languages Article
Acquisition of L2 French Object Pronouns by Advanced Anglophone Learners Julia Herschensohn 1, * and Randall Gess 2,3 1 2 3
*
Department of Linguistics, University of Washington, Box 352425, Seattle, WA 98195-2425, USA Department of French, Carleton University, 1125 Colonel By Drive, Ottawa, ON K1S 5B6, Canada;
[email protected] School of Linguistics and Language Studies, Carleton University, 1125 Colonel By Drive, Ottawa, ON K1S 5B6, Canada Correspondence:
[email protected]; Tel.: +1-206-543-2046
Received: 22 December 2017; Accepted: 19 April 2018; Published: 3 May 2018
Abstract: The role native language transfer plays in L2 acquisition raises the question of whether L1 constitutes a permanent representational deficit to mastery of the L2 morphosyntax and prosody or if it can eventually be overcome. Earlier research has shown that beginning and low intermediate Anglophone L2 French learners are insensitive to French morphosyntactic and prosodic constraints in using in situ pronouns transferred from the L1. The prosodic transfer hypothesis (PTH) proposes that native prosodic structures may be adapted to facilitate acquisition of L2 prosodic structure. Our study presents new evidence from three Anglophone advanced learners of L2 French that indicates ceiling performance for pronoun production (99% accuracy in 300 tokens over nine interviews) and grammaticality judgment (98% accuracy). This native-like performance demonstrates target French morphosyntax and prosody, built—as predicted by the PTH—by licensing pronominal free clitics in a new pre-verbal L2 position distinct from post-verbal L1. Furthermore, the learners’ data confirms accurate prosody by way of appropriate prominence patterns in clitic + host sequences, correct use of clitics with prefixed verbs, use of stacked pronouns, as well as correct prosodic alternations involving liaison and elision. These results counter impaired representation approaches and suggest early missing inflection may be overcome. Keywords: second language acquisition; native language transfer; French object pronouns; advanced learners; prosodic transfer hypothesis; French object clitics; strong pronouns
1. Introduction The role of the native language in L2 acquisition has been a topic of inquiry for decades (e.g., Lado 1957; Schwartz and Sprouse 1996) and is of particular interest in cases where the two languages differ in several dimensions. This notion of transfer raises the question of whether the native language constitutes a permanent representational deficit to mastery of the new morphosyntax (Hawkins and Chan 1997; Tsimpli and Dimitrakopoulou 2007) or if it can eventually be overcome (Lardiere 2007; Rothman 2011). In the arena of prosody, one proposal concerning native language influence is the prosodic transfer hypothesis (PTH, Goad and White 2006, 2008, 2009), which argues that learners’ native language prosodic structures may be used to accommodate L2 functional material. The acquisition of French pronouns is a challenge for Anglophone second language (L2) learners, since the two languages differ morphosyntactically and prosodically for object pronouns.
Languages 2018, 3, 15; doi:10.3390/languages3020015
www.mdpi.com/journal/languages
Languages 2018, 3, 15 Languages 2018, 3, 15
1.
2.
a.
I see John
b.
I see him
c.
*I him see
d.
I see’m
a.
Je vois Jean
b.
Je le vois
c.
*Je vois le
2 of 21 2 of 20
termsofofmorphosyntax, morphosyntax, English pronouns located in the canonical object position InInterms English pronouns areare located in the canonical object position of theof the full noun phrase (1a, b) and may optionally be right-cliticized to the verb (1d). In contrast, full noun phrase (1a, b) and may optionally be right-cliticized to the verb (1d). In contrast, French French object pronoun clitics are left-adjoined to the inflected verb notcanonical in the canonical object pronoun clitics are left-adjoined to the inflected verb (2b), and(2b), are and not are in the direct direct object position (2a, c). The acquisition of L2 Romance pronouns has been a subject object position (2a, c). The acquisition of L2 Romance pronouns has been a subject of investigation toof investigation to test morphosyntactic transfer hypotheses (White 1996; and Franceschina 2004; test morphosyntactic transfer hypotheses (White 1996; Hawkins andHawkins Franceschina 2004; Granfeldt Granfeldt and Schlyter Arche and Domínguez 2011)that in studies that mastery of clitic and Schlyter 2004; Arche2004; and Domínguez 2011) in studies indicatethat thatindicate mastery of clitic pronouns 1 Beginning learners of L2 French clearly 1 pronouns varies substantially according to proficiency level. varies substantially according to proficiency level. Beginning learners of L2 French clearly transfer transferstructures English structures in producing in situ pronouns in (2c) (Herschensohn 2000; Hawkins 2001), English in producing in situ pronouns as in (2c)as(Herschensohn 2000; Hawkins 2001), while while intermediate proficiency learners transitional responses, behavioral andcortical, cortical,toto intermediate proficiency learners showshow transitional responses, bothboth behavioral and ungrammatical in situ German et al.et2015). Little Little research has been done on done advanced ungrammatical situpronouns pronouns(Sneed (Sneed German al. 2015). research has been on 2 learners of L2 French with respect torespect object clitic pronouns. advanced learners of L2 French with to object clitic pronouns. 2 Norhave haveobject objectclitics cliticsbeen beenexamined examinedwith withrespect respecttotothe theprosodic prosodictransfer transferhypothesis. hypothesis.InInterms terms Nor of prosodic structure, English nouns and pronouns and French nouns constitute independent prosodic of prosodic structure, English nouns and pronouns and French nouns constitute independent prosodic words,whereas whereasFrench Frenchobject objectpronouns pronounsconstitute constitutefree freeclitics cliticsattached attachedtotothe theleft leftofofthe theverb, verb,so sothat that words, the prosodic structure of (2b) is as shown in (3c), where the clitic is attached as a sister to the prosodic the prosodic structure of (2b) is as shown in (3c), where the clitic is attached as a sister to the prosodic word(PWd) (PWd)and andaadaughter daughtertotothe theprosodic prosodicphrase phrase(PPh) (PPh)(Goad (Goadand andBuckley Buckley2006). 2006).English Englishpronouns, pronouns, word normallyindependent independentprosodic prosodicwords, words,may mayalso alsocliticize, cliticize,but butasasright-branching right-branchingaffixal affixalclitics, clitics,so sothat that normally the prosodic structures of (1a/b) and (1d) are as shown in (3a) and (3b) (Selkirk 1996). the prosodic structures of (1a/b) and (1d) are as shown in (3a) and (3b) (Selkirk 1996).
1
2
1
2
Even early bilinguals may show variability. Perpiñán (2017) describes non target-like responses by Spanish-dominant bilinguals for Catalan clitics non-existent in Spanish, indicating a divergent Catalan grammar despite very early exposure (before age four) to Catalan. Hoover and Dwivedi (1998) and Duffield et al. (2002) study the use of clitics (in reading and sentence matching tasks, respectively) in French students that they label as advanced.
Even early bilinguals may show variability. Perpiñán (2017) describes non target-like responses by Spanishdominant bilinguals for Catalan clitics non-existent in Spanish, indicating a divergent Catalan grammar despite very early exposure (before age four) to Catalan. Hoover and Dwivedi (1998) and Duffield et al. (2002) study the use of clitics (in reading and sentence matching tasks, respectively) in French students that they label as advanced.
Languages Languages2018, 2018,3,3,1515 Languages 2018, 3, 15
3. 3.
English English a. a.
3 3ofof2120 3 of 21
PPh PPh
PWd PWd see see
b. b.
PWd PWd John/him John/him PWd PWd
PWd PWd see see
‘m ‘m
French French c. c.
The French clitic is both directionally (left, not right) and prosodically (free, not affixed) distinct The French clitic is both directionally (left, not right) and prosodically (free, not affixed) distinct is both directionally not right) and prosodically (free, not languages affixed) distinct fromThe the French Englishclitic pronominal clitic. In their(left, nominal morphosyntax, however, both share from the English pronominal clitic. In their nominal morphosyntax, however, both languages share a from the English pronominal clitic. In their nominal morphosyntax, however, both languages share a similar prosodic free clitic structure, that of the leftward article plus noun as in (4). similar prosodic free clitic structure, that of the leftward article plus noun as in (4). a similar prosodic free clitic structure, that of the leftward article plus noun as in (4). 4. French/English 4. French/English
PPh PPh PWd PWd le livre le livre the book the book The availability of this structure is interesting, given the PTH’s proposal that target L2 The availability of this structure is interesting, given the PTH’s proposal that target L2 prosody prosody can be built when L1 structures are licensed in new positions (“minimal adaptation”, The availability of this structure is interesting, given the(“minimal PTH’s proposal that target prosody can be built when L1 structures are licensed in new positions adaptation”, GoadL2 and White Goad and White 2004). can be built when L1 structures are licensed in new positions (“minimal adaptation”, Goad and White 2004). In this article, we explore implications of the PTH for Anglophone acquisition of L2 French clitics 2004).In this article, we explore implications of the PTH for Anglophone acquisition of L2 French clitics by examining new data from advanced learners with extensive experience in the target environment. In this article, explore implications of thewith PTHextensive for Anglophone acquisition of L2 environment. French clitics by examining new we data from advanced learners experience in the target The first background section describes French and English pronouns (Section 2.1), prosodic structure by new data from describes advancedFrench learnersand with extensive experience in the environment. Theexamining first background section English pronouns (Section 2.1),target prosodic structure and the prosodic transfer hypothesis (Section 2.2), and previous research on the L2 acquisition of French The first background section describes French and English pronouns (Section 2.1), prosodic structure and the prosodic transfer hypothesis (Section 2.2), and previous research on the L2 acquisition of French clitics (Section 2.3), which leads to our research questions for advanced learners (Section 2.4). The next and the(Section prosodic transfer (Section 2.2), and previous research learners on the L2(Section acquisition French clitics 2.3), whichhypothesis leads to our research questions for advanced 2.4).of The next sections present new evidence from advanced learners of L2 French (participants and methodology clitics (Section 2.3), which leads from to ouradvanced research questions learners (Section The next sections present new evidence learners offorL2advanced French (participants and 2.4). methodology described in Sections 3.1 and 3.2, respectively), coming from three case studies that include data on sections new evidence from advanced learners L2 three Frenchcase (participants and methodology describedpresent in Sections 3.1 and 3.2, respectively), coming of from studies that include data on production and grammaticality judgments (results presented in Section 3.3). We discuss the new described Sections 3.1 and 3.2,judgments respectively), coming from three case studies thatdiscuss include the datanew on productioninand grammaticality (results presented in Section 3.3). We evidence in light of the PTH (Section 4), arguing that the data support the hypothesis in demonstrating production and grammaticality judgments (results presented in Section 3.3). We discuss the new evidence in light of the PTH (Section 4), arguing that the data support the hypothesis in demonstrating the learners’ shift to L2 French prosody. Our analysis adopts the concept of minimal adaptation of evidence in light theL2PTH (Section 4), arguing that the adopts data support the hypothesis in demonstrating the learners’ shiftofto French prosody. Our analysis the concept of minimal adaptation of native prosodic structures to accommodate L2 functional material. We close with some brief concluding the learners’ shift to L2 French prosody. Our analysis adopts theWe concept of minimal adaptation of native prosodic structures to accommodate L2 functional material. close with some brief concluding remarks (Section 5). native prosodic structures to accommodate L2 functional material. We close with some brief concluding remarks (Section 5). remarks (Section 5). 2. Background 2. Background 2.1. French and English Object Pronouns 2.1. French and English Object Pronouns
Languages 2018, 3, 15
4 of 20
2. Background 2.1. French and English Object Pronouns Cross-linguistically, pronouns fall into strong, weak and clitic classes (Cardinaletti and Starke 1999). Strong pronouns project the most complex syntactic, semantic and phonological structure, followed by weak pronouns and finally by clitics.3 French pronouns include parallel sets of strong and clitic pronouns in six persons, while English pronouns are strong unless cliticized to the verb as in (1d) Languages 2018, 3, 15 4 of 19 (Table 1).4 and clitic pronouns in six persons, while English pronouns are strong unless cliticized to the verb as Table 1. Pronouns of English and French. in (1d) (Table 1). 4 1st s.
2nd s. of English and 3rdFrench. s. 1st p. Table 1. Pronouns French 2nd s. 3rd s. 1st p. 2nd p. moi toi vous lui, elle nous je tu vous il, elle, on nous toi vous lui, elle nous vous me te vous le, la nous tu vous il, elle, on nous vous me te vous lui nous te vous le, la nous vous te vous lui English nous vous I you you he, she, it we me us you you youyou he,you she, it him, her, itwe you you him, her, it us you
1st s. FrenchStrong Cl nom. Strong moi Cl acc. Cl nom. je Cl dat. Cl acc. me Cl dat. me Stg nom. English Stg obj. I Stg nom. Stg obj. me
2nd p.
3rd p.
3rd p. vous eux, elles vous ils, elles eux, elles vous les ils, elles vous leur les leur you they you they them them
Strong pronouns in French and English are phonetically free standing and fully referential, Strong pronouns in French and English are phonetically free standing and fully referential, whereas clitics in French are morphosemantically (person, number whereas clitics in French are morphosemantically minimalminimal (person, gender andgender number and features) and features) and structurally deficient, requiring phonological attachment a host. Strong pronouns structurally deficient, requiring phonological attachment to a host.toStrong pronouns act like fullact like full determinerphrases phrases (DPs) in French in they that are they are syntactically mobile and phonologically determiner (DPs) in French in that syntactically mobile and phonologically stressed. stressed. They, pronouns, cancan appear in coordination (5), in (5), isolation (6) and in They, but butnot notclitic clitic pronouns, appear in coordination in isolation (6)prepositional and in prepositional phrases areare quite frequent in left dislocation (8) and(8) areand usually human. human. phrases(7). (7).They They quite frequent in and left right and right dislocation are animate/ usually animate/ 5.
Elle et le prochain, je les ai écoutés. *La et le prochain, je les ai écoutés. ‘Her and the next one, I listened to them.’
6.
Qui a-t-il aidé?
‘Who did he help?’ Lui/Toi/Elles ‘him/you/them.’ *Le/*Te/*Les ‘him/you/them.’ 7.
Toi
seul
va
au cínema
avec lui.
*Tu seul avec le. va au cínema ‘You alone go to the movies with him.’ 8.
Moi, je trouve qu’il est stupide *Me, je trouve qu’il est stupide ‘Me, I think he’s stupid.’
In contrast, clitic pronouns are unstressed, may not remain in situ, and refer to both animate and inanimate referents. Roberts (2010, p. 41) analyzes French clitics as “simultaneoulsy maximal and 3 minimal elements”, bundles featuresfor(number, gender, person, case)beginning in the head D of(1975). the See recent Romance clitic syntax has beenofdiscussed decades within generative theory, with Kayne summaries by Roberts (2010) they and Grüter Generally, there are two approaches, one group clitics as agreement Determiner Phrase in which occur,(2008). whose defective nature requires cliticization totreating an inflected on the verb (Grüter 2008; Culbertson 2010), and the other arguing that clitics must move from within the Verb verbmarkers or infinitive. English strong pronouns resemble French ones in syntactic distribution and stress, Phrase (Kayne 1975; Roberts 2010; Cardinaletti and Starke 1999; De Cat 2005). as translations differ English and French, however, in terms of a loro ‘we 4 the Our focus here isindicate. on strongThe andclitic clitic forms pronouns thatbetween are relevant for French and English. In Italian in Telefono placement, morphological variation, feature specification, as well as prosody, as outlined in the next phone them’ the pronoun is strong, in Telefono loro it is weak, and in Gli telefono it is a clitic (Ordoñez and Repetti 2014). section. To be able to use pronouns, Anglophone learners need to master the implicit knowledge that French object pronouns must be placed in non-canonical position to the left of the verb; that strong 4
Our focus here is on strong and clitic pronouns that are relevant for French and English. In Italian in Telefono a loro ‘we phone them’ the pronoun is strong, in Telefono loro it is weak, and in Gli telefono it is a clitic (Ordoñez
Languages 2018, 3, 15
5 of 20
In contrast, clitic pronouns are unstressed, may not remain in situ, and refer to both animate and inanimate referents. Roberts (2010, p. 41) analyzes French clitics as “simultaneoulsy maximal and minimal elements”, bundles of features (number, gender, person, case) in the head D of the Determiner Phrase in which they occur, whose defective nature requires cliticization to an inflected verb or infinitive. English strong pronouns resemble French ones in syntactic distribution and stress, as the translations indicate. The clitic forms differ between English and French, however, in terms of placement, morphological variation, feature specification, as well as prosody, as outlined in the Languages 2018, 3, 15 5 of 19 Languages 2018, 3, 15 5 of 19 next section. To be able to use pronouns, Anglophone learners need to master the implicit knowledge that and weak pronouns are in syntactic complementary distribution (nominative and objective and weak pronouns in be syntactic distribution andthat objective French object pronounsare must placed incomplementary non-canonical position to the(nominative left of the verb; strong obligatory clitic); that French pronouns are marked for person, number and gender (as English), but obligatory clitic); that French pronouns are marked for person, number and gender (as English), but and weak pronouns are in syntactic complementary distribution (nominative and objective obligatory that gender is grammatical, not semantic (and thus applies to every inanimate noun as well as that gender is grammatical, not semantic (and thus applies to every inanimate noun as well as clitic); that French pronouns are marked for person, number and gender (as English), but that gender animate). Learners also need to prosodically integrate object pronouns in a way compatible with L2 animate). Learners need (and to prosodically pronouns wayascompatible with L2 is grammatical, not also semantic thus appliesintegrate to everyobject inanimate noun in as awell animate). Learners target structures. target structures. also need to prosodically integrate object pronouns in a way compatible with L2 target structures.
2.2. Prosodic Structure 2.2. 2.2. Prosodic Prosodic Structure Structure Which prosodic representations accommodate object pronouns in Which in English English and and French? French? English English Which prosodic prosodic representations representations accommodate accommodate object object pronouns pronouns in English and French? English object pronouns are typically realized as full prosodic words (9a), but may cliticize as right-branching object object pronouns pronouns are are typically typically realized realized as as full full prosodic prosodic words words (9a), (9a), but but may may cliticize cliticize as as right-branching right-branching affixal clitics (9b) affixal (9b) (Selkirk (Selkirk 1996). 1996). affixal clitics clitics (9b) (Selkirk 1996). When an object pronoun in English is realized as aa full prosodic word (9a), it attaches as a PWd When an object pronoun When an object pronoun in in English English is is realized realized as as a full full prosodic prosodic word word (9a), (9a), it it attaches attaches as as aa PWd PWd sister to the verb, itself a PWd, with both of these linking to aa phonological phrase node. A different sister sister to to the the verb, verb, itself itself aa PWd, PWd, with with both both of of these these linking linking to to a phonological phonological phrase phrase node. node. A A different different structure is projected under phonological reduction In this case (9b), failing to project structure reduction of of the the pronoun. pronoun. structure is is projected projected under under phonological phonological reduction of the pronoun. In In this this case case (9b), (9b), failing failing to to project project its own PWd node, the pronoun is instead incorporated into an embedded prosodic word structure its own PWd node, the pronoun is instead incorporated into an embedded prosodic word structure its own PWd node, the pronoun is instead incorporated into an embedded prosodic word structure as as aa structurally deficient sister to PWd verb. as structurally deficient sister to the the PWd verb. a structurally deficient sister to the PWd verb. 9. 9. Prosodic Prosodic structure structure of of English English object object pronouns pronouns a. b. a. b. PPh PPh PPh PPh || PWd PWd PWd PWd PWd PWd see John/him see John/him PWd PWd see see
‘m ‘m
In In French, French, on on the the other other hand, hand, (as (as for forall all“morphologically-free “morphologically-free functional functional items items [in [in French] French] which which In French, on the other hand, (as for all “morphologically-free functional items [in French] which precede their host” (Goad and Buckley 2006, p. 114)), object pronouns are prosodically left-branching precede their host” (Goad and Buckley 2006, p. 114)), object pronouns are prosodically left-branching precede their host” (Goad and Buckley 2006, p. 114)), object pronouns are prosodically left-branching free free clitics clitics (10). (10). free clitics (10). 10. 10.
Prosodic Prosodic structure structure of of French French object object pronouns pronouns PPh PPh
PWd PWd vois vois
le le
The free clitic—which can be separated from its host by other elements—differs from its affixal The clitic—which can separated host other from its counterpart its direct relation the PPh:from “theits function sister to PWd and daughter to The free freeby clitic—which can be be to separated from its host by byword other iselements—differs elements—differs from its affixal affixal counterpart by its direct relation to the PPh: “the function word is sister to PWd and daughter to phonologicalbyphrase” (Selkirk 1996). ThePPh: English clitic cannot be separated its host (11), counterpart its direct relation to the “theaffixal function word is sister to PWd from and daughter to phonological phrase” (Selkirk 1996). The English affixal clitic be separated from host whereas the French free clitic can, as we in examples clitic in Tense Phrases (12). phonological phrase” (Selkirk 1996). Thesee English affixal involving clitic cannot cannot bestacking separated from its its host (11), (11), whereas whereas the the French French free free clitic clitic can, can, as as we we see see in in examples examples involving involving clitic clitic stacking stacking in in Tense Tense Phrases Phrases (12). (12).
11. 11.
12. 12.
a. a.
II saw’m saw’m
b. b.
*I *I saw’m’n’er saw’m’n’er
[I [I saw saw him him and and her] her]
a. a.
J’ai J’ai
le le cadeau cadeau
II have have
offert offert
offered offered the-M the-M gift gift
le
vois
The free clitic—which can be separated from its host by other elements—differs from its affixal counterpart by its direct relation to the PPh: “the function word is sister to PWd and daughter to Languages 2018, 3, 15 of 20 phonological phrase” (Selkirk 1996). The English affixal clitic cannot be separated from its host 6(11), whereas the French free clitic can, as we see in examples involving clitic stacking in Tense Phrases (12). 11.
12.
a.
I saw’m
b.
*I saw’m’n’er
[I saw him and her]
a.
J’ai
le cadeau
b.
Je
offert
I have offered ‘I offered the
the-M gift
ne l‘ai NEG have ‘I didn’t offer it.’
Languages 2018, 3, 15 Languages 2018,I 3, 15
pas NEG
offert offered
6 of 19 6 of 19
c. c.
Je ne le lui ai pas offert Je ne le lui ai pas offert I NEG it to him have NEG offered I NEG it to him have NEG offered ‘I didn’t offer it to him.’ ‘I didn’t offer it to him.’ InIn(12a) ‘I’ attached the (12a)the thesubject subjectclitic cliticje ‘I’(which (whichelides elidesbefore beforethe thevowel-initial vowel-initialauxiliary) auxiliary)is attachedto the In (12a) the subject clitic jeje‘I’ (which elides before the vowel-initial auxiliary) isisattached totothe inflected verb; in (12b, c) it is separated from the verb by additional pronominal and negative clitics inflected verb; in (12b, c) it is separated from the verb by additional pronominal and negative clitics inflected verb; in (12b, c) it is separated from the verb by additional pronominal and negative clitics that prosodic alternations alternations includeelision elisionofof vowels before following vowels thatcan can intervene. intervene. French French prosodic vowels before following vowels and that can intervene. French prosodic alternationsinclude include elision of vowels before following vowels and liaison of final consonants in function words when they precede vowel-initial content liaison of final consonants in function words when they precede vowel-initial content words/PWds. and liaison of final consonants in function words when they precede vowel-initial content words/PWds. and verbalin structures in which and elision are the PPh, may which The nominalThe andnominal verbal structures which liaison andliaison elision occur areoccur the PPh, which be words/PWds. The nominal and verbal structures in which liaison and elision occur are the PPh, which may be iterated (Buckley 2005). Acquisition of French prosody is partially diagnosed by the learner’s iterated (Buckley 2005). Acquisition of French prosody is partially diagnosed by theby learner’s use of may be iterated (Buckley 2005). Acquisition of French prosody is partially diagnosed the learner’s use of these prosodic alternations. these prosodic alternations. use of these prosodic alternations. With is noteworthy is one Withrespect respectto tothe thefree freeclitic cliticstructure structurein inFrench, French,it noteworthythat thatit onewhich whichoccurs occursin With respect to the free clitic structure in French, ititisisnoteworthy that ititisisone which occurs inin English, albeit in a different context—in article + noun sequences. In other words, English articles are, English,albeit albeitininaadifferent differentcontext—in context—inarticle article++noun nounsequences. sequences.InInother otherwords, words,English Englisharticles articlesare, are, English, like French object pronouns, prosodicallyleft-branching left-branchingfree freeclitics clitics (13). Note that in both English like French object pronouns, prosodically (13). Note that in both English and like French object pronouns, prosodically left-branching free clitics (13). Note that in both English and French, other elements be inserted adjectives) between the free article andhead the French, other elements maymay be inserted (e.g.,(e.g., adjectives) between the free cliticclitic article and the and French, other elements may be inserted (e.g., adjectives) between the free clitic article and the head noun in the DP, just as clitics may be stacked before the verb in the TP. It is well known that nounnoun in theinDP, asjust clitics stacked before the verbthe in the TP.inItthe is well that articles are head thejust DP, as may cliticsbemay be stacked before verb TP. known It is well known that articles arebefore acquired before object clitics inL2A L1A(Prévost and L2A (Prévost 2009; Paradis et al. 2014). acquired object clitics in L1A and 2009; Paradis et al. 2014). articles are acquired before object clitics in L1A and L2A (Prévost 2009; Paradis et al. 2014). 13. Prosodic structure of English articles 13. Prosodic structure of English articles
PPh PPh
the the
PPh PPh
PWd PWd book book book book
Accordingto to the the prosodic et et al. al. 2003; Goad andand White 2006,2006, 2008,2008, 2009), According prosodic transfer transferhypothesis hypothesis(Goad (Goad 2003; Goad White According to the prosodic transfer hypothesis (Goad et al. 2003; Goad and White 2006, 2008, learners have difficulty with L2 prosody distinct from L1. Goad and White (2004, 2006, 2008, 2009) 2009), learners have difficulty with L2 prosody distinct from L1. Goad and White (2004, 2006, 2008, 2009), learners have difficulty with L2 prosody distinct from L1. Goad and White (2004, 2006, 2008, and Goad et al. (2003) study L1 Mandarin and L1 Turkish learners of L2 English to test the PTH with 2009) and Goad et al. (2003) study L1 Mandarin and L1 Turkish learners of L2 English to test the PTH 2009) and Goad et al. (2003) study L1 Mandarin and L1 Turkish learners of L2 English to test the PTH respect to distinctive L2 prosodic structures in the past tense does notdoes marknot tense andtense does with respect to distinctive L2 prosodic structures in the past(Mandarin tense (Mandarin mark with respect to distinctive L2 prosodic structures in the past tense (Mandarin does not mark tense not allow affixal prosody) and articles (neither Mandarin nor Turkish have articles and do not allow and does not allow affixal prosody) and articles (neither Mandarin nor Turkish have articles and do and does not allow affixal prosody) and articles (neither Mandarin nor Turkish have articles and do freeallow affixes-left). The series of series articlesofsuggests that learners accommodate the L2 English not free affixes-left). The articles suggests that learners accommodate the L2 prosodic English not allow free affixes-left). The series of articles suggests that learners accommodate the L2 English structures either by either omission (of tense or adapting L1 prosodic realization. example, prosodic structures by omission (ofarticle) tense or or by article) or byanadapting an L1 prosodic For realization. prosodic structures either by omission (of tense or article) or by adapting an L1 prosodic realization. some of the L1 Turkish participants (Goad and White 2009) used an L1-licit prosodic structure of a For example, some of the L1 Turkish participants (Goad and White 2009) used an L1-licit prosodic For example, some of the L1 Turkish participants (Goad and White 2009) used an L1-licit prosodic stressed article in L2 English when L2 stressing was not This tendency is confirmed structure of a stressed article(even in L2 English (even when L2appropriate). stressing was not appropriate). This structure of a stressed article in L2 English (even when L2 stressing was not appropriate). This by Snape and Kupisch (2010) in their phonetic analysis (using Praat software, www.praat.org) of an L1 tendency is confirmed by Snape and Kupisch (2010) in their phonetic analysis (using Praat software, tendency is confirmed by Snape and Kupisch (2010) in their phonetic analysis (using Praat software, Turkish L2 English learner’s article production in which she overused stressed articles. www.praat.org) of an L1 Turkish L2 English learner’s article production in which she overused www.praat.org) of an L1 Turkish L2 English learner’s article production in which she overused Wearticles. have seen that English pronouns vary in prosodic realization between prosodic words and stressed stressed articles. affixal clitics-right, whereas French consistently pronouns to beprosodic free clitics-left. The We have seen that English pronouns vary in requires prosodicobject realization between words and We have seen that English pronouns vary in prosodic realization between prosodic words and two languages differ, therefore, not only in the directionality of object clitic placement (driven by affixal clitics-right, whereas French consistently requires object pronouns to be free clitics-left. The affixal clitics-right, whereas French consistently requires object pronouns to be free clitics-left. The morphosyntax) but also in their prosodic (affixal vs. free clitics).clitic Theplacement PTH offers(driven a meansby by two languages differ, therefore, not only structure in the directionality of object two languages differ, therefore, not only in the directionality of object clitic placement (driven by morphosyntax) but also in their prosodic structure (affixal vs. free clitics). The PTH offers a means by morphosyntax) but also in their prosodic structure (affixal vs. free clitics). The PTH offers a means by which a distinct new L2 prosodic structure might be built by licensing L1 structures in new positions which a distinct new L2 prosodic structure might be built by licensing L1 structures in new positions (“minimal adaptation”, Goad and White 2004). Goad and White (2009, p. 9) point out two conditions (“minimal adaptation”, Goad and White 2004). Goad and White (2009, p. 9) point out two conditions for minimal adaptation: “(a) when they [structures] can be built through combining pre-existing (L1) for minimal adaptation: “(a) when they [structures] can be built through combining pre-existing (L1) licensing relations; or (b) when they involve L1 structures being licensed in new positions.” Given licensing relations; or (b) when they involve L1 structures being licensed in new positions.” Given
Languages 2018, 3, 15
7 of 20
which a distinct new L2 prosodic structure might be built by licensing L1 structures in new positions (“minimal adaptation”, Goad and White 2004). Goad and White (2009, p. 9) point out two conditions for minimal adaptation: “(a) when they [structures] can be built through combining pre-existing (L1) licensing relations; or (b) when they involve L1 structures being licensed in new positions.” Given the availability to our learners of the left-branching free clitic structure in (13), we assume that minimal adaptation in this case would involve the extension of the L1 free clitic-left structure for nominal DP Languages 2018, 3, 15 7 of 19 morphosyntax, to verbal TP configurations. This process is illustrated in (14).
14.
Minimal adaptation in the L2 acquisition of object clitics PPh PWd book
the
PPh
PPh PWd
PWd vois
le
TP
DP
The possible prosodic transfer illustrated in (14) is similar to the one discussed in Goad and White The possible prosodic transfer (14) is similar to the one discussed in Goadarticle and (2009), which involved some of the illustrated L1 Turkishinparticipants’ inappropriate use of a stressed White (2009), whichininvolved some of except the L1 that Turkish use of awould stressed prosodic structure their L2 English, in theparticipants’ case at handinappropriate the prosodic transfer be 5 article prosodic structure in their L2 English, except that in the case at hand the prosodic transfer facilitative. would be facilitative. 5 2.3. L2A of Clitics 2.3. L2A of section Clitics reviews previous research on the L2 acquisition of French clitic pronouns, and then This looks at investigations ofprevious prosodicresearch structure, English determiners. Over the pastand 40 years, This section reviews onmainly the L2 of acquisition of French clitic pronouns, then there at have been severalofstudies of French pronouns, indicating that beginning learners of L2 looks investigations prosodic structure, mainly of English determiners. Over the past 40 French years, manifest types of pronoun in situ placement of that strong and cliticlearners pronouns use of there havefour been several studies realizations: of French pronouns, indicating beginning of (15), L2 French null pronouns (16),of past participle cliticization (17)placement and correct clitic placement (Herschensohn manifest four types pronoun realizations: in situ of left strong and clitic pronouns (15), use 2000; and Schlyter 2004;(17) Herschensohn 2004; Prévost 2009).6(Herschensohn of nullHawkins pronouns2001; (16),Granfeldt past participle cliticization and correct left clitic placement
2000; Hawkins 2001; Granfeldt and Schlyter 2004; Herschensohn 2004; Prévost 2009). 6 15.
5 6
In situ
We thank a reviewer for pointing out that the syntactic literature on clitics has used the distinction phonological versus a. J’ai vuto the English elle clitic-right(=type je versus l‘ the Frenchai vue) type (cf. Kayne (Herschensohn 2004) syntactic clitics to refer clitic-left 1975; Auger 1994). Herschensohn (2004) reports on Chloe, one of the subjects of the current study, at an earlier stage of her L2 French acquisition, I seen her-STG (I her-CL have seen) at which she makes numerous production and comprehension errors.
‘I saw her.’
b.
16.
Il veut les He wants them-CL ‘He wants them still.’
Null pronoun
a.
T‘ as placé [e] You-SG have placed ‘You put it on the bed.’
b.
J‘ai I have ‘I ate it.’
mangé [e] eaten
encore still
sur le lit on the-M
(= les (= them-CL
veut) wants
(Selinker et al. 1975)
(= tu l’as placé) (= you-SG it have placed)
(Herschensohn 2004)
(= je l’ai mangé) (= I it have eaten)
(Adiv 1984)
This section section reviews reviews previous previous research research on on the the L2 L2 acquisition acquisition of of French French clitic clitic pronouns, pronouns, and and then then This looks at investigations of prosodic structure, mainly of English determiners. Over the past 40 years, looks at investigations of prosodic structure, mainly of English determiners. Over the past 40 years, there have have been been several several studies studies of of French French pronouns, pronouns, indicating indicating that that beginning beginning learners learners of of L2 L2 French French there manifest four types of pronoun realizations: in situ placement of strong and clitic pronouns (15), use manifest four types of pronoun realizations: in situ placement of strong and clitic pronouns (15), use Languages 2018, 3, 15 8 of 20 of null null pronouns pronouns (16), (16), past past participle participle cliticization cliticization (17) (17) and and correct correct left left clitic clitic placement placement (Herschensohn (Herschensohn of 6 2000; Hawkins Hawkins 2001; 2001; Granfeldt Granfeldt and and Schlyter Schlyter 2004; 2004; Herschensohn Herschensohn 2004; 2004; Prévost Prévost 2009). 2009).6 2000; 15.. 15
16. 16.
In situ situ In a. a.
J’ai vu J’ai vu seen II seen ‘I saw saw her.’ her.’ ‘I
b. b.
Il veut les Il veut les He wants them-CL He wants them-CL ‘He wants wants them them still.’ still.’ ‘He
Null pronoun pronoun Null
a. a.
T‘ as placé placé [e] [e] T‘ as You-SG have have placed placed You-SG ‘You put it on the bed.’ ‘You put it on the bed.’
b. b.
J‘ai J‘ai have II have Languages 2018, 3, 15 ‘I ate it.’ it.’ ‘I ate 17.
elle elle her-STG her-STG
(= je je (= (I (I
l‘ l‘ her-CL her-CL
encore encore still still
(= les les (= (= them-CL (= them-CL
sur le le lit lit sur on the-M the-M on
mangé [e] [e] mangé eaten eaten
ai vue) vue) ai have seen) seen) have
veut) veut) wants wants
(Herschensohn 2004) 2004) (Herschensohn
(Selinker et et al. al. 1975) 1975) (Selinker
(= tu tu l’as l’as placé) placé) (= (= you-SG you-SG it it have have placed) placed) (=
(Herschensohn 2004) 2004) (Herschensohn
(= je je l’ai l’ai mangé) mangé) (= (= II it it have have eaten) eaten) (=
(Adiv 1984) 1984) (Adiv
(Herschensohn 2004)
9 of 21
Past participle a.
Vous
avez
la
pris
(= vous l’avez pris)
You-PL
have
it-F
taken
(= you-PL it have taken)
‘You have taken it.’ b.
J’ai
le
vu
(= je l’ai vu)
(Granfeldt and Schlyter it-M seen out that the syntactic (I it haveliterature seen) 5 We thank thankI have reviewer for pointing pointing on clitics clitics has has2004) used the the distinction distinction 5 We aa reviewer for out that the syntactic literature on used ‘I saw it.’ phonological versus syntactic clitics to refer to the English clitic-right type versus the French clitic-left type phonological versus syntactic clitics to refer to the English clitic-right type versus the French clitic-left type (cf. Kayne 1975; Auger 1994). The most prevalent errors in early interlanguage areuse theofuse null pronouns (16)—similar (cf.The Kayne 1975; Auger 1994). most prevalent errors inChloe, early interlanguage are the nullof pronouns that 6 Herschensohn (2004) reports on one ofL1 the subjects of2004; the current current study, at an an (16)—similar earlier stage oftoher her L2 6 Herschensohn (2004) reports on Chloe,asone of the subjects of the study, at earlier stage of L2 to that of children acquiring French an (Hamann Hulk 2004; Paradis et al. 2014)—and of children acquiringatFrench as makes an L1 numerous (Hamannproduction 2004; Hulk 2004; Paradis et errors. al. 2014)—and in situ French acquisition, which she and comprehension acquisition, at which she makes errors. in French situ pronouns (15). Paradis et al. numerous (2014, p. production 170) note and for comprehension L1A “regarding incorrect structures,
pronouns (15). Paradis et al. (2014, p. 170) note for L1A “regarding incorrect structures, over 90% were over 90% were omitted In situ French reflect native transfer of English omitted objects”. In situ objects”. French pronouns reflectpronouns native language transferlanguage of English morphsyntactic morphsyntactic while null objects could represent a well-documented tendency to missing structure, while structure, null objects could represent a well-documented tendency to missing inflection and inflection and missing functional items (Prévost and White 2000; Lardiere 2007; Arche and Domínguez missing functional items (Prévost and White 2000; Lardiere 2007; Arche and Domínguez 2011). As 2011). Asadvance learners in advance in proficiency theymore shownative-like more native-like command of pronouns learners proficiency they show command of pronouns (Arche(Arche and and Domínguez 2011; Cuza et al. 2013; Grüter and Crago 2012; Sneed German et al. 2015). Domínguez 2011; Cuza et al. 2013; Grüter and Crago 2012; Sneed German et al. 2015). Acquisition of of L2 L2 Romance Romance pronouns pronounshas hasbeen beenaasubject subjectof ofinvestigation investigationtototest testmorphosyntactic morphosyntactic Acquisition transfer hypotheses (Granfeldt 2005; Granfeldt and Schlyter 2004; Arche and Domínguez 2011). transfer hypotheses (Granfeldt 2005; Granfeldt and Schlyter 2004; Arche and Domínguez 2011). Impaired representation (Snape et al. assume that adult will havewill persistent Impaired representationapproaches approaches (Snape et 2009) al. 2009) assume that learners adult learners have difficulties mastering grammatical (uninterpretable) features of an L2, and that the syntactic persistent difficulties mastering grammatical (uninterpretable) features of an L2, and thatdeficit the will be represented persistent morphosyntactic In contrast, missing inflection approaches syntactic deficit willbybe represented by persistenterrors. morphosyntactic errors. In contrast, missing (Prévost and White 2000; Lardiere 2007)2000; attribute morphological errors to other problems inflection approaches (Prévost and White Lardiere 2007) attribute morphological errors tosuch otheras processing overload or prosodic difficulties. In any case, evidence from earlier studies indicates problems such as processing overload or prosodic difficulties. In any case, evidence from earlier that mastery varies to proficiency beginning learners of L2learners French of transfer from studies indicates thataccording mastery varies accordinglevel: to proficiency level: beginning L2 French English in producing in situ pronouns, whereas intermediate learners show transitional responses, transfer from English in producing in situ pronouns, whereas intermediate learners show transitional both behavioral cortical,and to ungrammatical in situ pronouns (Sneed German et al.German 2015). et al. responses, both and behavioral cortical, to ungrammatical in situ pronouns (Sneed 2015).Sneed German et al. (2015) use both behavioral measures and event related potentials (ERPs) to investigate comprehension object by twomeasures groups ofand intermediate Anglophone Sneed German et al. (2015)ofuse bothclitics behavioral event related potentials learners (ERPs) toof L2 French and native speakers. Event related potentials, which record electrophysiological investigate comprehension of object clitics by two groups of intermediate Anglophone learnerscortical of L2 activity and usingnative scalp electrodes, have been used to determine reactions language phenomena in real French speakers. Event related potentials, which recordtoelectrophysiological cortical time, with temporal resolutionhave to the millisecond (Friederici et al. 1993; Hagoort and Brown 1999; activity using scalp electrodes, been used to determine reactions to language phenomena in real Osterhout al. 2004, resolution 2006; Steinhauer Lexico-semantic morpho-syntactic anomalies time, with et temporal to the 2014). millisecond (Friedericiand et al. 1993; Hagoort and Brownreliably 1999; Osterhout et al. 2004, 2006; Steinhauer 2014). Lexico-semantic and morpho-syntactic anomalies reliably elicit two distinct native responses, the N400 and the P600 respectively. 7 For example, sentence (18b), would elicit a negative trending electrophysiological wave 400 ms after the event of interest (cream vs. batteries), compared to (18a). 18.
a.
I drink my tea with cream
Sneed German et al. (2015) use both behavioral measures and event related potentials (ERPs) to Sneed German et al. (2015) use both behavioral measures and event related potentials (ERPs) to investigate comprehension of object clitics by two groups of intermediate Anglophone learners of L2 investigate comprehension of object clitics by two groups of intermediate Anglophone learners of L2 French and native speakers. Event related potentials, which record electrophysiological cortical French and native speakers. Event related potentials, which record electrophysiological cortical activity using scalp electrodes, have been used to determine reactions to language phenomena in real activity using scalp electrodes, have been used to determine reactions to language phenomena in real time, with temporal resolution to the millisecond (Friederici et al. 1993; Hagoort and Brown 1999; Languages 2018, 3, 15 9 of 20 time, with temporal resolution to the millisecond (Friederici et al. 1993; Hagoort and Brown 1999; Osterhout et al. 2004, 2006; Steinhauer 2014). Lexico-semantic and morpho-syntactic anomalies Osterhout et al. 2004, 2006; Steinhauer 2014). Lexico-semantic and morpho-syntactic anomalies reliably elicit two distinct native responses, the N400 and the P600 respectively. 77 For example, reliably elicit two native distinct native responses, the the N400 and the P600 7respectively. For example, elicit two distinct responses, thetrending N400 and P600 respectively. For example, (18b), sentence (18b), would elicit a negative electrophysiological wave 400 ms aftersentence the event of sentence (18b), would elicit a negative trending electrophysiological wave 400 ms after the(cream event of would elicit a negative trending electrophysiological wave 400 ms after the event of interest vs. interest (cream vs. batteries), compared to (18a). interest (cream vs. batteries), compared to (18a). batteries), compared to (18a). 18. a. I drink my tea with cream 18. a. I drink my tea with cream b. I drink my tea with batteries b. I drink my tea with batteries InIncontrast, contrast,(19b) (19b)would wouldelicit elicitaaapositive positivegoing goingwave wave600 600ms msafter afterthe thearea areaof ofinterest interestcompared compared In contrast, (19b) would elicit positive going wave 600 ms after the area of interest compared toto(19a). (19a). to (19a). 19. a. I drink my tea with cream 19. a. I drink my tea with cream b. I drinks my tea with cream b. I drinks my tea with cream Eventrelated relatedpotentials potentialsresearch researchwith withL2 L2learners learnershas hasdocumented documentedmixed mixedresults, results,with withbeginners beginners Event Event related potentials research with L2 learners has documented mixed results, withwith beginners usually insensitive to violations, but showing increasingly native-like cortical responses greater usually to violations, but showing increasingly native-like cortical responses with greater Languages insensitive 2018, 3, 15 9 of 19 usually insensitive to violations, but showing increasingly native-like cortical responses with greater proficiency(Foucart (Foucart2008; 2008;McLaughlin McLaughlinetetal. al.2010; 2010;Tanner Tanner et et al. al. 2013). 2013). proficiency proficiency (Foucart 2008; McLaughlin et al. 2010; TannerAnglophone et al. 2013). Givenextensive extensive production documenting L2pronouns, French pronouns, Given production datadata documenting Anglophone in situ in L2 situ French Sneed Sneed German et al. (2015) tested reactions to ungrammatical in situ strong and clitic pronouns (20a, German et al. tested reactions to ungrammatical in situ strong and clitic pronouns (20a, b), as well as 8 8 as a baseline. b), as well assentences grammatical that serve grammatical (20c)sentences that serve(20c) as a baseline. 7 7
For discussion, see Gouvea et al. (2010); Steinhauer et al. (2010). 20.For a. Après seeleGouvea match elleSteinhauer a rencontré discussion, et al. (2010); et al. (2010).*le./ *les After the-M match she has met him-CL/them-CL
dans
la
rue
in
the-F
street
‘After the match she met him/them in the street.’ b.
Après
le
match
elle
a rencontré
*lui/ *eux
dans
la
rue
After
the-M
match
she
has met
him-STG/ them-STG
in
the-F
street
‘After the match she met him/them in the street.’ c.
Après
le
match
elle
l’a/les a
rencontré(s)
dans
la
rue
After
the-M
match
she
him-CL/them-CL
has met
in
the
street
‘After the match she met him/ them in the street.’
Froma prosodic a prosodic perspective, strong pronouns in a PWd stressed PWd(20b), position (20b), From perspective, strong pronouns are licitare in alicit stressed position but they but they are morphosyntactically ungrammatical. In contrast, the in situ clitics (20a) are both are morphosyntactically ungrammatical. In contrast, the in situ clitics (20a) are both morphosyntacitcally and prosodically illicit. Participants did two behavioral tasks (grammaticality morphosyntacitcally and prosodically illicit. Participants did two behavioral tasks (grammaticality judgmentsand andgeneral generalproficiency) proficiency)and andtheir theircortical cortical reactions were measured. The native French judgments reactions were measured. The native French speakers showed predictable behavioral results (near ceiling accuracy) and a robust P600 speakers showed predictable behavioral results (near ceiling accuracy) and a robust P600 to to ungrammaticalininsitu situpronouns, pronouns,both bothstrong strongand andclitic. clitic.The The high-intermediates’ results were ungrammatical high-intermediates’ results were qualitativelysimilar similarin in behavioral responses to natives and significantly better than qualitatively behavioral andand ERPERP responses to natives and significantly better than lowlow-intermediates. In contrast, the low-intermediates’ results on the behavioral grammaticality intermediates. In contrast, the low-intermediates’ results on the behavioral grammaticality judgments(GJ) (GJ)task task showed chance judgement inclitics, situ clitics, (rejected 49%time), of the time), judgments showed chance judgement of in of situ (rejected 49% of the while while ungrammatical strong pronouns were dismissed only of thesuggesting time, suggesting that ungrammatical strong pronouns were dismissed only 23% of 23% the time, that low9 9 low-intermediates are not sensitive to the ungrammaticality of these constructions. intermediates are not sensitive to the ungrammaticality of these constructions. An important important inference thethe response distinction between strong An inference for forfurther furtherinvestigation investigationconcerns concerns response distinction between and clitic for the two groups intermediates. Recall that the English uses strong andpronouns clitic pronouns for the two of groups of intermediates. Recall that thepronoun English system pronoun strong pronouns as the default and differs in morphosyntactic form and word order from the French system uses strong pronouns as the default and differs in morphosyntactic form and word order from the French system, in which object pronouns may only be clitic. Both groups of intermediates showed higher accuracy and more French-like cortical responses to the clitic violations than the strong 7 pronouns. The less group inettheir acceptance of prosodically licit strong pronouns, For discussion, seeproficient Gouvea et al. (2010);persists Steinhauer al. (2010). 8 Given the incremental appearance of the clitic le, the reader’s first reaction to (20a) would be toconclude process le asthat an article while showing an emerging implicit understanding of object clitics; one might whileof an anticipated noun phrase. still apparently using their L1 English settings for pronoun placement, their cortical response 9 Buckley (2005), in a phonological analysis of a clitic production task given to intermediate French learners, found a small demonstrates sensitivity to the French particular, be demonstrating sub-group ofemerging her participants who stress clitics;prosody. most of theIn participants did learners not assignmay prominence to the clitics. sensitivity to the apparent prosodic deficiency of the clitic pronouns in L2 input. This in turn may be part of a general emerging sensitivity to French schwa (the vowel of highly frequent object clitics me, te, and le). If this is the case, part of leaners’ higher intolerance for the clitic pronoun in situ results from an emergent awareness that the schwa vowel is incompatible with a prominent position. This may facilitate the acquisition of morphosyntax, while not necessarily driving it. The more proficient group, while showing some residual behavioral acceptance of strong
Languages 2018, 3, 15
10 of 20
system, in which object pronouns may only be clitic. Both groups of intermediates showed higher accuracy and more French-like cortical responses to the clitic violations than the strong pronouns. The less proficient group persists in their acceptance of prosodically licit strong pronouns, while showing an emerging implicit understanding of object clitics; one might conclude that while still apparently using their L1 English settings for pronoun placement, their cortical response demonstrates emerging sensitivity to French prosody. In particular, learners may be demonstrating sensitivity to the apparent prosodic deficiency of the clitic pronouns in L2 input. This in turn may be part of a general emerging sensitivity to French schwa (the vowel of highly frequent object clitics me, te, and le). If this is the case, part of leaners’ higher intolerance for the clitic pronoun in situ results from an emergent awareness that the schwa vowel is incompatible with a prominent position. This may facilitate the acquisition of morphosyntax, while not necessarily driving it. The more proficient group, while showing some residual behavioral acceptance of strong pronouns, already were approaching the French natives in their responses to the ungrammatical in situ clitics. The research cited earlier on beginning and intermediate Anglophone learners of L2 Languages 3, 15 evidence that beginners and low intermediates transfer native pronoun word 10order, of 19 French 2018, provides but with increasing proficiency, learners are able to gain mastery of L2 French morphosyntax, as comprehension, production and cortical measures. We measures. infer that L2 can features and indicated by comprehension, production and cortical Welearners infer that L2acquire learners can acquire word order different from different the L1 system, butL1 need to further investigate prosody. features and word order from the system, but need to further investigate prosody.
2.4. 2.4.Research ResearchQuestions Questionsfor forAdvanced AdvancedLearners Learners Turning now to a proficiency levellevel thatthat has has been lessless explored, that of of advanced L2L2 French, Turning now to a proficiency been explored, that advanced French,we weare are led ledtotothe thefollowing followingresearch researchquestions. questions.Recall Recallthe theprosodic prosodicstructures structuresinvolved involvedininFrench Frenchand andEnglish English clitic cliticconstructions constructions((9) ((9)isisrepeated repeatedhere): here): 9.
Prosodic structure of English object pronouns PPh
PPh |
PWd
PWd
see
John/him
PWd PWd see
‘m
English full pronouns are PWds under PPh, whereas their reduced clitic forms are affixal clitics English fullto pronouns PWds PPh,stages whereas their reduced clitic forms are affixal right-attached the PWdare verb. Inunder the early of L2A, the in situ pronouns that haveclitics been right-attached to the PWd verb. In the early stages of L2A, the in situ pronouns that have been observed in production can be analyzed prosodically as PWd, as in (9a). There is no phonological observed production can be analyzed prosodically as PWd, as in an (9a). Thereclitic is no phonological evidencein (phonological reduction or prosodic weakness) to suggest affixal (9b) analysis. In evidence (phonological reduction or prosodic weakness) to suggest an affixal clitic (9b) analysis. contrast, French object clitics are left free clitics under PPh, just as French and English articles are. In contrast, French L2 object clitics are leftpersist free clitics under PPh,transfer just as with French and placement; English articles are. Beginning French learners in Anglophone in situ intermediate Beginning French learners persist distinguishing in Anglophone with in in situ learners appearL2 to gain prosodic sensitivity, licittransfer strong pronouns situplacement; from illicit intermediate learners appear to gain prosodic sensitivity, distinguishing licit strong pronouns in situ clitics. We investigate advanced learners of L2 French pronouns to determine if they have adapted from illicit clitics. We investigate advanced learners of L2 French pronouns to determine if they have their object clitics to a French-like prosodic representation of left free clitics (left sister to PWd, daughter adapted objectpersist cliticsin tousing a French-like prosodic representation of left freetoclitics to of PPh),their or if they an English-like PWd structure (right sister PWd).(left Wesister pose the PWd, daughter of PPh), or if they persist in using an English-like PWd structure (right sister to PWd). following research questions. We pose the following research questions. Research questions (RQs) Research questions (RQs) In order to determine that advanced learners have adapted to French prosody in their In order toand determine that advanced learners have adapted to French prosody in their comprehension production of pronominal clitics, we will need to prove that they are not using the comprehension and production of pronominal clitics, we will need to prove that they are not using the English-like representations in (9). 10 A lack of prosodic prominence on object clitics (RQ2) will rule out a PWd analysis, while clitic “stacking” (RQ3) will constitute evidence against an affixal clitic analysis. The production of clitics before verbs with prefixes will constitute further evidence against an affixal analysis (RQ4) (Buckley 2005, p. 127, citing Hannahs (1995)). We pose the following research questions.
RQ1.
Do advanced Anglophone learners of L2 French show any evidence of prosodic
Languages 2018, 3, 15
11 of 20
English-like representations in (9).10 A lack of prosodic prominence on object clitics (RQ2) will rule out a PWd analysis, while clitic “stacking” (RQ3) will constitute evidence against an affixal clitic analysis. The production of clitics before verbs with prefixes will constitute further evidence against an affixal analysis (RQ4) (Buckley 2005, p. 127, citing Buckley (2005)). We pose the following research questions. RQ1. Do advanced Anglophone learners of L2 French show any evidence of prosodic transfer in their comprehension (grammaticality judgments) or production (morphosyntax) of object pronouns? RQ2. Do advanced Anglophone learners of L2 French produce object pronouns with any prosodic prominence? RQ3. Do advanced Anglophone learners of L2 French produce object pronouns in “stacked” sequences? RQ4. Do advanced Anglophone learners of L2 French produce object pronouns before verbs with prefixes? 3. New Evidence from Advanced Learners In this section we present L2 French data on production and GJ drawn from oral interviews with Chloe (age of acquisition onset 13), Eleanor (age of onset 17) and Max (age of onset 48), all post-puberty L2 learners.11 As noted earlier, French object clitic usage can be problematic for L2 learners, as indicated by several studies cited in the previous section. We examine the use of French object clitics and strong pronouns in three interviews for each individual conducted before, during, and after a period abroad of seven to nine months (a total of nine interviews over a nine-month period). The following section provides a description of the subjects and the data collection.12 3.1. Participants Max (age 59–60 at interview) began studying French at age 48, first on his own (completing the “French in Action” video program of first-year French) and later (with his wife Eleanor) with the help of a native French tutor who met with them one hour weekly for conversational exchanges over a period of 11 years prior. Eleanor (age 53 at interview) studied for six years in high school and college and at age 28 spent two months with a family in France. She and Max vacation annually in France and do extensive reading, independent vocabulary/grammar study, audio listening, and television viewing in French for 16-18 h per week at home. At the time of the interviews, Max and Eleanor were spending four months in Paris and three months in Lyon, where they had daily contact with French in a variety of contexts. Chloe, (age 22–23 at interview), studied French for nine years in high school and college; spent six months in a family in France at age 16 (Herschensohn 2001, 2003) and four months in study abroad at age 20. Subsequently, she became an “assistante d’anglais” (English teaching assistant) in the French overseas department of Réunion for nine months. Table 2 summarizes these points. Table 2. Characteristics of subjects.
10 11
12
Subject
Age of Onset
Age, Interview
Immersion
Max Eleanor Chloe
48 17 13
59–60 53 22–23
13 months 15 months 19 months
An anonymous reviewer notes “in order to show that clitics are free clitics in interlanguage French, it must be shown that no other analysis is possible.” We thank this reviewer for suggesting the following diagnostics. This research protocol was approved by the University of Washington IRB committee, in compliance with human subjects guidelines. All subjects gave their informed consent to participate in interviews and grammaticality tasks. Pseudonyms are used. Chloe’s L2 French mastery at an earlier stage is reported in Herschensohn (2001, 2003, 2004); the three advanced learners are documented (for TP and DP phenomena) in Herschensohn and Arteaga (2007, 2009, 2016).
Languages 2018, 3, 15
12 of 20
3.2. Materials and Methods Although the project was ostensibly longitudinal (over seven to nine months), the interviews showed that the learners were sufficiently advanced to have reached a fairly steady state, and so the differences among the interviews were rather minimal. No notable change was observed over time in the error rate (but Chloe’s production increases over the nine month period), and for this reason, we have collapsed the results from the three interviews together. Interviews were conducted by one of the authors (who has been certified by the American Council on the Teaching of Foreign Languages), who informally evaluated the three as being at an advanced level according to the American Council on the Teaching of Foreign Languages (ACTFL) Guidelines (cf. http://www.actfl.org/i4a/pages/index.cfm? pageid=4236). All three were able to use a range of tenses and moods, to discuss hypothetical situations and to carry out conversations on abstract and theoretical topics. They nevertheless made errors, particularly of gender assignment and aspect (Herschensohn and Arteaga 2009, 2016). Conducted in a university office or in the learner’s residence in France, each interview included elicitations of the following areas: present tense, descriptions of everyday routines and environments; past and future tenses; hypothetical situations; role play including a problem (e.g., confronting a dry cleaner for damage or introducing a speaker at a lecture). The interviewer gave leading questions to carry forward the conversation, but it was the interviewees who provided most of the dialogue. After the interview, subjects did written grammaticality tasks related to pronouns and other topics. The GJ task was adapted from Hawkins et al. (1993) and modified to have three versions for the three interviews. Comprising 50 sentences, half of which were ungrammatical, it had been used in the earlier studies reported (Herschensohn 2001, 2003, 2004; Herschensohn and Arteaga 2007, 2009, 2016). It included sentences that targeted both pronoun use (20% of the sentences) and verb placement (80% of the sentences); representative sentences for the pronouns are included in Appendix A. Participants were asked to mark the sentences as G (grammatical), NS (not sure) or * (ungrammatical) and to correct the latter. The data from the interviews were transcribed and checked for accuracy by three linguists fluent in French. The pronoun production data were coded and compiled by a graduate research assistant. The evidence was in the present case evaluated for all obligatory contexts of non-nominative pronouns, and for all uses of non-nominative pronouns, which were coded as strong or clitic. The production data from the nine interviews included 17,000 words, of which 217 were object clitics and 91 strong pronouns. 3.3. Results from Production and Grammaticality Judgments The evidence from the interview production and the grammaticality judgments indicates near ceiling performance on object pronoun production and the grammaticality judgment task. Table 3 documents the suppliance in obligatory contexts (the denominator in Table 3) of the 217 accurate object clitics, Table 4 portrays the correct use of the 91 strong pronouns (plus one error), while Table 5 gives the results of the GJ task. The three interviewees produced no prosodic errors of clitics in situ or strong pronouns as preverbal; more importantly, they produced consistent and accurate elision and liaison of function words, indicating ease with the L2 prosodic alternations (Tables 3 and 4). Table 3. Object clitic suppliance in obligatory contexts.
Chloe Max Eleanor
Interview I
Interview II
Interview III
5/5 24/24 20/20
21/21 24/24 28/29
25/25 44/44 26/27
The only errors in object clitic suppliance or word order were Eleanor’s in the second and third interviews (21).
Languages Languages2018, 2018,3,3,15 15 Languages 2018, 3, 15
14 14ofof21 21 13 of 20
21. Eleanor (II,III) III) 21. Eleanor Eleanor(II, (II, III) a.a. vous nous aimez pas vous nous n’ n’ aimez pas you-PL NEG you-PL us us NEG like-2PL like-2PL NEG NEG ‘You do not like us.’ ‘You do not like us.’ b.b. dis àà dis tell-2SG tell-2SG to to ‘Tell him.’ ‘Tell him.’
(= (=vous vousne nenous nousaimez aimezpas) pas) (= (=you-PL you-PLNEG NEGus uslike-2PL like-2PLNEG) NEG)
lui lui him-ST him-ST
(= (=dis-lui) dis-lui) (tell (tellhim-DAT-CL) him-DAT-CL)
Chloe made lexical errors of case assignment (22), which were not counted as lack of supply or Chloe Chloemade madelexical lexicalerrors errorsof ofcase caseassignment assignment(22), (22),which whichwere werenot notcounted countedas aslack lackof ofsupply supplyor or wrong morphological form. wrong wrongmorphological morphologicalform. form.
22. Chloe (II, II,III) III) 22. Chloe Chloe(II, (II,II, II, III) a.a. me visitent me visitent me-CL visit-3PL me-CL visit-3PL ‘They visit me.’ ‘They visit me.’
(= (=me merendent rendentvisite) visite) (= (=me-ACC me-ACCpay payvisit)) visit))
b.b. les répondre les répondre them-ACC them-ACC to toanswer answer ‘To ‘Toanswer answerthem.’ them.’
(= (=leur leurrépondre) répondre) (them-DAT (them-DATto toanswer) answer)
c.c. leur aider leur aider them-DAT to them-DAT tohelp help ‘To ‘Tohelp helpthem.’ them.’
(= (=les lesaider) aider) (them-ACC (them-ACCto tohelp) help)
As for the suppliance indicated inin Table 4 as As most were used in PPs (obligatory suppliance indicated Table 44 Asfor forthe the strong strongpronouns, pronouns,most mostwere wereused usedin inPPs PPs(obligatory (obligatory suppliance indicated in Table the denominator); in addition, the three participants used freestanding strong pronouns (indicated by as the denominator); in addition, the three participants used freestanding strong pronouns (indicated as the denominator); in addition, the three participants used freestanding strong pronouns (indicated “+” in the thesethese werewere supplemental, not required). These These correct tokens numbered 91. Eleanor by in the not 91. by “+” “+” in table; the table; table; these were supplemental, supplemental, not required). required). These correct correct tokens tokens numbered numbered 91. mistakenly used tu for toi in interview II (23), that is the additional “+*1” indicated in her Interview Eleanor Eleanormistakenly mistakenlyused usedtu tufor fortoi toiin ininterview interviewIIII(23), (23),that thatisisthe theadditional additional“+*1” “+*1”indicated indicatedin inher her II column.IIIIcolumn. Interview Interview column. Table 4. Strong pronoun use. Table Table4.4.Strong Strongpronoun pronounuse. use.
Chloe Chloe Max Max Eleanor Eleanor
Chloe Max Eleanor
Interview II I Interview Interview 3/3 PP 3/3 3/3 PP PP 3/3 3/3 66 + 6 3/3PP PP++PP 6/6 PP 6/6 6/6PP PP++44 + 4
23. 23. 23. Strong Strongpronoun pronounerrors errors(Eleanor (EleanorII) II) vous, tu et Michel vous, tu et Michel you-PL you-PL you-SG-CL you-SG-CL and andMichel Michel ‘Y’all, ‘Y’all,you youand andMichel.’ Michel.’
Interview Interview II II Interview II
III InterviewInterview III Interview III 3/3 PP + 4 4/4 PP + 1 PP 3/3 3/3 PP + 4 +4 4/4 PP + 1 4/4 PP + 1 3/33/3 PPPP + 2++22 14/14 PP +14/14 4 PP 3/3 PP 14/14 PP++44 15/15 PP + 5 + *1 11/11 PP + 3 15/15 11/11 15/15PP PP++55++*1 *1 11/11PP PP++33
(= (=vous, vous,toi toietetMichel) Michel) (= you-PL you-SG-STG (= you-PL you-SG-STGand and Michel) Michel)
The errors include mainly lexical errors (21b, the verb dire ‘to say’ requires an indirect object The Theerrors errors include includemainly mainlylexical lexicalerrors errors (21b, (21b, the theverb verb dire dire ‘to ‘to say’ say’ requires requires an an indirect indirect object object clitic, not a PP complement), and two case reversals on the object clitics (22b, c). Eleanor misaligns clitic, clitic,not notaaPP PPcomplement), complement),and andtwo twocase casereversals reversalson onthe theobject objectclitics clitics(22b, (22b,c). c).Eleanor Eleanormisaligns misaligns her negative cliticne ne(21a) (21a)and andmisuses misusesstrong strongand and clitic pronouns (21b, 24). The word order error her clitic pronouns (21b, 24). The word order error in hernegative negativeclitic clitic ne (21a) and misuses strong and clitic pronouns (21b, 24). The word order error in in 21a is noteworthy in that the object clitic is placed (and “stacked”) to the left of the negative clitic 21a 21aisisnoteworthy noteworthyin inthat thatthe theobject objectclitic clitic isisplaced placed(and (and“stacked”) “stacked”)to tothe theleft leftof ofthe thenegative negativeclitic clitic particle ne—a placement incompatible with an affixal clitic analysis of object pronouns on the part of particle particlene—a ne—aplacement placementincompatible incompatiblewith withan anaffixal affixalclitic cliticanalysis analysisof ofobject objectpronouns pronounson onthe thepart partof of this speaker. this speaker. this speaker. As the first section has shown, beginning and intermediate Anglophone learners of L2 French As Asthe thefirst firstsection sectionhas hasshown, shown,beginning beginningand andintermediate intermediateAnglophone Anglophonelearners learnersof ofL2 L2French French transfer native pronoun word order, but the evidence here indicates that these advanced learners transfer native pronoun word order, but the evidence here indicates that these advanced learners transfer native pronoun word order, but the evidence here indicates that these advanced learners appropriately use word order. The pronominal errors produced by by the appropriately pronominal morphosyntax and word order. The pronominal errors produced appropriatelyuse usepronominal pronominalmorphosyntax morphosyntaxand and word order. The pronominal errors produced by
Languages 2018, 3, 15 Languages 2018, 3, 15
14 of 20 15 of 21
the interviewees do resemble not resemble of beginners; no in situ or clitics or missing interviewees do not thosethose of beginners; there there are noare in situ clitics missing objects.objects. Rather, Rather, the few mistakes that do appear are almost all lexical errors or wrong case. the few mistakes that do appear are almost all lexical errors or wrong case. As the grammaticality grammaticality judgments judgments(Table (Table5), 5),the theadvanced advancedlearners’ learners’allall indicated rejection As for for the indicated rejection of of in in situ clitics and across the board acceptance of properly placed object clitics. There were no errors situ clitics and across the board acceptance of properly placed object clitics. There were no errors for for Eleanor, response in Interview for Max for ungrammatical sentence 13 13 Eleanor, andand oneone not not suresure response in Interview I forI Max for ungrammatical sentence (24).(24). 24.
*Ils deux sont partis They two are gone ‘They both left at noon.’
à midi at noon
(= tous les deux, ils sont partis)
Table Table 5. 5. Accuracy Accuracy of of pronouns pronouns in in GJT. GJT.
Chloe Max Eleanor Languages 2018, 3, 15
Chloe Max Eleanor
Interview I Interview I NA NA 9/10 9/10 10/10 10/10
Interview II 7/10 7/10 10/10 10/10 10/10 10/10
Interview II
Interview III 9/10 9/10 10/10 10/10 10/10 10/10 14 of 19
Interview III
In In contrast, contrast, Chloe Chloe did did not not complete complete the the GJT GJT for for Interview Interview I, I, and and made made wrong wrong judgements judgements for for the the In contrast, Chloe did not complete the GJT for Interview I, and made wrong judgements for the sentences in (25) in the other sessions. sentences in (25) in the other sessions. sentences in (25) in the other sessions. 25. 25. a. Eve-Anne s’est s’est brossé brossé les les cheveux cheveux (NS, (NS, Int Int II) II) a. Eve-Anne ‘EA ‘EA brushed brushed her her hair.’ hair.’ (G) (G) b. b.
Est-ce Est-ce que que vous vous leur leur avez avez parlé parlé ?? (*, (*, Int Int II) II) ‘Have you spoken to them?’ (G) ‘Have you spoken to them?’ (G)
c. c.
Ils Ils deux deux sont sont partis partis àà midi. midi. (G, (G, Int Int II, II, III) III) ‘They both left at noon.’ (*) ‘They both left at noon.’ (*)
The The three three participants participants correctly correctly rejected rejected in in situ situ clitics clitics and and accepted accepted correctly correctly placed placed clitics clitics (cf. (cf. The three participants correctly rejected in situ clitics and accepted correctly placed clitics (cf. Appendix A). In contrast, Chloe and her peer Emma, taking the same GJT at an earlier stage Appendix A). A). In In contrast, contrast, Chloe Chloe and and her her peer peer Emma, Emma, taking taking the the same same GJT GJT at at an an earlier earlier stage stage of of Appendix of development development (Herschensohn (Herschensohn 2004), 2004), averaged averaged less less than than 70% 70% accuracy accuracy on on the the task, task, with with mistaken mistaken development (Herschensohn 2004), averaged less than 70% accuracy on the task, with mistaken judgements to past participles. judgements about about in in situ situ clitics clitics and and those those ungrammatically attached judgements about in situ clitics and those ungrammatically ungrammatically attached attached to to past past participles. participles. 4. Discussion Discussion 4. 4. Discussion
Recall earlier. Recall the the research research questions questions posed posed earlier. earlier. Do Anglophone L2 show any evidence of RQ1.RQ1. Do advanced Anglophone learners of learners L2 Frenchof any evidence prosodic transfer in their RQ1. Do advanced advanced Anglophone learners ofshow L2 French French show of any evidence of prosodic prosodic transfer in (grammaticality judgments) or comprehension judgments) or production (morphosyntax) of object pronouns? transfer(grammaticality in their their comprehension comprehension (grammaticality judgments) or production production (morphosyntax) of pronouns? (morphosyntax) of object object pronouns? RQ2. Do advanced Anglophone learners of L2 French produce object pronouns with any RQ2. Do advanced Anglophone RQ2. advanced Anglophone learners learners of of L2 L2 French French produce produce object object pronouns pronouns with with any any prosodic Do prominence? prosodic prominence? prosodic prominence? RQ3. Do advanced Anglophone learners of L2 French produce object pronouns in “stacked” sequences? RQ3. Do Anglophone learners of L2 produce object pronouns in Do advanced advanced Anglophone learners L2 French French produce objectpronouns pronounsbefore in “stacked” “stacked” RQ4.RQ3. Do advanced Anglophone learners of L2ofFrench produce object verbs sequences? sequences? with prefixes? RQ4. Do RQ4. Do advanced advanced Anglophone Anglophone learners learners of of L2 L2 French French produce produce object object pronouns pronouns before before verbs verbs with prefixes? Do Max, Eleanor and Chloe indicate in their use of L2 French pronouns that they have adapted with prefixes? theirDo object clitics to a French-like prosodic in representation of left free pronouns clitics (left sisterthey to PWd, daughter Do Max, Max, Eleanor Eleanor and and Chloe Chloe indicate indicate in their their use use of of L2 L2 French French pronouns that that they have have adapted adapted of PPh), or do they persist in using the English-like PWd structure (right sister to PWd)? As discussed their their object object clitics clitics to to aa French-like French-like prosodic prosodic representation representation of of left left free free clitics clitics (left (left sister sister to to PWd, PWd, daughter of PPh), or do they persist in using the English-like PWd structure (right sister to PWd)? daughter of PPh), or do they persist in using the English-like PWd structure (right sister to PWd)? As As discussed discussed earlier, earlier, if if the the clitics clitics are are never never stressed, stressed, the the PWd PWd analysis analysis is is eliminated, eliminated, and and if if clitics clitics are are 13 Sentence (25) was not varied for the three versions; it is ungrammatical since the subject clitic is stressed, but can only be “stacked”, or “stacked”, or used used before before prefixed prefixed verbs, verbs, the the affixal affixal analysis analysis is is invalid. invalid. corrected through paraphrase. Our three learners show no evidence of prosodic prominence (stress) on object clitics that are preverbal; the only pronouns used in the stressed position (freestanding and in PPs) are strong 13 Sentence (25) was not varied for the three versions; it is ungrammatical since the subject clitic is stressed, pronouns. We eliminated PWd analysis (which may be appropriate for beginning learners but can onlytherefore be corrected through the paraphrase. who transfer in situ pronouns from English). As we saw earlier, the advanced learners’ GJs all indicated rejection of in situ clitics and across the board acceptance of properly placed object clitics. Furthermore, Max, Eleanor and Chloe show evidence of stacked pronouns (see Appendix B) as in (26).
RQ4.
sequences? Do advanced Anglophone learners of L2 French produce object pronouns before verbs with prefixes?
Do Max, Eleanor and Chloe indicate in their use of L2 French pronouns that they have adapted Languages 2018,clitics 3, 15 to a French-like prosodic representation of left free clitics (left sister to PWd, 15 of 20 their object daughter of PPh), or do they persist in using the English-like PWd structure (right sister to PWd)? As discussed earlier, if the clitics are never stressed, the PWd analysis is eliminated, and if clitics are earlier, if the clitics are never stressed, the PWd analysis is eliminated, and if clitics are “stacked”, or “stacked”, or used before prefixed verbs, the affixal analysis is invalid. used before prefixed verbs, the affixal analysis is invalid. Our three learners show no evidence of prosodic prominence (stress) on object clitics that are Our three learners show no evidence of prosodic prominence (stress) on object clitics that are preverbal; the only pronouns used in the stressed position (freestanding and in PPs) are strong preverbal; the only pronouns used in the stressed position (freestanding and in PPs) are strong pronouns. We therefore eliminated the PWd analysis (which may be appropriate for beginning learners pronouns. We therefore eliminated the PWd analysis (which may be appropriate for beginning learners who transfer in situ pronouns from English). As we saw earlier, the advanced learners’ GJs all indicated who transfer in situ pronouns from English). As we saw earlier, the advanced learners’ GJs all indicated rejection of in situ clitics and across the board acceptance of properly placed object clitics. Furthermore, rejection of in situ clitics and across the board acceptance of properly placed object clitics. Furthermore, Max, Eleanor and Chloe show evidence of stacked pronouns (see Appendix B) as in (26). Max, Eleanor and Chloe show evidence of stacked pronouns (see Appendix B) as in (26). 26. a. Je ne le connais pas (Chloe II) ‘I don’t know him.’ b.
il s’en fichait (Max III) ‘he doesn’t care a whit.’
c.
je ne l’ai jamais visité (Eleanor III) ‘I have never visited it.’
Theirability abilitytotostack stackclitics cliticsshows showsthat thatthey theyare arenot notusing usingaaflipped flippedaffixal affixalclitic cliticanalysis analysis(which (which Their permits no stacking), but rather are treating these leftward clitics attached to inflected and infinitival permits no stacking), but rather are treating these leftward clitics attached to inflected and infinitival verbsas asfree freeclitics. clitics.Finally, Finally,Max Maxatatleast leastuses usespreverbal preverbalobject objectclitics cliticswith withverbs verbscontaining containingprefixes prefixes verbs such as ré-, and dé(Appendix B). such as ré-, and dé- (Appendix B). The accuracy of both production and comprehension (GJs) indicates mastery of the L2 French morphosyntax of clitic pronouns. As Arche and Domínguez (2011, p. 305) note, “if learners have acquired the morphosyntactic properties of Spanish clitics, then high rates in both the production and comprehension tasks will be observed.” We answer the first question in the negative, as the learners have overcome the transfer of English word order and morphosyntax; their only errors are lexical. We consequently see no impaired representation of their L2 syntax with respect to the use of strong and clitic pronouns in French. Although beginning and low intermediate French L2 learners make persistent errors with in situ clitics in behavioral responses and non-native neural responses, advanced Anglophone learners of L2 French master French pronoun word order and morphosyntax. These advanced learners demonstrate target-like mastery of French pronoun word order and prosody in production, and sensitivity to ungrammatical uses in grammaticality judgments. Similarly to Gabriele et al. (2013, p. 225) observations, we believe “our results show that by an advanced level of proficiency, learners can process features not instantiated in the native language similarly to native speakers”. As for prosodic stucture, recall that English pronouns vary in prosodic realization between prosodic words and affixal clitics-right, while French consistently requires object pronouns to be free clitics-left. However, we noted the availability of leftward free clitics in both L1 and L2 DPs. It is our point that learners are incapable of gaining the free clitic construction for French clitic pronouns for quite a long time. Anglophone learners are able to gain free clitic articles at the beginning and intermediate stages, so the free clitic prosodic structure is presumably available to beginning–intermediate learners in L1 English and L2 French articles (which are morphologically identical to third person clitic pronouns). The difficulty that the Anglophone learners have with French object pronouns is clearly not a difficulty with the prosody at the beginning and early intermediate stages, since they master the article plus noun left-clitic structure early on (indeed, they can simply transfer the L1 to L2 DP prosody). Our point is that a combination of factors that differentiate English and French pronouns—the contrasts noted by Cardinaletti and Starke (1999) in terms of semantic, morphological, syntactic and phonological properties—seem to converge to inhibit command of French object pronouns by early Anglophone learners. We propose that it is not until high intermediate (cf. Sneed German et al. 2015) and advanced levels that learners finally gain the implicit mastery that enables them to transfer this article free clitic prosody from DP to TP. Despite its availability, the
Languages 2018, 3, 15
16 of 20
beginning and intermediate learners were unable to adapt that structure to TP in L2 French. Eventually, though, at the advanced level, learners attain the ability to adapt the DP structure of prosodic free clitics to the TP domain, supporting the PTH (transfer of the L1 free clitic article) and the related concept of minimal adaptation (adapting the DP clitic to a TP domain) of native prosodic structures to construct the new L2 structures. The prosodic transfer hypothesis maintains that target prosody can be built when L1 structures can be licensed in new positions, and it is clear that the advanced learners have mastered the leftward clitics, licensing the pronominal objects in L2 preverbal instead of L1 postverbal positions. A lack of stressed preverbal clitics, together with the presence of stacked clitics and clitics before prefixed verbs, demonstrates that the correct surface word order corresponds to the correct prosodic structure as well; the clitics are prosodified as free clitics, and not as PWds or affixal clitics. Together with previous research, our results show that target prosody achieved via minimal adaptation, that is the transfer of the free clitic construction (article + noun) to a new context (pronoun + verb), cannot take place in an immediate transfer, but rather, a certain threshold of proficiency must be reached before the prosodic structures are adapted. Morphosyntax and prosody must apparently develop in parallel. 5. Conclusions Beginning and low intermediate Anglophone L2 French learners are insensitive to French prosodic constraints in using in situ pronouns transferred from the L1, as proposed by the prosodic transfer hypothesis.14 The PTH proposes that native prosodic structures may be adapted to facilitate the acquisition of L2 prosodic structure. This study has presented new evidence from three Anglophone advanced learners of L2 French that indicates ceiling performance for pronoun production (99% accuracy in 300 tokens over nine interviews) and grammaticality judgments (98% accuracy). This native-like performance demonstrates that target French morphosyntax and prosody, are built—as predicted by the PTH—by licensing pronominal free clitics in a new pre-verbal L2 position distinct from post-verbal L1. Furthermore, accurate prosody is confirmed by correct pronoun placement, pronoun use in stacked sequences and with prefixed verbs, as well as target-like production of prosodic alternations (liaison, elision) in the learners’ data. These results counter impaired representation approaches and suggest that early missing inflection may be overcome. Author Contributions: The authors contributed equally to this work. Acknowledgments: We thank Deborah Arteaga, Elisa Sneed German, Cheryl Frenck Mestre, Neal Snape and audiences at EuroSLA 2016 and GALA 2017 for comments and questions. We thank Kristen Piepgrass for transcribing and coding the data. Conflicts of Interest: The authors declare no conflict of interest.
Appendix Examples of Grammaticality Judgment Sentences with Pronouns/Clitics* 1. 2. 3. 4. 5. 6.
La television? Nous la regardons tous les jours. (‘The television? We watch it every day.’) *Marie? Est-ce que vous avez vu la? (=vous l’avez vue?) (‘Marie? Have you seen her?’) Eve-Anne s’est brossé les dents. (‘Eve-Anne brushed her teeth.’) *Marc a lave se avec du savon de Marseille (=Marc s’est lavé). (Marc washed with soap from Marseille.) Est-ce que tu leur as parlé? (‘Did you speak to them?’) *Marie et Paul, je la et le connais. (=je les connais) (‘Marie and Paul, I know them.’) * indicates ungrammaticality; the correct sentence follows.
14
We thank a reviewer for pointing out that in order to show that the difficulty is for prosodic reasons, further research should include the three levels of proficiency using the same experimental tasks.
Languages 2018, 3, 15
17 of 20
Appendix Examples of Stacked Clitics and Clitics before Prefixed Verbs Stacked clitics 1. 2. 3. 4. 5. 6.
Je ne le connais pas. (Chloe II) (‘I don’t know him.’) Je vois aucune façon de s’en sortir. (Chloe III) (‘I see no way of getting out of here.’) Tout le monde s’y intéresse. (Max II) (‘everyone is interested in that.’) Il s’en fichait. (Max III) (‘he didn’t care a whit about it.’) Le soleil ne s’est levé jusqu’à 8h30. (Eleanor II) (‘the sun didn’t come up until 8:30.’) Je ne l’ai jamais visité. (Eleanor III) (‘je have never visited it’)
Clitics before prefixed verbs 1. 2. 3.
Mais tout le monde est en train de se déplacer. (Max I) (‘ . . . everyone is moving’) Pour assurer qu’aucun étudiant ne se démarque pas par rapport aux autres. (Max II) (‘to make sure that no student stands out with respect to the others.’) Son désir est de se réintégrer. (Max III) (‘her desire is to reintegrate herself.’)
References Adiv, Ellen. 1984. Language learning strategies: The relation between L1 operating principles and language transfer in L2 development. In Second Languages: A Cross-Linguistic Perspective. Edited by Roger Andersen. Rowley: Newbury House, pp. 125–42. Arche, María J., and Laura Domínguez. 2011. Morphology and syntax dissociation in SLA: Evidence from L2 clitic acquisition in Spanish. In Morphology and Its Interfaces. Edited by Alexandra Galani, George Tsoulas and Glyn Hicks. Amsterdam and Philadelphia: John Benjamins, pp. 291–320. Auger, Julie. 1994. Pronominal Clitics in Québec Colloquial French: A Morphological Analysis. Ph.D. dissertation, University of Pennsylvania, Philadelphia, PA, USA. Buckley, Meaghen. 2005. Prosodic constraints and the syntax-phonology interface: The phonology of object clitics in L2 French. In BUCLD 29 Proceedings. Edited by Alejna Brugos, Manuella R. Clark-Cotton and Seungwan Ha. Somerville: Cascadilla Press, pp. 122–33. Cardinaletti, Anna, and Michal Starke. 1999. The typology of structural deficiency: A case study of the three classes of pronouns. In Clitics in the Languages of Europe. Edited by Henk van Riemsdijk. Berlin: Mouton de Gruyter, pp. 145–233. Culbertson, Jennifer. 2010. Convergent evidence for categorical change in French: From subject clitic to agreement marker. Language 86: 85–132. [CrossRef] Cuza, Alejandro, Ana Teresa Pérez-Leroux, and Liliana Sánchez. 2013. The role of semantic transfer in clitic drop among simultaneous and sequential Chinese-Spanish bilinguals. Studies in Second Language Acquisition 35: 93–125. [CrossRef] De Cat, Cécile. 2005. French subject clitics are not agreement markers. Lingua 115: 1195–219. [CrossRef] Duffield, Nigel, Lydia White, Joyce Bruhn de Garavito, Silvina Montrul, and Philippe Prévost. 2002. Clitic placement in L2 French: Evidence from sentence matching. Journal of Linguistics 38: 487–525. [CrossRef] Foucart, Alice. 2008. Grammatical Gender Processing in French as a First and Second Language. Unpublished Doctoral dissertation, 2008, Université de Provence & University of Edinburgh, Provence, France, Edinburgh, Scotland. Friederici, Angela D., Erdmut Pfeifer, and Anja Hahne. 1993. Event related brain potentials during natural speech processing: Effects of semantic, morphological and syntactic violations. Cognitive Brain Research 1: 183–92. [CrossRef] Gabriele, Alison, Robert Fiorentino, and José Alemán Bañón. 2013. Examining second language development using event-related potentials: A cross-sectional study on the processing of gender and number agreement. Linguistic Approaches to Bilingualism 3: 213–32. [CrossRef] Goad, Heather, and Meaghen Buckley. 2006. Prosodic structure in child French: Evidence for the foot. Catalan Journal of Linguistics 5: 109–42. [CrossRef]
Languages 2018, 3, 15
18 of 20
Goad, Heather, and Lydia White. 2004. Ultimate attainment of L2 inflection: effects of L1 prosodic structure. In EuroSLA Yearbook 4. Edited by Susan H. Foster-Cohen, Michael Sharwood Smith, Antonella Sorace and Mitsuhiko Ota. Amsterdam: J. Benjamins, pp. 173–200. Goad, Heather, and Lydia White. 2006. Ultimate attainment in interlanguage grammars: A prosodic approach. Second Language Research 22: 243–68. [CrossRef] Goad, Heather, and Lydia White. 2008. Prosodic structure and the representation of L2 functional morphology: A nativist approach. Lingua 118: 577–94. [CrossRef] Goad, Heather, and Lydia White. 2009. Prosodic transfer and the representation of determiners in Turkish-English interlanguage. In Representational Deficits in SLA: Studies in Honor of Roger Hawkins. Edited by Neal Snape, Yan-kit Ingrid Leung and Michael Sharwood Smith. Amsterdam and Philadellphia: J. Benjamins, pp. 1–26. Goad, Heather, Lydia White, and Jeffrey Steele. 2003. Missing inflection in L2 acquisition: defective syntax or L1-constrained prosodic representations? Canadian Journal of Linguistics 48: 243–63. [CrossRef] Gouvea, Ana. C., Colin Phillips, Nina Kazanina, and David Poeppel. 2010. The linguistic processes underlying the P600. Language and Cognitive Processes 25: 149–88. [CrossRef] Granfeldt, Jonas. 2005. The development of gender attribution and gender concord in French: A comparison of bilingual first and second language learners. In Focus on French as a Foreign Language. Edited by Jean-Marc Dewaele. Clevedon: Multilingual Matters. Granfeldt, Jonas, and Suzanne Schlyter. 2004. Cliticisation in the acquisition of French as L1 and L2. In The Acquisition of French in Different Contexts: Focus on Functional Categories. Edited by Philippe Prévost and Johanne Paradis. Amsterdam and Philadelphia: J. Benjamins, pp. 334–70. Grüter, Theres. 2008. When learners know more than linguists: (French) direct object clitics are not objects. Probus 20: 211–34. [CrossRef] Grüter, Theres, and Martha Crago. 2012. Object clitics and their omission in child L2 French: The contribution of processing limitations and L1 transfer. Bilingualism: Language and Cognition 15: 531–49. [CrossRef] Hagoort, Peter, and Colin M. Brown. 1999. Gender electrified: ERP evidence on the syntactic nature of gender processing. Journal of Psycholinguistic Research 28: 715–28. [CrossRef] [PubMed] Hamann, Cornelia. 2004. Comparing the development of the nominal and the verbal functional domain in French Language Impairment. In The Acquisition of French in Different Contexts: Focus on Functional Categories. Edited by Philippe Prévost and Johanne Paradis. Amsterdam and Philadelphia: J. Benjamins, pp. 109–46. Hannahs, S.J. 1995. The phonological word in French. Linguistics 33: 1125–44. [CrossRef] Hawkins, Roger. 2001. Second Language Syntax. Oxford: Blackwell. Hawkins, Roger, and Cecilia Yuet-hung Chan. 1997. The partial availability of Universal Grammar in second language acquisition: The ‘failed functional features hypothesis’. Second Language Research 13: 187–226. [CrossRef] Hawkins, Roger, and Florencia Franceschina. 2004. Explaining the acquisition and non-acquisition of determiner-noun gender concord in French and Spanish. In The Acquisition of French in Different Contexts. Focus on Functional Categories. Edited by Philippe Prévost and Johanne Paradis. Philadelphia and Amsterdam: John Benjamins, pp. 175–207. Hawkins, Roger, Richard Towell, and Nives Bazergui. 1993. Universal Grammar and the acquisition of French verb movement by native speakers of English. Second Language Research 9: 189–233. [CrossRef] Herschensohn, Julia. 2000. The Second Time Around: Minimalism and L2 Acquisition. Philadelphia and Amsterdam: John Benjamins. Herschensohn, Julia. 2001. Missing inflection in L2 French: Accidental infinitives and other verbal deficits. Second Language Research 17: 273–305. [CrossRef] Herschensohn, Julia. 2003. Verbs and rules: Two profiles of French morphology acquisition. Journal of French Language Studies 13: 23–45. [CrossRef] Herschensohn, Julia. 2004. Functional categories and the acquisition of object clitics in L2 French. In The Acquisition of French in Different Contexts: Focus on Functional Categories. Edited by P. Prévost and J. Paradis. Philadelphia and Amsterdam: John Benjamins, pp. 207–42. Herschensohn, Julia, and Deborah Arteaga. 2007. Marquage grammatical des syntagmes verbaux et nominaux d’un apprenant avancé. AILE 25: 159–78.
Languages 2018, 3, 15
19 of 20
Herschensohn, Julia, and Deborah Arteaga. 2016. Parameters, features and re-assembly in L2 French DP. In The Acquisition of French in Multilingual Contexts. Edited by Pedro Guijarro-Fuentes, Katrin Schmitz and Natascha Müller. Bristol, Buffalo and Toronto: Multilingual Matters, pp. 215–33. Herschensohn, Julia, and Deborah Arteaga. 2009. Tense and Verb Raising in advanced L2 French. Journal of French Language Studies 19: 291–318. [CrossRef] Hoover, Michael L., and Veena D. Dwivedi. 1998. Syntactic processing by skilled bilinguals. Language Learning 48: 1–29. [CrossRef] Hulk, Aafke. 2004. The acquisition of the French DP in a bilingual context. In The Acquisition of French in Different Contexts: Focus on Functional Categories. Edited by Philippe Prévost and Johanne Paradis. Amsterdam and Philadelphia: J. Benjamins, pp. 243–74. Kayne, Richard. 1975. French Syntax. Cambridge: MIT Press. Lado, Robert. 1957. Linguistics across Cultures: Applied Linguistics for Language Teachers. Ann Arbor: University Michigan Press. Lardiere, Donna. 2007. Ultimate Attainment in Second Language Acquisition: A Case Study. Mahwah: L. Erlbaum. McLaughlin, Judith, Darren Tanner, Ilona Pitkänen, Cheryl Frenck-Mestre, Kayo Inoue, Geoffrey Valentine, and Lee Osterhout. 2010. Brain potentials reveal discrete stages of L2 grammatical learning. Language Learning 60 S2: 123–50. [CrossRef] Ordoñez, Francisco, and Lori Repetti. 2014. On the morphological restrictions of hosting clitics in Italian and Sardinian dialects. L’Italia Dialettale: Rivista di Dialettologia Italiana 75: 173–200. Osterhout, Lee, Judith McLaughlin, Albert Kim, Ralf Greenwald, and Kayo Inoue. 2004. Sentences in the brain: Event-related potentials as real-time reflections of sentence comprehension and language learning. In The On-Line Study of Sentence Comprehension: Eyetracking, ERP, and Beyond. Edited by Manuel Carreiras and Charles E. Clifton Jr. New York: Psychology Press. Osterhout, Lee, Judith McLaughlin, Ilona Pitkänen, Cheryl Frenck-Mestre, and Nicola Molinaro. 2006. Novice learners, longitudinal designs, and event-related potentials: A paradigm for exploring the neurocognition of second-language processing. Language Learning. 56: 199–230. [CrossRef] Paradis, Johanne, Antoine Tremblay, and Martha Crago. 2014. French-English bilingual children’s sensitivity to child-level and language-level input factors in morphosyntactic acquisition. In Input and Experience in Bilingual Development. Edited by Theres Grüter and Johanne Paradis. Amsterdam and Philadelphia: J. Benjamins, pp. 161–80. Perpiñán, Silvia. 2017. Catalan-Spanish bilingualism continuum: The expression of non-personal Catalan clitics in the adult grammar of early bilinguals. Linguistic Approaches to Bilingualism 7: 477–513. [CrossRef] Prévost, Philippe. 2009. The Acquisition of French: The Development of Inflectional Morphology and Syntax in L1 Acquisition, Bilingualism, and L2 Acquisition. Amsterdam and Philadelphia: J. Benjamins. Prévost, Philippe, and Lydia White. 2000. Accounting for morphological variation in second language acquisition: Truncation or missing inflection? In The Acquisition of Syntax: Studies in Comparative Developmental Linguistics. Edited by Marc-Ariel Friedemann and Luigi Rizzi. Harlow: Longman, pp. 202–35. Roberts, Ian. 2010. Agreement and Head Movement: Clitics, Incorporation and Defective Goals. Cambridge: MIT Press. Rothman, Jason. 2011. L3 syntactic transfer selectivity and typological determinacy: The Typological Primacy Model. Second Language Research 27: 107–27. [CrossRef] Schwartz, Bonnie D., and Rex A. Sprouse. 1996. L2 cognitive states and the full transfer/full access model. Second Language Research 12: 40–72. [CrossRef] Selinker, Larry, Merrill Swain, and Guy Dumas. 1975. The interlanguage hypothesis extended to children. Language Learning 25: 139–52. [CrossRef] Selkirk, Elisabeth. 1996. The prosodic structure of function words. In Signal to Syntax: Bootstrapping from Speech to Grammar in Early Acquisition. Edited by James L. Morgan and Katherine Demuth. Mahwah: Lawrence Erlbaum, pp. 187–213. Snape, Neal, Yan-kit Ingrid Leung, and Michael Sharwood Smith, eds. 2009. Representational Deficits in SLA: Studies in Honor of Roger Hawkins. Amsterdam and Philadelphia: J. Benjamins. Snape, Neal, and Tanja Kupisch. 2010. Ultimate attainment of second language articles: A case study of an endstate second language Turkish-English speaker. Second Language Research 26: 527–48. [CrossRef] Sneed German, Elisa, Julia Herschensohn, and Cheryl Frenck-Mestre. 2015. Pronoun processing in Anglophone late L2 learners of French: Behavioral and ERP evidence. Journal of Neurolinguistics 34: 15–40. [CrossRef]
Languages 2018, 3, 15
20 of 20
Steinhauer, Karsten. 2014. Event related potentials (ERPs) in second language research: A brief introduction to the technique, a selected review and an invitation to reconsider critical periods in L2. Applied Linguistics 35: 393–417. [CrossRef] Steinhauer, Karsten, John E. Drury, Paul Portner, Matthew Walenski, and Michael T. Ullman. 2010. Syntax, concepts and logic in the temporal dynamics of language comprehension: Evidence from event-related potentials. Neuropsychologia 48: 1525–42. [CrossRef] [PubMed] Tanner, Darren, Judith McLaughlin, Julia Herschensohn, and Lee Osterhout. 2013. Individual differences reveal stages of L2 grammatical acquisition: ERP evidence. Bilingualism: Language and Cognition 16: 367–82. [CrossRef] Tsimpli, Ianthi Maria, and Maria Dimitrakopoulou. 2007. The Interpretability Hypothesis: Evidence from wh-interrogatives in second language acquisition. Second Language Research 23: 215–42. [CrossRef] White, Lydia. 1996. Clitics in L2 French. In Generative Perspectives on Language Acquisition. Edited by Harald Clahsen. Amsterdam and Philadelphia: J. Benjamins, pp. 335–68. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
