the lines of *il poet le/ lui veir. It is surely significant that where analogical errors independent of phonological form were conceivable, as here, they did not occur. The procedure adopted in this study was to investigate a single rather tightly circumscribed syntactic area, and our findings on this point naturally cannot provide a definitive account of the linguistic status of later Anglo-Norman. The variable chosen was nevertheless, we feel, a revealing one. Abstract syntax, such as clitic pronoun positioning, is generally considered to be below the level of first language metalinguistic awareness and a source of difficulty to second language learners; it is therefore, we believe, a valid diagnostic of the unmonitored syntactic competence of language users, such as the AN writers sampled here. Nativelike performance, which is largely the result we obtained on the target variable, even in the later period, is strong evidence of non-divergence from the continental mainstream, at least on the basis of the sample. The practitioners of AN using this highly targetlike grammar of object pronouns seem to have participated in that linguistic mainstream, and closely followed its evolution. How they acquired this near-nativelike, if not nativelike, competence is an intriguing issue. The existence in medieval England of pedagogical mentors linguistically equipped to formulate and teach the complex rules of Old French object pronouns may reasonably be doubted: certainly their quite inadequate formulation in the French version of the Orthographica Gallica, as mentioned in footnote 4, is not encouraging in this regard. Richard Ingham 104 This being so, it is all the more curious that later AN is prone to what appear to be frank elementary grammatical errors, e. g. with noun gender or verb conjugation, as has often been noted. How this state of affairs can have comported with the high degree of syntactic accuracy displayed in the data we have analysed is an issue that goes well beyond the confines of this study, but one which we feel will be important for a better understanding of the nature and transmission of French in later medieval England. It may well be (Legge/ Holdsworth 1934) that they largely concern phonologically determined traits and perhaps thus indirectly testify to its transmission as a spoken variety, though in circumstances rather different from those in which L1 English was passed on in medieval England. 7. Conclusion The overall implications of this study may now be stated. The results sit well with a view of AN in which it maintained close contacts with continental models of French, perhaps by means of personal travel and correspondence with continental users of French, and by new infusions of native speakers as language tutors. If AN had been a linguistic backwater, we would not have expected the French users sampled here to shadow so closely the linguistic usage of the continent. In the domain studied, the syntax of object pronouns, it is particularly striking that no effect of English influence is perceptible, even in the latest subperiod investigated (1320-62). This suggests that conceptualising the 14 th century as that of the onset of the «final decline» of French in England (cf. Berndt 1972) should not be seen in qualitative terms, pertaining to the level of French proficiency of its practitioners, but perhaps more as a matter of receding use. Even then, for much of the 14 th century there seems to be evidence for increasing use of French.As pointed out by Lusignan (2004: 216-17), it was during 14 th century that French had the most impact on the linguistic life of England, prompting the question as to whether we schould not be at least as much concerned with its status at this time, when the attractiveness of French was at its greatest, as with the question of when French ceased to be spoken in England as a native language. However this may be, those who continued to employ French in England in the 14 th century were capable, it seems, of mastering an abstract syntactic system as it underwent change on the continent, and accurately reproducing its formal exponents in just the appropriate contexts. At the same time those exponents, such as the morphological forms of pronouns, had been undergoing considerable phonological change within the insular context, and this may explain why writers in French on occasion used some, though not all, tonic forms in clitic contexts. Such an outcome suggests that directions for research into the transmission of AN should consider ways in which this intriguing combination of local change and adherence to a supralocal linguistic mainstream interacted. Birmingham Richard Ingham The status of French in Medieval England 105 Key to primary sources Middle English AW: Tolkien, J. (ed.) 1962: Ancrene Wisse, ed. from Corpus ms. Cambs, London PC: Clark, C. (ed.) 1958: Peterborough Chronicle 1070-1154, London Hali Meid.: Furnivall, F. (ed.) 1922: Hali Meidenhad, rev. ed., London Continental French Charles V: Delisle, L. 1874: Mandements et actes divers de Charles V recueillis dans les collections de la Bibliothèque Nationale, Paris F-B: Funck-Brentano, F. 1896: Philippe le Bel en Flandres, Paris Pir.: Pirenne, H. 1900: Le soulèvement de la Flandre maritime de 1323-28. Documents inédits publiés avec une introduction, Bruxelles B: Beugnot, A. (ed.) 1842: Les coutumes du Beauvoisis de Philippe de Beaumanoir, Paris Law French SCEC: Hemmant, M. (ed.) 1933: Select Cases in the Exchequer Chamber 1377-1461, London AN verse CB: Macbain, W. (ed.) 1964: Clemence of Barking, The Life of St. Catherine, Oxford Ip.: Holden, A. J. (ed.) 1979: Hue de Rotelande, Ipomedon, Paris JF: Johnston, R. (ed.) 1981: Jordan Fantosme’s Chronicle, Oxford Ad.: Studer, P. (ed.) 1918: Le Mystère d’Adam, Manchester Pet. Pl.: Merrilees, B. S. (ed.) 1970: Le Petit Plet, London Prot.: Holden, A. J. (ed.) 1991-93: Hue de Rotelande, Protheselaus, 3 vol., London SR: Jenkins, T. A. et al. (ed.) 1943: La Seinte Resureccion, London AN Prose Bozon: Toulmin Smith, L./ Meyer, P. (ed.) 1889: Les Contes moralisés de Nicole Bozon, Paris Brev. Plac.: Turnerand, G. J./ Plucknett, T. F. T. (ed.) 1951: Brevia Placitata, London Britton: Nichols, F. M. (ed.) 1865: Britton. The French Text Carefully Revised with an English Translation, Introduction and Notes, 2 vol., Oxford Burt.: Luard, H. (ed.) 1864: Annales Monasterii de Burton, 1004-1263, London CCL: Chapple, G. 1938: Correspondence of the City of London, unpublished University of London PhD thesis H. III: Shirley, W. (ed.) 1866: Royal and other historical letters illustrative of the reign of Henry III, vol. 3, 1236-72, London Holk.: Pickering, F. (ed.) 1971: The Anglo-Norman text of the Holkham Bible picture book, London Jerarch.: Legge, M. D. 1942: «John Pecham’s Jerarchie», MAe 11: 77-84 Lanc.: Arnould, E. (ed.) 1940: Le livre de seyntz medicines. The unpublished devotional treatise of Henry of Lancaster’, Oxford [Book I only] LC: Sheppard, J. B. (ed.) 1887-89: Literae Cantuarienses. The Letter Books of the Monastery of Christ Church, Canterbury, 3 vol., London LRE: Glover, J. (ed.) 1865: Livere des reis de Engleterre, London NMP: Fraser, C. M. (ed.) 1966: Ancient Petitions Relating to Northumberland, Durham NR: Raine, J. (ed.) 1873: Historical papers and letters from the northern registers, London OakBk. Sth.: Studer, P. (ed.) 1910-11: Oak Book of Southampton, Southampton Peck.: Martin, C. (ed.) 1885: Registrum epistolarum fratris J. Peckham, vols. 1-3, [s. l.] Plac. Cor.: Kaye, J. (ed.) 1966: Placita corone or la corone pledee devant justices, London Richard Ingham 106 Rot. Parl.: Record Commission 1767-77: Rotuli Parliamentorum, vol. 1-4, London Rot. Parl. Ined.: Richardson, H. G./ Sayles, G. O. (ed.) 1935: Rotuli Parliamentorum Anglie hactenus inediti, 1272-1373, London RP 1305: Maitland, F. (ed.) 1893: Records of Parliament of 1305 holden at Westminster, London Walter: Oschinsky, D. 1971: Walter of Henley and other Treatises on Estate Management and Accounting, Oxford Wenl.: Harvey, B. (ed.) 1965: Documents illustrating the rule of Walter de Wenlok, abbot of Westminster 1283-1307, London References Berndt, R. 1972: «The period of the final decline of French in Medieval England (14 th and early 15 th centuries)», Zeitschrift für Anglistik und Amerikanistik 20: 341-69 Boxtel, S./ van Bongaerts, T./ Coppen, P.-A. 2005: «Native-like attainment of dummy subects in Dutch and the role of the L1», International Review of Applied Linguistics 43/ 4: 355-80 Brereton, G. 1939: «Some grammatical changes made by two revisers of the Anglo-Norman version of Des Grantz Geantz», in: Studies in French language and mediæval literature presented to Mildred K. Pope by pupils, colleagues and friends, Manchester: 21-28 Buridant, C. 2000: Nouvelle Grammaire de l’ancien Français, Paris Chapple, G. 1938: «Correspondence of the City of London 1298-1370», Unpublished University of London PhD thesis Clanchy, M. 1993: From memory to written record. England 1066-1307, Cambridge (Mass.) Foulet, L. 1930: Petits syntaxe de l’ancien français, Paris Gundel, J./ Tarone, E. 1983: «Language transfer and the acquisition of pronominal anaphora», in: S. Gass/ L. Selinker (ed.), Language transfer in language learning, Rowley (Mass.): 281-96 Hirschbühler, P./ Labelle, M. 2001: «La position des clitiques par rapport au verbe à l’imperatif dans l’évolution du français», Recherches Linguistiques de Vincennes 30: 13-37 Johnston, R. C. 1987: Orthographia Gallica, London Kibbee, D. 1991: For to speke Frenche trewely. The French language in England 1000-1600: its status, description and instruction. Amsterdam Kibbee, D. 1996: «Emigrant languages and acculturation: the case of Anglo-French», in: H. Nielsen/ L. Schoesler (ed.), The origins and development of emigrant languages. RASK supplement No. 6, Odense: 1-20 Kibbee, D. 2000: «Historical perspectives on the place of Anglo-Norman in the history of the French language», French Studies 54: 137-53 Labelle, M./ Hirschbühler, P. 2005: «Changes in the clausal organization and the position of clitics in Old French», in: M. Batllori/ M.-L. Hernanz/ C. Picallo/ F. Roca (ed.), Grammaticalization and Parametric Change, Oxford Legge, M. D. 1950: «The French language and the English cloister», in: Ruffer, V./ Taylor, A. J. (ed.): Medieval Studies presented to Rose Graham, Oxford: 146-62 Legge, M. 1980: «Anglo-Norman as a spoken language», Anglo-Norman Studies 2: 108-17 Legge, M./ Holdsworth, W. 1934: Year Books 10 Edw. II 1316-17. London Lusignan, S. 2004: La langue des rois au Moyen Âge. Le français en France et en Angleterre, Paris Marchello-Nizia, Chr. 1997: La langue française au XIV e et au XV e siècles. Paris Ormrod, M. 2003: «The use of English: language, law and political culture in fourteenth-century England», Speculum 78: 750-87 Paradis, J./ Crago, M./ Genesee, F. 2005/ 6: «Evidence from bilingual children’s acquisition of object pronouns», Language Acquisition, 13/ 1: 33-62 Pope, M. 1934: From Latin to Modern French, with especial consideration of Anglo-Norman: phonology and morphology, Manchester Richter, M. 1979: Sprache und Gesellschaft im Mittelalter. Untersuchungen zur mündlichen Kommunikation in England von der Mitte des elften bis zum Beginn des vierzehnten Jahrhunderts, Stuttgart The status of French in Medieval England 107 Rothwell, W. 1976: «The role of French in thirteenth-century England», Bulletin of the John Rylands University Library of Manchester 58: 445-66 Rothwell, W. 1996: «Playing follow my leader in Anglo-Norman studies», Journal of French Language Studies 6: 177-210 Sorace, A. 2003: «Near-nativeness», in: C. Doughty/ M. Long (eds.), Handbook of Second Language Acquisition, Oxford: 130-51 Suggett, H. 1946: «The use of French in England in the later Middle Ages», Transactions of the Royal Historical Society 28: 61-83 Trotter, D. 2003a: «Not as eccentric as it looks: Anglo-French and French French», Forum for Modern Language Studies 39: 427-38 Trotter, D. 2003b: «L’anglo-normand: variété insulaire ou variété isolée? », in: Grammaires du vulgaire. Médiévales 45, Paris: 43-54 Vance, B. 1997: Syntactic change in medieval French, Dordrecht Völker, H. 2003: Skripta und Variation. Untersuchungen zur Negation und zur Substantivflexion in altfranzösischen Urkunden der Grafschaft Luxemburg (1237-81), Tübingen. White, L. 1996: «Clitics in L2 French», in: H. Clahsen (ed.), Generative perspectives on language acquisition, Amsterdam: 335-68 Wilshere, A. (1993): «A plea for syntax», in: I. Short (ed.), Anglo-Norman anniversary essays, London: 395-404 Wilson, R. (1943): «English and French in England 1100-1300», History 28: 37-60