The Acquisition Of Spanish In Understudied Language Pairings — facebook.comLinguaLIB


Чтобы посмотреть этот PDF файл с форматированием и разметкой, скачайте его и откройте на своем компьютере.
Volume 3
�e Acquisition of Spanish in Understudied Language Pairings
Edited by T椞any Judy and Silvia Perp槱án
Issues in Hispanic and Lusophone Linguistics (IHLL)
ᴜsn
‚21ᠭ3᜗7
IHLL aims to provide a single home for the highest quality monographs and edited volumes
pertaining to Hispanic and Lusophone linguistics. In an eort to be as inclusive as possible, the
series includes volumes that represent the many sub-elds and paradigms of linguistics that
Jason Rothman
Jennifer Cabrelli Amaro
University of Reading
University of Illinois at Chicago
Editorial Board
Sonia Colina
University of Arizona
João Costa
Universidade Nova de Lisboa
I滪s Duarte
Universidade de Lisboa
Sia Frota
Universidade de Lisboa
Ángel J. Gallego
University of Barcelona
María del Pilar Gar揭a Mayo
Universidad del País Vasco
Anna Gavarró
Universitat Automa de Barcelona
Kimberly L. Geeslin
Indiana University
Michael Iverson
Macquarie University
Paula Kempchinsky
University of Iowa
Juana M. Liceras
University of Ottawa
John M. Lipski
Pennsylvania State University
Gillian Lord
University of Florida
Jairo Nunes
Universidade de São Paulo
Acrisio Pires
University of Michigan, Ann Arbor
Pilar Prieto
Universitat Pampeu Fabra
Liliana Schez
Rutgers University
Ana Lúcia Santos
Universidade de Lisboa
Carmen Silva-Corval
University of Southern
California
Juan Uriagereka
University of Maryland
Elena Valenzuela
University of Ottawa
Bill VanPatten
Michigan State University
e Acquisition of Spanish
in Understudied Language Pairings
Edited by
T椞any Judy
Wake Forest University
Silvia Perp槱án
�e University of Western Ontario
John Benjamins Publishing Company
Amsterdam
Philadelphia
the
American National Standard for Information Sciences – Permanence
of Paper for Printed Library Materials,
ᐛsi
�ḝ
“ᬮᰛ2᤯ihll⸘
Cataloging-in-Publication Data available from Library of Congress:
��ḝ“n21n38᠘ᘠ⠕aጝሩ “n21n10᨜ᠠ⠐ⴏ55ഩ
ᴗᘕ—2ጠᐛ‒ᨒ’ጛሠs
ᴗᘕ—2ጠᐛ‒ᨒ‐ᐛጠa
��ษ
ꤠ4nsᤠ阠John Benjamins B.V.
No part of this book may be reproduced in any form, by print, photoprint, micro�lm, or any
other means, without written permission from the publisher.
John Benjamins Publishing Co. · P.O. Bo砠36ᨚሠ뜠ᤑᨑ
Table of contents
Introduction
�e importance of crosslinguistic comparison in the study
of the acquisition of Spanish
Ti�any Judy & Silvia Perpiñán
in Quechua-Spanish Bilingualism
Liliana Sánchez
Verbal agreement in the L2 Spanish of speakers of Nahuatl
Alma P. Ramírez-Trujillo
& Joyce Bruhn de Garavito
Early coda production in bilingual Spanish and Basque
Maria-José Ezeizabarrena & Alaitz Alegria
�e locative paradigm in the L2 Spanish of Catalan native speakers
Silvia Perp槱án
Combining Spanish with German, French and Catalan
Laia Arnaus Gil & Natascha Müller
Knowledge and processing of subject-related discourse properties
in L2 near-native speakers of Spanish, L1 Farsi
Ti�any Judy
Subject pronouns in the L2 Spanish of Moroccan Arabic speakers:
Evidence from bilingual and second language learners
Aurora Bel & Estela Gar揭a-Alcaraz
�e construal of goal-oriented motion events by Swedish speakers
of L2 Spanish:
Encoding of motion endpoints and Manner of motion
Alejandra Donoso & Emanuel Bylund
�e Acquisition of Spanish in Understudied Language Pairings
Spanish as an L2 in an instructional context
Object drop in L2 Spanish, (complex) feature reassembly, and
L1 pre-emption:
Comparing English, Chinese, European and
Brazilian Portuguese learners
Michael Iverson & Jason Rothman
�e acquisition of dierential object marking in Spanish
by Turkish speakers
Silvina Montrul & Ay
e Gürel
Copula choice in adjectival constructions in Dutch L1 Spanish L2
Manuela Pinto & Alexia Guerra Rivera
Typological proximity in L2 acquisition:
�e Spanish non-native grammar
of French speakers
Juana M. Liceras & AnahAlba de la Fuente
Index
e importance of crosslinguistic comparison
in the study of the acquisition of Spanish
T椝any Judy & Silvia Perp槱án
Wake Forest University / e University of Western Ontario
�e aim of this volume is to present to both the reader and the eld a collection
of empirical studies examining the acquisition of Spanish, currently one of the
world’s most spoken and studied languages, in combination with languages other
than English. Despite the multitude of speakers of varied L1s that acquire Spanish,
no collection of understudied language combinations, such as those that comprise
this volume, has thus far been compiled⸠�e research that appears herein includes
European Portuguese) in conjunction with two non-Romance languages (
glish
and Mandarin Chinese). e purpose of this chapter is to account for some
observed dierences in the developmental sequences and ultimate attainment in
�椝any Judy & Silvia Perp槱án
that it demonstrates, through comparison of the four aforementioned language
pairings, that a simple view of L1 transfer alone does not adequately explain the
摩ierent behavior from the L1 groups. Instead, Iverson and Rothman propose that
�e importance of crosslinguistic comparison in the study of the acquisition of Spanish
Still other pairings from the same language family, namely Romance lan
�椝any Judy & Silvia Perp槱án
degree
�ce the emphasis has largely been placed on describing implicit gram
texts:
�uechua-Spanish in Peru; Nahuatl-Spanish in Mexico; Basque-Spanish in
the Basque Country; and Catalan-Spanish in Catalonia. Section 2 contains four
chapters that present research conducted on naturalistic aquirers of Spanish, be
adult classroom learners of Spanish: an overview chapter that includes
razilian
Portuguese,
uropean Portuguese and English classroom learners of Spanish as
well as Mandarin Chinese-Spanish child, teenage and adult naturalistically acquir
ing bilinguals in Peru; Turkish-Spanish learners in Turkey; Dutch-Spanish learners
�e collection of studies included in this section describes the grammar of bilin
�e importance of crosslinguistic comparison in the study of the acquisition of Spanish
the unique linguistic reality of places such as the Basque Country and Peru, for
example, where the speakers included in the four chapters outlined below live in
communities where two languages enjoy varying levels of presence and prestige in
studies were carried out in bilingual communities of Latin
merica (Sánchez in
Quechua-Spanish bilingual speech communities in the Southern Andes of Peru and
Bolivia and from the region of Lamas in the Amazonian region of Peru; Ra淭rez
Trujillo and Bruhn de Garavito in a predominantly Nahuatl-speaking village in the
state of Tlaxcala, Mexico). e other two studies took place in bilingual regions
of Spain in which both languages have o᭣ial status and the minority languages
have comparable prestige 阠albeit not presence 阠to that of the majority language
(Ezeizabarrena and Alegria in Basque-Spanish bilinguals in the Basque Country,
and Perp槱án in Catalan-Spanish bilinguals in Catalonia).
A case in point is Sánchez’s overview chapter, which goes through previously-
published studies on Spanish-Quechua bilingualism and demonstrates the
existence of crosslinguistic inuence in the Spanish spoken by Quechua speakers.
Furthermore, she provides an explanation for the diculties that these bilinguals
experience in terms of problems with the mapping of functional features onto
morphology. Building from the Feature Reassembly Hypothesis (
Lardiere 1998
), Sánchez proposes a ne-grained account to explain crosslinguistic
楮�uence in Quechua-Spanish speakers by dening and teasing apart functional
interference from functional convergence. Sánchez surveys a series of studies
and linguistic phenomena, paying special attention to the linguistic phenomena
placed at the lexicon/morphology/syntax interface. She documents innovative
�椝any Judy & Silvia Perp槱án
Trujillo and Bruhn de Garavito argue that this dierence is not only due to quan
tity of input, but also to quality of input, since the speakers that work or travel
Ezeizabarrena and Alegria’s chapter investigates the emergence of codas, a well-
known milestone in phonological development, in the spontaneous speech of a
Basque-Spanish bilingual child (1;09–2;01). is study is of great import because
the authors show that the two phonological systems of the early bilingual, despite
the punctual presence of crosslinguistic interference, develop separately and
according to the developmental stages of each language⸠�ey also show that the
bilingual child produced more codas in the language with more frequent and/or
varied types of codas (i.e. Basque) compared to his production in Spanish, a lan
guage with less frequent and fewer types of codas, corroborating previous ndings
�e importance of crosslinguistic comparison in the study of the acquisition of Spanish
�e last chapter of this section is devoted to the study of the Spanish grammar
Spanish knowlege, as measured by a standard prociency test, is indistinguish
able from that of Spanish native speakers. However, when these Catalan-dominant
bilinguals are thoroughly tested in their production and grammatical intuitions
in Spanish, they display signicant 摩ierences in their acceptance of the
d攞nite
eὥct
in existential constructions, underuse of
estar
to express location, and over
acceptance of
ser
to locate objects, as employed in Catalan grammar⸠�ese results
suggest that, even in the case of highly functional bilinguals, we may still nd
crosslinguistic interference in certain linguistic domains, and signicant dier
ences in the implicit grammars of the bilinguals compared to that of monolinguals.
�椝any Judy & Silvia Perp槱án
to study abroad for a certain period of time, but in reality, if the study abroad
session is not very long, this may more accurately be considered an instructional
environment. And nally, a more radical way in which a language can be acquired
is via rather continuous exposure to it through immigration to a country in which
that language is spoken, as in the cases that Judy, Bel and Gar揭a Alcaraz, and
Donoso and Bylund portray⸠�ese cases share the characteristic that the speak
ers were rst exposed to the L2 in adulthood, which might render very dissimi
lar outcomes, from fossilization to full convergence depending on the linguistic
property and the language pairing. 䑩ierently from the studies reported on in
Sectio渚3, it is widely assumed, given certain conditions like integration with the
楮�uence in the bilingual Spanish-German children, who have more diculties
assigning permanent properties (
ser
to the predicate than temporal properties
estar
⤮‟ese and other similar ndings lead the authors to propose that if the
�e importance of crosslinguistic comparison in the study of the acquisition of Spanish
of

�ultilingualism because it
hows that the acquisition of complex derivations
can be expedited if children are exposed to several languages with these complex
constructions, and that trilingual speakers can be more successful than bilingual
speakers in the acquisition process.
Next, Jud禒s chapter examines convergence on the discourse-constrained dis
tribution of null and overt referential subject pronouns in Spanish by near-native,
L1 Farsi speakers. Data from both oᡩne and online tasks examining Contrastive
Focus, Topic Shi and Topic Maintenance were collected and reported on in order
to test the Interface Hypothesis鈠prediction (
Sorace 2011
) that even near-
native speakers of facilitative language pairings will evidence processing dier
ences on external interface-conditioned properties. While evidence of native-like
knowledge and processing was found for some conditions, dierences were also
found highlighting the need for further empirical investigation of the tenability of
these claims. Importantly, and in line with the goals of this volume, Jud禒s chapter
contributes to the debate surrounding the IH’s claims by examining a facilitative
language pairing⸠�at is, much work conducted on the distribution of subject pro
nouns has focused on non-facilitative language pairings such as English-
panish
or French-Spanish, where transfer of L1 values does not immediately result in con
vergence⸠�us, if divergence in non-facilitative language pairings is found, it is
not possible to tease apart the eects of L1 transfer from the increased processing
burden experienced purportedly by bilinguals (IH). However, nding divergence
in speakers for whom L1 transfer is facilitative lends credence to the IH’s claims.
Without amassing research on both types of language combinations, it is not pos
sible to comment on this prediction.
but also contributes new empirical data regarding the Position of
ntecedent
Hypothesis (PAH,
Carminati 2002
) and microvariation. Bel and Gar揭a-Alcaraz
found that the PAH is only partly applicable in Moroccan Arabic and that while
native speakers鈠preferences with null subjects patterned with results found for
Spanish, Italian, and Catalan, preferences for overt subjects diered from
talian
and Catalan, instead patterning with Spanish preferences. Even so, a further dif
rabic
preferences, indicating a further level of variation. e lack of signicant dier
�椝any Judy & Silvia Perp槱án
resolution strategies would have proved facilitative but was not evidenced and
no research currently exists on this linguistic property for Swedish-
panish
acquisition – the quantity and quality of the input – is eliminated, or, at min
imum, reduced, thus providing researchers a window into the upper limits of
non-native language acquisition. For example, in Jud禒s study, the adult near-
native Farsi-Spanish speakers demonstrated native-like processing of Topic S桩1
tokens in a Self-Paced Reading task, a property for which other researchers have
�e importance of crosslinguistic comparison in the study of the acquisition of Spanish
study to study may have some bearing on the results. is may especially be true
of Arnaus Gil and
üller’s study on bi/trilingual children who had certainly
Spanish as an L2 in an instructional context
�e ᱮal section of this volume is devoted to learners exposed to explicit L2
instruction. While classroom learners are likely the most commonly studied
L2 population in the United States, this acquisition context has not consistently
proved itself to be neither maximally eective nor ecient in terms of learner
outcomes, perhaps as a result of the quantity and quality of the primary linguistic
�椝any Judy & Silvia Perp槱án
stages of language acquisition and can serve as a point of comparison for studies
conducted on speakers that are exposed to the L2 in naturalistic environments like
Iverson and Rothman, actually contains both naturalistic speakers in a non-
above:
�razilian Portuguese-Spanish, European Portuguese-Spanish, English-
own right,
�verson and Rothman’s chapter speaks to one of the main goals of the
�e importance of crosslinguistic comparison in the study of the acquisition of Spanish
A classic topic investigated in instructional SLA is the role of the L1. As dis
cussed above, the L1 can be both facilitative and non-facilitative, and its 攝ects
can be long-lasting in the development of the L2. Pinto and Guerra Rivera’s
chapter highlights the importance of analyzing the two languages at hand in
of existing data on the seven linguistic properties mentioned above,
iceras and
Alba de la Fuente demonstrate that proximity does not equal similarity and that
�椝any Judy & Silvia Perp槱án
that, by systematically examining L1 French and L2 Spanish, the authors provide
Concluding remarks
�e importance of crosslinguistic comparison in the study of the acquisition of Spanish
the nature of interlanguage grammars, one of the main purposes of SLA research.
generalizable ᱮdings.
�e collection of studies included here has allowed us to gain ground on our
understanding of L1 transfer in the sense that several studies have pointed out that
it is selective and by no means random. L2 speakers create coherent interlanguage
grammars that are not always copies of the L1 grammar, although they might
�椝any Judy & Silvia Perp槱án
variability in acquisitional outcomes remains a core characteristic of SLA and bi/
multilingualism.
In summary, this volume took as its central goal the expansion of the lan
guage pairings combined with Spanish in generative language acquisition
research as we believe much is to be learned from a wider and more representa
tive body of research. e results of the 12 contributing chapters, individually
and collectively, speak to the usefulness of examining less commonly-studied
Garavito, Alejandro Cuza, María del Pilar Gar揭a Mayo, Pedro Guijarro-
uentes,
Alberto Hijazo-Gas揳n, Pilar Larrañaga, Conxita Lleó, Karen Miller, Antje
Muntemdam, Öner Ö竧elik, Maria Carme Parasto Couto, Claire Renaud, Ana
Lucia Santos, Cristina Schmitt, Ludovica Serratrice, Roumyana Slabakova, Julio
Villa-Gar揭a, Christiane von Stutterheim, and two reviewers who wish to remain
anonymous. eir comments and insights have greatly shaped each of the chapters
and we are very appreciative of their eorts. We are also indebted to our editorial
assistant, Ansley Byers, for her hard work and precision, especially in the nal
moments of editing. Lastly, we are grateful to Jason Rothman and Cornelis H. J.
Vaes for their expert guidance and support.
References
panish
children: Eects of prominence and frequency of exposure.
Probus,
DOI: 10.1515/probus-2013-0001
Bruhn de Garavito, J., & Guijarro-Fuentes, P. (2002). L2 acquisition of indenite object drop
in Spanish. In J. Costa & M.J. Freitas (Eds.),
GALA 2001 Proceedings
(pp. 60–67). Lisbon:
Associ懧ão Portuguesa de Lin杵ística.
Carminati, M.N. (2002). e Processing of Italian Subject Pronouns. Unpublished Ph.D. dis
DOI: 10.2307/2273965
�e importance of crosslinguistic comparison in the study of the acquisition of Spanish
Chomsky, N. (1995). Minimalist inquiries: �e framework.
MIT Occasional Papers in Linguis
tics
DOI: 10.1515/prbs.2003.010
Long, M.H. (2007).
Problems in SLA
. Mahwah, NJ: Lawrence Erlbaum Associates.
DOI: 10.1080/09500780802152622
Polo, N. (2011). Notas sobre la adquisición fono泳gica bilingüe del español: El desarrollo de las
codas en español como lengua materna en hablantes bilingües del País Vasco.
Oihenart,
Prévost, P., & White, L. (2000). Missing surface inection or impairment in second language
acquisition? Evidence from tense and agreement.
Second Language Research, 16
Schmitt, C.J. (1997).
Aspect and the Syntax of Noun Phrases
. Unpublished Ph.D. dissertation.
University of Maryland, College Park.
Schmitt, C.J. (2005). Semi-copulas: Event and aspectual composition. In P. Kempchinsky
& R⸚
�labakova (Eds.),
Syntax, Semantics and the Acquisition of Aspect
(pp. 121–145).
Ne眚York, NY: Springer.
Schwartz, B.D., & Sprouse, R.A. (1996). L2 cognitive states and the full transfer/full access
model.
Second Language Research, 12
DOI: 10.1177/026765839601200103
Slabakova, R. (2009). What is easy and what is hard to acquire in a second language? In
䴮2Bowles, T. Ionin, S. Montrul, & A. Tremblay (Eds.),
Proceedings of the 10th
Generative
Approaches to Second Language Acquisition Conference
GASLA 2009
) (pp. 280–294).
Sorace, A. (2011). Pinning down the concept o映“interface” in bilingualism.
Linguistic Approaches
to Bilingualism, 1
DOI: 10.1075/lab.1.1.01sor
Sorace,
Pinning down the concept o映“interface” in bilingualism: A reply to peer com
mentaries.
Linguistic Approaches to Bilingualism, 2,
DOI: 10.1075/lab.2.2.04sor
�椝any Judy & Silvia Perp槱án
Trahey, M., & White, L. (1993). Positive evidence and preemption in the second language class
room.
Studies in Second Language Acquisition, 15
VanPatten, B. (1985). e acquisition of ser and estar by adult learners of Spanish: A preliminary
DOI: 10.2307/342218
VanPatten, B. (1987). Classroom learners鈠acquisition of ser and estar: Accounting for develop
mental patterns. In B. VanPatten, T.R. Dvorak, & J.F. Lee (Eds.),
Foreign Language
eaching:
A Research Perspective
(pp. 61–75). Cambridge, MA: Newbury House.
VanPatten, B. (2010). Some verbs are more perfect than others: Why learners have diculty
with ser and estar and what it means for instruction.
Hispania
White, L. (2003a).
Second Language Acquisition and Universal Grammar.
Cambridge: CUP.
DOI: 10.1017/cbo9780511815065.002
White, L. (2003b). On the nature of interlanguage representation: Universal Grammar in second
language. In C. Doughty & M. Long (Eds.),
se Handbook of Second Language Acquisition
(pp. 19–42). Oxford: Blackwell.
DOI: 10.1002/9780470756492.ch2
White, L. (2008). 䑩ierent? Yes. Fundamentally? No. Deniteness 攝ects in the L2 English of
Mandarin speakers. In R. Slabakova, J. Rothman, P. Kempchinsky, & E. Gavruseva (Eds.),
Proceedings of the 9th
Generative Approaches to Second Language Acquisition Conference
(pp. 251–261). Somerville, MA: Cascadilla Proceedings Project.
White, L. (2011). Second language acquisition at the interfaces.
Lingua,
DOI: 10.1016/j.lingua.2010.05.005
8ᘕᐠ6
Crosslinguistic inuences in the mapping
of functional features in Quechua-Spanish
Bilingualism
Liliana Sánchez
Rutgers University
Ὡs paper discusses evidence of crosslinguistic inuence that involves the
emergence of new patterns of feature-morphology mapping in Spanish as spoken
by native speakers of Quechua. It builds on the notions of functional interference
and functional convergence (
叡nchez 2003
) and of feature reassembly
Lardiere 2003
) in order to account for crosslinguistic inuence at: (a) the
syntactic level (feature reassembly and the emergence of non-argumental clitics)
(b) the syntax/morphology interface (feature reassembly) (c) syntax/lexicon
interface (feature reassembly and changes in argument structure), and (d) the
syntax/pragmatics interface (licensing of null objects and focus fronting). e
evidence shows that activation of features, feature reassembly and mapping onto
morphology is an important source of crosslinguistic inuence.
Introduction
�e understanding of second language acquisition and bilingualism from a
generative perspective has been characterized in the last decades by multiple
approaches. ey have ranged from earlier studies that provided strong empiri
cal support to the availability of universal language acquisition mechanisms and
the autonomy of syntax (
Flynn 1987
Liceras 2010
Schwartz & Sprouse 1996
White 1989
�iliana Sánchez
Paradis & Genesee 1996
) and have emphasized the role of crosslinguistic inu
ment (
Ullman 2005
). is perspective takes into account the separate and dis
tinct nature of lexical and syntactic knowledge. At the same time, it recognizes
that interactions across components (
Jackendo 1997
Slabakova 2008
) take place
when language representations and lexical items are activated for production and
comprehension purposes. e evidence from bilingual data represents an invalu
able window into how these interactions take place. Such evidence can not be eas
ily obtained from monolingual data.
In this context, the need to study bilingual and second language acquisition
in naturalistic contexts (
Meisel 2008
Serratrice, Sorace & Paoli 2004
) has become
cesses involved in the development of a grammatical representation in
tuations
in which input and processing for comprehension and production are more fre
quent than in instructional contexts.
osslinguistic inuences in Quechua-Spanish bilingualism
In this paper, I will focus on some of the major ndings and contributions
that the study of Quechua-Spanish bilingualism from a generative perspective
has brought to the ᱥld of second language acquisition studies. In this paper,
䤚adopt a perspective that assumes modularity and interactions across components
Jackendo 1997
⤮‟e contributions include evidence of crosslinguistic inuence
in the Spanish of Quechua speakers: (a) at the core syntax in the projection of
non-argumental clitics (
Camacho, Paredes & Sánchez 1995
Escobar 1991
Kalt
叡nchez 2003
) (b) at the lexico-morphosyntactic interface in
the mapping of functional features onto overt morphemes or independent words
Kalt 2009
叡nchez 2004
); and (c) at the syntax-pragmatics interface in the inter
Crosslinguistic inuence in Quechua-Spanish Bilingualism:
Feature reassembly, functional convergence, and the lexicon
�iliana Sánchez
know that both memory systems interact (
Ullman 2005
⤮‟e same is true of the
role of access to the L1 lexicon by L2 learners (
Kroll & Stewart 1994
Schwartz &
Kroll 2006
). We know that access to items in the L2 lexicon can be characterized
by suppression of their corresponding lexical equivalents in the L1. ere is also
current research that assumes non-selective access to the lexicon in each of the lan
guages spoken by a bilingual individual but focuses on syntactic cues as predictors
of crosslinguistic activation of the lexicon, namely, the possibility of higher activa
It supposes frequent activation of the L2 lexicon for
mprehension and
production purposes. e nature of L2 acquisition in language contact situations
�e examples presented in this article come from previous research on bilingualism in
Quechua and Spanish among bilingual speech communities in the Southern Andes of Peru
and Bolivia and from the region of Lamas in the Amazonian region of Peru. Most of the
Quechua examples belong to Southern Quechua varieties especially from Cuzco Quechua as
described in
Cerrón-Palomino (1987
) and
Sánchez (2010
).‟e examples in Spanish are from
non-contact General Spanish unless they come from a speci�c study in Quechua-Spanish
bilingualism.
osslinguistic inuences in Quechua-Spanish bilingualism
allows us then to address issues such as the activation of lexical and functional
Lardiere
related to the activation of lexical items and functional features and to the
apping
�is particular phenomenon will be discused in detail in the next section of the paper.
�iliana Sánchez
of these features onto morphology. In doing so my purpose is to highlight the need
Main morphosyntactic and syntactic characteristics
of Quechua and Spanish
Some of the main syntactic and morphological characteristics of Quechua and
Spanish are partially shared. Both Quechua and Spanish are null subject languages
with subject agreement marking on the verb, as exemplied in the following
sentences:
�e discussion of the data in this paper will not center around constraints on structures
or on general principles of economy, as they are not relevant to the data.
osslinguistic inuences in Quechua-Spanish bilingualism
Quechua
Papa-ta
mikhu-n-mi

Potato-
ᘒ4
eat-
᜕0m

�S/He is eating (a) potato(-es).’
叡nchez 2010
, p. 24)
Spanish
Com-e
papa-s

Eat-3.
potato-

�S/He eats potatoes.’
As these examples show, both languages share morphological marking of subject
person and number on the verb. is means that in both languages acquisition
Cerrón-
Palomino 1987
Tayta
churi-n-man
asi-ta

�ather
son-3.
9ᘔ
house-
ᘒ4
give-3.

��e father gives the house to his son.’
叡nchez 2010
, p. 13)
Spanish
El
�adre
le
asa
ijo

father
give
the
house
to
his
son

El
�adre
ijo
la
asa

father
give
to
his
son
he
house

�e father gives his son the house.’
�iliana Sánchez
As (4a) and (4b) show, Spanish marks indirect objects with the phonologically
independent preposition
ₑto鈠that has been argued to mark deniteness and/
or specicity similar to the clitic specication (
Bleam 1999
Spanish also dier greatly. Whereas in Quechua third person direct objects are not
marked as morphemes on transitive verbs and they may be null, in Spanish they
are marked as clitics if their referent is denite and/or specic:
Cuzco Quechua
Huwan-ta
riku-rqa-nki-chu?

Huwan-
ᘒ4
see-
8༔
ᘔa

Did you see Huwan?’

Ar
riku-rqa-ni.

Yes,
see-
8༔
ᘔa

Yes, I saw (him).’
Spanish
Vi-ste
Juan?

See-
8༔
ᜋ�f
to
Juan

Did you see Juan?’

S
lo-

Yes,
ᘒ4
see-
8༔
ᜋ�f

Yes, I saw him.’
�ere is, however, some direct, indirect and oblique object marking on the verb in
Quechua involving rst and second persons that does not involve features such as
deniteness or specicity. Quechua also allows for overt pronouns as independent
words marked with case:
Pusa-wa-sqa

bring-1.
ᄉj
᜖ea
ᔋ8
uܓ9

�S/he brought me.’
Cusihuamán 2001
, p. 163)
Mariya-m
ay-ta)
iku-n.

Mariya-
਑4
uܓ9
(s/he-
ᘒ4
see-3.

�Mariya sees (him/her).’
叡nchez 2010
, p. 27)
Notice also that there are signicant 摩ierences in the way in which tense and
person features are mapped onto morphemes in both languages. In Examples (5a)
and (5b) the morpheme -
rqa
encodes past tense and attested evidentiality and the
morphemes -
nki
and -
encode person and number whereas in Spanish the mor
pheme -
encodes tense, aspect, person, and number.
osslinguistic inuences in Quechua-Spanish bilingualism
In addition to dierences with respect to argumental morphological marking
on the verb, Quechua and Spanish dier with respect to the relevance of deri
vational morphology to syntax. As noted by
Cerrón-Palomino (1987
), Quechua
languages have a complex morphological system that includes derivational and
inectional morphemes. �e basic sequence of a᭸ation proposed by
Cerrón-
Palomino (1987
) in Quechua languages is:
Mikhu-naya-n-mi

Eat-
ഋe
਑4
uܓ9

�S/he is about to eat.’

�oot + Derivational morpheme+I渞ectional morpheme+Sentence level
morpheme
Cerrón-Palomino 1987
, p. 267)
While it is traditionally the case to think of derivational morphemes as involv
ing only the generation of new lexical items and of inectional morphemes as
being related to the syntax, in Quechua the separation is not as clear cut. Van
de Kerke (1996) notes that in Quechua, there are at least six derivational s甛xes
Mikhu-n.
Eat-3.

e) eats.’
Mikhu-naya-n.
Eat-
ഋe

e) wants/is about to eat.’
ased on
Cerrón-Palomino 1987
In its desiderative meaning, the sux -
naya
introduces the agent of the desire
as an addition to the agent of the eating but in its imminent meaning it does not
María
quie-re
om-er.
Maria
want-3.
eat-

�Maria wants to eat.’
�iliana Sánchez
As will be shown in the next section, this dierence is relevant to the understand
ing of crosslinguistic inuence in feature-morpheme mapping and the projection
of argument structure in Quechua-Spanish bilinguals.
Chay
allqu-cha-kuna-pis
phiña-ru-ku-n-ña
�at
ഓm
8อᘍ9
ᜋ�f
�ଊ5
ഓe
chay
chiku-cha-n-ta.
that
boy-
ഓm
8ᄏe
ᘒ4

A third type of important crosslinguistic 摩ierences in the mapping of features
onto morphemes comes from le dislocated elements. In this case, the informa
Wasi-ta-qa
Pirdu-m
uwa-rqa-n.

House-
ᘒ4
ᐑ8
Pirdu-
96ᔋv
਑4
build-
8༔

��e house, Pirdu built (it).’
叡nchez 2010
, p. 31)
Spanish
La
casa
Pedro
�a-
izo.

�e
house
Pedro
ᘒ4
made

�As for the house, Pedro made it.’
�is example comes from a study for which data were collected using a picture-based
story task in Cuzco in 2005 currently in preparation.‟e speaker is an adult female.
osslinguistic inuences in Quechua-Spanish bilingualism
With respect to le dislocated focalized constituents, in Quechua they are mor
phologically marked, but they are not in Spanish and it is possible that they may
occur without a sentence-internal clitic.
Quechua
Kustal-ta-m
unqa-rqu-n
hay
chu�r.

Package-
ᘒ4
਑ሯuܓ9
forget-
ᜋ�f
that
iver

�It was the package that that driver forgot.”
叡nchez 2010
, p. 30)
Spanish
El
�aquete
lvidó
el
hofer.
�e
package
forgot
the
driver

�It was the package that the driver forgot.’
As we will see in the next section, crosslinguistic inuence at the syntax/informa
tional structure interface might aiect some aspects of Merge but it may not neces
sarily aiect structural restrictions.
Functional interference, feature re-assembly and functional
convergence: Evidence of crosslinguistic inuence
in Quechua-Spanish bilingualism studies
Crosslinguistic inuence at the syntactic level: Functional interference,
feature re-assembly and the emergence of non-argumental clitics
One of the sources of evidence for functional interference that results in new pat
terns of L2 mapping onto morphology is the emergence of new non-argumental
clitics in Quechua-Spanish bilinguals. As noted by
Camacho, Paredes and
nchez
) and
Kalt (2009
), a possessor clitic has been found at dierent stages of
acquisition of Spanish by Quechua speakers from Southern Peru, as shown in
Example (19):
amarran
su
ata
del
condor

�ie

of-the
ndor
como
estuviera
montando.
as
was
iding’

��ey tie the condor’s leg as if it was riding.’
Camacho, Paredes & Sánchez 1995
, p. 135)
In this sentence the clitic is coindexed with the genitive expression
del condor
ₑof
the condor鈠and not with the whole direct object
su pata del condor
ₑthe condor’s
leg鈠as it would be expected if this were a case of direct object clitic doubling. is
�iliana Sánchez
*Lo-/*le-
oqué
ano
del
iño
ᘒ4
9ᘔ
touched
the
hand
f-the
boy

�I touched the hand of the boy.’
Camacho, Paredes & Sánchez 1995
, p. 135)
Camacho, Paredes and 叡nchez (1995
) show evidence from
-extraction that the
clitic
is not associated with the full direct object DP but only with the genitive
expression
del condor
‘of the condor鈠as it can coexist with the
-extraction of the
possesed constituent, as shown by the question in (21). is is consistent with a
more general pattern of clitic doubling found in L2 Spanish:
Q痩
�arran
del
cdor

what
ᘒ4
tie
f-the
condor]

�What (part) of the condor do they tie?’
Camacho, Paredes & Sánchez 1995
, p. 136)
�e emergence of this possessor clitic is indicative of a process of activation of
L1 genitive features and of their association with L2 pronominal clitics. What
makes the reconguration or feature reassembly possible is the fact that a simi
lar structure can be found with non-direct objects in General Spanish (
Camacho,
Paredes& Sánchez 1995
Kalt 2012
b). In fact,
Kalt (2012
b) notes that, in general
Spanish, structures with oblique or applicative clitics are characterized by a chain
in which the clitic is related to a possessor but is preceded by the prepositon
ₑt澒
and not by the preposition
ₑof’, as shown by the grammaticality of (22) and the
ungrammaticality of (23):
Mar
�obó
el
inero
uan
Maria
9ᘔ
stole
the
money
to
Juan

�Maria stole the money from Juan.’
*Mar
�obó
inero
Juan

Maria
9ᘔ
stole
the
money
of
uan

�Maria stole Juan’s money.’
Kalt 2012
b, p. 174)
Following
Masullo (1992
), Kalt proposes that, in General Spanish, some oblique
osslinguistic inuences in Quechua-Spanish bilingualism
Camacho, Paredes and Sánchez (1995
) propose that the doubling in (19) and (21)
is possible because the clitic is marked in the L2 for genitive features. e crucial
aspect of this type of proposal is that in L2 Spanish the preposition
is not a case
assigner but the spell out of genitive features. e reassembly of features in this
case involves the mapping of the Quechua case marking sux -
shown in (24)
onto the Spanish preposition
deprived of its case assigning features.
Kuntur-pa
chaki-n.
condor-

��e condor’s leg.’
Camacho, Paredes & Sánchez 1995
, p. 142)
Ὡs possibility combined with the availability of preverbal dative clitics in dou
bling structures in the L2 allows for the mapping of genitive features onto an dative
clitic, a feature not associated with clitics in general Spanish.
Mi
�re
le
�arⷡ
libro
Ana
My
mom
9ᘔ
give-3.
ਅa
the
book
(to
Ana)

�My mom will give the book to Ana.’
Mi
�re
�arⷡ
ibro
na
para mí

My
mom
give-3.
ਅa
the
book
to
Ana
(for me

�My mom will give Ana the book for me.’
Mama-y
na-man
libru-ta
aywa-pu-wa-nqa.
Mom-1
8ᄏ
Ana-
9ᘔ
book-
give-
ᄉj
ਅa

�My mom gave Ana the book for me.’
Kalt 2012
b, p. 175)
Example (27) shows that the aspectual sux -
introduces a benefactive mean
ing and coexists with the marking of ᱲst person object features on the verb (-
�ese features are not associated with an indirect object but with an oblique object
�iliana Sánchez
intransitive verbs and transitive verbs with reexive and oblique objects expressed
by clitics using a picture selection task in which the children listened to a sentence
and then were asked to choose from several pictures the one that r攞ected the
meaning of the sentence. She also asked the children to describe pictures. She
tested sentences with and without clitics. She found that Quechua-Spanish bilin
Ana
pone
�hompa.
9ᘔ
puts
�he
sweater

�Ana puts his sweater on.’
Kalt (2012
Ana
chumpa-ta
hura-0-n.
Ana
sweater-
ᘒ4
put-3.
ᄉj
eԉj

�Ana put the sweater (on somebody specied in discourse/ on somebody
unspecied/ there/ somewhere).’
Kalt 2012
b, p. 183)
�ese data point in the direction of a process of functional interference under
stood as the activation of features from the L1 such as oblique or locative case and
their mapping onto L2 morphology even in contexts in which the clitic
is only
the spell out of dative features in general Spanish.
�e emergence of non-argumental clitics such as genitive clitics, oblique and
locative clitics in the L2 Spanish of Quechua speakers are processes that highlight
osslinguistic inuences in Quechua-Spanish bilingualism
the relevance that activation of L1 features and their mapping onto L2 morphology
has in the development of L2 syntax.
Crosslinguistic inuence at the syntax-morphology interface:
Functional interference, feature reassembly and functional
convergence in tense, aspect and evidentiality
Functional interference and feature reassembly can also be found in the tense,
aspect and evidentiality systems.
叡nchez (2004
) presents evidence of func
tional interference and convergence in a study of oral production of 38 Southern
Quechua-Spanish bilingual children (ages 10–16) for whom Spanish is a second
language. In the Spanish spoken by the children there is evidence of the mapping
of reported evidentiality and past tense features onto the Spanish morphological
forms that spell out aspect and tense features. In Quechua, past tense features
Manku
�apaq-qa
itiqaqa
qucha-manta-s
luqsimu-sqa.
Manku
Qhapaq-
ᐑ8
Titikaka
ake-
ጌ9ጕ
emerge-3.
8༔
ᔋ8

�Manku Qhapaq emerged from the Titicaca Lake.’
Cusihuamán 2001
, p. 161)
�e sux
-sqa
Huwan-mi
ariya-ta
qhawa-rqa-n.
Huwan-
ഓ�
Mariya-
ᘒ4
see-
᜖ea

�Huwan saw Mariya.’
叡nchez 2004
, p. 149)
In Spanish, on the other hand, past tense and aspectual features are mapped onto
single morphemes as shown in (32–33).
Compr-é.
Buy-1.
8༔
ᜋ�f

�I bought.鈠⡰erfective)
Compr-aba.
Buy-1.
8༔
ጆ8କf

�I used to buy.’ (imperfective)
�iliana Sánchez
叡nchez (2004
) notes that the imperfective forms of Spanish share some meanings
Anda, ¡sabía nadar!
Hey, knew to swim

�Hey, (s/he) knew how to swim!’
叡nchez 2004
, p. 149)
叡nchez (2004
) proposes that this partial similarity serves as a trigger to generate
Había
na
viejita,
ice.
Have-
8༔
ጆ8କf
an
�ld woman,
say

ere) was an old woman, (they) say.’
Había
embr-ado
ma.
Have-
8༔
ጆ8କf
sow-
᜖�a
corn

e) had sowed corn.’
Ella
sembr-aba
a.
She
sow-
8༔
ጆ8କf
corn

�She sowed corn.’
Le-
�a戭ía
contr-ado
un
pajarito
amarillo.
9ᘔ
have-
8༔
ጆ8କf
ᱮd-
8ᘕa
birdy
ellow

�She found a yellow bird.’
�e-
abía
lev-ado
su
asa
and
9ᘔ
have-
8༔
ጆ8କf
take-
8ᘕa
to
her
home

�And (she) took him home.’
le
abía
cho
And
9ᘔ
have-
8༔
ጆ8କf
make-
᜖�a
osslinguistic inuences in Quechua-Spanish bilingualism
comer
trigo,
gua,
an.
eat
�heat,
water,
bread

�And (she) made him eat wheat, water and bread.’
叡nchez 2004
, p. 157–158)
Ὡs sequence of sentences is not typical of a Spanish narrative. Firstly, it lacks
convergence in features.”
叡nchez 2004
, p. 150)
�e FHC predicts that among Quechua-Spanish bilinguals the activation of evi
dentiality features in Quechua aiects bilingual Spanish representations and activa
tion of aspectual features in Spanish aiects the bilingual Quechua representations.
Evidence for the mapping of evidentiality onto Spanish morphology was found in
the consistent use of pluperfect and imperfective morphemes in the Spanish data.
�e bilingual Quechua results also exhibited convergence because rather than
showing a consistent use of reportative past forms, attested past morphemes were
also found in the narratives. eir distribution was similar to the distribution of
perfective and imperfective forms in Spanish.
Like the emergence of oblique clitics, the emergence of reportative evidential
ity in the Spanish of Quechua speakers is also evidence in favor of a view of cross
linguistic inuence that is triggered by the activation of functional features in both
languages and a common pattern of feature-morphology mapping.
Crosslinguistic inuence at the syntax-lexicon interface: Functional
interference, functional convergence, aspect and argument structure
Another piece of evidence in favor of functional interference and convergence can
also be found in the mapping of some grammaticalized Quechua aspectual fea
associated with
olitive and
�iliana Sánchez
imminent modal/aspectual features is mapped onto the Spanish modal verb
querer
‘to want’⸠�e study of picture-based narratives of 30 Lamas Kechwa-Spanish bilin
Miku-naya-yka-n.
Eat-
ഋe
᜕0m

e) wants to/is about to eat.’
叡nchez 2006
, p. 540)
Está
queriendo
comer.
Is
want-
eat-

e) wants to eat/She is about to eat.’
�e latter form conveys a desiderative/imminent aspectual meaning absent in the
narratives of a comparison group of 25 Spanish-speaking children. e results of
the study showed evidence of functional interference⸠�e narratives showed that
bilingual children used the desiderative form in Quechua:
Sap-itu
rma-naya-n
yaku-pi.
Toad-
ഓm
fall-
ഋe
water-
ฑ4

��e toad wants/is about to fall into the water.’
叡nchez 2006
, p. 544)
�ey also used it with progressive forms as in:
Kay
�chku
uku-chi-naya-yka-n
ay sap-itu-ta.
Ὡs
it-
ሖԏ
ഋe
᜕0m
toad-
ഓm
ᘒ4

�Ὡs dog is wanting to have this toad bit.’
叡nchez 2006
, p. 544)
In the Spanish narratives, modal progressive forms with a desiderative/imminent
reading were found:
Un
wamr-illu
stá
queriendo
agarrar
apo.
boy-
ഓm
�nting
o) grab
his
toad

�A boy wants/is about to grab his toad.’
叡nchez 2006
, p. 545)
In these cases, the progressive marker contributes to the imminent reading of
Spanish. Notice that, in Quechua, the grammaticalized aspectual imminent fea
of these
�uechua-speaking children there is mapping of both features onto an
osslinguistic inuences in Quechua-Spanish bilingualism
independent modal verb in Spanish. is type of mapping in its imminent inter
Evidence of crosslinguistic inuence at the syntax/pragmatics interface
In addition to new patterns of feature-morphology mapping, the study of the L2
Spanish of Quechua speakers has also shown evidence of transfer of mapping
strategies at the syntax/pragmatics interface involving the spell out of functional
features.
Camacho, Paredes and Sánchez (1997
) have shown that the Spanish
of Southern Quechua speakers is characterized by the pervasive nature of null
objects with denite antecedents. In their study of 33 adult L1 Southern Quechua-
L2 Spanish speakers, they found evidence of null objects with denite antecedents
in the speech of all prociency subgroups. e following are examples of the types
¿Y Uds. preparaban el desayuno
para los pensionistas?

‘And did you prepare breakfast for the guests?’

Sí, mis hermanas e
preparaban.

‘Yes, my sisters prepared it.’
뽅xtrañas mucho a tu papa

‘Do you miss your father?’

Sí, se
extraño.

‘Yes, yes I miss him.’
¿Q痩
pasó?- me dijo- Lo
había atropellado una moto

‘What happened? -he told me-. A motorcycle has run her over.’

le dije, entonces:

‘I told him, then:’
�iliana Sánchez

Hemos llevado e
ya al Hospital del Niño, ya e

‘�en we took her to the hospita氻’

hemos llevado al Emergencia

We took her to the Emergency room.’
¿Q痩
hace, la mata
el lobo
o no la mata a la oveja

‘What does the wolf do? Does it kill the sheep?’

Smata e
, smata e

‘Yes, it kills (it), yes it kills it.’
Camacho, Paredes & Sánchez 1997
, p. 59)
Notice that, in all these cases, the antecedents of null objects were denite.
Camacho, Paredes, and Sánchez (1997
) note that the pervasive nature of null
objects can be accounted for by a topic operator that licenses the null object. is
type of licensing takes place despite the availability of clitics in the L2 Spanish of
these speakers as exemplied in sentences (48a) and (48b). In sentence (48a), the
discourse topic is introduced as a clitic, but a1erwards, a null object is used to refer
to that topic (48b). is type of licensing is also found in Quechua as illustrated in
sentences (5a) and (5b). Notice that the null object cannot be accounted for by a
general absence of clitics in L2 Spanish. I would like to propose that, in these cases,
there is a mapping strategy that is transferred from one language to the other.
�ere is no overt object third person verbal morphology in Quechua and a null
object appears in contexts in which there is a third person topic object as in(5b).
It is the mapping strategy of Quechua, namely the lack of third person object overt
morphology that is at the source of the denite null objects in L2 Spanish. Inde
pendent evidence for such a mapping has also been found in the L2 Spanish of
Chinese speakers (
Cuza, Pez-Leroux & Sánchez 2013
). In this case, crosslinguis
tic inuence is shown by the absence of third person object morphology in the L2
verb following the L1 patterns in null topic contexts.
Evidence of null objects with denite antecedents in the Spanish of two
groups of Central Quechua-Spanish and Southern Quechua-Spanish bilingual
children (ages 9–13) was also found by
叡nchez (2003
⤮‟e picture-based nar
ratives of both groups of bilingual children showed that the second most fre
quent type of direct object in Quechua was a null object, the most frequent type
being an overt DP. is was not the case in the Spanish narratives. ese had a
higher frequency of clitics and structures with clitic doubling than of null objects.
Null objects, however, had overt denite DPs previously mentioned in the text as
their antecedents and also antecedents not present in the text but shown in the
pictures, a pattern also found in the Quechua narratives. is pattern is consis
tent with the idea that the null objects correspond to a discourse salient element,
namely, a topic.
osslinguistic inuences in Quechua-Spanish bilingualism
I would like to discuss a nal piece of evidence of crosslinguistic inuence
at the interface of syntax and the pragmatic component reported in a study on
word order, focus, and weak crossover in the Spanish of adult Quechua speak
ers (
Muntendam 2013
Mayqen
wawa-ta-taq
mama-n

which
child-
7ርq
mother-3
8ᄏ
apa-mu-chka-n
achay wasi-man?
bring-
ഓ�
᜕0m
R-3
school-
ഓ�

�Which child does his mother bring to school?’
Sapa
wawa-ta
mama-n

each
child-
ᘒ4
mother-3
8ᄏ
apa-mu-chka-n
achay wasi-man.
bring-
ഓ�
᜕0m�
school-
ഓ�

�His mother brings each child to school.’
qué
�ño
trajo
�u
madre
scuela?

To
which
boy
rought
his
mother
to
the
school

�Which child did his mother bring to school?’
�a
ni
trajo
�u
madre
scuela.

To
each
�hild
rought
his
mother
to
the
school

�His mother brought each child to school.”
Muntendam 2013
, p. 119)
Muntendam (2013
) also reports results of other tasks but I will only discuss the ones rel
evant to the argument presented.
�iliana Sánchez
�e results of the grammaticality judgment task showed that no bilingual speak
ers showed sensitivity to sentences with focus fronting (48) and questions (47) in
Quechua revealing no weak cross over攝ects. However, in Spanish, 10 out of the
15 speakers showed sensitivity to weak crossover 攝ects with questions, and 11 out
of 15 showed sensitivity with focus fronting sentences in a proportion almost as
high as that of the control groups of speakers of non-Andean Spanish. At the same
time, the naturalistic data examined showed 18.5 % of OV structures, a percentage
Concluding remarks
In this paper, I have discussed evidence of crosslinguistic inuence that involves
the activation of functional features that generates new patterns of feature-
morphology mapping across languages that includes crosslinguistic inuence at:
⡡⤚the syntactic level (feature reassembly and the emergence of non-argumental
clitics) (b) the syntax/lexicon interface (feature reassembly and changes in aspec
tual congurations and argument structure), and (c) the syntax/pragmatics inter
face (licensing of null objects and focus fronting but no weak crossover eects). At
the syntactic level, feature reassembly is represented in the emergence of new non-
argumental clitics in Quechua-Spanish bilinguals because Quechua oblique case
marking is mapped onto the Spanish dative clitic
. is new mapping allows for
an extension of verbal agreement from arguments to non-argumental
nstituents
osslinguistic inuences in Quechua-Spanish bilingualism
such as benefactives. is in turn generates a new syntactic conguration in which
rather than being merged as prepositional phrases such as
para/por mi
ₑfor mi鈠as
in Example (26), benefactives are directly merged as morphological markers on
. In J. M.
�uthier,
B. Bullock, & L. Reed (Eds.),
Formal Perspectives on Romance Linguistics
(pp⸚21–37).
Amsterdam: John Benjamins.
DOI: 10.1075/cilt.185.04ble
Camacho, J. (1999). From SOV to SVO: e grammar of interlanguage word order.
Second Lan
guage Research, 15
DOI: 10.1191/026765899673532714
Camacho, J., & 叡nchez, L. (2002). Explaining clitic variation in Spanish. In M. Amberb敲2&
P. Collins (Eds.),
Formal Approaches to Language Universals and Language Variation
(pp.21–40). Norwood, CT: Ablex.
�iliana Sánchez
Camacho, J., Paredes, L., & Sánchez, L. (1995). e genitive clitic and the genitive construction
in Andean Spanish.
Probus, 7
DOI: 10.1515/prbs.1995.7.2.133
Camacho, J., Paredes, L., & Sánchez, L. (1997). Null objects in Bilingual Andean Spanish. In
䔮2Hughes, M. Hughes, & A. Greenhill (Eds.),
Proceedings of the 21st
annual Boston Univer
sity Conference on Language Development, Vol 1.
(pp. 55–66). Somerville: MA: Cascadilla
Press.
Cerrón-Palomino, R. (1987).
Lingﳭstica Quechua
. Cusco, Peru: Centro de Estudios Rurales
Andinos Bartolomé de las Casas.
Chomsky, N. (2000). Minimalist inquiries: e framework. In R. Martin, D. Michaels, &
J⸚Uriagereka (Eds.),
Step by Step: Essays on Minimalist Syntax in Honor of Howard Lasnik
(pp. 89–155). Cambridge, MA: e MIT Press.
DOI: 10.1017/s0022226702271625
Cusihuamán, A. (2001).
Gramática Quechua Cuzco-Collao
. Lima: Instituto de Estudios Peruanos.
Cuza, A., Pez-Leroux, A.T., & Sánchez, L. (2013). e role of semantic transfer in clitic-drop
among Chinese L1-Spanish L2 bilinguals.
Studies in Second Language Acquisition,
DOI: 10.1017/s0272263112000691
Escobar, A.M. (1991).
Los bilingües en el Perú
. Lima, Peru: Instituto de Estudios Peruanos.
DOI: 10.1017/s0022216x00000882
Dordrecht: Reidel.
DOI: 10.1007/978-94-009-3387-3_10
Giorgi, A., & Pianesi, F. (1997).
Tense and Aspect
. Oxford: OUP.
DOI: 10.1515/ling.2003.019
Gﱲel, A.
(2004). Selectivity in L2-induced L1 attrition: A psycholinguistic account.
Journal of
Neurolinguistics,
DOI: 10.1016/s0911-6044%2803%2900054-x
Hintz, D. (2011).
Crossing Aspectual Frontiers
. Los Angeles, CA: University of California Press.
Hopp, H. (2009). e syntax-discourse interface in near-native L2 acquisition: O-line and on-
line performance.
Bilingualism, 12
DOI: 10.1017/s1366728909990253
DOI: 10.1177/0267658312461803
ntax
and pragmatics.
Bilingualism: Language and Cognition,
DOI: 10.1017/s1366728900000353
Jackendoi, R. (1997).
se Architecture of the Language Faculty
. Cambridge, MA: e MIT Press.
Jackendoi, R. (2002).
Foundations of Language: Brain, Meaning, Grammar, Evolution
. Oxford:
OUP.
DOI: 10.2298/j0561213i
Kalt, S. (2000). Non-direct object agreement in the second language Spanish of Southern
Quechua speakers. In B. Swierzbin, F. Morris, M.E. Anderson, C.A. Klee, & E. Tarone
(Eds.)
Social and Cognitive Factors in Second Language Acquisition: Selected Proceedings
of the 1999. Second Language Research Forum
(pp. 222–242)
Somerville, MA: Cascadilla
Press.
Kalt, S. (2002).
Second Language Acquisition of Spanish Morpho-Syntax by Quechua-Speaking
Children.
Unpublished Ph.D. dissertation. University of Southern California at Los Angeles.
Kalt, S. (2009). Bilingual children’s object agreement and case marking in Cusco Quechua. In
䠮2Bliss & E. Gerard. (Eds.),
University of British Columbia Working Papers in Linguistics,
, (pp. 126–142). Vancouver, BC: University of British Columbia.
Kalt, S. (2012b). Cambios morfosintticos en castellano impulsados porel quechua hablante⸠2In
P. Dankel, V.F. Mallat, J.C. Godenzzi, & S. P曤nder (Eds.),
El espal andino. Espacios comu
nicativos y cambios gramaticales, 41
, (pp. 165–192). Neue Romania.
osslinguistic inuences in Quechua-Spanish bilingualism
Kerke, S.C., van de. (1996).
Ax Order and Interpretation in Bolivian Quechua
. Leiden: CNWS.
Kroll, J.F., & Bogulski, C.A. (2013). Cognitive second language acquisition: Organization of the
second language lexicon. In C.A. Chapelle (Ed.),
se Encylopedia of Applied Linguistics
(pp. 4322–4330). Oxford: Blackwell.
DOI: 10.1002/9781405198431.wbeal0886
Kroll, J.F., & Stewart, E. (1994). Category interference in translation and picture naming:
Journal
of Memory and Language
DOI: 10.1006/jmla.1994.1008
Kroll, J.F., Van Hell, J.G., Tokowicz, N., & Green, D.W. (2010). e Revised Hierarchical Model:
A critical review and assessment.
Bilingualism: Language and Cognition,
DOI: 10.1017/s136672891000009x
Lardiere, D. (1998). Dissociating syntax from morphology in a divergent L2 end-state grammar.
Second Language Research,
DOI: 10.1191/026765898672500216
Lardiere, D. (2003). Second language knowledge of [±Past] vs. [±Finite]. In J.M. Liceras,
䠮2Zobl, & H. Goodluck. (Eds.),
Proceedings of the 6th
Generative Approaches to Second
Language Acquisition Conference
GASLA 2002
) (pp. 176–189). Somerville, MA: Casca
dilla Proceedings Project.
Proceedings Project.
DOI: 10.1177/0142723711403980
Marull, C. (2013, April). Language-specic syntax constrains cross-linguistic activation. Paper
presented at the 43rd Linguistic Symposium on Romance Languages, CUNY Graduate
Center, New York.
Masullo, P. (1992).
Incorporation and Cas攠seory in Spanish: A Crosslinguistic Perspective
Unpublished Ph.D. dissertation. University of Washington, WA.
Meisel, J.M. (1997). e acquisition of the syntax of negation in French and German: Contrast
ing rst and second language development.
Second Language Research, 13
DOI: 10.1191/026765897666180760
Meisel, J. (2008). Child second language acquisition or successive rst language acquisition? In
B. Haznedar & E. Gavruseva (Eds.),
Current Trends in Child Second Language Acquisition:
A Generative Perspective
(pp. 55–80). Amsterdam: John Benjamins.
DOI: 10.1017/s0272263109990076
Montrul, S. (2010). Dominant language transfer in Spanish L2 learners and heritage speakers.
Second Language Research,
DOI: 10.1177/0267658310365768
Müller, N., & Hulk, A. (2001). Crosslinguistic influence in bilingual language acquisi
tion: Italian and French as recipient languages.
Bilingualism: Language and Cognition,
DOI: 10.1017/s1366728901000116
Muntendam, A. (2013). On the nature of cross-linguistic transfer: A case study of Andean Span
ish.
Bilingualism: Language and Cognition, 16
DOI: 10.1017/s1366728912000247
�iliana Sánchez
Muysken, P. (1995). Focus in Quechua. In K. Kiss (Ed.),
Discourse Congurational Languages
(pp. 375–393). Oxford: OUP.
Paradis, M. (1993). Linguistic, psycholinguistic, and neurolinguistic aspects o映“interference” in
bilingual speakers: e activation threshold hypothesis.
International Journal of Psycholin
guistics
Paradis, M. (2009).
Declarative and Procedural Determinants of Second Languages
. Amsterdam:
John Benjamins.
DOI: 10.1017/s0272263110000306
Paradis, J., & Genesee, F. (1996). Syntactic acquisition in bilingual children.
Studies in Second
Language Acquisition
DOI: 10.1017/s0272263100014662
Pladevall, E. (2010). Child L2 development of syntactic and discourse properties of Spanish
subjects.
Bilingualism: Language and Cognition
DOI: 10.1017/s1366728909990447
�rolegomenon to modeling heritage language grammars.
Linguistic Approaches to Bilin
gualism
DOI: 10.1075/lab.3.4.04put
叡nchez, L. (2003).
Quechua-Spanish Bilingualism: Functional Interference and Convergence
Amsterdam: John Benjamins.
DOI: 10.1075/lald.35
叡nchez, L. (2004). Functional convergence in the tense, evidentiality and aspectual systems of
Quechua-Spanish bilinguals.
Bilingualism: Language and Cognition
DOI: 10.1017/s136672890400149x
叡nchez, L. (2006). Kechwa and Spanish bilingual grammars: Testing hypotheses on functional
interference and convergence.
International Journal of Bilingual Education and Bilingual
ism
DOI: 10.2167/beb379.0
叡nchez, L. (2010).
se Morphology and Syntax of Topic and Focus: Minimalist Inquiries in the
Quechua Periphery
. Amsterdam: John Benjamins.
DOI: 10.1075/la.169
Schwartz, A., & Kroll, J. (2006). Bilingual lexical activation in sentence context.
Journal of Mem
ory and Language,
DOI: 10.1016/j.jml.2006.03.004
Schwartz, B.D., & Sprouse, R.A. (1996). L2 cognitive states and th攠“full transfer/full access鐠
model.
Second Language Research
DOI: 10.1177/026765839601200103
Serratrice, L., Sorace, A., & Paoli, S. (2004). Crosslinguistic inuence at the syntax-pragmatics
interface: Subjects and objects in Italian-English bilingual and monolingual acquisition.
Bilingualism: Language and Cognition
DOI: 10.1017/s1366728904001610
Slabakova, R. (2008).
Meaning in the Second Language
. Berlin: Mouton de Gruyter.
DOI: 10.1017/s0272263110000124
Sorace, A. (2000). Syntactic optionality in L2 acquisition.
Second Language Research
DOI: 10.1191/026765800670666032
Sorace, A. (2005). Syntactic optionality at interfaces. In L. Cornips & K. Corrigan (Eds.),
Syntax
and Variation: Reconciling the Biological and the Social
(pp. 46–111). Amsterdam: John
Benjamin.
DOI: 10.1017/s1360674307002316
Sorace, A., & Serratrice, L. (2009). Internal and external interfaces in bilingual language devel
opment: Beyond structural overlap.
International Journal of Bilingualism, 13
DOI: 10.1177/1367006909339810
Ullman, M. (2001). e neural basis of lexicon and grammar in rst and second language: e
declarative/procedural model.
Bilingualism: Language and Cognition
DOI: 10.1017/s1366728901000220
osslinguistic inuences in Quechua-Spanish bilingualism
Ullman, M. (2005). A cognitive neuroscience perspective on second language acquisition:
�e declarative/procedural model. In C. Sanz (Ed.),
Mind and Context in Adult Second
Language Acquisition: Methods, seory, and Practice
(pp. 141–178). Washington, DC:
White, L. (1989).
Second Language Acquisition and Universal Grammar
. Amsterdam: John
Benjamins.
DOI: 10.1177/026765839100700306
White, L. (2003).
Universal Grammar and Second Language Acquisition
. Cambridge: CUP.
DOI: 10.1017/cbo9780511815065.002
White, L. (2011). Second language acquisition at the interfaces.
Lingua,121
DOI: 10.1016/j.lingua.2010.05.005
Verbal agreement in the L2 Spanish
of speakers of Nahuatl
Alma P. Ra淭rez-Trujillo
& Joyce Bruhn de Garavito
Emory and Henry College &
�e University of Western Ontario
Ὡs paper examines the production of subject/verb agreement in the L2
Introduction
Ὡs paper describes the interlanguage grammar of second language (L2) learn
ers of Spanish whose rst language (L1) is Nahuatl, an indigenous language spo
indigenous peoples in Mexico and other parts of Latin America either
licitly
We wish to than欠�e University of Western Ontario for an International Research Award
to Joyce Bruhn de Garavito and a New Research and Scholarly Initiative Award (ADF) to
Joyce Bruhn de Garavito and Silvia Perpiñán.‟e data reported on here form part of Alma P.
Ramírez-Trujillo’s Ph.D. thesis.
We also wish to express gratitude to the people of San Isidro Buensuceso for their patient
help and hospitality, particularly Don Fidencio and his extended family. Finally, thanks are
due to Jeremías Cabrera who produced the translations to and from Nahuatl.
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
Lastra (1992
the standard
panish of the region, while those that do not have this contact will
erbal agreement in the L2 Spanish of speakers of Nahuatl
Sociolinguistic background
Mexicano in this region. When we visited, around 10 years a1er the publication
of Hill and Hill’s book and perhaps 20 years a1er the original research was carried
out, we found villages where adults spoke to each other in Nahuatl and to their
children in Spanish, a sure sign that the language is endangered. In contrast, in
San Isidro Buensuceso, where our participants live, Nahuatl is the language of the
village, heard everywhere, and the speakers are proud of their language, although
they do not seem to have any idea that it was once the language of an empire.
�e economic situation described by Hill and
Hill (1986
) has not improved.
It is not possible for people to continue to live oᴠthe land as they had in the past,
and most younger members of the village now travel to the large city of Puebla
to work. is situation of economically-forced migration is r攞ected across the
country; in fact, one of the greatest concentrations of Nahuatl speakers is found
in Mexico City. is means that Spanish is becoming more and more important,
leading to a generational divide. In San Isidro, the older members of the commu
nity are either monolingual in Nahuatl or second language learners of Spanish.
Middle-aged people may also be second language learners of Spanish, but they will
have a great deal more contact with the language⸠�e younger generation is o1en
bilingual (See
Ra淭rez-Trujillo 2013
�e situation regarding schooling is also complicated, as our research
has shown. Many of the participants in our study are illiterate or semi-literate.
Elementary school gives children the opportunity to learn to read and write
Nahuatl, although not in all classes. High school is generally in Spanish. However,
few members of the community, particularly girls, nish high school, and of the
older generations few even nished elementary school.
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
It is clear from the above that these speakers are not used to being asked about
their language or to being tested (see
Rice, Libben, & Derwing (2002
) for a dis
cussion of testing illiterate speakers). Within the generative approach, besides
the authors of the present paper, only work by
MacSwan (1998
), who
looked at code switching, has considered the structure of Nahuatl. However, he
did not conduct an experiment; rather Nahuatl speakers served as informants
regarding code-switching intuitions. Given the population studied and the rst
language of the speakers, the present paper is clearly breaking new ground, not
only because of the language combination but also because this paper is one of the
Agreement in Spanish and Nahuatl
Spanish
Spanish is a pro-drop language that exhibits rich subject agreement on the verb.
In fact, all verbs in Spanish are bound roots that must attach to either a non-nite
ending or to person and tense s甛xes. e innitive, for example, is made up of
the bound root (e.g.
compr
ⴠ‘buy’), the thematic vowel (
compr-a
), and the inni
tival ending -r (
compr-a-r
‘to buy’). Tense s甛xes generally do not attach directly
to the verb but rather to the thematic vowel that is used by linguists to categorize
verbs into rst, second or third class. e most common class has -
as the the
matic vowel, with around 80% of verbs belonging to it. e second class thematic
vowel is -
, and the third is -
. In these two classes the thematic vowel is not as con
sistent as in the rst class, as you see in Table 1. Table 1 illustrates the innitive and
the conjugation in the present tense indicative of the three classes, Table 2 provides
the simple past. In each case we nd the root, followed by the thematic vowel,
with the tense, aspect and mood (TAM) morphology following, and the person/
number agreement morphemes in nal position. We have excluded the second
person plural as this form is not used in Latin America. It is important to note
that the third person singular of the present tense ends only in the thematic vowel,
with no overt person agreement. It is for this reason that it has been argued that
the third person is the default (
Bruhn de Garavito 2003
McCarthy 2006
erbal agreement in the L2 Spanish of speakers of Nahuatl
Table 1.
Present tense of three classes of verbs
Innitive (stem + theme
vowel + innitival ending)
compr-a-r (to buy)
com-e-r (to eat)
viv-i-r (to live)
Person (pronoun)
Stem + theme vowel + person/number
(yo)
compr-o
com-o
viv-o
(t﨩
compr-a-s
com-e-s
viv-e-s

él/ella)/2
formal (usted)
compr-a
com-e
viv-e
(nosotros/as)
compr-a-mos
com-e-mos
viv-i-mos
(ellos/ellas)/2
formal and informal (Uds.)
compr-a-n
com-e-n
viv-e-n
Table 2.
Innitive (stem + theme
vowel + innitival ending)
compr-a-r
com-e-r
viv-i-r
Person (pronoun)
Stem + theme vowel + TAM + Person/Number
(yo)
compr-é
co洭í
vi瘭í
(t﨩
compr-a-ste
com-i-ste
viv-i-ste

él/ella)/2
formal (usted)
compr-ó
com-i-ó
viv-i-ó
(nosotros/as)
compr-a-mos
com-i-mos
viv-i-mos
(ellos/ellas)/2
formal and informal (Uds.)
compr-a-r-on
com-ie-r-on
viv-ie-r-on
Although the present tense may be used in Spanish to refer to the present time
Miro la tele
ₑI am watching TV’) in many places the present progressive is used
instead (
Estoy mirando la tele
). is tense, which does not have identical uses to
the English progressive but is somewhat similar, is formed with the auxiliary
estar
‘to be鈠and the gerund of the main verb. We will see below that the present progres
sive was o1en used by our participants instead of the simple present.
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
in subject position with which the verb agreed. Recent reformulations within the
Minimalist Program (
Chomsky 1995
) argue that
pro
is an unnecessary category.
Rather, verbal agreement s甛xes are considered pronominal elements that include
the feature [+D] which checks the EPP
feature by merging with the root of the
verb (
Speas 1994
Alexiadou & Anagnostopoulou 1998
). is is illustrated in (1).
It is interesting to note that this brings Spanish more in line with some analyses
regarding Nahuatl, as we will see below.
comemos

come
-mos
&#x/MCI; 38;3 0;&#x/MCI; 38;3 0;VP&#x/MCI; 38;4 0;&#x/MCI; 38;4 0; [&#x/MCI; 38;5 0;&#x/MCI; 38;5 0;DP&#x/MCI; 38;6 0;&#x/MCI; 38;6 0; mos
&#x/MCI; 38;9 0;&#x/MCI; 38;9 0;V&#x/MCI; 38;@ 0;&#x/MCI; 38;@ 0; come]]]
In those cases in which the subject pronoun is expressed, it is assumed that it is in
topic position or somehow focalized (
Mallén 1992
Nahuatl
EPP or Extended Projection Principle originally referred to the requirement that a Tense
Phrase project a Speci�er (the subject). Currently this requirement is said to be due to unin
terpretable EPP features on T that attract the subject to its speci�er.
�e paradigms and examples used here were collected by the authors in the village of San
Isidro Buensuceso with the help of informants.
erbal agreement in the L2 Spanish of speakers of Nahuatl
Wedo not include in Table 3 the honoric forms as they are beyond the scope
of this paper.
Table 3.
Present tense of intransitive (cry) and transitive (have) verbs in Nahuatl
Pronouns
Cry
Have a house
Subject person agreement +
stem (+number)
Subject person agreement + object
agreement + stem (+number)
(ne)
�hoka
༉j
�y
ia
lli
༉j
ᄉj
ᘐ�
�have
house
(te)
ti
-choka
༉j
-cry
ti
ia
kalli
༉j
ᄉj
ᘐ�
�have
house
(ye)
�oca
�y
�i
ia
kalli
ᄉj
ᘐ�
-have
�ouse
(tehuan)
ti
�hoca
༉j
�y
ti
ia
kalli
༉j
�.
ᄉj
-have
house
(amehuan)
An
�hoca
༉j
-cry
An
�i
ia
kalli
༉j
ᄉj
ᘐ�
�ave
house
(yehuan)
�hoca
cry
�i
ia
kalli
ᄉj
ᘐ�
�ave
house
�e past tense is formed by the addition of the prex
to the le periphery
of the verb and its obligatory agreement morphemes. In other words and simplify
ing somewhat, we nd tens�e subject agreemen�t object agreemen�t verb root,
as illustrated in (2). In short, the tense morpheme precedes the person agreement
morphemes, and all are pr攜xes. e morpheme denoting plurality goes at the
end⸠�ere are some s甛xes that have traditionally been associated with tense, for
example the sux
is said to signify future. We believe, however, that it is prob
ably a marker of
irrealis
mood.
ia
me
kaltin
8༔
ᄉj
7ပ
-have
two
ouses

�You had two houses.’
�ere is a great deal of variability regarding the spelling in Nahuatl. It is still common to
use Spanish spelling, but for some there is an eiort to move away from this.‟e word ‘house’,
for example, can be found as
calli
(Spanish spelling)
or
kalli
, as we have spelt it here.
Pronouns can be realized in diierent forms according to the degree of formality. For
example, th攠�rst person ‘ne’ (or neh) is produced as ‘nehua’ in a more formal setting or
‘nuhuatzin’ to show respect.
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
and also as suggested by an anonymous reviewer, the two languages,
panish
erbal agreement in the L2 Spanish of speakers of Nahuatl
and it may well be the case that the languages 摩ier to such an extent that the
Second language acquisition of agreement and tense
Assuming we are on the right track, what does the Nahuatl learner of Spanish L2
Garavito 2009
�e acquisition of agreement and tense morphology has been the subject of a
great deal of research, both within the domain of DP and within the domain of TP.
�e one thing that is generally agreed upon is that second language learners have
problems producing accurate morphology and, although they exhibit fewer errors
in comprehension, comprehension is not perfect either (
Lardiere 1998a
White, Valenzuela, Kozlowska-Macgregor, & Leung 2004
⤮‟e ques
tion that arises revolves around the causes for the problems related to morphology,
or, put another way, what the lack of production and perhaps comprehension of
accurate morphology in second language learners tells us about the mental repre
sentation of second languages.
�e answer to the above question is roughly represented by two approaches
(see
White 2003
for a comprehensive summary of this debate). On the one hand
it is claimed that morphology drives syntax (
Rohrbacher 1999
) and, therefore, a
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
decit in explicit morphology is the reex of a decit at the representational level.
Among others,
Meisel (1997
) argues that Universal Grammar (UG) is not
available in adult L2 and, as a result, the mental representation and the acquisi
tional process are dierent for older learners. Other scholars argue for a local
ized impairment. According to Hawkins and colleagues (
Hawkins & Franceschina
Hawkins & Hattori 2006
; see also
Tsimpli & Dimitrakopoulou 2007
), second
language learners are not able to instantiate in their second languages features that
are not present or are valued dierently in their L1. More recently,
Hawkins (2009
suggests that L2 learners, unlike L1 learners, rely on statistical information exclu
sively, which explains, for example, their reliance on a default in their production
of features such as gender concord.
Hawkins and Casillas (2008
), which looked at
verbal morphology in L2, claims that, although L2 grammars are organized in the
same way as rst language grammars, at the initial stages L2 vocabulary entries
摩ier because the overt morphological items are ident検ed by way of the nodes
with which they co-occur, and not, as in native speaker grammars, activated by the
nodes themselves into which they are inserted. Evidence for this claim is the pro
duction of
be + bare V
(I am read)
in early stages of acquisition. In more advanced
erbal agreement in the L2 Spanish of speakers of Nahuatl
learners have many more cues to identify lexical items, such as the written word
and rich context, and therefore need not pay much attention to transitional prob
abilities and co-occurrence relations. As a result, they form weaker associations. In
other words,
�ill Nahuatl early learners of Spanish dier from late learners in their sup
pliance of appropriate subject verb agreement and tense morphology? If age
�ill speakers with substantial contact with native (monolingual) Spanish
speakers dier from those with little contact in their suppliance of appropri
ate subject verb agreement and tense morphology? If quantity and type of
�hat type of errors will the Spanish L2 speakers commit? ere is evidence
that L2 learners rely on the least specied forms as a default (
Bruhn de
Garavito 2003
McCarthy 2007
). In Spanish the least specied form of the
verb is the third person and, interestingly, this is also the case for Nahuatl.
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
Description of participants
It was impossible t漠�nd more early bilinguals in the same village.
erbal agreement in the L2 Spanish of speakers of Nahuatl
language. Organizing the data along these lines results in a grouping that is dif
Description of the testing procedures
Because our interviews showed the participants in this study were mostly illit
entence to be translated:

osllegó de Estados Unidos. ¿Lo quieres ir a ver?

Josarrived from the United States. Do you want to go to see him?’

�.
ranslation given:

hto:s-
neki
law

0ई⸖ပ
-say
nt
-come

e:lpo:chtsi:n
we:noh
tonces
tla:
aw

8ᄏe
-boy
ell
hen
-come

o:tsa,

0ई⸖ပ
8༔
speak,

no:tsa

0ई⸖ပ
past-speak
nd
ere

ta- s

0ई⸖ပ
-see-
ਅa

It means, this boy came, then if he comes, they will tell me and I am
going to see him.’
It is interesting to note that the introduction to the above translation, ‘it means鈠is
in fact a word for word translation of the Spanish
quiere decir
ₑit wants to sa禒. A1er
500 years of contact with Spanish there is a great deal of inuence of this language
on Nahuatl.
On the basis of the translation task we excluded two speakers who were unable
to carry it out, but we were fairly liberal in what counted as an acceptable transla
tion. is is clearly not a ne-grained instrument.
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
Besides the translation task, there were two main tasks that tested subject verb
agreement: an elicited production task and a spontaneous production task. We
only report on the elicited production task due to space limitations (see
Ra淭rez-
Trujillo 2013
for other results).
�e objectives of the production task were to elicit subject verb agreement and
also object pronouns, which we do not report on here. It consisted of 38 pairs of pic
tures, in the rst of which someone was carrying out an action with an object, and
in the second a dierent action with the same object. For example, the speaker saw
the following picture and was asked in Spanish to describe the actions of the person.
�e expected answer was along the lines of ‘She is cutting the melon.鈠�en the per
son saw the second picture and was asked: and a1erwards/before with the melon?
礚después, con el melón?
), with the expected answer ‘She ate it.鈠Because of the way
the second question was framed the response was always expected to refer to the past.
Figure 1.
Elicited Production Task
Of the 38 pairs of pictures, 10 were designed to elicit the rst person singular
(one picture in the present, one in the past), 10 the second person singular and
10 the third person. ere were 8 distracters. In the case of the rst person the
participants were asked to imagine they were in the picture, and in the case of the
second person that the interlocutor was in the picture⸠�ey were told not to use
the formal
usted
given that this would make the second and third person indistin
guishable regarding agreement. Distracters included the verb
gustar
ₑto like’⸠�e
responses were recorded and then transcribed⸠�ey were coded for person and
for agreement, and within agreement for tense. Agreement could be matched, or
Results
erbal agreement in the L2 Spanish of speakers of Nahuatl
bilinguals. e answer is negative. A repeated measures ANOVA
failed to pro
1 Past2 Past3 Past
Figure 2.
Mean accuracy rates for each of the three groups according to sentence type
Results of a repeated measures ANOVA in which the independent variables
are Group (high contact, low contact, control) and Sentence Type (person/tense)
We realize these results are only tentative given the number of subjects in the bilingual
group.
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
shows a main eect for Group (
= .0001), a main eect for the
Sentence Type (
Error types
It is clear from the group and individual results that the highest accuracy rate
for the bilinguals is in the third person, particularly in the past. e high contact
group produced only two errors in the third person past (98% accuracy), and 8 in
the third person present, 6 of which were produced by the same person (L2-19)
(92% group accuracy). e low contact group produced 14 errors in total in the
third person past (90.66% accuracy) and 33 errors in the third person present
(78% accuracy). As Figure 1 shows, the accuracy rate for rst and second persons
is much lower in the low contact group, below 50% in the present and below 70%
in the past. Given the hypothesis of the third person as underspecied default, we
erbal agreement in the L2 Spanish of speakers of Nahuatl
and third persons by the third. As Table 4 shows, this is not the case. Both the ger
und and the innitive were used more frequently as a default.
Table 4.
Raw number of errors and percentages organized according to type of answer
produced
Present
Past
Gerund
Innitive
3Ps
Other
Gerund
Innitive
3Ps
Other
High contact

6 (7%)
Low contact
se of third person for rst person.

xperimenter –
Imagina que tú eres el de la foto, ¿quhaces t蘒aqu

‘Imagine that you are the one in the photo, what are you doing here?’

articipant:
Lo
e獴á
osiendo

it
is-3.
᜕e
sewing-

I (he?) am sewing it.’
�se of the gerund for third person.

xperimenter
隿Quhace esta s旱ora?

‘What is this lady doing?’
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito

Participant
陃omiendo
melón.

eating-gerund
the
elon

She is eating the melon.’

xperimenter
ntes con el me泳n?

And before with the melon?’

�.
articipant
Cor瓡ndolo.

cutting-gerund-it

She is cutting it.’
�se of the innitive for the second person.

xperimenter
Imagina que tú eres quien está en la

otografía.
뽑ué haces aq痭?

Imagine that you are who is in the photograph. What are you
doing here?’

articipant
Lucir
sombrero.
(correct in standard Spanish)

how-oᴭ
�at

xperimenter
隿夠antes con el sombrero?

‘And before with the hat?’

�.
Participant
陔ejerlo.

Weave-
-it

She was weaving it.’
Morphology and syntax
�e sentences in the elicitation task were quite simple. However, object clitic pro
nouns were produced frequently (see
Ra淭rez-Trujillo 2013
), including when a
single innitive or present participle was used. is is important because the posi
tion of the clitic is an indication of the correct positioning of the verb. Following
Kayne (1991
; see also
Uriagereka 1995
; among others) we assume the clitic is in
a xed position and the conjugated verb in Spanish has moved to T, while non-
ᱮites move to some higher position, probably C. Ὡs explains why object clitics
precede conjugated verbs (7), while they follow non-nite forms (8).
Lo
como.

it
eat-
᜕e

I eat it.’
It is beyond the scope of this paper to examine the complete quite complex system of clitic
object pronouns in Spanish.
erbal agreement in the L2 Spanish of speakers of Nahuatl

*Comolo.

Eat-
᜕e
it (=(7a))
Para comerlo.

in order to eat-
-it

In order to eat it.’

*Para lo comer

rder to it eat-

Com槩ndolo.

eating-
-part-it

Eating it.’

*Lo comiendo.

t eating-
In total in this task we found 36 responses that consisted of an innitive with a
pronoun, as in (6b) and 4 in which the gerund was followed by a pronoun, as in
(5b). ere were no errors in the position of the pronoun relative to the verb; that
is, the pronoun was correctly positioned following the non-nite form. We also
found no errors in the position of the pronoun in the many cases of conjugated
verbs and pronouns. It is clear that the grammar of these learners distinguishes
Discussion and conclusions
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
extensive contact with the Spanish spoken in the region were signicantly less
likely to make errors in agreement and/or tense, no matter what the age at which
they were exposed to Spanish. However, we argue that it is not only the quantity
of input, but also the quality, in that those who worked outside the town would
encounter standard Spanish, while those that did not leave would have heard
�e need to form dictionary entries has led to the inclusion of certain endings in order
to constitute citation forms. For example, verbs borrowed from Spanish are overwhelmingly
made up of the Spanish in�nitive with the Nahuatl ending
-oa
(e.g.
trabajar-oa
).
erbal agreement in the L2 Spanish of speakers of Nahuatl
non-nite forms may appear in positions s where a nite form is expected.
rguing
on the basis of this, they suggested the Missing Surface I渞ection Hypothesis
(MSIH) according to which learners have appropriate representations of niteness
and agreement in the L2 syntax but have a problem mapping the correct forms.
Recall that, in this case, the forms are what the learners have to acquire. Evidence in
our data supporting the MSIH comes from the position of clitics. We have assumed,
following
Kayne (1991
) that the nite verb in Spanish climbs to T where it adjoins
to the clitic, while the non-nite forms climb over the clitic to a position higher
in the derivation, perhaps in C. is accounts for the fact that clitics in Spanish
precede conjugated verbs and obligatorily follow non-nite forms when there is no
auxiliary/main verb. We found absolutely no errors in the position of clitics, either
with conjugated verbs or with non-nite forms, evidence that the speakers are cor
to its
�ventual disappearance? ere is evidence that variability may be the norm
in certain individuals and communities, even in the case of monolingual speakers
Bruhn de Garavito & Atoche 2006
). In second language research we usually take
variability as an indication of some sort of problem either of mapping or at the
representation level (
Bley-Vroman 2009
). is cannot be the case if it is the norm
of monolingual native speakers. We believe the answer to some of these questions
is relevant to the study of languages in contact, language change, second language
acquisition, and the other well-known case of variability in production: heritage
languages.
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
References
Ac痱a, M.L., & Menegotto, A.C. (1992). Dativo sin
y verbos pronominales sinse: Rasgos
dialectales del español de la zona mapuche.
Revista Argentina de Lingüística
Ac痱a, M.L., & Menegotto, A.C. (1996). El contacto lin柼ístico español/mapuche en la
Argentina.
Signo y S旱a,
DOI: 10.1023/a:1006090432389
Baker, M. (1996).
se Polysynthesis Parameter.
Oxford: OUP.
DOI: 10.1017/s0022226798227113
Bley-Vroman, R. (2009). e evolving context of the fundamental dierence hypothesis.
Studies
in Second Language Acquisition,
DOI: 10.1017/s0272263109090275
Spanish L2 and agreement morphology. In J. M. Liceras, H. Zobl, & H. Goodluck (Eds.),
Generative Approaches to Second Language Acquisition
GASLA
6, University of Ottawa
(pp. 17–23). Somerville, MA: Cascadilla Press.
Bruhn de Garavito, J. (2009). Eventive and stative passives: �e role of transfer in the acqui
sition of ser and estar by German and English L1 speakers. In J. Collentine, M. Gar揭a,
B⸚
�aiordⰚ& F. Marcos Marín (Eds.),
Selected Proceedings of the 11th
Hispanic Linguistics
Symposium
(pp. 27–38). Somerville, MA: Cascadilla Proceedings Project.
Bruhn de Garavito, J., & Atoche, C. (2006). Variability in contact Spanish: Implications for
second language acquisition. In C. Lefebvre, L. White, & C. Jourdan (Eds.),
L2 Acquisition
and Creole Genesis. Dialogues
(pp. 353–369). Amsterdam: John Benjamins.
DOI: 10.1075/lald.42.07bru
Bruhn de Garavito, J., & White, L. (2002). L2 acquisition of Spanish DPs: e status of gram
matical features. In A. T. Pez-Leroux & J. Liceras (Eds.),
se Acquisition of Spanish
Morphosyntax: e L1/L2 Connection
(pp. 151–176). Dordrecht: Kluwer.
DOI: 10.1007/978-94-010-0291-2_6
Chomsky, N. (1981).
Lectures on Government and Binding
. Dordrecht: Foris.
DOI: 10.2307/2273965
Chomsky, N. (1995).
se Minimalist Program
. Cambridge, MA: e MIT Press.
DOI: 10.1017/s0272263197241070
español – euskera.
Unpublished Master thesis. �e University of Western Ontario, London,
Ontario, Canada.
Escobar, A.M. (2000).
Contacto social y lingüístico. El español en contacto con el quechua en el
Perú
. Lima, Fondo Editorial de la Ponticia Universidad Católica del Perú.
DOI: 10.1017/s0022216x00014930
Franco, J. (1993).
On object agreement in Spanish.
Unpublished Ph.D. dissertation. University of
Southern California.
Grüter, T., Lew-William, C., & Fernald, A. (2012). Grammatical gender in L2: A production or a
real-time processing problem?
Second Language Research,
DOI: 10.1177/0267658312437990
Hawkins, R. (2009). Statistical learning and innate knowledge in the development of second lan
guage prociency: Evidence from the acquisition of gender concord. In A. G. Benati (Ed.),
Issues in Second Language Pro�ciency
(pp. 63–78). London: Continuum.
erbal agreement in the L2 Spanish of speakers of Nahuatl
Hawkins, R., & Casillas, G. (2008). Explaining frequency of verb morphology in early L2 speech.
Lingua,
DOI: 10.1016/j.lingua.2007.01.009
Hawkins, R., & Franceschina, F. (2004). Explaining the acquisition and nonacquisition of
(Eds.),
se Acquisition of French in Di�erent Contexts
(pp. 175–205). Amsterdam: John
Benjamins.
DOI: 10.1075/lald.32.10haw
DOI: 10.1191/0267658303sr212oa
Hill, J.H., & Hill, K.C. (1986).
Speaking Mexicano: Dynamics of Syncretic Language in Central
Mexico
. Tucson, AZ: e University of Arizona Press.
DOI: 10.1017/s0142716400008134
DOI: 10.1177/0267658312461803
Kayne, R.S. (1991). Romance clitics, verb movement, and PRO.
Linguistic Inquiry
Lardiere, D. (1998a). Case and tense in th攠‘fossilized鈠steady state.
Second Language Research,
DOI: 10.1191/026765898674105303
Lardiere, D. (1998b). Dissociating syntax from morphology in a divergent end-state grammar.
Second Language Research
DOI: 10.1191/026765898672500216
Lardiere, D. (2000). Mapping features to forms in second language acquisition. In J. Archibald
(Ed.),
Second Language Acquisition and Linguisti挠seory
(pp. 102–129). Oxford: Blackwell.
DOI: 10.1017/s0022226702211925
Lardiere, D. (2003). Second language knowledge of [± past] vs. [±nite]. In J. M. Liceras, H.ZoblⰚ&
H. Goodluck (Eds.),
Proceedings of the 6th
Generative Approaches to Second Language Acqui
sition Conference
GASLA 2002
(pp. 176–189)
Somerville, MA: Cascadilla Press.
Lardiere, D. (2009). Some thoughts on the contrastive analysis of features in second language
acquisition.
Second Language Research,
DOI: 10.1177/0267658308100283
Amsterdam: John Benjamins.
DOI: 10.2307/416129
DOI: 10.1017/s1366728900000122
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
Mallén, E. (1992). Subject topicalization and inection in Spanish.
seoretical Linguistics
18:
DOI: 10.1515/thli.1992.18.2-3.179
McCarthy, C. (2006). Default morphology in second language Spanish: Missing inection
or underspecied inection? In J. P. Montreuil & C. Nishida (Eds.),
New Perspectives on
Romance Linguistics: Selected Papers from the 35th
Linguistic Symposium on Romance Lan
guages
(pp. 201–212). Amsterdam: John Benjamins.
DOI: 10.1075/cilt.275.16mcc
McCarthy, C. (2007).
Morphological Variability in Second Language Spanish.
Unpublished Ph.D.
dissertation. McGill University, Montreal, Canada.
Meisel, J. (1997).
First and Second Language Acquisition
. Cambridge: CUP.
DOI: 10.1017/s0272263112000770
Meisel, J.M. (2011). Bilingual language acquisition and theories of diachronic change:
Bilingualism as cause and eect of grammatical change.
Bilingualism: Language and Cogni
tion,
DOI: 10.1017/s1366728910000143
Prévost, P., & White, L. (2000). Missing surface inection or impairment in second language
acquisition? Evidence from tense and agreement.
Second Language Research,
DOI: 10.1191/026765800677556046
Ra淭rez-Trujillo, A.P. (2013).
Producción de la concordancia en español por hablantes de náhuatl.
Unpublished Ph.D. dissertation. e University of Western Ontario, London, Ontario,
Canada.
Rice, S., Libben, G., & Derwing, B. (2002). Morphological representation in an endangered,
DOI: 10.1006/brln.2001.2540
Rohrbacher, B. (1999).
Morphology Driven Syntax: A eory of V to I Raising and Pro-drop.
Amsterdam, John Benjamins.
DOI: 10.1075/la.15
叡nchez, L. (2004).
Quechua-Spanish Bilingualism. Interference and Convergence in Functional
Categories
. Amsterdam: John Benjamins.
DOI: 10.1075/lald.35
Speas, M.J. (1994). Null arguments in a theory of economy of projection.
University of Massa
chusetts Occasional Papers in Linguistics
DOI: 10.1017/cbo9780511815065.002
White, L., Valenzuela, E., Kozlowska-Macgregor, M., & Leung, Y.K.I. (2004). Gender and num
ber agreement in nonnative Spanish.
Applied Psycholinguistics
DOI: 10.1017/s0142716404001067
erbal agreement in the L2 Spanish of speakers of Nahuatl
Appendix 1
Participants
Occupation
Contact with Spanish
Language they feel most
comfortable speaking
Mason’s helper
Bosses speak Spanish
Both languages
Student
Secondary school is in
Spanish
Both languages
Civil judge
Works in the capital of
Tlaxcala
Both languages
Mason’s helper
Bosses speak Spanish
Both languages
Factory worker, lived
in the US for two
years
People at work speak
Spanish
Both languages
Assistant to the
president of the
village of San Isidro
Speaks Spanish frequently
but doesn’t leave the village
Both languages
Factory worker
People at work speak
Spanish
Both languages
Housewife/
Independent work
Sells shoes via catalogue
and travels frequently
Both languages
Student
Secondary school is in
Spanish
Both languages
Housewife
Family speaks Spanish only
Spanish
Housewife
Never leaves the village
Nahuatl
Works in a large
�lma P. Ra淭rez-Trujillo & Joyce Bruhn de Garavito
Appendix 2
Accuracy rates for the two bilingual groups on agreement in the three persons and the two
tenses. Accuracy for the control group was 100% across the board.
Participant
Contact level
Present
Present
Present
Past
Past
Past
High
Low
High
High
High
High
High
High
Low
Low
Low
Low
Low
Low
Low
High
Low
Low
Low
Low
Low
High
Low
Low
Low
Early coda production in bilingual
Spanish and Basque
Maria-José Ezeizabarrena & Alaitz Alegria
University of the Basque Country
�e study of the codas in the spontaneous speech of a Basque-Spanish bilingual
child reveals that they are produced early and continue to develop gradually,
in both languages, during the period studied (1;09–2;01). e inter-linguistic
摩ierences attested regarding the inventory of segments, the frequency of the
Introduction
�ere is quite an extended consensus that the presence of codas (consonant(s)
following the nucleus of the syllable (1)) in early speech is one of the developmen
tal milestones in the acquisition of phonological complexity. However, studies do
diverge regarding the model chosen to account for the development of the syllable
structure, such as
Fikkert’s (1994
parametric
approach or
Demuth and Fee‘s (1995
Minimal Word
view; additionally, the age and frequency of codas throughout the
ᱲst stages of language production show cross-linguistic variability. Children
acquiring English are reported to produce codas by 1;08 and Catalan children pro
duce over 50% of the expected codas at age 1;09 (
�aria-José Ezeizabarrena & Alaitz Alegria
Factors of early coda production
�us, children exposed to languages with high coda frequencies, such as
glish,
German or Catalan, produce them at a higher rate (
Bernhardt & Stemberger
Kehoe & Stoel-
Gammon 2001
the former and based on the coda patterns of 35 languages, these authors
redict
that the earliest segments produced in coda positions will be coronals (earlier
than labials and dorsals) and sonorants (earlier than obstruents). Many languages,
including Spanish (
Lleó 1997
Polo 2011
), have provided data consistent with the
former hypothesis. For Basque,
Saizar (2005
) reports on a 1;09 year old mono
lingual child producing dierent segment types in coda positions: (alveolar and
arly coda production in bilingual Spanish and Basque
predorsal) nasals and sibilants (100%) and laterals /l/ (40%), whilst
Barr旱a (2003
reported on two monolingual children producing alveolar nasal codas from 1;04,
been considered to aiect child coda production in dierent languages (
Freitas
Freitas, Miguel, & Hub Faria 2001
Bosch-Baliarda 2006
). Moreover, the earlier production of codas in word nal
rather than internal positions (
Goad & Brannen 2000
Rose 2000
) in languages
like French (accent on the last syllable), and English (frequent ᱮal accent in
low-prominence nature of the mostly unstressed fricative coda (
Ezeizabarrena
) whereas the morphological content of some non-prominent syllables is
For instance, in the diachronic processes from Latin to Spanish, vowel shortening and
vowel omission were mostly attested only in unstressed syllables (
Lloyd 1993
�aria-José Ezeizabarrena & Alaitz Alegria
the same segments are realized productively earlier in non-morphological codas
than in morphological ones in Spanish (
Polo 2013
Coda production in early bilingualism
Finally, bilingual language acquisition and the potential inter-linguistic inuence
extend the list of factors aiecting the early production of codas. Similarly to what
is observed in the lexical and grammatical domains, the debate on the shared
Celce-Murcia 1978
Deuchar & Clark 1996
Leopold 1949, 1971
Vogel 1975
) ver
sus the separated phonological systems (
Paradis (1996
) for English-German and
Paradis (2001
) for English-French;
Johnson & Lancaster (1998
) for Norwegian-
English;
Garlant (2001
) for Spanish-English a.o.) has promoted the amount of
research in early bilingual acquisition. For instance,
Schnitzer and Krasinski (1994
proposed a unitary system of consonants but a separated one for vowels in Spanish
and English, whereas
Ingram (1981, 1982
) argues for a separated system in both
domains by age 2. More specically, the research on the bilingual acquisition of
codas has provided interesting conclusions related to the input. Under the assump
tion that the input the child is exposed to has an 楮�uence on her development,
many authors (
Kehoe & Stoel-Gammon 2001
Lleó 2003
�osch-Baliarda
) predict that bilingual children will produce more CVC syllables in the lan
guage in which they are more frequently exposed.
oreover, among the dierent
Notice that absolutive case is marked with a zero morpheme in Basque.
arly coda production in bilingual Spanish and Basque
forms of inter-linguistic inuence in bilingual language development (e.g. feature
transfer, acceleration or delay (
Paradis & Genesee 1996
Pearson, Fernandez &
Oller 1993
Basque and Spanish codas
Basque and Spanish share many phonological properties, one of which is syllabic
structure. Both languages may have from zero to a maximum of two consonants
NucleusCoda
Figure 1.
Syllabic structure in child Basque and Spanish input
Nevertheless, some instances of two consonant codas can be found in both
languages. In Basque, complex codas are only found in a reduced list of words
such as
[bart] ‘last night鈠or
bel
[belts
崠‘black鈠and a some proper nouns
like
Erlan
[erlants
]. Even the complex coda of the seemingly frequent adjective
ltz
ₑblack鈠cannot be considered frequent in the input, since in most cases it
�aria-José Ezeizabarrena & Alaitz Alegria
As far as segmental inventories go,
Hualde (2003
) indicates that Basque codas
present more segmental diversity in nal position than in the medial one. In medial
position, the inventory is restricted to nasals that share their dento-alveolar (apical
to predorsal) place of articulation with many of the following consonants: liquids
/l/, /r/, sibilant alveolar fricatives like the apico-alveolar /s
/, the lamino-alveolar
/ and the pre-dorsal //, where inter-morphemic boundaries may turn into /ns
mendira
ko
‘bound for the mountain鈩⁡nd /ls
ₑmake or become black鈩
sequences. Stops and airicates are not allowed in medial codas. In contrast, in
addition to the continuants /n/, /l/, /r/, /s
/ and // observed in medial position,
ᱮal codas may include the airicate consonants /ts
/, /t/ and the stops /t/, /k/.
In some cases, the sequences /nts
antz
ₑresemblance’), /lts
beltz
ₑblack’), /rts
umezurtz
ₑorphan’), /rts
erts
ₑto close’), /rt/ (
idort
ₑto dry out鈩⁡nd /nt/ (
galant
‘gallant鈩ay also occupy coda positions.
Spanish has a more reduced inventory: /n/, /m/, /l/, /r/, /s
and // in medial
position and /n/, /l/, /r/, /s
/, // and // in nal ones, though complex codas such
as /ns
/, /ls
/, /rs
/ and /bs
/ may also appear in nal position with infrequent words
such as
transporte
ₑtransport’,
solsticio
ₑsolstice’,
perspicaz
ₑacute or perceptive鈠or
ob獴áculo
ₑobstacle’.
Noticeably, some (intra)dialectal variation can be observed in nasals and sibi
lants regarding their place of articulation and/or voice as a consequence of assim
ilative processes, such is the case of
hanka
� [aŋka] ‘foot鈻
enborra
� [embora]
‘trunk鈻
esne
� [ezne] ‘milk’ in Basque and
mismo
[mizmo] ‘same’ in Spanish.
As for prominence, stress is not phonologically contrastive in Basque⸠�e
prominence-based and frequency-based hypotheses, children acquiring Spanish
are predicted to show some preference for the production of codas in nal sylla
Usually transcribed as /s/ in Spanish regardless of its laminal (in some varieties) or the
(more) apical pronunciation such is the case of the variety of Spanish the child is exposed to.
arly coda production in bilingual Spanish and Basque
One more fact to be taken into account is the morphological nature of codas.
Both Spanish and Basque are richly inected languages in which codas are very
frequently (alone or as part of) overt expressions of morphosyntactic features such
Codas in the early production of Spanish and Basque
Predictions
Some predictions can be made about the early simultaneous acquisition of Basque
and Spanish based on the acquisition literature and the specic features of codas
in these languages.
�requency. Children acquiring languages in which closed syllables are not very
frequent like in Basque (25%–30% of closed syllables, about 60% in nal posi
tions) and Spanish (about 30% of closed syllables) as shown by
Jauregi (2003
are expected to produce codas at a similar age, and later than in languages with
frequent codas such as English, German or Catalan (
Bernhardt & Stemberger
Grijzenhout & Joppen 1998
Kehoe & Stoel-Gammon 2001
). However,
languages with varied kinds of codas in terms of segments (Basque) and/or
complexity are expected to show an earlier development in the acquisition of
codas than languages with less segmental (Spanish) and structural variability
and productivity (
�aria-José Ezeizabarrena & Alaitz Alegria
b.
frequency-based hypotheses, the child is expected to produce (more or earlier)
�rammar. Morphological codas may show higher production rates than
non-morphological ones due to their high frequency in both richly inected
languages, in which frequent morphological marking such as plural in the
d.
�e exposure to two languages with similar syllabic structures and coda rates
predicts a similar development in the acquisition of codas. However, the pres
ence of dierent segments in coda position in the two languages may improve
the bilingual’s sensitivity to this syllabic structure and cause that the bilingual
child acquire/produce closed syllables at earlier ages than the monolinguals
acquiring any of the languages.
�e current study is based on partial data available from a longer longitudinal case
arly coda production in bilingual Spanish and Basque
⠹2in Spanish and 8 in Basque), in which the child interacted with (at least) one of the
Gasteiz, Spain, and the University of Hamburg in Germany.
Phonology was not
the main goal of the materials recorded in the project mentioned, and therefore, all
the recordings included in the analysis for the current study were re-transcribed
ez daki
pintzen
ₑI don’t know (how) to put it’

uto拺s
ₑthe bu玒
�e production of onomatopoeic expressions, pieces of songs and proper nouns
Mikel, Maialen, Charly, Mosku
) were included, as well as cognates and Basque
loan words of Romance or Spanish origin such as
asustau/asustar
ₑscar斒 or
aber/
a ver
Results
�ere is some evidence that Mikel produced some CVC syllables during his one-
word stage (
han
ₑthere鈠at 1;07, MLU = 1) even before the rst recording of the
present study at age 1;09, but the precise age at which he produced his ᱲst coda is
not available to us.
We are very thankful to the child and his family, to the promoters of both research groups,
Itziar Idiazabal and Jürgen M. Meisel, respectively, and to the collaborators: A. Barreña,
M.Almgren, A. Mahlau, M.P. Larrañaga, A. Hernández-Cembel泭n, J. Mercado, J. Cárdenes,
Nuria Acacio and Luis Moreno.
It should be noted that glides produced in contexts of target consonants have been in
cluded in the study as (target-deviant) codas.
�aria-José Ezeizabarrena & Alaitz Alegria
spontaneous conversations during th攠sve months of the study.
Table 1.
AGE
Basque
Spanish
Recorded
time
min.
Utterance
Context
Produced
Recorded
time
min.
Utterance
Context
Produced
Total
�e data in Figure 2 indicate an increase in the production of codas in both
languages during the studied period, with a consistently higher rate in Basque
than in Spanish throughout this time. With the exception of the rst recordings,
in which the number of contexts was quite low in Spanish, a steady increase in the
1;091;101;112;002;01
Figure 2.
Mikel’s rates of codas produced out of the intended (C)VC syllables in both
languages by month (1;09–2;01)
Both, target-like and target-deviant CVC productions by the child are included in this
category. For instance, the target-like /
pur
/ and the deviant /
pul
/ would be counted as a “pro
duced CVC” when the child is attempting to produce
ₑbreak’).
arly coda production in bilingual Spanish and Basque
From the rst recording, the bilingual child produces over 20% of the expected
codas in Basque, a rate he achieves in Spanish one month later. However, his coda
production increases from age 2 onwards in Basque, whereas in Spanish there is
an increase at age 1;10, but no increase a1erwards.
Table 2.
Mikel’s number and rates of codas produced in the two languages out of the
intended (C)VC syllables
Age
Total
Basque
Spanish
Di�erence
As Table 2 shows, the number of closed syllable contexts is higher in the
Basque sample (810 contexts) than in the Spanish one (400 contexts). Based on
the data described, a general tendency to omit codas is attested in the two lan
guages of the bilingual during his earliest stages of language production, though
the longitudinal data suggest that the developmental pattern observed seems to be
language specic: Spanish rates remain at around 30%, whereas Basque produc
tion rates reach 50% at 2;1.
Segment inventory
�e inventory of segments in coda position produced by Mikel during the studied
period is quite wide, though it diers from Basque to Spanish. e presence of
egment inventory in Basque codas: /n/, /l/, /r/, /s

�.
egment inventory in Spanish codas: /n/, /m/, /l/, /r/, /s
�e data in Table 3 indicate that the production of codas varies across segments in
both languages. e fact that the sample is larger in Basque (1537 utterances) than
�aria-José Ezeizabarrena & Alaitz Alegria
in Spanish (1325 utterances) could aiect the higher total coda numbers in the for
Table 3.
Nasals
Liquids
Sibilants
Stops
Notice that in the dialect the child is exposed to, the apical/laminal distinction is not
phonological.
arly coda production in bilingual Spanish and Basque
stops (4f, j) and alveolar fricatives (4i). In some cases, the vibrant /r/ (4a, c) and the
alveolar nasal /n/ (4b) are also replaced by th攠�ap /
/, the lateral /l/ (4a, c) and the
(pre)palatal /
/. In Spanish (4m-q) Mikel produces 9 substitutions. Among them
7 correspond to segments not included in (3b), such as the palatal glide /j/ (4m),
the �ap /
/, the voiceless velar /x/ (4p), and the dental /d/ (4q) produced instead
of other consonants, mostly liquids. �e �ap /
/ and the (pre)palatal /
/ are also
attested in place of the vibrant /r/ and the alveolar nasal /n/ respectively.
�ome substitutions in Basque (a–o) and Spanish (m–q)

al.tu] for
sartu
&#x/MCI; 67;6 0;&#x/MCI; 67;6 0;�&#x/MCI; 67;7 0;&#x/MCI; 67;7 0;artu]

�.
�‘a.lal] for
holan
[olan]

�’pul.tu]
for
apurtu
[apurtu]

�.
�] [j]: [‘ej.ta.a.’pul.tu] for
ez da apurtu
&#x/MCI; 67;Q 0;&#x/MCI; 67;Q 0;i&#x/MCI; 67;R 0;&#x/MCI; 67;R 0;tapurtu]

�.
�oj.te.’re.ta] for
boltereta
[bolte

�.
ux] for
zuk
&#x/MCI; 67;c 0;&#x/MCI; 67;c 0;i&#x/MCI; 67;d 0;&#x/MCI; 67;d 0;uk]

�.
�or
nik
[nik]

hamen
[amen]

i.
�] [x]: [ax.’ka.tu] for
askatu
[as
katu]

�.
�‘awn.ts
eg] for
ahuntzek
[awnts
ek]

ien da] for
zer da
&#x/MCI; 67; 0;&#x/MCI; 67; 0;i&#x/MCI; 67;‘ 0;&#x/MCI; 67;‘ 0;erda]

�.
�antaw] for
soltau
[soltaw]

�‘mi.kej] for
Mikel
[mikel]

]: [mikes
] for
Mikel
[mikel]

�.
�nd [x]: [a.’�l], [a.’ßej], [a.’ßex] for
a ver
[aßer]

�.
�‘u.ka:.’lox] for
un caracol
[unka
akol]

�.
�nd [
] [pin.’tad], [pin.’ta
] for
pintar
[pintar]
�ough it is a much less frequent phenomenon than omission in coda positions,
Examples in (3) indicate that
substitution
is also attested in the coda production of
this child in both languages (19 items in Basque and 9 in Spanish). It occurs in dif
ferent contexts and aiects dierent segments as shown in (3). us, since in most
�aria-José Ezeizabarrena & Alaitz Alegria
in the preceding period⸠�e rest of consonants are produced quite rarely, ranging
2;002;01
1;091;101;11
Figure 3.
1;091;10
1;112;002;01
Figure 4.
Placement
One of the goals of the study was to look at the potential eect of syllable (nal,
medial) word placement on the production of codas. Since, in the ow of oral
arly coda production in bilingual Spanish and Basque
lexical word was codied in the case of mono-lexemic utterances, but in multi-
lexemic prosodic words, the (medial or nal) placement of the syllable was codi
ᱥd regarding its position inside the prosodic word. Consequently, codas codied
as codas in nal position of the utterance were all at word nal position (see
Examples(5a) and (5c) for Basque and (5b) and (5d) for Spanish), whilst codas
in medial position of the prosodic word included word internal codas ((6a) for
Basque, (6b) for Spanish) as well as codas which, despite their word nal position,
are placed at some position inside the prosodic word ((6c) for Basque, (6d) for
Spanish).
hame
酨ere’

zapatilla
‘shoes’

hau kendu gura do
‘I want to put this out’

�l tre
‘in the train’
ₑ⡴o) repare’

ₑthi玒

dau
ₑwhere is it?’

se ha caído e
tren
ₑhe fell out of the train’
�e frequent presence of the negation
ₑno’ in Basque may distort (C)VC tal
lying. When appearing in nal positions, this particle is produced accurately for
the most part (Table 4) and the rest of the monosyllabic words are also produced
in over 50% of the contexts:
bat
[bat]. Medial positions are more problematic,
though, as the omission of codas even in these same particles is as frequent as
their production. Similarly, nasal and vibrant codas are inconsistently produced in
medial positions, as in
apu
tu egi
da
['pu.tu.'in.da] ‘it is broken’, where the codas
are only attested in some of the stressed positions.
Table 4.
Production rates of codas in Basque negation
&#x/MCI; 69;6 0;&#x/MCI; 69;6 0;i&#x/MCI; 69;7 0;&#x/MCI; 69;7 0;]ₑn澒
Total
N Produced/target
In Spanish, the production of medial codas is quite irregular: 0% to 8% by
age 2;00 and over 70% a1er that age. See for instance,
班 a
cua
&#x/MCI; 69;f 0;&#x/MCI; 69;f 0;�&#x/MCI; 69;g 0;&#x/MCI; 69;g 0;i.al.'ta.to]
⡍21;11;02);
túnel
[al.’tu.nel] (M 2;00;00);
a
cide
['un.'te:n.te] (M 2;00;20).
Proclitic contractions
�aria-José Ezeizabarrena & Alaitz Alegria
Stress and word length
Stress is another of the features studied. Multisyllabic words as well as prosodic
words may have more than one closed syllable. In many cases the child tends
to produce codas in the stressed syllable and to omit (at least) the coda of the
unstressed one⸠�e examples in (7) illustrate this preference in Basque words and
the same is illustrated in (8) for Spanish.
C contexts in polysyllabic (prosodic) words:

hemendik doie
['men.ji.'o.je], ['men.di.'o.je]

it goes from here’

hemendik
�.'men.di]

from here’

hemendik
['men.t
i] ['man.t

from here’

konpondu
�un.du]

repair’

bazkaldu
['kal.du]

to have lunch’

kontuz
on.tu]

be careful’

g.
bazkaldu
�a.'kal.du], [bo.'kal.du]

to have lunch’
sí durmiendo
i.nu.'nen.do]

yes, sleeping’

asahse engancha
�j.ta.'kan.ta]

so, there can be tied’

un coscorr
o.ko.'ko:n]

a bump on the head’
Anecdotally, a few instances (
= 4) were found in Basque in which codas were
omitted in the stressed (primary) syllable, but produced in the stressed (
econdary)
syllable of the same word (9). No similar examples are found in Spanish.
hau ipintzen da
['a.'pi.cen.'ða]

this is put in this wa禒

bestien
['be.cen]

in the other one’

bazkaltzen
[ke.'ci.cen]

aving lunch’

ez dot ikusten
[ta.'ßu:.cen]

I don’t see’
arly coda production in bilingual Spanish and Basque
In general, stress seems to facilitate the general production of codas in Basque, as
indicated by their rate of over 40% in stressed syllables as compared to a rate of
13% in unstressed ones. In contrast, no such dierence is observed in the Spanish
productions, in which coda production rates oscillate around 20% regardless of
the stressed or unstressed nature of the syllable (See Table 5).
Table 5.
Coda production in stressed and unstressed syllables in Basque and
Spanish. Mikel 1;09–2;01
Syllable
Basque
Spanish
Stressed
Unstressed
Stressed
Unstressed
Prod/Target
Rate
1;091;101;112;002;01TOTAL
Figure 5.
CVC production rates in (un)stressed syllables, by month and language. Mikel
�aria-José Ezeizabarrena & Alaitz Alegria
A look to the length of words in syllables indicates that codas are produced
more frequently in monosyllabic than in polysyllabic words in both languages
(Table 6).
Table 6.
Coda production in monosyllabic and polysyllabic words in Basque and
Spanish. Mikel 1;09–2;01
Basque
Spanish
Monosyllabic
Polysyllabic
Monosyllabic
Polysyllabic
Prod/Target
Rate
�e dierence is higher in Basque (31%) than in Spanish (11%), though
monosyllabic words show a similar pattern from 1;10 onwards in Basque (range
34%–65%) and in Spanish (range 20%–53%), as do polysyllabic words in Basque
(range 24%–42%) and in Spanish (range 20%–26%). See Figure 6.
1;091;101;11
2;012;00
Figure 6.
Coda production rates in monosyllabic and polysyllabic words by month and
language. Mikel 1;09–2;01
Data in Figure 7 show that word length, position and stress, all interact with
each other in Basque, as codas are produced the most frequently in monosyllabic
ᱮal words (mean 77.2% and rates over 70% in all recordings), followed by mono
syllabic words in medial positions (mean 41.6%, range 15%–59% across record
ings), as well as stressed syllables in medial (mean 43%, range 7%–55%) and
stressed syllables in nal positions inside polysyllabic words (mean 43%, range
0% to 50%). e lowest ones correspond to unstressed syllables inside polysyllabic
arly coda production in bilingual Spanish and Basque
words regardless of their nal (mean 13%, range 0% to 26%) or medial position
(mean 14% range 0% to 25%).
1;091;101;112;002;01
Figure 7.
Rates of coda production in Basque by syllable type (stress and placement) and age
In Spanish, the production of codas seems to be slightly favored in monosyl
labic words (mean over 30% regardless of position) as compared to polysyllabic
words, (around 20%), though rates of coda production show more variation in
monosyllabic nal (mean 32%, range 0% to 100%) than in medial monosyllabic
words (mean 30%, range 0%–43%) as shown in Figure 8. In polysyllabic words
coda production rates are lower, of around 20%, regardless of stress and position
(21% in nal stressed, 22% unstressed nal syllables and 21% and 19% in medial
stressed and unstressed syllables respectively). Figures 7 and 8 do not suggest any
�aria-José Ezeizabarrena & Alaitz Alegria
syllable in the production of codas in Spanish by this bilingual child, at least dur
ing the studied period.
Morphological codas
A nal classication was conducted distinguishing morphological and non-
morphological closed syllables in order to test the potential 攝ect of the gram
matical nature (and consequently, the eect of frequency) in coda production.
ₑwith鈠are pseudo-
cliticized to the following nominal element. Except for the prepositional function
words, the other grammatical codas correspond to the plural sux /-s/ in most
cases,
�ollowed by the innitive markers
-ar/er/ir
and person marking. No clear
1;091;101;112;002;01
Figure 8.
Rates of coda production in Spanish by syllable type (stress and placement) and age
arly coda production in bilingual Spanish and Basque
摩ierences were observed in the production of grammatical codas in stressed vs.
unstressed syllables in Spanish.
In Spanish, a steady increase is observed in the low rates of coda production
1;091;101;112;002;01
Medial -morf
Final +morf
Final -morf
Medial +morf
Figure 10.
1;091;101;112;002;01
Figure 9.
�aria-José Ezeizabarrena & Alaitz Alegria
Discussion
�e simultaneous Basque-Spanish bilingual child studied in the period from
1;09 to 2;01 frequently omits codas in monosyllabic CVC words in his spon
taneous production, for instance /'xa/ for
jan
ₑeat’ /xan/, or /'ba/ for
bat
/bat/
酯ne’ (
= 3), /'a/ for
han
/an/
‘there鈠(
= 22) in Basque or /'ma/ for
淡s
‘mor斒
㴚12) in Spanish. But, at the same time, the codas produced, among others,
in the 36 instances of the deictic adverbial /an/ ‘there鈠along the study (preceded
by a very early instance of this monosyllabic CVC word in an earlier recording
at 1;07) are evidence for the early production of codas, at the one-word stage, at
arly coda production in bilingual Spanish and Basque
Spanish).
In fact, the data plotted in Table 7 (reported by
Table 7.
Longitudinal data of coda production in child language
Pro�le language
José
Mono Spanish
María
Mono Spanish
Miguel
Mono Spanish
Simon
Bi Span(-Ger)
Stefan
Bi Span(-Ger)
Robert
Bi Span(-Ger)
Mikel
Bi Span(-Ger)
Mikel
Bi Basque(-Span)
*: data not available.
�e nature of the segments, their position and their prominence seems to have
an eect on the earlier/later consistent coda production in Basque and
panish. As
attested in many other linguistic levels such as syntax, morphology and vocabu
lary, the ᱲst production of any phonological segment in the spontaneous linguistic
production of a child implies neither its adult-like knowledge nor its use at any
phono-syntactic position (
Barr旱a 2003
Lleó 1997
Notice that along this study not only lexical but also functional words are included in the
counting of target and produced codas. For this reason coda rates in the current study may not
be directly comparable to other studies based on lexical items (
Polo 2011
Prieto 2006
; etc.) or
(pseudo)words (
Borràs-Comes & Prieto 2013
�aria-José Ezeizabarrena & Alaitz Alegria
Basque,
Barr旱a (2003
) observes that the chronological order in the production of
arly coda production in bilingual Spanish and Basque
categories for which a shared
epresentation seems to facilitate the inter-linguistic
楮�uence and consequently the similar production pattern in the two languages,
the bilingual child seems to make use of language specic strategies in the pro
duction of sibilants: mostly substitutions in Basque (allophonic or non-allophonic
variants which share some of their features with the segment intended) such as
&#x/MCI; 75;‡ 0;&#x/MCI; 75;‡ 0;�&#x/MCI; 75;ˆ 0;&#x/MCI; 75;ˆ 0;], [es&#x/MCI; 75;‰ 0;&#x/MCI; 75;‰ 0;i&#x/MCI; 75; 0;&#x/MCI; 75; 0;], e[] or [e] instead of
&#x/MCI; 75;“ 0;&#x/MCI; 75;“ 0;i&#x/MCI; 75;” 0;&#x/MCI; 75;” 0;]) and mostly omissions in Spanish ([
ete
instead of [es
te] ‘this鈩.
�e prosodic prominence hypothesis predicts that stressed syllables will be
Saceda (2005
) reports on diierences between the coda rates produced in stressed over
unstressed syllables in the earliest production of codas of one child up to age 1;9. No clear dif
ferences can be observed between stressed and unstressed syllables aser this age.
�aria-José Ezeizabarrena & Alaitz Alegria
Basque (always morphemic s甛xes) or of [s
], which is more frequently omitted
in Spanish (morphological and non-morphological) than in Basque (always non-
morphological). In Spanish codas, the distribution of their production/omission
the
�uisition of codas in the other language. We are nevertheless aware that the
arly coda production in bilingual Spanish and Basque
scarcity of studies on child coda production in Basque and the limitations of the
current sample present an obstacle to the co渜rmation/refutation of this hypoth
esis and the statement of any conclusive remark in this regard. Further research is
required in order to test this hypothesis.
Conclusions
�e longitudinal study of the coda productions in Basque and Spanish by a bilin
gual child during the period 1;9–2;1 reveals that:
�espite an initial frequent omission of codas, their production rate increases
gradually in both languages from 1;09 onwards, though faster in Basque than
in Spanish, and earlier for some segments (nasals) than for others (sibilants
and plosives) in both languages.
�e language specic development in the inventory and the production rates
of the most productive segments during the earliest production of codas,
�aria-José Ezeizabarrena & Alaitz Alegria
Bernhardt, B.H., & Stemberger, J.P. (1998).
Handbook of Phonological Development. From a
Non
linear Constraints-based Perspective.
San Diego, CA: Academic Press.
DOI: 10.1017/s0952675799003723
panish
children: Eects of prominence and frequency of exposure.
Probus,
DOI: 10.1515/probus-2013-0001
Celce-Murcia, M. (1978). e simultaneous acquisition of English and French in a two-year-
old child. In E. Hatch (Ed.),
Second Language Acquisition: A Book of Readings
(pp. 38–53).
Rowley MA: Newbury House.
Demuth, K., & Fee, E.J. (1995).
Minimal Words in Early Phonological Development
. MS, Brown
University, Providence, RI & Dalhousie University, Halifax, CA.
Deuchar, M., & Clark, A. (1996). Early bilingual acquisition of the voicing contrast in English
and Espanish.
International Journal of Bilingual Education and Bilin
gualism
DOI: 10.1080/13670050902783528
Fikkert, P. (1994).
the Acquisition of the Prosodic Structure
. Dordrecht: ICG Printing.
Freitas, M.J. (2001). Syllabic constituency and segmental emergence: Evidence from the acquisi
tion of European Portuguese. In M. Almgren, A. Barr旱a, M.J. Ezeizabarrena, I.
diazabalⰚ&
B. MacWhinney (Eds.),
Research on Child Language Acquisition
(pp. 45–57). Somerville:
Cascadilla Press.
DOI: 10.1075/lald.24.04fre
tax: Plurals within codas in the acquisition of European Portuguese. In J.
eissenbor渚&
B⸚Höhle (Eds.),
Approaches to Bootstrapping: Phonological, Lexical, Syntactic and Neu
rophysiological Aspects of Early Language Acquisition,
II (pp. 45–57). Amsterdam: John
Benjamins.
Garlant, M. (2001).
Spanish Phonological Patterns of Young Spanish-English Bilinguals
. Unpub
lished Master thesis. University of Arizona, Tucson.
arly coda production in bilingual Spanish and Basque
Kehoe, M.M., & Stoel-Gammon, C. (2001). Development of syllable structure in English-
speaking children with particular reference to rhymes.
Journal of Child Language,
DOI: 10.1017/s030500090100469x
Leopold, W. (1949/71).
Speech Development of a Bilingual Child, a Linguists鈠Record,
(Vols. 1–4).
New York: AMS Press.
Levelt, C.C., Schiller, N.O., & Levelt, W.J. (1999/2000)
�e acquisition of syllable types.
Lan
guage Acquisition
DOI: 10.1207/s15327817la0803_2
Lleó, C. (1997).
La adquisición de la fonología de la primera lengua y lenguas extranjeras
. Madrid:
Visor.
DOI: 10.5209/rev_rfrm.2013.v30.n1.42607
Lleó, C. (2003). Prosodic licensing of coda in the acquisition of Spanish.
Probus 15
DOI: 10.1515/prbs.2003.010
Lleó, C., Kuchenbrandt, I., Kehoe, M., & Trujillo, C. (2003). Syllable nal consonants in
panish
and German monolingual and bilingual acquisition. In N. Müller (Ed.), (
vulnerable
Domains in Multilingualism
(pp. 191–220). Amsterdam: John Benjamins.
DOI: 10.1075/hsm.1.08lle
Lloyd, P.M. (1993).
Del latín al español.
Madrid: Gredos.
Jo桮2
Benjamins.
DOI: 10.1017/s0022226704293239
Paradis, J. (1996). Phonological dierentiation in a bilingual child: Hildegrad revisited. In
䄮2Stringfellow, D. Chana-Amitay, E. Hughes, & A. Zukowski (Eds.),
Proceedings of BUCLD
(pp. 528–539). Somerville, MA: Cascadilla Press.
Paradis, J. (2001). Do bilingual two-year-olds have separate phonolgy systems?
International
Journal of Bilingualism, 5
DOI: 10.1177/13670069010050010201
Paradis, J., & Genesee, F. (1996). Syntactic acquisition in bilingual children: Autonomous or
interdependent?
Studies in Second Language Acquisition, 18,
DOI: 10.1017/s0272263100014662
Pearson, B.Z., Fernandez, S.C., & Oller, D.K. (1993). Lexical development in bilingual infants
and toddlers: Comparison to monolingual norms.
Language Learning, 43
DOI: 10.1111/j.1467-1770.1993.tb00174.x
Polo, N. (2011). Notas sobre la adquisición fono泳gica bilingüe del español: El desarrollo de las
codas en español como lengua materna en hablantes bilingües del País Vasco.
Oihenart,
Polo, N. (2013). Adquisición del componente fono泳gico y morfo泳gico del español como len
gua materna. Talk at the
II Jornadas de Lingüística Hispánica
. Lisbon, Universidade de
Lisboa, 15th April 2013.
DOI: 10.1177/00238309060490020501
�aria-José Ezeizabarrena & Alaitz Alegria
Saceda, M. (2005).
Adquisición prosódica en español peninsular septentrional: la sílaba y la
palabra prosica.
Unpublished Master thesis. Universitat Autònoma de Barcelona,
Barcelona.
Saizar, M. (2005).
Euskararen jabekuntza: silaba itxiak edo kontsonantez amaitzen diren koda-
egiturak
. Unpublished Master thesis. University of the Basque Country, Vitoria-Gasteiz.
Salidis, J., & Johnson J. S. (1997). �e production of minimal words: A longitudinal case study
of phonological development.
Language Acquisition, 6
DOI: 10.1207/s15327817la0601_1
Schnitzer, M.L., & Krasinski, E. (1994). e development of segmental phonological production
in a bilingual child.
Journal of Child Language
DOI: 10.1017/s0305000900009478
Vogel, A. (1975). One system or two: An analysis of a two-year-old Romanian-English bilin
gual’s phonology.
Papers and Reports on Child Language Development, 9
Zamuner, T.S., Gerken, L.A., & Hammond, M. (2005). e acquisition of phonology based on
input: A closer look at the relation of cross-linguistic and child language data.
Lingua
DOI: 10.1016/j.lingua.2004.06.005
e locative paradigm in the L2 Spanish
of Catalan native speakers
Silvia Perp槱án
�e University of Western Ontario
Ὡs chapter explores the expression of locative and existential constructions
in the Spanish of native speakers of Catalan (
Introduction
Ὡs study investigates the use of Spanish as a second language by native speakers
of Catalan. Spanish and Catalan are the two co-o᭣ial languages of Catalonia,
and their amount of use mainly varies according to the geographical area where
it is spoken as well as issues such as language choice and national identity. Given
the sociolinguistic reality of Catalonia, all Catalan native speakers are, to dier
ent degrees, bilingual in both Catalan and Spanish. is study investigates the
Spanish of Catalan native speakers who declare themselves dominant in Catalan,
�ilvia Perp槱án
Recently, there has been an increase in the much-needed description of the
Spanish language spoken by Catalan bilinguals (see
Blas Arroyo 2011
Sinner 2004
Sinner & Wesch 2008
for overviews). Practically all of these studies are enclosed in
the eld of contact linguistics and/or a variationist framework, with a sociolinguist
focus. e purpose of this chapter is to look at the same overall phenomenon from
a 摩ierent perspective: contact linguistics as the byproduct of second language
acquisition. Whereas it does not aim to deny the relevance of sociolinguistic vari
ables, or the ᱥld of language contact, here we are interested in the outcomes of
bilingualism as instances of language development. erefore, we believe that the
atalan
Spanish. Section 3 describes the linguistic phenomenon and its properties in the
two languages.
ection 4 reviews the main studies on SLA on this linguistic topic.
Section7 discusses the main ndings and conclusions.
e Spanish of Catalan speakers
According to the last survey on language uses in Catalonia,
approximately
3.4million people identify Spanish as their rst language while slightly less than
㈚million identify Catalan as their rst language. Only 236,000 people acknowl
edge having both languages as their rst languages. erefore, even though bilin
gualism is an essential feature of the identity of Catalonia and Catalonians, and
nearly all Catalonians can understand and speak both languages, language choice
e locative paradigm in the L2 Spanish of Catalan native speakers
and language identity is a key element in the linguistic reality of these bilinguals
Boix-Fuster & Sanz 2008
). Moreover, the family origin and the neighborhood
in which the bilingual is born and raised are crucial components that ultimately
condition their level of prociency in the two languages. For instance, according to
the same survey, the average percentage of the population that has Catalan as their
ᱲst language is 40.4%. However, this gure multiplies 1.5 to 2 times in central
Catalonia (
comarques centrals
), whereas the percentage is signicantly lower in
�ilvia Perp槱án
speakers use the verb
ser
to express the location of objects or people, as it is used
in SC. However, GS requires
estar
in these contexts. e following examples are
from
Serrano (1996
Mamá
es
asa
la
buelita.

-Ser
�t
home
of
he
grandma

Mom is at grandma’s house.’

El
�an
la
esa.

bread
-Ser
�t
the
table.

�e bread is on the table.’
e locative paradigm in the L2 Spanish of Catalan native speakers
forms not so much to express aspectual dierences, but to express evidentiality,
�eature that is present in Quechua morphology but not in Spanish morphology.
e linguistic phenomenon: e expression of existentials
and locatives
Assuming a unied account for locative and existential constructions (
Freeze
Lyons 1967
Rigau 1997
), there is a universal locative paradigm with three
possible surface structures related syntactically: (a) the predicate locative, with
examples in(2); (b) the existential structure, as in (3); and (c) the possessive con
struction with
have
, exemplied in (4). GS employs the verb
estar
for the predicate
locative (2a) except for eventive themes, in which case
ser
is used
(2c). SC, on the
other hand, employs
ser
with object and event subjects
(2b, 2d),
although
estar
is
constantly gaining ground to locate objects, especially in some areas of Barcelona,
�redicate Locative

OPULA
LOC
ATIVE

Las llaves
stán
l ca櫳n.
ᤘ8ᘕ
Les claus
ón
l calaix.
᠙�

‘�e keys are in the drawer.’

L愠�esta
asa de Joan.
᠙�
La festa
�a casa d’en Joan.
᠙�

‘�e party is at Joan’s place.’
Existential

OPULA
LOC
ATIVE

�nas llaves
l ca櫳n.
hᘇu

�.
*(Hi)
ha
nes claus
l calaix.
hᘇu

�ere are some keys in the drawer.’
Possessive

�engo las llaves.

�inc les claus.

I have the keys.’
According to
Brucart (2012
�ilvia Perp槱án
and
in Catalan and Spanish (2a,
2c vs. 2b, 2d). Brucart argues that GS
Estar + en
Ser + a
, the main dier
ence being that Catalan preposition
can express the notion of path, which in
Ell *
⃩s.
་�

CL BE

&#x/MCI; 84;s 0;&#x/MCI; 84;s 0;e&#x/MCI; 84;t 0;&#x/MCI; 84;t 0;] &#x/MCI; 84;u 0;&#x/MCI; 84;u 0;e獴á
ଏaᘕ

He is (there).’
No
�i
�a
pa.
�i
ha.

ot
has
bread
ot

part
has

No
�ay
pan.

�ay.

ot
has
read


�ere is no bread.’
�e technical details have been omitted since they are not relevant to the focus of this
study.
Longa et al. (1998
) assumed an AgrP to which
hi
adjoins. In AgrP, the combination of
the person features produces the derivation to crash.
e locative paradigm in the L2 Spanish of Catalan native speakers
No
�i
�avia
el
*No
l’
avia.

ot
have
he
ot
loc
have

No
�abía
un
perro

�abía.

ot
had

ot
had

�ere wasn’t the/a dog.’
Furthermore, as (7) shows, Catalan, unlike Spanish or English, does not present
the
d攞niteness eect
(DE). is restriction is mainly based on the semantics of
the theme.
Milsark (1977

�EFINITESS EFFECT
Hi
�a
�es
laus
al
alaix.

loc
has
the
keys
he
drawer

#Hay
las
�laves
el
a櫳n.

as
he
keys
he
drawer

�ere are the keys in the drawer.’
�e contrast is even clearer with proper nouns:
Hi
�a
�l
oan
asa.

loc
has
the
Joan
at
home

*Hay
�oan
casa

as
Joan
at
ome

�ere is Joan at home.’
Lyons (1999
) characterizes denite nouns as more familiar, more individualized,
and more ident検able than indenites. eir reference points towards a unique
element, which the hearer and the speaker share.
Freeze (1992
) and
White,
Belikova, Hagstrom, Kupish an搠Ö竧elik (2012
), among others, suggested that
the deniteness eect is a universal restriction observed in most natural lan
guages, particularly in a᭲mative sentences. is restriction, not applicable to
list-readings, has been explained in semantic terms (
Milsark 1977
), in syntactic
terms (
Saᱲ 1985
), and in pragmatic terms (
Lumsden 1988
). �e basic pragmatic
�ilvia Perp槱án
idea, defended by
Holmback (1984
) is that existential constructions introduce a
new entity in the discourse; therefore the new element needs to be unrelated to
the preceding discourse, that is, not previously dened (i.e. indenite). In any
case,
Lyons (1999
) considers it a semantics-pragmatic constraint, not a syntactic
one, with a broad crosslinguistic representation, more present in some languages
than in others. Catalan, as well as Italian, are examples of two languages that do
not seem to observe this constraint.
�ses
ser
instead of
estar
locative constructions, although it has been
argued that this dierence is progressively disappearing, displaying large con
vergence in some dialectal areas.
b.
�eeds the overt clitic
hi
in the existential construction, either with
haver
or with
ser
. Moreover, GS has an impoverished clitic system and expresses
partitive meanings with accusative clitics or with an empty element. SC, on
the other hand, has a rich clitic system that needs to be used maximally.
�es not present the deniteness eect, whereas GS does.
e acquisition of locatives and existentials in L2 Spanish
�e acquisition of the Spanish
ser/estar
contrast is one of the characteristic di
culties for speakers whose native language has a unique
system. Some stud
ies have been devoted to the acquisition of this contrast in Spanish L2 speakers,
e locative paradigm in the L2 Spanish of Catalan native speakers
particularly in English-speaking learners (
Ryan & Laiord 1992
VanPatten 1985
), but practically no other language combination has been investigated. Most
of the studies have explored the acquisition of the copulas when combined with
adjectives (
Bruhn de Garavito & Valenzuela 2006
Geeslin 2002
Schmitt 1992
Schmitt & Miller 2007
), but not so much in other contexts, with the exception
of the studies on child psychology carried out by Sera and colleagues (
Sera 1992
Sera, Bales & del Castillo 1997
). �e studies with English-speaking learners have
shown that
estar
for locative contexts appears late in the development of the copu
las.
VanPatten (1985
) proposed sve developmental stages for the acquisition
of the
ser/estar
contrast. First, there is a short period of omission of the copula, fol
lowed by a period of overproduction of
ser
for all contexts (second developmental
stage).
Estar
appears in the third stage, but only as an auxiliary for progressives. It
is not until the fourth developmental stage that
estar
is used for locative contexts
(later according to
Ryan & Laiord 1992
⤮‟e nal developmental stage is when
both
ser
and
estar
are used with adjectives.
Ryan and Laiord (1992
) tested English-
speaking students in 3 dierent moments, within a four and a half month study
abroad program. Overall, they found that the use of
ser
was much more accurate
(90%) than that of
estar
(40%–70% accuracy). e patterns of errors wer攚also dif
ferent. With
ser
the most common mistake was to omit the copula, while with
estar
the most common mistake was to substitute it for
ser
, particularly in the locative
contexts.
Despite the late appearance of
estar
in the language of L2 learners, several
variationist studies have documented a progressive historical replacement of
ser
with
estar,
especially when the copulas are combined with adjectives. is phe
nomenon has been attested in monolingual contexts (
Gut槩rrez 1992
), bilingual
acquisition (
Gut槩rrez 2003
Ortiz-López 2000
Silva-Corvalán 1986
) and L2
acquisition (
Geeslin 2001
). In the case of Catalan, this gradual substitution
of
ser
by
estar
has also been attested and explained as an internal change acceler
ated by contact with Spanish (
Sanz & Go湺ález 1995
Solà 1987
). For instance,
Solà (1987
) summarizes
Silva-Corvalán’s (1986
) study about the Spanish in Los
Angeles and suggests that a similar process of an accelerated expansion of
estar
an already ongoing internal change, is happening in present Catalan as the result
of language contact.
Two sociolinguistic studies have investigated the copula selection in Catalan-
Spanish bilingualism. Sanz and Go湺ález (
) investigated
ser
and
estar
�ilvia Perp槱án
Sanz and Go湺ález (1995
) found extension of
estar
over
ser
in both contexts, with
adjectives and locatives, despite the prescriptive rules proposed by Catalan gram
marians; the overextension is larger in younger speakers. In locative contexts, they
did not nd a single instance of
ser
with animate or inanimate subjects, displaying
Table 1.
Rates of copula selection, data from
Geeslin and Guijarro-Fuentes (2008
Both
Estar
Ser
Total
Catalans
Valencians
Monolinguals
Despite the small dierences among the percentages (44.9 vs. 44.5 or 43.8),
the authors found strong signicant dierences in the use of
estar
�ese authors seem to treat Valencian as a diierent ‘regional language鈺 “we have no
evidence that Catalan and Valencian diier in their rules of copula choice, but […] we have
chosen to maintain the distinction in light of government eiorts to distinguish the two and in
the absence of sociolinguistic research on copula contrast in Valencian.” (
Geeslin & Guijarro-
Fuentes 2008
, p. 366). While I acknowledge the research advantages of including another lan
e locative paradigm in the L2 Spanish of Catalan native speakers
bilinguals and the monolinguals, and concluded that Catalans and Valencians
used
estar
signicantly less frequently. �us, the authors argued that bilingualism
does not necessarily increase the frequency of
estar
, and that the language contact
situation does not automatically imply simplication of the copulas, as previously
argued. In the case at hand, it seems that the contact with Catalan and Valencian
has the opposite eect: it slows the extension of
estar
in the Spanish of these areas.
Table 2.
Locative, existential and possessive constructions in L2 Spanish,
data from
Gràcia, Crous and Garganta (2008
Group
% of errors in
total production
Types of errors
Omission (%)
Substitution (#)
Moroccan
Arabic
Estar
instead of
haber
Chinese
Tener
instead of
haber
Romanian
Estar
instead of
ser
�e authors conducted an analysis of errors and found that the existential
construction was the one that posed more diculties to the learners, and the most
common error was omission of the copula (or
haber
), particularly in Chinese and
Arabic speakers. Per L1 group, in the Arabic group, 70% of their errors consti
tuted a case of omission, especially in present tense. Also, they replaced the verb
�ilvia Perp槱án
estar
with
haber
(3/34), not obeying the deniteness eect such as in
Sí, hay el mar
cerca de m
‘Yes, there is the sea close to me’. As for the Chinese speakers, 76% of
their errors were verb omissions. e remaining errors were substitutions of the
copula with the verb
tener
(10/84), particularly in existential constructions, such
as in
es que tengo una chica de baile muy bien,
(literallyⰠ‘is that I have a girl of dance
very well), meanin朠‘�ere is a girl that dances very well’⸠�ere are some instances
of replacement of
estar
with
ser
(6/84) in locative constructions:
cuando yo era en
China
‘when I was in China’. Interestingly, the Chinese group, due to the analytical
morphology nature of their L1, committed an error that no other group did: they
Lardiere 2004
White 2003
). is feature has been less investigated in
existential constructions, with the recent exception of
White (2008
) and White
整2al. (2009,
⤮‟e main argument in White and colleagues鈠studies is that
even though the acquisition of articles is far from being errorless, L2 learners
remarkably produce very few DE violations, a related semantic phenomenon.
For instance, in their two case studies (a Chinese fossilized L2 speaker of English
and a Turkish L2 English speaker)
Lardiere (2004
) and
White (2003
) reported
article omission but no problems with DE. ese problems with article omission
e locative paradigm in the L2 Spanish of Catalan native speakers
ὥ study
Participants
�e speakers investigated in this study are sequential Catalan-Spanish bilinguals
= 20), with a mean age at the time of testing of 23.6 years old (
= 4.1). ey
reported starting their exposure to Spanish once they entered into the school
one
�rgentinean and one Peruvian. 11 of these Spaniards were born and raised
in monolingual areas of Spain, but 4 of these Spaniards were born and raised
In Catalonia, the school system is obligatorily in Catalan; Spanish language is not intro
duced until
3rd
grade.
�ilvia Perp槱án
in Catalonia, although none of them lived in Catalonia at the time of testing.
�ese 4 Catalan Spaniards come from very Spanish-speaking neighborhoods of
arcelona. e second group of Spanish native speakers
consisted of monolingual speakers (
=21), 16 from the central area of Spain and
of the superior level of DELE (Diploma de Español como Lengua
xtranjera),
欚㴚
23. Spanish speakers and Catalan speakers had exactly the same mean accu
racy in their prociency results: 94% of correct responses.
�ere were two main experimental tests, an Oral Production Task (OPT) and
a web-based Acceptability Judgment Task (AJT). e OPT consisted of a ‘Spot
the Dierence Task’, with 5 pairs of very similar pictures that participants had to
e locative paradigm in the L2 Spanish of Catalan native speakers
Ser/Estar with Objects
El
�ibro
s/está
cima
de
la
esa.

�e
book
་�
ଏaᘕ
on-top
�f
he
table

�e book is on the table.’
Ser/ *Estar with Events
La
�eunión
s/*está
el
otel
ajestic.

�e
ଏaᘕ
�he
hotel
Majestic

Haber with *d攞nites and ind攞nites.
Hay
�as/unas
laves
encima
de
la
esa.

hᘇu
�he/some
keys
n-top
he
table

�ere are the/some keys on the table.’
Estar with d攞nites and
indeṮites
El/
un
�ibro
stá
cima
de
la
esa.

ὥ/
book
ଏaᘕ
on-top
�f
he
table

�e/ a book is on the table.’
Relative Clauses:
Haber/Estar with d攞nite antecedents
Las
�laves
que
hay/están
uerta
son
mías.

�e
�eys
hat

hᘇu
ଏaᘕ
on
the
door
re
e.

�e keys that are at the door are mine.’
Results
Results of the Oral Production Task
All utterances produced in the OPT were transcribed and checked by two native
speakers. Later, these were coded with the CLAN program from the CHILDES
system
http://childes.psy.cmu.edu/clan
by two Spanish-speaking researchers. All
Figure 1.
Sample of a ‘Spot the Dierence Task’
�ilvia Perp槱án
predicates that expressed location or an existential meaning were coded⸠�at is,
all occurrences of
haber
with an existential meaning, as well as the occurrences of
estar
and
ser
accompanied by a locative adverb or a prepositional phrase express
ing location, and
llevar
‘to carry, to wear’, and
tener
‘to have鈠as possessives. Verbs
such as
existir
‘to exist’,
aparecer
ₑto appear’, their negative counterparts such as
fal
tar
or
hacer falta
‘to lack, to miss’, if followed by a locative expression were included
under the categor礠‘other’. Also verbs such as
ver
‘to see鈠with a presentational/
existential meaning in sentences such as
veo una ardilla
en la foto A
⢑I see a squir
rel in photo A鈩⁷ere also included under th攠‘other鈠class. In addition, verbs were
In the case of the
panish
speakers, the next most frequent verb was
tener
, the possessive one, but in the
case of the Catalan speakers,
llevar
, a verb rarely used by Spanish speakers,
was
more common than
tener
. A Chi-Square test with the counts per group and verb
haber, estar, tener, llevar
Figure 2.
Frequency of lexical verb in locative and existential contexts
by type of speaker (in %)
e locative paradigm in the L2 Spanish of Catalan native speakers
A multivariate ANOVA was conducted with the percentages of verb use as the
dependent variables, and group as the independent variable⸠�e results indicated
signicant dierences in the use of
estar F
haber F
p = .
llevar F
tener F
p = .
047 and
ellipsis F
030, but not in the use of
ser
or
other
verbs (
�us, despite the overall similar tendency in the use of the locative verbs, we can
see signicant dierences in the use of
haber
, which Catalan speakers used more
frequently (42% vs. 29 %), and in the use of
estar,
which they used signicantly
less frequently (17% vs. 24%). e use of
ser,
a previously reported characteristic of
Catalan Spanish for locative constructions was not found signicant (.2% vs. .8%);
only 3 tokens were found in the Catalan speakers as in (13a), and 8 in the Spanish
speakers (13b).
�ses of SER in locatives

Bajando
or
el
estido,
otra
iferencia
el
olor

Descending
by
he
dress,
ther
摩ierence
་�
�he
color

de
la
�a.

he
gemstone

Going down by the dress, there is another dierence in the gemstone.’
Catalan speaker # 129

De
las
�res
anzanas
que
ay
la
oto
la

�he
three
apples
hat
hᘇu
�he
photo
he

foto
son
�atro
la
arte
arriba
derecha.

photo
་�
four
he
part
pper
ight

Of the three apples that are in the picture A, in picture B there are four
in the upper right part.’
Spanish speaker # 58
With respect to the dierences in the clitic system, we coded not only the use of
accusative clitics, but also the reference to the theme in the second part of the
�ilvia Perp槱án
El
�ey
�ntes
enía
cuatro
símbolos
arriga,
ay,

the
king
before
had
our
mbols
the
belly,
ops,

antes
�o
tenía
tres
�ímbolos
arriga
ahora
tiene

before
not
has
hree
symbols
the
belly
nd
now
as

tres
símbolos
la
arriga

three
symbols
belly

‘Before, the king had four symbols on the belly, oh, no, before he did
not have three symbols on the belly and now he has three symbols on
the belly.’
Catalan Speaker # 132

Antes
�abía
una
casa
in
alos
ejado

before
脖9
�ouse
without
sticks
on
the
roof
nd

ahora
tiene.

now
yes

Before, there was a house with no sticks on the roof, and now it does
have them.’
Catalan Speaker # 123
In Example (14a), the last
tres símbolos en la barriga

‘the last three symbols on the
belly’) was considered a repeated NP, and coded as such, since the speaker could
have chosen to refer to it with a negative polarity item such as
nada
⢑nothing’), with
an accusative clitic
lo
⢑it鈩⁡s in (14b), or with an empty element. With respect to the
verbs,
había
in (14b) was considered rst mention, but
tiene
was considered second
mention. Similarly, the rst two
te滭a
from (14a) are considered rst mention verbs,
but
tiene
was coded as second mention. Table 3 includes the tokens and percentages
of locative and existential verbs used in rst and second mention as well as the use of
accusative clitics and repeated DPs. e percentages were calculated over the total of
second mention verbs, the context in which a clitic would be legitimized.
Table 3.
Frequency and distribution of locative verbs, clitics, and repeated DPs
1st
mention verbs
2nd
mention verbs
Accusative CL
Repeated NP
L1 Catalan
Tokens and %
L1 Spanish
Tokens and %
�e percentages of clitic and repeated DPs use per participant were submitted
to an independent sample
-test and results indicated that these variables are not
statistically signicant at the.05  level, but both were marginally signicant (Per
centage of Repeated DPs use,
(2-tailed)
= .076; Percentage of Clitic
use,
(2-tailed)
�e deniteness of the
was also analyzed in the production data. Per
centages of appearance of
haber
e locative paradigm in the L2 Spanish of Catalan native speakers
the two groups (
= 28, 4.6% in the Catalan speakers vs.
= 11, 4.4% in the
Spanish speakers,
� .1). e percentages were calculated from the overall use
of
haber
En
el
�ntes
stá
ájaro
pero
hay
omo
un

�he
before
not
ଏaᘕ
the
bird
ut
hᘇu
like

cuadro
na
ventana
la
magen
de
ahora
sólo

painting
or
indow
and
he
image
ow
nly

hay
�l
烡jaro

hᘇu
the
bird

�e bird is not there in the before (picture) but there is like a painting
or a window and in the image of now, there is only the bird.’
Catalan Speaker # 20
En
la
�oto
el
�atón
hay
�l
del

In
�he
photo
the
mouse
that
hᘇu
to-the
f-the

príncipe
está
erca.

prince
ଏaᘕ
more
close

In picture A, the mouse that is next to the prince is closer.’
Spanish Speaker # 58
Interestingly, two linguistic features were found in the speech of the Catalan speak
ers that were not produced by the Spanish speakers: the occasional presence of the
Catalan locative clitic
, and agreement errors with the verb
haber
. Four Catalan
speakers produced 9 instances of the clitic
. is clitic was produced with the
past form of
haber, había
only one instance of this clitic was found with
estar
(16b). e fact that the clitic was hardly produced with a dierent verb, and
the almost homophonic nature of
había
and
e獴án
with their Catalan counterparts
havia, estan
could indicate that this is a case of language transfer or even code-
switching rather than a linguistic innovation into their L2 Spanish.
I owe this observation to an anonymous reviewer.
�ilvia Perp槱án
Antes
hi
�拭a
cuatro
pájaros
encima
de
la
oto

before
loc
脖9
�our
�irds
n-top
he
photo
to

la
�quierda,
ahora
hay
ólo
res
ájaros.

the
ow
hᘇu
only
three
birds

Before there were four birds on top of the picture to the le, now there
are only three birds.’

Antes
�o
abían
estos
res
alos
ctualmente

ue

before
not
脖9
�hese
three
sticks
and
nowadays
that

e獴án
estos
tres
alos.

loc
ଏaᘕ
these
three
sticks.

Before there were not these three sticks and now they are there, these
three sticks.’
Catalan speaker # 132
Results of the Acceptability Judgment Task
.001). When we look at the mean percentages, shown in Figure 4, we can
e locative paradigm in the L2 Spanish of Catalan native speakers
Figure 4.
Mean acceptability rates of
haber
and
estar
uses and the deniteness eect
Finally, the deniteness eect on relative clauses (12), as the ones produced
by Spanish speakers (15b) was also considered⸠�e repeated measures ANOVA
indicated a main eect of verb (
haber
estar
= .009, but no
*SerEstarSer*Estar
Figure 3.
Mean acceptability rates of
ser
and
estar
uses and the eventiveness eect
�ilvia Perp槱án
Figure 5.
Mean acceptability rates on relative clauses with a denite antecedent by verb
Discussion
�e oral production data showed a general underuse of
estar
by the Catalan speak
ers compared to the use of Spanish speakers. is nding is consistent with the data
found in typical adult L2 learners which showed a delay/underuse in the acquisi
tion of
estar
Perp槱án, 2014
Ryan & Laiord 1992
VanPatten 1987
). is nd
ing is also compatible with an explanation in terms of L1 inuence, since
estar
is
not used for locative purposes in Standard Catalan. Similarly to what
Geeslin and
Guijarro-Fuentes (2008
) found, we can also conclude that the presence of Catalan
produces a reduced use of
estar
in Spanish in locative constructions, despite the
documented overextension of this verb in Catalan (
Solà 1994
). We can only specu
late these speakers have not adopted
estar
as their regular locative verb in
atalan,
Spanish. Nevertheless, when we examined the intuitional
ata collected through
e locative paradigm in the L2 Spanish of Catalan native speakers
the AJT we can see that there is, indeed, a signicant dierence in the implicit
grammar of these speakers. Catalan speakers did not strongly reject sentences
with the verb
ser
to locate inanimate individuals, showing that in their bilingual
mind,
ser
can still function as a locative verb, independently of the preposition.
Following
Brucart (2012
), Catalan
ser
can appear in locative contexts when com
bined with the preposition
(Spanish
Estar + en
= Catalan
Ser + a
Ὡs variable
ser
+ object) in the AJT had 5 tokens, each with a dierent preposition or locative
adverb, in particular:
encima de
⢑on top of鈩
, cerca de
⢑close to鈩
, entre
�ese are the sentences included in this variable:
El libro es encima de la mesa; el estadio
de fútbol es cerca del río; la biblioteca es entre el hotel y la universidad; la puerta es a la izqui
erda de la casa; la televisión es en el salón.
�ilvia Perp槱án
existential sentences with a denite NP, and moderately accepted them in the AJT
㴚44), showing that their grammars worked dierently in that respect from the
Spanish monolingual grammars. Unlike
Russian, and assume that, somehow, the theme in existential constructions in
Catalan can escape the domain of existential closure, and then it can be inter
panish in this respect lies on a
e locative paradigm in the L2 Spanish of Catalan native speakers
Conclusions
To recapitulate, this study has shown some divergent acquisition of L2 Spanish by
L1 Catalan speakers. ese dierences are more evident when we employ experi
DOI: 10.1002/9781444393446.ch18
Boix-Fuster, E., & Sanz, C. (2008). Language and identity in Catalonia. In M. N槱o Murcia &
J⸚Rothman (Eds.),
Bilingualism and Identity: Spanish at the Crossroads with other
anguages
(pp. 87–106). Amsterdam: John Benjamins.
DOI: 10.1075/sibil.37.07boi
Brucart, J.M. (2010). La alternancia ser/estar y las construcciones atributivas de localización. In
A. Avellana (Ed.),
Actas del V encuentro de gramática generativa
(pp. 115–152). Neuquén:
Editorial Universitaria del Comahue.
Brucart, J.M. (2012). Copular alternation in Spanish and Catalan attributive sentences.
Revista
de Estudos Linguísticos da Universidade do Porto
Bruhn de Garavito, J. & Valenzuela, E. (2006). �e status of ser and estar in late and early bilin
gual L2 Spanish. In C. A. Klee & T. L. Face (Eds.),
Selected Proceedings of the 7th
Conference
on the Acquisitino of Spanish and Portuguese as First and Second Languages
(pp. 100–109).
Sommerville, MA: Cascadilla Proceedings Project.
Freeze, R. (1992). Existentials and other locatives.
Language
DOI: 10.2307/415794
�ilvia Perp槱án
Gallego, A.J., & Uriagereka, J. (2011). From ser to estar: A study of the Spanish copulas. Ms.Uni
versitat Autònoma de Barcelona & University of Maryland.
Geeslin, K.L. (2001). Linguistic simplication: Past, present and future links to second language
acquisition. In X. Bonch-Bruevich, W.J. Crawford, J. Hellermann, C. Higgins, & H. Nguyen
(Eds.),
Selected Proceedings of the 2000 Second Language Research Forum
(pp. 271–291).
Somerville, MA: Cascadilla Press.
Geeslin, K.L. (2002). e acquisition of Spanish copula choice and its relationship to language
change.
Studies in Second Language Acquisition
DOI: 10.1017/s0272263102003030
Geeslin, K.L., & Guijarro-Fuentes, P. (2008). Variation in contemporary Spanish: Linguistic pre
dictors of estar in four cases of language contact.
Bilingualism: Language and Cognition
DOI: 10.1017/s1366728908003593
Gràcia, L., Crous, B., & Garganta, L. (2008). Typological dierences and second language acqui
sition 2: Attributive, locative, existential, and possessive constructions.
Revista Española de
Lingﳭstica
Gut槩rrez, M.J. (1992). e extension of estar: A linguistic change in progress in the Spanish of
Morelia, Mexico.
Hispanic Linguistics
Gut槩rrez, M.J. (2003). Simplication and innovation in US Spanish.
Multilingua
DOI: 10.1515/mult.2003.009
Haznedar, B., & Schwartz, B.D. (1997). Are there optional innitives in child L2 acquisition?
Proceedings of the Annual Boston University Conference on Language Development
Second Language Acquisition of Spanish Morpho-Syntax by Quechua-
Speaking Children
. Unpublished Ph.D. dissertation, University of Southern California,
Los Angeles, CA.
Kayne, R. S. (1994)
se Antisymmetry of Syntax
. Cambridge, Mass: MIT Press.
Lardiere, D. (1998). Dissociating syntax from morphology in a divergent L2 end-state grammar.
Second Language Research
DOI: 10.1191/026765898672500216
Lardiere, D. (2004). Knowledge of deniteness despite variable article omission in second lan
guage acquisition.
Proceedings of the Annual Boston University Conference on Language
Development
Lardiere, D. (2008). Feature assembly in second language acquisition. In J. M. Liceras,
�oodluck, & H. Zobl (Eds.),
se Role of Formal Features in Second Language Acquisition
(pp.106–140). Mahwah, NJ: Lawrence Erlbaum Associates.
Lardiere, D. (2009). Some thoughts on the contrastive analysis of features in second language
acquisition.
Second Language Research
DOI: 10.1177/0267658308100283
Longa, V.M., Lorenzo, G., & Rigau, G. (1998). Subject clitics and clitic recycling: Locative sen
tences in some Iberian Romance languages.
Journal of Linguistics
DOI: 10.1017/s0022226797006853
Lumsden, M. (1988).
Existential Sentences: eir Structure and Meaning
. London: Croom Helm.
DOI: 10.1017/s0022226700012329
e locative paradigm in the L2 Spanish of Catalan native speakers
Lyons, J. (1967). A note on possessive, existential, and locative sentences.
Foundations of Lan
guage
Lyons, C. (1999).
De�niteness
. Cambridge: CUP.
DOI: 10.1017/cbo9780511605789
Milsark, G.L. (1977). Toward an explanation of certain peculiarities of the existential construc
tion in English.
Linguistic Analysis
Ortiz-López, L. (2000). La exten獩ón de estar en contextos de ser en el español de Puerto Rico:
뽅volución interna o contacto de lenguas?
Boletín de la Academia Puertorriqueña de la
Lengua Española
Perp槱án, S. (2014). Locative and existentials in L2 Spanish: e acquisition of semantic con
trasts among
ser
estar
and
haber
Second Language Research, 30
DOI:
Prévost, P., & White, L. (2000). Missing surface inection or impairment in second language
acquisition? Evidence from tense and agreement.
Second Language Research
DOI: 10.1191/026765800677556046
Rigau, G. (1994). Catalan presentational sentences and the properties of Agr nodes. In G.Cinque,
J. Koster, J.-Y. Pollock, L. Rizzi, & R. Zanuttini (Eds.),
Paths towards Universal Gram
mar: Studies in Honor of Richard S. Kayne
(pp. 343–359). Washington, DC:
University Press.
Rigau, G. (1997). Locative sentences and related constructions in Catalan: ésser/haver alter
Morphology-
syntax Interface
(pp. 395–421). Bilbao & San Sebastián: Universidad del País
Vasco & Diputación Foral de Gipuzkoa.
Ryan, J., & Laiord, B. (1992). Acquisition of lexical meaning in a study abroad environment:
Ser and estar and the Granada experience.
Hispania
DOI: 10.2307/344152
Saᱲ, K.J. (1985).
Syntactic chains
. Cambridge: CUP.
DOI: 10.1017/s0022226700011932
叡nchez, L. (2003).
Quechua-Spanish Bilingualism: Interference and Convergence in Functional
Categories
. Amsterdam: John Benjamins.
DOI: 10.1075/lald.35
叡nchez, L. (2004). Functional convergence in the tense, evidentiality and aspectual systems of
Quechua Spanish bilinguals.
Bilingualism: Language and Cognition
DOI: 10.1017/s136672890400149x
叡nchez, L. (2015). Crosslinguistic inuences in the mapping of functional features in
Quechua-Spanish Bilingualism. In T. Judy & S. Perp槱án (Eds.),
se Acquisition of Spanish
in Understudied Language Pairings
(pp. 21–47). Amsterdam: John Benjamins.
Sanz, C., & Go湺ález, M.J. (1995). Ser and estar in Tortosi Catalan: Language contact, language
variation, and language change.
Sintagma
Schmitt, C. (1992). Ser and estar: A matter of aspect. In K. Broderick (Ed.),
Proceedings of NELS
(pp. 411–426). Amherst, MA: GLSA.
Schmitt, C., & Miller, K. (2007). Making discourse-dependent decisions: �e case of the copulas
“ser鐠an搠“estar” in Spanish.
Lingua
DOI: 10.1016/j.lingua.2006.11.007
Sera, M. (1992). To be or to be: Use and acquisition of the Spanish Copulas.
Journal of Memory
and Language
DOI: 10.1016/0749-596x%2892%2990021-o
Sera, M.D., Bales, D.W., & del Castillo, J. (1997). Ser helps Spanish speakers identif礠“real鐠prop
erties.
Child Development
DOI: 10.2307/1132035
Serrano Vquez, M. del C. (1996). Interferencias léxicas y semánticas en una situación de con
tacto entre dos lenguas, catalán y castellano.
D槡logos His烡nicos
�ilvia Perp槱án
Silva-Corvalán, C. (1986). Bilingualism and language change: e extension of estar in Los
Angeles Spanish.
Language
DOI: 10.1353/lan.1986.0023
Silva-Corvalán, C. (1994). e gradual loss of mood distinctions in Los Angeles Spanish.
Lan
guage Variation and Change
DOI: 10.1017/s095439450000168x
Sinner, C. (2004).
El castellano de Cataluña: Estudio empírico de aspectos licos, morfosintácti
DOI: 10.1515/9783110933871
Sinner, C., & Wesch, A. (Eds.). (2008).
El castellano en las tierras de habla catalana
. Madrid
Frankfurt: Iberoamericana
&Vervuert.
DOI: 10.1558/sols.v5i1.177
Solà, J. (1987).
Qüestions controvertides de sintaxi catalana
(1st ed.). Barcelona: Edicions 62.
Solà, J. (1994).
Sintaxi normativa: Estat de la qüestió
(1st. ed.). Barcelona: Editorial Emp全ies.
�omason, S.G., & Kaufman, T. (1988).
Language Contact, Creolization, and Genetic Linguistics
Berkeley CA: University of California Press.
DOI: 10.1017/s0022226700014791
Vann, R. (2002). El castellcatalanitzat a Barcelona: Perspectives lin柼ístiques i culturals.
Cata
lan Review
VanPatten, B. (1985). e acquisition of ser and estar by adult learners of Spanish: a preliminary
DOI: 10.2307/342218
VanPatten, B. (1987). Classroom learners’ acquisition of ser and estar: accounting for develop
mental patterns. In B. VanPatten, T.R. Dvorak, & J. Lee F. (Eds.),
Foreign Language Teach
ing: A Research Perspective
(pp. 61–75). Cambridge MA: Newbury House.
Wesch, A. (1997). e Spanish spoken in Barcelona and Catalan inuence. An outline of a
research program.
Verba
White, L. (2003). Fossilization in steady state L2 grammars: Persistent problems with inec
tional morphology.
Bilingualism: Language and Cognition
DOI: 10.1017/s1366728903001081
White, L. (2008). Deniteness eects in the L2 English of Mandarin and Turkish speakers.
Proceedings of the Annual Boston University Conference on Language Development
White, L., Belikova, A., Hagstrom, P., Kupisch, T., & Ö竧elik, Ö. (2009)
. Restrictions on Denite
ness in L2 English.
Proceedings of the Annual Boston University Conference on Language
Development, 33
White, L., Belikova, A., Hagstrom, P., Kupisch, T., ☠Ö竧elik, Ö. (2012). Restrictions on denite
ness in second language acquisition: A᭲mative and negative existentials in the L2 English
of Turkish and Russian speakers.
Linguistic Approaches to Bilingualism
DOI: 10.1075/lab.2.1.03whi
Zagona, K. (2012). Ser and estar: Phrase structure and aspect. In C. Nishida & C. Russi (Eds.),
Building a Bridge between Linguistic Communities of the Old and the New World.
urrent
Research in Tense, Aspect, Mood and Modality.
(Vol. XIV, pp. 303–327). Amsterdam:
Rodopi.
8ᘕᐠጓ
e acquisition of Spanish in a bilingual
Introduction
�uestion (
Schmeißer,
Hager,
Arnaus
Gil, Jansen, Geveler, Eichler, Patuto &
Müller,
to appear)). If this is true, then the language combination should
atter
�oss-linguistic inuence is to be studied (
Müller & Patuto 2009
Patuto 2013
).
In this chapter, the empirical examination of Spanish as one of two/three rst
�aia Arnaus Gil & Natascha Müller
languages contributes to the theory of bilingual L1 acquisition by providing
important information regarding the selection of copula verbs and the position
of attributive adjectives in early multilingual children. e multilingual children
acquire Spanish in dierent language combinations so they are presented with
syntactic structures that dier according to the complexity required in the pro
cess of syntactic derivation. If confronted with derivations with dierences in
complexity, the young monolingual or bilingual child will prefer the less complex
analysis (
Müller& Hulk 2001
). It is in this respect that language acquisitionists
German-Spanish;
Vila & Cort 1991
for Catalan-Spanish;
Barnes 2011
for Basque-
Spanish-English trilinguals;
Montanari 2009
for Spanish-English-
agalog). Little
attention has been paid to some linguistic phenomena, such as the ones under
study in this paper⸠�e focus of the present paper lies on two linguistic phenom
ena which relate intimately the clause and the nominal level: the use of adjectives
in verbal and nominal expressions. Spanish adjectives within nominal expressions
can generally appear before or a1er the noun, or both. At the clause level,
panish
adjectives allow either only one of two copula verbs or both. Catalan shares most
of these properties with Spanish (for dierent uses, see
Perp槱án,
this volume).
German and French, on the other hand, dier from Spanish by providing only one
phonologically realized copula form and by restricting the positional possibilities
for adjectives within the noun phrase,
with German only exhibiting prenominal
adjectives. We will see examples of these structures in the following section. In
order to study copula selection and adjective placement, we focus on some multi
lingual language combinations, all sharing Spanish as one of the L1s. Specically,
we examine early bilinguals of Spanish-German and Spanish-French as well as
early trilinguals of Spanish-Catalan-German.
French actually has both pre- and postnominal adjectives in the DP. Yet, French adjectives
tend to make use exclusively of one position. For example, color adjectives are always post
nominal and short adjectives such as
petit
only occur prenominally (
Bouchard 2002
�e copula verb(s)
se dual Spanish and Catalan system
In this section, we br楥�y describe the distribution of the copula verbs in Spanish
ther
quella
variable
matemática

nstante.

at
riable
mathematic
is-
eକ
/*is-
ଏaᘕ
constant

�.
uando
juega
el
arça,
la
udad
*es/está
ta.

hen
lays
he
Barça,
the
city
་�
/is-
ଏaᘕ
deserted
ilvia
legre,
da
al

ilvia
is-
་�
happy,
�t doesn’t
matter

enga
buen
al
擭a.

he) has
ood
or
bad
ay

�.
omás
el
egalo

Tomás
Cl-Dat
has
arrived
he
present,

or
eso

legre.

hat
s why (he)
ଏaᘕ
happy
As can be seen in Example (1a), SER is the only copula allowed in combination
with adjectives that can only express a permanent property. By contrast, ESTAR
has to be selected when the predicate expresses a temporal property (1b). �e
examples in (2) illustrate the dierent meaning of the adjective when accompa
nied by one of the copula verbs: when SER is selected, an inherent property is
assigned to the adjective (2a); when ESTAR is chosen, a temporal reading of the
�aia Arnaus Gil & Natascha Müller
�e examples in (1) and (2) reveal that there are two groups of adjectives in
Spanish. In lexical adjectives, the expression of a permanent or a temporal property
is encoded in the lexical entry (cf. the examples in 1). e syntactic derivation will
be chosen correspondingly. A second group of adjectives does not have any lexi
cal specication of a permanent/temporal property (cf. the examples in(2) above
and (3) below). Both the temporal and the permanent property will be expressed
via a particular syntactic derivation. In order to derive the dierent constructions,
Arnaus Gil (2013
) proposes an analysis for Spanish SER in which T selects a
P as
in structure (3a).
uan
apo

uan
is-
་�

�.
rofesora
está
apa
SLP

�e
teacher
oks-
ଏaᘕ
esguapo
Evidence for this assumption comes from Spanish passive constructions, where
SER is used (
La canc槳n ha sido escuchada por la banda
ₑ�e song has been-SER
heard by the band’), i.e. the form of the copula which activates a
P according
to
Arnaus Gil (2013
). Further evidence for the existence of a
P layer for SER
comes from causatives and gerundive constructions. Spanish SER can be used in
causatives as in
El profesor hace ser más valiente al alumno
⢑the teacher makes be-
SER more courageous to the student’), thus activating a
P layer. SER in
must
move further to T in order to check tense and agreement features within the TP.
In gerundive constructions, SER can also be used in its gerundive form, accom
panied by the auxiliary ESTAR as in
El vendedor está siendo muy pesado

‘�e
salesperson is-ESTAR being-SER very boring鈩⸠In contrast to SER, ESTAR can
neither be passivizer (*
La canción ha estado escuchada por la banda
ₑ⨟e song
has been-ESTAR heard by the band鈩
nor does it play a role in causativization
Su madre hace estar descalza a María
‘*Her mother makes be-ESTAR barefoot
to Maria鈩
nor can it be used in its gerundive form (*
Juan e獴á estando a曳nico
鄪Juan is-ESTAR being voiceless鈩
Arnaus Gil
) deduces from these obser
vations that ESTAR is directly inserted in T, thus a
P layer can remain inactive
�ere are adjectival passives which are indeed built with ESTAR as in
El pollo estaba
bien cocinado
‘the chicken was-ESTAR good cooked’ (
Bruhn de Garavito 2009
, p. 1) and are
claimed to be built in the lexicon (Wasow 1977, as cited in
Bruhn de Garavito 2009
). However,
De Miguel (1999
) and
Marín (2004
) suggest that there is actually a case of a passive con
struction which is built with ESTAR and needs the preposition
por
to incorporate the activity
of the agent as in
Las alergias están provocadas por muchas causas
鄨the) allergies are-ESTAR
caused by several causes’ (
De Miguel 2004
, p. 24).
Sentences such as the one above and
El maquillaje hace estar guapa a Mar
ₑ�e make-up
makes be-ESTAR pretty to Maria’ as well as
La madre hace estar quietos a los niños
ₑὥ mother
makes be-ESTAR quiet to the children’ seem to elicit variability in the responses of native
speakers’ grammaticality judgments.‟erefore, we could speak here of dialectal dierences in
the use or avoidance of causative structures with ESTAR.
�aia Arnaus Gil & Natascha Müller
they convey a path. is implies that neither the copula nor the verbal passive have
quella
variable
matemàtica

nstant.

at
riable
mathematic
is-
透eକ
�s-
ଏaᘕ
constant

�.
uan
arça,
la
utat

hen
plays
the
Barça,
the
city

està
erta.

透eକ
is-
ଏaᘕ
deserted
ílvia
legre,
igual

ilvia
is-
透eକ
happy,
(it)
esn’t
matter

bon
al
dia.

has
ood
or
bad
ay

�.
omàs
rribat
l regal,

omás
Cl-Dat
has
arrived
the present,

això

legre.

hat
s why (he)
ଏaᘕ
happy
Generally speaking, the copulas ÉSSER and ESTAR are allowed in the same con
texts as in Spanish. However, there are also some dierences: e Catalan adjec
tive
deserta
ₑdeserted鈠allows both ÉSSER and ESTAR (4b), whereas Spanish
only selects ESTAR (1b). Several Catalan adjectives seem to behave like
deserta
‘deserted鈠and therefore it can be claimed that property assignment in Catalan
One reviewer poses the question as to whether assigning an IL/SL quality aser insertion of
the copula in
(SER) or T (ESTAR) matters. In fact, property assignment takes place in a head-
complement relation, both for SER and for ESTAR, as transitive verbs assign accusative case
to their complements; therefore, a琠�rst sight, no diierence should be observed in this respect.
takes place
ntactically far more frequently than in Spanish. Along the same lines,
Catalan locative expressions can also pattern with adjectives like
deserta
, as
Ramos
Alfarín (2008
) observes:
eva

is-
透eକ
at
the
his
ouse

�.

eva

is-
ଏaᘕ
at
the
his
ouse
�e Catalan locative examples in (6) show that both copulas are allowed in certain
contexts and/or dialects, whereas in Spanish only ESTAR is permitted in this con
text (for more on Catalan locatives, see
Perp槱án,
this volume). As in (5) an搚(6),
oan
està
apo
SLP.

�e
Joan
looks-
ଏaᘕ

�.
oan
apo
ILP.

�e
Joan
is-
透eକ
handsome
estàguapo
�aia Arnaus Gil & Natascha Müller
It is thus the syntactic derivation of ESTAR which is more complex than ÉSSER in
Catalan, since the former involves more movement steps.
In what follows, the copula systems of German and French, SEIN an搠ÊTRE
respectively, will be presented and compared to the Spanish and Catalan copula
verbs SER/ÉSSER and ESTAR.
se unitary Germanic/French system
German (8) and French (9) only have one copula verb, SEIN and ÊTRE respec
tively⸠�e Spanish examples presented in (1) and (2) have translations that use
only one copula verb:
mathematische
Variable
ist
onstant.

at
athematic
riable
is-
eଓp
constant

�.
arça
spielt,
ist
tadt
leer.

hen
Barça
plays,
eଓp
the
city
erted

ilvia
ist
lﱣklich,
ist

ilvia
is-
eଓp
happy,
esn’t
matter

guten
oder
schlechten
Tag
hat.

she
ood
ay
as

�.
omás
hat
as
Geschenk
erhalten,

omás
has
the
resent
eceived,

egen
st
glﱣklich.

hat is why
eଓp
he
so
happy
variable
mathatique
est
nstante.

at
riable
mathematic
êaᔋ
constant

�.
uand
Barça
joue,
ille
est
erte.

hen
arça
plays,
the
city
êaᔋ
deserted

ilvia
est
ntente,
n’importe

ilvia
is-
êaᔋ
�appy,
t doesn’t matter

u’
lle
assé
une
bonne
ou
mauvaise
journée.

she
as
had
ood
ay

�.
adeau
our
Tomás
est
rrivé,
鈠est

present
for
omás
has
arrived,
that is
the

ison
our laquelle
ntent.

eason
why
is-
êaᔋ
happy
As the German and French examples show, both SEIN an搠ÊTRE can be employed
unitary copula systems such as German or French will assign the corresponding
permanent/temporal property to predicates by means of a dierent syntactic posi
tion in the derivation, as the following structures in (10) show, taken from
Arnaus
Gil (2013
, p. 108).
as
Auto
ist
putt
SLP.

car
eଓp
broken

�.
ans
ist
telligent
ILP.

Hans
is-
eଓp
intelligent

La
oiture
assée
SLP.

�e
car
êaᔋ
broken

�.
ans
est
intelligent
ILP.

Hans
is-
êaᔋ
intelligent
In German and French, SEIN an搠ÊTRE gure in passive constructions:
Die Haare
sind gefärbt/les cheveux sont coloriés
ₑthe hair is coloured’⸠�ese constructions
�aia Arnaus Gil & Natascha Müller
Stage-Level-property.
Arnaus Gil (2013
) takes this observation as evidence in favor
of the assumption that a
P layer is necessary in order to derive the copulas SEIN
an搠ÊTRE for temporal properties. In this respect, French and German resemble
the dual copula system of Catalan in that they provide a syntactic derivation which
contains a
Previous studies on the early acquisition of the Spanish SER and ESTAR
�e rst studies on Spanish copula acquisition date from the 1970s and 1980s.
In her early studies,
Hernández Pina (1979
) examines the acquisition of
Spanish by a monolingual child up to 3;0. e rst copula productions found in
the data are with ESTAR to express a state of the subject referent as in
nena tá
cota da
(la nena está acostad愠‘the girl is sleeping’). During this very rst period,
the studies from
Hernández Pina (1979
) and
Corpas (2007
) also observe
that copula omissions are frequent, although it seems that SER is more prone to
omission. Since adjectives are those predicative contexts which tend to capture
most of the omissions,
Sera, Bales and del Castillo Pintado (1997
) carried out an
Remember that French use猠ÊTRE for both passive constructions and state passives and
that is why, under this proposal, a
P layer is always needed.
experiment with L1 Spanish and L1 English children that found evidence for pos
tulating that the dual copula system ben攜ted the children by helping them dif
Holtheuer & Miller 2004
Sera 1992
Liceras, Fernández Fuertes, de la Fuente &
ercedor
叡nchez 2010
Silva-Corvalán & Montanari 2008
), whereas other investigations suggest that it
is the lexical transparency of SER for permanent and ESTAR for temporary prop
erties that favors copula acquisition in English, measured here by means of BE
omissions (
Fernández Fuertes & Liceras 2008
Fernández Fuertes & Liceras 2010
*El
�to
estaba
alo
tiene mal carter
�icolás 2;7,16

cat
ଏaᘕ
�ad
�as a bad character)
�e study by
Arnaus Gil (2013
) investigates early bilinguals who acquire Spanish in
combination with either a Germanic or a Romance language, i.e. German-Spanish
�aia Arnaus Gil & Natascha Müller
and French-Spanish and observes similar results as with the early Spanish-English
bilinguals. Our present study on the early acquisition of Spanish by Spanish-
German bilinguals will include the data analyzed in
Arnaus Gil (2013
). For this
reason, we summarize the results of
Arnaus Gil (2013
Attributive adjectives
se Spanish adjective system
Scalar or qualitative adjectives such as
famosa
rubia
Bouchard 1998
Gumiel-
Molina & Pez-Jimez 2012
) are the most frequent adjective type in the speech
of young children. �ey assign a certain property to the objects denoted by the
noun. From a syntactic perspective, they are allowed in both pre- and postnomi
nal position in Spanish and Catalan (
Picallo 2002
, p. 1655), although the post
nominal position is more frequent. Adjectives that allow both positions are called
non-
�ategorical adjectives. e dierent adjectival positions found in Spanish and
Catalan produce meaning dierences.
Bouchard (1998
, p. 146) observes that post
the rst map properties onto individuals and thereby produce a new extension of
the noun with which they occur, the latter do not modify the extension of N but
predicate the whole group of objects denoted by the N. Crucially, the change of
meaning is restrictive in the case of postnominal adjectives and non-restrictive in
the case of prenominal adjectives.
Still, not all Spanish adjectives allow both positions. As
Demonte (2008
) and
other linguists have observed, one group of Spanish adjectives is only allowed
either exclusively before or a1er the noun, such as the classifying or relational
adjectives
e泩ctrico
‘electric鈠or
nuclear
‘nuclear鈠and modal or epistemic adjectives
such as
presunto
‘alleged’, as the examples in (12) and (13) show:
able
e泩ctrico/*el
泩ctrico
cable

he
cable
electric/*the
electric
able

�.
ergía
nuclear/*la
uclear
energía

he
energy
uclear/*the
nuclear
energy
�l
resunto
揳mplice/*el
mplice
resunto

�e
alleged
mplice/*the
accomplice
alleged
Adjectives which occur in only one position are generally called categorical adjec
tives. As for the syntactic derivation of adjectives, some linguists have proposed
an adjunction analysis to the le or to the right within NP (
Abney 1987
). Later
analyses have postulated a common base position for all adjectives. e dierent
surface positions are derived by means of N and/or Adj movement (
Alexiadou
Cinque 1995
Demonte 2008
Kayne 1994
DP [D la] [
AP pálida] [
IP [NP luz] [
AP pálida
the pale light’
DP [D la] [
F luz] [
AP pálida] [
N luz
IP [NP luz
I [AP pálida
�the light pale’
�e rst movement step of the scalar adjective is similar for a pre- and a post
nominal position, namely to Spec CP. For adjectives that appear a1er the noun,
other derivational steps are needed⸠�e next derivational step involves movement
�aia Arnaus Gil & Natascha Müller
CP. F stands for Functional category and might equal Number. In other words,
prenominal adjectives require AP-movement only, whereas postnominal ones are
the result of more derivational steps (AP- and NP-movement). us, the syntactic
derivation of prenominal adjectives is more economical than the one of postnomi
nal adjectives.
Spanish and Catalan scalar adjectives are mainly postnominal in French
and exclusively prenominal in German, therefore, the following question arises:
how do French and Germanic languages convey a restrictive and non-restrictive
meaning when the language only allows one position? One proposal is provided
by
Demonte (2008
) who postulates N-movement a1er PF to convey the restrictive
Previous studies on the acquisition of Spanish adjectives
Adjective placement is not problematic in monolingual Spanish children: post
nominal and prenominal placement is mastered by the children from early on
(cf.
Predictions for the acquisition of copula selection and
adjective placement
Research in the eld of early child bilingualism has shown that the child’s two lan
guages may inuence each other (cross-linguistic inuence). According to
Müller
and Hulk (2001
Hulk and Müller (2000
) and
Müller and Patuto (2009
), cross-
linguistic inuence is a child-internal process.
Hulk and Müller (2000
) and
Müller
and Hulk (2001
) have dened two conditions under which cross-linguistic inu
ence is likely to occur, one of which is shared by many researchers (cf.
Serratrice
& Sorace 2003
Serratrice, Sorace & Paoli 2004
; among others). e rst condition
claims that
the vulnerable grammatical phenomenon is an interface property, e.g.
located at the syntax-semantics interface. As such, we are faced with a
rammatical
phenomenon located at an internal interface. According to
Sorace and Serratrice
), cross-linguistic inuence is not expected here, but, as we will see, this does
not seem to be the case.
�e second condition for cross-linguistic inuence is that the vulnerable
grammatical phenomenon must be expressed in the two languages in such a way
as to show overlap of the two constructions at the surface.
Müller and Patuto
, p. 23) have r攜ned this condition in the following way㨠“�e surface strings
of the two languages A and B are derivable in terms of the syntactic derivation of
one language (which is less complex).”
�e direction of the cross-linguistic 楮�uence is related to computational
complexity (i.e. a syntactic analysis in language A is more complex if it requires
more movement steps than the derivation in language B; cf.
Gavarró 2003
Hul欚&
Zuckerman 2000
Zuckerman 2001
⤮‟erefore, we hypothesize that, when con
fronted with a grammatical phenomenon which has a complex and a less complex
derivation available, the bilingual child will use the computationally less complex
derivation available for both languages.
A1er all, the two linguistic phenomena under investigation, copula selec
tion and adjective placement, show diverging syntactic structures among the
languages, which makes it possible to observe cross-linguistic 楮�uence. In addi
tion, the grammatical structures in the studied languages exhibit dierences in the
computational complexity of their syntactic derivations. is enables us to predict
�or the acquisition of Spanish copula verbs, the child will frequently make use
of the less complex analysis. Regarding production, we expect an overgener
alization of the Spanish copula verb ESTAR to SER contexts, since ESTAR is
syntactically less complex, following our proposal (cf. (3)). Since the analysis
of the Catalan data enables us to underline the plausibility of the syntactic
analysis of the copula verbs in Spanish, we formulate the inverse predication
for the Catalan data, namely that in Catalan, ÉSSER will take over ESTAR
contexts for exhibiting the less complex derivation (cf. (13)).
�or the acquisition of adjective placement in Romance, the bilingual child has
a less (A+N) and a more complex syntactic operation (N+A) available. We
predict that s/he will make use of the less complex analysis for those contexts
in which the more complex, postnominal analysis is required.
In what follows, the longitudinal production data analyzed in this study is pre
sented with the main results on acquisition of copula verbs and adjectives in the
Romance languages in early bilingual and trilingual children.
�aia Arnaus Gil & Natascha Müller
Participants
In Table 1, the relevant information concerning the multilingual children is pre
sented. In addition to the data from ten multilingual children,
two monolingual
Spanish children from the CHILDES database (
MacWhinney 2000
) were also
included as a control group. Both the bilingual and trilingual children were raised
simultaneously with two/three languages and have received input of at least two
of their L1s at home⸠�e languages we focus on are Spanish (Sp), German (Ger),
French (Fr) and Catalan (Cat). e subjects were recorded in regular intervals
(mostly every two weeks) for half an hour in each language, generally starting
before the age of 2 until the 5th birthday. is allowed us to examine the linguistic
data of approximately 50 recordings per language (cf. Table 1). e children who
lived in Germany were visited at their homes by a research team formed by two
monolingual native speakers of the languages in which the recordings took place.
In the Romance countries, the parents carried out these recordings. e interac
tions with the children took place in monolingual situations: while one person was
playing with the child, the other was video-recording.
Table 1.
Monolingual and multilingual child data
Child
Language
combination
Age
Number of
recordings
Total number
of utterances
Arturo
Sp- Ger
Teresa
Sp- Ger
Lucas
Sp- Ger
Erik
Sp- Ger
Nora
Sp- Ger
�e research projects were funded by the
German Science Foundation
(DFG) by grants
given to Natascha Müller from 2005–2013. 1. Die Architektur der frühkindlichen bilingualen
Sprachfähigkeit: Italienisch-Deutsch und Franzisch-Deutsch in Italien, Deutschland und
Frankreich im Vergleich (Wuppertal 2005–2008); 2. Code-Switching bei bilingual aufwach
senden Kindern in Deutschland, Italien, Frankreich und Spanien: Italienisch-Deutsch,
Franzisch-Deutsch, Spanisch-Deutsch, Italienisch-Franzisch, Italienisch-Spanisch,
Franzisch-Spanisch (Wuppertal 2009–2013). For a detailed description of the data, cf.
Cantone, Kupisch, Müller and Schmitz (2008
);
Müller, Kupisch, Schmitz and Cantone (2011
);
Hauser-Grüdl, Arencibia-Guerra, Witzmann, Leray and Müller (2010
) and Müller (in prep.).
Continued
Child
Language
combination
Age
Number of
recordings
Total number
of utterances
Syca-In
Sp-Fr
Frank
Sp-Ger-Cat
Milena
Sp-Ger-Cat
Eric
Sp-Ger-Cat
Kilian
Sp-Ger-Cat
Emilio
Irene
As can be seen from Table 1, the multilingual data includes early bilingual
Spanish-German/French and early trilingual children, the latter being raised with
Spanish, Catalan and German in Barcelona. is valuable data is unique.
�e data on the acquisition of Spanish copula verbs
Figure (1) shows the overall use of the two Spanish copula verbs SER and ESTAR
in all possible predicative contexts (i.e. NP/DP, PP, AP) for monolingual, bilin
gual Spanish-German as well as Spanish-French and trilingual children separately.
According to
Liceras, Fernández Fuertes, de la Fuente & Tercedor Sánchez (2010
), this
Copula omissions that are also found in adult speech (e.g. answers to questions using the
copula as in (i) and enumerations as in (ii)) were not included as target-deviant omissions in
child’s speech.

ónde estel libro? El libro, aq痭
ₑWhere is the book? e book, here.’

l osito aquí y el panda aq痭⸠‘�e little bear here and the panda here.’
Table 1.
(Continued)
�aia Arnaus Gil & Natascha Müller
aría es cansada.

�.
edro está un doctor.
l coche Ø vacío.

�.
uan Ø inteligente.
64162479211
Figure 1.
ones who concentrate the highest rate of copula omissions, reaching 60% of the
total amount of their SER productions. Concerning the overall grammatical use
of both SER and ESTAR on the one hand, and the overall ungrammatical use of
Examples(1a) and (1b) of the (lexical) adjectives
constante
and
desierto
, and a much
more free variation as the cases in (2) of the (lexical-syntactic) adjectives
alegre
on the other hand. In this regard, adjectives are a good testing ground to examine
copula acquisition, since they allow us to observe the semantic and syntactic selec
tion mechanisms that children might employ in order to assign the corresponding
property to the predicate through the selection of SER or ESTAR. Table 2 focuses on
the multilingual children and their use of the appropriate copula verb with a certain
adjective type, particularly that of the kind
alegre
that allows both SER and ESTAR.
Table 2.
Distribution of (un)grammatical productions with Spanish lexical-syntactic
adjectives (adapted from
Arnaus Gil 2013
, p. 287)
Lexical-syntactic IL
Lexical-syntactic SL
o (=este)
mu
uy)
ueno.
eresa (3;1,0)

ଏaᘕ
very
�ood

he child shows the adult a crocodile and the adult asks her to put the
animal away because she is scared⸠�e child then produces the sentence
and means that the crocodile has a good heart.]
�aia Arnaus Gil & Natascha Müller

�.
sto
arrón.
ucas (3;0,17)

*is-
ଏaᘕ
brown

he child and his father are drawing several animals in a book. e child
realizes that the lion is already drawn and has the color brown. en, he
produces the sentence.]
sto
amb槩n
ojo.
ik (2;1,4)

s (is-
་�
also
ed

he child is playing with Legobricks that are dierent colors and is
collecting the red ones with the aid of his father.]

�.
ol
rande.
rank (2;10,9)

e) sun (is-
་�
big

摩ierent objects in the picture and tells his mother what he sees.]
Table 3 presents the (un)grammatical productions of SER and ESTAR accompa
nying adjectives like the ones of the type
constante
or
desierto
. Recall that these
adjectives exclusively select one copula verb.
Table 3.
Distribution of (un)grammatical productions with Spanish lexical adjectives
(adapted from
Arnaus Gil 2013
, p. 287)
Lexical IL
Lexical SL
Interestingly, these data show that bilingual and trilingual children consider
lexical adjectives and the quality they are assigned to quite dierently, at least
when temporal properties are concerned: While bilinguals seem to have captured
Since the total amount of lexical IL adjectives for the trilingual group is very low (8 cases)
and the target-like and target-deviant productions are equally low (4 cases each), we will not
take these productions into consideration.
stán
uy
tan
fr.
rturo (3;2,10)

ey) *are-
ଏaᘕ
very
�o

�.
se
uy
pido.
ucas (2;1,14)

s (is-
་�
very
�st
ucio.
rank (3;1,4)

ere
already
*is-
་�
dirty

�.
⫉se
roto.
rank (2;5,2)

s (is-
ଏaᘕ
broken
Keeping Table 3 and the Examples (17)–(20) above in mind, we have observed
that trilinguals seem to evidence some diculties in assigning the temporal prop
Omissions and commissions have been considered ungrammatical productions, since
they deviate from adult speech. Yet, we d漠�nd omissions in the longitudinal production data
that have been considered grammatical because they represent copula ellipses and therefore
they do not belong to the group of target-deviant omissions. For the grammatical cases, we
show in (i) an omission of Arturo (3;10,13):

dult: y se enfadmucho con el gato/hm/¿estaba contento el gato? /

hild: no /

dult:₿no?/¿questaba? /

ild: triste
�aia Arnaus Gil & Natascha Müller
those
�panish adjectives considered to belong to the lexical adjectival group (e.g.
inteligente
europeo
cansado
cerrado
Examples(21) are extracted from the trilingual data.
Figure 2.
(Un)grammatical productions of Catalan copulas with
lexical
adjectives
by trilinguals

globo
tallat

ଏaᘕ
) (the)
balloon
cut?

�.
anguera
ollada.
ilian (3;2,5)

ose (is-
ଏaᘕ
�iled up.


o.
ic (1;10,11)

ଏaᘕ
good.

he child takes an apple and starts eating it. He wants to give his
father a bite and oiers him the apple. Later on, he continues eating it.]
As Figure 2 shows, trilingual children have, in fact, some diculties when using
ESTAR in Catalan, that is to say, when assigning a temporal property to the adjec
�e data on the acquisition of Romance attributive adjectives
�e same multilingual children examined in Section 4.2 are examined again here.
In this case, we aim to illustrate the Romance pattern for adjective placement for
(un)grammatical productions of all bilinguals and trilinguals.
‟is is seen in
(Sp-
�er-Cat), Catalan is the language represented, which belongs to the Spanish-
German-Catalan trilinguals.
For two trilingual children (Eric and Kilian), there are only Catalan and German
recordings available.‟is is why the Spanish language of these children is not represented in
Figure3.
�aia Arnaus Gil & Natascha Müller
SP (Sp-Ger-Cat)FR (Fr-Sp)SP (Fr-Sp)SP (Sp-Ger)
Figure 3.
(Un)grammatical adjective placement for all Romance languages of the multilingual
children
French (Fr-Sp)Spanish (Fr-Sp)Spanish (Sp-Ger)Catalan (Sp-Ger-Cat)
Figure 4.
Adjective placement in Romance depending on (non)categorical placement; bilin
gual and trilingual children
As Figure 3 shows, Spanish-German bilingual children produce the highest
order in a nearly error-free way⸠�e trilinguals always produce Catalan adjectives
according to the adult system, and they perform at ceiling in Spanish, too.
French (Fr-Sp)Spanish (Sp-Ger)Spanish (Fr-Sp)
Figure 5.
Ungrammatical productions of adjectives in Spanish, pre- and postnominal
�aia Arnaus Gil & Natascha Müller
Figure 5 above illustrates that this is exactly what happens. We observe that
Discussion and conclusions
Ὡs study has provided data concerning the early acquisition of the Spanish copula
verbs SER and ESTAR⸠�e children’s productions have shown that monolinguals
and multilinguals behave 摩ierently with respect to copula assignment to perma
nent/temporal properties of Spanish adjectives. We have observed that whereas
monolingual Spanish children demonstrate relatively few problematic instances
of SER and ESTAR assignment (cf. Figure 1), multilinguals, and
panish-German
assignment to the predicative adjective also takes place syntactically, that is, SEIN
(o爠ÊTRE) is inserted either in T or
to assign the corresponding characteristic
to the adjective (a temporal or permanent property, respectively). Following this
idea, condition 1 for cross-linguistic inuence (
Hulk & Müller 2000
Müller &
Hulk 2001
) is fu氜lled, namely that the linguistic phenomenon of SER and ESTAR
�aia Arnaus Gil & Natascha Müller
converge. Prediction (2) in Section 3 proposed that bilingual children have a less
(A + N) and a more (N + A) complex analysis available in Romance. As such, it was
predicted that the bilingual child would make use of the less complex analysis for
contexts in which the more complex, postnominal analysis was expected.
placement mainly concerns adjectives that are allowed in two positions in
panish.
�f language A exhibits a complex syntactic derivation (SynD+) and a less com
plex syntactic derivation (SynD), and language B exclusively provides the less
complex syntactic derivation (SynD), the multilingual child is more likely to
use SynD in language A in contexts in which SynD+ is expected.
Notice that the trilinguals had a small advantage over the bilinguals regarding
�f language A exhibits a complex (SynD+) and a less complex syntactic deriva
tion (SynD) and language B is similar to language A in providing SynD and
SynD+, the multilingual child is more likely to use SynD+ as required by the
If this tentative generalization is on the right track, it would thus predict a faster
development of Catalan and Spanish copula verbs and adjective placement by
bilingual Catalan-Spanish children, than what is expected for their monolingual
peers, since in these two languages both the less and the more complex syntactic
derivation are available.
A nal note with respect to the trilingual children is in order here: these chil
dren are trilinguals from birth and they outperform the bilinguals. is observa
Amsterdam: John Benjamins.
DOI: 10.1017/s0022226702211986
Arche, M.J. (2006).
Individuals in Time
. Amsterdam: John Benjamins.
DOI: 10.1017/s0022226708005215
Arnaus Gil, L. (2013).
La selección copulativa y auxiliar: Las lenguas romances
español –
italiano2– catalánₖ fra湣és
y el alemán en contacto. Su adquisición en niños bilingües y
trilingües.
Tübingen: Narr.
Arnaus Gil, L. (To appear). Acquisitional advantages of simultaneous 3L1 trilingual children:
�e Spanish copulas SER and ESTAR.
Educational Linguistics.
Arnaus Gil, L., Eichler, N., Jansen, V., Patuto, M., & Müller, N. (2012). e syntax of code switch
ing and the le periphery of DPs: Mixed DPs containing an adjective. Evidence from bilin
gual child language. In K. L. Geeslin & M. Díaz-Campos (Eds.),
Proceedings from the His
panic Linguistics Symposium 2010
. (pp. 242–257). Somerville, MA: Cascadilla Proceedings.
Austin, J. (2001).
Language Di�erentiation and the Development of Morphosyntax in Bilingual
Children Acquiring Basque and Spanish.
Unpublished Ph.D. dissertation. Cornell Univer
sity, Ithaca, NY.
Barnes, J. (2011). e inuence of child-directed speech in early trilingualism.
International
Journal of Multilingualism
DOI: 10.1080/14790711003671861
Barr旱a, A. (1997). Desarrollo diferenciado de sistemas gramaticales en un niño vasco-español
bilin柼e. In A. T. Pez-Leroux & W. Glass (Eds.),
Contemporary Perspectives on the Acqui
sition of Spanish
(pp. 55–74). Somerville, MA: Cascadilla Press.
Becker, M. (2000).
se Development of the Copula in Child English: e Lightness of be.
Unpub
lished Ph.D. dissertation. University of California, Los Angeles.
Bernstein, J. (2001). e DP hypothesis: Identifying clausal properties in the nominal domain.
In M. Baltin & C. Collins (Eds.),
se Handbook of2Contemporary Syntactic“eory
(pp.536–561). Malden, MA: Blackwell.
DOI: 10.1002/9780470756416.ch17
�aia Arnaus Gil & Natascha Müller
. Linguistic Inquiry
onographs.
Cambridge, MA: e MIT Press.
DOI: 10.7551/mitpress/9780262014168.001.0001
Corpas, R. (2007). ‘Ser鈠an搠‘estar’ in early child Spanish. Paper presented in
Ontario Dialogues
on the Acquisition of Spanish
. University of Ottawa, March 2 2007.
Crain, S., & �ornton, R. (1998).
Investigations in Universal Grammar: A Guide to Experiments
on the Acquisition of Syntax and Semantics
. Cambridge, MA: e MIT Press.
DOI: 10.1017/s0022226701238651
de Miguel, E. (1999). El Aspecto Léxico. In I. Bosque & V. Demonte (Eds.),
Gra淡tica Descrip
tiva de la Lengua Española,
Vol. 2 (pp. 2977–3060). Madrid: Espasa Calpe.
de Miguel, E. (2004). Q痩 signican aspectualmente algunos verbos y q痩 pueden llegar a sig湩s
car. In J. L. Cifuentes & C. Marimón (Eds.),
Estudios de Lingüística: El verbo
(pp⸚167–206)
Alicante: Estudios de Lin柼ística de la Universidad de Alicante.
Demonte, V. (2008). Meaning-form correlations and adjective position in Spanish. In
�enned礚& L. McNally (Eds.),
se Semantics of Adjectives and Adverbs
(pp. 71–100).
Oxford: OUP.
Ezeizabarrena, M.J. (1996).
Adquisición de la morfología verbal en euskera y castellano por niños
bilingües
. Vitoria: Servicio Editorial del País Vasco.
Ezeizabarrena, M.J. (1997a). Innitivos verbales en el lenguaje de los niños bilingües. In L.䓭慺2&
C. Pez (Eds.),
Views on the Acquisition and Use of a Second Language
(pp⸚177–190).
Barcelona: Universidad Pompeu Fabra.
Ezeizabarrena, M.J. (2002). Root innitives in two pro-drop languages. In A. T. Pez-
eroux
& J. Liceras (Eds.),
se Acquisition of Spanish Morphosyntax. e L1/L2 Connection
(pp.35–66). Dordrecht: Kluwer.
DOI: 10.1007/978-94-010-0291-2_2
Fernández Fuertes, R., & J. Liceras. (2008). Copula omission in the English developing gram
mar of English/Spanish bilingual children. Paper presented in
International Conference on
Models of Interaction in Bilinguals
. Bangor, Wales, UK.
DOI: 10.1080/13670050.2010.488285
Fernández Fuertes, R., & Liceras, J. (2010). Subject and copulas: e directionality of interlin
guistic 楮�uence. Paper presented in
16. Wuppertaler Linguistisches Kolloquium. Languages
in Contact: Cross-linguistic Iniuence on Di�erent Domains in Bilingual First Language
Acquisition
. Bergische Universität Wuppertal, Germany.
Fernández Leborans, M.J. (1999). La predicación: Las oraciones copulativas. In I. Bosque &
V⸚Demonte (Eds.),
Gramática descriptiva de la Lengua Española
(pp. 2358–2460). Madrid:
Espasa Calpe.
Gallego, A., & Uriagereka, J. (2011).
se Lexical Syntax of Ser and Estar
. Ms. Universitat
Autónoma de Barcelona & University of Maryland.
Gavarró, A. (2003). Economy and word order patterns in English-Dutch acquisition.
Bilingual
ism: Language and Cognition,
DOI: 10.1017/s1366728903001044
Gumiel-Molina, S., & Pez-Jimez, I. (2012). Aspectual composition in ser/estar + adjec
tive structures: Adjectival scalarity and verbal aspect in copular constructions.
Borealis
DOI: 10.7557/1.1.1.2321
Hauser-Grﱤl, N., Arencibia Guerra, L., Witzmann, F., Leray, E., & Müller, N. (2010). Cross-
linguistic inuence in bilingual children: Can input frequency account for it?
Lingua,
DOI: 10.1016/j.lingua.2010.06.008
Hernández Pina, F. (1979). Etapas en la adquisición del lenguaje infantil.
Infancia y Aprendizaje,
DOI: 10.1080/02103702.1979.10821759
Hernández Pina, F. (1984).
Teorías psicosociolingüísticas y su aplicación a la adquisición del espa
ñol como lengua materna.
Siglo XXI, Madrid.
DOI: 10.2307/342474
Holtheuer, C. (2009).
Learning Ser and Estar: A Study of Production and Comprehension in Chil
ean Spanish.
Unpublished Ph.D. dissertation. Australian National University, Canberra.
DOI: 10.1017/s1366728900000353
DOI: 10.1075/sic.4.1.07gar
�aia Arnaus Gil & Natascha Müller
Larrañaga, M.P. (2000).
Ergative Sprachen, akkusative Sprachen. Der Erwerb des Kasus bei bilin
gualen Kindern
. Frankfurt: Vervuert.
DOI: 10.1515/zrph.2004.757a
Lema, J. (1995). Distinguishing copular and aspectual auxiliaries: Spanish ser and estar. In
DOI: 10.1075/cilt.123.19lem
Liceras, J., Fernández Fuertes, R., de la Fuente, A.A., & Tercedor Sánchez, M. (2010). Lexi
cally-based interlinguistic inuence at the syntax-semantic interface: Copula omission
in the English grammar of English-Spanish bilinguals. In C. Borgonovo, M. Español-
Echevarrí愬2& P. Prévost (Eds.),
Selected Proceedings of the 12th
Hispanic Linguistics Sympo
sium
(pp.183–193). Somerville, MA: Cascadilla Proceedings Project.
Lleó, C. (2001a). �e interface of phonology and syntax. �e emergence of the article in the early
acquisition of Spanish and German. In J. Weissenborn & B. Höhle (Eds
Approaches to
Bootstrapping. Phonological, Lexical, Syntactic and Neurophysiological Aspects of Early Lan
guage Acquisition
(pp. 23–44). Amsterdam: John Benjamins.
DOI: 10.1075/lald.24.03lle
Lleó, C. (2001b). e transition from prenomina氠sllers to articles in German and Spanish.
In M.Almgren, A. Barr旱a, M.J. Ezeizabarrena, I. Idi慺ábal, & B. McWhinney (Eds.),
Research on Child Language Acquistion. Proceedings of the 8th
Conference of the Inter
national Association for the Study of Child Language
(pp. 713–737). Somerville. MA:
Cascadilla Press.
MacWhinney, B. 2000.
se CHILDES Project: Tools for Analyzing Talk
. Mahwah, NJ: Lawrence
Erlbaum Associates.
DOI: 10.1177/014272370002006006
Marín, R. (2004).
Entre ser y estar
. Madrid, Arco Libros.
Montanari, S. (2009). Multi-word combinations and the emergence of dierentiated ordering
patterns in early trilingual development.
Bilingualism: Language and Cognition,
DOI: 10.1017/s1366728909990265
Müller, N., & Hulk, A. (2001). Crosslinguistic inuence in bilingual language acquisition: Italian
and French as recipient languages.
Bilingualism: Language and Cognition,
DOI: 10.1017/s1366728901000116
V⸚
�gerland, & J. Granfeld (Eds.),
Mélanges plurilingues o�erts  Suzanne Schlyter 
occasion de son 65ème anniversaire
(pp. 299–319), Études Romanes de Lund 85. Lund:
DOI: 10.1515/zrs.2011.052
Müller, N. (Ed.) (in prep.).
Code-Switching: Spanisch, Französisch, Italienisch. Eine Einführung
Tübingen: Gunter Narr.
Patuto, M. (2013).
Der Erwerb des Subjekts in
Nicht-
Nullsubjektsprachen. Die Rolle des Sprache
neiniusses und der Sprachdominanz bei bilingual deutsch-italienisch, deutsch-spanisch und
französisch-italienisch aufwachsenden Kindern
. Tübingen: Gunter Narr.
Perp槱án, S. (2015). �e locative paradigm in the L2 Spanish of Catalan native speakers. In T.
Judy & S. Perp槱án (Eds.),
se Acquisition of Spanish in Understudied Language Pairings
(pp. 105–132).
Amsterdam: John Benjamins.
Barcelona: Emp全ies.
MüllerⰚN. (To appear). Related but 摩ierent: �e two concepts of language dominance and
language prociency. In C. Silva-Corvalán & J. Tr攝ers-Daller (Eds.),
Language Dominance
in Bilinguals: Issues of Operationalization and Measurement
. Cambridge: CUP.
Schmitt, C. (1992).
Ser
and
estar
: A matter of aspect.
NELS 22 Proceedings
, 411–426. Amherst,
Schmitt, C., & Miller, K. (2007). Making discourse dependent decisions: �e case of the copulas
ser and estar in Spanish.
Lingua,
DOI: 10.1016/j.lingua.2006.11.007
Schmitt, C., Holtheuer, C., & Miller, K. (2004). Acquisition of ser and estar in Spanish: Learning
lexico-semantics, syntax and discourse. In A. Brugos, L. Micciulla, and C. E. Smith,
Pro
ceedings of the 28th
annual Boston University Conference on Language Development
Online
Supplement
. Somerville, MA: Cascadilla Press.
Sera, M.D. (1992). To be or to be.
Journal of Memory and Language, 31
DOI: 10.1016/0749-596x%2892%2990021-o
Sera, M.D., Bales, D.W., & del Castillo Pintado, J. (1997). Ser helps speakers identif礠“real鐠prop
erties.
Child Development, 68,
DOI: 10.2307/1132035
Serratrice, L. & Sorace, A. (2003).Overt and null subjects in monolingual and bilingual lan
guage acquisition. In B. Beachley, A. Brown, & F. Conlin (Eds.),
Proceedings of the 27th
Annual Boston University Conference on Language Development
(pp. 739–750). Somerville,
MA: Cascadilla Press.
Serratrice, L., Sorace, A., & Paoli, S. (2004). Crosslinguistic inuence at the syntax-pragmatics
interface: Subjects and objects in Italian-English bilingual and monolingual acquisitio渮2
Bilingualism Language and Cognition
DOI: 10.1017/s1366728904001610
Silva-Corvalán, C. (2003). Linguistic consequences in reduced input in bilingual rst language
acquisition. In S. Montrul & F. Or擳ñez (Eds.),
Linguisti挠seory and Language Develop
ment in Hispanic Languages
(pp. 375–397). Somerville, MA: Cascadilla Press.
Silva-Corvalán, C., & Montanari, S. (2008). e acquisition of ser, estar (and be) by a
Spanish-English bilingual child: e early stages.
Bilingualism: Language and Cognition,
DOI: 10.1017/s136672890800357x
Sorrace, A., & Serratrice, L. (2009). Internal and external interfaces in bilingual language devel
opment: Beyond structural overlap.
International Journal of Bilingualism
DOI: 10.1177/1367006909339810
Vila, I., & Cortés, M. (1991). Aspectos relativos al desarrollo lexical y morfosinttico de los
bilingües familiares. In I. Idi慺ábal (Ed.),
Adquisición del lenguaje en niños bilingües y
monolingües – Hizkuntz Jabekuntza Haur Elebidun eta Elebakarretan
(pp. 109–127). San
Sebastián: Servicio Editorial de la U.P.V.
�aia Arnaus Gil & Natascha Müller
Wexler, K. (1994). Optional innitives, head movement and the economy of derivations. In
D⸚Lightfoot & N. Hornstein (Eds.),
Verb Movement
(pp. 305–350). Cambridge: CUP.
DOI: 10.1017/cbo9780511627705.016
DOI: 10.1016/s0024-3841%2898%2900029-1
Zagona, K. (2009). Ser and estar: Phrase structure and aspect.
Proceedings of 2008. Chronos
Cahiers Chronos
(pp. 1–25). Austin, TX: University of Texas.
Zuckerman, S. (2001).
se Acquisition of Optional Movement.
Unpublished Ph.D. dissertation.
Knowledge and processing of subject-related
discourse properties in L2 near-native speakers
of Spanish, L1 Farsi
T椝any Judy
Wake Forest University
Ὡs chapter investigates the discourse-constrained distribution of referential
subject pronouns (RSP) by native Farsi-speaking, adult second language (L2)
Spanish speakers. Results from oᡩne and online tasks measuring participants鈠
knowledge and processing of overt and null subjects in Contrastive Focus,
Topic Shi and Topic Maintenance contexts were mixed, showing divergence
and convergence⸠�e results are discussed in light of the Interface Hypothesis
Sorace 2011
), which predicts divergence on the processing of
external interface-conditioned properties like RSP distribution. While some
Introduction
Much generative research conducted in previous decades focused on grammatical
representation, as this was the assumed point of divergence in adult second lan
guage (L2) acquisition (e.g.
Hawkins & Chan 1997
Tsimpli &
trakopoulou
Vainikka & Young-Scholten 1994
Serratrice 2009
) has proven an inuential account attempting to reconcile some
�椝any Judy
convergence on certain domains of language, on the other. In some ways, the IH
has spurred a shi from examining non-native language representation to exam
ining the integration of linguistic and cognitive information (in early instantia
tions) and processing (currently) of non-native languages. is is so because the
IH assumes that convergence on syntax is possible (though not guaranteed), but
predicts divergence for interface-conditioned properties. Specically, proper
ties requiring integration of linguistic and cognitive information (i.e. external
interface-conditioned) are predicted points of divergence (
Sorace 2011
),
as they are deemed costly for processing. Results from early IH-related work,
Sorace &
�erratrice 2009
Sorace,
Serratrice, Filiaci, & Baldo 2009
), the IH currently claims that divergence results
from bilingualism itself and not (only) L1-L2 dierences (
Sorace 2011
): since
bilinguals must inhibit the non-relevant language (
Bialystok 2009
) while activat
ing the other, nite cognitive resources are divided and processing dierences
obtain.
nowledge and processing of subject-related discourse properties
Subject Distribution in Spanish and Farsi
Spanish and Farsi are null-subject languages, meaning that
pro
is licensed and
ident検ed per
Rizzi’s (1982
) requirements. Along with
Alexiadou and Anagnos
topoulou (1998
) and
Goodall (2001
), I assume that the Extended Project
In Spanish, present, preterite and future tense morphology fo爠�rst and second person sin
gular and plural subjects is distinct; for the imperfect and conditional tenses, some compound
tenses and subjunctive moods, �rst and third person singular are connated. In Farsi, all person
and number morphology is distinct.
Regarding Topic Shiᰠand Topic Maintenance contexts, see
Alonso-Ovalle, Fernández-
Solera, Frazier and Clison (2002
) and
Filiaci (2011
) for Spanish,
Filiaci (2011
) for Italian and
Bel and Gar揭a-Alcaraz (this volume) for Moroccan Arabic. In general, these studies found
that null subjects were osen interpreted as co-referential with subjects (and thus an instance
of Topic Maintenance), while more varied results obtained for overt subjects.
�椝any Judy
distinct for the second and third person singular subjects (
digas
versus
diga
Spanish and
beguyi
versus
beguyam
in Farsi), the absence of the overt subject ren
ders the sentence infelicitous.

�r
you-
9ᘔ
it-
ᘒ4
tell-1.
᜕ue
�r
�ou.
9ᘔ
it-
ᘒ4

lla.

tell-3.
᜕ue

�Either I will tell you or she will tell you.”

uiero
me
ú.

�.
uiero
me
pro

want-1.
᜕ue
that
me-
9ᘔ
it-
ᘒ4
tell-2.
᜕ue
eԉj
(you)

I want you to tell it to me [and not her].”
�an
be
to
eguyam
be
to
eguyad.

ou
tell-1.
᜕ue
or
to
ou
tell-3.
᜕ue

�Either I tell you or she tells you.”

an
man
beguyi.

�.
an
mixam
pro
be
man
beguyi.

അ�
-want.1.
that
you
to
ell-2.
᜕ue

I want you to tell me.”
Next, Examples (3) and (4) show Topic Shi contexts in which the topic (here, the
subjects) changes within the discourse. Specically, the topic changes from
yo/
man
“I鐠to
ellas/unha
“they,鐠respectively. Consequently, the overt subject
ellas
unha
“the禔 is preferred over
pro
to identify the matrix subject. Again, this is true
despite the recoverability of the subject via verbal morphology.

lmor揩
oy.
pro
not
eat-1.
᜕ua
today

�I did not eat lunch today.”

llas
iensan
engo
ambre
ahora.

pro
�iensan
engo
ambre
ahora.

(they)
hink-3.
᜕ue
that
pro
have-1.
᜕ue
hunger
�ow

�ey think I am hungry.”
�an
em ruz
nahar
nakhordam.

oday
unch
-eat-1.
᜕ua

�I did not eat lunch today.”
In
Rothman (2009
), the judgment of this sentence is listed as ungrammatical (*). Since
this sentence is entirely grammatical, the judgment has been changed to # to represent the
infelicity of the null subject in this context.‟e same is true of (4).
nowledge and processing of subject-related discourse properties

nha
ekr
konand
oshneh
basham.

pro
�ekr
konand
oshneh
basham.

(they)
hought
അ�
-do.3.
that
pro
hungry
be-1.
᜕ue

�ey think that I am hungry.”
Finally, the subjects of the contexts and follow-up sentences in Examples (5)
an搚(6) below both refer to the same women:
María
and
Hilda
(5) and
Shabnam
and
Leila
(6). us, the topic is maintained across the discourse rendering the
overt subject
ellas
unha
ₓthey” less preferable than
pro
�aría
ilda
no
lmorzaron
oy.

�ary
nd
Hilda
not
eat-3.
᜕ua
today

�Mary and Hilda did not eat lunch today.”

llas
en摲án
ucha
hambre.

pro
�en摲án
ucha
hambre.

hey)
have-3.
ਅa
much
�unger

�ey must be hungry.”
�habnam
eila
em ruz
nahar
nakhordand.

�habnam
and
Leila
today
unch
-eat-3.
᜕ua

�Shabnam and Leila did not eat lunch today.”

nha
ayad
goshneh
bashand.

pro
bayad
goshneh
bashand.

hey)
must
ungry
᜕ue

�ey must be hungry.”
Table 1 below summarizes the discourse-constrained distribution of RSPs for
Spanish and Farsi.
Table 1.
Subject distribution preferences in Spanish and Farsi
Spanish
Farsi
Contrastive Focus
Overt
Overt
Topic S桩1
Overt
Overt
Topic Maintenance
Null
Null
�at overt and null subject distribution is similar in Spanish and Farsi, two
null-subject languages, is not surprising.
But, see
Kissock (1995
) for Telagu and Duarte (1993, 1995) for Brazilian Portuguese for
evidence showing rather distinct distributions of overt and null subject pronouns in these
null-subject languages.
�椝any Judy
demonstrated that even typologically similar null-subject languages such as
Catalan, Italian and Spanish may dier subtly with respect to overt subject resolu
tion and subject distribution (
Filiaci, Sorace, & Carreiras 2013
Prada Pez 2009
),
judgments of overt and null subjects were obtained from 39 native Farsi speakers
for the three contexts above⸠�e data from
Judy and Feizmohammadpour (2012
demonstrate a similar distributional pattern in that the Farsi speakers (1) rated far
more overt than null subjects as good in Contrastive Focus contexts (5.82 vs. 0.92
o昚6), (2) rated more overt than null subjects in Topic Shi contexts as good (5.69 vs.
4.31 of 6) and (3) rated far more null subjects in Topic Maintenance contexts than
Previous acquisition studies
Subject distribution is perhaps the most widely studied linguistic phenomena
But see
Liceras, Fernández Fuertes and Alba de la Fuerte (2012
) for results suggesting that
the overuse seen in
Paradis and Navarro (2003
) results from the input.
nowledge and processing of subject-related discourse properties
of overt subjects (including subject pronouns) in an English-Spanish bilingual
(1;8–2;7 years) as compared to two child monolinguals (~17% and 20%). Simi
larly, in a study testing subject distribution in Topic Shi contexts,
See
Bini (1993
) for similar results in Spanish-Italian adult bilinguals.
Diierently from the majority of studies examining subject use,
Tsimpli and Sorace (2006
found overuse of null subjects across three pro�ciency groups of L1 Russian naturalistic
learners of L2 Greek.
�椝any Judy
Rothman (2008
) and
Rothman (2009
), which investigated L1 English speakers of
L2 Spanish. e intermediate prociency group of
Rothman (2008
) demonstrated
syntactic knowledge of subjects, but did not converge on subject distribution:
they rated overt subjects favorably where null subjects were preferred in the judg
ment task while they underused overt subjects in the production task. Advanced
prociency learners, however, demonstrated convergence on the syntactic and
discourse-distribution of subjects. Similar ᱮdings obtained in
Rothman (2009
) in
that only the advanced group showed knowledge of related syntax-semantics and
syntax-discourse properties. While 28 of 38 intermediate prociency participants
of the study showed knowledge of the syntax-semantics property tested therein,
they accepted and used more overt and null subjects than the control group, indi
cating that L2 speakers converge on external interface-conditioned properties
later, but that convergence is possible. Finally,
Gﱲel (2006
) showed that L1 English
learners of L2 Turkish (presumed near-native speakers since they lived in
urkey
for at least 10 years) converged on the pragmatic use of overt and null subjects.
Overall, the experimental studies examined above provide some evidence sup
porting the IH in that external interface-conditioned properties were subject to
divergence in child language acquisition and in some adult L2 studies. However, the
three studies outlined in the previous paragraph reported convergence on external
interface-conditioned properties, providing counterevidence. Importantly, both
bodies of research indicate that syntax is not the source of delay or divergence on
external interface-conditioned properties. is nding is in line with the current
state of the IH which claims that syntax is not inevitably vulnerable. �e studies
Ὡs study examines the explicit judgments and processing of the discourse-
constrained distribution of RSPs in Spanish. As described in Section 2, while both
nowledge and processing of subject-related discourse properties
overt and null subjects are permissible in Spanish, their distribution is conditioned
�o near-native L2 speakers converge on the discourse-constrained distribu
tion of RSPs in Spanish in the oᡩne task?
�o near-native L2 speakers exhibit processing dierences as compared to the
native speakers in the online task?
Recall from the introduction that the IH claims that processing dierences will
Participants
Twenty-four native Spanish speakers and 8 near-native
L2 Spanish speakers par
ticipated in this study⸠�e native speakers were born in Argentina to native River
Plate Spanish-speaking parents and were exposed to this dialect since birth, thus
controlling for dialectal variation. Most native speakers reported knowledge of an
Classi�cation as a near-native speaker and participation in the study was conditioned
upon scoring a minimum of 45/50 on an abbreviated version of the
Diploma del español como
lengua extranjera
(DELE 2002) that is widely used in the US (see work by Montrul, Slabakova
and White, for example) and is comprised of a vocabulary and a cloze section.
All received formal instruction in school and some continued to study the Spanish lan
guage at university.
�椝any Judy
quantitatively insignicant. While all bilingual participants are fully integrated
Context-matching felicitousness task (CMFT)
�e purpose of the CMFT was to examine the judgments of overt and null subjects
in three discourse-constrained contexts. is task contained 36 tokens (n㴚12
Contrastive Focus, n = 12 Topic Shi, n = 12 Topic Maintenance). Participants

�uando salimos a cenar, mi novia preere comer platos livianos, pero yo
preero comer algo sustancioso.

�When we go out to eat, my girlfriend prefers to eat light dishes, but I prefer

sí que ella come ensaladas y yo como milanesas en los restaurantes.

�.
sí que
pro
come ensaladas y
pro
como milanesas en los restaurantes.

So, she eats salads and I eat breaded meats in restaurants.”
Similarly, all Topic Shi contexts introduced two subjects such that a shi in ref
erence could be made⸠�e third person singular feminine subject was always
pro
appears for expository purposes only and was not present in the actual tokens.‟is
is true of all other tokens.
nowledge and processing of subject-related discourse properties
introduced in the rst sentence of the context. en, the rst person singular sub
ject pronoun was introduced in the second and any subsequent sentences. Mini
mally, three verbs conjugated for rst person singular separated the rst sentence

�i hija quiere ser autora y no tiene otros intereses. Yo creo que es mejor
tener varios intereses y sugiero otras actividades, pero no importa lo que
diga yo.

�My daughter wants to be an author and she has no other interests. I think
that it is best to have various interests and I suggest other activities, but it
doesn’t matter what I say.”

almente ella escribe cuentos y pasa todo el día en su cuarto.

�.
almente escribe cuentos y pasa todo el día en su cuarto.

In the end, she writes stories and she spends the whole day in her
room⺔
䑩ierently from the two previous contexts, Topic Maintenance contexts introduced
only one third person singular DP feminine subject (i.e.
mi hija
“my
daughter鐩
which was initially introduced in the rst sentence of the context. From there, the
context contained a minimum of three verbs conjugated for third person singular
with null subjects. Half the Topic Shi tokens contained overt pronouns (9a) and
half contained null pronouns (9b). Recall from Section 2 that null pronouns are
preferred in Topic Maintenance contexts.

�i c痱ada es muy sociable. Tiene muchos amigos y por eso va a muchas
cenas a la canasta donde tiene que contribuir con algo.

�My daughter-in-law is very social. She has a lot of friends and for that
reasons, she goes to a lot of potluck dinners where she has to share

sí que ella lleva postres y comparte todo con sus amigos.

�.
sí que lleva postres y comparte todo con sus amigos.

So, she takes desserts and shares everything with her friends.”
While preferred, more variation has been shown with Topic Shiᰠand Topic Maintenance
than Contrastive Focus (e.g.
Bentivolgio 1987
Enríquez 1984
Filiaci et al., 2013
Lapidus
Shi渚& Cairns 2011
Lubbers Quesada & Blackwell 2009
Otheguy, Zentella, & Livert 2007
Prada Pérez 2009
Silva-Corvalán 1994
).
�椝any Judy
In addition to the care taken with the creation of each context type described
above, all subjects, matrix clause verbs and direct objects (DO) were counter
balanced across token type and subject type to allow for maximal structural
similarity. is is especially important for the online version of this task in that
statistical comparisons of the Reaction Times (RT) to the regions of interest
(matrix clause verb and DOs) are made. Twelve matrix clause transitive verbs
come
“eat,鐠
hace
“makeⲔ
prepara
“prepare,”
toma
“drink,鐠
ofrece
“oier,鐠
lleva
“takeⲔ
vende
“sellⲔ
escribe
“write,”
lee
“read,鐠
compra
“buyⲔ
pinta
“paint,鐠
dibuja
“dra瞔⤠and 12 matrix clause DOs (
ensaladas
“salads,鐠
alfajores
“cookies with
dulce de leche,鐠
empanadas
“savory pastries,鐠
agua
“water,鐠
café
“coiee,鐠
postres
“desserts,鐠
revistas
ₓmagazines,”
cuentos
“stories,鐠
poemas
ₓpoems,”
esculturas
“sculptures,鐠
paisajes
鍬andscapes,鐠
�guras
“gures鐩 were distributed evenly
across the token and subject types. Finally, the oᡩne task was taken a1er the
online task to avoid priming.
Self-paced reading task (SPRT)
�e SPRT (
Aaronson & Scarborough 1976
Mitchell & Green 1978
) tested par
ticipants鈠processing of RSP distribution in the same discourse-constrained con
texts described in Section 4.2.
‟e purpose of this task, which also contained
㌶2tokens (n=12 Contrastive Focus, n = 12 Topic Shi, n = 12 Topic Maintenance)
and employed E-prime so1ware (
Schneider, Eschman & Zuccolotto 2002
), was to
Just, Woolley &
arpenter 1982
was employed to mimic normal reading as closely as possible and to
easure
�e 36 contexts employed in the o1ine task were employed in the online task.‟ey were
counterbalanced such that participants saw each context once with a target sentence with an
overt subject and once with a null subject.
�is presentational format begins with a blank screen containing several underscored
sentence items whose content becomes visible as the participant presses a button to advance
along word-by-word or region-by-region (i.e. in a non-cumulative and linear fashion as
opposed to the
centered
technique). Example (i) provides a short English example of this tech
nique. Each line represents the visua氠�eld the participant would see as they progress through
the sentence, which starts with all regions hidden and ends with th攠�nal word.



ittle ___ _ ____

oy _ ____


ired.
nowledge and processing of subject-related discourse properties
entence,
participants answered a short meaning-based yes/no comprehension question
designed to focus their attention on the content of the experimental stimuli and
not its felicitousness.

�uando salimos a cenar, mi novia preere comer platos livianos, pero yo
preero comer algo sustancioso.

�When we go out to eat, my girlfriend prefers to eat light dishes, but I prefer

sí que/ella/come/ensaladas/y/yo/como/milanesas/en los restaurantes.

�.
sí que/come/ensaladas/y/como/milanesas/en los restaurantes.

So, she eats salads and I eat breaded meats in restaurants.”

omprehension Question) ¿Mi novia come ensaladas?

Does my girlfriend eat salads?”
Example (11) shows a Topic Shi token. e regions of interest for overt subject
(11a) and null subject tokens (11b) alike are the verb
escribe
ₓwrites鐠and the DO
cuentos
“stories.”

�i hija quiere ser autora y no tiene otros intereses. Yo creo que es mejor
tener varios intereses y sugiero otras actividades, pero no importa lo que
diga yo.

�My daughter wants to be an author and she has no other interests. I think
that it is best to have various interests and I suggest other activities, but it
doesn’t matter what I say.”
�椝any Judy

almente/ella/escribe/cuentos/y/pasa/todo/el día/en su cuarto.

�.
almente/escribe/cuentos/y/pasa/todo el día/en su cuarto.

In the end, she writes stories and she spends the whole day in her
room⺔

omprehension Question) ¿Mi hija pasa todo el día en la biblioteca?

Does my daughter spend the whole day in the library?”
Lastly, the regions of interest examined in Topic Maintenance overt subject (12a)
and null subject tokens (12b) are the verb
lleva
ₓtakes鐠and the DO
postres
“des
serts鐠of the matrix clause.

�i c痱ada es muy sociable. Tiene muchos amigos y por eso va a muchas
cenas a la canasta donde tiene que contribuir con algo.

�My daughter-in-law is very social. She has a lot of friends and for that
reasons, she goes to a lot of potluck dinners where she has to share

sí que/ella/lleva/postres/y/comparte/todo/con sus amigos.

�.
sí que/lleva/postres/y/comparte/todo/con sus amigos.

So, she takes desserts and shares everything with her friends.”

omprehension Question) ¿Mi c痱ada tiene muchos amigos?

Does my daughter-in-law have a lot of friends?”
As described at the close of Section 4.2, all subjects, matrix clause verbs and DOs
were counterbalanced across token and subject type⸠�e online task was com
Results
Context-matching felicitousness task
�e contingency table below shows that the participants鈠ratings from the oᡩne
task largely corresponded to the ends of the scale (either 1 (=100% good) or 4
(=100% bad)). As such, ratings of 1 and 2 have been combined a猠“good鐠and
ratings of 3 and 4 have been combined a猠“bad鐠for all token types. e statistical
analyses described below are based on this binary distinction.
�e group average “good鐠ratings for the Contrastive Focus with overt (CFO)
and Contrastive Focus with null (CFN) tokens are shown in Figure 1 below. Both
the native and L2 groups rated a high number of CFO tokens a猠“good” (5.92 and
5.75 out of 6, respectively) while also rating considerably fewer CFN tokens as
“good” (0.71 and 0.25, respectively). Importantly, all participants rated more CFO
nowledge and processing of subject-related discourse properties
Figure 1.
CMFT group results: Contrastive Focus. CFO = Contrastive Focus token with overt
subject; CFN = Contrastive Focus token with null subject
Figure 2 below shows the group averages for Topic Shi tokens. For both the
native and L2 groups, the average number of TSO tokens rated a猠“good” was quite
high (5.54 and 5.75 out of 6, respectively). A higher number of TSN tokens, as
compared to the CFN tokens, was rated as “good鐠across both groups (2.75 and
4.38, respectively). All but one native and two L2 speakers rated more TSO than
TSN tokens a猠“good”.
Lastly, Figure 3 below shows the group averages for Topic Maintenance tokens.
Both the native and L2 group’s average number o映“good” ratings was rather high
for TMO (4.58 and 5.50 out of 6, respectively) and TMN tokens (5.54 and 6.00,
Table 2.
Distribution of ratings
Rating
1 = 100% good
2 = more or less good
3 = more or less bad
4 = 100% bad
�椝any Judy
Figure 2.
CMFT group results: Topic Shi. TSO = Topic Shi token with overt subject;
TS业㴚Topic Shi token with null subject
Figure 3.
CMFT group results: Topic Maintenance. TMO = Topic Maintenance token with
overt subject; TMN = Topic Maintenance token with null subject
nowledge and processing of subject-related discourse properties
respectively). Only 11 of 24 native and 3 of 8 L2 speakers rated more TMN than
TMO tokens a猠“good”. Of the 13 native speakers that did not make the expected
distinction, 11 rated an equal number of TMO and TMN tokens a猠“good”⸠�攠sve
L2 speakers that did not make the expected distinction all rated an equal number
of TMO and TMN tokens a猠“good.”
A Repeated Measures ANOVA with Bonferroni post-hoc analyses (
= .05)
was conducted for the CMFT with Subject Type (overt or null), Context Type
(Contrastive Focus, Topic Shi and Topic Maintenance) and Group (native or
L2 speaker) as factors. Signicant main eects for Subject Type (
(1, 1140)
2㴚.020), Context Type (
.001) and Group (
(1, 25)
Bonferroni post-hoc pairwise t-tests showed that both the native and the L2
�椝any Judy
Self-paced reading task (SPRT)
RT data from the SPRT was treated as follows. First, data points corresponding to
incorrect answers to comprehension questions were excluded. Second, RT data
was trimmed such that data points slower than 2 seconds
were excluded. Finally,
the mean RT and standard deviation for each region of interest were calculated
across the SPRT. Data points higher than 2 standard deviations from the mean
were replaced with the cuto value (mean RT + (standard deviation X 2)).
For
Regions 2 and 3, respectively, Tables 4 and 5 below provide the average RTs for
each token type per participant, where S and F in the rst column represent native
Spanish and native Farsi speakers, respectively. Per participant, the average RT to
�is data, which constituted a mere 0.313%, was excluded on the grounds that such high
RTs (i.e. slower than 2 seconds) likely represent external distractions.
See
Hopp (2007
, p. 224) for similar data treatment. 3.4% and 3.7% of data was replaced
for Regions 2 and 3, respectively.
Participants F2, F3 and F7 did not distinguish between #TMO and TMN tokens in either
region, but did so in the other context types.
nowledge and processing of subject-related discourse properties
Table 4.
SPRT individual RTs to Region 2 (verb)
Part.
#TSN
TSO
#TMO
TMN
Ave.
All values in Tables 4 and 5 are rounded to whole numbers (e.g. original values for native
speaker #5 were 284.25ms (#CFN)-421.60ms (CFO) = 137.35ms.)
Continued
�椝any Judy
Part.
#TSN
TSO
#TMO
TMN
Ave.
However, this comparison only shows that the native group average RT is faster
Table 5.
SPRT individual RTs to Region 3 (DO)
Part.
#TSN
TSO
#TMO
TMN
Continued
Table 4.
SPRT individual RTs to Region 2 (verb)
ontinued)
nowledge and processing of subject-related discourse properties
Part.
#TSN
TSO
#TMO
TMN
Ave.
Ave.
Ὡs is referred to as the average dierential RT, which is the dierence
Contrastive focus
Figure 4 below shows the average dierential RT for Regions 2 and 3 for the CFN
versus CFO tokens repeated below in (13). e regions of interest are bolded.

�uando salimos a cenar, mi novia preere comer platos livianos, pero yo
preero comer algo sustancioso.

�When we go out to eat, my girlfriend prefers to eat light dishes, but I prefer

sí que/ella/
come/ensaladas
/y/yo/como/milanesas/en los restaurantes.

�.
sí que/
come/ensaladas
/y/como/milanesas/en los restaurantes.

So, she eats salads and I eat breaded meats in restaurants.”
No comparison can be made for th攠�rst region of interest, the subject
ella
“she,” since it is
necessarily present in the overt subject tokens but necessarily absent in the null subject tokens.
Table 5.
(Continued)
�椝any Judy
Figure 4.
SPRT comparative group results (verb and DO) with S.E. bars. CFN-CFO = group
average RT to CFN tokens minus group average RT to CFO tokens
�e average dierential RT was calculated by subtracting the group average
RT to the felicitous CFO from the group average RT to the infelicitous CFN token
type. For Region 2 (verb), it is clear that this equation resulted in a negative aver
age dierential RT for both the native and L2 groups. e native group dierence
in average RT for CFN-CFO was greater than that of the L2 group (
11.94ms
versus
4.37ms, respectively). e negative result indicates that, on average, the
Topic shia
Next, the average dierential RTs to Topic Shi tokens is examined. An example is
repeated below in (14) with the regions of interest bolded.
nowledge and processing of subject-related discourse properties

�i hija quiere ser autora y no tiene otros intereses. Yo creo que es mejor
tener varios intereses y sugiero otras actividades, pero no importa lo que
diga yo.

�My daughter wants to be an author and she has no other interests. I think
that it is best to have various interests and I suggest other activities, but it
doesn’t matter what I say.”

Finalmente/ella/
escribe/cuentos
/y/ella/pasa/todo/el día/en su cuarto.

�.
#Finalmente/
escribe/cuentos
/y/pasa/todo el día/en su cuarto.

In the end, she writes stories and she spends the whole day in her
room⺔
Figure 5.
SPRT comparative group results (verb and DO) with S.E. bars. TSN-TSO = group
average RT to TSN tokens minus group average RT to TSO tokens
Again, the average dierential RT was calculated by subtracting the group
average RT to the felicitous TSO token from the group average RT to the infelici
tous TSN token type for Regions 2 and 3. Figure 5 shows that for Region 2 (verb), a
positive average dierential RT for both the native and L2 group obtained (21.39ms
�椝any Judy
�e independent samples t-test revealed no statistically signicant dierence
Topic maintenance
Finally, this section presents a comparison of the RTs to felicitous and infelicitous
Topic Maintenance tokens, an example of which is repeated below in (15) with the
second and third region of interests bolded.

�i c痱ada es muy sociable. Tiene muchos amigos y por eso va a muchas
cenas a la canasta donde tiene que contribuir con algo.

�My daughter-in-law is very social. She has a lot of friends and for that
reasons, she goes to a lot of potluck dinners where she has to share

sí que/ella/
lleva/postres
/y/comparte/todo/con sus amigos.

�.
sí que/
lleva/postres
/y/comparte/todo/con sus amigos.

So, she takes desserts and shares everything with her friends.”
Figure 6.
SPRT comparative group results (verb and DO). TMO-TMN = group average RT to
TMO tokens minus group average RT to TMN tokens
Across both comparisons, slower RTs are expected for the overt subjects
since these are less preferred than null subjects in Topic Maintenance contexts.
Figur攚6 shows that the average dierential RT for Region 2 (verb) is negative in
both groups (
6.20ms and
15.57ms, respectively). e negative result indicates
that, on average, the infelicitousness of TMO tokens was not evidenced in this
region. For Region 3 (DO), both groups demonstrated a positive group average
nowledge and processing of subject-related discourse properties
摩ierential RT (0.46ms for the native group and 37.84ms for the L2 group). is
Discussion & conclusion
�o near-native L2 speakers converge on the discourse-constrained distribu
tion of RSPs in Spanish in the oᡩne task?
�o near-native L2 speakers exhibit processing dierences with respect to the
discourse-constrained distribution of RSPs in the online task?
Regarding the rst research questions, two statistical analyses were conducted
�椝any Judy
does not have an L1-based explanation as the L1 Farsi speakers accepted only 2
of 6 TMO tokens (compared to 5.77 of 6 TMN tokens in
nowledge and processing of subject-related discourse properties
eliciting the expected responses, but that the dierences regarding which region
the distinction was evidenced gives the impression that few distinctions were
made in the online version.
�椝any Judy
DOI: 10.1023/a:1006090432389
Alonso-Ovalle, L., & D’Introno, F. (2000). Full and null pronouns in Spanish: the zero pronoun
hypothesis. In H. Campos, E. Herburger, A. Morales-Front, & T.J. Walsh (Eds.),
Hispanic
Linguistics at the Turn of the Millennium
(pp. 400–414). Somerville, MA: Cascadilla Press.
Alonso-Ovalle, L., Fernández-Solera, S., Frazier, L., & C汩1on, C. (2002). Null vs. overt pronouns
and the topic-focus articulation in Spanish.
Rivista Di Linguistica, 14
Bel, A., & Gar揭a-Alcaraz, E. (2015). Subject pronouns in the L2 Spanish of Moroccan Arabic
speakers: Evidence from bilingual and second language learners. In T. Judy & S.Perp槱án
(Eds.),
se Acquisition of Spanish in Adult and Child Bilingualism. Data from Understudied
Language Pairings
(pp. 201–232). Amsterdam: John Benjamins.
DOI: 10.1075/eurosla.4.06bel
Caracas: Universidad Central de Venezuela.
DOI: 10.1017/s0047404500014214
Bialystok, E. (2009). Bilingualism: e good, the bad, and the in摩ierent.
Bilingualism: Lan
guage and Cognition, 12
DOI: 10.1017/s1366728908003477
Bini, M. (1993). La adquisición del italiano: Más allde las propiedades sintticas del pa狡
DOI: 10.2307/2273965
Chomsky, N. (1995).
se Minimalist Program
. Cambridge, MA: e MIT Press.
DOI: 10.1017/s0272263197241070
Chomsky, N. (2000).
On Nature and Language
. Cambridge: CUP.
Chomsky, N. (2001). Derivation by phase. In M. Kenstowicz (Ed.),
Ken Hale: A Life in Language
(pp. 1–52). Cambridge, MA: e MIT Press.
DOI: 10.1017/s0022226704322747
DELE (2002). Spanish Embassy, Washington, DC.
�ervantes [e subject personal pronoun in the Spanish language spoken in Madrid].
Fernández-Soriano, O. (1989). Strong pronouns in null subject languages and the Avoid Pro
noun Principle.
MIT Working Papers in Linguistics,
DOI: 10.1080/01690965.2013.801502
nowledge and processing of subject-related discourse properties
Filiaci, F., Sorace, A., & Carreiras, M. (2013). Anaphoric biases of Null and Overt Subjects in Italian
and Spanish: A cross-linguistic comparison.
Language and Cognitive Processes
(7), 825–843.
DOI: 10.1080/01690965.2013.801502
Goodall, G. (2001). e EPP in Spanish. In W. Davies & S. Dubinsky (Eds.),
Objects and
Other Subjects: Grammatical Functions, Functional Categories and Con�gurationality
(pp.193–223). Dordrecht: Kluwer.
Goodall, G. (2002). On preverbal subjects in Spanish. In T. Satterᱥld, C. Tortora, & D. Cresti
(Eds.),
Current Issues in Romance Languages: Selected Papers from the
29th
Linguistics Sym
posium on Romance Languages
LSRL
) (pp. 95–109). Amsterdam: John Benjamins.
DOI: 10.1075/cilt.220.08goo
Grinstead, J. (2004). Subjects and interface delay in child Spanish and Catalan.
Language,
DOI: 10.1353/lan.2004.0024
Gﱲel, A. (2006). L2 acquisition of pragmatic and syntactic constraints in the use of overt and
null subject pronouns. In R. Slabakova, S. Montrul, & P. Prévost (Eds.),
Inquiries in Lin
guistic Development: In Honor of Lydia White
(pp. 259–282). Amsterdam: John Benjamins.
DOI: 10.1075/z.133.15gur
Hawkins, R., & Chan, Y–H. (1997). e partial availability of universal grammar in second lan
guage acquisition: 攠“failed functional features hypothesis”.
Second Language Research,
DOI: 10.1191/026765897671476153
Hopp, H. (2007).
Ultimate Attainment at the Interfaces in Second Language Acquisition: Gram
mar and Processing
Groningen: Grodil Press.
Just, M.A., Woolley, J.D., & Carpenter, P.A. (1982). Paradigms and processes in reading compre
hension.
Journal of Experimental Psychology: General, 111
Judy, T., & Feizmohammadpour, A. (2012). When parsing failures fail: Referential subjects
and anaphora resolution in Farsi learners of L2 English. Poster presented at: Gen
erative Approaches to Language Acquisition North America 5. October 11–12, 2012;
Lawrence,KS.
Karimi, S. (2005).
A Minimalist Approach to Scrambling: Evidence from Persian
. Berlin: de
Gruyter.
DOI: 10.1515/9783110199796
Kissock, M. (1995).
R攝exive-middle Constructions and Verb Raising in Telugu
. Unpublished
Ph.D. dissertation. Concordia University.
Lapidus Shin, N., & Cairns, H.S. (2011). �e development of NP selection in school-age chil
dren: Reference and Spanish subject pronouns.
Language Acquisition
DOI: 10.1080/10489223.2012.633846
Liceras, J.M., Fernández Fuertes, R., & Alba de la Fuente, A. (2012). Overt subjects and copula
omission in the Spanish and the English grammar of English-Spanish bilinguals: On the
locus and directionality of interlinguistic inuence.
First Language, 32
DOI: 10.1177/0142723711403980
Lozano, C. (2006). Focus and split-intransitivity㨠�e acquisition of word order alternations in
non-native Spanish.
Second Language Research, 22
DOI: 10.1191/0267658306sr264oa
Lubbers Quesada, M., & Blackwell, S. (2009). �e L2 acquisition of null and overt Spanish sub
ject pronouns: A pragmatic approach. In J. Collentine, M. Gar揭a, B. Laiord, & F.M. Mar
(Eds.),
Selected Proceedings of the 11th
Hispanic Linguistics Symposium
(pp. 117–130)
Somerville, MA: Cascadilla Proceedings Project.
Luján, M. (1987). Los pronombres implícitos y explícitos del español.
Revista Argentina de
Lingﳭstica,
�椝any Judy
Luján, M. (1999). Expresión y omi獩ón del pronombre personal. In I. Bosque & V. Demonte,
(Eds.),
Gramática descriptiva de la lengua española
(pp. 1275–1315). Madrid: Espasa Calpe.
Margaza, P., & Bel, A. (2006). Null subjects at the syntax–pragmatics interface: Evidence from
Spanish interlanguage of Greek speakers. In M.G. O’Brien, C. Shea, & J. Archibald (Eds.),
Proceedings of GASLA 2006.
(pp. 88–97). Somerville, MA: Cascadilla Press.
Mitchell, D.C., & Green, D.W. (1978). e eects of context and content on immediate process
ing in reading.
Quarterly Journal of Experimental Psychology, 30
DOI: 10.1080/14640747808400689
Montrul, S. (2004). Subject and object expression in Spanish heritage speakers: A case of mor
phosyntactic convergence.
Bilingualism: Language and Cognition,
DOI: 10.1017/s1366728904001464
Montrul, S., & Louro-Rodríguez, C. (2006). Beyond the syntax of the null subject param
DOI: 10.1353/lan.2008.0019
Paradis, J., & Navarro, S. (2003). Subject realization and crosslinguistic interference in the bilin
gual acquisition of Spanish and English: what is the role of input?
Journal of Child Lan
guage, 30
DOI: 10.1017/s0305000903005609
Picallo, M.C. (1994). Catalan possessive pronouns: e Avoid Pronoun Principle revisited.
Natural Language and Linguistic“eory,
DOI: 10.1007/bf00993146
Longman.
DOI: 10.9793/elsj1984.10.1
Rothman, J. (2008). How pragmatically odd! Interface delays and pronominal subject distribu
tion in L2 Spanish.
Studies in Hispanic and Lusophone Linguistics
Rothman, J. (2009). Pragmatic decits with syntactic consequences?: L2 pronominal subjects
and the syntax-pragmatics interface.
Journal of Pragmatics,
DOI: 10.1016/j.pragma.2008.07.007
Schneider, W., Eschmann, A., & Zuccolotto, A. (2002). E-Prime [Computer so1ware].
ittsburgh,
PA: Psychology So1ware Tools Inc.
Silva-Corvalán, C. (1994).
Language Contact and Change: Spanish in Los Angeles.
Oxford: OUP.
DOI: 10.1017/s0047404500020753
Sorace, A. (2006). Gradedness and optionality in mature and developing grammars. In G.
anselow,
C. Fy, R. Vogel, & M. Schlesewsky (Eds.), Gradience in Grammar (pp. 106–123). Oxford:
OUP.
DOI: 10.1093/acprof:oso/9780199274796.003.0006
Sorace, A. (2011). Pinning down the concept o映“interface” in bilingualism.
Linguistic Approaches
to Bilingualism, 1
DOI: 10.1075/lab.1.1.01sor
Sorace,
(2012). Pinning down the concept o映“interface” in bilingualism: A reply to peer com
mentaries.
Linguistic Approaches to Bilingualism, 2,
DOI: 10.1075/lab.2.2.04sor
nowledge and processing of subject-related discourse properties
Sorace, A., & Filiaci, F. (2006). Anaphora resolution in near-native speakers of Italian.
Second
Language Research,
DOI: 10.1191/0267658306sr271oa
Sorace, A., & Serratrice, L. (2009). Internal and external interfaces in bilingual language devel
opment: Beyond structural overlap.
International Journal of Bilingualism, 13
DOI: 10.1177/1367006909339810
Sorace, A., Serratrice, L. Filiaci, F., & Baldo, M. (2009). Discourse conditions on subject pro
noun realization: Testing the linguistic intuitions of older bilingual children.
Lingua
DOI: 10.1016/j.lingua.2008.09.008
DOI: 10.1177/0267658307076546
Tsimpli, I.-M., & Sorace, A. (2006). Dierentiating interfaces: L2 performance in syntax-
semantics and syntax-discourse phenomena. In D. Bamman, T. Magnitskaia, & C. Zaller
(Eds.),
Proceedings of the 30th
Annual Boston University Conference on Language Develop
ment
(pp.653–664). Somerville, MA: Cascadilla Press.
Tsimpli, I.-M., Sorace, A., Heycock, C., & Filiaci, F. (2004). First language attrition and syntactic
subjects: A study of Greek and Italian near-native speakers of English.
International Journal
of Bilingualism, 8
DOI: 10.1177/13670069040080030601
Vainikka, A., & Young-Scholten, M. (1994). Direct access to X
-theory: evidence from Korean
and Turkish adults learning German. In T. Hoekstra & B. Schwartz (Eds.)
Language Acqui
sition Studies in Generative Grammar: Papers in Honor of Kenneth Wexler from the 1991.
GLOW Workshops
(pp. 265–316).
Amsterdam: John Benjamins.
DOI: 10.1075/lald.8.13vai
Vainikka, A., & Young-Scholten, M. (1996). Gradual development of L2 phrase structure.
Second
Language Research,
DOI: 10.1177/026765839601200102
White, L. (2003).
Second language acquisition and Universal Grammar
. Cambridge: CUP.
DOI: 10.1017/cbo9780511815065.002
White, L. (2009). Grammatical theory: interfaces and L2 knowledge. In E. Ritchie & T.
hatia
(Eds.),
se New Handbook of Second Language Acquisition
(pp. 49–68). Bingham, UK:
Emerald.
Subject pronouns in the L2 Spanish
of Moroccan Arabic speakers
Evidence from bilingual and second language learners
Aurora Bel & Estela Gar揭a-Alcaraz
Universitat Pompeu Fabra
�e goal of this study is to shed light on the conditions that regulate the
Introduction
nominal); conversely, nominal categories are used to refer to less salient or acces
sible referents.
Psycholinguistic research has also examined the factors that might inu
ence the choice of an antecedent by the dierent types of pronominal
naphoric
expressions and have found that speakers tend to use pronouns to refer to a
prominent or salient discourse entity (
Garvey, Caramazza & Yates 1975
). In
�urora Bel & Estela Gar揭a-Alcaraz
ambiguous referential contexts, it has also been observed that null subject pro
nouns refer back to topic antecedents and overt pronouns tend to express topic
change (
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
developmental studies, the mastery of the pragmatic conditions that regulate the
Anaphora resolution in null subject languages: Spanish and
Moroccan Arabic
Spanish, as a null subject language with rich agreement marking on verbs, has
two types of pronouns: an overt pronoun and a phonologically null pronoun
pro
⤮‟eir use, however, is not fully optional: they carry information that con
Luis
telefoneaba a su madre cuando pro
llegaba tarde.
[–topic shi]

Luis telephoned his mother when
pro
was late.’

Ana vio a su vecina
mientras ella
se bañaba en el mar.
[+topic shi]

Ana saw her neighbour while she was bathing in the sea.’
It is worth mentioning that the use of a null pronoun instead of the overt one (
ella
�urora Bel & Estela Gar揭a-Alcaraz
third person pronoun may refer directly to an unambiguous antecedent generat
ing redundancy as in (2) below:
Julia
dice que ella
confesla verdad.

�Julia says that she confessed the truth.’
In the last decade, some studies (
Montrul 2004
Sorace & Filiaci 2006
Sorace
整2al. 2009
) have revealed that speakers have antecedent preferences for pronouns
when the context is potentially ambiguous. e most inuential work has been
Carminati’s (2002
) study in which she formulated a processing account for the
�oberto ha insultato Ugo quando lui/pro era ubriaco.

�Roberto insulted Ugo when he/pro was drunk.’

Roberto era ubriaco

‘Roberto was drunk.’

Ugo era ubriaco

‘Ugo was drunk.’
In (3) there are two potential antecedents for the overt pronouns (
) and for the
null pronoun (
pro
⤮‟e results showed that Italian speakers are signicantly more
likely to choose the null pronoun when referring to the subject (3a) and the overt
nominal preferences in two sentence discourses in Spanish. Following
Carminati’s
) PAH, they predicted that, while a null pronoun would prefer a subject
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
antecedent, the overt pronoun would not. Unlike the experimental sentences of
Carminati’s (2002
) work, these authors used inter-sentential contexts in experi
mental sentences. An example from
Sara abrazó a Teresa. pro está emocionada.

‘Sara embraced Teresa. pro is excited.’

Sara abrazó a Teresa. Ella está emocionada.

‘Sara embraced Teresa. She is excited.’
In contexts as the one in (4a), in 73.2% of the cases the null pronoun was linked to
the subject,
Sara
, while in examples such as (4b) the overt pronoun was assigned
to either
Sara
or
Teresa
(50.2% of the total cases were linked to the subject). Even
though
Spanish. However, unlike the preceding research, the second sentence was manip
ulated so as to control the semantics of the predicate and thus favor either topic
continuity (as in 5a) or topic change (as in 5b), disambiguating towards the pre
ceding subject or object, respectively:
Cuando Ana visitó a María en el hospital, ella/pro le llevun ramo
de rosas.

‘When Ana visited Mary in the hospital, she/pro brought her a bunch
of roses.’

Cuando Ana visitó a María en el hospital, ella/pro ya estaba fuera
de peligro.

‘When Ana visited Mary in the hospital, she/pro was already out of
danger.’
Filiaci (2011
) measured reading times on these sentences and found that, in
the
�ull subject condition, participants were faster in topic continuity contexts
than in topic change contexts, which suggests that null pronouns are associ
ated with the 宭topic shi] feature, thus supporting the PAH. On the contrary,
in the overt pronoun condition reading times were similar across the two con
texts, a result which is in line with what
�urora Bel & Estela Gar揭a-Alcaraz
among Italian speakers and results supported
Carminati’s (2002
⤠ᱮdings, hence
co渜rming the attested division of labor of Italian pronouns. As for clause order,
although
Carminati (2002
and subordinate-
ain clause orders,
Filiaci (2011
) did. Our study also attempts to
shed some light on this question.
To sum up, previous research on the antecedent preferences of null and overt
pronouns in Romance languages is far from showing a denite picture. On the
one hand, null pronouns seem to have a specic role: they are clearly specialized
in establishing co-reference with elements in subject position, giving continuity
to the element (the subject) that initiated the discourse and that is the current
topic. And this is so irrespective of the sentence type: subordinate intra-senten
tial contexts as well as inter-sentential contexts. On the other hand, the use of
overt pronouns does not oier a clear picture since varying degrees of anteced
ent
�references are shown across languages. More precisely, while in Italian overt
pronouns visibly signal a switch in subject reference, this connection seems to be
much weaker in Spanish, thus suggesting the existence of microvariation among
null subject languages.
Given the observation that there is cross-linguistic variation in the usage and
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
ktab.

e-boy
read
book

�.
ktab.

ead
the-boy
book

tab
ld.

ead
book
the-boy

�e boy read a book.’
studies based on empirical ndings that address this issue in MA and other
rabic
dialects. Only a few studies have examined the topic in Egyptian Arabic, EA
Farghaly 1982
Schulte-Nafeh 2004
). Looking at subordinate contexts, such as
those analyzed in the present research study,
Farghaly (1982
) proposed that third
person overt pronouns convey contrast and emphasis functions; he also addressed
Studies on the acquisition of pronominal anaphora
Sorace and collaborators initiated the current growing body of research on the
acquisition of pronominal anaphora. Based on intra-sentential contexts, their
work focused in how Italian [null-subject language]-English [non-null-subject lan
guage] bilinguals and L2 learners managed this phenomenon in Italian according to
Carminati’s (2002
) PAH. Among their work, the most inuential and widely cited in
�urora Bel & Estela Gar揭a-Alcaraz
the bilingual acquisition literature is perhaps
pronominal subjects in [±topic shi] contexts. e performance of
glish-Italian
and Spanish-Italian children aged 6–7 and 8–10 was compared to that observed
among age-matched monolingual children and monolingual adults. e authors
considered the potential impact of language exposure and, consequently, two sub
groups were included: (1)
glish-Italian bilinguals living in Italy and (2) English-
Italian bilinguals living in the UK. An acceptability judgement task that followed a
short video clip was devised⸠�e experiment dealt with two conditions each having
two levels: type of pronoun (null vs. overt) and discourse function (+topic s桩1
�opic shi). For
talian, the patterns of results yielded more pragmatically inap
propriate overt subject pronouns among younger monolingual and bilingual chil
dren than among older monolingual children and adults. Moreover, in line with
previous ndings (
Montrul & Rodríguez Louro 2006
Rothman 2009
), the trouble
shown by previous work, the scope of the overt pronoun in
panish seems to be
wider than in
talian. e overuse of overt pronouns in bilingual children has also
been attested in Dutch-Italian bilinguals (
Pinto 2006
), German-Italian bilinguals
Müller, Kupisch, Schmitz &
antone 2006
), English-Italian bilinguals (
Serratrice
Serratrice, Sorace & Paoli 2004
), Spanish-English bilinguals (
Paradis &
Navarro 2003
), and Turkish-English bilinguals (
Haznedar 2006
As far as L2 is concerned, many studies also reported a notable proportion
of pragmatically infelicitous overt subject pronoun (
), and null subject pronouns (
Keating,
VanPatten & Jegerski 2011
Lafond, Hayes & Bhatt 2001
Montrul 2004
Montru氚&
Rodriguez Louro 2006
Rothman 2009
Rothman (2009
) examined the syntax-before-discourse hypothesis through
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
Also, more recently,
Research questions
�s the PAH operative in non-biased ambiguous intra-sentential contexts
in Spanish? Does sentence clause order have an eect on the resolution of
pronouns?
�urora Bel & Estela Gar揭a-Alcaraz
ties of null and overt third person subject pronouns in Spanish?
�s there any cross-linguistic inuence with respect to this issue due to cross-
linguistic microvariation?
�o MA-Spanish bilinguals (i.e. early sequential bilinguals) dier from L2
learners in their mastery of these properties? Furthermore, have the dierent
amounts of language exposure that bilinguals and L2 learners receive had an
impact on their knowledge of null and overt pronouns?
�dditionally, in order to perform an eective comparison and discover poten
tial cross-linguistic eects, what are the main factors that aiect pronoun reso
lution preferences in ambiguous intra-sentential contexts in MA?
We ᱲst address question 5 and report the corresponding results for MA in the
next section.
Experiment 1: Anaphora resolution pre-test in Moroccan Arabic
Design and procedure
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
two previous referents. Participants were asked to answer who was th攠‘doer鈠of the
aria
anan
melli

aria
surprised-3s-F
ħanan
when
pro
married-3s-F

kun
we

Who
ᘏ8
married?

anan B. Maria

anan B. Maria



Hi
allem
abdelwaħed
melli
owa
afa

am
�aught -3s-M
abdelawħed
hen
he
ᘏ8

kun li
afa
men

ecovered-3s-M
the accident
Who
ᘏ8
�ecovered
he accident?

abdelwaħed
. Hi

abdelwaħed
. Hi

‘Hi
am taught
abdelwaħed when he was recovered from the accident.
Who was recovered from the accident? A.
abdelwaħed B. Hi
a涒

Melli
san
at
an,
dat
tebki.

hen
䦀san
picked up-3s-F
‚izlan,
pro
started-3s-F
crying

kun
�da
ibki?

ᘏ8
started
crying?

san
. ‚izlan

san
. ‚izlan

When I€san picked up ‚izlan, she started crying.

ho started crying? A. I€san B. ‚izlan’
�urora Bel & Estela Gar揭a-Alcaraz

�.
elli
atiħa
gat
adia,
hiya
bdat
tedħak.

hen
Fatiħa
interrupted-3s-F
Nadia,
she
tarted-3s-F
laughing

kun
�da
idħak?

ᘏ8
started
laughing?

adia
. Fatiħa

adia
. Fatiħa

When Fatiħa interrupted Nadia, she started laughing.

ho started laughing? A. Nadia B. Fatiħa’
Main ᱮdings
Proportions of responses were evaluated. Comparing mean scores (see Figur敳21
and 2), dierent proportions of responses were attested depending on the type of
pronoun. Subjects were more frequently chosen as antecedents for null pronouns.
Ὡs preference was even more pronounced in the subordinate-main order (66.89%
subordinate-main vs. 58.37% main-subordinate). As for overt pronouns, subjects
and objects were selected as antecedents, regardless of sentence order (44.33% of
subjects in main-subordinate vs. 46.67% of subjects in subordinate-main).
Figure 1.
Percentage of antecedent preferences of pronouns in MA (native speakers);
main-subordinate order
�e data were submitted to a 2 × 2 repeated measures ANOVA with type
of pronoun (null vs. overt) × sentence order (main-subordinate vs. subordinate-
main) as within-subject factors on the basis of mean proportions. e analysis
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
yielded a main eect of type of pronoun, (F(1, 44) = 18.983,
= .000), no main
). However, although
Figure 2.
Percentage of antecedent preferences of pronouns in MA (native speakers);
subordinate-main order
�urora Bel & Estela Gar揭a-Alcaraz
behaviour regarding null pronouns in MA does not dier from other null subject
languages that have been examined within the framework of the PAH (Italian,
and L2 Spanish
Participants
Twenty-six early sequential MA-Spanish bilinguals and 26 MA L2 learners of
Table 1.
Early sequential MA-Spanish bilinguals鈠sociolinguistic background
Speakers information
Mean
Range
Age (years)
Arrival age (years)
Self-reported Spanish level (out of 100)
Self-reported MA level (out of 100)
Average daily use of Spanish (out of 100)
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
�e L2 learners were students of Spanish at the
Instituto Cervantes
of
Marrakech (Morocco) (see Tab汥22). eir rst language is MA and all partici
pants were placed in general Spanish courses for a B2 level. As we mentioned ear
lier, Morocco is a multilingual country so our learners had knowledge of various
languages: in addition to French, MSA, Spanish and MA (their rst language),
they also report dierent levels of knowledge of other languages such as English,
among others. Importantly, none of them reported high prociency in a dierent
null subject language (besides MA and Spanish).
Table 2.
L2 learners鈠sociolinguistic background
Age (mean in years)
Procedure and materials
�urora Bel & Estela Gar揭a-Alcaraz
the continuation according to a four-value Likert scale, ranging from 1 (totally
unacceptable) to 4 (perfectly acceptable).
Table 3.
Implicit causality of experimental verbs (bias towards subject in percentages;
from
Verb
Bias towards
subject
Verb
Bias towards
subject
Abandonar
(to abandon)
Investigar
(to investigate)
Aguantar
(to withstand)
Recoger
(to pick up)
Asustar
(to scare)
Saludar
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
and randomized across two lists. Participants were randomly assigned either to
list1 or list 2. Table 4 provides an example of each experimental condition.
Table 4.
Main-subordinate clause. Null
pronoun. Subject antecedent
Iker evita a Iván cuando tiene problemas. Iker tiene
problemas.
‘Iker avoids Ivan when
pro
has problems. Iker has
problems’
Main-subordinate clause. Null
pronoun. Object antecedent
Ángel asustó a Héctor mientras entraba en la habita捩ón.
Héctor entraba en la habita捩ón.
‘Angel scared Héctor while
pro
came in the room. Hector
came in the room’
Main-subordinate clause. Overt
pronoun. Subject antecedent
Fernando investigó a Antonio cuand漠él trabajaba para el
gobierno. Fernando trabajaba para el gobierno.
‘Fernando investigated Antonio when he worked for the
government. Fernando worked for the government’
Main-subordinate clause. Overt
pronoun. Object antecedent
Sebastián sorprendió a Marcos mientras él se casaba.
Marcos se casaba.
Results
In this section we report the general results of our study. We ᱲst describe the
�urora Bel & Estela Gar揭a-Alcaraz
Table 5.
Mean rates (and SDs) in the acceptability preference of pronouns in Spanish
(maximum score = 4)
Group
Main-subordinate
Null pronoun
Overt pronoun
Subject
Antecedent
Object
Antecedent
Subject
Antecedent
Object
Antecedent
Spanish native (control)
Early sequential bilinguals
Spanish L2 learners
Subordinate-main
Null pronoun
Overt pronoun
Subject
Antecedent
Object
Antecedent
Subject
Antecedent
Object
Antecedent
Spanish native (control)
Early sequential bilinguals
Spanish L2 learners
Starting with main-subordinate sentence order, Spanish native speaker con
trols rated subject and object antecedents as equally probable; from Table 5, a near
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
distinction), the two experimental groups remain rather neutral in their anteced
ent choices in this specic condition (a mere
0.08 distinction for bilinguals and
0.23 for L2 speakers).
We performed a repeated measure ANOVA on the overall 2 × 2 × 2 × 3 design
with sentence order (main-subordinate vs. subordinate-main) × type of pronoun
(null pronoun vs. overt pronoun) × antecedent choice (subject vs. object) and group
Figure 3.
Antecedent preferences for main-subordinate order (native speakers)
�urora Bel & Estela Gar揭a-Alcaraz
Figure 4.
Antecedent preferences for subordinate-main order (native speakers)
Independent analyses inside each order revealed an eect both for pronoun
= .006) and antecedent (F(1, 16) = 32.859,
= .000) in main-
subordinate order, whereas only antecedent (F(1, 16) = 7.095,
= .017) – as well
regardless of sentence order. On the other hand, null pronouns only show a pre
ference pattern when the subordinate clause precedes the main clause. Surpris
ingly, this is not in accordance with previous ndings for Spanish (
Alonso-Ovalle
整2al. 2002
Filiaci 2011
). We can attribute this divergence to the dierent nature of
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
our task; recall that our participants are not forced to make a choice as in Alonso
Figure 5.
Antecedent preferences for main-subordinate order
On the other hand, in subordinate-main order a main eect of group was
found in the two remaining conditions. Signicant results were found for null pro
nouns referring to subject antecedents (F(2, 40) = 8.628,
= .001) and for overt
pronouns referring to object antecedents (F(2, 40) = 4.417,
= .018). Post hoc
�urora Bel & Estela Gar揭a-Alcaraz
Tukey tests co渜rmed that L2 learners diered from both the natives (
and the bilinguals (
= .019) in the null pronoun-subject antecedent condition
and from natives in the overt-pronoun-object antecedent condition (
= .020) (see
Figure 6).
Figure 6.
Antecedent preferences for subordinate-main order
In summary, the eect of pronoun type is signicant for all groups; this eect is
stronger in subordinate-main clause order where null and overt pronouns display
a marked division of labour. In this specic clause order, some dierences have
Discussion and conclusions
We rst address research question 1 concerning the results of our native speak
their
�verage answer rating displayed a clearer pattern when the sentence order
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
was subordinate-main. In this case, their answers were remarkably biased as they
subordinate-main order.
�urora Bel & Estela Gar揭a-Alcaraz
evaluated. Even though all these studies deal with ambiguous anaphora, none of
them controlled for implicit causality of verbs. is could be a source of the dis
questio渚5). e pre-test on anaphora resolution conducted for MA provided new
empirical data and revealed that the PAH is only partially applicable in this lan
guage. Null pronouns show a strong tendency towards topic continuity since they
prefer to establish co-reference with elements in subject position regardless of sen
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
approached from the perspective of the PAH (Italian, Spanish and Catalan). On
the other hand, overt pronouns show a more unpredictable behavior since they
were assigned to either subject or object antecedents. us, the pattern observed
for MA is more in line with
Carminati (2002
), for Italian, and
Mayol and Clarck (2010
) for Catalan, which
to be a robust and steady phenomenon among 摩ierent null-
ubject languages,
even typologically distinct (Romance and Arabic), whereas overt pronouns are
more unpredictable.
In light of the IH (
Sorace 2011
), pronominal anaphora resolution, as a phe
nomenon at the syntax-discourse interface, is a vulnerable domain that could also
be aiected by cross-linguistic inuence according to the proposal by
Hulk and
ller (2000
), as there is structural overlap in the two languages involved. Given
that pronoun preferences seem to work in similar directions in MA and
panish
(with the mentioned exception of null pronouns in main-subordinate clause
order), one would expect a positive transfer, thus reinforcing the favorite options
Sorace & Filiaci 2006
�urora Bel & Estela Gar揭a-Alcaraz
both null and overt pronouns. For the opposite clause order, main-subordinate,
MA and L2
panish in order to explore a potential transfer eect, the L2 learners
do not seem to take advantage of their rst language: if they would have projected
the choices of MA for null pronouns – the element that shows most consistent
choices–, higher rates of acceptance should have been attested in L2 Spanish. Null
pronouns in the subordinate-main order are particularly striking and illuminat
ing: null pronouns display the most extreme choice towards subject antecedents
both in L1 MA (almost 70%; see Figure 2) and in L1 Spanish (3.5 mean rate out
o昚4; see Figure 6) but L2 learners behave dierently from natives (see Figure 6);
thus they do not seem to apply the L1 MA rules to L2 Spanish. Bilinguals, on the
other hand, pattern typically with natives, except in the case of overt pronouns. We
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
Another important issue to highlight is the overuse of overt pronouns in
Montrul and
odriguez
Louro (2006
), who found that both null and overt pronouns were aiected by
overproduction/overacceptance in advanced developing grammars. No over
acceptance of either null or overt pronouns was found for L2 learners in the pres
ent study. �e question of why overextention of overt pronouns especially aiects
bilinguals still remains an open question.
Next, we address research question 4: Is there an eect of the dierent amount
of exposure on the acquisition of pronominal anaphora resolution in Spanish?
�urora Bel & Estela Gar揭a-Alcaraz
Of course, it is impossible to identify the amount of inpu琠‘exemplars鈠needed
for a successful acquisition. However, as pointed out above, if the appropriate use
of pronouns is constrained by a number of complex features, then it appears rea
sonable to think that a notable frequency of occurrences will be necessary. What is
more, it is well attested that overt third person pronouns are relatively infrequent
in natural input whereas null pronouns, on the other hand, are numerous. e
scarcity of pronouns in spontaneous data has been invoked to explain the delay in
the L1 acquisition of overt pronouns (
Shin & Cairns 2012
⤮‟us, in the case of our
L2 learners, the underrepresentation of a linguistic element whose acquisition is
highly dependent on input practice must denitely have some cost. As a nal note,
it should be added that neither formal instruction nor instructional input address
this issue explicitly.
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
Acknowledgments
Ὡs research received nancial support throug栚FFI2009-09349 an搚FFI2012-
35058grants (Aurora Bel) and the FPU program pre-doctoral grant (Estela
Gar揭a-Alcaraz) from the Spanish Ministry of Education. Special thanks are due
to A. R. Lara for her linguistic advice and to two anonymous reviewers for their
useful comments on an earlier version of this chapter. We would like to sincerely
than欚the two editors fo爚their suggestions and valuable comments while editing
this boo欮2Finally, our de数2gratitude goes to the students who participated in the
experiments and to th攚
Instituto Cervante猚
of Marrakech for its kind and collab
orative support.
References
Alhawary, M. (2007). Null subjects use by English and Spanish learners of Arabic as an L2. In
E. Benmamoun (Ed.),
Perspectives on Arabic Linguistics XIX
(pp. 217–245). Amsterdam:
JohnBenjamins.
DOI: 10.1075/cilt.289.16alh
Alhawary, M. (2009).
Arabic Second Language Acquisition of Morphosyntax.
New Haven, CT:
Yale University Press.
Alonso-Ovalle, L., Fernández-Solera, S., Frazier, L., & C汩1on, C. (2002). Null vs. overt pronouns
and the topic-focus articulation in Spanish.
Rivista di Linguistica
Aoun, J., Benmamoun, E., & Choueiri, L. (2010).
se Syntax of Arabic
. Cambridge: CUP.
DOI: 10.1017/cbo9780511691775
Ariel, M. (1990).
Accessing Noun Phrase Antecedents
. London: Routledge.
DOI: 10.1017/s0022226700015504
Ariel, M. (2001). Accessibilit礠�eory: An overview. In T. Sanders, J. Schilperoord & W⸚Spooren
(Eds.),
Text Representation: Linguistic and Psycholinguistic Aspects
(pp. 29–87).
msterdam:
John Benjamins.
Arnold, J., Brown-Schmidt, S., & Trueswell, J. (2007). Children’s use of gender and order-of-
mention during pronoun comprehension.
Language and Cognitive Processes
DOI: 10.1080/01690960600845950
Arnold, J., Eisenband, J., Brown-Schmidt, S., & Trueswell, J. (2000). e rapid use of gender
discourse: Data from Catalan.
Written Language and Literacy
DOI: 10.1075/wll.13.2.03bel
�urora Bel & Estela Gar揭a-Alcaraz
Bever, T., & Townsend, D. (1979). Perceptual mechanisms and formal properties of main and
subordinate clauses. In W. Cooper & E. C. T. Walker (Eds.),
Sentence Processing: Psycholin
guistic Studies Presented to Merrill Garrett
(pp. 159–226). Hillsdale, NJ: Lawrence Erlbaum
Associates.
Bialystok, E. (2001).
Bilingualism in Development: Language, Literacy and Cognition
Cambridge:
CUP.
DOI: 10.1017/cbo9780511605963
Carminati, M.N. (2002).
se Processing of Italian Subject Pronouns.
Unpublished Ph.D. disserta
owers
(Eds.),
Discourse Analysis: eory and Application Proceedings of the Second National Sym
posium on Linguistics and English Language Teaching
(pp. 60–69). Cairo: Center for Devel
oping English Language Teaching, Ain Shams University.
Filiaci, F. (2011).
Anaphoric Preferences of Null and Overt Subjects in Italian and Spanish: a Cross-
linguistic Comparison
. Unpublished Ph.D. dissertation. University of Edinburgh.
DOI: 10.1080/01690965.2013.801502
Benjamins.
DOI: 10.1017/s0022226700010434
DOI: 10.3758/brm.40.3.760
Gordon, P., Grosz, B., & Gilliom, L. (1993). Pronouns, Names, and the Centering of Attention in
Discourse.
Cognitive Science, 17
DOI: 10.1207/s15516709cog1703_1
Haznedar, B. (2006).
Cross-linguistic interference in the bilingual acquisition of Turkish and
English: e overuse of subjects in bilingual Turkish
. Paper presented at GALANA 2,
Montreal.
DOI: 10.1017/s1366728900000353
Jegerski, J., VanPatten, B., & Keating, G. (2011). Cross-linguistic variation and the acquisition of
pronominal reference in L2 Spanish.
Second Language Research
DOI: 10.1177/0267658311406033
Keating, G., VanPatten, B., & Jegerski, J. (2011). Who was walking on the beach? Anaphora reso
lution in Spanish heritage speakers and adult second language learners.
Studies in Second
Language Acquisition
DOI: 10.1017/s0272263110000732
DOI: 10.1007/978-94-010-0351-3_8
ubject pronouns in the L2 Spanish of Moroccan Arabic speakers
Lafond, L., Hayes, R., & Bhatt, R. (2001). Constraint demotion and null subjects in Spanish L2
acquisition. In J. Camps & C. Wiltshire (Eds.),
Romance Syntax, Semantics and L2 Acquisi
tion
(pp. 121–135). Amsterdam: John Benjamins.
DOI: 10.1075/cilt.216.11laf
DOI: 10.1191/0267658306sr264oa
Margaza, P., & Bel, A. (2006). Null subjects at the syntax-pragmatics interface: Evidence from
Spanish interlanguage of Greek speakers. In M. Graham O’Brien, C. Shea & J. Archibald
(Eds.),
Proceedings of the
8th
Generative Approaches to Second Language Acquisition Con
ference
GASLA 2006
(pp. 88–97). Somerville, MA: Cascadilla Proceedings Project.
Mayol, L., & Clark, R. (2010). Pronouns in Catalan: Games of partial information and the use of
linguistic resources.
Journal of Pragmatics, 42
DOI: 10.1016/j.pragma.2009.07.004
Montrul, S. (2004). Subject and object expression in Spanish heritage speakers: A case of
morpho-syntactic convergence.
Bilingualism: Language and Cognition
DOI: 10.1017/s1366728904001464
Montrul, S. (2008).
Incomplete Acquisition in Bilingualism: Re-examining the Age Factor.
Amsterdam: John Benjamins.
DOI: 10.1075/sibil.39
Pinto, M. (2006). Subject pronouns in bilinguals: Interference or maturation. In L.
scoba爚&
V⸚Torrens (Eds.),
se Acquisition of Syntax in Romance languages
(pp. 331–352).
Amsterdam: John Benjamins.
DOI: 10.1075/lald.41.16pin
Pires, A., & Rothman, J. (2009). Acquisition of Brazilian Portuguese in late childhood: Impli
cations for syntactic theory and language change. In A. Pires & J. Rothman (Eds.),
Mini
malist Inquiries into Child and Adult Language Acquisition: Case Studies Across Portuguese
(pp.129–154)
Berlin: Mouton De Gruyter.
DOI: 10.1515/9783110215359.1.129
Roberts, L., Gullberg, M., & Indefrey, P. (2008). Online pronoun resolution in L2 discourse:
L1 inuence and general learner eects.
Studies in Second Language Acquisition
DOI: 10.1017/s0272263108080480
Rothman, J. (2009). Pragmatic decits with syntactic consequences? L2 pronominal subjects
and the syntax-pragmatics interface.
Journal of Pragmatics
DOI: 10.1016/j.pragma.2008.07.007
�urora Bel & Estela Gar揭a-Alcaraz
Schulte-Nafeh, M. (2004).
Overt Subject Pronouns in Cairene Arabic: Pragmatic and Syntactic
Functions
. Unpublished Ph.D. dissertation. e University of Arizona.
DOI: 10.1017/s1366728904001610
Shin, N.L., & Cairns, H.S. (2012). e development of NP selection in school-age children:
reference and Spanish subject pronouns.
Language Acquisition: A Journal of Developmental
Linguistics
DOI: 10.1080/10489223.2012.633846
Sorace, A. (2004). Native language attrition and developmental instability at the syntax-
DOI: 10.7551/mitpress/9780262015127.003.0004
Unsworth, S. (2013). Assessing the role of current an搚cumulative
exposure in simultaneous
bilingual acquisition: e case of Dutch gender.鈚
Bilingualism: Language and Cognition,
DOI: 10.1017/s1366728912000284
e construal of goal-oriented motion events
by Swedish speakers of L2 Spanish
Encoding of motion endpoints and Manner
of motion
Alejandra Donoso
& Emanuel Bylund
Stockholm University &
Linnaeus University
�e current study investigates motion event construal in Swedish speakers of
L2 Spanish. In particular, the study examines the encoding of motion endpoints
and manner of motion through elicited video clip descriptions of everyday
motion event situations. e results show that Swedish learners of Spanish
exhibit the same, high endpoint frequencies as their monolingual Swedish peers,
thus deviating from the Spanish native pattern. Moreover, the learners used the
same amount of manner verbs as Spanish natives, but were more prone to give
additional manner information in periphrastic constructions. ese ndings are
Introduction
In the past two decades, there has been a steady increase in crosslinguistic research
on the selection and organisation of information in discourse. An ever-growing
body of research shows that speakers of dierent languages dier in the way
they structure information about, for example, time, space, and causation (e.g.
Berma渚& Slobin 1994
Hasse汧ård, Johansson, Behrens & Fabricius-Hansen 2002
Strömqvist & Verhoeven 2004
). One domain that has received focused attention
in this line of research is motion events. In a generic denition, a motion event is
a situation that involves physical displacement whereby an entity occupies dier
ent spatial positions at dierent temporal intervals. Current research on motion
events has shown that languages across the world are very selective in the way
they choose and structure (or
construe
) information about motion. Whereas some
�lejandra Donoso & Emanuel Bylund
languages regularly provide information about the Manner in which motion is
approaches to the study of motion. e approach concerned with the encod
-
ing of motion endpoints will be labelle搠‘the Grammatical Aspect approach’,
as its underlying assumption is that crosslinguistic dierences in the encoding
of grammatical aspect induce crosslinguistic dierences in endpoint encoding
(e.g.
Athanasopoulos & Bylund 2013
Bylund 2008
Carroll & von Stutterheim
Schmiedtová, von Stutterheim & Carroll 2011
von Stutterheim & Nüse
⤮‟e other approach is concerned with how information regarding Man
ner and Path (the trajectory of motion) is encoded in dierent languages (e.g.
Donoso 2013
Naigles&
�errazas 1998
Slobin 1996
Talmy 1985
). We will call this
approach to motio渠‘the Manner and Path approach’⸠�e Grammatical Aspect
e construal of goal-oriented motion events by Swedish speakers of L2 Spanish
combine the frameworks oiered by the Grammatical Aspect approach and the
Manner and Path approach in order to gain a more nuanced picture of the vari
ables that govern L2 learning in this domain. Even though it is not uncommon that
languages exhibit such ‘double鈠contrasts, there is, to the best of our knowledge, no
study to date that has attempted to explain learner behaviour in the domain of
motion by integrating the Aspect approach with the Manner and Path approach.
Background
Motion endpoints and grammatical aspect in L1 and L2 speakers
Human beings o1en ‘engage’ in motion events with the intention to reach a dif
ferent location than the one we started from. ere is, in other words, o1en a spe
cic goal or an endpoint to our motion. Research within the Grammatical Aspect
approach has shown that speakers of dierent languages dier with regards to
the extent to which they actually mention endpoints when talking about motion.
For example, when showing a scene with two people walking towards a house,
Athanasopoulos and Bylund (2013
) found that Swedish speakers were more prone
to mentioning the endpoint than were English speakers. e following examples
were typical of the two language groups’ descriptions of the scene:
Two people are walking
Två
�änniskor
查r
ill
ett
hus
two
humans
lk-
᜕ue
to
�ouse

�two
umans walk to a house’
Apart from the obvious dierence concerning endpoint encoding, the two exam
ples above also dier with respect to the selection of the time span during
which
the event unfolds: in the English example, the predicate is marked with progres
sive aspect, and thus the ongoing phase of the event is conveyed. In the Swedish
sentence, in contrast, there is no aspectual morphology.
Evidence from a number of crosslinguistic investigations show that speakers
of English, Modern Standard Arabic, Russian, and Spanish are less prone to men
tioning, allocating visual attention to, and remembering motion endpoints than
are speakers of Afrikaans, Dutch, Norwegian, and Swedish. �ese clusters of end
point behaviour are consistent with the clusters given on the basis of grammatical
aspect in these languages. Even though the aspectual systems of Arabic, English,
Russian, and Spanish may dier with regards to the exact semantic boundaries
�lejandra Donoso & Emanuel Bylund
are not present in the Arabic imperfective, such as habituality), their common
denominator is that they mark the predicate for ongoingness (imperfective and/
or progressive aspect) in a productive and obligatory fashion. Afrikaans, Dutch,
Norwegian, and Swedish, in contrast, do not have grammatical aspect. is means
that these languages do not have the same structural means nor the obligation to
encode ongoingness. Of course, ongoingness may be optionally coded through
lexical circumlocution or serial verbs, but the use of such means in motion event
descriptions is not obligatory and has proven scarce (
Bylund, Athanasopoulos &
Oostendorp 2013
Carroll, von Stutterheim & Nüse 2004
). Crucially then, ‘aspect
languages鈠an搠‘non-aspect languages’ do not necessarily dier in what they are
able to express, but in the relative ease and frequency with which they can express
a concept such as ongoingness.
In aspect languages, the concept of ongoingness
has a higher degree of
codability
than in non-aspect languages.
von
Stutterheim
), for example, examined endpoint encoding patterns among advanced
English speakers of L2 German, and advanced German speakers of L2 English.
No information was given about the learners鈠current place of residence (L1 or
Henceforth, we will use the term “aspect language” to refer to languages in which the
aspectual category of ongoingness is grammaticised, and “non-aspect language” to refer to
languages in which this notion is not grammaticised.
e construal of goal-oriented motion events by Swedish speakers of L2 Spanish
German learners of English were prone to mentioning endpoints to the same
extent as
�erman monolinguals, and English learners of German behaved like
English monolinguals. In view of their ndings, von Stutterheim and colleagues
have suggested that L1 conceptual patterns are highly resistant to restructur
ing (e.g.
Carroll & von Stutterheim 2006
) (see however
Bylund 2009
Bylund &
Athanasopoulos 2014
�e type of transfer documented in these studies is commonly referred to as
conceptual transfer (
Jarvis & Pavlenko 2008
⤮‟e kind of knowledge that is being
transferred is, in other words, language-specic patterns of selecting and structur
ing information relative to motion. ese patterns of event construal have been
acquired in the course of L1 acquisition, whereby the speaker has learnt to attend
to and encode certain aspects of reality in accordance with the lexical and gram
matical categories available in the L1 (
Slobin 1996
). When acquiring a L2 that has
distinct event construal patterns from the L1, the learner has to overcome the reli
ance on L1 patterns when presenting information and instead attain the new pat
Path and Manner of motion in L1 and L2 speakers
Inspired by the work of
Talmy (1985
), the second main approach to the
study of motion events is th攠“Manner and Path approach鐠⡥.g.
Berman ☚
Slobin
Yangklang 2004
). is approach investigates how languages encode information
about the fashion (
Manner
) in which dierent objects move in space and how their
trajectories (
Path
) are conveyed by linguistic means. Path and Manner do not have,
however, equal importance in the expression of motion events. While Path is the
core schema in a motion event, Manner is an optional semantic component. is
typology divides the languages of the world into two main groups depending on
how they encode Path. �e ᱲst group is Satellite-framed languages (S-languages)
and the second Verb-framed languages (V-languages).
In the rst group, Path is
expressed in a satellit攠⢓a grammatical category of any constituent other than a
A third group, equipollently-framed languages (e.g.‟ai) has also been identi�ed. See
Slobin (2004
) for further discussion.
�lejandra Donoso & Emanuel Bylund
noun-phrase or prepositional phrase complement that is in a sister relation to the
verb root”,
Talmy 2000
, p. 102); in the latter, into the main verb.
Several scholars have proposed that Swedish belongs to the group of S-lan
guages (e.g.
Ragnars擳ttir & Strömqvist 2004
) in which Path is typically expressed
in satellites, for example
ut
(out) or
ner
(down), and Manner is encoded in the
main verb. Spanish, on the other hand, would belong to the second group, in
which Motion and Path co渞ate in the main verb, whereas Manner tends to be
conveyed separately, frequently by an adjunct, a gerund, a noun phrase or another
kind of expression (
Stam 2010
Talmy 1985
Flaskan
Ỷt ut
från grottan.
⡆泶t: Manner and Motion; ut: Path)

��e bottle oated out from the cave.’
La botella
sal槳„otando
de la cueva.
(Salió: Path and Motion; otando: Manner)

��e bottle exited oating from the cave.’
In the Swedish example, Manner is expressed in the main verb, i.e.
Flaskan
�
ut
från grottan
⢑the bottle
�oated
out of the cave’), and Path is conveyed in a satellite
‘out’). Spanish, on the other hand, lexicalizes Path in the main verb,
salir
⢑to
go out’), and Manner is encoded in a gerund (
ṯtando
‘ṯating鈩.
Although Manner in Spanish may be expressed in a gerund, it is not a core
element (as is Path), and can therefore be le out in speech. However, when cog
nitively relevant, Manner can be expressed by dierent means, even in the main
verb, though this encoding option may be limited to situations in which the trajec
tory of the moving entity does not include a telic Path phrase (see below about the
“boundary-crossing constraint鐩.
Still, Manner verbs tend to be less numerous in V-languages than in
anguages
Jovanovic & Martinovic-Zic 2004
). Even more, and especially regarding
panish,
explicit Manner descriptions are scarce compared to those of S-languages and
other V-languages, such as Basque (
Cadierno 2010
Pedersen 2012
Stam 2010
) represents a central factor that 摩ierenti
ates both language groups. is restriction, originally proposed by
Aske (1989
), states
that V-languages do
“not generally allow Manner-of-motion verbs to take a direc
tional complement and to function as bounded predicates”
La bailarina bailó en la habitación.
(With directional meaning)

��e dancer danced in the room.’
e construal of goal-oriented motion events by Swedish speakers of L2 Spanish
Because of this constraint, V-languages would be less exible to generate phrases
that both co渞ate Manner and Path in the main verb (
bailar,
‘to dance鈩⁡nd that
are directly tied to telic endpoints (
en la habitación,
‘in the room’, with directional
meaning). In other words, when crossing a boundary, the usual way to describe
it in Spanish would be by using a bare Path verb (without a satellite) plus a telic
prepositional phrase, as in
la mujer entró en la habitación
⢑the woman entered the
room’). However, it is important to take into consideration that in some cases it
is actually possible to express a bounded event (a motion situation containing a
direction and an endpoint) by using a Manner verb (e.g.
Cadierno 2010
Pedersen
To summarize, Swedish and Spanish dier considerably in how they encode
Manner and Path. Overall, Swedish presents fairly complex Path descriptions with
an extensive use of directional satellites, endpoint encoding and a wider range
of Manner verbs than Spanish (
Bylund 2008
Donoso 2013
Sjtröm 1990
Moreover, Swedish L1 speakers seem to follow a highly schematic pattern for
motion encoding, which is usually represented by the structure of [SUBJ V OBL]
Pedersen 2012
). is schema would apparently facilitate motion descriptions
containing information not only about Manner and Path, but also possibly about
the Ground (i.e. a reference point in space to which the Figure moves), in the form
of a bounded predicate.
Several studies on L2 Spanish have investigated how dierences in the L1 may
have an impact in learning in both speech (e.g.
Cadierno 2004
Cadiern漚& Ruiz
Hijazo-Gas揳n 2011
Hohenstein, Esisenberg & Naigles 2006
Larrañaga,
Tr攝ers-Daller, Tidball & Ortega 2012
) and gesture (
Negueruela, Lantolf,
orda渚&
Gelabert 2004
⤮‟e general results speak both for and against a possible cross-
linguistic inuence, which can take two dierent shapes that may be interrelated
or act separately⸠�e ᱲst mode involves transfer that originates at the level of lin
guistic knowledge and implies that structures from the L1 are simply transferred
to the L2; the second one originates at the conceptual level and implies that the
way speakers conceptualize and encode motion events in their L1 aiects how they
express the same type of events in their L2.
As earlier SLA research has shown (
Ijaz 1986
Jarvis & Odlin 2000
Odlin
), L2 learners are prone to transfer dierent aspects of their L1, includ
ing linguistic means and patterns of thought. For instance, Cadierno’s work (
on written narratives of the book
Frog, Where Are You?
Mayer 1969
found that
Danish learners of Spanish used fewer motion verb types than the Spanish L1
speakers; second, some of the learners (in particular, those at the intermediate
level) exhibited a “satellization鐠of the Spanish locative constructions, i.e. the use
of redundant and anomalous Path particles not found in the Spanish speakers鈠
data; third, the learners added more Ground adjuncts to the motion verbs (which
�lejandra Donoso & Emanuel Bylund
could include the source, the medium, and the goal of movement). Consequently,
Cadierno assumes that the L1 patterns for motion encoding may still play a role in
adult L2 learning, and that prociency levels seem to interact with the impact of
crosslinguistic inuence. In sum, her results show that the learners’ L1 seems to be
present in some traits of their L2 use but not in others: the Danish learners proved
to be able to encode both Manner and event co渞ation in the same fashion as the
Spanish natives but they were still inuenced by their L1 when encoding Path and
Ground (
Cadierno 2004
, p. 42).
In a similar vein,
Navarro and Nicoladis (2005
) using oral narratives of a
Pink Panther
cartoon among procient English speakers of L2 Spanish (most
of them students at University level), found that learners and native Spanish
speakers use Path verbs to the same extent. Secondly, they found that L2 speak
ers produce more Path-intransitive verbs followed by a post-verbal phrase,
which could contain information about goal, source and location. is latter
using somewhat more Manner verbs. At the opposite end, the native
panish
speakers resorted more to bare Path intransitive verbs and produced fewer
Manner verbs, which they balanced by including a larger number of post-verbal
adverbials and gerunds. In sum, the authors conclude that L2 Spanish speakers
had almost fully achieved the L1 Spanish pattern to describe motion events,
which is consistent with the above mentioned ndings of
Cadierno (2004
).
Similar results have been obtained by
Hijazo-Gas揳n (2011
) on German speak
ers of L2 Spanish, and
Cadierno and Ruiz (2006
) on Danish learners of Spanish.
Surprisingly, and contrary to these results,
ous diculties, even among learners who had been living in a
panish-speaking
country prior to data collection. �ese subjects were prone to use
anner verbs
in combination with a Path or locative preposition (e.g. the preposition
which in Spanish does not entail a directional meaning) thus evading the use of
the gerund and co渞ating Manner and Motion when confronted to a boundary
crossing situation.
e construal of goal-oriented motion events by Swedish speakers of L2 Spanish
Aims and scope of the present study
Against this background, the overall aim of the current study is to investigate
how Swedish speakers of L2 Spanish construe goal-oriented motion. As seen
above, Swedish and Spanish dier with regards to the encoding of motion events.
Swedish, being a non-aspect language, presents a higher tendency than Spanish to
encode endpoints for goal-oriented motion events. Moreover, since Swedish is an
S-language, it has a greater repertoire of Manner verbs than Spanish, and
wedish
speakers allegedly use Manner verbs more frequently and in slightly dierent con
texts than Spanish speakers.
With these crosslinguistic dierences in mind, the study seeks to address
�o Swedish speakers of L2 Spanish encode motion endpoints more o1en than
Spanish L1 speakers?
�n view of the previous ndings on endpoint encoding patterns in L2 speak
ers (e.g.
von Stutterheim 2003
), it could be expected that the Swedish learners
will be more prone than Spanish L1 speakers to taking a holistic perspective
of goal-oriented motion, according to which endpoints are included.
�o Swedish speakers of L2 Spanish encode information about Manner of
motion to the same extent as Spanish L1 speakers?
�f transfer from the L1 occurs in this regard, the high frequency with which
Manner of motion is encoded in Swedish could be expected to prompt the
learners to encode Manner information to a higher degree than Spanish L1
speakers.
�oes the encoding of information relative to Manner of motion vary as a
function of endpoint encoding?
�panish and Swedish exhibit dierent typological traits with regard to the
encoding of Manner in combination with an event endpoint (as seen above).
If the Swedish learners combine endpoints and Manner information in the
same way as Spanish L1 speakers, then they can be said to have assimilated the
�lejandra Donoso & Emanuel Bylund
Participants
Seventeen Swedish speakers of L2 Spanish participated in the study⸠�ese indi
viduals were living in Chile at the time of testing, and Spanish was their primary
language of communication. �eir mean age was 35 years and their mean length
of residence in Chile was 8 years. Seven of these individuals had received some
kind of formal instruction in Spanish, either in Sweden prior to departure or in
Chile. All of them had learnt Spanish as adults.
Two groups of L1 Spanish speakers and L1 Swedish speakers were included
for comparative purposes. e Spanish speakers (
= 12) were originally from
Argentina and Chile, and were short-term visitors in Sweden at the time of testing.
�e Swedish speakers (
= 12) were living in Sweden at the time of testing. Both
native speaker groups were in their mid-thirties.
Material
�ategory III). In addition, there were 14 distractor clips depicting people or ani
mals involved in activities void of trajectory-based motion, such as a ma渠�y-
sshing by a river⸠�e distractors were not included in the analyses.
�e rst category (I), consisting of scenes with a high degree of orientation,
showed how the endpoint of the trajectory was reached, for example, a seagull
landing on a perch. Given that the arrival at the endpoint of these events was
overtly shown, speakers of 摩ierent languages are not expected to 摩ier sig
nicantly in the degree to which they mentioned endpoints in these clips (
von
Stutterheim, Andermann, Carroll, Flecken & Schmiedtová 2012
). Half of the clips
in this category contained boundary-crossing motion (e.g. a dog walking into a
greenhouse).
�e second category (II), with an intermediate degree of goal-orientation,
contained scenes in which there was a visible, possible endpoint for the motion,
e construal of goal-oriented motion events by Swedish speakers of L2 Spanish
but the arrival at it was not overtly shown. �e subjects鈠references to these events
did not need to include a mention of the goal of the motion. An example of this
is the situation of a man on a horse carrier on the countryside: on this item, the
subjects could say either a man riding on a horse-drawn carriage or a man riding
on a horse-drawn carriage to a village.
Finally, category III with a low degree of goal-orientation, consisted of scenes
depicting the motion of an entity along a trajectory without a clearly discernable
endpoint. An example of this is the clip with a jeep driving in the desert: in this
scene, there is no conceivable endpoint except for perhaps a mountain range at
the horizon.
Procedure
�e subjects were administered the tests individually by a native speaker of the rel
evant language⸠�e participants were shown the clips on a computer monitor and
were asked to describe what was happening in each scene, following the prompt
Qupasa en la escena?

‘What happens in the scene鈩r
Vad händer i scenen?
⢑What happens in the scene?鈩⁡s soon as they recognised the type of situation.
Moreover, they were told to focus on the event and not to describe the scene in
terms of, for example, weather conditions or the protagonist’s clothing.
Each participant’s descriptions of the video clips were audio-recorded and
transcribed⸠�e descriptions were then classied according to endpoint encoding
and perspectivation. Locative expressions containing the moving entit禒s arrival at
or intention of arriving at a goal were counted as endpoint verbalisations. Exam
ples of such expressions in Spanish are
ir
a una parada de autobuses
⢑to go
to a
bus stop
caminar
hacia un convento
⢑to walk
towards a convent
�lejandra Donoso & Emanuel Bylund
Manner can be conveyed is through a verbal periphrasis, such as in the sentence
un perro entra corriendo

‘a dog enters running’, i.e⸠‘a dog runs in’), where
cor
riendo
is the gerund of the verb
correr

‘to run鈩⸠Ὡs type of information about
Manner of motion was labelle搠‘periphrastic Manner’.
Results
�e ndings will be presented in the order of the research questions posed in
the section above. Firstly, the results on endpoint encodings will be presented.
Secondly, the patterns of use of Manner verbs and periphrastic encoding of
and Manner of motion is assessed.
Endpoint encoding in L2 Spanish
An analysis of the participants鈠scene descriptions showed that the Spanish L1
speakers encoded event endpoints 27.2 % (SD 10.2) of the time, whereas the
Swedish L1 speakers did so 50 % (SD 16.4) of the time⸠�e L2 Spanish speakers
When confronted with a scene showing, for example, two women walking towards
a building, the L2 speakers were more prone than the Spanish L1 group to describe
the event as
dos mujeres caminando
hacia un edi�cio

‘two women walking
towards
a building
�e endpoint encodings were further analysed according to category of end
point orientation (i.e. I = high endpoint orientation; II = medium endpoint ori
e construal of goal-oriented motion events by Swedish speakers of L2 Spanish
Post-hoc tests (Tuke禒s) revealed that the monolingual L1 Spanish and L1
wedish
speakers diered signicantly from each other in each category (p .02). As for
the L2 speakers, it was found that they encoded signicantly more endpoints than
the Spanish L1 speakers only in category II (p
.02) but not in categories I or III
.05). is means that the L2 endpoint frequencies diverged with the L1 fre
quencies in the clip category where there was a movement towards a goal, but the
goal was not reached.
Table 1.
Endpoint frequencies (%) according to endpoint orientation category
Category I
Category II
Category III
Spanish L1
5.2 (8.4)
Spanish L2
Swedish L1
Manner of motion in L2 Spanish
�e next main step in the analysis involved examining the scene descriptions with
regards to the encoding of Manner of motion. To this end, the amount of Manner
verbs was calculated across all three groups. �e Swedish L1 speakers were shown
to use Manner verbs in 79.2 % (SD 14.0) of their motion descriptions, whereas the
Spanish L1 speakers did so in 29.0 % (SD 7.4) of theirs. For example, when con
fronted with a scene of a car driving on a country road, the Swedish speakers were
more prone to say
en bil kör plandet

‘a car drives on the countryside’), whereas
the Spanish speakers would say
un coche yendo por el campo

‘a car going on the
countryside鈩⸠Ὡs result aligns with previous ndings showing that speakers of
�lejandra Donoso & Emanuel Bylund
S-languages are more prone to encode Manner information into the main verb
than are speakers of V-languages. �e L2 speakers indeed adhered to the Spanish
pattern, using Manner verbs in only 26.5 % (SD 8.8) of their motion descriptions.
Manner information in endpoint encodings
the proportion of Manner verbs in those scene descriptions containing
dpoint
e construal of goal-oriented motion events by Swedish speakers of L2 Spanish
encodings. It turned out that 77.3% (SD 16.2) of the Swedish L1 endpoint encod
ings contained Manner verbs, whereas the corresponding number for the L1
Spanish encodings was 33.5 % (SD 26.2). As for the L2 Spanish speakers, 26.7 %
(SD 12) of their endpoint encodings contained Manner verbs. ese distributions
are, in other words, very similar to the ones documented above on general use of
Manner verbs in the three groups. An ANOVA indicated that these dierences
were statistically signicant,
.001, such that the L1 and L2
Spanish patterns were signicantly dierent from the L1 Swedish pattern (
2㰚.01)
but not from each other.
Next, the Spanish L1 and L2 groups were examined with regards to their use
of periphrastic Manner in scene descriptions that contained endpoint encodings.
�e native speaker group was found to convey this kind of Manner information in
9.7 % (SD 5.9) of their endpoint encodings, whereas the L2 speaker group did so
in 21.2 % (SD 8.6) of theirs. An independent samples
-test co渜rmed that these
摩ierences were statistically signicant,
.05. is means that the L2
speakers were more likely to produce endpoint encodings in which information
about Manner was given periphrastically (e.g.
en bicicleta
ₑon bicycle’,
en auto
ₑin
was similar to that of L1 speakers in terms of the extent to which they used
anner
verbs. However, the L2 speakers tended to encode periphrastic Manner more o1en
in combination with endpoints than did the Spanish L1 speakers.
Discussion
Ὡs paper asked three questions with regards to motion event construal in
wedish
speakers of L2 Spanish: First, do Swedish speakers of L2 Spanish encode motion
endpoints more o1en than Spanish L1 speakers? Second, do Swedish speakers of
L2 Spanish encode information about Manner to the same extent as Spanish L1
speakers? And third, does the encoding of information relative to Manner vary as
�lejandra Donoso & Emanuel Bylund
a function of endpoint encoding? In what follows, we will discuss the results with
regards to each of these questions.
Starting with the issue of endpoint behaviour, the nding that the L2 speakers
encoded motion endpoints to the same elevated extent as monolingual Swedish L1
speakers is taken as evidence of L1-based conceptual transfer⸠�at is to say, when
speaking Spanish, the L2 speakers are more prone to adopting an event perspec
tive by which motion endpoints are included. is they do to the same extent as
their Swedish monolingual peers. Interestingly, however, it was also found that
even though the L2 speakers in general diered from the Spanish L1 speakers, a
more nuanced picture of the L2 endpoint behaviour emerged when the endpoint
encoding were analysed according to category of endpoint orientation. Here, it
was found that the learners really only diered from the L1 Spanish speakers in
the category that shows movement towards a goal, but the goal is not reached
e construal of goal-oriented motion events by Swedish speakers of L2 Spanish
Spanish speakers exhibited elevated endpoint frequencies in spite of their pros
ciency with the gerund, then it would suggest that it is not the mere presence of
used Manner verbs to the same extent, which is in line with the results of
Cadierno
Cadierno and Ruiz (2006
) and
Hijazo-Gas揳n (2011
), but not with the
outcomes presented in the study of
Navarro and Nicoladis (2005
), who found
that English learners of Spanish use somewhat more Manner verbs than
panish
natives, and the ones of
�lejandra Donoso & Emanuel Bylund
the encoding patterns of both Spanish L1 and Swedish L1. is was observed rst
in the extent to which they encoded an endpoint, following the Swedish L1 typical
construal pattern of motion events; second, in the extent to which they used ger
und forms to express ongoingness, adapting to the Spanish pattern; and third, in
the signicant use of periphrastic Manner, which is a typical feature of the Spanish
language but which was not overtly used by the Spanish L1 speakers in this study.
Regarding the third nding, and as earlier explained, Spanish presents a syn
tactic restriction which limits the use of Manner verbs associated with telic comple
ments and therefore the range of Manner verbs and their use might be reduced both
in the Spanish L1 and L2 data in these type of scenes. However, the reason behind
Conclusions
e construal of goal-oriented motion events by Swedish speakers of L2 Spanish
References
Aske, J. (1989). Path predicates in English and Spanish: A closer look.
Proceedings of the Berkeley
Linguistics Society, 15
Athanasopoulos, P., & Bylund, E. (2013). Does grammatical aspect aiect motion event cogni
tion? A cross-linguistic comparison of English and Swedish speakers.
Cognitive Science,
DOI: 10.1111/cogs.12006
Berman, R.A., & Slobin, D.I. (1994). Development of linguistic forms: English. In R.A. Berman
& D.I. Slobin (Eds.),
Relating Events in Narrative: A Crosslinguistic Developmental Study
Hillsdale, NJ: Lawrence Erlbaum Associates.
DOI: 10.1017/s0272263100014443
Bylund, E. (2008). Procesos de conceptualización de eventos en sueco y en español: Diferencias
translin柼ísticas [Event conceptualization processes in Swedish and Spanish: Crosslinguis
tic dierences].
Revue Romane
DOI: 10.1075/rro.43.1.02byl
Bylund, E. (2009). e eects of age of L2 acquisition on L1 event conceptualization principles.
Bilingualism: Language and Cognition, 12
DOI: 10.1017/s1366728909990137
Bylund, E., & Athanasopoulos, P. (2014). Language and thought in a multilingual context: e
case of isiXhosa.
Bilingualism: Language and Cognition
DOI: 10.1017/s1366728913000503
Bylund, E., & Jarvis, S. (2011). L2 攝ects on L1 event conceptualization.
Bilingualism: Language
and Cognition, 14
DOI: 10.1017/s1366728910000180
Bylund, E., Athanasopoulos, P., & Oostendorp, M. (2013). Motion event cognition and gram
matical aspect. Evidence from Afrikaans.
Linguistics, 51
DOI: 10.1515/ling-2013-0033
Cadierno, T. (2004). Expressing motion events in a second language: A cognitive typological
perspective. In M. Achard & S. Niemeier (Eds.),
Cognitive Linguistics, Second Language
Acquisition, and Foreign Language Teaching
(pp. 13–49).
SOLA Studies on Language
Acquisition. Berlin: Mouton de Gruyter.
DOI: 10.1515/9783110199857.13
Cadierno, T. (2010). Motion in Danish as a second language: Does the learner’s L1 make a dif
ference? In Z. Han & T. Cadierno (Eds.),
Linguistic Relativity in SLA:“inking for Speaking
(pp. 1–33). Bristol: Multilingual Matters.
DOI: 10.1017/s0272263111000428
Cadierno, T., & Ruiz, L. (2006), Motion events in Spanish L2 acquisition.
Annual Review of
Cognitive Linguistics, 4
DOI: 10.1075/arcl.4.08cad
Carroll, M., & von Stutterheim, C. (2006). e impact of grammaticalized temporal catego
ries on ultimate attainment in advanced L2 acquisition. In H. Byrnes (Ed.),
Educating for
Advanced Foreign Language Capacities
DOI: 10.1515/9783110894028.183
Carroll, M., Weimar, K., Flecken, M., Lambert, M., & von Stutterheim, C. (2012). Tracing tra
jectories: Motion event construal by advanced L2 French-English and L2 French-German
speakers.
Language, Interaction and Acquisition
DOI: 10.1075/lia.3.2.03car
Donoso, A. (2013). Camino, base y manera en español y sueco. Un estudio contrastivo [Path,
Ground and Manner in Spanish and Swedish. A contrastive study].
Revue Romane, 48
DOI: 10.1075/rro.48.1.01don
�lejandra Donoso & Emanuel Bylund
Hasse汧ård, H., Johansson, S., Behrens, B., & Fabricius-Hansen, C. (Eds.). (2002).
Information
Structure in a Cross-linguistic Perspective
. Amsterdam: Rodopi.
DOI: 10.1075/lic.7.2
Hendriks, H., & Hickmann, M. (2011). Space in second language acquisition. In V. Cook &
In J. Guo, E. Lieven, N.
udwig,
S. Ervin-Tripp, K. Nakamura, & S. Ö竧alişkan (Eds.),
Crosslinguistic Approaches to the
Psychology of Language: Research in the Tradition of Dan Isaac Slobin
(pp. 403–414).
Ne眚York, NY: Psychology Press.
Discourse across Languages and Cultures
(pp. 211–226).
Amsterdam: John Benjamins.
DOI: 10.1075/slcs.68.12jov
Langacker, R. (2000).
Grammar and Conceptualization
. Berlin: De Gruyter.
DOI: 10.1515/9783110800524
Larrañaga, P., Tr攝ers-Daller, J., Tidball, F., & Ortega, M. (2012). L1 transfer in the acquisition
of manner and path in Spanish by native speakers of English.
International Journal of Bilin
gualism
DOI: 10.1177/1367006911405577
Mayer, M. (1969).
Frog: Where are you?
New York, NY: Dial Press.
Naigles, L., & Terrazas, P⸚(1998). Motion-verb generalizations in English and Spanish: �e
楮�uence of language and syntax.
Psychological Science, 9
(5), 363–369.
DOI: 10.1111/1467-9280.00069
Naigles, L., Eisenberg, A., Kako, E., Highter, M., & McGraw, N. (1998). Speaking of motion:
Verb use in English and Spanish.
Language and Cognitive Processes
DOI: 10.1080/016909698386429
Navarro, S., & Nicoladis, E. (2005). Describing motion events in adult L2 Spanish narratives.
In D. Eddington (Ed.),
Proceedings of the
6th
Conference on Acquisition of Spanish and
Portuguese as First and Second languages
(pp. 102–107). Somerville, MA: Cascadilla
Press.
Negueruela, E., Lantolf, J.P., Jordan, S., & Gelabert, J. (2004). 攠“private function鐠of gesture in
second language speaking activity: A study of motion verbs and gesturing in English and
Spanish.
International Journal of Applied Linguistics
DOI: 10.1111/j.1473-4192.2004.00056.x
Odlin, T. (1989).
Language Transfer: Cross-linguistic Iniuence in Language Learning.
Cambridge:
CUP.
DOI: 10.1017/cbo9781139524537
e construal of goal-oriented motion events by Swedish speakers of L2 Spanish
Odlin, T. (2003). Crosslinguistic inuence. In C. Doughty & M. Long (Eds.),
se Handbook of
Second Language Acquisition
(pp. 436–486). Oxford, UK: Blackwell.
DOI: 10.1002/9781405198431.wbeal0292
Pedersen, J. (2012). Spanish expressions of directed motion revisited 阠variable type framing. In
H. Boas & F. Go湺álvez-Gar揭a (Eds.),
Constructional Approaches to Romance Languages
(pp. 269–304). Amsterdam: John Benjamins.
Ragnars擳ttir, H., & Strömqvist, S. (2004). Time, space and manner in Icelandic and Swedish. In
S. Strömqvist & L. Verhoeven (Eds.),
Relating Events in Narrative: Typological and Contex
tual Perspectives
(pp. 113–141). Hillsdale, NJ: Lawrence Erlbaum Associates.
Schmiedtová, B., & Flecken, M. (2008). e role of aspectual distinctions in event encoding:
Implications for second language acquisition. In S. de Knop & T. de Rycker (Eds.),
Cogni
tive Approaches to Pedagogical Grammar
(pp. 357–384). Berlin: De Gruyter.
DOI: 10.1017/s0272263109990428
Schmiedtová, B., von Stutterheim, C., & Carroll, M. (2011). Language-specic patterns in event
construal of advanced second language speakers. In A. Pavlenko (Ed.),
sinking and Speak
ing in Two Languages
(pp. 112–141). Clevedon: Multilingual Matters.
Sjtröm, S. (1990).
Spatial Relations. Towards a eory of Spatial Verbs, Prepositions and Pro
nominal Adverbs in Swedish
. Unpublished Ph.D. dissertation. Gothenburg University,
Sweden.
Slobin, D. (1996). Two ways to travel: Verbs of motion in English and Spanish. In M. Shibatani &
S. �ompson (Eds.),
Grammatical Constructions. eir Form and Meaning
(pp. 195–219).
Oxford: Clarendon Press.
Slobin, D. (2003). Language and thought online: Cognitive consequences of linguistic relativity.
In D. Gentner & S. Goldin-Meadow (Eds.),
Language in
Mind
(pp. 157–191). Cambridge,
MA: e MIT Press.
Slobin, D. (2004). How people move. In C.L. Moder & A. Martinovic-Zic (Eds.),
Discourse across
Languages and Cultures
(pp. 195–210). Amsterdam: John Benjamins.
DOI: 10.1075/slcs.68.11slo
Stam, G. (2010). Can an L2 speaker’s patterns of thinking for speaking change? In Z. Han &
T⸚Cadierno (Eds.),
Linguistic Relativity in SLA:“inking for Speaking
(pp. 59–83)
Bristol:
Multilingual Matters.
DOI: 10.1017/s0272263111000428
Strömqvist, S., & Verhoeven, L. (2004).
Relating Events in Narrative, Vol. 2: Typological and Con
textual Perspectives
. London: Taylor & Francis.
Talmy, L. (1985). Lexicalization patterns: Semantic structure in lexical forms. In T. Shopen (Ed.),
Language Typology and Semantic Description, Vol. 3: Grammatical categories and the lexi
con
(pp. 36–149)
Cambridge: CUP.
Talmy, L. (1991). Path to realization: A typology of event co渞ation. In
Proceedings of the 17th
Annual Meeting of the Berkeley Linguistics Society
(pp. 480–519). Berkeley, CA: Berkeley
DOI: 10.1162/coli.2000.27.2.309
von Stutterheim, C. (2003). Linguistic structure and information organisation. e case of very
advanced learners. In S. Foster-Cohen & S. Pekarek Dohler (Eds.),
EuroSLA Yearbook
(pp.183–206). Amsterdam: John Benjamins.
DOI: 10.1075/eurosla.3.11stu
von Stutterheim, C., & Nüse, R. (2003). Processes of conceptualization in language production:
Language-specic perspectives and event construal.
Linguistics
DOI: 10.1515/ling.2003.028
�lejandra Donoso & Emanuel Bylund
von Stutterheim, C., Andermann, M., Carroll, M., Flecken, M., & Schmiedtová, B. (2012). How
grammaticized concepts shape event conceptualization in language production: Insights
from linguistic analysis, eye tracking data, and memory performance.
Linguistics
DOI: 10.1515/ling-2012-0026
Zlatev, J., & Yangklang, P. (2004). A third way to travel: e place o映�ai in motion event typol
ogy. In S. Strömqvist & L. Verhoeven (Eds.),
Relating Events in Narrative. Typological and
Contextual Perspectives
(pp. 159–190).
Hillsdale, NJ: Lawrence Erlbaum Associates.
Cambridge, MA: e MIT Press.
8ᘕᐠጓ6
Spanish as an L2 in an instructional context
Object drop in L2 Spanish, (complex) feature
reassembly, and L1 pre-emption
Comparing English, Chinese, European and
Brazilian Portuguese learners
Michael Iverson & Jason Rothman
Macquarie University / University of Reading
L2 learners from dierent L1s may dier in L2 development and ultimate
Introduction
�ichael Iverson & Jason Rothman
if it is reasonable to account for these dierences without hypothesizing that the
mechanisms responsible for successful acquisition in childhood are inaccessible
to adults. Attempting to do so seems warranted in light of ample acquisition evi
Accounting for lack of success in L2 acquisition
bject drop in L2 Spanish, (complex) feature reassembly, and L1 pre-emption
Schwartz and Sprouse’s FT/FA model, more recent work within generative SLA
has acknowledged the limitations of assuming solely full transfer (e.g.
Schwartz
) and has taken seriously the task of trying to articulate more precisely the
dynamic nature of the causes behind protracted delays and failures in adult second
language acquisition.
Herein, we focus on two of several eorts in the literature attempting to
account for some L2 dierences while at the same time problematizing simplis
tic notions of L1 transfer: (1) L1 pre-emption eects and (2) feature reassembly.
Although initial L1 transfer does not always block new L2 acquisition given access
to UG, it is also the case that new L2 syntactic acquisition does not always result in
learning task, but rather acquisition proper is further complicated by the
resence
of the L1 due to the complexities of re-mapping procedures. Indeed, these are
independent tasks. In the case where no new L2 features need to be acquired, the
With this understanding of pre-emption, L2 acquisition scenarios susceptible to the
Subset Principle (
Manzini & Wexler 1987
) would serve as potential cases of a pre-emption
problem. Distinct from the Subset Principle, however, pre-emption problems need not neces
sarily stem from contrasting L1/L2 settings of the same linguistic parameter.
�ichael Iverson & Jason Rothman
relevant features must still be combined and mapped to the appropriate lexical
items and learned to be used in felicitous discourse contexts. Considering that
linguistic features may be instantiated overtly (i.e. morphologically) or not, and
Portuguese) examining the acquisition of syntactic and semantic properties
related to the licensing of Spanish object drop, we will try to oier a principled
account of why these learners at comparable levels of Spanish prociency dis
bject drop in L2 Spanish, (complex) feature reassembly, and L1 pre-emption
Object expression in Spanish
Spanish employs a paradigm of object clitics for pronominal object expression.
Accusative object clitics are inected for person, number, and additionally for gen
der in the 3rd person. While not generally considered a true null object language,
Trajiste
cara?

brought-you
the
camera

Did you bring the camera?’

S
traje.

yes,

it
�rought-I

Yes, I brought it.’
Trajiste
galletas?

brought-you
cookies

Did you bring cookies?’

S
*las
traje.

yes,

it
rought-I

Yes, I brought them/some.’

Bruhn de Garavito & Guijarro-Fuentes 2002
, p. 60)
De�niteness and speci�city are properties that refer to the state of knowledge of both
the speaker and hearer, and to the saliency or noteworthiness of the entity (as determined
by the speaker), respectively (see e.g.
Ionin 2006
). Br楥ny and informally, denite entities are
those identi�able by or known to both the speaker and hearer, and speci�c entities are those
which the speaker considers to have some noteworthy quality (regardless of the hearer’s
knowledge of this entity).
�ichael Iverson & Jason Rothman
In (1), a denite, specic
object may be replaced by an object clitic, but cannot be
omitted in the answer. In contrast, as seen in (2), an indenite, nonspecic object
must be dropped in the response.
Similarly, this alternation is seen in Clitic L攙
Dislocation and topicalization structures, as in (3) and (4) respectively:
Estos
�apatos,
los
�ompré
emana
pasada

�ese
shoes,

hem
bought-I
the
week
ast

��ese shoes, I bought last week.’
Caf
*lo
tomo
�odas
añanas

�oiee,
(*it)
ink-I
every
the
mornings

�Coiee, I drink every morning.’

Cuza, Pez-Leroux & Sánchez 2013
, p. 97)
Campos (1986
), following
Huang’s (1984
) analysis for Chinese and Raposo’s
(1986) for European Portuguese, claimed that the dropped object in cases like (2)
is an operator that has moved to the le periphery. By invoking movement, this
claim makes further predictions that dropped objects in Spanish are subject to
subjacency constraints. is is borne out in the data: while indenite object clitics
can be dropped in embedded clauses, as in (5a), they cannot be dropped from DP
islands (5b), CP islands (5c), or adjunct islands (5d).
uan trajo cerveza a la esta?

Did Juan bring beer to the party?’

Su novia me dijo que
*la
trajo.

His girlfriend told me that he brought (some).’

�.
Existe el rumor de que *
trajo.

�ere exists the rumor that he brought (some).’

Que *
trajo es obvio.

�at he brought (some) is obvious.’

�.
Sí, todos nos emborrachamos porque *
trajo.

Yes, we all got drunk because he brought (some).’

apted from
Campos 1986
, p. 355)
So, in addition to the semantic restriction on deniteness, the realization of
dropped objects in Spanish also depends on the syntactic environment. In
eneral,
In addition to de�niteness, dropped objects in Spanish also depend on the speci�city of
the referent, and are limited to nonspeci�c objects. A full treatment of speci�city, including
In contrast to
Campos’ (1986
) claims, the clitic in examples like (2) is optionally available
to some speakers. We note this here, but as the consequences are minimal for the studies we
bject drop in L2 Spanish, (complex) feature reassembly, and L1 pre-emption
for native-like knowledge of dropped objects, learners of Spanish must associate
the object clitic with the deniteness of the referent, learn that only indenite
objects may be dropped (the semantic restriction), and learn that dropped objects
are a result of movement in the grammar, and subject to subjacency (the syntactic
restriction).
Acquisition of object drop in L2 Spanish by speakers of various L1s
L1 English:
Bruhn de Garavito & Guijarro-Fuentes (2002
Bruhn de Garavito and Guijarro-Fuentes (2002
) examined the acquisition of
dropped objects by L1 English learners of L2 Spanish (as well as L1 European
Portuguese learners of L2 Spanish, discussed below). English is distinct from
Spanish in that it does not have object clitics and does not allow indenite object
drop. English learners of L2 Spanish must develop a representation of the Spanish
clitic paradigm in addition to learning the semantic and syntactic restrictions on
dropped objects.
Participants in this study were 18 L1 English speakers at an intermediate level
Table 1.
Item types in
Bruhn de Garavito and Guijarro-Fuentes (2002
) and their
grammaticality in Spanish
Item type
Referent
Structure
Clitic
Grammaticality
Denite
Simple
Overt
Denite
Simple
Null
Indenite
Simple
Null
Denite
Complex DP
Overt
Indenite
Complex DP
Null
Denite
Embedded CP
Null
Indenite
Sentential CP
Null
Denite
Adjunct
Overt
Indenite
Adjunct
Null
�ichael Iverson & Jason Rothman
Post-hoc tests following up a signicant ANOVA revealed that the L1
glish
L1 European Portuguese:
Bruhn de Garavito & Guijarro-Fuentes (2002
In the same study described in the previous section, Bruhn de Garavito and
Guijarro-Fuentes also tested L1 European Portuguese learners of L2 Spanish. Sim
ilar to Spanish, European Portuguese has object clitics marked for person, number
and gender (in the 3rd person). However, unlike Spanish, European Portuguese
allows for denite null objects, as seen in (6):
�oana
viu
na
�V
�ntem.

�e
Joana
aw
on
TV
yesterday

�Joana saw them/him/her/it on TV yesterday.’
Raposo 1986
, p. 373)
While the semantic restriction on dropped objects diers from Spanish, the syn
tactic restrictions are the same. Dropped objects are subject to subjacency con
straints, and European Portuguese examples analogous to the Spanish examples in
(5b–d) would be ungrammatical. Because dropped objects show the same restric
tions on movement as those in Spanish, they are claimed to have the same under
lying syntactic representation (
Raposo 1986
). Given that L1 European Portuguese
learners of Spanish already have knowledge of clitics and the syntactic represen
tation of dropped objects, and assuming L1 transfer (in the vein of
Schwar瑺2&
Sprouse 1996
), the learning task for these learners would consist only of narrowing
the licit semantic space, limiting dropped objects to exclusively those with in摥s
nite antecedents.
bject drop in L2 Spanish, (complex) feature reassembly, and L1 pre-emption
�e European Portuguese group (
11) in this study were at an advanced
L1 Brazilian Portuguese:
Rothman & Iverson (2013
Rothman and Iverson (2013
) tested L1 Brazilian Portuguese learners of L2
panish.
While Brazilian Portuguese (BP) has a full paradigm of accusative clitics like those
in Spanish, it is distinct from Spanish with respect to dropped objects. First, there
is no restriction on deniteness,
as seen in question-answer pairs (7) vs. (8):
�armen
trouxe
orvete
ntar?

�e
Carmen
brought-3.
the
ice cream
to-the
dinner?

�Did Carmen bring ice cream to the dinner?’

Ela
trouxe,
im.
Trouxe,
im.

She
rought-3.
yes.

brought-3.
yes.

Yes, she brought it.’
Yes, she brought (it).’
�ablo
erviu
erveja
na
esta?

�he
Pablo
serve-3.
beer
he
party

�Did Pablo serve beer at the party?’
We note here that we are restricting our discussion to inanimate objects. Animate objects
are subject to diierent constrains (see
Bianchi & Figueiredo 1993
Schwenter 2006
for details).
Experimental tasks used by Bruhn de Garavito & Guijarro-Fuentes and Rothman & Iverson
consisted of only inanimate objects.
�ichael Iverson & Jason Rothman

Ele
serviu,
im
Serviu,
im.

served-3.
yes.
served-3.
yes.

Yes, he served it.’
Yes, he served (it).’

Rothman & Iverson 2013
, p. 596)
Carlos trouxe vinho para a festa?

Did Carlos bring beer to the party?’

A sua namorada disse que
trouxe.

His girlfriend said that he brought (some).’

�.
Não conheço a pessoa que
trouxe.

I don’t know the person that brought (it).’

Que
troux攠é
claro.

�at he brought (some) is obvious.’

�.
Sim, todos nós„camos⁢êbados porque
trouxe.

Yes, we all got drunk because he brought (some).’

apted from
Rothman & Iverson 2013
, p. 597)
In contrast to Spanish, objects can be omitted in DP islands (9b), CP islands (9c),
and adjunct islands (9d).
Farrell (1990
) observed that because these examples lack
subjacency 攝ects, syntactic movement must not be involved in the derivation of
omitted objects in BP, and claimed that they are true null objects (i.e.
pro
). is
syntactic representation of null objects in BP and its counterpart in Spanish and
ravito
and Guijarro-Fuentes, expanding it to more thoroughly examine the potential
bject drop in L2 Spanish, (complex) feature reassembly, and L1 pre-emption
acquisition of the deniteness restriction. e participants, L1 BP learners of L2
Spanish (
Table 2.
Item types in
Rothman and Iverson (2013
) and their grammaticality in Spanish
Item type
Referent
Structure
Clitic
Grammaticality
Denite
Simple
Overt
Denite
Simple
Null
Indenite
Simple
Null
Denite
Complex DP
Overt
Denite
Complex DP
Null
Indenite
Complex DP
Overt
Indenite
Complex DP
Null
Denite
Sentential CP
Overt
Denite
Sentential CP
Null
Indenite
Sentential CP
Overt
Indenite
Sentential CP
Null
Denite
Adjunct
Overt
Denite
Adjunct
Null
Indenite
Adjunct
Overt
Indenite
Adjunct
Null
For simple syntactic structures, BP learners showed sensitivity to deniteness,
rating the grammatical items (realization of a denite object clitic and omission
of an indenite object) signicantly higher than the ungrammatical item (omis
sion of a denite object). ey were less successful with items testing the syntac
tic restriction in various island types. ey did rate indenite dropped objects in
complex DPs signicantly lower than indenite dropped objects in simple syntac
�ichael Iverson & Jason Rothman
denite objects were similarly rejected, but dropped indenite objects were rated
as acceptable.
Rothman and Iverson claim that these results show robust knowledge of the
semantic restriction, and, given a lack of comparable performance demonstrat
ing knowledge of the syntactic restriction, suggest that the semantic and syntac
tic portions of this property can be acquired independently of each other⸠�e
results here also corroborate Bruhn de Garavito and Guijarro-Fuentes’ conclusion
that the deniteness constraint can be acquired even by those learners whose L1
L1 Chinese:
Cuza, Pez-Leroux & Sánchez (2013
Zhangsan
shuo
Lisi
eshi

�angsan
ay
isi
not
know

�Zhangsan said that Lisi doesn’t know (him).’
Huang 1984
, p. 541)
bject drop in L2 Spanish, (complex) feature reassembly, and L1 pre-emption
Huang (1984
) notes that the null object in Chinese examples like (10) cannot refer
to the matrix subject (here,
Zhangsan
), but rather refers to some external indi
Neige
ren
�angsan
huo
Lisi
eshi

�at
an
Zhangsan
say
isi
not
know

��at man, Zhangsan said that Lisi doesn’t know (him).’

Huang 1984
, p. 541)
Based on these observations,
Huang (1984
) proposed that null objects in Chinese
can be analysed as a variable bound by a (null) topic. is, then, is similar to the
case of Spanish (and indeed his analysis was the starting point for the analysis of
dropped objects in Spanish), with the dierence being that null objects are not
restricted to indenites in Chinese, similar to European Portuguese.
�ichael Iverson & Jason Rothman
contexts in which the clitic could be omitted were those in which the antecedent
was indenite and nonspecic.
We focus here on the results from the adult L1 Chinese learners of L2 Span
ish for comparability with the other studies. On the acceptability judgment task
these learners tended to accept all item types, with group average ratings greater
than 4 out of 5. As expected, this group allowed for dropped objects in all con
texts. Contrary to the predictions of the authors, however, the adult Chinese
learners also accepted an object clitic in indenite contexts. Results from other
tasks were consistent with those of the acceptability judgment task. Adult learn
ers reliably used null objects in favour of clitics in elicited production, did not
Discussion
bject drop in L2 Spanish, (complex) feature reassembly, and L1 pre-emption
Table 3.
Results from each L1 Group
Knowledge of restrictions?
Semantic
Complex DP
Sentential CP
Adjunct
English
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Chinese
Not tested
Not tested
Not tested
�e English, European Portuguese and Brazilian Portuguese groups all showed
knowledge of the semantic restriction on deniteness, while the Chinese group did
not. Which groups showed knowledge of the syntactic restrictions was less clear,
uropean
Portuguese group respected subjacency constraints in all conditions, the English
group did so with both Complex DP islands and Sentential CP islands (but not with
adjunct islands), and the Brazilian Portuguese group did so with only Complex DP
islands.
parison, this is unfortunate. One could speculate that since
hinese and
panish
have the same syntactic instantiation (i.e. an operator), the group would show no
problems with the syntax, akin to what was shown for the European Portuguese
group, but this would need to be empirically co渜rmed.
Although each study does a good job at attempting to explain the perfor
mance of the group on which it focuses, the observable discrepancy across dier
ent L1s acquiring the same properties in L2 Spanish is worthy of consideration. L1
transfer seems a likely possibility to explain the dierences, however, the question
remains as to what exactly from the L1 inuences the dierences noted. Recall
We do not ignore the fact that Brazilian Portuguese does not regularly use third person
object clitics, although they have them (and these learners are competent speakers of the stan
�ichael Iverson & Jason Rothman
not instantiate zero-objects, while the other three L1s do. However, in none of
these languages do the zero-objects have the same distribution as they do in Span
ish. In all three languages, they are not restricted to only indenite antecedents,
but also allow denite antecedents. Both Chinese and European Portuguese show
subjacency eects (and thus are assumed to have the same syntactic representation
as in Spanish), but Brazilian Portuguese does not.
In addition to these properties that seem straightforwardly related to Spanish
clitics and dropped objects, we may also want to consider a property that is less
directly related – the instantiation of deniteness in these languages. Deniteness
can be dened as the property of being able to be ident検ed by both the speaker
and the hearer (e.g.
Ionin, Ko & Wexler 2004
). According to
Lyons (1999
), lan
uropean
Portuguese and Brazilian Portuguese, which all have denite/
denite article
systems, serve as prototypical cases of languages in which deniteness has been
grammaticalized. In languages like these, deniteness is obligatorily and unambig
uously marked using unique functional morphology. Chinese, on the other hand,
may be considered a language in which deniteness has not been grammaticalized
and is expressed through more semantic or pragmatic means. Chinese indicates
deniteness through (an interaction of) context, syntactic position or mo摩s
cation of the nominal with a classier or plural morpheme (
Chen朚& Sybesma
). Furthermore, marking a noun as denite or indenite is not obligatory, and
representation and semantic constraints of dropped objects. e L1
hinese have
bject drop in L2 Spanish, (complex) feature reassembly, and L1 pre-emption
perhaps the greatest challenge: they must acquire the Spanish clitic system, restrict
the semantic space in which dropped objects are available, and, if a grammatical
ization of deniteness is necessary for the acquisition of object drop, they must
also acquire this feature.
Table 4.
L1 Properties relevant for L2 Spanish
L1 Properties
Grammatical
deᵮiteness
Clitics
Ø-Objects
(Type)
宖deᵮite]
only
Subjacency
Span
Yes
Yes
Yes (Op)
Yes
Yes
Eng
Yes
No (n/a)
Yes
Yes
Yes (Op)
Yes
Yes
Yes
Yes (
pro
Yes (Op)
Yes
�ichael Iverson & Jason Rothman
�e case of the Brazilian and European Portuguese groups is distinct from that
parsing failure and subsequent grammatical restructuring.
razilian
ortuguese
See
Rothman and Iverson (2013
) for further explanation.
bject drop in L2 Spanish, (complex) feature reassembly, and L1 pre-emption
As
Rothman and Iverson (2013
) argue the Brazilian Portuguese performance with
Spanish-
�peaking country and had unrestricted access to abundant input that was
both native and natural. As was seen though, with respect to the semantic con
straint, the European Portuguese group (as well as the Brazilian Portuguese group)
succeeded while the Chinese group did not. Furthermore, the English group, who
were unaided by linguistic transfer, also outperformed the Chinese group. Appeal
ing to L1 transfer for this domain alone, these discrepancies are not predicted,
morphosyntactically realized in both the L1 and L2, but requires the
eassembly
�ichael Iverson & Jason Rothman
of features. Harder still is the acquisition of a property which is contextually
expressed in the L1, but morphosyntactically expressed in the L2. Combining this
with
Lyons’ (1999
) idea that languages dier in how they mark deniteness, we
may be able to paint a more satisfactory picture of the acquisition task and its
relative diculty for the learner groups examined here. To acquire object drop
in Spanish, all groups must rst acquire clitics and their associated features (e.g.
deniteness and specicity). Both European and Brazilian Portuguese have clit
ics, while English and Chinese do not. In spite of the similarity in this respect
and indenite dropped objects, and both successfully acquire the correct
panish
constraint – but rather at the syntactic level. In this respect, European
ortuguese
speakers have nothing new to acquire, as their syntactic representation of dropped
objects is what is found in Spanish. e Brazilian Portuguese speakers must
acquire this representation, which is distinct from that of their L1. However, this
cannot be the entire story either, because the English group, who similarly must
acquire a representation for object drop that they do not have in their L1 gram
bject drop in L2 Spanish, (complex) feature reassembly, and L1 pre-emption
that a comparable L1 property will no longer be entertained as a possibility in
the L2 grammar, thus leading to apparent variation (
Rothman & Iverson 2011
Trahey & White 1993
). In the following section, we bring the discussion of
Conclusion
We have provided a macro-analysis that accounts for the whole of the existent data
on the acquisition of object drop in L2 Spanish in an eort to address a greater
concern of L2 theory, that of overall explanatory adequacy of the processes of
�ichael Iverson & Jason Rothman
DOI: 10.1515/ling.2004.42.6.1129
Cheng, L.L., & Sybesma, R. (1999). Bare and not-so-bare nouns and the structure of NP.
Linguis
tic Inquiry, 30,
DOI: 10.1162/002438999554192
Chomsky, N. (1986).
Barriers.
Cambridge, MA: e MIT Press.
DOI: 10.9793/elsj1984.6.213
Clahsen, H., & Hong, U. (1995). Agreement and null subjects in German L2 development: New
evidence from reaction-time experiments.
Second Language Research, 11
DOI: 10.1177/026765839501100103
Cuza, A., Pez-Leroux, T., & Sánchez, L. (2013). e role of semantic transfer in clitic drop
among simultaneous and sequential Chinese-Spanish bilinguals.
Studies in Second Lan
guage Acquisition, 35
DOI: 10.1017/s0272263112000691
bject drop in L2 Spanish, (complex) feature reassembly, and L1 pre-emption
Gabriele, A. (2009). Transfer and transition in the SLA of aspect: A bidirectional study of learn
ers of English and Japanese.
Studies in Second Language Acquisition, 31
DOI: 10.1017/s0272263109090342
Hamann, C. (2003). Phenomena in French normal and impaired language acquisition and their
implications for hypotheses on language development.
Probus, 15,
DOI: 10.1515/prbs.2003.006
Huang, J. (1984). On the distribution and reference of empty pronouns.
Linguistic Inquiry, 15,
Ionin, T. (2006). is is denitely specic: Specicity and deniteness in article systems.
Natural
Language Semantics, 14,
DOI: 10.1007/s11050-005-5255-9
Ionin, T., Ko, H., & Wexler, K. (2004). Article semantics in L2 acquisition: the role of specicity.
Language Acquisition, 12,
DOI: 10.1207/s15327817la1201_2
Kupisch, T., & Müller, N. (2009). Relating Italian articles and clitic object pronouns in bilingual
children acquiring German and Italian. In C. Dimroth & P. Jordens (Eds.),
Functional Cat
egories in Learner Language
(pp. 307–342). Berlin: De Gruyter.
DOI: 10.1515/9783110216172.307
Lardiere, D. (2009). Some thoughts on the contrastive analysis of feature in second language
acquisition.
Second Language Research, 25
DOI: 10.1177/0267658308100283
Lyons, C. (1999).
De�niteness
. Cambridge: CUP.
DOI: 10.1017/cbo9780511605789
DOI: 10.1075/lald.31
Meisel, J. (1997). e acquisition of the syntax of negation in French and German: Contrasting
ᱲst and second language development.
Second Language Research, 13,
DOI: 10.1191/026765897666180760
Meisel, J. (2011).
First and Second Language Acquisition: Parallels and Di�erences.
Cambridge:
CUP.
DOI: 10.1017/cbo9780511862694
Ramchand, G., & Svenonious, P. (2008). Mapping a parochial lexicon onto a universal seman
tics. In M.T. Biberauer (Ed.),
Limits of Syntactic Variation
(pp. 219–245). Amsterdam: John
Benjamins.
DOI: 10.1075/la.132.08ram
Raposo, E. (1986). On the null object in European Portuguese. In O. Jaeggli & C. Silva-Corvalán
(Eds.),
Studies in Romance Linguistics
(pp. 373–390). Dordrecht: Foris.
Rothman, J. (2008). Why not all counter-evidence to the Critical Period Hypothesis is equal or
problematic: Implications for SLA.
Language and Linguistics Compass, 2
DOI: 10.1111/j.1749-818x.2008.00098.x
Rothman, J., & Iverson, M. (2008). Poverty-of-the-stimulus and L2 epistemology: Considering
L2 knowledge of aspectual phrasal semantics.
Language Acquisition: A Journal of Develop
mental Linguistics, 15
DOI: 10.1080/10489220802352206
Rothman, J., & Iverson, M. (2011). L1 preemption and L2 learnability㨠�e case of object drop
in Brazilian Portuguese native learners of L2 Spanish. In N. Danis, K. Mesh, & H. Sung
(Eds.),
BUCLD 35: Proceedings of the 35th
Annual Boston University Conference on Lan
guage Development
(pp. 296–307). Somerville, MA: Cascadilla Press.
Rothman, J., & Iverson, M. (2013). Strong islands and null objects in L2 Spanish of Brazilian
Portuguese natives: Do you know the learners who drop ____?
Studies in Second Language
Acquisition, 35
�ichael Iverson & Jason Rothman
叡nchez, L. (2004).
Object Agreement and the Checking of Focus in Spanish.
Unpublished ms,
Rutgers University.
Schwartz, B. (1998). e second language instinct.
Lingua,
DOI: 10.1016/s0024-3841%2898%2900032-1
Schwartz, B. (2013, April).
At what level, “transfer”?
Plenary paper presented at Generative
Approaches to Language Acquisition 12, University of Florida, Gainesville, FL.
Schwartz, B., & Song, H. (2009). Testing the Fundamental Dierence Hypothesis: L2 adult, L2
child, and L1 child comparisons in the acquisition of Korean
-constructions with nega
tive polarity items.
Studies in Second Language Acquisition, 31
Schwartz, B., & Sprouse, R. (1996). L2 cognitive states and the full transfer/full access model.
Second Language Research, 12
DOI: 10.1177/026765839601200103
Schwartz, B., & Sprouse, R. (2000). When syntactic theories evolve: Consequences for L2 acqui
sition research. In J. Archibald (Ed.),
Second Language Acquisition and Linguisti挠seory
(pp. 156–186). Oxford: Blackwell.
Schwenter, S. (2006). Null objects across South America. In T.L. Face & C. Klee (Eds.),
Selected
Proceedings of the
8th
Hispanic Linguistics Symposium
(pp. 23–36). Somerville, MA:
Cascadilla Press.
Slabakova, R. (2006). Is there a critical period for semantics?
Second Language Research, 22,
DOI: 10.1191/0267658306sr270oa
Slabakova, R. (2008).
Meaning in the Second Language.
Berlin: De Gruyter Mouton.
DOI: 10.1017/s0272263110000124
DOI: 10.1177/0267658308100291
Slabakova, R., & Montrul, S. (2003). Genericity and Aspect in L2 Acquisition.
Language Acquisi
tion, 11,
DOI: 10.1207/s15327817la1103_2
Trahey, M., & White, L. (1993). Positive evidence and preemption in the second language class
room.
Studies in Second Language Acquisition, 15,
DOI: 10.1017/s0272263100011955
Westergaard, M. (2003). Unlearning V2: Transfer, markedness and the importance of input cues
in the acquisition of word order in English by Norwegian children.
EUROSLA Yearbook,
DOI: 10.1075/eurosla.3.07wes
White, L. (1989).
Universal Grammar and Second Language Acquisition.
Amsterdam: John
Benjamins.
DOI: 10.1177/026765839100700306
White, L. (2003).
Second Language Acquisition and Universal Grammar
. Cambridge: CUP.
DOI: 10.1017/cbo9780511815065.002
Yuan, B. (2001). e status of thematic verbs in second language acquisition of Chinese: Against
inevitability of thematic-verb raising in second language acquisition.
Second Language
Research, 17
DOI: 10.1191/026765801680191505
e acquisition of dierential object marking
in Spanish by Turkish speakers
Silvina Montrul & Ay
e Gﱲel
University of Illinois Urbana-Champaign / Bo…aziçi University
�e obligatory use of the prepositio渠‘
with animate, specic direct objects
in Spanish (
Juan conoce
María
ₑJuan knows Maria鈩⁩s a well-known instance
of Dierential Object Marking (DOM). is study investigates the acquisition
of DOM in Spanish by native speakers of Turkish, a language that exemplies
DOM on the basis of case marking. Twenty native speakers of Spanish and 32
Introduction
In second and third language acquisition, inectional morphology (tense, aspect,
�ilvina Montrul & Ay
e Gﱲel
the relevant abstract
eatures or their values, or they can be due to the fact that the
feature may be part of their grammars but the learner lacks the relevant knowl
edge of the conditions for expressing it. An additional complication for second
and third language acquisition is previous knowledge of a language. It has been
shown that the rst language constitutes the initial state of L2 acquisition (see
Schwar瑺2& Sprouse 1996
Full Transfer/Full Access Hypothesis and the research
it generated). As a result, learners already bring an entrenched system of morpho
syntactic features already assembled into lexical items. �ese are the morphemes
associated with the functional categories in their language. If appropriate and
sucient positive evidence exists, L2 learners may be able to restructure their
�ohn saw Maria/the woman.
glish (no DOM)
Juan vio
María/la mujer
panish DOM

Juan saw DOM Maria/the woman.’
e acquisition of dierential object marking in Spanish by Turkish speakers

Juan vio
María/la mujer.

‘Juan saw Maria/the woman’

Juan saw Maria/the woman.’
In recent years, Spanish DOM has received attention in the acquisition litera
ture because it appears to be a very vulnerable area in language contact situa
tions. �e phenomenon is acquired early, before age 3 years, by monolingual
children (
Rodríguez-Mondoñedo, 2008
), but it is very dicult to master in L2
acquisition by speakers of languages that do not have DOM (
Bowles & Montrul
Guijarro-Fuentes 2011
Guijarro-Fuentes & Marinis 2007
Spanish
DOM has also been the focus of extensive investigation in early bilingualism
Guijarro-Fuentes & Marinis 2011
Ticio in press
) and in language attrition when
the dominant language (e.g. English, French) does not exhibit DOM (
Girard 1995
Grosjean& Py 1991
Montrul 2004
Montrul & Sánchez-Walker 2013
In this study, we assume the Full Transfer/Full Access Hypothesis (
Schwar瑺2&
Sprouse 1996
) subsumed under the Feature Reassembly Hypothesis (
Lardiere
) to investigate the acquisition of Spanish DOM by Turkish speakers.
Both Spanish and Turkish have DOM but although they are very similar in many
respects the two languages also dier in the conditions under which the relevant
features operate in each language⸠�ere are several existing studies of Dierential
Object Marking in Spanish as a L2 especially by speakers whose rst language
(L1) does not have DOM. Our study is unique in providing new empirical evi
dence of the acquisition of this property of Spanish by speakers whose language
instantiate DOM.
Since both Spanish and Turkish exhibit DOM, we expect Turkish speakers
Guijarro-
Fuentes 2011
Guijarro-Fuentes & Marinis
). Although we do not include data from English speakers in this study, we
will show that Turkish speakers learning Spanish do not seem to experience the
same level of diculty as reported of English speakers acquiring Spanish DOM
in the existing literature. We provide additional support for the hypothesis that
the L1 plays a crucial role in the initial and intermediate stages of additional
language acquisition as assumed in the Feature Reassembly Hypothesis. We
�ilvina Montrul & Ay
e Gﱲel
䑩nerential object marking in Spanish and in Turkish
Spanish
As shown in (2) and below in (3), in Spanish animate and specic (denite) direct
objects are marked with the prepositio渠‘
, the same as the marker for dative case,
as in
Roberto le dio un libro
Patricia
⢑Roberto gave Patricia a book.’). All other
direct objects such as animate, non-specic (4), inanimate-specic (5), and inani
mate non-specic (6) receive no marking. Deniteness and specicity coincide
in these examples: specic objects are expressed with a denite article and non-
specic objects are expressed with an indenite article. Spanish DOM is mor
phologically instantiated by the preposition ‘
鈠which carries the semantic formal
features [+animate, +specic].
Marina
vio
ujer
nimate, +specic

�arina
saw
9ᄆ
the
woman

�Marina saw the woman.’
Marina vio una mujer
nimate, –specic

�Marina saw a woman (any woman).’
Marina vio la casa
nimate, +specic

�Marina saw the house.’
Marina vio una casa
nimate, –specic

�Marina saw a house.’
According to
Torrego (1998
) properties of the nominal object and of the predi
cate, such as deniteness, specicity, aspect, topicality, agentivity, and aiectedness,
Spanish.
However, in this study, we do not consider features related to
opicality,
Although animate and speci�c/de�nite direct objects must be marked in Spanish, it is
less clear when other objects are marked. Nonspeci�c quanti�ers like
alguien
ₑsomebody’ and
nadie
‘nobody’
always require
酡’
Cono揭
alguien
‘I met somebody’,
No vi
nadie.
ₑI didn’t
see anybody.鈩. It is also possible to occasionally mark inanimate objects for disambiguation
purposes if the subject is also inanimate (
El submarino hundió
l barco
‘�e submarine sank
the ship.鈩. Nonhuman (animal) direct objects exhibit variability with ‘
’ marking depending
on the type of animal (
Mató el
/al
mosquito.
‘He/she killed the mosquito.鈩. Based on a crosslin
guistic comparative study,
Aissen (2003
) notes that in many languages DOM is characterized
by a great deal of apparent fuzziness.‟e exact semantic, syntactic and pragmatic condi
tions regulating when accusative objects should be marked with the preposition ‘
in Spanish,
especially with human inde�nites, are quite complex and not entirely clear in the linguistics
literature (
Leonetti 2004
Torrego 1998
Weissenrieder 1990
Zagona 2002
e acquisition of dierential object marking in Spanish by Turkish speakers
the telicity of the predicate, or agentivity of the subject: we only focus on the
semantic features of the object [+animate,+specic]. Several analyses within gen
erative grammar (
Lidz 2006
López 2012
Rodríguez-Mondoñedo 2007
Torrego
) propose that DOM objects check specicity or accusative case, depending
on the analysis, in a higher functional projection dierent from the projection for
unmarked objects. us, overtly marked objects are structurally more complex
than unmarked objects.
Although a seemingly simple marker, Spanis栠‘
with direct objects presents
signicant challenges for its acquisition. In the rst place ‘
may be dicult to
ceive, especially when the verb ends in the vowel [a] as well, as in the third person
singular present, as in (7), where the two /a/ sounds are practically collapsed in
Marisa llam
Juana

�Marisa calls Juana.’
Marisa llam
Juana
. [oa], [ua]

�Marisa called Juana.’
Llamamo
/llamaro
Juana
. [sa], [na]

�We/they called Juana.’
Note that in some sentences, the presence o映‘
鈠is crucial to understand who is
doing what. Compare the minimal pair in (10).
Llamó Juan

Juan called.’

Llamó a Juan

He/she called Juan.’
�e example in (10a) has a postverbal subject and no object (V–S), whereas (10b)
has the structure S-V-O, except that the subject is a null pronoun (
pro
V-DOM-O).
ὥₑ
in front of
Juan
indicates that
Juan
is the object and not the subject.
In addition to its low acoustic salience in these contexts, ‘
is a polyfunctional
preposition and a case marker. It is used as a locative or directional preposition,
as in (11); it is the dative marker appearing with indirect objects (regardless of
animacy), as in (12), and it is also the dative marker appearing with dative experi
encer subjects in
gustar
-type verbs, as in (13).
�ilvina Montrul & Ay
e Gﱲel
Mario fue a Mexico

�Mario went to Mexico.’
Juan
pididirecciones a la s旱ora

�Juan asked directions to the lady.’
A Pedro le gustla película

�edro-dat liked the movie.

�Pedro liked the movie.’
In all these constructions, the marker ‘
is obligatory, and not subject to ani
macy or deniteness eects with indirect objects and dative subjects. However,
-marking with direct objects looks “optional鐠because it is regulated by several
semantic and pragmatic constraints or features (animacy, deniteness, specicity,
topicality, among many others), as discussed previously. Given the semantic com
plexity and ambiguity with respect to its use, learners need to gure out from the
input how to extract the precise syntactic and semantic conditions that regulate
DOM. DOM presents diculty to learners possibly because it engages multiple
interfaces in the sense of
Sorace (2011
); see also discussion in
Montrul (2011
Turkish
Turkish is a pro-drop Altaic language with SOV as the canonical word order
Erguvanl† 1984
). However, six dierent word order patterns are possible under
certain morphosyntactic, phonological, and pragmatic conditions: OSV, OVS,
SOV, SVO, VOS and VSO. Case plays a large part in this word order exibility.
Subjects in main clauses are not marked (i.e. they are in nominative) but sub
jects of the nominalized embedded clauses are marked with genitive case (
Kornslt
). Direct objects may appear bare or with the accusative case marker –(
depending on the specicity/deniteness of the noun. Turkish does not have a
denite article, but it has a numeral form,
bir
ₑone’, which is generally analyzed as
an indenite article (
Kor渜lt 1997
Lewis 1967
Underhill 1976
). With respect to
marking indirect objects and directions, unlike Spanish which uses ‘
鈠for accu
sative dative and direction, Turkish uses a separate dative case morpheme. For
example, the Turkish sentences corresponding the Spanish sentences in (11) and
(12) require dative case-marked indirect objects. Furthermore, Turkish does not
mark dative experiencer subjects as in (13).
�e examples in (14) illustrate how the accusative case morpheme indicates
the referential property of the direct object and accusative case-word order inter
action (
Aygen 2007
Von Heusinger & Kor渜lt 2005
, p. 5):
Ben
kitap
�ku-du-m.
(incorporated)

ook
ead-
8༔

I did book-reading.’
e acquisition of dierential object marking in Spanish by Turkish speakers

Ben
kita戭6
ku-du-m.
(denite)

ook-
ᘒ4
read-
8༔

I read the book.’

Ben
bir
kitap
ku-du-m.
onspecic, indenite)

ook
ead-
8༔

I read a book.’

Ben
bir
kita戭6
ku-du-m.
pecic, indenite)

ook-
ᘒ4
read-
8༔

I read a certain book.’

�.
kitap
ku-r-um.

st
ook
ᘑ�

�.
kitap
oku-r-um.

ook
st
ead-
ᘑ�

�.
kita戭ı
oku-r-um.

ook-
ᘒ4
fast
�ead-
ᘑ�

I read a book fast.’
As can be seen in the above examples, in Turkish, specicity is marked on object
DPs. All denite DPs are specic (
Enç 1991
⤮‟us, an accusative case marked
object DP is denite, hence specic. If a direct object is in a position other than
the immediately preverbal position, it has to carry the accusative case sux (see
Examples (14e) through (14g)) (
Aygen 2007
Furthermore, all denite NPs such as names, pronouns, denite descriptions,
and demonstrative NPs are obligatorily marked with accusative case irrespective
of animacy (
Enç 1991
, p. 9), as shown in (15).
Zeynep
Ali-yi
da洭6
asa-y6
r-dﰮ
Zeynep
Ali-
ᘒ4
s/he-
ᘒ4
man-
ᘒ4
that
table-
ᘒ4
see-
8༔

�Zeynep saw Ali/her-him/ the man/ that table.’
Turkish is considered to be a DOM language on the basis of accusative case
marking on direct objects (
Aissen 2003
). It has been suggested that in
urkish
DOM depends on specicity and information structure (typically expressed
by word order). Unlike in Spanish, animacy, however, does not play a role in
Turkish DOM.
A topicalized direct object, in most cases, is required to be specic but the
presence of accusative case alone may not be sucient to make the direct object
specic enough to qualify as a topic. Consider the examples in (16) from
Von
Heusinger and Kor渜lt (2005
, p. 12):
*Bir
kita戭6
urat
aceleyle
ku-yor

ook-
ᘒ4
Murat
�urriedly
read-
᜕0m

�ntended meaning: ‘Murat is hurriedly reading a /some book.’
�ilvina Montrul & Ay
e Gﱲel

Mavi
kapl6
ir
kita戭6
urat
aceleyle
ku-yor

blue
�vered
ook-
ᘒ4
Murat
�urriedly
read-
᜕0m

�ntended meaning: ‘Murat is hurriedly reading a certain blue-covered book.’
With respect to the idea that the function of DOM is to disambiguate subject from
object, the prediction is that atypical subjects (i.e. inanimate, indenite, nonspecic)
are to be marked as the mirror image of DOM. Turkish, however, marks specic
subjects only, as in (17a), and does not mark nonspecic ones, as in (17b). Note the
genitive-marked embedded subjects below (
Von Heusinger & Kor渜lt 2005
, p. 15):
Kö礭ü
ir
haydut-un
as-t6ር6
-n6
duy-du-m

Village-
ᘒ4
�obber-
raid-
ᘒ4
hear-
8༔

I heard that a certain robber raided the village.’

Kö礭ü
aydut
bas-t6ር6
-n6
uy-
du-m

Village-
ᘒ4
robber
�id-
ᘒ4
hear-
8༔

I heard that robbers raided the village.’
In sum, as illustrated in the above examples, DOM in Spanish and in Turkish
panish,
DOM marks animacy and specicity and carries the features [+animate, +
pecic],
whereas in Turkish DOM marks specicity [+specic] and is blind to animacy. In
other words, among the factors that govern DOM, Turkish is said to be sensi
tive only to referential features (specicity/deniteness) and information structure
(word order and topicality), while animacy does not restrict the use of the case suf
Von Heusinger & Kor渜lt 2005
, p. 13). Our study does not include examples
testing topicality, which is relevant in Spanish and Turkish, and for this reason we
do not represent topicality in Table 1 below⸠�e properties of DOM that are the
focus of our study are summarized in Table 1.
Table 1.
Feature specication of Spanish and Turkish DOM
DOM
Morpholexical form
Formal features
Spanish
+animate, +specic/denite
Turkish
+specic/denite
Indeed, as noted by
Von Heusinger and Kor渜lt (2005
), the semantic and
e acquisition of dierential object marking in Spanish by Turkish speakers
Previous studies on the acquisition of dierential object marking
in Spanish
�e only available published study of Spanish DOM in L1 acquisition is
Rodríguez-Mondoñedo (2008
), who conducted an analysis of the spontaneous
production of four Spanish-speaking children (ages 0;9 – 2;11) from the
data base. All sentences containing V–O structures were analyzed. From a total of
991 exemplars of animate and inanimate objects the children made a total of 17
errors – 98.38% accuracy before age 3 – as
Rodríguez-Mondoñedo (2008
) reports.
But when accuracy with DOM is taken into account by object, it is 85% for ani
mate (45/53) and 99% (929/937) for inanimate objects. is study suggests that
Spanish-speaking children acquire the semantic constraints on DOM very early,
at least with core cases.
Research on early bilingualism, especially of child and adult Spanish heritage
speakers in the United States has shown that DOM is not fully acquired in bilin
gual school-age children and young adult heritage speakers, and it is even subject
to L1 attrition in rst generation immigrants (
Montrul 2004
Montrul & Bowles
Montrul & Sánchez-Walker 2013
Ticio in press
Grosjean and Py (1991
and
Girard (1995
) (both reported in
Grosjean 2008
) also found that rst and sec
ond generation Spanish immigrants in French-speaking Switzerland showed ero
sion of Spanish DOM.
Spanish DOM seems to be even more problematic in L2 acquisition. DOM
takes a long time to be mastered by adult English-speaking L2 learners of Spanish,
whose L1 does not mark DOM, despite its frequency in the L2 input
(Bowles &
Montrul 2009; Guijarro-Fuentes & Marinis 2007; Guijarro-Fuentes 2011, 2012)
Guijarro-Fuentes (2011
) tested English-speaking learners of Spanish rang
�ilvina Montrul & Ay
e Gﱲel
have DOM, such as Brazilian Portuguese (
Halloran & Giancaspro 2012
Montrul,
Dias & Santos 2011
) and Catalan (
Perp槱án 2013
). What no study has investigated
ὥ study
Our study investigates the acquisition of Spanish DOM by Turkish speakers, whose
language also manifests DOM in a dierent way. Assuming the Full Transfer/
Full Access Hypothesis (FT/FA) (
Schwartz & Sprouse 1996
), we predict that
Turkish speakers will have little diculty with DOM in Spanish even at earlier
stages of interlanguage development. We follow
Guijarro-Fuentes (2011
assuming that
Lardiere’s (2009
Feature Reassembly Hypothesis
(FRH) provides a
e acquisition of dierential object marking in Spanish by Turkish speakers
direct objects. In Turkish, the feature [+specic] is bundled in the accusative case
marker (
, but the feature [+animate] is not. us Spanish DOM bundles two
features in the lexical item ‘
’ [+animate, +specic] whereas Turkish -(
bundles
only one [+specic]. However, when ‘
appears with indirect objects and dative
experiencer subjects, it carries the feature [+dative], since it is a marker of dative
Participants
�e participants in our study were a group of native speakers of Spanish tested in
Mexico (
= 20) and a group of Turkish speakers learning Spanish in Istanbul,
Turkey (
= 32). e Mexican native speakers (7 male, 13 female), ages 18–25
(mean 21.05) were attending the Universidad de Guanajuato⸠�ey knew some
English as a L2, with age of acquisition ranging from 12 to 24 years.
�e Turkish speakers were university students of similar age tested in
stanbul,
Turkey. All of them knew English and three of them knew German as well. is
makes Spanish a third language rather than a second. Because English is an oblig
atory foreign language taught in Turkish schools (from primary through high
school), it is not possible to nd L1-Turkish speakers learning Spanish as a foreign/
second language prior to English.
�ese Turkish participants all began learning Spanish a1er puberty (mean
19.2, range 14–21). ey have been exposed to Spanish for less than one year to
5 years with a mean length of exposure of 2.43 years. One of the Spanish instruc
tors reported that the construction under investigation is not explicitly taught in
Spanish classes.
�e native speakers and the Turkish speakers were administered a writ
ten prociency test in Spanish, consisting of a cloze passage (with three
multiple-choice response options for each blank) from a version of the
Diploma
de
�spañol como Lengua Extranjera
(DELE) and a multiple choice vocabulary
�ilvina Montrul & Ay
e Gﱲel
part from an MLA placement test. e maximum possible score on the pros
ciency test (both sections combined) was 50. Basic information about the two
groups is shown in Table 2.
Table 2.
Information about the participants
Groups
Mean age/range
Mean/range of AoA
Spanish
Spanish proiciency score
Spanish native
speakers
Turkish speakers
learning Spanish
�e native speakers and the Turkish speakers did not dier in age (
(51)㴚0.908,
= 0.34), but they diered in prociency (
0.0001). e Turkish
group was divided into two prociency levels for the data analysis: low intermedi
ate (
= 18) with scores ranging from 16–30, mean 24.05 (SD 4.16) and intermedi
ate (
= 14) with scores ranging from 31–42, mean 34.5 (SD 3.58).
Tasks
Our study included three tasks: a written production task, a written comprehen
sion task, and a bimodal acceptability judgment task.
se written production task
�e written production task (WPT) was used in
Montrul and Bowles (2010
classroom-
Participants were given three words – a noun, a verb in the innitive, and another
animate direct object

�rompt:
estudiante/
visitar/
profesora

tudent
isit
rofessor

�rammatical response:
El estudiante visi瓳
la profesora

‘�e student visited the professor.’
e acquisition of dierential object marking in Spanish by Turkish speakers
inanimate direct object

�rompt:
Patricio/
visitar/
Museo
del

atricio
isit
Museum

�rammatical response:
Patricio visitel Museo del Prado

‘Patricio visited the del Prado Museum.’
indirect object

�rompt:
Armando/
enviar/
amiga/
ṯres

rmando
end
riend
ṯwers

�rammatical response:
Armando
enviiores
su amiga

Armando sent owers to her friend.’
Gustar
-type verbs with animate dative experiencer

�rompt:
Juan/
gustar/
Patricia

uan
like
atricia

�rammatical response:
Juan le gusta Patricia/ Patricia le gusta
Juan

Juan likes Patricia.’
Gustar
-type verbs with inanimate dative experiencer

�rompt:
Francisco/
gustar/
música
rock

rancisco
like
ock
usic

�rammatical response:

Francisco le gusta la música rock/La música rock
le gusta
Francisco

Francisco likes rock music.’
�e objective of this task was to see if the L2 learners incorrectly omi琠‘
with
human or animate specic direct objects in written production, when they sup
posedly have time to compose and write their responses. We also wanted to inves
se written comprehension task
pictureA, Juan is calling somebody by phone, in picture B Juan is receiving a call,
�ilvina Montrul & Ay
e Gﱲel
and in picture C Mary and Jane are calling Juan on the phone. At the same time,
participants saw one sentence at a time, such as
Llamó a Juan
ₑHe/she called Juan.鈩
and had to indicate which picture matched the sentence, by pressing the A, B or
C on the keyboard. In this case, B is correct. A1er the participant made a choice,
the responses disappeared from the computer screen: the survey was programmed
so that it was not possible for participants to compare sentences and go back and
change answers.
�e same pictures appeared two more times, with two other sentences
Lla淳
Juan
ₑJuan called.鈠and
Llamaron
a Juan
ₑ�ey called Juan.鈠If L2 learners do not
assign meaning to the DOM marker, they will be more accurate on sentences with
V–S order than on V–O order. But if L2 learners also have diculty with postver
bal subjects in V陓 sentences, then they will be equally inaccurate in V–O and V陓
sentences. (Sentences with postverbal subjects were also tested in the acceptability
Llamó Juan

�Juan called.’
Llamó a Juan

�She called Juan.’
Llamaron a Juan

��ey called Juan.’
Figure 1.
Sample picture and sentences used in the aural/written comprehension task.
Accusative condition
se acceptability judgment task
�e acceptability judgment task presented grammatical and ungrammatical sen
tences wit栠‘
鈠marking and omission with dierent predicates. e task included
140 sentences (75 grammatical, 65 ungrammatical) divided into 28 types, wit栠sve
e acquisition of dierential object marking in Spanish by Turkish speakers
token sentences per type, depending on the structure. Fourteen sentence types are
the focus of the study and the rest wer攠sllers and control structures that we will
not discuss. We included sentences with animate/inanimate, denite/
denite
direct objects, indirect objects and dative subjects, where the presence and omis
sion of the prepositio渠‘
ten
�panish and with an audio player below, with bimodal stimulus presentation.
Participants were instructed to read each sentence and play the soun搠sle before
Table 3.
�nimate direct objects

te with DOM
Marina vio a Madonna
‘Marina saw DOM Madonna.鈠2

b.
te with no DOM
Julia vio Shakira
‘Julia saw Shakira.’

�denite with DOM (optional)
Mi abuelo conoció a unos pintores
‘My grandfather knew DOM some painters.’

d.
�denite with no DOM
Antonia vio una gitana
‘Antonia saw a gipsy.’
Continued

�reguntó Susana

�Susana asked.’
Preguntó a Susana

�She asked Susana.’
Preguntaron a Susana

��ey asked Susana.’
Figure 2.
Sample picture and sentences used in the aural/written comprehension task. Dative
condition
�ilvina Montrul & Ay
e Gﱲel
�nimate direct objects

te with DOM
Josefa eligial auto rojo
‘Josefa chose DOM the red car.’

b.
te with no DOM
Ignacio visitó la universidad
‘Ignacio visited the university.’

�denite with DOM
Adriana siguió a unas direcciones
‘Adriana followed DOM some directions.’

d.
�denite with no DOM
Pablo escuchun concierto
‘Pablo listened to a concert.’
�ther sentences

�ostverbal subject
Lla淳 Ignacio.
‘Ignacio called.’

�.
�ronominal subject and no DOM
Él invitJuan
‘He invited Juan.”
�e three tasks were administered through a web-based program called Sur
vey Gizmo⸠�e order of administration of the tasks did not follow a strict pre
sentation order and varied by participants. e 32 Turkish L2 learners of Spanish
Results
�e Written Production Task
Overall accuracy on the production task was high: the mean for the native speak
ers was 98.7%, 85.4% for the 8 intermediate speakers and 69.5% for the 15 low
prociency speakers. Figure 3 shows the results by sentence types.
We ran a repeated measures ANOVA with sentence types as within-subjects
Table 3.
(Continued)
e acquisition of dierential object marking in Spanish by Turkish speakers
of native speakers. Table 4 shows the number and percentage of Turkish-speaking
Table 4.
Written Production Task. Individual participants and percentage per group who
scored above 60% with each sentence type
L2 groups
Human
Inanimate
Indirect
object
Psych
animate
Psych
inanimate
Turkish interm
Turkish low
We see that all the intermediate speakers (8 of 8 or 100%) knew that human
objects tak攠‘
鈠and inanimate objects do not. More than 60% of the sample of low
intermediate participants (
=15) were accurate with human and inanimate direct
objects, but about 40% omitte搠‘
鈠with human objects and overgeneralize搠‘
to inanimate objects. With the exception of 2 low intermediate participants, the
Figure 3.
Mean accuracy in the written production task by verb and object type
�ilvina Montrul & Ay
e Gﱲel
�e written comprehension task
Figure 4 shows the results of sentences with direct objects marked with DOM,
postverbal subjects, and plural sentences. We rst ran an ANOVA with repeated
measures contrasting the three sentence types with direct objects by group.
�ere was a main eect for sentences (
.0001), for group
(2,39)㴚4.21
= .022), and a sentence by group interaction (
(2,39) = 3.29,
2㴚.004). e native speakers did not dier from the intermediate level speakers,
according to Tukey post hoc tests (
.10), but the group dierence was due to
the fact that the low intermediate Turkish speakers were signicantly 摩ierent
from the native speakers (
.017). All participants were very accurate with sen
tences with DOM marking, but had diculty with VS sentences, suggesting that
they know that Spanish is a DOM language but they may not know that Spanish
allows postverbal subjects.
Figure 4.
Mean accuracy on the written comprehension task. Sentences with direct objects
Figure 5 shows the results of sentences with indirect objects. A repeated mea
sures ANOVA showed no signicant main 攝ect for sentences and no sentence by
group interaction, although the dierence by group was signicant (
.028). All the Turkish-speaking learners knew that indirect objects in Spanish
are marked with the prepositio渠‘
⸠�e low intermediate learners also had some
diculty with sentences with postverbal subjects.
�ere were 7 individuals in the two Turkish-speaking groups who scored
lower than 70% on sentences with postverbal subjects in this task, but all partici
e acquisition of dierential object marking in Spanish by Turkish speakers
�e bimodal acceptability judgment task
In this task we asked participants to express acceptability judgments on a 4-point
scale on a list of grammatical and ungrammatical sentences. A rating of 4 meant
Figure 5.
Mean Accuracy on the Written Comprehension task. Sentences with indirect objects
Figure 6.
Mean acceptability ratings on human direct objects by group
�ilvina Montrul & Ay
e Gﱲel
results of sentences with human, denite and indenite direct objects. Sentences
with human, denite and specic direct objects require DOM human, denite
objects without DOM are ungrammatical. Sentences with indenite human objects
e acquisition of dierential object marking in Spanish by Turkish speakers
prociency Turkish speakers rated inanimate objects with DOM more acceptable
than the other two groups (
Figure 7.
Mean acceptability ratings on inanimate direct objects by group
Figure 8.
Mean acceptability ratings on postverbal subjects (VS) and sentences with
pronominal subjects and unmarked human objects (Pron VO)
Figure 8 shows the results of sentences with postverbal subjects (VS) (gram
matical) and sentences with pronominal subjects and unmarked human objects
(ungrammatical). Sentences with postverbal subjects were prone to errors by
the Turkish speakers in the comprehension task. In this task as well, there were
�ilvina Montrul & Ay
e Gﱲel
.0001) and with
pron VO
sentences (
.0001).
�e low intermediate and the intermediate Turkish speakers did not dier from each
other on their ratings of VS sentences (
.5), but the intermediate level speak
ers, who did not dier from the native speakers, were more accurate at rejecting
ungrammatical sentences without DOM than the low level speakers (
.0001).
To summarize, the results of our three tasks showed that low and intermedi
ate prociency Turkish-speaking learners of Spanish know tha琠‘
鈠is required with
animate, specic direct objects in Spanish, as shown in the production, compre
hension and acceptability judgment tasks. Some participants had diculty inter
Discussion and conclusion
Existing studies on the acquisition of dierential object marking in Spanish sug
gest that DOM is very dicult to acquire in a second language when the ᱲst
language of the learners does not mark DOM (
Bowles & Montrul 2009
Guijarro-
Fuentes 2011
Turkish and Hebrew the dative is not the source of DOM (
Aissen 2003
According to
Aissen (2003
, p. 437) there is room for language particular varia
tion among languages that exhibit DOM. at is, even when DOM exists in two
languages, the dimensions that dene DOM and the cut oᴠpoint for DOM may
vary. In Spanish, animacy and specicity are the main dimensions that regulate
e acquisition of dierential object marking in Spanish by Turkish speakers
We assumed the Feature Reassembly Approach (
Lardiere 2009
), according
to which L2 learners must gure out how to bundle features in lexical items in
their L2. In Spanish, DOM has two main features [+animate, +specic] whereas
in Turkish DOM has only one feature [+specic]. Animacy and specicity can
be bundled in a DP. Animacy is lexically encoded, whereas specicity is not lexi
cally encoded in nouns and can be obtained from contextual information and
鈠is required with animate, specic direct objects in
panish.
More than 60% of the participants produced the DOM marker with 100% accu
racy in the production task. ey were also above 80% accurate in the compre
hension task, where they had to discriminate DOM marked animate and specic
objects from postverbal subjects. Although we did not test topicality in this study,
perhaps Turkish speaking learners of Spanish may have also noticed that the fea
ture topicality is relevant in Spanish DOM as it is in Turkish, which may have
also contributed to their initial high performance in Spanish.
Guijarro-Fuentes
) found that all the English-speaking learners tested in his study gave similar
ratings to grammatical and ungrammatical sentences with DOM, regardless of
prociency level⸠�ere was some evidence in his studies that the English-speaking
learners had developed some sensitivity to Spanis栠‘
鈠as a marker of animacy
with direct objects, but had not bundled specicity⸠�ey correctly accepted gram
matical sentences but were very inaccurate rejecting ungrammatical sentences
with no DOM. By contrast, we found in our study that our low intermediate and
intermediate level learners of Spanish whose ᱲst language is Turkish gave statisti
cally dierent ratings to grammatical and ungrammatical sentences with DOM
task. is may be due to the fact that Spanish does not overtly mark
ominative
�ilvina Montrul & Ay
e Gﱲel
acceptability judgment task may support this hypothesis. �at is, some
urkish
speakers may still analyze Spanish DOM as a marker of specicity only as in
e acquisition of dierential object marking in Spanish by Turkish speakers
References
Aissen, J. (2003). Dierential object marking: Iconicity vs. economy.
Natural Language and Lin
guistic“eory,
Aygen, G. (2002).
Finiteness, Case and Clausal Architecture
. Unpublished Ph.D. dissertation,
Harvard University.
Aygen, G. (2007). Specicity and subject-object positions/scope interactions in Turkish.
Dil ve
Edebiyat Dergisi/Journal of Linguistics and Literature
Bardel, C., & Falk, Y. (2007). e role of the second language in third language acquisition: e
case of Germanic syntax.
Second Language Research,
DOI: 10.1177/0267658307080557
Bossong, G. (1991). Dierential object marking in Romance and beyond. In D. Wanner &
D⸚Kibbee (Eds.),
New Analyses in Romance Linguistics. Selected Papers from the XVIII Lin
guistic Symposium on Romance Languages
(pp.143–170). Amsterdam: John Benjamins.
DOI: 10.1075/cilt.69.14bos
Bowles, M., & Montrul, S. (2009). Instructed L2 acquisition of dierential object marking in
Spanish. In R. P. Leow, H. Campos, & D. Lardiere (Eds.),
Little Words
seir History, Pho
nology, Syntax, Semantics, Pragmatics and Acquisition
(pp. 199–210). Washington DC:
DOI: 10.1017/s1366728912000144
Guijarro-Fuentes, P., & Marinis, T. (2007). Acquiring the syntax/semantic interface in L2
Spanish: �e personal preposition ‘a’. In L. Roberts, A. Gﱲel, S. Tatar, & L. Mart蘠(Eds.),
Eurosla Yearbook 7
(pp. 67–87). Amsterdam: John Benjamins.
DOI: 10.1075/eurosla.7.06gui
Guijarro-Fuentes, P., & Marinis, T. (2011). Voicing language dominance: e acquisition of
interfaces by English/Spanish bilingual adolescents. In K. Potowski & J. Rothman (Eds.),
Bilingual Youth: Spanish in English-speaking Societies
(pp. 227–248). Amsterdam: John
Benjamins.
DOI: 10.1075/sibil.42.14gui
Halloran, B., & Giancaspro, D. (2012, October). Examining L3 transfer models: e acquisition
of dierential object marking in L3 Brazilian Portuguese. Poster presented at GALANA,
University of Kansas, Lawrence.
Kor渜lt, J. (1997).
Turkish
. Descriptive Grammars Series. London: Routledge.
�ilvina Montrul & Ay
e Gﱲel
Kor渜lt, J. (2003). Scrambling, subscrambling, and case in Turkish. In S. Karimi (Ed.),
Word
Order and Scrambling
, (pp. 125–155). Malden, MA: Blackwell.
DOI: 10.1002/9780470758403.ch6
Kural, M. (1997). Postverbal constituents in Turkish and the linear correspondence axiom.
Lin
guistic Inquiry, 28
DOI: 10.1353/lan.2006.0054
López, L. (2012).
Ind攞nite Objects. Scrambling, Choice Functions and Di�erential Marking
Lin
guistic Inquiry Monograph
. Cambridge, MA: e MIT Press.
DOI: 10.1017/s0022226713000376
Montrul, S. (2004). Subject and object expression in Spanish heritage speakers: A case of
morpho-syntactic convergence.
Bilingualism: Language and Cognition,
DOI: 10.1017/s1366728904001464
Montrul, S., & Sánchez-Walker, N. (2013). Dierential object marking in child and adult
panish
heritage speakers.
Language Acquisition,
DOI: 10.1080/10489223.2013.766741
Montrul, S., Dias, R., & Santos, H. (2011). Clitics and object expression in the L3 acquisition of
Brazilian Portuguese. Structural similarity matters for transfer.
Second Language Research,
DOI: 10.1177/0267658310386649
Pensado, C. (1995). El complemento directo preposicional: Estado de la cuest槳n y bibliografía
comentada. In C. Pensado (Ed.).
El complemento directo preposicional
(pp. 11–59). Madrid:
Visor.
DOI: 10.2307/474952
Perp槱án, S. (2013, April). Direct object expression and its semantic properties in Catalan-
Spanish bilingualism. 43rd
Linguistics Symposium on Romance Languages
, City University
of New York (CUNY Graduate Center, April 19).
Rodríguez-Mondoñedo, M. (2007).
se Syntax of Objects: Agree and Di�erential Object Mark
ing.
Unpublished Ph.D. dissertation. University of Connecticut.
Rodríguez-Mondoñedo, M. (2008). e acquisition of dierential object marking in Spanish.
Probus, 20
DOI: 10.1515/probus.2008.004
e acquisition of dierential object marking in Spanish by Turkish speakers
Schwartz, B.D., & Sprouse, R.A. (1996). L2 cognitive states and the full transfer/full access
model.
Second Language Research, 12
DOI: 10.1177/026765839601200103
Sorace, A. (2011). Pinning down the concept o映“interface” in bilingualism.
Linguistic Approaches
to Bilingualism, 1
DOI: 10.1075/lab.1.1.01sor
Ticio. E. (in press). Dierential object marking in Spanish-English early bilinguals.
Linguistic
Approaches to Bilingualism
Torrego, E. (1998).
se Dependency of Objects
. Cambridge, MA: e MIT Press.
DOI: 10.1017/s0022226799298088
Underhill, R. (1976).
Turkish Grammar
. Cambridge: MA: e MIT Press.
DOI: 10.1017/s0041977x0010864x
Von Heusinger, K. (2005). e evolution of Dierential Object Marking in Spanish. In K. Von
Heusinger, G. Kaiser, & E. Stark (Eds.),
Proceedings of the Workshop Speci�city and the
Evolution/Emergence of Nominal Determination Systems in Romance.
Fachbereich Sprach
wissenscha der Universität Konstanz, Arbeitspapier Nr. 119. (pp. 33–70). Konstanz:
University of Konstanz.
Von Heusinger, K. (2008). Verbal semantics and the diachronic development of DOM in
Spanish.
Probus,
DOI: 10.1515/probus.2008.001
Von Heusinger, K., & Kor渜lt, J. (2005). e case of the direct object in Turkish: Semantics,
syntax and morphology.
Turkic Languages, 9
Weissenrieder, M. (1990). Variable uses of the direct-object marker A.
Hispania, 73
DOI: 10.2307/343010
Zagona, K. (2002).
se Syntax of Spanish
. Cambridge: CUP.
DOI: 10.1017/cbo9780511613234
Copula choice in adjectival constructions
in Dutch L1 Spanish L2
Manuela Pinto
& Alexia Guerra Rivera
Utrecht University
/ Universidad Austral de Chile
L2 learners of Spanish have diculties choosing the right copula in
ser/estar
+ adjective constructions. Beyond the eect of processing load of
information pertaining to dierent cognitive modules (
Sorace’s (2011
Interface Hypothesis), this study aims at pinpointing the specic properties
Introduction
Ὡs paper aims to contribute to the debate on the accessibility of inherent
semantic properties in L2 acquisition by examining a specic construction of
Spanish grammar, the predicative construction formed by the copula
ser
or
estar
and an adjective⸠�e majority of Spanish adjectives can take
ser
as well as
estar
guals, Heritage
peakers, and L2 learners with English L1 show instead persistent
optionality in the choice of the right copula (Bruhn de Garavito & Valenzuela
2008; among others). e problem may be ascribed to the interface nature of
these constructions:
tegration of information pertaining to 摩ierent cogni
tive domains implies a higher processing load, making these areas of second
�anuela Pinto & Alexia Guerra Rivera
panish copula + adjective constructions and pinpoint their
vulnerable aspects in the L2 acquisition process. e experimental data presented
here show that Dutch L1-Spanish L2ers do not have problems with all ambiguous
copula + ADJ constructions, but just with a subclass of adjectives, the so-called
irreversible scalar gradable
adjectives. ese ndings suggest that L2ers may have
selective access to some inherent semantic properties of the copula.
Ὡs paper is organized as follows: the next section oiers a description of
adjectival constructions in Spanish and gives a brief review of how these construc
tions have been examined in syntactic and in semantic terms. e distribution
of the two Spanish copulas in adjectival constructions is also compared to Dutch
copulas
zijn
and
worden
. In Section 3 we comment on some studies on L2 acquisi
tion of
ser/estar
, focusing on ambiguous constructions. In Section 4 we present the
Ser
and
estar
with adjectives
Spanish has two copular verbs,
ser
and
estar
, whereas a language like English, has
only one form:
. In Spanish, copulas link noun phrases to predicate structures,
noun phrases, adjective phrases, and prepositional phrases. is can be seen in
Examples (1) to (5) below.
Miguel
es
studiante.
Miguel
ser-3
student

�Miguel is a student.’
opula choice in adjectival constructions in Dutch L1 Spanish L2
Miguel
e獴á
nfermo.
Miguel
estar-3
enfermo-
mᘏ

�Miguel is sick.’
Miguel
es
ubio.
Miguel
ser-3
rubio-
mᘏ

�Miguel is blond.’
Viviana
está
Chile.
Viviana
estar-3
�hile

�Viviana is in Chile.’
La
�esta
casa
María.

�e
party
ser-3
at
�ouse
of
aria

��e party is at Maria’s house.’
In Example (1), the referent Miguel is linked to the NP
estudiante
ₑstudent’. In
Examples (2) and (3), Miguel is linked to the AP
enfermo
ₑsick’. In Examples (4)
and in (5), the subjects Viviana and
�esta
are linked to the P倠‘in Chile鈠an搠‘at
María’s house鈠respectively. As the examples show, in all these cases English uses
is
(from
Mi primo es delgado.
(with
ser

�My cousin is slim.’
Mi primo e獴á delgado.
(with
estar

�My cousin is slim (nowadays, or than before).’
In Example (6),
ser
refers to properties that are relatively stable, and that dene
characteristics at an individual level, whereas
estar
, in Example (7), denotes a more
temporary, changing property. Both sentences translate into English a猠‘my cousin
�anuela Pinto & Alexia Guerra Rivera
Ser
and
estar
opula choice in adjectival constructions in Dutch L1 Spanish L2
of referents having a minimum amount of a given property. In
Camacho (2013
),
L2 acquisition of
Ser
and
Estar
A majority of the studies on copula choice in Spanish L2 have been carried out
students pass:
�bsence of copula in learner speech.
b.
�election of
ser
to perform most copula functions.
�ppearance of
estar
with progressive.
d.
�ppearance of
estar
with locative
e.
�ppearance of
estar
with adjectives of condition.
STUDIES that followed examined two variables, L2 prociency and L1 interference.
Geeslin (2001
) and
Geeslin and Guijarro-Fuentes (2006
) tested
�glish-speaking
learners of Spanish at 摩ierent prociency levels (from beginner to intermediate)
and found that prociency positively aiected the use of
estar
; although for
ser
it was
�anuela Pinto & Alexia Guerra Rivera
In another study
Geeslin and Guijarro-Fuentes (2005
) examined the eects of
a number of factors on the acquisition of Spanish
ser
and
estar
. One of these factors
was the linguistic background of the participants. is variable was considered
erman,
English) played a role in Spanish L2 copula choice. Crucially, none of the L1 lan
guages had a similar copula system as the Spanish one⸠�e authors claimed that
the L1 did not play a relevant role in the acquisition of the Spanish copulas, as the
frequency of use of
ser/estar
and their sample did not vary across the dierent L1s.
More recently, research on Spanish copulas in L2 acquisition has s桩1ed the
focus of interest to the aspectual properties of copulas, and to how L2 learners
(henceforth L2ers) deal with these properties.
Bruhn de Garavito and Valenzuela
) investigated the acquisition of
ser
and
estar
in verbal and adjectival passives
and in individual and stage level predicates by two dierent L2 populations: early
and late bilinguals (English L1 and Spanish L2). �e main aim of this study was
problems incorporating pragmatic cues into the copula choice⸠�e
ulties
opula choice in adjectival constructions in Dutch L1 Spanish L2
with the Spanish copulas seem to increase in specic grammatical contexts,
namely, adjectival phrases and locatives. L2 prociency seems to positively cor
relate with the use of
estar
, but gives unclear results for
ser
. As with regard to
the possible eects of the L1,
Geeslin and Guijarro-Fuentes (2005
) argued that,
on a whole, L2 acquisition of
ser/estar
is not inuenced by previously learned
languages.
Ser
estar
in Dutch L1 Spanish L2
To the best of our knowledge,
Hanegreefs (2004
) is the only publication that
addresses the aspectual properties of Dutch copulas. In a contrastive corpus
analysis of the dierent possible translations in Spanish of the Dutch copula
worden
�anegreefs shows that the choice of the Spanish copula is related to a
摩ierent conceptualization of the process of change. Referring to
Delbecque,
Masschelein and Vanden Bulcke (1995
), she argues that
worden
, when used as a
copula, is normally related to a concept of change.
Worden
can also have the function of auxiliary in compound tenses and in passives.
�anuela Pinto & Alexia Guerra Rivera
Table 1.
Aspectual properties of copulas in Dutch and in Spanish
Dutch
Spanish
no aspectual properties
Ser
change of stat�e process
WORDEN (present perfect)
Estar
change of stat�e result
Zijn
Estar
Worden
estar
As Table 1 shows, both Dutch
worden
and Spanish
estar
are assumed to have
inherent aspectual properties: they are both marked for [state]. ey both surface
as a copula containing this specic aspectual feature that has to match the seman
tic and pragmatic features of the adjective. is kind of information is also avail
able in the input and indeed L2ers do not have problems with
ser
and
estar
when
the adjective clearly requires just one of the two copulas. �e problems for L2ers
seem to involve constructions in which adjectives may take either
ser
or
estar
In the sense of Lardiere (2000).
opula choice in adjectival constructions in Dutch L1 Spanish L2
In this section we report of two experiments we ran with Dutch L1 Spanish L2
speakers with an advanced prociency level⸠�e rst experiment, a Grammatical
peaking country on
Following common practice of the sources we relied on, we did not take any independent
measure of the prociency level of the control group. However, we agree with one of the
reviewers that this may give a more reliable baseline measure.
�anuela Pinto & Alexia Guerra Rivera
version of the
Diploma de Español como Lengua Extranjera
(DELE), and consisted
of a cloze passage with three multiple choice options for each blank, as well as a
multiple choice vocabulary section, adapted from a Modern Language Association
Task 1: Grammaticality Judgment Task
�e Grammaticality Judgment Task was a pen-and-paper task. �e task included
thirty six Spanish sentences. Twelve sentences contained
ser
and twelve
ntained
Table 2.
Linguistic background questionnaire of Dutch L1 Spanish L2 subjects
Participant ID
Age in years
Proiciency
score (max. 50)
Other Foreign
languages
Age started
learning Spanish
Time spent
abroad (Spain or
South America)
English-German
2 months
English-German-French
6 months
English-German-French
7 months
English-German-French
1 months
English-Catalan-French
9 months
English-French-German
12 months
English-French
14 months
English-French
12 months
English-German
12 months
English-German-French
8 months
English-German-French
8 months
English
24 months
English-German-French
18 months
English-German-French
8 months
opula choice in adjectival constructions in Dutch L1 Spanish L2
estar
, equally distributed over 12 grammatical sentences and 12 ungrammati
cal ones. Moreover, 6 grammatical and 6 ungrammatica氠sller sentences were
included and randomly distributed through the task. e participants were asked
to judge the grammaticality of each sentence on a Likert scale, indicating 1 when
totally unacceptable and 5 when totally acceptable⸠�e test sentences were divided
into four dierent types. Each type of sentence appeared six times during the task.
�e dierent sentences types used in this task are described below:
Ser
+ AP (grammatical):
El hijo de María
es cruel
‘María’s son is cruel.’
Ser
+ AP (ungrammatical):
*Los nuevos empleados
están muy capaces
‘�e new workers are very capable.’
Estar
+AP (grammatical):
El vaso
e獴á lleno
de agua.
‘�e glass is full of water.’
Estar
+ AP (ungrammatical):
*El entrenador de fútbol
es satisfecho
‘�e football trainer is sati猜ed.’
�e adjectives used in this task were of two types. (1) adjectives that unambigu
ously combine with one specic copula. For example,
satisfecho
ₑsati猜ed’,
lleno
‘full鈠always select
estar
capaz
ₑhand禒,
cruel
ₑcruel鈠always select
ser
. (2) adjectives
that can be used with both copulas, like
rubio
ₑblond’,
iaco
ₑthin’. In this case, how
ever, we manipulated the context in such a way that only one of the two possible
Results of the GJT
A Multivariate Analysis of Variance (MANOVA) statistical test conducted on the
average number of responses per condition shows that a signicant dierence in
the performance of the two groups is only present in one of the tested conditions,
namely, ‘
estar
ungrammatical’. �e L1 group performed similarly to the L2 group
in the three other conditions and only signicantly dierent in this specic condi
tion. In addition, the L2 group had more troubles in judging ungrammatical test
sentences than grammatical ones; see Figure 1 below.
�anuela Pinto & Alexia Guerra Rivera
Discussion of the GJT
�e picture that emerges from this study co渜rms that integration of morpho-
syntactic information with semantics and pragmatics may be one of the major
factors aiecting the acquisition of Spanish copulas, in line with Sorace’s Interface
Task 2: Fill-in-the-Gap Task
�e next step is an analysis of the inherent semantics of Dutch copulas compared
to
ser
and
estar
. Two hypotheses are being tested: (1) Inherent aspectual proper
ties of copulas may not be visible in the input that L2ers receive. If this is
rrect,
Estar – GRAMEstar–UNGRAMSer–GRAMSer–UNGRAM
Grammaticality judgment task
Mean score
Groups
Figure 1.
Mean scores of correct responses
opula choice in adjectival constructions in Dutch L1 Spanish L2
Dutch L1 Spanish L2ers may fail to see that
worden
diers from
estar
in just one
aspectual property. is would predict a superᱣial transfer of Dutch
worden
for all cases of Spanish
estar
⸠�e other hypothesis says instead that (2) inher
ent aspectual properties of copulas may be available to L2ers. In this case Dutch
Fill-in-the-Gap Task
Table 3.
scalar gradable
non-scalar gradable
adjectives
Scalar gradable
Non-scalar gradable
Flaco (thin)
Romántico (romantic)
Largo (long)
Elegante (elegant)
Grande (tall)
Pelirroja (red haired)
Viejo (old)
Revoltoso (naughty)
Gordo (fat)
Simpático (nice)
�anuela Pinto & Alexia Guerra Rivera
Each sentence was introduced by a short story that provided contextual infor
mation eliminating ambiguity:
El abuel漠Ángel siempre ha sido delgado y eso no ha cambiado incluso cuándo
se ha casado con Maria, quien es una gran cocinera

�Grandp愠Ángel has always been thin and he hasn
t changed, not even when
he married Maria, who is a great cook.’
Results of the Fill-in-the-Gap Task
An independent
-test revealed no signicant dierence in the performance of the
two groups:
Ser Sc. GradSer Non Sc GradEstar Sc. GradEstar Non Sc. Grad
Fill in the gap task
Mean score
Groups
Figure 2.
L1 and L2 mean-scores in terms of correct vs. incorrect answers
Figure 2 shows that L1 and L2ers performed similarly with the
ser
condition,
but dierently with the
estar
opula choice in adjectival constructions in Dutch L1 Spanish L2
signicant, these results are consistent with previous ndings: also in this task the
L2ers show more trouble with
estar
than with
ser.
When looking at dierent adjectives, a t-test revealed a signicant dierence
non-scalar gradable
ones.
Discussion of Fill-in-the-Gap Task
�e results of the Fill-in-the-Gap Task sheds light into the internal grammar that
Dutch L1 speakers build of their L2. ese Spanish L2ers correctly choose
estar
for change-of-state constructions with
non-scalar gradable
adjectives. Note that
General discussion
�e analysis of the inherent semantic properties of
estar
in comparison with
the properties of Dutch
worden
showed that although both copulas denote a
Figure 2 also shows that the L2 speakers are better than the L1 in the ‘
ser
scalar gradable
condition’. An item analysis revealed that this result was caused by a Chilean/Spanish diier
ence in the use of the adjective
grande
�anuela Pinto & Alexia Guerra Rivera
change-of-state, they crucially dier in the type of information they provide
about the modality of this change.
Worden
focuses on the process subsumed by
the change, whereas
estar
emphasizes its end result. A Fill-in-the-Gap Task tested
Table 4.
Choice of the Copula in Dutch and in Spanish According
to the Adjectival Class
ADJ
Dutch
Spanish
Non-scalar gradable
ESTAR
Reversible scalar gradable
WORDEN
ESTAR
Irreversible scalar gradable
WORDEN
ESTAR
opula choice in adjectival constructions in Dutch L1 Spanish L2
Conclusions
Ὡs paper presented a study on the acquisition of the Spanish copulas in adjectival
constructions by Dutch L1-Spanish L2ers at an advanced level⸠�e rst experi
ment showed that the diculties with Spanish copulas are not across the board,
but can be localized in ambiguous constructions and seem to involve mainly
copula
estar
. In the second experiment we explored the inherent semantics of
information L2ers access about L2 grammar. is preliminary study shows that
�anuela Pinto & Alexia Guerra Rivera
Diesing, M. (1990).
se Syntactic Roots of Semantic Partition.
Unpublished Ph.D. dissertation.
DOI: 10.1017/s0272263102003030
Geeslin, K. (2003a). A comparison of copula choice in advanced and native Spanish.
Language
Learning, 53
DOI: 10.1046/j.1467-9922.2003.00240.x
Geeslin, K. (2003b). e role of adjectival features in the second language acquisition of copula
choice. In P. Kempchinsky & C. P槱eros (Eds.),
seory, Practice and Acquisition: Papers
from the 6th
Hispanic Linguistics Symposium and the 5th
Conference on the Acquisition of
Spanish and Portuguese
(pp. 332–351). Somerville, MA: Cascadilla Proceedings Project.
Geeslin, K., & Guijarro-Fuentes, P. (2005). e acquisition of copula choice in instructed
Spanish: e role of individual characteristics. In D. Eddington (Ed.),
Selected Proceedings
of the
6th
Conference on the Acquisition of Spanish and Portuguese as First and Second Lan
guages
(pp. 66–77). Somerville, MA: Cascadilla Proceedings Project.
Geeslin, K., & Guijarro-Fuentes, P. (2006). e second language acquisition of variable struc
tures in Spanish by Portuguese speakers.
Language Learning, 56
DOI: 10.1111/j.0023-8333.2006.00342.x
Gómez-Soler, I. (2009).
se Morphosyntax-Lexicon Interface Breakdown: An Aspectual Account
of the L2 Acquisition of Ser and Estar by L1 English Speakers.
Unpublished Master’s thesis.
University of North Carolina, Chapel Hill.
Gunterman, G. (1992). An analysis of interlanguage development overtime, Part II: Ser and
estar.
Hispania, 75,
DOI: 10.2307/344396
Hanegreefs, H. (2004). e Dutch change-of-state copula
worden
and its Spanish counterparts.
A matter of aspect and voice.
Annual Review of Cognitive Linguistics, 2
DOI: 10.1075/arcl.2.01han
Kennedy, C. (1999).
Projecting the Adjective: e Syntax and Semantics of Gradability and Com
parison
. New York, NY: Routledge.
DOI: 10.4324/9780203055458
Kennedy, C. (2007). Vagueness and grammar㨠�e semantics of relative and absolute gradable
adjectives.
Linguistics and Philosophy,
DOI: 10.1007/s10988-006-9008-0
Mesa Alonso, M., Do淭nguez Herrera, M., Padrón Sánchez, E., & Morales Aguilera, N. (1993).
Ser y estar: Consideraciones sobre su uso en español.
Islas, 104
DOI: 10.1017/s0272263103000159
Prévost, P., & White, L. (2000). Missing surface inection or impairment in second language
acquisition? Evidence from tense and agreement.
Second Language Research, 16
DOI: 10.1191/026765800677556046
opula choice in adjectival constructions in Dutch L1 Spanish L2
Rothman, J., & Slabakova, R. (2011). e mind-context divide: On acquisition at the linguistic
interfaces.
Lingua
DOI: 10.1016/j.lingua.2011.01.003
Ryan, J.M., & Laiord, B.A. (1992). Acquisition of lexical meaning in a natural environment:
“Ser鐠an搠“estar鐠and the Granada Experience⸚
Hispania, 75
DOI: 10.2307/344152
Schmitt, C. (1992). Ser and estar: A matter of aspect.
Proceedings of the North East Linguistic
Society, 22
Schmitt, C. (2005). Semi-copulas. In P. Kempchinsky & R. Slabakova (Eds.),
Aspectual Inquiries
(pp. 121–145). Dordrecht: Kluwer.
DOI: 10.1007/1-4020-3033-9_6
Schmitt, C., & Miller, K. (2007). Making discourse-dependent decisions: �e case of the copulas
ser and estar in Spanish.
Lingua, 117
DOI: 10.1016/j.lingua.2006.11.007
Sorace, A. (2011). Pinning down the concept of interface in bilingualism.
Linguistic Approaches
to Bilingualism
DOI: 10.1075/lab.1.1.01sor
Uriagereka, J. (2001). Adjectival clues, keynote speech at
Acquisition of Spanish & Portugues/
Hispanic Linguistics Symposium
, University of Illinois at Urbana-Champaign (USA),
October 11–14.
Ursini, F.A. (2011). On the syntax and semantics of ser and estar. Macquarie University Research
Online.
VanPatten, B. (1985). e acquisition of ser and estar in adult second language learners: A pre
DOI: 10.2307/342218
VanPatten, B. (1987). e acquisition of ser and estar: Accounting for developmental patterns.
In B. VanPatten, T.R. Dvorak, & J.F. Lee (Eds.),
Foreign Language Learning: A Research
Perspective
(pp. 61–75). New York, NY: Newbury House.
Winter, Y. (2005). Cross-categorial restrictions on measure phrase modication.
Linguistics and
Philosophy
DOI: 10.1007/s10988-004-1469-4
Zagona, K. (2012). Ser and estar: Phrase structure and aspect. In C. Nishida & C. Russi
(Eds.),
Building a Bridge between Linguistic Communities of the Old and the New World.
Current Research in Tense, Aspect, Mood and Modality
(pp. 303–327). Amsterdam: Rodopi.
Typological proximity in L2 acquisition
�e Spanish non-native grammar of French speakers
Juana M. Liceras & Anahí Alba de la Fuente
University of Ottawa / Université de Montréal
Typologically-close languages such as French and Spanish share many
Introduction
For more than three decades now, typological proximity understood as member
ship to the same language family – or lack thereof – has played a central role in the
study of second language learning. In fact, we can trace what we cal氠‘the linguis
tic approach鈠to second language acquisition (L2A) to the so-called Contrastive
Analysis Hypothesis (CAH), which was in perfect harmony with typology.
How
ever, the concept of typological proximity is rather elusive because close languages
may display striking dierences. A case in point is the one we discuss in this chap
ter: the acquisition of Spanish by L1 French speakers.
While French and Spanish can be taken as an obvious example of typological
See
Ellis (1994
),
Liceras (1986
) or
Larsen-Freeman & Long (1991
), among many others.
�uana M. Liceras & Anahí Alba de la Fuente
To date, fossilization has been related by many L2 researchers to the presence of L2 con
structions that show permanent resistance to incorporate L2 input (change L1 structures or
idiosyncratic IL structures) and thus fail to progress towards the target (
Han 2013
). Some
researchers (
Liceras 1986
), taking a similar approach to the one portrayed by the Competing
Grammars Hypothesis (
Kroch 1994
), de�ned fossilization as the permanent co-
xistence
of the two options of a given parameter.
�e issue of typological proximity plays a central role in L3 acquisition.‟e Typological
Primacy Model (
Rothman 2011
in press
) states that at the initial stages of multilingual
ypological proximity in L2 acquisition
Typological Primacy Model states is that the lexical or grammatical level of underlying or true
Baker (1996
) maintains that the diversity found in human language from a morpho-
syntactic perspective falls within the boundaries of three types of languages and the potential
combination of the various categories, as depicted in Table 1.
Table 1.
ὲee-way typology underlying human language from a morphosyntactic
point of view (
Baker, 1996
, p. 5).
Morphological type
Isolating
Dependent marking
Head marking
Word order type
Head initial
Head nal
Free
Exemplar
English
Japanese
Mohawk
�uana M. Liceras & Anahí Alba de la Fuente
to the
that-t
“lter.
‟e two languages are also dierent in that they have been
as
�panish and French may dier in terms of a number of formal universals or
According to
Keenan and Comrie’s (1977
) Accessibility Hierarchy, if a language can
relativize a subject position, there is no guarantee that it will relativize a genitive position.
However, if it can relativize the latter, it will relativize all the positions which are above in the
hierarchy, which they characterize as shown in (1).


U(bjec�t) D(irect)O(bjec�t) I(ndirect) O(bjec�t) O(bject) of P(reposition)
� G(enitiv�e) O(bject) of C(omparison).
Chomsky and Lasnik’s (1977)
en-GB&#x/Lan;&#xg en;&#x-GB/;&#xMCID;&#x 247;U 0;&#x/Lan;&#xg en;&#x-GB/;&#xMCID;&#x 247;U 0;COMP
“lter states that the complementizer cannot be

’est le livre *Ø/que je viens d鉡cheter

s el libro _*Ø/que_(yo) acabo de comprar

It is the book (that) I’ve just bought.’
However, French and Spanish are dierent with respect to the so-called *[
that-t
“lter since
French, but not Spanish, requires the presence of a subject marker in a subject restrictive relative
clause such as (3a) versus (3b).

’est le facteur *que/qui parle l鉡rabe

s el cartero que/*quien habl愠árabe

It is the mailman that/who speaks Arabic.’
ypological proximity in L2 acquisition
occupy a dierent place in terms of Accessibility Hierarchies;
(3) typologically
distant languages such as Spanish and Arabic may share typological or formal uni
versals (i.e. have post-nominal adjectives or allow null subjects) but we maintain
that this does not make them typologically more similar to Spanish than English,
a language that does not share these two properties with Spanish; (4) the distinc
Both Spanish and French have plain passives as in (1) and (2) and renexive passives as
in (3) and (4) but only Spanish has impersonal passives as in (5).‟e diierence between the
Spanish example and the French example in (5) is that French cannot use a
-morpheme
there but
which, unlike Spanish
, functions as the subject of the sentence. English only
has plain passives.

sta catedral fue construída en la Edad Media

ette cathédrale a eté construite dans la Moyen Age

�is cathedral was built in the Middle Ages.’

sta catedral se construyen la Edad Media

ette cathedral s’a construit dans la Moyen Age

�is cathedral was built in the Middle Ages.’

Aq痭
a los
investigadores

respect
hercheurs

Here one respects researchers.’/‘Here researchers are respected.’
We use the terms interlanguage, non-native grammar and L2 grammar interchangeably in
spite of the fact that the term ‘interlanguage’, as coined by Selinker, was meant to imply that a
non-native or L2 grammar (it could also be L3 or Ln) is fundamentally diierent from a native
grammar. We use the term without linking it to the ‘interlanguage hypothesis’ or the ‘fun
damental dierence hypothesis’ (
Bley-Vroman 1990
Slabakova 2009
, among many others),
namely, as being synonymous with the terms non-native or L2 grammar.
�uana M. Liceras & Anahí Alba de la Fuente
L2Sp-L1Fr structures that neither occur in the L1 French nor in the L2 Spanish
grammar (#
demasiado nis
versus
demasiados niños
trop d’enfants,
酴oo many
children’); and (2) identify and account for L2Sp-L1Fr structures that are not pos
sible in Spanish but surface as an 攝ect of negative transfer or interference from
the L1 (*
hombres ranas
as in
hommes grenouilles
versus
hombres rana
Ⱐ‘frogmen,
divers鈩.
e Spanish grammar of L1 French speakers: Typological proximity
versus typological similarity
It has systematically been shown that, while non-native grammars are shaped by
the L1 and the L2 (
Selinker 1972
, among others), they also display what can be
labelled idiosyncratic constructions that they share with natural languages other
than the L1 or the L2 (
Adjian 1976
Adjian & Liceras 1984
Liceras 1986
Schwartz & Sprouse 1994
, among others). It has also been shown that there is
ypological proximity in L2 acquisition
such but rather as variability or optionality.
With this in mind, we take as our
point of departure the assumption that optionality will be an intrinsic character
istic of IL systems. Furthermore, we argue that the variability that characterizes
the L2Sp-L1Fr IL is, in fact, a r攞ection of the optionality that may stem from
Object clitics in L2Sp-L1Fr
Spanish object clitics – apart from the actual lexical dierences – relate to word
order in the case of innitival complements and to the position of direct and indi
rect object clitics in clitic clusters.
�e order of direct object/indirect object clitic clusters in French and
panish
is the opposite. Namely, in Spanish third person clusters, the indirect object
(dative) always precedes the direct object (accusative), as shown in (1), while the
opposite is the case in French, as shown in (2):
�os
libros?
[DAT]
[ACC]
enviaré
mana

books?
to him/her
them [I]
will send
omorrow
�es
vres?
[ACC]
lui
[DAT]
�verrai
demain

�e
books?
hem
o him/her
will send
tomorrow

��e books? I will send them to him/her tomorrow.’
While in the L2A �eld optionality and variability may be used and have been used
indistinctly (
Pérez-Leroux & Liceras 2002
), within the generative grammar approach, the
term optionality has taken preference because variation has been de�ned at the level of com
petence.‟e Competing Grammars Hypothesis is meant to address optionality resulting from
the speakers’ access to two diierent rules or parametric options (two competing grammars).
In our characterization of the L2Sp-L1Fr IL, we will try to account for relevant instances of
optionality resulting from the speakers’ access to more than one grammar and will use the
terms optionality and variability indistinctly.
Tarone (1982
) uses the term
vernacular
to refer to IL structures that are neither L1 nor
L2-like and
Goodluck (1986
) discusses the fact that any child L1 as well as L2 structures
should exist in a given natural language, namely, that there are no “wild” grammars.
�uana M. Liceras & Anahí Alba de la Fuente
It has also been shown that the two languages dier, as we will illustrate in the
forthcoming sections, in terms of the restrictions that apply to clitic cluster combi
nations (
Alba de la Fuente 2010
quiero
ver

�.
veux
oir
-la

�nt
o see-her

I want to see her.’
la
�uiero
�er

�.
*Je
�eux
�oir

nt
o see

I want to see her.’
quiero
ver

�.
Je
voir

nt
o see

I want to see her.’
A superᱣial approach to transfer would predict that L1 French speakers would
place Spanish clitic pronouns before the innitival form. �us, in spite of typo
logical proximity, this dierence could shape L2Sp-L1Fr in a way that would not
be expected in the case of other Spanish ILs (e.g. L1 English IL). In turn, if learn
ers have access to UG and therefore to constructions available in any natural lan
guage, L2Sp-L1Fr could display idiosyncratic clitic structures. In other words,
it would display clitics that are dierent from those of French and Spanish but
were possible in Old Spanish, and are also possible in languages such as Czech,
as in (7) below. However, why would these IL speakers produce or accept those
constructions?
One possible answer to that question could come from linguistic theory, as
Liceras (1985
ypological proximity in L2 acquisition
Old versus modern Spanish clitics
Old Spanish
Modern Spanish
-Interpolation
-No interpolation
-Precede or follow the inected verb
-Only precede inected verbs
-No clitic doubling
-Clitic doubling
�ese properties of Old Spanish are exemplied in (6)–(8).
Interpolation
otro
queles
este
buen
andado
Moysen

�nother
day
that them
this
ood
directions
said
oysen

�the day a1er Moses had given them directions’

Fontana 1997
, p. 229)
�-CL and CL-V orders with iniected verbs
Rogaron
que
les
diesses
llave

�hey) begged
him that
them
(you) give
the
key

��ey asked him to give them the key.’

Fontana 1997
, p. 228)
�itics and NPs in complementary distribution
�el
lamaban
otrossi
amosis

�o him (they)
alled
lso
mosis

��ey also called him Amosis.’

Fontana 1994
, p. 89)
�e properties of Modern Spanish, which it shares with French, are depicted in(9)
to (11).
�o Interpolation
Se
�yer

it
gave
yesterday

I gave it to him/her/you yesterday.’

Se
�yer

it
yesterday
ve

I gave it to him/her/you yesterday.’
�uana M. Liceras & Anahí Alba de la Fuente
�rder before iniected verbs
Lo
trajo
ayer

t (he/she)
brought
yesterday

He/she brought it yesterday.’

�.
*t狡jo
ayer

e/she) brought it
esterday
�litic doubling
Les
contaron
cuento
ptional)

hem)
(they)told
tory
the
children)

�ey told a story to the children.’

�.
les
contaron
cuento
bligatory)

the
children
them
(they)told
tory

�ey told a story to the children.’
Perales and Liceras (2010
) analyzed oral and written data produced by 30
intermediate-
vanced L1 French speakers of Spanish. e only IL property that
was compatible with Old Spanish was the presence of instances of clitics in prever
bal and post-verbal position with inected verbs. �us, the authors did not ᱮd
clear evidence for a phrasal (Old Spanish) account of IL clitics. However, since the
authors found instances of post-verbal clitics with tensed verbs, which are neither
possible in French nor in Spanish, they speculated that learners may generate clit
ics in both argument and non-argument positions. In other words, the IL grammar
Spanish.
However, there is a recent feature account of object pronouns that makes
As we have seen, this was possible in Old Spanish and it is possible in languages such as
Czech and Serbo-Croatian.
It could be argued, under a ‘psychotypology’ approach as in
Kellerman (1979
) or the
Typological Primacy Model (
Rothman 2011
), that French learners might perceive
Spanish object pronouns as not being typologically similar to French object pronouns. While
this is a logical possibility, we will not discuss ‘psychotypology’ in this paper.
ypological proximity in L2 acquisition
French and Spanish closer to each other than to languages, like English (
Camacho
Taboada 2006
), but not identical.
Perales and Liceras (2010
) and
Liceras (2014
draw from this account to provide a feature denition of object pronouns in terms
of the [±] combination of two features which relate to their phonological nature
宱phonological] and their phrasal [±XP] nature.
Table 2.
A feature account of object pronouns: From full pronouns to agreement
markers
Full pronoun�s
imple clit�ics
litics
�ategorial clit�ics
greement markers
English
nglish
ech, Old Sp
talian, French
panish, Bulgarian
宖phon]
hon]
[+phon]
[+XP]
As Table 2 shows, French clitics are not classied as agreement markers but
as categorial clitics. is implies that typological proximity is a relative term and
cannot be automatically equated to typological similarity⸠�e feature combina
tion approach allows us to account, among other things, for the fact that only
languages whose clitics are agreement markers have clitic-doubling. is in itself
does not seem to be the only reason why clitic-doubling has systematically been
found to be problematic for all Spanish learners (Bruhn de Garavito 1999b;
Liceras
Perales & Liceras 2010
Zobl & Liceras 2006
; among others), but it separates
French from Spanish when it comes to object pronouns. us, L2Sp-L1Fr may
Null/overt subjects
�ere have been analyses of French (
Authier 1992
Roberge 1986
) that clas
sify it as a [+null subject] language like Spanish or Italian because its subject clitics
are analyzed as person agreement morphology. However, the majority of syntacti
cians consider French a [–null subject] language like English or German, and it
has in fact been used to illustrate diachronic change from [+null subject] to [–null
subject] (
Adams 1987
Roberts 1993
⤮‟us, while being a Romance language
makes French typologically close to Spanish, there is no typological similarity
Second position (2P) clitics have been related to Verb Second phenomena.‟is wa猠�rst
pointed out by Wackernagel (1892). See Anderson (1993) for a discussion of both phenomena.
�uana M. Liceras & Anahí Alba de la Fuente
�issing Subjects
elefonearmañana

�.
é泩phonerai demain
Je té泩phonerai demain)

ill call tomorrow

I will call tomorrow.’
�ree Subject-Verb Inversion
ea ha telefoneado/Ha telefoneado Lea

�.
ea a télephoné/*a té泩phonLea

Lea a té泩phoné鈩

ea has phoned/*has phoned Lea

Lea has phoned.’
ᱡt-trace
攟ects
뽑uién
dice Lea
que
acaba de llegar?

�.
ui dit Lea
que
vient d’arriver?

ui dit Lea
qui
vient d’arriver?)

*Who
does Lea say
that
has just arrived?

ho does Lea say __ t
has just arrived?)

Who does Lea say has just arrived?’
�ese examples show that the equivalent of the Spanish sentences is ungrammati
cal in both French and English, which in principle explains why L1 Spanish speak
ers (mainly those with lower prociency) accepted many of the ungrammatical
sentences listed in (13) and (14) as grammatical, while the French speakers did
not. However, a closer look at the construction in (14) shows that French and
English 摩ier in the way in which they avoid the
that-t
sequence (
that
is not real
ized in English whereas French requires
qui
to mark the subject position).
‟is
implies that transfer may not work the same way in English and French when it
comes to their respective Spanish ILs. In other words, English and French adopt
摩ierent solutions to avoid the sequence
que+t
to comply with the
&#x/MCI; 25;̹ ;&#x/MCI; 25;̹ ;that+t
崠slter
proposed by
Chomsky and Lasnik (1977
). is is another example of how typo
logical proximity cannot be equated to typological similarity, either in terms of
�is
qui
has been analyzed as the result of the fusion of the complementizer
que
and
the subject pronoun
il
which took place in XIV century French (
Kayne 1975
), and it is
referred to as the
que
qui
que + il
) rule.
ypological proximity in L2 acquisition
in the case of French, we are dealing with a degree of typological proximity that
almost reaches typological similarity. is is not the case for English.
Furthermore, even if French and English select the [–null subject] option,
of
�panish, since the rejection of overt subject pronouns that are redundant in
French group while the English-speaking group showed less consistency.
econd,
uién dices ___ estudia español contigo?

�ho say (you) ___ study (3rd. p.) Spanish with you

�Who do you say ___ is studying Spanish with you?’
Participants either inserted an overt subject (
Ⱐ‘you鈩r erased the verb
decir
. In
) or specic morpho-
phonological realizations of the properties that interfere with assigning French
and Spanish to either the same or the opposite option of the [±null subject]. None
theless, the typological proximity, albeit not similarity, which stems from verbal
e Spanish grammar of L1 French speakers: Morphosyntactic issues
�e available analyses of the Spanish grammar of L1Fr speakers that we will dis
cuss are word formation (compounding), the plural feature in nouns, clitic cluster
constraints, quantiers and passive constructions. We have chosen to discuss these
areas of grammar for two reasons. First, they allow us to show how typological
�uana M. Liceras & Anahí Alba de la Fuente
proximity shapes the L2Sp-L1Fr IL at dierent structural levels and in relation to
Productivity, head directionality and distinct phono-morphological
features in French and Spanish compounds: Acceptance versus
production
�omme-grenouille

�ombre
rana

�an frog ‘frogman’
�uvre-boîte

abrelata-
open can/s
can open-
As the English equivalent shows, French and Spanish N–N compounds are le-
headed, while English N–N compounds are right-headed. is is also the case for
deverbal compounds, which are VO in French and Spanish but OV in English. e
deverbalization process is also similar in French and Spanish. Specically, regard
Spanish and the lack of productivity of N–N compounds in both languages are
also a r攞ection of their typological proximity.
However, in spite of this high
In fact, many English N–N compounds or deverbals are realized as Noun Phrases or as
Nouns with derivative morphology in both French and Spanish, as shown in (1) and (2).

oll house

asa de mucas

aison de poupée
ypological proximity in L2 acquisition
ascanuec-
es
unos cascanuec-
b.
omme-grenouille
des homme
-grenouille-
b.
�ombre rana
unos hombre-
rana
imbre-poste
des timbre-
-poste

�.
ello/unos sello-
Pomerleau (2000
) compared the status of these compounds in the Spanish L3 of
French speakers with English as an L2. In the production task, none of the French
speakers had problems with directionality or with the morphology of N–N com
pounds, but they had problems with the morphology of deverbal compounds,
which seems to indicate that the L2, English, did not play a relevant role. In other

lothes line

endedero

When the noun complement is a mass noun, it looks as if both options (with and without
plural marker) are possible:

un chasse-neige
des chasse-neiges

atch snow
ome catch snows

a snow catcher’
ows catcher

�n chasse-neige
des chasse-neige
An anonymous reviewer points out that in some Spanish compounds mass nouns occur
in the singular (i.e.
tragaluz -
swallow light- “skylight鐩. We do not think that this is relevant
here because we are discussing N–N compounds in order to show that only the head takes the
plural morpheme.‟us, the Spanish equivalent of
light house
would be
casa luz
and, in plural,
light houses
would correspond to
casas luz
but not *
casas luces
. If
maison lumière
were pos
sible in French, then
maisons lumières
would be (and in fact is) the expected plural.
�ere is no N–N equivalent in Spanish.
�uana M. Liceras & Anahí Alba de la Fuente
words, the typological proximity of the L1 won over a potential recency eect
⡩渞uence from the second language acquired). However, in the
rammaticality
judgment task all participants accepted some N–N right-headed compounds.
Pomerleau argues that when these subjects have to produce compounds they sys
tematically produce le-headed ones, while they seem to need substantial expo
sure to Spanish to be able to categorically reject right-headed compounds. Since
A prosody account of the realization of plural in L2Sp-L1Fr nouns
Unlike the case with the generic
of deverbal compounds, French and Spanish
share not only the plural marker
but also the noun-adjective word order.
Bruhn
de Garavito (2008
) assumes that this word order results from a number projection
with a [+strong] number feature that is available for DPs to which nouns raise,
resulting in the noun–adjective word order⸠�e author investigates the acquisi
tion of Spanish plural
�is is a potential candidate for fossilization because learners systematically alternate
ypological proximity in L2 acquisition
observes that, if one compares English- and French-speaking learners of Spanish,
one would expect an advantage for the French speakers in regard to the no畮–
adjective order, which French shares with Spanish, and that, taken at this supers
cial level, neither group should experience diculty with plural morphology, as it
exists in both L1s– see (21) – and is frequent in the L2 input data.
�as
luce
roja
lumière
rouge

*the
lights
red

�the red lights’
However, as the author points out, French 摩iers from Spanish in several respects:
(i) French doesn’t have word markers linked to the formation of the plural; (ii) in
French a postvocalic consonant is only pronounced when it is a sonorant; (iii) a
ᱮal
can only be realized if it attaches to a following syllable or if it is marked in
the lexicon and (iv) the syllab検cation of a nal obstruent is not saved by a rule of
epenthesis.
Now you see it, now you don’t: Plural and case marking
in L2Sp-L1Fr quantiers
�uana M. Liceras & Anahí Alba de la Fuente
examined the L2 acquisition of Number specication on Spanish quantiers by
French speakers.
In French and Spanish, quantiers can combine with NPs by means of the
preposition
, as in (22), or directly, as in (23):
�lgo
vino

�ome
of
ine

�ome win斒

�ssez
vin

�ough
of
ine

�enough wine’
�rios
erros

�lusieurs
chiens

�several dogs’
�e need to value the case feature of the NP when Number agreement is not pos
sible accounts for the use of
as a case marker. While the general pattern is the
one in (23) for Spanish, in French most quantiers require
. In fact, only quanti
ᱥrs that can establish Number agreement with the Noun can modify the Noun
directly. Quantiers such as
bastante
(enough),
mucho
(much/many), or
demasi
ado
(too much/too many) establish Number agreement with the Noun in Spanish
but require
in French, as shown in (24).
�astantes/muchos/demasiados perros

�ssez de/beaucoup de/trop de chiens

�enough/many/too many dogs’
Androutsopoulou, Español-Echevarría and Prévost (20
10) administered a gram
maticality judgment task and a production task to 9 intermediate and 17 advanced
French-speaking learners of Spanish at the University of Laval (Quebec, Canada).
�ey found that these learners performed poorly on plural inection with some
quantiers, and rejected the use of
. is suggests that they had acquired Num
ber specication on these quantiers, which allows case marking on the following
noun, but did not produce the appropriate morphology⸠�e authors argue that
nor
�panish quantiers can occur without plural marking when they are directly
followed by nouns (
demasiado perros
). �us, the IL of these speakers seems to
ypological proximity in L2 acquisition
include bare quantiers (without a plural marker agreeing with the Noun) that are
neither possible in French nor in L1 Spanish but can be found in other Romance
languages such as Catalan. is provides us with an idiosyncratic L2Sp-L1Fr con
struction that cannot be accounted for by L1 transfer from French or by overgener
alization of a construction available in Spanish but that occurs in other languages.
Restrictions on clitic clusters: Typological proximity and
the representation/processing divide
�e 攝ects of typological similarity are explored in
Alba de la Fuente’s (2012
) study
of clitic cluster constraints, which provides data from two experiments admin
istered to advanced L2 speakers of Spanish whose L1s are English, French and
Romanian.
Te me
presentaste en la esta [Spanish]

Mi te

*Tu
te m
’es prentla fe [French]

You introduced yourself to me at the party.’
Given the 摩ierent versions of clitic combinations that are available in natural lan
guages, this phenomenon provides a scenario that allows us to explore the eects
of typological similarity in contrast with typological proximity.
Generally speaking, both French and Romanian may be considered to be
typologically-close to Spanish, as they are all Romance languages. However, as we
Since the second experiment does not provide clear-cut answers with respect to the
typological similarity/proximity dichotomy, only the results of experiment 1 are reported here.
According to
Alba de la Fuente’s (2010
) Narrow Plural-Blocking Eiect, which would
apply in both Spanish and Romanian, “in a combination of
1st
and
2nd
person clitics with a
dative, the non-dative cannot be plural” (p.213). As such, combinations of two singular clitics
are allowed and combinations of two plural clitics are rejected.
�uana M. Liceras & Anahí Alba de la Fuente
L1 Romanian speakers would be at an advantage in this case, since the restrictions
in their L2 match those of their L1. In turn, L1 French speakers would have to
restructure the restrictions available to them in their L1 to accommodate to the
Passive constructions are not dierent when it comes
to dierential object marking
Tremblay (2006
) investigated the L2 acquisition of Spanish reexive passives, as in
(26), and reexive impersonals, as in (27), by French- and English-speaking adults
at an advanced level of prociency.
�sos
isos
[ACC]
�nstruyeron
hace

�es
ppartements
se
[ACC]
sont
�nstruits
uis
cles

�ese
apartments
ere
uilt
1er
wo
turies

��es攠�ats were built two centuries ago.’
ta
udad
se
[NOM]
�uede
entrar
con
mucha facilidad

�ans
ille
[NOM]
peut
�trer
r
lement

�n
his
ty
ne
an
ter
ery
easily

�In this city one can enter very easily.鈯‘One enters this city very easily.’
ypological proximity in L2 acquisition
impersonal passives checks accusative case and must be preceded by the so-called
personal
that precedes Spanish [+animate] [+specic] direct object comple
ments, as in (28).
�q痭
se
[nom.]
�os

�ci
especte
nts

�Here one respects children.鈯‘Here children are respected.’
In principle, as Tremblay herself hypothesized, typological proximity would pre
dict that native French speakers would evidence convergence on these properties
because neither of the two constructions exist in English and even if imper
sonal passive does not exist in French, this construction and the reexive passive
construction are superᱣially similar (
V DP) to the French reexive passive
construction.
Tremblay (2006
) administered a grammaticality judgment task to 16 French,
found
�hat the two non-native groups 摩iered signicantly from the native group.
As for the two non-native groups, typological proximity played a role in that
the French group signicantly outperformed the English group on grammatical
reexive passives with a pre-verbal [-animate] DP that had to agree with the verb.
Grammatical and ungrammatical impersonal test items involving [+animate]
DP猚阠preceded or not by the object-marking preposition
– were particularly
problematic, as L2 learners judged them both as grammatical. is implies, once
Diierential object marking is not particular to Spanish but occurs in natural languages
such as Hindi, Yiddish, Persian or Turkish (Montrul & Gürel, this volume), which are typo
logically distant from both Spanish and French (
Aissen 2003
Leonetti 2004
Torrego 1998
�uana M. Liceras & Anahí Alba de la Fuente
e Spanish grammar of L1 French speakers: Beyond morphosyntax
vio
uan
mientras
�ontaba

saw
9ᄆ
Juan
while
was riding
on
bike

Alex saw Juan while he was riding a bike.’

�.
vio
uan
mientras
él
�ontaba

saw
9ᄆ
Juan
while
was riding
on
bike

Alex saw Juan while he was riding on bike.’
In other words, null pronouns prefer the antecedent in Spec IP (the subject posi
tion) while overt pronouns prefer to be coindexed with the antecedent in object
position, regardless of them being in forward or backward anaphora construc
tions, as we can see in (29) versus (30).
ientras
�ontaba
Alex
vio a
Juan

hile
was riding
on
ike
saw
�an

�.
Mientras
�ontaba
Alex
io a
Juan

while
he
was riding
on
ike
aw
uan

While he was riding on bike Alex saw Juan.’
�e PAH has been tested using experimental data in Italian (
Sorace 2007
Sorace & Filiaci 2006
), Spanish (
Alonso-Ovalle, Fernández-Solera,
Frazier & C汩1on 2002
), and Arabic (Bel & Gar揭a-Alcaraz, this volume), among
other languages.
�ese experimental data have shown that native speakers have rather cate
gorical processing preferences in the case of null subjects as in (29a) and (30a), in
ypological proximity in L2 acquisition
contrast with the less clear-cut picture oiered by overt subjects (29b) and (30b), in
which the overt pronoun can also take an outside referent (neither Alex nor Juan
but rather someone else). As for non-native speakers (L1 English speakers), the
processing preferences were native-like for the anaphora resolution of null sub
jects but signicantly dierent from the native preferences for the anaphora reso
lution of overt subject pronouns.
panish
group, the results were compatible with the authors鈠proposal that Spanish overt
pronouns can have a pragmatic value but they can also be weak pronouns,
�uana M. Liceras & Anahí Alba de la Fuente
and object referents were not straightforward, as the results revealed dierences
In sum, the anaphora resolution preferences displayed by the native
panish
speakers of this experiment do not follow Carminati’s PAH when it comes to null
subject forward anaphora–shown in (29a). However, the native group diers sig
Conclusions
been considered a [–null
ubject] language like English, the fact that, unlike Eng
lish, French has clitic subject pronouns and adopts a 摩ierent solution to abide
by the so-called
that-trace
“lter distances this language from English, a fact that
ypological proximity in L2 acquisition
markers, may facilitate the
acquisition of Spanish by L1 French speakers. In this
respect, typological proximity may
upersede the fact that these two languages
�uana M. Liceras & Anahí Alba de la Fuente
is problematic for the L2Sp-L1Fr speakers, but rather the processing of null and
overt pronouns, in the sense that they do not seem to capture the marked value
of Spanish overt pronouns. We are aware of the fact that more research is needed
to tease apart diculties encountered by non-native speakers at the pragmatic-
�OI: 10.1111/j.1467-1770.1976.tb00279.x
Adjian, C., & Liceras, J.M. (1984). Universal grammar, the intake component and L1:
Accounting for adult acquisition of relative clauses. In F. Eckman, L. Bell, & D. Nelson
(Eds.),
Universals of Second Language Acquisition
(pp. 101–118). Rowley, MA: Newbury
House.
Aissen, J. (2003).
䑩ierential object marking: Iconicity vs. economy
Natural Language and
Linguistic seory,
DOI: 10.1023/a:1024109008573
Alba de la Fuente, A. (2010). More on the clitic combination puzzle: Evidence from
panish,
Catalan and Romanian. In S. Colina, A. Olarrea, & A. M. Carvalho (Eds.),
Romance
Linguistics 2009. Selected papers from the 39th
Linguistics Symposium on Romance Lan
guages
LSRL
, Tuscon, Arizona, March 2009.
(pp. 203–215). Amsterdam: John Benjamins.
DOI: 10.1075/cilt.315.12alb
Alba de la Fuente, A. (2012).
Clitic Combinations in Spanish: Syntax, Processing and Acquisition.
Unpublished Ph.D. dissertation. University of Ottawa, Ottawa.
Alonso-Ovalle, L., Fernández-Solera, S. Frazier, L., & C汩1on. C. (2002). Null vs. overt pronouns
and the topic-focus articulation in Spanish.
Journal of Italian Linguistics,
Anderson, S.R. (1993). Wackernagel’s revenge: Clitics, morphology, and the syntax of second
position.
Language,
DOI: 10.2307/416416
Androutsopoulou, A., Español-Echevarría, M., & Prévost, P.H. (2010). e syntax/
orphology
interface in Spanish L2 acquisition: Focus on quant検ed DPs.
ᱥ Canadian Journal of
Linguistics,
DOI: 10.1353/cjl.2010.0005
DOI: 10.1017/s0022226798227113
ypological proximity in L2 acquisition
Bley-Vroman, R. (1990). e logical problem of foreign language learning.
Linguistic Analysis,
DOI: 10.1017/cbo9781139524544.005
DOI: 10.2307/2273965
Chomsky, N. (1986).
Knowledge of Language: Its Nature, Origin, and Use
. New York, NY: Praeger.
DOI: 10.9793/elsj1984.6.213
Chomsky, N. (1995).
se Minimalist Program
. Cambridge MA: e MIT Press.
DOI: 10.1017/s0272263197241070
Chomsky, N. & Lasnik, H. (1977). Filters and control.
Linguistic Inquiry, 8
DOI: 10.7551/mitpress/9780262062787.003.0002
Stechow, W. Sternefeld, & T. Vennemann (Eds.),
Syntax: An International Handbook of
Contemporary Research, Vol.1
(pp. 506–569). Berlin: Walter de Gruyter.
Desrochers, A., Liceras, J.M., Spradlin, K.T., & Fernández, R. (2003). Can an on-line response
DOI: 10.1017/s0272263100014479
Fontana, J. (1994). A variationist account of the development of the Spanish clitic system. In
䬮2Beals (Ed.),
Papers from the
13th
Regional Meeting of the Chicago Linguistic Society,
Vo氮22: e Parasession on Variation in Linguisti挠seory
(pp. 87–100). Chicago IL: Chicago
DOI: 10.1017/cbo9780511620683.005
�uana M. Liceras & Anahí Alba de la Fuente
Greenberg, J. (1963).
Universals of Language
. Cambridge, MA: e MIT Press.
DOI: 10.1017/s0022226700001109
DOI: 10.1162/002438905774464322
Kayne, R. (1975).
French Syntax
. Cambridge, MA: e MIT Press.
DOI: 10.1017/s0022226700005272
Kayne, R. (1976). French relative
que
. In M. Luján & F. Hensey (Eds.),
Current Studies in
Romance Linguistics
London: Longman.
DOI: 10.1177/026765839200800205
judgments on the anaphoric value of null and overt Spanish subject pronouns.
Liceras, J.M., & Fernández-Fuertes, R. (2013). Subject omission/production in child bilingual
English and child bilingual Spanish: e view from linguistic theory. Workshop on
Cross-
linguistic inᵵence in language contact
. 19th International Congress of Linguistics (ICL
2013) – Geneva, July 2013.
Liceras, J.M., Mongeon, C., Cuza, A., Senn, C., & Spradlin, K.T. (2004). La adquisición en el aula
sin input formal: Los compuestos exo揩ntricos de las interlenguas del español.
Red-ELE,
Martez-Sanz, C. (2011).
Null and Overt Subjects in a Variable System: e Case of Dominican
Spanish.
Unpublished Ph.D. dissertation. University of Ottawa, Ottawa.
ypological proximity in L2 acquisition
DOI: 10.1017/s1366728909990071
Newmeyer, F.J. (1990). Explaining language universals.
Journal of Linguistics,
DOI: 10.1017/s002222670001450x
Perales, S., & Liceras, J.M. (2010). Looking for universals in the acquisition of L2 Spanish object
clitics. In P. Guijarro-Fuentes & L. Dominguez (Eds.),
New Directions in Language Acqui
sition: Romance Languages in the Generative Perspective
(pp. 419–452). Newcastle upon
Tyne: Cambridge Scholars.
Pez-Leroux, A.T., & Liceras, J.M. (2002).
se Acquisition of Spanish Morphosyntax: e L1/L2
Connection.
Dordrecht: Kluwer.
DOI: 10.1007/978-94-010-0291-2
Pomerleau, J. (2000).
La adquisición del español en el aula: Los compuestos nominales de los fran
ceses de Canadá.
Unpublished Master thesis. Universitat de Barcelona, Barcelona.
Bulgarian and Old Spanish. In A. Kemenade & N. Vincent (Eds.),
Parameters of Morpho
syntactic Change
(pp. 170–206). Cambridge: CUP.
Rivero, M.L. (2004). Spanish quirky subjects, person restrictions and the person-case constraint.
Linguistic Inquiry, 35
DOI: 10.1162/ling.2004.35.3.494
Rivero, M.L. (2008). Oblique subjects and person restrictions in Spanish: A morphological
approach. In R. D’Alessandro, S. Fischer, & G. Hrafnbjargarson (Eds.),
Agreement Restric
tions
(pp. 215–250). Berlin: De Gruyter.
DOI: 10.1515/9783110207835.215
Roberge, Y. (1986). Subject doubling, free inversion and null argument languages.
Canadian
Journal of Linguistics,
Roberge, Y. (1990).
se Syntactic Recoverability of Null Arguments
. Montreal: McGill-Queen’s
University Press.
DOI: 10.7202/602763ar
Roberts, I. (1993).
Verbs and Diachronic Syntax: A Comparative History of English and French
Dorcrecht: Kluwer.
DOI: 10.1007/978-94-011-2910-7_2
Rothman, J.
DOI: 10.1177/0267658310386439
Rothman, J.
Cognitive economy, non-redundancy and typological primacy in L3 acqui
sition: evidence from initial stages of L3 Romance.
In S. Baauw, F.A.C. Drijkoningen, &
䴮2Pinto (Eds.),
Romance Languages and Linguistic“eory
(pp. 217–247). Amsterdam:
John Benjamins.
DOI: 10.1075/rllt.5.11rot
Rothman, J. (In press). Linguistic and cognitive motivations for the typological primacy model
of third language (L3) transfer: considering the role of timing of acquisition and pros
ciency in the previous languages.
Bilingualism: Language and Cognition
DOI: 10.1017/s136672891300059x
Schwartz, B., & Sprouse, R. (1994). L2 cognitive states and the full transfer/full access model.
Second Language Research, 12
DOI: 10.1177/026765839601200103
Selinker, L. (1972). Interlanguage.
IRAL, 10
DOI: 10.1515/iral.1972.10.1-4.209
Selinker, L., & Lamendella, J. (1978). Two perspectives on fossilization in interlanguage learning.
Interlanguage Studies Bulletin,
DOI: 10.1111/j.1467-1770.1979.tb01075.x
�uana M. Liceras & Anahí Alba de la Fuente
Sheehan, M. (2006).
se EPP and Null Subjects in Romance.
Unpublished Ph.D. dissertation.
Newcastle University, UK.
Slabakova, R. (2009). L2 fundamentals.
Studies in Second Language Acquisition,
DOI: 10.1017/s0272263109090263
Sorace, A., & Filiaci, F. (2006). Anaphora resolution in near-native speakers of Italian.
Second
Language Research,
DOI: 10.1191/0267658306sr271oa
Tarone, E. (1982). Systematicity and attention in interlanguage.
Language Learning,
DOI: 10.1111/j.1467-1770.1982.tb00519.x
Torrego, E. (1998). Nominative subjects and pro-drop I渞.
Syntax,
DOI: 10.1111/1467-9612.00008
Tremblay, A. (2006). On the second language acquisition of Spanish reexive passives and
reexive impersonals by French- and English-speaking adults.
Second Language Research,
DOI: 10.1191/0267658306sr260oa
Valenzuela, E., Liceras, J.M., & López-Morelos, L.P. (2011). Ambiguous anaphora in L2 English
and L2 Spanish. Paper presented at the
BUCLD
[+STATE]
Accessibility Hierarchy
Accusative
case
�nᨙᰝ3n
i᠘ᤜi᜞ᤜ
i᜛ᤜ�n8
see also
Dierential Object
Marking (DOM)
clitic
7᜙ᰞḚᤜḝ�
ᴖ᜙ᰝ��
object
i᜛ᤜ἟a
Activation
ᴖᤜ�ᴙᰟn
�ᐙs
ἕᤜ἗ᤜni
ᬟᤜḚ᠙ᰞᴕᤜ�nᨙs
ḕ�
ḕ3
Airicate
��ᤜᔗ
᠞ᤜ3a
Afrikaans
iἔ
Alveolar
��ᤜ᠚
see also
Apico-alveolar,
Dento-alveolar
Anaphora resolution
ᴚἙs
iḚᤜᴝn
ᴝ�ᤜ἟ᬙs
ἔ�
Animacy
i᠝
i3Ἑᰝ3a
i᜚ᤜ
Ἒᨙᰟ2i
�᨟
Animate
Ḟᬙᰝ7ᐙᰝ᠞
i᠝1s
i᠛
i3ᐙᰝ᠘
i8Ἑᰝ᜔1s
i8�ᤜἚᨙᰟ2i
Ἒᬙs
�n᜙ᰟᐟ
Inanimate
Ḟᬙᰞᴕᤜḝ᜙ᰝ7ᐙs
i3ᬙᰝ᠘
i᠗ᤜi᜞
ᴗ�1s
ᴗa
i8�ᤜἚ2
Ἒᴙᰟ2n
Apico-alveolar
��ᤜ᠚ᤜ᠖
see also
Alveolar
Arabic
Ḟᐙᰞ�ḙᰝ2ḙᰝ2Ἑs
ᴚ�ᤜᴚ8
iḚᤜ
ᴝn
ᴝᘙᰝ�a
�Ἕ
἟Ἑᰟa2
Moroccan Arabic
Ḟᐙᰞ�ḙs
ᴚḙᰝ2Ἑᰝᨖᤜᴚ8
iḚᤜ
ᴝᬙᰝᴖ
Modern Standard
Arabic
ᴚᘙᰝ�a
Darija
Aspect
i᠙ᰟᨙᰟἙᰟᐙᰟ�ᤜaᴙs
᜛ᤜᴟn
ᴟᘙᰝnḙᰝnᬙᰝ᠞ᤜ
i3ᬙᰟḝᤜ἞a
Aspectual
content
delimitation
摩ierence
feature
iᬙᰟn
�ᐙᰟ�
�᠙s
nᴙᰟḝᤜ἞7
information
markers
morphology
iἔ
property
�἗ᤜi᠗ᤜἚ᜙s
἞ᘙᰟᴚ
�iḙᰟin
s甛x
Attributive adjective
�ἔ
�἖ᤜ
�ᬖᤜḔ�ᤜḖi
Bantu languages
i᠝
Basque
ᐖᤜ�a
�ᘙᰕ3
�᨞
Bilingual acquisition
�ᘙᰕ᠙s
ḞἙᰞa�ᤜᴚ3
Bilingual rst language
acquisition
Brazilian Portuguese (BP)
ḕἙs
ᴖᨙᰝ7ᐙᰝ��
ᴕ7
Catalan
�ᘙᰘḙᰗᘙᰗ᜙s
ḝ᜙ᰞ�a
�ἕᤜ
�ᬚ
�nᴙᰞnᬙᰞᬖᤜ�ᬘ
Ḕᴙs
Ḕ᜙ᰞᘞᤜḖἙᰞ�ᬙs
iḛ
iḔᤜᴝᐙᰝ᜚ᤜ἞᠙ᰟn�
Case
Ἒᤜ�i
�ᬙᰛᴙᰕ᠙ᰘḙs
ḝḙᰞnᨙᰞn᜙ᰝ᠞ᤜi3n
i᠘ᤜ
i᜞ᤜi᜛ᤜἚ�
Ἒᬙᰟᬔ
�ᬗ
Change of state
�ᬟᤜ἞Ἑs
἞ᘙᰟᴟ
Ἕn
CHILDES
Ḟ᜙ᰞaᨙᰝ᠗
Chinese
nᨙᰞḔ
Ḟᘙᰝa�ᤜᴖᨙs
ᴖᴙᰝᘘ
ᴕ�
Mandarin Chinese
(MC)
Clitic
i᠙ᰟ2
�ᬙᰟ�ᤜnᨙs
nᴙᰔᘙᰖᘙᰖ᜙ᰞᨔᤜḚ᠙s
ḞᨙᰞḝᤜḞ�ᤜḝ�
ḝἙᰞᴕᤜ
iᘝ
ᴖ�ᤜᴖ8
ᴕἙᰝ�ᘙs
�ἔ
἟᜙ᰟn�ᤜ�a�
ἔ�
cluster
�ἔ
἟ᘙᰟnḙᰟn�
doubling
�ᴙᰛᨙᰟ἖
non-argumental
iḙᰝἙᰟḙs
἟ᤜni
see also
Object clitic
Coda
complex
consonant
Contrastive Analysis
Hypothesis (CAH)
Ἕ8
Contrastive Focus
Ḗ᜙ᰞ�ḙs
ḕ�
ḕᬙᰞ�3
ḘἙᰞ3a
Convergence
iḙᰝἙᰝa
἞ᤜ�ᐙᰟ�ᤜ἗ᤜni
�ᨕ
Ḛ᜙ᰞḝᤜḝ᜙ᰞ78
ḕa
ḕ�ᤜḗn
ḗᐙᰟn8
Coronal
�ᘙᰘḙᰘ7
Copula
Ḟᘙᰞᴕᤜ
�ἔ
�ᬖᤜ�ᬘ
�n᜙ᰞa�
Ḕ�ᤜ
Ḗḙᰞ7Ἑᰟᨗ
἞�ᤜ
�iḙᰟᴟ
Ἕa
Copula + ADJ
Cross-linguistic
楮�uence
iᐙᰟ2
἞1s
�ᐙᰟ�
nᴙᰞ�ᐙᰞᬔ
�ᬖ1s
�ᬘ
�n᜙ᰞa�ᤜḖ2
Ḗᴙs
ᴚᴙᰝᨗ
iḚᤜᴝa
ᴝᘙs
ᴝ᠙ᰝ἗ᤜinᨙᰝ��
3Ἑᰗ7
CVC
�ᘙᰕ᠙ᰘ�
3ᐙᰘ8
Dative
ᴕᤜ἟
�ᬙᰛᴙᰞᨘᤜ
i᠛
i᠖ᤜi᜞
i᜔ᤜi8�ᤜ
�᨝
Ἒᬙᰟ�ᐙᰟn�
marker
i3ᐙᰟ2i
�᨟
Default
aᨙᰔᴙᰔᘙᰔ3
7ᐙᰖ᠙ᰞḚᤜḞᴙᰞ�ᐙs
ḗᬙᰟin
Deniteness
i᠙ᰞᨔᤜ
ḝᐙs
ḝ3
ḝ᜙ᰝᘞ
ᴖἙᰝ7ᐙs
iᘕ
ᴕἙᰝ�ᘙs
i3ᴙᰝ3ᬙᰝ᠖ᤜi᠘ᤜἚi
Denite
i᠙ᰟᨙᰟ8
nᨙs
Index
Index
ḝᬙᰞᴖ
ḝ᜙ᰝᨖᤜ
ᴖ᜙ᰝᔝ
ᴕᘙᰝ3ᴙs
i3ᬙᰝ᠖
i᠘ᤜi᜚
i᜞ᤜ
ᴗa
i᜖ᤜἚ2
Deniteness Eect
ḞᴙᰞḖᤜḞ᠙ᰞiᐙᰞi᜙s
i᠖
Indenite
ḝᨙᰞin
ḝᐙᰞi᠙s
ᴕᨙᰝᔝᤜᴕn
ᴕᘙs
i3ᬙᰝ᠖
i᠘ᤜi᜔
i᜖ᤜ
ᴗ8
�᨞
Dento-alveolar
Diploma de Español como
Lengua Extranjera
(DELE)
Ḟ᠙ᰞ��ᤜi᜞ᤜ἞3
䑩ierential Object Marking
(DOM)
i᠞
ᴗ�1sᴗ�
i᜖ᤜ
ᴗ3
Ἒᬙᰟᬘ
�n᜙ᰟᐟ
Direct Object
i᠙s἞
�ᴙsnᨙs
ᐖᤜ7᜙ᰝ᠞
i3ᴙᰝ3n
i᠘ᤜ
ᴗ2
i8Ἑᰝ᜔ᤜi8�
ἚἙs
἟ḙᰟ�ᐙᰟn8
marking
�ᬙᰝ3n
Doubling
see
Clitic doubling
Dutch
ᴚ᠙ᰝᴖᤜᴟa
�iḙᰟᴟ
Ἕa
Empty complementizer
slter
English
a�ᤜ�a
ᔗ1s
᠞ᤜ᜖ᤜḞḙᰞ�ἙᰞḖᤜḝ᠙s
�἖ᤜ�ᬔ
�ᬖᤜḔ�ᤜḖᨙs
ḕa
ḕ�ᤜḘᨙᰝ2i
ᴚἙs
iᨕ
iḚᤜiḔᤜᴝᘙs
iἔ
ᴟ�ᤜinᨙᰝᬘ
ia2ᤜ
ia�ᤜᴖᨙᰝ7�
ᴖᐙᰝᘘ1s
ᴕ�
ᴕ�ᤜi3i
i᠟1s
i᠗
i᜞ᤜἚ�
Ἒᬙs
἞ḙᰟ��
἞᠙s
�ᬔ1s
Ἓ�
�n᜙ᰟa�
ἔi
Endpoint encoding
ᴟn
ᴟ᜙ᰝnḙᰝᬟ
inᬙs
iᬖ
iᐚ
Ergative
�᠙ᰘ�
3ᴙᰗn
Estar
ᐟᤜa�ᤜ7ᐙᰞᨔ
ḝḙᰞᴟ
ḝ�ᤜḝ᜙s
ḟ�
�nᴙᰞnn
�ᬖᤜ�n᜙s
Ḕ�
Ḕ�ᤜḖ2
Ḗᴙᰟᨗ
Ἕa
Être
�ᬝ
�nᬙᰞᘞ
European Portuguese
iaᔙs
ᴖᨙᰝ7i
Ḗ᜙s
i᠞
i᠕
Event construal
ᴟἙᰝἕᤜinḙs
in�ᤜia2
Event type property
Eventive eect
Evidentiality
iᐙᰝ᠙s
ἕᤜḚ8
Farsi
�ᘗ
ḕᬙᰞ��
ḕ᠙ᰞ᠖ᤜ
�ᜟ
ḗᬙᰝ3i
Feature Reassembly
Hypothesis
iḙᰝἙᰝa
�ᴙᰟᐙᰛ�
nᴙᰞᨘᤜḖ᜙ᰝa�ᤜ
iᐚ
ᴖᨙᰝᔝᤜᴕa
ᴕ�ᤜ
i᠝
i3Ἑᰝ᜚ᤜἚ�
Fill-in-the-Gap Task
἞�ᤜ�ᴚ
Ἕn
French
a�ᤜ��
�᠙ᰞ�ᬙᰞ�a
�἖ᤜ
�ᬝ
�nᬙᰞᬖᤜ�ᬘᤜḔ2
Ḕᴙs
Ḗḙᰞ��ᤜᴚᘙᰝḚᤜiḔᤜ
iaᨙᰝ3i
i3Ἑᰝ᠗ᤜ἞ᬙs
἞᠙ᰟi8
ἔ�
Focus
iḙᰝ�
iᬙᰝᘙᰟᨙᰛ�
aᨙᰔᘙᰖ᠙ᰞᨖ
Ḛ�ᤜḞᨙs
�἖ᤜ�n8
ḔᨙᰞᐟᤜḖ᜙ᰞ�ḙs
ḕ�
ḕᬙᰞ��
ḘἙᰞ3a
ḗḙᰞ8�
ḗᐙᰝ�ᬙᰝᬟᤜ
iᐗᤜᴕᨙᰝ3Ἑᰝ3ᐙᰝ᠘ᤜ
i᜚ᤜi᜔ᤜ἞ḙᰟ�ᬙᰟiḙᰟa�
see also
Contrastive Focus
Formal universal
Fossilization
i᠗ᤜ἞ᬙᰟἚᤜ἟ᬙs
�nᬙᰟᐟ
Full Transfer/Full Access
Hypothesis
iᐘᤜi3i
i3Ἑs
ᴗ2
ᴗ�
Frequency
ᴝᤜnᨙᰔ᜙ᰕa
�ᔙs
3ᴙᰗḙᰗᬙᰗᘙᰗ3
Ḛḙs
Ḛ�ᤜḞᐙᰞḕᤜḝᨙᰞᴝᤜ�nḙs
Ḕᴙᰝi᠙ᰝ἖ᤜinḙᰝnᬙs
iᬖᤜiᬘ
iaᨙᰝ�ᐙᰝ᠗ᤜ἞n
Fricative
��ᤜ᠚
Functional Convergence
Hypothesis (FCH)
Functional convergence
iḙᰝἙs
iᐙᰟḙᰟᐙᰟ�ᤜḚᐙs
Ḛ᜙ᰞi8
Functional
feature
iḙᰝ�
ᴕᤜἕᤜ἗ᤜ
interference
iḙᰝἙs
�ᐙᰟ�
�᠙ᰛᴙs
Genitive
�ᬙᰝ᠖ᤜi᠘ᤜ἟i
Grammatical aspect
ᴟn
Grammaticality Judgment
Task
nᴙᰝ᠗ᤜ἞ᨙᰟ�ᬙs
἞᠙ᰟᴚᤜ�nᬙᰟᬖᤜ
�ᬘ
�n᜙ᰟa�
Scalar Grammaticality
Judgment Task
ᴖἙᰝ7�
Greek
ḕᐙᰝ2i
Haber
7ᐙᰞᨔᤜḞᨙᰞḔ
Hebrew
i3ᴙᰟ2i
Heritage speaker
i᠗ᤜi᜝ᤜ
Ἒ᜙ᰟn8
Hindi
i3ᴙᰟ2ᴙᰟn8
Imperfective
ἕᤜ᜛ᤜ
iἔ
ᴟᘙᰝᬘ
Implicational hierarchy
Indian Spanish
aᨙᰖ8
Indirect object
i᠙s
�ᬙᰝ3a
i᠖ᤜi᜞
i᜔1s
i᜕
i᜗ᤜἚi
Ἒᬙᰟ἞1s
἟ᐙᰟa2
Individual-Level
Predicate
�ᬖ
see also
Stage-Level
Predicate
Interface
iḙᰝ�
iᬙᰝᘙᰟḙs
�ᐙᰟ�ᤜ἗ᤜn�
ᬟᤜ�ᬘ1s
Ḗḙᰞ78
ḕᨙᰞ�n
ḕᘙs
ḗᐙᰝ2i
ᴚἙᰝiᐙᰝᴕ1s
἞ᨙᰟᴚᤜ἟ᬙs
�aḙᰟᐟ
ἔn
Interface Hypothesis
Ḗ᜙ᰝ2ᴙs
἞ᨙᰟᴚ
Interlanguage Hypothesis
Interpolation
Italian
Ḟᴙᰞi᠙ᰞᘚᤜḕḙs
�ᔛ
ḕᘙᰝ2ᴙᰝ2n
i�ᬙᰝᴟᤜᴝᐙᰟ�ᴙᰟ἗ᤜ
ἔ�
Island
iᘝ
ᴖᬙᰝᘖ
ᴕᴙᰝ�n
ᴕa
Index
Japanese
�ᘙᰟ἞
�Ἕ
L2 acquisition
iᬙᰔ᜙ᰞ�Ἑᰞᔚᤜ
inḙsia�
iᐗᤜᴖḙsᴕ�
ᴕ᠙s
i᠝
i3Ἑᰝ᠗
i᜚ᤜ
἞ᨙᰟ��
἞ᐙᰟi8
἟ᬙᰟ�᠙ᰟᬖᤜ�ᬘ
Language contact
ᴝᤜiᬙᰟ�ᤜ
ḚᘙᰞᨘᤜḞ�
Ḟᐙᰝ3�
Language identity
�ᨕ
Language Specic Grammar
Hypothesis
Latent Linguistic
Structure
Latent Psychological
Structure
Lexicon
iᬙᰝᘙᰟᨙᰟ�ᤜ἗ᤜ
ᬟᤜᐘᤜᘘᤜ�἗ᤜi᠞ᤜ
἟�ᤜ�ᬔ
Lexico-morphosyntactic
interface
Liquid
᠞ᤜ3a
Locative
�ᬙᰘḙᰞᨔᤜḚ�
ḝ�ᤜḝ᜙ᰞnḙs
�ᬔᤜḔᴙᰝ἗
inᨙᰝᬟᤜi3ᐙs
἞ἙᰟḔ
expression
�ᬙᰞᴚᤜ�nḙᰝᬟ
Manner
ἕᤜḞ�ᤜḔᬙᰝ἟
ᴟᐙs
inḙᰝᬟ
inᔙs
iᬗ
iaᨙᰝ��ᤜi3�
Mapping
iḙᰝ�
ᴕᤜἚ
�ᐙᰟ�
ᬟᤜᐖᤜᐘᤜ7᜙s
Ḛ᠙ᰞḖᤜḝ᠙ᰝᐗ
�ᬖᤜ�a�
Mapudungun
�nḙᰟᐟ
Microvariation
ᴚḙᰝᨖᤜ
iḚᤜᴝ7
Minimal Word
Minimalist Program
Missing Surface I渞ection
Hypothesis
7᜙ᰞḖᤜ἞n
Modularity of language
Mood
aᴙᰔᐙᰞ�ḙᰝ᠞
Morphological coda
�᠙ᰘᴙs
Motion event
ᴟ�ᤜ
inᴙᰝn�ᤜia2
N-N compound
�ᬝ
�ᬛ
Nahuatl
aᴙᰔn
a�ᤜᐗ
7Ἑᰖ�
ᘘᤜ��
Nasal
��ᤜ᠚
᠞ᤜ3a
�᨞
Near-native
�ᘗ
ḕᨙᰞ�n
ḕ᠙s
ḗἙᰞ᜔
Nominative
i᠖ᤜἚ�
Norwegian
�ᘙᰕ᠙ᰝ�a
Nucleus
�ᐙᰕ᜙ᰝ3a
Null
object
iḙᰟ8
nᨙᰛᴙᰝ7ᬙs
ᴕᨙᰝᔝᤜᴕn
pronoun
i᨟
iḕ
iḗᤜᴝ�
ᴝᬙᰝ3ᐙᰟa2
subject
ᴕᤜᐟᤜḕ�
ḕἙs
ḕᐙᰞ᠚
Ḙᬙᰞ᠗
ḗᐙᰝ2i
ᴚἙᰝᨔ
ᴚ᜙ᰝ�n
iḔᤜᴝᨙs
ᴝᬙᰝᴖᤜ἟i
�nḙᰟa�
ἔ�
Number
i᠙ᰟᨙᰔi
ᐔ1s
7ᐙᰖ�ᤜ7᜙ᰘ�
᜖1s
��ᐙᰞᬘᤜḔᨙᰞaᴙᰞᐘ
Ḕ᜙s
ḖἙᰞ�ḙᰞ3i
ḘἙᰞ3ᐙᰞ8Ἑs
ᴚᘙᰝḚᤜi�Ἑᰝi᠙ᰝ�n
ᴟᐙs
inᨙᰝᬖ
in�ᤜᴖḙᰝ7ᬙᰝ7᜙s
ᴕḙᰝ᠞ᤜi᜚ᤜi8�ᤜ἞ᨙᰟ�ᬙs
἞᜙ᰟ�ᴙᰟ἖ᤜ�nᬙᰟᬖ
Oto淭
aᨙᰔ7
Oblique
i᠙ᰟi
�ᬙᰟ�ᤜnᴙᰞᨘ
Object
agreement
clitic
see also
clitic
drop
see also
null object
Old Spanish
἟3
Ongoingness
ᴟᘙᰝᬘ
iᐚ
ᘞᤜᔗᤜ3Ἑᰗ᠙ᰝḔᤜ
ᴝ�ᤜi3a
Optionality
aᨙᰞḚᤜᴚᴙs
ᴝᬙᰝᴖᤜἚᨙᰟᨗᤜ�i᜙s
἟n
἟ᐙᰟ἗ᤜ�nᴙᰟᐟ
Overt pronoun
i᨟
ᴚ᠙ᰝḚᤜ
iḔᤜiḕ
ᴝḙᰝᴟ
ᴝᬙs
ἔi
Past tense
ᴟᤜi᠙ᰟᨙᰟᐙᰔn
7ᬙᰘ�
Path
ḞᨙᰞnᨙᰞᬘᤜḖἙᰝ�n
ᴟᐙs
inᨙᰝᬖᤜin᜙ᰝ��
Perfectivity
Permanent property
ḟ�
��᠙s
�nḙᰞᬔ
�ᬖᤜḔᴙᰞᐔᤜḔ�ᤜ
Ḗḙᰟḝᤜ�in
Person
i᠙ᰟᨙᰟ�
�ᬙᰛᨙs
ᐖᤜᐗᤜ7i
3ᴙᰗᬙᰗ᜙ᰞ2�ᤜḞᨙᰞḘᤜ
Ḕᨙᰞ��
ḕᴙᰞ�ᬙᰞ�3
ᴚḙᰝ2ᬙᰝᨖ
ᴚ�ᤜiḚᤜ
ᴝ᠙ᰝ἖ᤜinᴙᰝᘞᤜᴖᬙs
ᴖᘙᰝ�ḙᰝ3ᐙᰟḞᤜ἟ᐙᰟ἗ᤜ
�nᴙᰟn�
Person-Case constraint
(PCC)
Ἓ�
Phonological
process
�ᐙᰞᨚ
�᨞
system
�ᘙᰞᨚ
Pluperfect past tense
Plural marking
��ᤜ�ᬖᤜ�ᐟ
Position of Antecedent
Hypothesis
ᴚᬙᰟa2
Praat
Pre-emption
ia�ᤜiᐗ
ᴕ7
ᴕ�
Predicative adjective
��ᐙᰞᐟᤜ
Predorsal
��ᤜ᠚
Preposition
i᠙ᰟᴙᰞḚᤜḝ�ᤜi᠞
i3ᴙs
i3ᬙᰝ᜛
i᜔ᤜi᜘ᤜ�n8
἟ᤜ�ᬖ
pro
ᐛᤜᐖᤜḕ�
ḕἙᰞ�᠙s
i᨟
ᴚᐙᰝᨗᤜiḞᤜiḕ1s
ᴕᬙs
i᠔
i᠖ᤜἚn
pro
-drop
aᴙᰔᘙᰝᨗᤜi᠖ᤜ
�ᨛ
Processing
ᴝᤜḖ8
ḕᨙᰞ�ᬙs
ḕ7
ḕ�ᤜḘᨙᰞ8�
ḗᐙᰝ2ᴙs
ᴚᬙᰝin
ᴝᐙᰝᘘᤜi᠞1s
Ἒ᜙ᰟi᜙ᰟ�ᬙᰟn�1s
�aᴙᰟᐛ
load
Production of morphology
aᨙs
Index
Progressive
�᠙ᰔἙᰞ�Ἑᰞnᨙs
iἔ
ᴟᘙᰝᬟᤜiᬘᤜ἞�
aspect
iἔ
Quantier
i3ᬙᰟnḙᰟᬔ
�nᔙs
ἔ�
Quechua
iḙᰝ�
Ḛ᜙ᰞḕᤜḝ�
Resyllab検cation
ᔗᤜ3�
Rhyme
Romanian
Ḟᘙᰝ3ᴙᰟ2ᴙs
Ἓ�
�ᬘ
Russian
Ḟᘙᰞi᠙ᰞ�ᐙᰝ�ᐙᰝ3ᴙs
�Ἕ
Satellite-framed language
ᴟ�ᤜin8
Scalar gradable adjective
἞ᨙᰟḕᤜ�iḙᰟᴟ
Ἕn
Irreversible scalar
gradable
἞ᨙᰟḕᤜ�ᴟ
Ἕn
Non-scalar gradable
�iḙs
Ἕ�
Ἕn
Reversible scalar
gradable
Ἕ�
Ἕn
Sein
a�ᤜ�ni
�nᬙᰞᘞ
Selective access
Non-selective access
Ser
a�ᤜḚa
ḝḙs
ḝn
ḝ�ᤜḝ᜙ᰞἕ
�nᨙᰞnᴙs
�ᬛ
�ᬖᤜ�n᜙ᰞa�
Ḕ�ᤜḖ2
ḖᴙᰞᔗᤜḘḙᰞ᜞ᤜ
Ἕa
Ser/estar
+ adjective
Sibilant
�ᐙᰕ�ᤜ᠚ᤜ3a
Ḛḙᰞ2�
Spanish
iḙᰝ�
ᬟᤜn8
ᘛᤜ
7᜙ᰕἙᰕa
Ḛḙs
ḝἙᰞia
ḝ᜙s
�἟ᤜ��a
�nᴙᰞnn
ḖἙs
�ᘗ
ḕ᠙ᰞ3a
Ḙᘙᰞ8�
ḗᬙs
i᨞
iḚᤜi�n
iḖᤜiḘᤜᴝᨙᰝᴝ
ᴝ᜙s
ᴟᐙᰝ�3
iaᨙᰝᐔ1s
ia�ᤜᴖ2
ᴕ�ᤜi᠞
i᜝1s
ᴗa
i᜖ᤜi᜘ᤜἚᨙs
�᨝
Ἒᬙᰟᨗ
Ἕ�1s
Ἕ�
�iᐙᰟi8
ἔn
Spec
ḛ�
�n�1sἔ2
Specicity
i᠙ᰞi᠙s
ᴖᴙᰝᔚ
ᴕḙᰝ�ᘙs
i᠝
i᠘ᤜi᜚ᤜἚᨙs
�᨝
�ᨛ
Non-specic
ᴕᨙᰝ3n
Specic
iᬙᰝ᠙ᰛᴙᰕa
�ᔙs
᠞ᤜ3ᐙᰘ�ᤜ᜗ᤜḚᨙᰞi᠙ᰞἕᤜ
ḕᨙᰝᨖᤜiḔᤜiḗᤜᴝᴙᰝiᐙs
ᴝ�ᤜᴟᐙᰝἕᤜin�ᤜiᐘᤜᴖḙs
ᴖᴙᰝ7᜙ᰝ��ᤜi᠞ᤜi3ᴙs
i3ᬙᰝ3ᐙᰝ᠕
i᜞ᤜi8Ἑs
Ἒᨙᰟ2ᴙᰟ2ᬙᰟᨗᤜ἞a
἞�ᤜ
἞᜙ᰟᴚᤜ�iᬙᰟi᜙ᰟnḙs
�ᬛ
�ᬖᤜ�n᜙ᰟai
ἔ�
Stage-Level Predicate
�nᬙᰞᬖ
Stop
��ᤜ᠚
᠞ᤜ3a
᠘ᤜ᜘ᤜḚ�
Stress
�᠙ᰘᨙᰘᴙᰗ2
Ḛḙᰝn8
Stressed
�᠙ᰘᨙᰘᴙᰘᐙs
8ἙᰟḔ
Unstressed
�᠙ᰘᨙᰘᴙs
3ᐙᰗ2
8Ἑᰗᐙᰗ8
Subject
agreement
ᴖᤜaᴙᰔa
overt
ḕ�
ḕᐙᰞ᠚
Ḙᬙᰞ᠗ᤜ
ᴚᐙᰝᨘᤜ�nḙᰟa�
ἔi
pronoun
ᐛᤜḖ᜙ᰞᔚ
ḕḙs
�ᔛ
ḕᐙᰞᔗᤜḘᐙᰝ2�
ᴚἙs
iḚᤜi�ᬙᰝi᠙s
�ᬚ
�nḙᰟa�
ἔi
Subject-verb inversion
�ᬚ
Ἓ�
iᐗᤜᴖa
Swedish
ᴟᘙᰝ�3
ᴛ�
iaᨙᰝ3i
Syntactic complexity
Ḕḙᰞᘚ
Syntax-pragmatics
interface
Syntax-before-discourse
hypothesis
TAM
Temporal property
ḟ�
��᠙s
�ᬟ
�ᬖᤜḔḙᰞᐛ
Ḕ�ᤜ
Tense
ᴟᤜiᐙᰝ᠙ᰟᨙᰟᐙᰔi
a�ᤜᐗ1s
7ᬙᰖ�
ᘘᤜ᠞ᤜḞᨙᰞḔ1s
��᠙ᰞ�ḙᰝᨖᤜiᬘ
in᜙ᰝ᠞
�at-trace eect
἟ᴙᰟn2
�nḙs
ἔi
ἔ�
Topic
iᬙᰟᨙᰛᨙᰔᬙᰞᨖᤜ
Ḗ᜙ᰞ��
ḕᐙᰞ��
Ḙᐙᰞ᜚ᤜ
ḗᐙᰝ2i
ᴚ᜙ᰝ�Ἑs
ᴝᬙᰝᴕ
ᴝ᠙ᰝᬘᤜᴖ᜙s
i᠕ᤜ἞ᴙᰟa2
Topic S桩1
᜙ᰞᨙᰞ7᜙s
ḕ�
ḕᐙᰞ��
Ḙḙᰞ3ᬙᰞ3ᐙs
ḗᨙᰞ8�
ḗᐙᰝ2Ἑᰝᨔᤜᴚ᠙s
Topic Maintenance
Ḗ᜙ḕḙᰞ�Ἑᰞ�ᬙᰞ�3
�᠝
Ḙᐙᰞ᜝ᤜḗᬙᰞ᜔
Turkish
Ḟᘙᰞi᠙ᰞ�ᘙᰝᨘᤜ
ᴝᘙᰝ᠞
i3ᬙᰝ᠖
i᠘ᤜ
ᴗ2
i8Ἑᰝ᜖
i᜘ᤜ
Ἒᬙᰟn8
Typological
proximity
Ἕ8
�ᬔᤜ
Ἓ�
�n᜙ᰟai
ἔ�
similarity
�i᜙ᰟ἞
἟ᐙs
�nḙᰟnᬙᰟn�
�n᜙s
ἔi
ἔ�
Universals
�i᜙ᰟ἞
Universal Base Hypothesis
ḛ�
Universal Grammar
iᐘᤜ἟2
Hypothesis
�ᘙᰘ�
Verb root
ᐔᤜᴟ3
Verb-framed language
Vibrant
Visibility
Voiceless
��ᤜ3ᴙᰘ7
᠕ᤜ᜘ᤜ
Ḛᨙᰞ἗
Weak crossover eect
nḙᰛi
Word order
ᴕᤜἚᤜnḙᰔᬙᰔᘙs
Ḕ᜙ᰞ�ᐙᰝᨖᤜi᠖
i᠘ᤜ
Ἒᬙᰟ἞ᤜ἟ᐙᰟnn
Word length
Worden
἞ᘙᰟiḙs
Ἕ�
Ἕn
Zijn
἞ᘙᰟiḙs
Ἕ�
Ἕn
�e Acquisition of Spanish in Understudied Language Pairings

Приложенные файлы

  • pdf 15765918
    Размер файла: 3 MB Загрузок: 0

Добавить комментарий