HLS 2013 University of Ottawa ABSTRACTS
HLS 2013
University of Ottawa
ABSTRACTS
2
The Proper Formulation of the Overt Pronoun Constraint: Evidence from Spanish
Calixto Aguero-Bautista
Université de Québec à Trois-Rivières
—The Problem: The overt pronoun constraint (OPC) (Montalbetti 1984) states that an overt pronoun
cannot be construed as a bound variable in contexts where it alternates with a null pronoun, as shown in
the Spanish example in (1). There are contexts, however, where the null /overt alternation does not
obtain and yet some overt pronouns cannot have a bound variable interpretation in the given contexts.
Thus, while the overt pronoun él ‘him’ in (2) cannot have a bound variable interpretation, sus ‘his/her’
in (3) can. This difference seems related to the fact that in the relative clause in (4), the overt pronoun
sus does not trigger a weak crossover violation, but él in (5), does. The question then is: why does él,
but not sus, resist a bound variable interpretation in the contexts in (1)-(5) and what goes wrong in
narrow syntax or any of the interfaces when pronouns of the same type as the former are used?
— roposal will ar ue, ollowin ero-Bautista (2013) (hereon AB), that in the bound-variable
interpretation of a determiner pronoun, the NP complement of the pronoun is a copy of the NP part of
the antecedent that binds it. The copy of the antecedent NP is merged with the pronoun before major
work on the working space begins in a process akin to what Nuñez (2001) calls sideward movement.
Since the NP part of the pronoun is a copy of the NP of the antecedent, a PF-interface strategy of
deletion will delete the complement o the pronoun in order to avoid a potential violation o Kayne’s
1994 LCA when the chains are finally linearized (along the lines of Nuñez). However, since the
phonetic of the Φ-features of the deleted NP survive in the phonetic realization of the pronoun, an LCA
violationwith respect to the phonetic realization of the Φ-features will still occur. I argue, however, that
the PF component can only see the violation if the pronoun DP is linearized in the same spell-out
domain as one of the copies of its antecedent (Aguero-Bautista 2012), as that would entail a domain-
internal violation of the LCA. I assume with AB that this is the proper formulation of the WCO
constraint.
—Predictions: The analysis predicts that only overt determiner pronouns can lead to WCO violations in
that their complement NPs will be copies of the NPs of their antecedents that need to be linearized and
comply with the LCA. By contrast, pronominal elements belonging to the category of adjectives or
nouns, or phrasal reflexives (DPs in which the NP part is filled with a nominal of the self type) will not
induce a violation of WCO as the NP in their DPs cannot be copies of the NPs of their antecedents,
hence no violation of the LCA is possible in such cases. Because of their potential for violating the
LCA, overt determiner pronouns are costly. Their use requires that the structure be inspected for
compliance with the LCA. No similar inspection of the structure is required in the case of Non-
determiner overt pronouns and null pronouns, since there’s no potential or violatin the LC in such
cases. The grammar thus prefers the latter type of pronouns over the former. It follows then that overt
determiner pronouns can only be used in the grammar as a last resort. That is, when there are no non-
determiners or null pronouns that could be used in their place. This is, I argue, the right formulation of
the Overt Pronoun Constraint. This analysis further predicts that only pronouns that trigger WCO will
be subject to the overt pronoun constraint and that overt determiner pronouns will be blocked not only
by null pronouns, but also by overt non-determiner pronouns. These predictions are confirmed by the
examples in (1)-(5). The contrast in (4) and (5) shows that él ‘him’ leads to WCO in relative clauses,
whereas sus ‘his/her’does not. While él is a determiner pronoun, su can be considered a possessive
adjective, as shown by the fact that in some dialects of Spanish it can co-occur with a preceding
determiner or demonstrative (e.g., esta su casa ‘this his house’) ( icallo and Ri au 1999). There ore,
only su is predicted to be exempt from WCO. Notice now that el in (2) cannot have a bound-variable
interpretation, despite the fact, that no null-pronouns are acceptable as complements of prepositions
3
(Montalbetti 1984). The un rammaticality o (2) is unexpected under Montalbetti’s O C, but expected
under the current analysis. Because o the possibilities o examples like (7) and (8), with su or suyo ‘o
his’, in place o él, the latter pronoun is excluded. I will show that this analysis can explain cases of
unexpected (un)grammatical occurrences of overt pronouns in languages like Japanese and Turkish.
—The Proper Formulation of the Overt Pronoun Constraint: Evidence from Spanish.
(1) Nadiei cree que proi/el*i/j es inteligente
Nobody believes that pro/him is smart
‘Nobody believes that he is smart’
(2) Nadiei habló de los amigos de él*i/j
Nobody spoke of the friends of him
‘Nobody spoke o his riends’
(3) Nadiei habló de susi problemas
Nobody talked of his problems
(4) Ningún niñoi a quieni sui madre maltrate será acceptado sin examinar
No childi to whomi hisi mother mistreat will-be accepted without examining
‘No child to whom his mother mistreat will be accepted without examinin him’
(5) Ningun niñoi a quieni la madre de él*i/j maltrate sera acceptado sin examinar.
No childi to whomi the mother of him*i/j mistreat will-be accepted without examine
(6) Nadiei habló de suis amigos
Nobody spoke of his friends
(7) Nadiei habló de los amigos suyois
Nobodyi spoke of the friends hisi
Selected References
Aguero-Bautista, Calixto. 2013. (Weak) Crossover Phenomena and Language Design. Ms, University
of Quebec at Trois-Rivieres.
Kayne, Richard S. 1994. The antisymmetry of syntax. Cambridge, Mass.: MIT Press.
Montalbetti, Mario. 1984. After binding. Doctoral dissertation, Cambridge, Mass.: MIT.
Nuñez, Jairo. 2001. Sideward movement. Linguistic Inquiry, 32: 303-344.
4
Processing ungrammaticalities: Evidence from native and non-native Spanish
Anahí Alba de la Fuente
Université de Montréal
Research within the generative framework has identified a number of restrictions with respect to the
combinatorial possibilities of clitic clusters across languages. Languages such as Spanish and
Romanian accept combinations of 1st and 2nd person clitics (1a), which contrasts with languages such
as French, where these combinations are not acceptable (Bonet 1991). In turn, not all 1st/2nd
combinations are possible in Spanish and Romanian, since such combinations are ungrammatical when
both clitics are plural (Nevins and Săvescu, 2010; lba de la Fuente, 2010).
(1) a) Tú te nos presentaste en la fiesta. [SP] // *Tu te nous es présenté à la fête. [FR]
'You (pl) introduced yourselves to us at the party'
b) *Vosotros os nos presentasteis en la fiesta.
'You (pl) introduced yourselves to us at the party'
(2) *Vosotros te nos presentasteis en la fiesta.
'You (pl) introduced yourself to us at the party'
In contrast with sentences such as (2), whose ungrammaticality is due to an agreement mismatch
between the reflexive clitic and its antecedent, sentences such as (1b) meet all agreement requirements
and, yet, they are ungrammatical. Thus, whereas the nature of the ungrammaticality is clear in (2), the
nature of the one in (1b) is still under study.
Thus, in order to further investigate the combinatorial restrictions displayed in (1), in view of the
different nature of the ungrammaticalities in (1) and (2), and given the fact that different languages
display different clustering restrictions, this study explores the following research questions: 1) whether
the ungrammaticalities in (1b) and (2) are processed differently, and 2) whether there are processing
differences between native and non-native speakers with respect to these two types of
ungrammaticalities.
The results of a self-paced reading task administered to four groups of participants (L1 speakers of
Peninsular Spanish and three groups of L2 speakers of Spanish (L1English, L1French and
L1Romanian, respectively) show significant differences both between the native and non-native
speakers and between the two types of ungrammaticality. Specifically, the native group displayed the
longest reading times (RTs) for the ungrammaticality shown in (1b), whereas the longest RTs in the L2
groups corresponded to the purely syntactic ungrammaticality expressed in (1a). These results answer
affirmatively to both research questions.
To conclude, considering that much of the debate on clitic restrictions revolves around the issue of
whether they are syntactic or morphological in nature, these results are of particular interest, as they
constitute the first attempt to discriminate between this type of ungrammaticality and a purely syntactic
violation, such as an agreement mismatch between a reflexive clitic and its antecedent.
References
Alba de la Fuente, A. (2010). ‘More on the clitic combination puzzle Evidence rom Spanish, Catalan
and Romanian’. n Colina, S., Olarrea, A and Carvalho, A. M. Romance Linguistics 2009. Selected
papers from the 39th Linguistics Symposium on Romance Languages (LSRL), Tucson, Arizona,
March 2009. Amsterdam: John Benjamins, 203-215.
Bonet, E. (1991). Morphology after syntax: Pronominal clitics in Romance. Tesis doctoral, MIT.
Nevins, . and Săvescu, O. (2010) ‘ n pparent Number Case Constraint in Romanian The Role o
Syncretism’. Romance Linguistics 2008. Selected papers from the 38th Linguistic Symposium on
Romance Languages (LSRL), Urbana-Champaign, April 2008 (pp. 185-200). Amsterdam: John
Benjamins.
5
The rise of dative experiencers: Historical changes in Spanish collocations with entrar and state
nouns
Josep Alba-Salas
College of the Holy Cross
In Modern Spanish the verb entrar ‘enter’ combines with nouns desi natin physical or psycholo ical
states, such as miedo ‘ ear’, celos ‘jealousy’, hambre ‘hun er’ and calor ‘heat’. n these structures the
state noun controls verb agreement on entrar, and the person or entity experiencing the state designated
by the noun (the experiencer) is realized as a dative clitic that may optionally co-occur with a co-
referential NP introduced by the preposition a ‘to’. This is illustrated in (1), which eatures the plural
state noun celos ‘jealousy’ (literally ‘jealousies’).
(1) Al verlo con Ana, (a Evai) lei entraron muchos celos.
upon see-him with Ana to Eva to-her -DAT entered-3rdPL many jealousies
‘When she saw him with na, Eva ot very jealous.’
Structures of the type le entraron celos ‘she ot jealous’ (literally, ‘jealousies entered to her’) have
been analyzed as inchoative collocations in which the state noun (the base of the collocation) lexically
selects entrar (the collocate) to indicate the moment in which an entity begins to experience the
physical or psychological state designated by the noun (Alonso Ramos 2004: 108). These inchoative
collocations with entrar are particularly interesting from a diachronic perspective because throughout
the history of Spanish they have been associated with two other syntactic schemas besides the dative
experiencer construction illustrated in (1), i.e. structures with a subject experiencer (e.g. el valeroso
caballero entró en temor ‘the coura eous kni ht ot scared’, literally ‘the coura eous kni ht entered
into ear’), and structures with locative experiencers (e. . entró temor en el valeroso caballero ‘the
coura eous kni ht ot scared’, literally ‘ ear entered into the coura eous kni ht’).
Using data from the Corpus del español (Davies 2002-), my study examines the historical
distribution of these three types of entrar collocations from the 1200s to the year 2000. The results
show that the locative experiencer structure predominated between the 1300s and the 1500s, and the
subject experiencer construction prevailed in the 1600s and 1700s, but dative experiencers became the
norm starting in the 1800s. This historical evolution is interpreted as the result of other, more general
lexical and grammatical changes in Spanish, including the demise of many collocations of the type en
‘in’ + state noun (e. . en temor ‘in ear’) and, more importantly, the rise o dative experiencers as the
default option in constructions indicating involuntary physical or mental processes in Spanish (cf.
Melis 1998, Vázquez Rozas y Rivas 2007, Rivero 2010, Elvira 2011).
References
Alonso Ramos, Margarita (2004). Las construcciones con verbo de apoyo. Madrid: Visor Libros.
Davies, Mark (2002-). Corpus del Español (100 million words, 1200s-1900s). Available at
<http://www.corpusdelespanol.org>. Elvira, Javier (2011). “Constructions o Uncontrolled State or
Event The ncrease in roductivity o a New r ument Structure in Old Spanish”. Constructions and
Frames 3(2), 184–207. Melis, Chantal (1998). “Sobre la historia sintáctica de gustar”. n Claudio
García Turza, Fabián González Bachiller and José Javier Mangado Martínez (eds.), Actas del IV
Congreso Internacional de Historia de la Lengua Española, vol. 2. La Rioja: Universidad de la Rioja.
295-306. Rivero, María Luisa (2010). “Los verbos psicoló icos con experimentante dativo en español
y el cambio histórico”. Cuadernos de la ALFAL (Asociación de Lingüística y Filología de la América
Latina), Nueva Serie 1, 167–193. Vázquez Rozas, Victoria, and Elena Rivas (2007). “Un análisis
construccionista de la diacronía de gustar”. In Iraide Ibarretxe-Antuñano, Carlos Inchaurralde and
Jesús Sánchez-García (eds.), Language, Mind, and the Lexicon. Frankfurt: Peter Lang. 143-164.
6
Degree modification and indefiniteness in Spanish: a diachronic link
Patricia Amaral
University of North Carolina at Chapel Hill
This paper analyzes the morpho-syntactic and semantic change undergone by the indefinite pronoun
with inanimate reference algo ‘somethin ’, as in (1), in the history o Spanish. The ocus is on the
relation between the properties of the indefinite pronoun and the emergence of the degree adverb algo
meanin ‘a bit, a little’, exempli ied in (2). show how the reanalysis o the indefinite pronoun took
place in constructions with a partitive interpretation. In order to explain the semantic and syntactic
change, this paper explores in particular (i) the role of the quantity implicature associated with the
algseries of indefinite pronouns in Ibero-Romance, and (ii) the relation between the mereological
properties of classes of nouns and gradability in the emergence of the degree interpretation.
The degree interpretation of algo developed from the semantic component of the indefinite pronoun
(existential quantification) and the conventionalization of a generalized conversational implicature, by
pragmatic enrichment (Haspelmath 1997). This implicature was associated with the alg-series of
indefinite pronouns in Ibero-Romance, specifically, the quantity implicature ‘somethin but not
everythin ’ ( uilar-Guevara et al. 2010; Eckardt 2006). In Ibero-Romance, the form algo developed
as part of a series of pronominal forms, together with algún ‘some’ and alguien ‘somebody’ (Malkiel,
1948). These forms participate in a Horn scale formed by existential,
proportional, and universal quantifiers, as in (3), adapted from Gutierrez-Rexach (2010: 689, ex. [31]).
Corpus data show that diachronically, the indefinite pronoun occurs initially (13th-15th centuries) in a
range of constructions in which it is modified, in particular in a partitive structure with the preposition
de ‘o ’ and an N . These constructions are syntactically and semantically ambi uous, in that either
algo may be analyzed as a head (a pronominal element modified by a PP), or the sequence [algo de]
may be analyzed as a modifier of a nominal element (i.e. as the adjunct of an NP), as in (4a,b).
Sequences of the type algo de + Noun could denote both a part of a whole and the degree of a property.
The reanalysis of the pronominal element as a degree modifier was facilitated by analogy with nominal
degree expressions common in Ibero-Romance, e.g. bastante de + Noun, muchas/os de + Noun
(Camus Bergareche 2009, Keniston 1937) and by the contiguity of semantic features (scale structure)
shared by the nouns that occurred in the partitive constructions and other categories, like adjectives.
Once the degree interpretation emerges, the degree modifier algo extends its distribution from nouns to
adjectives and gradable verbs (Doetjes 2008, Kennedy and McNally 2005).
These diachronic findings show that the quantificational expression is the source of the degree word,
not the other way around (corroborating the hypothesis of Doetjes 2008). The account I propose
provides an explanation for this direction of change: the strengthening and conventionalization of the
quantity implicature generally associated with quantificational expressions underlies the reanalysis of
algo to a degree interpretation. In this respect, algo shows a similar development to that of degree
words resulting from measure constructions containing nouns, e.g. a bit from a bite (of) in English
(Traugott 2008). Furthermore, the collocations of the indefinite pronoun in diachrony provide a
privileged test case to study the semantic notion of gradability across lexical categories and the possible
paths undergone by indefinites in language change.
(1) Si el marido quisiere dar algo ala mu ier (Corpus del Español, 13th century) ‘ the husband wants
to ive somethin to [his] wi e,…’
7
(2) Su madre se excusó; estaba algo cansada. (Corpus del Español, 20th century)
‘Her mother apolo ized she was a bit tired’
(3) < todos ‘all’, la mayoría ‘most’, algunos ‘some’>
(4) la materia…sera çitrina con algo de vermejura (Corpus del Español, 15th century)
a) ‘[The matter comin out o the bladder] will be yellow-green with some element or other that is
red’ ( nde inite pronoun interpretation)
b) ‘[The matter coming out of the bladder] will be yellow- reen with a bit o red’ (De ree
modifier interpretation)
The corpus data are from the Corpus del Español: http://www.corpusdelespanol.org
References
Aguilar-Guevara, A. et al. 2010. “Semantics and pra matics of indefinites: methodology for a
synchronic and diachronic corpus study”. Available at: staff.science.uva.nl/maloni/DGfS2011.pdf .
Camus Bergareche, B. 2009. “Cuanti icadores . Los cuanti icadores propios”. Sintaxis histórica de la
lengua española: La frase nominal ed. by C. Company Company, 883-956. Mexico: Universidad
Nacional Autónoma de México.
Doetjes, J. 2008. “ djectives and de ree modi ication”. Adjectives and Adverbs. Syntax, Semantics,
and Discourse ed. by L. McNally and C. Kennedy, 123-155. Oxford: Oxford University Press.
Eckardt, R. 2006. Meaning change in grammaticalization: An Enquiry into semantic reanalysis.
Oxford: Oxford University Press.
Gutierrez-Rexach, J. 2010. “ nde inites in Spanish”. Language and Linguistic Compass, 8(4):680-693.
Haspelmath, M. 1997. Indefinite pronouns. Oxford: Oxford University Press.
Keniston, H. 1937. The syntax of Castilian prose. The Sixteenth Century. Chicago: The University of
Chicago Press.
Kennedy, C. & McNally, L. 2005. “Scale structure and the semantic typology of gradable predicates”.
Language 81:345-381.
Malkiel, Y. 1948. Hispanic algu(i)en and related formations: A study of the stratification of the
Romance lexicon in the Iberian Peninsula. Berkeley: University of California Press.
Traugott, E. 2008. Grammaticalization, constructions and the incremental development of language:
Suggestions from the development of degree modifiers in English. In Regine Eckardt, Gerhard Jäger,
and Tonjes Veenstra, eds., Variation, Selection,
Development--Probing the Evolutionary Model of Language Change. Berlin/New York: Mouton de
Gruyter, 219-250.
8
‘Matar’ and ‘Morir’: a formal analysis of root allomorphy
Grant Armstrong
Jason Doroga
University of Wisconsin-Madison
Corominas & Pascual identify muerto as an early suppletive participle of matar (1980-91: s.v. matar).
This form is found in the earliest documentation of Old Spanish, continuing through the medieval
period. Though the participle muerto is preferred in perfective constructions with haber and analytic
passives with ser in Old Spanish shown in (1) and (2), it is considered to be a perfective adjective on
par with seco and vacío in the modern language. In this paper we argue that the traditional idea that
muerto is a suppletive form of matado should be extended more enerally to ‘matar’ and ‘morir.’ Usin
the model of Distributed Morphology and root allomorphy outlined in Siddiqi (2006, 2009), we argue
that vocabulary items associated with a single conceptual (root) domain, mainly ‘bein dead’, such as
matado, matar, morir, muerto, murió, etc, compete for insertion into a syntactic context based on the
features of the functional heads above the conceptual domain. We contend that this model explains two
unexpected distributional facts about morir, which are discussed below.
Our first argument, following the observation in Corominas & Pascual, is based on the behavior
‘matar’ and ‘morir’ in the analytic passive construction. As shown in (3), the analytic passive of matar
is formed, preferably for most speakers, with the participle of morir (= muerto) rather than matar (=
matado). This would be unexpected if matar were listed as a transitive verb and morir as an intransitive
one. It is important to note that the passive in (3) is agentive and permits a by-phrase, which would rule
out treating muerto as a kind of special, agentless passive. Out o 210 occurrences o ‘was killed’ in a
newspaper corpus of modern Spanish, 194 (92.4%) are ‘ser+muerto’ and 16 (7.6%) are ‘ser+matado.’
In order to account for this unexpected fact, we claim that the participle in passives contains a v,
Voice[+Ag, -ExtArg] and Asp layer (Alexiadou 2010; Embick 2004) as shown in (4a). The root that is
embedded under this structure is simply a numerical index (= √234) that accesses a correspondin list
of vocabulary items. Following Siddiqi (2006, 2009), we claim that the numerical index fuses with the
heads dominating it, forming a single node for vocabulary insertion. The vocabulary insertion
mechanism then matches the vocabulary item that is specified for the greatest number of features on the
complex head. Grammar 1, the most common one in Spanish, calls for the insertion of muerto in this
context (4c), while Grammar 2, far less frequent, calls for the insertion of matado (4d).
Our second argument is based on the pronominal use of morir (5). This particular use of morir is
puzzling if it is treated as a pure unaccusative verb. The RAE (2009) distinguishes at least two types of
causative alternations: a lexical one (6a), where the two variants are realized by different verb roots,
and a morphological one (6b), where the intransitive variant is marked by SE. As can be observed in
(6a), morir appears to be an exception to the generalization that an intransitive change of state is either
realized by a different root than the transitive variant or marked by SE, but not both. We argue that
morir-se is simply an anticausative of matar that involves the reflexivization of a stative causer
(Chierchia 2004; Koontz-Garboden 2009), which is realized morphologically by SE in Voice[-AG, -ExtArg]
(Alexiadou 2010), shown in (7a). This explains why morir-se is interpreted as “death by an internal
cause”, rather than by some external force. The mechanism of vocabulary insertion is the same as
outlined above. A numerical index that corresponds to the conceptual domain o ‘bein dead’ (= √234)
is fused with a verbalizing layer and a Voice[-AG, -ExtArg] head. This particular feature set is matched with
an appropriate vocabulary item from the list in (7b).
We conclude that these particular acts re ardin the distribution o ‘matar’ and ‘morir’ in Spanish
provide a new argument in favor of a non-lexicalist approach to syntax and specifically one in which
9
roots, in addition to functional vocabulary items, are inserted after syntactic operations (Siddiqi 2006,
2009).
(1) et mando [thamaris] adozir un odre lleno de sangre de omnes e por que auie este rey ciro fechas muchas
batallas & muertas tantas yentes que eran sin cuenta & gozaba siempre de matar omnes (Estoria de España, c.
1270)
(2) los del consulto senatorio luego hizieron tres emperadores: pupieno albino y el tercero de quien hablamos.
empero los dos dichos arriba en el segundo año de su imperio fueron muertos por sus caualleros.(Tratado de
Roma, 1498)
(3) PASSIVES
El ladrón fue {muerto / % matado} por el policía (% = variable acceptability) The thief was dead / killed by the
policeman
(4) a. [TP T [vP SER [AspP Asp [VoiceP Voice[+AG, -ExtArg] [vP [DP] v [ √234 ]]]]
b. L ST or √234 mat, mor, muer, mur
c. [√234 [ v [ Voice[+AG, -ExtArg] [ sp ]]]] → Spell out this chunk as “muerto” ( rammar 1)
d. [√234 [ v [ Voice[+AG, -ExtArg] [ sp ]]]] → Spell out this chunk as “matado” ( rammar 2)
(5) PRONOMINAL USE OF ‘MORIR’
Juan se murió ?? en un accidente / *fusilado / de una enfermedad
Juan SE.3s died in an accident / by firing squad / from a sickness
(6) a. Transitive and intransitive are different lexical roots: matar-morir(se); parir-nacer(*se), acrecentar-
crecer(*se), quemar-arder(*se)
b. Transitive is unmarked and intransitive is marked with SE: secar-secarse, acostar-acostarse, fundir-
fundirse, romper-romperse
(7) a. [TP T [VoiceP Voice[-AG, -ExtArg] [vP [DP] v [ √234 ] ] ] ]
b. L ST or √234 mat, mor, muer, mur
c. [√234 [ v [ Voice[-AG, -ExtArg] [ T[pst] ]]]] → Spell out this chunk as “murió” + “SE”
References
Alexiadou, A. 2010. On the morpho-syntax of (anti-)causative verbs. In M. Rappaport Hovav, E.
Doron & I. Sichel (eds.) Syntax, Lexical Semantics and Event Structure, 177-203. Oxford, UK: Oxford
University Press; Chierchia, G. 2004. A Semantics for Unaccusatives and its Syntactic Consequences.
In A. Alexiadou, E. Anagnostopoulou & M. Everaert (eds.). 2004. The Unaccusativity Puzzle:
explorations of the syntax-lexicon interface, 22-69, Oxford, UK: OUP; Corominas, J. & Pascual, J.
1980-91. Diccionario crítico etimológico castellano e hispánico. Madrid: Gredos. Embick, D. 2004.
On the structure of resultative participles in English. LI 35: 355-392; Koontz- Garboden, A. 2009.
Anticausativization. NLLT 27: 77-138. Real Academia Española. 2009. Nueva gramática de la lengua
española. Madrid: Espasa-Calpe. Siddiqi, D. 2006. Minimize exponence: economy effects on a model
of the morphosyntactic component of the grammar. PhD Diss, University of Arizona; Siddiqi, D. 2009.
Syntax within the word: Economy, allomorphy and argument selection in Distributed Morphology.
Amsterdam: Benjamins.
10
The voice quality of Peninsular Spanish-speaking females – phonetic and socioindexical
observations
Meghan Armstrong
University of Massachusetts-Amherst
Nicholas Henriksen
University of Michigan
In this paper we use production data to investigate the acoustic correlates of voice quality differences in
young Peninsular Spanish- (PS) speaking women. We then examine how naïve listeners might use
voice quality differences as socioindexical markers. We designed a production experiment and a
perception experiment to determine the acoustic correlates of phonation type variation in PS, and the
perceived social categories that the phonation types might map onto. For the production experiment, six
females (aged 18-23) were recorded. Three speakers that produced what we perceived to be a “hoarse”
voice quality in speech, and three other speakers whose speech we did not perceive as “hoarse” were
recorded. All speakers per ormed an oral narration o the children’s picture book story Frog, Where
Are You? By Mercer Mayer (1969). The first 50 stressed /a/, /e/ and /o/ vowels with a minimum
duration of 50ms were extracted and submitted to acoustic analysis usin the “Spectral Tilt Script for
raat” (DiCanio, 2012), which outputs H1-H2 (spectral tilt), H1-A1 and harmonics to noise ratio
(HNR). For each measure, a mixed effects model was fit to the data in R with Vowel and Group (the
speakers perceived as hoarse vs. those that were not) as possible predictors, and Speaker as a random
actor. The actor “Group” was not chosen as a predictor of H1-A1 or HNR values, but it was
significant for H1-H2 values: the speakers chosen for exhibiting a hoarse quality produced significantly
lower H1-H2 values (p<0.01) than other “unmarked” roup.
In the perception experiment, 48 listeners from various parts of Spain listened to 5-second audio clips
from each speaker. Listeners were asked to provide adjectives to describe each speaker. Because we
had initially perceived hoarseness, listeners we also asked whether they thought the speaker smoked or
not. There were 8 adjectives that occurred most frequently. As shown in Figure 2, those with lower H1-
H2 values tended to be judged as pasota (apathetic), fuerte (strong) and chula (cool) more than the
group with higher values, and those with the higher H1-H2 values tended to be judged as joven
(young), inteligente (intelligent) and femenina (feminine) more often. The groups were judged as
interesante about the same. A t-test revealed that speakers from the group with significantly lower H1-
H2 values were rated as sounding like they smoked significantly more than speakers from the other
group (t=-13.17, df = 255.973, p<0.01). This was noteworthy since only one of the three speakers in
this group actually smoked.
In sum, this exploratory study shows new evidence for the type of social information indexed through
phonation type in Peninsular Spanish, since listeners showed clear tendencies when providing social
categories for speakers that differed in voice quality based on the H1-H2 measure. As Gordon &
Ladefoged (2001) point out, phonation types considered pathological and uncontrollable in one
language may have social or phonemic meaning in another. Our study is yet another example of this,
since even members of the speech community tended to associate speakers with low H1-H2 values as
participating in an activity that affects the respiratory system, smoking. The listeners also had clear
intuitions about describing different speakers based on glottal control. Finally, by including perception
data, the present results represent an initial metric for determining socially normative vs. individual
voice quality distinctions in PS. This is a methodological challenge in any study of voice quality
distinctions (Esling & Edmondson, 2011).
11
References
DICANIO, C.T. retrieved 10 December 2012. Spectral tilt script for Praat. Retrieved from
http://www.linguistics.berkeley.edu/~dicanio/Scripts.html.
ESLING, J. & J. EDMONDSON. 2011. Acoustical analysis of voice quality for sociophonetic
purposes. In M. Di Paolo & M. Yaeger-Dror (eds). Sociophonetics: A student's guide (pp. 131-148).
New York: Routledge.
GORDON, M & P. LADEFOGED. 2001. Phonation types: a cross-linguistic review. Journal of
Phonetics 29.383-406.
MENDOZA, E., N. VALENCIA, J. MUÑOZ & H. TRUJILLO. 1996. Differences in voice quality
between men and women: Use of the long-term average spectrum (LTAS). Journal of Voice 10.59-66.
PODESVA, R.J. 2007. Phonation type as a stylistic variable: The use of falsetto in constructing a
persona. Journal of Sociolinguistics 11.4: 478-504.
SZAKAY, A. 2012. Voice quality as a marker of ethnicity in New Zealand: from acoustics to
perception. Journal of Sociolinguistics 16.382-97.
12
Three types of result modification in Spanish
Grant Armstrong
University of Wisconsin-Madison
It is well known that Romance languages like Spanish do not permit the kind of secondary resultative
predicates of the type I hammered the metal flat that productively exist in Germanic languages and
there a number of works that have tackled this issue from different theoretical perspectives (Mateu
2000; Snyder 1995). That said, Spanish and other Romance languages have adjectival modifiers that
target the resultant state of the direct internal argument of the verb, which are commonly labeled
‘pseudo-resultatives’ (Bosque 1990; Demonte 1991; Demonte & Masullo 1999), shown in (1). In this
paper, two main points about Spanish pseudo-resultatives are made. First, a unifying characteristic of
pseudo-resultative adjectives is that they are always the focus of the sentence. We claim that pseudo-
resultatives have a focus feature in a functional projection, FP, that must be checked by intonation,
degree modifiers, diminutive morphology, or a combination of these strategies. This feature is meant to
explain the obligatory presence of a degree modifier or diminutive morphology (or both). Second,
based on recent advances in the typology of resultative constructions (Levinson 2010) it is argued that
pseudo-resultatives do not constitute a homogeneous syntactic or semantic class: (1a), (1b) and (1c)
have different structural representations and interpretations. Expressions like muy finito in (1a) modify
an implicit created noun (the result of chopping, a chunk), expressions like bien seco in (1b) represent
an overt pronunciation of the state that is already entailed by a change of state verb and expressions like
bien caminados in (1c) are aspectual projections (participles) that modify the VP, and may appear with
virtually any verb-object pair.
The first part of the proposal is that all pseudo-resultatives, whether they are DegP (adjectives) or AspP
(adjectival participles) contain an FP with a focus feature that must be checked by some element in the
c-command domain of F or by emphatic intonation. This accounts for why the adjectival modifiers of
results in (1) generally require a degree/manner modifier, diminutive morphology, emphatic intonation,
or a combination of these in order to be acceptable. The proposed internal structure of each
pseudoresultative phrases in (1) is shown in (2). The second part of the proposal concerns the
attachment site of FP. We claim that there are three possible attachment sites, which are shown in (3).
For verbs that have null objects of creation such as picar (= chop), cortar (= cut), anudar (= knot),
apilar (= pile), the FP modifies the null object, which we claim is the complement of a preposition of
terminal coincidence (Hale & Keyser 2002; Levinson 2010). Thus, the meaning of chop/cut/knot/pile
can be paraphrased rou hly as ‘to recon i ure x into a chunk/slice/knot/pile’, where the null and N
incorporate into V through head movement. This proposal captures provides a straightforward answer
to the question of what the adjectives in constructions like (1a) actually modify. Note that muy finito in
(1a) does not modify el tomate but rather the object that is created by chopping. For change of state
verbs that take an implicit adjectival complement that describes the result state of the object, it is
argued that FP is the complement of the verb. This proposal captures the fact that pseudo-resultatives
like (1b) are only possible for change of state verbs and the adjectives that may appear as their heads
are root identical to the verb, indicating that they must describe a result state that is entailed by the verb
itself (Armstrong 2012; Bosque 1990; Demonte 1991). Finally, FPs that contain participles like (1c)
modify the VP. These have no lexical semantic requirement other than that there must be direct object
in the VP. In such cases, bien (= well) is interpreted as a manner modifier of the verbal layer in the FP
rather than a degree modifier. This explains why these are possible with manner verbs like caminar,
which do not entail a result state.
Some implications of this proposal are outlined in (4), where the same verb can have multiple uses, all
of which can be illustrated by the different types of pseudo-resultatives discussed above. The verb
13
cortar may mean ‘make x into a cut/slice’ (implicit created object) when its object is a food item such
as carne or tomate but it may also mean to shorten when its object is pelo or pasto (a type of change of
state called a ‘de ree achievement’ in the literature). Both o these uses o the verb can be modi ied by
different types of pseudo-resultatives as shown in (4a) and (4b), but these are not mutually
interchangeable. Note that the lexical syntax of the structure that permits (4a) would not permit a
modifier such as bien corta. The same goes for (4b). When something is shortened, there is no implicit
created entity, so substituting bien corto for muy fino or muy grueso would not make sense. Finally, the
prediction based on the structures in (3) is that when a participle is used it should not matter what type
of VP is being modified so long as there is a direct object. This is precisely what we see in (4c) and
(4d).
(1) a. ??Pica el tomate fino / Pica el tomate muy finito (Demonte 1991) Chop the tomato fine / chop the tomato very fine.DIM
(= the created chunks are fine, not the tomato)
b. secarlo *(bien) seco / limpiarlo *(bien) limpio (Bosque 1990)
dry.it (really) dry / clean.it (really) clean
c. Caminó los tres kilómetros *(bien) caminados. (Demonte 1991)
She walked the three kms (well) walked
(2) a. Structure of (1a): [FP F[+FOC] [DegP [muy] Deg [AP finito]]]
b. Structure of (1b): [FP F[+FOC] [DegP [bien] Deg [AP seco]]]
c. Structure of (1c): [FP F[+FOC] [AspP Asp [vP v[-act] [VP bien caminados]]]]
(3) Attachment sites of FP
a. Juan cortó la carne muy gruesa (implicit created object, create a cut/slice)
Juan cut the meat very thick
b. Me cortaron el pelo bien corto (change of state verb, shorten)
DAT.1s they cut the hair very short
c. Juan cortó la carne bien cortada (implicit created object, create a cut/slice)
Juan cut the meat well cut
14
d. Me cortaron el pelo bien cortado (change of state verb, shorten)
DAT.1s they cut the hair well cut
References
Armstrong, G. 2012. On the adjectival component of change of state verbs in Spanish. The End of
Argument Structure?, 13-41. Bingley, UK: Emerald. Bosque, I. 1990. Sobre el aspecto en los adjetivos
y en los participios. Tiempo y aspecto en español, 177-214. Madrid: Cátedra; Demonte, V. 1991.
Temporal and Aspectual Constraints on Predicative APs. Current Studies in Spanish Linguistics, 165-
200. Washington, DC: Georgetown University Press. Demonte, V & Masullo, P.J. 1999. La
predicación: los complementos predicativos.) Gramática descriptiva de la lengua española, Volumen 2,
2461-2524. Madrid: Espasa; Hale, K. & Keyser, S. J. 2002. Prolegomenon to a Theory of Argument
Structure. Cambridge, MA: MIT Press; Levinson, L. 2010. Arguments for pseudo-resultative
predicates. NLLT 28: 135-182; Mateu, J. 2000. Why Can’t We Wipe the Slate Clean?: A Lexical-
Syntactic Approach to Resultative Constructions. Catalan Working Papers in Linguistics 8: 71-95
Snyder, W. 1995. A Neo-Davidsonian Approach to Resultatives, Particles and Datives. Proceedings of
NELS 25: 457-471.
15
Unexpected agreement patterns with parecer
Adolfo Ausín
Michigan State University
Marcela Depiante
University of Wisconsin-Eau Claire
The purpose of this paper is to establish the attestation of examples with unexpected agreement patterns
with parecer, such as (1), to determine their syntactic pattern, and to provide an account for them.
Assuming standard accounts of raising constructions in Spanish (Torrego 1996, among others), we
suggest that some speakers are taking advantage of the mixed A/A-bar nature of subject position in
Spanish (Villa-Garcia 2012, and references therein) and generating unexpected agreement patterns.
Sentences such as (1) are typically considered ungrammatical (see also Fernández‐ Salgueiro 2011),
and (2) are suggested as grammatical alternatives. The Nueva Gramática (RAE/AALE 2009:2834)
illustrates the standard attitude towards sentences like (1) when it states that in this type of construction
“the initial topic does not a ree with the verb Algunos novelistas jóvenes {parece ~ *parecen} que son
muy buenos escritores.” When examples like (1) are actually addressed, they are considered an error
(Diccionario Panhispánico de Dudas (RAE/AALE 2005)), a case of false agreement (Gómez Torrego
1998:42), or a case of backtracking (Davies 1997:51).
To try to determine the availability of examples like (1), we performed a series of Google searches on
the Internet. We searched or the exact sequences “parece que” and “parecen que” and obtained the
results in (3). These results confirm that the pattern in (2a) is preferred over the one in (1), but they also
show that examples like (1) are attested. Examples that contain the sequence “parecen que” appear in
(4).
We also used Goo le’s N ram viewer to search Goo le’s book database and to obtain the relative
requency o the “parece(n) que” sequences. The approximate requency values or the 1960-2000 time
frame appear in (5). Again there is a strong preference for the standard pattern (2a) over the
nonstandard pattern (1) (a 266 to 1 ratio), but still the non-standard option is attested. The examples in
(4) illustrate two of the most common types of the nonstandard examples. In these examples parecer
agrees with a plural topicalized subject as in (4a) or with a plural relativized subject as (4b). In order to
determine whether examples like these should be considered a true syntactic pattern that needs to be
accounted for or some type of performance error, we decided to focus on the pattern in (4b) and
searched for the sequences in (6). To keep the number of examples manageable, we decided to restrict
our searches to a one month period on a newspaper site: www.elpais.com. We found approximately 80
sequences o “que parece que”, out o which 7 had a relativized plural subject as in (7). For the same
period we ound 2 instances o “que parecen que” as in (8). That is, out o the 9 examples o the
sequence “ lural N + que parece(n) que”, two o them showed the nonstandard a reein parecen.
We take the observed relative frequency of the nonstandard parecen in the “ lural subject + que
parece(n) que” context (approximately 1 in 5) to be evidence o a syntactic pattern that deserves to be
accounted for. To do that we propose that examples like (8) follow the abstract pattern illustrated in (9),
where the null relative pronoun moves through the subject position of parecer where agreement is
established yielding the unexpected agreement pattern.
We argue that movement of the null relative pronoun through the subject position of parecen (an
instance of improper movement) is possible in Spanish due to the A-bar properties of Spec,IP in
Spanish.
16
Cross-linguistic evidence for this analysis comes from English. Kimball and Aissen (1971) and Kayne
(1989) discuss examples such as (10) with agreement between the verb think and the relativized
subject, which follows a pattern similar to (9). It could be possible that if the nonstandard agreement
pattern that we are discussing generalizes, Spanish could evolve into a situation similar to what we
found in Brazilian Portuguese examples like (12) from Martins and Nunes (2010).
To conclude, we show that an agreement pattern with parecer that is normally considered absent or a
mistake of some sorts is attested and has some regularity, and we provide an account for it that
crucially relies on the mixed properties of Spec,IP in Spanish.
Examples
(1) a. *Los niños parecen que tienen prisa. (Brucart 1987, 24)
b. Se encontrarán con un vídeo de esos que parecen que son un montaje.
(www.elpais.com)
(2) a. Parece que los niños tienen prisa.
b. Los niños parecen tener prisa. (Brucart 1987, 24)
(3) a. pproximate number o hits or “parecen que” 985,000
b. pproximate number o hits or “parece que” 145,000,000
(4) a. Algunos parecen que mienten en sus comentarios
b. Hay relaciones de amistad que parecen que sean de amor
(5) a. “parece que” 0.004%
b. “parecen que” 0.000015%
(6) Plural NP + que parece(n) que.
(7) Lo que sucede es que son informaciones sobre eventuales apuntes contables que parece
que no constan (en la contabilidad oficial). (www.elpais.com)
(8) se encontrarán con un vídeo de esos que parecen que son un montaje (www.elpais.com)
(9) NPPL [CP Opi que [IP ti parecen + [CP ti que [IP ti V…]]]]
(10) Mark knows the people who Clark think are in the garden.
(12) Os meninos parecem que viajaram ontem.
the boys seem-3PL that traveled-3PL yesterday
‘The boys seem to have traveled yesterday.’
References
Brucart, José M. 1987. La elisión sintáctica en español. Bellaterra: Universidad Autónoma de
Barcelona.
Davies, Mark. 1997. “ Corpus Based nalysis o Subject Raisin in Modern Spanish.” Hispanic
Linguistics 9: 33–63.
Fernández‐Salgueiro, Gerardo. 2011. “Romance Null Subjects at the Sensory‐Motor nter ace.” Syntax
14 (1): 1–28.
Gómez Torrego, Leonardo. 1998. La impersonalidad gramatical: Descripción y norma. Madrid:
Arco/Libros.
Kayne, Richard S. 1989. Notes on English agreement. Reprinted in Parameters and Universals, Oxford
University Press, 2000.
17
Kimball, John, and Judith issen. 1971. “ think, you think, he think.” Linguistic Inquiry 2 (2): 241–
246.
Martins, Ana Maria, and Jairo Nunes. 2010. “ pparent hyper-raising in Brazilian Portuguese:
reement with topics across a inite C .” n hoevos ana iotidis (edh.) The Complementiser Phase:
Subjects and operators. New York: Oxford University Press.
Real Academia Española, and Asociación de Academias de la Lengua Española. 2005. Diccionario
panhispánico de dudas. Santillana.
———. 2009. Nueva gramática de la lengua española. Madrid: Espasa Libros.
Torrego, Esther. 1996. “Experiencers and Raisin Verbs.” Current Issues in Comparative Grammar:
101–120.
Villa-García, Julio. 2012. “Spanish Subjects Can Be Subjects: Acquisitional and Empirical Evidence.”
Iberia 4 (2): 124–169.
18
Light verb constructions in stative versus eventive passives in contact Spanish
Osmer Balam
Ana de Prada Pérez
University of Florida
Light verb constructions (LVCs), as in (1), have been attested in such a wide variety of contact
situations that they have been proposed as a potential universal property of code-switching (Edwards
and Gardner-Chloros 2007; González-Vilbazo and López 2012). They consist of a fully inflected
‘do/make’ li ht verb rom the recipient lan ua e, which typically co-occurs with an infinitive lexical
verb from the source language (Edwards and Gardner-Chloros 2007; Muysken 2000). To date, a large
body of work has explored LVCs in Indic languages and/or Indic contact situations (Annamalai 1989;
Bhattacharja 2010; Butt 1993; Hook 1974; Muysken 2000; Ritchie and Bhatia 1996, Romaine 1986;
Sankoff, Poplack and Vanniarajan 1990, inter alia); however, less research has been carried out vis-à-
vis LVCs in Spanish contact situations.
In Spanish-English contexts, the LVC exhibits very limited geographical distribution, previously
documented only in Southwest U.S. (Jenkins 2003; Silva-Corvalán 1994; Pfaff 1979; Reyes 1982;
Wilson Vergara 2012) and Belize (Hagerty 1996; Fuller-Medina 2005). Both variationist and formalist
researchers have examined the LVC. The extant variationist work has examined variables/contexts that
restrict the use of the LVC, concluding that it is constrained by FREQUENCY (Reyes 1982) and/or
STATIVITY of the lexical verb (Fuller-Medina 2005; Jenkins 2003). Formalist approaches analyze hacer
‘do/make’ as a LV, thus, bein projected in little v. It has been contended that in monolingual and
bilingual speech, passives are incompatible with LVCs because V does not incorporate into v, and thus,
the internal argument cannot raise over PRO (Gonzalez-Vilbazo and López 2011).
This study expands on a previous project on Northern Belizean Spanish (NBS), where Authors
(2013) found canonical LVCs to be very productive in NBS. Also attested in the oral production data
were very infrequent yet innovative LVCs in control structures and stative passives, as in (2). The
current comparative study sou ht to investi ate NBS and U.S. Spanish speakers’ jud ments o LVCs in
stative (3) versus eventive passives (4), a verbal difference encoded in Spanish but not in English
(Bruhn de Garavito & Valenzuela 2008; Lujan 1981). lso o interest was examinin speakers’
maintenance of long distance gender agreement (LDGA) across two levels of distance.
46 NBS speakers from Orange Walk, Belize and 57 U.S. Spanish speakers (46 from Florida and 11
from New Mexico) completed a language background questionnaire and an acceptability judgment task
(AJT), whose test items were manipulated for passive structure (STATIVE vs. EVENTIVE), gender
marking in the LV (MASC -o vs. FEM -a) and LDGA across two levels of distance (DP + LVC vs. DP
+ relative clause + LVC). Statistical analysis of the AJT data revealed that the three groups
significantly differed from each other, F(2, 100) = 74.739, p = .000. In particular, N.B. bilinguals
accepted both stative and eventive passives across the two levels of distance, whereas FL bilinguals but
not NM bilinguals categorically rejected the LVCs. Overall, no distance effects were found in any of
the groups. Furthermore, NB bilinguals rated the masculine conditions higher than the feminine ones, a
pattern consistent with the previously collected NBS oral production data where the masculine default
was pervasively used.
Thus, contra González-Vilbazo and Lopez, the current study reports passives to be compatible with
LVCs (as did Annamalai 1989; Backus 1996; Chatterjee, in progress). The data support the notion that
speakers build syntactic congruence to facilitate code-mixing (Edwards and Chloros 2007; Sebba 1998,
2009). The innovative LV hecho ‘done’ in passives and the masculine default in NBS are analyzed as
strategies used to make both component languages more congruent in bilingual discourse. We further
contend that LVs can be used as canonical carriers of tense, aspect and mood features, and as markers
of grammatical gender in Spanish-English code-mixed verbal phrases. This project is innovative in
19
that it contributes new data that can inform both formalist and variationist approaches to contact
phenomena.
(1) Hicieron rent un golf cart
Do-3rd
pl rent.inf a golf cart
‘They rented a ol cart.’
(2) En Sixth Form, ‘tas hecho considered un adult
In Sixth Form, are.2nd
sing done considered.pastpart an adult
‘ n Sixth Form, you are considered an adult.’
(3) Patricia se molestó porque la cocina no estaba hecho cleaned bien.
‘ atricia ot mad because the kitchen wasn’t cleaned well.’
(4) Cristina se molestó porque la casa no fue hecho decorated bien.
‘Cristina ot mad because the house wasn’t decorated well.’
Selected References
Annamalai, E. (1989). The language factor in codemixing. Internationl Journal of the Sociology of
Language, 75, 47–54.
Bhattacharja, S. (2011). Benglish Verbs: a Case of Code-Mixing in Bengali. In Annual Review of South
Asian Languages and Linguistics, Rajendra Singh, Ghanshyam Sharma (Eds.), 17–33. Berlin: De
Gruyter.
Bruhn de Garavito, J. and Valenzuela, E. (2008). Eventive and stative passives in Spanish L2
acquisition: a matter of aspect. Bilingualism: Language and Cognition. Special Issue, Guijarro-
Fuentes, P. and Geeslin, K (Eds.), 11(3), 323–336.
Edwards, M., and Gardner-Chloros, P. (2007). Compound Verbs in Code-switching: Bilinguals Making
Do?
International Journal of Bilingualism, 11(1), 73–91.
González-Vilbazo, K. and López. L. (2011). Some properties of light verbs in code-switching, Lingua,
121 (5), 832–850.
----------. (2012). Little v and parametric variation. Natural Language and Linguistic Theory, 30(1), 33–
77.
Jenkins, D. L. (2003). Bilingual verb constructions in southwestern Spanish. Bilingual Review, 27(3).
Muysken, P. (2000). Bilingual speech: A Typology of Code-Switching. Cambridge: Cambridge
University Press.
Reyes, R. (1982). Language Mixing in Chicano Spanish. In Spanish in the United States:
Sociolinguistics Aspects, ed. Amstae and L. Elías-Olivares, 154-65. New York: Cambridge
Romaine, S. (1986). The syntax and semantics of the code-mixed compound verb in Panjabi/English
bilingual discourse. In Language and Linguistics. The interdependence of theory, data and
application. ed. Tannen, D. and J. E. Alatis, 35-50. Washington DC: Georgetown University Press.
Sebba, M. (1998). A congruence approach to the syntax of codeswitching. International Journal of
Bilingualism, 2(1), 1–19.
Silva-Corvalán, C. (1994). Language Contact and Change: Spanish in Los Angeles. Oxford:
Clarendon Press.
Wilson, Damian V. In Press. One construction, two source languages: hacer with an English infinitive
in bilingual discourse. Proceedings of the 6th Workshop on Spanish Sociolinguistics. Somerville,
MA: Cascadilla Proceedings Project.
20
Peak alignment in Veneto-Spanish bilinguals
Hillary Barnes
College of Charleston
Jim Michnowicz
North Carolina State University
Numerous studies have focused on various aspects of Spanish intonation in several dialects of the
language (Sosa 1999, Quilis 1993, Face 2002, Hualde 2002, Willis 2003, among many others).
Intonation patterns, such as peak alignment, have been shown to be susceptible to contact influence
(O’Rourke 2005, lvord 2006, Colantoni 2011, Colantoni & Gurlekian 2004, Elordieta 2003, amon
others). The present project examines intonation patterns in an understudied contact variety of Mexican
Spanish: Chipilo Spanish (CS), in contact with Veneto, a Northern Italian language. Chipilo is a small
Veneto-Spanish bilingual community near Puebla, Mexico, founded in 1882 by Veneto speaking
immigrants from Northern Italy. Today, Veneto is still spoken alongside Spanish and is the first
lan ua e o many bilin uals in the community. Chipileños are o ten noted to have an ‘accent’ that
differs from surrounding monolingual Spanish communities, and this difference may stem from
variation in intonation patterns (see figure 1). Studies have shown that pre-nuclear pitch accents in
Italian languages frequently occur in stressed syllables whereas in Spanish they generally occur in the
post-tonic syllable (see Colantoni & Gurlekian 2004). Colantoni and Gurlekian (2004) find early peak
alignments in Argentinian Spanish, which is not consistent with declaratives in other varieties of
Spanish, a feature which they attribute to Italian substrate influence. Therefore, in the case of Veneto-
Spanish bilinguals, it is possible that bilingual Chipilo Spanish will also demonstrate early peak
alignment in declarative utterances. Importantly, as Colantoni & Gurlekian (2004: 108) note, the
contact situation between an Italian variety and Argentine Spanish has disappeared, thus analysis of CS
can provide insight not only into questions of contact-induced intonation in Chipilo, but also into the
development of other contact varieties (such as Buenos Aires Spanish).
Previous research by the authors has found early peaks in semi-spontaneous, naturalistic bilingual
CS speech. The current study continues to examine CS by focusing on laboratory speech, addressing
the following questions: Does CS show intonation patterns that a) distinguish it from other varieties of
Spanish; and b) could be attributable to language contact (with Veneto)? In order to provide a tentative
answer to these questions, 30 bilingual Spanish-Veneto speakers from ages 18-76 were asked to read a
set of 24 sentences, including 12 broad focus declaratives and 12 distractor sentences, taken from
O’Rourke (2005). Declarative utterances were analyzed or overall F0 track and peak alignment, two
eatures that have been shown to di er based on lan ua e contact in other varieties (O’Rourke 2005,
Alvord 2006, Colantoni & Gurlekian 2004, among others). A monolingual group of Mexican Spanish
speakers from the neighboring towns of Cholula and Puebla served as a control group.
Initial analysis of CS shows that it displays a higher percentage of early peaks than many other
varieties of monolingual Spanish (25% for Castilian Spanish, see Face 2002: 123). This is consistent
with reported studies of peak alignment in Italian varieties (Ladd 1996, Grice 1995) and indicates that
this variation in peak alignment may be contact induced. Further results and conclusions will be
discussed.
21
Figure 1: Chipilo Spanish (bilingual Spanish-Veneto speaker)
Selected References
Alvord, S.M. (2006). Spanish intonation in contact: The case of Miami Spanish bilinguals. Diss.
University of Minnesota.
Colantoni, L. (2011). Broad-focus declarative sentences in Argentine Spanish contact and Non-contact
varieties. In Gabriel and Lleó (eds.), Intonational Phrasing in Romance and Germanic: Cross-Linguistic
and Bilingual Studies. Amsterdam: J. Benjamins. pp. 183-212.
Colantoni, L. & Gurlekian, J. (2004). “Conver ence and intonation Historical evidence from Buenos
ires Spanish”, Bilin ualism Lan ua e and Co nition 7(2), 107-119.
Elordieta, G. (2003). The Spanish intonation of speakers of a Basque pitch-accent dialect. Catalan
Journal of Linguistics 2.67-95.
Grice, M. (1995). The intonation of interrogation of Palermo Italian. Tubingen: Niermeyer
Face, T.L. (2002). Intonational marking of contrastive focus in Madrd Spanish. Munich: Lincom
Europa.
Face, T. L. (2003). Intonation in Spanish declaratives: differences between lab speech and spontaneous
speech. Catalan Journal of Linguistics 2.115-131.
Hualde, J. I. (2002). Intonation in Spanish and the other Ibero-Romance languages: overview and status
quaestionis. Romance Philology and Variation ed. by Caroline Wiltshire and Joaquim Camps, 101-115.
Philadelphia: John Benjamins.
Ladd, R. (1996). Intonational phonology. Cambridge: Cambridge University Press.
O’Rourke, E. (2005). ntonation and lan ua e contact case study o two varieties o eruvian
Spanish. Diss. University of Illinois at Urbana Champaign.
Quilis, A. (1993). Tratado de fonología y fonética españolas. Madrid: Gredos.
Sosa, J.M. (1999). La entonación del español: su estructura fónica, variabilidad y dialectología.
Madrid: Catedra.
Willis, E. W. (2003). The intonational system of Dominican Spanish: Findings and analysis. Ph.D.
diss., University of Illinois at Urbana-Champaign
22
Hispanic Heritage Speakers’ Interpretations: Quantified antecedents and the OPC Ewelina Barski
The College at Brockport, SUNY
Studies looking at the language of heritage speakers are interested in learning about the stability of
language before the critical period and how grammar develops under reduced input conditions
(Benmamoun et al. 2010). In this study the aim is to investigate the impact of reduced input on a
component within the Null-Subject Parameter - the Overt Pronoun Constraint (OPC) (Montalbetti
1984) - by focusing on the mental representation of the OPC in Hispanic heritage speakers (2nd
generation immigrants of Hispanic background) living in Canada. The goal in this study is to probe
into the interpretations that bilingual heritage speakers assign to overt pronouns in the subordinate
clause with quantified and wh-word antecedents.
The OPC restricts the possible antecedents that an overt pronoun can have. Specifically, it states the
restrictions on this pronoun when it has a quantified expression (someone, no one) or a wh-phrase (who,
which) as its antecedent. As a null subject language, Spanish allows the speaker to omit the subject of
the sentence. Examples (1a) and (1b) below show that unlike the null pronoun, the overt pronoun in
the subordinate clause can never bind with the quantified expression or wh-phrase: The overt pronoun
needs to refer to a third person within the discourse. On the other hand, if the pronoun is covert, the
sentence becomes ambiguous and allows for a less restrictive interpretation.
Following Montalbetti (1984), I assume that all quantifiers will be treated equally. Moreover, following
a generative framework, it is assumed that that the Null Subject Parameter is set early in the grammars
of these null-subject heritage languages (Chomsky, 1981; Jaeggli, 1982; Rizzi, 1982), and thus they
will demonstrate understanding of the interpretative restrictions found with subordinate overt pronouns
with quantified antecedents.
20 Hispanic heritage speakers participated in the experiment. Participants were asked to complete a
picture matching task, which looked at a forced interpretation of the OPC and a sentence selection task,
which allowed participants to provide their own interpretation of the sentence by choosing between two
pictures. Both tasks tested interpretation of the implicit knowledge of the OPC with quantified
antecedents.
Results for the Picture-Matching task show that advanced heritage speakers understand the
interpretative contrast present with overt and null pronouns within OPC contexts. However, heritage
speakers appear to have more difficulty in the Sentence-Selection task: They do not differentiate
between null and overt pronouns. Results suggest lower-proficiency participants have difficulty with
the reading/comprehension component of the task, but the OPC remains in their grammars.
Examples:
Overt (1a) Nadiei cree que élj/*i va a ganar
No one believes that he will win.
Null (1b) Nadiei cree que Øi/j va a ganar
No one believes that pro will win.
23
References
Beghelli, F. & T. Stowell. (1997). Distributivity and Negation: The Syntax of each and every. In Anna
Szabolcsi (ed.), Ways of Scope Taking (p. 71-108). Dordrecht: Kluwer Academic Publishers.
Benmamoun, E., S. Montrul & M. Polinsky. (2010). White Paper: Prolegomena to Heritage
Linguistics. Harvard University Press.
Montalbetti, M. (1984). After binding. PhD dissertation, MIT.
24
Morphosyntactic, Semantic and Working Memory Effects in L2 Anaphora Processing
Aurora Bel
Universitat Pompeu Fabra
Nuria Sagarra
Rutgers University
Juan Pablo Comínguez
Rutgers University
Estela García-Alcaraz
Universitat Pompeu Fabra
Late second language learners have difficulty acquiring discourse constraints on the distribution of
overt and null subject pronouns in pro-drop languages, (Tsimpli & Sorace, 2006; Sorace, 2011). In
addition to developing the representation of these syntax-discourse properties, they also have to acquire
an online parsing bias — Carminati’s (2002) osition o the ntecedent Strate y ( S). The PAS
constrains the resolution of intrasentential anaphoras in pro-drop languages by assigning matrix
subjects as the antecedents of embedded null subject pronouns as pro in (1a), and matrix objects as the
antecedents of their overt counterparts as lei in (1b) (Italian: Carminati; Catalan: Mayol, 2009).
Nevertheless, other pro-drop languages assign embedded overt subject pronouns to either the matrix
subject or object antecedent (Spanish: Filiaci, 2011; Moroccan Arabic: Bel & García-Alcaraz,
forthcoming). As per non-null subject languages, ambiguous embedded subject pronouns refer to
matrix subjects (English: Arnold et al., 2000; Gordon et al. 1999).
L2 studies on intrasentential anaphora resolution employ offline techniques, use a non-null-subject L1
(English), and are inconclusive. Some advance instabilities at the syntax-pragmatics interface
acquisition (i.e., non-native over-extension of overt subject pronouns to the matrix subject antecedent)
(English-Italian near-natives: Sorace & Filiaci, 2006; Belletti et al., 2007; English-Spanish intermediate
and advanced learners: Jegerski et al., 2011; Keating et al., 2011). In contrast, others claim native-like
performance (English-Spanish advanced learners: Valenzuela et al., 2011). In these studies, it is
difficult to determine what modulates these opposite outcomes. This is because they did not manipulate
the role of L2 proficiency, the L1-L2 pair, sensitivity to morphosyntactic information, and cognitive
limitations in the resolution of intrasentential anaphoras. The present study is the first to address the
online operativeness of the PAS by examining the role of L1 (Moroccan Arabic vs. English), L2
proficiency (low vs. high), morphosyntactic bias (null vs. overt subject pronouns), semantic bias
(subject vs. object), and cognitive demands (working memory) in the L2 online processing of Spanish
subject pronouns.
L1 English (48 low, 31 high proficiency) and L1 Moroccan Arabic (36 low, 25 high proficiency)
learners of L2 Spanish and 32 Spanish native speakers completed a language background
questionnaire, a proficiency test based on the DELE, a non-linguistic working memory test, and a non-
cumulative word-by-word selfpaced reading task with a yes-no question after each of the 111
sentences. The 24 experimental sentences had 4 conditions: null/overt pronoun and subject/object
semantic bias (El músico saluda al bombero mientras Ø /él lleva un casco/disco en la mochila ‘The
musician greets the fireman while Ø /he carries a helmet/disc in the backpack’). Experimental
sentences were controlled for length; matrix subjects/objects and embedded objects were controlled for
length, gender (half ms, half fs), and only appeared once in the entire experiment; and matrix verbs had
neutral implicit causality for interpreting the pronoun as co-referent with either the subject or the object
(bias range =59%-47%, based on Goikoetxea et al., 2008). Preliminary results reveal morphosyntactic
bias effects: faster RTs for overt than null pronouns in the embedded object and the PP in all
participants. Considering the fact that we controlled for implicit causality, we argue that the potential
25
disambiguating effect of the pronoun is attenuated. Furthermore, there was an interaction of L1 and
semantic bias: the Arabic learners were faster processing object than subject antecedents. They seem to
rely on the closest antecedent that is arguably less costly. Also, there were working memory effects for
all participants: faster RTs in the embedded verb and embedded object, the critical disambiguating
region that is more cognitively taxing, in higher than lower span subjects. Finally, there were no L2
proficiency effects, apart from the obvious finding that intermediate learners were slower
than advanced learners, and the latter slower than the natives. Considering that intrasentential anaphora
is acquired at the end of the L2 acquisition spectrum albeit it shows residual optionality even in near-
natives (Sorace 2011), probably our learners were not proficient enough to process the anaphora native-
like. To conclude, L2 processing of intrasentential anaphora is modulated by the pronoun
(morphosyntactic bias), the embedded object (semantic bias), the L1, and working memory.
(1)
a. La vecchiettai saluta la ragazzaj quando proi/?j attraversa la strada.
b. La vecchiettai saluta la ragazzaj quando leij attraversa la strada.
“The old woman reets the irl when ø/she crosses the road.”
(Sorace, 2011: 2)
Selected References
Arnold, J., Eisenband, J., Brown-Schmidt, S., & Trueswell, J. (2000). The rapid use of gender
information: Evidence of the time course of pronoun resolution from eyetracking. Cognition, 76(1),
B13-B26.
Belletti, A., Bennati, E., & Sorace, A. (2007). Theoretical and developmental issues in the syntax of
subjects: Evidence from near-native Italian. Natural Language and Linguistic Theory, 25, 657-689.
Carminati, M.N. (2002). The Processing of Italian Subject Pronouns. Unpublished doctoral
dissertation. University of Massachusetts at Amherst.
Filiaci, F. (2011). Anaphoric Preferences of Null and Overt Subjects in Italian and Spanish: a
Crosslinguistic Comparison. Unpublished doctoral dissertation. University of Edinburgh.
Gordon, P., Grosz, B., & Gilliom, L. (1993). Pronouns, Names, and the Centering of Attention in
Discourse. Cognitive Science, 17(3), 311-347.
Goikoetxea, E., Pascual, G. & Acha, J. (2008). Normative study of the implicit causality of 100
interpersonal verbs in Spanish. Behavior Research Methods, 40 (3), 760-772.
Jegerski, J., VanPatten, B. & Keating, G. (2011). Cross-linguistic variation and the acquisition of
pronominal reference in L2 Spanish. Second Language research, 27(4), 481-507.
Keating, G., VanPatten, B., & Jegerski, J. (2011). Who was Walking on the Beach? Anaphora
Resolution in Spanish Heritage Speakers and Adult Second Language Learners. Studies in Second
Language Acquisition, 33, 193-221.
Mayol, L. & Clark, R. (2010). Pronouns in Catalan: Games of Partial Information and the Use of
Linguistic Resources. In Journal of Pragmatics, 42, 781-799.
Sorace, A. (2011). Pinning down the concept o “inter ace” in bilin ualism. Linguistic Approaches to
Bilingualism, 1, 1-33.
Sorace, A. & Filiaci, F. (2006). Anaphora resolution in near-native speakers of Italian. Second
Language Research, 22, 339–368.
26
The effect of short-term study abroad on the acquisition of Spanish phonology
Silvina Bongiovanni
Avizia Long
Megan Solon
Indiana University
Despite growing interest in the role of context of learning on second language (L2) acquisition,
research on the impact of study abroad (SA) on phonetic or phonological development has lagged
behind that focusing on the effect of SA on other areas of language learning such as morphosyntactic
development, pragmatic competence, or L2 fluency (e.g., Collentine, 2004; Collentine & Freed, 2004;
DeKeyser, 1986; Freed, 1995; Lafford, 1995, 2004; Shively, 2008). Additionally, those studies that
have explored the impact of SA on the developing L2 sound system have found modest and
inconsistent, albeit generally positive, effects for time abroad (e.g., Díaz-Campos, 2004, 2006; Lord,
2006, 2010; Simōes, 1996). The present study aims to add to the literature on the role of context of
learning and, specifically, on its impact on L2 Spanish phonological development, by examining the
development of four sets of Spanish segments by two groups of learners: one participating in a short-
term SA program and one studying in the at-home (AH) context. It builds upon previous literature in its
expansion of the sounds under investigation and in its use of acoustic analyses and acoustic
characterization to account for L2 phonetic development abroad and at home.
The participants were 42 undergraduate English-speaking learners of Spanish from a large
Midwestern university. All participants were enrolled in an introductory Hispanic linguistics summer
course; 22 were completing the course during a 4-week SA program in the Dominican Republic,
whereas the other 20 were completing the course in the classroom at their home university. Participants
completed (a) a language background questionnaire, (b) a 25-item grammatical proficiency test, and (c)
an oral production task that involved a spontaneous narrative based on a picture book, a paragraph
reading activity, and a sentence reading activity. Participants completed the oral production task twice:
once prior to departing for the Dominican Republic (SA) or during the first week of their course (AH),
and again during the final week of the course (both groups).
Phonetic development was examined in the production of four groups of segments known to
present difficulties to English-speaking learners of Spanish: (a) word-initial voiceless stops /p t k/
(Díaz-Campos, 2004, 2006; Zampini & Green, 2001), (b) intervocalic voiced stops /b d g/ (Díaz-
Campos, 2004, 2006; Face & Menke, 2009), (c) intervocalic rhotics (Face, 2006), and (d) word-final
laterals (Díaz-Campos, 2004). Each production of a target token (N = 8,442) was extracted and
analyzed in Praat (Boersma & Weenik, 2013). Development was measured based on the segment in
question. The voice onset time of /p t k/ was measured in ms. Intervocalic /b d g/ were characterized
along a continuum from a nontargetlike occluded production to a targetlike spirantized production
(Face & Menke, 2009). Intervocalic rhotics were coded according to their duration, number of
occlusions, and presence of frication, and were further characterized using the labels of Face (2006)
and Rose (2010). Finally, the degree of velarization of word-final laterals was measured via the second
formant (e.g., Simonet, 2010). Recordings at Time 1 and Time 2 were compared for each segmental
phenomenon under investigation.
Preliminary results suggest that the SA program had a positive, albeit marginal, effect on phonetic
production gains as compared to the AH context: More individual learners in the SA group showed
positive gains on more of the sounds than learners in the AH group. These results appear to be
mediated both by the sound in question and by individual difference factors such as number of years
studying Spanish and attitudes toward pronunciation accuracy (as measured in the background
questionnaire). We discuss the implications of learner individual variation on SA outcomes as they
pertain to phonetic and phonological development.
27
Selected references
Collentine, J., & Freed, B. (2004). Learning context and its effects on second language acquisition:
Introduction. Studies in Second Language Acquisition, 26, 153-172.
Díaz-Campos, M. (2004). Context of learning in the acquisition of Spanish second language
phonology. Studies in Second Language Acquisition, 26, 249-273.
Díaz-Campos, M. (2006). The effect of style in second language phonology: An analysis of segmental
acquisition in study abroad and regular-classroom students. In C. A. Klee & T. Face (Eds.),
Selected Proceedings of the 7th Conference on the Acquisition of Spanish and Portuguese as first
and second languages (pp. 26-39). Somerville, MA: Cascadilla Proceedings Project.
Face, T. L. (2006). Intervocalic rhotic pronunciation by adult learners of Spanish as a second language.
In C. A. Klee & T. L. Face (Eds.), Selected proceedings of the 7th conference on the acquisition of
Spanish and Portuguese as first and second languages (pp. 47-58). Somerville, MA: Cascadilla
Proceedings Project.
Face, T. L. & Menke, M. R. (2009). Acquisition of the Spanish Voiced Spirants by Second Language
Learners. In J. Collentine et al. (Eds.), Selected Proceedings of the 11th Hispanic Linguistics
Symposium (pp. 39-52). Somerville, MA: Cascadilla Proceedings Project.
Freed, B. (1995). What makes us think that students who study abroad become fluent? In Barbara
Freed (Ed.) Second language acquisition in a study abroad context (pp.123-48) Philadelphia:
Benjamins.
Lafford, B. (1995). Getting into, through and out of a survival situation: a comparison of
communicative strategies used by students studying Spanish abroad and at home. Barbara Freed
(Ed.) Second language acquisition in a study abroad context (pp. 97-121) Philadelphia:
Benjamins.
Lafford, B. (2004). The Effect of the Context of Learning on the Use of Communication Strategies by
Learners of Spanish as a Second Language. Studies in Second Language Acquisition, 26, 201-225.
Lord, G. (2006). Defining the indefinable: Study abroad and phonological memory abilities. In C. A.
Klee & T. L. Face (Eds.), Selected Proceedings of the 7th conference on the acquisition of Spanish
and Portuguese as first and second languages (pp. 40-46). Somerville, MA: Cascadilla
Proceedings Project.
Lord, G.. (2010). The combined effects of immersion and instruction on second language
pronunciation. Foreign Language Annals, 43, 488-503.
Rose, M. (2010). Intervocalic tap and trill production in the acquisition of Spanish as a second
language. Studies in Hispanic and Lusophone Linguistics, 3, 379-419.
Shively, R. (2008). The cquisition o [β], [ð], and [ɣ] in Spanish: Impact of Experience, Linguistic
Environment, and Learner Variables. Southwest Journal of Linguistics, 27, 79-114.
Simōes, . R. M. (1996). honetics in second lan ua e acquisition n acoustic study o luency in
adult learners of Spanish. Hispania, 79, 87-95.
Simonet, M. (2010). Dark and clear laterals in Catalan and Spanish: Interaction of phonetic categories
in early bilinguals. Journal of Phonetics, 38, 664-679.
Zampini, Mary & Green, Kerry. (2001). The voicing contrast in English and Spanish: The relationship
between production and perception. In J. Nicol (Ed.), One mind, two languages: Bilingual
language processing (pp. 23-48). Cambridge, MA: Blackwell.
28
Learning word order in the Spanish classroom: lexico-semantics and pragmatics
Joyce Bruhn de Garavito
The University of Western Ontario
According to the Interface Hypothesis (IH, Sorace 2011) areas where linguistic modules interact with
each other are the locus of optionality in different types of acquisition, particularly when external
modules such as pragmatics are involved. It is argued that the syntax may be acquired from positive
evidence alone, and the problem arises from the interplay of syntax with pragmatics due to the higher
processing load of interfaces. Although it is not made explicit, it seems to be also assumed that the
pragmatics itself is unproblematic, probably because all languages structure information along the same
lines (new information, old, focus, etc.). However, besides the high demands on processing, it is also
possible that learning how a particular language makes use of the options provided by the
computational system to organize the information may not be easily accessible from the input.
Information may be structured in all languages, but how the language goes about doing this may vary.
For example, English and Spanish differ when it comes to signaling the focus of the sentence. English
relies almost exclusively on intonation, Spanish uses word order as well (Zubizarreta 1998).
Spanish allows both SV and VS word order. It has been shown the VS word order is contingent on
two factors, one dependent on the type of verb and the other pragmatic. VS order is canonical for
unacusative verbs such as llegar ‘arrive’ (Llegaron los estudiantes ‘arrived the students). With all types
of verbs VS is used to indicate focus (¿Quién gritó? Gritó un niño. ‘Who screamed? Screamed a child).
Studies such as Lozano (2006) have shown that, as predicted by the IH, the syntax/pragmatic interface
is the most difficult to acquire. Lozano found evidence that the contrast between the canonical VS
order with unaccusatives and SV with uner atives was part o the learners’ competence, the e ects o
focus for all verbs was not. The question is, can we help students recognize the role of pragmatics by
focusing their attention on the importance of word order?
This paper reports on an empirical study that examined the effects of teaching on the acquisition of
pragmatics and word order, properties of Spanish that are almost never taught. The methodology
consisted of the traditional pre-test, teaching intervention, post-test. The subjects were 35 students
enrolled in a third year Spanish course. The pre- and post-tests were identical to Lozano’s 2006 test, a
series of scenarios that established the focus of the sentence. These were followed by two sentences,
one with a SV order and one VS, each of which were judged by the learners on a scale. The teaching
intervention touched on how new and old information was organized, and examined also null subjects
and object pronouns. The different types of verbs in relation to word order was not touched on.
Results for the pre-test showed that learners rejected outright the order VS equally for both verb
types. After the intervention, the acceptance rate for VS was considerably higher on almost all
categories. Interestingly, for the post-test, an ANOVA repeated measures showed no main effect for
focus/non focus, but a main effect for verb type. In other words, knowledge of pragmatics, which was
taught, was not learned, but exposure to VS word orders seems to have triggered a recognition of the
role of verb types. What was not taught was nevertheless acquired.
References
Lozano, C. (2006). "Focus and split-intransitivity: the acquisition of word order alternations in non-
native Spanish." Second Language Research 22(2): 145-187.
Sorace, A. (2011). "Pinning down the concept of "interface" in bilingualism." Linguistic Approaches to
Bilingualism 1(1): 1-33.
Zubizarreta, M. L. (1998). Prosody, Focus and Word Order. Cambridge, Mass., MIT Press (Linguistic
Inquiry Monograph: 33).
29
The perception and reproduction of coda [s] in rural Dominican Spanish
Barbara E. Bullock
Almeida Jacqueline Toribio
University of Texas at Austin
A stigmatized trait of popular Dominican Spanish is the sporadic insertion of a preconsonantal or final
[s] into a lexical item in which it does not belong (e.g., Jacqueline > Jasqueline). The predominant
explanation among phonologists for this uniquely Dominican phenomenon is that the uneducated are
‘lost-s’ speakers who do not know which words contain lexical /s/ so that they randomly insert an [s]
into any syllable coda when attempting to speak in a higher register (Terrell 1986, Harris 2002, Bradley
2006). However, a quantitative sociolinguistic analysis of 37 rural Dominicans speakers conducted by
the present investigators revealed that the vast majority of the ‘[s]- ul’ tokens realized in spontaneous
speech were correct, rather than hypercorrect, productions. Both linguisticinternal and -external
(gender, literacy) factors were shown to bear significantly on correct [s] maintenance. The results of the
quantitative analysis effectively debunked the idea that Dominicans are ‘lost-s’ speakers whose
distribution of [s] is haphazard. The present analysis reports the results of a follow-up, controlled study
that was designed to assess whether speakers could perceive and correctly reproduce an [s] equally well
across different phonological contexts, or whether internal and external factors (gender, literacy, age)
affect how a speaker lexically codes and produces novel ‘s- ul’ tokens. For example, upon hearin a
token such as Metonís, what factors determine whether a speaker repeats a correct (Metonís), deleted
(Metoni), or hypercorrect (Mestonís, Mestoní, etc.) variant?
25 children (12 F; 13M ages 7-15) and 23 adults (11F; 12M ages 21-64), literate and nonliterate, from
the same community as the original study, participated in a “name ame”experiment where they
listened through headphones to a carrier phrase prompt, recorded by a Dominican speaker who is a
member of the community: Mi prima tiene un nombre muy especial: [name] ¿Cómo se llama mi
prima? Participants were instructed to respond to the prompt with the phrase Tu prima se llama
[name], repeating the name they heard, following the paradigm of Guion (2003), which minimizes
mimicry and requires speakers to attempt a lexical coding. The proper name stimuli were novel but
plausible Dominican female names, which are known to be fanciful. The stimuli (28 items) were all 3-
syllable words, with the position of lexical [s] balanced across words to appear in final position or
word-internally before a voiceless stop, voiced stop, fricative, nasal, or liquid in stressed and unstressed
position. There were also 28 distracter items containing no [s]. Auditory stimuli were presented in
random order, using SuperLab running on a laptop. The speakers wore a headmounted mic and their
productions were digitally recorded at 16-bit/48kHz in the field. We coded each token auditorily (with
spectrogram confirmation if required) for whether /s/ was correct, deleted, moved, duplicated, or
resyllabified, and for the phonological context of each realization or deletion.
The data were analyzed in R and fitted to logistic regression models with [s] realization, /s/ deletion, or
[s] insertion as the dependent variables and internal (context) and external (gender, literacy, age)
factors as independent variables. Speaker and token were treated as random variables. The results
reveal that external as well as internal factors appear to guide these speakers’ lexical codin o [s] as
reflected by production. Literate speakers were significantly more accurate than others and, for all
speakers, voiceless stops act as unequivocal attractors of [s] in correct and in hypercorrect productions.
Intriguingly, word final [s] appears to be either not detected in perception or is treated as optional in
production across all speakers. These results imply that the lexical coding of /s/ is affected by social
factors such as literacy, gender, and community norms.
30
L3 influence on L1 vs. L2 phonological systems: The case of vowel reduction
Jennifer Cabrelli Amaro
United States Air Force Academy
The present study tests how truly native-like adult-acquired phonological systems that are claimed to be
such are, using a unique methodology that examines the extent to which early- and late-acquired
systems are equally resilient to influence (regressive transfer) from an L3. We test strong and weak
versions of the Phonological Permeability Hypothesis (Cabrelli Amaro & Rothman, 2010), which states
that phonological perception and production systems acquired in adulthood, even when evidencing
nativelike target constraint rankings (assuming an Optimality Theoretic framework), are either
representationally different from systems acquired in childhood with regards to the stability of
constraint rankings (strong version) or more susceptible to processing-based influence (weak version).
We examined the L3 acquisition of Brazilian Portuguese (BP) and its effects on the Spanish
phonological systems of successive bilinguals (L1 English/L2 Spanish n=26; L1 Spanish/L2 English
n=11), and simultaneous English/Spanish bilinguals (n=16). A cross-sectional study testing Spanish
and BP perception/production by intermediate/advanced BP learners was implemented to observe BP >
Spanish influence throughout BP development. To test the hypothesis, we investigated potential effects
of BP vowel reduction on Spanish word-final vowels. BP unstressed syllables are less prominent than
stressed syllables, evidenced by higher, shorter unstressed vowels. However, in Spanish, syllables are
equally prominent regardless of stress and position. Crosswhite (2001) models vowel reduction via
interaction of prominence reduction constraints with feature faithfulness constraints, such that highly
ranked prominence reduction constraints yield reduced vowels. In Spanish, however, feature
faithfulness constraints outrank reduction constraints. Thus, as L3 learners converge on the BP ranking
via demotion of faithfulness constraints, the strong version of the PPH suggests that the BP system
could affect the Spanish ranking, yielding variable outputs in Spanish. Support for the weak version of
the PPH could take shape in the form of asymmetry in perception and production results; in the case of
data indicating stable perception across groups, we proposed that variation in production might be due
to issues of inhibitory control (see e.g., Green, 1998). Such variation could be explained via the
Constraint Fluctuation Hypothesis (CFH, Goodin-Mayeda, Rothman, & Renaud, 2011), which claims
that failed inhibition of a simultaneously activated system with different constraint rankings (i.e., BP)
can give rise to variability.
Spanish/BP data from a delayed repetition task were measured for F1, F2, F2-F1, duration, and
intensity. Perceptual accuracy and reaction time were measured via a forced-choice naturalness
preference task. In BP, all experimental groups performed similarly to the BP controls for perception
and production, indicating convergence on the L3 ranking. In Spanish,no differences were found
among groups for the perception task (Figures 1 and 2) or production task (Figure 3), suggesting
stability independent of age of acquisition of Spanish at the levels of mental representation, processing,
and articulation. However, while these data do not support the PPH, evidence from a longitudinal case
study of an L1 English/L2 Spanish/L3 BP speaker will be presented that reveal important differences at
the individual level which bring much to bear on generalizability of the findings.
31
References
Cabrelli Amaro, J. & Rothman, J. (2010). L3 acquisition and phonological permeability: A new test
case for debates on the mental representation of non-native phonological systems. International Review
of Applied Linguistics, IRAL, 48(2-3), 275-296.
Crosswhite, K. (2001). Vowel reduction in Optimality Theory. New York: Routledge.
Goodin-Mayeda, E., Rothman, J., Renaud, J. (2011). Optimality theoretic L2 reranking and the
Constraint Fluctuation Hypothesis: Coda nasals in the L2 English of L1 Spanish speakers. In M.
Pirvulescu et al. (Eds.), Selected Proceedings of GALANA 2010, pp. 66-77. Somerville, MA: Cascadilla
Proceedings Project.
Green, D. Mental control of the bilingual lexico-semantic system. Bilingualism:
Language and Cognition, 1, 67-81.
32
The role of gender agreement in the acquisition of DP structure in Spanish
José Camacho
Alena Kirova
Rutgers University
Studies generally agree that the acquisition of L2 gender agreement presents difficulties, particularly
for learners of an L2 with gender whose L1 is ungendered (cf. Bruhn de Garavito and White 2002;
Hawkins and Franceschina 2004; Grüter et al. 2012 a.o.). In this paper we address the issue from a
slightly different angle: what is the impact of gender when both the L1 and the L2 display it overtly? In
this paper we focus on the role of grammatical gender in the overall acquisition of nominal structures,
particularly from a determinerless L1 to an L2 with determiners and with bare NPs. If both languages
have gender agreement across the nominal phrase, can overt gender provide a positive clue to the
structure of DPs in the L2?
Russian and Spanish both have overt gender specification in animate nominals and agreement between
nominals and other elements in the nominal domain (demonstratives, possessives, adjectives, cf. (1a-
b)). In Spanish, animate Ns can have overtly marked gender (enfermero/enfermera ‘nurse-F/M’) or
unmarked gender, with gender appearing on the determiner (el/la cantante ‘the.M/F sin er’).
Additionally, Spanish nominals typically appear with an overt determiner and, when they are bare, they
are only postverbal, plural and existentially interpreted (cf. (2a)). Crucially, bare NPs in Spanish have
been analyzed as having a null determiner (cf. Contreras 1986; Benedicto 1998) that needs to be
licensed by a c-commanding verb, hence the word order restriction. In other words, Spanish always has
DPs, as in (3). Russian, on the other hand, generally lacks DPs (cf. (2b)), and bare NPs are not subject
to word order and/or interpretive constraints of the same type (cf. Kallestinova 2007). We assume that
Russian only projects to N (c . Boškovi´c 2008), as in (4) and that NPs may be interpreted as
arguments in determinerless languages (cf. Chierchia 1998).
Since both languages have overt gender agreement, will the availability of overt gender morphology
and generalized gender agreement (particularly with deteminers) in Spanish lead the L2 learner to
postulate that Spanish always has a DP, even when D is null? If so, will this cue improve the
acquisition of the word order and interpretation constraints on bare NPs? Our hypotheses are the
following.
Hypotheses
[H1]: Russian L1 speakers will acquire the overall structure of DP in Spanish.
[H2]: Russian L1 speakers will acquire the word order constraints in the use of bare NPs in Spanish.
[H3]: Overtly marked gender will help the acquisition of null determiners in Spanish.
The study. 38 native speakers of Russian participated in a time-limited acceptability task. they were
asked to judge the acceptability of a randomized set of sentences with animate DPs and bare NPs in
different word orders, on a scale of -2 to 2. The sentences were varied along three dimensions: 1) overt
morphological gender marking (-o/-a vs. -e Ns); 2) word order (SV vs. VS) and 3) DP vs. bare NPs.
Results. Ratings for DP subjects were slightly higher (M=.59, SE=.10) than ratings on NP subjects
(M=.56, SE=.09, t(35)=.40, p=.68), confirming H1 that L2 speakers have acquired the overall DP
structure of Spanish (i.e. they recognize overt Ds). Preverbal DP subjects were rated more acceptably
(M=.81, SE=.11) than NPs (M=.55, SE=.10, t(35)=2.451, p<.05), whereas postverbal NPs had higher
ratings (M=.58 SE=.11) than DPs (M=.38, SE=.10, t(35)=-2.3, p<.05), suggesting that overall word
33
order constraints on the use of bare NPs are acquired (H2), since there is a bias in favor of NPs in
postverbal position (as opposed to preverbal). Bare NPs were accepted more postverbally (M=.59,
SE=.11) than preverbally (M=.55, SE=.10, not significant t(35)=-.35, p=.72), also consistent with with
H2. In preverbal position, bare NPs with overt gender marking (-a/-o) were judged worse (M=.48,
SE=.10) than those with underdetermined marking (-e, M=.68, SE=.15), although not statistically
significantly (t(35)=-1.49, p=.14). This tendency is expected if hypothesis 3 is correct. In postverbal
position, -o/-a bare NPs were also judged worse than -e NPs (M=.49, SE=.12 and M=.7, SE=.13
respectively), but this time statistically significantly (t(35)=-2.08, p=.04).
Discussion. Our study suggests that Russian L1-Spanish L2 by speakers can acquire the overall DP
structure in Spanish, that bare NPs are analyzed and represented as DPs with a null D (otherwise we
would expect no differences between bare NPs and DPs with overt D), and that overt morphological
gender is plausibly relevant for the acquisition of that structure. As a consequence, we tentatively
propose a differential representation for gender in animate nouns (cf. (5), where those with overt
gender morphology have a lexically specified gender feature, whereas those with underdetermined
morphology have an unvalued gender feature.
In terms of the accessibility debate, we examine the three possible representation scenarios for nominal
structures in (6), and conclude that (6b) is the most compatible with the data. These results are
consistent with constrained access to UG hypotheses, although the results have little to say as to the
nature of those constraints.
34
References
Benedicto, E., 1998. Verb movement and its effects on determinerless plural subjects, in: Schwegler,
A., Tranel, B., Uribe-Etxebarria, M. (Eds.), Romance Linguistics. Theoretical Perspectives. John
Benjamins, Amsterdam, pp. 25–40.
Boškovi´c, Z., 2008. What will you have, D or N ?, in El ner, E., Walkow, M. (Eds.), NELS 37
Proceedings of the 37th
Annual Meeting of the North East Linguistic Society. GLSA, Amherst, MA,
pp. 101–114.
Chierchia, G., 1998. Reference to kinds across language. Natural Language Semantics 6, 339–405.
10.1023/A:1008324218506.
Contreras, H., 1986. Spanish bare nps and the ecp, in: Bordelois, I., Contreras, H., Zagona, K. (Eds.),
Generative Studies in Spanish Syntax. Foris. Studies in Generative Grammar: 27, pp. 25 – 49.
Bruhn de Garavito, J., White, L., 2002. L2 acquisition of spanish dps: the status of grammatical
features, in: Teresa, P.L.A.,
Liceras, J.M. (Eds.), The Acquisition of Spanish Morphosyntax: The L1/L2 Connection. Springer, pp.
153–178.
Grüter, T., Lew-Williams, C., Fernald, A., 2012. Grammatical gender in l2: A production or a real-time
processing problem? Second Language Research 28, 191 – 215.
Hawkins, R., Franceschina, F., 2004. Explaining the acquisition and non-acquisition of determiner-
noun gender concord in french and spanish, in: Prévost, P., Paradis, J. (Eds.), The acquisition of French
in different contexts. John Benjamins, Amsterdam, p. 175–206.
Kallestinova, E.D., 2007. Aspects of word order in Russian. Ph.D. thesis. University of Iowa.
35
Prosodic and segmental effects on Spanish sibilant voicing assimilation
Rebeka Campos-Astorkiza
The Ohio State University
This study experimentally analyzes sibilant voicing assimilation in Spanish, i.e., the assimilation in
voicing of /s/ before a voiced consonant ([rázgo] vs. [rásko]). We argue that it is only by exploring the
effects of both prosodic and segmental material, and their interplay, that the nature of this assimilation
can be understood. We show that the variable degree of sibilant voicing is dependent upon a series of
factors including the location of stress, the manner of articulation of the triggering consonant, and the
presence/absence of prosodic boundaries. Based on our results, this assimilatory process is modeled as
the result of blending of the glottal gestures for the adjacent voiceless and voiced sounds, and for this
reason, the degree of voicing is subject to influence by segmental and supra-segmental information.
Some earlier studies on the phenomenon show the variable and gradient character of this process
(Romero 1999, Schmidt & Willis 2010), although the precise conditionings have not been thoroughly
explored. Furthermore, previous theoretical accounts of the process, although allowing for variability,
do not discuss or incorporate any provisions to model this variability (Bradley 2005, Colina 2009).
Here, we adopt the theory of Articulatory Phonology (AP) to guide our methodology and formally
frame our results. Assimilation within AP derives from increased gestural overlap among adjacent
gestures due to changes in gestural magnitude and timing (Browman & Goldstein 1989). This
conceptualization makes certain predictions about what factors might be relevant in voicing
assimilation. We explore the role of stress and prosodic boundaries, which influence the duration and
magnitude of gestures, and the manner of articulation of the consonant initiating the assimilation,
which has been shown to influence the degree of voicing blending (Recasens & Mira 2012). Blending
refers to increased gestural overlap among gestures specified for the same articulator, in the case of
voicing, the glottis.
Two experiments were designed in order to obtain production data from 10 speakers of northern
Peninsular Spanish. The stimuli consisted of a series of sentences, which included the target /s/ and
triggering consonants in different environments: pre-tonic, tonic or post-tonic; within word, across
words or across an intonational phrase boundary (see some sample stimuli in [1]). In addition, the
triggering consonant could be an obstruent, a nasal or a lateral. Three acoustic cues to voicing were
measured: preceding vowel and fricative duration, and voicing during frication. The percentage of
voicing during frication was used to categorize each token as unvoiced, partially voiced or fully voiced.
The results show that the three factors under study, stress, manner of articulation and prosodic
boundaries, influence the degree of /s/ voicing. Due to space constraints, let us focus on the results for
stress, which make a significant contribution to our general understanding of the interaction between
stress and segmental information in Spanish.
Previous studies on the effect of stress on other (de)voicing phenomena in Spanish failed to get any
results using a binary stressed vs. unstressed distinction (Gonzalez 2002, Torreira & Ernestus 2011). In
the current study, different stress configurations are analyzed (see [1]), allowing us to distinguish
between different patterns and explore stress effects in more detail. Within words, where /s/ can be in
the tonic or pre-tonic syllable, there are no effects of this factor on the amount of voicing. However,
across words, where /s/ in post-tonic positions can also be analyzed, stress shows a significant effect on
the degree of voicing. /s/ in post-tonic positions, i.e., with a preceding stressed syllable and a following
unstressed syllable (e.g. tápas durítas), shows the highest amount of voicing, significantly more than all
other stress configurations. We argue that this is the case because post-tonic positions are the weakest
positions, with correspondingly shorter gestures, and are more prone to greater overlap from adjacent
36
gestures (cf. Cole et al. 1999). Our results are evidence that in Spanish, the relevant position to see any
effects of stress on segmental material is a truly post-tonic location, right after a stress and without a
following one - this is the articulatoryly weakest position.
References
Browman, C. P. & Goldstein, L. M. 1989. Articulatory gestures as phonological units. Phonology 6
201-252.
Cole, J., J.I. Hualde & K. Iskarous. 1999. Effects of prosodic and segmental context on /g/-lenition in
Spanish. Proceedings of the Fourth International Linguistics and Phonetics Conference, ed. By O.
Fujimura, B.D. Joseph & B. Palek, 575-589.
Colina, S. 2009. Sibilant voicing in Ecuadoran Spanish. Studies in Hispanic and Lusophone Linguistics
2(1): 3-29.
Bradley, T. G. 2005. Sibilant voicing in Highland Ecuadorian Spanish". Lingua(gem) 2.2:9-42.
González, C. 2002. Phonetic variation in voiced obstruents in North-Central Peninsular Spanish.
Journal of the International Phonetic Association 32. 17-31.
Recasens, D. & M.Mira. 2012. .Voicing assimilation in Catalan two-consonant clusters. Journal of
Phonetics 40. 639–654
Romero, J. 1999. The effect of voicing assimilation on gestural coordination. ICPhS99.
Schmidt, L. & E. Willis. 2010. Systematic investigation of voicing assimilation of Spanish /s/ in
Mexico City. Presented at Laboratory Approaches to Romance Phonology 2010.
Torreira, F. & M. Ernestus. 2011.Voicing of intervocalic /s/ in conversational Spanish. Presented at
Phonetics and Phonology in Iberia, Tarragona, Spain.
37
Morphological adjectival intensifier variation in Lima, Peru
Álvaro Cerron-Palomino
Arizona State University
This variationist study analyzes the alternation between the adjectival morphological intensifiers –ísimo
and–azo in the Spanish of Lima, Peru. On the one hand, –ísimo is the traditional Pan-Hispanic
adjectival intensifier; on the other, there is the suffix –azo, which has functioned traditionally as a noun
intensifier or augmentative in most Spanish varieties. This innovative use is shown in examples (1)-(2).
(1) Bueno, y la mayoría de veces he comprado reventa, menos cuando he querido estar aseguradazo,
porque normalmente las sacaban más baratas de reventa...hasta que me sucedió: fui a Kiss, confiadazo
de que iba a haber entrada en general y no había, y la reventa estaba al triple (qué rateros) y me quedé
afuera.
‘Well, most o the time, have bou ht re-sells [tickets], except for when I had really wanted to be
assured, because normally, they were cheaper when re-sold… until it happened to me: I went to [see]
Kiss, very confident that there were going to be general tickets, and there were none, and the re-sells
[prices] had tripled (what a bunch of rip-o s!) and had to stay outside’.
(2) Katia está bronceadaza, por eso es que se cortó el velo.
‘Katia is very tan, that’s why she cut her veil’.
It has been pointed out that the use of –azo as an adjectival intensifier is more common in Hispanic
America than in Spain (Falcinelli, 2007: 24). Interestingly, this use has become more extended in the
Spanish of Peru, concretely, in Lima, which is the paradigmatic variety for the whole country.
Unlike many other variationist studies, the data used for this one were online corpora (CREA1 and
CORDE2), online journals and magazines, and e-networks (Facebook, Twitter, YouTube, and
comments of the aforementioned online journals and magazines), all of them pertaining exclusively to
the variety spoken in Lima. The analysis of the data was partially based on Ito & Tagliamonte (2003),
and Tagliamonte (2008).
The statistical analysis with Goldvarb X shows that e-networks, participial adjectives, male speakers,
and predicative function are the factors favoring the occurrence of the non-standard variant. The factors
that were discarded in the statistical analysis were a positive/negative value of the adjective, the
emotional or lack of emotional character assigned to the noun modified by the adjective, the semantic
classification of the adjective, and the existence/absence of a feminine gender marker for the adjective.
The extremely scarce presence of -azo in both the CREA and CORDE corpora and its abundant
occurrence in the e-networks underscores the colloquial nature of the innovative form. By the same
token, the almost exclusive occurrence of -ísimo in the online corpora and its very small rate in the e-
networks suggests it is a rather formal morpheme. Another revealing finding related to its colloquial
nature is the fact that male speakers prefer the non-standard variant, which can be explained in terms of
covert prestige.
1 Corpus de Referencia del Español Actual, Real Academia Española.
2 Corpus Diacrónico del Español, Real Academia Española.
38
References
Falcinelli, Alessandro (2007). Valores y funciones del sufijo –azo en el español actual y sus
equivalencias en italiano. Linguae. Rivista di Lingue e culture moderne 1, 23-54.
Ito, Rika and Sali Tagliamonte (2003). Well weird, right dodgy, very strange, really cool: Layering and
recycling in English intensifiers. Language in Society 32, 257-279.
Tagliamonte, Sali (2008). So different and pretty cool! Recycling intensifiers in Toronto, Canada.
English Language and Linguistics 12, 361-394.
39
La pluralización de haber presentacional en las capitales antillanas. Un acercamiento desde la
gramática de construcciones y la sociolingüística comparada
Jeroen Claes
Universidad de Amberes (Bélgica)
En español estándar, el verbo haber presentacional se comporta como un impersonal: sólo se conjuga
en la tercera persona del singular y el único grupo nominal, e.g. mangos en (1), desempeña la función
de objeto directo. Sin embar o, parece estar expandiéndose el uso concordado (c . D’ quino-Ruiz,
2008), también conocido como 'la pluralización de haber' (véase (2)). En términos construccionistas
(cf. Goldberg 1995, 2006), podemos considerar que estamos ante una alternancia entre dos
construcciones: además del formato impersonal canónico (Locativo haberIMPERSONAL Objeto Directo)
también se da el formato personal (Locativo haberPERSONAL Sujeto). Planteamos la hipótesis de que, en
ciertas variedades, la construcción personal podría estar ganando terreno sobre la impersonal.
(1) Sí, había mangos (SJ14H22 /SJ1672).
(2) En mi época habían unos veinticinco, treinta alumnos por aula (LH01H22/LH17).
Para evaluar esta hipótesis, proponemos un análisis variacionista de la alternancia entre uso concordado
y no concordado. En tres muestras recientes del español de San Juan, Santo Domingo y La Habana
examinamos la distribución gramatical y discursiva de haber impersonal y personal así como su
distribución social para averiguar si y en qué medida la variación observada apunta hacia una
competencia entre dos variantes de la construcción presentacional con haber y un posible cambio
lingüístico en marcha ‘desde abajo’.
En términos generales, los resultados indican que el hablante caribeño pluraliza el verbo en
proporciones similares (San Juan: 41.4% N=682/1649; Santo Domingo: 46.7% N= 859/1841; La
Habana: 44.6% N=934/2093). El análisis VARBRUL de efectos mixtos (Johnson, 2009) muestra poca
variación en cuanto a los condicionantes gramaticales: en las tres capitales, impulsan la pluralización
los SSNN inherentemente agentivos y el uso de la variante personal en el discurso previo, así como
todos los tiempos y perífrasis verbales, menos el presente y el pretérito de indicativo. La única
diferencia estriba en que sólo en San Juan el uso concordado se ve favorecido por la ausencia de
negación.
En cambio, los significados indexicales de la alternancia (cf. Silverstein, 2003) varían más según la
ciudad estudiada. El análisis sociolingüístico comparado depara los resultados siguientes: en San Juan y
Santo Domingo la pluralización se relaciona con la clase media y el género femenino, lo cual sugiere
un cambio en pro reso ‘desde abajo’ (c . Labov, 2001 188, 292). En La Habana, por lo contrario, la
pluralización se vincula con la clase baja y los jóvenes de género masculino, lo cual indica que no
progresa como tendencia (cf. Labov, 2001: Cap. 3).
Estos resultados parecen favorables a la hipótesis de que las tres variedades se hallan en etapas
diferentes de la evolución hacia el uso concordado. Sin embargo, en La Habana, las connotaciones
sociales de clase social baja y género masculino parecen contrarrestar la propagación de la construcción
personal (cf. Labov, 1972: 178-179).
Palabras clave: Gramática de construcciones - sociolingüística comparada - español caribeño -
pluralización de haber presentacional.
40
Referencias bibliográficas
D'aquino-Ruiz, Giovana. (2008). El cambio linguistico de haber impersonal. Núcleo, XX (25), 103-124.
Goldberg, Adele. (2006). Constructions at work. The nature of generalization in language. Oxford:
Oxford University Press.
Goldberg, Adele. (1995). Constructions. A construction-grammar approach to argument-structure.
Chicago: Chicago University Press.
Johnson, Daniel Ezra. (2009). Getting off the GoldVarb standard: Introducing Rbrul for mixed-effects
variable rule analysis. Language and Linguistics Compass, III (1), 359-383.
Labov, William. (2001). Principles of linguistic change. Volume 2: social factors. Oxford: Blackwell.
Labov, William. (1972). Sociolinguistic patterns. Philadelphia: University of Pennsylvania Press.
Silverstein, Michael. (2003). Indexical order and the dialectics of sociolinguistic life. Language &
Communication, XXIII, 193–229.
41
Constraints on front mid-vowel gliding in Spanish
Laura Colantoni
University of Toronto
José Ignacio Hualde
University of Illinois at Urbana-Champaign
Variable syllabification of eV or iV sequences was already attested in Vulgar Latin (VINEA>VINIA;
LILIUM> LILEUM, Appendix Probi). Among Romance varieties, Spanish has clearly favored
diphthongization (Lloyd 1993), and this process is still active synchronically in many varieties, where
sequences of non-high vowels, which phonologically are syllabified in hiatus (e.g. te-a-tro ‘theater’;
po-e-ta ‘poet’), are realized as diphthongs in casual speech (e.g. teatro [tjátro], poeta [pwéta]) (e.g.
Garrido 2008, 2013, Hernández 2009). This phonological process gives rise to neutralization with
phonological diphthongs, and may result in hypercorrection in spelling, e.g. *copear for copiar ‘to
copy’. Whereas this tendency to reduce hiatus sequences to diphthongs has been widely reported,
curiously, the fact that there are clear differences among words with the same phonological sequence in
their tendency to diphthongize has, to our knowledge, been largely overlooked (but see Hernández
2009, Souza 2010). For instance, reduction to diphthong in peleemos [peljémos] ‘let’s i ht’ seems
much more likely than in leemos *[ljémos] ‘we read’. lthou h this casual speech phenomenon is
more or less stigmatized depending on the variety, speakers’ intuitions appear to be clear; and thus, we
seek to find here the main factors that condition gliding of mid-front vowels. On the basis of
observation and introspection, we propose three phonological hypotheses: (1) Minimality condition:
diphthongization is dispreferred when the result is a monosyllabic word (e.g. apear [apjár] ‘to step
down’ >> mear *[mjár] ‘to pee’, see Cabré & rieto 2004 or Catalan sequences with hi h vowels); (2)
Stress condition: the possibility of diphthongization increases with distance to stress (e.g. idealismo
[idjalísmo] >> ideal ?[idjál]); (3) Onset condition: diphthongization is dispreferred when the first vowel
in the sequence is preceded by a palatal consonant or a cluster (e.g. jalear [xaljár] >> centellear
*[senteʒjár], emplear ?[empljár]. In addition, there appear to be purely morphological conditions, (4)
diphthongization is dispreferred in second conjugation verbs, e.g. pelear [peljar] ‘to i ht’ >>
leer *[ljer] ‘to read’. These hypotheses are tested with a paper-and-pencil test administered to
participants from two different areas where the phenomenon has been attested, Argentina (N=48) and
Mexico (N=54). For each of the 41 target words, participants are asked to indicate whether a
pronunciation with /i/ instead of /e/ seems normal to them in casual speech. Results show that
hypotheses (1) and (4) are generally supported by the data, while (2) and (3) receive mixed support,
with higher rates of diphthongization reported in Mexican than in Argentine Spanish. Although part of
the variability obtained for (3) could be motivated by the different realization of the onset consonant in
both dialects, we propose that rather than a hard constraint against diphthongization in specific contexts
(i.e. no diphthongization in stressed syllables), which was already active in Vulgar Latin, there are
softer constraints that predict when diphthongization is more likely to happen (i.e. unstressed syllables
that are further away from the left-edge). Thus, diphthongization and the opposite phenomenon, /ia/ >
/ea/, may have the same explanation. Both tendencies reveal that there is indeed confusion regarding
the realization of the weakest vowel in the sequence, and this confusion is motivated by the acoustic
characteristics of those vowels. Vowels /i/ and /e/ in the sequence are shorter creating a quick transition
towards the next vocalic target (see Borzone de Manrique 1979; Hualde et al. 2008). The articulation of
these sequences of vocalic gestures crucially depends on the rate of speech and other phonological
factors (i.e. distance to stress). Since these factors vary across individuals in a speech community
differences in the intuitions about the realization of the sequences under study here are also expected.
42
References
Anonymous, Appendix Probi. In: Díaz y Díaz, M. (1962). Antología del latín vulgar. Madrid: Gredos.
Cabré, T. & Prieto, P. (2004): Prosodic and analogical effects in lexical glide formation in Catalan.
Probus 16.2: 113-150.
Borzone de Manrique, A. M. (1979). Acoustic analysis of Spanish diphthongs. Phonetica 36:194-206.
Garrido, M. (2008). Diphthongization of Non-high Vowel Sequences in Latin American Spanish.
Doctoral dissertation, University of Illinois at Urbana-Champaign
Garrido, M. (2013) “ Hiatus resolution in Spanish: Motivating forces, constraining factors, and research
methods”. Language and Linguistics Compass 7/6: 339–350.
Hernández, E. (2009). Resolución de hiatos en verbos -ear: un estudio sociofonético en una ciudad
Mexicana. PhD dissertation. Columbus: The Ohio State University.
Hualde, J.I., Simonet, M. & Torreira, F. (2008). Postlexical contraction of non-high vowels in Spanish.
Lingua 118: 1906-1925.
Lloyd, P. (1993). Del latín al español. Madrid: Gredos.
Souza, B. (2010). Hiatus resolution in Spanish: an experimental study. PhD dissertation. University
Park: The Pennsylvania State University.
43
On onset clusters in Spanish: voiced obstruent underspecification and /f/
Sonia Colina
University of Arizona
Onset clusters in Spanish consist of an obstruent or /f/ plus a liquid, e.g., blanco ‘white’, abrir ‘to
open’, aplastar ‘to latten’, apretar ‘to squeeze’, dragón ‘dra on’, egresar ‘to raduate’, sufrir ‘to
su er’, afligir ‘to a lict’, with the exception of /tl/ and /dl/ which are ill formed in some dialects or in
all. While this descriptive generalization is uncontroversial, the phonological account of onset clusters
has been a topic of debate amongst phonologists for decades (Harris 1983, 1989a, 1989b; Martínez-Gil
1997, 2001; Colina 2009). This paper provides additional evidence for the proposal in Martínez-Gil
(2001) regarding the sonority of /f/; it also introduces refinements of the onset cluster generalization
(Martínez-Gil 2001, Colina 2009), with regard to the point at which the generalization applies.
Most analyses agree that the main factor driving the well-formedness of the cluster is sonority: the first
member is drawn from of the group of the least sonorous consonants permissible in the onset (voiced
and voiceless stops) and the second member from the most sonorous ones (liquids). One crucial issue is
the behavior of /f/ and why this is the only fricative that is grouped with the stops as a permissible first
member of the cluster. Martínez-Gil (2001) convincingly shows that /f/ is grouped with the voiceless
stops and voiced stops because, like them, /f/ does not have [+continuant] in its representation:
voiceless stops are [-continuant], voiced stops are argued to be underspecified for continuancy (Lozano
1979, Mascaró 1984, Hualde 1989), and /f/ is not specified for continuancy because labiodentals do not
contrast in continuancy. Martínez-Gil argues that it is the presence of [+continuancy] in the underlying
representation that contributes to sonority. Accordingly, he proposes a division of the sonority scale in
which voiceless stops, voiced obstruents and /f/ belong to the least sonorous group; the resulting
eneralization is that “Spanish onsets consist of at most two consonants that differ maximally in
sonority rank (2001 219).” crucial question arises now with regard to the point at which this
condition is operative, especially since voiced obstruents are often surface approximants. In a
derivational framework, the condition is operative at the level of the underlying representation
(Martínez-Gil 2001). OT analyses (Martínez-Gil 1997, Colina 2009) have claimed that the condition
(i.e., constraint) is violated under domination by a more highly ranked constraint. This paper, however,
argues that the onset condition is a phonological condition that must be satisfied at the output of the
phonology. I present additional evidence for Martínez-Gil’s proposal that voiceless stops, voiced
obstruents and /f/ lack a [+continuant] feature; in particular, I propose that the underspecification of
voiced obstruents [B,D,G] persists through the output of the phonology (i.e., output underspecification,
Hale & Kissock 2007, from Keating 1988). As result, the onset condition is operational throughout the
phonology and no constraint violations are incurred for the dialects with voiced approximants in the
onset. The evidence in favor of output underspecification rests on phonetic variation. Phonetic studies
report a great degree of variation in aperture (Cole, et al. 1999; Ortega 2004), which can be easily
explained if voiced obstruents are [B, D, G]) at the output of the phonology, realized variably
according to the surrounding segments, i.e., in the phonetic component.
In addition to presenting additional evidence for the grouping of /f/ with the least sonorous consonants
on the sonority scale, this analysis offers an elegant and straightforward account of the Onset Condition
as it does not need to specify or restrict the point at which it applies (i.e., in the phonology; in OT, at
the output of the phonology where all restrictions apply). Although an OT analysis could solve the
problem through constraint violation, the alternative explanation provided here offers a simpler
phonological description of the relevant group of sounds. This is independently necessary to explain
why /f/ is grouped with the obstruents.
44
Finally, some dialects with voiced obstruent vocalization (e.g., some Chilean Spanish varieties) appear
to violate the constraint that requires that onsets differ maximally in terms of sonority, since only onset
voiced obstruents are vocalized (i.e., *po[B]re po[w]re ‘poor’; le[p]ra *le[w]ra ‘leprosy’) (Martínez-
Gil 1997). While violation of the Onset Condition through domination (of the need to spirantize) is
possible, I argue for an alternative in which the feature [+voice] separates voiced obstruents from /f/,
making [B, D, G] more sonorous than /f/. [+voice] is known to contribute to sonority in obstruents for
some languages.
References
Cole J., et al. 1999. Effects of prosodic and segmental context on /g/-lenition in Spanish. In: O.
Fujimura, O., Joseph, B.D., and Palek B. (Eds.), Proceedings of the Fourth International Linguistics
and Phonetics Conference. Charles University, Prague, The Karolinum Press, pp. 575-589.
Colina, S., 2009. Spanish Phonology. Washington, DC: Georgetown University Press.
Hale, M., Kissock, M., 2007. The phonetics-phonology interface and the acquisition of perseverant
underspecification. In: Ramchand, G., Reiss, Ch. (Eds.), The Oxford Handbook of Linguistic
Interfaces. Oxford, UK, Oxford UP, pp. 81-101.
Harris, J. W., 1969. Spanish Phonology. Cambridge, MA: MIT Press.
Harris, J. W. 1984. La espirantización en castellano y la representación fonológica autosegmental.
Estudis Gramaticals 1, 149-67.
Harris, James W. 1989a. Sonority and Syllabification in Spanish. In: Kirschner, C. and
DeCesaris,J.(Eds.), Studies in Romance Linguistics, ed. Amsterdam and Philadelphia: John Benjamins,
pp. 139-153.
Harris, James W. 1989b. Our resent Understandin o Spanish Syllable Structure.” In: Bjarkman, P.
and Hammond. R.M. (Eds.), American Spanish Pronunciation. Washington, DC: Georgetown
University Press, pp. 151-169.
Hualde, J.I. 1989. Procesos consonánticos y estructuras geométricas en español. Lingüística 1: 7-44.
Keating, P. A., 1988. Underspecification in phonetics. Phonology 5, 275-292.
Lozano, M. C. 1979. Stop and spirant alternations: fortition and spirantization processes in Spanish
phonology. Bloomington, IN: Indiana University Linguistics Club.
Martínez-Gil, Fernando. 1997. Obstruent vocalization in Chilean Spanish: A serial versus a constraint-
based approach. Probus 9: 165-200.
Martínez-Gil, Fernando. 2001. Sonority as a primitive phonological feature. In: Herschensohn, J.,
Mallén, E. & Karen Zagona, K. (Eds). Features and interfaces in Romance. Amsterdam and
Philadelphia: John Benjamins, pp. 203-222.
Mascaró, J. 1984. Continuant spreading in Basque, Catalan and Spanish language sound structure. In:
Aronoff M. & R. T. Oerhle, (Eds.), Language Sound Structure: Studies in phonology. Cambridge, MA:
MIT Press, pp. 287-98. Ortega-Llebaria, M. 2004. Interplay between phonetic and inventory constraints
in the degree of spirantization of voiced stops: Comparing intervocalic /b/ and intervocalic /g/ in
Spanish and English. In: Face, T. (Ed.), Laboratory Approaches to Spanish Phonology. Mouton de
Gruyter, pp. 237-255.
45
Puerto Rican Spanish as a partial pro-drop language: evidence from wh-questions
Juan Cominguez
Rutgers University
A well-known source of syntactic variation between Caribbean Spanish (CS) and non-Caribbean
Spanish (nCS) is the existence of pronominal preverbal subjects in CS wh-questions, but not in nCS, as
illustrated by the contrast between (1) and (2). However, the extent to which CS wh-questions exhibit
non-pronominal preverbal subjects is still a matter of controversy (Suñer 1994; Toribio 2000; Ordóñez
& Olarrea 2006; Gutiérrez-Bravo 2008; among others).
This paper reports experimentally obtained data on the existence of non-pronominal preverbal subjects
in Puerto Rican Spanish (PRS) wh-questions like in (3). These data were obtained by running an
acceptability judgment task on interrogative wh-movement and the position of non-pronominal subjects
(full DPs like María or la tormenta). Results from this experiment which included twenty-nine native
speakers of PRS living in the island (18 females; mean age = 26) show that argumental and non-
argumental wh-questions with non-pronominal preverbal subjects are highly accepted (78.78% and
82.6%, respectively). Argumental and non-argumental wh-questions with non-pronominal postverbal
subjects obtained slightly higher acceptability rates (93.93% and 90.13%, respectively). A repeated
measures two-way ANOVA performed on the results revealed a significant effect regarding subject
position (p < .001). No significant effects were found regarding the argumental status of the wh-
operator (p = .996), or the interaction between subject position and argumental status of the wh-
operator (p = .061). These results show that non-pronominal subjects can surface preverbally in PRS
regardless of the argumental status of the wh-operator. Additionally, these findings contrast with
Ordóñez and Olarrea’s (2006) study o Dominican Spanish, suggesting a source of microvariation in
the CS macrodialect. Their results found only pronominal subjects in preverbal position in
whquestions, which they accounted for by treating pronouns as weak pronouns that pied-pipe with the
IP when the latter undergoes remnant movement to the Comp Field. Because our results also found full
DPs preverbally, their analysis cannot be extended.
Alternatively, in this paper the experimentally obtained PRS data on wh-questions is accounted for in
terms of PRS becoming a partial pro-drop lan ua e. Followin Camacho’s (2012) approach to the
syntax of pro-drop languages, null subjects are the result of two features in Inflo, namely, a nominal
[D] feature which satisfies the EPP requirement (all clauses have to have a subject), and an [R] feature
which identifies the reference of Inflo. In accordance with Camacho’s (2012) analysis o partial pro-
drop languages, we will show that whereas PRS still keeps [D] in Inflo, as (4) illustrates, it is
undergoing a change according to which the [R] feature in Inflo is underspecified and is becoming [-
R]. With a [-R] feature, Inflo needs to obtain its reference from a DP with which it enters an Agree
relationship in [Spec, INFLP]. This leads to a significantly higher use of overt subject pronouns as
shown in (5), a well-attested difference between PRS and other Spanish varieties (Morales 1999). The
underspecification of [R] in PRS makes possible the occurrence of preverbal subjects in interrogative
wh-movement via the projection of [Spec, INFLP] in order to value n lo’s [-R] in a Spec-Head
con i uration. n lo’s [-R] is not valued by a DP in [Spec, INFLP], the derivation crashes. By
contrast, the [Spec, INFLP] position is not projected in General Spanish (GS) wh-questions because
Inflo’s [R] does not need to be valued. In GS, the only cases in which preverbal subjects occur in
[Spec, INFLP] are at the beginning of discourse with all-focus sentences (López 2009: 133), or with
desiderative/exhortative sentences (Villa-García 2012). According to these researchers, non-dislocated
preverbal subjects in GS undergo movement to [Spec, INFLP] in order to value features related to the
syntax-discourse interface in non-interrogative sentences.
46
The slight asymmetry in the acceptability of preverbal and postverbal non-pronominal subjects found
in our experiment is advanced to be the result of an epiphenomenon related to the discursive
consequences of having to value Inflo’s [-R] in a Spec-Head configuration. Whereas both pronominal
and non-pronominal DPs can do the valuing job, the former are used to express topic continuity as in
(5), as well as topic shift like in (6). By contrast, full DPs only express topic shift like in (7), in which a
full DP is used to reintroduce a previously mentioned discourse entity, their not being able to express
topic continuity as is well attested in non-pro-drop languages (Prince 1992). This results in a more
frequent use of overt subject pronouns than full DPs, which is reflected in corpora as the ones analyzed
by Suñer and Lizardi (1992), Lizardi (1993), and Gutiérrez-Bravo (2008). The underspecification of
[R] in PRS is further supported by data from non-interrogative wh-movement as in (8) showing the
existence of preverbal subjects, as well as the use of overt subject pronouns to refer to inanimate
entities like in (9). Finally, preliminary findings from fieldwork with native informants reveal that only
subject DPs (pronominal and non-pronominal) can intervene between the wh-operator and the verb as
shown in (10). This evidence further confirms that [Spec, INFLP] is projected in PRS interrogative
whmovement, casting doubts on the approach proposed by Gutiérrez-Bravo (2008). According to his
analysis, preverbal subjects in PRS interrogative wh-movement are the result of a topicalization
operation triggering movement to a position that can also be occupied by other constituents like
CLLDs. These preliminary data seem not to support that approach, strengthening the hypothesis that
PRS is becoming a partial pro-drop language.
47
Selected References
Camacho, J. (2012). “The Null Subject arameter revisited. The evolution from null subject Spanish
and ortu uese to Dominican Spanish and Brazilian ortu uese.” Ms. Rut ers University. Gutiérrez
Bravo, R. (2008). “Topicalization and reverbal Subjects in Spanish wh-interro atives.” n Selected
Proceedings of the 10th Hispanic Linguistics Symposium, ed. Joyce Bruhn de Garavito and Elena
Valenzuela, 225-236. Somerville, MA: Cascadilla Proceedings Project. www.lingref.com, document
#1799. Lizardi, C. (1993). Subject Position in Puerto Rican wh-questions: Syntactic, Sociolinguistic
and Discourse Factors. Ph.D. dissertation, Cornell University. López, L. (2009). A derivational syntax
for information structure. Oxford : Oxford University Press. Morales, A. (1999). “ nteposición de
sujeto en el español del Caribe.” n El Caribe Hispánico: Perspectivas Lingüísticas Actuales, Ortiz
López, L. (Ed.), 77-98. Vervuert Verlag, Frankfurt. Morales, A. and Vaquero, M. (1990) El Habla
Culta de San Juan. Materiales para su estudio. Río Piedras, Editorial de la Universidad de Puerto Rico.
Ordóñez, F. and Olarrea, A. (2006). “Microvariation in Caribbean/ non Caribbean Spanish
nterro atives.” Probus: An International Journal of Latin and Romance Linguistics 18: 59–96, 2006.
Suñer, M. and Lizardi, C. (1992). “Dialectal Variation in an Argumental/Non-Argumental=
symmetry in Spanish.” n mastae, Jon, Grant Goodall, Mario Montalbetti and M. Phinney.
Contemporary Research in Romance Languages, El Paso/Cd Juárez 1995, Amsterdam: John
Benjamins. Villa-García, J. (2012). “Spanish subjects can be subjects cquisitional and empirical
evidence.” Iberia: An International Journal of Theoretical Linguistics 4(2): 124-169, 2012.
48
/p/ e /t/ a /k/ a: Adult novice learners’ production and perception of L2 Spanish voiceless
stops
David Counselman
Ohio Wesleyan University
Verónica González López
Arizona State University
This study offers evidence supporting the SLM (Flege 1995) postulate regarding perceptual distance
among the L1 and L2 phones and extends its predictions to the late acquisition of an L2. Many studies
focus on the end state of L2 pronunciation in early and late bilinguals in a variety of language pairings
(Baker and Trofimovich 2005; Flege 1995; Flege and Eefting 1988), but more research on the
emergence of L2 phonetic categories is needed (but see González López 2012; Hurtado and Estrada
2010; Zampini 1998; Zampini and Green 2001). This experiment examines the /ptk/ production and
perception of two groups of L1 English first-semester Spanish students, and tests the following
hypotheses: H1) instruction on articulatory phonetics at beginning levels has a positive effect on L2
phone production, and H2) students will exhibit early signs of category formation only for L2 phones
they perceive as different from their L1 counterparts. The target group (N=18) received explicit
training in articulatory phonetics while the control group (N=16) did not. Recordings of /ptk/
productions and a perception task were conducted at the beginning and end of a 14-week semester.
Each participant produced 90 counterbalanced /ptk/ tokens in word-initial position in stressed syllables
in both English and Spanish, totaling 180 tokens. In each language, 30 tokens were located in single
words., and the other 60 were embedded in complete sentences. Half of these 60 were in sentence-
initial position and the other half were in sentence-medial position to control for phrasing and speech
rate effects (Allen et al. 2003; Cho and McQueen 2005; Fougeron 1999; Kessinger and Blumstein
1998; Miller and Baer 1983; Miller et al. 1986; Pind 1995; Theodore et al. 2009). The perception task
was conducted on a separate day in which participants listened to 60 instances of /ptk/ (4 voice onset
time (VOT) measurements x 3 places of articulation x 5 repetitions each) followed by a neutral vowel
in which the aspiration had been adjusted following results from previous studies that reported a
gradual increase of VOT value with each different place of articulation (Cho and Ladefoged 1999;
Docherty 1992; Klatt 1975; Lisker and Abramson 1964, 1967; Volaitis and Miller 1992; Zampini et al.
1999): /p/= 7 ms., 30 ms., 50 ms., and 70 ms; /t/= 10 ms., 35 ms., 55 ms., and 75 ms.; and /k/= 15 ms.,
40 ms., 60 ms., and 80 ms.
Findings show significantly shorter VOTs for L2 Spanish voiceless stops than L1 English. Both groups
significantly improved their production of /k/ on the posttest. Contrary to H1, there were no significant
differences between the two groups. However, linear regressions revealed a significant correlation
between the perception and production of Spanish /k/, confirming H2 and offering evidence supporting
the SLM. Overall, these findings suggest that category formation may begin with initial exposure as
long as L2 sounds are perceived as different from the L1 phones, but the effects of instruction are not
noticeable in the first semester.
References
Allen, J. S., J. L. Miller and D. DeSteno. (2003). Individual talker differences in voice-onset time.
Journal of the Acoustical Society of America 113(1), 544-552.
Baker, W. and P. Trofimovich. (2005). The effect of perceived phonetic similarity on non-native sound
learning by children and adults. In S. Fisch, B. Skarabela and A. H.-J. Do (Eds.), Proceedings of the
26th Boston University Conference on Language Development, pp. 36-47. Somerville, MA: Cascadilla
Press.
49
Cho, T. and P. Ladefoged. (1999). Variation and universals in VOT: Evidence from 18 languages.
Journal of Phonetics 27, 207-229.
Cho, T. and J. M. McQueen. (2005). Prosodic influences on consonant production in Dutch: Effects of
prosodic boundaries, phrasal accent and lexical stress. Journal of Phonetics 33, 121-157.
Docherty, G. J. (1992). The Timing of Voicing in English Obstruents. Berlin: Foris.
Flege, J. E. (1995). Second-language speech learning: Theory, findings, and problems. In W. Strange
(Ed.), Speech Perception and Linguistic Experience: Issues in Cross-Language Research, pp. 229-273.
Timonium: York Press.
Flege, J. E. and W. Eefting. (1988). Imitation of a VOT continuum by native speakers of English and
Spanish: Evidence for phonetic category formation. Journal of the Acoustical Society of America 83(2),
729-740.
Fougeron, C. (1999). Prosodically conditioned articulatory variations: A review. UCLA Working
Papers in Phonetics 97, 1-74.
González López, V. (2012). Spanish and English word-initial voiceless stop production in
codeswitched vs. monolingual structures. Second Language Research 28(2), 243-263.
Hurtado, L. M. and C. Estrada. (2010). Factors influencing the second language acquisition of Spanish
vibrants. The Modern Language Journal 94, 74-86.
Kessinger, R. H. and S. E. Blumstein. (1998). Effects of speaking rate on voice-onset time and vowel
production: Some implications for perception studies. Journal of Phonetics 26, 117-128.
Klatt, D. H. (1975). Voice onset time, frication, and aspiration in word-initial consonant clusters.
Journal of Speech and Hearing Research 18, 686-706.
Lisker, L. and A. Abramson. (1964). A cross-language study of voicing in initial stops: Acoustical
measurements. Word 20, 384-422.
Lisker, L. and A. Abramson. (1967). Some effects of context on voice onset time in English stops.
Language and Speech 10, 1-28.
Miller, J. L. and T. Baer. (1983). Some effects of speaking rate on the production of /b/ and /w/.
Journal of the Acoustical Society of America 73(5), 1751-1755.
Miller, J. L., K. P. Green and A. Reeves. (1986). Speaking rate and segments: A look at the relation
between speech production and speech perception for the voicing contrast. Phonetica 43, 106-115.
Pind, J. (1995). Speaking rate, voice-onset time, and quantity: The search for higher-order invariants
for two Icelandic speech cues. Perception and Psychophysics 57(3), 291-304.
Theodore, R. M., J.L. Miller and D. DeSteno. (2009). Individual talker differences in voice-onset time:
Contextual influences. Journal of the Acoustical Society of America 125(6), 3974-3982.
Volaitis, L. E. and J. L. Miller. (1992). Phonetic prototypes: Influence of place of articulation and
speaking rate on the internal structure of voicing categories. Journal of the Acoustical Society of
America 92, 723-735.
Zampini, M. (1998). The relationship between the production and perception of L2 Spanish stops.
Texas Papers in Foreign Language Education 3, 85-100.
Zampini, M. and K. P. Green. (2001). The voicing contrast in English and Spanish: The relationship
between perception and production. In J. Nicol (Ed.), One Mind, Two Languages: Bilingual Language
Processing, pp. 23-48. Malden, MA: Blackwell.
Zampini, M., C. M. Clarke and K. P. Green. (1999). Language experience and the perception of stop
consonant voicing in Spanish: The case of late English-Spanish bilinguals. In R. P.
Leow and C. Sanz (Eds.), Spanish Applied Linguistics at the Turn of the Millenium, pp. 194-209.
Somerville, MA: Cascadilla Press.
50
Variation in Tense and Mood in Nominal Clauses in Peruvian Spanish
Claudia Crespo de Rio
University of Illinois at Urbana-Champaign
Spanish nominal clauses require a verb in subjunctive with specific verb types in the main clause (e.g.
querer "to want") and that both verbs have the same tense. In many Latin American dialects, contrary
to this standard requirement, present tense in the nominal clause is common with main verbs in past
tense (Sessarego 2010, Suñer and Padilla-Rivera 1987), but it is even more frequent in Peruvian
Spanish (Sessarego 2008), as in Ex. (1). In this dialect, this variation can be found with many verb
types: with verbs flexible about this requirement (factive-emotive, such as alegrarse "to be happy" and
lamentar"to be sorry") and with the verb type that is one of the strictest (desire, such as querer "to
want" and desear "to wish").
Within Peruvian Spanish varieties, it has been found that both Andean and non-Andean dialects share
many grammatical features (Escobar 2000, 2007, 2010; Caravedo and Klee 2005; O'Rourke 2005;
Muntendam 2009), due to the intense dialectal contact in the country. This contact is reinforced by the
constant migrations from the Andes to the coast, happening since the 1950's. Thus, in this study, three
groups are considered: two of Peruvian Spanish, where one refers to L2 Spanish spoken by Quechua
speakers in the Andes (the main amerindian language in the country), and the other one refers to non-
Andean Spanish spoken in Lima, the capital. The third group, included as a different Latin American
variety, are speakers of Mexican Spanish living in Mexico City. Therefore, our research questions are
(i) whether tense variation is higher in Peruvian dialects than in the Mexican one and (i)i whether it is
found in both Peruvian varieties in a similar way.
School students (N=66) completed a production task with sentences using factive-emotive and desire
verbs in the main clause. Results in Table 1 show that although past subjunctive answers were
preferred in all three groups, there were still significant differences between them, p<.05. Present
subjunctive answers were also used in all groups, but in a higher percentage in the two Peruvian
groups. On the other hand, the bilingual group from the Andes showed another significant difference
when compared to the other groups (p<.05): not only they alternated between past and present in the
subjunctive form, but also there were many answers in indicative mood (see Table 1), as the one in Ex
(2). Based on the Spanish proficiency test administered to this group, the subgroup of intermediate
speakers were the ones who produced more indicative answers, whereas the advanced group performed
native-like (similar to the non-Andean group). Two main conclusions emerge: (i) both varieties of
Peruvian Spanish have a similar frequency of tense variation (due to intense dialectal contact), which is
higher than tense variation in Mexican Spanish (ii) the Andean group also produced mood variation,
which is related to the Spanish proficiency of bilinguals.
51
References
Arellano, R. and Burgos, D. 2010. Ciudad de los Reyes, de los Chávez, de los Quispe... Lima: Planeta.
Caravedo, R. and Klee, C. 2005. "Contact-Induced Language Change in Lima, Peru: The Case of Clitic
Pronouns".Selected Proceedings of the 7th Hispanic Linguistics Symposium. Ed. by David Eddington.
Somerville, MA: Cascadilla Proceedings Project. 12-21.
Escobar, A. M. 2010. Selection and diffusion in the spatiotemporal continuum: the case of
Quechua/Spanish contact variants. 46th meeting of the Chicago Linguistics Society. Forthcoming.
Escobar, A. M. 2007. "Migración, contacto de lenguas encubierto y difusión de variantes lingüísticas".
Revista
Internacional de Lingüística Iberoamericana. 17, 93-107.
Escobar, A. M. 2000. Contacto social y lingüístico: el español en contacto con el quechua en el Perú.
Lima: PUCP.
Montrul, S. 2004. The Acquisition of Spanish: Morphosyntactic Development in Monolingual and
Bilingual L1
Acquisition and in Adult L2 Acquisition. Amsterdam: John Benjamins.
Muntendam, A. 2009. Linguistic Transfer in Andean Spanish: Syntax or pragmatics?.Doctoral
dissertation, University of Illinois at Urbana-Champaign.
O’Rourke, E. 2005. Intonation and Language Contact: A Case Study of Two Varieties of Peruvian
Spanish. Doctoral dissertation, University of Illinois at Urbana-Champaign.
Sessarego, S. 2010. Temporal concord and Latin American Spanish dialects: a genetic blueprint.
Revista Iberoamericana de Lingüística. 5. 137-169.
Sessarego, S. 2008. Spanish Concordantia Temporum: an old issue, new solutions. Selected
proceedings of the 4th
workshop on Spanish sociolinguistics, ed. by Maurice Westmoreland and Juan
Antonio Thomas, 91-99. Somerville, MA: Cascadilla Proceedings Project.
Suñer, M. and Padilla-Rivera, J. 1987. Sequence of tenses and the subjunctive. Hispania 70, 3. 634-
642.
52
The sum is more than its parts: Acquisition of clitic clusters in Spanish
Maria Cristina Cuervo
Ana Teresa Pérez Leroux
University of Toronto
Spanish has three types of object clitics: dative, accusative, reflexives (a broader category
encompassing argumental, lexical, aspectual, impersonal, etc.). Clitics encode case, person, gender and
number, but full contrasts are expressed in 3rd
person only. These object markers can co-occur in
clusters subject to a series of complex morphosyntactic and surface constraints.
How do children master the ability to produce clitic clusters? The general consensus about the
acquisition of Romance is that clusters in children are produced early, possibly constrained (low
productivity), but seemingly free of errors in the sequence (Monachesi, 1995 for Italian; Connors &
Nuckle 1986 for French).
In Spanish, clitic sequences also emerge early; errors might appear in gender features ‘Sacamelo la
toalla’ (María 2;02, Lopez Ornat 1990), but otherwise these sequences appear to have the ri ht orderin
(Gonzalez 1970, Ezeizabarrena 1997). Errors in young children are cited anecdotally (me se cayó) but
are rare under experimental conditions (2 out of 561 cases in Eisenchlas 2002 elicited imitation study).
This latter work confirms a systematic pattern of accusative omission: while omissions represent 32 of
all repetition errors, two thirds of omission errors involve dropping the accusative but preserving the
dative. No study has yet directly examined what constraints the initial clusters in acquisition: What is
the structure of early sequences, and, are these initial clusters restricted to specific constructions or
featural combinations?
To answer these questions, we conducted a corpus analysis of longitudinal spontaneous data from
María (1;9-2;11, López-Ornat 1994) and Emilio (1;10-3;0, Vila 1990) in CHILDES.
(MacWhinney 2000). We analyzed the properties of the clusters actually produced, as well as those in
which one or two clitics had been omitted. A search for all verbs contexts in the timeframe under study
isolated 691 potential cluster contexts for Emilio, and 352 for Maria. These were classified by
construction into ditransitive (1), aspectuals/accusative (2), and inchoative/dative (3) contexts. Most of
the utterances involved other grammatical options other than a cluster (e.g., DPs). Overall, about a third
of these contexts yielded production of clusters, which were unevenly distributed across context types
(Table 1). There were very few errors in linear order, no illicit featural combinations, and even the
opaque spurious se was produced appropriately by both children. A small proportion of the data
consisted of omissions (1% for Emilio, 6% for Maria). Omissions occurred uniquely with ditransitives
(only in 1 case was a clitic omitted in aspectual/accusative) and, contrary to previous results, most
omissions were of the dative clitic (all utterances with omission involved the dative for Emilio; and
19/22 for María), as illustrated in (4).
n terms o case, Maria’s patterns mirror omission patterns in sin le clitic contexts Maria omits both
accusatives and datives, with the highest omission rates for referential (as opposed to reflexive) datives
in ditransitives, and in inchoative/dative contexts. For Emilio accusatives in single contexts are omitted
at 20% vs. 8% dative omission.
Given that omissions in clusters involve datives in the first position of ditransitive sentences but not
datives in the second position of inchoative/dative constructions, results clearly suggest that clitic
production is sensitive to syntactic context beyond the featural properties of individual clitics. Omitted
clitics are not limited to one person either, omissions involving all first, second and third person target
forms. Thus, patterns of omissions in clusters are not accounted for in terms of the dichotomy between
53
first/second person and third person clitics. Overall, results suggest that patters of omission in clusters
may arise from syntactic properties of the constructions involved and cannot be accounted for solely in
terms of the same phenomena underlying omissions in single clitic structures.
(1) Ahora te lo doy. Ditransitive
Now Cl.2s.Dat Cl.3sm give.1s
‘ ’ll ive it to you in a second’ (Emilio 2;7, e24)
(2) Me lo como yo. Aspectual/ accusative
Cl.1s.Ref Cl.3sm eat.1s I
‘ will eat it mysel ’. (Emilio 2;5, e21)
(3) Se me va a enfriar lo(s) pie(s) Inchoative/dative
Inch Cl.1s.Dat go to get.cold the feet
‘My eet are oin to et cold’ (María 2;2, section 408)
(4) Ø lo ha quitado. Dative omission
Cl.1s.Dat Cl.3sm has taken.away I
‘He has taken it away rom me’. (Emilio 2;0, e16)
Table 1. Proportion of success at producing clusters in each context type for María and Emilio
Maria Emilio
Ditransitives 0.59 0.41
Aspectual 0.26 0.04
Inchoativedatives 0.15 0.01
Overuse 0.00 0.01
Selected references
Connors, K. & L. Nuckle (1996) Morphosyntax of French personal pronouns and the
learning/acquisition dichotomy. In Studies in Romance linguistics, edited by O.A. Jaeggli, and C.
Silva Corvalán, 225-282, Dordrecht, Foris.
Eisenchlas, S. 2002. What do r entinian children know about clitics that lin uists don’t? Doctoral
Dissertation, U. of Queensland.
Ezeizabarrena, M.J. (1997). Morfemas de concordancia con el sujeto y el objeto en el lenguaje infantil.
In Pérez-Leroux, A. T., and W. R. Glass, (eds). Contemporary perspectives on the acquisition of
Spanish, volume 1: developing grammars. Somerville: Cascadilla Press
Gonzalez, G. (1971). The Acquisition of Spanish Grammar by Native Spanish Speakers. Doctoral
Dissertation, U. of Texas at Austin.
Monachesi, P. 1995. A grammar of Italian clitics. Doctoral Dissertation, Tillburg University.
54
Developmental asymmetries in early child bilinguals: Evidence from differential object marking
and aspectual distinctions in Spanish
Alejandro Cuza
Lauren Miller
Mariluz Ortiz
Purdue University
Differential object marking (DOM) and preterite/imperfect distinctions in Spanish are stable by age 4;0
in normally developing monolingual children (Pérez-Pereira, 1989; RodríguezMondoñedo, 2008).
However, these two areas of the grammar have been found to be challenging for L2 learners and
Spanish heritage speakers (Montrul, 2002; Montrul & Bowles, 2009). Some researchers suggest
incomplete development during childhood; others claim acquisition of a contact variety (Rothman,
2007). However, this previous research has not examined data from heritage bilingual children and
their parents to confirm one position or the other. We address this gap by reporting data on elicited
production. Can input factors, bilingual dominance and cross-linguistic influence account for the
deficits shown?
14 Spanish-English bilingual children born in the US (6;2 to 10;8) and 5 parents participated in the
study. The children were divided into two groups: younger children (n=8, 6;2-7;7, mean, 7;03, SD,
0.60), and older children (n=6, 8;5-10;8, mean, 9;5, SD, 0.79). Parents were of Mexican background
and completed reports on patterns of language use and child linguistic dominance. Older children were
reported to be English-dominant or balanced bilinguals; younger children were split across English-
dominant, balanced or Spanish-dominant. Aspectual distinctions were tested in eight different
combinations of situation and predicate types (1-2), and DOM was tested in animate and inanimate
contexts within matrix (3) and CLLD (4) structures. Testing occurred at the school setting in two
sittings.
Results show low production of required DOM in animate contexts among the younger children in
matrix sentences (35%), and null use in CLLD structures (0%). The older children, however, showed a
higher level of DOM production with matrix sentences (80%) but not with CLLD structures (0.07%).
With inanimate contexts, children behaved target-like. Regarding aspectual use, the younger children
showed low performance with the use of the imperfect across lexical classes (range, 8%-50%), with a
heavy reliance on the present tense, but better performance with the preterite. This pattern, however,
does not hold for the older children who did much better with the imperfect (range, 33% to 80%),
despite difficulties in some contexts. Parents behaved target-like in all conditions. An individual
analysis showed a positive correlation between target performance and bilingual ability. The balanced
bilinguals showed an advantage over the English dominant and the Spanish dominant (Figures 1 and 2).
We found difficulties in DOM use and preterite/imperfect distinctions in early bilingual development.
Deficits, however, are substantially overcome with age, although some lingering bilingualism effects
remain in contexts where the input may not be relevant enough to activate the necessary L1 options.
These results suggest a protracted morphosyntactic development stemming from inconsistent L1 input
and output as well as reliance on default/less marked representations (Meisel, 2007). Developmental
delays do not seem to be attributed to transfer from English, as has been found for other
morphosyntactic structures (Pirvulescu et al., 2012). English dominant children often outperformed
Spanish dominant children. Incomplete acquisition or acquisition of a contact variety in early childhood
is not supported by the data. Results are discussed in relation to input factors, patterns of language use
and the role of language dominance in early child bilingual development (de Houwer, 2007; Simard et
al., 2013).
55
[1] Ayer, Dora tocó la guitarra. “Yesterday, Dora played the uitar”
[2] De pequeña, Dora visitaba a su abuela. “ s a child, Dora visited her randmother”
[3] Dora estaba visitando a su padre. “Dora was visitin her ather”
[4] Felipe nunca juega con chicas de su escuela, pero a Marta siempre la invita a jugar
“Felipe never plays with school irls but he always invites Marta to play.”
References
De Houwer, A., (2007). Parental Language Input Patterns and Children's Bilingual Use. Applied
Psycholinguistics 28, 411424.
Meisel, J., (2007). The Weaker Language in Early Child Bilingualism: Acquiring a First
Language as a Second Language. Applied Psycholinguistics 28, 495-514.
56
Montrul, S. (2002). Incomplete acquisition and attrition of Spanish tense/aspect distinctions in
adult bilinguals. Bilingualism: Language and Cognition, 5 (1), 39-68.
Montrul, S., & Bowles, M. (2009). Back to basics: Incomplete knowledge of Differential Object
Marking in Spanish heritage speakers. Bilingualism: Language and Cognition, 12(3), 363-
383.
Pérez-Pereira, M. (1989). The acquisition of Morphemes: some evidence from Spanish. Journal of
Psycholinguistic Research, 18, 289-312.
Pirvulescu, M., A. Pérez-Leroux, and Y. Roberge. (2012). A bidirectional study of object
omissions in French-English bilinguals. Multilingual Individuals and Multilingual
Societies, ed. by Kurt Braunmüller, Christoph Gabriel and
Barbara Hänel‐Faulhaber. Amsterdam, John Benjamins. 171-188.
Rodriguez-Mondonedo, M. (2008). The acquisition of differential object marking in Spanish.
Probus, 20(1), 111-145.
Rothman, J. (2007). Heritage speaker competence differences, language change and input type:
inflected infinitives in heritage Brazilian Portuguese. International Journal of Bilingualism
11(4), 359–389.
Simard, D., Veronique, F., Foucamert, D. (2013). Measuring metasyntactic ability among heritage
language children. Bilingualism: Language and Cognition, 16(1), 19-31.
57
The effect of stigmatization on the imitation of phonological dialectal variants in Spanish
Mandy Daniluk
Bethany MacLeod
Carleton University
Trudgill (1986) predicts that phonological dialectal variants that are stigmatized will be more salient
than those that are not (Trudgill 1986: 11). He also predicts that speakers will be more likely to
accommodate towards more salient variables. The purpose of this study is twofold: 1. To determine
how stigmatized four phonological variables of Caribbean Colombian Spanish are, an 2. To determine
whether the degree of stigmatization of these variable affects the extent to which speakers of the more
conservative dialect spoken in the Andean region of Colombia will imitate the variables in a
conscious imitation task.
The Andean dialect is considered the standard, prestige, conservative variety (Lipski 1994, Hualde
2005, Garrido 2007a). In contrast, the Caribbean dialect is more innovative and is often considered
“careless” by Andean speakers (Garrido 2007a: 31, Garrido 2007b). A conscious imitation of a dialect
is created out o a speaker’s perception o a particular variety of a language, which largely stems
from the shared societal impression of that variety (Brunner 2010). Kerswill (1994) found that the
degree of stigmatization of individual linguistic features can differ. Our hypothesis is that the greater
the degree of stigmatization of a particular dialectal variant, the greater the extent that speakers will
imitate the variable.
The 4 phonological differences between the two dialects investigated here reflect differences in the
application of phonological processes. For example, we included the treatment of intervocalic /d/ in the
past participle marker –ado, which is frequently weakened to the point of deletion in Caribbean
Spanish, but not in the Andean variety (Lipski 1994, Hualde 2005: 23). Three other similar types of
differences were also included, all of which were predicted to be at least somewhat stigmatized (Hualde
2005).
To determine the degree of stigmatization and imitation of these variables, we conducted an experiment
in which 10 speakers of Andean Spanish participated. The imitation component was performed first.
During this component, the participants were asked to read aloud a short story, which included words
exemplifying the four dialectal differences under investigation, in their native dialect. Next,
participants watched a video of a Caribbean speaker reading the same story twice. Finally, the
participants were asked to read the story again, this time attempting to do so while imitating the
Caribbean dialect. To quantify the degree of stigmatization of the four dialectal differences, the
participants listened to pairs of recordings of Spanish words produced once by a Caribbean speaker and
once by an ndean speaker and were asked to decide which o the two recordin s was more “proper”
in their opinion. Each of the words contained only one dialectal difference. The proportion of Andean
tokens containing each dialectal di erence chose as more “proper” was taken to re lect the de ree o
stigmatization of the Caribbean variant.
Acoustic analyses of the native dialect reading of the story and the imitated version were performed
using Praat (Boersma & Weenink 2013). The particular acoustic measures taken depended on
which dialectal difference was present. For example, for the weakening of /d/ in –ado, the duration and
intensity of the following vowel were measured to quantify the degree of weakening
(Colantoni & Marinescu 2010, MacLeod 2012). Statistical analyses using mixed effects
models determined whether there was a significant effect of the stigmatization of the four variables on
the extent to which the participants imitated as compared to their native dialect reading of the story.
58
The preliminary results suggest that although speakers differed in their imitation performances, there
was a significant positive relationship between degree of stigmatization and the extent to which the
variables were imitated, providin support or our hypothesis and or Trud ill (1986)’s assertion that
more stigmatized variables will be more salient and therefore more subject to accommodation.
This study provides a means of quantifying the stigmatization of variables at the individual speaker
level and shows that more stigmatized variables are more likely to be produced in conscious imitations.
While conscious imitation has not received much attention in the literature as compared to more
“natural” speech styles, exploring how conscious imitation as a type of performance speech is used in
particular sociocultural contexts may help us improve our understanding of cross-dialectal
communication and sound change as well as in situations of language endangerment (e.g. Schilling-
Estes 1995).
References
Boersma, P. & Weenink, D. (2013). Praat: doing phonetics by computer [Computer program]. Versi
on 5.3.51, retrieved 30 May 2012 from http://www.praat.org/
Brunner, E.G. (2010). Imitation, awareness, and folk linguistic artifacts (Doctoral thesis). Rice Univ
ersity, Houston, TX.
Colantoni, L & Marinescu, I. (2010). The scope of stop weakening in Argentine Spanish. In Marta
Ortega-‐ Llebaria (ed.), Selected proceedings of the 4th Conference on Laboratory Approaches
to Spanish Phonology, 100-‐114. Somerville, MA: Cascadilla Proceedings Project.
Garrido, M. (2007a). Diphthongization of mid/low vowel sequences in Colombian Spanish. In Proce
edings of Third Workshop on Spanish Sociolinguistics. Temple University: Cascadilla Proceedings
Project. 30-‐37
Garrido, M. (2007b). Language attitude in Colombian Spanish: Cachacos vs. Costeños.
LL Journal, 2(2). Hualde, J. I. (2005). The Sounds of Spanish. New York, NY: Cambridge Univers
ity Press.
Kerswill, P. (1994) Dialects converging: Rural speech in Urban Norway. Oxford, NY: Oxford Univ
ersity Press. Lipski, J. (1994). Latin American Spanish, London, New York: Longman
MacLeod, B. (2012). Investigating L2 acquisition of Spanish vocalic sequences. Studies in Hi
spanic and Lusophone Linguistics 5(1): 103-‐148
Schilling-‐Estes, N. (1995). Production, perception and patterning: “ erformance” speech in an endangered dial
ect variety. Penn Working Papers in Linguistics 2(2): 117-‐131.
Torstensson, N., Eriksson, E. & Sullivan, K. (2004). Mimicked accents –
Do speakers have similar cognitive prototypes? Proceedings of The 10th Australian International Co
nference on Speech Science & Technology (pp.271-‐276). Sydney, Australia: Australian Speech Science & Technology Association Inc.
Trudgill, P. (1986). Dialects in contact, Oxford: Basil Blackwell .
59
Microvariación dialectal: el orden de palabras en el Caribe y en México
Ashlee Dauphinais Civitello
Luis Ortiz López
Universidad de Puerto Rico- Río Piedras
El español permite cierta flexibilidad en cuanto al ordenamiento de palabras, como resultado de ciertas
restricciones sintáctico-semánticas (hipótesis inacusativa / inergativa, Burzio 1986; Levin/Rappaport-
Hovav 1995) y sintáctico-pragmáticas (hipótesis de foco/tópico, Sorace 2003, 2005), que rigen el
parámetro de sujeto nulo (inversión sujetoverbo). Sin embargo, en el español caribeño (EC) se ha
observado una tendencia hacia un orden más fijo de Sujeto-Verbo (SV), en comparación con los
dialectos del español nocaribeños (EnC) (Camacho 2011, 2012, en prensa; Morales 1999; Ordóñez y
Olarrea 2006; Ortiz López 2010; Suñer 1983; Toribio 1994), como se puede ver en (1) y (2).
Suñer (1994) ha postulado que en el EC, el llamado Argumental Agreement Licensing (AAL) no
forma parte de su gramática, mientras que Ordoñez y Olarrea (2007) han atribuido esta diferencia a que
el español tiene pronombres de sujeto débiles. Más recientemente, Goodall (2008) y Camacho (en
prensa) argumentan que en el EC (y también en el portugués brasileño, según Camacho), los
pronombres han perdido su peso informacional y sus rasgos de foco y de contraste. Hasta la fecha, son
pocos los estudios cuantitativos sobre este fenómeno, y mucho menos con datos naturales. No sabemos
con precisión qué entornos (sintácticos/semánticos/pragmáticos) impulsan el orden SV, y cuáles lo
detienen (VS).
En este trabajo, investigamos mediante datos naturales qué factores lingüísticos, sociales y
dialectales promueven o impiden el orden SV (1b, 2b), frente a VS (1a, 2a), en el EC frente al español
mexicano, dentro de una perspectiva sociolingüística, variacionista y de cambio lingüístico
(Tagliamonte, 2012). Examinaremos dos corpus de habla semiespontánea grabados in situ: uno de la
República Dominicana y otro de Ciudad de México. El corpus de la República Dominicana incluye 17
participantes de 4 zonas geolectales del país, y de tres sociolectos, mientras que el de Ciudad de
México contiene 12 participantes de la capital, distribuidos en dos sociolectos. Para el análisis
estadístico, según Goldvarb, tomamos en cuenta las variables lingüísticas: animacidad, tipo de sujeto
(pronombre vs. no pronombre), extensión de sujeto no pronominal, especificidad del sujeto, persona y
número del verbo, transitividad del verbo y clase semántica del verbo, y como variables
extralingüísticas, incluimos el género, la edad, el nivel educativo, y el dialecto del hablante. En la
primera parte del estudio, comparamos cuantitativa y cualitativamente el orden de palabras de los
sujetos pronominales y no-pronominales (SV/VS) de los dos dialectos. En la segunda parte,
investigamos la variación del fenómeno en el español dominicano. Los resultados revelan que ambos
dialectos se comportan estadísticamente distintos (p < .001) en cuanto al orden de palabras. Los
hablantes dominicanos muestran una clara preferencia por el orden SV (89.4%), en comparación con
los mexicanos (66.9%). El análisis estadístico de Goldvarb, por un lado, prueba que la pronominalidad,
la transitividad de los verbos, y la clase semántica de los verbos son los factores que mayor peso tienen
en el orden SV en el español dominicano y, por el otro, evidencia que ninguna variable extralingüística,
más allá del dialecto, influye sobre el orden SV/VS. Estos hallazgos demuestran que en el español
dominicano, los pronombres de sujeto y el ordenamiento de palabras han perdido funciones discursivas
(peso informacional, foco y contraste, Sorace 2003, 2005) y semánticas (hipótesis
inacusativa/inergativa (Burzio 1986; Levin/Rappaport-Hovav 1995). Estos resultados también
evidencian que el español dominicano y el EC en general (Ortiz López 2010; Comínguez en progreso)
muestran un cambio tipológico, que reta muchas de las propuestas sintácticas para el español (Gooddall
2008; Ordóñez & Olarrea 2006; Suñer 1992).
60
Ejemplos:
1 (a) ¿Qué quieres tú? (EnC) 2 (a) Llegó Sofía a la fiesta. (EnC)
(b) ¿Qué tú quieres? (EC) (b) Sofía llegó a la fiesta. (EC)
Referencias bibliográficas
Burzio, Luiggi (1986): Italian syntax: A goverment-binding approach. Dorderecht: Foris. Camacho, J.
(en prensa). Null Subjects. New York: Cambridge University Press.
Camacho, J. (2012) The Null Subject Parameter revisited. The evolution from null subject Spanish and
Portuguese to Dominican Spanish and Brazilian Portuguese, ms. (under review).
Camacho, J. (2011). "On the Position of Preverbal Subjects in Spanish. A View from the Discourse
Perspective", ms. Rutgers University (under revision)
Comínguez, J. (en progreso). Microvariation in Caribbean Spanish interrogative whmovement. Rutgers
University.
Goodall, G. (2008). “The limits o syntax in inversion.” Proceedings of the 41st Annual Meeting of the
Chicago Linguistic Society. 161-174.
Levin, Beth/Rappaport-Havav, Malka (1995): Unaccusativity: at the syntax-lexical semantics interface.
Cambridge, MA: MIT Press.
Morales, A. (1999). Anteposición de sujeto en el español del Caribe. In Luis A. Ortiz López (ed.), El
caribe hispánico: perspectivas lingüísticas actuales, 77-98.
Frankfurt and Madrid: Vervuert and Iberoamericana.
Ordóñez, F. y Olarrea, A. (2006). “Microvariation in Caribbean/ non Caribbean Spanish
nterro atives,” robus 18 (2006), 59–96.
Ortiz López, L. (2010). El español del Caribe:orden de palabras a la luz de la interfaz léxico-sintaxis y
sintaxis-pragmática. Revista Internacional de Lingüística Iberoamericana (RILI 2010) 14, 75-
94.
Suñer, M. (1983): Subjects of infinitives in standard and Caribbean Spanish. Ithaca: Cornell University
Press.
Suñer, M. (1992). Subject clitics in the Northern Italian vernaculars and the matching hypothesis.
Natural Language and Linguistic Theory, 10, 641-672.
Suñer, M. (1994). V-movement and the licensing of argumental wh-phrases in Spanish. Natural
Language & Linguistic Theory, 12(2), 335-372.
Sorace, A. (2005). Syntactic optionality at interfaces. In L. Cornips and K. Corrigan (eds). Syntax and
Variation: Reconciling the Biological and the Social , 46-111. Amsterdam: John Benjamins.
Sorace, . (2003) “Near-nativeness”, en Dou hty, Catherine / Lon , Michael (eds.) Handbook of
second language acquisition. Oxford: Blackwell, 130-151.
Tagliamonte, S. (2012). Variationist sociolinguistics change, observation, interpretation. Malden,
Mass.: Wiley-Blackwell.
Toribio, J. (1994) “Dialectal variation in the licensin o null re erential and expletive subjects”, en
Parodi, Claudia et al. (eds.): Aspects of Romance linguistics: Selected papers from the
Linguistics Symposium on Romance Languages XXIV. Washington D.C.: Georgetown
University Press, 409-432.
61
The Vulnerability Hypothesis (VH)
Ana de Prada Perez
University of Florida
Formal proposals within the fields of bilingualism and language contact attempt to delimit cross-
linguistic influence. Weinreich (1953), for instance, develops a hierarchy of permeability for the
different linguistic components (e.g. lexicon is easier to borrow than morphology than syntax). In a
more recent account, couched within the generative tradition, Sorace (2000, 2011) identifies
restrictions on the syntactic component, whereby the core or narrow syntax is impervious to cross-
linguistic effects and these are restricted to the syntax interfaces with other modules (the Interface
Hypothesis, Sorace & Filiaci 2006). While this proposal accounts for numerous data and has been
regularly updated to account for problematic data, this paper enumerates a number of limitations in the
proposal’s assumptions and develops a new hypothesis, the Vulnerability Hypothesis (VH), which
establishes a categorical-variable continuum of permeability, i.e. variable phenomena, where a form is
used close to 50% of the time in a specific context, are permeable while categorical phenomena, where
a form is used close to 100% of the time in a specific context, are not. In this paper we review the
following relevant monolingual and bilingual data in favor of the VH, and against the IH:
(i) Facts: Spanish is a NULL SUBJECT language where null subjects are used, rather categorically, in topic
maintenance (TM) and overt subjects, variably, in topic shift (TS) and, near categorically, in contrastive
focus (CF) contexts. Predictions: IH predicts equal difficulty with TS and CF (interface phenomena), while
VH predicts difficulty with TS (variable), and not with CF (near-categorical). Results: Several studies
(Author 2009, 2010b, submitted, Belletti et al. 2007, Rothman 2009) reveal that bilinguals exhibit more
cross-linguistic effects in TS than in CF, which are both at the syntax-pragmatics interface while only TS is
variable. Theory validated: VH.
(ii) Facts: SUBJECT POSITION in Spanish can be pre or post-verbal depending on focus, which lies at the external
interface, and predicate type, which lies at the internal interface. Importantly, subject position is most
categorical with narrow focus. Predictions: IH predicts more cross-linguistic influence in subject position
across the broad vs narrow focus distinction, while VH predicts less cross-linguistic influence in narrow
focus. Results: Both studies reviewed (Spanish HSs in the US and a study of Catalan-Spanish bilinguals)
reveal that bilinguals are more monolingual-like in narrow focus, which is not predicted by the IH since it is
at the external interface. Theory validated: VH.
(iii) Facts: SUBJECT-TO-SUBJECT RAISING is blocked in Spanish (Torrego 1996) across an experiencer, which is
regulated by the core syntax (specifically by the featural specification of T) and variable. Predictions: the IH
predicts no cross-linguistic influence and the VH predicted cross-linguistic influence. Results: Campos et al.
(submitted) reported that Spanish L2ers exhibit more variability in their judgments than the monolingual
group, which is not predicted by the IH. Theory validated: VH.
This paper presents the VH and reviews relevant monolingual and bilingual data from Spanish-
English and Spanish-Catalan bilinguals with regards to subject expression, position, and subject raising
to contrast it with the IH. As a consequence, this paper contributes significantly to current discussions
in the fields of language contact and bilingualism.
62
C, inversion, and Spanish clause structure: Evidence from code-switching
Shane Ebert
University of Illinois at Chicago
A long-standing debate centers on the roles of C and T in triggering inversion in Spanish
interrogatives. The phenomenon of inversion can be seen in the contrast between example (1a), where
a postverbal subject is grammatical, and (1b), where a preverbal subject renders the sentence
ungrammatical. Interrogative inversion of this type occurs in most dialects of Spanish with simple wh-
phrases that are internal arguments of the verb, including those examined here.
Roughly put, some scholars argue that wh-phrases establish a dependency with T and consequently
move to Spec,T (Barbosa 2001, Goodall 2001, Gutiérrez-Bravo 2005, Zubizarreta 2012, inter alia),
forcing the subject to remain in situ, hence the VS order. Others maintain the more traditional view that
wh-phrases establish dependencies only with C (Chomsky 2000, Pesetsky and Torrego 2001 et seq,
McCloskey 2006), as is the case in many other languages. I present novel evidence from code-
switching (CS) suggesting that it is exclusively the properties of C that determine the presence of
inversion in Spanish. Additionally, I consider the implications of this conclusion for our analyses of
inversion and clause structure more generally.
In code-switching, the language of individual words, and therefore their associated features, can be
independently manipulated. In Spanish, the language of T, expressed on the finite verb or an auxiliary,
is visible and therefore subject to manipulation. I designed and carried out an experimental study of
wh-movement in code-switching in which T was in Spanish while the wh-phrase was drawn from the
lexicon of a language that does not trigger inversion in subordinate clauses (English, German, see
examples (5a) and (5b)). The prediction of the Spec,T approach to inversion in Spanish is that in these
code-switched sentences inversion would be obligatory. This prediction was not confirmed.
I collected data from two groups of participants: (a) 30 Spanish-English bilinguals and (b) 1 Spanish-
German bilingual consultant. Participants provided acceptability judgments of written code-switching
sentences on a 1–5 (group (a)) or 1–7 (group (b)) Likert scale. All participants were highly proficient
speakers in both languages that began acquiring them by the age of 6.
Results revealed that a preverbal subject was grammatical in a number of structures even though the
verb, and therefore T, was in Spanish. This was true with both Spanish-German CS, as in (2a) and (2b),
and Spanish-English CS, as in (3a) and (3b).
If the T-based account of inversion were correct and Spanish T triggered inversion through the
formation of a wh-dependency, then preverbal subjects would not be grammatical in these examples.
Notice that the ungrammaticality of inversion in (2b) and (3b) cannot be due to an incompatibility
between the language of the wh-phrase and T, as VS word order is grammatical with a German or
English wh-phrase and a Spanish T in other structures, as in (4a) and (4b). This suggests that the
alternative account, in which C determines the presence of inversion, is correct. We can explain the
grammaticality of preverbal subjects in (2a) and (3a) if we adopt the hypothesis that C always triggers
movement and further assume that C is in English or German. My conclusions are aligned with
González-Vilbazo and López’s (2013) indin s that C also determines rammatical properties o
negation and null subjects, supporting a broader hypothesis that C determines grammatical properties
of its complement.
We have seen original evidence that it is the C head that triggers inversion in Spanish and not the T
head. This in turn has important implications not only for the structure of wh-questions in Spanish but
63
also clause structure more generally. First, inversion in Spanish occurs not only in wh-questions but
also other structures, including those with negative operators and focused constituents, so the present
CS data also raise questions about inversion more broadly. Second, the C-based account has important
ramifications for the debate concerning the position of preverbal subjects and the EPP in Spanish (cf.
Villa-García, To appear). Finally, a C-based account conflicts with Extended Projection accounts
(Grimshaw 1997, 2000; Gutiérrez-Bravo 2005) that limit the maximal project of most clauses to TP. I
will discuss these and other implications in more detail in my talk.
Examples
(1) a. No sé qué compró Juan. (Spanish)
no know what bou ht Juan ‘ don’t
know what Juan bou ht.’ b. *No sé qué
Juan compró.
(2) a. No sé wen Juan amenazó. (Spanish-German)
no know whom Juan threatened “ don’t
know whom Juan threatened.” b. *No sé wen
amenazó Juan.
(3) a. I don't remember what your cousin probó en el restaurante. (Spanish-English)
tried in the restaurant
“ don’t know that your cousin tried in the restaurant.”
b. *I don't remember what tomaron your brothers esta semana.
(4) a. No sé wie viele Bücher vendió su tío. (Spanish-German) no know how many
books sold your uncle
“ don’t know how many books your uncle sold.”
b. No sé how many texts ha enviado Gimena hoy. (Spanish-English)
no know has sent Gimena today “ don’t know
how many texts Gimena has sent today.”
(5) a. Ich weiß nicht, was Hans verkauft hat. (German)
I know not what Hans sold has
“ don’t know what Hans has sold.”
a. don’t know what John bought. (English)
References
Barbosa, P., 2001. On Inversion in Wh-questions in Romance. In: Pollock, A. H. . J. I. (Ed.), Subject
Inversion in Romance and the Theory of Universal Grammar. Oxford (New York): Oxford
University, pp. 20–59.
Chomsky, N., 2000. Minimalist inquiries: The framework. In: Roger Martins, D. M., Uriagereka, J.
(Eds.), Step by Step. The MIT Press.
González-Vilbazo, K., López, L., 2013. Phase switching. Presented at the Centre for General
Linguistics (ZAS), Berlin, Germany.
Goodall, G., 2001. The EPP in Spanish. Objects and other Subjects , 193–223.
Grimshaw, J., 1997. Projection, Heads, and Optimality. Linguistic Inquiry 28, 373–422.
Gutiérrez-Bravo, R., 2005. Structural Markedness and Syntactic Structure. Routlegde/Taylor and
Francis, New York.
64
McCloskey, J., 2006. Questions and Questioning in a Local En lish. n Ra aella Zanuttini, H ector
Campos, E. H. P. H. P. (Ed.), Crosslinguistic Research in Syntax and Semantics: Negation, Tense,
and Clausal Architecture. Georgetown University Press, pp. 87–126.
Pesetsky, D., Torrego, E., 2001. T-to-C: Causes and Consequences. In: Kenstowicz, M. (Ed.), Ken
Hale: A Life in Language. Cambridge: MIT Press.
Villa-García, J., To appear. On the Status of Preverbal Subjects in Spanish: Evidence for a Dedicated
Subject Position. In: Proceedings of the 42nd Annual Meeting of the North East Linguistic Society
(NELS 42). Amherst, MA.: GLSA.
Zubizarreta, M. L., 2012. A Note on the Spanish Left Periphery. In: Functional Heads: The
Cartography of Syntactic Structures, vol. 7. Oxford University Press, USA.
65
'¡Pues largate!': Complaining Among Mexican-American Bilinguals
Vanessa Elias
Indiana University
Complaints, characterized by LaForest (2002), are expressions “o dissatis action addressed by an
individual to an individual B concernin behavior on the part o B that eels is unsatis actory,” (p.
1596) have not been extensively studied in Spanish (Bolívar, 2002; Márquez Reiter, 2005; Pinto &
Raschio, 2008). Only one such study exists analyzing Spanish in the U.S.; however, no study on
complaints in this dialect has been done utilizing role-plays.
Due to the lack of studies analyzing Spanish complaints among second generation Mexican-
American bilinguals in the United States, role-plays were collected from 21 participants, ten males and
eleven females, who interacted with a female interlocutor. The data was analyzed using Spencer-
Oatey's (2005) Rapport Management in order to gain a better understanding of the politeness strategies
used by this population when complaining. In addition to acting out the role-plays, the participants
were asked to fill out a Language Experience and Proficiency Questionnaire (LEAP-Q), in order to
assess language proficiency. Upon completion of the role-plays, the participants completed a post role-
play questionnaire, which evaluated their impressions of the interactions. The strategies used in the
complaints included, but were not limited to: complaining/accusing, reason/explanation/ justification,
moralizing, expressing gratitude, suggesting/requesting/commanding, requesting information, offering
solution, and providing information. The results showed that the participants preferred the use of
reason/explanation/justification. In the gender analysis the male and female participants chose to utilize
different strategies; however, upon completion of a difference in proportions test this difference was
not significant. In addition, the participants chose to respect the association principle, however, this
result was not statistically significant. Furthermore, the participants chose to respect the involvement
component within the association principle. This result demonstrates that despite the fact that this is a
face-threatening situation, the participants still want to engage in activities with the interlocutor. With
respect to face sensitivities, the participants chose to enhance the interlocutor's identity face when
complaining. Additionally, one of the goals of this study was to determine if any gender differences
exist among this population when forming a complaint. The results demonstrated that no significant
gender differences exist in this population when complaining. This study concludes that due to the fact
that the participants chose to respect the involvement component of the association principle and
demonstrated a tendency toward respecting the association principle, the participants show a tendency
toward positive politeness, which falls in line with other Hispanic cultures such as Cubans (Ruzickova,
1998), Spaniards (Bravo, 2001, 2004; Haverkate, 2004; Hernández Flores, 2004), Argentineans
(García, 2004), Uruguayans (Márquez Reiter, 1997; 2000), Peruvians (García, 2002), and Venezuelans
(García, 2002). Although this study adds to the literature of Spanish in the U.S. pragmatics, further
study of this population is needed.
Referemces
Bolívar, A. (2002). Los reclamos como actos de habla en el español de Venezuela. In M. E. Placencia
& D. Bravo (Eds.), Actos de habla y cortesía en español (pp. 37-53). Munich, Germany: LINCOM
Europa.
Bravo, D. (2001). Sobre la cortesía lingüística, estratégica y conversacional en español. Oralia, 4, 299-
314.
66
Bravo, D. (2004). Tensión entre universalidad y relatividad en las teorías de la cortesía. In D. Bravo &
A. Briz (Eds.), Pragmática sociocultural: Estudios sobre el discurso de cortesía en español (pp. 15-
37). Barcelona: Ariel.
García, C. (2002). La expresión de camaradería y solidaridad: Cómo los Venezolanos solicitan un
servicio y responden a la solicitud de un servicio. In M.E. Placencia & D. Bravo (Eds.), Actos de habla
y cortesía en español (pp. 55-88). Germany: Lincom Europa.
García, C. (2004). Coercion and cooperation: A case study of Argentinean reprimands and responses to
reprimands. In R. Márquez Reiter & M.E. Placencia (Eds.), Current trends in the pragmatics of
Spanish (pp. 231-264). Amsterdam: John Benjamins.
Haverkate, H. (2004). El análisis de la cortesía comunicativa: Categorización pragmalingüística de la
cultura española. In D. Bravo & A. Briz (Eds.), Pragmática sociocultural: Estudios sobre el discurso
de cortesía en español (pp. 55-65). Barcelona: Ariel.
Hernández Flores, N. (2004). La cortesía como la búsqueda del equilibrio de la imagen social. In D.
Bravo & A. Briz (Eds.), Pragmática sociocultural: Estudios sobre el discurso de cortesía en español
(pp. 95-108). Barcelona: Ariel.
Márquez Reiter, R. (1997). Politeness phenomena in British English and Uruguayan Spanish: The case
of requests. Miscelanea: A Journal of English and American Studies, 18, 159167.
Márquez Reiter, R. (2000). Linguistic politeness in Britain and Uruguay: A contrastive study of
requests and apologies. Amsterdam: John Benjamins.
Márquez Reiter, R. (2005). Complaint calls to a caregiver service company: The case of desahogo.
Intercultural Pragmatics, 2(4), 481-513.
Pinto, D., & Raschio, R. (2008). Oye, ¿qué onda con mi dinero?An analysis of heritage speaker
complaints. Sociolinguistic Studies, 2(2), 221-249.
Ruzickova, E. (1998). Cuban facework: Politeness strategies in requests, apologies and compliment
responses. Paper presented at the Linguistics Colloquium. Columbus, Ohio.
Spencer-Oatey, H. (2005). (Im)Politeness, face and perceptions of rapport: Unpackaging their bases
and interrelationships. Journal of Politeness Research, 1, 95-119.
67
Fine-tuning a quantitative typology of code-switching: The Deuchar-Muysken-Wang model
under the magnifying glass of Guaraní-Spanish mixing
Bruno Estigarribia
UNC-Chapel Hill
Muysken’s (2000) ramework distin uishes between NSERT ON L (1a–b), ALTERNATIONAL
(1c), and CONGRUENT LEXICALIZATION (1d) code-switching (CS). Deuchar, Muysken and Wang
(2007, henceforth DMW) provided a quantitative implementation. We discuss improvements to DMW
in (a) determination of switches, and (b) scoring criteria, highlighted by our analysis of Guaraní-
Spanish CS (jopara), from the novel Ramona Quebranto (Ayala de Michelagnoli 1989). DMW restrict
the unit of analysis to the bilingual clause, for which they first determine the Matrix Language (ML:
Myers-Scotton 2002; Jake, Myers-Scotton & Gross 2002). All elements in the Embedded Language
(EL) are switched material. This leaves interclausal switches unaccounted for. For these and for clauses
without a clear ML, elements in the language that occurs SECOND are considered switched (sequential
criterion). Then, each switch is scored on four groups of properties: constituency of the switch, type of
element switched, kind of switch site, and further properties of switches. These are weighted negatively
or positively for each pattern of CS. When added, the pattern with the maximum score is dominant in
the corpus.
We first argue that jopara is true CS (not a mixed language or a variety of Guaraní) and hence that
DMW is applicable. Of note, Jopara has frequent composite ML clauses with either Spanish subject-
verb agreement but Guaraní case-marking (2a, b), or Guaraní inflected verbs in an otherwise Spanish
clause (3). Assuming for the former that the Spanish nouns receiving Guaraní marking are insertions is
clearly inaccurate. Assuming a sequential approach for both creates problems in a corpus where entire
sequences of clauses have composite MLs. We propose that, in corpora heavy in composite ML cases,
the priority should be to minimize the number of switches identified in the clause, especially when a
disambiguating sequential approach fails or gives counterintuitive results (e.g., by identifying most of a
clause as switched).
Furthermore, DMW’s sequential criterion ives undesirable results with “weak” main clauses in
Jopara (4). Such CPs may be syntactic main clauses, but they usually have epistemic interpretations.
Hence, we consider them discourse marker switches and not a case of interclausal switching. This case
is prima acie an issue or any other lan ua e pair with such “weak” clauses (e. . Sp. “parece que”). t
also highlights an issue for future research: We propose a switch-determination algorithm that codifies
these insights EXPLICITLY (and can eventually be partially automatized).
Finally, some have noted that DMW introduce scoring redundancies (Bakaeva 2013). The second
part of this talk presents a combinatorial analysis indicating that DMW is biased in favor of insertion
and congruent lexicalization. For example, summing all possible combinations of scores of the factors
associated with the type of element switched gives overall scores of 196 for insertion and congruent
lexicalization and of -216 for alternation. In fact, this group identifies alternation patterns only in 7 out
of 126 possibilities. When these scores are added to the overall sum for a corpus, they overwhelmingly
disfavor alternation. We propose a reweighting of the model that assigns scores separately for the four
groups of properties first, and then combines them to obtain an unbiased score for the sentence that
reflects the confidence with which it reflects one of the three CS patterns. These scores are then added,
obtaining a classification for the whole corpus. Surprisingly, congruent lexicalization (which is
supposed to happen mainly with two related languages or varieties), is the second most predominant
pattern in this corpus, very closely following insertions. This prompts us to reevaluate, to conclude,
Muysken’s (2000) hypothesized relationships between sociolinguistic factors and structural patterns.
Examples from our corpus (Guaraní morphemes boldfaced, SV agr italicized, argument marking
underlined)
68
1. a. ¡Oheka otro kuña avei!
3Sg-seek other woman too
‘He looks or other women too!’
b. porque no saca su(s) ojo(s) de mi retyma
because no he.takes his eyes of my leg
‘because he does not take his eyes o my le s’
c. ¡Ma(e)rã piko che mombe’u si no podemo(s) remediá(r)!
Why INTERR 1Sg tell if no we.can remedy
‘Why are you tellin me i we can’t ix it!’
d. Nd-a-negá-i voi que che gu(s)tá
NEG-1Sg-deny-NEG EMPH that 1Sg like
‘ do not deny that like him’
e. Cuando salí(s) la iglesia-gui
When youSG.go.out theFEM churchFEM-ABLATIVE
‘When you walk out o church’
f. vemo(s) nue(s)tro casita-pe día a día entra agua
we.see our little.house-LOCATIVE day after day enters water
‘We see that water comes into our little house, day in day out’
g. Su cabeza oiko en otra parte
Her head is elsewhere
‘Her head is elsewhere’
h. Gua’u que e(s)tá haciendo e(s)tione(s) para…
False that is doing negotiations to
‘ t is alse that she is ne otiatin to…’
References
Ayala de Michelagnoli, Margot. 1989. Ramona Quebranto. Asunción, Paraguay: Editorial Arandurã.
Bakaeva, Natalia. 2013. A quantitative approach to defining code-switching patterns. Paper presented
at the Linguistic Evidence 2013, University of Tübingen.
Deuchar, Margaret, Pieter Muysken & Sung-Lan Wang. 2007. Structured Variation in Codeswitching:
Towards an Empirically Based Typology of Bilingual Speech Patterns. International Journal of
Bilingual Education and Bilingualism 10(3). 298–340.
Jake, Janice L., Carol Myers-Scotton & Steven Gross. 2002. Making a minimalist approach to
codeswitching work: Adding the Matrix Language. Bilingualism: Language and Cognition
5(01). 69–91.
Muysken, Pieter. 2000. Bilingual Speech: A Typology of Code-Mixing. Cambridge, UK: Cambridge
University Press.
Myers-Scotton, Carol. 2002. Contact linguistics!: bilingual encounters and grammatical outcomes.
Oxford-New York: Oxford University Press.
69
The role of Executive Function in L2 processing of morphosyntax
María Fionda
University of Mississippi
One aspect of research in second language (L2) acquisition considers the role of cognitive abilities,
namely individual differences in mental capacity, in language processing (Doughty & Long, 2007). The
present study seeks to further our understanding of the relationship between one general cognitive
system, Executive Function (EF), and Spanish L2 online processing of clitic placement errors in the
object clitic + finite verb construction involving the pronouns lo, la, los, las, le and les.
EF is a complex neuropsychological construct whose definition includes the notion of mental
flexibility, the ability to filter interfering information, as well as practice goal-directed behaviours
(Ardila, 2008). While there is considerable research regarding the role of EF in native language
processing (e.g., Biegler, Crowther & Martin, 2008; Ye & Zhou, 2009), little is known about the
relationship between EF and adult L2 processing (e.g., Bialystok & Feng, 2009; Hernandez &
Meschyan, 2006), particularly that of morphosyntax.
Specifically, then, this study examines whether individual differences in EF abilities predict native
English learners' perception of Spanish clitic placement errors when the erroneous word order mirrors
English syntax. Crucially, English does not have a system equivalent to that of the Spanish clitic
pronouns, as they differ both in grammatical category and syntax.
Participants were 29 intermediate-level L2 Spanish learners who took part in two EF abilities tests
(Stroop and Trails Making) and an eye tracking experiment that measured their reading processing of
clitic placement in Spanish. The results show predictions that suggest lower EF abilities translate into
delayed integration of information and clitic placement error detection. The findings are discussed in
terms of the relationship between general cognition, L2 online processing and acquisition.
References
Ardila, A. (2008). On the evolutionary origins of executive functions. Brain and Cognition, 68(1), 92-
99.
Bialystok, E., & Feng, X. (2009). Language proficiency and executive control in proactive interference:
Evidence from monolingual and bilingual children and adults. Brain and Language, 109(2-3),
93-100.
Biegler, K. A., Crowther, J. E., & Martin, R. C. (2008). Consequences of an inhibition deficit for word
production and comprehension: Evidence from the semantic blocking paradigm. Cognitive
Neuropsychology, 25(4), 493-527.
Doughty, C.J. & Long, M. H. (Eds.). (2007). The Handbook of Second Language Acquisition. Malden,
MA: Blackwell Publishing.
Hernandez, A. E., & Meschyan, G. (2006). Executive function is necessary to enhance lexical
processing in a less proficient L2: Evidence from fMRI during picture naming. Bilingualism:
Language and Cognition, 9(2), 177-188.
Ye, Z., & Zhou, X. (2009). Executive control in language processing. Neuroscience and Biobehavioral
Reviews, 33(8), 1168-1177.
70
Conditions for cross-linguistic influence: Evidence from SPAN and ENG nominal compounding
Joshua Frank
The University of Texas at Austin
Key phrases: surface overlap, cross-linguistic influence, nominal compounding, head direction,
compounding parameter, word marker, recursive, productive, container-content relation
Surface Overlap (SO) theory considers environments where one of the two available options in
language A is equivalent to the single option in language B (e.g., Hulk & Müller, 2000; Yip &
Matthews, 2009). Cross-linguistic influence (CLI) will then be in the direction of the structurally
similar option in language A. Yip and Matthews (2009) find evidence that overlap at the word order
level (i.e., isomorphic strings), and not at the abstract level (i.e., features/parameters), conditions CLI.
In the present study I find support for this claim by investigating Spanish and English nominal
compounding strategies via a written dehydrated sentence task (Example 1).
Nominal compounding strategies in Spanish and English diverge in several important ways. In English,
compounding is productive, recursive, and right-headed. Contrastingly, Spanish compounding is
characterized by a lack of both productivity and recursivity, and by leftheadedness. These differences
are not only characteristic of Spanish and English, but of Romance and German language groups in
general (Varela, 2012). For a theoretical analysis, refer to the influential work on the [+/- compounding
parameter] and the [+/- word marker morpheme] (e.g., Harris, 1991a; Piera, 1995; Slabakova, 2002;
Snyder, 2001).
In this study, I specifically analyze the treatment of the container-content relation. English represents
this relation with both a nominal compound (N2-N1) and a noun followed by a prepositional phrase
complement construction (N1-P-N2; e. ., “bottle o / or wine” and “wine bottle”). n Spanish, this
particular relation is only produced as N1-P-N2 (e.g., botella de/para vino). Thus, English is the two
option language, where head directionality is dependent on the construction, while Spanish is the one
construction language. This is represented in terms of SO theory in Table (1).
Results from binary logistic regression analysis indicate a main effect for both language (SPAN and
ENG) and population (L2, L1; Table 2). A post-hoc analysis shows that (n=19) advanced L2 ENG (L1
SPAN) participants produce more ENG N1-P-N2 constructions than (n=16) L1 ENG (advanced L2
SPAN) participants (Table 3). When tested in SPAN, these participants do not display significantly
different compounding strategies (Table 3).
As in Liceras and Díaz (2000), I argue that nominal compounding strategies in the L2 are triggered by
head directionality (i.e., word order). Importantly, the abstract measures of [+/- word marker
morpheme] or [+/- compounding parameter] do not explain the main effect found for language. Head
directionality, when combined with SO theory, does (Table 1). Specifically, CLI persists in the
rammar o advanced L2 En lish speakers because the Spanish “le t” headed construction matches one
of the two available options in English. This interpretation of the results is further supported by the
evidence that SO is a word order level phenomenon, as well as the notion that L2A is driven by shallow
word level processing and not syntactic algorithms or abstract parameters (e.g., Yip & Matthews, 2009;
Clahsen & Felser, 2006; respectively).
Example 1: Sample tokens from ENG and SPAN dehydrated sentence task
71
(1) a. John left the empty bottle in the trash can. (ENG token)
/John/to recycle /bottle/ wine/
My sentence: ______________________________________________.
b. Sara dejó la botella vacía en el basurero. (SPAN token)
/Sara/botar/botella/vino/
Mi oración: ________________________________________________.
Table 1: Surface overlap at the level of head directionality and word order
SPAN ENG
N2-N1
N1-P-N2 N1-P-N2
Table 2: Binary logistic regression analysis for compounding strategy
Independent Variable Coef. Z-score DoF Significance
1. Language (SPAN, ENG) -4.27 -8.30 1 p<0.001
2. Population (L1, L2) 0.27 2.20 1 p=0.03
Table 3: Chi-square post hoc analysis for SPAN and ENG compounding strategy
Group Category Chi-square DoF Significance
1. L1 ENG L2 ENG N2-N1 N1-P-N2 7.118 1 p=0.007
2. L1 SPAN L2 SPAN N2-N1 N1-P-N2 2.631 1 p=0.10
Selected References:
Clahsen, H. and Felser, C. (2006). How native-like is non-native language processing?. Trends in
Cognitive Sciences, 10(12), 564.
Harris, J. (1991a). The exponence of gender in Spanish substantives. Linguistic Inquiry, 22 (1) 27-62
Hulk, A. and Müller, N. (2000). Bilingual first language acquisition at the interface between syntax and
pragmatics. Bilingualism: Language and Cognition, 3(3), 227-244.
Liceras, J. M. and Díaz, L. (2000). Triggers in L2 Acquisition: The case of Spanish N-N compounds.
Studia Linguistica, 54 (2). 197-211.
Piera, C. (1995). On Compouding in English and Spanish. Washington: Georgetown University Press.
Slabakova, R. (2002). The compounding parameter in second language acquisition. Studies in Second
Language Acquisition, 24(4), 507-540.
Snyder, W. (2001). On the nature of syntactic variation: Evidence from complex predicates and
complex word-formation. Language, 324-342.
Varela, S. (2012). Derivation and Compounding. In J. Jualde, A. Olarrea, and E. O’Rourke (Eds.), The
Handbook of Hispanic Linguistics, (pp. 209-226). Wiley-Blackwell.
Yip, V. and Matthews, S. (2009). Cross-linguistic influence in bilingual and multilingual contexts.
Invited paper at the 2009 International Symposium on Bilingualism (ISB7). University of Utrecht,
The Netherlands.
72
The syntax of [hacer + V]: Haciendo borrow in Belizean Spanish
Nicté Fuller Medina
University of Ottawa
This paper examines one kind of verbal borrowing observed in the speech of Belizean Spanish speakers
and proposes a minimalist account of the phenomenon. These borrowings typically involve the Spanish
verb hacer (“to make” or “to do”) in conjunction with an element taken from English as illustrated in
(1).
(1) Este pobre hizo invest
This poor man did/make invest
This poor man invested (money)
(Fuller Medina 2005)
This strategy is productive having been noted in other Spanish-English contact situations (Jenkins
2003; Pountain 2001; Reyes 1976) and has numerous counterparts in other language pairs (see
Wichmann & Wohlgemuth 2005 for an inventory).
The general consensus is that the do-verb, in this case, hacer, does not function as a main verb but
rather as an auxiliary or light verb (e.g. González-Vilbazo & López 2008, 2012). I show that in
Spanish-English data these do-verb constructions exploit vP and, as such, hacer is the spell out of little
v, where v is a functional head (Chomsky 1999). This means that hacer does not originate in VP.
Second, while the borrowed element has been variously referred to as a bare infinitive, a nominal, or
nominalized form, I propose that it cannot be nominal since, for example, as (2) illustrates, the
borrowed element cannot take modification:
(2) *Nos hacían cinco/el penalize
to us did 3PL IMP five/the penalize
Instead, I propose that the borrowed element is verbal and further that it must be a verbal root since its
nearest c-commanding licenser is little v (Harley & Noyer 1999:2). Further to the proposal that hacer is
the spell-out of little v, I suggest that it is the spell out of different flavors of little v in the sense of Folli
and Harley (2007) as the verbal complex can be transitive, intransitive or unaccusative, illustrated in
(3a-c):
(3) a. Mi hermano hizo retire intransitive
My brother did/made retire
(Jenkins 2003:196)
b. Tú estás haciendo relay la información transitive
You are making/doing relay the information
(Fuller Medina 2005)
c. Marta se hizo drop down unaccusative
Marta herself made drop down
Marta fell down
(Fuller Medina 2005)
This proposal allows us to demonstrate the utility of this kind of data in informing theoretical
approaches as we are able to account for an observed phenomena in bilingual data, and, in fact, we
could also account for the prevalence of these constructions cross-linguistically, since these
constructions can be said to appeal to a linguistic universal i.e. little v.
73
References
Chomsky, N. 1999. Derivation by phase. MIT Occasional Papers in Linguistics, 18. MIT.
Edwards, M. & Gardner-Chloros, P. 2007. Compound verbs in codeswitching: Bilinguals making do?
International Journal of Bilingualism, 11(1), 73-91.
Folli, R., & Harley, H. 2007. Causation, obligation, and argument structure: On the nature of little v.
Linguistic Inquiry, 38(2), 197-238.
Fuller Medina, N. 2005. El uso de hacer como estrategia de préstamo: La perífrasis ‘hacer +V’en el
español beliceño. MA Thesis. University of Western Ontario, London, Ontario.
González-Vilbazo, K. E. & López, L. 2008. The little v hypothesis: Evidence from code-switching.
Linguistic Symposium on Romance Languages. University of Illinois at Urbana-Champaign, April 4-6
2008.
---. 2012. Little v and Parametric Variation. Natural Language and Linguistic Theory, 30, 33-77.
Harley, H. & Noyer, R. 1999. State-of-the-article: Distributed Morphology. GLOT International, 4, 3-
9.
Jenkins, D. 2003. Bilingual verb constructions in southwestern Spanish. Bilingual review/Revista
bilingüe, 27(3), 195-204.
Pountain, C. J. 2001. A history of the Spanish Language Through Texts. London; New York:
Routledge.
Reyes, R. 1976. Language mixing in Chicano bilingual speech. In J. Bowen and J. Ornstein, Eds.
Studies in Southwest Spanish. Rowley, Massachusetts: Newbury House, 183-188.
Wichmann, S. & Wohlgemuth, J. 2005. Loan Verbs in a Typological Perspective. In T. Stolz, R.
Palomo and D. Bakker, Eds. Proceedings of “Romanisation worldwide”, Bremen.
74
Silabificación del grupo consonántico /tl/ en español: Un estudio de variación dialectal
Iraida Galarza
Elizabeth Juárez-Cummings
Beatriz Sedó
Gibran Delgado-Díaz
Indiana University-Bloomington
Estudios previos indican que el estatus silábico del grupo consonántico /tl/ está sujeto a variación. En
el español de México, este grupo consonántico se encuentra en diversos préstamos provenientes del
náhuatl en donde puede aparecer en posición inicial, interior y final de palabra (tlacuache, Atlixco,
Popocatépetl) (Schwegler et al. 2010); mientras que en el español general sólo es posible en interior de
palabra. Esta secuencia puede ser tautosilábica (a.tle.ta) o heterosilábica (at.le.ta), según el dialecto
(Mora et al. 2000; Hualde, 2005; Hualde y Carrasco, 2009). Se afirma que en algunos dialectos
peninsulares /tl/ es producida en sílabas diferentes, mientras que en la mayoría de los dialectos
latinoamericanos (a excepción de Puerto Rico) y en dialectos del norte de España y Canarias, se
produce como ataque silábico (Mora et al. 2000; Hualde y Carrasco, 2009). Sin embargo, esta
evidencia a favor de la variación en la silabificación de /tl/ no proviene de análisis experimentales sino
de intuiciones. Hualde y Carrasco (2009) y Mora et al. (2000), en contraste, exploran algunos índices
acústicos de silabificación de /tl/ que demuestran variación dialectal. Hualde y Carrasco (2009)
encuentran que en el español mexicano la duración de /tl/ no se diferencia significativamente de aquella
de /pl/ y /kl/, lo que indica que /tl/ forma un grupo consonántico de ataque, al igual que /pl/ y /kl/. Sin
embargo, en el español de Castilla, se encuentra evidencia de una organización heterosilábica de la
secuencia /tl/, ya que la duración de /tl/ es significativamente menor que la de /pl/ y /kl/, además de que
/t/ está sujeta a debilitamiento, produciéndose frecuentemente como fricativa o aproximante. Mora et
al. (2000) encuentran que en el español venezolano, aunque la duración de /t/ es significativamente
mayor que en el español peninsular, presenta una alta incidencia de ensordecimiento de /l/, lo cual es
consistente con un efecto de coarticulación. Por ello, estos autores concluyen que el español
venezolano presenta una organización tautosilábica, mientras que en el español peninsular /tl/ se
organiza en sílabas diferentes. No obstante, los estudios previos demuestran algunas limitaciones al
comparar pocos dialectos, tener pocos participantes y no tomar en consideración una gama amplia de
índices acústicos de silabificación.
El presente estudio tiene como propósito investigar acústicamente la variación dialectal previamente
observada en la silabificación del grupo consonántico /tl/, tomando en consideración varios índices
acústicos (duración de la secuencia; duración del tiempo de inicio de sonoridad, VOT; evidencia de
debilitamiento de /t/; y presencia y duración de vocal epentética). Para ello, se grabó a 24 hablantes de
tres dialectos diferentes (México, España y Puerto Rico), leyendo una serie de palabras con la
secuencia [oclusiva+líquida], tanto en posición inicial (tlacuache), como en interior de palabra (atleta),
en el contexto de frase portadora. Posteriormente se procedió a un análisis acústico con Praat (Boersma
y Weenik, 2013).
Los resultados preliminares muestran que la secuencia /tl/ en posición media es significativamente más
larga en el español mexicano, ya que en el español peninsular y el puertorriqueño se evidenciaron
diferentes procesos de debilitamiento de /t/ (sonorización, fricativización y elisión). En contraste, en
posición inicial de palabra, la duración del segmento fue mayor en España y Puerto Rico que en
México, debido a la presencia de vocal epentética. Sugerimos que tanto el debilitamiento de /t/
observado en posición media como la inserción de la vocal epentética encontrada en posición inicial de
palabra proveen evidencia de estrategias de evasión del grupo consonántico /tl/ en el español de España
y Puerto Rico, y, por lo tanto, apoyan la organización heterosilábica de la secuencia. Se discute que esta
75
diferencia dialectal que distingue el español de México del de España y Puerto Rico se debe a la alta
frecuencia de este grupo consonántico en el español mexicano en vocablos provenientes del náhuatl.
Key Words: interdialectal production, /tl/ cluster phonetics, syllabification, Nahuatl legacy.
Referencias bibliográficas
Boersma, P. y Weenik D. (2013). Praat: doing phonetics by computer [Computer program]. Version
5.1.49, retrieved May 2013 from http://www.fon.hum.uva.nl/praat/.
Hualde, J. I. (2005). The Sounds of Spanish. Cambridge: Cambridge University Press.
Hualde, J. I. y Carrasco P. (2009). /tl/ en español mexicano. ¿Un segmento o dos? Revista de fonética
experimental XVIII, 175-191.
Mora, E., Cavé, C., Meunier, C., y Meynadier, Y. (2000). El grupo consonántico [tl] en el español y su
aspecto silábico fonológico. Lingua Americana IV, 34-46.
Schwegler, A., Kempff, J. y Ameal-Guerra, A. (2010). Fonética y fonología españolas. Hoboken, NJ:
Wiley.
76
Eye-tracking evidence of null and overt pronouns processing in Spanish
Estela García-Alcaraz
Universitat Pompeu Fabra
Aurora Bel
Universitat Pompeu Fabra
Llorenç Andreu
Universitat Oberta de Catalunya
Elisa Rosado
Universitat de Barcelona
Mònica Sanz-Torrent
Universitat de Barcelona
In recent years, there has been a growing body of research aimed at exploring the antecedent
preferences of referential elements. Pronouns are particularly interesting from this perspective because
they allow observing how speakers integrate morphosyntactic forms with different cues that help in
anaphora resolution. Findings for null-subject and non-null-subject languages have shown that speakers
employ different processing strategies for linking subject pronouns to antecedents. For Italian,
Carminati (2002) formulated the Position of Antecedent Hypothesis (PAH) according to which null
pronouns display a tendency to select subject antecedents whereas overt pronouns prefer non-subject
antecedents. The PAH was also shown to apply to Spanish by means of an offline task (Alonso Ovalle
et al 2002). In addition, experimental studies using eye-tracking techniques (Arnold et al 2000) have
revealed that English speaking adults find it easier to interpret a subject pronoun as coreferential with
the subject of the preceding sentence while English speaking children (Arnold et al 2007) do not show
this first-mention bias because they rather rely on other cues such as gender.
The goal of the present study is to identify the cues employed by Spanish speakers in processing
null and overt pronouns both in gender ambiguous and unambiguous complex sentence contexts. To
address this question, two experiments on a group of undergraduate students, native speakers of
Spanish (N=44, mean age=20.5 years) were conducted. Eye movements of participants were monitored
while watching a picture with two cartoon characters of same and different gender and listening to a
two sentence short story (La abuela recoge a la niña cerca del colegio. Ella / Ø tiene un perrito en los
pies que parece perdido y sin dueño. ‘The grandmother picks the boy up near the school. She/Ø has a
do at her eet that seems to be lost and lonely’).
Experiment 1 tested two conditions (pronoun type & antecedent position, in same gender
context) with two-levels each, and experiment 2 tested two conditions (gender & antecedent position,
with overt pronoun constant). In order to avoid verbs with strong implicit causality bias the verbs of the
first sentence we selected from a normative study (Goikoetxea et al 2008) (verb bias range= 35% to
65%).
An ANOVA was conducted on proportion of looks every 200 ms from first sentence offset. In
experiment 1 eye movements revealed a rapid identification of the pronoun referent suggesting a strong
tendency to link a null pronoun to an antecedent in first position. On the other hand, the overt pronoun
did not show such bias. In experiment 2 a gender effect was identified: gender affected the initial
process of pronoun resolution even though this effect was not very strong. Thus Spanish speakers seem
to rely more on syntactic (function of the antecedent) than on semantic (gender) cues in their choice of
a potential antecedent for a pronoun.
77
La cortesía lingüística en el discurso político: El caso de fíjese, mire y oiga
Inmaculada Garnes
University of Georgia
El presente trabajo tiene como objetivo analizar tres marcadores discursivos de control de
contacto en el discurso político del español peninsular dentro de la cortesía lingüística de Brown y
Levinson (1987): Los enfocadores de la alteridad fíjese, oiga y mire. La función de control de contacto
sirve para manifestar la relación entre los participantes de la conversación y la relación de éstos con sus
enunciados (Martín Zorraquino y Portolés 1999). La forma vocativa de estos marcadores garantiza la
presencia del oyente en el enunciado y esto tiene consecuencias sobre la imagen de los interlocutores
como muestra este trabajo. Estos tres marcadores tienen otras características en común: su origen se
encuentra en el campo semántico de la percepción, se trata de formas imperativas formales con carácter
fático y conativo (Rojas Edgardo 2011), y suelen estar relacionados con el registro informal, pero en
este trabajo se analizan en un registro formal: El debate político.
Veamos primero los siguientes ejemplos para comparar los significados canónicos y no canónicos
de estas partículas:
(1) a. Fíjese en el cuadro y dígame si observa algo extraño
b. Fíjese, en aquella época, claro, no había móviles, también poníamos el teléfono público.
(2) a. Mire este video y después lo comentamos.
b. Mire, es que yo no conozco que haya privatización de la sanidad en España.
(3) a. Oiga bien lo que le voy a decir.
b. Oiga, nosotros hemos hecho siete presupuestos, señor Rajoy.
Los ejemplos en a. muestran el valor canónico de la forma de imperativo singular de los verbos de
percepción fijarse, mirar y oír. En cambio, los ejemplos en b. muestran su significado no canónico, es
decir, el significado procedimental como elementos que sirven para “ uiar de acuerdo con sus distintas
propiedades semánticas y pra máticas, las in erencias que se realizan en la comunicación” ( ortolés
2001: 25-26). Este trabajo se centra en el significado no canónico de estas tres partículas y el objetivo
del estudio es doble, por un lado 1) analizar la función mitigadora que estas partículas realizan en el
enunciado y 2) demostrar que un mismo marcador puede usarse tanto como mecanismo cortesía
negativa como positiva siguiendo a Chodorowska (2008).
El corpus que sirvió de base para el análisis del presente estudio de las partículas fíjese, oiga y mire
es el último debate político entre los candidatos a la presidencia del gobierno y consta de 21.502
palabras. El debate tuvo lugar a pocos días de las elecciones generales en España del 20 de noviembre
de 2011 disponible en la página web www.rtve.es.
Aunque estos debates se consideran como intercambios de enunciados o un diálogo, son en realidad
pequeños monólogos en los que los candidatos emiten enunciados preparados con anterioridad sobre
los diferentes temas a tratar (política exterior, economía, inmigración, etc.). El desacuerdo es la tónica
general entre los participantes en un debate político. Este desacuerdo amenaza la imagen positiva de
ambos interlocutores, por lo que los hablantes utilizan una serie de estrategias para evitarlo (Flores
Acuña 2007: 223), como los marcadores de control de contacto.
78
Los resultados del análisis muestran los usos de fíjese, mire y oiga como estrategias de cortesía
lingüística, como función de focalizadores del enunciado y como partículas catafóricas de un enunciado
despreferido (crítica, opinión contraria a la del interlocutor, etc.) como se puede observar en los
siguientes ejemplos:
(4) Fíjese usted que el presupuesto, la deuda de las diputaciones provinciales es de 3000
millones de euros, señor Pérez Rubalcaba, y la deuda de España es de más de 700.000.
(5) Esto es capital, señor Pérez Rubalcaba. Que es que tú, eh este mismo año nos estuvo
diciendo que en el segundo semestre del año 2011 se iba otra vez a volver a crear empleo.
Mire, hay que decir la verdad y no engañar a la gente. Engañaron con la crisis y engañaron
con muchas más cosas.
(6) Oiga, usted por lo menos debería tener un mínimo de coherencia, o por lo menos debía
decir unas cosas eh que hayan tenido que ver con lo que han hecho.
A pesar de que los tres marcadores comparten funciones mitigadoras, en el análisis se observa que
la partícula oiga introduce enunciados que suponen actos que amenazan la imagen con mayor fuerza
que los anteriores, con los que se puede ofender al oyente de una forma más abierta.
(7) Oiga, usted se ha quedado en el siglo XIX.
(8) Oiga, ¿y por qué no las hizo antes? ¿Es que se le han ocurrido hace media hora?
Por último, cabe destacar que como indicó Chodorowska (2008) para el marcador verás, las
estrategias positiva y negativa se solapan en el mismo marcador, ya que fíjese, mire y oiga invocan al
mismo tiempo cortesía positiva por la inclusión del oyente en la partícula (segunda persona del
singular), y cortesía negativa que aparece codificada en la forma de la partícula (un mandato es una
imposición).
Referencias
Brown, P. y S. Levinson. (Eds.), (1987). Politeness: Some Universals in Language Usage. Cambridge,
University Press, Cambridge.
Chodorowska-Pilch, M. (2008). Verás in Peninsular Spanish as a grammaticalized discourse marker
invoking positive and negative politeness. Journal of Pragmatics, 40, 8, 1357-1372.
Flores Acuña, E. (2007). Los marcadores de control de contacto en el español hablado contemporáneo:
estudio contrastivo español/italiano. Partículas=Particelle. Estudios de lingüística contrastiva
español e italiano, 217-232.
Martín Zorraquino, M.A. y J. Portolés (1999). Los marcadores del discurso. En Bosque, I. y V.
Demonte (eds.). Gramática descriptiva de la lengua española, Madrid, Espasa Calpe, 3, 4051-
4213.
Portolés, J. (2001). Marcadores del discurso, Madrid, Ariel
Rojas Edgardo, G. (2011). La metáfora visual y el control del contacto indicadores de complejidad
sistémica en la fijación de partículas conversacionales. Ponencia en la I Jornada de Jóvenes
Lingüistas (UBA). Disponible en:
http://www.textosenlinea.com.ar/monografias/La%20met%C3%A1fora%20visual%20y%20el%20cont
rol%20del%20contacto.pdf
79
Non-native trends in the acquisition of split intransitivity and its relation to variable input
Inmaculada Gómez Soler
University of Memphis
Background: This study focuses on the acquisition of the Spanish unaccusative/unergative distinction
by (L1 English) L2 learners. Both types of predicates are intransitives but while unaccusatives take
atheta-marked deep-structure object (1a), unergatives take a theta-marked deep-structure subject (1b)
(Perlmutter, 1978).
(1) a. [VP lle o yo] ‘I arrive’ / b. yo [V corro] ‘I run’
I will contrast the predictions of two theories with regard to this phenomenon: the Interface
Hypothesis (IH), which claims that structures that hinge on the interfaces between syntax and other
cognitive systems (i.e. external interfaces) are more problematic for learners than those that do not
depend on this interface (i.e. internal interfaces/narrow syntax) (see Sorace, 2011 for a literature
review) and the Variability Hypothesis (VH) (dePrada Pérez, in progress), which states that categorical
phenomena are easier to acquire than variable phenomena.
Research questions: (1) Are L2ers more target-like with the internal than the external-
interfaceproperties of the Spanish unaccusative/unergative distinction? (a la IH) (2) Are L2ers more
native- like with categorical phenomena regardless of the type of interface to which they belong? (a
la VH)
Methodology and results: I conducted three scalar grammaticality judgment tasks in which 101
participants judged properties associated with these predicates on a Likert scale. The first experiment
tests an internal-interface property: in ‘out o the blue’ contexts, unaccusative verbs have VS order (2a)
and unergative predicates present SV order (2b) (Contreras, 1976; Suñer, 1982).
(2) a. Desaparece María ‘María disappears’ / b. María corre ‘María runs’
Figure 13 shows that even the near-native and advanced learners seem to be insensitive to the
intransitivity split as it is reflected in the syntax (contra IH). The second experiment deals with asyntax-
pragmatics property: when unaccusatives and unergatives appear in subject-focused contexts (e.g. Who
disappeared?/ who runs?) they both exhibit VS order (Zubizarreta, 1998).
(3) a. Desaparece María ‘María disappears’/ b. Corre María ‘María runs’
Only the near-native group performs at a native-level (Figures 2) showing their understanding of the
pragmatic properties that regulate these verbs. Finally, the third experiment deals with a different
internal property: the use or misuse of anticausative se with change-of-state unaccusatives, (4a)
unaccusatives (4b) and unergatives (4c) (Toth, 2000). All groups showed a tendency in the right
direction in this experiment in contrast with experiment 1 and 2.
(4)a. El vaso se rompió ‘The glass broke’/ b.*María se llegó ‘María arrived’/ c. *María se lloró
‘María cried’
Discussion and conclusion: The IH predicts that learners would be more accurate in experiment 1 (in
line with Lozano’s (2006) indin s on the acquisition o split intransitivity) and 3 since they
testinternal-interface properties. On the other hand, the VH predicts that L2ers would be more target-
like in experiment 3 since it tests the most categorical phenomenon. My findings clearly support the
VH: a comparison of experiments 1 and 3 shows that sensitivity to the syntactic reflexes of split
intransitivity does not develop monolithically for all internal-interface properties alike. Instead, some
properties are understood at a native-like level (experiment 3) while others are not (experiment 1). The
key to this asymmetry lays in the consistency of the native input which is fairly variable in the first and
3 Due to space constraints, I will only present the results for experiments 1 and 2 with unaccusative
verb.
80
second experiment (Figure 1 and 2) but not in the third one (Figure 3). We can take this conclusion a
step further and propose that delay in the pragmatic experiment is also related to variable native input
(Figure 2) but I leave this question to future research. In conclusion, inconsistency in the native input
causes a delay in the acquisition of the reflexes of split intransitivity whether they are of a syntactic or
of a pragmatic nature. Thus, stability of native input becomes a key factor to predict L2 acquisition
success at all levels as stated in the VH.
Selected References
De Prada Pérez, A. (in progress). The Variability Hypothesis.
Lozano, C. (2006). Focus and split intransitivity: The acquisition of word order alternations innon-
native Spanish. Second Language Research, 22: 1-43
Perlmutter (1978). Impersonal passives and the unaccusative hypothesis. Proceedings of the Fourth
Annual Meeting of the Berkley Linguistics Society, 157-189.
Sorace, . (2011). innin down the concept o “inter ace” in bilin ualism. Linguistic Approaches to
Bilingualism , 1 (1), 1-33.
Toth, P.D. (2000). The Interaction of Instruction and Learner-Internal Factors in the Acquisition of L2
Morphosyntax. Studies in Second Language Acquisition , 22 (2), 169-208.
Zubizarreta, M. L. (1998). Prosody, Focus and Word Order. Cambridge, MA: MIT Press.
81
On the weakness of Spanish pronouns: Evidence from Spanish/English code
switching
Kay Gonzalez-Vilbazo Bryan Koronkiewicz
University of Illinois at Chicago
Based on Cardinaletti and Starke (1999) Spanish personal pronouns can be classified as either
strong or weak. Strong pronouns project a Determiner Phrase (DP), whereas weak pronouns do
not. The presence vs absence of the DP layer in the projection of pronouns has various syntactic,
prosodic and phonological consequences.
For example, consider the Mexican-‐Spanish sentences in (1). We can see that the pronoun él
‘he’ can be coordinated (1a), modified (1b) and prosodically stressed (1c). According to
Cardinaletti and Starke’s (1999) proposal, these are all examples of a strong pronoun, i.e.
projecting a full DP. However, in (1d) the case is less clear.
The pronoun in (1d) has the same phonetic realization as (1a-‐c), namely él, suggesting it might
also be a strong pronoun. However, in (1d) the pronoun is neither coordinated, modified, nor
stressed. Cardinaletti and Starke (1999) specifically argue that strong and weak pronouns can be
homophonous in some languages, i.e. the pronoun él could be either strong or weak. Based on
the criteria the authors provide, there is not a straightforward reason to label the pronominal form
in (1d) as either strong or weak. An unstressed subject pronoun in canonical position does not
violate any of the criteria provided for either type.
We turn to code-‐switching to provide evidence that the pronoun in (1d) is actually weak. In
order to do that we show that code-‐switching weak pronouns results in unacceptable structures,
whereas switching strong pronouns seems to be acceptable. Then we show that the pronoun in
(1d) cannot be switched and conclude that this type of pronoun is weak. At least since Timm (1975) and through recent research (Jake, 1994; van Gelderen and
MacSwan, 2008) it has been noted that there is a strong restriction against code-‐switching
pronouns. This contrasts sharply with switching lexical DPs, which are more generally
acceptable, as exemplified in (2). However, we have found that it is acceptable to switch pronouns if
they are coordinated (3a), modified (3b) or prosodically stressed (3c), i.e. if they are strong.
We present the findings of a pilot study that has been run testing all of the various structures in
(2-‐3), both from Spanish-‐to-‐English code-‐switching and vice versa. The data obtained are
grammaticality judgments on a seven-‐point Likert scale and the stimuli were presented in written
form. All participants were either simultaneous or early-‐sequential Spanish-‐English bilinguals of
Mexican descent living in the US. Our methodology follows the standards discussed in González-‐Vilbazo et al. (2013).
Our data analysis shows that normally a code-‐switch between a pronominal form and the finite verb leads to lower grammaticality ratings (as found in previous literature). However, when these pronouns are coordinated, modified or prosodically stressed such a switch is no longer restricted, i.e. These latter pronoun code-‐switches show ratings comparable to those of lexical DPs like in (2b).
We propose that the distinction with respect to code-‐switching pronouns is in line with the
notion of strong and weak forms as proposed by Cardinaletti and Starke (1999). First, pronouns
that are coordinated, modified or prosodically stressed can be switched due to their DP-‐like
structure, confirming their status as strong pronouns. Furthermore, we are able to categorize the
previously unclear pronoun in (1d) as weak. An unstressed pronoun in standard subject position
cannot be switched, which only follows for a weak pronoun as it lacks a DP shell.
82
Examples
(1) a. Él y Julia tienen mucho dinero. 3SG.MASC and have a-‐lot money ‘He and Julia have a lot of money.'
b. Él de la camisa roja tiene mucho dinero.
3SG.MASC of the shirt red has a-‐lot money
‘Him in the red shirt has a lot of money.’
c. Ellos son pobres, pero ÉL tiene mucho dinero.
3PL.MASC are poor but 3SG.MASC has a-‐lot money
‘They’re poor, but HE has a lot of money.’
d. Él tiene mucho dinero.
3SG.MASC has a-‐lot money
‘He has a lot of money.’
(2) * Él works all the time. b. Ese hombre works all the time.
3SG.MASC that man
‘He works all the time.’ ‘That guy works all the time.’
(3) a. Él y Julia work all the time. 3SG.MASC and
‘He and Julia work all the time.’
b. Él de la camisa roja works all the time. 3SG.MASC of the shirt red
‘Him in the red shirt works all the time.’
c. Ellos son perezosos, pero ÉL works all the time. 3PL.MASC are lazy but 3SG.MASC ‘They’re lazy, but HE works all the time.’
References
Cardinaletti, Anna and Michal Starke. 1999. The typology of structural deficiency a case study
of the three grammatical classes. In Clitics in the Languages of Europe, ed. Henk van
Riemsdijk. Berlin: Mouton de Gruyter.
González-‐Vilbazo, Kay, Laura Bartlett, Sarah Downey, Shane Ebert, Jeanne Heil, Bradley
Hoot, Bryan Koronkiewicz, and Sergio E. Ramos. 2013. Methodological considerations in code-‐switching research. Studies in Hispanic and Lusophone Linguistics 6 (1), 118-‐138.
83
Jake, Janice L. 1994. Intrasentential code-‐switching and pronouns: On the categorial status of
functional elements. Linguistics 32-‐2, 271–298.
Timm, Lenora A. 1975. Spanish–English co d e -‐switching: El p o rqué and h o w -‐not-‐to.
Romance Philology 28, 473–482.
van Gelderen, Elly and Jeff MacSwan. 2008. Interface c o n d i t i o n s a n d c o d e -‐switching:
Pronouns, lexical DPs, and checking theory. Lingua 118, 765–776.
84
Spanish future of probability: teaching and learning
Irina Goundareva
University of Ottawa
In this study I attempt to answer the following questions: what is future of probability, how it is
acquired and how it is treated in the current teaching materials. The complexity of the Spanish future of
probability (SFP) for language learners has been brought to the attention of linguists and language
instructors because of its special position at the interface of morphology, syntax, semantics, and
discourse/pragmatics. Spanish future morphology can be interpreted as future time event or present
time probability, i.e. how probable the speaker considers the event or action in the present time. I am
particularly interested in the place of the SFP in foreign language teaching and learning.
Previous research on Second Language Acquisition of SFP (e.g. Bruhn de Garavito and
Valenzuela 2007) by English speaking learners of Spanish supports the claim that learners are able to
acquire the epistemic interpretation of future morphology in spite of the ambiguity of the input. Soto
(2008) also researched how learners acquire the Spanish future morphology along with its various
interpretations and uses. Based on the previous research, I conclude that the future of probability
provides a rich testing ground since it involves aspects of the language that are taught in the classroom
(future time) and aspects that are not explicitly taught (epistemic interpretation).
My project investigates the differences between the two main forms of instruction: form-
focused (explicit grammar and usage instruction) and meaning-based (some grammar instruction in rich
context, examples, without explicit usage instruction) of Spanish future of probability for intermediate
learners in order to develop didactic materials for the classroom. The two instructional groups received
two types of instruction and were tested before the instruction, immediately after and four weeks after
the instruction. In order to measure the results of the teaching materials, I developed two experimental
tasks: an untimed grammaticality judgment task and a limited written production task. The reason for
such investigation is a lack of clear explanation and variety of exercises in current Spanish textbooks
available for classroom instruction and which follow the Common European Framework of Reference
for Languages.
The results suggest that the meaning focused instruction has better results than the form focused
instruction in both short and long term. Both instructional groups outperformed the control group on
both interpretation and production tasks.
85
The Eventivity Constraint and Child Spanish Root Infinitive Verbs
John Grinstead
The Ohio State University
Paij Lintz
The Ohio State University
Juliana de la Mora
Universidad Autónoma de Querétaro
Myriam Cantú-Sánchez
Universitat Autònoma de Barcelona
Blanca Flores
Centro Nacional de Rehabilitación
It has been suggested that some child languages, such as Spanish, do not have a root infinitive stage
(e.g. Wexler 1998). However, a range of evidence now suggests the contrary. Diverse data collection
techniques, including elicited production with real words (Bedore & Leonard 2001, Grinstead et al
2009), elicited production with nonce words (Kernan & Blount 1966, Pérez-Pereira 1989),
grammaticality judgment (Grinstead et al 2009) and spontaneous production (Lintz 2012) all indicate
that monolingual typically-developing child Spanish-speakers pass through a stage during which they
use an array of verb forms that do not mark tense. Similarly, child Spanish-speakers diagnosed with
SLI appear to pass through an Extended Optional Infinitive Stage, in the sense of Rice & Wexler
(1996), to the extent that child Spanish-speakers with SLI can be identified with fair to good sensitivity
and specificity on the basis of their tense marking (Grinstead et al 2013). Another characteristic of
many languages in which children less controversially pass through a root infinitive stage is what
Hoekstra & Hyams (1998) refer to as the Eventivity Constraint. This constraint is argued to hold of
child French and Dutch, for example, and it consists of children restricting their nonfinite forms
asymmetrically to predicates that are “eventive”, which is to say telic or atelic, while inite orms tend
to occur primarily on stative predicates. In this study, we examine whether child Spanish grammars are
restricted by the Eventivity Constraint. If so, we take it to be further evidence of the root infinitive
stage in child Spanish. To test this hypothesis, we examined the language of 38 monolingual Spanish-
speakers in Mexico, 19 of whom are diagnosed with SLI (mean age = 67 months; mean MLUw = 2.97)
and 19 of whom are age-matched controls (mean age = 67 months; mean MLUw = 5.49). We coded the
spontaneous production o each child’s 20-30 minute spontaneous production sample for tense,
grammatical aspect and lexical aspect. Of the 276 root infinitives that these children used, significantly
more were eventive (69%) than stative (31%), consistent with the Eventivity Constraint (Wilcoxon
Signed Rank test, Z score = -3.934, p < .001). This contrasts with the total set of finite and nonfinite
verbs the children produced, in which proportions of eventives (1219) and statives (1019) were not
significantly different from each other (p > .05). In addition, the children in the sample were also given
a Grammaticality Choice task to measure their knowledge of tense. The results of this tense measure
correlate positively with children’s proportions o all verbs that occur on stative predicates (r = .297,
p=.034), which should be primarily finite, according to the Eventivity Constraint. Finally, children with
SLI use a significantly larger proportion of eventive predicates (12%) than do age-matched typically-
developing children (3%), as the Eventivity Constraint would seem to predict (Z score = -3.114, p =
.002), given the documented tendency of children with SLI to use more root nonfinite verbs than same-
age peers.
86
References
Bedore, L., & Leonard, L. (2001). Grammatical Morphology Deficits in Spanish-speaking children
with specific language impairment. Journal of Speech, Language, and Hearing Research, 44(4),
905-924.
Grinstead, J., Baron, A., Vega-Mendoza, M., De la Mora, J., Cantú-Sánchez, M., & Flores, B. (2013).
Tense Marking and Spontaneous Speech Measures in Spanish SLI: A Discriminant Function
Analysis. Journal of Speech, Language, and Hearing Research, 56(February), 1-12.
Grinstead, J., De la Mora, J., Pratt, A., & Flores, B. (2009). Temporal Interface Delay and Root
Nonfinite Verbs in Spanish-Speaking Children with Specific Language Impairment: Evidence from
the Grammaticality Choice Task. In J. Grinstead (Ed.), Hispanic child languages: typical and
impaired development (pp. 239-263). Amsterdam: John Benjamins.
Grinstead, J., De la Mora, J., Vega-Mendoza, M., & Flores, B. (2009). An Elicited Production Test of
the Optional Infinitive Stage in Child Spanish. In J. Crawford, K. Otaki & M. Takahashi (Eds.),
Proceedings of the 3rd Conference of Generative Approaches to Language Acquisition - North
America (pp. 36-45). University of Connecticut, Storrs: Cascadilla Press.
Hoekstra, T., & Hyams, N. (1998). Aspects of Root Infinitives. Lingua, 106(1-4), 81-112.
Kernan, K. T., & Blount, B. G. (1966). The Acquisition of Spanish Grammar by Mexican Children.
Anthropological Linguistics, 8(9), 1-14.
Lintz, P. (2012). Overt Subject Licensing and Inflectional Verb Morphology in Child Spanish Speakers
with SLI. Unpublished MA Thesis, The Ohio State University, Columbus.
Perez-Pereira, M. (1989). The Acquisition of Morphemes: Some Evidence from Spanish. Journal of
Psycholinguistic Research, 18(3), 289-312.
Rice, M. L., & Wexler, K. (1996). Toward Tense as a Clinical Marker of Specific Language
Impairment in English-Speaking Children. Journal of Speech and Hearing Research, 39(6), 1239-
1257.
Wexler, K. (1998). Very Early Parameter Setting and the Unique Checking Constraint: A New
Explanation of the Optional Infinitive Stage. Lingua, 106(1-4), 23-79.
87
Focus at the syntax-discourse interface: optionality reconsidered
Timothy Gupton
University of Georgia
Much recent generative research focuses on optionality in developing grammars, which has been
attested in L1 acquisition (e.g. Poeppel & Wexler 1993; Wexler 1994, 1998), in adult L2 acquisition
(e.g. White 1992; Sorace 1999, 2000), and in child L2 acquisition (e.g. Pladevall 2010). It has been
proposed that structures involving the interface between narrow syntax and external modules (e.g.
discourse) are more likely to exhibit residual instability (e.g. Sorace 2000, 2004, 2005; Tsimpli et al
2004). Research on the acquisition of word order among advanced L2 learners in particular has
suggested that these learners experience problems at the syntax-discourse interface. Some studies (e.g.
Hertel 2003; Lozano 2006) claim that residual optionality is symptomatic of interface instability, in
accordance with the Interface Hypothesis (IH, e.g. Sorace & Filiaci 2006), while others (e.g.
Domínguez & Arche 2008) claim that problems at the syntax-discourse interface result from a purely
syntactic deficit causing a restructuring of the grammar – essentially overgeneralizations based on
ambiguous input on the part of the learner. Although Sorace (2011) has suggested that developmental
trends in SLA are an unwarranted extension of the IH, I follow White (2011) in supporting the need for
research on developmental trends in interface phenomena.
The current study seeks to further explore the syntax-discourse interface by examining the L2
acquisition of Spanish word order variation with transitive and unaccusative predicates in broad focus,
narrow-focus and contrastive focus contexts. I present data from the following participant groups:
intermediate learners, a subset of which completed the same tasks at an interval of ten weeks, advanced
non-natives who use Spanish for professional purposes, and two groups of Spanish native speakers, one
group that has spent extended periods of time in English-speaking countries and one group that has not.
I present data from a contextualized Appropriateness Judgment Task and a contextualized Word Order
Preference Task. For both tasks, participants read contexts designed to construct a variety of discourse
scenarios and then evaluated audio stimuli read by native speakers in order to control for intonation and
prevent accommodation of an undesired reading. The results of the study suggest that learners exhibit
early delays in the acquisition of the narrow syntactic and discourse-conditioned word order variation.
Advanced learners acquire core syntax in a near native-like fashion but also display levels of interface
instability with respect to two focus-related properties in Spanish: 1) contrast, which, unlike English,
may be expressed on direct objects by intonation alone in situ (1a), or by dislocation in addition to
intonation (i.e. focus fronting, 1b), and 2) narrow focus, which has been claimed to target the rightward
clause edge (Zubizarreta 1998, bold-face subject for expositional purposes). Unexpectedly, native
speakers exhibited a higher level of optionality than the learner groups, who were comparatively more
categorical in their judgments. The learners in this study then exhibit interface instability in not
recognizing optionality in a native-like fashion. In fact, these advanced learners display a lesser degree
of optionality than their native counterparts (see also Prévost 2011 on native-speaker optionality).
Curiously, word order optionality among native-speaker grammars was more heavily pronounced in
speakers lacking exposure to the learner L1 (English). These results, however unexpected, suggest that
greater care be taken with native-speaker control data, especially when making claims about residual
optionality in highly advanced and near-native learners.
Examples
(1) a. Compró UNOS PANTALONES.
b. UNOS PANTALONES compró.
He bought SOME PANTS (not a t-shirt)
.
88
(2) Context: ¿Quién ha comprado esto?
Lo ha comprado Paco.
Context: Who bought this?
Paco bought it.
References
Domínguez, Laura & María J. Arche. 2008. Optionality in L2 grammars: the acquisition of SV/VS
contrast in Spanish. BUCLD 32 Proceedings, 96-107. Somerville (MA): Cascadilla Press.
Hertel, Tammy. 2003. Lexical and discourse factors in the second language acquisition of Spanish
word order. Second Language Research 19(4), 273-304.
Lozano, Cristobal. 2006. Focus and Split intransitivity: the acquisition of Word order alternations in
non-native Spanish. Second Language Research 22(2), 145-187.
Pladevall Ballester, Elisabet. 2010. Child L2 development of syntactic and discourse properties of
Spanish subjects. Bilingualism: Language and Cognition 13(2), 185-216.
Poeppel, David & Kenneth Wexler. 1993. The full competence hypothesis of clause structure in early
German. Language 69, 1-33.
Prévost, Philippe. 2011. The Interface Hypothesis: What about optionality in native speakers?
Linguistic Approaches to Bilingualism. 1(1). 79-83.
Sorace, Antonella. 1999. Initial states, end-states, and residual optionality in L2 acquisition. BUCLD 23
Proceedings. Somerville (MA): Cascadilla Press.
Sorace, Antonella. 2000. Syntactic optionality in non-native grammars. Second Language Research 16.
93-102.
Sorace, Antonella. 2004. Native language attrition and developmental instability at the syntax-
discourse interface: data, interpretations and methods. Bilingualism: Language and Cognition 7.
143-145.
Sorace, Antonella, 2005. Syntactic optionality at interfaces. In L. Cornips and K. Corrigan (eds.).
Syntax and Variation: Reconciling the biological and the social. 46-111. Amsterdam: John
Benjamins.
Sorace, ntonella. 2011. innin down the concept o “inter ace” in bilin ualism. Linguistic
Approaches to Bilingualism. 1(1). 1-33.
Sorace, A. and F. Filiaci. 2006. Anaphora resolution in near-native speakers of Italian. Second
Language Research 22. 339-368.
Tsimpli, I., Sorace, A., Heycock, C. and F. Filiaci. 2004. First language attrition and syntactic subjects:
a study of Greek and Italian near-native speakers of English. International Journal of
Bilingualism 8. 257-277.
Wexler, Kenneth. 1994. Optional infinitives, head movement, and the economy of derivations in child
grammar. In D. Lightfoot & N. Hornstein (eds.). Verb movement. 305-350. Cambridge (MA):
Cambridge University Press.
Wexler, Kenneth. 1998. Very early parameter setting and the unique checking constraint. Lingua 106.
23-79.
White, Lydia. 1992. Long and short verb movement in second language acquisition. Canadian Journal
of Linguistics 1. 337-360.
White, Lydia. 2011. The Interface Hypothesis: How far does it extend? Linguistic Approaches to
Bilingualism 1(1), 108-110.
89
Monoptongación de [aw] primario y secundario en español medieval
César Gutiérrez
Purdue University
Como consecuencia de diversos procesos fonéticos, en castellano surgieron diptongos [aw] que no
existían en latín (véase 1a-c). Estos diptongos secundarios tendieron a monoptongar en [o] como los
[aw] primarios que poseía originalmente la lengua latina (v. gr., CAUSAM > cosa, TAURUM > toro). Se
ha propuesto que esta convergencia de ambos tipos de diptongos en el mismo resultado se debió a dos
procesos de monoptongación independientes, uno ocurrido durante el periodo latino, el cual actuó
sobre los [aw] primarios, y otro ocurrido durante el periodo romance, que actuó sobre los [aw]
secundarios (Menéndez Pidal, 1941; Lapesa, 1981; Lloyd, 1993).
Esta explicación se apoya en dos argumentos. Por un lado, en que las grafías utilizadas para
representar los diptongos secundarios son diferentes de las utilizadas para representar los diptongos
primarios, y por otro, en que las grafías que reflejan diptongo secundario no dejan paso a las que
reflejan monoptongo hasta muy tarde (siglo XII) (véase 2a-b) (Menéndez Pidal, 1956).
Partiendo de estos planteamientos, en esta presentación se mostrará que en castellano el proceso
de monoptongación de los diptongos secundarios fue el mismo que afectó a los primarios en latín y que
dicha monoptongación tuvo lugar muy tempranamente en el periodo de orígenes. Para ello se utilizarán
evidencias procedentes de la cronología relativa (interacción de la monoptongación con la vocalización
de consonantes en coda silábica y con la aparición de la yod 4ª) y de la cronología real (presencia de
testimonios escritos fechables en la documentación medieval). Esto permitirá comprobar que tanto la
existencia de diferentes grafías para los [aw] primarios y secundarios como la tardía imposición de
grafías con monoptongo no respondieron a diferentes pronunciaciones en castellano, sino al uso de un
sistema de escritura logográfico durante el periodo altomedieval.
(1) a. Metátesis
HABUIT > [ˈawbi] > ove
b. Síncopa
AMAUIT > [aˈmawt] > amó
c. Vocalización
SALTUM > [ˈsawto] > soto
(2) Grafías para los diptongos primarios y secundarios en iberorromance
a. Diptongos primarios:
<au>, <ou>, <ob>, <op>, <ab>, <oc>, <o>
b. Diptongos secundarios
<au>, <auc>, <ao>, <oc>, <o>, <u>
Bibliografía Lapesa, R. (1981), Historia de la lengua española, Madrid, Gredos.
Lloyd, P. M. (1993), Del latín al español, I: fonología y morfología históricas de la lengua española,
Madrid, Gredos.
Menéndez Pidal, R. (1941), Manual de gramática histórica española, Madrid, Espasa-Calpe.
– (1956), Orígenes del español. Estado lingüístico de la Península Ibérica hasta el siglo XI, Madrid,
Espasa-Calpe.
90
Task effects on comprehension: Acceptability judgments versus comprehension questions
Rosa Guzzardo Tamargo
Universidad de Puerto Rico
Paola Dussias
Penn State University
Psycholinguistic studies that focus on sentence comprehension have shown that the way in
which participants process sentences is heavily influenced by the particular task that they are given
(Hahne & Friederici, 2002; Macizo & Bajo, 2006; Newman et al., 2009; Williams, 2006). However,
this topic is still much understudied in other linguistic arenas. In this study, we extend the examination
of task effects to the code-switching context. Specifically, we examined if different tasks, namely an
acceptability judgment task and a comprehension task, influenced the processing patterns of Spanish-
English bilinguals during their comprehension of code-switches.
Two types of switches were compared here: switches between the Spanish auxiliary estar
‘to be’ and an En lish present participle (“los actores están rehearsin their lines”), which are more
frequent in naturalistic data, and switches between the Spanish auxiliary haber ‘to have’ and the
En lish past participle (“los actores han rehearsed their lines”), which are less requent in code-
switching corpora (Guzzardo Tamargo, 2012; Pfaff, 1979; Poplack, 1980).
Forty-two early bilin uals rom New York were recruited or the study. articipants’ eye
movements were recorded as they read code-switched sentences on a computer screen. The
experimental stimuli comprised 96 item sets, each consisting of four versions of the same sentence,
corresponding to four experimental conditions. Conditions 1 and 2 were code-switched conditions with
the pro ressive structure. n Condition 1 (C1), the switch occurred at a phrasal boundary (“los actores
are rehearsin their lines”) and in Condition 2 (C2), the switch occurred between the auxiliary and the
participle (“los actores están rehearsing their lines”). Conditions 3 (C3) and 4 (C4) were analo ous to
C1 and C2, but they involved the per ect structure (“los actores have rehearsed their lines” and “los
actores han rehearsed their lines,” respectively).
Participants completed two reading blocks. In one block, after reading each sentence, they were
asked to answer a comprehension question about the content of the sentence. In the other block, they
performed an acceptability judgment on the sentence.
A two-way repeated measures ANOVA was conducted for each reading block, with auxiliary
type (estar versus haber) and switch site (auxiliary versus participle) as within-subjects factors. In the
comprehension question block, results revealed a significant interaction of auxiliary type and switch
site for gaze duration [Means: C1-347ms, C2-384ms, C3-316ms, C4427ms, F(1,41) = 5.30, p = .026]
and for total time [Means: C1-691ms, C2-750ms, C3-650ms, C4-944ms, F(1,41) = 9.49, p = .004]. The
regression path time results exhibited a main effect of switch site [Means: C1-526ms, C2-606ms, C3-
518ms, C4-699ms, F(1,41) = 17.05, p < .001]. In the acceptability judgment block, results displayed
main effects of switch site for all three reading measures: gaze duration [Means: C1-339ms, C2-442ms,
C3-331ms, C4-444ms, F(1,41) = 30.07, p < .001], regression path time [Means: C1-581ms, C2-719ms,
C3-590ms, C4-860ms, F(1,41) = 21.14, p < .001], and total time [Means: C1-778ms, C2-1130ms, C3-
820ms, C41281ms, F(1,41) = 21.14, p < .001].
The results indicate that the two tasks brought about different reading patterns. When reading
for comprehension, participants displayed processing difficulties with haber+English participle
switches compared to estar+English participle switches, suggesting that production frequency of the
two types of switches is linked to the processing patterns they entail in comprehension. When
91
performing the acceptability judgment task, participants exhibited processing difficulties with both
types of switches, such that their reading patterns no longer mirrored natural production patterns. These
results suggest that code-switching findings that hinge solely on judgment data should be re-evaluated.
References
Guzzardo Tamargo, R. E. (2012). Linking comprehension costs to production patterns during
the processing of mixed language. (Unpublished doctoral dissertation). The Pennsylvania State
University, University Park, PA.
Hahne, A. & Friederici, A. D. (2002). Differential task effects on semantic and syntactic
processes as revealed by ERPs. Cognitive Brain Research, 13, 339-356.
Macizo, P. & Bajo, M. T. (2006). Reading for repetition and reading for translation: Do they
involve the same processes? Cognition, 99(1), 1-34.
Newman, S. D., Lee, D., & Ratliff, K. L. (2009). Off-line sentence processing: What is involved
in answering a comprehension probe? Human Brain Mapping, 30, 2499-2511.
Pfaff, C. W. (1979). Constraints on language mixing: Intrasentential code-switching and
borrowing in Spanish/English. Language, 55, 291-318.
Poplack, S. (1980). Sometimes ’ll start a sentence in Spanish y termino en español Toward a
typology of code-switching. Linguistics, 18(7-8), 581-618.
Williams, J. N. (2006). Incremental interpretation in second language sentence processing.
Bilingualism: Language and Cognition, 9(1), 71-88.
92
PRO = pro
Peter Herbeck
University of Salzburg
Background: Recent studies have revealed that subjects can receive a phonetic shape in several
nonfinite configurations (see e.g. Szabolcsi 2009). Spanish and Catalan present an example of two
languages that only allow morphologically pronominal elements to appear as overt subjects of
obligatorily controlled complements (see (1) – (2)). lthou h ‘emphatic pronouns’ have o ten been
analyzed as adjoined to V and, hence, doubles o the ‘real’ thematic subject RO, there is evidence
that they do not have the same properties as Englishtype emphatic adjunct anaphors (cf. (3); Solà 1992,
Barbosa 1995). Thus, the complementary distribution of emphatic pronouns with inverted subjects (and
objects) indicates that we are in fact dealing with a real thematic subject in its base position rather than
an adjoined element (Barbosa 1995). Further evidence or the existence o ‘pronounced RO’ comes
from the personal infinitive (see (4) – (5)). This indicates that Spanish-type non-finite configurations
allow pronounced subjects in a similar vein as they allow instances o ‘pronounced pro’ in inite
domains. Final evidence for a PRO/pro unification in Spanish comes from finite control configurations
(see (6) – (7)).
Proposal: We would like to argue that the PRO vs. pro distinction does not exist as a lexical
differentiation. Both types of empty categories are D elements with fully variable phi-features at the
moment o External Mer e (they are “re erence variables” (Si urðsson 2008)) whose status with
respect to Binding Theory is determined by means of Agree with T (see (8) – (10)). Following Barbosa
(1995), T in Romance pro-drop is associated with interpretable phifeatures. We argue that
valued/interpetable phi-features on T license D[φ _] in Spec,v. This strategy can also be observed in
Spanish/Catalan control infinitives, the difference to finite structures lying in the (in)-completeness of
the feature make up of T: phi-de iciency tri ers ree with a matrix ar ument in control (c . Borer’s
1989 anaphoric AGR; Landau 2004). doptin ‘late insertion’ (Halle & Marantz 1993 and
subsequent work), a morphophonolo ical make up is provided or D postsyntactically. ‘Emptiness’
derives from a failure to equip D[φ _] with a value or φ be ore trans er to S-M: T[φ de ] cannot assi n a φ
value to D within the embedded, weak CP phase, resulting in null pronunciation of D. However, since
CP is not a (phi-complete) strong phase, valuation of D[φ _] by means of Agree can be delayed until the
completion of matrix vP, linking the interpretation of D to an antecedent. We argue that overt subjects
arise from a focus feature assigned to D[φ _], which instructs the computational system to delay
morphophonological insertion until the construction of the first strong phase, where φ-valuation takes
place (see (11) – (12)). Thus, morphophonological insertion always applies as soon as possible (even at
weak phase edges); only if there is an explicit trigger will insertion be delayed until the next (strong)
phase boundary.
Some evidence for this approach will be provided in that overt subjects in infinitives crucially
appear in discourse sensitive positions: Overt subjects inside infinitives cannot appear in preverbal
position (see (13)) in Spanish or Catalan because of left peripheral defectiveness, as witnessed by the
general unavailability of left peripheral fronting (cf. (14)). Thus, the possibility of overt subjects in
Spanish and Catalan infinitives is crucially tied to the availability of a low discourse sensitive position
(Belletti 2004), where [focus] or [emphasis] can trigger delay of insertion of morphophonological
material (adopting a model in the vein of López 2002). In finite pro-drop, morphophonological
insertion correlates with a contrastive interpretation of D, functioning as a trigger for the delay of
phonolo ical insertion until ‘phase slidin ’ (Galle o 2010), correlated with head movement of v, and
φ-valuation of D in Spec,T has taken place (derivation in (15) – (16)). Thus, the licensing of the
93
overt/covert alternation in Spanish and Catalan infinitives crucially depends on the availability of
discourse-sensitive information at the phase edges as triggers for delay of morphophonological
insertion: if insertion applies before phi-feature valuation, null pronunciation arises as a consequence of
the absence of a phi-specification; if insertion is delayed until phi-valuation, pronunciation arises,
triggered by discourse sensitive requirements.
(1) Juliai quería [telefonear ellai]. Julia wanted
to-phone she
(Spanish; Piera 1987)
(2) Odia [jugar (*Pablo) a las cartas] Hates to-play
Pablo at the cards
(Spanish; Torrego 1998)
(3) Ho ha fet en Joani (*elli / en persona). It
has made the Joan he / in person
(Catalan; Solà 1992)
(4) Antes de actuar Caballé, cantó Carreras.
Before of to-act Caballé sang Carreras
(Spanish; Rigau 1995)
(5) Salir yo de casa fue un error. (Spanish)
To-leave I from home was an error
(6) Obligué a Pacoi que (___i/*j / *María) saliera conmigo.
(I)-forced Paco that (hei/*j)/*Maria goes-out-SUBJ with-me (Suñer 1984: 266) (7) Ordené al
generali que (___i/j / las tropas) dejara(n) de disparar.
(I)-ordered the general that (hei/j) / the troops stop-SUBJ firing (Suñer 1984: 266)
(8) a. interpretable/complete phi-features → T in pro-drop; D of pronominals
b. interpretable/defective phi- eatures → T in Control; D o anaphors
(9) Juan dice [CP que [TP T[φ: compl]-durmieron [vP D[φ:_] v-durmieron [VP …
(10) Juan intentó [CP Cdef [TP T[φ: def]-dormir [vP D[φ:_] v-dormir [VP dormir…
(11) a. [CP Cdef [TP T[φ: def]-dormir [vP D[φ:_] v-dormir [VP …
b. Insertion at vP phase: D[φ _] = Ø
(12) a. [CP Cdef [TP T[φ: def]-dormir [#P D[φ:_]/[focus] [vP D[φ _] v-dormir [VP ...
b. Marta intentó [CP Cdef [TP T[φ: def]-dormir [#P D[φ:_]/[foc] [vP ...
c. Insertion at matrix TP phase (after Phase Sliding): D[φ 3.s . ] = /eʎa/
(13) Julia quería [(*ella) telefonear (ella)]. (Spanish; Piera 1987)
(14) *? Quiero, los libros, leerlos. (Spanish *CLLD; Gallego 2010)
Want-I the books to-read-them
(15) a. [CP C [TP T [vP D[φ:_] v-duermo [VP … b. Insertion at vP phase: D[φ _] = Ø
(16) a. [CP C [TP T[φ 1.s .]-duermo [#P D[φ:_]/[Contrast] [vP D[φ:_] v-duerme [VP …
b. Insertion at TP (after Phase Sliding): D[φ 1.s .] = /ɟo/
94
References
BARBOSA, P. (1995). Null Subjects. PhD dissertation, MIT. BELLETTI, A.
(2004). “ spects o the low area”. The structure of CP and IP. L. Rizzi (ed.), 16-51.
Ox ord University ress. BORER, H. (1989). “ naphoric GR”, O. Jea li & K. Sa ir (eds.).
GALLEGO, Á. (2010). Phase Theory. Amsterdam: John Benjamins. HALLE, M. & A. MARANTZ
(1993). “Distributed Morpholo y and the ieces o n lection”. The view from Building 20. Hale &
Keyser (eds.). Cambridge: MIT Press. 111-76. LANDAU, I. (2004).
“The Scale o Finiteness and the Calculus o Control”. NLLT 22: 811-77. LÓ EZ, L. (2002). “Toward
a Grammar without Top or Foc ” Georgetown Working Papers in Linguistics 2:
181-219. PIERA, C. (1987). “Sobre la estructura de las cláusulas de in initivo”. Demonte &
M. Lagunilla (eds.). Sintaxis de las lenguas románicas. Madrid: El arquero. 148-166. RIGAU,
G. (1995). “The roperties o the Temporal n initive Constructions in Catalan and Spanish”.
Probus 7: 279-301. S GURðSSON (2008). “The Case o RO”. NLLT 26: 403-450. SOLÀ, J. (1992).
Agreement and Subjects. PhD dissertation, UAB. SUÑER, M. (1984). “Controlled pro”. Papers from
the XIIth LSRL. Baldi, hilip (ed.). msterdam John Benjamins. SZ BOLCS , . (2009). ). “Overt
Nominative Subjects in n initival Complements CrossLin uistically”.
http://ling.auf.net/lingBuzz/00044. TORREGO, E. (1998). “Nominative Subjects and ro-Drop n l”.
Syntax 1/2: 206-219.
95
A variationist study of the discourse marker bueno in Border Spanish
José Esteban Hernández
Michelle Madrid
University of Texas Pan American
Despite the fact that we now have an extensive array of studies on discourse markers, few attempts
have been made to understand these linguistic elements from a sociolinguistic perspective. We lack, for
instance, studies that are solely focused on explaining the variable distribution of markers as their main
objective. No study that we know of has attempted to undertake an analysis of the distribution of a
marker in corpus data, perhaps because of the inherent limitations that such a task entails. In a way, the
pragmatic functions that markers exert on discourse have partly halted the application of a variationist
analysis to determine internal or external variables that rule their alternation in discourse. The problem
lies partly in that variable analysis is perceived in binary terms; thus, variation calls for two or more
linguistic variants that compete in speech for the same semantic space. Let us think about variables of a
phonological or morphological nature, with all the problematic that this last type may involve. The
problem is sometimes resumed in that markers are discretionary elements in discourse in Brinton’s
(1996) own words, "optional rather than obligatory features". However, options are the hallmark of
speech production and alternation is the raison d’être o variation theory. s a more pressin drawback,
for a marker to be treated as a sociolinguistic variable, variants must be clearly recognizable if a
context of application is to be established.
In the present study, we have undertaken a variationist analysis of the discourse marker bueno, as
in examples (1a) and (1b) below. The data come from two corpora of border Spanish: a) the Spanish in
Texas Project Corpus (Toribio & Bullock 2012) and b) the Reynosa Corpus. The systematic application
of a variationist approach put the accent on accountability of all occurrences of the variable under
consideration at hand in the corpus data, as well as those instances that could have materialized, but did
so as a null variant (cf. Poplack and Tagliamonte 2001: 89). We argue that discursive bueno has two
variants: an expressed and a null variant. Previous work by Serrano (1984) suggests that bueno is a
turn-initial marker, one that helps interlocutors formulate a response in conversation exchanges. The
proposal is that in discourse, every response or switch in turn-taking in conversation constitutes the
variable context of turn-initial bueno. Our study has two interrelated objectives: to add to our
understanding of markers and to advance variationist theory. Adhering to variationist principles, we
suggest that expression or absence of bueno is conditioned by linguistic (verb form, tense, mode),
pragmatic variables (previous utterance type, type of information), and social (sex, age, origin). The
statistical analysis of bueno was done using the statistical program GoldVarb X (Sankoff, Tagliamonte
and Smith 2005), which determined that four variables have a significant influence on variation: tense,
mode, previous utterance type, and type of information. These results indicate that it is feasible to apply
a variationist analysis to the study of makers, an area in which new investigations can certainly add
more to our understanding of discourse markers.
96
Examples:
(1a) A: -Bueno, um...ya cuando com...comenzó como el primer grado que también aprendió
ella inglés, hablaba con ellos...mis...mi mamá aprendía también por ella porque le decía
cosas en inglés y luego este...ella también aprendió español.
‘Well, um...since, when she sta--started like, the first grade she also learned English,
and spoke to them...my...my mom would also learn for her because she would say things
in En lish, and then um...she also learned Spanish.’
(1b)
Q: -¿Y tienes algún tipo de familia en este país, en esta ciudad quizás?
Q: -‘ nd do you have any amily in this country, in this city perhaps?’
A: - Ø Tengo familia en Houston, y tengo familia pues, en Brownsville y este...en
Puerto Rico es parte de Estados Unidos también, a lo mejor cuenta, y este...pero fuera
de eso ya es todo.
A: - Ø ‘ have amily in Houston, and have well, amily in Brownsville and
um...Puerto Rico is part of the United States, maybe it counts, and um...but other than
that, that’s it.’
References
Brinton, L. J.. 1996. Pragmatic markers in English. Grammaticalization and discourse functions.
Berlin: Mouton de Gruyter.
Sankoff, David, Sali Tagliamonte and Eric Smith. 2005: GoldVarb X: A Variable rule application for
Macintosh and Windows. Toronto, Canada, Department of Linguistics, University of Toronto.
Available in: http://individual.utoronto.ca/tagliamonte/Goldvarb/GV_index.htm
Serrano, María José. 1999. Bueno como marcador discursivo de inicio de turno y contraposición:
estudio sociolingüístico. International Journal of the Sociology of Language 140. 115-133.
Toribio, Almeida J., and Barbara E. Bullock. 2012. Spanish in Texas Project.
http//:coerll.utexas.edu/coerll/project/Spanish-texas-0. University of Texas at Austin.
97
The effect of code switching in the variation of copulative predicates ser and estar among Cuban
heritage speakers
Andrea Hernández
University of Florida
Copula use among Spanish-English bilinguals differs from monolingual use, a variation likely due
to numerous factors. Among these factors is cross-linguistic influence. The Spanish copulas determine
two states of being (1)–ser has been described as a stage-level predicate; estar as an individual level
predicate (Arche, 2006, Cortés-Torres 2004, Escandell-Vidal and Leonetti 2002; a.o., however, see
Camacho 2011)– while this distinction is not manifested in English (2). As a consequence, the
interpretive definitions of the Spanish copulas present a challenge among bilinguals (Bruhn de Garavito
and Valenzuela 2006). It is, however, not obvious which copula will be extended to more contexts.
Aguilar-Sánchez (2012) (see also Silva-Corvalán 1986) proposed one outcome is the overextension of
estar, which seems to be spreading to ser contexts in monolingual Mexican Spanish (Gutiérrez 1992),
an outcome attested with certain adjectival predicates among Cuban heritage speakers in Alfaraz
(2012). A different outcome, however, is conceivable: ser is more morphologically similar to English is
which might make a case for an extension of ser. Thus, this paper examines possible changes in copula
use in HS of Spanish of Cuban descent in order to examine copula use among bilinguals.
In very recent years, research on code switching (CS) has also taken an interest on the effects that
CS has on variable phenomena (Toribio 2004, Torres-Cacoullos and Travis 2011). To the best of our
knowledge, however, to date only subject expression (the distribution of overt and null subjects in
Spanish) has been examined. While Toribio (2004) and Torres-Cacoullos and Travis (2011) anticipate
an increase in overt pronominal subjects in Spanish-English CS, they attribute this increase to different
causes: Toribio ascribes it to convergence, and Torres-Cacoullos and Travis (2011) attribute it to
priming from English. This paper examines copula use in CS. Both theories, however, would predict
convergence with/ priming from English, where ser is more similar. Extending Silva-Corvalán’s (1986)
and Aguilar-Sánchez’s (2012) idea of acceleration of an ongoing change to CS, one could argue that in
CS the extension of estar is advancing faster to more ser contexts.
In an attempt to clarify which cause seems more probable, this paper examines second-generation
Cuban bilin uals’ copula choice across bilingual modes (Spanish-only and CS conditions). In
particular, the research questions addressed here are: if contact with English accelerates the extension
of estar in Spanish, is there also an extension of estar during CS? And if there is an extension of estar
during CS, is it overextended in comparison to their judgments of monolingual sentences? In order to
answer these questions, to date 12 participants completed a language background questionnaire, an
independent measure of proficiency (a section of the DELE), both used for participant selection and
profiling, and a contextualized 4-point Likert scale acceptability judgment task (AJT), based on
l araz’s (2012) with a Spanish-only and 2 CS conditions: one with Spanish adjectives and one with
English adjectives (n=32 each; 4 tokens x 4 adjective types—physical property, value, age, and
dimension x 2 contexts—favoring either ser or estar).
Preliminary results indicate that in monolingual Spanish conditions, participants either maintained
the copula distinction (physical property and dimension), lost copula distinction (age) or extended ser
to contexts where estar was favored (value). In CS English adjective conditions, participants either
maintained copula distinction (physical property and value), lost copula distinction (dimension) or
extended ser (age). Lastly, with CS Spanish adjectives, participants had a loss of distinction (age and
dimension) or extended ser (physical property and value). Therefore, participants exhibited a CS effect
with English and Spanish dimension adjectives by losing copula distinction, as compared to copula
distinction in the monolingual Spanish condition. With respect to value and age adjectives, there is a
CS effect only with CS English adjectives, where participants exhibit a copula distinction with value
98
adjectives and exhibit a ser extension with age adjectives. Regarding CS effects with physical property
adjectives, ser extension was present in CS with Spanish adjectives conditions. Crucially, an extension
of estar, to date, is not present in our data.
This paper contributes to the fields of bilingualism and contact linguistics in two ways: it examines
the debated topic of the extension of estar in bilingual speech and to the new CS focus on the effect on
CS on variable phenomena by examining the variable copula choice in Spanish.
Examples:
(1) a. María está alta.
b. María es alta.
(2) María is tall.
CS AJT task:
1. Favoring ser: Roberto y Pepe are going to play baloncesto. t’s the irst time that epe plays y nunca
ha visto una pelota de baloncesto and he asks Roberto to describe la pelota. Roberto le contesta
a. La pelota es round.
1 2 3 4 *No sé /No estoy segura
b. La pelota está round.
1 2 3 4 *No sé /No estoy segura
2. Favoring estar: Sandra is going to have un bebé. Durante el progreso de su embarazo, her sister
Maria Elena tells her
a. ¡Pero que round es tu barriga!
1 2 3 4 *No sé /No estoy segura
b. ¡Pero que round está tu barriga!
1 2 3 4 *No sé /No estoy segura
3. Favoring ser: Roberto y Pepe are going to play baloncesto. t’s the irst time that epe plays y nunca
ha visto una pelota de baloncesto and he asks Roberto to describe la pelota. Roberto le contesta
c. La pelota es round.
1 2 3 4 *No sé /No estoy segura
d. La pelota está round.
1 2 3 4 *No sé /No estoy segura
4. Favoring estar: Sandra is going to have un bebé. Durante el progreso de su embarazo, her sister
Maria Elena tells her
a. ¡Pero que redonda es tu stomach!
1 2 3 4 *No sé /No estoy segura
b. ¡Pero que redonda está tu stomach!
1 2 3 4 *No sé /No estoy segura
99
Variable agreement with Spanish binominal constructions
Chad Howe
Inma Garnes
University of Georgia
This paper analyzes a subcase of the phenomenon known as ad sensum agreement in Spanish
binominal constructions, typically schematized as NP1 de NP2 (see Traugott 2008a/b; Gili Gaya
1993). In these cases, a verb whose subject is a binominal construction agrees either with the higher
NP1, show in example (1a), or with the lower NP2, as in example (1b). It has been argued that this
pattern reflects a shift in the syntactic head in these constructions, particularly with high frequency
NP1 tokens undergoing semantic change (see, e.g., Brucart 1997, Berg 1998). Verveckken and
Cornillie (2012) discuss different factors that have been claimed to favor ad sensum agreement in
Spanish, arguing that a proper account of these patterns must consider both the 'degree' to which the
NP1 has grammaticalized and the effects of "lexical persistence" (following Hopper 1991). The current
analysis revisits this phenomenon focusing on a wider selection of binominal constructions and
demonstrates that agreement is highly sensitive not only to degree of semantic/structural change but
also to the different frequencies associated with each of the nominal elements in the construction.
Among the tokens analyzed by Verveckken and Cornillie (2012) are a number NP1 elements
ranging from highly grammaticalized examples like montón, exemplified in example (2), to cases with
similar quantifier function like pila or aluvión, shown in (3). Their analysis highlights several factors
that are important in determining the attested patterns of agreement, including clause type (main vs.
relative), presence of articles, verb type (e.g., distributive vs. non-distributive), position of the
(binominal) subject with respect to the verb, and distance between the subject and the verb.
Verveckken and Cornillie conclude that ad sensum agreement with these constructions is precipitated
by a confluence of factors and thus should be understood as "a reflection of the conceptual intentions
of the speaker" (2012:249, translation mine).
In this study, tokens of binominal constructions were extracted from both the written and oral
portions of the Corpus del Español (CdE) and the Corpus de Referencia del Español Actual (CREA).
In addition to the more grammaticalized quantifier expressions with montón, pila, aluvión, and alud,
and others (total of 9), there were three other groups of NP1 elements that were extracted: (i) NP1
tokens denoting quantity or collectivity, including cantidad (see example 4), grupo, parte, serie, and
mayoría, (total of 15, see also Van Doorslaer 2009); (ii) NP1 tokens that make indefinite reference,
illustrated in example (5) with tipo (total of 5); and (iii) a small control sample of relational nouns
(e.g., familia de, total of 5). Of the 34 distinct NP1 elements that were targeted, a total of 1,245 tokens
occurring as verbal subjects were coded for agreement type (i.e. NP1 or NP2) and for a list of different
independent factors predicted to influence agreement. Regarding the independent factors, this group
included those that were observed by Verveckken and Cornillie (e.g., verb type, presence of articles)
and, crucially, a list of factors related to the frequency of occurrence of both the NP1 and NP2
elements. Following Tagliamonte and Baayen (2012), this study employs a mixed-effects model and
random forests (using R) to explain the interactions between the target variable (agreement type) and
the independent factors.
The results confirm that the group of quantifier expressions including montón displays the highest
overall frequency of NP2 agreement (i.e. ad sensum agreement), followed by the indefinite nouns. This
result is consistent with the argument that these expressions are likely more grammaticalized than basic
quantity expressions (like cantidad) and certainly more than relational nouns. The mixed-effects
model, while confirming Verveckken and Cornillie's claim that ad sensum agreement is the result of
several different factors, reveals a strong effect for frequency, particularly in high frequency collocates
such as montón de personas and tipo de cosas. Moreover, though clause type is indeed shown to be a
significant factor, as discussed by Verveckken and Cornillie, it is in fact the nature of the NP1 (i.e.
100
quantifier, indefinite, etc.) that has the largest effect on ad sensum agreement. In sum, this study makes
two important observations: first, language change is a highly gradient phenomenon (see also Traugott
and Trousdale 2010, among others); and second, gradience in grammar is best examined by looking at
the interaction of multiple factors associated with the structural variability.
Examples
(1) a. [La mayoría] de los estudiantes asistió la clase.
b. La mayoría de [los estudiantes] asistieron la clase.
(adapted from Gili Gaya 1993)
(2) a. Mira, estaba la gallina, pone los huevos pon-pon-pon, sale [un montón] de huevecitos
(CREA, oral)
b. llueve a las once y media de la mañana y coge a la gente en la calle sin el paraguas y
entonces se compran un montón de [paraguas] (CREA, oral)
(3) a. Estoy seguro de que [el aluvión] de estudios que se realiza en todo el mundo nos traerá
aun mejores noticias en los próximos años (CREA, written)
b. al quedarse sin medios de subsistencia se van a engrosar el aluvión de [los desplazados]
de Colombia (CREA, written)
(4) a. hay [una cantidad] de cosas que nos ha impuesto la sociedad de consumo (CdE, oral)
b. Me faltan cantidad de [cosas] en la casa (CdE, oral)
(5) a. Pero ese [tipo] de cosas sí me interesaría (CdE, oral)
b. Creo que este tipo de [cosas] son muy interesantes (CdE, oral)
References
Berg, T. 1998. The resolution of number agreement conflicts in English and German agreement
patterns. Linguistics 36:41–70.
Brucart, J. M. 1997. Concordancia ad sensum y partitividad en español. Contribuciones al estudio de la
lingüística hispánica. Homenaje al profesor Ramón Trujillo, Tomo I, ed. by M. Almeida and J. Dorta,
157-183. Tenerife: Montesinos.
Corbett, G. G. 2006. Agreement. Cambridge University Press, Cambridge.
Gili Gaya, S. 1993. Curso superior de la sintaxis española. Barcelona: Biblograf.
Hopper, P. 1991. On some principles of grammaticization, Approaches to grammaticalization, ed. by
E. C. Traugott and B. Heine, 17-36. Amsterdam/Philadelphia: John Benjamins
Tagliamonte, S. and R. H. Baayen. 2012. Models, forests, and trees of York English: Was/were
variation as a case study for statistical practice. Language Variation and Change 24: 135- 178.
Traugott, E. C. 2008a. Grammaticalization, constructions and the incremental development of
language: Suggestions from the development of Degree Modifiers in English. Variation, Selection,
Development: Probing the Evolutionary Model of Language Change, ed. by R. Eckardt, G. Jäger,
and T. Veenstra, 219-250. Berlin/New York: Mouton de Gruyter.
101
Traugott, E. C. 2008b. The grammaticalization of NP of NP patterns. Constructions and Language
Change, ed. by Alexander Bergs and Gabriele Diewald, 23-46. Berlin/New York: Mouton de Gruyter.
Traugott, E. C. and G. Trousdale (eds). 2010. Gradience, gradualness and grammaticalization.
Amsterdam/Philadelphia: John Benjamins.
Van Doorslaer, J. 2009. Concordancia de número. Un análisis cognitivo-funcional de la
concordancia entre verbo y regente [N1 + de + N2]. Masters Thesis, University of Leuven,
Leuven. Verveckken, K. and B. Cornillie. 2012. Un análisis cognitivo-funcional de la concordancia
verbal con construcciones (pseudo)partitivas del tipo [N1 + de + N2]. Revue Romane 47: 219–257.
102
The within-subjects measurement of L2 Spanish sentence processing with two different subject-
verb agreement paradigms Jill Jegerski
University of Illinois at Urbana-Champaign
The Shallow Structure Hypothesis (Clahsen & Felser, 2006) claims that intrinsic processing
deficits can be the key to understanding why there is a general lack of success in adult second language
acquisition. In this model, abstract and hierarchical elements of syntax in particular are purportedly
absent from L2 sentence processing, as is non-local agreement between items separated by linear
distance. Previous psycholinguistic study of subject-verb number agreement was initially consistent
with this prediction (Jiang, 2004; Ojima et al., 2005), though more recent work has suggested that non-
native processing can become nativelike in this regard (Foote, 2011; Rossi et al., 2006). However, it is
not clear to what extent the discrepancy in previous research may be due to processing behavior
developing over time, as higher L2 proficiency is acquired, because prior investigations each included
only one L2 participant group at a single level of proficiency. It is also possible that the type of
sentence comprehension paradigm used to measure online processing may have affected experimental
outcomes, as nearly all investigations used stimuli with grammar violations (1), but one employed the
more subtle broken agreement paradigm (Jiang, 2004), in which reading time effects are induced
without ungrammaticality (2). The present study addressed these two issues by including participants
at different L2 proficiency levels and by including two experiments, one with grammar violation
stimuli and one with the more subtle broken agreement stimuli.
A total of 53 native English speakers residing in Spain were divided into three L2 Spanish
proficiency groups, based on their scores on a DELE test. These three non-native groups, plus a group
of 39 native Spanish speakers, then completed a noncumulative linear selfpaced reading task that
included 20 stimuli for each of the two experiments, plus 120 fillers, as illustrated below. Each
stimulus was followed by a meaningful comprehension question.
(1) Grammaticality Stimulus (Grammatical, Ungrammatical)
a. Los olores / del almendro / llegan / al patio / detrás de la casa.
b. El olor / del almendro / *llegan / al patio / detrás de la casa. “The odor(s)
o the almond tree reach the patio behind the house.”
c.
(2) Broken Agreement Stimulus (Singular-Singular-Singular, Singular-Plural-Singular)
a. El testigo / del abogado / tiene / mala fama / en el barrio.
b. El testigo / de los abogados / tiene / mala fama / en el barrio. “The
witness o the lawyer(s) has a bad reputation in the nei hborhood.”
Reading time data from the grammaticality stimuli showed immediate online sensitivity to non-
local number agreement at stimulus Region 3 among all participant groups, but only the three L2
groups showed sustained sensitivity that extended to Regions 4 and 5. Data from the broken agreement
stimuli, on the other hand, revealed no reliable reading time effects among any participant group, not
even the L1 participants, which stands in contrast with previous research using very similar materials
(Bock et al., 2012). In addition to challenging the Shallow Structure Hypothesis with evidence of
online sensitivity to non-local agreement in L2 processing, these results also suggest that the definition
of target L2 processing behavior is not always straightforward and that experimental results can vary
103
according to the method used. The importance of including a local comparison group of L1
participants is also highlighted.
References
Bock, J. K., Carreiras, M., & Meseguer, E. (2012). Number meaning and number grammar in English
and Spanish. Journal of Memory and Language, 66, 17-37.
Clahsen, H., & Felser, C. (2006). Grammatical processing in language learners. Applied
Psycholinguistics, 27, 3-42.
Foote, R. (2011). Integrated knowledge of agreement in early and late English–Spanish bilinguals.
Applied Psycholinguistics, 32, 187–220.
Jiang, N. (2004). Morphological insensitivity in second language processing. Applied
Psycholinguistics, 25, 603-634.
Ojima, S., Nakata, H., & Kakigi, R. (2005). An ERP study of second language learning after
childhood: Effects of proficiency. Journal of Cognitive Neuroscience, 17(8), 1212-1228.
Rossi, S., Gugler, M. F., Friederici, A. D., & Hahne, A. (2006). The impact of proficiency on syntactic
second-language processing of German and Italian: Evidence from eventrelated potentials. Journal of
Cognitive Neuroscience, 18(12), 2030-2048.
104
Explaining the variation between voseo and tuteo negative commands in Argentinian
Spanish
Mary Johnson
The Ohio State University
Throughout the Spanish-speaking world, there is variation between the use of the 2nd person
singular pronouns tú and vos and their corresponding verb forms. In Argentinian Spanish (AS),
speakers alternate between the tuteo negative imperative (TNI) and the voseo negative imperative
(VNI), as shown in examples (1) and (2):
(1) ¡No cantes! 'Don't sing!' (TNI)
(2) ¡No cantés! 'Don't sing' (VNI)
Previous research has shown that the VNI and the TNI is AS differ in meaning.
Fontanella de Weinberg (1979) notes that the VNI is more forceful, while the TNI is more polite.
Johnson & Grinstead (2010) found the VNI to be more pragmatically restricted than the TNI, and
propose that the VNI expresses urgency. The current study consists of an analysis of an online survey
of native speakers of AS in order to form a more accurate description of the meaning difference
distinguishing these negative imperatives, and how this meaning difference comes about both
pragmatically and socially.
The survey explores the effect of immediacy on the choice of imperative used. A context is
immediate if the interlocutor is already performing the action that the speaker is requesting they not do.
A context is neutral if the action is not yet in progress. Immediate contexts were further divided into
contexts in which the speaker was angry and those in which the speaker was not. Participants were
provided with contexts eliciting a negative imperative response, and were asked to choose between the
VNI and TNI for each context.
Statistical tests including a generalized mixed-effects logistic regression and conditional inference
trees were run to determine the relevant factors (social and linguistic) predicting the choice of
VNI/TNI. Results indicate that the VNI was used significantly more in immediate contexts than in
neutral ones (p<.001). They also indicate that immediate contexts that were also angry yielded
significantly more use of VNI than those that were immediate but not angry (p<.001). Dialectal and
social differences emerged, such that inside Buenos Aires, males chose the VNI significantly more than
females (p=.006) in all contexts but angry ones. Outside of Buenos Aires, the only contexts in which
the VNI is significantly more likely to be used are the angry ones (p=.019), and there is no effect of
gender. These results point to a pragmatic differentiation of the TNI and VNI, and a difference in
licensin across dialects o S. The VN conveys in ormation re ardin the speaker’s perceived
immediacy of the situation. Anger, presumably, is enticed by the immediacy of the situation.
Differences among males and females are representative of the social manifestation of the pragmatic
meaning difference. Commentaries provided in the survey responses corroborate this association of the
VNI to masculinity, or to other masculine traits such as aggression.
The results from this survey provide evidence that the VNI encodes information about the speaker.
This information may be social (as in the case o ender di erences) or pra matic (the speaker’s
perception of the immediacy of the context).
105
References
FONTANELLA DE WEINBERG, M. B. 1979. La oposición <<Cantes/Cantés>> en el Español de Buenos
Aires. Thesaurus 34, 72-83.
JOHNSON, MARY & JOHN GRINSTEAD. 2010. “The pra matic variation between voseo and tuteo
ne ative imperatives in r entine Spanish.” aper presented at the 2010 Hispanic Lin uistics
Symposium, Bloomington.
106
Knowledge and Processing of Subjects in Farsi-Spanish Adult Bilinguals Tiffany Judy
Wake Forest University
Previous research on the acquisition of subjects has largely shown that adult second language
learners are able to converge on the syntactic distribution of subjects; yet research on their discourse-
constrained distribution has provided mixed results (e.g. Belletti, Bennati & Sorace 2007; Gürel 2006;
Rothman 2009; Sorace & Filiaci 2006). One current theory, the Interface Hypothesis (IH; Sorace 2011,
2012), claims that this differential convergence is a product of the increased difficulty associated with
processing properties at the syntax-discourse interface, which, as opposed to purely syntactic
properties, requires integration of both syntactic and discourse information. These claims stem from
evidence suggesting that even the nonrelevant language is activated in bilinguals (Bialystok 2009).
Since bilinguals must simultaneously activate the relevant language while inhibiting the other, working
memory and executive function are stressed, ultimately causing differences in language processing.
Unlike previous accounts, the IH takes the presence of two grammars to be the source of divergence.
As a consequence of this assumption, differences between monolinguals and bilinguals are predicted
regardless of whether transfer from the native language would be facilitative.
With this theoretical back round in mind, this study tests the H’s claims that (a) di erences
between monolinguals and bilinguals at the syntax-discourse interface result from processing
differences and that (b) language pairing is not deterministic. While subject pronouns in Romance
languages have been extensively studied (cf. Belletti et al. 2007; Prada Pérez 2010; Rothman 2009;
Sorace, Serratrice, Filiaci & Baldo 2009), this study fills a gap in the literature in that it presents both
offline and online data and it examines a novel language pairing: FarsiSpanish. Apart from being an
understudied language pairing, examining Farsi-Spanish bilinguals allows us to disentangle the effect
of processing from native language transfer. This is so because while Farsi and Spanish share
analogous discourse distributions of referential subject pronouns, the underlying syntax of subjects is
different in each language.
Eight near-native, naturalistic Farsi-Spanish bilinguals who have been living in a Spanish-
speaking environment for over 30 years completed two offline and two online tasks examining
convergence on the syntax- and discourse-constrained distribution of subjects in Spanish. The results
showed that no statistically significant differences were found between the native Spanish speakers and
the Farsi-Spanish bilinguals with respect to their performance on the offline syntactic task (a
Grammaticality Judgment/Correction Task) nor their Reaction Time (RT) to any of the regions of
interest in the online version (a Self-Paced Reading task).
Regarding the discourse-distribution, results from the offline Context-Matching Felicitousness Task
show differences in three of six categories (higher ratings of null subjects in Topic Shift contexts and
higher ratings of both overt and null subjects in Topic Maintenance contexts). No differences were
found with overt or null subjects in Contrastive Focus contexts nor with overt subjects in Topic Shift
contexts. Regarding the online Self-Paced Reading task, only one difference was found in the
counterbalanced Contrastive Focus tokens (in the spillover region). Crucially, no differences in RT
were found between the counterbalanced Contrastive Focus tokens for the verb region, nor for the
Topic Shift and Topic Maintenance tokens in either the verb or spillover region. Taken together, data
from both offline and online tasks are interpreted as indicating that the bilinguals have converged on
the syntactic and discourse distribution of subject pronouns in Spanish, providing counter-evidence to
the IH.
107
Expletive Subjects (obligatory null subject)
(1) a. Los vecinos dicen que pro siempre llueve mucho en esa región.
the neighbors say that pro always rains a lot in that region b. *Los
vecinos dicen que ello llueve mucho en esa región. the neighbors
say that it rains a lot in that region
“The nei hbors say it rains a lot in that re ion.”
Contrastive Focus (overt subject felicitous)
(2) Cuando salimos a cenar, mi novia prefiere comer platos livianos, pero yo
prefiero comer algo sustancioso.
“When we o out to eat, my irl riend pre ers to eat li ht dishes, but pre er to eat somethin
substantial.”
a. √ sí que ella come ensaladas y yo como milanesas en los restaurantes.
b. #Así que pro come ensaladas y pro como milanesas en los restaurantes.
“So, she eats salads and eat breaded meats at the restaurant.”
References
Belletti, A., Bennati, E., & Sorace, A. (2007). Theoretical and developmental issues in the syntax of subjects: evidence from near-native Italian. Natural Language and Linguistic Theory, 25, 657-
689.
Bialystok, E. (2009). Bilingualism: The good, the bad, and the indifferent. Bilingualism: Language and
Cognition, 12(1), 3-11.
Gürel, 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 linguistic
development: in honor of Lydia White (pp. 259-282). Amsterdam: John Benjamins.
Prada Pérez, A. de (2010). Subject Position in Spanish in Contact with Catalan: Language Similarity
vs. Interface Vulnerability. In M. Iverson, I. Ivanov, T. Judy, J. Rothman, & R. Slabakova
(Eds.), Proceedings of the 2009 Mind/Context Divide Workshop (pp. 104115). Somerville, MA:
Cascadilla Proceedings Project.
Rothman, J. (2009). Pragmatic Deficits with Syntactic Consequences?: L2 pronominal subjects and the
Syntax-Pragmatics Interface. Journal of Pragmatics, 41(5), 951-973.
Sorace, . (2011). innin down the concept o “inter ace” in bilin ualism. Linguistic Approaches to
Bilingualism, 1(1), 1–33.
Sorace, A. (2012). Pinning down the concept of "interface" in bilingualism: a reply to peer
commentaries. Linguistic Approaches to Bilingualism, 2, 209-216.
Sorace, A., & Filiaci, F. (2006). Anaphora resolution in near-native speakers of Italian. Second
Language Research, 22, 339-368.
Sorace, A., Serratrice, L. Filiaci, F., & Baldo, M. (2009). Discourse conditions on subject pronoun
realization: Testing the linguistic intuitions of bilingual children. Lingua, 119,
460–477.
108
El futuro es perifrástico: Un análisis sociolingüístico de la expresión de futuridad en dos
comunidades mexicanas
Kendall Kyzar
Rafael Orozco
Louisiana State University
Este estudio comparativo de la expresión de futuridad en el español de mexicanos en los EE.UU. y
residentes de la, sociolingüísticamente inexplorada, ciudad de Xalapa, México tiene propósitos
múltiples. Exploramos la distribución de las variantes del futuro (futuro morfológico, futuro
perifrástico, presente simple) y los constreñimientos que condicionan la ocurrencia de cada una de
ellas. Usamos datos extraídos de dos corpus socialmente estratificados basados en entrevistas
sociolingüísticas con respectivos grupos de informantes de cada comunidad bajo estudio. Realizamos
una serie de análisis de regresión estadística para explorar los efectos de diez constreñimientos sociales
y nueve lingüísticos que funcionan a tres niveles morfosintácticos: la cláusula entera, el sujeto y el
predicado.
La distribución de variantes indica que el futuro perifrástico registra la tasa más alta de ocurrencia con
un 69.3% mientras que el uso del futuro morfológico es desfavorecido a expensas de las otras variantes.
En México, el futuro perifrástico y el presente indicativo se han convertido en las principales
alternativas empleadas para expresar la futuridad en detrimento del futuro morfológico. Los resultados
muestran que la expresión de la futuridad está condicionada por un complejo amalgamiento de
constreñimientos lingüísticos y sociales. Los condicionamientos lingüísticos revelan los efectos de
factores como tipo de verbo, distancia temporal, persona gramatical y especificación adverbial. Esto
concuerda con lo que ocurre en otras partes del mundo hispánico (cf. Claes & Ortíz López 2011,
Orozco 2007a) sugiriendo que aparte de diferencias en la frecuencia de uso de las diferentes variantes,
en diversas partes del mundo hispánico, el futuro está bajo condicionamientos similares. Los efectos de
factores sociales se ajustan a los principios de cambio lingüístico (Chambers 2002:349, Guy 1990:52,
Labov 2001:437), ya que las innovaciones son promovidas por los hablantes más jóvenes. En cuanto al
Sexo del Hablante, los resultados revelan que en México, los hombres se comportan como innovadores
lingüísticos, mientras que en Luisiana son las mujeres que demuestran esta conducta (cf. Claes & Ortíz
López 2011, Orozco 2007b). Los resultados también le otorgan validez a la hipótesis de Silva-Corvalán
(1994a: 208) ya que el futuro morfológico muy poco se utiliza en la variedad bilingüe, lo que sugiere
que esta innovación lingüística latente ya estaba en marcha antes del inicio del contacto intensivo con el
inglés. Al tomar en consideración otras comunidades de habla, estos resultados también indican que los
mexicanos están en un nivel más avanzado en el cambio en progreso que favorece la variante analítica.
El uso preferencial del futuro perifrástico a expensas del morfológico refleja el efecto gradual de la
gramaticalización (cf. Bybee 2003; Bybee, Perkins & Pagliuca 1994), la cual va de la mano con el
proceso supralingüístico de ciclicidad. Como parte de este proceso, la lengua cambia de ser
principalmente sintética a predominantemente analítica, llegando a ser sintética otra vez (cf.
Fleischman 2009: 152, Silva-Corvalán 1994: 52), Gutiérrez 1995: 214). Estos resultados proporcionan
información importante que ayuda a comparar las fuerzas sociolingüísticas que condicionan la
variación y cambio en las comunidades bajo estudio con otras comunidades hispánicas.
109
On the representation of neuter in the mental lexicon: A study of Spanish-German
bilinguals
Rachel Klassen
University of Ottawa
The representation of L1 and L2 grammatical gender systems in the mental lexicon has been the focus
of a significant body of research and remains a matter of debate. Prominent models of speech
production seem to converge on the notion that grammatical gender is represented by syntactic nodes
linked to the lexical entry of each word in the L1 (Levelt et al, 1999; Caramazza, 1997). The
representation of the L2 gender system and its relationship to that of the L1, however, is not yet clear.
According to the gender-integrated representation hypothesis (Salamoura & Williams, 2007), the L1
and L2 have a shared grammatical gender system represented as gender nodes common to both
languages. In contrast, the gender autonomous representation hypothesis (Costa et al, 2003) maintains
that the L1 and L2 gender systems are independent and therefore gender nodes are language-specific.
Since Schriefers (1993) found an effect of gender congruency between target and distractor nouns in a
pictureword interference task, this gender congruency effect has been replicated in monolingual studies
and has been extended to bilinguals studies through the L2 picture naming paradigm. In contrast, other
studies have found no effect of gender congruency or an effect that varies according to whether gender
is borne by an NP or a bare noun. In addition to these conflicting results, previous studies have focused
almost exclusively on similar grammatical gender systems and thus there is currently no information
available regarding the representation of gender classes not present in the L1 (neuter for L1 Spanish
speakers). This study brings new evidence to inform the L2 gender representation debate and also
addresses the limitations of previous work by investigating the representation of gender systems
mismatched in number of gender classes (Spanish: 2-class; German: 3-class).
19 L1 Spanish-L2 German speakers as well as 25 L1 German controls participated in this experiment.
Participants performed an online L2 picture naming task, producing the corresponding bare noun or NP
in German (Fi ure 1). articipants’ verbal responses were recorded and a custom device was desi ned
to mark stimuli onsets with a photodiode to avoid voice key accuracy issues. Stimuli were line
drawings of inanimate, concrete nouns of the same gender in Spanish and German (gender congruent),
different genders (gender incongruent) and neuter nouns in German (Figure 2).
Our analysis of the results shows an effect of gender congruency for L1 Spanish-L2 German speakers
in both bare noun and NP naming conditions (Figure 3). Naming latencies were approximately 30 msec
shorter for gender congruent nouns than incongruent ones. Interestingly, neuter nouns were named
faster than gender congruent nouns, especially in the NP condition (bare nouns: 40 msec faster; NP:
100 msec faster).
These results support the gender-integrated representation hypothesis o the bilin ual’s rammatical
gender systems, even for those that are non-similar. A gender effect in both naming conditions suggests
that gender is always available upon lexical access, contrary to prominent models of speech production.
Shorter naming latencies for neuter nouns indicate that the gender not present in the L1 has a unique
representation that interferes with the response significantly less than when the incongruency is
between genders present in both languages (masculine-feminine mismatches).
110
References
Caramazza, A. (1997). How many levels of processing are there in lexical access? Cognitive
Neuropsychology, 14, 177–208.
Costa, A., Kovacic, D., Franck, J., & Caramazza, A. (2003). On the autonomy of the grammatical
gender systems of the two languages of a bilingual. Bilingualism: Language and Cognition, 6(3), 181-
200.
Levelt, W. J., Roelofs, A., & Meyer, A. S. (1999). A theory of lexical access in speech production.
Behavioral and Brain Sciences, 22, 1–38; discussion 38–75.
Salamoura, A., & Williams, J. N. (2007). The representation of grammatical gender in the bilingual
lexicon: Evidence from Greek and German. Bilingualism: Language and Cognition, 10(3), 257–275.
Schriefers, H. (1993). Syntactic processes in the production of noun phrases. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 19(4), 841–850.
111
Widening the envelope of variation: A corpus-based analysis of no saber si
Stephanie Knouse
Furman University
In variationist methodology, it is widely sustained that researchers must carefully circumscribe the
variable context, incorporating all possible variants and non-occurrences in the analysis, in order to
accurately account for the factors conditioning speaker selection (Bayley 2002:124; Labov 1982:30;
Tagliamonte 2006:72, 86). Previous quantitative investigations of verb choice in Spanish after no saber
si ‘do not know i ’ have limited the scope o the analysis to the present indicative (1) or present
subjunctive (2) as possible variants (cf. Hall 1999; Moreno de Alba 1978; Serrano Montesino 1992),
and have disregarded the synthetic future (3) as a viable choice in this variable context.
(1) […] no sé si soy (IND) bueno; pero sé decir que no soy el malo. (CDE) ‘[…] don’t know
i ’m ood; but know to say that ’m not bad.
(2) Eh... yo no sé si sea (SUB) cierto; no, yo personalmente me baño todos los días. (CDE) ‘Uh…
don’t know i it is true; no, personally bathe everyday.’
(3) […] no sé si serán (FUT) de su agrado. (CDE)
‘ don’t know i they are o his likin .’
I contend that one must reconsider the envelope of variation when analyzing no saber si and include the
future tense as a variant for the following reasons. First, functional analyses of no saber si attest that the
synthetic future commonly follows this proposition of uncertainty (Bybee, Perkins, & Pagliuca 1994;
Klein-Andreu 1995) and it is the preferred form in certain geographical regions (Lope Blanch 1953;
Urdiales 1966), even though prescriptive accounts have determined that only the present indicative is
the ‘correct’ choice ollowin no saber si and the present subjunctive is an aberration in this context
(Bosque 1990; Foster 1982; inter alia). Second, upon analyzing the grammaticalization paths of mood
and modality, Bybee et al. (1994 235) a irm that a “ uture ram could take on subjunctive uses,
especially as a new uture develops,” which is the case with the competin utures in Spanish (c .
Aaron 2006). Furthermore, it has been documented that the synthetic future after no saber si takes on
epistemic meaning in the present, as opposed to futurity (Bybee et al. 1994:225).
Consequently, in this diachronic, corpus-based study the variable context of no saber si was broadened
to include the present indicative, present subjunctive, and the synthetic future, as in examples (1), (2),
and (3). 2031 tokens were extracted from the Corpus del Español (CDE) (Davies 2002) from the 13th
to
20th
centuries and coded for the following independent variables: form of saber, grammatical person in
the subordinate clause, co-reference, type of negation, verb class, and century. Token and type
frequency were also included in the analysis. The data were analyzed using Goldvarb X (Sankoff,
Tagilamonte & Smith 2005). Preliminary statistical results indicate that beginning in the 17th
century,
the synthetic future began to systematically replace the subjunctive form after no saber si. This finding
not only ali ns with Bybee et al.’s contention that the synthetic uture orm can encroach on
subjunctive terrain as it competes with the periphrastic future, but it also corroborates other researchers’
assertions that the subjunctive is “slowly disappearin rom spoken Spanish” (Hall 1999 54). n
addition, the subjunctive form was strongly favored after no sé ‘ don’t know,’ when co-referenced, and
with the verbs decir ‘say’ and llamar ‘call,’ indicating that subjunctive use following no saber si is
highly constrained, as in other Romance languages (Poplack 2001:414).
112
References
Aaron, Jessica Elena. 2006. Variation and Change in Spanish Future Temporal Expression.
Unpublished Doctoral Dissertation, University of New Mexico.
Bayley, Robert. 2002. The quantitative paradigm. In The Handbook of Language Variation and
Change, eds. P. Trudgill and N. Schilling-Estes J.K. Chambers, 117-41. Oxford: Blackwell.
Bosque, Ignacio. 1990. Las bases gramaticales de la alternancia modal. In Indicativo y subjuntivo, ed.
I. Bosque, 165. Madrid: Taurus Universitaria.
Bybee, Joan, Revere Perkins and William Pagliuca. 1994. The Evolution of Grammar: Tense, Aspect,
and Modality in the Languages of the World. Chicago: University of Chicago Press.
Davies, Mark. 2002. Corpus del Español: 100 million words, 1200s-1900s. Available online at
http://www.corpusdelespanol.org.
Foster, William. 1982. Internal contradictions of a Spanish subjunctive. International Review of
Applied Linguistics 20(1), 131-137.
Hall, Ronda. 1999. Mood choice after no saber si in Mexican Spanish. Unpublished Doctoral
Dissertation, Oklahoma State University.
Klein-Andreu, Flora. 1995. The painless subjunctive. In Studies in Second Language Learning and
Spanish Linguistics in Honor of Tracy D. Terrell, eds. P. Hashemipour, R. Maldonado, and M. van
Naerssen, 419-431. New York: McGraw Hill.
Labov, William. 1982. Building on empirical foundations. In Perspectives on Historical Linguistics,
eds. W. Lehmann and Y. Malkiel, 17-92. Amsterdam: John Benjamins.
Lope Blanch, Juan M. 1953. Observaciones sobre la sintaxis del español hablado en México.
México: Publicaciones del Instituto Hispano Mexicano de Investigaciones Científicas.
Moreno de Alba, José G. 1978. Valores de las formas verbales en el español de Méjico. México:
UNAM.
Poplack, Shana. 2001. Variability, frequency, and productivity in the irrealis domain of French. In
Frequency and the emergence of linguistic structure, eds. J. Bybee and P. J. Hopper, 405430.
Amsterdam: John Benjamins.
Sankoff, David, Sali Tagliamonte and Eric Smith. 2005. Goldvarb X: A variable rule application for
Macintosh and Windows. Department of Linguistics, University of Toronto.
Serrano Montesino, María José. 1992. El subjuntivo en Canarias y América. Revista de Filología y su
Didáctica (Sevilla), 241-251. Available online at
http://cvc.cervantes.es/literatura/cauce/pdf/cauce14-15/cauce14-15_16.pdf
Tagilamonte, Sali A. 2006. Analysing Sociolinguistic Variation. Cambridge: Cambridge University
Press.
Urdiales, J. M. 1966. El hablo de Villacidayo (León). Anejos del Boletín de la Real Academia
Española 13. Madrid.
113
A sociophonetic analysis of trill production in Panamanian Spanish
Delano Sydney Lamy
SUNY Geneseo
Several accounts of standard production define the trill as a segment consisting of two or three
brief occlusions between the apex of the tongue and the alveolar ridge (Martínez Celdrán, 1997;
Hualde, 2005). However, via a combination of acoustic characteristics, various researchers have
observed other forms, such as a pre-aspirated trill, a pre-breathy trill, an approximantized trill, an
assibilated trill, and a uvular trill (Lewis, 2004; Colantoni, 2006; Willis, 2006; Díaz-Campos, 2008;
Henriksen & Willis, 2010; Bradley & Willis, 2012). The present study seeks to add to this body of
research by analyzing trill production using variationist methods in Panamanian Spanish spoken in the
urban areas of Ciudad de Panamá, Panamá.
Few studies explore rhotic production in the Panamanian variety, and of the descriptions
available, most focus on the alveolar tap in coda position, (cf. Cedergren et al., 1986; Broce & Torres
Cacoullos, 2002). Therefore, this study contributes to the previous literature by focusing particularly on
the production of the trill. The analysis includes sociolinguistic interviews with 11 native speakers of
Panama City Spanish (PCS). 608 tokens of the trill were extracted and submitted to acoustic analyses
using PRAAT (Boresma & Weenink, 2012). Cues such as presence of occlusions, waveform amplitude
reduction, transition in formant structure, and intensity (cf. Bradley & Willis, 2012) were considered in
the acoustic characterization of trills. Subsequently, multivariate analyses were carried out in Rbrul
(Johnson, 2009) with three dependent variables: (i) normative/non-normative alternation, (ii) mean
occlusions, and (iii) mean duration. This study is the first to analyze the latter two variables in multiple
regression analyses.
The acoustic analysis reveals 8 variants in PCS, which include the normative trill (with 2 or 3
occlusions and generally followed by r-coloring or approximantization), an assibilated trill, an
approximantized trill, a voiced tap, a pre-assibilated tap, a post-assibilated tap, a post-approximantized
tap, and a perceptual tap. The most frequent variant is the normative trill, especially with two
occlusions (cf. Bradley & Willis, 2012). The multivariate analyses further reveal that variable trill
production is conditioned by certain linguistic and extralinguistic factors. Syllable stress, grammatical
category, position in the word, age and sex are selected as significant predictors for trill production in
terms of duration. The trill segment is longer in posttonic and tonic syllables, in adjectives, and in
word-initial position (cf. Henriksen & Willis, 2010). Furthermore, males and speakers between the ages
of 27 and 35 favor longer segments. With regard to mean occlusions, position in the word, syllable
stress, preceding segment and age are selected as significant factor groups, where a greater mean is
favored in tonic and posttonic syllables, in word-initial position, when preceded by a consonant or
vowel, and among speakers between 27 and 35 (cf. Henriksen & Willis, 2010). Lastly, in terms of
normative/non-normative alternation, preceding segment, syllable stress, number of syllables, position
in the word, age and sex are the most important predictors of trill production. Normative variants are
favored with preceding consonants and vowels, in posttonic and tonic syllables, in words containing 1-
3 syllables, in word-initial position, among speakers between 27 and 35, and among males (cf. Díaz-
Campos, 2008). The results for trill production in PCS reveal an inherent systematic variability as is
observed in other Latin American and Peninsular varieties. This variability is conditioned by prosodic
factors, such as word position and stress, and extralinguistic factors, such as age and sex, which has
important implications for literature concerned with phonetic, phonological, and variationist
sociolinguistic theory in the Spanish-speaking world.
114
References
Boersma, P., & Weenink, D. 2012. Praat: doing phonetics by computer (Version 5.3.04) Retrieved
from http://www.praat.org/.
Bradley, T.G. & E. Willis. 2012. Rhotic variation and contrast in Veracruz Mexican Spanish. Estudios
de Fonética Experimental, 21, 43-74.
Broce, M. & Torres Cacoullos, R. 2002. ‘Dialectolo ía urbana’ rural la estrati icación social de (r) y
(l) en Coclé, Panamá. Hispania, 85/2, 342–354.
Cedergren, H.J., Rousseau, P. & Sankoff, D. 1986. La variabilidad de /r/ implosiva en el español de
Panamá y los modelos de ordenación de reglas. In Núñez Cedeño, R., I. Páez Urdaneta & L.
Guitart (eds.), Estudios sobre la fonología del español del Caribe, 13–20. Caracas: Ediciones
La Casa de Bello.
Colantoni, L. 2006. Micro and macro sound variation and change in Argentine Spanish. In N. Sagarra
& A.J. Toribio (eds.), Selected Proceedings of the 9th
Hispanic Linguistic Symposium, 91-102.
Somerville, MA: Cascadilla Proceedings Project.
Díaz-Campos, M. 2008. Variable production of the trill in spontaneous speech: Sociolinguistic
implications. In L. Colantoni & J. Steele (eds.), Selected Proceedings of the 3rd Conference on
Laboratory Approaches to Spanish Phonology (pp. 47-58). Somerville, MA: Cascadilla
Proceedings Project.
Henriksen, N. & E. Willis. 2010. Acoustic characterization of phonemic trill production in Jerezano
Andalusian Spanish. In M. Ortega-Llebaria (ed.), Selected Proceedings of the 4th Conference
on Laboratory Approaches to Spanish Phonology, 115-127. Somerville, MA, Cascadilla
Proceedings Project.
Hualde, J.I. 2005. The sounds of Spanish. Cambridge: Cambridge University Press.
Johnson, D. E. 2009. Getting off the GoldVarb standard: Introducing Rbrul for mixed-effects variable
rule analysis. Language and Linguistics Compass 3, 359-383.
Lewis, A. 2004. Coarticulatory effects on Spanish trill production. In A. Agwuele, W. Warren, & S-H.
Park (eds.), Proceedings of the 2003 Texas Linguistics Society Conference, 116-127.
Somerville, MA: Cascadilla Proceedings Project.
Martínez Celdrán, E. 1997. El mecanismo de producción de la vibrante apical múltiple. Estudios de
Fonética Experimental, 8, 85-97.
Willis, E.W. 2006. Trill variation in Dominican Spanish: An acoustic examination and comparative
analysis. In N. Sagarra & A.J. Toribio (eds.), Selected Proceedings of the 9th Hispanic
Linguistics Symposium, 121-131. Somerville, MA: Cascadilla Proceedings Project.
115
Dialect identification and listener attributes: Do you hear la tonada?
Jennifer Lang-Rigal
University of Texas at Austin
Listeners may rely on a few, linguistic features to identify the dialect of a given speaker. Which cues
are the most relevant, and how do they vary between populations of listeners? This study investigates
these questions by applying speech perception methodology to Spanish dialect variation in examining a
particularly salient prosodic feature of the Spanish spoken in Córdoba, Argentina, often referred to as
“la tonada cordobesa” (Fontanella de Weinber 1971; Yorio 1973). For instance, in Standard Spanish,
or the word “tonada”, the tonic syllable /na/ would be lon er in duration and may also have hi her
pitch and an increase in intensity relative to the other syllables (Hualde 2002). Whereas, in la tonada
cordobesa, these correlates of stress might apply to the pre-tonic syllable /to/ o ‘tonada’, resultin in a
realization /to:náda/, when in phrase-final position. However, the perception of this dialect feature has
not been examined, despite the widespread popular attention that this dialect receives, and the
uniqueness of pre-tonic vowel lengthening across languages. This leads to the goals of the present
study: To determine if the feature of vowel duration alone is enough for a speaker to be perceived as
being from Córdoba (as opposed to being from another dialect region of Argentina), and to investigate
the differences of perception between the in-group (listeners from Córdoba) and the out-group
(listeners from Buenos Aires). This is done by use of a matched-guise perception test.
The methodology for this test is fairly straightforward: Listeners perform a dialect identification task
presenting short tokens (4-9 syllables) of naturalistic speech from speakers of three different dialect
regions (Buenos Aires, Córdoba, Tucuman). The question accompanyin these stimuli, “where is this
speaker most likely rom?” presents these same three dialect re ions to the listener, with a three-point
scale o certainty, where the speaker is “de initely” or “maybe” rom the dialect re ion selected. Each
token consists of one intonational phrase (declarative neutral) with a multisyllabic word in phrase-final
position, since this is where pre-tonic lengthening is observed to occur. This is a matched-guise
perception test since each of the naturalistic tokens were submitted to Praat for manipulation of the pre-
tonic vowel duration, creating a duplicate of the natural token (Figures 1,2). The resulting tokens
(n=80) represent three speaker origins (Córdoba, Tucuman, or Buenos Aires) and two pre-tonic vowel
durations (a natural and manipulated (long or short) duration for each token, see Table 1). The
perception test is accompanied by a written questionnaire investigating the language experience of each
listener subject; includin the subject’s re ional a iliation and that of their family, friends, and
coworkers; the subject’s history o travel within and outside o r entina, and in particular to Córdoba;
their knowledge of and experience with Spanish dialects and other languages, and their beliefs towards
language use and variation.
Preliminary results suggest that longer pre-tonic vowel length influences perception of Córdoba origin,
even when the speaker voice and all other features stayed the same in one-word tokens in which
pretonic vowel duration was manipulated for a speaker from Córdoba and one from Buenos Aires
(LangRigal 2012). Additionally, this study aims to expose the effect of listener experience and origin
on the sensitivity to the linguistic features characteristic of their same dialect group (Peters et al. 2002),
and the underlying listener percepts that associate a particular linguistic feature, (eg. pre-tonic vowel
lengthening) to a speaker characteristic (eg. region of origin).
116
Table 1. Token types based on speaker origin and nature pre-tonic vowel duration.
Regional origin/token type Natural Manipulated duration
Córdoba Experimental Experimental
Buenos Aires Control Control
Tucuman Distractor Control
Figure 1. Naturally produced token, waveform and durations (ms) for experimental word
segments of ‘Argentina’.
0 0.8833 Time (s)
Figure 2. Manipulated token, waveform and durations (ms) for experimental word segments of
‘Argentina’, in which duration of pre-tonic vowel /e/ has been shortened.
0 0.8833 Time (s)
Fontanella de Weinberg, M. B. (1971). La entonación del español de Córdoba (Argentina) Thesaurus:
Boletín del Instituto Caro y Cuervo, 26(1),11-21. Hualde, J. (2002). Intonation in Romance. Probus:
International Journal of Latin and Romance Linguistics, 14(1), 1-204. Lang-Rigal, J. (2012). Pre-
tonic vowel lengthening in the Spanish of Córdoba, Argentina. Unpublished Manuscript. Peters, J.,
Gilles, P., Auer, P., & Selting, M. (2002). Identification of Regional Varieties by Intonational Cues:
An Experimental Study on Hamburg and Berlin German. Language And Speech, 45(2), 115-139.
Yorio, C.A. (1973). “ honolo ical style in the dialect o Spanish o Córdoba, r entina.” Dissertation
University of Michigan.
117
The Realization of Word-Final, Preconsonantal /s/ in the Spanish of Mexico City
Greg Lewis
University of North Carolina - Wilmington
The realization of syllable-final /s/ has become one of the most studied processes in Spanish
phonology. Therefore, the lack of studies involving the relationship between certain linguistic factors
and the realization of syllable-final /s/ comes as a surprise. The current study analyzes the relationship
between word stress, following consonant, and sex on the realization of word-final, preconsonantal /s/
in the Spanish of Mexico City. Historically, a fundamental differentiating characteristic of the variety
of Spanish spoken in the Valley of Mexico has been the full retention of syllable-final /s/ (Henríquez-
Ureña 1921; Hualde 2005). However, the results of the present study demonstate surprisingly high rates
of /s/ deletion among thirteen (7 female, 6 male) native Spanish speakers from the Mexico City
metropolitan area. Each participant was either a current or ormer student o a Master’s in ublic
Health program at the Universidad Autónoma Metropolitana – Xochimilco; the age range for all
participants was 24 to 33 years. All of the data was gathered by the first author in July of 2012 at the
UAM – Xochimilco. The data to be analyzed consisted of 44 sentences, each containing the target
sequence of verb (word-final /s/) followed by a noun beginning with a consonant. The initial
consonants of the noun were voiced obstruents (/b, d, g, ʝ/), liquid (/l/), nasals (/m, n/), or voiceless
obstruents (/p, t, k, f/). One of four syllabic stress conditions was possible in every verb with final /s/
noun sequence (1. penultimate syllable stressed in verb and first syllable stressed in noun: vendes
gatos; 2. penultimate syllable stressed in verb and second syllable stressed in noun: llevabas camisa; 3.
final syllable stressed in verb and first syllable stressed in noun: beberás leche; 4. final syllable stressed
in verb and first syllable stressed in noun: observerás delfines). Both researchers independently
analyzed the realization of each /s/ target via spectrogram pulses, which indicate vibration of the vocal
chords, in Praat (Boersma & Weenik 2012). Tokens were either coded [s], [z], or Ø (deletion). There
were no instances of aspirated /s/ in the current data. In order to test whether the observed results for
the linguistic and extralinguistic factors were statistically significant, a linear mixed model was run
using SPSS v19. Main effects and interactions were tested, and a Bonferroni adjustment for multiple
comparisons was applied for posthoc analyses. Main effects were found for following consonant (F(10,
550) = 18.233, p = .000), following consonant class (F(3, 560.0) = 59.443, p = .000), following
consonant voicing (F(1, 570) = 104.569, p = .000), and sex (F(1, 564) = 10.906, p < .01). A significant
interaction was found for following consonant class and stress position (F(3, 560.0) = 4.352, p < .01),
and for following consonant class and sex (F(3, 560.0) = 4.59, p < .01). Apart from providing evidence
that /s/ deletion occurs in the Spanish spoken in central Mexico, the results of the present study also
support the earlier findings of Muñiz-Cachón and Cuevas Alonso (2003) and Schmidt and Willis
(2011) that /s/ before a voiced consonant is far more vulnerable to variation than /s/ before a voiceless
consonant.
118
Figure 1. Realization of /s/ based on speaker sex and following consonant class.
Figure 2. Realization of /s/ by following consonant.
References
Boersma, Paul, and David Weenik. Praat: Doing Phonetics by Computer. Computer software. Vers.
5.3.05. 19 Feb. 2012. Web. 25 Feb. 2012.
Henríquez-Ureña, Pedro. Observaciones sobre el español en América y otros estudios filológicos.
Buenos Aires: Academia Argentina de Letras, 1921. Print.
Hualde, José Ignacio. The Sounds of Spanish. Cambridge: Cambridge UP, 2005. Print.
IBM Corp. IBM SPSS Statistics for Windows: Computer software, Vers. 19.0. 2010.
Muñiz Cachón, Carmen, and Miguel Cuevas Alonso. "Grados de sonorización de la consonante /s/ en
el espàñol de Asturias." Revista de filoloxía asturiana 3-4 (2003): 291-304. Print.
Schmidt, Lauren B., and Erik W. Willis. "Systematic Investigation of Voicing Assimilation of Spanish
/s/ in Mexico City." Selected Proceedings of the Fifth Conference on Laboratory Approaches to
Romance Phonology. Ed. Scott M. Alvord. Somerville, MA: Cascadilla Proceedings Project, 2011. 1-
20. Print
0%10%20%30%40%50%60%70%80%90%
100%
voicedconsonant
voicedconsonant
voicelessconsonant
voicelessconsonant
males females males females
% R
eal
izat
ion
Following Consonant Class by Speaker Sex
[s]
[z]
0
0%
20%
40%
60%
80%
100%
b d g ʝ l m n p t k f
% R
eal
izat
ion
Following Consonant
[s]
[z]
deletion
119
Tense and Lexical Semantics Connections in Child Spanish SLI Spontaneous Production
Paij Lintz
The Ohio State University
John Grinstead
The Ohio State University
Juliana de la Mora
Universidad Autónoma de Querétaro
Myriam Cantú-Sánchez
Universitat Autònoma de Barcelona
Blanca Flores
Centro Nacional de Rehabilitación
The Aspect-First Hypothesis (e.g. Bloom, Lifter & Hafitz 1980) observes that children tend to associate
tense and grammatical aspect marking with the lexical semantics of their predicates. Wagner (2009)
contends that this is a function of information processing, in the sense that telic predicates, by virtue of
having inherent endpoints, are most easily associated with perfective grammatical aspect and past
tense, while atelic predicates, lacking inherent endpoints, require more information to be marked as
past, and are thus more likely to be marked with imperfective grammatical aspect and present tense, by
adults as well as children. Because tense is a critical part of the Aspect-First phenomenon, it is
interesting to ask whether children with SLI, who have been documented to have severe delays in
production and comprehension of tense marking (e.g. Rice, Wexler & Hershberger 1998), will follow
prototypical patterns. A clinically-relevant theoretical question raised by these facts is whether the SLI
difficulty with tense is limited to morphosyntax, or whether it extends to the lexical semantics
implicated in the Aspect-First phenomenon, as well. Leonard et al (2007) show in an elicited
production experiment that English-speaking children with SLI are less consistent than controls in their
Aspect-First prototypical tense-aspect groupings. Similarly, Leonard & Deevy (2010) demonstrate in a
comprehension experiment that English-speaking children with SLI are worse than age-matched peers
at using the same type of grouping. The Aspect-First prototypes have also been shown to hold of
Spanish in spontaneous production (Jackson-Maldonado & Maldonado 2001) and in comprehension
with typically-developing children (Grinstead, Pratt & McCurley 2009). Child Spanish-speakers with
SLI, however, appeared to not follow Aspect-First prototypes in comprehension (Grinstead et al 2013).
In the current project, we study the spontaneous production data of 38 monolingual Spanish-speakers in
Mexico, 19 of whom are diagnosed with SLI (mean age = 67 months; mean MLUw = 2.97) and 19 of
whom are age-matched controls (mean age = 67 months; mean MLUw = 5.49). We coded the
spontaneous production o each child’s 20-30 minute language sample for tense, grammatical aspect
and lexical aspect to determine: 1) whether each group followed Aspect-First prototypical tense-aspect
groupings and 2) whether children in the SLI and TD groups were different from one another in their
tense-aspect distributions. With respect to question 1, children in both groups used more telics than
either atelics or statives in the past and more statives than telics in the present, following the Aspect-
First prototypes, in contrast to the Grinstead et al (2013) comprehension study. With respect to question
2, SLI children were not significantly different from TD children in their distribution of predicates, as a
function of tense, contra results from English. Results are discussed in terms of comprehension-
production differences.
120
References
Bloom, L., Lifter, K., & Hafitz, J. (1980). Semantics of verbs and the development of verb inflection in
child language. Language, 56, 386-412.
Grinstead, J., McCurley, D., Pratt, T., Obregon, P., & Flores, B. (2013). The Semantics of the Tense
Deficit in Child Spanish SLI. In M. Becker, J. Grinstead & J. Rothman (Eds.), Generative
Linguistics and Acquisition: Studies in Honor of Nina M. Hyams (pp. 107-128). Amsterdam:
John Benjamins.
Grinstead, J., Pratt, A., & McCurley, D. (2009). Comprehension of Prototypical Tense and Aspect
Combinations in Child Spanish. Studies in Hispanic and Lusophone Linguistics, 2(2), 435-450.
Jackson-Maldonado, D., & Maldonado, R. (2001). Determinaciones semánticas de la flexión verbal en
la adquisición temprana del español. In C. Rojas-Nieto & L. De León Pasquel (Eds.), La
adquisición de la lengua materna: español, lenguas mayas, euskera (pp. 165-200). México
D.F.: UNAM-CIESAS.
Leonard, L., Deevy, P., Kurtz, R., Chorev, L. K., Owen, A. J., Polite, A., et al. (2007). Lexical Aspect
and the Use of Verb Morphology by Children with Specific Language Impairment. Journal of
Speech, Language, and Hearing Research, 50(June), 759-777.
Leonard, L. B., & Deevy, P. (2010). Tense and aspect in sentence interpretation by children with
specific language impairment. Journal of Child Language, 37(2), 395-418.
Rice, M. L., Wexler, K., & Hershberger, S. (1998). Tense over Time: The Longitudinal Course of
Tense Acquisition in Children with Specific Language Impairment. Journal of Speech,
Language, and Hearing Research, 41(6), 1412-1431.
Wa ner, L. (2009). ’ll Never Grow Up Continuity in spect Representations. Linguistics, 47(5),
1051-1074.
Past Present Z Score p
Stative .08 .36 -5.156 < .001
Atelic .04 .11 -4.039 < .001
Telic .20 .13 -2.255 .024
Table 1 - Proportions of Lexical Aspect Categories in Present and Past Tense, Comparisons Using Wilcoxon Signed Rank
Test, alpha = .05.
121
Figure 1 – Distribution of Predicate Types As a
Proportion of All Verbs, in Both Groups (SLI
& TD), Different in the Past vs. the Present, By
Friedman Test and Post-Hoc Wilcoxon Signed
Rank Tests (p < .05)
Figure 2 - Past Tense Distributions of Lexical
Aspect Between SLI and TD Groups, No
Difference By Friedman Test and Post-Hoc
Wilcoxon Signed Rank Tests (p > .05)
122
Task-related effects on the production of L3 VOT
Raquel Llama
University of Ottawa
Walcir Cardoso
Concordia University
Laura Collins
Concordia University
Several studies within the subfield of Third Language (L3) Phonology seem to have found a
more marked effect of the second or non-native language (L2) than that of the first or native language
(L1) on L3 pronunciation (Tremblay, 2007). While this may be the case, a predominant influence from
the L1 has also been attested (Llisterri & Poch, 1987). Moreover, some authors have pointed to the
possibility of more than one source of cross-linguistic influence (Wrembel, 2011; Wunder, 2011), or
combined CLI, as an explanation for their findings. Among the factors that may cause one of the
previously-learnt languages to override influence from the other, we may find: typological proximity to
the L3 (Wrembel, 2012), L2 status, L2 proficiency or the type of elicitation task (Hammarberg &
Hammarberg, 1993).
It is the aim of this work to investigate the relative influences of typology and L2 status on L3
phonology within two different tasks at an intermediate level of L3 proficiency. More specifically, our
study was set up to address the following question: Which is a stronger predictor in the selection of a
source language for phonological influence in L3 acquisition in a formal (reading list) versus a more
informal (running speech) context: L2 status or typology?
Two groups of L3 Spanish learners were tested: Group A (n=10) consisted of L1 English-L2
French speakers, whereas Group F (n=10) consisted of L1 French-L2 English speakers. An ANOVA
compared the rates of presence (English-influenced) or absence (French-influenced) of aspiration
(longer VOT) in the production of the word-initial voiceless stops /p t k/ in the two groups. Although
there were diverging trends for the two tasks, our findings suggest a more determinant role for the L2
than for typology in the production of /p/, /t/, and /k/ in stressed onset positions in Spanish for both
groups. However, the presence of combined CLI, as evident from the hybrid values found, cannot be
ruled out.
References
Hammarberg, B., & Hammarberg, B. (1993). Articulatory re-setting in the acquisition of new
languages. Reports from the Department of Phonetics, University of Umea, Phonum, 2, 61-67.
Llisterri, J., & och, D. (1987). honetic inter erence in bilin ual’s learnin o a third lan ua e. n
Proceedings of ICPhS XI, The Eleventh International Congress of Phonetic Sciences (pp. 134-
137), Tallinn, Estonia. Estonia: Academy of Sciences of the Estonian SSR.
Tremblay, M. C. (2007). “L2 in luence on L3 pronunciation Native-like VOT in the L3 Japanese of
English-French bilin uals.” aper presented at the Satellite Workshop of ICPhS XVI, Freiburg,
Germany.
Wrembel, M. (2011). Cross-linguistic influence in third language acquisition of voice onset time. In L.
Wai Sum & E. Zee (eds.), Proceedings of the Seventeenth International Congress of Phonetic
Sciences ICPhS (pp. 2157-2160). Hong Kong: City University of Hong Kong (CD Rom).
123
Wrembel, M. (2012). Foreign accent ratings in third language acquisition: the case of L3 French. In E.
Wanick-Klimczak & L. Shockey (eds.), Teaching and Researching English Accents in Native
and Non-native speakers (pp. 29-45). Berlin / Heidelberg: Springer Verlag.
Wunder, E. M. (2011). Crosslinguistic Influence in Multilingual Language Acquisition: Phonology in
Third or Additional Language Acquisition. In G. de Angelis, & J. M. Dewaele (Eds.), New
Trends in Crosslinguistic Influence and Multilingualism Research (pp. 105-128). Clevendon:
Multilingual Matters.
124
The role of experience, saliency and auditory enhancement in Spanish and English voicing
Fernando Llanos
Olga Dmitrieva
Alexander L. Francis
Purdue University
The present study examines the role of language experience, saliency and auditory enhancement in the
processing of Spanish and English voicing. Although the main cue to voicing distinctions in stop
consonant voicing is Voice Onset Time (Lisker, 1978), onset f0 (f0 at the onset of voicing) has been
proven to be another reliable cue (Idemaru & Holt, 2004) in languages like English in which voiced
stops ([p]) exhibit a short VOT and a low onset f0 whereas voiceless stops (i.e. [ph]) exhibit a long
VOT and a high onset f0. For some authors the perceptual use of onset f0 as a cue to voicing is
determined by listeners’ experience with its correlation with VOT across voicin cate ories (Holt et al.,
2001). However, other studies suggest that the effect of onset f0 on voicing perception emerges directly
from the way in which it is processed by the auditory system (Kingston et al. 2008). According to this
hypothesis, a lower onset f0 enhances the amount of low frequency energy characteristic of voiced
categories.
As in English, onset f0 correlates with VOT across Spanish voicing categories: Spanish voiced stops
(i.e. [b]; voicing starts well before the consonant release) exhibit a low onset f0 whereas voiceless stops
([p]) exhibit a high onset f0 (Cassie, 1982). However, Llanos et al. (submitted) demonstrated that, in
contrast to English, Spanish listeners do not rely on onset f0 to identify voiced stops. This lack of an
onset f0 effect in the perception of Spanish voiced stops could be the result of a lower degree of
correlation between VOT and onset f0 across productions of the Spanish voicing categories (consistent
with an experience-based explanation). On the other hand, consistent with the auditory enhancement
hypothesis, the contribution of onset f0 to the perception of Spanish voiced stops could be undermined
by the salient prevoicing that characterizes the production of such categories. According to this view,
the auditory contribution of onset f0 decreases when VOT is very salient. Given that English voiceless
stops are also characterized by a salient VOT, this view predicts that the onset f0 effect in Spanish and
English will be concentrated in the range in which VOT is acoustically less prominent in signaling the
voicing feature (VOT values that are not too long).
In this study we present the results of a new experiment in speech production and a new perceptual
analysis of data from Llanos et al., (submitted) to test these two possibilities. The production task
consisted of a series of recordings of Spanish and English words beginning with different voiced and
voiceless stops from 24 native speakers of Spanish tested in Spain and 30 native speakers of English
tested in the USA. Results indicate that the correlation of onset f0 across voicing categories in Spanish
(r= 0.51) is much larger than in English (r=0.34) (Figure 1). This suggests that Spanish listeners' failure
to use onset f0 in identifying voiced stops cannot be the result of lack of experience with a correlation
between onset f0 and VOT in their native language.
For the perceptual analysis, 10 English and 9 Spanish native listeners tested in Spain and US
(respectively) classified tokens varying in VOT (from -60 to +60 ms) and onset f0 (from 90 to 150 Hz)
in a 2AFC identi ication task (i.e. “B” vs. “ ”). n order to measure the onset 0 e ect or each VOT
value, the VOT range was subdivided into 30 ms analysis windows (10 ms step) and for each window
each group of responses (Spanish/English) was submitted to two different logistic models, one with one
predictor (VOT) and another with two predictors (VOT and onset f0). The onset f0 effect at each point
was then taken as the difference between the residual deviances of these two models (the lower the
onset f0 effect, the lower the difference between residual deviances; Hosmer & Lemeshow, 2004). As
125
predicted, the onset f0 effect in both Spanish and English is concentrated in the range of VOT in which
VOT is less salient (Figure 2). This suggests the existence of a language-independent (i.e. perceptual)
mechanism of cue integration in which the contribution of the secondary cue (onset f0) to the
perception of voicing is determined by the saliency of the primary cue (VOT).
Figure 1 Production task results (blue dots = voiced; red dots = voiceless)
Production task results for native
speakers of Spanish (N=24, left panel)
and English (N=30, right panel), tested
in Spain and the US, respectively.
Correlations between VOT (measured
in ms) and onset f0 (measured in Hz,
expressed in semitones with respect to
each speaker's average f0) across both
voicing categories were r=0.51
(Spanish) and r=0.34 (English).
Figure 2 Perceptual analysis
Perception task results showing relative importance
of onset f0 in the voicing decision by Spanish (N=9, top panel) and English (N=10, bottom panel)
listeners. Onset f0 effect is calculated as the difference between the residual deviances of a
logistic model with a single predictor (VOT) and
one with two predictors (VOT and onset f0) within successive 30 ms windows of the VOT range (10
ms step). Larger values indicate greater reliance on onset f0.
References
Caisse, M. (1982). “Cross-linguistic differences in fundamental frequency perturbation induced by
voiceless unaspirated stops. Berkeley,” M. . thesis, Univ. of California-Berkeley.
Holt, L. L., Lotto, . J., and Kluender, K. R. (2001). “ nfluence of fundamental frequency on
stopconsonant voicin perception case o learned covariation or auditory enhancement?” J. Acoust. Soc. Am. 109, 764–774.
Hosmer, D. W., & Lemeshow, S. (2004). Applied logistic regression (Vol. 354). WileyInterscience.
demaru, K., and Holt, L. L. (2011). “Word reco nition reflects dimensionbased statistical learnin ,” J.
Exp. Psychol. Hum. Percept. Perform. 37(6), 1939–1956.
126
Kin ston, J., Diehl, R. L., Kirk, C. J., and Castleman, W. . (2008). “On the internal perceptual
structure o distinctive eatures The [voice] contrast,” J. honet. 36, 28–54.
Lisker, L. (1978). “ n quali ied de ense o VOT,” Lan and Speech, 21(4), 375-383.
Llanos, F., Dmitrieva, O., Francis, L. A., and Shultz, A. (submitted). Auditory enhancement explains
Spanish and English weighting of secondary voicing cues. J. Acoust. Soc. Am.
127
Mitigation in L2 Spanish: Does conversational ability survive a change in context?
Kelly Lovejoy
Rutgers University
All speakers engage in arguments in day-to-day activities (Antaki, 1994; Schiffrin, 1985). Thus, the
ability to make a point, to state a position, and to question another person's position is necessary for all
language users. In both Spanish and English, this ability hinges upon the use of mitigation and
indirectness, or words and behaviors that reduce the negative impact of what is said. It has been
demonstrated that classroom language learners, in developing pragmatic proficiency, often lack
mitigation or employ it in ways that diverge from the target language norm (e.g., Félix-Brasdefer, 2007;
Shively and Cohen, 2008).
This paper presents results obtained in a study of the use of mitigation in conversational arguments
produced in Spanish by intermediate- and advanced-level learners of Spanish as a second language and
native Spanish speakers. Fourty four participants completed two dyadic tasks that were designed to
elicit spontaneous arguments: a ranking conversation and a film narration. The results of a quantitative
analysis using SPSS and a qualitative micro-analysis using methods of Conversation Analysis point to
two key issues relating to learner language: levels of acquisition and context. Regarding acquisition,
intermediate L2 Spanish learners use mitigation with less frequency than advanced learners. Regarding
context, the use of mitigation by both L2 learner groups varies significantly according to the
conversational task, with the film narration yielding fewer argument interactions and significantly
fewer tokens of mitigation by both L2 learner groups. Of particular interest are learners' use of tag and
challenge questions (e.g., No es cierto? 'Isn't that right?') and discourse markers (e.g., o sea 'that is', no
sé 'I don't know') to mitigate the argument interactions.
These findings are discussed in terms of institutional talk, defined as spontaneous language that is
produced by speakers who are sensitive to the social context they are in (Bardovi- Harlig and Hartford,
2005). In other words, the framework of institutional talk allows us to consider discourse in terms of
the situation in which it occurs, by taking into consideration the individual identities of the speakers as
well as the influence of the context in which they interact. It is argued that there are limits to L2
learners' ability to mitigate in a conversational argument, and that this ability is dependent on the
context of the interaction (e.g., in a conversation versus a cooperative narration) and their status as
language novices.
References
Antaki, C. (1994). Explaining and Arguing: The Social Organization of Accounts. London: Sage
Publications.
Bardovi-Harlig, K., and Hartford, B. (2005). Institutional discourse and interlanguage pragmatics
research. In K. Bardovi-Harlig and B. Hartford (Eds.), Interlanguage Pragmatics: Exploring
Institutional Talk, 7-36. Mahwah, NJ: Lawrence Erlbaum.
Félix-Brasdefer, J.C. (2007). Pragmatic development in the Spanish as a FL classroom: A
crosssectional study of learner requests. Intercultural Pragmatics 4(2), 253-286.
Schiffrin, D. (1985). Everyday argument: The organization of diversity in talk. In T. van Dijk (Ed.),
Handbook of Discourse Analysis. New York: Academic Press.
Shively, R.L., and Cohen, A.D. (2008). Development of Spanish requests and apologies during study
abroad. Íkala: Revista de Lenguaje y Cultura, 13(20), 57-118.
128
An acoustic investigation of Peruvian Amazonian Spanish “long” vowels
Miguel García
The Ohio State University
While it has been generally claimed that in Spanish there are only short vowels and that contrast
between short and long vowels is non-existent (Navarro Tomás 1967, Hualde 2005), Face 2000
(among others) shows that phonetically longer vowels are found in narrow focus declaratives. The
present study looks at vowels of Peruvian Amazonian Spanish (PAS), and shows that, in broad
focus declaratives, vowels, specifically pre-tonics and tonics, are phonetically longer than in other
Spanish varieties previously surveyed in the literature. The main goal of this study is to analyze
vowel duration in PAS and contrast it with that of the Lima variety, and moreover relate these
results to previous PAS intonational descriptions.
The data for the current study comes from seven speakers from Pucallpa, Peru, and seven
speakers from Lima, Peru, who were recorded during the summer of 2012. All participants were
native Spanish monolinguals; college students; and their ages ranged between 20-25 years. The five
Spanish vowels were examined in four different contexts (stressed vs. unstressed syllable, and non-
final vs. final position syllable within a target word) in a reading task of Subject-Verb-Object
declaratives. The target sentences
(e.g. Mariana donaba las monedas ‘Mariana was donatin the coins’) containin the tar et words
were elicited by a question which prompted a broad focus response, such as ¿(Sabes) qué pasó?
‘(Do you know) what happened?’ ll tar et words were trisyllabic; urthermore, the vowel duration
analysis focused only on verb and object position, and pre-tonic and tonic syllables. As a way to
accurately contrast the measurements in both varieties, duration values were normalized by dividing
the duration of the target vowel by the duration of the entire sentence (following Estebas
Vilaplana’s 2010 study on r entinian Spanish). While there have been multiple works on vowel
duration on Spanish (Marín Gálvez
1994, Morrison & Escudero 2007, Chládková et al. 2011), the present study differs from those
foundational works in (1) the use of more common words, and (2) their inclusion in broad focus
declaratives (thus avoiding possible narrow focus effects).
Results of this study show that in verb and object positions, PAS vowels are significantly
longer than Lima vowels in both pre-tonic and tonic positions. For tonic vowels, the normalized
duration is 58 ms in PAS and 40 ms in Lima Spanish (raw values: 98 ms vs. 62 ms). For pre-tonic
vowels, the normalized duration is 41 ms in PAS and 34 ms in Lima (raw values 74 ms vs. 56 ms).
These differences were statistically significant (p<.01). While in both varieties stressed vowels are
phonetically longer in final sentence position, PAS vowels were significantly longer than in Lima:
PAS (normalized: 68 ms; raw: 115 ms) and Lima (normalized: 42 ms; raw: 66 ms). Further analysis
looked at vowel duration in post-tonic position, and their ratio to vowels in pre-tonic and tonic-
position. Results show that in non-final sentence position PAS tonic vowels are on average more
than twice as long as post-tonic vowels whereas in Lima tonic vowels are only slightly longer than
post-tonic vowels (see Figure 1).
These findings suggest that PAS exhibits phonetically longer vowels in pre-tonic and tonic
positions than Lima Spanish, a “standard” eruvian variety. urther propose that this behavior
should not be considered a coincidence if we take into consideration Garcia 2011. That study
reported that in PAS the high tone tends to be aligned within the stressed syllable in pre-nuclear
accents in broad focus declaratives. Based on the results of the present study, I propose that this
intonational pattern is connected to phonetically longer stressed vowels in PAS, mainly in pre-
nuclear accents, by providing sufficient time for the high tone to align within the limits of the
129
stressed syllable. Hence, this study presents a possible explanation for the PAS intonational patterns
not yet provided in its previous descriptions.
Figure 1. Oscilogram and spectrogram of the paroxytone word pelaron ‘peeled’ produced by a Pucallpa speaker (left) and a Lima
speaker
(right) in the broad focus declarative Las señoras pelaron los melones ‘The ladies peeled the melons’. The normalized
duration of /e/, /a/ and /o/ in the word pelaron for the Pucallpa speaker (left) is 46ms, 63ms, and 27ms, respectively. The
normalized duration of /e/, /a/, and /o/ in the word pelaron for the Lima speaker (right) is 27ms, 40ms, and 33ms, respectively.
References
Chládková, Kateřina, aola Escudero, and aul Boersma. 2011. Context-specific acoustic
differences between Peruvian and Iberian Spanish vowels. Journal of Acoustic Society of America
130(1).416428.
Estebas Vilaplana, Eva. 2010. The role of duration in intonational modeling. A comparative study of
Peninsular and Argentinean Spanish. Revista española de lingüística aplicada 23.153-174.
Face, Timothy. 2000. Prosodic manifestations of focus in Spanish. Southwest Journal of Linguistics
19(1).45-62.
Garcia, Miguel. 2011. The Intonational Patterns of the Peruvian Amazonian Spanish (PAS). Stony
Brook University, New York. M.A. thesis.
Hualde, José I. 2005. The Sounds of Spanish. Cambridge, UK: Cambridge University Press.
Marín Gálvez, Rafael. 1994. La duración vocálica en español. Estudios de Lingüística de la
Universidad de Alicante 10.213-226.
Morrison, Geoffrey S., and Paola Escudero. 2007. A cross dialect comparison of Peninsular- and
Peruvian- Spanish vowels. In J. Trouvain & W.J. Barry (ed.), Proceedings of the 16th International
Congress of Phonetic Sciences. 1505-1508.
Navarro Tomás, T. 1967. Manual de pronunciación española. Madrid: CSIC.
130
Impersonal, passive and aspectual se: the PCC & the periphrastic passive
Jonathan MacDonald
Christopher Eager
University of Illinois at Urbana-Champaign
Introduction: In this paper, we discuss Spanish impersonal se, passive se and aspectual se and their
differing behaviors with respect to the Person Case Constraint (PCC) and ability to appear in the
periphrastic passive. The main claim is that due to different positions where these distinct ses merge in
the structure, systematic patterns emerge with respect to PCC violations and grammaticality in the
periphrastic passive.
The PCC: Repair strategies of varying types have been noted for PCC violation avoidance
(Anagnostopoulou 2005, Bonet 2008, Recaz 2007). Bonet (2008) observes for leísta dialects of Spanish
that the accusative direct object clitic pronoun that usually surfaces as le surfaces as lo to avoid a PCC
violation. This is illustrated in (1). Taking this as one diagnostic for PCC violations, consider aspectual
se with the verb conocer “know” in (2). (2a) illustrates the use of le for accusative direct objects in
leísta dialects; (2b) illustrates the use of aspectual se with conocer; (2c) illustrates that le can no longer
surface; lo must surface. The conclusion is that aspectual se is subject to the PCC. In this respect,
aspectual se contrasts with impersonal se, an example of which is given in (3a). Observe in (3b) that le
surfaces with impersonal se. Impersonal se is not subject to the PCC constraint. Finally, observe that
passive se is also not subject to the PCC, since le can surface grammaticality.
we adopt na nostopoulou’s (2005) analysis o the CC, whereby two elements enter a multiple
agree relation with v, we can take PCC violations as a diagnostic for Agree with v. Given this
assumption and the facts in (1) to (3), impersonal and passive se are entailed not to Agree with v, while
aspectual se is. Assuming that impersonal se merges above v, following work by Mendikoetxea (2008),
Ordóñez & Treviño (2011) and Rivero (2002) the fact that impersonal se is not subject to the PCC falls
out immediately on na nostopoulou’s multiple ree account. ssumin that the basic unction o
passive se is to remove v’s accusative Case assi nin abilities (applying assumptions about English en
from Boeckx & Hornstein 2004), it is natural that se does this as the head of v itself. In this position,
passive se cannot Agree with v, since it is not c-command by v, and thus passive se is not subject to the
PCC. Finally, assuming that aspectual se is lower in the verb phrase, minimally below v (Basilico 2010,
Campanini & Schäfer 2011, Armostrong 2011, MacDonald 2004, 2006), it is c-command by v, can
Agree with it and is thus subject to the PCC.
The Periphrastic Passive: A look at the periphrastic passive in combination with impersonal, passive
and aspectual se offers further support for the basic proposed contrasts among these ses. As has been
noted, impersonal se is the only se in Spanish that can appear in the periphrastic passive (see Sánchez
López 2002 and reference therein). This is illustrated in (5a). Why can impersonal se appear in the
periphrastic passive? Because it merges in a position above v, following Mendikoetxea (2008),
Ordóñez & Treviño (2011) and Rivero (2002), and does not interfere with the passive operation. The
inability of passive se to appear in the periphrastic passive follows directly from the assumption that
passive se and the participle morphology ado play the same role: they remove the accusative case
assigning ability of v (following Boeckx & Hornstein 2004 on English en). If ado is present, accusative
case has been removed from v, thus there is no Case for se to remove. Regarding aspectual se, observe
in (5b) that it is ungrammatical in the periphrastic passive. This is rather surprising, on the assumption
that aspectual se, being an unselected element, is dative—accusative already assigned to the direct
object—since overtly dative indirect object le, as illustrated in (6a) can appear in the periphrastic
passive. As has been known since Kayne (1975), observe that even indirect object se in a ditransitive
cannot appear in the periphrastic passive (6b). We assume that the proper account of indirect object se
131
in the periphrastic passive can extend straightforwardly to aspectual se in periphrastic passive. To date,
however, all accounts—those assuming that an external argument is required to license indirect object
se (Kayne 1975, Burzio 1986, and Pesetsky 1995) and those assuming indirect object se intervenes
between the subject and its empty category (Rizzi 1986)—fail to account for the grammaticality of
indirect object se in Romanian, as noted by Dobrovie-Sorin (2006:132), and illustrated in (7). We claim
here that the proper account is that both indirect object se and aspectual se in Spanish are accusative;
since there is no accusative in passive, se is not licensed. In this respect it contrasts with overtly dative
şi in Romanian. The fact that aspectual se is subject to the PCC and is out in the periphrastic passive
reduces to a single syntactic relation aspectual se establishes with v: Agree. Agree with v entails that it
is subject to the PCC and that it can check accusative case.
Conclusion: We have argued that impersonal, passive and aspectual se in Spanish are in structurally
distinct positions relative to v, the accusative case assigning head. These distinct structural positions
explain their distinct patterns with respect to PCC violations and their grammaticality in the
periphrastic passive, if basic assumptions regarding the PCC and periphrastic passive are adopted.
Impersonal, passive and aspectual se: the PCC & the periphrastic passive
(1) Te *le/lo dí.
(2) a. María le/*lo conoce muy bien.
b. María se conoce a Juan muy bien.
b. María se *le/lo conoce muy bien.
(3) a. Se conoce a Juan como un buen padre.
b. Se le/*lo conoce como un buen padre.
(4) a. María le dí un regalo a él.
b. Se le dio un regalo (a él).
(5) a. Se es perseguido por las fuerzas del mal. [Suñer 2002: 223]
b. *La cerveza se fue bebida
(6) a. El professor le ha sido presentado (a María)
b. El profesor (*se) ha sido presentado (a sí mismo).4
(7) on şi Maria şi-au fost prezentaţi anul trecut
4 A felicitous context here might be one where the professor has a clone that he meets for the first time.
132
References
na nostopoulou, E. 2005. ‘Stron and Weak erson Restrictions. Feature Checkin nalysis’. n
Heggie & Ordoñez (eds). Clitics and Affix Combinations: Theoretical Perspectives. Amsterdam:
John Benjamins.
Armstrong, Grant. 2011. Two classes of transitive verbs: Evidence from Spanish. Ph.D. Dissertation.
U. Georgetown.
Basilico, D. 2010. The se clitic and its relationship to paths. Probus 22.2: 271-302.
Bonet, Eulàlia. 2008. The Person-Case constraint and repair strate ies. n Roberta d’ lessandro,
Susann Fischer, Gunnar Hrafn Hrafnbjargarson (eds.) Person Restrictions, Mouton de Gruyter.
Burzio, Luigi. 1986. Italian Syntax: A Government-Binding Approach. Dordrecht: Reidel.
Dobrovie-Sorin, Carmen. 2006. The SE-Anaphor and its Role in Argument Realization. In The
Blackwell Companion to Syntax (Vol 4) Martin Everaert and Henk van Riemsdijk (eds.) 118-179.
Malden, MA: Blackwell.
MacDonald, Jonathan E. 2004. Spanish Reflexive Pronouns: a Null Preposition Hypothesis. In G.
Garding & M. Tsujimura (eds.) Proceedings of WCCFL 23. Somerville, MA: Cascadilla Press.
MacDonald, Jonathan E. 2006. Null Directional Prepositions in Romanian and Spanish. In New
Perspectives in Romance Linguistics Vol.1: Morphology, Syntax, Semantics, and Pragmatics.
Selected Papers from the 35th Linguistics Symposium on Romance Linguistics. J.P. Montreuil and
C. Nishida (eds), 169-185. Amsterdam: John Benjamins.
Mendikoetxea, maya, 2008. “Clitic impersonal constructions in Romance Syntactic features and
semantic
interpretation”, Transactions of the Philological Society 106:2, 290-336.
Ordóñez, Francisco & Esthela Treviño. 2011. Selected Proceedings of the 13th Hispanic Linguistics
Symposium, ed. Luis A. Ortiz-López, 314-324. Somerville, MA: Cascadilla Proceedings Project.
Pesetsky, David (1995). Zero Syntax: Experiencers and Cascades. Cambridge: MIT Press.
Rivero, María Luisa, 2002. “On impersonal re lexives in Romance and Slavic and semantic variation”,
in Joaquim Camps and Caroline R. Wiltshire (eds.), Romance Syntax, Semantics and L2
Acquisition: Selected Papers from the 30th Linguistic Symposium on Romance Languages,
Amsterdam: Benjamins, 169–195.
133
Analyzing the Relationship between Awareness of Derivational Morphology and
Vocabulary Size in Instructed L2 Spanish
Nausica Marcos Miguel
University of Pittsburgh
Awareness of derivational morphology (ADM) can help learners of a second language (L2) to infer the
meaning of unknown words (e.g., Haastrup, 2008; Koda, 2008) as well as to retain new words (e.g.,
Hu & Nassaji, 2012). It is, however, not clear how L2 ADM relates to vocabulary size. In L1
acquisition, there is a strong relationship between ADM and vocabulary knowledge (see Kuo &
Anderson, 2006). In L2 acquisition, whereas some studies have found a strong relationship between
ADM and vocabulary size (e.g., Medellín Gómez & Auza Benavides, 2008; Mochizuki & Aizawa,
2000; Schmitt & Meara, 1997), others have not (e.g., Author, year; Danilović, Dimitrijević Savić, &
Dimitrijević, 2013; Morin, 2003, 2006). Thus, this paper contributes to the literature on ADM by
showing a weak relationship between ADM and vocabulary size in Englishspeaking learners of
instructed L2 Spanish.
This paper explores and compares the development of receptive and productive ADM with the learning
of vocabulary in English-speaking learners of instructed L2 Spanish (n=209). These learners, who were
taking classes from a 2nd
to a 7th
semester at a large US college, completed five tests:
1) a Productive Awareness Test where learners derived the word families of six non-words;
2) a Receptive Awareness Test where learners chose from three options (a noun, a verb and an
adjective) the possible meaning of a pseudo-word with a biunique suffix, i.e., a suffix
marking only one syntactic category;
3) a Yes/No-Vocabulary Size Test including corpus frequencies and textbooks’ requencies;
4) a Word-Family Test where learners provided the correct derived forms of six words; and
5) a Proficiency Cloze Test.
Results indicate that proficiency, and not vocabulary size, is the main predictor of ADM. Moreover,
productive and receptive ADM evolve differently. Whereas vocabulary size better predicts receptive
awareness, word family knowledge better predicts productive awareness. Furthermore, high levels of
receptive awareness were reached at beginning proficiency levels, whereas mastery of productive
awareness was only achieved for the most advanced learners in the study. In spite of this, the number of
derivational suffixes that these L2 learners could actively manipulate was very limited. In conclusion,
the weak relationship between ADM and vocabulary size call for including derivational morphology
within regular classroom instruction so that L2 learners can benefit from this knowledge.
Table 1. Correlations between the Five Tests
1 2 3 4
1. Proficiency Cloze Test
2. Vocabulary Size Test
3. Word Family Test
4. Receptive Awareness Test
.671**
.522**
.493**
.494**
.468**
.329**
5. Productive Awareness Test .331** .203* .406** .135
*p<.01; **p<.001.
134
References
Author (year).
Danilović, J., Dimitrijević Savi, J., & Dimitrijević, M. (2013). Affix Acquisition Order in Serbian EFL
Learners. Romanian Journal of English Studies, 10(1), 77-88.
Haastrup, K. (2008). Lexical Inferencing Procedures in Two Languages. In D. Albrechtsen, K.
Haastrup & B. Henriksen (Eds.), Vocabulary and Writing in a First and Second Language. New York:
Palgrave. Macmillan.
Hu, H.-c. M., & Nassaji, H. (2012). Ease of Inferencing, Learner Inferential Strategies, and Their
Relationship with the Retention of Word Meanings Inferred from Context. The Canadian Modern
Language Review/La Revue canadienne des langues vivantes, 68(1), 54-77.
Koda, K. (2008). Impacts of prior literacy experience on second-language learning to read. In K. Koda
& A. Zehler (Eds.), Learning to Read Across Languages. Cross-Linguistic Relationships in First and
Second-Language Literacy Development (pp. 68-96). New York: Routledge.
Kuo, L.-J., & Anderson, R. C. (2006). Morphological Awareness and Learning to Read: A Cross-
Language Perspective. Educational Psychologist, 41(3), 161-180.
Medellín Gómez, A., & Auza Benavides, A. (2008). Influencia del tamaño del vocabulario en el
conocimiento morfológico de afijos. Estudios de Lingüística Aplicada, 48.
Mochizuki, M., & Aizawa, K. (2000). An affix acquisition order for EFL learners: an exploratory
study. System, 28, 291-304.
Morin, R. (2003). Derivational Morphological Analysis as a Strategy for Vocabulary Acquisition in
Spanish. The Modern Language Journal, 87(2), 200-221.
Morin, R. (2006). Building Depth of Spanish L2 Vocabulary by Building and Using Word Families.
Hispania, 89(1), 170-182.
Schmitt, N., & Meara, P. (1997). Researching Vocabulary through a Word Knowledge Framework.
Studies in Second Language Acquisition, 20, 17-36.
135
Perfecting the past
Joanne Markle LaMontagne
María Cristina Cuervo
University of Toronto
In studies on L1 acquisition of Spanish it has been observed that the Present Perfect appears together
with the Present tense, or immediately after it, and that Peninsular Spanish children use it productively
before the Preterit (Hernández Pina 1984; López Ornat 1994, 1997; Aguado 1995; Bel 2002, a.o.). A
gap in research remains with respect to the acquisition of tenses as a system within which each tense
carves out a particular space. In particular, it remains to be seen whether the acquisition of the perfect
as a form entails the acquisition of its complex temporal and aspectual meaning. The fact that the
resent er ect is produced be ore the reterit is surprisin iven the er ect’s complexity.
Morphologically, the Present Perfect consists of two verbs, i.e., the auxiliary verb haber in the Present
tense and the past participle (1). Semantically, the Preterit denotes an event that took place at some
point anterior to the moment of speech (2). Building on work on the Perfect by Iatridou (2003),
Iatridou, Anagnostopoulou & Pancheva (2003) and von Fintel & Iatridou (2005), as well as Rojo &
Vei a’s (1999) analysis o the Spanish tenses, we take the Spanish resent er ect —‘antepresente’ or
Bello (1847)—to express an event that is anterior to a reference point that is itself simultaneous with
the moment of speech. Additionally the Present Perfect introduces a time interval that extends from the
reference time (the Present) into the past, and whose left boundary can be set by an adverbial (desde,
‘since’). Thus, a crucial difference between the Preterit and Present Perfect is that the Preterit
establishes a direct relation with the moment of speech, whereas in the Present Perfect this relation is
achieved indirectly via reference to a point that is simultaneous with the moment of speech. This
complex form and meaning combination raises the question of whether children actually make the right
association between the two.
In order to investigate this question further, we conduct a CHILDES (MacWhinney 2000)
corpus study and analyse longitudinal spontaneous data from López-Ornat (1994) and Vila’s (1990)
databases of two L1 Peninsular Spanish children, María (1;9-2;11) and Emilio (1;9-2;12). We report
comparative frequencies for the emergence of the Present, Present Perfect, Preterit and Imperfect
tenses, as in Hernández Pina (1984), López Ornat (1994, 1997), Aguado (1995), Bel (2002), a.o.
dditionally, we classi y each instance o the resent er ect into one o the ollowin “uses” recent
past (REC) (3), perfect of result (RES) (4), experiential perfect (5), and perfect of persistent situation
(6) (adapted from Gathercole 1986, Howe 2013). Preliminary results show the most frequent Present
er ect “uses” are RES (María 63.2%; Emilio 61.5%) and REC (María 33.8%; Emilio 35.4%). The
perfect of persistent situation — atridou’s universal per ect—(María 2.6%; Emilio 3.1%) and the
experiential perfect — atridou’s existential per ect— (María 0.4%; Emilio 0%) are marginal. Crucially,
REC and RES do not involve a time span into the past; their central characteristic is “current
relevance”, and the location o the event as anterior to some relevant slice o the present (esta semana,
este año, recién). n these children, however, the relevant ‘present’ in this ‘antepresente’ seems to be
the moment of speech: there is no indication of reference to the Present tense as a linguistic concept. In
other words, the Present Perfect of these children appears to have a simpler meaning of anteriority in
direct reference to the situation of speech, thus competing with the Preterit. This idea of competition
for semantic space is initially supported by a trade-off tendency observed in the data between the
resent er ect and reterit. s María and Emilio’s use o the reterit and mper ect increases, their use
of the Present Perfect decreases. From 2;4-2;11, María’s reterit and mper ect use triples, while her
Present Perfect use drops slightly, as compared to 1;9-2;3. In Emilio during the period 2;7-2;12, his
Present Perfect use reduces in half, while his Preterit and Imperfect use increases slightly, as compared
to 1;10-2;6.
136
(1) ¿Has comido bien (hoy)?
‘Have you eaten well (today)?’
(2) ¿Comiste bien (ayer)?
‘Did you eat well (yesterday)?’.
(3) REC recent past
Esta mañana hemos ido a nadar.
This morning we-have gone to swim
‘We went swimmin this mornin ’.
(4) RES perfect of result
Se ha secado la tinta del bolígrafo.
SE has dried the ink of-the pen
‘The pen has dried up’.
(5) Experiential perfect (Existential Perfect)
Desde 2010 he visitado a mis padres dos veces.
since 2010 I-have visited my parents two times
‘Since 2010 have visited my parents twice’.
(6) Perfect of persistent situation (Universal Perfect)
He vivido en Bogotá desde enero de 2010.
I-have lived in Bogotá since January of 2010.
‘ have lived in Bo otá since January 2010’.
References Aguado, G. (1995). El desarrollo del lenguaje de 0 a 3 años. Madrid: Ciencias de la educación presescolar y
especial.
Bel, A. (2002). Early Verbs and the Acquisition of Tense Feature in Spanish and Catalan. In A-T Pérez-Leroux
& J.M. Liceras (Eds.), The Acquisition of Spanish Morphosyntax: The L1/L2 Connection. (pp. 1-34).
Dordrecht: Kluwer Academic Publishers.
Bello, Andrés. 1847. Gramática de la lengua castellana destinada al uso de los americanos. Madrid: Edaf (1984
edition).
Gathercole, V.C. (1986). The acquisition of the present perfect: explaining differences in the speech of Scottish
and American children. Journal of Child Language 13, 537-560.
Howe, C. (2013). The Spanish Perfects. Pathways of Emergent Meaning. UK: Palgrave MacMillan.
Iatridou, S. (2003). A little bit more on the English perfect. In A. Alexiadou, M. Rathert & A. von Stechow
(Eds.), Perfect Explorations, (pp. 133-152). Berlin: Mouton de Gruyter.
Iatridou, S., Anagnostopoulou, E. & Pancheva, R. (2003). Observations about the form and meaning of the
perfect. In A. Alexiadou, M. Rathert & A. von Stechow (Eds.), Perfect Explorations, (pp. 153-204). Berlin:
Mouton de Gruyter.
López Ornat, S. (1994). La adquisición de la lengua Española. Madrid: Siglo XXI.
Rojo, G. & Veiga, A. (1999). El tiempo verbal: Los tiempos simples. In I. Bosque & V. Demonte (Eds.),
Gramática descriptiva de la lengua española, vol. 2, (pp. 2867-2934). Madrid: Espasa Calpe.
Vila, I. (1990). Adquisición y desarrollo del lenguaje. Barcelona: Graó.
von Fintel, K. & Iatridou, S. (2005). Since. Unpublished manuscript, MIT.
137
Spirantization and the structure of Spanish complex onsets: a stratal OT solution
Fernando Martínez-Gil
The Ohio State University
Restrictions on complex (bisegmental) onsets in Spanish present a puzzling asymmetry when we
compare the word level and the phrasal domains. As illustrated in (1) with examples of intervocalic
complex onsets (see page 2; dots = syllable boundaries), in core (lexical) syllabification, complex
onsets are restricted to the combination of an underlying oral stop (voiceless (1a) or voiced (1b)), or /f/
(1c), plus a liquid. As in many (if not most) languages dental stop-lateral sequences are excluded (/dl/
onsets are systematically disallowed; /tl/ is marginal at best, and it is parsed heterosyllabically in most
Spanish dialects: at.las [áD.las]). Any combination of an underlying fricative or affricate plus a liquid
is systematically excluded: *{/s, T, č, ǰ, x/ + /l. r/}. The case of /f/ is special, for typological reasons:
no language apparently exhibits a labiodental stop, thus precluding a potential phonemic contrast with
a labiodental fricative; it follows that the feature [cont(inuant)] is universally non-contrastive for
labiodental consonants. These facts point at a robust generalization, labeled the Complex Onset
Generalization (COG) in Martínez-Gil (2001): in Spanish core syllabification a well-formed complex
onset consists of a liquid preceded by an underlying obstruent that lacks a [+cont] specification
(fricatives, except /f/, are excluded because they are [+cont], and so are affricates, which comprise a
sequentially-ordered [-con, +cont] contour; Steriade 1982, Sagey 1986, Lombardi 1991). In sharp
contrast with the lexical COG, complex onsets consisting of a voiced spirant/approximant followed by
a liquid are pervasive at the phrase level in Spanish, as illustrated in (3), both within words (3a), and
across word boundaries (3b). Responsible for this fact is the well-known postlexical process of
spirantization (SPIR), whereby voiced obstruents are realized as spirants/approximants after continuant
segments (see, e.g., Harris 1969, 1984, Navarro Tomás 1978, Mascaró 1984, Hualde 1991a, 2005,
Quilis 1998). Crucially, although the restriction banning underlying fricatives as the first members of
a complex onset also holds at the phrase level, the voiced spirants/approximants derived by SPIR in this
domain remain tautosyllabic with a following liquid. While such paradoxical state of affairs has been
overlooked in studies of Spanish syllable structure (e.g., Harris 1983, 1989a, 1989b, Hualde 1989,
1991b, Colina 2007), it can be readily handled in a rule-based approach, by simply ordering core
syllabification before SPIR. However, it presents an intractable challenge for standard Optimality
Theory (OT): because constraint evaluation applies exclusively to surface forms, whatever constraints
and constraint ranking disallow tautosyllabic fricative-liquid clusters (thus enforcing the COP), will
also incorrectly rule out the observed spirant/approximant-liquid complex onsets in (3).
This work presents a constraint-based solution to the complex-onset paradox within Stratal OT, a
modular version of OT which distinguishes two ordered domains or strata: the lexical stratum (word
phonology) and the postlexical domain (phrasal phonology) (Kiparsky 2000, 2008, 2009). Each stratum
is an OT grammar with its own constraint ranking, and the output of the lexical stratum constitutes the
input to the phrasal domain. Stratal OT provides a straightforward solution to the complex onset
paradox created by SPIR. In the lexical stratum the constraint(s) that enforce the COG dominate those
that favor SPIR, thus ensuring that underlying voiced stop-liquid clusters are parsed in the onset, while
at the same time excluding their underlying (voiceless) fricative-liquid counterparts. The ranking is
reversed in the postlexical stratum: the constraints that favor SPIR, and well as those that demand
faithfulness to the lexical syllabic parsing, dominate those that enforce the COG, thus ensuring that
complex onsets created in the lexical stratum survive postlexically, even when the first member is
rendered a spirant/approximant by SPIR in the phrasal domain, as in the case in do.[Bl]e, ma.[Dr]e, es.tá.
[Bl]an.do, etc. in (3). Finally, this analysis is shown to provide an insightful account of dialectal data
138
from Chilean Spanish (Martínez-Gil 1997), in which the constraints that compel the COG outrank
faithfulness to lexical parsing, and thus the spirant/approximant first member is parsed as the coda of
the preceding syllable, where it is subject to vocalization.
(1) a. /p, t, k/ + /l, r/: b. /b, d, g/ + /l, r/:
so.plar ‘to blow’ o.pri.mir ‘to press’ do.ble ‘double’; li.bro ‘book’
cua.tro ‘ our’ ma.dre ‘mother’
bu.cle ‘curl’; lu .cro ‘pro it’ si.glo ‘century’; ti.gre ‘ti er’
c. /f/ + /l, r/
a.fli.gir ‘to a lict’; su .frir ‘to su er’
(2) a. do.[Bl]e; li.[Br]o; ma.[Dr]e; si.[ƒl]o; ti.[ƒr]e
b. es.tá. [Bl]an.do ‘it is so t’; u.na. [Br]o.ma ‘a joke’; las.[Dr]o.gas ‘the dru s’; e.se. [ƒl]o.bo ‘that
balloon’ tie.ne.[ƒr]a.cia ‘it is unny’
References
Colina, Sonia. 2007. Spanish Phonology: A Syllabic Perspective. Washington D.C.: Georgetown University Press.
Harris, James W. 1969. Spanish Phonology. Cambridge, Massachusetts: MIT Press.
Harris, James W. 1983. Syllable Structure and Stress in Spanish. Cambridge, Mass.: MIT Press.
Harris, James W. 1984. La espirantización en castellano y la representación fonológica autosegmental. Working Papers in Linguistics 1.149-167. Universitat Autònoma de Barcelona.
Harris, James W. 1989a. Sonority and syllabification in Spanish. Studies in Romance Linguistics, ed. by Carl Kirschner and Janet DeCesaris. Philadelphia: John Benjamins, 139-153.
Harris, James W. 1989b. Our present understanding of syllable structure in Spanish. American Spanish Pronunciation, ed. by Peter C. Bjarkman and Robert M. Hammond. Washington D.C.: Georgetown University Press, 151-169.
Hualde, José I. 1989. Silabeo y estructura morfémica en español. Hispania 72. 821-831.
Hualde, José I. 1991a. Basque Phonology. New York: Routledge.
Hualde, José I. 1991b. On Spanish syllabification. Current Studies in Spanish Linguistics, ed. by Héctor Campos and Fernando Martínez-Gil, 475-493. Washington D.C.: Georgetown Univ. Press.
Hualde, José I. 2005. The Sounds of Spanish. New York: Cambridge University Press. Kiparsky, Paul. 2000. Opacity and cyclicity. The Linguistic Review 17.351-367.
Kiparsky, Paul. 2008. Fenno-Swedish quantity: contrast in Stratal OT. Rules, Constraints, and Phonological Phenomena, ed. by B. Vaux and A. Nevins. Oxford: Oxford Univ. Press, 185-219.
Kiparsky, Paul 2009. Reduplication in Stratal OT. In Reality, Exploration, and Discovery: Pattern Interaction in Language and Life, ed. by Linda Uyechi and Lian-Hee Wee. Stanford, CA: CSLI Publications, 125-141.
Lombardi, Linda. 1990. The nonlinear organization of the affricate. Natural Language and Linguistic Theory 8.375-425.
Martínez-Gil, Fernando. 1997. Obstruent vocalization in Chilean Spanish: a serial versus a constraintbased approach. Probus 9.167-202.
Martínez-Gil, Fernando. 2001. Sonority as a primitive phonological feature: evidence from Spanish complex onset phonotactics. Feature and Interfaces in Romance, ed. by Julia Herschenson, Enrique Mallén and Karen Zagona. Philadelphia: John Benjamins, 203-222.
Mascaró, Joan. 1984. Continuant spreading in Basque, Catalan and Spanish. Language sound structure, ed. by Mark Aronoff and Richard T. Oehrle. Cambridge, Mass.: MIT Press, 287–298.
139
Navarro Tomás, Tomás. 1977. Manual de pronunciación española. Madrid: C.S.I.C.
Quilis, Antonio. 1999. Tratado de fonología y fonética españolas (2ª ed.). Madrid: Gredos.
Sagey, Elisabeth. 1986. The Representation of Features and Relations in Non-Linear Phonology. Doctoral dissertation, MIT, Cambridge, Mass.
Steriade, Donca. 1982. Greek Prosodies and the Nature of Syllabification. Doctoral dissertation, MIT, Cambridge, Mass.
140
Syntax as a Language Cue
Crystal Marull
Rutgers University
How a bilingual stores and activates lexical representations in their two languages has been a central
interest in psycholinguistic studies of bilingualism. Two competing theories of bilingual lexical access
have been put forth – the language-selective access theory and the language nonselective access theory.
The former proposes that only items from the intended language are activated and available for
selection during speech planning and comprehension (e.g., Costa, Santesteban, & Ivanova, 2006). The
latter proposes that both languages receive activation and compete for production (e.g., Kroll, Bobb, &
Wodniecka, 2006). Within the framework of the non-selective access theory, the ability to negotiate
cross-linguistic competition may come from language cues that reliably signal the L1 or L2 in
conjunction with an inhibitory control mechanism (e.g. Green, 1998). This study investigates whether
language-specific syntax – the syntactic positioning of a target word in a determiner phrase (DP) – can
act as a language cue to modulate cross-linguistic activation in bilingual sentence comprehension and if
competition for production is mediated by an inhibitory control mechanism.
To examine cross-linguistic activation twenty-three Spanish native speakers, with advanced
English proficiency living in Argentina, completed a cross-modal priming paradigm followed by a
lexical-decision task (figure 1) using SuperLab software (Cedrus Corp.; Haxby, Parasuraman, Lalonde
and Abboud, 1993). The stimuli were comprised of English sentences containin D ’s that received
genitive or dative case. In English, these structures have two acceptable positions (see examples); one
which is linearly congruent with Spanish syntax (canonical dative construction and ofgenitive) and one
which is linearly incongruent (dative double object construction and‘s-genitive). In task 1, after reading
each stimulus sentence, the participants were presented with a word that was either the Spanish
translation-equivalent of the target word in the DP or a control word. The participants pressed a “yes”
or “no” button to indicate i the word was a real Spanish word. The response times were recorded in
milliseconds rom the onset o the lexical decision task stimulus to the button press. Occasional “yes-
no” comprehension questions were included to control or participants’ attention. n task 2, the
participants performed the same activity as in task 1, except they repeated the stimulus out loud prior to
the lexical decision task. Again, response times were recorded in milliseconds.
The results of task 1(fig. 2) revealed that response times in the congruent condition (canonical
dative construction and of-genitive) were significantly faster (p = .001*) than response times in the
incongruent conditions (dative double object construction and ‘s-genitive) providing evidence that
linearly congruent syntax results in cross-linguistic activation and that language-specific syntax
(linearly incongruent) can reduce such activation. The results of Task 2 also showed that the
Congruent condition was significantly faster than the Incongruent condition (p = .02*). However, in
comparison to Task 1, the difference between the two conditions had decreased indicating that the
highly activated translation-equivalent competitor had been inhibited to allow for target production.
Overcoming such inhibition to complete the lexical decision task incurred a cost in the response time,
reducing the advantage of the Congruent condition. This finding is in line with the asymmetrical
language-switch costs reported by Meuter and Allport (1999).
The results from this study support the language non-selective access theory and provide
evidence that language-specific syntax can be a language cue that modulates cross-linguistic activation.
Furthermore, they suggest that remaining cross-linguistic competition may be resolved by an inhibitory
control mechanism in speech production. Such results refine current theories of crosslinguistic
activation and provide a better understanding of the role of inhibition in bilingual language
comprehension and production.
141
Examples of stimuli sentences:
1) Congruent Dative (canonical dative construction):
The man gave some water to the donkey in the lot. (target: burro) 2) Incongruent Dative (dative double object construction):
The worker in the street gave the dog a little space. (target: perro) 3)
Congruent Genitive (of-genitive):
The teacher drew the top of the tree with a marker. (target: árbol) 4)
Incongruent Genitive (‘s-genitive):
The young girl wanted to paint the chair’s bottom bright green. (target: silla)
Fig. 1: Example of cross-modal paradigm with a lexical-decision
142
References
Costa, M., Santesteban, M., & Ivanova, I. (2006). How do highly proficient bilinguals control their
lexicalization process? Inhibitory and language-specific selection mechanisms are both functional.
Journal of Experimental Psychology: Learning, Memory, and Cognition, 32(5), 1057-1074.
Green, D. (1998). Mental control of the bilingual lexico-semantic system. Bilingualism: Language and
Cognition, 1, 67-81.
Kroll, J. F., Bobb, S. C., & Wodniecka, Z. (2006). Language selectivity is the exception not the rule:
Arguments against a fixed locus of language selection in bilingual speech. Bilingualism: Language and
Cognition, 9, 119–135.
Meuter, R. F., & Allport, A. (1999). Bilingual language switching in naming: Asymmetrical costs of
language selection. Journal of Memory and Language, 40, 25-40.
Haxby J.V., Parasuraman R., Lalonde F., Abboud H. (1993). SuperLab: General-purpose Macintosh
software for human experimental psychology and psychological testing. Behavior Research Methods,
Instruments, & Computers, 25, 400–405.
143
Sequence of tenses in complementation structures: lexical restrictions and effects on language
acquisition Rui Marques
University of Lisboa Purificação Silvano University of Porto Anabela Gonçalves University of Lisboa
Ana Lúcia Santos University of Lisboa
A large body of literature has been devoted to the study of the temporal relations between main and
complement verbal clauses, both within the framework of Formal Semantics (cf. Abusch 1997, Ogihara
1996, a.o.) as within the framework of Generative Syntax (cf. Meireles & Raposo 1983, Kempchimsky
1986, Ambar 1992, a.o.). Formal Semantics has been focused on deriving the correct interpretation for
embedded tenses, while Generative Syntax has paid a lot of attention to the restrictions on sequence of
tenses imposed by some verbs, but not others (cf. (1) and (2)). To explain these restrictions on sequence
of tenses, the concept of temporal dependence (TD) has been critical. The basic idea is that TD is a
lexical property, being imposed by some verbs of sentential complementation that require the presence
of an anaphoric tense in their complement, the infinitive and the subjunctive being anaphoric
morphemes, contrary to the indicative, which would be specified for tense features.
The relevance of TD in infinitival complement clauses has been questioned by Duarte, Gonçalves &
Santos (2012), who propose that the relevant lexical property is, instead, temporal orientation (i.e.,
whether the main verb specifies the temporal location of the embedded proposition as being in a
relation of anteriority, overlapping or posteriority with the situation described by the main clause).
In this paper, we will focus on finite complement clauses, with two purposes in mind: (i) to explain the
restrictions on possible sequence of tenses (SOT) in complement clauses; (ii) to observe (though with
preliminary data) the acquisition of SOT in relation to the acquisition of subordination structures.
We observe that, even though the temporal orientation imposed by some verbs is, obviously, crucial
for SOT, it is not the only issue that restricts the tenses that might occur in embedded clauses. Thus,
tough the contrast between (1) and (2) may be accounted for by the fact that querer (‘want’) is a
prospective verb, imposing a future reading on the complement clause, contrary to pensar (‘think’), the
issue of temporal orientation does not explain the ungrammaticality of PRES in (3), where the main
verb does not impose any temporal orientation on the complement. Moreover, the classical idea that the
subjunctive, like the in initive (and contrary to the indicative), is anaphoric doesn’t seem tenable, the
same restrictions on SOT being observed with subjunctive and indicative complements (the equivalents
of, e.g., think, an indicative ruler, and ask¸ subjunctive ruler, do not impose restrictions on the tense of
the complement, contrary to, e.g., want, doubt, subjunctive rulers, or decide, suppose, indicative rulers;
cf. (1)-(6)).
Our proposal is that both the subjunctive and the indicative morphemes express temporal reference on
their own and that the complement clause may exhibit a tense that creates a new temporal domain only
if the denotation of the complement clause necessarily includes the real world, otherwise, the sequence
PAST + PRES being ruled out.
After explaining our proposal for the lexical restrictions on SOT in verbal complement clauses, we
will present preliminary data on the acquisition of embedded tenses in (European) Portuguese, on the
144
basis of a corpus of spontaneous child language production (3 monolingual children). The observed
data suggests that the first child productions of embedded clauses involve deictic tenses, and only in
later stages children produce constructions where the tense of the embedded clause has the main clause
fixing its temporal perspective point. Such data suggests that children begin by interpreting the main
and embedded clauses independently. In the line suggested by Hollebrandse et al. 2001, it seems
reasonable to assume that the acquisition of complementation and the capability of using tenses that
take as their temporal perspective point the situation defined by the main clause are part of the same
process. Thus, it is expectable that the absence of acquisition of complementation structures forces the
preference for using speech time as the only temporal perspective point, as the observed data suggests.
Examples:
(1) ele pensa-PRES que a Ana está-PRES-IND / estava-PAST-IMPERF-IND / esteve-PAST-PERF-IND / estará-FUT-IND em casa
he thinks that the Ana is was- PAST-IMPERF-IND was-PAST-PERF-IND be-FUT-IND at home (2) ele quer-
PRES que a Ana esteja-PRES-SUBJ / *estivesse-PAST-IMPERF-SUBJ / *tenha estado-PAST-PERF-IND em casa he wants
that the Ana be--PRES-SUBJ be-PAST-IMPERF-SUBJ be-PAST-PERF-IND at home
(3) ele supôs-PAST que a Ana *está-PRES-IND no comboio he
supposed that the Ana is in.the train
(4) ele decidiu que *foi-PAST-PERF-IND / ia-PAST-IMPERF-IND / vai-PRES-IND
/ irá-FUT-IND ao mercado he decided that go -PAST-PERF-IND
go-PAST-IMPERF-IND go-PRES-IND go-FUT-IND to.the market (5) ele
duvidou que a Ana *esteja-PRES-SUBJ / estivesse-PAST-IMPERF-SUBJ no
comboio he doubted that the Ana be-PRES-SUBJ be-PAST-IMPERF-
SUBJ in.the train (6) ele pediu que tu lhe telefones-PRES-SUBJ
/ telefonasses-PAST-IMPERF-SUBJ
he asked that you him call -PRES-SUBJ call-PAST-IMPERF-SUBJ
References
busch, Dorit 1997, “Sequence o tense and temporal de re”, Linguistics & Philosophy 20, 1–50.
mbar, Manuela 1992 “Temps et structure de la phrase en portu ais”, in Hans Obenhauer & Anne
Zribi-Hertz (eds.) Structure de la phrase et Théorie du Liage, Saint-Denis: Presses Universitaires de
Vincennes, 29-49.
Duarte, I., A. Gonçalves & A. L. Santos (2012) Control, agreement and temporal (in)dependence: an
argument for a theory of Control as Agree. 38th
Incontro di Grammatica Generativa. Verona, Italy,
23-25 Fevereiro.
Hollebrandse, Bart, Denis Del iito, n eliek van Hout & ndréa de Vroe (2001) “ talian sequence o
tense Complementation or imper ectivity?”, in Anna H.-J. Do, Laura Domínguez & Aimee
Johansen (eds.), Proceedings of the 25th Annual Boston University Conference on Language
Development, Somerville: Cascadilla Press, 343-352.
Kempchinsky, aula 1990, “Más sobre el e ecto de re erencia disjunta del subjuntivo”, in Ignacio
Bosque (ed.) Indicativo y subjuntivo, Madrid: Ediciones Taurus, 234-258.
Meireles, José & Eduardo Raposo 1983, “Subjunctives and disjoint re erence in ortu uese Some
implications or the Bindin Theory”, communication presented to the 1er
Colloqui Internacional de
Lingüistica Teòrica i Llengues Romàniques, Sitges.
Ogihara, Toshiyuki: 1996, Tense, Attitudes, and Scope, Kluwer.
Raposo, Eduardo (1987) Case Theory and Infl-to-Comp: The inflected inftinitive in European
Portuguese. Linguistic Inquiry 18 (1), pp. 85-109.
145
Focus on the dialects: The syntactic variation of focus marking in Caribbean Spanish
Catalina Méndez Vallejo
College of William and Mary
Manuel Delicado Cantero
The Australian National University
Maria Carmen Parafita Couto
Leiden University
In this paper we study focus marking in colloquial Caribbean Spanish with focus adverbs and the
focus verb ser (‘to be’). Unlike other varieties o Spanish, which typically mark ocus throu h
intonation patterns (c . Zubizarreta’s 1998 Focus-Prosody Correspondence Principle), several
Caribbean varieties of Spanish mark focus with ser (1) (Bosque 1999; Sedano 2003; Toribio 2002;
Bosque 1999; Curnow & Travis 2004; Camacho 2006; author 2009).
The phenomenon illustrated in (1), usually referred to as the Focalizing Ser construction (FS),
has been thoroughly analyzed in Colombia and Venezuela (e.g. Sedano 2003; Curnow & Travis 2004;
author 2009). While this construction is typical of some Caribbean varieties (i.e. Colombian,
Venezuelan, Panamanian, Dominican), to our knowledge, no previous study has delved into the micro-
differences between the dialects in the region.
n order to obtain data about speakers’ perception o FS, we distributed an anonymous online
survey to both Venezuelan and Panamanian speakers, and a written questionnaire to Dominican
speakers. In the online survey and in the written questionnaire, participants were asked to evaluate
specific sentences containing either FS or the more standard pseudo-cleft form (2). These sentences
were presented in short mini-dialogues that provided a context and participants rated the sentences
based on how (un)familiar they seemed to them (3).
The results of these acceptability judgment tests show that there is some clear variation among
these dialects. For instance, Panamanian and Dominican speakers tend to accept FSfocused embedded
CPs (4), and FS-focused direct object DPs (5) at a much higher rate than Venezuelan speakers.
Furthermore, based on the results of a similar test conducted among Colombian speakers (author 2009),
it becomes evident that in Colombia FS-focused subject DPs (6) are more commonly accepted than in
Venezuela and in the Dominican Republic.
Drawing from the data obtained, we show that the use of focus ser may be more or less
restricted depending on the type of constituent being focused (e.g. subject DPs, object DPs, etc.) and
the variety of Spanish observed.
This paper contributes to the understanding of the typology of Focus and dialect syntax,
particularly necessary for pluri-centric languages such as Spanish.
Examples
(1) a. Mi mamá no toma agua. Toma es pura gaseosa My mom
not take.3sg.pres water. Take.3sg.pres be.3sg.pres pure soda
‘My mom doesn’t drink water. She drinks only soda’ (Curlow & Travis 2003 8)
146
b. Hay es solo cuatro empleadas Have.3sg.pres be.3sg.pres only four
employees
‘ t is only our employees that there are’ (author 2009: 102)
c. *Solo hay es cuatro empleadas
only have.3sg.pres be.3sg.pres four employees
‘ t is our employees that there are only’ (author 2009: 101)
d. *Hay solo es cuatro empleadas
have.3g.pres only be.3sg.pres four employees
‘ t is our employees that there are only’ (author 2009: 102)
(2) a. Había era una cama vieja
Have.3sg.imp be.3sg.imp a bed old
‘ t was an old bed that they had’
b. Lo que había era una cama vieja
rel pron have.3sg.imp be.3sg.imp a bed old
‘What they had was an old bed’
(3) Juan: ¿En el patio no había una mesa de plástico?
‘Was there not a plastic table in the backyard?’
María: No, allá había era una cama vieja
‘No, it was an old bed that they had there’
(4) Empezaron fue a tirar piedras por todas partes
‘ t was throwin stones everywhere that they started to do’
(5) Tenemos es dos gatos y un perro
‘ t is two cats and one do that we have’
(6) Salió fue Lucía
‘ t was Lucía who le t’
References
Author 2009. Reference anonymized.
Bosque, I. 1999. On focus vs. wh-movement: The case of Caribbean Spanish. Sophia Linguistica 44-
45: 1-32.
Camacho, J. 2006. In situ focus in Caribbean Spanish: Toward a unified account of focus. Proceedings
of the 9th
Hispanic Linguistics Symposium. Somerville, MA: Cascadilla Press.
Curnow, J. & C. Travis. 2004. The emphatic es construction of Colombian Spanish. In: Moskovsky,
Christo (ed.): Selected Papers from the 2003 Conference of the Australian Linguistic Society: 1-11.
Sedano, M. 2003. Más sobre seudohendidas y construcciones con verbo ser focalizador en el habla de
Caracas. In Lengua, variación y contexto: Estudios dedicados a Humberto López Morales. Madrid:
Arco/Libros. 823-847.
Toribio, A. J. 2002. Focus on clefts in Dominican Spanish. In K. Lee and C. Clements (eds.): Structure,
meaning, and acquisition of Spanish. Somerville, MA: Cascadilla Press. 130-146.
Zubizarreta, M.L. 1998. Prosody, Focus, and Word Order. Cambridge, MA: The MIT Press.
147
TP-internal focus: a view from two varieties of Spanish
Catalina Méndez Vallejo
Jennifer Giuffrida
College of William and Mary
Similar to other languages, Spanish may have topic and focus projections inside CP (e.g. Casielles-
Suárez 2004). Aside from left dislocation (1), clefting (2) is commonly used in Spanish to focalize
arguments within CP. Thus, the affirmative sentence in (3) can be transformed into a cleft sentence as
in (4), where the direct object is the focalized element of the sentence, is ccommanded by the copula
verb fue 'was' (> ser ‘to be’), and seems to receive a narrow ocus.
In some Spanish varieties (i.e. Venezuelan, Ecuadorian, Panamanian, Colombian, and Dominican), an
apparently similar sentence to the one in (4) can also be created without the REL PRON and the
COMP que (‘that’), with a structure o ten called focalizing ser (FS), as shown in (5). FS has been
previously examined as an incomplete form of a cleft structure (e.g. Sedano, 1990; Toribio, 2002), but
later research shows that FS is syntactically unrelated to clefts (e.g. Bosque, 1999; Curnow & Travis,
2004; Méndez Vallejo, 2009). In fact, while clefts occur in the external periphery of the clause (at the
CP-level), FS is a clause-internal phenomenon. Given the focusproperties of FS, this clearly shows that
the interface with the informational structure cannot be confined to the highest periphery (in line with
Belleti 2004).
In the present analysis, we show data collected within two distinct communities of practice (cf. Eckert
& McConnell-Ginet, 1992), from two Dominican cities (Santiago de los Caballeros and Santo
Domingo). A total of 75 participants evaluated 93 sentences containing both FS and cleft structures.
These sentences were part of pre-recorded mini-dialogues (6) and participants were asked to evaluate
the last sentence they heard based on how (un)familiar it sounded to them. The results of these
acceptability judgment tests are compared with the results from a similar test conducted in Colombia in
2008 (Méndez Vallejo, 2009).
Based on the results collected, we ar ue that there is clear variation in the way speakers’ perceive the
use of FS. First, Dominican speakers tend to evaluate certain cases of FS-focused subject and object
DPs (7) at a much higher rate than Colombian speakers. Second, Colombians find some FS-focused
impersonal constructions (8) and FS-focused embedded CPs more acceptable than Dominicans. Third,
while Dominicans highly accept FS in sentence-final position, ser must occur in the preterit form (9).
Fourth, although FS-focused subject DPs (10) have been reported as ungrammatical in Dominican
Spanish (Toribio, 1992), our data show that Dominican speakers accept them as much as Colombian
speakers.
Our study shows that the use of FS is much more accepted by Dominican speakers than we had
initially anticipated. Furthermore, it is evident that speakers’ perceptions o the use o FS vary
depending on the type of constituent being focused and the variety of Spanish observed. Despite the
dialectal differences observed, the data collected provide further support to the claim that FS is a TP-
internal focus structure, generated between T and vP (11).
(1) A: ¿Quién trajo el vino? B: [El vino-TOP [lo trajo Julián]]
‘Who brou ht THE W NE?’ ‘THE W NE, Julian brou ht it’
(2) A: ¿QUIÉN trajo el vino? B: [JULIÁN-FOC fue quien [lo trajo]]
‘WHO brou ht the wine?’ ‘JUL N was the one who brou ht it
148
(3) Julián trajo VINO
Juan brou ht some wine’
(4) Lo que Julián trajo fue VINO REL PRON that Julián bring-3SG-
PAST to be-3SG-PAST wine ‘What Julián brou ht was wine’
(5) Julián trajo fue VINO
Julián bring-3SG-PAST to be-3SG-PAST WINE
‘ t was wine what Julián brou ht’
(6) Juan: ¿En el patio no había una mesa de plástico?
‘Was there not a plastic table in the backyard?’
María: No, allá había era una cama vieja
‘No, it was an old bed that they had there’
(7) Tendrá dinero será la familia
‘ t is his amily who may have some money’
(8) Hay es un tipo en el jardín
‘ t is some uy who is in the arden’
(9) Llovió en la sierra fue
‘ t was in the mountain that it rained’
(10) Salió fue Lucía
‘ t was Lucía who le t’
(11) [TP [T' ([TopP [Top°]) [FocP [Foc°] … VP […] ]]]]]]
References
BELLETTI, A. 2004. Aspects of the low IP area. In: The structure of IP and CP. The Cartography of
Syntactic Structures. Ed: Rizzi, L. 16-51. New York: Oxford University Press.
BOSQUE, I. 1999. On focus vs. wh-movement: The case of Caribbean Spanish. Sophia Linguistica 44-
45: 1-32.
CASIELLES-SUÁREZ, E. 2004. The Syntax-Information structure interface. Evidence from Spanish
and English. New York, Routledge.
CURNOW, J. & C. TRAVIS. 2004. The emphatic es construction of Colombian Spanish. In:
Moskovsky, Christo (ed.): Selected Papers from the 2003 Conference of the Australian Linguistic
Society: 1-11.
ECKERT, P. & MCCONNELL-GINET, S., 1992, Communities of practice: where language, gender,
and power all live, En Kira Hall, et al. (eds.) Locating Power: Proceedings of the Second Berkeley
Women and Language Conference, 89-99, Berkeley, CA: Women and Language Group.
MÉNDEZ VALLEJO, D.C. 2009. Focalizing ser (‘to be’) in Colombian Spanish. Ph.D. dissertation,
Indiana University.
SEDANO, M. 1990. Hendidas y otras construcciones con ser en el habla de Caracas. Caracas: UCV.
TORIBIO, A.J. 2002. Focus on clefts in Dominican Spanish. In: J. F. Lee, K. L. Geeslin & J. C.
Clements (eds), Structure, meaning, and acquisition in Spanish. Somerville: Cascadilla Press, 130-
146.
TORIBIO, A.J. 1992. Proper government in Spanish subject relativization. Probus 4: 291-304.
149
What is the initial state in child L2 acquisition?
Silvina Montrul
Alexandra Morales
Tania Ionin
Melissa Bowles
Andrei Cimpian
University of Illinois at Urbana Champaign
How different sources of knowledge drive initial acquisition and development of different sub
modules of grammar (phonology, morphology, syntax and semantics) (Epstein, Flynn &
Martohardjono 1996; Schwartz & Sprouse 1996; Vainnikka & Young-Scholten 1996) has been the
focus of much research on the initial state in adult L2 acquisition. The vast majority of empirical
evidence shows that the L1 plays a significant role at early stages of development in L2 adults.
Yet very little is known about how children start their lexical and grammatical development in a
second language, how they go on to acquire different structural properties, and how their age at onset
of L2 acquisition impacts their learning. L2-acquiring children already have knowledge of an L1, like
L2 adults; but unlike adults they may deploy more implicit language learning mechanisms operative at
their younger age. An issue that is not clear is whether transfer from the L1 is as evident and strong in
children as it appears to be in adult L2 acquisition, or whether the extent of L1 transfer depends on the
age and cognitive development of the child (Gavruseva & Haznedar 2008, Ionin 2008, Schwartz 2004).
As it turns out, questions specifically related to the initial state in child L2 acquisition have not really
been addressed. The main objective of this project is to document the initial state of L2 acquisition of
Spanish morphosyntax and semantics and the nature of transfer from the L1 in 5 to 8 year-old children
whose primary language is English and are enrolled in a foreign language program at school. We
followed the emergence and early development of basic structures as the children interact with the
teachers and peers in class activities. We investigate whether the L1 plays a significant role at earliest
stages of development as in adult L2 acquisition, or whether L2 children follow the route of
development of L1 Spanish-speaking children already documented in the literature. Our study tested
grammatical features that are present in both English and Spanish (i.e., plural agreement), but the main
focus was on structures that present differences in Spanish and English and are ideal to test language
transfer. These included gender agreement in the noun phrase and interpretation of determiners (in
particular the definite determiner).
Fourteen children ages 5-8 enrolled in first year Spanish classes at school (Grades K and 1)
participated in the study. To assess receptive vocabulary, we administered the standardized Peabody
Picture Vocabulary Test (PPVT) in English and its Spanish version (TVIP) one month after beginning
of instruction and six months later, when the experiments were conducted. Three comprehension
experiments were designed to test number and gender in nouns and the interpretation of determiners. In
the plural experiment, children were presented with an aural sentence (Pablo tiene perro or Pablo tiene
perros) and two pictures (one plural, one singular) and had to point to the picture that matched the
sentence. In the gender task, the children were shown three objects. Two objects shared the same color
but differed in gender. The third one was plural. The experimenter asked questions with noun drop,
such as Muéstrame/ Señala la/el rojo/a. “Show me/ oint to the red one.” the children know the
gender of the target noun they would pick up the object with the correct gender. To test the generic
interpretation of definite articles we adapted the task from Cimpian, Meltzer and Markman (2011) to
Spanish. The children were shown pictures of atypical animals or objects (such as blue elephants) and
were asked questions such as ¿Los elefantes son azules? (Are elephants blue?), which in Spanish can
have a generic meaning. If they say No, it means children interpret the definite article as generic in
Spanish, like most monolingually raised Spanish speakers. The control questions had demonstratives
¿Estos elefantes son azules? (Are these elephants blue?), and the expected response was yes. The three
150
tasks included a practice sessiosn, 4-6 target items, and distractors. The tasks were administered to 10
age-matched monolingual children from Puerto Rico to make sure they were age-appropriate and
reliable.
Our preliminary findings show that the children performed well above chance with the
comprehension of plural marking on nouns (83% singular, 76% plural) and half of the children
achieved 100% accuracy. By contrast, the group results of the gender experiment and of the generic
interpretation of determiners was close to chance: 52% masculine, 58% feminine; 43% generic, 37%
demonstrative. However, there were a few children who performed above chance. Our study suggest
that at this early stage of development, children transfer their knowledge of plural in English but have
not yet acquired specific properties of Spanish, like gender or the generic interpretation of definite
determiners.
Experiment 1: Plural in nouns Experiment 2: Gender in nouns
Juan tiene perro/Juan tiene perros
Muéstrame la roja.
Muéstrame el negro.
Experiment 3: Determiners and genericity
Los elefantes son azules.
Estos elefantes son azules.
References
Cimpian, ., Meltzer, T. & Markman, E. (2011). reschoolers’ use o morphosyntactic cues to identi y
generic sentences: indefinite singular noun phrases, tense and aspect. Child Development 82, 5,
1561-1578.
Epstein, S., Flynn, S., & Martohardjono, G. (1996). Second language acquisition: Theoretical and
experimental issues in contemporary research. Behavioral and Brain Sciences, 19, 677-714.
Haznedar, B. and E. Gavruseva (Eds.) (2008). Current Trends in Child Second Language Acquisition:
A Generative Perspective. Amsterdam: Benjamins.
151
Ionin, T. (2008). Progressive aspect in child L2-English. In B. Haznedar and E. Gavruseva (Eds.),
Current Trends in Child Second Language Acquisition: A Generative Perspective. Amsterdam:
Benjamins, 17-53.
Schwartz, B. (2004). On child L2 development of syntax and morphology. Lingue e Linguaggio 3: 97-
132.
Schwartz, B., & Sprouse, R. (1996). L2 cognitive states and the Full Transfer/Full Access model.
Second Language Research, 12, 40-72.
Vainikka, A. and Young-Scholten, M. (1996). The early stages of adult L2 syntax: additional evidence
from Romance speakers. Second Language Research, 12, 140-176.
152
Spanish rhotics acquisition by child L2 learners of Spanish
Alexandra Morales
UIUC
One of the major challenges that L2 learners face in acquiring the target language is to master its
sound system. In adult L2 learners, although there may be exceptions, foreign accent persists despite
being exposed to the target language for a long period of time. By contrast, children are found to be
more successful in achieving a native-like accent and to do it with relatively ease (Long 1993). For L2
learners of Spanish, two sounds that are highly problematic are the rhotics: the tap [ɾ] and the trill [r].
English-speaking adults learning Spanish as their L2 show difficulties with these sounds and tend to
transfer the English [ɹ] in contexts where these are required (Face 2006; Olsen 2012). Nevertheless,
learners’ pronunciation can improve with reater exposure to the lan ua e. dvanced learners show
high accuracy in producing the tap and although they are not highly successful with the trill, they more
commonly substitute this sound with the Spanish tap, which indicates that they are no longer relying on
their L1 to produce the sound. These findings suggest that L2 learners initially rely on their L1 to
produce the Spanish rhotics. However, how children learning Spanish as L2 would produce these
sounds is practically known.
The objective of this study is to evaluate the production of the rhotics in child L2 learners of
Spanish. In particular, we want to analyze whether child L2 learners show developmental patterns
similar to those observed in child L1 acquisition or whether instead their performance more closely
resembles that exhibited in adult L2 acquisition and they show transfer from their L1. In order to do
this, and as part of a more comprehensive study about Spanish L2 acquisition by children, a small
sample of children learning Spanish as their second language (N = 5) was tested. They completed a
picture-naming task. In this task they were asked to name objects with names that had either a tap or a
trill sound. The children’s a es ran e rom 5 to7 years old. They are native or dominant speakers o
English and are all learning Spanish at a school in the United States. They are exposed to Spanish one
or two hours, four days a week.
The results revealed that although the children could accurately produce both the tap and the trill,
they also showed developmental patterns found in child L1 acquisition. For example, the children
substituted both the tap and the trill for [l] or they omitted the sound. Both of strategies have been
reported for L1 acquisition (Anderson 1987; Goldstein & Iglesias 1996) but have been rarely found in
adult L2 acquisition (Face 2006). However, there were a few cases where the children used the English
[ɹ]. Thus, the indin s su est that, althou h the child’s L1 may have an e ect, child L2 learners’
strategies for learning the Spanish rhotics parallel those also observed in L1 acquisition. The results
also revealed that child L2 learners can be highly accurate producing the sounds of the L2 in a
relatively short period of time, but also showed that there may be a lot of individual variation. Taking
into account the results, we concluded that phonology acquisition in child L2 learners although similar
it does not completely mirror child L1 acquisition. These findings support the proposal that in child L2
phonolo y acquisition there is an interaction between the child’s L1 and developmental strate ies
found in L1 acquisition (Anderson 2004; Ball, Müller & Munro 2001; Hecht and Mulford 1982).
References
Anderson, R. T. (1987). Phonological development of two-year-old monolingual Puerto Rican
Spanish-speaking children. Journal of Child Language 14, 57-78.
Anderson, R. T. (2004). Phonological acquisition in preschoolers learning a second language via
immersion: a longitudinal study. Clinical Linguistics & Phonetics 18, 183-210.
Ball M.J., Müller N. & Munro S. (2001) The acquisition of the rhotic consonants by Welsh-English
bilingual children. International Journal of Bilingualism 71 , 71-86.
153
Face T. L. (2006). Intervocalic rhotic pronounciation by adult learners of Spanish as a second
language. In C. A. Klee & T.L. Face (Eds.) Selected Proceedings of the 7th Conference on the
Acquisition of Spanish and Portuguese as First and Second Language. Somerville, MA: Cascadilla
Proceedings Project.
Goldstein, B.A. & Iglesias, A. (1996). Phonological Patterns in Normally Developing Spanish
Speaking 3- and 4-Year-Olds Puerto Rican Descendents. Language, Speech, and Hearing Services in
School 27 , 82-89.
Hecht & Mulford.(1982). The acquisition of a second language phonology: interaction of transfer and
developmental factors. Applied Psycholinguistics 3, 313-328.
Long, M. H. (1993). Second language acquisition as a function of age: research findings and
methodological issues. In K. Hyltenstam and A. Viberg (Eds), Progression and Regression in
Language: Sociocultural, Neuropsychological, and Linguistic Perspectives Cambridge: Cambridge
University Press.
Olsen M. K. (2012) The L2 acquisition of Spanish rhotics by L1 English speakers: The effect of L1
articulatory routines and phonetic context for allophonic variation. Hispania 95, 65-82.
154
Linguistic Interference?: Overt Subject Realization by Spanish-English Bilinguals in Spanish
Catherine Nussbaum
John Grinstead
The Ohio State University
In their model of crosslinguistic interference for bilingual language development, Hulk & Müller
(2000) propose that constructions are likely to undergo interference if they occur at the discourse-
syntax interface and if the constructions under consideration are superficially overlapping between the
two languages. For example, overt subjects in southern Romance languages, such as Italian and
Spanish, have been shown to be susceptible to this type of interference from overt subject languages
such as English (Paradis & Navarro 2003; Serratrice, Sorace & Paoli 2004). In particular, a Spanish-
English bilingual child uses more overt subjects in her Spanish than do children in a monolingual
control group. The same appears to be true for Italian-English bilingual children. While Paradis &
Navarro (2003) provide evidence of interference from English in the subject use of the Spanish-
speaking bilingual child they studied, it is not clear whether the higher proportions of overt subjects in
the bilin ual child’s Spanish could be due to the in luence o En lish or to the act that the Caribbean
Spanish she was exposed to uses higher percentages of overt subjects than does the Peninsular Spanish
of the comparison children. In the current case study we address the same question of whether English
interferes with Spanish in subject position, however, both the bilingual children and the children of the
monolingual comparison group, whose language development we consider, are speakers of the same
(Peninsular) variety of Spanish. In particular, we analyze the spontaneous speech of two Spanish
monolinguals (Irene from the Llinàs Grau-Ojea corpus and María from the López-Ornat corpus –
López-Ornat 1994), and two Spanish-English bilinguals (from the Liceras et al 2008 corpus), obtained
throu h the CH LDES database (MacWhinney 2000). The monolin ual and bilin ual children’s
spontaneous production data was divided into two MLUw-matched stages: MLUw Stage 1 (1.5-2.0)
and MLUw Stage 2 (2.0-3.0). Across the four participants, a total of 2,687 verbal utterances were
analyzed, 707 of which were realized with overt subjects. The data were analyzed for
presence/absence of overt subjects, type of overt subject (personal pronouns vs. lexical DPs) and
pragmatic function of the overt subject. Analyses of variance, covariance and correlations revealed that
bilinguals and monolinguals use overt subjects differently. Bilingual and monolingual children at this
early stage of development did not differ in their overall rate of overt subject use in Spanish, but did
differ in the types of overt subjects they used: bilinguals using significantly more personal pronouns
(p=.027 ) and monolinguals using significantly more lexical DPs (p=.034). We take this to be evidence
o trans er in the children’s syntax in the sense that bilin ual children appeared to use the En lish
strategy of using overt pronominals instead of the Spanish option, which allows either overt or null
pronominals. Further, as bilin uals’ Spanish develops, they use more overt subjects that serve no clear
pragmatic purpose, when compared to monolinguals (p=.002), suggesting transfer on the pragmatic
level. Our results support the conclusion that Spanish-English bilingual children undergo
crosslinguistic transfer from English to Spanish in their realization of overt subjects, consistent with
Hulk & M ller’s crosslin uistic trans er account.
References
Hulk, A., & Muller, N. (2000). Bilingual First Language Acquisition at the Interface between Syntax
and Pragmatics. Bilingualism: Language and Cognition, 3(3), 227-244.
López Ornat, S. (1994). La adquisición de la lengua española. [The acquisition of Spanish.]. Madrid,
Spain: Siglo XXI.
MacWhinney, B. (2000). The CHILDES Project. Mahwah, New Jersey: Lawrence Erlbaum.
155
Paradis, J., & Navarro, S. (2003). Subject Realization and Crosslinguistic Interference in the Bilingual
Acquisition of Spanish and English: What Is the Role of the Input? Journal of Child Language, 30(2),
371-393.
Serratrice, L., Sorace, A., & Paoli, S. (2004). Crosslinguistic Influence at the Syntax-Pragmatics
Interface: Subjects and Objects in English-Italian Bilingual and Monolingual Acquisition.
Bilingualism: Language and Cognition, 7(3), 183-205.
156
Examining the Input in Heritage Speaker Bilingualism: The Case of Gustar-like Verbs
Diego Pascual y Cabo
Texas Tech University
Herita e Speaker (HS) bilin ual competence di ers rom their monolin ual counterparts’ in
ways not always predictable and/or well-understood. Left to be determined is whether attrition
(Polinksy 2011), incomplete acquisition (Montrul 2008), and/or input delimited differences (Pires and
Rothman 2009) explain the documented HS linguistic outcomes. To differentiate among these sources,
this study examines the syntax of reverse psych-predicates (gustar “to like”) in HS Spanish. Thou h
highly frequent, gustar-like verbs exhibit a somewhat exceptional word order in Spanish. Moreover,
they are stative verbs and differ from regular agentive verbs in that they subcategorize for a postverbal
<theme> that controls verb agreement and a preverbal <experiencer> (the logical subject), which is
obli atorily doubled by a clitic and preceded by the dative marker ‘a’ (1a&b). These ar uments,
however, can be fronted for discourse-pragmatic reasons (1c&d). Not surprisingly, this structural
opacity poses an acquisitional problem not only for L2ers (Montrul 2000; White et al 1999) but also for
HSs (de Prada Pérez & Pascual y Cabo 2011; Toribio & Nye 2006).
To provide an explanatory analysis of HS outcomes in this domain, I propose that gustar-like
verbs are being reanalyzed in HS Spanish, taking an optional stative-agentive alternation (2), a use
deemed ungrammatical in monolingual grammars. With this innovation, gustar-like verbs would match
the well-documented hybridity displayed by other psych-verbs (e.g., asustar “to ri hten” (3)). on the
right track, HSs should allow for gustar to (optionally) appear in structures that would only be
grammatical if (and only if) the emerging stative-agentive alternation stated above were available in
their grammars (e.g., passives).
To test this hypothesis, ollow olinsky’s (2011) methodolo y, to isolate possible incomplete
acquisition and attrition effects. Moreover, I examine L1 immigrants to determine the extent to which
(attrited) input differences may be explanatory of HL development. 110 participants (see 4) completed
(i) an elicited production task (EPT) that aimed to elicit production of gustar-like verbs in different
contexts/forms, and (ii) a 1-4 scalar grammaticality judgment task (GJT), which was presented in video
modality to control or any potential literacy e ect. The GJT tested the participants’ acceptance o
(un)grammatical items in relation to the syntactic reflex of passivization stated above. The GJT critical
tokens were distributed equally among the following counterbalanced conditions: (i) passives with
transitive verbs (5); (ii) * passives with intransitive verbs (6); (iii) passives with legitimate Spanish
stative-agentive alternating verbs (7), and (iv) * passives with gustar (8).
The data reveal a sharp contrast between HSs’ tar et-deviant preferences regarding their ratings
of gustar with an agentive use and their ceiling-level accuracy regarding their ratings of other types of
verbs (figure-1). That said, neither attrition nor incomplete acquisition can be supported since the data
reflect continued HS development into adulthood (figure-2). Furthermore, in the EPT, HS adults and 1st
generation immigrants revealed comparable results in regards to the absence/presence of the dative
marker ‘a’ ( i ure-3) and verb agreement (figure-4), but differed significantly to those of true
monolingual adult speakers. This is interpreted in ways that are consistent only with input delimited
differences.
157
(1) a. A Teo le gusta esto b. Le gusta esto
To Teo him.DAT.CLI likes this. Him.DAT.CLI. likes this.
“Teo likes this” “S/He likes this”
c. Esto le gusta a Pau d. Esto le gusta
This him. DAT.CLI pleases to Pau This Him DAT.CLI likes
“This pleases au” “S/He likes this”
(2) a. A Carlos le gusta el chocolate (unintentionally Carlos is NOT the agent)
b. * A Carlos le gusta el chocolate (intentionally Carlos is the agent)
“Carlos likes chocolate”
(3) a. Diana asustó a Whitman (unintentionally Diana is NOT the agent)
b. Diana asustó a Whitman (intentionally Diana is the agent)
Diana.NOM scare. 3SG. PRES. to. Whitman.ACC.
“Diana scared Whitman”
(4) To control for dialect, all informants were Cuban (in the case of the monolinguals and 1st generation immigrants) or
Cuban-Americans (in the case of the heritage speakers).
Heritage Speakers 1
st Generation
Immigrants Monolingual Speakers
16 Children (HC) 49 adults (HS) 16 adults (AA) 13 Children (MC) 16 Adults (MA)
(5) Transitive sample (TRANS) Esa carta fue escrita por Laurie “That letter was written by Laurie”
(6) Unaccusative sample (UNACC) *Elena fue llegada a la casa “Elena was arrived home”
(7) Class II sample (PP) Pau fue asustado por Teo “ au was scared by Teo”
(8) Class III sample (RPP) * La pizza fue gustada por Jason “ izza was liked by Jason”
158
Spanish speakers’ use of prosody for sentence disambiguation
Matthew Patience
University of Toronto
The present study investigates the prosodic characteristics of Spanish sentences with ambiguous
attachment of PPs, as illustrated in (1) and (2).
(1) High attachment: El niño vio [NP al ladrón] [PP con el telescopio]]
(2) Low attachment: El niño vio [NP al ladrón[PP con el telescopio]] ‘the child
saw the thie with the telescope’
As context is not always sufficient to determine the correct interpretation, other strategies are often
necessary, such as prosodic cues including short pauses or changes in pitch. Studies on the prosody of
ambiguous sentences in English have found that a longer pause before the PP tends to indicate a high
attachment reading (1) whereas a shorter pause, or no pause at all, indicates low attachment ((2);
Snedeker & Casserly, 2010). The few studies that have investigated the prosody of ambiguous Spanish
sentences have arrived at contradictory findings. Both Reyes (2007) and Teira & Igoa (2007) examined
the ambiguous attachment of relative clauses (RC) and found that pauses were often present directly
before the RC in sentences with high attachment. Teira & Igoa did not observe pauses in any other
position; rather, they found a more prominent pitch change in sentences with high attachment occurring
between the tonic vowel of the lower NP and the final tonic vowel of the RC. Contrary to Teira &
Igoa, Reyes found that speakers also tended to place a pause after the higher NP in sentences with low
attachment. The results of these studies suggest that Spanish speakers rely on both the use of pauses
and changes in pitch to disambiguate sentences.
Studies have yet to examine the prosody of ambiguous PP attachment in Spanish. In the present study,
I analyze the prosodic characteristics of ambiguous sentences produced by 10 native Spanish speakers
performing a repetition task and an elicitation task. In the repetition task, participants listened to a
recorded context, followed by a question (Example A). They subsequently heard the words of the
target answer in the correct order produced by a synthesized robotic voice devoid of intonation. The
participants were then asked to repeat the answer, providing their own intonational model. In the
elicitation task, participants listened to a short context while looking at a picture (Figure B) and were
asked a question that was designed to elicit the target response. Each task consisted of 10 pairs of
ambiguous sentences varying in NP length (2 or 4 syllables). The duration of the boundaries between
constituents was measured, as were the changes in pitch between the tonic vowels of the NP
complement and the PP adjunct.
Preliminary results of two Mexican Spanish speakers indicate that pauses are not present in the data
which implies that, in contrast to English speakers, Spanish speakers do not use syntax-mapped pauses
to mark intonational boundaries as a strategy to disambiguate sentences. This contrasts with the results
of previous work on ambiguous RC attachment in Spanish, and may be attributable to the fact that
sentences with a RC are usually longer, which necessitates multiple intonational phrases. With regards
to pitch, the overall average change was similar for sentences with high and low attachment (Figure A).
However, when the data were examined individually for each ambiguous sentence pair, it was observed
in several instances of both tasks that the pitch change was greater with high attachment. These data
suggest that Spanish speakers use pitch changes to mark intonational boundaries, which serves as a
strategy to disambiguate sentences.
159
Example A – Example of the context, question, and answer for a sentence with low attachment for
task #1.
Context: Tres ladrones salieron corriendo de una tienda. Uno tenía una bolsa, uno tenía un televisor,
y el otro tenía un telescopio. El ladrón con el telescopio corrió en la dirección opuesta de los
dos primeros ladrones, y casi se chocó con Sandra. ‘Three thieves ran out of a store. One had
a bag, one had a television, and the other had a telescope.
The thief with the telescope ran in the opposite direction of the two first thieves, and almost
ran into Sandra.’
Question: ¿A qué ladrón vio Sandra? ‘Which thief did Sandra see?’
Answer in robot voice: Sandra vio al ladrón con el telescopio
‘Sandra saw the thie with the telescope’
Example B – Example of the images used for task #2
Low attachment
Question: ¿A qué ladrón vio Sandra?
‘Which thie did Sandra see?’
Target response: Sandra vio al ladrón con el telescopio ‘Sandra saw the thie with the telescope’
Figure A – Average pitch change for sentences with high and low attachment. Pitch change was
measured by subtracting the F0 of the tonic vowel of the PP complement from the F0 of tonic vowel
of the NP complement. Minimum and maximum F0 values were analyzed.
Task #1 Task #2
- 5
15
35
55
75
95
High Low High Low
Participant #1 Participant #2
F0 ( Hz )
max F0 min F0
- 5
15
35
55
75
95
High Low High Low
Participant #1 Participante #2
F0 ) Hz (
max F0 min F0
160
References
Reyes, Y. . (2007). “Determinación del patrón prosódico de las oraciones ambi uas con adjunción alta
de la cláusula de relativo” en Onomázein, Revista de Lingüística, Filología y Traducción. 16.
139-158.
Snedeker, J., & Casserly, E. (2010). “ s it all relative? effects of prosodic boundaries on the
comprehension and production o attachment ambi uities.” Language and Cognitive
Processes, 25(7-9), 1234-1264.
Teira, C., oa, J. M. (2007). “Relaciones entre la prosodia y la sintaxis en el procesamiento de
oraciones”. Anuario de psicología.38.1. 45-70.
161
Stimulated recalls and the complexity of learner language
Meghann M. Peace
University of Minnesota
Non-native expression of the Spanish direct object is a frequently studied topic in second language
acquisition research (e.g., Geeslin et al, 2010; Lee, 2003; Malovrh, 2008; Malovrh & Lee, 2010;
Montrul, 2010; VanPatten, 1990; Zyzik, 2004, 2006). In general, non-native speakers have been shown
to be less capable than native speakers when dealing with the appropriate forms of the direct object.
However, in the majority of these studies, the evaluations of non-native speech were made based
principally on the data, with minimal to no input from the speakers themselves.
Gass and Mackey (2000) explain the importance of a stimulated recall protocol to second language
research methodology. It has an advantage over other types of post-hoc interviews in that it involves
the use of stimuli to prompt the memory and that the participants do not need to have been previously
trained on how to do it. Its importance has to do with the fact that it allows the researcher to base
his/her conclusions on in ormation taken directly rom the participants, rather than the researcher’s own
intuitions and inferences.
In this study, non-native speakers of Spanish participated in a narration task, in which they retold
the events of a cooking video (Malovrh, 2008) to fellow non-native speakers. These listeners, in turn,
were responsible for writing a summary of the narration that would be given to the researcher. These
tasks were repeated by native speakers of Spanish for the purpose of providing a control group.
Immediately following the tasks, every speaker and listener took part in a stimulated recall. The recalls
were done separately, using the stimuli of the recorded narrations, and they were conducted in the
participants’ native lan ua e.
The results that were obtained from the narration alone presented a pattern of direct object use that
coincided with the results rom the previous studies. The participants’ level seemed to be the principal
factor that affected direct object production. However, the results from the stimulated recall revealed a
picture that is more complex than the other studies would lead one to believe. The participants’ speech
is affected not only by their language abilities, but also by a host of other factors. These factors include:
The speakers’ belie s about their listeners’ lin uistic abilities
The speakers’ belie s about their listeners’ back round knowled e
The speakers’ desire to communicate in ormation that is comprehensible
The speakers’ desire to communicate in ormation that is use ul to the listeners’ task
The speakers’ comprehension (or lack thereo ) o the events in the video
The speakers’ exposure to country-specific cultural and linguistic elements
Other personal factors
The results from the native speaker control group indicate that the native speakers share similar
pragmatic, cognitive and personal constraints.
Learner speech is complex and occurs at the intersection of many competing factors. The
complexity of learner speech is certainly a well-known tenet of second language research (e.g., Geeslin,
2011; Tarone, 2008), but by not includin the learners’ own viewpoints and contributions to the
analysis, this complexity may not be fully uncovered. The evidence from this study reveals that a
stimulated recall is indeed a necessary tool of second language research.
162
References
Gass, S. M. & Mackey, A. (2000). Stimulated Recall Methodology in Second Language Research.
Mahwah, NJ: Lawrence Erlbaum Associates.
Geeslin, K. L., Garcia-Amaya, L. J., Hasler-Barker, M., Henriksen, N. C., & Killam, J. (2010). The
SLA of direct object pronouns in a study abroad immersion environment where use is variable. In
Borgonovo, C. et al. (eds.) Selected Proceedings of the 12th
Hispanic Linguistics Symposium (246-
259). Somerville, MA: Cascadilla Proceedings Project.
Geeslin, K. (2011). The acquisition of variation in second language Spanish: How to identify and catch
a moving target. In M. Diaz-Campos (Ed.) Handbook of Spanish Sociolinguistics. Malden, MA:
Blackwell.
Lee, J. F. (2003). Clitics: Cognitive and linguistic perspectives on the acquisition of object pronouns in
Spanish as a second language. In B. A. Lafford & R. Salaberry (eds.), Spanish Second Language
Acquisition: State of the Science (pp. 98-129). Washington, D.C.: Georgetown University Press.
Malovrh, P. (2008). A Multifaceted Analysis of the Interlanguage Development of Spanish Direct-
object Clitic Pronouns Observed in L2-learner Production. (Doctoral dissertation). Indiana University,
Bloomington, IN.
Malovrh, P. A. & Lee, J. F. (2010). Conections between processing, production and placement:
Acquiring object pronouns in Spanish as a second language. Research in Second Language Processing
and Parsing, 231-256.
Montrul, S. (2010). How similar are L2 learners and heritage speakers? Spanish clitics and word order.
Applied Psycholinguistics 31, 167-207.
Tarone, E. (2008). Sociolinguistic approaches to second language acquisition research – 1997-2007.
The Modern Language Journal 91, 837-848.
VanPatten, B. (1990). The acquisition of clitic pronouns in Spanish: Two case studies. In B. VanPatten
& J. F. Lee (eds.), Second Language Acquisition – Foreign Language Learning (pp. 118-139).
Clevedon: Multilingual Matters.
Zyzik, E. (2004). Encoding Meaning with Polyfunctional Forms: The Acquisition of Clitics in L2
Spanish. (Doctoral dissertation). University of California Davis, Davis, CA.
Zyzik, E. (2006). A look at missing objects in L2 Spanish. In N. Sagarra & A. J. Toribio (eds.), Selected
proceedings of the 9th
Hispanic Linguistics Symposium (pp. 192-202). Somerville, MA: Cascadilla
Proceedings Project.
163
Eventiveness and copula selection in L2 Spanish Silvia Perpiñán
Itziri Moreno
Diana Botero
The University of Western Ontario
Selection of copula verbs in Spanish is a classic challenging area for L2 learners. Even so, it has
received only moderate attention in SLA research, and most of the studies have focused on the
acquisition of the semantic and pragmatic distinctions between ser and estar, particularly when
combined with adjectives (Bruhn de Garavito & Valenzuela, 2006; Geeslin, 2002; 2003; Schmitt &
Miller, 2007; among others). These studies explored the acquisition of semantic features such as aspect,
temporality or genericity, yet very few looked at the expression of location with the copular alternation.
This study aims to fill that gap by exploring the expression of location in [±eventive] subjects. To our
knowledge, this area has only been investigated in L1 and early bilingualism (Sera, 1992), but not in
adult SLA.
Recently, it has been proposed that estar is a more complex copula than ser, including the
merging of a terminal coincidence preposition: estar = ser + PT (Gallego & Uriagereka, 2011; Zagona,
2010). Following this idea, Brucart (2012) proposed a feature checking system by which the abstract
element (the preposition) of terminal coincidence has an uninterpretable feature that needs to be
checked. This [uPT] can be checked either by estar -because this verb, but not ser, includes an
interpretable feature of terminal coincidence-, or by eventive or path nominals in the specifier position.
Hence, eventive nominals seem to have a special status because they can legitimate that uninterpretable
feature, and as a result they need to appear with ser to express the meaning that typically would be
conveyed by estar. This theoretical analysis makes interesting predictions for language acquisition. On
the one hand, it considers ser to be the default copula, and as a consequence, acquired first. On the
other hand, eventive nominals, which require ser, are somehow equivalent to estar in the sense that
they imply an extra checking operation to legitimate the [uPT] feature. Thus, we predict that this extra
layer of complexity will result in a delayed acquisition.
The present study investigates the production and comprehension of locatives with ser and estar
in English-speaking learners of L2 Spanish (n = 36) with an intermediate proficiency. A Picture
Matching Task and an Elicited Production Task targeted location contexts with objects and events. In
the Picture Matching Task, participants had to choose between pairs of pictures according to the
meaning of the sentence provided (see example in 1). Crucially, the nominals employed could have a
double interpretation—as an event or as an object—and the use of the verb—ser or estar—
disambiguated the meaning. There were 6 pairs of pictures, with two interpretations each (one with ser
and one with estar), a total of 12 target responses plus 12 distracters. In the written production,
participants were presented with a map of a Hispanic city and had to locate a list of objects or events (k
= 8), such as la cita (the date), las palomitas
(popcorn), etc. Participants had one example with ser and one with estar as models.
Results indicated that intermediate level English-speaking learners of Spanish have weak
intuitions about the semantic entailment of the ser/estar distinction in locatives. Accuracy responses
were higher for events with ser (56% of correct responses) than for objects with estar (40% accuracy).
In the written production task, only the uses of ser and estar were analyzed. This task found similar
results: correct use of ser with events was relatively high (60% accuracy), but the location of objects
with estar was correct only 42% of the time. We interpret these higher accuracy responses of ser with
events not as the complete acquisition of the eventiveness effect but as a delayed acquisition of estar to
express location. These results con irm Van atten’s (1987) developmental sta e in the acquisition o
164
the copulas in which ser is overgeneralized for an extended period of time. Overall, these results
demonstrate a general
delay in the acquisition of estar, as the theoretical analyses would predict, and even a longer acquisition
process to master the semantic differences that the two copulas entail.
Examples:
(1) a. El examen final es en el aula 205. (target response: B)
b. El examen final está en el aula 205 (target response: A)
A B
References
Brucart, J. M. (2012). Copular alternation in Spanish and Catalan attributive sentences. Revista de
Estudos Linguísticos da Universidade do Porto, 7, 9–43.
Bruhn de Garavito, J. B. de, & Valenzuela, E. (2006). The Status of Ser and Estar in Late and Early
Bilingual 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.
Gallego, A. J., & Uriagereka, J. (2011). From ser to estar: A study of the Spanish copulas. Ms.
Universitat Aut noma de Barcelona & University of Maryland.
Geeslin, K. L. (2002). The Acquisition of Spanish Copula Choice and Its Relationship to Language
Change. Studies in Second Language Acquisition, 24(3), 419-450.
Geeslin, K. L. (2003). A Comparison of Copula Choice: Native Spanish Speakers and Advanced
Learners. Language Learning, 53(4), 703-764.
Schmitt, C., & Miller, K. (2007). Making discourse-dependent decisions The case o the copulas “ser”
and “estar” in Spanish. Lingua, 117(11), 1907-1929.
Sera, M. (1992). To Be or to Be: Use and Acquisition of the Spanish Copulas. Journal of Memory and
Language, 31(3), 408–427.
Van atten, B. (1987). Classroom learners’ acquisition o ser and estar accountin or developmental
patterns. In B. VanPatten, T. R. Dvorak, & J. Lee F. (Eds.), Foreign Language Teaching: a
Research Perspective (pp. 61–75). Cambridge: Newbury House Publishers.
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. (Vol. XIV). Amsterdam/New York: Rodopi.
165
Acoustic cues, vowel inventory size and bilingual perception strategies
Ryan Platz
Pennsylvania State University
coustic properties o the “same” vowels in particular lan ua es have been shown to be di erent (e. .
Bradlow 1994). Language learners, then, must adjust their acoustic awareness by using language-
specific perception strategies that perhaps were not relevant in L1. Highly competent bilinguals would
plausibly be the most perceptually attuned to relevant L2 acoustic cues. This language-specific
perception discrepancy might be a reflection of the different roles that f0 and F3 play in particular
languages that differ in vowel space size and distribution (e.g. Halberstam and Raphael 2004). To
address these issues, two ideal candidates for comparison are Spanish and English, languages that have
very different vowel spaces in terms of overall size as well as distribution. This study compares the L1
and L2 perception strategies of 6 Spanish-English and 6 English-Spanish bilinguals. Participants
listened and responded to four different variants of 5 Spanish words and 9 English words, each variant
representing an acoustically different version of that word: f0 removed (whispered-unfiltered), F3
removed (phonated-filtered), f0 and F3 removed (whispered-filtered) or no removal of any formant
(phonated-unfiltered). The specific research questions of this study were 1) is there a correspondence
between the size and distribution o a lan ua e’s vowel space and the acoustic cues used in its
perception?, and 2) do bilinguals of languages with contrasting vowel spaces adjust their perception
strategies in L2 to mimic those of native speakers? Results show that Spanish and English require
different analysis of the acoustic information provided by these formants: the former requires formant-
to-formant analysis, the latter requires individual formant analysis. Findings also show that (at least
highly proficient) bilinguals can eventually perform like natives in their non-native language, showing
an ability for L2 speakers to adjust their perception strategies used on the same acoustic information in
their L1.
Selected References
Bradlow, A. (1994). A comparative acoustic study of English and Spanish vowels. Journal of the
Acoustical Society of America 97(3). 1916-1924.
Disner, S. (1983). “Vowel quality The relation between universal and lan ua e speci ic patterns”.
UCLA Working Papers in Phonology. 58.
Halberstam, B. & Raphael L. J. (2004). Vowel normalization: the role of fundamental frequency and
upper formants. Journal of Phonetics, 32, 423-434.
Mendez, A. (1982). Production of American English and Spanish vowels. Language and Speech, 25(2),
191-197.
166
Attitudes towards Spanish, English, and Code-Switching in Two Texas Border Towns
Natalie Rangel
University of Texas at Austin
María Irene Moyna
Texas A&M University
Verónica Loureiro-Rodríguez
University of Manitoba
This study investigates language attitudes towards English, Spanish and code-switching in two
Texas border cities (Laredo and Edinburg) by means of a matched guise test with three attribute
dimensions (solidarity, status, and personal appeal). It was found that there were no significant overall
differences between attitudes in the two cities. As anticipated, code-switching received the lowest
ratings in all dimensions; English and Spanish were matched for status, and Spanish received the
highest scores for solidarity and personal appeal. The main finding of the study was that when the
variable of gender was considered (both of raters and speakers), then differences in ratings did emerge,
pointing to a complex interplay of the three varieties in mediating gender roles.
Situated 150 miles apart, Laredo and Edinburg share features such as their size, their majority
Hispanic ethnic and linguistic composition, and the vicinity to Mexico. However, whereas Laredo was
a fully constituted community when the Anglos arrived in the 19th century, and Laredoans succeeded
in retainin their lands and their political presti e (Hickey Cavazos, 2012), Edinbur ’s Hispanic
population was made up of dispossessed farm laborers (Alonzo, 1998). This study was designed to
ascertain whether these historical differences had linguistic consequences.
To measure attitudes towards Spanish, English, and Spanish/English code-switching, we
employed the matched guise test, developed by Lambert et al. (1960) to unearth covert attitudes
towards English and French in Montreal. In this technique, bilingual speakers record several controlled
samples in different linguistic varieties and then research subjects are asked to rate each voice they hear
(i.e., each guise), unaware that each speaker has spoken more than once. Since its development, the
matched guise test has been used to investigate language attitudes in multilingual and multicultural
contexts such as Brittany (Hoare, 2001), Mallorca (Pieras-Guasp, 2002), Galicia (Loureiro et al., 2013)
and Nigeria (Ihemere, 2006).
For our study, we prepared Spanish and English versions of an originally Spanish/English code-
switched spontaneous speech. Scripts were checked for grammaticality and then recorded by four
bilingual speakers (2M, 2F) of Mexican Spanish. Research subjects (96 bilinguals from Laredo and 91
from Edinburg) were asked to answer a demographic questionnaire and then listen and rate each voice
based on a list of attributes grouped according to dimensions of status, solidarity, and personal appeal.
It was found that in both cities code-switching received the lowest scores for all rating
dimensions (Tables 1-3). Additionally, English and Spanish exhibited no significant differences in the
status dimension (Table 2), but Spanish was ranked higher than English on solidarity (Table 1) and
personal appeal (Table 3). Subtle differences in the evaluation of the varieties emerged when we
considered the variables of speaker and rater gender. For example, female raters from Edinburg rated
male speakers who code-switched higher than Laredo female raters. Male raters in Edinburg rated
female English guises higher than those from Laredo. Yet, they rated male English speakers lower than
their Laredo counterparts. This complex pattern suggests that in Edinburg speakers have more tolerance
for English and code-switching between genders, but see Spanish as a covert solidarity marker among
males.
167
These findings confirm the long-term positive evaluation of Spanish in all social spheres in the
border region (Anderson-Mejías, 2005). Moreover, they show that even in nearby cities with similar
language usage, there may be differences in the role languages have in indexing gender relations.
References
Alonzo, A. C. (1998). Tejano legacy: Rancheros and settlers in South Texas 1734-1900. Albuquerque::
University of New Mexico Press.
Anderson-Mejías, P. L. (2005). Generation and Spanish language use in the Lower Rio Grande Valley
of Texas. Southwest Journal of Linguistics, 24(1-2), 1-12.
Hickey Cavazos, C. M. (2012). Spanish language use and linguistic attitudes in Laredo,Texas, between
1860 and 1930. (Ph.D.), Texas A&M University, Unpublished Dissertation.
Hoare, R. (2001). An Integrative Approach to Language Attitudes and Identity in Brittany. Journal of
Sociolinguistics, 5(1, Feb), 73-84.
168
Ihemere, K. U. (2006). An Integrated Approach to the Study of Language Attitudes and Change in
Nigeria: The Case of the Ikwerre of Port Harcourt City. Paper presented at the 36th Annual
Conference on African Linguistics.
Lambert, W. E., Hodgson, R. C., Gardner, R. C., & Fillenbaum, S. (1960). Evaluational reactions to
spoken languages. The Journal of Abnormal and Social Psychology, 60(1), 44-51.
Loureiro-Rodríguez, V., Boggess, M. M., & Goldsmith, A. (2013). Language attitudes in Galicia: using
the matched-guise test among high school students. Journal of Multilingual and Multicultural
Development, 34(2), 136-153.
Pieras-Guasp, F. (2002). Direct vs. indirect attitude measurement and the planning of Catalan in
Mallorca. Language Problems & Language Planning 26(1), 51-68.
169
The Subjunctive in Emotive and Epistemic Contexts as Lack of Syntactic Sentience
Elizabeth Rathmann
University of Iowa
Background. Following Stowell (1993), it is often assumed that subjunctive clauses fall into one of
two semantic classes: intensional subjunctive, including volitionals and directives, and polarity, or
operator-triggered, subjunctive. Subjunctive complement clauses to emotives in Spanish and other
Romance languages, however, have proven difficult to characterize because of their factive
(presupposed) status. To account for subjunctive mood distribution, syntactic proposals include special
licensing mechanisms contributed by the subjunctive (Laka, 1990, Motapanyane 1994, Poletto 1993,
Kempchinsky 1997), or as clauses with indefinite tense (Manzini 1994, Santos 1997). The main
objections against these approaches are the lack of tense dependencies and unexpected binding
properties found under some predicates. Semantic accounts have been based on modality (Farkas 1985,
et. seq.), non-veridicality (Giannakidou 1994, et. seq.), model shift (Quer 1998), illocutionary force
(Rivero and Terzi 1995), situation semantics (Portner 1997), or a null ordering source akin to epistemic
modals (Giorgi and Pianesi 1998). Nevertheless, semantic analyses have also failed to adequately
capture the entire spectra of properties that the subjunctive and indicative Mood alternations display
cross-linguistically.
Proposal. The current analysis proposes a distinct binary distribution of subjunctive clauses in Spanish.
The first is the nonveridical class; i.e. the proposition expressed by the subjunctive clause may not be
believed to be true by the speaker at the time of utterance (Giannakidou 1994, et. seq.). This class
includes subjunctive complements to volitionals, directives and negated cognitive-factives/verbs of
perception. The second class consists of subjunctive complement clauses to emotives and negated
epistemics, which, ar ue, are due to lack o syntactic ‘point o view’ anchorin in the subordinate
clause CP. The working assumption here is that sentience is syntactically represented through c-
command and co-indexation of 3 sentient arguments in the Speech Act and Sentience Phrases of the
subordinate CP (Speas and Tenny 2003). The current proposal is that the subjunctive complement
clauses to emotives and negated epistemics lack a Speech Act Phrase, while the latter also lack a
Sentience Phrase, in the embedded CP.
1. Similarities between Negated Epistemics and Factive Emotives. The subjunctive complement
clauses to negated epistemics cannot be attributed to non-veridical semantics, as is often assumed.
Unlike other negation triggered subjunctive, such as under negated verbs of perception or cognitive-
factives (1a), negated epistemics may be temporally modified (1b), allow a negation-raising
interpretation (2), and allow speaker disagreement with respect to the content of the subordinate clause
(3). Subjunctive complement clauses to negated epistemics and emotives share many properties; both
clause types disallow speaker oriented adverbs in embedded CP (4). They both show locality of
movement effects; i.e. they are weak islands to extraction (5a), allow only a high construal reading of
+wh interrogatives (6a) and disallow reconstruction (7a), as opposed to the subjunctive clauses selected
by volitionals in the (b) examples. Furthermore, they are the only subjunctive clauses that allow a
semantically de re, or reportative, interpretation (8). Other subjunctive clauses allow only de dicto
(quotative) arguments (8b).
2. Syntactic Sentience. Subjunctive clauses to emotives (like volitionals) display a subject disjoint
reference effect, meaning the matrix subject cannot co-refer to the subordinate clause subject (9a),
while the complement clauses to negated epistemics do not (9b). This is argued to stem from the co-
indexing of a null logophoric pronoun in the external argument position of the embedded Sentience
Phrase of emotives and volitionals with the matrix clause subject. Co-reference between clausal
complements is disallowed due to Principle B of Binding Theory; the logophoric pronoun in the
embedded CP may not be co-indexed with the complement clause subject. Because negated epistemics
170
lack a Sentience Phrase, no violation is incurred. The proposed structure for the clausal complement
CPs to indicative-selecting and volitionals is given in (10a), negated epistemics in (10b) and emotives
in (10c). It is also assumed, following Tenny and Speas (2003), that Speech Act Phrase is the locus of
illocutionary force; without this projection, no assertion feature is present, accounting for the
subjunctive mood in complement clauses to emotives and negated epistemics. Finally, it is proposed
that Speech Act Phrase constitutes a phase, creating an escape hatch for movement to the higher clause.
This accounts for the island status of subjunctive clauses to emotives and negated epistemics (11a).
Neither the volitional subjunctive nor indicative clauses constitute islands because their embedded CPs
contain a Speech Act Phrase (11b).
3. Clausal Referentiality. Recently, de Cuba (2007 et. seq.) has proposed that many predicates can
select a referential complement clause, resulting from a deficient CP; the structure is shown in (12).
Referential clauses may not be used to initiate a discourse, disallow Recomplementation (double
complementizer constructions) (13) and disallow Main Clause Phenomena (Haegeman & Ürögdi
2010); the subjunctive clauses in question share all three properties. The current proposal argues that
these clauses are the quintessential referential clauses, stemming specifically from lack of sentience
projections in the subordinate CP. The current analysis more elegantly captures the extraction data,
binding facts and pragmatic properties of the subjunctive in complements to emotives and negated
epistemics than previous analyses.
Data
(1) a. Juan no vio/entendió que llegara su profesor *hasta las 8.
‘John didn’t see/understand that his pro essor arrived (SUB) *until 8.’
b. Juan no pensó que llegara su profesor hasta las 8.
‘John didn’t think that his pro essor arrived (SUB) until 8.’
(2) a. Juan no pensó que llegara/llegó su profesor. =
Juan pensó que no llegó su profesor.
‘John didn’t think that his professor arrived (SUB/ ND)’
John thou ht that his pro essor didn’t arrive.
b. Juan no vio que llegara/llegó su profesor ≠
Juan vio que no llegó su profesor.
‘John didn’t see that his pro essor arrived (SUB/ ND).’
John saw that his pro essor didn’t arrive.
(3) a. Juan no vio que su mamá estuviera allí. #Pero yo, sí, lo vi.
‘John didn’t see that his mother was (SUB) there. #But did see it.’
b. Juan no pensó que su mamá estuviera allí. Pero yo, sí, lo pienso.
‘John didn’t think that his mom was (SUB) there. But do think so.’
(4) a. María piensa/dice que francamente/suertemente/supuestamente/evidentamente lo hizo.
Mary thinks/says that frankly/luckily/supposedly/evidently he did ( ND) it.’
b. *No piensa/lamenta que francamente/suertemente/supuestamente/evidentamente lo hiciera.
‘He doesn’t think/re ret that rankly/luckily/supposedly/evidently he did (SUB) it.’
(5) a. *Quién lamentas/no piensas que __ vaya a llamar a ella? -- island effects
‘Who do you re ret is oin (SUB) to call her?’
b. Quién querías que ___ llamara a ella? -- no island effects
‘Who did you want to call (SUB) her?’
(6) a. Cuándo querías que María reparara el coche <cuándo>?
‘When do you want Mary to repair (SUB) the car?’
171
b. *Cuándo no pensaste/lamentaste que María reparara el coche <cuándo>?
‘When didn’t you think/re ret that Mary repaired (SUB) the car?’
(7) a. Cuál de sus*i/j tías lamenta/no piensa Juanj que cada chicoi quiera más?
‘Which o his*i/j aunts does Johnj regret/not think that every boyi loves (SUB)
most?
b. Cuál de susi/j tías quiere Juanj que cada chicoi quiera más?
‘Which o hisi/j aunts does Johnj want every boyi to love (SUB) most?’
(8) a. Juan no pensó/lamenta que la ventana rota se hubiera roto.
‘John didn’t think/re ret that the broken window had (SUB) broken.’
b. *Juan quería/no vio que la ventana rota se hubiera roto.
‘John didn’t want/see that the broken window had (SUB) broken.’
(9) a. Maríai desea/lamenta que pro*i/j haya llevado el libro de matemáticas.
‘Mary desires/re rets that s/he (someone else/*hersel ) has (SUB) brou ht
the math book.’ b. Maríai no piensa que proi/j haya llevado el libro de
matemáticas.
‘Mary doesn’t think that s/he (someone else/*hersel ) has (SUB) brou ht the
math book.’
(10) a. [Sub que SA Sen Fin]
b. [Sub que Fin]
c. [Sub que Sen Fin]
(11) a. *[CP Cuándo [TP lamentaste [SUB que [TP María reparara [VP el coche <cuándo>]]]]]
b. [CP Cuándo [
(12) a. Non-referential CP: [cP que1 [TopP Top [FocP Foc [CP C ] ] ] ]
b. Referential CP: [FocP Foc [CP C ] ] ] ]
(13) a. Juan pensó/oyó que el filme (que) lo viste ayer.
John thought/heard that the film (that) you saw it
yesterd
c. Juan no pensó/lamenta que el filme (*que) lo viste
ayer.
John didn’t think/re ret that the filme (*that) you saw it
yesterday.
Selected References
de Cuba, C. 2007. On (Non)Factivity, Clausal Complementation and the CP-field. Ph.D. dissertation,
Stony Brook University.
Speas, M. and C. L. Tenny. 2003. Configurational Properties of Point of View Roles. In Anna
Maria Di Scuillo (ed.), Asymmetry in Grammar. Amsterdam: John Benjamins, 315-344.
TP querías [ SUB que [ SaP < cuándo> [ TP María reparara [ VP el coche <cuándo>]]]]]]?
172
No Superiority, no Intervention Effects: The Spanish puzzle
Lara Reglero
Florida State University
Emma Ticio
University of Syracuse
The puzzle: As argued extensively in the literature (Beck 1996; Pesetsky 2000; Cable 2010), there is a
clear correlation between Superiority effects and Intervention effects across languages. According to
Cable (2010 132), “in any lan ua e L, the in situ wh-words of a multiple wh-question of L are subject
to Intervention Effects if and only if the multiple wh-questions of L are not subject to Superiority
E ects”. This eneralization, which is considered to be universal, captures the behavior o lan ua es
such as English or Tinglit, on the one hand, and German and Navajo, on the other hand. As shown in
(1), non-D-linked in situ wh-phrases in English exhibit Superiority Effects but not Intervention Effects.
In contrast, in situ wh-phrases in German do not exhibit Superiority Effects but they are subject to
Intervention Effects, as in (2). Jaeggli (1982) noted for the first time that Spanish lacks Superiority
Effects, as shown in (3a-b). Based on the universal generalization above, the prediction here is that in
situ wh-phrases in Spanish should exhibit Intervention Effects. Interestingly, this is not the case. As
illustrated in (3c-d), Spanish does not display Intervention Effects.
Assumptions: Our analysis combines the following first three assumptions, independently motivated in
the literature, with a fourth one, motivated by cross-linguistic differences. First, previous work on wh-
phrases (Reglero 2007) has conclusively shown empirical evidence against overt wh-movement in wh-
in-situ constructions in Spanish. As the contrast in (4a-b) illustrates, Spanish wh-in-situ phrases can
appear in island contexts, which points to a non-movement analysis of these constructions. Second,
typological research has shown that languages display different lexical properties regarding the
requirements to be satisfied at Force/CompPhrase in wh-questions (Pesetsky 2000; Cable 2010). Third,
different types of movements (i.e., phrasal and non-phrasal/featural) are available to satisfy the
ForcePhrase lexical requirements (Pesetsky 2000). Fourth, non-phrasal whmovement is banned if it
crosses an o endin operator in its movement, adaptin Beck’s (1996) ori inal insi hts that
characterized Intervention Effects as a particular type of island.
Proposal: Languages displaying superiority effects must satisfy the wh-feature requirement of Force
multiple times (as many times as there are wh-phrases in the structure), obeying usual locality
constraints (i.e. non-D-linked wh-questions in English, as in (1)). Languages without superiority effects
can either satisfy the wh-feature requirement via non-phrasal movement (i.e. German, as in (2);
Pesetsky 2000) or they can select from their lexicon a Force head that triggers only one whfeature
requirement, if available (i.e. Spanish, as in (3), Reglero & Ticio 2008). Under this proposal, only
languages with non-phrasal wh-movement will display Intervention Effects. In other words, the major
claim of the analysis is that the in situ expressions in languages with Intervention Effects must undergo
an illicit movement.
Our proposal assumes that Spanish has available in its lexicon a Force head that can satisfy its
requirements by checking its wh-features only once. This is supported by the fact that there is no
movement in Spanish wh-in-situ constructions ((4)), and explains the lack of Intervention Effects since
phrasal movement does not trigger Intervention Effects.
Predictions: Two pieces of evidence confirm this analysis for Spanish Intervention Effects. First, some
questions allow non-phrasal wh-movement. These are similar to the separation constructions (wh-
questions in which the operator appears to move overtly without its restriction) also available in French
173
and German (Beck 1996). In those cases, the presence of an intervener triggers ungrammaticality even
though the movement is overt, since it is non-phrasal movement. The prediction is confirmed, cf. (5).
Second, assuming recent proposals on D-linked wh-phrases (Shields 2008), D-linked in situ wh-phrases
need to move to satisfy their specificity requirements. The current analysis predicts that those questions
should display Intervention Effects in Spanish, which is indeed the case, cf. (6). This suggests that the
ban on non-phrasal movement over another operator affects not only wh-features associated with wh-
elements, but that it is a broader phenomenon affecting the interaction of operators.
Examples
(1) a. Who bought what?
b.*What did who buy?
c. Who didn’t buy what?
(2) a. Wer hat was gekauft?
who has what bought
b. Was hat wer gekauft?
c.??Wer hat niemanden wo angetroffen?
who has nobody where met
(3) a. Quién compró qué? who bought what
b. Qué compró quién?
c. Quién no compró qué?
who neg bought what
d. Qué no compró quién?
(4) a. Te has enamorado del hombre que vive con quién?
you have fallen-in-love of-the man who lives with who
b.*Con quién te has enamorado del hombre que vive?
(5) a. Cómo de caliente quieres la sopa? how of hot
you-want the soup
b. Cómo quieres la sopa de caliente?
c. Cómo de caliente no quieres la sopa? how of hot neg you-want the soup
d.*Cómo no quieres la sopa de caliente?
(6) ??Qué regalo no pidió qué niño?
which present neg asked which child
References
Beck, Sigrid. 1996. Quantified structures as barriers for LF movement. Natural Language Semantics 4:
1-56.
Cable, Seth. 2010. The Grammar of Q: Q-Particles, Wh-Movement and Pied-Piping. Oxford
University Press.
Jaeggli, Osvaldo. 1982. Topics in Romance Syntax. Foris: Dordrecht.
Pesetsky, David. 2000. Phrasal Movement and its Kin. Cambridge, MA: MIT Press.
174
Reglero, Lara. 2007. Wh-in-situ interrogatives in Spanish. Probus 19(2): 267-297.
Reglero, Lara and Emma Ticio. 2008. Wh-in-situ and the Spanish DP Movement or no movement?’,
University of Pennsylvania Working Papers in Linguistics 14(1), J. Tauberer, A. Eilam, and L.
MacKenzie (eds), p. 311-324.
Shields, Rebecca. 2008. Intervention and Rescue: Arguments for Representational Constraints on
Syntactic Dependencies. Ph.D. dissertation. University of Wisconsin-Madison.
175
Exploring the actuation problem: Fricative allophony in Latin American Spanish
Jeff Renaud
University of Iowa
1. Introduction. One of the central issues in sound change is the actuation problem (Weinreich et al.
1968), which asks, if coarticulatory effects are ubiquitous, why are sound changes that are claimed to
result from coarticulation language- and dialect-specific? I.e., why is it that only some cases of
phonetic variability (and not others) result in phonologization only in some languages (and not others)
(see Garrett and Johnson 2013 for a general review)? In this paper I explore these questions by
examining three fricative allophonic properties o Latin merican Spanish / /→[x] velarization,
/ /→[ɸ] bilabialization and /x/→[ç] palatalization. ollow recent work by Baker et al. (2011) who
argue that sound change depends on the confluence of a context-dependent phonetic effect subject to
speaker variability with particular social factors that facilitate dispersion. Concerning the Spanish data,
I argue that, although all three processes are articulatorily motivated and manifest speaker variability, it
is the social aspect of each that differentiates the presence versus absence of the processes in a given
speaker’s rammar.
2. Data. In particular dialects throughout Latin America, Spanish /f/ may have as possible allophones
either the velar [x] or bilabial [ɸ] fricatives when followed by the back vowels [u] and [o] (e.g., fui ‘
went/was’ [x i] or [ɸ i]), but not in the contexts of [a, i, e] (falla ‘ ail re’ [ a.ʝa], *[x]/*[ɸ], (1) and
(2)). Velar /x/, on the other hand, fronts to palatal [ç] before [i] and [e] (gema ‘ emstone’ [çe.ma]), but
not [a, u, o] (jabón ‘soap’ [xa. on], *[ç], (3)). In terms of phonetic context, all three processes are
assimilatory in nature–the fricative assumes the place of the following vowel: the backness of [u, o]
retracts /f/ to velar [x]; the roundedness of [u, o] bilabializes /f/ to [ɸ]; and the frontness of [i, e]
advances /x/ to palatal [ç].
3. Empirical results. To assess variability concerning the degree of coarticulation, I conducted three
production tasks on 26 speakers from Chile and 9 from Mexico. Participants responded to ten questions
during a 15-minute sociolinguistic interview, read 200 sentences and described 85 pictures. Responses
were recorded and analyzed in Praat. The data suggest that variability with respect to degree of
coarticulation between fricative and the following vowel obtains in all three processes. In that case, if,
all else being equal, the assimilatory nature and the phonetic variability that Baker et al. argue to be
prerequisites for change are present in the three processes, the fact that /x/ palatalization is a standard
allophonic process in the regions listed above while /f/ allophony is relegated to the lower classes is
unexpected. I address this question by exploring the social implications associated with each process.
4. Hypothesis. I argue that palatalization has propagated due to the fact that no social stigmatization
accompanies it; /x/ velarization is present in all social strata where it forms part of the regional dialect.
The two /f/ allophonic processes, on the other hand, do carry social stigma, and their presence (or lack
thereof) correlates with social status: /f/ allophony is absent in the upper classes, but quite frequent in
the lower classes. As a corollary with social class, I also argue that orthography (i.e., literacy)
rein orces / / aith ulness. / /→[x] is avoided in the upper classes due to the e ect o spellin in the
sense that the grapheme <f> is associated exclusively with the phonetic segment [f]; velar [x] is
represented orthographically by <gi, ge, j>. The unique written representations prevent /f/ from being
realized as [x] in the literate upper classes; the lower classes (i.e., less education in general) do not
benefit from such reinforcement. The output of /x/ palatalization, [ç], is, on the other hand, strictly
176
allophonic, and causes no phonemic neutralization, so spelling has no role in its avoidance irrespective
of class.
(1) /f/ velarization
a. fumar [xu.maɾ] ‘to smoke’ (Piñeros 2009: 283)
b. forma [xoɾ.ma] ‘ orm’ (Piñeros op. cit.)
c. falla
(2) /f/ bilabialization
*[xa.ʝa] ‘ ail re’ (Perissinotto 1975: 55)
a. futuro [ɸ .t .ɾo] ‘ t re’ (Perissinotto op. cit.)
b. fuerte [ɸ eɾ.t e ‘stron ’ (Resnick 1975: 24)
c. fácil
(3) /x/ palatalization
*[ɸa.sil] ‘easy’ (Perissinotto op. cit.)
a. gema [çe.ma] ‘ emstone’ (Piñeros 2009: 284)
b. jefe [çe.fe] ‘boss’ (Piñeros op.cit.)
c. jabón *[ça. on] ‘soap’ (Resnick 1975: 38)
References
Baker, Adam, Diana Archangeli and Jeff Mielke. 2011. Variability in American English s-retraction
suggests a solution to the actuation problem. Language Variation and Change 23, 347–374.
Garrett, Andrew and Keith Johnson. 2013. Phonetic bias in sound change. In A. C. L. Yu (ed.), Origins
of Sound Change: Approaches to Phonologization, 51–97. New York, NY/Oxford, UK: Oxford
University Press.
Perissinotto, Giorgio. 1975. Fonología del español hablado en la ciudad de México [Phonology of the
Spanish spoken in Mexico City]. Guanajuato, Mexico: El colegio de Mexico.
Piñeros, Carlos-Eduardo. 2009. Estructura de los sonidos del español [Structure of the sounds of
Spanish]. Upper Saddle River, NJ: Pearson Prentice Hall.
Resnick, Melvyn C. 1975. Phonological Variants and Dialect Identification in Latin American
Spanish. The Hague, Netherlands: Mouton de Gruyter.
Weinreich, Uriel, William Labov and Marvin Herzog. 1968. Empirical foundations for a theory of
language change. In W. Lehmann and Y. Malkiel (eds.), Directions for Historical Linguistics, 95–195.
Austin, TX: University of Texas Press.
177
Clitic climbing in bilingual Spanish: testing language contact outcome hypotheses
Adrián Rodríguez Riccelli
University of Florida/University of Texas at Austin
Kelley Woodfine
University of Florida
Sarah Rogers
University of Florida
In Spanish, clitics can appear attached to the non-finite verb (1) or to the main conjugated restructuring
verb (2), a phenomenon known as clitic climbing (ClC). In the syntactic-theoretical literature, two main
analyses have been put forth: the Incorporation approach (Kayne 1989, 1991, 1994, Roberts 1991,
1994, 1997) and the Restructuring approach (Cardinaletti & Shlonsky 2004, Cinque 2001, 2002, Rizzi
1982). These analyses converge on (i) the restriction of ClC to restructuring predicates and (ii) the
optionality of ClC. Interestingly, variationist studies indicate that this optionality varies across
predicates depending on their level of grammaticalization (Davies 1995, 1998). Building on previous
literature, this project examines monolingual and bilingual judgments of ClCed and non-ClCed
sentences in Spanish, which informs formal contact linguistics theories. In particular, this project tests
the Interface Hypothesis (IH) (Sorace & Filiaci 2006) whereby language contact effects are observed at
the syntax interfaces with other modules but, crucially, not at the narrow syntax. ClC is relevant to test
the IH because the availability of ClC lies at the narrow syntax. Therefore, English is not predicted to
borrow ClC from Spanish in a contact situation and Spanish is not predicted to lose ClC in this contact
situation. At the same time, ClC in Spanish lies at an internal interface as it is regulated by the type of
predicate (restructuring vs non restructuring predicates). Thus, some contact effects are predicted for
Spanish, in terms of a reduction of use of ClC and a loss of the sensitivity to predicate types that allow
for ClC. An alternative approach, the Vulnerability Hypothesis (VH, Prada Pérez, in progress), predicts
that language contact targets variable phenomena, independent of the area of the grammar in question.
The VH establishes a categorical-variable continuum of permeability, i.e. variable phenomena, where a
form is used close to 50% of the time in a specific context, are permeable while categorical
phenomena, where a form is used close to 100% of the time in a specific context, are not. As such, it
contrasts with the IH as it takes into account the differential variability across restructuring verbs and
predicts that restructuring verbs with different ClC frequencies should be differentially affected by
language contact. Thus, verbs such as poder, which are highly variable with regards to ClC should be
more susceptible to language contact effects as opposed to verbs like esperar which seldom allows
ClC.
In order to compare the IH and the VH, a total of 16 monolingual Spanish (MS) speakers from
Valladolid, Spain, and 72 heritage speakers (HSs) completed a scalar grammaticality judgment task
(GJT). The GJT tested speakers’ acceptability o ClCed and non-ClCed sentences controlled for clause
type (all main clauses), verb form (simple present forms only), and predicate factuality (factual
predicates only), and manipulated for VERB TYPE (noClC verb esperar, frequent ClC ir + a, infrequent
ClC intentar), CLITIC TYPE (accusative vs. reflexive) and ANIMACY (animate vs. inanimate Cl referent),
with a total of 96 tokens (4 tokens per condition). Preliminary results indicate that both monolingual
and bilingual speakers are sensitive to VERB TYPE, CLITIC TYPE and ANIMACY of the referent.
Monolingual speakers, however, use ClC in a more restricted manner by rating ClC significantly higher
with infrequent ClC verbs than with no ClC verbs and higher yet with frequent ClC verbs. Bilinguals
follow the same trends but there are significant differences in the acceptance of ClC with infrequent
ClC verbs, as compared to monolinguals. They reject sentences with ClC with infrequent ClC verbs
significantly more than monolinguals, as predicted by the VH.
This paper contributes to several fields. It builds on syntactic-theoretical and variationist analyses
providing a better understanding of both the monolingual and the bilingual grammar. Additionally, it
178
provides novel data not considered in the bilingualism literature (ClC). Lastly, it helps elucidate
restrictions on the outcomes of language contact.
Examples
(1) No ClC
Quiero verlo
want-1SG see-it
‘ want to see it’
(2) ClC
Lo quiero ver
It want-1SG see
‘ want to see it’
(3) No English ClC
I have seen it
*I it have seen
References
Cardinaletti, A. and Shlonsky, U. (2004). Clitic positions and restructuring in Italian. Linguistic
Inquiry 35: 519-557.
Cinque, G. (2001). “Restructurin ” and unctional structure. University of Venice working papers in
linguistics, 11: 45 – 127. Venice: University of Venice.
Cinque, G. (2002). A note on restructuring and quantifier climbing in French. Linguistic Inquiry 33:
617 – 636.
Davies, M. (1995). Analyzing syntactic variation with computer-based corpora: The case of Modern
Spanish clitic climbing. Hispania, 78, 2: 370-380.
Davies, M. (1998). The evolution of Spanish clitic climbing: A corpus-based approach. Studia
Neophilologica, 69, 2: 251-263
Kayne, Richard. (1989a). Null Subjects and Clitic Climbing. In Osvaldo Jaeggli and Ken Safir,
(eds.) The Null Subject Parameter, pp. 239-261. Dordrecht:
Kayne, Richard. (1991). Romance clitics, verb movement, and PRO. Linguistic Inquiry 22.4: 647 –
686.
Kayne, Richard. (1994). The antisymmetry of syntax. Cambridge, MA: The MIT Press.
Rizzi, Luigi. (1982). Issues in Italian Syntax. Dordrecht: Foris.
Roberts, Ian. (1991). Excorporation and minimality. Linguistic Inquiry, 22: 209- 218.
Roberts, Ian. (1994). Two types of verb movement in Romance. In David Lightfoot and Norbert
Hornstein (eds.) Verb Movement, pp. 207 – 242. Cambridge, UK: Cambridge University
Press.
Roberts, Ian. (1997). Restructuring, head movement, and locality. Linguistic Inquiry, 28.3: 423-460.
Sorace, A., & Filiaci, F. (2006). Anaphora resolution in near-native speakers of Italian. Second
Language Research, 22, 339–368
179
Intermediate and Heritage Processing of Spanish SE Structures in Self-Paced Reading
Estrella Rodríguez
Florida State University
This project contrasts the processing of Spanish SE structures in an online task by a group of intermediate
learners of Spanish with L1 English and a group of heritage Spanish speakers. Montrul & Bowles (2009) have
viewed heritage learners as “early bilin uals who have been exposed to the Spanish lan ua e at home in a
naturalistic settin ”. Herita e rammar acquisition may be incomplete, since these learners’ ormal lan ua e o
instruction has been English. It was interesting to see if the mental grammar of heritage speakers while
processing SE sentences resembled the one of intermediate learners. SE is a multi- unctional structure o Spanish. This project adopted Mendikoetxea’s (2008) proposal
that Spanish SE structures differ as to verbal features. The verb of impersonal SE sentences is defective
or devoid of person and number features, as it is not in agreement with any element of the sentence, as
seen in (3) and in structure (6). Mendikoetxea (2008) views passive SE sentences as partially defective
structures, with only a number, but no person feature; as seen in (2) and in structure (5). Her analysis of
impersonals and passives was adapted to reflexives to contrast how reflexive sentences with SE do
contain person and number features, as the verb agrees at all times with the subject of the sentence.
This is seen in (1) and in structure (4).
The features of the 3 SE structures were manipulated in a self-paced reading task. The statistical
results suggest that heritage processing of reflexive, passive and impersonal SE structures differs
significantly from intermediate processing at the Verb region, as heritage learners took more time to
read reflexive and passive constructions in the [+ agreement] condition, as seen in (1) and (2). This is
something with which they patterned with monolingual speakers of Spanish. The intermediate
participants with L1 English, on the other hand, were slow in reading reflexive sentences in the [-
agreement] condition, as seen in (7). Experimental results indicate that when processing multi-
functional SE structures in real time, intermediate processing does not resemble heritage processing,
especially at the Verb region.
Examples (1) En el baño la niña se seca las manos todas las mañanas.
Is the following information included in what you read? The girl dries her hands in the morning. A. Yes B. No
(2) En la oficina se pierden los papeles todo el tiempo.
Is the following information included in what you read? The
papers are misplaced in the zoo. A. Yes B. No
(3) En las cocinas se seca los platos después de lavarlos.
Is the following information included in what you read? People dry the windows in the kitchen. A. Yes B. No
(4) [T D la niña [T [T’ se seca [v [v’ [v se s e c a [V [V’[ V s e s e c a [DP las manos ]]]]]]]]]]
[+ person/ + number] [+ person/ + number] [NP Plural]
(5) [CLP [CL se [T [T [T’ pierden [v G-pro [v’ [ v p i er d e n [ V [V’ [ Vpier de n [DP los
papeles]]]]]]]]]]]] [0 person] [0person/Plural] [0 person/Plural] [0 Person/ Plural] [NP Plural]
(6) [CLP [CL se [T [T [T’seca [v G-pro [v’[ v s e c a [ V [V’ [V s ec a [ DP los
platos]]]]]]]]]]]] [0 person] [0 person/0 number] [0 person/0 number] (defective) [0 pers./ 0 numb] [NP Plural]
180
(7) *En el baño la niña se secan las manos todas las mañanas. Is the following information
included in what you read? The girl dries her hands at night. A. YES B. NO
(8)*En la oficina se pierdes los papeles todo el tiempo. Is the
following more or less what you just read?
The papers are misplaced in the office. A. YES B. NO
(9)*En las cocinas se secamos los platos después de lavarlos. Is
the following more or less what you just read?
People dry the dishes in the kitchen. A. YES B. NO
References
Bruhn de Garavito, J. (1999). The Syntax of Spanish multifunctional clitics and near native ompetence.
Unpublished doctoral dissertation. Department of Linguistics. McGill University.
Montrul, S. (2011). Morphological errors in Spanish second language learners and heritage speakers.
SSLA, 33, 163-192.
Montrul, S., & Bowles, M. (2009). Back to basics: Incomplete knowledge of Differential Object Marking
in Spanish heritage speakers. Bilingualism: Language and Cognition, 12, 363-383.
Mendikoetxea, A. (2008). Clitic impersonal constructions in Romance: Syntactic features and semantic
interpretation. Transactions of the Philological Society, 106, 290-336.
181
Do all bilinguals switch the same way?
Testing patterns of non-habitual Spanish-English code-switchers
Eva Rodríguez-González
Miami University
Maria Carmen Parafita-Couto
Leiden University
Background: Codeswitching is a widespread phenomenon in bilingual speech. Most researchers have
paid attention to sociological interpretation and discourse functions of codeswitching. Only in recent
years has bilingualism research focused on the study of the processes takin place in the speaker’s
mind (MacWhinney, 2005; Pienemann et al., 2005). In this regard, results from controlled
psycholinguistic laboratory experiments have provided the field of bilingualism with fresh
perspectives. The study of mixed noun phrases within Spanish-English codeswitching research is
relevant since gender and word order in nominal constructions operate differently in both languages.
Spanish shows postnominal adjectives while English shows prenominal adjectives. Recent research
accounts for a preferred Determiner-Adjective-Noun word order (Myers-Scotton, 2002). As for gender,
Spanish shows a masculine-feminine distinction while English shows no gender agreement. Masculine
is claimed to be the default gender in Spanish-English switched nominal constructions (Dussias et al.,
2009).
Present Study: The present study examined codeswitching patterns within the noun phrase in both
naturalistic and experimental settings. Thirty Spanish-English bilingual speakers participated in the
study. Different from previous studies, all participants in the study were non-habitual switchers
recruited in Ohio, USA and exhibited different ranges of language use, experience and exposure to both
languages (see figure 1). The study investigated to what extent similar patterns are produced by non-
habitual language switchers as compared to habitual switchers in Spanish-English switched nominal
constructions. The study consisted of a linguistic questionnaire (information about language
dominance, experience and exposure and other social factors), a director-matcher task (cf. figure 2)
that elicited nominal constructions (Gullberg, Indefrey & Muysken 2009) and an acceptability
judgment task (VanDulm 2007). The acceptability task consisted of 72 sentences presented on a
computer screen. Participants were asked for a scalar response that ranged between absolute
unacceptable sentences to completely acceptable ones. Participants were also provided with an option
that indicated whether they could not decide for a specific acceptability rating. Reaction times were
recorded to measure metalinguistic knowledge as a reflection of language awareness.
Results: English was by far the language that most of the participants had received their input in from
their mother, father, guardian and in school settings. The social network was also geared towards
English use. Sixty three per cent of participants agreed that they kept languages separate, while twenty
three per cent stated that languages should be kept separate. There was an overall preference of one
switch over two switches in the acceptability judgment task. More specifically, prenominal adjective
positions were preferred when noun and adjective were in the same language. One switch was also
more common in the toy task (“The candela roja” vs. “La candle roja”). Postnominal adjectives were
dominant in the toy task (“El plane rosa”). The results also support recent previous indin s that
indicate that masculine is the default gender for Spanish and English bilinguals in both naturalistic and
experimental settings. Feminine determiners were used when the translation equivalent of the noun was
feminine. In summary, the findings obtained in the present study from non-habitual switchers show that
their switching patterns within the noun phrase are not different from those previously observed in
habitual code-switchers.
182
3. Acceptability Judment Task: Stimuli (72 sentences with Codeswitching) The table roja es nueva Det Noun Post Adj (Matrix Language=Spanish)
I love the fresco cheese Det PreAdj Noun (Matrix Language= English)
4. References
Dussias, G., Chip C., Valdés Kroff, J., Gullifer, J. and R. Guzzardo Tamargo. (2009). Asymmetrical
use of gender information during the processing of unilingual and code-switched speech.
Bangor University, UK, March 12 2009.
Gullber , M., nde rey, . & Muysken, . (2009). “Research techniques for the study of code-
switchin ”. n Bullock, B. E. & . J. Toribio (eds.). The Cambrid e Handbook o Lin uistic
Code-switching. Cambridge Cambridge University Press, 21-39.
MacWhinney, B. (2005b). A unified model of language acquisition. In J. F. Kroll & A. M. B. de Groot
(Eds.), Handbook of bilingualism: Psycholinguistic approaches (pp. 49-67). New York: Oxford
University Press.
Myers-Scotton, C. (2002). Contact linguistics: Bilingual encounters and grammatical outcomes.
Oxford and New York: Oxford University Press.
183
Pienemann, M., Di Biase, B., Kawaguchi, S. and Hakansson, G. 2005. Processing Constraints on L1
Transfer. In J.F. Kroll and A.M.B. de Groot (Eds.), Handbook of Bilingualism: Psycholinguistic
Approaches (pp. 128-153). Oxford, New York: Oxford University Press.
Van Dulm, O. (2007). The grammar of English-Afrikaans code switching: a feature checking account.
PhD Thesis.
184
Transfer, Proficiency and Working Memory Effects in Processing S-V Agreement in L2 Spanish
Nuria Sagarra
Crystal Marull
Rutgers University
Native speakers process and produce speech rapidly and efficiently in real time (cf. McCarthy,
2008). Native procedures develop along with grammatical competence during first language (L1)
acquisition (Kuhl, 2004), eventually permitting the mature speaker automatized and rapid
comprehension and production. Non-native (L2) adults must likewise develop representational
knowledge and procedural routines, but they start their task already possessing the grammar and
processing strategies of their L1. While L2 adult learners benefit from L1-L2 morphological similarity
(e.g., Hopp, 2007; Tokowicz & MacWhinney, 2005; Sabourin & Stowe, 2008), they still show a
persistent difficulty with inflectional morphology. Various proposals have been put forward to account
or adult L2 learners’ poor per ormance in morpholo ical representation and computation. This study
examines the role of L1 (English-impoverished morphology vs. Romanian/Spanish-rich morphology),
L2 (different proficiency levels), and cognitive control (working memory) in the processing of subject-
verb number agreement, using eye-tracking.
A group of 236 English and Romanian learners of Spanish of different proficiency levels and
English, Romanian and Spanish monolinguals read 85 sentences (5 practice, 32 experimental—k = 8
per conditgion—) in the L2 (learners) or the L1 (monolinguals) with subject-verb
agreement/disa reement with explicit subject (…el hombre bebe/*beben… ‘the man drinks/*drink’)
and without explicit subject (…bebe/beben… ‘drinks/drink’). En lish monolin uals only read sentences
with explicit subject. After each sentence, participants chose one of four pictures: [+/-grammatically
congruent, +/- semantically congruent]. Semantic changes affected the subject (one-third of
experimental trials), the verb (another third), and the object (another third). Proficiency was determined
via the DELE test, and working memory was assessed through a number-letter sequencing test.
Preliminary results reveal that (a) all participants—except the lowest proficiency ones— are sensitive
to violations, (b) that native speakers of morphologically rich languages and more proficient learners
rely more on verbal inflections than native speakers of morphologically poor languages and less
proficient learners, respectively (L1 and L2 effects), and (c) that beginning learners with more working
memory are more sensitive to agreement violations. These findings lend support to theories in favor of
transfer effects, and are in line with studies showing working memory effects on L2 morphosyntactic
processing in beginners.
185
Semantic and Pragmatic Properties Of Venir And Andar + Gerund
Lorena Sainzmaza-Lecanda
Meagan Horn
The Ohio State University
Spanish periphrastic constructions have been widely studied from a semantic point of view
(Laca 2004b, 2006) but to a much lesser extent from a pragmatic perspective (Amaral, forthcoming).
To the best of our knowledge, no study exploring the semantic and pragmatic differences between two
Spanish verbs of motion, andar and venir, in construction with a gerund has yet been carried out.
Therefore, this paper takes the lead in providing an analysis of these two periphrastic constructions in
Peninsular Spanish through syntactic, semantic and pragmatic tests with native speaker insights.
Prior work (Bergareche 2004) has defined these two constructions in terms of their aspectual
restrictions, such that both carry frequentative or habitual meanings. We show however that that the
entailments conveyed by the pluractional meaning of each periphrasis illustrate a semantic difference
between these two constructions. In addition, we demonstrate that, together with this semantic
difference, these constructions can also be distinguished on the basis of two pragmatic properties:
speaker intention towards accomplishing a goal and the level of speaker engagement.
An example of each construction is presented below:
(1) ’m a member o the Betis soccer team and ’m an unconditional supporter of the team. Every
season, soccer and Betis are the only things I can think about and I never miss a game.
(1a) Vengo asistiendo a los partidos del Betis desde hace 50 años.
(1b) # Ando asistiendo a los partidos del Betis desde hace 50 años.
(2) My ather is obsessed with the Betis soccer team and he is an unconditional supporter, whereas ’m
not. Sometimes, though, I like to spend some time with him so I join him in going to a game.
(2a) # Vengo asistiendo a los partidos del Betis desde hace 50 años. (2b)
Ando asistiendo a los partidos del Betis desde hace 50 años.
The examples (1) and (2) suggest that both periphrases have frequentative meaning. But whereas venir
+ gerund entails that the gaps of time between each repetition are relatively short, andar + gerund
entails longer durations between each repetition relative to the maximal potential of the repetition of the
event. These semantic distinctions are closely related to their pragmatic facets: venir + gerund
implicates that the speaker has a clear intention to accomplish a goal and is highly engaged in
accomplishing that particular goal, whereas andar + gerund implicates that the speaker has an uncertain
or indifferent intention towards the goal accomplishment and is passively engaged in accomplishing it.
Therefore, we demonstrate that the frequentative meaning of these two periphrases varies regarding
the time gaps between each repetition of the event. In addition, following Grice (1975), we postulate
that the pragmatic differences observed for andar and venir + gerund in Peninsular Spanish should be
understood as Generalized Conversational Implicatures. Although context-dependent, the implicatures
are attached to the verb of motion + [gerund] form and they do not need to be computed repeatedly
with every relevant utterance. Our analysis has linguistic implications with regard to the existing
pragmatic variation underlying the meaning of Spanish frequentative periphrases cross-dialectally (cf.
Amaral, forthcoming).
186
References
Amaral, Patricia. To appear. On how to live and keep dying. In Howe, C., Blackwell, S., Quesdada, M.
Selected Proceedings of the Hispanic Linguistics Symposium 2011. Cascadilla Press.
Amaral, Patricia. Forthcoming. The pragmatics of number: The evaluative properties of vivir +
V[Gerund]
Birner, Betty J. 2013. Introduction to Pragmatics. Chichester, West Sussex: Wiley-Blackwell.
Camus Ber areche, Bruno. 2004. “ erí rasis verbales y expresión del aspecto en español”, L. García
Fernández y B. Camus Bergareche (Eds.), El pretérito imperfecto, Madrid: Gredos, 511-572.
Grice, H.P., 1975. The logic of conversation. In Cole, P., Morgan, J.L. (Eds.), Syntax and Semantics 3:
Speech Acts. Academic Press, New York, 41–58.
Huang, Yan. 2007. Pragmatics. Oxford: Oxford UP.
Laca, Brenda. 2004b. “ ro ressive, luractionals and the Domains o spect.” n O. Crouzet, H.
Demidache, and S. Wauquier (Eds.) Domain(e)s, Proceedings of the Journée de Linguistique de Nantes
2004. 87-92.
Laca, Brenda. 2006. Indefinites, quantifiers, and pluractionals. What scope effects tell us about event
pluralities. Vogeleer, S. & L. Tasmowski (Eds.), Non-definiteness and Plurality, John Benjamins,
Amsterdam/Philadelphia, 191–217.
Sedano, Mercedes. 2000. La perífrasis de gerundio en Caracas y otras ciudades hispanohablantes.
Nueva Revista de Filología Hispánica, 48, 2, 253-274.
Torres Cacoullos, Rena. 2001. From lexical to grammatical to social meaning. Language in Society 30,
443-478.
187
Futurates: the present progressive as interpreted by L1 English L2 Spanish speakers
Covadonga Sanchez
University of Massachusetts, Amherst
The Full Transfer/Full Access Hypothesis (FTFA), proposed by Schwartz and Sprouse, states that the
initial state of second language acquisition is the final state of the L1 at which all the properties of the
L1 are available to the learner (White, 2003). The theory of Multiple Grammars (Amaral and Roeper,
2012), being one of the theories that follow the steps of FTFA, posits that speakers then have access to
di erent “parallel rule-sets” to which di erent rates o productivity are assi ned based on rules
available to them. It also claims that rules are not deleted but their productivity might be blocked.
Therefore, it is predicted that the productivity of the L1 rules will prevent learners from blocking
ungrammatical readings in their target language.
One grammatical aspect that can be considered from a Multiple Grammars perspective is the use of the
present progressive. Future readings with present progressive constructions are allowed in English, but
not in Spanish (Quirk, 1985; Copley, 2009; RAE, 2009). This study has two goals: (1) examine the
acceptability of future readings by L1 English L2 Spanish speakers (see examples 1-2); (2) determine
whether these readings are influenced by lexical aspect, by comparing activity (example 3) and
accomplishment verbs (example 4). Previous studies (Robinson, 1994; Muñoz and Gilabert, 2011;
Gabriele and Canales, 2011) have looked at L1
Spanish L2 En lish speakers’ production and acceptability o uture readin s dependin on the verb
type. They have shown that the acceptability of future readings was higher in the interpretation tasks
when a context was provided. Conversely, learners had difficulty accepting both types of future
readings in the grammaticality judgment tests.
The questions that guide this study are: (1) will native speakers of English learning Spanish block the
interpretation of future readings? (2) If there is an extension of future readings with present progressive
constructions, will this be more visible with activity or accomplishment verbs? The hypothesis is that
learners will allow for future interpretations in a considerable amount of cases and that the future
interpretation of activity verbs will be blocked less often than that of accomplishments, due to the
tendency of activities to express progressivity. In order to test this hypothesis, an experiment consisting
of a picture-matching interpretation task was run with a total of 24 intermediate speakers of Spanish.
The answers provided were compared with those of native speakers.
The results obtained show that L2 speakers accept future readings significantly more often than native
speakers not only with activity verbs (p= 0.0007) but also with accomplishments (p=0.003). On the
other hand, no statistical difference was found in the acceptability rates between activities and
accomplishments, being that both were accepted by approximately 7080% of the non-native speakers
(see figure). This suggests that the rules that are productive in their L1 remain so in their second
language. Therefore, speakers at this level of proficiency fail to assess the productivity of the rules
governing the use of the present progressive in the target language. In respect to lexical aspect, neither
of the two categories considered seem to play any crucial role in this matter, since both favor the
acceptability of future readings.
Examples:
(1) My grandmother is visiting us tomorrow. (2) *Mi abuela nos está visitando mañana.
My grandmother us is visiting tomorrow
188
“My grandmother is visiting us tomorrow”
(3) Este fin de semana, (María) está viajando a México This weekend, (María) is travelling to México
“María is travelling to Mexico” (4) Paula está corriendo una maratón benéfica esta tarde.
Paula is running a marathon charitable this afternoon
“Paula is running a charitable marathon this afternoon” (5) A Marisa le fascina el mar. Desde pequeña, siempre quiso tener su propio barco para navegar,
pero nunca pudo permitírselo. Marisa acaba de mudarse a Miami y allí ha conocido a un rico empresario
que tiene un lujoso yate. Éste la ha invitado a un viaje en su yate así que María por fin está navegando este
fin de semana. “Marisa is ascinated by the sea. When she was a kid, she always wanted to have her own boat to sail, but
she couldn’t ever a ord it. Marisa has just moved to Miami, and has met a rich businessman with a luxurious
yatch. He has invited her on a trip on his yatch, so Marisa is inally sailin this weekend”
[-future] [+future]
Figure:
activities and accomplishments [+future]
100%
50%rejected
0%accepted
Native L2 Native L2 ACTIVITIES ACCOMPLISHMENTS
References:
M R L, L and T. Roeper. 2012. “Multiples Grammars and Second Lan ua e Representation” (in
press) Copley, B., 2009. The Semantics of the Future. Taylor & Francis Group, London.
GABRIELE, A., and A. Canales. 2011. "No time like the present: Examining transfer at the interfaces
in second language acquisition". Lingua. 121 (4): 670-687.
MUÑOZ C., and Gilabert R. 2011. "More evidence concerning the aspect hypothesis: The acquisition
of English progressive aspect by Catalan-Spanish instructed learners". IRAL - International Review of
Applied Linguistics in Language Teaching. 49 (3): 241-263.
QUIRK, R., 1985, A Comprehensive Grammar of English Language, Longman, London; New York
REAL ACADEMIA ESPAÑOL, AND ASOCIACIÓN DE ACADEMIAS DE LA LENGUA
ESPAÑOLA. 2009. Nueva gram tica de la lengua espa ola. Madrid Espasa Libros.
ROBINSON, Richard Earl. 1994. "Aspectual Marking in English Interlanguage: A Cross-Sectional
Study". Dissertation Abstracts International. 54 (11): 4073A-74A.
189
SALABERRY, M. Rafael, and Yasuhiro Shirai. 2002. The L2 acquisition of tense-aspect morphology.
Amsterdam: J. Benjamins Pub.
WH TE, Lydia (2003). “On the Nature of Interlanguage Representation: Universal Grammar in Second
Lan ua e”. n Dou hty, C. and Michael, L. The Handbook of Second Language. P. 19-42. Malden –
MA: Blackwell
Publishers
190
Gender Agreement of Overt and Covert Nouns
Tiago Schulz
Danielle Muilenberg
Diego Sevillano
Sam Hursey
University of Florida
The acquisition of gender agreement in Spanish as a second or heritage language has spurred
significant research (Alarcón 2004, 2009, 2010, 2011, Gruter, Lew-Williams and Fernald 2011,
Montrul 2008, Kurinski and Sera 2010,). A distinction is made between gender assignment, or the
identification of the gender of a noun, and gender agreement, or gender matching of the
adjective/determiner and the noun (Alarcón 2004 and references therein). In particular, research has
reported L2ers that adequately demonstrate gender agreement also adequately demonstrate gender
assignment, but the inverse is not necessarily true (Alarcón 2004) even though when gender assignment
of a noun is congruent, there is a higher probability of gender agreement with overt and covert nouns
(1a-b), semantic and non-semantic nouns (2a-b), and masculine and feminine (3a-b)(Alarcón 2010).
Crucially, research on gender agreement has reported differential difficulty in gender acquisition
depending on language-external variables: PRESENTATION MEDIUM, where heritage speakers (HSs) are
more accurate with gender assignment in oral than written production (Montrul 2008) while L2ers
display the opposite pattern (Gruter, Lew-Williams and Fernald 2011), CULTURAL ASPECTS, where
L2ers select the gender, not for linguistic association, but for the relationship between language and
cognition (Kurinski and Sera 2010), PROFICIENCY, where advanced learners are more sensitive to
gender agreement, arguably native-like (Alarcón 2011) as well as language-internal variables: LEXICAL
ASSIGNMENT (Gruter, Lew-Williams and Fernald 2011), where L2ers difficulty in assigning gender is a
lexical (i.e., not morphosyntactic) representation issue, NOUN CLASS, or whether gender is semantic or
non-semantic in a noun (4a-b), the GENDER of the head and attractor nouns,(5a-b), even though
evidence remains inconclusive as to whether gender affect Reaction Time (Alarcón 2009) HEAD NOUN
MORPHOLOGY, (6a-b) where L2ers do not process overt and covert nouns without conscious effort
(Alarcón 2009), and noun class and gender CONGRUENCIES, (7a-b) of which L2ers process semantic
attractor nouns (8a-b) the fastest (Alarcón 2009).
In spite of the great advances in the study of second language gender agreement, there are still some
questions that remain to be answered, mainly with regards to HS acquisition and its comparison to L2
acquisition and with regards to agreement of covert nouns. This paper addresses these questions,
analyzing how second language learners and heritage speakers acquire gender without the
morphological cue in the noun (covert nouns). We anticipate that (i) regarding GENDER of the noun:
both groups will be more target-like with masculine than with feminine nouns, (ii) regarding NOUN
TYPE: gender agreement of overt nouns will be more target-like than covert nouns, (iii) regarding
ELLIPSIS TYPE: gender agreement is expected to be more target-like in adjectival than in prepositional or
clausal ellipsis, (iv) regarding SPEAKER TYPE: HSs will be more target-like than L2ers, and (v)
regarding PROFICIENCY: both groups will be more target-like with increased proficiency. In order to test
these hypotheses, to date 19 (Advanced, N=9; Intermediate, N= 5; Low, N= 5) HSs and 27 (Advanced,
N=5; Intermediate, N=9; Low, N=13) L2ers completed a picture selection task manipulated for NOUN
TYPE, GENDER of the noun and ELLIPSIS TYPE (adjectival, prepositional, or clausal). Preliminary results
indicate that (i) both groups are sensitive to gender disagreement but more so with masculine nouns,
(ii) neither group performs significantly better with overt than covert nouns, (iii) no ellipsis type
facilitates gender assignment, (iv) no significant differences were found across speaker groups, and (v)
low proficiency speakers are less target-like than intermediate and advanced speakers, however, no
differences were found between intermediate and advanced speakers.
191
This study contributes to the field of bilingualism as it compares two groups of bilingual speakers
that differ in the age and mode of acquisition to further understand the effect that internal factors have
on language contact outcomes.
Examples:
(1) a. Overt Noun
Mesa “table”, suelo “ loor”, ventana “window”
b. Covert Noun
Noche “ni ht”, pared “wall”, reloj “watch
(2) a. Semantic Noun
Compañero “classmate”, pintor “painter”, hombre “man”
b. Non-Semantic Noun
Fiesta “party”, Flor “ lower”, país “country”
(3) a. Masculine
Cuchillo “kni e”, plato “plate, arból “tree
b. Feminine
Taza “cup”, cuchara “spoon”, verdura “ve etable”
(4) a. Semantic
Agente
b. Non-Semantic
Lámpara
(5) a. Head noun
El asistente..
b. Attractor noun
…del president
(6) a. Overt head
La chica del vecino
b. Covert head
El reloj de la camarera
(7) a. Masculine head/ Masculine attractor
“El almuerzo del secretrario”
b. Masculine head/ Feminine attractor
“El almuerzo de la secretaria”
c. Feminine head/ Masculine attractor
La cena del secretario
d. Feminine head/ Feminine attractor
La cena de la secretaria
(8) a. Semantic Attractor nouns
“El billete del pasajero...”
192
The Social Stratification of Loanwords:
A Computational and Corpus-based Approach to Anglicisms in Argentina
Jacqueline Serigos
University of Texas at Austin
As a highly salient outcome of language contact, loanwords often carry social meaning within the
communities they are found. For example, longstanding contact between Spanish and English speakers
in the US has resulted in highly stigmatized loanwords, such as bildin 'building', and lonch 'lunch'
(Othguey and Stern 2011). Paradoxically, in Argentina, Anglicisms such as tips, marketing and
feedback are commonly incorporated into the quotidian lexicon to si nal speakers’ elevated education,
refinement and worldliness (Nielsen 2003). Though Argentine Spanish has been cited as having one of
the highest influxes of Anglicisms in the Spanish-speaking world, there is a paucity of scholarly work
on these highly regarded English-Spanish contact features (Bordelois 2011). To remedy this gap, the
present study aims to analyze the role of socio-economic status in loanword adoption through the use of
written corpora and proposes a computational method to identify loanwords. Automated methods to
process bilingual texts are of increasing importance for language contact researchers to allow them to
efficiently process large data sets, yet very few methods exist (Solorio 2008). Additionally this paper
seeks to address the role of indirect language contact to the field of contact linguistics.
In order to analyze the Anglicisms currently used in Argentine Spanish, a 1.8 million-word corpus of
newspaper articles from 2012 to 2013 was created. Three prominent national newspapers were selected
to represent the spectrum of social groups within the society: La Nación -high socioeconomic status-
Clarín -middle socioeconomic status- and Crónica -low socioeconomic status- (Boczkowski and de
Santos 2007). To identify Anglicisms in the corpus, I employed a combination of automatic and
manual processing. First, the raw corpus was processed using TreeTagger, a program that annotates
each token with its part of speech and corresponding lemma (see Table 1)(Schmid 1995). Second, all
tokens in the corpus were checked to see if their corresponding lemma appeared in a modified English
dictionary. The modified English dictionary is the Unix English dictionary minus all proper nouns, eg.
China and José, and all homographs between the two languages, eg. chocolate and admirable. The
lemmas for each newspaper were then manually inspected to remove incorrectly identified tokens. For
example, the band Kiss, a proper noun incorrectly identified as a loanword, was removed.
This processing method fails to capture any loanword that has been orthographically altered, such as
suéter < sweater. Nevertheless, two factors led me to choose this method: a large majority of loanwords
in this corpus retained English orthography (judged from manually inspecting 3,000 words from the
corpus) and I was able to automate the majority of the loanword identification process. I propose that
automated methods may not only greatly improve efficiency in language contact research but also more
importantly will avoid human error and demand a greater transparency for criteria used by researchers
to identify contact phenomena.
Upon analyzing the 1,968 lemmas of 7,593 tokens identified as Anglicisms in the corpus, a significant
pattern emerged; both loanword tokens and unique lemmas appear with greater frequency in the
newspaper associated with the highest socio-economic status (see Table 2). The differences between
newspapers demonstrate that loanwords are not evenly distributed throughout Argentine society; rather
they are more densely concentrated in the higher socio-economic status group and thus may be
considered to be prestige markers. This distribution may be in part related to the fact that there is not
much direct contact with English, in contrast with many bilingual setting in the US where Spanish
speakers and English speakers interact in person. Most contact with English in Argentina comes
through indirectly from the media and internet. As access to English is greater for those with more
193
access to technology and media, the middle and upper socio-economic groups are the groups more
likely to adopt contact features. As advances in technology and globalization expand the ways
languages come into contact, the type of contact: direct or indirect, is increasingly significant to contact
linguistics as it appears to affect sociolinguistic factors of contact outcomes.
Table 1: Sample Output from Treetagger
TOKEN POS LEMMA
hicieron NP <unknown>
un NNS <unknown>
cover VBP cover
fatal JJ fatal
de NP de
Kiss NP Kiss
Table 2: Raw Counts from the 2012-2013 Argentine Newspaper Corpus
Anglicisms Unique Lemmas
La Nación 2894 764
Clarín 996 240
Crónica 809 153
References
Boczkowski, Pablo J., and Martin de Santos. 2007. “When More Media Equals Less News atterns o
Content Homo enization in r entina’s Leadin rint and Online Newspapers.” Political
Communication 24 (2): 167–180.
Bordelois, Ivonne. 2011. El país que nos habla. Sudamericana.
Nielsen, Paul Maersk. 2003. “En lish in r entina Sociolin uistic ro ile.” World Englishes 22 (2):
199–209.
Othe uy, Ricardo, and Nancy Stern. 2011. “On So-called Span lish.” International Journal of
Bilingualism 15 (1): 85–100.
Schmid, Helmut. 1994. “ robabilistic art-of-speech Ta in Usin Decision Trees.” Proceedings
of International Conference on New Methods in Language Processing, 12:44–49.
Solorio, Thamar, and Yan Liu. 2008. “ art-of-speech Tagging for English-Spanish Code-switched
Text.” Proceedings of the Conference on Empirical Methods in Natural Language Processing, 1051–
1060.
Weinreich, Uriel. 1953. Languages in Contact, Findings and Problems. Publications of the Linguistic
Circle of New York no. 1. New York: Linguistic Circle of New York.
194
Afro-Peruvian Spanish and the Spanish Creole Debate
Sandro Sessarego
University of Wisconsin-Madison
This study considers the linguistic and sociohistorical aspects of Afro-Peruvian Spanish (APS), an
Afro-Hispanic language spoken in the province of Chincha (coastal Peru) by the descendants of the
slaves taken to this region to work on sugarcane plantations during the colonial period. It attempts to
explore the unclear origin of APS and to shed further light on the long-lasting Spanish creole debate
(cf. Lipski 2005: ch.9).
According to McWhorter (2000), APS would be a “missin Spanish creole”, a contact variety that
did not evolve into a creole, even though the sociohistorical conditions for a creole to emerge were well
in place in coastal Peru at the time of its formation: 1) massive African importation; 2) harsh working
conditions; 3) big plantation settings; 4) rare manumission; 5) few chances of learning the Spanish
lan ua e throu h church indoctrination since “reli ion was withheld even to the point o denyin
slaves their last rites” (McWhorter 2000 37). Given this socioeconomic analysis of coastal Peru, as
well as the historical information this author provides for several other regions of Latina America (i.e.,
Chocó, Colombia; Chota Valley, Ecuador; Veracruz, Mexico; Barlovento, Venezuela), McWhorter
claims that the lack of access to the lexifier language could not have been the reason for the
development of creoles in the New World, since in all of the above mentioned cases the slaves had
little access to Spanish, but yet Spanish creoles are not found in such regions.
McWhorter believes that the American plantation creoles used to be pidgins which were
expanded into fully referential languages by either children or adults (McWhorter 1997, 2000). As a
result, he does not a ree with Chaudeson’s (1979, 1992) and Mu wene’s (1996) accounts, which depict
creole languages as transformed varieties of their European lexifiers. According to the latter scholars,
the distance between certain creole languages and their lexifiers may be modeled as the result of the
cyclical application of language-acquisition phases, where each new wave of slaves would end up
tar etin a ‘square approximation’ (c . Chaudeson 2001 132) o the lan ua e tar eted by the previous
generation; thus resulting in a progressive differentiation between the means of communication used in
the plantation and the European variety. This model is unacceptable in McWhorter’s view, since it
would exclude the pidgin stage by presupposing a gradual drift apart from the lexifier.
On the contrary, in line with his Afrogenesis hypothesis of creole formation, McWhorter claims
that since Spanish never pidginized on the West African coasts, the linguistic bases were missing for
the establishment of a full-fledged Spanish creole in the Americas, in contrast with what happened for
French and English.
The present paper provides new linguistic and sociohistorical data showing that the Peruvian
plantation settings were quite different from the picture offered by McWhorter, thus indicating that
access to Spanish was not so limited. Most importantly, this study presents a new perspective on the
long lasting Spanish creole debate. This work carries out a comparative analysis of slavery in the
Americas, showing that Spain not only diverged from other European powers in that it lacked African
colonies (as pointed out by McWhorter); rather, one of the most prominent differences between Spain
and the other countries involved in the colonization of the Americas had to do with the legal position of
the slave. In fact, the juridical figure of the slave had been received by the Spanish legal system in
ancient time, from the Roman Corpus Juris Civilis; it was gradually developed into the Siete Partidas
and then further modified in the Leyes de India. On the contrary, such a juridical figure followed a
significantly dissimilar evolutionary path in the other European codifications, resulting in remarkably
different socioeconomic and sociolinguistic outcomes (Watson 1989; Andrés-Gallego 2005). The
current forensic account and its sociohistorical implications add a new and original dimension to the
existing linguistic debate concerning the genesis and evolution of Afro-Hispanic languages in the
Americas.
195
Phonetic Interactions in Code-Switching by Spanish-English Naturalistic Code-Switchers
Whitman Suarez
Diana Pedraza
University of Texas at Austin
Most of the literature on the production of voiceless occlusive stops consistently agree that
bilinguals are able to produce almost native-like voice onset time (VOT) values in monolingual
contexts (e.g. Antoniou et al., 2010, Flege & Eefting 1987, Bullock & Toribio, 2009), On the other
hand, the effects of code-switching on the production of VOT values differ considerably across studies
revealing (i) dominance of the first language (L1) over the second language (L2) (Antoniou et al.,
2011), (ii) a bidirectional effect of the two languages either by convergence (Bullock & Toribio, 2009)
or divergence (Flege & Eefting 1987), and (iii) no crosslinguistic effects (Grosjean & Miller 1994).
One possible explanation for the different results in the previous literature, which we explore here,
might be the heterogeneity of the participant groups.
The present study examines and analyzes the speech production of the voiceless stop /k/ by a
group of highly proficient US-Spanish heritage speakers. These informants, referred to here as
naturalistic code-switchers (NCS) due to the nature of their bilingualism, regularly engage in
spontaneous code-switching interactions. Their linguistic behavior is ideal for the investigation into
how the bilingual mind deals with language (de)activation and the implication of language interaction
in their phonetic realizations.
To study the effects of code-switching on the phonetic realization of /k/, 12 participants
completed a reading task (1) and a story-retelling task in English/Spanish monolingual modes and in
code-switching mode. Their results were compared with those of Spanish and English L1 dominant
speakers. The analysis consisted of five independent variables, a) language, b) speaker type, c) mode,
d) switch site (pre, during, post), e) task, and one dependent variable, VOT.
Results showed that NCS’ VOT values resemble those o their L1-dominant counterparts in
both English and Spanish. Statistical analyses also revealed that code-switching did not significantly
affect the production of VOTs regardless of the position of the switch. While there was no significant
difference between tasks in Spanish, both controls and NCS produced shorter VOTs in reading than in
retelling in English. The fact that the methodology combined both controlled and spontaneous speech
allowed us to find this outcome. We suggest along with the prediction that stems from the Variability
Hypothesis (de Prada, in prep) that in language contact situations more categorical phenomena (i.e.
Spanish VOTs) is less vulnerable to crosslinguistic influence than more variable phenomena (i.e.
English VOTs). This argument could explain similar data attested in other code-switching studies (e.g.
Antoniou et al. 2011).
The findings also suggest that NCS have better control of their phonetic realizations than other
bilinguals who do not regularly engage in code-switching. Initial analyses of the full data set are in line
with this trend. The different type of population recruited could explain the results from previous
studies (e.g. Grosjean & Miller, 1994 and Bullock et al. 2006).
Keywords: Code-switching, Voice Onset Time, Voiceless stops, English, Spanish.
(1) Examples of sentences from the reading task. All critical segments were in word-initial position in
disyllabic words with penultimate stress.
a) Spanish
Anoche vendí una casa preciosa.
‘Last ni ht sold a beauti ul house.’
196
b) English
Parking on campus after ten is not permitted.
c) Code-Switching (English-Spanish)
The Campbell family cambia de casa todos los años.
‘The Campbell amily chan es homes every year.’
d) Code-Swiching (Spanish-English)
La tía de Carlos Pérez carries lots of candies on Halloween.
‘Carlos érez’s aunt carries lots o candies on Halloween.’
References
Antoniou, M., Best, C. T., Tyler, M. D., & Kroos, C. (2010). Language context elicits native-like stop
voicing in early bilinguals' productions in both L1 and L2. Journal of Phonetics, 38, 640-653.
Antoniou, M., Best, C. T., Tyler, M. D., Kroos, C. (2011). Inter-language interference in VOT
production by L2-dominant bilinguals: Asymmetries in phonetic code-switching. Journal of Phonetics,
39(4), 558 – 570.
Bullock, B.E.; Toribio, A. J.; González, V., & Dalola, A. (2006). Language dominance and
performance outcomes in bilingual pronunciation". Proceedings of the 8th Generative Approaches to
Second Language Acquisition Conference (GASLA 2005) (pp. 9-16). Somerville, MA: Cascadilla.
Bullock, B. E., & Toribio, A. J. (2009). Trying to hit a moving target: On the sociophonetics of code-
switching. In I. Ludmila, D. Winford & K. de Bot (Ed.), Multidisciplinary approaches to code
switching (pp.189-206). Amsterdam: Jonn Benjamins.
Flege, J.E. & Eefting, W. (1987). Production and perception of English stops by native Spanish
speakers. Journal of Phonetics, 15, 67–83.
De Prada, A. (submitted). The Variability Hypothesis. Unpublished manuscript.
Grosjean, F. & Miller, J. (1994). Going in and out of languages: an example of bilingual flexibility.
Psychological Sciences, 5, 201–206.
197
Investigations into the calculation of scalar implicatures by Spanish-English bilingual children
Kristen Syrett
Rutgers University - New Brunswick
Anne Lingwall
Rutgers University - New Brunswick
Silvia Pérez-Cortes
Rutgers University - New Brunswick
Jennifer Austin
Rutgers University - Newark
Liliana Sánchez
Rutgers University - New Brunswick
Scalar implicatures (SIs) ([1]) are a linguistic phenomenon of particular interest, because they
represent a pragmatic layer of language usage above and beyond pure logical (semantic) content.
Although (1) is true if each boy in a group scores a goal, a speaker delivering this utterance is
understood as indicating that some, but not all, of the boys scored a goal.
1. Some boys scored a goal.
Cross-linguistic evidence shows that children generally do not demonstrate adult-like ability to
compute SIs ([2]-[6]), although some task manipulations improve performance ([6]-[8]).
One notable exception is Spanish-speakin children, who appear to calculate a ‘somebut-not-all’
implicature by age six ([9]-[10]). Spanish has two words for some: while algunos encodes the
implicature, unos does not. This contrast between Spanish and other languages led us to wonder
how Spanish-English bilingual children fare with SIs, asking whether they pattern with their
monolingual counterparts (calculating the implicature – Spanish – or failing to – English) or if
they exhibit a unique pattern. Our experimental results argue in favor of the second option.
Experiment 1: 27 children (3;4-5;5 M=4;3) participated in a Forced-Choice Selection Task in
which four scenes were pitted against each other. In one scene, each object in a set shared the
same property (‘whole set’); in another, only some o the objects had the property (‘subset’).
(Figure 1) Test sentences were statements with algunos, unos, and todos (n=3 each). Children’s
choices were almost evenly distributed between ‘whole set’ and ‘subset’ scenes. (Table 1) To
further examine whether children distinguished between these determiners, we targeted
experiments providing additional contextual support.
Experiment 2: 18 children (4;2-5;1 M=4;9) participated in a Truth-Value Judgment Task ([11]).
Test items (n=8) included todos or algunos in stories presentin ‘whole set’ or ‘subset’ scenarios,
rendering todos statements True/False, and algunos statements True but Infelicitous/Felicitous.
Children distinguished scenarios for todos sentences as expected, but accepted algunos statements
in both scenarios. (Figure 2)
Experiment 3: 16 children (3;9-5;5 M=5;0) were shown a series of brief video-recorded scenes in
which a Speaker requested that the Hearer perform an action with objects, using unos, algunos, or
todos. While the Hearer always complied, children had to judge whether the action was
performed as requested. Results showed that children not only sharply distinguished between
True/False scenarios for todos, but also for Felicitous/Infelicitous scenarios for algunos and unos,
though they did not seem to discriminate between algunos and unos. (Figure 3)
198
Experiment 4: 24 children (4;2-5;3 M=4;9) participated in an adaptation of [5]’s sentence
judgment task for preschoolers, designed to assess their ability to recruit real-world knowledge.
Children consistently rejected false sentences as raro, but accepted true sentences as bien,
regardless of the felicity status. (Table 2)
Improved performance with tasks highlighting speaker-hearer interactions may reveal bilingual
children’s sensitivity to the status o their interlocutor ([12]-[13]). The comparison between
monolingual and bilingual Spanish-speaking children may also show the Principle of Contrast
([14]) at work in the acquisition of Spanish determiners. The results invite further exploration of
SIs with a wider range of bilingual populations.
Table 2: Experiment 4 (Sentence Judgment task)
n=2 each False and ‘Bizarre’ Felicitous (algunos, unos) /
True (todos) Infelicitous / False (todos)
algunos Algunos caballos cantan.
Some horses sing Algunos gatos son negros.
Some cats are black Algunos perros tienen ojos.
Some dogs have eyes
% bien 23.3% 56.6% 61.6%
unos Unos pájaros bailan.
Some birds dance Unas flores son amarillas.
Some flowers are yellow Unas casas tienen puertas.
Some houses have doors
199
% bien 13.3% 45.0% 65.0%
todos Todas las llaves comen chocolate.
All keys eat chocolate Todas las mariposas tienen alas.
All butterflies have wings Todas las vacas son café.
All cows are brown
% bien 10.0% 56.6% 38.3%
References
[1] Grice, 1975; [2] Guasti et al., 2005; [3] Huang & Snedeker, 2009; [4] Hurewitz et al, 2006;
[5] Noveck, 2001; [6] Papafragou & Musolino, 2003; [7] Foppolo et al., 2012; [8] Miller et al.,
2004; [9] Schmitt & Miller, 2004; [10] Vargas-Takuda et al., 2008; [11] Crain & McKee, 1985;
[12] Lanza, 1992; [13] Valdés, 2003; [14] Clark, 1987
200
¿Cómo pueden salir de la isla? Non-native speakers of Spanish and their interpretation and
production of wh- islands
Maria Turrero-Garcia
University of Massachusetts Amherst
The large body of research focusing on wh- movement in the fields of First and Second Language
Acquisition tackles different areas of both its production and its interpretation (de Villiers 1997, de
Villiers, de Villiers & Roeper 2005, Yamane 2003, Yusa 1995 et al.). Specifically for Spanish, research
on L1 has looked both into interpretation and production (Pérez-Leroux 1993, Gutiérrez 2006), finding
that children have a similar pattern of acquisition to that of other languages with the same movement
patterns such as English (Thornton 1990) or French (Oiry and Demirdache 2006). In the area of SLA,
most research focuses on the difference between movement and non-movement languages (Yamane
2003, Yusa 1995). The studies that focus on the acquisition of a movement language by native speakers
of another movement language tend to look into the differences between these languages (Dussias
2007, Montrul et al. 2008, Cuza 2011), and to the best of our knowledge no researchers have addressed
the issue of what happens with structures that have a similar syntactic behavior in both the L1 and the
target language.
For that reason, this research addresses the acquisition of the wh- island constraint in structures like
“¿Cuándo dijiste cómo iba a venir a la iesta?” “When did you say how [s/he] was going to come to the
party”) that has very similar characteristics in Spanish and English.
In order to investigate this phenomenon, a situation interpretation task (see example 1) was carried out.
18 adult L2 and 10 native speakers of Spanish participated in the experiment, which consisted of a
series of past narratives presented on a screen. After reading each situation, they were asked a question
that contained a wh- island orally, and they were asked to write down the answer to it.
A statistically significant difference was found between the experimental group and the control group
(p<,000). The experimental group responded to the medial wh- word in a 15% of occasions and they
gave a LD reading to the fronted wh- word 30% o the time (see example 2). The control roup’s
responses pattern with previous descriptions o adult Spanish speakers’ behavior in wh- questions
(Torrego 1984).
Non-native speakers of Spanish differ from both their L1 representation of island constraints and from
adult and children native speakers of Spanish. The data does not adhere to transfer accounts of L2
acquisition (White 2000, Clahsen and Hong 1995), as there is no evidence for a facilitation of the
acquisition o this structures comin rom the subjects’ L1; but the data cannot be attributed to the
native developmental path of Spanish either, as non-developmental errors occur in the L2 data that do
not occur during the course of L1 acquisition of Spanish (Pérez-Leroux 1993, Gutiérrez 2006), nor
during the course of acquisition of English, as English and Spanish pattern alike in the kind of
developmental errors that occur in wh- islands.
Example 1: Test item for interpretation experiment
Elena robó unas joyas anteayer y las enterró en su jardín para esconderlas. Después se
Helen stole some jewels before yesterday and CLITDO buried in her garden to hide-CLITDO Later CLIT
sintió culpable y fue a la iglesia a confesar que había escondido las joyas robadas en el jardín.
felt guilty and went to the church to confess that had hidden the jewels stolenfem pl in the garden
Helen stole some jewels the day before yesterday and she buried them in her backyard to hide them. Later she felt guilty and
she went to church to confess that she had hidden the stolen jewels in the backyard.
201
¿Dónde confesó cómo había escondido las joyas?
hWhere confessed how had hidden the jewels?
Where did she confess how she had hidden the jewels?
Example 2: Possible responses to target question:
¿Dónde le dijo Pedro al médico cuándo se había extendido el dolor?
Where clitdat said Pedro to-the doctor when clitref had spread the pain?
Where did Pedro tell the doctor when the pain had spread?
Medial response Long-distance to fronted wh- Target response
Al día siguiente The following day
Se había extendido al pecho It had spread to the chest
Se lo dijo en el hospital
He told him in the hospital
References
CUZ , ., “Trans er E ects at the Syntax-Semantics Interface: The Case of Double-que Questions in
Herita e Spanish”, in Herita e Lan ua e Journal 8, 2011
CL HSEN, H. and HONG, U. (1995). “ reement and null subjects in German L2 development
New evidence from reaction-time experiments”. Second Lan ua e Research 11.
DUSSIAS, P. “The e ect o exposure on syntactic parsin in Spanish-En lish bilin uals”, in
Bilingualism: Language and Cognition, vol. 10, issue 1, 2007
GUT ERREZ, M. J., “ cquirin lon -distance wh- questions in Spanish”, in V. Torrens, ed., The
acquisition of syntax in Romance languages, 2006, Amsterdam
MONTRUL, S., FOOTE, R. and ER Ñ N, S., “Knowled e of Whmovement in Spanish
L2 Learners and Heritage Speakers”. Selected roceedin s o the 10th Hispanic Lin uistics
Symposium, ed. Bruhn de Garavito, J. and Valenzuela, E., 2008
O RY, M. and DEM RD CHE, H., “Evidence rom L1 acquisition or the syntax o wh-scope
markin in French”, in V. Torrens, ed., The acquisition o syntax in Romance lan ua es, 2006,
Amsterdam
PEREZ-LEROUX, A. T., Empty categories and the acquisition of wh- movement, U. of
Massachusetts, Amherst, 1993
THORNTON, R. Adventures in long-distance movement: the acquisition of complex wh- questions,
doctoral dissertation, U. of Connecticut, 1990
TORREGO, E. (1984), “On nversion in Spanish and Some o ts E ects”, in Lin uistic nquiry, Vol.
15, No. 1
De V LL ERS, J. “Empty cate ories and complex sentences the case o wh- questions”, in . Fletcher
and B. McWhinney, ed. The handbook of child language, 1995, Oxford
De VILLIERS, J., De VILLIERS, L. and ROE ER, T. “ nswerin hard questions : Wh-movement
across dialects and disorder”, pub. in pplied sycholin uistics, vol. 29 issue 1, 2007
WH TE, L. (2000). “Second Lan ua e cquisition rom initial to inal state”. n rchibald, J. Second
Language Acquisition and Linguistic Theory. Malden, MA
YAMANE, M. On interaction of first-language transfer and universal grammar in adult second
language acquisition: WH-movement in L1-Japanese/L2-English interlanguage, doctoral dissertation,
U. of Connecticut, 2003
YUS , N. “Multiple-Specifiers and Wh- sland E ects in L2 cquisition”, in E. Klein and G.
Martohardjono, eds., The development of second language grammars: a generative approach, 1998,
Philadelphia
202
Prosody and question particles as evidential strategies in Majorcan Catalan
Maria del Mar Vanrell1, Meghan E. Armstrong
2 & Pilar Prieto
3
1Freie Universität Berlin,
2UMASS-Amherst,
3I-CREA-Universitat Pompeu Fabra
All languages have some way of marking information source, though different parts of the grammar
may be recruited to do so depending on the language. Lexical means for specifying source of
information is probably universal. For instance, the lexical item es veu is used in Catalan in (1) to
indicate that the speaker does not have firsthand knowledge of the proposition. Some languages have
true evidential systems (inflectional systems with morphemes that have source-marking at the core of
their semantics and are obligatory), whereas in other languages, evidentiality specifications are
“scattered” throu hout the rammar ( ikhenvald 2004). The only work we are aware o that has
examined prosodic marking of evidential strategies is Hara & Kawahara (2012), who showed that
deaccenting is used in Japanese biased polar questions when the speaker expects a positive answer
based on public evidence.
Majorcan Catalan (henceforth Maj_Cat) is a variety of Catalan with several strategies for forming
polar questions based on bias. Speakers may choose from different pitch accents as in (2) or may head
questions with question particles (QP) such as que (complementizer ‘that’) or o (conjunction ‘or’), as in
(3). We asked what role intonation might play in encoding information source in QP que + LH* L%
questions. If it is the case that information source is being encoded through the use of QP que + LH*
L%, we also asked what type of evidential information the speaker might be conveying.
We tested this experimentally using three different tasks: a) the Discourse Completion Test
(Billmyer & Varghese 2000), b) a contextualized acceptability task and c) a multiple choice task. In the
first task we created a set of situations that contained two evidential conditions (visual evidential and
non-visual evidential) plus a non-evidential/non-epistemic situation, which triggered a neutral (info-
seeking) question. 15 speakers of Maj_Cat participated in this production experiment. In the second
task (perception), 40 Maj_Cat listeners were asked to rate the acceptability of a question produced in a
specific context using a 7-point Likert scale. In the third task (perception), 40 Maj_Cat listeners were
asked to answer 2 multiple choice questions about the information source and the temporal information
related to the intonation and QP of some yes-no questions.
The results of the production experiment indicate that the rising-falling nuclear configuration LH*
L% is the most preferred contour used for polar questions produced in contexts for which the speaker
had visual evidence or the question’s propositional content (Fi ure 1). For experiment 2, Q que +
LH* L% questions produced in a visual context were rated most natural by listeners (Figure 2). We
argue that this combination of QP and intonation is used to mark sensory information source in polar
questions. In experiment 3, listeners inferred that the speaker had sensory information, and in addition
made an inference about when that information became available to the speaker -just prior to the time
of utterance (tu). Thus we present evidence that QP que + LH* questions function as a construction
(learned pairing of form + meaning – Goldberg 1996, 2006) conveying sentence modality, source-
marking and temporal information. Our results show that without recruiting a specific part of grammar
(intonation), source-marking and temporal information are no longer available to the listener.
Therefore, we confirm that intonation is indeed a part of the grammar available for conveying
evidential meaning, in this case working in tandem with the QP que.
(1) S’han quedat sense llum, a Girona. Es veu que hi ha nevat molt. (González 2012)
‘The power is out in Girona. There must have been a lot o snow’.
(2)Teniu mandarines?¡H+L* L% ‘Do you have tan erines?’ vs. Que hi ha gana?L+H* L% ‘ re you hun ry?’
(3) Que encara no ha vengut, s’electricista? ‘The electrician hasn’t arrived yet?’ vs. O no estàs bo?
‘ ren’t you well?’ ( rieto and Cabré (eds.) 2007-2012)
203
References
AIKHENVALD, A. Y. (2004) Evidentiality. Oxford: Oxford University Press. BILLMYER, K. & VARGHESE, M.
(2000). Investigating instrument-based pragmatic variability: Effects of enhancing discourse completion tests.
Applied Linguistics 21(4), 517-552. HARA, Y. & KAWAHARA, S. (2012) The prosody of public evidence in
Japanese: A rating study. In: Jaehoon Ch., Hogue, A.; Punske, J.; Tat, D.; Schertz , J.; Trueman, A. (eds.),
Proceedings of the 29th West Coast Conference on Formal Linguistics, 353-361. Somerville, MA: Cascadilla
Press. GOLDBERG, A. (1995). Constructions. Chicago: University of Chicago Press. GOLDBERG, A. (2006).
Constructions at work. Cambridge: Cambridge University Press. GONZÁLEZ, M. (2012). Indirect evidence in
Catalan: A case study. In: Payrató, Ll.; Cots, J.M. (eds.), The Pragmatics of catalan, 146-172. Berlin: Mouton de
Gruyter. Prieto, P. & Cabré, T. (eds.) (2007-2013). Atles interactiu de l'entonació del català.
[http://prosodia.upf.edu/atlesentonacio/]
0
20
40
60
0
oth
er
QP 0
oth
er
neutral nonvisual
visual
¡H+L* L%
H+L* L%
L+H* L%
other
0
10
20
30
40
50
¡HL*
qu
e_¡H
L*
qu
e_LH
*
¡HL*
qu
e_¡H
L*
qu
e_LH
*
auditory visual
1
2
3
4
5
6
7
Figure 1. Graph showing % of occurrence of different
combinations of prosodic markers (¡H+L* L%, H+L*
L%, L+H* L% and other) and lexical markers (0= no
lexical marker, QP = presence of question particle, other
= other syntactic mechanisms) for each evidential
condition (neutral, non visual and visual)
Figure 2. Graph showing the acceptability judgments
(7-point Likert scale) of different combinations of QP
and intonation (¡H+L* L%, que_¡H+L* L% and
que_L+H*) with different evidential conditions
(auditory or visual).
204
Clitic directionality in Andalusian and Asturian Spanish infinitival imperatives
Julio Villa-García
Villanova University
I present And(alusian) and Astur(ian) Sp(anish) data that show a contrast in clitic directionality in 2nd
-
person plural imperatives displaying infinitival verb forms: whereas positive imperatives involve
postverbal clitics in both dialects (cf. (2)), negative imperatives involve enclisis in AndSp (cf. (3)) and
proclisis in AsturSp (cf. (4)). I argue for a PF-merger+copy-and-delete approach à la Miyoshi (2002)
and Bošković (2001 et seq.), i.a., whereby imperatives involve a null F head which is an affix that must
merge with a host under PF adjacency.
Positive infinitival imperatives (cf. (1a)) are attested in general colloquial, spoken Spanish, but
negative variants such as (1b) are restricted to dialects including AndSp and AsturSp; other dialects,
such as standard Spanish, use (surrogate) subjunctive forms in negative contexts (cf. (13)). What is
puzzling is that when we introduce clitics into the picture, we find an asymmetry in terms of clitic
directionality between negative infinitival imperatives in AndSp, which display the neg.+inf.+cl. word
order (cf. (3a)), and their AsturSp counterparts, which exhibit the NEG.+cl.+inf. word order (cf. (4a)),
with stress on negation, which is represented by the use of upper-case letters. This contrast has so far
gone unnoticed in the literature.
Two major competing accounts will be considered here. On the one hand, Kayne’s (1992 et seq.)
analysis of a parallel construction in Italian assumes that the negative head no ‘not’ selects a null modal
to which the clitic climbs, yielding the neg.+cl.+inf. word order, as in (5). The problem with this
approach for the AsturSp case is that clitic climbing is optional with standard-issue modal+inf.
constructions, which raises the question of why it is obligatory in this particular dialect (cf. (4a)/(5)).
Similarly, a different syntax would have to be assumed for the AndSp case (cf. (3a)). Finally, the
infinitival imperatives in question are rather high in the structure (cf. (6) vs. (7)), much like adult root
infinitives (Grohmann and Etxepare 2003), contrary to what (5) assumes. On the other hand, the
account pioneered by Miyoshi (2002) adopts (i) the proposal that imperatives involve a null
F(unctional) head (possibly C) whose affixal nature requires PF merger (i.e. affix hopping) with a host
(i.e. a verb/prosodic word) under adjacency and (ii) Franks’ (1998) su estion that a low copy of a
non-trivial chain can be pronounced provided that convergence so demands. In this paper I argue for
Miyoshi’s analysis.
On this view, the sentences in (2) receive the simplified derivation in (9a), which features
imperative F, and a copy of the clitic above and below the verb. Under Frank’s proposal, the hi hest
copy of the clitic chain is pronounced unless this pronunciation causes a PF violation. To illustrate this
approach, consider Romanian (8). In this language, all wh-items must be fronted, as in (8a). However,
there is a constraint against homophonous forms, exemplified in (8b). In order to avoid the ensuing PF
violation in such cases, the low copy of the whitem is chosen in PF, as in (8c). Returning now to
Spanish, recall that F must be adjacent to the verb/prosodic word. In (2), the PF adjacency requirement
between the null head F and the verb is met as long as the low copy of the clitic is favored in PF (cf.
(2a)), as in (9b), which instantiates the pronounce-a-low-copy-to-avoid-a-PFviolation approach (cf.
(8c)). However, if the clitic were to surface preverbally (i.e. if the high copy of the clitic were
pronounced), PF adjacency between F and the verb would be disrupted, yielding the ungrammatical
(2b), which includes a stranded PF affix (i.e. F) without a host, thus incurring a PF violation, as shown
in (9c).
205
As for negative infinitival imperatives, I propose that the negative head in such cases raises to the
head F in both dialects, as in (10a)/(11a). In AndSp (cf. (3)), negation is not stressed and does not
constitute a prosodic word; hence, it cannot properly support F phonologically. In this case, only the
verb can serve as a host. Thus, a low copy of the clitic is pronounced, since F can only hop onto the
verb provided that the two entities are PF adjacent (cf. (10b)). Pronouncing the high copy of the clitic
would disrupt adjacency between F and the verb (cf. (10c)), resulting in an illegitimate stranded PF
affix (i.e. F) (cf. (*3b)). In AsturSp (cf. (4)), by contrast, NO bears stress, making it a host that can
support F. Therefore, F does not need to hop onto the verb. Since nothing goes wrong in PF if the
highest copy of the clitic is pronounced, the high copy is chosen (cf. (11b)). If a low copy of the clitic
were favored in PF instead (cf. (11c)), the derivation would crash (cf. (*4b)), which is now explained
as a violation of Last Resort, since there would be no PF reason to pronounce the low copy of the clitic
in place of the high one. It is of note that this account allows us to treat the two dialects in exactly the
same way syntactically (cf. (2)-(4)). Put differently, there is no need to posit a different syntax between
ndSp and sturSp in initival imperatives, in line with Bošković’s (2001 et seq.) approach to PF word
reordering, on which PF considerations are allowed to affect word order but without actual PF
movement, much like in (8).
Lastly, the analysis currently pursued correctly predicts that negative imperatives cannot be
accomplished via a preverbal negative constituent (e.g. nunca ‘never’) in either dialect (c . (12)), since
such cases do not involve head movement of negº to Fº (note that such configurations are licit with
subjunctive forms, as in (13)). As expected, the counterparts of (12a,b) with Xº negation and postverbal
nunca are grammatical (cf. (14a,b)).
(1) a. ¡Venir! (comeinf. – ‘Come2.PL!’) b. ¡No venir! (not comeinf. - ‘Don’t2.PL come!’)
(2) AndSp and AsturSp 2-pl. positive imperatives
a. ¡Llamarla! (callinf. cl. – ‘Call2.PL her!’) b. *¡La llamar! (cl. callinf.)
(3) AndSp 2-pl. negative imperatives
a. ¡No llamarla! (not cl. callinf. - ‘Don’t2.PL call her!’) b. *¡No la llamar!
(4) AsturSp 2-pl. negative imperatives
a. ¡NO la llamar! (not cl. callinf. - ‘Don’t2.PL call her!’) b. *¡NO llamarla!
(5) [neg. [cl. + modal (…) [VP infinitive cl.]]] (cf. (4a))
(6) a. *¡A menudo comprar manzanas! b. ¡Comprar a menudo manzanas!
often buyinf. apples buyinf. often apples
Both ‘Buy2.PL apples o ten!’
(7) a. Pedro a menudo compra manzanas b. Pedro compra a menudo manzanas
Peter often buysindic. apples Peter buysindic. often apples Both:
‘ eter o ten buys apples.’
(8) a. Cine ce precede? [Romanian, rom Bošković 2002] who what precedes
‘Who precedes what?’
b.*Ce ce precede? what what
precedes
‘What precedes what?’
c. Ce precede ce?
what precedes what
(9) Positive 2-pl. infinitival imperatives in both AndSp and AsturSp
a. F la llamar la b. F la llamar la c. * F la llamar la
(cf. (2)) (cf. (2a)) (cf. (2b))
206
(10) Negative 2-pl. infinitival imperatives in AndSp
a. no F la llamar la b. no F la llamar la c. * no F la llamar la
(cf. (3)) (cf. (3a)) (cf. (3b))
(11) Negative 2-pl. infinitival imperatives in AsturSp
a. NO F la llamar la b. NO F la llamar la c. * NO F la llamar la
(cf. (4)) (cf. (4a)) (cf. (4b))
(12) a. *¡Nunca llamarla! (never callinf. cl. – ‘Don’t2.PL ever call her!’) [AndSp]
b. *¡Nunca la llamar! (never cl. callinf. – ‘Don’t2.PL ever call her!’) [AsturSp]
(13) ¡Nunca la llaméis! (never cl. callSubj. – ‘Don’t2.PL ever call her!’) [Standard Sp]
(14) a. ¡No llamarla nunca! (not callinf. cl. never – ‘Don’t2.PL ever call her!’) [AndSp]
b. ¡NO la llamar nunca! (not cl. callinf. never – ‘Don’t2.PL ever call her!’) [AsturSp]
207
Production differences between Spanish-dominant and Heritage speakers reflect a mapping
deficit
Alvaro Villegas
Penn State University
Paola E. Dussias
Penn State University
Kara Morgan-Short
University of Illinois at Chicago
Past studies have reported differences between monolingually raised and Spanish heritage speakers in
the production and interpretation of the Spanish mood system (e.g., Montrul, 2007; Silva-Corvalán,
1994). Yet, few studies have examined whether heritage speakers are sensitive to the presence of the
subjunctive mood during online processing. This gap is significant given that past research has
demonstrated that L2 learners are able to process some aspects of L2 syntax even when their accuracy
in the production of such structures is near chance (Tokowicz & MacWhinney, 2005). The goal of this
study, then, is to examine the production and online processing of the Spanish subjunctive in heritage
speakers.
Twenty-six second generation heritage speakers from a large US institution were recruited. They
reported having acquired Spanish from birth, having switched dominance to English later in their
childhood, and using Spanish in their daily lives with friends and family. Additionally, 39 L1 Spanish–
L2 Catalan speakers living in Tarragona (Spain) were also recruited as a bilingual control group (L1
Spanish-dominant group hereafter). They reported having acquired Spanish from birth, being dominant
in Spanish, and using Spanish on a daily basis with friends and family. Participants completed a
language history questionnaire and a battery of language proficiency tests (to tap into linguistic
knowledge), a computerized sentence completion task to measure productive use of the subjunctive,
and an eye-tracking study to investigate online processing of the subjunctive. For the sentence
completion task, participants were presented with 48 main clauses and were asked to complete them
freely. Sixteen sentences were experimental sentences that contained verbs in Spanish (e.g.,
aconsejar/suggest) that obligatorily subcategorized for the use of the subjunctive mood. To examine
online processing of the subjunctive mood, participants were presented with the same verbs that
obligatorily subcategorized for the presence of the subjunctive mood (e.g., aconsejar) in sentences that
were syntactically ambiguous between a sentence-complement reading (i.e., the subjunctive condition;
Condition 1, see page 2) and a relative clause reading (i.e., the indicative condition; Condition 2, see
page 2). The ambiguity was resolved via the morphological marking of the verb in the first subordinate
clause (i.e., comieron/comieran).
If participants access lexical information associated with the main verb (aconsejó) that signals the
presence of a subjunctive verb in the embedded clause, they should show longer fixation durations at
the verb in the first subordinate clause (in bold below) in Condition 2 than in Condition 1, because the
verb in the indicative condition does not meet the lexical requirements of the main verb.
A one-way ANOVA examining correct production of the subjunctive showed that heritage speakers
produced verbs in the subjunctive form significantly less accurately than the L1 Spanish-dominant
group (heritage: M=9.31, SD=5.01; L1 Spanish dominant: M=14.79, SD=1.54; F(1,63)=41.21,
p=.000), replicating past findings in the literature. For the eye-tracking study, a 2(subjective v.
indicative) x 2(L1 Spanish dominant v. heritage) ANOVA revealed no significant differences in early
and late measures of processing (i.e., gaze duration and total reading times) between the heritage
speakers and the L1 Spanish-dominant group (gaze duration: p=.922; total time; p=.246). Both groups
208
read the first verb in the subordinate clause slower in the indicative form than in the subjunctive form
(gaze duration: M= 584.69, SD= 21.96 for the indicative; M= 564.64, SD= 21.16 for the subjunctive;
total reading times: M=1025.72, SD= 40.28 for the indicative; M= 892.63, SD= 39.17 for the
subjunctive). Our results show that heritage speakers are like L1 Spanish-dominant speakers when
processing the subjunctive, despite the fact that their production is not monolingual-like (although the
production of the L1 Spanish-dominant group is). Results suggest that the source of production
difficulties reported in past studies with heritage speakers comes about in the mapping of linguistic
knowledge onto the production mechanism.
Materials
El general les aconsejó a los soldados que comieran carne y que bebieran cerveza.
The general suggested the soldiers that they eat meat and drink beer.
Condition 2- Indicative condition:
El general les aconsejó a los soldados que comieron carne que bebieran cerveza.
The general suggested the soldiers who ate meat that they drink beer.
References
Montrul, S. (2007). Interpreting mood distinctions in Spanish as a heritage language. In Spanish in
Contact: Policy, Social and Linguistic Inquiries, ed. K. Potowski and R. Cameron, 23-41.Philadelphia:
John Benjamins.
Silva-Corvalán, C. (1994). Language contact and change: Spanish in Los Angeles. New York: Oxford
University Press.
Tokowicz, N., & MacWhinney, B. (2005). Implicit and explicit measures of sensitivity to violations in
second language grammar: An event-related potential investigation. Studies in Second Language
Acquisition 27, 173–204.
209
Regional Dialectology: Vowel Duration of Mexican Spanish Prosody
Shontael Wanjema
The Ohio State University
In the history of regional dialectology of Mexican Spanish there is precious little known about
variation in the vowel system and prosody. As early as 1938 salient differences were noticed by
linguists concerning vowel duration and the cadence (prosody) of central Mexican Spanish. Henríquez
Ureña (1938) notes that, “es característica la cadencia inal de la rase enunciativa, muy distinta de la
cadencia usual en Castilla (Madrid)…la penúltima es lar a, la inal muy breve.” “The inal cadence is
very different rom Castillian (Madrid)…the penultimate is lon and the inal very short.” The current
study is an acoustic analysis of the production of vowel duration and the prosody of multisyllabic
words.
Ten native speakers of Mexican Spanish living in the US from two different geographic regions
in Mexico (south 3, coast 1 and central 6, all male) were recruited and interviewed in Columbus Ohio.
Each participant provided limited demographic and social information and answered questions about
Mexican culture.
Production data was extracted from participant answers to the questions about Mexican culture.
Analysis was conducted on multisyllabic words whose lexical stress is on the penultimate syllable. The
production data was automatically aligned, duration was automatically extracted in Praat and the
duration ratio between the ultimate and penultimate syllable was calculated for each word.
Production differences suggest three patterns for vowel duration in this context: speakers who
have a longer penultimate than ultimate vowel (central), speakers who have nearly equal duration on
both vowels (central, coast), and speakers who have a longer ultimate than penultimate (south).
Mixed effects linear regression models were fit with duration ratio as the dependent variable.
Predictors were If the word occurs at an Intonational Phrase (IP) boundary, Open or Closed Syllable,
speaker region, and an interaction between IP and syllable with random effects of speaker, word, and
speaker+IP. The final model included Speaker Region (south, central) and random effects of speaker
and word. Speaker region south was a significant predictor of vowel duration (p=.02, n=80).
This acoustic analysis of salient prosodic variation provides a detailed description of a previously
observed linguistic phenomena and shows a correlation with geographic region.
Reference
Henríquez Ureña, edro. 1938. “Mutaciones articulatorias en el habla popular”, en El español en
Méjico, los Estados Unidos y la América Central. Buenos Aires: Universidad de Buenos Aires,
pp. 329-379.
210
The production of Spanish stop-rhotic sequences by L2 learners
Christine Weissglass
Carolina González
Florida State University
Previous research regarding the production of Spanish rhotics by non-native speakers has primarily
focused on intervocalic position (e.g., Face, 2006; Major, 1986; Rose, 2010 but cf. Colantoni & Steele,
2006). The present study aims to expand our understanding of the ways in which non-natives
pronounce Spanish rhotics in a different context, namely stop-rhotic sequences (e. ., ‘prisa’ hurry).
In Spanish stop-rhotic sequences, the rhotic is typically realized as a tap [ɾ] or an approximant [ɹ] and
a vocalic element of variable duration known as a svarabhakti vowel (henceforth, SV) frequently
emerges between the two consonants (Blecua, 2001; Hualde, 2005; Weissglass, 2011). In comparison,
in English stop-rhotic sequences (e. ., ‘problem’) the rhotic tends to be realized as an approximant [ɹ]
or a ricative [ř] and SVs are not present (Colantoni & Steele, 2006). revious research has shown that
native English speakers learning Spanish (henceforth, L2 learners) experience difficulty in achieving
native-like pronunciation of stop-rhotic sequences. For example, Colantoni & Steele (2006) found that
only 1 out of 10 L2 learners exhibited native-like pronunciation of these sequences in real Spanish
words.
The present study examines acoustically the production of stop-rhotic sequences in 6 Spanish non-
words (Table 1) by 26 advanced L2 learners. Three native Spanish speakers served as a control group
(henceforth, CG). Non-words were used to ensure that familiarity with the words did not impact
pronunciation. Additionally, the use of non-words facilitated the investigation of the role of place of
articulation and voicing of the preceding consonant. The tokens were integrated into a list of 47
Spanish words and non-words. Participants were asked to pronounce each item twice while being
recorded in a laboratory setting. Each participant produced a total of 12 tokens (6 tokens*2 repetitions).
A total of 348 tokens (12 tokens*29 speakers) are analyzed acoustically using Pratt (Boersma &
Weenink, 2012). The independent variables are (i) stop place of articulation (bilabial, dental, velar) and
(ii) stop voicing (voiced, voiceless). The dependent variables are (i) rhotic realization (tap,
approximant, fricative, trill) and (ii) SV occurrence (present, absent). Examples of each rhotic
realization as produced by an L2 learner are provided in Figures 1-4.
Chi-square tests conducted in SPSS (IBM Corp., 2012) for 6 of the L2 learners and the CG revealed
that the L2 learners produced significantly more fricatives than the CG, which in turn produced
significantly more approximants than the L2 learners (p=.001). The L2 learners produced more taps
(53%) than fricatives (26%), approximants (17%) and trills (4%) and the CG produced more taps (64%)
than approximants (36%). Furthermore, the L2 learners produced significantly less SVs than the CG
(67% vs. 100%) (p<.001). For the L2 learners, approximants were significantly more common after
voiced stops whereas fricatives were significantly more common after voiceless stops (p=.001). After
voiced stops, the L2 learners produced more taps (56%), than approximants (31%), fricatives (8%) and
trills (5%). They produced more taps (50%), than fricatives (44%), approximants (3%) and trills (3%)
after voiceless stops. Finally, the L2 learners produced significantly more SVs after voiced stops than
after voiceless stops (86% vs. 47%) (p<.001). There were no significant effects for stop voicing for the
CG. Stop place of articulation was not a significant factor for either group.
These preliminary results suggest that L2 learners exhibit transfer from English resulting in the
realization of the rhotic in Spanish stop-rhotic sequences as the ricative [ř]. This indin is in line with
theories of L2 phonological acquisition that highlight the relevance of transfer such as the Speech
211
Learning Model (Flege, 1987), which claims that learners begin L2 phonological acquisition with the
phonetic categories that exist in their native language. In this talk, we will discuss results from a total of
26 L2 learners and the implications that they have for the Speech Learning Model and other theories of
L2 phonological acquisition.
212
References
Blecua, B. 2001. Las vibrantes en español: manifestaciones acústicas y procesos fonéticos. Ph.D. diss., Universitat Autónoma de Barcelona.
Boersma, P. & Weenink, D. 2012. Praat: doing phonetics by computer. v. 5.3, http://www.praat.org.
Colantoni, L, & Steele, J. 2006. Native-like attainment in the L2 acquisition of Spanish obstruent-liquid
clusters. Selected proceedings of the 7th CASPSLP, 59-73.
Face, T. 2006. Intervocalic rhotic pronunciation by adult learners of Spanish as a second language.
Selected proceedings of the 7th CASPSLP, 47-58.
Fle e, J. 1987. The production o “new” and “similar” phones in a orei n lan ua e Evidence or the
effect of equivalence classification. Journal of phonetics, 15, 47-65.
Hualde, J. 2005. The Sounds of Spanish. Cambridge University.
IBM Corp. 2012. IBM SPSS Statistics for Windows, Version 21.0. Armonk, NY: IBM Corp.
Major, R. 1986. The ontogeny model: Evidence from L2 acquisition of Spanish r. Language learning,
36, 453-504.
Rose, M. 2010. Intervocalic tap and trill production in the acquisition of Spanish as a second language.
Studies in Hispanic and Lusophone linguistics, 3, 379-419.
Weissglass, C. 2011. An acoustic study of rhotics in onset clusters in La Rioja. Romance Linguistics
2010, 49-62.
213
Towards a typology of exclamative phrases: Evidence from the Spanish expression No me digas
Zachary Wilkins
UNC Chapel Hill
There exists in Spanish a set of expressions in which the notion of not telling expresses a
heightened interest on the part of one of the interlocutors towards a certain piece of information in the
discourse, notably No me digas (≈‘Don’t tell me’). Such an expression o ten makes a discourse
contribution comparable to so-called “exclamative” phrases (e. . Wow!). As Zannuttini and Portner
(2003) have noted, this comprises a category heavily under-studied in all languages, especially in
Spanish. Close examination of this Spanish phrase helps lend a better understanding of the different
forms that exclamative expressions may take and aims to make a contribution to typological studies of
this clause type.
This paper analyzes the pragmatic function of the phrase No me digas (hence orth, “the phrase”) as in
(1) and asserts its similarity in behavior to “wh-exclamatives” (e. .
What a surprise!, ¡Qué increíble!) (Castroviejo, 2008 and Andueza & Gutiérrez-Rexach, 2010). The
dialectal focus of this study is on the Spanish of the River Plate region, yet as my results suggest, the
phenomenon can likely be applied to the rest of the Spanishspeaking world.
(1) A: Tu nieto salió campeón en la competencia. [≈Your grandson
placed first in the competition.] B: ¡No me digas! Qué bueno.
[≈lit. Don’t tell me! That’s great.]
My results show that in fact No me digas is used non-literally more often than literally, and that the
pragmatic function is almost always lost when it is syntactically modified. In addition, my research
proves that this expression behaves similarly to a whexclamative in at least two of the criteria
established by Zannuttini and Portner (2003). The failure in my application of the phrase against their
third test calls into question the presence of factivity that linguists have often associated with
exclamative phrases. Unless otherwise cited, all of the following examples are gleaned from Mark
Davies’ Corpus del Español which provides an orthographic transcription of spoken dialogue between
two interlocutors from most major cities across the Spanish-speaking world (Davies, 2002).
I will divide the analysis into three parts. First, I will identify the inherent properties of the
phrase by testing how it behaves in several syntactic environments and run a quantitative analysis of a
corpus data set, establishin the phrase’s pra matic function as distinctly exclamative. Before applying
Zannuttini and ortner’s model, will do a neo-Gricean analysis of the pragmatic function to show that,
certainly for the dialect in question, the phrase triggers a generalized conversational implicature
indicating behavior equivalent to that of a wh-exclamative (Grice, 1975 and Horn, 2004). Finally, I will
dedicate a section to applying the test proposed by Zanuttini and Portner (2003) to discriminate
exclamative phrases and analyze why these tests and No me digas interact in the way they do.
References
Alonso Cortés, Ángel (1999). La exclamación en español: estudio sintáctico y pragmático. Madrid:
Minerva Ediciones, S.A.
Andueza, Patricia and Gutiérrez-Rexach, Javier (2010). “Ne ation and the nterpretation o Spanish
Rhetorical Exclamatives.” Selected Proceedings of the 12th Hispanic Linguistics Symposium, ed. by
Claudia Borgonovo et. al. Somerville, MA. 17–25.
214
Austin, John (1962). How to do things with words. New York: Oxford University Press.
Castroviejo, Elena (2008). “Deconstructin exclamations.” Catalan Journal of Linguistics 7. 41-90.
Bajo érez, Elena (2000). “Suscribo rotundamente lo que estoy diciendo. No me adhiero en absoluto.”
Cuestiones de actualidad en lengua española, ed. by J. Borrego et al. University of Salamanca/Instituto
Caro y Cuervo. 23-31.
Davies, Mark. (2002-) Corpus del Español: 100 million words, 1200s-1900s. Available at
http://www.corpusdelespanol.org.
Ducrot, Oswald (1980). Les échelles argumentatives. Paris, les Éditions de Minuit.
Grice, aul (1975). “Lo ic and Conversation.” Studies in the Way of Words. Cambridge, MA: Harvard
University Press. 22-40.
Grimshaw, Jane (1979). “Complement Selection and the Lexicon”. Linguistic Inquiry 10: 279-326.
Horn, Laurence (2004). “ mplicature.” Handbook of Pragmatics, ed. by L. Horn and G. Ward. Malden,
MA: Blackwell. 3-28.
----- (1984). “Towards a new taxonomy o pra matic in erence Q-based and R-based implicature.”
Meaning, form and use in context, ed. by Deborah Schiffrin. Washington, DC: Georgetown University
Press. 11-42.
Levinson, Stephen (2000). Presumptive Meanings: The Theory of Generalized Conversational
Implicature. MIT Press, Cambridge, MA.
Real Academia Española (2001). Diccionario de la lengua española (2001). Vigésima segunda edición.
Madrid, España.
Searle, John (1965). “What is a speech act?” Philosophy in America, ed. by Max Black. Cornell
University Press, Ithaca. 221–239.
Zanuttini, Ra aella, and ortner, aul. (2003). “Exclamative clauses at the syntaxsemantics inter ace.”
Language 79. 39-81.
215
Spanish Dative Arguments in Passive Constructions
Nakyung Yoon
UIUC
Introduction: This paper offers an account of the unacceptability and acceptability of certain datives in
the periphrastic passives. Demonte (1995) observes that core (argumental) dative clitics can appear in
the passive, while non-core (non-argumental) datives cannot, as in (1) and (2). In this paper, I argue
that core and non-core datives should receive a distinct analysis, in order to account for their distinct
behaviors in the periphrastic passives. In concrete, I argue that the passivization, the nominalization and
the sluicing data in ditransitives and monotransitives demonstrate that the argument structure properties
of core and non-core datives are different and that certain non-core datives can be passivized via
complex predicate formation.
Background: Gutiérrez (1999) observes the different acceptability with respect to the verb types: the
acceptability of the nominalization of ditransitives, such as (3) and the unacceptability of
nominalization of monotransitives, such as (4). This suggests that the argument structure properties of
core and non-core datives are distinct. On an applicative account such as Cuervo’s these data are
problematic. Pujalte (2010) observes that ditransitives can license implicit datives, as in (5), but
monotransitives cannot, as in (6). This also suggests that the argument structure properties of core and
non-core datives are distinct. Cuervo’s account cannot make a ri ht prediction because all the datives
are applied arguments, which will alter argument structure, and because there is no distinction between
the structure of core and non-core datives.
Previous account ujalte ar ues that Cuervo’s account comes up short with respect to the data
discussed above, and o ers the ollowin proposal. ujalte’s approach to core and non-core datives
discusses that the core dative argument of a ditransitive predicate is directly associated to the verbal
Root, while non-core datives are introduced by a low applicative head. Specially, she assumes that the
low applicative head (Appl) which introduces non-core datives is fully defective and values structural
Case by feature inheritance from v; regarding unnacusatives where v lacks φ-features, Case is inherited
by T. In both unaccusatives and passives, there is no involvement of accusative Case assignment. Thus
the same Case analysis should hold for passives. Her analysis does not predict that the sentences in (2)
should be ungrammatical.
Present approach: I claim that passives can be explained by recourse to the argument structure, on the
assumption that the passive operation can apply to both a simple and a complex predicate in Spanish in
a way that the passive operation suppresses the accusative Case assignment ability (Osvaldo 1986 and
Baker et al. 1989). I also suggest here that certain complex predicate formation of the periphrastic
passive forms a very close unit for argument structure projections, thus generally there is no
involvement of movement for syntactic properties such as Case assignment. As illustrated in (7) from
Masullo (1992), monotransitives with possessor dative can undergo passivization in which the theme of
the original base verb is postverbal, as in (7b) and (7c), but they cannot in such way that the theme of
the original base verb is preverbal, as in (7a). Manning (1992), however, points out that the complex
predicate formation with periphrastic construction such as terminar de ‘ inish’+ in initive can under o
passivization in which the position of the theme of the original verb is preverbal, as in (8a). This seems
challenging to the current analysis because the theme undergoes movement from the original position,
unlike that o the verb tomar ‘take’ with possessor datives. This can be explained by the assumption
that the theme can undergo movement to the preverbal position if the element is focalized, as in (8a). In
this way, the Case assignment is realized in the same way as with both sentences, and thus the analysis
can conform Spanish to the present passivization approach.
216
(1) a. El premio Nobel (le) fue concedido a Cela.
the prize Nobel CL- DAT was awarded to Cela
‘The Nobel prize was awarded to Cela last year’
(2) a. *La casa le fue pintada a Juan anteayer. the house CL-DAT was
painted to Juan the day before yesterday
‘The house was painted or Juan the day be ore yesterday’
b. */La bicicleta le fue robada a Pablo ayer.
The bicycle CL-DAT was stolen to Pablo yesterday
‘ ablo’s bicycle was stolen yesterday’
c. *Los hijos le fueron admirados a Carolina en la escuela.
The children CL- DAT were admired to Carolina at the school
‘Carolina’s children were admired at school’
(3) a. El envío de un salmon ahumado en Uga a sus amigos.
‘The shippin o a smoked salmon in U a to his/her riends’
(4) a. ?/* El lavado de la ropa a Pepe.
‘The clothes washin or epe’
b. ?/*La copia del examen a la compañera.
‘The copy o the exam to the classmate’
(5) a. Juan mandó la carta, pero no sé a quién (le mandó la carta).
Juan sent the letter, but not know to who CL-DAT-3SG sent the letter
‘Juan sent a letter, but don’t know who to’
b. Juan entregó la hoja, pero no sé a quién (le entregó la hoja).
Juan submitted the paper, but not know to who CL-DAT-3SG sent the paper.
‘Juan submitted the paper, but don’t know who to’
(6) a. *Juan cortó el pasto, pero no sé a quién (le cortó el pasto).
Juan cut the grass, but not know to who CL-DAT-3SG cut the grass.
‘Juan cut the rass, but don’t know who to’
b. *Juan contruyó la casa, pero no sé a quién (le contruyó la casa).
Juan built the house but not know to who CL-DAT-3SG built the house ‘Juan
built the house, but don’t know to who’ (7) a. *El pulso le ue tomado al paciente.
‘The patient’s pulse was taken’
b. Le fue tomado el pulso al paciente.
c. Al paciente le fue tomado el pulso.
(8) a. Estas paredes les están siendo terminadas de pintar a los dueños.
these walls CL-DAT-3PL are being finished to paint for the landlords
‘These walls are bein inished to paint or the landlords’
Selected References
Cuervo, Cristina (2003). Datives at large, doctoral dissertation, Massachusetts: MIT. Demonte,
Violeta (1995). Dative alternation in Spanish, Probus 7:5-30. Masullo, Pascual (1992). Incorporation
and the Case Theory in Spanish: A crosslinguistic Perspective, Doctoral dissertation, University of
Washington. Pujalte, Mercedes (2010). (Non)added Datives in Spanish. Presented at NELS 41.
217
A variationist account of trill /r/ usage in the Spanish of Málaga
Sara Zahler
Danielle Daidone
Indiana University
The purpose of this study is to analyze trill variation in Andalusian Spanish through a corpus of 12
sociolinguistic interviews with speakers from Málaga, Spain in order to address two main research
goals. The first is to provide a description of trill frequency in this dialect. This question is pertinent
due the high amount of variability observed in trills across Spanish dialects (Canfield, 1962; Cárdenas,
1958; Navarro Tomás, 1977, etc.) and within dialects as well as within individuals (e.g. Willis, 2006).
The second research goal is to determine which factors constrain said variation of the /r/. Previous
research has demonstrated that various linguistic and social factors can condition use of the trill variant
in comparison to other realizations (e.g. Adams, 2002; Diaz-Campos, 2008; Diaz-Canseco, 1997;
Lastra & Martín Butragueño, 2006).
Using 12 sociolinguistic interviews with Spanish speakers from Málaga (Múñoz, Manuel, Lasarte
& Villena, 2008), 6 men and 6 women divided into 3 age groups, 963 occurrences of the trill were
extracted, acoustically analyzed with Praat (Boersma & Weenink, 2012), and coded as either the
standard alveolar trill production consisting of two or more occlusions or as a non-canonical variant.
The tokens were subsequently coded according to thirteen linguistic factors (preceding segment point,
preceding segment mode, following segment point, following segment mode, word position, lexical
accent, number of syllables, grammatical category, lexical frequency, corpus frequency, number of
phonological neighbors, number of high frequency phonological neighbors, and average frequency of
phonological neighbors) and two social factors (gender and age). The data were analyzed using
Goldvarb X (Sankoff, Tagliamonte & Smith, 2005) and R (Ihaka & Gentleman, 1996).
Results show that variants other than the canonical trill with 2 or more occlusions account for more
than half of the data (56.1%), although an alveolar trill with 2+ occlusions is the single most common
realization (43.9%). Five linguistic factors condition the variation: the point of articulation of the
following segment, word position, lexical frequency, corpus frequency, and number of phonological
neighbors. Trill phonemes followed by central and posterior vowels (26% of the data) slightly favor the
standard trill variant, whereas following anterior vowels (74% of the data) disfavor it. Consequently,
frequency plays a particularly important role: while the most frequent segments (/r/ + anterior vowels)
favor the reduced non-canonical variants, all other types of frequency (lexical, corpus, and number of
phonological neighbors) correspond to a higher use of the alveolar trill. Reasons for this patterning will
be discussed in greater detail. While not statistically significant, other linguistic constraints present
expected tendencies. With regard to social factors, women and older speakers prefer the trill, providing
evidence that it can be considered a prestigious variant. Overall, this study contributes to the body of
knowledge on the significant variation in trill productions evidenced throughout the Spanish-speaking
world, while also addressing the complications of the interaction of frequency with other linguistic
factors.
218
References
Adams, C. (2002). Strong assibilation and prestige: a sociolinguistic study in the central valley of Costa
Rica. Diss., University of California Davis.
Boersma, P., & Weenink, D. (2012). Praat: doing phonetics by computer (Version 8.4.10). Retrieved
from www.praat.org
Bradley, T. & Willis, E. (2008). Rhotic contrast and neutralization in Veracruz Mexican Spanish. Paper
presented at Linguistic Symposium on Romance Languages. Urbana-Champaign, Indiana.
Canfield, D. (1962). La pronunciación del español en América. Bogotá: Instituto Caro y Cuervo.
Cárdenas, D. (1958). The geographic distribution of assibilated R and RR in Spanish America. Orbis.
7:407-410.
Díaz-Campos, M. (2008). Variable production of the rrill in spontaneous speech: Sociolinguistic
implications. In L. Colantoni & J. Steele (Eds.), Selected Proceedings of the 3rd Conference on
Laboratory Approaches to Spanish Phonology (pp. 47-58). Somerville, MA: Cascadilla
Proceedings Project.
Diez-Canseco, S. (1997). Language variation: the influence of speakers’ attitudes and gender on
sociolinguistic variables in the Spanish of Cusco, Peru. Ph.D. Dissertation: University of
Pittsburgh.
Hammond, R., (2000). “The multiple vibrant liquid in U.S. Spanish.” n . Roca’s (ed.) Research on
Spanish in the U.S. Somerville, MA: Cascadilla Press, 333-347.
Henriksen, N. and Willis, E. (2010). “ coustic characterization o phonemic trill production in
Jerezano ndalusian Spanish.” n Orte a-Llebaria, M. (Ed.) Selected Proceedings of the 4th
Conference on Laboratory Approaches to Spanish Phonology, 115-127. Somerville, MA:
Cascadilla Proceedings Project.
Ihaka, R., & Gentleman, R. (1996). R: A language for data analysis and graphics. Journal of
computational and graphical statistics, 5(3), 299-314.
Lastra, Y. & Martín Butragueño, P. (2006). Un posible cambio en curso: el caso de las vibrantes en la
ciudad de México. In A.M. Cestero Mancera (Ed.), Estudios sociolingüísticos del español de
España y América. Madrid: Arco Libros.
Múñoz, A., Manuel, A., Lasarte, M., & Villena, J.A. (eds.) (2008) El español hablado en Málaga II.
Málaga: Sarriá
Navarro Tomás, T. (1977). (19th ed.) Manual de pronunciación española. Madrid: Consejo Superior de
Investigaciones Científicas.
Sankoff, D., Tagliamonte, S. A., & Smith, E. (2005). "Goldvarb X: A multivariate analysis
application." Dept. of Linguistics, Univ. of Toronto.
http://individual.utoronto.ca/tagliamonte/Goldvarb/GV_index.htm
Willis, E. (2006). Trill variation in Dominican Spanish: an acoustic examination and comparative
analysis. In N. Segarra and A. J. Toribio (eds.), Selected proceedings of the 9th Hispanic
Linguistic Symposium, 121-131. Somerville, MA: Cascadilla Proceedings Project.
Willis, E. (2007). n acoustic study o the “pre-aspirated trill” in narrative Cibaeño Dominican
Spanish. Journal of the International Phonetic Association 37(1). 33-49.
Willis, E. & Bradley, T. (2008). Contrast maintenance of taps and trills in Dominican Spanish: Data
and analysis. In Laura Colantoni and Jeffrey Steele (eds.) Selected Proceedings of the 3rd
Conference on Laboratory Approaches to Spanish Phonology, 87-100. Somerville, MA:
Cascadilla Proceedings Project.