acl/cla 2011

résumés/abstracts

Case-Reservation in the analysis of the Indirect Attribute adjectives Ahmad Assiri (Memorial University)

Generally, adjectives in Modern Standard Arabic follow the nouns they modify. Adjectives usually show agreement with their modified nouns in Case, phi-features, and definiteness (i.e., when used attributively); however, adjectives in Arabic can appear before the nouns they modify, thus forming what is known as the 'Indirect Attribute'. The adjective in this structure forms a Construct State-like (CS) construction with the following noun, which bears the genitive Case value (see, the bracketed words in (1)) as a reflection of the relationship between the adjective and the noun. While the adjective (semantically) modifies the following noun, it agrees with the preceding noun (i.e., in phi-features, Case, and definiteness):

(1) ratay-tu fataat-a-n [jamiil-at-a l-wajh-i] saw-I girl-ACC-INDEF beautiful-F.S.-ACC DEF-face-GEN 'I saw a girl with a beautiful face' or 'I saw a girl beautiful of face'

(2) [CP [TP [vP -tu [VP ratay [nP fataat-a-n [DP D [aP l-wajh [A jamiil .... (3) ratay-tu fataat-a-n [jamiil-a-n wajh-u-ha] saw-I girl-ACC-INDEF beautiful-ACC-INDEF face-NOM-her 'I saw a girl with a beautiful face' or 'I saw a girl whose face is beautiful' (4) [CP [TP [vP -tu [VP ratay [aP fataat-a-n [DP D [nP wajh-u-ha [A jamiil .... Another sub-type of the Indirect Attribute is formed by an adjectival element and a following noun (see the bracketed words in (3)). Like (1), the adjective in (3) semantically modifies the following noun; however, it shows a different type of agreement with the preceding as well as the following nouns: It agrees with the following noun in Gender, but agrees with the preceding one in Case and definiteness. As for Number, the adjective is always singular.

The schematic representations in (2) and (4) for the sentences in (1) and (3), respectively, show that the adjective originates lower than the noun it semantically applies to. They also show that the adjective executes multiple movements before it reaches the point (i.e., the head n) where values for its phi-features are provided (or shared), thus allowing its unvalued Case feature to receive valuation (in the sense of Chomsky, 2000, 2001, 2008). Concretely, the adjective raises first to the head a of aP, and then to the head D before it reaches the head n.

Moreover, the data shows that phi-and Case features operate independently. This has implications for how we view the features present on adjectives from the numeration.

The Agree-based analysis proposed for this construction accounts for Case and agreement properties, and equally offers new insights into the ample syntactic literature on Semitic CS constructions (including the typical, nominal CS) by proposing a principled and Minimalist-driven explanation for CS, and for the general noun-adjective relationship.

References: Chomsky, N. (2000). Minimalist inquiries: The framework. In R. Martin, D. Michaels &

J. Uriagereka (Eds.), Step by step: Essays on minimalist syntax in honour of Howard Lasnik (pp. 89-155). Cambridge, MA: MIT Press.

Chomsky, N. (2001). Derivation by phase. In M. Kenstowicz (Ed.), Ken Hale: A life in language (pp. 1-52). Cambridge, MA: MIT Press.

Chomsky, N. (2008). On phases. In R. Freidin, C. P. Otero & M. L. Zubizarreta (Eds.), Foundational issues in linguistic theory (pp. 133-166). Cambridge, MA: The MIT Press.

Featural mobility and the phonology-morphology interface Degif Petros Banksira (Memorial University)

Chaha, one of the Semitic languages of Ethiopia, displays morphemes (e.g., the impersonal subject) that are sometimes expressed by labialization, palatalization or both (Polotsky 1951, Lelau 1967, Hetzron 1971, Banksira 2000). These processes affect phonemes that are located at various positions in a word. Due to this, the language has been considered to have ‘mobile morphemes’ (McCarthy 1983, Akinlabi 1996, Rose 1997, Piggott 2000). The two main characteristics of such morphemes are that i) their phonological content is less than a full segment – it is a feature or bundle of features – and ii) the position of these features is variable. In this presentation, I will argue that these so-called mobile morphemes do not satisfy any of these conditions and that Chaha in fact does not have mobile morphemes at all.

From an empirical point of view, I will present data showing that the phonological signals of the putative mobile morphemes are not necessarily mobile nor less than a full segment – they sometimes surface as independent segments and their position is fixed. This is shown in (1a) where the impersonal suffix surfaces independently and occupies a fixed position like any other fixed suffix such as (1b).

(1) a. čəәnəә-wi-m b. čəәnəә-či-m come-impersonal-past come-she-past One came. She came. I will then propose that the independent and fixed -wi of (1a) is the output of morphology even in cases where mobility is at issue. This output of morphology, namely the underlying /-wi/ is submitted to phonological rules which may give rise to mobile phonological features depending on the nature of the stem. In other words, in addition to establishing dominance relations between morphemes, morphology establishes a fixed precedence relationship between them, this relationship is asymmetric, namely if ‘a precedes b’ is true then the inverse is not. However, phonological rules may alter this asymmetry by manipulating the phonological features associated with /-wi/ without manipulating their morphosyntactic features. These rules are blind to morphosyntactic features of the impersonal subject and they can apply within morphemes. Thus, mobility is a function of phonology – not of morphology – and phonological rules alone are responsible for featural mobility. If this claim proves correct, it will have two important consequences. First, it will provide evidence that analyzing Chaha labialization and palatalization as mobile morphemes is descriptively inadequate. Thus, gradient Optimality Theoretic constraints such as “align morphological categories to an edge” proposed by Akinlabi (1996) to compute the movement of these morphemes can be dispensed with (see Piggott 2000 for similar proposals). Second, it will solve the puzzle that the placement of a morpheme in Chaha depends on its morphosyntactic content. Notice that the phonological signals of well-known and better understood morphemes cross-linguistically have a variety of shapes: C, V, CV, CVC, CVCC, and so on, or their moraic equivalents. Yet, such morphemes are

not known to be mobile. So, theories that subscribe to mobile morphemes must attribute the special behaviour of mobility only to morphemes whose phonetic expression can be (partly) featural. In such analyses the morphosyntactic contents of morphemes (e.g. the impersonal in (1a) vs. the 3fem. sing. in (1b)) will dictate their positions. On the contrary, the claim being made here is that the placement of a morpheme is blind to its morphosyntactic content and that the mobility of phonological features has nothing to do with morphology. References Akinlabi, A. (1996). Featural affixation. Journal of Linguistics 32, 239-289. Banksira, D. P. (2000). Sound Mutations: The Morphophonology of Chaha. Amsterdam:

John Benjamins Publishing Company. Hetzron, R. (1971). Internal labialization in the tt-Group of Outer South-Ethiopic.

Journal of the American Oriental Society 91, 192-207. Leslau, W. (1967). The impersonal in Chaha. To honor Roman Jakobson. Essays on the

occasion of his seventieth birthday, 1150-1162. The Hague: Mouton. McCarthy, J.J. (1983). Consonantal morphology in the Chaha verb. The proceedings of

the second West Coast Conference on Formal Linguistics, edited by M. Balow, D. Flickinger & M. Wescoat, 176-188. Palo Alto: Stanford Linguistics Association.

Piggott, G. (2000). Against featural alignment. Linguistics 36, 85-128. Polotsky, J.H. (1951). Notes on Gurage grammar. Jerusalem: Israel Oriental Society. Rose, S. (1997). Theoretical issues in comparative Ethio-Semitc phonology and

morphology. Doctoral dissertation, McGill University.

On Wh-Movement in Onondaga Michael Barrie (University of Ottawa)

Nutshell: In this investigation I argue that wh-constructions in Onondaga (Northern Iroquoian) raise to a dedicated SpecCP position (contra Koenig and Michelson, 2008) and that this movement obeys well known constraints on movement (Huang, 1982, Ross, 1967). Introduction: Onondaga exhibits both local and long-distance obligatory wh-movement, although the exact placement of the wh-phrase is somewhat free (though see Baker, 1996). The wh-phrase can appear in any preverbal position (not all combinations shown for lack of space).

Wh-Movement: The fluidity in the placement of the wh-phrase suggested by the data in (1) is only apparent. I argue that expressions that appear before the wh-phrase are in a topic position above SpecCP (in the sense of Rizzi, 1997, although I continue to use the general label 'SpecCP' for convenience). Evidence for this proposal is offered by the following data.

Since quantifiers cannot be topicalized (Cinque, 1990, Rizzi, 1986), the ungrammatical version of (3) (with pre-verbal everyone) can be attributed to the quantifier appearing in topic position above SpecCP, a position which is otherwise available to R-Expressions, (1).

Onondaga also obeys well-known constraints on A-bar movement. For reasons of space, I show here only a relative clause island.

Conclusion: Onondaga exhibits both local and long-distance obligatory wh-movement and exhibits standard wh-island effects. Assuming a SpecCP position for wh-phrases accounts these facts and for the absence of quantifiers to the left of the wh-phrase. Such a position is the target of topicalization, which is not available to quantifiers.

References Baker, M. C. 1996. The Polysynthesis Parameter. Oxford: Oxford University Press. Cinque, G. 1990. Types of A-Bar Dependencies. Cambridge, MA: MIT Press. Huang, C.-T. J. 1982. Logical Relations in Chinese and the Theory of Grammar,

Department of Linguistics and Philosophy, MIT: Ph.D. Dissertation. Koenig, J.-P., and Michelson, K. 2008. Revisiting the realization of arguments in

Iroquoian. Paper presented at Annual Meeting of the Linguistic Society of America, Chicago, IL.

Rizzi, L. 1986. On the Status of Subject Clitics in Romance. In Studies in Romance Linguistics, ed. O. Jaeggli, 391-419. Dordrecht: Foris.

---. 1997. The Fine Structure of the Left Periphery. In Elements of Grammar: A handbook in generative syntax, ed. L. Haegeman, 281-337. Dordrecht: Kluwer Academic Publishers.

Ross, J. 1967. Constraints on Variables in Syntax, Department of Linguistics and Philosophy, MIT: Ph.D. Dissertation.

Les propriétés prosodiques des clitiques en français spontané Veranika Barysevich & François Poiré (University of Western Ontario)

Les dernières décennies montrent un grand intérêt des chercheurs pour l’étude des

clitiques, mots prosodiquement déficients comme les déterminants le, la, les, un(e), des, mon, ma, mes, quel(le)(s), etc. et/ou les pronoms personnels comme je, tu, il, elle, etc. en français. La question de la représentation phonologique des clitiques provoque beaucoup de discussions vu qu’ils se comportent soit comme des affixes qui appartiennent aux mots phonologiques, soit comme des mots indépendants qui appartiennent au syntagme phonologique (Nespor et Vogel 1986).

L’identification des mots portant la valeur [+clitique] (Mertens 1993) ne provoque pas moins de discussions. Les positions sont différentes selon les approches théoriques. Au niveau morphologique, le caractère clitique d’un mot dépend en gros de sa catégorie grammaticale (Wunderli et al. 1978). Au niveau syntaxique, le caractère accentogène [c’est-à-dire non clitique] est une propriété de certaines types syntaxiques (Garde 1968 : 69). Au niveau phonologique, la propriété [+/-clitique] dépend de la possibilité de former ou non d’une façon autonome un groupe intonatif, c’est-à-dire porter l’accent final ou non en français (Rossi 1987 : 21). Malgré de nombreuses approches, l’approche morphologique semble prédéterminer dans une large mesure le regroupement des mots selon la propriété [+/-clitique]. Néanmoins, les classements des clitiques selon leur catégorie grammaticale diffèrent d’une étude à l’autre.

L’hypothèse générale de notre travail de recherche est qu’un clitique est un élément prosodiquement déficient par définition monosyllabique qui ne devrait se distinguer dans sa durée des autres syllabes inaccentuées (sauf peut-être en étant plus court) et qui ne devrait jamais porter de sommet mélodique. Nous testons cette hypothèse dans le cadre de la phonologie prosodique (Nespor et Vogel 1986, Selkirk 1995), en comparant les propriétés prosodiques (principalement durée et hauteur mélodique) des syllabes des mots clitiques à celles des syllabes des mots pleins sur un échantillon de six locuteurs européens (France métropolitaine, Belgique francophone et Suisse romande) du corpus C-PROM (Avanzi et al. 2010). Ce corpus disponible en ligne contient plus de 70 heures de parole segmentée et codé pour la proéminence syllabique (par accords inter-juges).

Nous poursuivons donc les objectifs suivants : (i) la description de la durée des clitiques suite à l’analyse comparative d’une syllabe clitique avec la même syllabe non clitique dans un mot monosyllabique et/ou polysyllabique ; (ii) la description de l’alignement mélodique propre aux syllabes clitiques ; (iii) la description du comportement des clitiques présentés par les verbes auxiliaires (avoir, être, etc.) et/ou modaux (pouvoir, devoir, etc.) monosyllabiques et/ou polysyllabiques ; (iv) la description du comportement des clitiques dans les cas de pause (hésitation, fusion vocalique) ; (v) l’examen des propriétés prosodiques des clitiques en corrélation avec le codage en proéminence effectué par les auteurs du corpus C-PROM.

Après avoir examiné les profils de durée et d’alignement mélodique des clitiques dans la parole de ces six locuteurs, nous sommes arrivés aux résultats suivants : (a) les clitiques sont par définition monosyllabiques ; (b) l’analyse comparative du comportement prosodique des syllabes clitiques et des syllabes non clitiques selon la position distributionnelle dans un mot a montré que les clitiques se comportent plutôt

comme des éléments internes d’un mot prosodique ; (c) les syllabes clitiques à ton haut et les syllabes clitiques perçues comme proéminentes sont des propriétés idiosyncratiques de certains locuteurs ; (d) la réalisation phonologique des clitiques en français ne peut pas être expliquée à partir de leur catégorie grammaticale. Ces résultats préliminaires sont discutés dan le cadre de la phonologie prosodique. Références Avanzi, M., A.-C. Simon, J. P. Goldman et A. Auchlin. 2010. C-PROM. An annotated

corpus for French prominence studies. Dans Proceedings of Prosodic Prominence: Perceptual and Automatic Identification, Proceedings of Speech Prosody 2010 Satellite Workshop. Chicago, Illinois, USA.

Garde, P. 1968. L’accent. Paris: P. U. F. Mertens, P. 1993. Accentuation, intonation et morphosyntaxe. Dans Travaux de

Linguistique 6, 21-69 Mertens, P., J.-Ph. Goldman, E. Wehrli et A. Gaudina. 2001. La synthèse de l'intonation

à partir de structures syntaxiques riches. Traitement Automatique des Langues 42 (1), 142-195

Nespor, M. et I. Vogel. 1986. The Clitic Group. Dans Prosodic Phonology, sous la direction de M. Nespor et I. Vogel, 145-163. Dordrecht et Riverton : Foris Publications.

Rossi, M. 1987. Peut-on prédire l'organisation prosodique du langage spontané? Etudes de Linguistique Appliquée 66, 20-48.

Selkirk, E. 1995. The prosodic structure of function words. Dans Papers in Optimality Theory, sous la direction de J. Beckman, L. Walsh Dickey et S. Urbanczyk, 439-470. Amherst, MA: GLSA Publications.

Wunderli, P., K. Benthin et A. Karash. 1978. Französische Intonationsforshung Kritische bilanz und Versuch einer Synthese. Tübingen: G. Narr Verlag.

Le complémenteur dans les relatives sujets 3PL en français acadien : [qui] vs [que] Louise Beaulieu (University de Moncton, campus de Shippagan) &

Wladyslaw Cichocki (UNB)

En français acadien du nord-est du Nouveau-Brunswick (dorénavant FANENB), l’élément qui occupe la position du complémenteur dans les relatives sujets est réalisé comme [ki] ou [k]. (1) a. [des gars [[ki] ont ça]] b. [des gars [[k] ont ça]]

Dans les relatives sujets de 3e personne du pluriel (dorénavant 3PL), deux analyses

peuvent être proposées pour rendre compte de [ki] vs [k]. La première stipule que la position du complémenteur est remplie par le morphème [qui], comme en (2a) ; la deuxième que cette position est occupée par le morphème [que] suivi du marqueur de sujet i-(2b).

(2) a. [les femmes [[qui] [ø font ça d’ même]]] b. [les femmes [[qu’] [i font ça d’ même]]]

Il est généralement accepté que dans la relative sujet du français de référence, l’élément présent dans la position du complémenteur est [qui] comme en (2a). En français parlé, il semble qu’on trouve aussi des relatives sujets dont la position du complémenteur est remplie par le morphème [que] suivi du clitique sujet (Auger 1994, Bouchard 1982, Gadet 1989, Lefebvre & Fournier 1978), tel qu’en (2b).

Le but de cette communication est d'examiner certains arguments sociolinguistiques qui permettent de vérifier la validité de ces deux analyses quand il s’agit de rendre compte des relatives sujets 3PL du FANENB. Les données proviennent d'un corpus d'entrevues sociolinguistiques réalisées au début des années 90 auprès de 16 locuteurs natifs de FANENB. Dans 24 heures de données (1,5 heures par locuteur) en situation informelle provenant de ce corpus, 1353 occurrences de relatives sujets ont été extraites dont 450 sont des relatives sujets 3PL.

Un premier examen de la distribution de [ki] -[k(əә)] montre qu’en FANENB, on ne relève aucun exemple qui atteste de l’indépendance syntaxique de [k] et [i]. De plus, la présence du complémenteur [que] suivi d’un sujet clitique de 1e

ou de 2e

personne dans

les relatives sujets et dans les pseudo clivées est pratiquement nulle et il en va de même pour son occurrence dans les constructions dans lesquelles l’extraction à distance d’un élément devrait laisser le complémenteur [qui].

On note aussi que malgré la fréquence élevée de [ki] dans les relatives sujets 3PL, la variante acadienne portant le marqueur d’accord mixte i-...-ont (i-parl-ont) n’apparaît jamais dans ces constructions.

Des analyses (GOLDVARB, Rand et Sankoff 1990) montrent que le patron de variation de [ki] vs [k] n’est pas associé à celui de i-marqueur d’accord ou à celui des autres marques d’accord verbal 3PL. Les contraintes linguistiques sur la variation de [ki] vs [k] diffèrent grandement de celles qui jouent un rôle dans la présence vs l’absence du marqueur de sujet idans les autres constructions 3PL ou celles qui ont un effet sur les marques d’accord verbal dans les relatives sujets 3PL. Par exemple, on note

que le facteur Segment vocalique qui suit la variante joue un rôle important dans la variation de [ki] vs [k]. Les contraintes sociales sont aussi très différentes : la plupart de ces facteurs ont un effet négligeable sur l’alternance [ki] vs [k]. Une comparaison entre les différents patrons de variation nous amène à conclure que dans les relatives sujets 3PL en FANENB, le complémenteur est le morphème [qui].

Références Auger, J. 1994. Pronominal clitics in Québec Colloquial French: A morphological

analysis. Thèse de doctorat, University of Pennsylvania. Bouchard, D. 1982. Les constructions relatives en français vernaculaire et en français

standard. In La Syntaxe comparée du français standard et populaire: approches formelle et fonctionnelle, sous la dir. de C. Lefebvre, vol.1, 103-133. Québec: Gouvernement du Québec, Office de la langue française.

Gadet, F. 1989. Le français ordinaire. Paris : A. Colin. Lefebvre, C. et R. Fournier. 1978. Les relatives en français de Montréal. In Cahiers de linguistique 8, Syntaxe et Sémantique du français, 273-295. Montréal: Les Presses de l’Université du Québec. Rand, David, et David Sankoff. 1990. GoldVarb, version 2.1 : A variable rule

application for the Macintosh. Montréal : Centre de recherches mathématiques de l'Université de Montréal.

GEN and PART case valuation: a contrastive analysis Rostyslav Bilous (University of Toronto)

All Romance languages seem to make a clear distinction between GEN and PART cases. Some Slavic languages, however, are characterized by intricate irregularities in the distribution of endings serving to mark the so-called Partitive Genitive (PART-GEN) (Pliushch 1983, Valkova 1999, inter alia). This study, based on generative and functionalist typology of cases, offers a contrastive analysis of GEN and PART case valuation in Ukrainian and French. In our investigation we draw a clear demarcation line between GEN and PART in Slavic targeting those functions of GEN case that are related to the direct object (DO) position – the function of a DO marker (Ex. 1a-b) and the function of copular verb transitivizer (Ex. 1c-d): (1) a. Petryk sluxajet’sia mamy. Petryk obeys-CL mother-GEN ‘Petryk obeys his mother.’ b. Sashko ne čytaje knyhy. Sasha not read book-GEN ‘Sasha does not read a/the book.’ c. Tsia divčyna – dvorians’koho rodu. This girl noble-GEN descent-GEN ‘This girl is of noble descent.’ d. Vony dijšly zhody They went-PERF agreement-GEN ‘They achieved an agreement.’ e. Andrij zjiv zupy. Andrew ate-PERF soup-PART ‘Andrew ate (some) soup.’ We propose that in Ukrainian the difference between GEN and PART cases can be seen at two levels: semantic and syntactic. Semantically speaking, GEN cannot have a partitive reading, unlike the PART case. Syntactically speaking, we distinguish between three types of GEN: 1) GEN marking a post-copular noun and assigned as an inherent lexical case by the (null) lexical head P (cf. Ex. 1c/d), realizing the feature of characterization or transitivizing an unaccusative verb; 2) GEN marking a DO of a transitive (or transitivized) verb, assigned (obligatorily or optionally) by the lexical head v as an inherent lexical case (cf. Ex. 1a); 3) Genitive of Negation (GEN-NEG) checked by a quantifying preposition embedded in vQP in the head-complement configuration [P–DetP] (after being licensed by negation). The last type of GEN, GEN-NEG, is similar to PART (cf. Ex. 1e) sharing with it the same checking domain. However, PART case is valued by Asp, if the verb is or becomes vQ (i.e. one which quantifies its DO). This is a modified version of Brown’s (1999) proposal on GEN-NEG and PART checking and of Harves’s (2002) proposal on GEN-NEG checking. We conclude that in spite of important differences between Ukrainian (a non-configurational system with rich case morphology) and French (a generally configurational system with impoverished case morphology) in terms of case valuation,

both codes share two instances of differential object marking – GEN-NEG and PART cases, representing a mix of structural and inherent lexical case properties. References Brown, S. 1999. The syntax of negation in Russian: a minimalist approach. Stanford,

CA: Center Study Language & Information, xii+154p. Harves, S. A. 2002. Unaccusative syntax in Russian. Doctoral Dissertation: Princeton

University, 330p. Valkova, Z. 1999. Semantics of Case: The Partitive Genitive in Russian. Doctoral

Dissertation: Dissertation Abstracts International, The Humanities and Social Sciences, 61, 3, 166p.

Pliushch, M. J. 1983. Katiegorija padiezha v siemantika-sintaksichieskaj strukturie priedlazhenija (‘The category of case in semantico-syntactic structure of a sentence’). Doctoral Dissertation: Kyiv State Pedagogical Insitute after Gorkij, 412p.

The Syntax of Auxiliaries From a Cross-linguistic Perspective Bronwyn Bjorkman (MIT)

Verbs like be are often assumed to be default in some sense. Looking primarily at

auxiliary be, this paper argues that be is a morphologically default verb. It is not present in the syntactic derivation, but is instead inserted in the morphological component to realize “stranded” inflectional material. This analysis sheds light on the typology of auxiliary-participle constructions, and more generally on the syntactic mechanisms that manipulate verbal inflection.

I present two types of auxiliary constructions: in the first, auxiliaries are used only in certain combinations of inflectional categories; in the second, certain inflectional categories always require an auxiliary. Latin and Kinande (Bantu) exemplify the first category. In Latin both the perfect (amavi ‘I loved/have loved.’) and the passive (amor ‘I am loved’) had synthetic tensed forms, but the perfect-passive required an auxiliary-participle construction (amatus sum ‘I was loved/have been loved’). In Kinande, similarly, single verbs can be inflected for past tense (1a), or for certain aspects (1b), but to express both past tense and aspect on a single verb requires an auxiliary followed by an aspectually-inflected participle (1c):

(1) a. tu-nému-húma, ‘We are hitting’ b. tw-á-húma, ‘We hit (recently, not today)’ c. tw-á-bya i-tu-nému-húma, ‘We were (recently, not today) hitting.’

If auxiliaries were present in the syntax, the auxiliary distribution in Latin and Kinande would indicate that some inflectional combinations were something more than the sum of their parts, with extra structure corresponding to the auxiliary. I argue that a simpler syntactic analysis is available if auxiliaries are not represented in the syntax, but are instead inserted post-syntactically (as in Distributed Morphology; Halle and Marantz, 1993, 1994, et seq.). This is done to realize inflection that is insufficiently local to the lexical verb to combine with it via morphological merger or lowering (Embick and Noyer, 2001), but which must nonetheless occur on a verb: Vocabulary Insertion therefore inserts an “elsewhere” verb, i.e. be.

This analysis extends to the second type of auxiliary construction, where auxiliaries always occur with certain inflectional categories (e.g. English, French, Hindi, Basque, Finnish, among many others). Though the syntactic representation is not straightforwardly simplified by this analysis (though see Shütze, 2003, Cowper, 2010), it provides insight into reduced relative clauses, which have been observed to be possible only when the relative’s participle would have taken auxiliary be (Iatridou et al., 2003: in English, The cake eaten by the children, The children eating the cake, but *The children eaten the cake). If auxiliary be is a morphological realization of inflection unable to combine with a main participial verb, and reduced relatives lack higher inflectional structure, then this is exactly the environment in which we would expect an auxiliary not to occur.

The difference between Latin/Kinande-type languages and English-type languages is argued to be the position of the participial verb. In English-type languages, the participial verb remains very low, and is thus able to combine with only one higher

inflectional category; in Latin/Kinande-type languages it is slightly higher, and so can potentially express two inflectional categories on a single verb, but requires an auxiliary beyond that.

This analysis has broader implications for the syntax of verbal inflection. Syntactic approaches often assume diverse mechanisms for the manipulation of verbal inflection, some of which can operate long-distance (i.e. Agree, Chomsky, 1998), with different mechanisms instantiated in different languages. This is at odds, however, with the typological commonalities of auxiliary-participle constructions, which can be accounted for only by a cross-linguistically unified theory of verbal inflection. References Chomsky, Noam. 1998. Minimalist inquiries: the framework. MIT Occasional Papers in

Linguistics 15. Cowper, Elizabeth A. 2010. Auxiliary Verb Insertion. Handout from ACL/CLA

Conference. Embick, David, and Rolf Noyer. 2001. Movement operations after syntax. Linguistic

Inquiry 32:555–595. Halle, Morris, and Alec Marantz. 1993. Distributed Morphology and the pieces of

inflection. In The View from Building 20: Essays in Linguistics in Honor of Sylvain Bromberger, ed. Ken Hale and Samuel J. Keyser. Cambridge, MA: MIT Press.

Halle, Morris, and Alec Marantz. 1994. Some key features of Distributed Morphology. MIT Working Papers in Linguistics 21:275–288.

Iatridou, Sabine, Elena Anagnostopoulou, and Roumyana Izvorski. 2003. Observations about the form and meaning of the Perfect. In Perfect explorations, ed. Artemis Alexiadou, Monika Rathert, and Arnim von Stechow, volume 2 of Interface Explorations, 153–204. Berlin: Mouton de Gruyter.

Shütze, Carson T. 2003. When is a verb not a verb? Nordlyd 31:400–415.

One of those situations where a relative pronoun becomes a complementizer Marisa Brook (University of Toronto)

There exists a certain colloquial use of where in North American English that has so far largely escaped notice:

1. I found a lot of couches where they wouldn’t fit through that door. (2010) 2. Either way you’re going to end up with something where you don’t want it.

(2010) 3. You ever have that thing where you get mad at someone in a dream and then

you’re still mad when you wake up? (2009) 4. There were some people at my high school where they just couldn’t write legibly.

(2008) Aside from brief mentions of this pattern made by Comrie (1999), Massam (1999), and Pullum (2008), there appears to be nothing on it at all. Yet this where is very distinct both semantically and syntactically from the original relative-adverb where: in terms of its head nouns, it prefers classes and categories rather than locations and metaphorically spatial entities; and, curiously enough, its subordinate clauses are invariably gapless. Not only could this newer where not have moved from a trace position (as it would when acting as a locative relative adverb: the town where I live ____ ), there are no traces in its relative clauses at all. I argue, therefore, that this new where occupies the position of C, but unlike the that/zero complementizer does not license covert WH-movement. If it did, there would be no accounting for the resumptive pronouns in examples such as (1) and (4) (subject relativization) and (2) (object relativization). Resumptive pronouns do not have much of a place in English as it stands: they tend to be a semi-grammatical last-resort kind of strategy used to salvage complex subordinate clauses and syntactic islands that arise in spoken language. Here, however, they prove to be found systematically, as they are obligatory (* I found a lot of couches where wouldn’t fit through that door; * Either way you’re going to end up with something where you don’t want). In spite of the fact that this is all decidedly atypical behaviour for standard English relativizers of any sort (relative pronoun, relative adverb, relative complementizer), the where-complementizer construction is widespread in spontaneous speech. My stance is that this is attributable to a number of unique properties it has. First of all, it is capable of applying to its head noun a stand-alone clause without any co-referent at all, as in (3). Second, because the where-complementizer does not trigger anything in the way of traces or movement under any circumstances, there are no syntactic loose-ends needing to be dealt with in the process of uttering the relative clause after it (gaps, pied-piping, and so on). This gives it an unmatched level of versatility, especially in spontaneous speech. Curiously enough, evidence suggests that in several other Germanic languages - Afrikaans, Yiddish, and Bavarian - a small number of relative complementizers emerged from secondary usages of relative pronouns: in most cases the equivalents of ‘who’, but sometimes - of all things - ‘where’. (Lowenstamm 1984; Bayer 1984; Harbert 2007). Therefore, the shift taking place in colloquial North American English is interesting not

just for its own sake but also due to the possibility that it will be able to illuminate potential intermediate stages through which the earlier parallel changes in other Germanic languages could have progressed. References Bayer, J. (1984). COMP in Bavarian syntax. The Linguistic Review, 3, 209-274. Comrie, B. (1999). Relative clauses: structure and typology on the periphery of standard

English. In P. Collins and D. Lee (eds.), The clause in English: in honor of Rodney Huddleston, 81-91. Philadelphia/Amsterdam: John Benjamins.

Harbert, W. (2007). The Germanic Languages. New York/Cambridge: Cambridge University Press.

Lowenstamm, J. (1977). Relative clauses in Yiddish: a case for movement. Linguistic Analysis, 3(3), 197-216.

Massam, D. (1999). Thing-is constructions: the thing is, is what’s the right analysis? English Language and Linguistics, 3(2), 335-352.

Pullum, G. K. (2008). Some discipline where nobody knows what the hell it is. Language Log. <http://languagelog.ldc.upenn.edu/nll/?p=823>

Investigating the differences between the English way-construction and the fake reflexive resultative construction

Liz Christie (Carleton University)

The English way-construction (1) and the English fake reflexive resultative construction (2) are distinct constructions which have many commonalities in their use and meaning. (1) John elbowed his way into the room. (2) John laughed himself silly. The two constructions can be used in the same situation to talk about the same event, as shown in (3) and (4). (3) The rat chewed his way through the wall. (4) The rat chewed himself through the wall.

Authors have either proposed to treat the two constructions in the same manner requiring little or no variation in their treatment (Jackendoff, 1990; Marantz, 1992; Levin, 1993), or to treat the two constructions as distinct phenomena requiring separate analyses (Goldberg, 1997; Asudeh et al., 2008; van Egmond, 2009). However, even in these latter approaches, the differences between the two constructions have not been carefully investigated. According to Goldberg (1997), interpretation of the way-construction crucially involves the creation of the path that is traversed by the subject (p. 170), whereas the fake reflexive construction focuses the final state of the participants. The way-construction is able to convey both a means and manner interpretation for the motion involved in the sentence (Jackendoff, 1990; Goldberg, 1995; Asudeh et al., 2008), whereas the fake reflexive construction is not able to convey the manner interpretation (Goldberg, 1997; van Egmond, 2009). This paper will provde an analysis which relates the two constructions while also capturing their differences more thoroughly than has been the case before. Given their differences in informational content, and the fact that the way-construction is able to take a non-agent subject as demonstrated in Asudeh et al.’s (2008) example (33), the two constructions will be given separate treatments, as shown in (5) and (6).

For the treatment of the way-construction in (5), the function F is either the means or manner of the GO event, which places the subject of the movement, Y, along the path denoted by Z. For the treatment of the fake reflexive construction, on the other hand, the function F must be the means of the agent X coming to be in the state denoted by Z. References Asudeh, Ash, Dalrymple, Mary and Toivonen, Ida. 2008. Constructions with Lexical In-

tegrity: Templates as the lexicon-syntax interface. In Miriam Butt and Tracy Holloway-King (eds.), Proceedings of the LFG08 Conference, pages 68–88, CSLI Publications.

Goldberg, Adele E. 1995. Constructions. The University of Chicago Press. Goldberg, Adele E. 1997. Making one’s way through the data. In Alex Alsina, Joan

Bresnan and Peter Sells (eds.), Complex Predicates, Stanford, CA, USA: Center for the Study of Language and Information.

Jackendoff, Ray. 1990. Semantic Structures. MIT Press. Levin, Beth. 1993. English Verb Classes and Alternations. University of Chicago Press. Marantz, Alec. 1992. The way-construction and the semantics of direct arguments in

English: A reply to Jackendoff. In Tim Stowell and Eric Wehrli (eds.), Syntax and Semantics 26: Syntax and the Lexicon, pages 179–188, Academic Press.

van Egmond, Marie-Elaine. 2009. Two way-constructions in Dutch: motion along a path and transition to a location. VDM Verlag.

Polysynthetic word-internal adjectives and verb-like adjectives in Inuit Richard Compton (University of Toronto)

While Eskimoan languages have been claimed to lack a lexical category of adjectives (Fortescue 1984, de Reuse 1994, Sadock 2003), I argue based on evidence including (1) stacking, (2) variable ordering, and (3) noun-incorporation that Inuktitut and related languages possess a small class of strictly attributive adjectives, as well as a class of verb-like adjectives which exhibit properties that differentiate them from stative intransitive verbs. The examples below from the Baffin Island dialect of Inuktitut were elicited by the author: (1) qarisauja-ralaa-kulu-tuqa-nnguaq

computer-small-adorable-old-pretend(ABS.SG) ‘old adorable small pretend computer’ (e.g. in a toy store)

(2) a. iglu-tsiava-kuluk

house-good-adorable(ABS.SG) ‘good adorable house’

b. iglu-kulu-tsiavaq house-adorable-good(ABS.SG) ‘good adorable house’

(3) ‘Barbie-doll’ iglu-tsiava-IIua-qaq-tuq

B.(ABS.SG) house-good-fake-have-DEC.3SG ‘Barbie has a nice [toy] house.’

These adjective classes have previously been overlooked due to Lexicalist assumptions that lexical categories should (i) instantiate phonologically independent words, not subparts of polysynthetic words, and (ii) bear a distinct inflectional paradigm. Employing Distributed Morphology’s proposal for “syntactic hierarchical structure all the way down” (Halle & Marantz 1993), I argue for little aP modifiers inside polysynthetic words. Furthermore, I examine a semantic constraint on the class of strictly attributive adjectives that prevents members with intersective denotations. All members of this class possess subsective, privative, or plain nonsubsective denotations. I argue that this is due to a lack of Predicate Modification (PM) in the language which blocks the creation of intersective attributive adjectives. Evidence for this analysis includes the widespread use of appositives (which I argue are composed using Potts 2005’s Conventional Implicature application), and, crucially, that relative clauses, verb-like adjectives, and (derived) adverbials all appear to be nominalized case-bearing DPs in apposition with the elements they modify. Support for the (parametric) existence of PM will be drawn from the findings on adjective ordering in Svenonius (2008) and Truswell (2009). This research bears on the topic of parametric variation in semantics and is also relevant to debates on the cross-linguistic universality of lexical categories (Dixon 2004, Baker 2004).

References Baker, Mark C. (2004). Lexical Categories. Cambridge: Cambridge University Press. Cinque, Guglielmo (1994). On the evidence for partial N movement in the Romance DP.

Paths toward Universal Grammar: Studies in Honor of Richard S. Kayne, ed. by Guglielmo Cinque, Jan Koster, Jean-Yves Pollock, Luigi Rizzi, and Raffaella Zanuttini, 85-110. Washington, DC: Georgetown University Press.

Creider, Chet (1978). The Syntax of Relative Clauses in Inuktitut. Inuit Studies 2.95-110. de Reuse, Willem J. (1994). Siberian Yupik Eskimo: the language and its contacts with

Chukchi. Salt Lake City: University of Utah Press. Dixon, Robert M. W. (2004). Adjective Classes in Typlogical Perspective. Adjective

Classes: A Cross-Linguistic Typology, ed. by Robert M. W. Dixon and Alexandra Y. Aikhenvald, 1-49. Oxford: Oxford University Press.

Ernst, Thomas (2002). The Syntax of Adjuncts. Cambridge: Cambridge University Press. Fortescue, Michael (1980). Affix Ordering in West Greenlandic Derivational Processes.

International Journal of American Linguistics 46:4.259-278. Fortescue, Michael (1984). West Greenlandic. London: Croom Helm. Halle, Morris and Alec Marantz (1993). Distrubuted Morphology and the pieces of

inflection. The view from building 20, ed. by Kenneth Hale and S. Jay Keyser, 111-176. Cambridge, MA: MIT Press.

Heim, Irene and Angelika Kratzer (1998). Semantics in Generative Grammar. Malden, MA: Blackwell.

Johns, Alana (2007). Restricting noun incorporation: root movement. Natural Language & Linguistic Theory 25.535-576

Kennedy, Christopher and Louise McNally (2005). Scale structure, degree modification, and the semantics of gradable predicates. Language 81:2.345-381.

Potts, Christopher (2005). The Logic of Conventional Implicatures. Oxford: Oxford University Press.

Sadock, Jerold (2003). A Grammar of Kalaallisut (West Greenlandic Inuttut). Muenchen: Lincom Europa.

Svenonius, Peter (2008). The position of adjectives and other phrasal modifiers in the decomposition of DP. Adjectives and Adverbs: Syntax, Semantics, and Discourse, ed. by Louise McNally and Christopher Kennedy, 16-42. Oxford: Oxford University Press.

Truswell, Robert. 2009. Attributive adjectives and nominal templates. Linguistic Inquiry 40:3.525-533.

L2 learners’ processing of agreement dependencies in French: Proficiency and working-memory capacity effects

Caitlin Coughlin & Annie Tremblay (University of Illinois at Urbana-Champaign)

Agreement can be difficult for second/foreign language (L2) learners, even if the native language has the corresponding agreement features (e.g., Hopp, 2010). Whereas earlier research on this topic focused on production (e.g., Lardiere, 1998; Prévost & White, 2000), more recent studies have begun examining comprehension (e.g., Foote, 2011; Hopp, 2006, 2010; Jiang, 2004, 2007; Keating, 2009; Sabourin & Stowe, 2008; Sato & Felser, 2008). Two types of accounts have sought to explain L2 learners’ variable processing of agreement dependencies: represen-tational accounts (e.g., Clahsen & Felser, 2006; Jiang, 2004, 2007; Keating, 2009; Sato & Felser, 2008), which claim that L2 learners lack/cannot deploy the structure necessary to process agreement; and computational accounts (e.g., Hopp, 2010; McDonald, 2000, 2006), which attribute L2 learners’ difficulty with agreement dependencies to reduced processing efficiency, due to L2 learners’ limited decoding abilities and working memory (WM) capacity. To shed further light on this debate, the present study investigates the roles of proficiency (established by cloze test scores) and WM capacity (reading span) in the processing of short and long number-agreement dependencies by English L2 learners of French. It focuses specifically on L2 learners’ processing of object clitics. In both French and English, object pronouns agree in number with their antecedents, but in French, these pronouns are pre-verbal clitics and they have been analyzed by some as verbal affixes (e.g., Auger, 1994; Cummins & Roberge, 1994). This study examines L2 learners’ sensitivity to number-agreement violations between object clitics and their left-dislocated antecedents in online sentence comprehension. 35 L2 learners at mid and high proficiencies (experimental group) and 10 native French speakers (control group) completed a non-cumulative region-by-region self-paced reading task containing object clitics (Experiment 1). In the experimental conditions, the clitic was manipulated for grammaticality (agrees vs. disagrees with its antecedent), number (singular vs. plural), and distance from its antecedent (short vs. long) (1). A true-or-false comprehension question unrelated to the number of the clitic (2) followed each sentence in the experiment. To ensure that L2 learners had explicit knowledge of clitic agreement rules, they also completed an acceptability-judgment task containing the same sentences, in which they corrected ungrammatical sentences (Experiment 2). After Experiments 1-2, all the participants completed a WM task in French, and the L2 learners completed an equivalent but different WM task in English. Both WM tasks were developed based on Waters and Caplan (1996), and L2 learners’ WM scores in the two languages were correlated. Importantly, the L2 learners’ proficiency scores were not correlated with their overall WM scores in French.

The results of Experiment 1 show that with short agreement dependencies (1a), both mid-and high-level L2 learners had marginally significant slower residual times at the critical region (underlined) when the clitic did not agree with its antecedent than when it did; with long agreement dependencies (1b), only the high-level learners showed this asymmetry, despite all L2 learners reaching ceiling on Experiment 2. A structural-equation model with proficiency and WM in French as predictors of L2 learners’ asymmetry between ungrammatical-grammatical sentences (in Experiment 1) yielded

reliable estimates of their performance, with stronger estimates for sentences in which the clitic was distant from its antecedent. The same model but with WM in English as predictor did not reveal reliable estimates of their performance. These findings suggest that processing inefficiency rather than incomplete/impaired representations is responsible for L2 learners’ difficulty with agreement dependencies.

References Auger, J. (1994). Pronominal clitics in Québec colloquial F rench: A morphological

analysis. Unpublished PhD dissertation, University of Pennsylvania. Clahsen, H., & Felser, C. (2006). Grammatical processing in language learners. Applied

Psycholinguistics, 27, 3–42. Cummings, S., & Roberge, Y. (1994). A morphosyntactic analysis of Romance clitic

constructions. In M. L. Mazola (Ed.), Issues and theory in Romance linguistics (pp. 239–257). Washington, DC: Georgetown University Press.

Hopp, H. (2006). Syntactic features and reanalysis in near-native processing. Second Language Research, 22, 369–397.

Hopp, H. (2010). Ultimate attainment in L2 inflection: Performance similarities between native and non-native speakers. Lingua, 120, 900–931.

Jiang, N. (2004). Morphological insensitivity in second language processing. Applied Psycholinguistics, 25, 603–634.

Jiang, N. (2007). Selective integration of linguistic knowledge in adult second language learning. Language Learning, 57, 1–33.

Keating, G. D. (2009). Sensitivity to violations of gender agreement in native and nonnative Spanish: An eye-movement investigation. Language Learning, 59, 503–535.

Lardiere, D. (1998). Dissociating syntax from morphology in a divergent end-state grammar. Second Language Research, 14, 359–375.

McDonald, J. L. (2000). Grammaticality judgments in a second language: Influences of age of acquisition and native language. Applied Psycholinguistics, 21, 395–423.

McDonald, J. L. (2006). Beyond the critical period: Processing-based explanations for poor grammaticality judgment performance by late second language learners. Journal of Memory and Language, 55, 381–401.

Prévost, P., & White, L. (2000). Missing surface inflection or impairment in second language acquisition? Evidence from tense and agreement. Second Language Research, 16, 103–133.

Sabourin, L., & Stowe, L. (2008). Second language processing: When are first and second languages processed similarly? Second Language Research, 24, 397–430.

Sato, M., & Felser, C. (2008). Sensitivity to morphosyntactic violations in English as a second language. Ms., University of Essex.

Waters, G. S, & Caplan, D. (1996). The measurement of verbal working memory capacity and its relation to reading comprehension. The Quarterly Journal of Experimental Psychology,49A, 51–79.

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

4: 5(!,"&%Y- #+1%&%#"%-.Z 2,)4 5(!,)A: H)*%)' #+1%&%#"%-.Z &) [,)\5!&) [A)B-\5 (36() [=#&Y\

] #: !&) ! ^_`abc4: &) ! c^bdefdcA: (36() ! ga__db

h ci_ag^e!!!

"""ga__db c^bdefdc

^_`abcji_ag^e %. # ')'?.>#*%#& #'#&), )0 B)*%)'5 #'6 ga__db #'6 c^bdefdc -'A)6- */- 0#B%&%#" R)B#'A-6%.*%'A*%)' 4-*=--' 6%"-A*-6?B)*%)' #'6 B#''-"?)0?B)*%)' 3-"4.: I_`abc -1>"-..-. )"%-'*#*%)' *)?=#"6 */- 6-%A*%A A-'*"-: F. #+1%&%#"%-.5 FGH. .>-&& )+* )'&2 !7 =/-' */-2 #>>-#" #. B#%' 3-"4.5 */-2#&.) .>-&& )+* # 3-"4#& "))* 6-')*%', A)'A"-*- B)*%)' %' .>#A-?*%B-: R->"-.-'*#*%3- B#%'?3-"4 +.-.#"- ./)=' %' 7k9: #36()5 #. -1>-A*-65 A#'')* *#Y- # .>-A%<A ,)#& ((5 =/%&- */- B)*%)' -1>"-..-6 42!&) %. )"%-'*-6 *)=#"6. */- 6-%A*%A A-'*"-:l #: mn&- 'o) !"#$% '# 0-.*#: [L- 6%6'\* ,)p=#&Y *) */- >#"*2:\ 7I'*-'6-69

/- ')* #'6#"?>"-*?N., %'?*/- >#"*24: q *%) &'( r#'*#" A)').A): [O'A&- %. A)B%', *) )+" /)+.- 0)" 6%''-":\

*/- +'A&- A)B-?>"-.?N., 6%'-?%'0 =%*/?+.A: )*'+5 '-B s+- *-'/# s+- +* # >t: [I\&& ,)5 -3-' %0 I /#3- *) =#&Y:\

%"?0+*?8., -3-' */#* *-"?>"-..4r?8., */#* ,)?%'0 *) 0))*F. #+1%&%#"%-.5 FGH. #>>-#" %' .&%,/*&2 6%Q-"-'* A)'*-1*.5 #. >"-6%A*-6 42 */-%" 6%Q-"-'* 0-#*+"#& A)'?*-'*: F&& B#2 *#Y- */- "36/ ,-"+'65 ,%3%', # "-#6%', &%Y- 8,,9 /3 +"&3:: u/- 6%Q-"-'A- 4-*=--' */%.#'6 */- )"6%'#"2 >"),"-..%3- A)'.*"+A*%)' =%*/ ,.2() %. 6+- *) */- 0-#*+"- ci_ag^e A#""%-6 42 #&& */"--FGH.: u/- A/)%A- #B)', */- */"-- %. .)B-=/#* T-1%4&-: ;) %. */- B).* A)BB)'5 =/%&- A#.-. =%*/!&) #"- #",+#4&2 )"%-'*-6 *)=#"6. */- 6-%A*%A A-'*"-: q'&2 &) #>>-#". #. #' #+1%&%#"2 =%*/ #' %'<'%*%3#&A)B>&-B-'*5 =%*/ # B-#'%', &%Y- */#* )0 */- n',&%./ -,":/&3:"2/ 0+*+"-: @%'A- )"%-'*#*%)' 0")B */->"-.-'* *)=#"6 # 0+*+"- -3-'* %. '-A-..#"%&2 #=#2 0")B */- 6-%A*%A A-'*"-5 !&) %. %B>)..%4&-5 =/%&- (36()&#AY. c^bdefdc #'6 */+. A#'')* -1>"-.. )"%-'*#*%)' #* #&&:

v)* #&& %'.*#'A-. )0 FGH. =%*/ %'<'%*%3#& )" ,-"+'6%3- A)B>&-B-'*. #"- #+1%&%#"%-.5 /)=-3-": F&&*/"-- /#3- B#%'?3-"4 +.-. =%*/ 6->%A*%3- .-A)'6#"2 >"-6%A#*-. 7w95 B)">/)&),%A#&&2 %6-'*%A#& *) */-#+1%&%#"2 +.-. 7E9: u=) A"+A%#& 6%Q-"-'A-. 6%.#B4%,+#*- */- .*"+A*+"-.Z */- 6->%A*%3- A)'.*"+A*%)'.#&=#2. %'3)&3- A)'A"-*- B)*%)' %' .>#A-p*%B-5 >"-6%A*-6 %0 */-2 %'A&+6- */- "))* B-'*%)'-6 #4)3-5 #'6=/-' */- 6->%A*%3- %. # ,-"+'65 */- .-'*-'A- &#AY. */- #.>-A*+#& >")>-"*%-. )0 */- >"),"-..%3- .%'A- */-B#%' 3-"4 )0 */- A&#+.- %. ')* >"),"-..%3-:x #: ,"#!( -*$.%*!"#$ #&,+tB s+- .#%4# "+..): [u/-2 #"- &))Y%', 0)" .)B-)'- =/) Y')=. R+..%#':\

#'6#"?N>& .--Y?'6) .)B-)'- */#* Y')=?.4r?N., R+..%#'4: q 6)-'*- &!+ ('/0$*!"#$ #). >)+A).: [u/- >#*%-'* %. %B>")3%', &%**&- 42 &%**&-:\

*/- >#*%-'* %"?>"-.?N., %B>")3-?'6) #*?*/- &%**&-y #: ,"#'+ *)B#'6) A-"3-r# A)B ). #B%,).: [I =-'* #")+'6 6"%'Y%', 4--" =%*/ B2 0"%-'6.:\

#'6#"?>"-* *#Y-?'6) 4--" =%*/ */- 0"%-'6.4: n&- 1$+ -'*"#'6) >-&# T)"-.*# #6-'*"): [L- =-'* )' %'*) */- 0)"-.*:\

L- %"?>"-*?N., -'*-"?'6) 0)"?*/- 0)"-.* *)?%'.%6-I A)'A&+6- */#*5 r+.* #. */-"- #"- T#3)+". )0 */- A#+.#*%3-p#,-'*%3- ! 7J)&&% K L#"&-2 CDDE95 !"VW7M+-"3) CDDN9 #&.) A)B-. %' T#3)+".: @%'A- */- T#3)+". )0 !VW #>>-#" 4)*/ #. #+1%&%#"%-. #'6 #. 7>#"*)0 9 B#%' 3-"4.5 %* =)+&6 4- %'*-"-.*%', *) .-- =/-*/-" */- .#B- %. *"+- )0 )*/-" ! /-#6.:

!"

M)=>-"5 n&%$#4-*/: CDDz u/- ,-)B-*"2 )0 %'*-">"-*#4&- 0-#*+"-.Z I'T %' n',&%./ #'6 @>#'%./: <(3":$(:, {8X: 8D|k}

M+-"3)5 H#"%# M"%.*%'#: CDD~ =(2&!,. (2 '():,: (/:j: u/-.%.5 H#..#A/+.-**. I'.*%*+*- )0 u-A/')&),25M#B4"%6,-5 HF:

J)&&%5 R#Q#-&&# K L-%6% L#"&-2: CDDz J&#3)". )0 !Z M)'.+B%', "-.+&*. %' I*#&%#' #'6 n',&%./: I'(#+&# �-B>A/%'.Y2 K R)+B2#'# @&#4#Y)3# 7-6.:95 #.9,>2$(' &3?$&)&,.: j)"6"-A/*Z @>"%',-":

L#"&-25 L-%6% K n&%$#4-*/ R%**-": CDD� (-".)' #'6 '+B4-" %' >")')+'.Z # 0-#*+"-?,-)B-*"%A #'#&2.%.:<(3:$(:, �{h: k{C|EC}

@A/B%**5 M"%.*%'#: CDDz @-B%?A)>+&#.: I' (#+&# �-BA/%'.Y2 K R)+B2#'# @&#4#Y)3# 7-6.:95 #.9,>2$('&3?$&)&,.5 8C8|8k} @>"%',-":

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

=> '$" $()*" +* !"+./ !)+,10&> 2$" +* %33+&"-0#> 2$" $%* !)+,1 1$" $()*"0

I +> J<KL;MNK6 LOKKP '$" $()*" 4%* !)+,10=> J<KL;LMQR6 LOKKP '$" $()*" 4%* !)+,-+./0&> J<KL;MNKP 2$" 4%* %33+&"-0#> J<KL;LMQRP 2$" 4%* S%33+&+./ 5 !)+,-+./ 1$" $()*"T0

U&V+##", W <1"C$+#.' @DXXY6 DX3XA +*-'" ()+( $, ()" 0*"FZ2 0"*$.#6 =.() ()" !"#0"*9"&( @B)$&)(70$&+117 .&&'**"# B$() ',+&&'%+($8"%A +,# ()" $%&"#0"*9"&( B"*" 0'*"17 *"%'1(+($8"6 +,# 1+&/"# ()" (":F0.*+1 :"+,$,- .9 +,("*$.*$(7 &+**$"# =7 ()" [\2 0"*9"&( &.,%(*'&($.,> ?)" $%&"#0"*9"&( %'=%"]'",(17+&]'$*"# ()" +,("*$.*$(7 *"+#$,-6 +,# ()"*"=7 "C0+,#"# $(% &.,("C(% (. $,&1'#" 0+%( &.',("*9+&('+1%@B)$&)6 6"3 E+(*$#.' DXXX6 *"]'$*" #.'=1" :+*/$,- .9 +,("*$.*$(7A> "̂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`.() <%0"&( )"+#% +*" %0"11"# .'( =7 ()" 0+*($&$0$+1 :.*0).1.-76 +,# ()"["*9 )"+# $% %0"11"# .'( =7 $%&"> ?)" 0'aa1" ()+( U&V+##", W <1"C$+#.' @DX3XA 1"+8" .0", $% B)76+9("* &.F"C$%($,- B$() ()" $%&"#0"*9"&( 9.* %"8"*+1 )',#*"# 7"+*%6 ()" *"%'1(+($8" !"#0"*9"&( "%%",($+117#$%+00"+*"# +*.',# ()" ",# .9 ()" 34() 5>

<,.()"* &)+,-" ()+( )+00","# +*.',# ()" %+:" ($:" B+% ()" 1.%% .9 B)+( ^$%%"* @3bcGA &+11"#()" 6%**+&%, @3+A> ?)$% &.,%(*'&($., B+% &.::., %$,&" +( 1"+%( ()" 3G() 56 0"*%$%($,- +% ()" 0*"9"**"#"C0*"%%$., .9 +, $:0"*9"&($8" 0+%%$8" &1+'%" ',($1 ()" "+*17 34XX%6 B)", ()" 0*.-*"%%$8" 0+%%$8" @3=A="-+, (. (+/" .8"*> ?)"*" +*" ()*"" 9'*()"* 0'aa1"% )"*"; B)7 #$# ()" 0+%%$8+1 9+11 .'( .9 '%"6 B)7 )+#()" 0*.-*"%%$8" 0+%%$8" 0*"8$.'%17 ="", ',8+$1+=1"6 +,# ).B #$# $( ="&.:" -*+::+($&+1d

!" 0*.0.%" ()+( ()"%" &)+,-"% 9.11.B"# 9*.: + *"+,+17%$% .9 ()" *"%'1(+($8" <%0"&( )"+# @<KL;MNKA>[*$.* (. ()" 1+(" 34() 56 ()$% )"+# B+% ()" :+*/"# :":="* .9 + 0+$* .9 <%0 )"+#%6 B$() ()" ',:+*/"#)"+#6 %0"11"# .'( =7 #+./7 ="+*$,- ()" 9"+('*" [*.&"%% *+()"* ()+, e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a" (B. 8"*%$.,% .9 ()" %+:" <%0 )"+#6 +,# +*" ()'% $,&.:01":",(+*7 #$%(*$='($.,>

<*.',# ()" ",# .9 ()" 34() 56 <KL;MNK +,# <KL;LMQR %01$( $,(. (B. 9',&($.,+1 0*.f"&($.,%6 B$()<KL;MNK .=1$-+(.*$17 &+**7$,- ()" LOKK 9"+('*"6 +,# <KL;LMQR ="&.:$,- + 0'*"17 +%0"&('+1 )"+# B$(),. +*-':",(F%(*'&('*" 0*.0"*($"%> <KL;MNK ()'% ="&+:" B)+( $% &.::.,17 /,.B, +% ()" 0+%%$8" 1$-)(8"*=6 B)$1" <KL;LMQR ="&+:" + 8$"B0.$,( +%0"&( )"+# ",&.#$,- $:0"*9"&($8$(7> <, $:0"*9"&($8" 0+%%$8"B.'1# ,.B )+8" <KL;LMQR +,# LOKK ., %"0+*+(" 0*.f"&($.,%6 +,# B.'1# ()'% =" %0"11"# .'( +% $, @3=A*+()"* ()+, @3+A6 B$() #". *"+1$a$,- LOKK +,# #+./ *"+1$a$,- <KL;LMQR> ?)" &.,&.:$(+,( #$%+00"+*+,&".9 ()" ,.,F0+%%$8" <KL;MNK 1$/"B$%" :+#" @3&A ',+8+$1+=1"6 +,# %. ()" *"01+&":",( .9 ()" !"F0"*9"&(=7 ()" $%&"F0"*9"&( B+% &.:01"("#>

2:=$&/6 \+8$#> DXXg Z, ()" %(*'&('*" .9 *"%'1(+($8" 0+*($&$01"% $, 2,-1$%)> 8+./)+*1+9 :.;)+3< Ghi>GhhjGbk

E+(*$#.'6 l+=$,"> DXXX> ?)" -*+::+($&+1 $,-*"#$",(% .9 &.',("*9+&('+1$(7> 8+./)*1+9 :.;)+3< G3I>DG3jGbk

U&V+##",6 ?).:+% W <*(":$% <1"C$+#.'> DXXm <'C$1$+*7 %"1"&($., +,# &.',("*9+&('+1$(7 $, ()")$%(.*7 .9 2,-1$%) +,# n"*:+,$&> E, o'((+ U> p+*(:+,, W qr%a1s U.1,r*t @"#%>A6 =(>6%3%1+&"*1)-+"* +. ?"3>%.+9 *<.1%@6 DGcjDYk <:%("*#+:; o.), `"u+:$,%>

U&V+##",6 ?).:+% W <*(":$% <1"C$+#.'> DXv ["*9"&(%6 *"%'1(+($8"%6 +,# +'C$1$+*$"% $, "+*1$"* 2,F-1$%)> 8+./)+*1+9 :.;)+3< _3i> G4bj_Dw

^$%%"*6 V> ?)> 3bcx A. $+*1(3+9%, *<.1%@ (B 1$" C./,+*$ ,%./)%/"6 8.1> EEE6 k q"$#",; 2> o> `*$11>

A Pronominal Feature Geometry for Cayuga Elijah Deer (University of Western Ontario)

The morphology of Cayuga (Canada, Iroquoian) includes bound, obligatory pronominal marking on verbs, which inflect for the phi-features of person (1, 2, 3), number (SG, DU, PL) and gender (M, F, N), and the grammatical relations of subject and object as well as exclusivity distinctions. The phi-features and exclusivity features are combined to give 14 “base” distinctions (Sasse and Keye 1998:44), these combinations in turn being used in the various combinations of subject and object. (1) provides a brief paradigm.

In spite of the combinatorial potential of these features not all of the possible feature combinations in Cayuga have a one-to-one correspondence with a unique morpheme. For example, although dual number is expressed in first-and second-person morphemes (1d), it is not expressed in third-person morphemes, the plural feature being used for all third-person numbers over one (1b); in the case of a masculine object, there is only one morpheme used regardless of whether the subject is masculine or feminine—there are not two unique morphemes individually encoding each subject feature; and finally, the question marks in (1f) show that the morpheme encodes several different combinations, which only take on relevance in context— this particular morpheme can have referents resulting in glosses such as ‘you (DU/PL) saw her’ and ‘you (DU/PL) saw them (DU/PL)’, among others.

Theoretically, it’s desirable to have a representation that accounts for the combinations and syncretisms exemplified in the Cayuga pronominal paradigm. The current research accounts for the Cayuga facts by claiming that they are constrained through the pronominal feature hierarchy as developed by, among others, Harley and Ritter (2002). Though this research is ongoing, it is predicted that the Cayuga facts will support the geometric organization of features proposed in that work.

References Harley, H. and Elizabeth Ritter. 2002. Person and number in pronouns: A feature-

geometric analysis. Language 78:482-526. Sasse, Hans-Juergen, and Alfred Keye. 1998. Far more than one thousand verbs of

Gayogohono (Cayuga): A handbook of Cayuga morphology. Unpublished ms. Universität zu Köln, Köln, Germany and Woodland Cultural Centre, Brantford, Ontario.

Les asymétries koj ‘qui’ -kakvo ‘quoi’ en bulgare Elena Dimova (Université de Montréal)

L’objectif de cette communication sera de proposer une nouvelle explication au comportement syntaxique différent des syntagmes interrogatifs koj ‘qui’ et kakvo ‘quoi’ en bulgare. Traditionnellement, pour rendre compte des différences dans l’ordre de déplacement dans les questions à mouvement multiple, on fait appel au caractère [+/-animé] des syntagmes antéposés. Comme l’illustrent les exemples ci-dessous, le sujet [+ animé] koj ‘qui’ doit occuper la première position, alors que les deux ordres sont possibles si le sujet est exprimé par le syntagme [– animé] kakvo ‘quoi’ :

(1) a. Koj kade padna? qui où tombé

b. * Kade koj padna? où qui tombé ‘Qui est tombé où?’

(2) a. Kakvo kade padna?

quoi où tombé b. Kade kakvo padna?

où quoi tombé ‘Qu’est-ce qui est tombé où?’

À l’aide de faits nouveaux, nous montrerons qu’une telle approche n’arrive pas à expliquer toutes les données du bulgare et nous examinerons d’autres propriétés des syntagmes interrogatifs, telles que la discursivité et la référentialité. Nous proposerons que la possibilité pour un syntagme Qu- de se positionner au début d’une question multiple est attribuable non pas à son caractère [+/-animé] ni à des conditions purement syntaxiques, mais d’abord et avant tout à sa capacité d’être interprété comme lié au discours. Nous regarderons de plus près la structure interne des syntagmes simples koj‘qui’ et kakvo ‘quoi’, ainsi que des syntagmes complexes de type koj N ’quel N’ et kakvo N ’quelle sorte de N’ et nous discuterons la façon dont ces syntagmes peuvent ou non être analysables comme D-liés.

L’étude proposée permettra d’éclaircir certains faits du bulgare qui n’ont pas trouvéd’explication dans les recherches précédentes, d’expliquer le statut apparemment « particulier » du syntagme interrogatif koj et de justifier la présence d’une positioninitiale de topique dans la périphérie gauche pouvant recevoir les syntagmes liés au discours.

Comparing Acceptability Judgments of Serbo-Croatian Monolinguals and Bilinguals

Vedran Dronjic (University of Toronto) & Rena Helms-Park (University of Toronto Scarborough)

This study investigated differences in acceptability judgments between Serbo-Croatian-speaking monolinguals living in Serbia (n = 20) and Serbo-Croatian/English bilinguals living in North America (n = 20). The bilinguals were born and educated in a Serbo-Croatian speaking country, had moved to North America as adults, and maintained Serbo-Croatian as their dominant language. The acceptability judgment instrument consisted of 20 sentences produced by bilinguals and 20 produced by monolinguals. The participants (a different group of speakers from the ones who produced the sentences) were asked to decide whether the sentences were well formed and, if not, to correct them. The bilingual items diverged from the monolingual standard at various levels of linguistic structure, as documented in various language-contact situations (e.g., Laufer, 2003; Jarvis, 2003). Quantitative and qualitative analysis of the data revealed that the bilinguals rejected more sentences than the monolinguals. The bilinguals’ tendency to reject well-formed sentences, possibly an indicator of a reduction in grammatical/stylistic range and/or of heightened metalinguistic awareness, was considerably stronger than their tendency to accept malformed utterances. Consistent with the findings of numerous other studies, the observed differences in bilinguals’ judgments were predominantly lexical. An important collateral finding of this study was that monolinguals’ judgments can be so inconsistent as to render a substantial number of both target and distractor items useless for normative linguistic purposes. Acceptability judgment tasks are viewed as potentially tapping both linguistic competence and metalinguistic knowledge (Ellis, 2005; see also Schütze, 1996). In this regard the results obtained here could be taken to imply that the bilingual participants here diverged from the monolingual controls both in their implicit and explicit L1 knowledge. In follow-up studies, two methods by which the precise nature of the observed differences could be ascertained would be through complementing the acceptability judgment task with on-line psycholinguistic (reaction-time) and neurolinguistic (ERP, MEG, fMRI) measures. References Cook, V. (2003). Introduction: The changing L1 in the L2 user’s mind. In Cook, V. (Ed.).

Effects of the second language on the first. Clevedon: Multilingual Matters. Ellis, R. (2005). Measuring implicit and explicit knowledge of a second language: A

psychometric study. Studies in Second Language Acquisition, 27(2), 141-172. Jarvis, S. (2003). Probing the effects of the L2 on the L1: A case study. In Cook, V. (Ed.).

Effects of the second language on the first. Clevedon: Multilingual Matters. Laufer, B. (2003). The influence of L2 on L1 collocational knowledge and on L1 lexical

diversity in free written expression. In Cook, V. (Ed.). Effects of the Second Language on the First. Clevedon: Multilingual Matters.

Schütze, C. T. (1996). The empirical base of linguistics: Grammaticality judgments and linguistic methodology. Chicago and London: The University of Chicago Press.

Inaccusativité et télécité en français medieval Monique Dufresne (Université Queen's) & Fernande Dupuis (UQAM)

Cette communication porte sur un changement qui a affecté les constructions inaccusatives (CI) dans l’histoire du français : la sélection de l’auxiliaire. Ces constructions posent des problèmes intéressants dans les langues romanes et germaniques. De nombreux travaux y ont été consacrés et attestent la difficulté à en fournir une analyse qui pourrait refléter leurs propriétés syntaxiques, leur aspect et leur structure d’évènement. Rappelons que le choix de l’auxiliaire constitue une caractéristique définitoire des langues qui possède deux auxiliaires, avoir et être, pour les temps composés. Selon Legendre&Sorace (2003), le choix de l’auxiliaire constitue un critère fiable pour déterminer la classe de verbes de mouvement et de changement d’état, soit ceux qui forment le ‘cœur’ de cette classe. Nous pouvons dès lors poser deux hypothèses : 1) nous devrions toujours rencontrer l’auxiliaire estre dans les CI; 2) si nous observons un changement d’auxiliaire nous ne sommes pas en présence d’une CI. Toutefois en français médiéval, un même verbe peut s’employer ave l’un ou l’autre des auxiliaires. À cet égard, Moignet (1976 :183) souligne que « la syntaxe de l’ancien français a moins de rigueur que la moderne. Beaucoup de verbes admettent l’un et l’autre auxiliaire. » (1) … qu'il avoit semons dis tanz de gent qu’il n’avoit venu a ses noces Graal, 128, 2. qu’il avait invité beaucoup plus de gens qu’il n’en était venu à ses noces

Une question se pose alors : cette variation est-elle libre ou reflète-t-elle un changement dans la phrase ? Buridant souligne que « pour les verbes de mouvement au sens large, la prise en compte de la durée de procès entraine le verbe avoir. » Cette durée est généralement marquée par un « adverbe quantificatif » (positif ou négatif) ou par un complément « indiquant une mesure de distance ou de temps », comme on peut le constater un (1). Il souligne aussi qu’un changement d’auxiliaire s’accompagne généralement d’une modification dans la structure argumentale ou sémantique, comme en (2) : (2) Il n’orent mie .iiii. liues alé qu’anmi la voie ont .i. vilain trové. Char.Nîm, 875

Ils n’avaient pas parcouru quatre lieues lorsqu’ils ont vu un vilain au milieu de la voie.

Pour rendre compte de cette alternance et de sa disparition en français moderne,

nous proposons que le choix de l’auxiliaire être ne peut être attribué aux seules propriétés non agentives de ces constructions inaccusatives, comme le suppose les approches par projection lexicale. Notre analyse se situe ans la lignée des travaux récents sur les propriétés des verbes inaccusatifs en grammaire générative et nous voulons démontrer l’importance fondamentale du rôle des catégories fonctionnelles dans l’identification de ces constructions. Notre modèle partage avec celui de Borer (2005) les présupposés que le système de projection est défini sur la structure évènementielle de l’ensemble du SV dans laquelle le verbe apparait, plutôt que d’être projeté seulement à partir des propriétés du verbe, définies dans le lexique. Nous fournirons des arguments pour montrer que la

dérivation des inaccusatives est un phénomène syntaxique qui implique une projection exprimant la télicité dans le sens des analyses de Borer et de van Hout (2004). Notre hypothèse sur le rôle de la télicité dans les CI permet de rapprocher deux constructions aux temps composés qui, de prime abord, semblent faiblement apparentées : les inaccusatives impersonnelles et les constructions transitives avec avoir. Bibliographie Belletti, Adriana (1988). The Case of Unaccusatives, LI 19 (1): 1-34. Borer, Hagit (2005). Structuring Sense, 2: The Normal Course of Events, New York:

OUP. Chomsky, Noam (2006) On phases, dans Robert Freidin &al. (réds), Foundational Issues

in Linguistic Theory, Cambridge: MIT Press. Legendre, G. & A. Sorace (2003). Auxiliaires et intransitivité en français et dans les

langues romanes, dans D. Godard (réd.) Les langues romanes: problèmes de la phrase simple. Paris: CNRS éditions. 185-233.

Van Hout, A. (2004) Unaccusativity as Telicity Checking dans Alexiadou, A. & al. (réds). The Unaccusativity Puzzle: Explorations of the Syntax-Lexicon Interface, Oxford: OUP.

Resource constraints, scope and underspecification Veena D. Dwivedi (Brock University), Leanne Angus (SUNY Buffalo) &

Matthew Goldhawk (Brock University)

We present results from 3 self-paced reading experiments that examine the role that verbal working memory span (WM) plays in sentence comprehension. Specifically, we show that certain stimuli (ie those with no preference in interpretation, as shown via off-line tasks), eg (i) Every jeweler appraised a diamond, are processed deeply by high span individuals, in contrast to other sentences with a similar scope ambiguous structure but which have a heavy bias off-line. That is, sentences that could be ambiguous (ii) Every kid climbed a tree but where off-line studies indicate a heavy pragmatic bias for a particular interpretation, are left as underspecified constructions by high span individuals. In other words, when lexical-pragmatic biases are present, high span readers do not pay attention to sentence interpretation—but low span individuals do. When these biases are absent, such that only structural considerations play a role in interpretation, high spanreaders do the work of interpreting scope ambiguous sentences. These findingscorroborate ERP work by Nakano, Saron & Swaab (2010) and Bornkessel et al. (2004), which showed that high span vs. low span individuals produced qualitatively differentERP components when perceiving similar linguistic stimuli.

In the experiments discussed below, a verbal WM task (Daneman & Carpenter, 1980) was administered to participants prior to the on-line study. In Exps 1 & 2, scope ambiguous stimuli as in (ii) were selected where off-line tests indicated a heavypreference for surface scope or plural continuation, (iii) The trees were in the park. In contrast, Experiment 3 used stimuli where off-line norming indicated no preference forsurface or inverse scope; these stimuli were truly ambiguous, as in (i).

In Experiment 1, 80 participants’ only task was to read 24 2-sentence ambiguousdiscourses as in (ii) as well as Unambiguous conditions (iv) Every kid climbed that/those trees (in addition to filler stimuli). Results indicated no difference in reading times forcontinuation sentences in any region, in any contexts between WM groups. The lack of difference for plural vs. singular continuations indicated that both high span and low spanreaders interpreted scope ambiguous sentences as underspecified. These findingshighlighted the fact that shallow processing is not the result of a parser that lacks theallocational resources to conduct a detailed parse but instead shows thatunderspecification is a strategic use of computational resources. Experiment 2 used thesame stimuli as Experiment 1, except that now critical trials were followed by a questionsuch as (v) How many trees were climbed? (all filler trials were also followed by comprehension questions). In other words, now participants (N=48) were given theexplicit goal of computing scope (Swets et al., 2008). Recall that these stimuli wereheavily biased for the plural continuation. Results indicated that only the low span groupdid the work of interpreting these sentences, as they showed longer RTs for singular(dispreferred) vs. plural continuation sentences after ambiguous contexts, whereas no differences were observed for the high span WM group. Finally, in Exp 3, 24 differentstimuli (see (i)) were used where the same design as Exp 2 was employed. Preliminary findings (N=36) indicate that now high span readers exhibit significantly longer RTs forsingular vs. plural continuation sentences that follow scope ambiguous vs. unambiguous contexts. In other words, high span readers pay attention to sentence

interpretation whensemantic interpretation is compositional, not conceptual. References Bornkessel, I. D., Fiebach, C., & Friederici, A. D. (2004). On the coast of syntactic

ambiguity in human language comprehension: An individual differences approach. Cognitive Brain Research, 21, 11-21.

Daneman, M., & Carpenter, P. A. (1980). Individual differences in working memory and reading. Journal of Verbal Learning & Verbal Behavior, 19, 450-466.

Nakano, H., Saron, C, & Swaab, T. (2010). Speech and span: Working memory impacts the use of animacy but not of world-knowledge during spoken sentence comprehension. Journal of Cognitive Neuroscience, 22, 2886-2898.

Swets, B., Desmet, T., Clifton, C., & Ferreira, F. (2008). Underspecification of syntactic ambiguities: evidence from self-paced reading. Memory & Cognition, 26, 201-216.

When Epistemic Modals Need Imperfective Aspect Ewelina Frackowiak (University of Ottawa)

Polish (and Russian) imperfective constructions (1B) support the idea that the imperfective is a semantically vacuous morpheme whose presence is guaranteed by specific conditions that rule out the perfective. The paper enriches the recent discussions concerned with the puzzle of how to account for various readings of the imperfective (Cipria & Roberts 2000, Hacquard 2006) by providing a semantic analysis of an interpretation not attested in languages discussed so far (Romance).

The construction in (1B), Existential Factual Imperfective (EFI), (Grønn 2003), entails the existence of a past event but also makes a claim about “a current state of affairs” (Grønn 2003: 26). Consider that (1B) requires a discourse topic (question (1A)) to be interpreted. Given (1A), (1B) states that in view of what is known Piotr is the best candidate to advice Tomek. The perfective, (2), is infelicitous as an answer to the question in (1A). The puzzle: what property of the imperfective is responsible for the reading in (1).

I propose that EFIs carry a silent epistemic modal (see (3) and (4)). I follow Hacquard (2006, 2010), in assuming that modals are relative to an event of evaluation and that a modal operator has an event variable that must be bound locally. The modal is restricted by a modal base that is relativized to the speaker and the speech time. This is captured in (4) where the variable e0 from the modal base is co-indexed with a speech event represented syntactically (Hacquard 2010). The CON(e0) picks out a set of propositions that are beliefs of the agent of e0 at the time of e0. In EFI this accessibility relation is further modified by a proposition denoted by a TP with imperfective AspP in its scope (TP1 in (4)). To the set of beliefs the proposition “Piotr broke up with a girlfriend” is added. In the nuclear scope of the Modal I locate the proposition that answers the question from the discourse topic: ‘Piotr be the best to advice Tomek’. If Piotr broke up with a girlfriend in all of the speaker’s doxastic alternatives, then in all the worlds compatible with what the speaker knows Piotr is the best person to give Tomek advice.

My analysis predicts that a perfective TP cannot replace the imperfective TP1 in (4). CON(e) is defined when e has a propositional content; hence, it must be modified by a proposition (type <s,t>). Any perfective construction is of type <t> since its proposition is bound by a topic situation.

References Bhatt, R. 2006. Covert Modality in Non-finite Contexts. Berlin: Mouton de Gruyter. Borik, O. 2006. Aspect and Reference Time. New York: Oxford University Press. Cipria, A. and Roberts, C. 2000. Spanish Imperfecto and Pretérito: Truth Conditions and

Aktionsart Effects in a Situation Semantics. Natural Language Semantics 8: 297-347.

Grønn, A. 2003. The Semantics and Pragmatics of the Russian Factual Imperfective. PhD Dissertation, University of Oslo.

Hacquard, V. 2006. Aspects of Modality. PhD Dissertation, MIT, Cambridge, MA. Hacquard, V. 2010. On the Event-Relativity of Modal Auxiliaries. Natural Language

Semantics. 18(1): 79-114. Kratzer, A. 2007. Situations in Natural Language Semantics. Stanford Encyclopedia of

Philosophy. http://plato.stanford.edu/entries/situationssemantics/ Percus, O. 2000. Constraints on Some Other Variables in Syntax. Natural Language

Semantics 8: 173-229. Portner, P. 2003. The (Temporal) Semantics and (Modal) Pragmatics of the Perfect.

Linguistics and Philosophy 26:511–545.

Immigrant Perspectives on Official Bilingualism in Canada Albert Galiev (University of Calgary)

Canada has been known as a country that is bilingual when it comes to language policy and multicultural when it comes to cultural diversity. In fact, the promotion of official bilingualism and linguistic duality has been one of the top priorities of the Government of Canada since the Official Languages Act in 1969. This study investigates the perceptions and experiences of a group of adult immigrant students learning one of Canada’s official languages regarding the status of bilingualism. It attempts to answer how promoted official bilingualism and the dominant English reality are reconciled and explained and how immigrants’ voices are placed in the discourses on bilingualism in the monolingual environment. It argues that the macro- level discourse on official bilingualism maintained and promoted by the government doesn’t relate to micro-level discourse produced by the Canadian public, especially new Canadians.

The research stipulates that while the immigration numbers have constantly been increasing in Canada (Statistics Canada, 2006), immigrants are often left out of the scope of discussion on languages, both official and heritage ones. In fact, the discourse and research on official bilingualism have largely been shaped by promoting bilingualism only among Anglophone and Francophone (Mady, 2008) communities. This is even more surprising in the light of the fact, that immigrants have consistently been more supportive for bilingualism and multiculturalism than any other group in Canada (Dasko, 2003; Parkin & Turcotte, 2004; The evolution of public opinion on official languages in Canada, 2006).

The following study aims to fill the gap by focusing on adult immigrant language learners by exploring how through language education experiences in Canada’s official languages immigrant students perceive and negotiate meanings of official, institutional and personal bilingualism and relate them to the monolingual environment of Alberta.

In-depth open-ended individual and focus group interviews in conjunction with an attitudinal survey were conducted with 50 ESL/French learners of immigrant descent. The results suggest that for newcomers it is often a challenge to match the promoted discourse on official bilingualism which they encounter through their pre-Canadian experiences and the real picture of bilingualism in a predominantly English speaking province. Frequently, despite strong federal discourse their perceptions of bilingualism are quite vague due to very fact of their experiences in a solely monolingual context.

The study tries to make possible recommendations regarding informing language policy makers of possible implications for accommodating immigrant voices in the official bilingualism discourse. References Dasko, D. (2003). Public attitudes towards multiculturalism and bilingualism in Canada.

In Canadian and French perspectives on diversity. Gatineau, Québec: Canadian Heritage.

Mady, C. (2008). Opening the doors to official language learning for allophones . In The state of French-Second-Language Education in Canada 2008 (pp. 14-15). Ottawa: Canadian Parents for French.

Parkin, A., & Turcotte, A. (2004). Bilingualism: part of our past or part of our future?

The CRIC Papers. Ottawa: Centre for Research and Information on Canada. Statistics Canada. (2006). Immigration in Canada: a portrait of the foreign-born

population, 2006 Census. Retrieved on Jan. 12, 2009 from http://www12.statcan.ca/censusrecensement/2006/as-sa/97-557/index-eng.cfm

The evolution of public opinion on official languages in Canada. (2006). Ottawa: Office of the Commissioner of Official Languages.

Language Barriers To Integration Of Adult ESL Learners Albert Galiev & Sepideh Masoodi (University of Calgary)

The Language Instruction for Newcomers to Canada (LINC) program, established in 1992, is a federally funded program with the aim of facilitating the social, cultural and economic integration of immigrants and refugees into Canada through language instruction in either English or French (Citizenship and Immigration Canada, 2004). By exploring discourses of 30 ESL immigrants regarding their language practices within and outside the program, the study argues that while acquiring language proficiency in English for LINC students may be essential for social interaction and consequently integration, such integration could not occur without interaction originating from native speakers.

The analysis of studies on LINC allows concluding that integration of immigrants through interaction is the prevailing topic of enquiry. Such discourse puts integration through interaction primarily as responsibility of immigrants and is mostly presented in governmentally funded reports aiming to assess and evaluate the efficiency of the LINC program (Citizenship and Immigration Canada, 2004). The concrete challenges of integration due to language competency are also pointed out by Citizenship, Immigration and Multiculturalism Minister Jason Kenney asserting that "people who arrive in Canada need to be able as quickly as possible to have competency in one of our two official languages as a pathway to economic and social integration” (2009).

However, little has been researched regarding immigrants own understanding of integration and their social interactional experiences within and outside the LINC program. Relevant studies suggest that immigrants language learning experiences can not be viewed and analyzed out of the social and cultural context and should consider a number of contributing factors that could either hinder or facilitate their language learning and integration (Derwing & Thomson, 2005; Kouritzin, 2000; Thomson & Derwing, 2004).

Through in depth semi-structured interviews conducted individually and in focus groups, immigrants reflect on their language learning and interactional experiences both within and outside the program. The findings demonstrate that while immigrants recognize the importance of language interaction and would like to pursue further integration for themselves and especially for children, the process of interaction is hindered by the fact that their language practices are limited to the LINC program and their local communities. Furthermore, often their attempts to extend their language practices to larger interactional circles fail due to language and cultural barriers.

The study suggests further exploration of the notion of immigrant integration through interaction and language acquisition by bringing the discourses of both sides together. It provides possible implications of fostering immigrant integration through interacting with native speakers. References Citizenship and Immigration Canada. (2004). Evaluation of the Language Instruction for

Newcomers to Canada (LINC) program. Citizenship and Immigration Canada. Retrieved from http://www.cic.gc.ca/english/resources/evaluation/linc/exec-summary.asp

Derwing, T. M., & Thomson, R. I. (2005). Citizenship Concepts in LINC Classrooms. TESL Canada Journal, 23(1), 44-62.

Kenney, J. (2009, March). Good Citizenship: The Duty to Integrate. Speech presented at the Huron University College’s Canadian Leaders Speakers’ Series, London, ON. Retrieved from http://www.cic.gc.ca/english/department/media/speeches/2009/2009-03-18.asp

Kouritzin, S. (2000). Immigrant Mothers Redefine Access to ESL Classes: Contradiction and Ambivalence. Journal of Multilingual and Multicultural Development, 21(1), 14-32. Thomson, R. I., & Derwing, T. M. (2004). Presenting Canadian Values in LINC: The Roles of Textbooks and Teachers. TESL Canada Journal, 21(2), 17-33.

Ellipsis, Focus and the Syntax-Phonology Interface: The case of the Empty Noun Construction in Persian

Ghaniabadi, Saeed (Sabzevar Tarbiat Moallem University)

In the Empty Noun Construction (ENC) in Persian, two conditions must be met: (i) there must be a discourse or linguistic antecedent (like in (1a) and (1b) respectively), and (ii) there must be at least one postnominal (1a) or prenominal modifying element (1b) in the noun phrase. The head noun can be left unpronounced when there is an attributive adjectival head modifier in the Ezafe domain. If there is no such modifier, the empty noun is licensed by a prenominal modifier (the nominal modifiers are in bold and the elided noun is struck through). (1) a. Empty nouns with postnominal modifiers

[A shopkeeper’s response to a customer who asks for green pens] [xodkâr sabz] na-dâr-am, [xodkâr âbi] mi-xây?

pen green NEG-have-1SG pen blue DUR-want.2SG ‘I don’t have a green [one], do you want a blue [one]?’

b. Empty nouns with prenominal modifiers hamiše avvalin jomle [ saxt-tar-in jomle]-e. always first sentence hard-COMP-SUP sentence-is ‘The first sentence is always the hardest.’

In this paper I investigate the syntactic and semantic properties of the ENC in

Persian and provide an analysis within the framework of Distributed Morphology (Halle & Marantz 1993, and subsequent related works). The underlying claim is that the non-pronunciation of the head noun is linked to the information-structural properties of the ENC (along the lines of Rooth 1992a, 1992b; Gengel 2007, among others). Under this view, the ENC is derived through the interaction between the following two information-structural features: (i) the E-feature (adapted from Merchant 2001), which ensures that the head noun is identical with its counterpart in the antecedent and specifies the head noun for non-pronunciation; (ii) the focus feature F, which specifies the remnant modifier as an element which is in some kind of contrastive relationship with its corresponding element in the antecedent. The interaction between these two features is implemented in the syntax in a phase-based derivation. It will be shown that the proposal offered here accounts for all instances of the ENC in Persian.

Given that research in the area of ellipsis in grammar has typically been closely connected with research at the syntax and semantics interface, my analysis of the ENC in Persian is innovative on the grounds that it looks at ellipsis using current theories of syntax/phonology interactions (e.g. Pak 2008). The idea behind my approach is that there has to be an articulated theory of what it means for material in syntactic structures not to be pronounced, just as there are articulated theories of what is actually pronounced in syntactic trees. References Halle, Morris and Alec Marantz. 1993. Distributed Morphology and the pieces of

inflection. In: Hale, Kenneth and Samuel Jay Keyser (eds.) The View from

Building 20: Essays in Linguistics in Honor of Sylvain Bromberger. Cambridge, Massachusetts: MIT Press, 111-176.

Rooth, Mats. 1992a. A theory of focus interpretation. Natural Language Semantics 1: 75-116.

Rooth, Mats. 1992b. Ellipsis redundancy and reduction redundancy. In: Steve Berman and Arild Hestvik (eds.) Proceeding of the Stuttgart Ellipsis Workshop, Report Series Sprachtheoretische Grundlagen fur die Computerlinguistik, no 29, SFB 340. IBM Heidelberg.

Gengel, Kirsten. 2007. Focus and ellipsis: a generative analysis of pseudogapping and other elliptical structures. Ph.D. dissertation, Institut für Linguistik/Anglistik der Universität Stuttgart.

Merchant, Jason. 2001. The syntax of silence: Sluicing, islands, and the theory of ellipsis. Oxford: Oxford University Press.

Pak, Marjorie. 2008. The postsyntactic derivation and its phonological reflexes. Ph.D. dissertation, University of Pennsylvania.

e Semantics and Pragmatics of the Skwxwú7mesh Evidential lhkwun

Introduction In this paper, we investigate the semantic and pragmatic properties of theSkwxwú7mesh evidential lhkwun. Evidentials are controversial, as they have been argued tocontribute to the truth conditions of the sentence (Mahewson et al 2007) or as notcontributing to the truth conditions of the sentence but instead being illocutionary operators(Faller 2002). We provide evidence that the Skwxwú7mesh evidential lhkwun contributes to thetruth conditions of the sentence, but is also sensitive to the discourse it is used in. We focus onone of the evidentials: lhkwun.

Simple assertions (1) may be modified by clitics that provide evidential information (1b).lhkwun indicates that “the speaker considers something possible or probable” (Kuipers1967:326).(1) a. Na7 ḵw’el ta sḵw’elám ̓. b. Na7 lhkwun ḵw’el ta sḵw’elám ̓.

rl ripe det berry rl evid ripe det berry‘e berries are ripe.’ ‘e berries are ripe.’

Semantics Like many other evidentials, lhkwun behaves like an epistemic modal (cf.Mahewson et al 2007). at is, they are sensitive to the type of evidence that the speaker hasfor making their statement. lhkwun may be used when the speaker has visual (2a), aural (2b) orolfactory (2c) evidence.(2) a. m ̓i lhkwun yiḵ. b. Na7 lhkwun wa nexwtí7 ta chíyatmixw.

come evid snow rl evid impf around det owl‘It's snowing.’ ‘e owl’s around.’Context: I can see snow on your shoes. Context: I can hear the owl.

c. Chexw lhkwun kwukw ti smeyts.2sg.sub evid cook det meatare you cooking the meat?Context: I can smell it

lhkwun is also like many other evidentials in that it can be translated as either havingexistential (3a) or universal (3b) force (cf. Mahewson et al 2007). (3) a. Chen lhkwun tel'-n-umut. b. Nilh lhkwun ta s7uxwen

1sg.sub evid find.out-tr-lcrefl focus evid det ice‘I guess I'll find out.’ (∃) ‘It must have been the ice.’ (∀)

Discourse In Skwxwú7mesh, not all discourse contexts are appropriate for evidentials. Forexample, the use of an evidential in (5) is deemed somewhat inappropriate given the closefamily relationship of the two.(5) a. # Chexw lhkwun kwukw ta smeyts. b. ✔Chexw u wa kwukw ta smeyts.

2sg.sub evid cook det meat 2s.sub pol impf cook det meat ‘You (must be) cooking meat.’ ‘Are you cooking meat?’

Context: One sister walks into the other sister’s house and smells something cooking.In this context, the speaker considers (5a) to be too indirect given the close familial ties she haswith her sister. She instead considers (5b), a direct yes/no question, the appropriate way toquestion her sister in this context.Implications Both semantic and pragmatic considerations are relevant to lhkwun. lhkwunprovides the sentence with epistemic modality (of variable force); however, the use of lhkwun isrestricted to only those discourse contexts that are deemed appropriate, such as between twonon-related speakers. Any analysis of lhkwun must take both into consideration.

ReferencesFaller, Martina. 2002. Semantics and Pragmatics of Evidentials in Cuzco echua. PhD thesis,

Stanford UniversityKuipers, Aert H. 1967. e Squamish Language: Grammar, Texts, Dictionary. e Hague: Mouton.Mahewson, Lisa, Hotze Rullmann and Henry Davis 2007. Evidentials as Epistemic Modals:

Evidence from St’at’imcets. e Linguistic Variation Yearbook 7.

Nasal Assimilation and Nasal Docking in Chiquitano: a preliminary analysis Girard, Raphael (University of British Colombia)

Nasal consonant harmony (NCH), whereby a voiced consonant becomes nasal in the environment of a non-adjacent nasal consonant, is a rare phenomenon, attested in some Bantu languages and a handful of other languages (Ao 1991, Odden 1994, Hyman 1995, Piggott 1996. Hansson 2001, Rose & Walker 2004). Chiquitano (isolate: Bolivia/Brazil) may constitute a new, independent, case of NCH, as the alternations (β~m, r~n) in 1-2 suggest. 1) β-akiba-ra-ti 2) m-ata'a:neka:-na-ti

3sing-hunt-TRANS-3sing.masc 3sing-hide-TRANS-3sing.masc “he hunts” “he is hiding” * m-akiba-na-ti * β-ata'a:neka:-ra-ti

However, it seems that at least some of the vowels that intervene between the trigger and the target consonants can surface as nasal, as shown in 3. 3) m-ata'ã:nekã:-na-ti ≈ (2)

3sing-sweep-TRANS-3sing.masc “he sweeps”

This casts some doubt on whether Chiquitano constitutes a genuine case of NCH, or a case of the more familiar nasalization harmony, where all vowels are nasalized. But it also raises the more important issue of “discontinuity”, and of what accounts for the nasalization of only some of the vowels of a domain. I argue that along with NCH, which targets only consonants, Chiquitano has two independent nasalization processes that target vowels: nasal assimilation (spreading) and nasal docking (from a deleted syllable). Nasal assimilation optionally applies to any long vowel that is adjacent to a nasal consonant, as can be seen in 2-3

Nasal docking obligatorily occurs to any (compensatorily lengthened) vowel, when a following CNV syllable is deleted, as shown in 4-6 (word-finally) and 7 (word-internally)

I show that these processes are independent from NCH and thus, that the nasalization of some of the vowels between the trigger and target consonants is to be accounted for separately. Therefore, I argue that Chiquitano constitute a genuine case of long-distance

NCH of the Bantu type; with an additional twist. Bibliography Ao, B. 1991. Kikongo nasal harmony and context-sensitive underspecification. Linguistic

Inquiry 22:193-196. Galeote Tormo, J. 1996. Manityana auqui besiro. Gramatica moderna de la Lengua

Chiquitana y vocabulario basico. Santa Cruz de la Sierra, Bolivia: Centro de estudios Chiquitanos, 2nd edition

Hansson, G. 2001. Theoretical and typological issues in consonant harmony. Berkeley: University of California, Berkeley dissertation.

Hyman, L.M. 1995. Nasal consonant harmony at a distance: the case of Yaka. Studies in African Linguistics 24:5-30.

Piggott, G.L. 1996. Implications of consonant nasalization for a theory of harmony. Canadian Journal of Linguistics 41.141 – 74.

Rose, S. & Walker, R. 2004 A typology of consonant agreement as correspondence. Language, 80:3 p. 475-531

Sans, P. 2010. Elements de sociolinguistique et de phonologie du besiro (chiquitano). Langue en danger des basses terres de Bolivie. Memoire de master 2, Universite Lumiere Lyon 2, Lyon, France.

Competing for the Standard Dialect: The Case of Modern Israeli Hebrew Elaine Gold (University of Toronto)

Contemporary mythology often describes standard modern Israeli Hebrew as

springing, Athena-like out of the head of one man, Ben Yehuda. This miraculous birth (or rebirth) is, in this view, unique, not to be framed within the patterns of revitalization and standardization found in societies around the world. However, more nuanced histories (eg. Harshav, Bar-Adon, Segal) recognize the variety of social, political, religious and historical factors that contributed to the development of this vernacular, and the large numbers of people involved. In this paper I focus on three dialects that were available in the early twentieth century as potential standards for the emerging Israeli Hebrew vernacular: the Jaffa or Judean dialect on the Mediterranean coast; David Yellin’s proposed composite dialect, which aimed for a one-to-one correspondence between written symbol and sound; and the Galilean accent centred around Rosh Pinnah in the north (Segal, Bar-Adon) . The rivalry between these dialects was highlighted in discussions of the Teacher’s Assemblies of 1903 and 1904 (Segal, Glinnert, Bar-Adon). I propose that the competition between these models is best understood using the three goals of language planning described by Fishman 1971: unification, authentification and modernization. I argue that the eventual dominance of the Jaffa dialect can in retrospect be seen as inevitable, both because that dialect was best at balancing the tension between these three factors and also in the light of the choice of language standards in other parts of the world. This discussion shows that the development of the standard of modern Israeli Hebrew was not a unique occurrence but has much in common with language planning processes that have occurred around the world with languages as disparate as Maori, Finnish, Shona, Turkish and Tok Pisn. References Bar-Adon, Aaron. 1975. The Rise and Decline of a dialect: A Study in the Revival of

Modern Hebrew. The Hague: Mouton. Blau, Joshua. 1981. The Renaissance of Modern Hebrew and Modern Standard Arabic.

Berkeley: University of California Press. Fellman, Jack. 1973. The Revival of a Classical Tongue: Eliezer Ben Yehuda and the

Modern Hebrew Language. The Hague: Mouton. Fishman, Joshua A. 1971. ‘The Impact of Nationalism on Language Planning’. In J.

Rubin & B. Jernudd, eds. Can Language be Planned: Sociolinguistic theory and practice for developing nations. Honolulu: University Press of Hawaii. pp. 39-51.

Glinert, Lewis. 1993 The First Congress for Hebrew, or When is a congress not a congress? In Joshua A Fishman, ed. . The Earliest Stage of Language Planning: The “First Congress” Phenomenon. Berlin: Mouton de Gruyter. pp. 85-116.

Harshav, Benjamin. 1993. Language in Time of Revolution. Berkeley: University of California Press.

Joseph, John Earl. 1987. Eloquence and Power: The Rise of Language Standards and Standard Languages. London: Frances Pinter.

Segal, Miryam. 2010. A New Sound In Hebrew Poetry: Poetics, Politics, Accent. Bloomington: Indiana University Press.

Spolsky, Bernard. 1996. Conditions for Language Revitalization: A Comparison of the

Cases of Hebrew and Maori. In Sue Wright, ed. Language and the State: Revitalization and Revival in Israel and Eire. Toronto: Multilingual Matters.

Introducing Transitions in Kwak’wala States Hannah Greene (University of British Colombia)

This research investigates lexical aspect in Kwak’wala, a North Wakashan

language spoken on the northern part of Vancouver Island and the adjacent mainland. Previous descriptions of the aspectual system are limited to work done by George Hunt and Franz Boas in the early part of the 20th century (Boas 1911, 1947). In this paper, I examine the function of the most widely-occurring aspectual morpheme in Kwak’wala, -x7id ‘change of state’, and its distribution within states. I propose that -x7id is a type of perfective that introduces transitions in the state verb class. I present evidence from individual and stage-level states to support the claim that -x7id introduces a transition, rather than operating on a lexically specified transition.

Stage-level states refer to transitory properties of various duration (Kratzer 1995). English examples include be angry, be married, or be sleepy. These states may have arbitrary transitions that need not be specified in discourse. In Kwak’wala, the initial transition into a state can be predictably specified when -x7id is added (1). Stage-level states unmarked by -x7id provide no information about the initial transition into, or the final transition out of that state (2).

(1) lhawis-7id-an abamp (2) lhawis-an abamp

angry-change.state-1poss mother angry-1poss mother ‘my mother got mad’ [just as I walked in the room] ‘my mother is/was mad’

Stage-level states do not provide sufficient evidence about the lexical nature of the transition within the verb class. In order to discover if -x7id introduces an initial transition, or operates upon an arbitrary, lexically present transition, I present evidence from Kwak’wala individual-level states.

I assume individual-level states are characteristic states without any transitional specification in their lexical representations. Examples include be tall, have brown hair, be good-natured. Such steady states do not have an arbitrary transition. If -x7id is an operator that requires a lexically present transition, I predict -x7id will be ungrammatical with individual-level states. On the other hand, if -x7id introduces a transition into a state, I predict that it will coerce an individual-level state into a stage-level state. The latter prediction is borne out in examples (3-4). Sentence (3) describes a person who has been tall her entire life. The change-of-state morpheme cannot be added to (3), and still refer to a person who has always been tall. In (4), a context was given in which a short actor steps upon a platform to appear taller than his female counterpart. In this case, -x7id introduces an initial transition to an individual-level state, in effect, coercing it into a stage-level state.

(3) galt’axsd-i Helena (4) galt’axsda-x7id-ox-da bagwanam tall-vis Helena tall-change.state-loc-det man I conclude that -x7id introduces initial transitions in states. This conclusion is a first step towards a greater understanding of the lexical aspect of verb classes in Kwak’wala.

References Boas, F. (1911). Kwakiutl. Bureau of American Ethnology, 40:423--557. Boas, F. (1947). Kwakiutl grammar with a glossary of the suffixes. Transactions of the

American Philosophical Society, 37:202--377. Republished by AMS Press 1976. Kratzer, A. (1995). “Stage-level and Individual-level Predicates.” In, The Generic Book.

Eds, Gregory N. Carlson and Francis Jeffry Pelletier. Chicago: University of Chicago Press.

Depictives and unergatives: a puzzle Monica-Alexandrina Irimia (University of Toronto)

This paper focuses on a previously unnoticed aspect regarding the syntax of depictives (DEPs). In English, and many other languages, DEPs are preferred with internal objects, while the external arguments of unergatives are generally inaccessible (as seen in 1). Three possible hypotheses are investigated in order to solve this puzzling restriction. One possibility would be to assume that their non-interaction with unergatives is due to the lexical-semantic specification of the latter. Preteoretically, what is described as the speaker’s involvement (Levin and Rappaport Hovav 1995) could be invoked; DEPs (similarly to other secondary predicates) require presuppositional readings on the argument they take, leading (among others) to specificity effects (see Williams 1983); unergative roots, on the other hand, might not necessarily permit this semantic composition, as they are seen to rather collapse features related to involuntariness. The major problem with such an assumption is that it cannot easily explain cross-linguistic variation. DEPs with unergatives, like in 1B, are possible across Romance, or Chinese; it is not clear how the semantic entry of unergatives can be parameterized in order to predict the split. A second possibility is to try to connect these facts to other domains in which unergatives appear to introduce combinatorial prohibitions. The same picture is found elsewhere in the domain of predication; for example, in some languages resultative secondary predicates (as in John pounded the metal flat) have been shown to be subject to Simpson’s law– a generalization that permits them to be only predicated of internal objects (Simpson 1983). It is also known that (lexical) causatives follow approximately the same pattern (see the example in 1c, and Halle and Keyser 1993, 1995, 2002, Doron 2000, Pylkkänen 2008, etc.). These facts preliminarily indicate that DEPs should not be treated as a completely separate class from resultatives (that is, the classical divide adjunct = depictives which do not affect/are insensitive to the lexical structure of the main verb vs. non-adjunct = resultatives) is not entirely on the right track.

(1) ENGLISH DEPICTIVES A. Depictives are allowed with internal objects (of unaccusatives): John arrived drunk/happy/tired; John ate the fish raw. B. Depictives cannot combine with unergative verbs *John smiled tired (happy) / *John laughed tired/happy// *John sneezed tired/sick. C. Lexical causatives cannot be constructed from unergatives: * I smiled him (for I made him smile).

Cross-linguistic investigation shows that DEPs should not be collapsed with causatives – there are languages which allow causatives from unergative roots, but do not allow depictives with unergatives (Finnish), or languages which do exactly the opposite (as in Romance). A third option, and the one argued for in this paper, is a syntactic account. The main claim is that secondary predicates (SPs) are similar to restructuring complexes, in that they involve a bare adjectival head (similarly to Wurmbrand’s 2001 restructuring non-finite predicates), which does not project a specifier layer. Argumental secondary predication is parasitic on main predication, and explained via a process of Multiple Agree (following Hiraiwa 2001, 2004, Chomsky 2008); the process is initiated by a functional restructuring head which flags these types of syntactic complex predicates

(altering ideas from Cinque 2005 regarding the nature of restructuring). It is further proposed that the hierarchy above results from a parametrization in the external argument position, combined with the mechanics of Agree. SPs provide strong evidence that subjects can be introduced both within the complex predicate domain, lower than the restructuring functional projection, or higher, above this projection. In order to initiate Multiple Agree (whose feature checking result is “predication”), the restructuring head has to send a probe upwards (Béjar and Rezac 2009) for those arguments introduced outside its domain. But this process is more marked than the downward phi-feature checking procedure, hence cross-linguistic degrees of variation.

Axial Parts in Inuktitut and Uzbeki Alana Johns & Brigid Thurgood (University of Toronto)

This paper presents new data from dialects of Inuktitut and the Afghanistan dialect of Uzbek, contributing to recent advances in the understanding of the typology of expressions encoding direction and location. It relates spatial expressions in these languages to extended PP structure in Generative Grammar, concluding that Genitive and agreement markings on spatial nominals demonstrate the underlying structure of the DP. Recent work has investigated the architecture of PPs, articulating it along the lines of CP, DP, and vP. One proposed projection (Svenonius 2006) is Axial Part, a head encoding a vector space (Zwarts 1997) that projects outward from the central axis of a Ground DP to a Figure DP. Elements which appear to be nominal are argued to have non-nominal properties and merged as AxPart. We argue that the data in (1) and (2) from Inuktitut and Uzbeki contain Axial Parts in Svenonius’ terms. (1) Inuktitut a. saniani ‘beside’ e. akungngani ‘among’ b. Kângani ‘on top of’ f. iluani ‘inside’ c. tunuani ‘behind’ g. isuani ‘the end of’ d. atâni ‘underneath’ h. kanani ‘down’ (2) Uzbeki a. ust ‘top, above’ g. toy ‘under’ m. nar ‘after, beyond’ b. ashog ‘below’ h. yon ‘beside’ n. jaqo’n ‘near, close’ c. ich ‘in, inside’ i. dash ‘out’ o. atraf ‘around’ d. qorsh ‘front’ j. keyn ‘back’ p. chevra ‘around’ e. old ‘front’ k. arqa ‘back’ q. pas ‘down’ f. dam ‘front’ l. uzog’ ‘distant, far’ r. bilant ‘up’ According to Svenonius (2006), a recurrent pattern seen with Axial Parts is Genitive marking on the Ground DP, which we find in both Uzbeki and Inuktitut. In these languages, the relationship between Ground and Axial Part appears to be identical to the relation between Possessor and Possessee: the Ground DP is marked for Genitive case, and is the source of agreement found on the AxPart, as shown in (3) and (4). (3) Inuktitut: a. Kimmi-p kiguti-nga b. nukkapiak- sâ-p sani-a-ni dog-GEN tooth-3s boy-ABS table-GEN side-3s-LOC ‘the dog’s tooth’ ‘The boy is under the table’ (4) Uzbeki: a. yog’och-in kuk-i b. tup uy-ing arqa-sin-da tree-GEN root-3s ball house-GEN back-3s-DAT ‘the tree’s roots’ ‘The ball is behind the house’ In Svenonius’ analyses, the AxPart reading is only available in oblique constructions. In Uzbeki, both nominal and AxPart readings are usually available in locative contexts; in Inuktitut contexts with structural case, only the literal meaning of the noun is available.

(5) Inuktitut atâ-nga nigua-sima-nngi-tuk bottom-3s paint-PERF-NEG-PTPL.3s ‘The bottom is not painted’ We will argue that the possessive relation is not trivial in these constructions and that the presence of a possessor is a critical component of the AxPart reading. References Svenonius, Peter. 2006. The emergence of Axial Parts. In P. Svenonius and M.

Pantcheva, eds., Adpositions. Special issue of Nordlyd: Tromsø Working Papers in Linguistics, 33.1, 49-77.

Zwarts, Joost. 1997. Vectors as relative positions: A compositional semantics of modified PPs. Journal of Semantics 14: 57-86.

Control, partial control, and tough constructions Michiya Kawai (Huron University College/University of Western Ontario)

The movement analysis of Control (Boeckx, Hornstein and Nunes 2010 and references cited therein) offers a minimalist account of control phenomena. For example, in (1a), Mary “moves” from the embedded subject to the matrix object, as in (1b). Partial control (PC) phenomenon, as in (2a), is an issue that is not adequately accounted for by this analysis, however. (1) a John persuaded Mary to leave. b. [persuaded Mary [ Mary to leave ] ] (2) a. John wants to meet at 6. b. John1 wants [PRO1+ to meet at 6 ]. c. * John met at 6. The interpretation of (2a) is (2b), given that meet requires a plural subject (2c). The movement analysis seems incapable of expressing (2b). Barrie and Pittman (B&P) (2004) propose an analysis that accounts for PC within the movement analysis of control. This paper examines B&P’s proposal in detail, concluding that their analysis is untenable. The result of the present study seems to favor the non-movement analysis of control, where interpretation rules apply at the interface.

According to B&P (2004), movement of a DP creates a chain that respects the standard theta-Criterion. Thus, a chain may not contain multiple theta-positions. A DP, moving from one theta-position to another as in (1), creates an ill-formed chain. Such a chain is “split” into two at the interface, so as to satisfy the theta-Criterion. The lower copy of the DP retains its identity with its antecedent, but pragmatic factors allow the lower copy to be interpreted as semantically plural.

B&P use the absence of the PC reading in (3a) as a confirmation of their analysis. They assume that the matrix subject of a tough-sentence is non-thematic; thus, tough-movement, from the embedded object to the matrix subject, creates a chain with only one theta-position; hence, the chain is not split, and no PC reading is available. This argument fails, however, because PC reading is not present in a degree-construction (4) and a “pretty”-construction (5), both of which arguably have a thematic matrix subject. Further, PC reading is absent in the control relation between the experiencer PP and the embedded subject in a degree construction (6). This is not predicted by B&P’s analysis.

(3) a. * This statue is easy to place in a circle. (B&P 2004: (52)) b. These statues are easy to place in circle. (4) a. * The sonata of Beethoven is too long to play together. b. The sonatas of Beethoven are too long to play together. (5) a. * This eel is slippery to hold together. b. These eels are slippery to hold together. (6) a. * Magic Flute is too long for Mary to watch together. b. Magic Flute is too long for John and Mary to watch together.

References Barrie, M., and C. M. Pittman. 2004. Partial control and the movement toward

Movement. Toronto Working Papers in Linguistics 22: 75–92. Boechx, C., N. Hornstein, and J. Nunes. 2010. Icelandic control really is A-movement:

Reply to Bobaljik and Landau. Linguistic Inquiry 41: 111–130.

Back in North American French: A Case Study of Reanalysis in Grammatical Change

Ruth King (York University)

French varieties spoken in contact with English typically display some degree of integration of the English intransitive preposition back. Through a combination of traditional dialectology, comparative sociolinguistics and recent syntactic theory, I trace its use in Acadian, Laurentian and Louisiana French from its earliest (1890) attestation through to its use in recently‐constructed sociolinguistic corpora.

In (1), back is used with the same locative meaning and in the same position as it is in English. Following Mougeon et al’s (1980) analysis of such data, I link the rise of back to the gradual loss in productivity of the re‐suffix (e.g. revenir “to come back”, refaire “to do again”) in the history of French. (1) Il s’en vient back où ce‐qu’était la vielle. (Nova Scotia; Aucoin 1953)

“He comes back to where the old woman was.” In situations of intense language contact, back takes on an iterative meaning, as shown in (2): (2) J’ai commencé à refumer back. (Lousiana; Rottet 2000)

“I started smoking again.” In some varieties, iterative back has been reanalyzed as an adverb, occupying the same syntactic as aspectual adverbs like souvent “often”, illustrated in (3) and (4): (3) Il m’a back frappé. (Southeast New Brunswick; Perrot 1995)

“He hit me again.” (4) Veux‐tu back faire ça? (Prince Edward Island; King 2000)

“Do you want to do it again?” Syntactic reanalysis may also extend to locative back, as shown in (5), even though French‐origin locative adverbs do not precede the past participle or the infinitive. (5) Il a back amené la tape. (Nova Scotia; Comeau 2007)

“He brought back the tape.” On the basis of both interdialectal comparisons and intradialectal real and

apparent time quantitative analyses, I identify five stages in the integration of back, involving variable presence/absence of the re‐suffix, meaning extension and syntactic reanalysis. References Aucoin, Gérald‐E. 1980/1953. L’oiseau de la vérité et autres contes de pêcheurs de l’Île

du Cap‐Breton, presentés par G.E. Aucoin. Montréal: Quinze.

Comeau, Philip. 2007. The integration of words of English origin in Baie Sainte-Marie Acadian French. Unpublished M.A. Major Research Paper, York University.

King, Ruth. 2000. The lexical basis of grammatical borrowing. Amsterdam & Philadelphia: John Benjamins.

Martineau, France. 1995‐2007. Corpus de français familier ancien. University of Ottawa.

Mougeon, Raymond, Cora Brent‐Palmer, Monique Bélanger & Walter Cicocki. 1980. Le français parlé en situation minoritaire. Vol. 1 : Emploi et maîtrise du français parlé par les élèves des écoles de langue française dans les communautés franco- ontariennes minoritaires. Toronto: Ontario Ministry of Education.

Perrot, Marie‐Eve. 1995. Aspects fondamentaux du métissage français/anglais dans le chiac du Moncton (Nouveau‐Brunswick, Canada). Unpublished doctoral dissertation, Université de la Sorbonne Nouvelle ‐Paris III.

Rottet, Kevin. 2000. The calquing of phrasal verbs in language contact. In Julie Auger & Andrea Word‐Allbritton, eds., The CVC of Sociolinguistics: Contact Variation, and Culture. Working Papers in Linguistics Vol. 2. Bloomington, Indiana: IULC, 109‐126.

Tepehua and Totonac consonants: contrastive hierarchy in action Loredana Andreea Kosa (University of Toronto)

The contrastive hypothesis maintains that the only features active in a language

are those features necessary to contrast between segments in the inventory. In other words, if a feature is not necessary to differentiate between two segments, that feature is not present in the feature inventory of the language. The contrastive hypothesis can thus account for how the same segment can pattern and behave differently from language to language. It is simply a result of different features active in different languages, and as a result different feature specifications for the same segment in different languages.

Accepting the contrastive hypothesis, it becomes crucial to establish how we determine contrast and how the resulting contrastive features are organized. The successive division algorithm by Elan B. Dresher (2008,2009) with the resulting contrastive hierarchy of features is a possible method of determining contrastive features. My paper attempts to show the benefits of this approach through the comparison of four languages spoken in Mexico, which, despite having very similar segment inventories, are shown to vary considerably in how the phonology groups the same segments in each of the languages. The current work focuses on four languages from the Totonacan language family: Tlachichilco and Huehuetla Tepehua, Misantla and Filomeno Mata Totonac. The processes involving obstruents in these languages point to key differences that are important in thinking about the contrastive hierarchy, pointing to similar inventories having different active features as well as different rankings of the importance of those features. The major areas of distinction revolve around affricates, the distinction between anterior and posterior obstruents, and the status of the lateral fricative /l/. Affricates in the four languages differ in how they pattern with fricatives and stops: in Tlachichilco Tepehua affricates pattern with fricatives, but never with stops, while in the other languages affricates pattern mainly with stops. Amongst the remaining three languages, Huehuetla Tepehua is different as affricates pattern only with stops, never with fricatives, unlike the Totonac languages in which affricates mainly pattern with stops, but also with fricatives in sibilant harmony. Affricates have, in the past, been considered segments with a dual specification for the [continuant] feature, either ordered, as in Sagey (1986), or unordered, as in Lombardi (1990). The contrastive hierarchy approach, instead, can separate affricates that pattern with stops from those that pattern with fricatives by placing the [continuant] feature above or below the [strident] feature. Thus, [continuant]>[strident] results in affricates that pattern with stops as [-continuant], while [strident]>[continuant] results in affricates that pattern with fricatives as [+strident]. At the same time, this system can accommodate cases such as Misantla Totonac where affricates pattern mostly as stops, but pattern with fricatives in sibilant harmony. This is handled by the presence of [+strident] specification for both affricates and the sibilant fricatives (which contrast with [-strident] non-sibilant fricatives). Even more unusual cases, such as that of affricates in Filomeno Mata Totonac which pattern as a group with the stop /t/, can be easily explained with the contrastive hierarchy. In this language, the place features are highly ranked within the [-continuant] group, resulting in /t/ grouping with all the affricates, separate from the labial stop /p/ and the dorsal stops /k,q/.

The contrastive hierarchies of the four languages also differ in the importance of the [anterior] feature, which is highly ranked and contrastive for all obstruents (even stops) in Tlachichilco Tepehua, but not in the other languages. Tlachichilco Tepehua is also unique in that the lateral fricative patterns as an approximant (or liquid) in this language, while in the other languages it patterns with the fricatives as an obstruent. Works cited Dresher, E. B. (2008). The Contrastive Hierarchy in Phonology. In P. Avery, E. B.

Dresher & K. D. Rice (Eds.), Contrast in Phonology (pp. 11-33). Berlin; New York: Mouton de Gruyter.

Dresher, E. B. (2009). The Contrastive Hierarchy in Phonology. Cambridge: Cambridge University Press.

Lombardi, L. (1990). The nonlinear organization of the affricate. Natural Language & Linguistic Theory, 8(3), 375-425.

Sagey, E. (1986). The Representation of Features and Relations in Nonlinear Phonology. phD, MIT, Cambridge, Massachusetts.

On mass nouns in Romance: Semantic markedness and structural underspecification

Ivona Kucerova & Anna Moro (McMaster University)

Even though the gender system in Romance is mostly binary some Italo-Romance and Ibero-Romance dialects exhibit properties of a three-way gender system. Remarkably, this happens only with a subset of nouns, namely, mass nouns (MN) and a certain productive class of deverbal nouns. We argue that to understand the properties of MN we must analyze them as NPs, i.e., as lacking a D layer encoding number and gender. If D is semantically or syntactically required, a structure extension takes place and triggers morphophonological alternations. The data bear on a more general question of the nature of nominal structures, their inherent morphological features and the relationship between a morphological structure and its semantic interpretation. The puzzle The surface realization of the MN/count distinction in many non-standard Romance varieties shows a wide range of variation displaying a special morphological realization of MN as a “third” gender category distinct from the masculine(m)/ feminine(f) opposition found in Standard Romance languages. Crucially, the MN category does not uniformly agree with the surrounding syntactic structure. While some agreement relations exhibit a default morphological agreement (typically m), others exhibit a marked mophological realization (either MN, or f; in some dialects realized as raddoppiamento sintattico). For example, in the dialects of the Rome-Ancona corridor, the affixes for MN are distinct from m.sg. Furthermore, MN have distinct determiners (Servigliano: kwístu MN, kwésto M.SG., kwésta F.SG. ‘this’). Crucially, one MN can trigger two distinct agreement patterns within the same structure, as in (1) (L’Aquila). (1) So kumbratəә ləә vinəә. Ləә so kumbratəә pəәrkɛ ɛ bbonəә. aux bought the.MN wine. It.MN aux bought because it’s good.M.SG The referential pronoun in (1) realizes the MN agreement while the predicative adjective exhibits a m.sg. agreement. This is rather puzzling because (i) the agreement pattern is not homogenous, and (ii) while in most cases discussed in the literature a feature mismatch in agreement triggers underspecification (.e.g., impoverishment), here the mismatch triggers either a marked value or a third value within a binary system. Even though this pattern has been extensively studied from the diachronic point of view (Maiden 1997, a.o.), no synchronic account is available. Here we investigate one aspect of the observed variation, namely, gender agreement properties.

The proposal We argue for a new empirical generalization: If MN may be predicative, it triggers a default vocabulary insertion. If MN must be referential, it triggers a “marked” vocabulary insertion. The generalization can be accounted for if we assume that MN are structurally deficient in the sense that they lack the D layer: MN is an NP and as such has no number projection (Borer2005, a.o.). Since gender is dependent on number, MN are necessarily genderless. Semantically, we follow Winter (2000) in that DPs are rigidly quantificational, while NPs are rigidly predicative. Type-shifting is freely available via choice function but crucially only if D’ is present in the structure. Since a structure

without D cannot get a referential interpretation (type e), a MN can function as type e only if some additional structure building process takes place. We argue that the observed agreement pattern reflects the structural and semantic duality of MN: if MN can be inter-preted as e, t, there is no gender and number feature present, thus no Agree valuation takes place and consequently the overt agreement is realized as the morphological default (m.sg., on predicative adjectives). In contrast, if the structure requires type e, e.g., a referential pronoun, additional structure must be built: the marked morphological realization is a direct reflex of this extra semantic process implemented as a structural adjustment. Finally, our hypothesis predicts that quantifiers as possible NP modifiers should pattern as predicative adjectives rather than demonstratives or definite articles. This prediction is indeed borne out: pokəә vinəә ‘little.M.SG. wine.MN’ (L’Aquila). References Borer, H. 2005. In name only. Oxford: Oxford University Press. Maiden, M. and M. Parry, eds. 1997. The dialects of Italy. NY: Routledge. Winter, Y. 2000. DP structure and flexible semantics. In M. Hirotani, A. Coetzee, N.

Hall, and J-Y. Kim, editors, Proceedings of the 30th Conference of the North Eastern Linguistic Society, University of Massachusetts at Amherst, 2000.

Quelques réflexions sur les paradigmes défectifs en russe Elena Kulinich (Université de Montréal)

Le but de cette communication est d’explorer la possibilité d’étendre l’analyse des verbes défectifs russes basée sur la notion de la fragmentation d’information sur la production (Albright 2006, à paraître) aux autres cas de la défectivité dans cette langue : soit dans les paradigmes nominaux (absence de génitif du pluriel de certains noms féminins, par exemple : mechta ‘rêve’ (N.sg.) -*mecht (Gén. pl.)) ou dans les paradigmes verbaux (absence de l’impératifs de certains verbes, par exemple : protivostojatj ‘résister’(inf.) -*protivostoi (impérat.)). Une autre question à examiner en rapport avec celle-ci est si la fragmentation d’information mène nécessairement à la défectivité.

À partir du moment où Halle (1973) a mentionné le phénomène des verbes défectifs en russe (verbes de 2e

conjugaison en une consonne coronale qui n’ont pas de

formes de la 1re personne du singulier au non-passé) en tant que problème linguistique,

ces verbes ont été analysés dans les travaux sur ce sujet comme étant accidentels («fortuites» dans la terminologie de Mel uk 1993) et lexicalisés (Baerman 2008, Dalland 2007). Cependant, notre étude de nouveaux verbes informatiques en russe et en ukrainien montre qu’ils ne sont ni accidentels, ni lexicalisés et que la présence de trous dans les paradigmes de ces verbes est lié à la fragmentation d’information sur la production causée par une alternance morphophonologique, i.e. la palatalisation d’une consonne finale du radical (d~ž, t~ , s~š, z~ž).

Une comparaison des résultats de notre enquête sur des verbes néologiques en russe et en ukrainien montre que la répartition des formes avec alternance (où la palatalisation s’applique) et sans alternance (où elle ne s’applique pas) est différente dans ces deux langues. On observe que des locuteurs du russe, ainsi que ceux de l’ukrainien, optent souvent pour des paradigmes qui leur permettent d’éviter de faire un choix entre ces deux possibilités (5.8% de réponses en russe et 30.6 % en ukrainien). Cependant, entre les formes avec ou sans alternance les locuteurs ukrainiens choisissent plus souvent la forme où la palatalisation s’applique (51% vs. 18.4%), tandis qu’en russe le partage des formes avec et sans palatalisation est presque égal (48.3% vs. 45.9%).

L’écart dans les résultats obtenus pour les deux langues en présence de conditions similaires pourrait, à notre avis, être expliqué par l’uniformité de l’alternance en ukrainien (/d ~ dž ~ dž / : narodyty ‘donner naissance’(inf.) -narodžu (1sg.) -narodžennyj (PP)) et par une fragmentation additionnelle de l’information sur l’alternance en russe causée par des emprunts du vieux-slave (/d ~ ž ~ ž/ ou /d ~ ž ~ žd/: posadit’ ‘planter’ (inf.) - posažu (1sg.) – posažennyj (PP) ou rodit’ (inf.) -rožu (1sg.) -roždennyj (PP)). Cette situation nous permet de faire la prédiction qu’il est peu probable que les verbes néologiques une fois adoptés en ukrainien seraient défectifs, tandis qu’en russe la répartition presque égale des formes avec et sans alternance nous indique clairement que des locuteurs affrontent une situation de choix sans avoir suffisamment d’information pour choisir une des deux possibilité ce qui mènerait à la défectivité de nouveaux verbes informatiques dans cette langue.

Cette explication rapproche notre travail des études sur la défectivité dans d’autres langues comme le français (Baronian 2009) ou l’espagnol (Albright 2003) dont l’analyse présente la défectivité comme phénomène liée au choix entre deux ou plusieurs possibilités morphologiques.

Références Albright, Adam. (2003). A quantitative study of Spanish paradigm gaps. In: G. Garding

and M. Tsujimura, eds., WCCFL 22 Proceedings. Somerville, MA: Cascadilla Press, pp. 1-14.

Albright, Adam. (2006, à paraître). Lexical and morphological conditioning of paradigm gaps. In Curt Rice (ed.), When nothing wins : Modeling ungrammaticality in OT. Equinox Publishing.

Baerman, Matthew. (2008). Historical observations on defectiveness: the first singular nonpast. Russian Linguistics 32/1. 81-97.

Baronian, Luc. (2009). Une analyse de verbes défectifs sans spécification lexicale. In: L. Baronian and F. Martineau, eds., Le français d’un continent à l’autre. Mélanges offerts à Yves Charles Morin. Québec: Presses de l’Université Laval, pp. 29-48.

Daland, Robert, Andrea D. Sims and Janet Pierrehumbert. (2007). Much ado about nothing: A social network model of Russian paradigmatic gaps. Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics in Prague, Czech Republic, June 24th-29th, 2007. Annie Zaenen and Antal van den Bosch, eds. Prague: Association for Computational Linguistics, 936-943.

Halle, Morris. (1973). Prolegomena to a theory of word formation. Linguistic Inquiry. vol. 4, no 1, p. 3-16.

Mel’čuk, Igor. (1993). Cours de morphologie générale, vol.1. Introduction et première partie : le mot. Montréal : Presses de l’Université de Montréal/Paris : CNRS.

The syntactic partition of the semantics of definiteness Maria Kyriakaki (University of Toronto)

In Greek, the definite DPs may contain multiple instances of the definite article and an adjective. These DPs are known as polydefinites and look as follows:

(1) To molivi to ksilino The.neu pencil.neu the.neu wooden.neu

‘The pencil the wooden one’

My purpose is to determine what makes polydefinites possible and examine whether polydefiniteness is a cross-linguistic phenomenon. I propose that the source of polydefiniteness is located in the partition of definiteness into two components, familiarity and uniqueness. A determiner that is underspecified for uniqueness is predicted to allow polydefinites. As I show, it is in this respect that polydefiniteness can be found cross-linguistically.

Syntactically, the multiple determiners and adjectives are restrictive DPs (Kolliakou 2004, Lekakou and Szendröi, 2007), which adjoin to nP and raise higher, if focused (Kyriakaki, 2010). Lekakou and Szendröi (2008) further claim that a language with rich inflection is predicted to have polydefinites. For them, the definite determiner is only inserted to spell out case, but is otherwise a semantic expletive. However, I argue that the determiner is not an expletive. Moreover, languages may have polydefinite constructions, even if the determiner does not carry any inflection (e.g. Scottish English). This phenomenon is therefore not the spell-out of inflection.

I propose that the source for polydefiniteness lies in the semantics of the definite determiner. Definiteness involves two components, familiarity (Heim, 1982) and uniqueness (Heim & Kratzer, 1998). Familiarity, or Fam, selects a contextually salient set of entities, and the iota operator that brings uniqueness picks out the unique entity out of that set. In Greek, the definite determiner does not pick out a unique entity. Rather, it only selects a salient set of entities. It is underspecified for uniqueness, and further restriction is thus possible. In particular, I propose that in languages with polydefinites, definiteness is mapped into two syntactic projections, Fam(iliarity)P and ι(ota)P. Furthermore, polydefinites involve the intersection of two sets: one containing a set of entities, and one containing predicative expressions. Fam, selects the contextually salient set of entities out of that set. In turn, the iota operator picks out the unique familiar entity out of that singleton set.

In Standard English (SE), definite DPs with the do not allow polydefinites. The determiner the spells out both uniqueness and familiarity, and thus there is no split in definiteness. Other definite expressions though, such as proper DPs (Ghomeshi and Massam, 2009) and pronominal possessive DPs, easily allow polydefinites. Expressions like my friend the linguist (not the psychologist) and John the writer (not the painter) are perfectly valid in SE. This suggests that the null D in proper DPs and the possessive D are underspecified for uniqueness. Like in Greek then, definiteness in proper and possessive DPs must be mapped into FamP and ιP.

In conclusion, in this work polydefiniteness is argued to be a direct effect of how definiteness is encoded in a language: if a definite determiner is semantically

underspecified, then, it is predicted to allow polydefinite constructions. Hence, under the proposed analysis, an interesting cross-linguistic parallel is revealed, while a new wayof exploring nominal modifiers is now offered. References GHOMESHI, JILA & MASSAM, DIANE. 2009. ‘The proper D connection’.

In Determiners, Universals and variation, Jila Gomeshi, Ileana Paul and Martina Wiltschko (eds.), vii, 247 pp. (pp 67-96).

HEIM, IRENE. 1982. ‘The Semantics of Definite and Indefinite Noun Phrases.’ Ph.D. Dissertation, University of Massachusetts at Amherst.

KRATZER, ANGELIKA & HEIM, IRENE. (1998). ‘Semantics in generative grammar’. Oxford: Blackwell.

KOLLIAKOU, DIMITRA. 2004. ‘Monadic definites and polydefinites: their form meaning and use’. InJournal of Linguistics, 40: 263 – 333.

KYRIAKAKI, MARIA. 2010. ‘What Greek DETs do: The restrictive DP’. In Proceedings ofthe 2010 annual conference ofthe Canadian Linguistic Association, Melinda Heijl (ed).

LEKAKOU, M. & K. SZENDRÖI. 2007. ‘Eliding the noun in close apposition or Greek polydefinites revisited’. Presented at UCLWPL.

LEKAKOU, M. & K. SZENDRÖI. 2008. ‘The cross-linguistic distribution of polydefinites: case and expletive determiners’. Presented at NELS.

!"#$%&'( )*'($"$+,-!"#$%&'()'*+" ,("-*.)'"&/ )%- '0" #*&&*%1 23)453)6) .$786)

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

="AOUV!WXYZ!W[\*%- !]ZB,9[[

$%'%&&3""'0")('^

*".-]ZOUV!WXY

/%0%([)()0^

[)()0 *& #: &3""'0")('A

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

="AOUV!WXYZ!W[\1%&%23U)&)4*

/W'5& U)&)4*A/ ?,($7$&"- 6*'` /U)&)4* *&R'0)'R$%"A/F

?OF ()*4,="AO!W[JXYZ!W[\

*3ZO!W[JXY

5%!!%(N)%%)0

6"78*%3*%Z,LV!

/N)%%)0 *& '0" 3*%%"(A/ ?,($7$&"- 6*'` /N)%%)0 30$ 3$% *&R'0)'R$%"R:$%-"(A/F

Q0"% '"&'"- )1)*%&' '0" <866 ()%1" $< "K8)'*+" &"%'"%."&> 0$3"+"(> '0*& )..$8%' 6")-& '$ &:%').'*.)%- &"#)%'*. #*&7("-*.'*$%&A W 7($7$&"> *%&'")-> '0)' '0"&" 7("-*.)'"& )(" '0" (")6*a)'*$%& $< '0""K8)'*+" .$786)> 30*.0 -*<<"( *% <$(# -8" '$ )1(""#"%' 3*'0 '0"*( &8=S".' ?30*.0 *& $<'"% 908F> )%-3" 0)+" #*&')4"% '0*& )1(""#"%' <$( '0"*( -"%$')'*$%A W% '0*& )..$8%'> '0" '3$ "K8)'"- "6"#"%'&<$(# ) &#)66 .6)8&"> 7($S".'"- <($# ) ,LV! 0")-> )%- '0" $+"(' .$786) *& ) 0*10"( 7("-*.)'" '$ 30*.0'0" &8=S".' ()*&"&A

.'/'0'(1'#['"70"% LA X%-"(&$%A 23)43)6) &:%'); )%- '0" 1$+"(%#"%'R=*%-*%1 '0"$(:A /'!:%. ; /),%!:3<&>BP`GBRREI> !". BC_DA

b()%a @$)&> N"6"%" @$)& c)#7$6&4:> )%- d"66*1 [ N)((*&A 23)4*8'6 1()##)( 3*'0 ) 16$&&)(: $<'0" &8<<*;"&A =0%!&%<:38!& 8> :() ?,)03<%! @(378&89(3<%7 /8<3):'A B)$ /)03)&> OE?OF`GHORROEE>!". BCDEA

B

U)(."6 -"% !*44"%A C)7%:80& %!6 D3!2)0&E =() /'!:%. 8> @0)63<%:38!A @0)63<%:) F!G)0&38!A %!6H89"7%&A UWJ ,("&&> GHHPA

L$1"( bA N*11*%&A =() @&)"68I<7)>: H8!&:0"<:38! 3! J!#73&(A ,0! '0"&*&> U)&&).08&"''& W%&'*'8'"$< J".0%$6$1:> BCEOA

@"('()%- L8&&"66A F!:086"<:38! :8 1%:(),%:3<%7 @(378&89('A Y$%-$%` X66"% )%- e%3*%> BCBCA

G

Second language acquisition of English question intonation by Koreans Danica MacDonald (University of Calgary)

Remnants of a speaker’s first language (L1) are often present on both the

segmental and supra-segmental features of their second language (L2). Numerous studies focus on the L2 acquisition of segmental features (Flege 1995, among others) but few studies focus on the L2 acquisition of prosody (Ueyama & Jun 1998). This paper looks at how native speakers of Korean acquire English intonational patterns on wh-questions and yes/no questions. Specifically, does the L1 intonation system affect L2 intonation patterns and, if yes, to what extent? Does a higher proficiency in the L2 improve intonation?

English and Korean intonational structures differ on numerous levels. First, the f0 contour of an English Intonational phrase (IP) is determined by pitch accents which are linked to stressed syllables while in Korean it is determined by a series of Accentual phrase (AP) tones. The Korean AP phrasal tones change depending on the size of the word. In English, the type and location of the pitch accents change a sentence’s meaning, while in Korean the location of the AP phrase boundary changes the meaning (Jun 1998).

In additional to different Intonational structures, English and Korean also differ as to how they distinguish between yes/no and wh-questions. In Korean, yes/no and wh-questions are syntactically the same, the only way they differ is in their Intonational phrasing (compare 1a and 1b). In 1a, we know that this is a yes/no question since there are three AP boundaries: one following the pre-wh-phrase, one at the end of the wh-word, and one following the verb. Whquestions (1b) are represented by two AP boundaries: one following the pre-wh-phrase and the second (which includes the wh-word) following the verb. In English these two items are distinguished by lexical item as well as intonation.

The stimuli for this experiment was based on the Korean stimuli from Jun and Oh

(1996). Twenty Korean sentences (containing the wh-word what /mwəә/ and who /nuka/) were selected so that each sentence could be interpreted in 2 ways: as a wh-question and as a yes/no question. The same stimuli were also presented in English for the L2 part of the study. I will present experimental data from native speakers of Korean who are at

various stages of acquiring English. I will also compare the intonational patterns to those of native English speakers and Korean L1 speakers using Praat. My preliminary results show that two of the native Korean participants do not seem to be aware of English intonational patterns, while the third (more advanced) speaker shows native-like intonational patterns on certain English question constructions. References Flege, J. (1995). Second-language speech learning: Theory, findings, and problems. In

W. Strange (ed.) Speech perception and linguistic experience: Issues in cross-language research.

Jun, S.-A. (1998). The accentual phrase in the Korean prosodic hierarchy. Phonology, 15(2): 189-226.

Jun, S.-A. & M. Oh. (1996). A prosodic analysis of three types of wh-phrases in Korean. Language and Speech, 39(1): 37-61.

Ueyama, M. & S.-A. Jun. (1998). Focus realization in Japanese English and Korean English intonation. In Japanese and Korean Linguistics, vol. 7: 629-645. CSLA, Cambridge University Press.

Near-identity and Laryngeal Harmony Sara Mackenzie (McGill University)

This talk presents an analysis of consonant harmony in laryngeal features in

Hausa and Tzutujil. Consonant harmony processes have figured prominently in theoretical work on the significance of similarity in the phonological grammar (e.g. Hansson 2001, 2010, Rose and Walker 2004). Hausa and Tzutujil belong to a subset of consonant harmony cases in which the result of harmony is total, segmental identity. In analyzing these data, I propose that interacting segments are near-identical, defined as differing in only a single feature. This talk further argues that only contrastive features are relevant to determining interacting segments in harmony. The contrastive status of feature specifications is determined by a hierarchic ordering of binary features (Dresher, Piggott and Rice 1994, Dresher 2003, 2009). According to this approach, segmental representations, and any evaluation of similarity or near-identity based on them, are influenced by both inventory shape and the ordering of features in the contrastive hierarchy.

Both Hausa and Tzutujil have a contrast between plain and glottalized stops and both languages have a ban on multiple glottalized segments within a morpheme (1a, 2a). Identical, glottalized segments are exempt from this general restriction (1b, 2b) and there is a further restriction against homorganic plain/glottalized segments occurring within a form (1c, 2c). Following previous work (Hansson 2001, 2010, Rose and Walker 2004), I analyze this pattern as consonant harmony in the feature [constricted glottis] which is parasitic on place of articulation. Only homorganic segments interact as targets and triggers of [constricted glottis] harmony.

Segments interacting in harmony are near-identical. They differ only in specification of the harmonic feature. The following data from Hausa show that homorganic segments may differ in [cg] if they also differ in [voice] (3).

(3) k’ugu ‘pelvis’ ɗata ‘a green tomato’ The proposal that only contrastive features are relevant to determining near-identity, considered in conjunction with the role of [voice] in [cg] harmony, requires that [voice] be contrastive in glottalized segments in Hausa. In both Hausa and Tzutujil, the single glottalized series is not uniform across place of articulation. Glottalized labial and alveolar stops are implosive whereas all other glottalized obstruents are ejective. Contrastive specification of implosives for the feature [voice] is therefore not predicted in

theories of contrast that depend on minimal pairwise comparisons to determine the contrastive status of features (see Archangeli 1988). According to the theory of the contrastive hierarchy, however, contrastive status is determined by a series of ordered, binary divisions. If the feature [voice] is ordered above the feature [cg], [voice] will be contrastive for implosives in Hausa. Differences between the Hausa and Tzutujil inventories require that, if [voice] is contrastive for implosives in Tzutujil, [cg] cannot be. This prediction is supported by differences in the patterning of restrictions against multiple [cg] segments in the two languages. References Archangeli, Diana. 1988. Underspecification in Phonology. Phonology 5: 183–207. Dresher, Elan. 2003. Contrast and asymmetry in inventories. in A.M. di Sciullo, ed.

Asymmetry in Grammar: Morphology, Phonology, Acquisition, 239-257. Amsterdam: John Benjamins.

Dresher, B. E., 2009. The Contrastive Hierarchy in Phonology. Cambridge: Cambridge University Press.

Dresher, B. Elan, Glyne L. Piggott, and Keren D. Rice. 1994. “Contrast in Phonology: Overview.” Toronto Working Papers in Linguistics 13: iii–xvii.

Hansson, Gunnar. 2001. Theoretical and Typological Issues in Consonant Harmony. Ph. D. Dissertation, University of California, Berkeley.

Hansson, Gunnar. 2010. Consonant Harmony: Long-Distance Interaction in Phonology. University of California Publications in Linguistics, 145. Berkeley, CA: University of California Press.

Newman, Paul. 2000. The Hausa language: an encyclopedic reference grammar. New Haven: Yale University Press.

MacEachern, Margaret. 1997 [1999]. Laryngeal Cooccurrence Restrictions. Doctoral dissertation, UCLA. [Published by Garland 1999].

Rose, Sharon and Rachel Walker. 2004. A typology of consonant agreement as correspondence. Language 80(3): 475-531.

Perceptual salience and cross-dialectal phonetic convergence in Spanish Bethany MacLeod (University of Toronto)

Behavioural alignment has been found to occur in many aspects of human behaviour, including speech (Chartrand & Bargh 1999). Within speech, choice of lexical item (Garrod & Doherty 1994), speech rate (Giles et al. 1991), pitch and vocal intensity (Goldinger 1998), and acoustic characteristics of segments (Babel 2009, Nielsen 2008) are all subject to alignment. Research on phonetic convergence (alignment of the acoustic-phonetic properties of segments) has found that talkers imitate the phonetic characteristics of the speech of another talker in non-social situations, such as in rapid shadowing (Goldinger 1998), and in social situations, such as during conversation (Evans et al. 2010, Pardo 2006). Much of the previous research has focused on convergence between speakers of the same dialect (and primarily of English); however, the pattern of phonetic convergence across two dialects may be distinct since it could involve the perceptual salience of the differences between the dialects. Trudgill (1986) suggests that the more salient a particular difference is, the more talkers will converge upon it. In contrast, Kim, Horton & Bradlow (submitted) found that the more distinct two talkers’ language backgrounds were, creating greater language distance between the talkers, the less phonetic convergence occurred. The present study investigates phonetic convergence during spontaneous conversation between speakers of Spanish from Madrid (MS) and Buenos Aires (BAS) in order to determine whether the extent to which speakers converge depends on the perceptual salience of the particular dialectal difference upon which they are converging. Six differences between the dialects were investigated including those predicted to be highly perceptually salient (such as the realization of orthographic <z>, <ci>, and <ce> as /θ/ by MS speakers and as /s/ by BAS speakers), moderately salient (such as the apical realization of /s/ in MS and the laminal realization in BAS), and least salient (such as the presence of the so-called exceptional hiatus (Hualde 1997, 1999) in the speech of MS speakers, but not BAS speakers) via a map task experiment (Anderson et al. 1991) testing 10 pairs of talkers (one talker from each dialect area). A perception task functioned as a metric to establish salience of the six dialectal differences for each participant and a repetition task determined that any lack of convergence would not be the result of an inability to articulate sounds in the contrasting dialect. All participants performed a sentence reading task before and immediately after exposure to the contrasting dialect.

Statistical analyses comparing the acoustic data collected in the sentence reading task pre-and post-conversation suggest that, while the more salient a particular difference is, the more it will be converged upon, there exists a salience threshold after which the difference is too great and convergence drops off sharply. Specifically, the dialectal differences found to be least perceptually salient only showed modest shifts in the speech of the participants towards each other (15% of the baseline discrepancy), while differences that were of intermediate salience were converged upon more (55% of baseline discrepancy), and those found to be highly salient converged very little in some cases (5%) or not at all in most cases.

This study helps clarify issues of perceptual salience in phonetic convergence contributing to our understanding of second dialect acquisition, and dialect contact and change in multi-dialectal regions. In addition, it provides the first examination of Spanish

in the emerging body of research on phonetic convergence in spontaneous conversation. References Babel, M. (2009). Phonetic and social selectivity in speech accommodation. Berkeley,

CA, University of California, Berkeley. PhD: 164. Chartrand, T. and J. Bargh (1999). "The chameleon effect: the perception-behavior link

and social interaction." Journal of Personality and Social Psychology 76(6): 893-910.

Evans, B. and W. Alshangiti (2010). "Investigating the effects of regional accent background on phonetic alignment in spontaneous speech. Paper presented at UCL Seminar Series."

Garrod, S. and G. Doherty (1994). "Conversation, co-ordination and convention: an empirical investigation of how groups establish linguistic conventions." Cognition 53: 181-215.

Giles, H., N. Coupland, et al. (1991). Accommodation theory: Communication, context, and consequence. In H. Giles, N. Coupland & J. Coupland (eds). . Contexts of Accommodation. Cambridge, Cambridge University Press: 1-68.

Goldinger, S. (1998). "Echoes of Echoes? An Episodic Theory of Lexical Access." Psychological Review 105(2): 251-279.

Hualde, J. I. (1997). "Spanish /i/ and Related Sounds: An Exercise in Phonemic Analysis." Studies in the Linguistic Sciences 27(2): 61-79.

Hualde, J. I. (1999). "Patterns in the Lexicon: Hiatus with unstressed high vowels in Spanish." In J.Gutierrez-Rexach & F.Martinez-Gil (Eds) Advances in Hispanic Linguistics: 182-197.

Kim, M., W. Horton, & A. Bradlow (submitted). "Phonetic Convergence in Spontaneous Conversations as a Function of Interlocutor Language Distance."

Nielsen, K. (2008). Word-level and Feature-level Effects in Phonetic Imitation, University of California, Los Angeles. PhD.

Pardo, J. (2006). "On phonetic convergence during conversational interaction." Journal of the Acoustical Society of America 119: 2382-2393. Trudgill, P. (1986). Dialects in Contact. Oxford, Basil Blackwell.

Can intonation be imitated? Spontaneous production vs. elicited imitation of Wh-questions

Olivia Marasco (University of Toronto)

Elicited imitation has been used in the study of both syntactic and phonological acquisition with a focus on first and second language (Munnich, Flynn, & Martohardjono, 1994;Markham, 1997). Very little, however, is known about the imitation of intonation in L2. Thus the current study seeks to explore whether intonation can be imitated and, if so whether imitation is a measure of L2 acquisition. In particular this study explores the acquisition of English Wh-questions by advanced Spanish speakers.

Both English and Spanish have an overall falling pattern in Wh-questions (Bartels, 1999;Sosa, 1999; English and Spanish respectively) however, further distinctions are made that characterize the two languages. English Wh-questions show a peak in the Wh-word but then show another high peak, possibly higher than the Wh-word, that will be perceived as the most prominent of the sentence (Ladd, 1997). The overall trend for Wh-questions in Spanish is falling but alternatives such as rise in final tonal group or rise-fall at the end are also common across varieties. In general the high tone of the Wh-word tends to be the highest tone of the utterance but does not always align within the stressed syllable of the word (Sosa, 1999).

For this study it was assumed that greater differences would emerge in the spontaneous production rather than the imitation. In particular it was predicted that a final falling contour would be easily imitated while the distribution and height of the tones within the phrase would be more problematic. Finally, peak alignment was expected to show the greatest amount of variation and transfer. In order to test these predictions a native speaker of English was recorded while he produced wh-questions. Five participants were asked to perform two tasks. First, they were presented with various scenarios plus a Q & A portion where the question part was blank. The participants were told to fill in the blank with the question that seemed most appropriate and their answers were recorded. For the second task they were informed that they would hear a possible question produced by someone else and they would have to repeat it. The word imitation was never used.

The preliminary results show that imitation does in fact take place. First of all, participants imitated contracted forms (where’s Martha) whenever the native English speaker recording contained such forms. Also, the participants, whose final tone was often a rise or rise-fall, imitated the final falling intonation of the native English speaker every time including those instances where the final falling tone had an extended low plateau. In pre-prompt production participants showed a tendency towards late peak alignment of the stressed syllable. In post-prompt production the participants aligned the peak with the stressed syllable, approximately50% of the time. The results suggest that it is possible to imitate intonation and that not all components are imitated in the same way and at the same rate. Boundary tones may be easier to imitate as they are at the end of the sentence while peak alignment may come with increasedlanguage use (Mennen, 2004). It seems that imitation is in fact a measure of L2 acquisition whenit is coupled with other elements such as the production of L2 segments.

References Bartels, C. (1999). The Intonation of English Statements and Questions. New York:

Garland Publishing Inc. Ladd, D., R. (1997). Intonational Phonology. Cambridge: Cambridge University Press. Markham, D. (1997). Phonetic Imitation, Accent, and the Learner. Lund, Sweden: Lund

University Press. Mennen, Ineke. (2004). Bi-directional interference in the intonation of Dutch speakers of

Greek. Journal of Phonetics, 32, 543-563. Munnich, E., Flynn, S., & Martohardjono, G. (1994). Elicited imitation and

grammaticality judgement tasks: What the measure and how they relate to each other. In E. E. Tarone, S. M. Gass & A. D. Cohen (Eds.), Research Methodology in Second-Language Acquisition (pp. 227-243). Hillsdale, NJ: Lawrence Erlbaum Associates Publishers.

Sosa, J. M. (1999). La entonación del español: su estructura fónica, variabilidad y dialectología. Madrid: Cátedra.

Native dialect effects in non-native vowel production: Cuban and Peninsular Spanish learners of English

Irina Marinescu (University of Toronto)

It has been shown that the native dialect influences L2 perception (e.g. Escudero and Boersma 2004, Cutler et al. 2005). However, no study has explored the potential differences in L2 vowel production determined by the learner’s native dialect. Thus, the present research investigates through two experiments how the English vowels /æ, ʌ, ɑ/ are produced by learners with distinct native dialects, Cuban and Peninsular Spanish. Experiment 1 compares the native vowel spaces to determine the extent of the cross-dialectal differences and Experiment 2 assesses the differences in L2 production of two groups of learners having different native varieties. Based on existing reports on Spanish dialects (e.g. Morrison and Escudero 2007, Guitart 1996), a more fronted vowel space is predicted for the Peninsular (PS) as compared to the Cuban variety (CS). Given the differences between the L1 varieties, it is predicted that the L2 English vowels /æ, ʌ, ɑ/ differ systematically for the two groups of learners and tend to be more fronted in the interlanguage of PS learners as compared to CS learners. Moreover, dialect-specific patterns of assimilation are predicted, with the /æ, ʌ/ vowels tending to overlap in the PS group’s L2 production and /ʌ, ɑ/ tending to overlap for the CS learners.

In Experiment 1, 40 Spanish native speakers (21 from PS, 19 from CS) read a series of pVpa and bVba real Spanish words (where V is one of the five vowels /i, e, a, o, u/) inserted in a carrier phrase. Acoustic measurements (F0-F3 formants, durations) were obtained from more than 800vowels and normalized. In Experiment 2, a subset of 19 participants from Experiment 1 (10 from PS, 9 from CS) who were advanced learners of English, were tested in a sentence completion task. They listened to 58 short definitions read by two English native speakers and were asked to finish the sentence with one word that suited the definition and then repeat the word inserted in a carrier phrase. The targets were CVC real English words (where V was one of the /æ, ʌ, ɑ/ vowels and C were stops or /h/). Six native English controls from Toronto were tested with the same procedure. More than 1,100 vowels for L2 production and 450 vowels for the target language were analyzed.

Twelve one-way ANOVAs were conducted for each of the /æ, ʌ, ɑ/ vowels to compare the effect of the linguistic variety (Canadian English, PS, CS) on the F1, F2, F3 and duration of the vowels. Then, within each dialect, L2 vowels were compared through a linear mixed model with repeated measures for vowels to determine whether the predicted pattern of category overlaps exists. Several significant effects at the p < .05 level were found.

Overall, L2 learners produce vowels that differ from the target along at least one variable (F1-F3 or duration). Moreover, L2 vowels differ significantly between the two groups of learners in F1 (height) and F2 (frontness) and duration (except for /æ/ in F1). Within the PS group, however, learners managed to produce all L2 vowels as distinct spectral categories. They did not produce though any duration distinction between tense and lax vowels. Within the CS group, as predicted, there is some category overlap between /ʌ, ɑ/ which are produced with no significant differences along F3. The remaining variables are significantly different. A closer look at the individual variation

indicates that within the PS group, 7 out of 10 subjects display the /æ, ʌ/overlap and within the CS group, 7 out of 9 subjects display the /ʌ, ɑ/ overlap, the remaining subjects having an unclear pattern of vowel categorization. Moreover, duration is not used to signal any contrast by L2 learners in PS but is used distinctively in CS. Based on these findings, we conclude that phonetic differences between native dialects influence non-native production. References Cutler A., Smits R., Cooper N. 2005. Vowel perception: Effects of non-native language

vs. nonnative dialect. Speech Communication 47. 32–42. Escudero, P., Boersma P. 2004. Bridging the gap between L2 speech perception research

and phonological theory. SSLA 26. 551-585. Flege, J.E. 1995. Speech Learning Theory, Findings, and Problems. In: Strange W. (ed.),

Speech perception and linguistic experience: Issues in cross-language research. York Press Timonium. 279–304.

Guitart, J. M. 1996. On the source of variance among Hispanophones with regard to the way they perceive certain American English vowels. Paper presented at the 1996 AATSP Biennial Northeast Regional Meeting, University of Amherst.

Morrison, G.S., Escudero, P. 2007. A cross-dialect comparison of Peninsular-and Peruvian-Spanish vowels. In: Proceedings of the ICPhS XVI Saarbrücken: 1505-1509. <www.icphs2007.de> Last access 30 Nov 2010.

Child Bilingual Language Acquisition of the Present Perfect-Preterit Distinction in Spanish

Joanne Markle LaMontagne (University of Toronto)

Acquisition of the preterit and imperfect aspectual tenses in Spanish has been examined extensively in studies on adult first language attrition (Cuza, 2008; Montrul, 2002) and in adult L2 acquisition (Montrul & Slabakova, 2002; 2003). In simultaneous bilingual child populations, the acquisition of aspectual properties of the present perfect-preterit distinction in Spanish is relatively unexplored. It is generally assumed that simultaneous bilingual children acquire two languages from birth and develop two separate grammatical systems from a very young age (De Houwer, 1990; Meisel, 1989), but these grammars are also said to be interdependent (Paradis & Genesee, 1996). The goal of this pilot study is to test the acquisition of aspectual properties in the interpretation of the present perfect-preterit distinction in Spanish by Spanish-English bilingual children, and to also consider the possible effect of transfer in bilingual grammars.

Spanish and English are two languages that make a structural and interpretational distinction between the present perfect and simple past tenses. This distinction however varies with respect to the range of meanings associated with it (Iatridou et al., 2003). The present perfect shares properties with both temporal and aspectual forms; many adverbs that one would expect possible are often disallowed. For example, in English certain past-oriented adverbs are incompatible with the present perfect (*John has arrived this morning), yet compatible in Spanish (Juan ha llegado esta mañana). This restriction is puzzling since the present perfect is a temporal configuration that expresses anteriority. Giorgi & Pianesi (1997) propose that because linguistic variation is encoded morphologically, morphology shapes tense interpretations. From a different perspective, De Swart (1998) posits that tense heads are sensitive to aspectual properties, such as states. Extending De Swart’s view, Schmitt (2001) proposes that tense receives its semantic interpretation by selecting a semantically appropriate eventuality. Schmitt argues that children may learn tense interpretations from other cues, such as adverbials, which provide a more reliable cue than morphology in the acquisition of tense interpretations.

In this study I wish to investigate the following questions: Is a child’s interpretation of the present perfect different in a bilingual context? More precisely, is a child’s interpretation of the present perfect in Spanish weakened because of influence from English? I predict that in a recent past context, Spanish-English bilingual children may transfer the selectional properties of the English present perfect into Spanish based on the typological differences between the two. The participants in this pilot study include two simultaneous bilingual Spanish-English children (6;0 and 3;10, respectively) from Toronto. The oldest child (SE1) attends an English elementary school and the youngest (SE2) a full-time English day care. Two native speakers of Spanish (1 Peninsular Spanish, 1 Mexican Spanish) serve as controls. A sentence preference task is used to test two temporal adverbial conditions: ya (already) and todavía no (not yet). In Spanish the present perfect is obligatory when an action is negated in the past by todavía no. Since in Spanish ya indicates realization, it is

obligatory with the preterit. In English, however, the use of both past tenses can occur with not yet and already.

Preliminary results show that when ya is used in a recent past context, SE1 prefers the use of the present perfect (75%) over the preterit (25%) while SE2 prefers the simple past (75%) over the present perfect. With respect to todavía no, SE1 shows a greater preference for the simple past (75%) and SE2 an equal preference for both tenses. These results suggest that in the case of SE1, influence from English follows from early and prolonged exposure to English in an educational setting. References Cuza, A. (2008). The L2 Acquisition and L1 Attrition of the Interpretation and Use of

Aspectual Properties in Spanish among English-speaking L2 Learners and Long-term Spanish Immigrants. Ph.D. dissertation, University of Toronto.

De Houwer, A. (1990). The acquisition of two languages from birth: A case study. Cambridge: Cambridge University Press.

De Swart, H. (1998). Aspect Shift and Coercion. Natural Language & Linguistic Theory 16, 347-385.

Giorgi, A. & F. Pianesi. (1997). Tense and Aspect: From Semantics to Morphosyntax. Oxford: Oxford University Press.

Iatridou, S., E. Anagnostopoulou, & R. Izvorski. (2003). Observations about the form and meaning of the Perfect. In A. von Stechow et al. (Eds.), Perfect Explorations (pp. 153-204). Berlin: Mouton de Gruyter.

Meisel, J. (1989). Early differentiation of languages in bilingual children. In K. Hyltenstam & L. Obler (Eds.), Bilingualism across the lifespan. Aspects of acquisition, maturity and loss (pp. 13-40). Cambridge: Cambridge University Press.

Montrul, S. (2002). Incomplete Acquisition and Attrition of Spanish Tense/Aspect Distinctions in Adult Bilinguals. Bilingualism: Language and Cognition, 5:1. 39-58.

Montrul, S. & Slabakova, R. (2002). The L2 Acquisition of Morphosyntactic and Semantic Properties of the Aspectual Tenses Preterite and Imperfect. In A.T. Pérez-Leroux & J. Liceras (Eds.), The Acquisition of Spanish Morphosyntax: The L1/L2 Connection (pp. 113-149). Dordrecht: Kluwer.

Montrul, S. & Slabakova. R. (2003). Competence similarities between native and near-native speakers: An investigation of the preterit/imperfect contrast in Spanish. Studies in Second Language Acquisition 25: 351-398.

Paradis, J. & F. Genesee. (1996). Syntactic Acquisition in Bilingual Children: Autonomous or Interdependent? Studies in Second Language Acquisition, 18, 1-25.

Schmitt, C. (2001). Cross-linguistic Variation and the Present Perfect: the Case of Portuguese. Natural Language & Linguistic Theory 19(2), 403-453.

Juxtaposed Complex Predicates in Niuean Diane Massam (University of Toronto)

This paper examines complex predicates (CPs) in Niuean (Oceanic) arguing for a syntactic (vs lexical) analysis, which accounts for the various argument sharing patterns and the juxtapositional nature of Niuean CPs. The analysis contributes rich new data and new analysis to the extensive literature on the properties of CP constructions cross-linguistically. The Oceanic family is known for its CPs (Bril and Ozanne-Rivierre 2004, Crowley 2003), but Niuean has not received attention. An example of a Niuean resultative CP is provided in (1). (1) Ne hifi kū e ia haaku ulu (P=proper)

Pst cut short ErgP 3Sg 1SgGen hair 'She cut my hair short.' (FW1:2010)

CPs are difficult to define cross-linguistically (Aikhenvald 2006). Following Sebba 1987 I consider cases where both members of the CP are verbs, and share one Tense-Aspect marker and there is no coordinator appearing between them. I examine only core [V+V] CPs.

Niuean CPs categorize descriptively into several types (e.g. first member is from a closed class, second member is from an open class, etc.) and these types will first be outlined in the talk.

It is a common property (perhaps a defining one) of CPs that they share arguments. For example, in (1) cut has two arguments, one of which is (hair) and the predicate be=short shares this argument. The CPs of Niuean divide into four types with respect to argument sharing.

(4) I. Share No Arguments (modifying/ambient verbs, some auxes, light verbs) II. Share All Arguments (conjunctives, intensifying light verbs) III. Share External Argument (some auxiliaries, derivational light verbs)

I account for the various Niuean classes as follows. (I assume a modified Pronominal Argument Hypothesis structure for Niuean in which predicate formation takes place prior to the merging in of arguments (Jelinek 1984, Massam 2010)). For the root ambient CPs that share no arguments and those that share all arguments (I, II) I propose a syntactic (v+root+verb) compounding analysis as in Harley (2008), formed by pseudo-incorporation (Massam 2001) rather than by incorporation, as proposed by Harley for English. This leads to an account of the juxtapositional nature of Niuean CPs allowing for a fuller understanding of compounding in isolating languages. [v [ V [v V]]

I consider the auxiliary and light verb CPs that share no arguments (I) to be higher verbs selecting an IP complement, which contains a complete clause. The verb fronts to the specifier of IP to yield V+VSO order. This is supported by the fact that some of these auxiliaries can optionally take a CP with an overt complementizer, suggesting their complements can be IP or CP, variously. There is no process of juxtaposition in these cases. [V [V S <V> O]]

I consider the auxiliary and light verb CPs that share no arguments (I) to be higher verbs selecting an IP complement, which contains a complete clause. The verb fronts to the specifier of IP to yield V+VSO order. This is supported by the fact that some of these auxiliaries can optionally take a CP with an overt complementizer, suggesting their complements can be IP or CP, variously. There is no process of juxtaposition in these cases. [V [V S <V> O]]

The complex predicates that share external arguments (III) are analyzed as selecting a vP complement, with V in its spec, which constitutes the top of rest of the predicate domain. The DP arguments are merged above this, hence both verbs are in the predicate to which voice relates the external argument, accounting for the shared argument. The two Vs are in a juxtapositional relation with each other syntactically, both within the predicate domain. [V [V v…]]

Finally, internal argument sharing CPs (IV) are analyzed as merging with the VP containing the Verb and the NP object (which can be incorporated or null and coindexed with the higher predicate-external DP object argument). The second predicate thus has a semantic relation with both the verb and with the internal argument, and forms a juxtapositional relation with the verb syntactically, since both are within the predicate domain. [VP V ].

The Niuean data and analysis will be compared with well-known structural analyses as well as examined from a typological point of view. References Aikhenvald, A. 2006. Serial Verb Constructions in Typological Perspective. In

Aikhenvald and Dixon (eds.) Serial Verbs: A Cross-linguistic Typology. Oxford University Press.

Bril Isabelle, and Françoise Ozanne-Rivierre, 2004. Complex predicates in Oceanic languages: Studies in the dynamics of binding and boundedness. Berlin: Mouton de Gruyter.

Crowley, Terry. 2003. Serial verbs in Oceanic: A descriptive typology. Oxford University Press.

Harley, Heidi, 2008. Compounding in Distributed Morphology. In Rochelle Lieber and Pavol Stekauer (eds.) The Oxford handbook of compounding. Oxford: Oxford University Press. pp.129-144. Jelinek, Eloise: 1984, 'Empty Categories, Case and Non-configurationality', Natural Language and Linguistic Theory 2, 39-76. Massam, Diane. 2001. Pseudo noun incorporation in Niuean. Natural Language and Linguistic Theory 19:153-197.

Massam, Diane. 2010. Deriving inverse order: The issue of arguments. in Raphael Mercado, Eric Potsdam & Lisa Travis (eds.), Austronesian and theoretical linguistics. Amsterdam/Philadelphia: John Benjamins. pp. 271-296.

Mosel, Ulrike. 2004. Complex predicates and juxtapositional constructions in Samoan. In Bril, Isabelle and Françoise Ozanne-Riviere, 2004. pp. 263-296.

Sebba, Mark. 1987. The syntax of serial verbs: an investigation into seralisation in Sranan and other languages. Amsterdam-Philadelphia: John Benjamins.

Noun incorporation and the new head movement Éric Mathieu & Michael Barrie (University of Ottawa)

Nutshell: Head movement, formerly shunned or relegated to PF (Boeckx and Stjepanović, 2001, Chomsky, 2001, Fanselow, 2003, Koopman and Szabolcsi, 2000, Mahajan, 2003) has enjoyed a resurgence as a syntactic operation (Baker, 2009, Roberts, 2010). We argue here that Noun Incorporation (NI), once a prime example of syntactic head movement (Baker, 1988), is unformulable under Roberts’ (2010) recent account of head movement. Instead, we suggest that an XP movement is required. Head Movement: Roberts (2010) recasts head movement as incorporation. His proposal is that head movement applies where the Goal of an Agree relation is defective (Roberts, 2010: 62). (1) A goal G is defective iff G’s formal features are a proper subset of those of G’s

Probe P. Roberts’ proposal handles cliticization nicely, given the natural assumption that

clitics are defective (Déchaine and Wiltschko, 2002), as a Case assigner, v* has [uφ] and the clitic is maximally a φP (crucially, not a DP) consisting solely of a [iφ] feature set, hence a subset of the features of v*. Roberts (2010) extends his proposal to noun incorporation. The main idea is that incorporated nouns are defective: they have an nP layer but no inflectional or DP layer. The derivation for NI goes as follows: inside nP, the √ incorporates with n. V incorporates with v*. The step of NI is then incorporation of n-√ into v*-V. Roberts argues that full agreement implies a [D] feature. Thus, v* in languages with rich object agreement has a ϕ-set as well as [D]. Since D is a type of N feature, v* and n-N are not featurally distinct.

Discussion: Consider first NI in Ojibwe (Algonquian). Free standing nouns such as bigw ‘pitch, gum’ and jiiman ‘boat’ incorporate together with their nominalizer, (2). Such forms are unproblematic for Roberts. However, material larger than a √+n complex can undergo NI in Algonquian languages (Ojibwe, (3)). Fox can also incorporate demonstratives (not shown for lack of space) (Michelson, 1917).

(2) N-gii-beengw-(i)naa-gan-ee. (Rhodes, 1976: 263) 1-PST-dry-dish-NZLR-VAI ‘I dried the dishes.’ (3) n-gii-gichi-gizhebaa-wiisin! p. 127 in Ningewance Talking Gookom’s Language 1SG-PST-big-breakfast-eat ‘I ate a big breakfast!’

Turning to Mohawk (Northern Iroquoian), we note that Roberts’ proposal relies crucially on rich object agreement, which Mohawk possesses. Fatally, however, NI constructions are both morphologically and syntactically intransitive (see also Koenig and Michelson, 2008). In transitives, subj and obj agreement do not vary wrt aspect. In intransitives, however, the single argument uniformly takes obj agreement in perfectives, as the following example shows.

(4) ro-ʔwahr-a-k-u (Deering and Delisle, 1976: 441) 3.SG.M.ACC-meat-EPEN-eat-PERF ‘He has eaten meat.’

Thus, NI in Algonquian contains material larger than a head and non-defective material; and Mohawk, crucially fails to agree with its object, as the [D] feature is absent. We conclude, then that the head movement analysis of Roberts cannot be maintained. Conclusion: Given the problems with head movement for NI, we conclude that NI must be derived via XP-movement. We derive NI in Mohawk as follows. The noun root merges with n and raises to SpecnP rather than to n. This nP merges with V and subsequently raises to SpecVP. This process proceeds in a roll-up fashion, giving rise to the structure below. NI in Ojibwe proceeds in a similar fashion, but with added complications not shown for reasons of space. (5) [vP [VP [nP Ni n ti]j V tj]k vtk] References Baker, M. C. 1988. Incorporation: A Theory of Grammatical Function Changing.

Chicago, IL: University of Chicago Press. ---. 2009. Is head movement still needed for noun incorporation? Lingua 119:148-65. Boeckx, C., and Stjepanović, S. 2001. Head-ing toward PF. LI 32:345-55. Chomsky, N. 2001. Derivation by Phase. In Ken Hale: A Life in Language, ed. M.

Kenstowicz, 1-52. Cambridge, MA: MIT Press. Déchaine, R. M., and Wiltschko, M. 2002. Decomposing Pronouns. LI 33:409-42. Deering, N., and Delisle, H. H. 1976. Mohawk: A Teaching Grammar. La Macaza, QC:

Manitou Community College. Fanselow, G. 2003. Münchhausen-Style Head Movement and the Analysis of Verb

Second. UCLA Working Papers in Linguistics 13:40-76. Koenig, J.-P., and Michelson, K. 2008. Revisiting the realization of arguments in

Iroquoian. Paper presented at Annual Meeting of the Linguistic Society of America, Chicago, IL.

Koopman, H., and Szabolcsi, A. 2000. Verbal Complexes. Cambridge, MA: MIT Press. Mahajan, A. 2003. Word Order and (Remnant) VP Movement. In Word Order and

Scrambling, ed. S. Karimi, 217-37. Malden, MA: Blackwell Publishers. Michelson, T. 1917. Notes on Algonquian languages. IJAL 1:50-7. Rhodes, R. A. 1976. The Morphosyntax of the Central Ojibwa Verb, Department of

Linguistics, University of Michigan: Ph.D. Dissertation. Roberts, I. 2010. Agreement and Head Movement: Clitics, Incorporation, and Defective

Goals. Cambridge, MA: MIT Press.

Null Object Constructions In Tagalog Lareina Milambiling (University of Western Ontario)

Empty object (EO) constructions vary from language to language (Massam & Roberge 1989). A French example is illustrated in (1): (1) La course a pied garde___en forme.

‘Running keeps in shape.” (Massam & Roberge 1989). Although (1) is ungrammatical in English, English EO constructions can be found in certain discourse contexts, such as in recipes or other instructions (Massam & Roberge 1989). In this squib I examine what can be dropped in Tagalog recipe contexts (RC) and I show that it is always the topic. Although in Tagalog the object is not dropped, I will continue to use the terms RCEO and EO for ease of comparison to English. Following Schachter (1976), I assume that Tagalog sentences contain a topic and not a subject. The noun phrase (NP) that is the topic is indicated by the morphology of the verb and is preceded by the morpheme “ang.” RCEOs can occur in Tagalog in the goal-topic (GT), beneficiary-topic (BT) and directional-topic (DT) voices (2). (2) Paghaluin ang giniling na karne, sibuyas,at paminta. Igulong _____

GT-mix T-the ground beef onion and pepper. BT-roll Mix the groud beef, onion and pepper. sa binating itlog at pabalutan ng breadcrumbs. Iprito _ in beaten egg and DT-wrap with breadcrumbs. BT-Fry Roll in beaten eggs and cover in breadcrumbs. ng lubog sa mainit na maitika. Hatiin ___ sa gitna bago ihain. submerged in hot oil. GT-cut in middle before serving. Deep-fry. Cut down the middle before serving.

We can see that different voices can be used to license Tagalog RCEOs. However,actor-topic (AT) voice cannot license a RCEO (3). (3) Isama ang mantikilya. . *Magbalot ________ sa asukal. BT-Add the butter AT-Wrap in sugar. Add the butter. Cover with sugar. I propose that this is the case because objects in AT must be indefinite, but RCEOs are necessarily definite (Bowen 1965). I also argue that unlike English EOs, Tagalog can allow EOs to occur outside of the recipe context. The topic can be dropped in Tagalog non-recipe or instruction contexts (4). (4)a. Madumi ang damit ko. Labhan mo ______. [are] dirty T-the pants my DT-wash you. My pants are dirty. Wash (them).

I propose that RCEOs are simply dropped topics, as we see elsewhere in Tagalog. I also show that, as in RC, AT verbs cannot license EOs in other contexts. (5)a Binili ng babae ang asukal. *Nagbalot siya _____ sa cellophane. GT-Bought the woman sugar. AT-wrapped she in cellophane. The woman bought sugar. She wrapped (it) in cellophane.

In summary, both English and Tagalog allow for null arguments in recipes, but

Tagalog does not follow the same pattern as English. I propose that in Tagalog what is dropped is always the topic while in English it is the object. We can see that topic-drop is licensed in Tagalog under certain conditions in multiple contexts, using goal-topic verbs, beneficiary-topic verbs and directional-topic verbs. References: Bowen, J. Donald,. Ed. Beginning Tagalog: A Course for Speakers of English. Los

Angeles: University of California Press, 1965. Print. Kain Na! Metro Manila: Anvil Publishing, Inc. 1993. Print. Massam, Diane, and Yves Roberge. “Recipe Context Null Objects in English.” Linguistic

Inquiry. The MIT Press. 20.1 (1989): 134-139. Print. Schachter, Paul. “The subject in Philippine Languages: topic, actor, actor-topic, or none

of the above.” Subject and Topic. Ed. C. Li. New York: Academic Press, 1976. 492-518. Print.

The role of input in the Acquisition of ser and estar by speakers of L1 Romanian Camelia Nunez (University of Western Ontario)

Research on the acquisition of the two Spanish copulas, ser and estar, has been extensive and from many different theoretical perspectives. Initially the use of the two verbs was explained through an extensive categorization of adjectives: those that accept ser, those that accept estar and those that accept both (Navas-Ruiz, 1963; Vañó-Cerdá, 1982). Syntactic approaches refer to the difference between the two as a matter of aspect (Luján, 1981; Schmitt, 1992) as well as the predicate type (Fernández Leborans, 1995; Leonetti, 1994). Although the debate on the use of the copulas is ongoing, a consensus has been reached in regards to their distribution: (1) copula + nominal (Shakira es una cantante famosa / Shakira is a famous singer); (2) copula + locatives (¿Dónde están las llaves? / Where are the keys?); (3) copula + eventives (Le fiesta es en tu casa. / The pary is at your house) and (4) copula + adjectivals. This last category is divided in three subcategories: (i) adjectives that only allow ser (Tú eres inteligente. / You are intelligent); (ii) adjectives that only allow estar (El vaso está vacío / The glass is empty) and (iii) adjectives that can be used with both copulas (Diego es aburrido / Diego is boring vs. Diego está aburrido / Diego is bored).

Given the complexity of ser and estar, it is not surprising that it poses great difficulty for those learning Spanish as a foreign language. Studies investigating the acquisition of the copula by formal learners of Spanish have found that the ser and estar contrast is acquired late. Nonetheless, research has shown that all formal learners of L2 Spanish go through similar stages of copula acquisition. (VanPatten, 1987; Ryan and Lafford, 1992; Ramirez-Gelpi, 1995). Little is known however, about the acquisition of the copula in an informal environment.

The current paper will report on an empirical study that examines the acquisition of the Spanish copulas by two groups of L1 Romanian speakers. The first group of participants, informal learners of Spanish (n = 15), has acquired the language exclusively from watching Latin American soap operas (telenovelas) with Romanian subtitles. The second group is comprised of formal learners studying Spanish at a university in Romania (n = 15). Results will be measured against a control group of native speakers (n = 15). Specifically, this paper will analyze the acquisition of the Spanish copula by informal learners (the telenovela viewers) and determine the role of input in the acquisition process.

Participants completed two main tasks: a standardized Comprehension task and a Sentence Selection Task. The Comprehension Task examined the global understanding of the Spanish language and determined the overall proficiency of the learners in the L2, while the Sentence Selection Task looked at whether L1 speakers of Romanian can correctly distinguish between the copulas ser and estar and whether they follow the acquisition stages of formal learners as established by previous research.

Results on the Sentence Selection Task for the first group showed that telenovela viewers have achieved some success in semantically distinguishing between the two copulas but

have not scored as high as the formal learners group. Preliminary findings of the current study showed that the informal learners follow the same stages of acquisition of the copula as those learning Spanish in a formal environment. Moreover, the higher scores of the formal group point out to the importance of formal instruction in the acquisition process of the Spanish copula. References Fernández Leborans, M. J. (1999). La predicación: Las oraciones copulativas. In I.

Bosque & V. Demonte (Eds.),Gramática descriptiva de la lengua española. (pp. 2357-2460). Madrid: Espasa.

Luján, M. (1981). The Spanish copulas as aspectual indicators. Lingua, 54, 165-210. Leonetti, M. 1994. Ser y estar: estado de la cuestión. Barataria 1:182-205. Navas Ruiz, R. (1963). Ser y estar: El sistema atributivo del español.Salamanca:

Publicaciones del Colegio de España. Ramírez-Gelpi, A. (1995). The Acquisition of Ser and Estar among Adult Native English

Speakers Learning Spanish as a Second Language. Los Angeles, CA: University of Southern California Dissertation.

Ryan, J. M., & Lafford, B. A. (1992). Acquisition of lexical meaning in a study abroad environment: Ser and estarand the Granada experience. Hispania, 75, 714-722.Schmitt C. (1992). Ser and estar: A matter of aspect. NELS, 22, 411-425.

Schmitt, C.(1992). Ser and estar: a matter of aspect. Proceedings of NELS, 22: 411-425. Vanpatten, B (1987) Classroom Learners’ Acquisition of ser and estar: accounting for

developmental patterns. In R. W. Andersen (Ed.) Cross-linguistic series on second language research (pp 61-75). Cambridge: Newbury House Publishers.

Vañó-Cerdá, A. (1982). Ser y estar + adjetivos: Un estudio sincrónico y diacrónico.Tübingen: Gunter Narr.

Subjects and Subject Clitics in Embedded Clauses in Jordanian Arabic Osama Omari (Memorial University)

In this study, I account for the derivation of preverbal subjects in Jordanian Arabic (JA) embedded clauses using a new model of syntactic movement (Branigan 2010). Four preverbal subject patterns are found in JA embedded clauses. In one, the subject of the embedded clause appears to be a pronominal clitic attached to the complementizer. This pronoun agrees with the embedded verb and must be realized in the accusative form, as illustrated in (1a). The second pattern occurs when the agreeing clitic is optionally ‘doubled’ by a coreferential lexical DP, as seen in (1b). The third possibility is when the lexical subject cooccurs with the non-agreeing clitic form -uh (henceforth the ‘default’ clitic), as shown in (1c). When the default clitic is used, the realization of a lexical DP becomes obligatory. Lastly, when the lexical subject is in postverbal position, the clitic attached to C is obligatorily –uh, the ‘default’ clitic, as shown in (1d). (1) a. Ali ʔaal ʔin-ha ʔij-at. Ali said.3ms that-her came-3fs

‘Ali said that she came.’ b. Ali ʔaal ʔin-ha (il-bint) ʔij-at.

Ali said.3ms that-her the-girl came-3fs c. Ali ʔaal ʔinn-uh il-bint ʔij-at.

Ali said.3ms that-def. the-girl came-3fs d. Ali ʔaal ʔinn-uh/*ha ʔij-at il-bint.

Ali said.3ms that-def./*her came-3fs the-girl I show that the JA complementizer ʔin is endowed with a provocative (P) feature. Therefore, it imposes the creation of an external copy of its goal projection. Then, the external copy created under provocation merges with C. Accordingly, the C in examples (1a) and (1b) finds a null referential pro and a lexical DP, respectively, as ‘appropriate’ goals. As a result of provocation, an ‘external’ copy of each of the ‘goals’ will be created. Given that a DP can have the form of a pronoun with a deleted NP (the NP deletion theory) (Elbourne 2001), I assume that the form of the external copy in (1a) and (1b) will be [DP[D –ha][NP]] and [DP[D –ha][NP bint]], respectively. It can be noted here that the ‘shape’ of the ‘external’ copy in (1a) is identical to the original copy, while the ‘external’ copy in (1b) is not. These ‘external’ copies merge with C as clitics deriving the complex form (C+clitic) ʔin-ha ‘that-her’ in (1a) and (1b). At the PF interface, either both the external copy (i.e., the subject clitic) and the original copy are given phonetic content, as in (1b), or only the external copy is spelled-out (1a). As for (1c), where the lexical subject cooccurs with a default clitic, the derivation of the ‘default’ subject clitic comes as a result of C having the option of satisfying its p-feature by finding a matching goal that is external to the phrase marker that it heads. Thus, C finds -uh which is an ‘external’ goal that is already generated with matching features. Thus, –uh merges with C deriving the ʔinn-uh form. In this case, the overt realization of the lexical subject becomes obligatory because the two copies are not identical. Since the ‘external’ copy in (1c) (i.e., the ‘default’ clitic) does not have phi-features, the spelling out of the phi-features of the internal ‘goal’ copy becomes obligatory. In (1d), the internal goal is not ‘local’ and thus

cannot be reached by C; therefore, C probes for –uh, an ‘external’ goal, as the only option. This analysis provides a unified account for the derivation of subject clitics and clitic ‘doubling’ in Arabic and provides a solution for the problem of ‘default’ clitic in the literature of Arabic (e.g., Mohammad 2000). References Branigan, Phil. 2010. Provocative Syntax. Massachusetts: MIT Press. Elbourne, Paul. 2001. E-Type Anaphora as NP-Deletion. Natural Language Semantics

9:241-288. Mohammad, Mohammad. 2000. Word Order, Agreement and Pronominalization in

Standard and Palestinian Arabic. Amsterdam: John Benjamins.

!""#$%&'()#$*%#''("#%$+&,#'(&-(.$%$/$'0(

!"#$%#&%'()%*+,-.++,.+/()%*+0,(.+,-"1#&,%#,(,#123+.,"-,&%--+.+#),/(#41(4+05,%#'/1&%#4,

6.+#'78,

9:;,<=(%,&+,>1"%,?'.%.+8,@A,7(*+,0"2+)7%#4,)",B.%)+,B%)78=,

C7+0+,-.++,.+/()%*+0,(.+,)DE%'(//D,(#(/DF+&,(0,GH0,.()7+.,)7(#,IH05,B7+.+,)7+,%#&+-%#%)+,(#&,

2"&(/,%#)+.E.+)()%"#0,(.%0+,-."2,(,'"23%#()%"#,"-,)7+,2().%J,*+.3,(#&,)7+,GH,9AF*".0K%,

:LLM5,N."01,(#&,O(#&2(#,:LLM5,G(E"#%4.",PQQR;8,S%2%/(.,'"#0).1')%"#0,%#,T(/(4(0D,%#*"/*+,7+(&/+00,.+/()%*+05,B7%'7,B+,(.41+,)",3+,IH0,9P$R;8,,

9P;,T(#(#(,,,U,%F(D,7",0(0(#(V,, (7"8, , 9R;,C(&%(*%K",,, , ,,,,U,%F(D,7",7(#%#(V8,, 7(*+,, , WXO,,AWW,B(078CC,, :SN9!YT;, , /""K$-".8CC8:SN9NX!;,,WXO,,,AWW,+()8CC,

, @A,7(*+,0"2+)7%#4,)",B(078=, , , , , @A,(2,/""K%#4,-".,0"2+)7%#4,)",+()8=,

T".+"*+.5,B+,07"B,)7(),)7+,%#&+-%#%)+#+00,'"2+0,-."2,)7+,2().%J,*+.3,(#&,2"&(/%)D,%0,(,

.+01/),"-,%..+(/%0,2(.K%#4,9!";,"#,)7+,+23+&&+&,*+.38,, A..+(/%0,7+(&/+00,.+/()%*+0,'(#,"''1.,(0,"3Z+')0,9P;,".,013Z+')0,9R;8,C7+D,(.+,%#)+.E.+)+&,

(0,%#&+-%#%)+,(#&,)(K+,#(.."B,0'"E+,1#&+.,#+4()%"#8,T".+"*+.5,)7+,%#)+.E.+)()%"#,7(0,(,

2"&(/,'"2E"#+#),9E"00%3%/%)D;[,9P;,2+(#0,)7(),)7+.+,%0,0"2+)7%#4,A,'(#,B(078,6%#(//D5,)7+0+,(.+,#"),-.++,.+/()%*+0,91#&+.,0)(#&(.&,&+-%#%)%"#0;[,)7+.+,%0,#",#!$E7.(0+8,

, \+,E."E"0+,)7(),)7+0+,7+(&/+00,.+/()%*+0,(.+,IH0,5,7+(&+&,3D,$%&'8,A#,7+(&+&,.+/()%*+05,$%&',%0,(.41(3/+,%#,G],9^++#(#,:L_P;5,31),7+.+,%),%0,(,I]8,X*%&+#'+,-".,%)0,&+)+.2%#+.,0)()10,

'"2+0,-."2,)7+,-('),)7(),%#,9P;,$%&',%0,"E)%"#(/5,21'7,/%K+,&+)+.2%#+.0,"-,"3Z+')05,31),%#,9R;,

$%&',%0,"3/%4()".D5,'"#-".2%#4,)",)7+,T(/(4(0D,.+>1%.+2+#),)7(),013Z+')0,3+,7+(&+&,3D,(#,"*+.),I]8,T".+"*+.5,GH,"3Z+')0,%#,T(/(4(0D,"3/%4()".%/D,+J).(E"0+,)",)7+,.%47),"-,)7+,013Z+')`,

9P;,07"B0,)7(),7+(&/+00,.+/()%*+,"3Z+')0,&",#"),9(#&,%#,-('),'(##");,+J).(E"0+8,

, \+,(/0",(.41+,)7(),)7+,0E+'%(/,2"&(/,%#)+.E.+)()%"#,"-,)7+0+,7+(&/+00,.+/()%*+0,%0,&1+,)",)7+,%..+(/%0,2(.K%#4,9!(a!"$;,"#,)7+,+23+&&+&,*+.38,b+(&/+00,.+/()%*+0,B%)7"1),%..+(/%0,(.+,

%#)+.E.+)+&,21'7,/%K+,-.++,.+/()%*+0,c,(0,&+-%#%)+,(#&,B%)7"1),2"&(/%)D8,,

9d;,C0D,)%(K",, , , ,,,,U,%F(D,#()("#DV8,, , 9e;,!%*%&D,U%F(D,7",0(0(#(V,, (7"8,

, !XN,/%K+8CC8:SN9NX!;,, WXO,,&"8CC8R9NX!;, , , 31D,, ,,WXO,,AWW,B(078CC,,:SN9!YT;,, @A,&"#=),/%K+,B7(),7+,&%&8=,, , , , , , @A,3"147),)7+,0)1--,)",3+,+()+#8=,

,C7+,2().%J,*+.3,(/0",E/(D0,(,."/+[,"#/D,*+.30,"-,E"00+00%"#,(#&,(00+.)%"#,"-,+J%0)+#'+,4%*+,

.%0+,)",)7+,%#&+-%#%)+,.+(&%#45,(0,0++#,3D,'"2E(.%#4,9P;,(#&,9e;8,

, Y1.,(#(/D0%0,07"B0,)7(),7+(&/+00,.+/()%*+0,'(#,7(*+,)7+,0(2+,.(#4+,"-,%#)+.E.+)()%"#0,(0,-.++,.+/()%*+05,&+0E%)+,)7+,&%--+.+#'+0,%#,'()+4".D8,T".+"*+.5,B+,(.41+,)7(),)7+,(30+#'+,

"-,-.++,.+/()%*+0,%#,T(/(4(0D,0)+20,-."2,)7+,(30+#'+,"-,"*+.),#!$2"*+2+#),)",USE+'5,GHV,

9H")0&(2,PQQf;8,T".+,4+#+.(//D5,B+,'"#0%&+.,7"B,7+(&/+00,.+/()%*+0,%#,T(/(4(0D,-%),9".,&",#"),-%);,B%)7,.+'+#),(#(/D0+0,"-,-.++,.+/()%*+0,9+848,G(E"#%4.",PQQR;8,\+,'"#'/1&+,B%)7,(,

'"2E(.%0"#,"-,)7+,+J(2E/+0,%#,9P;$9R;,B%)7,")7+.,%#0)(#'+0,"-,(EE(.+#),IH,'"#)."/,'"2E/+2+#)0,9H"/%#0KD,(#&,H")0&(2,PQQe`,!)+/%)7+"0,-".)7'"2%#4;8,

1#2#"#-3#'(

G(E"#%4."5,A*(#"8,PQQR8,)*++,-".,.",&/01,2-,.!+,/+3&-.$4/,"5,5*++,*+6&.$7+/,4*"//(6$-89$/.$4&66'8,H7I,&%00+.)()%"#5,gGOh8,

N."015,h/+J(#&.+,(#&,6.+&,O(#&2(#8,:LLM8,S).(#4+,.+/()%*+0,"-,)7+,)7%.&,K%#&8,:&.9*&6,;&-89&8+,<+3&-.$4/,f[,:Pe$:_Q8,

AF*".0K%5,W"12D(#(8,:LLM8,!"#$%#&%'()%*+,\b$'"2E/+2+#)0,"-,+J%0)+#)%(/aE"00+00%*+,

E.+&%'()+08,A#,H8!8,C(2(#Z%,(#&,^8,^1012")",9+&08;5,=*"4++>$-8/,"5,:?;<,@A5,h27+.0)5,Th[,NOSh8,:eL$:_R8,

^++#(#5,X&B(.&8,:L_P8,W+/()%*+,'/(10+,-".2()%"#,%#,T(/(4(0D,9(#&,0"2+,.+/()+&,(#&,#"),0",

.+/()+&,/(#41(4+0;8,A#,H8,H+.(#)+(1,9+&8;5,B!+,C!$4&8",#!$4!,!9-.8,G7%'(4"[,GOS8,:fL$:ML8,

!)+/%)7+"05,I%2%).%"08,6".)7'"2%#48,S13".&%#()%"#,)7."147,!"2%#(/%F()%"#[,I+),(0,G"2E,%#,T(/(4(0D,G"#)."/,G"2E/+2+#)08,A#,C&!$+*/,>9,;&D"*&."$*+,>+,E+4!+*4!+,/9*,6+,;&-8&8+,

F5,+&%)+&,3D,G8,i"&+/")5,b8,N.1+)$SK.(3(/"*(,(#&,68,C."1%//+1J8,G/+.2"#),6+..(#&5,

6.(#'+[,H.+00+0,1#%*+.0%)(%.+0,i/(%0+,H(0'(/8,H"/%#0KD5,T(.%(,(#&,X.%',H")0&(28,PQQe8,T(/(4(0D,'"#)."/,(#&,%)0,%2E/%'()%"#0,-".,/%#41%0)%',

)7+".D8,G;<,HI8,

H")0&(25,X.%'8,PQQf8,T".+,'"#'+(/+&,E0+1&"'/+-)0,(#&,)7+,'/(10(/,)DE%#4,7DE")7+0%08,;$-89&,::f[,P:ed$P:MP8,

,

The developmental challenge of nominal recursion Ana T. Pérez-Leroux, Diane Massam, Susana Bejar, & Anny Castilla

(University of Toronto)

Syntactic recursion as a property of human language has become an object of intense discussion in linguistics and adjacent fields in the cognitive sciences (Hauser, Chomsky & Fitch, 2002; Evan & Levinson, 2009; Pinker & Jackendoff; Arsenijevic & Hinzen, 2010, etc.), but surprisingly little is know about how it is acquired by children. Roeper & Snyder 2005 noted in their analysis of parent-child conversations, that children experience difficulty with genitive recursion in comprehension, but this has yet to be directly studied in production.

Syntactically, recursion refers to the general ability: i) to embed one phrase marker inside another (Merge as a recursive operation), ii) to allow a XP to dominate another instance of itself (“The ball with pink stripes in the basket”), or iii) to immediately dominate another instance of itself (“John’s mother’s neighbour’s daughter”). While merge is universal recursion operation, we expect for the second two possibilities to require learning of specific selectional patterns that are obviously parametrized. Option (ii) requires a category X to select a category Y capable of selecting for X, whereas (iii) entails learning that a category X can select another instance of itself. We know that (iii) varies cross linguistically. German has similar genitive constructions to English, but does not allow recursion. Same category recursion, as in the genitive case above, involves a selection of functional heads by identical functional heads. We explore the possibility that case contraints, possibly multiple case constraints such as in Bejar and Massam (1999), are involved in restrictions in variation. With the degree of selectional learning involved we hypothesize that a) recursive genitives and PPs will develop gradually, since one-level embedding does not directly lead to two-level embedding, and b) that direct (same category) recursion may present specific additional difficulty to child learners.

Using an elicited production task we tested the ability to recursively embed NPs inside NPs. Children were prompted to identify an object that can only be referenced uniquely by stacking 3 NPs, using either recursive genitives (Elmo’s sister’s ball) or recursive prepositional modification (the baby with the woman with the flowers. As control measures, we tested children’s ability to conjoin 3 NPs. Participants were 46 English-speaking preschoolers and 11 adults. Children were divided into a younger (n= 25; mean age= 43 months) and older group (n= 21; mean age = 61 months). We measured mean number of NP produced for the recursion task, independent of the method of joining, the mean number of recursion strategies (including possessive ‘s, prepositional phrases, relative clauses, etc), and the mean number of coordinated NPs for the control task. Our results suggested that although both children and adults can easily produce three coordinated NP [F(2,54)= 2.75, p= .073], there was substantial development in children’s ability to complete the recursion task. Statistically significant differences were found in the number of noun phrases expressed [F(2,54)= 35.71, p< .001] and the number of mixed recursion strategies used [F(2,54)= 44.87, p< .001] by the three age groups. However, we found no difference in the developmental patterns for direct vs. indirect recursion.

References Arsenijevic, B. & W. Hinzen (2010). Recursion as a Human Universal and as a Primitive.

Biolinguistics 4.2–3: 165–173. Bejar, S. & D. Massam. (1999) 'Multiple Case Checking.' Syntax 2.2:65-79. Hauser, M., Chomsky, C., & Fitch, W. T. (2002). The faculty of language: What is it,

who has it, and how did it evolve. Science, 298, (5598), 1569–1579. Roeper, T. and Snyder, W. (2005) "Language learnability and the forms of recursion." In

A.M. Di Sciullo and R. Delmonte (eds.) UG and External Systems: Language, Brain and Computation, 155-169. Amsterdam: John Benjamins.

Sauerland, U. & H. M. Gärtner (Eds.), Interfaces + Recursion = Language? (pp. 1-29). New York: Mouton de Gruyter.

Object clitic omission in child L2 French Mihaela Pirvulescu, Ana T. Pérez-Leroux, Yves Roberge, & Nelleke Strik

(University of Toronto)

Omission of arguments such as objects is commonly observed in children’s speech, but rates of omission vary according to the language acquired and the type of acquisition. For instance, it has been shown that French-speaking children omit objects in obligatory contexts longer than English-speaking children (Pérez-Leroux et al. 2008). For bilingual acquisition it has been argued that influence from a topic-drop language such as German or Cantonese, may increase rates of object omission in a non-topic-drop language such as French or English (cf. Müller et al. 1996, Müller & Hulk 2001, Yip & Matthews 2005). However, research on children acquiring simultaneously French and English, two non-topic-drop languages, shows higher rates of omission in both languages (Belzil et al. 2007). Similarly, Anglophone children aged 3 to 8 acquiring French as a L2 show high rates of null objects in French (White 1996, Paradis & Crago 2003, Grüter 2006, Prévost 2006). Influence from L1 English, in which the use of overt object pronouns is obligatory and the object omission stage is shorter, would predict higher rates of overt object realization. Belzil et al. (2007) proposed a Default Retention Hypothesis to explain object omission in bilingual development. This account is based on the hypothesis that a null N object is uniformly available across languages as an acquisition default provided by UG (cf. Pérez-Leroux et al. 2008). It implies that the added input variability in null object contexts in bilingual acquisition results in higher retention of this default null N in both languages. The hypothesis thus predicts a delay as compared to monolingual acquisition, regardless of the languages involved, and no influence from one language onto the other.

The present study evaluates the Default Retention Hypothesis based on results from the early L2 acquisition of French by children whose L1 is English. Eleven children (mean age 4;09, mean age of first exposure to French 4;0) participated in a picture elicitation task targeting the use of object clitics. The same task was also administered in English. The results are compared with bilingual and monolingual children in the same age group as given in Pirvulescu et al. (2010); see Table 1 & 2 below. Figure 1 Figure 2

No errors in clitic placement were found, in line with earlier data in child L2 acquisition and as opposed to previous research in adult L2 acquisition (Hawkins 2001). Some children used infinitives instead of a finite verb (15% of the responses). The use of root

infinitives is common in child L2 acquisition (cf. Prévost 2004) and in this, the L2 children differ from the English-French bilingual children.

In conclusion, the L2 children produced the same structures in French as monolingual and simultaneous bilingual children, but the course of their development is protracted. L2 children retain the default hypothesis for a longer time, in both their L1 and their L2, compared to balanced bilinguals, French-dominant bilinguals and monolinguals of approximately the same age. Interestingly, the relative degree of bilingualism of a young child appears to be correlated with the rate of object omission. Less exposure to French means more influence of the Default Retention as evidenced by higher rates of object omission. References Belzil, Isabelle, Ana Teresa Pérez-Leroux, Mihaela Pirvulescu, Yves Roberge and

Danielle Thomas. 2007. What is the bilingual doing with the object? Paper presentation, GALA, September 2007, Barcelona, Spain.

Grüter, Theres. 2006. Object (clitic) omission in L2 French: Mis-setting or Missing Surface Inflection? In Proceedings of the 8th Generative Approaches to Second Language Acquisition Conference (GASLA 2006), eds. M. O’Brien, C. Shea and J. Archibald, 63-71. Somerville, MA: Cascadilla Proceedings Project.

Hawkins, Roger. 2001. Second Language Syntax: A Generative Introduction. Oxford:Blackwell.

Müller, Natascha, Berthold Crysmann and Georg Kaiser. 1996. Interactions between the acquisition of French object drop and the development of the C-system. Language Acquisition 5:35–63.

Müller, Natascha and Aafke Hulk. 2001. Crosslinguistic influence in bilingual language acquisition: Italian and French as recipient languages. Bilingualism: Language and Cognition,4:1–21.

Paradis, Johanne and Martha Crago. 2003. What Can SLI Tell Us About Transfer in SLA?, In Proceedings of the 6th Generative Approaches to Second Language Acquistion Conference (GASLA 2002): L2 links, eds. J.M. Liceras, H. Zobl and H. Goodluck, 219-226. Somerville, MA: Cascadilla Press.

Pérez-Leroux, Ana Teresa, Mihaela Pirvulescu and Yves Roberge. 2008. Null Objects in Child Language: Syntax and the Lexicon. Lingua 118:370-398.

Prévost, Philippe. 2006. The phenomenon of object omission in child L2 French. Bilingualism: Language and Cognition 9(3):263-280.

White, Lydia. 1996. Clitics in L2 French. Generative Perspectives on Language Acquisition, ed. H. Clahsen, 335-368. Amsterdam: John Benjamins

Yip, Virginia and Stephen Matthews. 2005. Dual Input and Learnability: Null Objects in Cantonese-English Bilingual Children. Proceedings of the 4th International Symposium on Bilingualism, ed. J. Cohen, K.T. McAlister, K. Rolstad, and J. MacSwan, 2421-2431. Somerville, MA: Cascadilla Press.

Serial position effects in orthography-induced transfer Yasaman Rafat (University of Toronto)

Whereas recent studies have examined the effect of orthography on transfer in pronunciation (Erdner & Burnham, 2005; Rafat, 2010; Young-Scholten, 2002), most studies have focused on the effect of phoneme-sound incogruencies. This study, on the other hand, examines the role of working memory, specifically serial position/primacy recency effects at both word list and word levels on the rate of orthography-induced transfer in second language acquisition of Spanish sounds by novice English-speaking learners.

Based on the fact that primacy and recency effects shape word recall performance (e.g., Murdock, 1962; Foreit, 1976; Surprenant et at al., 1993) as well as the fact that sentence initial position (Barcroft &Van Patten, 1997) and word initial position exhibit a privilege in second language (L2) learning (e.g., Barcroft & Rott, 2010), the same effects are expected to be observed for orthography-induced transfer, where (1) the rate of orthography-induced transfer will be lower for list initial and list final positions in comparison with list medial positions and (2) a primacy effect with respect to transfer will also be observed at the word level. In other words, orthography-induced transfer will be lower in word initial position in comparison with word medial position.

40 novice English-speaking learners of Spanish were divided into four groups (4 x10).The conditions consisted of 3 different orthographic conditions and an auditory only condition. All together the stimuli consisted of 72 target words and 36 distracters. The target words either included a grapheme-to-sound correspondence that is the same (<d>-[d]) in Spanish and English or one that is different (e.g., <ll>-[j] and <v>-[b]). The stimuli were presented in 36 triplets (2 target stimuli & 1 distracter) and were controlled for positional effects at the time of learning and production. The following permutations were created for positional effects: first at learning and first at production (1*1), first at learning and third at production (1*3), second at learning and second at production (2*2), third at learning and first at production (3*1) and third at learning and first at production (3*3). Moreover, the stimuli with <ll>-[j] were controlled for position within the word (word initial vs. word medial).

A recency effect pattern with respect to position within the triplets emerged for all grapheme-tosound correspondences. For example, for <v>-[b], whereas1*1 resulted in .88 % transfer, 3*1 resulted in .63 % transfer. The findings also suggest a primacy effect within the word for <ll>-[j] in all three conditions. For example, in one of the orthographic conditions, whereas the word medial position resulted in 41 % transfer, the word initial position only resulted in .10 % transfer.

This study is important because it expands on previous literature to show that positional effects that have been implied in retrieval also modulate orthography-induced transfer. By highlighting the involvement of working memory in transfer, this study sheds light on the interplay between complex cognitive processes that underlie the acquisition of L2 pronunciation.

References Erdener, V. D., & Burnham, D. K. (2005). The role of audiovisual speech and

orthographic information in nonnative speech production. Journal of Learning Language, 55(2), 191-228.

Barcroft, J., & Rott, S. (2010). Partial Word Form Learning in the Written Mode in L2 German and Spanish. Applied Linguistics.

Barcroft, J., & VanPatten, B. (1997). Acoustic salience of grammatical forms: The effect of location, stress, and boundedness on Spanish L2 input processing. In Pérez-Leroux, A. & Glass, W. (Eds.) Contemporary perspectives on the acquisition of Spanish. Volume 2: Production, processing, and comprehension (pp. 109-122). Somerville, MA: Cascadilla Press.

Foreit, K. G. (1976). Short lived auditory memory for pitch. Perception & Psychophysics, 19, 368-370.

Murdock, B. B. Jr. (1962). The serial position effect in free recall. Journal of Experimental Psychology, 64, 482-488.

Rafat, Y. (2010, March). Orthography-induced transfer: the case of English speaking learners of Spanish. Paper presented at the Linguistic Symposium on Romance Languages, Seattle.

Surprenant A. M., Pitt M. A., Crowder R. G. (1993). Auditory recency in immediate memory. Quarterly Journal of Experimental Psychology , 46A, 193-223.

Young-Scholten, M. (2002). Orthographic input in L2 phonological development. In P. Burmeister, T. Piske & A. Rohde (Eds), An integrated view of language development: Papers in honor of Henning Wode (pp. 263-279). Trier: Wissenschaftlicher Verlag.

Resumptive pronouns in Esan Nicholas Rolle (University of Toronto)

I argue in this paper that in Esan (ISH) [Edoid, Benue-Congo, Niger-Congo: Nigeria] the co-indexed pronoun in the constructions in (1) and (2) represent resumptive pronouns. (1) Mei oi re khian giegie do igho mu 1.SG 3.SG although FUT quickly steal money carry “I was going to quickly steal money, (but...)” (2) eni awai [ni ei kpQlQ] [ni ei mQnse] DEF.PL dog REL 3.PL be big REL 3.PL be beautiful “the beautiful big dogs” In these examples, a pronoun is co-referential with an antecedent nominal in a normal matrix clause (ex. 1) or in a relative clause (ex. 2). This pronoun is invariably third person regardless of the antecedent, whose semantic contribution is not apparent for all speakers.

Within, I argue that these resumptive pronouns are true nominal arguments and not a manifestation of a predicational agreement system (i.e., an agreement marker). Evidence for this includes (1) the pronoun is not obligatory in all finite clauses which would be highly anomalous if it were a part of any verbal conjugation, (2) a contrast between a full/strong form and a reduced/weak form of a pronoun is still available, and (3) for some speakers the resumptive pronoun provides an implication of particular attention given to the co-referential nominal, suggesting topicalization. Topicalization is also suggested from the left peripheral position of the co-indexed nominal (Rizzi 1997, Ermisch 2007). I compare Esan to the related Edoid language Ivie (Emuekpere-Masagbor 1997), as well as various Romance varieties (Rizzi 1986, Roberge 1990, de Cat 2007, et al.), where similar constructions have been analyzed as agreement.

Further, I argue that these pronouns occur when a nominal is extracted from subject position (spec-IP) to a position in a higher clause (e.g. spec-TopP), leaving a trace (Chomsky 1995). This movement creates an A-bar chain between the extracted nominal and its trace (Cinque 1990). When this trace is in subject position, it is realized as a resumptive pronoun (i.e., an overt trace; Koopman & Sportiche 1986), resulting from a structural subject requirement. I formalize this requirement as a particularly strong manifestation of a principle akin to the Extended Projection Principle (EPP) (à la Chomsky 1995, Adesola 2005). This structural subject requirement is corroborated by independent evidence in the language including (1) a lack of pro-drop (2), the use of expletives and dummy subjects, (3) the presence of an impersonal subject in negative imperatives, and (4) possible raising of objects to subject position in certain causative/ existential clauses. Cross-linguistic evidence from other African languages is also considered.

Finally, I discuss the fact that resumptive pronouns in Esan are invariably third person, that is, there is feature matching with respect to number but not person. If we understand (1) these resumptive pronouns as spelling out traces of an extracted nominal, and assume (2) the “Vocabulary item that matches the most features of the node will be

inserted” under a distributed morphology (DM) approach (Halle & Marantz 1993), then this implies (3) there are no [Person] features present in the trace position left by the extracted nominal, otherwise they would be pronounced. This suggests, in Esan, extraction/movement of a nominal neutralizes any person specification, though not number. This is interesting for linguistic theory as previous studies of extraction triggering neutralization in African languages, e.g. Amayo 1975, Omoruyi 1989 for the related Edoid language Edo and Adesola 2005 for Yoruba, have shown complete neutralization (i.e., both number and person), rather than the partial neutralization shown in Esan.

References Adesola, O.P. 2005. “Pronouns and null operators -A-bar dependencies and relations in

Yoruba”. Doctoral Dissertation. Rutgers. Amayo, A. 1975. “The structure of verbal constructions in Edo (Bini)”. The Journal of

West African Languages, 10(1): 5-25. Chomsky, N. 1995. The minimalist program. Cambridge: The MIT Press. Cinque, G. 1990. Types of A-bar-dependencies. Cambridge: The MIT Press. de Cat, C. 2007. French dislocation: Interpretation, syntax, acquisition (Oxford Studies

in Theoretical Linguistics 17). Oxford: Oxford University Press. Emuekpere-Masagbor, G. 1997. “Preverbal subject markers in Ivie”. Doctoral

Dissertation. Université de Sherbrooke. Ermisch, S. 2007. Issues in the left periphery: A typological approach to topic and focus

constructions. Frankfurt am Main: Peter Lang. Halle, M. & A. Marantz. 1993. “Distributed morphology and the pieces of inflection”. In

Hale & Keyser (Eds.), The view from building 20. Cambridge: MIT Press. 111-176.

Koopman, H. & D. Sportiche. 1986. “A note on long extraction in Vata and the ECP”. Natural Language and Linguistic Theory, 4: 357-374.

Omoruyi, T. O. 1989. “Focus and question formation in Edo”. Studies in African Linguistics, 20 (3): 279-300.

Rizzi, L. 1986. “On the status of subject clitics in Romance.” In O. Jaeggle & C. Silva-Corvalan (Eds.), Studies in Romance Linguistics. Dordrecht: Foris Publications. 391-419.

Rizzi, L. 1997. “The fine structure of the left periphery”. In L. Haegeman (ed.), Elements of grammar. Dordrecht: Kluwer Academic Publishers. 281-337.

Roberge, Y. 1990. The syntactic recoverability of null arguments. Montreal: McGill and Queens University Press.

!"#$%#&'()&(*&$#&+(%#,-#*&)./(0*.()($#1#&'2(+&%)&'+.+%($)&'")'+(

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

R('5.4@&7*&7"..&+"'#/''&0"5%(..'&()+&'/##*''*'@&7"5$&5$*&$10*&5$(5&15$*-'&7"..&.*(-)&%-16&1/-&6"'5(2*'=&

Domains in Mainland Scandinavian DP: syntax and phonology Alexandra Simonenko (McGill University)

This paper investigates the role of domains in accounting for the properties of Mainland Scandinavian (MS) definite expressions. It proposes a unified account of syntactic and phonological contrasts between Norwegian (Norw.) and Swedish (Sw.) on the one hand and Danish (Dan.) on the other.

It has been proposed in Kester (1993) and Lohrmann (2008) that the so-called ‘double determination’ pattern characteristic of Norw.&Sw. modified definite expressions (Norw./Sw.: hest-en/häst-en horse-EN ‘the horse’ vs. den hvite/vita hest-en/häst-en DEN white horse-EN ‘the white horse’) involves a double DP, where the lower determiner position is realized as -EN and the higher one as DEN. This cannot be directly applied to Dan., where the two determiners do not co-occur (Dan.: den hvide hest DEN white horse ‘the white horse’ vs. *den hvide hest-en DEN white horse-EN). Julien (2005), who proposes a version of the two-domain model with -EN realizing the head no in Norw.&Sw., claims that in Dan. the two determiners realize the same head Do. This necessitates an idiosyncratic requirement for Dan. that the noun move to SpecDP when there is no modifier, even though Do

is already filled with -EN. It has passed unnoticed that the morphosyntactic contrast has a phonological

parallel. In Sw.&Norw., -EN does not belong to the phonological domain of the root with respect to pitch accentuation, root vowel lengthening and syncope. In Dan., -EN does not demonstrate any such clitic-like behaviour, as opposed to the plural morpheme. In contrast, suggesting a tighter phonological coherency, it can trigger stød (creaky-voice-like phonation) on the root. I argue that what has been identified as double DP in Norw.&Sw. should be crucially analyzed as a bi-phasal structure, where no heads the lower phase and Do the higher one. In Dan., on the other hand, the structure is mono-phasal, as no is not a phase head. The morphosyntactic contrast arises from the interaction between a feature checking mechanism common for all MS and no’s status as a phase head. I propose that no gets values for its uninterpretable ϕ-features (ϕs) from N under head-movement, whereas Do enters the derivation with valued ϕs. The latter assumption is warranted by the ‘nominal’ behaviour of DEN, as observed by Julien (2005), such as the ability to form definite expressions with modifiers in the absence of either a noun or a nominal antecendent in the previous discourse. I propose that ϕ-valuation is governed by a principle that rules out co-occurrence within the same phrase of two controllers of ϕs on a given head. Do, being a potential ϕ-controller for no under a mechanism akin to ‘feature sharing’ of Pesetsky and Torregp (2007) and thus ‘competing’ with N, cannot co-occur with no within one phrase. The morphosyntactic contrast in question then falls out on he assumption that in Dan., no is not a phase-head, and, consequently, would not be separated from Do by a phase boundary.

The proposal about a bi-phasal DP in Norw.&Sw. but a mono-phasal DP in Dan. suc-cessfully accounts for the phonological contrast. Following the proposal of Dobler et al. (2009) that spelled-out material tends to resist phonological changes, I account for the in-ability of -EN in Norw.&Sw. to induce changes on the root. Under the assumption that a phase head triggers spellout of its complement, it is expected for the root, located in the complement of no, to act as a separate phonological domain in relation to -EN. In Dan.,

where no is assumed not to be a phase head, we expect to find phonological evidence that

it is spelled out on the same cycle with its complement. This is supported by the stød data.

The paper contributes to the general study of domains by making an explicit distinction between a syntactic phase (a phase head plus its complement) and a spellout domain (head’s complement to the exclusion of the phase head itself), with representational features being sensitive to the former and the derivational processes such as spellout to the latter.

References Dobler, E., H. Newell, G. Piggott, T. Skinner, M. Sugimura, and L. deMena Travis

(2009). Narrow Syntactic Movement after Spell-Out. In Proc. of the Minimalist Approaches to Syntactic Loclity.

Julien, M. (2005). Nominal Phrases from a Scandinavian Perspective. John Benjamins Publishing Co.

Kester, E.-P. (1993). The Inflectional Properties of Scandinavian Adjectives. Studia Linguistica 47, 139–153.

Lohrmann, S. (2008). On the Structure of the Scandinavian DP. Working Papers of the SFB 732 Incremental Specification in Context 01 01, 83–102.

Pesetsky, D. and E. Torrego (2007). The Syntax of Valuation and the Interpretability of Features. In S. Karimi, V. Samiian, and W. K. Wilkins (Eds.), Phrasal and clausal architecture: Syntactic derivation and interpretation. In honor of Joseph E. Emonds, pp. 262–294. Amsterdam: John Benjamins.

!"#$%&'()*+,--,.&/,%)*$'#$+)011&('-$'.2)3411&,')5"4'1)6'#&'()&')7,%,.,%&8$#)9"'1"','.1

!"#$$#%&'#( ) *+,-+" ) #..&*,$+,% ) &. ) #, ) &,%+"+.%&,* ) /&+(- ) 0/ ) .%1-23 ) $#,2 ) (#,*1#*+. ) 4#5+)

*"#$$#%&'#()*+,-+")#,-)/+6)(#,*1#*+.)/0((06)%4+).#$+)7#%%+",)890":+%%);<<;=)>&'+)?@@ABC)D4&.)

7#7+")#&$.)%0)&,5+.%&*#%+)%4+),#%1"+)0/)%4+).2.%+$)%4#%)*05+",.)*"#$$#%&'#()*+,-+")#..&*,$+,%)&,)

>1..&#,C)E0"+).7+'&/&'#((2=)%4&.).%1-2)'0,'+,%"#%+.)0,),01,.)+,-&,*)6&%4)7#(#%#(&F+-)'0,.0,#,%.C)

G)7"+.+,%)%4+)/&,-&,*.)0/)#)'0"71.).+#"'4=)#,-)%4+,)$0-+()%4"++)-&//+"+,%)70..&:(+)*"#$$#".)&,)

H0+".$#I.)8;<<JB)!"#-1#()K+#",&,*)L(*0"&%4$)8!KLB=)&,)0"-+")%0)+5#(1#%+)%4+)70..&:(+)-+*"++)

0/)&,%+"#'%&0,):+%6++,)/+#%1"+.)&,)%4+)&,%"&'#%+)*+,-+")#..&*,$+,%).2.%+$C)

M+..+% ) 8?@@NB)7"+.+,%. ).%#%&.%&'#( )*+,+"#(&F#%&0,. )0/ )*+,-+" )#..&*,$+,% )-&.%"&:1%&0, )0/)

7#(#%#(&F+-)'0,.0,#,%O+,-&,*)>1..&#,),01,.=):#.+-)0,)#,)+P%+,.&5+)'0"71.).+#"'4)8Q#(&F,R#S=)

;<TTBC)D4+ ) .+% ) 0/ ) &,#,&$#%+= )$0"740(0*&'#((2 ) .&$7(+P ),01,. ) &. ) 4&*4(2 )5#"&#:(+= ) 2+% )M+..+% )

'#7%1"+.)%4+&").%#%&.%&'#()%+,-+,'&+.)6&%4)T)740,0(0*&'#((2O:#.+-)*+,+"#(&F#%&0,.C

G, ) %4&. ) 7#7+"= ) G ) :1&(- ) 0, )M+..+%I. )60"S ) :2 ) &,5+.%&*#%&,* ) +5+"2 ) /+#%1"+ ) 1.+- ) &, ) 4&.)

*+,+"#(&F#%&0,.)#.)#)*(0:#().2.%+$3)&,)0%4+")60"-.=)%4&.)7#7+")+5#(1#%+.)%4+).2.%+$)#.)#)640(+):2)

%#S&,*)&,%0)'0,.&-+"#%&0,)#(()%4+)&,%+"#'%&0,.):+%6++,)#(()%4+)/+#%1"+.C)G,)0"-+")%0)1,-+"%#S+)%4&.)

%#.S=)G)'0,-1'%+-)#)'0"71.).+#"'4)1.&,*)%4+)M#%&0,#()>1..&#,)90"71.)#,-)'0-+-)+#'4)60"-)/0")

+5+"2)/+#%1"+)/01,-)"+(+5#,%):2)M+..+%)8?@@NBC)

G)%4+,)$0-+()%4+)"+.1(%.)0/)%4+)'0"71.).+#"'4)1.&,*)%4"++)-&//+"+,%)70..&:(+)*"#$$#".U)%4+)

M+..+%)!"#$$#"=)%4+)E&""0")!"#$$#")#,-)%4+)H&,#"2)V+#%1"+)!"#$$#"C)D4+)M+..+%)!"#$$#")

+$7(02.)0,(2)%4+).7+'&/&')'0,.%"#&,%.),++-+-)%0)+P7(#&,)M+..+%I.)8?@@NB)-+%#&(+-)*+,+"#(&F#%&0,.C)

D4+)E&""0")!"#$$#"):1&(-.)0,)%4+)M+..+%)!"#$$#"):2)&,%"0-1'&,*)#).+%)0/)'0,.%"#&,%.)%4#%)#"+)

%4+ ) +P#'% ) 0770.&%+ ) 8W$&""0"I ) '0,.%"#&,%.= ) X",+.%1. ) #,- ) H##2+, ) ?@@NBC ) D4+ ) %4&"- ) #,- ) /&,#()

*"#$$#"=)%4+)H&,#"2)V+#%1"+)!"#$$#"=)-0+.),0%)1.+)%4+).7+'&/&')*+,+"#(&F#%&0,.)7"+.+,%+-)&,)

M+..+%=):1%)"#%4+")1.+.)$#,2).&$7(+=):&,#"2)/+#%1"+O:#.+-)'0,.%"#&,%.3)%4&.)*"#$$#")+5#(1#%+.)

+#'4)/+#%1"+)87(#'+= )$#,,+"= ),#.#(&%2= ).%"+..=)+%'CB )#% )#)*(0:#() (+5+()#,-) )#((06.)/0")'0$7(+%+)

&,%+"#'%&0,):+%6++,)#(()%4+)/+#%1"+.C

L(()%4"++)*"#$$#".)6+"+).1''+../1()&,)'0,5+"*&,*)0,)%4+)%#"*+%)*"#$$#")64+,)$0-+(+-)

&,)%4+)!KL=)6&%4)4&*4)'0""+(#%&0,.):+%6++,)%4+)!KL)01%71%)-&.%"&:1%&0,.)#,-)%4+)7"070"%&0,.)

0:.+"5+-)&,)%4+)'0"71.)8!)Y)@C<?N=)@C<Z<)#,-)@C<?T)8"+.7+'%&5+(2B=)6&%4)7[@C@@;BC)M0,+)0/)%4+)

'0""+(#%&0,. ) 6+"+ ) /01,- ) %0 ) :+ ) .&*,&/&'#,%(2 ) -&//+"+,% ) /"0$ ) +#'4 ) 0%4+"= ) 1.&,* ) V&.4+"\. ) QO

D"#,./0"$#%&0,)%+.%3)%4&.)&,-&'#%+.)%4#%)+#'4)*"#$$#")6#.)+]1#((2)#''1"#%+)&,)'0,5+"*&,*)0,)%4+)

7"07+")*"#$$#"C

D4&.)"+.1(%)&,-&'#%+.)%4#%)*"#$$#%&'#()*+,-+")#..&*,$+,%)&,)>1..&#,)&.)#)6+((O:+4#5+-)

.2.%+$=)+5+,)64+,)6+)'0,.&-+" )#(( ) /+#%1"+. ) &,) &,%+"#'%&0,)6&%4)+#'4)0%4+"C)V1"%4+"$0"+= ) %4+ )

'0$7#"#:(+)'0""+(#%&0,).'0"+.)0/)#(()%4"++)*"#$$#".)&.)"#%4+")1,+P7+'%+-C)D4+)M+..+%)!"#$$#")

#,-)%4+)E&""0")!"#$$#")#"+):0%4)4&*4(2O%#&(0"+-)*"#$$#".)6&%4)5+"2).7+'&/&')'0,.%"#&,%.)%4#%)

#"+)-+.&*,+-)%0)'0""+(#%+)+P#'%(2)6&%4)%4+)*+,+"#(&F#%&0,.)0/)%4+)'0"71.)-#%#C)D4+)H&,#"2)V+#%1"+)

!"#$$#"= )0,) %4+)0%4+" )4#,-= ) &. )'0$7"&.+-)0/ ):"0#-=),0,O.7+'&/&' )'0,.%"#&,%.= )64&'4)#"+),0%)

.7+'&/&'#((2)%#&(0"+-)%0)'#7%1"+)%4+.+)*+,+"#(&F#%&0,.C)G%)&.).1"7"&.&,*=)%4+,=)%4#%)%4+)H&,#"2)V+#%1"+)

!"#$$#" )7"+/0"$+-)#% )#, )+]1#( ) (+5+( )#,-)7"0-1'+-)'0$7#"#:(+ ) "+.1(%. )#. ) %4+ )4&*4O%#&(0"+-)

*"#$$#".C)

L$0,*)%4+)'0,'(1.&0,.)0/)%4&.).%1-2)&.)%4+)/#'%)%4#%)4&*4(2).7+'&/&+-)'0,.%"#&,%.)#"+),0%)

,++-+-)&,)^7%&$#(&%2)D4+0"+%&')*"#$$#".)7+"%#&,&,*)%0)>1..&#,)*"#$$#%&'#()*+,-+")-&.%"&:1%&0,)

0/ )7#(#%#(&F+-O+,-&,*),01,.3 )64+%4+" ) %4&. )7"070.#( )'#,):+)+P7#,-+-) %0)0%4+" )*+,-+"O:+#"&,*)

(#,*1#*+.)#,-)%0)0%4+")74+,0$+,0,)/01,-)&,),#%1"#()(#,*1#*+)&.)(+/%)/0")/1%1"+)"+.+#"'4C

:"+;1)9&.$#

H0+".$#=)_#1()8;<<JBC)V1,'%&0,#()_40,0(0*2U)V0"$#(&F&,*)%4+)G,%+"#'%&0,.)H+%6++,)L"%&'1(#%0"2)

#,-)_+"'+7%1#()`"&5+.C)`0'%0"#()`&..+"%#%&0,=)a,&5+".&%2)0/)L$.%+"-#$C

90":+%%=)!"+5&((+)8;<<;BC)!+,-+"C)9#$:"&-*+U)9#$:"&-*+)a,&5+".&%2)_"+..C

X",+.%1.=)E&"R#$)#,-)>C)b#"#(-)H##2+,)8?@@NBC)W_"+-&'%&,*)%4+)a,7"+-&'%#:(+U)G,%+"7"+%&,*)

M+1%"#(&F+-)c+*$+,%.)&,)`1%'4IC)K#,*1#*+)T<8;BU)AONJC

M+..+%=)D0"+)8?@@NBC)WD4+)L..&*,$+,%)0/)!+,-+")#,-)`+'(+,.&0,)%0)>1..&#,)M01,.)&,)c0/%)

90,.0,#,%.U)_"+-&'%#:&(&%2)#,-)>1(+)G,%+"#'%&0,IC)d01",#()0/)c(#5&')K&,*1&.%&'.);;8?BU)?JTO

N??C

>&'+=)91"%)8?@@ABC)W^7%&$&F&,*)>1..&#,)!+,-+"U)L)_"+(&$&,#"2)L,#(2.&.IC)V0"$#()L77"0#'4+.)%0)

c(#5&')K&,*1&.%&'.);NU)D4+)c01%4)9#"0(&,#)E++%&,*C)c%+5+,)V"#,S.)+%)#(C)8+-.BC)L,,)L":0"=)

EGU)E&'4&*#,)c(#5&')_1:(&'#%&0,.C)?eAO?TAC

Q#(&F,R#S=)LCLC)8;<TTBC)!"#$$#%&'4+.S&R).(05#"I)"1..S0*0)R#F2S#C)E0.'06U)>1..S&R)R#F2SC

Longueur phonologique et durée des occlusives orales et des voyelles en français d’Alsace

Myriam St-Maurice & François Poiré (University of Western Ontario)

Cette étude explore la variation attestée dans la production des occlusives orales en français régional d’Alsace (FA). Nous partons de l’hypothèse que cette variation est le résultat du contact linguistique entre deux variétés de structures grammaticales différentes, le français standard (FS) et l’alsacien de substrat alémanique. Nos résultats suggèrent qu’il existe en alsacien une corrélation de longueur phonologique et qu’elle est également active en FA, ce qui rend compte de la distribution des occlusives orales du FA et explique comment ces mêmes occlusives influencent et déterminent la variation attestée au niveau de la durée phonétique des voyelles en FA.

En FS, il est communément accepté que le facteur distinguant les occlusives orales est le voisement, permettant d’opposer /p t k / à /b d g /. Par contre, il existe dans certaines langues, dont l’alsacien, une distinction entre deux séries d’occlusives orales non voisées (et non aspirées), soit une série dite fortis et une série dite lenis. Il a été démontré que la durée de tenue de l’occlusion est un marqueur fiable de la distinction lenis (tenue plus courte) et fortis (tenue plus longue) (Willi 1996 ; Butcher 2004) et que la corrélation de durée est présente en position initiale, intervocalique et finale de mot (Kraehenmann 2001). L’importance de la longueur au niveau phonologique a été également discutée dans Fuchs (2005). Le fait qu’il existe dans l’inventaire phonémique de l’alsacien deux séries de voyelles qui se distinguent uniquement par la longueur (Philipp 1965, Jung 1983) est un indice supplémentaire du caractère distinctif de la longueur phonémique en alsacien, distinction absente en FS. Il semble de plus que la nature lenis ou fortis des occlusives orales du FA influence et détermine la variation également attestée au niveau de la longueur des voyelles adjacentes en FA. La direction de l’influence C → V est doublement motivée : d’une part, au niveau perceptuel, les résultats de Cho et al. (2001) suggèrent que l’information acoustique et perceptuelle d’une consonne s’étend à la voyelle subséquente, (rendant la qualité de celle-ci essentielle à l’identification de la consonne) et d’autre part, la qualité de la consonne (lenis ou fortis) semble jouer un rôle central dans certains phénomènes phonologiques. En coréen, par exemple, le schéma tonal d’un groupe accentuel serait déterminé par la nature lenis ou fortis de la consonne présente dans l’attaque de la syllabe initiale (Cho et al. 2001).

À partir de deux corpus audio/vidéo (6 loc. alsaciens parlant alsacien : 30 minutes tirées des Archives de l’Université de Strasbourg 2011 ; 6 loc. alsaciens parlant français : 30 minutes tirées de la chaine télévision France3 2011), codés pour le contraste de durée, nous avons identifié chaque occurrence d’occlusives orales lenis et fortis ainsi que leur contexte syllabique et leur environnement segmental immédiat. Nos résultats préliminaires confirment la présence d’une distinction basée sur la durée pour les occlusives orales, tant pour l’alsacien que pour le FA, ce qui va dans le sens de notre hypothèse. L'analyse détaillée de la distribution des voyelles longues et de leur interaction avec les occlusives fortis et lenis sera présentée et une discussion sur le système phonologique du FA s'en suivra.

Références Archives numérisées du département de dialectologie alsacienne et mosellane de

l’Université de Strasbourg. 2011. http://ala.u-strasbg.fr/index3.php?page= enquetes_sonores

Butcher, Andrew. 2004. ‘Fortis/Lenis’ revisited one more time: the aerodynamics of some oral stop contrasts in three continents. Clinical Linguistics & Phonetics 18:547-557.

Cho, Taehong, Sun-Ah Jun et Peter Ladefoged. 2001. Acoustic and aerodynamic correlates of Korean stops and fricatives. Journal of Phonetics 30(2):193-228.

France3. 2011. http://jt.france3.fr/player/regions/?id=l67a_locale Fuchs, Susanne. 2005. Articulatory correlates of the voicing contrast in alveolar

obstruent production in German. ZASPiL 41, August 2005. Jung, Edmond. 1983. Grammaire de l’alsacien : Dialecte de Strasbourg avec

indications historiques. Strasbourg: Editions Oberlin. Kraehenmann, Astrid. 2001. Swiss German stops: geminates all over the word.

Phonology 18:109-145. Philipp, Marthe. 1965. Le système phonologique du parler de Blaesheim : Étude

synchronique et diachronique. Nancy : Faculté des lettres et des sciences humaines de l’Université de Nancy.

Willi, Urs. 1996. Die segmentale Dauer als phonetischer Parameter von ‘fortis’ und ‘lenis’ bei Plosiven im Zürichdeutschen. Stuttgart: Steiner.

Acoustic measurement of cross-linguistic influence in L3 phonological acquisition Jeffery Steele (University of Toronto)

When acquiring a third language (L3), transfer may occur from both a learner’s first (L1) and second languages (L2). Previous research has revealed a number of factors that determine the nature of this cross-linguistic influence. Arguably the most cited of these is psychotypology: the greater the perceived L1-L3 or L2-L3 similarity, the greater the probability of transfer. Other recurring factors include non-native language proficiency, age at onset of L3 acquisition, and recency of L1-L2 use (see Cenoz, Hufeisen & Jessner 2001; De Angelis, 2007 for an overview).

Most L3 research to date has focussed on morphosyntax and the lexicon (e.g., Leung, 2005; Dewaele, 1998; Cenoz 2001). Indeed, relatively few studies have investigated cross-linguistic influence in L3 phonology. Accordingly, the present research seeks to contribute to our understanding of the role of psychotypology and L3 proficiency in shaping transfer in L3 phonological acquisition. This was accomplished via a study of the acquisition of voicing in French fricatives by beginner Mandarin-speaking learners who had previously acquired English. Whereas Mandarin possesses only voiceless fricatives, in both English and French, fricatives contrast for this feature (e.g. English sue /su/-zoo /zu/; French sous /su/ ‘under’-zoo /zu/ ‘zoo’). Two hypotheses were tested concerning cross-linguistic influence. First, if psychotypology plays a principal role in shaping L3 transfer, both beginner and more advanced Mandarin-speaking learners’ realization of French fricatives should resemble more closely those of their L2 English and should thus contrast for voicing. Second, as concerns L2 proficiency, more English-based transfer is predicted for learners with greater L2 English proficiency. To test these hypotheses, 9 adult learners of varying L2 English and L3 French proficiency were tested on their production of Mandarin, English, and French onset fricatives via sentence-reading tasks involving nonceword targets. The two most important acoustic parameters of fricative voicing (fundamental frequency; duration) were measured. The results supported the first hypothesis re psychotypology : for all learners, the L3 French fricative realizations were more like those of their L2 English in terms of the presence of a phonological voiceless versus voiced opposition. Moreover, for 5 of the 9 learners, there was no significant phonetic difference between their voiced L2 English and L3 French voiced fricatives. In contrast, the second hypothesis re the effect of L3 proficiency was not supported, as no significant correlation was found between L2 English proficiency and the mastery of L3 French voicing.

References Cenoz, J. (2001). The effect of linguistic distance, L2 status and age on cross-linguistic

influence in third language acquisition. In J. Cenoz, B. Hufeisen & U. Jessner (Eds.), Cross-linguistic influence in third language acquisition: Psycholinguistic perspectives (pp. 8-20). Clevedon: Multilingual Matters.

Cenoz, B. Hufeisen & U. Jessner (2001). Cross-linguistic influence in third language acquisition: Psycholinguistic perspectives. Clevedon: Multilingual Matters.

De Angelis, J. (2007). Third or additional language acquisition. Cleveland, UK: Multilingual Matters.

Hammarberg, B. (2001). Roles of L1 and L2 in L3 production and acquisition. In J.

Cenoz, B. Hufeisen & U. Jessner (Eds.), Cross-linguistic influence in third language acquisition: Psycholinguistic perspectives (pp. 21-41). Clevedon: Multilingual Matters.

Leung, I. 2005. L2 vs. L3 initial state: A comparative sstudy of the acquisition of French DPs by Vietnamese monolinguals and Cantonese-English bilinguals. Bilingualism: Language and Cognition 8(1): 39-61.

Genitive interrogative constructions are inherently discourse linked: Evidence from Iraqi Arabic

Laura Sterian (University of British Colombia)

This paper introduces a contrast which has not been investigated before between genitive interrogative constructions and discourse linked interrogative expressions. Based on the data from Iraqi Arabic, I demonstrate that genitive interrogative constructions are inherently discourse-linked and I explore the nature of this discourse-linking property.

In Iraqi Arabic, genitive constructions are realized via the construct state (Borer 1999; Fehri 1988; Shlonsky 2004), where the head noun is left-adjacent to a noun phrase. It has been observed (Pesetsky 1987; 2000) that some interrogative expressions are discourse linked, because they ask for answers in which the individuals that replace the interrogative expression are drawn from a set that is salient to both speaker and hearer. In Iraqi Arabic, the content question with the construct state in (1a) is felicitous, while the content question with bare interrogative pronoun (1b) is not. This suggests that the genitive interrogative construction in (1a) is discourse linked, while (1b) is much less natural given the context. 1. Context: a class of students is defending their theses; they all have different supervisors. The secretary of the board has to talk to each student's supervisor. a. content question with genitive interrogative construction

wu:ya: ʔusta:δ minnu: Hakat is-sikriti:ra with professor who spoke.3FS the-secretary.F 'Whose professor did the secretary talk to ?'

b. content question with bare interrogative pronoun

wu:ya: minnu: Hakat is-sikriti:ra with who spoke.3FS the-secretary.F #'Whom did the secretary talk to ?' I demonstrate that the claim that genitive interrogative constructions in Iraqi

Arabic are discourse linked is strongly supported by the following facts: (i) they behave like discourse-linked interrogative expressions with respect to superiority effects, (ii) they pattern in the same way as discourse-linked interrogative expressions with respect to the resumptive strategy and (iii) they have distributive interpretations.

Treating genitive interrogative constructions as discourse-linked may provide an explanation for why possessor extraction is the only argument extraction which allows both the gap strategy and the resumptive strategy in mono-clausal content questions in Iraqi Arabic. References Aoun, Joseph, and Lina Choueiri. 1999. Modes of interrogation. Benmamoun, E. (2000).

The feature structure of Functional Categories. Oxford University Press Borer, Hagit. 1999. Deconstructing the Construct. In Beyond Principles and Parameters.

ed. K. Johnson and I. Roberts. Dordrecht: Kluwer Publications. Pesetsky, David. 1987. Wh-in-situ: Movement and Unselective Binding. The

Representation of (In)definiteness. edited by Eric J. Reuland and Alice G. B. ter Meulen. MIT Press.

Pesetsky, David. 2000. Phrasal Movement and its Kin. MIT Press Ritter, Elizabeth. 1988. A head movement approach to construct-state noun phrases. In

Linguistics 26.6 909-929 Shlonsky, Ur. 2004. The form of Semitic Noun Phrases. In Lingua 114: 1465 – 1526. Wright, W. (1996). A Grammar of the Arabic Language, Librairie du Liban.

Predictions on markedness and feature resilience in the adaptation of loanwords Daniel Paul Stolzfus (Université Laval)

1. Research question: When a borrowed word enters into another language, the form of

the word is normally changed (adapted) in order to fit foreign elements to the phonological system of the borrowing language (L1); e.g. French ‘juge’ [ZyZ] -[dZUdZ] ‘judge’ in English. Previous research has shown that these changes are widely predictable and that they are conditioned by several principles including the Preservation Principle, which states that phoneme insertion has precedence over phoneme deletion, and the Minimality Principle, which predicts that an adaptation should be the least costly in terms of structure and processes as possible (Paradis and LaCharité 1997, among others). However, often more than one change can be considered minimal. Recently, some authors (e.g. Miao 2005; Steriade 2009, among others) have proposed that during the adaptation process of a foreign segment, the phonological features of manner of articulation are more resistant to change than are the phonological features of voice and of place of articulation. Steriade (2009) proposes that the least perceptible phonological feature will be changed. According to Miao, “[…] cross-linguistically, manner features (e.g. continuancy and nasality) are more resistant to change than place and voicing features.”

2. Hypotheses: Our hypothesis is that in loanword adaptation manner of articulation features are not more resilient to change than others. Instead, loanword adaptation eliminates marked features and results in less marked structures.

3. Objectives: The objectives of the presentation are to: a) verify the hypothesis that states that phonological features of manner of articulation (e.g. [±consonant], [±nasal], [±lateral], [±sonant], [±strident] and [±continuant]) are more resistant to modification during loanword adaptation than are place of articulation features (e.g. Labial, Dorsal, Coronal and [±anterior]) and laryngeal features (e.g. [±glottal constriction] and [±voice]); b) examine whether the phonological adaptation of problematic phonemes in loanwords results in less marked segments, i.e. determine if there is a movement towards the unmarked in phonological adaptations.

4. Results: First, our data demonstrates that manner of articulation features such as [±continuant] are just as likely to be involved in the adaptation process as are laryngeal features such as [±voice] and place of articulation features such as [±anterior]. For example, French /Z/ is ill-formed in onset in English. Adaptation options include /Z/ -[z] (change of place), /Z/ -[S] (change of voicing) and /Z/ -[dZ] (change of continuancy). The primary adaptation is /Z/ -[dZ] (e.g. French ‘juge’ [ZyZ] -[dZUdZ] ‘judge’). In Moroccan Arabic the primary adaptation of ill-formed /v/ in French loanwords is /v/ -[b] (change of continuancy), despite the availability of /v/ -[f] (change of voicing) (e.g. [savO)] -[sabUn] -*[safUn]). The feature [±continuant] is also modified at the same rate as the place feature [±anterior]. In our data, when a language had the choice between an adaptation targeting [±continuant] and [±anterior], in 180/365 cases (49.3%) the feature [±continuant] was changed as opposed to 185/365 cases (50.7%) where [±anterior] was modified. Instead of manner features being more resilient, we note that during loanword adaptation there is a movement towards the less complex. Languages overwhelmingly adapt with the goal of eliminating the complexities of the second language (L2) and change that involves

the elimination of marked features is preferred to the addition of marked features, or complexity. For instance, our data includes no case where a language adds the complexity of nasality or [+lateral] to a non-nasal or non-lateral even when the option of repair by insertion of [+nasal] or [+lateral] is available (e.g. in Lama /d/ -[z], *[n] and /b/ -[p], *[m]). This cannot be explained by a greater resilience of manner features, nor by the perceptual salience of features like [±nasal] compared to features such as [±voice] (in identification tasks [+voice] segments are rarely misidentified with [-voice] segments), but is due to the emergence of less marked features.

References Miao, Ruiqin. 2005. Loanword adaptation in Mandarin Chinese: perceptual, phonological

and sociolinguistic factors. Doctoral dissertation, Stony Brook University. Paradis, Carole and Darlene LaCharité. 1997. Preservation and minimality in loanword

adaptation. Journal of Linguistics 33: 379-430.

Distribution and Analysis of Adverbial Emphatic Reflexives in English Dennis Ryan Storoshenko (Yale University)

The Issue Beyond their use as markers of coargument reflexivity, English reflexive pronounsalsohavehavefocus-sensitive, ‘emphatic’ uses. This paper examines adverbial emphatic reflexives (K¨onig and Siemund, 2000), both exclusive (1) and additive (2).

(1) Did Mary have help getting that money? No, Mary earned that money herself. (2) Can you lend me some money? No, I am a broke myself.

Whereas prior work treats these as displaced DP-modifiers with a single origin, we claim that both the exclusive and additive reflexives are modifiers directly on the clausal spine, with differentbase-generated positions. Prior Analysis Gast (2006) provides an analysis in which (1) and (2) are derived from a single underlying form, with the reflexives adjoined to their antecedent DPs in [Spec, vP]. (1) is derived by stranding the emphatic reflexive in [Spec, vP], followed by an extra step of moving the entireVP to T. While this derives the correct string order for the emphatic reflexive, it predicts impossible instances of sentence-final quantifier stranding. Conversely, the sentence-final position of theadditive is ascribed to a ‘heavy’ shift rightward from a derived subject position. Corpus Evidence As ‘naturally-occurring’ examples are best found in a rich text, examples ofadditives and exclusives were culled from the 2009 section of the Corpus of Historical AmericanEnglish. A striking finding is that some additives are not sentence-final; rather, they can be followed by a temporal adverbial (3). This calls into question Gast’s derivation by means of a heavyshift, as these additives are not at the extreme right periphery. Turning to the exclusives, adjunctsare also found to the right of the emphatic, though a much broader range, including temporal, instrumental, and causal adjuncts (4). Gast does not predict such a contrast.

(3) And maybe I’ll try a Tooheys myself tonight. (4) ...the short brown hair he cuts himself every couple weeks with electric clippers.

Present Analysis Frey (2003) states that at the right periphery, the order of adjuncts is indicative of their relative syntactic positions. Demonstrating the interaction between emphatics and afrequency adverbial, (5) and (6) suggest that the additive has a higher position than the exclusive.

(5) I write a report myself every week. (6) I write a report every week myself.

For (5), the most natural reading is exclusive; only when the emphatic reflexive is sentence-finaldoes the additive reading emerge. Interaction with negation reinforces this idea that the additive ishigher than the exclusive. Gast himself notes that the exclusive reflexive scopes under sententialnegation, while the additive scopes over negation. A

final contrast comes in the antecedents forthe different emphatic reflexives. Additives are subject-oriented, regardless of the theta-role ofthe antecedent, whereas the exclusive emphatics have a tighter restriction to agency. Thus, thispaper argues that exclusive emphatics should be directly tied to the theta-position of the sententialsubject, bound from [Spec, vP], whereas the additive is base-generated at a higher position, above the vP domain and negation, bound from the derived subject position at [Spec, TP]. Turning backto the issue of temporal adjuncts, which could follow either the additive or the exclusive, Freydistinguishes two types of temporal adjunct: a reference-time adjunct which has a high position, and an event time positionwhich is low. This contrast is borne out in that where a temporal adjunctcan follow an additive, it is always a reference time; event times can only follow an exclusive. Conclusion This paper argues that the additive and exclusivereflexivesof English have different syntactic base positions. Both are adjoined to the clausal spine, though bound from differentpositions. This analysis better captures their interactions with other adjuncts, and their semanticstatus as propositional or event modifiers, rather than DP modifiers. References Frey, Werner. 2003. Syntactic conditions on adjunct classes. In Modifying adjuncts, ed.

EwaldLang, Claudia Maienborn, and Catherine Fabricius-Hansen, 163–209. Berlin: Mouton deGruyter.

Gast, Volker. 2006. The grammar of identity: Intensifiers and reflexives in Germanic languages. New York: Routledge.

König, Ekkehard, and Peter Siemund. 2000. Locally free self-forms, logophoricity, and intensification in English. English Language and Linguistics 4:183–204.

Les questions wh Longue Distance quez les enfants québécois Nelleke Strik (University of Toronto)

Peu est connu sur l’acquisition des questions wh en français québécois (voir De Cat 2002 pour quelques données concernant plutôt la dislocation). Cette étude se concentre sur les questions wh Longue Distance (LD), des questions wh directes contenant une subordonnée (voir (1), où le mot wh est antéposé, après avoir été extrait de la subordonnée par deux mouvements successifs).

(1) [Où tu penses [où que j’habite où ?]] Les questions LD sont rares en langage spontané, mais des données expérimentales d’enfants français en France ont montré qu’elles sont produites dès l’âge de 3 ans (Strik 2007, 2008, Jakubowicz & Strik 2008, Oiry 2008, Jakubowicz à paraître). Toutefois, elles deviennent plus productive à l’âge de 4 ans (dans Strik 2008, le taux de questions LD augmente de 45% à 84% entre 3 ans et 4 ans). En meme temps, diverses structures impliquant moins de mouvements syntaxiques sont produites à la place. Par exemple des questions à wh in situ et des questions avec un mot wh en position intermédiaire (voir (2)), mais aussi des questions wh racines ou des structures paratactiques. (2) a. Qu’est-ce que tu penses [où j’habite où ?]] (« mouvement partiel du mot wh »)

b. [Où tu penses [où j’habite où ?]] (« copie du mot wh ») Jakubowicz & Strik (2008), Strik (2008) et Jakubowicz (à paraître) justifient la présence de toutes ces structures par l’Hypothèse de la Complexité Dérivationelle, selon laquelle les structures impliquant moins de mouvement (visible) (ou un mouvement plus explicit, avec épellation de trace, dans le cas de la copie wh) sont préférées chez les (jeunes) enfants.

Une adaptation de la tâche de Strik (2008) a été utilisée avec 15 enfants francophones québécois (âge moyen 4;09, écart type 0;05), avec le but d’examiner les structures interrogatives chez les enfants québécois et de les comparer aux données des enfants français. La tâche contient 18 questions LD (6 wh/objet que/quoi, 6 wh/sujet qui et 6 wh/lieu où)

1.

Pour l’ensemble de la tâche, 135/210 (64%) questions LD ont été produites. Les réponses non LD sont majoritairement des questions wh racines. Le taux de réussite est plus bas que chez les enfants français de 4 ans, mais dans les types de réponses LD ainsi que dans leur distribution, les enfants québécois ressemblent aux enfants français (voir Tableau 1). Tableau 1 : Nombre moyen et écart type de questions LD chez les enfants québécois par type de mot wh

wh/objet wh/sujet wh/lieu Wh antéposé 2,9 (2,2) 2,3 (1,9) 2,7 (2,1) Wh intermédiaire 0,2 (0,4) 0,5 (1,3) 0,8 (1,1)

De manière descriptive, une chose qui frappe est que les questions à wh/lieu montrent un grand nombre (la moitié) de questions dans lesquelles le mot wh est suivi du complémenteur que (« Où que X a dit que…? »), construction courante en français

québécois (voir Tailleur 2010). Chez les enfants français ce type de réponse n’est pas utilisé.

En résumé, si une question LD est produite, le mot wh est antéposé dans la plupart des cas. Cependant, des questions LD à wh intermédiaire, ainsi qu’un nombre important de questions wh racines sont produites à la place. A part les structures typiquement québécoises, ces premiers résultats des enfants québécois confirment ceux des enfants français, mais semblent montrer une tendance plus forte à vouloir éviter le mouvement wh LD. La production des structures impliquant moins d’opérations syntaxiques est compatible avec l’Hypothèse de la Complexité Dérivationelle. Plus de données, seraient nécessaires pour creuser la question du développement des structures interrogatives en français québécois.

References De Cat, Cécile. 2002. French dislocation. Thèse de doctorat, University of York. Jakubowicz, Celia et Nelleke Strik. 2008. Scope-marking Strategies in the Acquisition of

Long Distance Wh-questions in French and Dutch. Language and Speech 51(1&2):101132.

Jakubowicz, Celia. À paraître. Measuring derivational complexity: New evidence from typically-developing and SLI learners of L1-French. Lingua.

Oiry 2008. L’Acquisition des Questions à Longue-Distance par des Enfants de Langue Maternelle Française. Stratégies à Dépendance Directe versus Indirecte et Questions Alternatives. Thèse de doctorat non publiée. Université de Nantes.

Strik, Nelleke. 2007. L’acquisition des phrases interrogatives chez les enfants francophones. Psychologie Française 52:27-39.

Strik, Nelleke. 2008. Syntaxe et acquisition des phrases interrogatives en français et en néerlandais: une étude contrastive. Thèse de doctorat non publiée. Université de Paris 8.

Tailleur, Sandrine. 2010. Does que equal est-ce que? The French wh-+ complementizer construction in détail. Ms. University of Toronto.

1 Un but supplémentaire était de tester une autre technique d’élicitation. Dans cette étude, uniquement les réponses résultant de la même technique que Strik (2008) sont prises en considération. Notez que ces réponses sont moins nombreuses que le nombre maximal de réponses possibles (15 enfants x 18 questions).

Is ideology enough? Building sociolinguistic identity in Nain, Labrador Jennifer Thorburn (Memorial University)

While Newfoundland English has been well studied, the English spoken in Labrador, the continental portion of the province, has received little attention from sociolinguists (e.g., Clarke 2010) even though its aboriginal communities offer the opportunity to study emerging varieties of English. The Labrador Inuit, for example, are experiencing rapid language shift as the population is becoming predominantly English speaking, with few people learning Inuttitut as their native language (Andersen and Johns 2005). Their linguistic situation allows us to examine identity construction and ideology in rapidly shifting communities, as well as contributing to the discussion of Indigenous English, a topic that has also received little attention from sociolinguists (Wolfram 1980, 1984, 1996; Miller 1996; Dannenberg and Wolfram 1998; Matsuno 1999; Wolfram and Dannenberg 1999; Wolfram and Sellers 1999; Torbert 2001; Dannenberg 2002). This paper investigates the use of a salient feature of Newfoundland English, the variable use of verbal -s in non-3sg constructions in present temporal reference, e.g., I loves it, in the Inuit community of Nain, Labrador. Residents of Nain overtly self-identify as not being Newfoundlanders; this study explores whether or not this ideology is expressed through their use (or non-use) of verbal -s. Specifically, if speakers show different linguistic and social conditioning to those observed in Newfoundland communities, they may well be using this feature as a sort of negative identity practice (cf. Bucholtz 1999); in contrast, showing similar constraints to Newfoundland English would suggest a mismatch between ideology and practice. Data from an age- and sex-stratified sample of 24 lifelong residents show a usage rate of 30.5% for verbal -s in Nain (N=1604); overall rates of usage in Newfoundland communities range from 5.6% (Petty Harbour; Van Herk et al. 2007) to 68% (Burin region; Clarke 1997). Unlike other communities in the province, this feature is not in decline in Nain; rather, a V-shaped pattern emerges, with older and younger generations showing greater usage of -s than middle-aged speakers. Interesting, women are more nonstandard than men in these generations, running counter to other communities in the province. Multivariate analysis suggests the linguistic constraints for verbal -s marking in Nain English are changing. Older speakers have a system governed by subject type, subject adjacency, and aspect; in contrast, younger speakers’ use of -s show no linguistic constraints. The data also show that verbal -s is far more frequent in the speech of residents whose first language is English, perhaps because younger speakers, who are predominantly L1 English, are also the most frequent users of -s. This paper explores the consequences of this weakening system and what impact these changes may have on speakers’ constructions of self, and whether verbal -s is a sufficient diagnostic tool for this type of study. References Andersen, Catharyn, and Alana Johns. 2005. Labrador Inuttitut: Speaking into the future.

Études/Inuit/Studies 29.1-2: 187-205. Bucholtz, Mary. 1999. “Why be normal?”: Language and identity practices in a

community of nerd girls. Language in Society 28.2: 203-22.

Clarke, Sandra. 1997. English verbal -s revisited: The evidence from Newfoundland. American Speech 72.3: 227-259.

Clarke, Sandra. 2010. Newfoundland and Labrador English. Edinburgh: Edinburgh University Press.

Dannenberg, Clare J. 2002. Sociolinguistic constructs of ethnic identity: The syntactic delineation of an American Indian English. Publication of the American Dialect Society 87. Raleigh, NC: Duke University.

Dannenberg, Clare, and Walt Wolfram. 1998. Ethnic identity and grammatical restructuring: Be(s) in Lumbee English. American Speech 73.2: 139-159.

Matsuno, Keiko. 1999. The English vowel system of the Ojibwe First Nation community of Garden River. M. A. thesis, Michigan State University.

Miller, Jason. 1996. Mixed sociolinguistic alignment and ethnic identity: R-lessness in a Native American community. M. A. thesis, North Carolina State University.

Torbert, Benjamin. 2001. Tracing Native American language history through consonant cluster reduction: The case of Lumbee English. American Speech 76.4: 361-387.

Wolfram, Walt. 1980. Dynamic dimensions of language influence: The case of American Indian English. In H. A. Giles, W. P. Robinson and P. M. Smith (eds.), Language: Social Psychological Perspectives, 377-388. Oxford/New York: Pergamon Press.

Wolfram, Walt. 1984. Unmarked tense in American Indian English. American Speech 59.1: 31-50.

Wolfram, Walt. 1996. Delineation and description in dialectology: The case of perfective I’m in Lumbee English. American Speech 71: 5-26.

Wolfram, Walt, and Clare Dannenberg. 1999. Dialect identity in a tri-ethnic context: The case of Lumbee Native American English. English World-Wide 20: 179-216.

Wolfram, Walt, and Jason Sellers. 1999. Ethnolinguistic marking of past be in Lumbee Vernacular English. Journal of English Linguistics 27.2: 94-114.

Van Herk, Gerard, Becky Childs, and Jennifer Thorburn. 2007. Identity marking and affiliation in an urbanizing Newfoundland community. In Papers from the 31st Annual Meeting of the Atlantic Provinces Linguistic Association/Actes du 31e Colloque annuel de l'Association de linguistique des provinces atlantiques, ed. Wladyslaw Cichocki, 85-94. Fredericton, NB.

On either in questions Lyn Tieu (University of Connecticut)

Background: There have been proposals about additive either in declarative sentences (Nathan 1999; Rullmann 2003), but less attention has been paid to either in questions. Rullmann (2003) presents some corpus examples of either in negatively biased questions, and suggests that here either is licensed pragmatically by a negative implicature. For Rullmann, either is a focus particle that in [α either] contributes the presupposition that there is at least one contextually salient proposition p∈⟦α⟧f–{⟦α⟧o} such that p is false. This predicts that so long as the presupposition is satisfied, any question with a negative implicature should license either. This paper presents an experiment designed to test this prediction, and proposes an account involving the use of VERUM to question the addition of a disjunctive implicature to the common ground. Experiment: We conducted an acceptability judgment task with 81 native speakers of English, using a series of dialogues with a 5-point Likert scale to test the acceptability of either in a variety of interrogative environments. The results suggest that satisfaction of the presupposition of either is a necessary, though not sufficient condition for acceptability of either in questions. A closer examination of certain test conditions that resulted in a bimodal distribution suggests that certain implicatures may play a crucial role in licensing either in questions, for example: (1) A: Shane is complaining that Alexia doesn’t study. B: Does Shane study either? Proposal: Some speakers are accessing a particular inference in the context that enables the either-question to give rise to a negative implicature; though this particular inference is crucial, it is a defeasible implicature, so there is no way to force all speakers to access it. On the basis of the bimodal data, we propose that either-questions involve a VERUM operator, along the lines of Romero & Han’s (2004) (R&H) treatment of (biased) negative questions. The crucial implicature in the context seems to involve a contrast between the proposition under question (Shane studies in (1)) and a salient alternative (Alexia studies), giving rise to the disjunction (P∨Q). In (1), B’s utterance questions the implicature that (P�Q); but assuming that both speakers are willing to add ¬Q (Alexia doesn’t study) to the common ground, B’s question ends up being interpreted as VERUM focus on P. We adopt R&H’s semantics for the VERUM operator, where: (i) Epix(w) is the set of worlds that conform to x’s knowledge in w; (ii) Convx(w’) is the set of worlds where all the conversational goals of x in w’ are fulfilled; (iii) CGw’’ is the Common Ground or set of propositions that the speakers assume in w’’ to be true. B’s question in (1) comes out as follows: (2) [CP [Q VERUM [IP Shane3 [VP t3 studies either]]] ⟦CP⟧(w0)=λq[q=λw.∀w∈Epix(w)[λw’’∈Convx(w’)[λw’’’.(study,(a,w’’’)∨(study(s,w’’’))∈CGw’’]]∨q=λw.¬∀w’∈Epix(w)[λw’’∈Convx(w’)[λw’’’.(study,(a,w’’’)∨(study(s,w’’’)) ∈CGw’’]]] = {“it is for sure that we should add to CG that either Alexia studies or Shane studies”,

“it is for sure that we should add to CG that either Alexia studies or Shane studies”} Assuming the speakers agree to add ¬(study,(a,w’’’)) to the CG, (2) gets reduced to: (3) … λq[ q=λw . ∀w ∈ Epix(w) [ λw’’ ∈ Convx(w’) [λw’’’ . (study(s,w’’’))∈CGw’’]] ∨ q=λw . ¬∀w’ ∈ Epix(w) [λw’’ ∈ Convx(w’)[λw’’’. (study(s,w’’’))∈CGw’’]]] = {“it is for sure that we should add to CG that Shane studies”, “it is for sure that we should add to CG that Shane studies”} This derives Rullmann’s negative implicature condition and the negative bias of either-questions. Conclusion: The results of this experiment are consistent with the view that strong NPIs such as either are sensitive to aspects of non-truth-conditional meaning in a way that weak NPIs are not (Gajewski 2010). Either in particular requires not only satisfaction of its presupposition, but also access to a disjunctive implicature, which is subsequently questioned using the VERUM operator. References Gajewski, Jon. 2010. “A note on licensing strong NPIs.” Under review. Nathan, Lance. 1999. Either: Negative polarity meets focus sensitivity. Unpublished

undergraduate honors thesis, Brown University. Romero, Maribel and Chung-Hye Han. 2004. “On negative yes/no questions.” Linguistics

and Philosophy 27:609-658. Rullmann, Hotze. 2003. “Additive particles and polarity.” Journal of Semantics 20:329-

401.

Underlying Representations of Surface Mid Tones in Siamou Carmela Toews (University of British Colombia)

In this paper, I present an analysis to account for some features of tone in Siamou (a Niger-Congo language of the Kru family). Syllables in Siamou can be described as having one of eight possible tone melodies (Thiessen, Paul, n.d.). The focus of this paper is the three tone melodies labeled M, MH, and M* (Table 1), which are all realized as level mid tones in isolation, but which may be distinguished in certain contexts (Table 2).Although there is no single context that differentiates M, MH and M*, I show three contexts which together show them to be contrastive. First of all, in the context of a noun followed by a high-tone verb, a M tone melody on the noun does not cause lowering of the H verb, while MH and M* do. Secondly, when a copular suffix follows a noun with a MH tone melody, it has a HM contour but when it follows a noun with a M or M* tone melody it has a mid tone. Finally, verbs with a M* tone melody are realized as low tone in the perfective while M and MH verbs are both realized as mid tone.

I develop a set of underlying representations for these tone melodies using two binary features, +/- Upper and +/- Raised (Yip, 2002), assuming that contour tones are composed of sequences of level tones (Leben, 1973). I argue that all three tonal melodies that surface as mid-tone contain a [-U, +R] tone which is linked to the tone-bearing unit, but that MH and M* have an additional element. MH has an unlinked [+U -R] tone and M* has an unlinked [-R] tone (Table 3). The [-R] feature present in MH and M* tone melodies causes lowering of H in context 1. The [+U, -R] features of the MH tone melody are responsible for the contour created in context 2. The [-R] feature of M*causes lowering in the perfective. This study tests and demonstrates the usefulness of binary features for explaining complex patterns of tone interactions. It also contributes primary data from an understudied language. Table 1 Tone Melody Minimal Set Gloss M kpar corner MH kpar monkey M* kpar umbilical cord Table 2 M MH M* Context 1:

Noun-H Verb no lowering of H

lowering of H lowering of H

Context 2: Noun-Copula

no contour contour created

no contour

Context 3: Perfective

no change no change tone becomes L

Table 3 M MH M* [-U, +R]

| σ

[-U, +R] [+U, -R] | σ

[-U, +R] [-R] | σ

References Leben, Will, 1973. Suprasegmental Phonology. PhD Dissertation, MIT. Thiessen, Paul, n.d. Tone Paper. unpublished. Yip, Moira, 2002. Tone. (Cambridge Textbook in Linguistics. Cambridge: Cambridge

University Press.

Composés N-A et N-de-N en français du 17e au début du 20e

siècle: la corrélation

entre la productivitéet la taille de leurs familles morphologiques Elena Voskovskaia (University of Toronto)

La productivité des mots composés est un domaine de recherche très peu exploré

(Krott, Schreider & Baayen : 1999 ; Fernández-Domínguez, Diaz-Negrillo & Štekauer:2007; Fernández-Domínguez: 2009) où la notion de productivité est surtout abordée sous un angle quantitatif (Baayen 1992,1993; Baayen et Lieber 199; Baayen et Renouf 1996). Récemment, une nouvelle approche en terme de la taille de famille morphologique a été proposée (de Jong et al. 2000; del Prado et al. 2004). En analysant le lien entre la productivité des affixes et la taille de la famille morphologique des dérivés en anglais, Baayen et Hay (2002) montrent que l’augmentation de la taille d’une famile catégorielle cause la croissance de la productivité morphologique des affixes.

Le présent travail vise à vérifier une corrélation potentielle entre la productivité et la taille de la famille morphologique des composés N-A et N-de-N en français du 17e

au

début du 20e siècle. La recherche est basée sur le corpus textuel Frantext avec plus de 4000 textes (environs 150 millions mots) selon quatre périodes importantes dans l’histoire de la langue française: 16061694; 1695-1798; 1799-1872 ; 1873-1920. La liste de 39 composés N-A (arc-boutant, becc-ornu, pied-bleu) et 27 composés N-de-N (belle-de-jour, mestre decamp, pou-de-soie) a été créée à partir du Dictionnaire de Littré (1877-1878).

Deux mesures ont été utilisées pour évaluer la productivité des composés N-A et N-de-N. La mesure P au sens strict basée sur les hapax (Baayen & Lieber 1991; Baayen 1992) est calculée comme P = n1/N (où P=la productivité; n1 = le nombre d’hapax legomenon; N= le nombre total d’occurrences dans le corpus). Cette mesure évalue la productivité du type des composés comme une structure figée en démontrant le taux de croissance du vocabulaire. La mesure de la taille de la famille catégorielle de Baayen et Hay (2002) se calcule en deux étapes : a) premièrement, on mesure la taille de la famille morphologique de la forme recensée (tous les dérivés et les composés formés à partir des mots-bases (par exemple, pour le composé chou-fleur, on relève tous les membres de la famille morphologique du constituant chou (chouchou, chouchouter, chou-blanc, chou-rave, coupe-chou, etc.) et du constituant fleur (fleurir, fleurissement, fleurette, fleuriste, fleuron, défleurir, refleurir, etc.); b) deuxièmement, on utilise une formule logarithmique log(F1)+log(F 2) + log(F3) etc. où F1, F2, F3, sont les tailles de la famille morphologique des constituants. Cette mesure calcule la fréquence des constituants du composé aussi bien que la fréquence des membres de leurs familles morphologiques.

Les résultats obtenus montrent que, globalement, il existe une corrélation inverse entre le niveau de la productivité morphologique et la taille catégorielle des composés analysés où le taux de productivité le plus élevé correspond à la taille catégorielle la plus basse. Ce phénomène est clairement présenté dans le type N-A où chaque abaissement du niveau de productivité est inversement corrélé à la croissance de la taille catégorielle de leur famille morphologique (P1=0.0348 vs T1=13.19; P2=0.0051 vs T2=15.57; P3=0.0012 vs T3=17.13). En ce qui concerne le type N-de-N, les données indiquent que : a) les chiffres sur la taille catégorielle varient d’une période à l’autre (T1=3.32 ; T2=2.28 ; T3=3.08 ; T4=3.80) ; b) le patron de corrélation inverse est présenté dans les périodes 1799-1872 (T3=3.08 vs P3=0.1000) et 1606-1694 (T1=3.32 vs P1=0.0180). Cependant la

lacune de données sur le P dans les années 1695-1798 et 1873-1920, ne permet pas d’analyser le rapport entre les deux mesures pour ce type de composés pendant ces périodes. Références Baayen, H.,Lieber, R. 1991. Productivity and English derivation: a corpus-based study,

Linguistics, 29, pp.801-43. Baayen, H. 1992. Quantitative aspects of morphological productivity, Yearbook of

Morphology,Kluwer Academic Publishers, , pp.109-149. Baayen, H. 1993. On frequency, transparency and productivity, Yearbook of

Morphology, Kluwer Academic Publishers, pp.181-208. Baayen, H., Renouf A. 1996. Chronicling the Times: productive lexical innovations in an

English newspaper, Language 72: 69-96. Dal, G. 2003. Productivité morphologique : définitions et notions connexes, dans Langue

française, v.140, Cedex, Larousse, Paris, pp.3-23. De Jong, N. H., Schreuder, R. and Baayen, R. H. 2000. The morphological family size

effect andmorphology, Language and Cognitive Processes 15, 329–365. del Prado, M., Bertram, R., Haikio, T., Schreuder, R., Baayen, H. 2004. Morphological

FamilySize in a Morphologically Rich Language: The Case of Finnish Compared With Dutch and Hebrew, Journal of Experimental Psychology: Learning, Memory, and Cognition, vol. 30, No. 6, p.1271–1278.

Fernández-Domínguez, J., Diaz-Negrillo, A., Pavol Štekauer, P. 2007. How is Low Morphological Productivity measured?. Atlantis, 29.1: 29-54.

Fernández-Domínguez, J., 2009, Productivity in English Word-formation: An approach to N+N compounding, European University Studies, Peter Lang Publishing.

Hay, J., Baayen, H. 2002. Affix Productivity and Base Productivity. Paper presented at the Morphological Productivity Seminar, ESSE 6, Aug 30-Sept 3, Strasbourg.

Hay, J., 2003. Causes and consequences of word structure. New York: Routledge. Krott, A., Schreuder, R., Baayen, R. H.1999. Complex words in complex words,

Linguistics, 37-5, pp. 905-926.

Generic uptake and the actuation of historical stylistics: a case study on indeterminate reference

Jeffery Wajsberg (University of British Colombia)

This poster presentation suggests ways in which Freadman’s (2002) notion of genre “uptake” can be adapted to the actuation problem of historical sociolinguistics. The problem that actuation poses to theories of language change was first described by Weinreich, Labov, and Herzog (1968): “Why do changes in a structural feature take place at a given time, but not in other languages with the same feature, or in the same language at other times?” (102). This line of inquiry has persisted in discussions of grammaticalization and even lexicalization (Brinton and Traugott 2005: 147), with no clear solution reached. One possibility, proposed by Milroy (1992), is the “social network” model of language change: placing an emphasis on individual rather than system-based interaction and change (22), Milroy reformulates the actuation question to ask “what . . . [social] conditions favouring or preventing[a] change might have been” (21). Milroy organizes these conditions into networks of speaker relationships, which reveal certain trends about linguistic innovation and diffusion: innovators, Milroy shows, tend to have weaker ties to a community, while early adopters will have stronger ties. The study of actuation, therefore, becomes one that examines the conditions by which changing linguistic structures move from the periphery of a speech community to the centre.

Where I find myself at odds with Milroy is in his dismissal of written language as a medium for the study of actuation (1992: 45) and, I thus infer, for social network theory itself. Perhaps this view is valid for the enterprise of phonology, the focus of his book, but I will contend that it is not the case at all levels of language. To do so, I invoke the notion of uptake—not as originally formulated by Austin (1962), for speech act theory, but as modified by Freadman (2002), for genre theory. Uptake—or, as Freadmen defines it, “the local event of crossing a boundary” (43)—offers an approach to changing writing practices that recovers their intertextual history, otherwise (here, resembling Milroy’s view of language change) lost in taxonomic description. The intertextual “memory” of generic uptake thus connects communities of writers in an analogous way that social network relations connect communities of speakers.

My poster presents a case study which explores to what extent this analogy can be pursued, particularly with an eye toward the conditions of actuation as defined above and how they can be applied to literary genres. Authors of the period 1790–1830 wrote at an important juncture in the history of the English novel, at a time when the novel underwent a considerable change from the epistolary form, or novel in letters, of the eighteenth century to the classic realism of the nineteenth century, characterized by third-person omniscient narration. Jane Austen was one such author, and my project treats her fiction as a case study of this shift because her own work enacts this change of form: her early work, Lady Susan (c1794), was a first-person epistolary novella, but her first-published novel, Sense and Sensibility (c1795), was revised into a third-person narrative. Like Milroy’s theory, which places innovation before quantification (1992: 171), uptake likewise occurs “minimally” between two texts (2002: 40), so my work begins by comparing these two novels. Specifically, I evaluate Austen’s status as an “innovator” of free indirect discourse (Lips 1926) by examining in both novels the “indeterminate

reference” characteristic of this style (Fludernik 1993: 143; Vandelanotte 2004: 494), whose effect in Austen is to create “the impersonal language of the crowd” (Lynch 1998: 236). If free indirect discourse is indeed already present in epistolary fiction, as Bray (2003) argues it is, then in Milroy’s terms it would be more accurate to label Austen an “early adopter” of the style. By analyzing this instance of uptake, I thus weigh in on enduring debates about whether Austen’s politics were “safe” or “radical” (e.g. Siskin 1998; Galperin 2003)—I do so through the language of actuation, judging if her style marks her as central or peripheral in her community of writers. References Austen, Jane. Complete Works and Letters. Charlottesville, Va: InteLex Corporation,

2008. Web. Austin, John Langshaw. How to Do Things with Words. Oxford: Clarendon Press, 1962.

Print. Bray, Joe. The Epsitolary Novel: Representation of Consciousness. New York:

Routledge, 2003. Print. Brinton, Laurel J. and Elizabeth Cross Traugott. Lexicalization and Language Change.

New York: Cambridge University Press, 2005. Print. Fludernik, Monika. The Fictions of Language and the Languages of Fiction: The

Linguistic Representation of Speech and Consciousness. New York: Routledge, 1993. Print.

Freadman, Anne. “Uptake.” The Rhetoric and Ideology of Genre. Eds. Richard Coe, Lorelei Lingard, and Tatiana Teslenko. Cresskill, N.J.: Hampton Press, 2002. 39–53. Print.

Galperin, William H. The Historical Austen. Philadelphia: University of Pennsylvania Press, 2003. Print.

Lips, Marguerite. Le Style Indirect Libre. Paris: Payot, 1926. Print. Lynch, Deidre Shauna. The Economy of Character: Novels, Market Culture, and the

Business of Inner Meaning. Chicago: University of Chicago Press, 1998. Print. Milroy, James. Linguistic Variation and Change: On the Historical Sociolinguistics of

English. Cambridge, USA: Blackwell, 1992. Print. Siskin, Clifford. The Work of Writing: Literature and Social Change in Britain, 1700–

1830. Baltimore: The Johns Hopkins University Press, 1998. Print. Vandelanotte, Lieven. “Deixis and Grounding in Speech and Thought Representation.”

Journal of Pragmatics 36 (2004): 489–520. Web. Weinreich, Uriel, William Labov, and Marvin I. Herzog. “Empirical Foundations for a

Theory of Language Change.” Directions for Historical Linguistics. Eds. Winfred Philipp Lehmann and Yakov Malkiel. Austin: University of Texas Press, 1968. 97–195. Print.

!"#$%&'#$&($)#*%+,-'$(&&*$.*%/,*/%#$+0$.+01'#$2&%3$%#,&10+*+&0$+0$4#%)-0$

!

! "#! $%!&'((!'%#)*($%+',!#+)#!-)#$.'!($%#'-'/%!)/'!'00$1$'-#!2%'/%!30!*3#+!%'45'-#)(!)-,!

%26/)%'45'-#)(!$-03/5)#$3-!&+'-!,$%)5*$42)#$-4!$-135$-4!%6''1+7!835'!%#2,$'%!$-,$1)#'!

#+)#! ($%#'-'/%! '96(3$#! )132%#$1:6+3-'#$1! 6/36'/#$'%! 30! #+'! %6''1+! %$4-)(! )%! %33-! )%! #+';!

*'135'!).)$()*('!<'747=!>'$-$%1+!'#!)(7=!?@A@B7!C#+'/!%#2,$'%!%244'%#!#+)#!($%#'-'/%!)(%3!6)/%'!

#+'!)132%#$1!%$4-)(!03/!53/'!)*%#/)1#!6/36'/#$'%=!%21+!)%!5'#/$1)(!033#!%#/21#2/'!<D35)+%!'#!

)(7=!?@@EB7!

! F+'!6/'%'-#!%#2,;!$-.'%#$4)#'%!+3&!.$3()#$3-%!30!5'#/$1)(!033#!%#/21#2/'!)00'1#!%$-4('!

&3/,!/'134-$#$3-!*;!-)#$.'!%6')G'/%!30!8#)-,)/,!H'/5)-7!I+$('!$#!$%!&'((!'%#)*($%+',!#+)#!

.$3()#$3-%! 30! 6/$5)/;! %#/'%%! $56','!&3/,! /'134-$#$3-! $-! %'.'/)(! ('9$1)(! %#/'%%! ()-42)4'%!

$-1(2,$-4!J-4($%+!<K336'/!'#!)(7=!?@@?B=!D2#1+!<D3-%'())/!'#!)(7=!?@@LB=!86)-$%+!<83#3:M)/)13!

'#! )(7=! ?@@AB! )-,! H'/5)-! <M/$',/$1+=! ?@@NB=! #+'! /3('! 30! 5'#/$1)(! 033#! %#/21#2/'! $-! &3/,!

/'134-$#$3-!$%!;'#!2-G-3&-7!

F+$/#;:3-'! -)#$.'! %6')G'/%! 30! 8#)-,)/,! H'/5)-! 6)/#$1$6)#',! $-! )! %'/$'%! 30! #+/''!

'96'/$5'-#%7! F+'! 0$/%#! '96'/$5'-#!&)%! )! ('9$1)(! ,'1$%$3-! #)%G! 13-,21#',! #3! ,'#'/5$-'! $0!

,$00'/'-#!#;6'%!30!%#/'%%!'//3/%!+).'!,$00'/'-#!'00'1#%!3-!&3/,!/'134-$#$3-7!F+'!%#$52($!&'/'!

#/$%;(()*$1!53-353/6+'5$1! H'/5)-! -32-%!&+3%'! #)/4'#! %#/'%%!&)%! '$#+'/! 3-! #+'! $-$#$)(=!

6'-2(#$5)#'!3/!0$-)(!%;(()*('7!F+'!)1#2)(!%#/'%%!3-!#+'!$#'5%!&)%!%;%#'5)#$1)((;!.)/$',7!M3/!

$-%#)-1'=!$#'5%!&$#+!13//'1#!$-$#$)(!%#/'%%=!($G'!!"#$%&'!()'!O4'-'/)#3/P=!&'/'!)(%3!6/3,21',!

&$#+!6'-2(#$5)#'!)-,!0$-)(!%#/'%%7!F+$%!)((3&',!#&3!63%%$*('!#;6'%!30!%#/'%%!'//3/%Q!8#/'%%!

3-!#+'!&/3-4!033#=!)%!$-!!"#$%&'!(*'!3/!)!.$3()#$3-!30!033#!%#/21#2/'!)%!$-!!"#'!%+$()'$,F+'!

6)/#$1$6)-#%!&'/'!)%G',!#3!($%#'-!#3!#+'%'!$#'5%!)-,!#3!$-,$1)#'!)%!R2$1G(;!)-,!)%!)112/)#'(;!

)%!63%%$*('!&+'#+'/!)-!$#'5!&)%!)!H'/5)-!&3/,!3/!-3#7!!

F+'!)11'6#)-1'!/)#'%!&'/'!$-!($-'!&$#+!#+'!+;63#+'%$%!#+)#!.$3()#$3-%!30!5'#/$1)(!033#!

%#/21#2/'! )/'! &3/%'! #+)-! %#/'%%$-4! #+'! &/3-4! 033#7! S3&'.'/=! #+'! /'%63-%'! ()#'-1$'%!

$-,$1)#',! '00'1#%! 30! ($-')/! 6/31'%%$-4! )%! &'((7! 86'1$0$1)((;=! /'%63-%'%! &'/'! %(3&'/! $0! #+'!

%#/'%%!'//3/!3112//',!)#!#+'!('0#!',4'!30!#+'!&3/,!#+)-!$0!$#!3112//',!)#!#+'!/$4+#!',4'7!

F+'! %'13-,! '96'/$5'-#! &)%! )! 13-#$-232%! /'134-$#$3-!5'53/;! #)%G! 13-,21#',! #3!

,'#'/5$-'! $0! '00'1#%! 30! %#/'%%! 5$%6()1'5'-#! 1)-! )(%3! *'! 3*%'/.',! &$#+! #/$%;(()*$1! -3-:

&3/,%7!"-!#+$%!'96'/$5'-#=!*3#+!)112/)1;!%13/'%!)-,!/'%63-%'!()#'-1$'%!$-,$1)#',!'00'1#%!30!

($-')/!6/31'%%$-4!*2#!-3!$-,'6'-,'-#!'00'1#%!30!5'#/$1)(!%#/21#2/'7!

M$-)((;=! #+'! #+$/,! '96'/$5'-#! &)%! )! 433,-'%%:/)#$-4! #)%G! $-! &+$1+! 6)/#$1$6)-#%!

($%#'-',!#3!#&3!,$00'/'-#(;!%#/'%%',!.'/%$3-%!30!#/$%;(()*$1!H'/5)-!-32-%!)-,!+),!#3!,'1$,'!

&+$1+! 3-'! %32-,',!53/'! -)#2/)(7! F+'! /)#$-4%! ,'53-%#/)#'! #+)#! #+'! 13//'1#! .'/%$3-!&)%!

6/'0'//',!3.'/!)-;!$-13//'1#!.'/%$3-!)-,!#+)#!)!.'/%$3-!&$#+!%#/'%%!3-!#+'!&/3-4!033#!&)%!

6/'0'//',!3.'/!)!.'/%$3-! $-.3(.$-4!)!.$3()#$3-!30!5'#/$1)(! 033#!%#/21#2/'7!S3&'.'/=! $0!*3#+!

.'/%$3-%!$-.3(.',!)!.$3()#$3-!30!5'#/$1)(!033#!%#/21#2/'=!#+'!6)/#$1$6)-#%!&'/'!2-,'1$,',7!!

! "!)/42'!#+)#!)132%#$1:6+3-'#$1!6/36'/#$'%!30!&3/,!%#/'%%!)-,!'00'1#%!30!5'#/$1)(!033#!

%#/21#2/'! 13-%'12#$.'(;! 0)1#3/! $-#3! #+'! /'134-$#$3-! 6/31'%%7! 8$-1'! )132%#$1:6+3-'#$1!

6/36'/#$'%! 30! &3/,! %#/'%%! *'135'! ).)$()*('! #3! #+'! ()-42)4'! 6/31'%%3/! ')/(;! $-! ($-')/!

6/31'%%$-4=! 0)%#!/'%63-%'%!)/'!53/'!)00'1#',!*;!%#/'%%:'//3/%!3112//$-4!)#!#+'! ('0#!',4'!30!

#+'!&3/,7!"0=!+3&'.'/=!%200$1$'-#!$-03/5)#$3-!$%!).)$()*('!#3!6)/%'!#+'!5'#/$1)(!%#/21#2/'!30!

#+'!%#$52(2%=!)%!$-!%(3&'/!/'%63-%'%!)-,!300:($-'!#)%G%=!.$3()#$3-%!30!5'#/$1)(!033#!%#/21#2/'!

+).'!)!%#/3-4'/!'00'1#!#+)-!%#/'%%$-4!#+'!3#+'/!033#7!

5#(#%#0,#.$

!

K336'/=!T7=!K2#('/=!U7!V!I)('%=!>7! <?@@?B7!K3-%#/)$-#%!30! ('9$1)(! %#/'%%!3-! ('9$1)(! )11'%%! $-!

J-4($%+Q!J.$,'-1'!0/35!T)#$.'!)-,!T3-:-)#$.'!W$%#'-'/%7!W)-42)4'!)-,!86''1+=!XL=!

?@Y:??E7!

K2#('/=! U7! <?@@LB7! W'9$1)(! %#/'%%7! "-! Z$%3-$=! D7! [7! V! >'5'\=! >7! J7! <J,%7B=! F+'!S)-,*33G! 30!

86''1+!Z'/1'6#$3-!<667?]X:?E^B7!_)(,'-=!_UQ![()1G&'((!Z2*($%+$-47!

D35)+%=!`7=!I$'%'=!>7=![3/-G'%%'(:81+('%'&%G;=!"7!V!81+('%'&%G;=!_7!<?@@EB7!F+'!6/31'%%$-4!

30!H'/5)-!&3/,!%#/'%%Q!'.$,'-1'!03/!#+'!6/3%3,$1!+$'/)/1+;7!Z+3-3(34;=!?L=!A:N]7!

D3-%'())/=!I7!.)-=!_7!a3%#'/!V!U7!K2#('/!<?@@LB7!J96(3/$-4!#+'!/3('!30!('9$1)(!%#/'%%!$-!('9$1)(!

/'134-$#$3-7!F+'!b2)/#'/(;!c32/-)(!30!J96'/$5'-#)(!Z%;1+3(34;=!LEU=!?LA:?YN7!

M/$',/$1+=!K7!<?@@NB7!Z/3%3,;!)-,!%63G'-!&3/,!/'134-$#$3-Q!['+).$3/)(!)-,!J>Z!13//'()#'%7!

`-62*($%+',!,31#3/)(!,$%%'/#)#$3-=!_)9!Z()-1G!"-%#$#2#'!30!K34-$#$.'!T'2/3%1$'-1'=!

W'$6\$47!

>'$-$%1+=!J7=!c'%%'=!U7!V!_1b2''-=!c7!_7!<?@A@B7!J)/(;!2%'!30!6+3-'#$1!$-03/5)#$3-!$-!%63G'-!

&3/,!/'134-$#$3-Q!W'9$1)(!%#/'%%!,/$.'%!';'!53.'5'-#%!$55',$)#'(;7!F+'!b2)/#'/(;!

c32/-)(!30!J96'/$5'-#)(!Z%;1+3(34;=!]N=!YY?:YEN7!

83#3:M)/)13=! 87=! 8'*)%#$d-:H)((e%! T7! V! K2#('/=! U7! <?@@AB7! 8'45'-#)(! )-,! %26/)%'45'-#)(!

5$%5)#1+!$-!('9$1)(!)11'%%7!c32/-)(!30!_'53/;!)-,!W)-42)4'=!XL=!XA?:N?7!

!

Nominal Licensing via Case or Deictic Anchoring

Many Indo-European make use of morphological case-marking to identify the grammatical role of a given nominal constituent. The fact that even in English, the effects of case-marking can be observed despite the virtual absence of morphological case-marking has lead to the assumption that case-marking functions as a universal but abstract nominal licensing strategy (cf. Vergnaud 1976/2006; Rouveret & Vergnaud 1980).

In this talk, I explore nominal licensing strategies in two First Nations languages: Blackfoot (Algonquian) and Halkomelem (Salish). Both these languages present us with an interesting puzzle. On the one hand, the distribution of overt nominal phrases is not subject to restrictions comparable to those imposed by case. i) Overt DP’s do not display fixed word order (Galloway 1993; Frantz 1991). ii) Neither Halkomelem nor Blackfoot show an infinitive effect: there are no embedded clauses obligatorily lacking overt subjects (Kroeber 1999; Frantz 1991). Consequently, these languages may be described as caseless, which is consistent with the absence of morphological case in both languages. Given that tense plays an important role for case assignment, their caselessness is consistent with their tenselessness (Ritter & Wiltschko 2009).

On the other hand, however, there are some phenomena in both Blackfoot and Halkomelem, which are best described in terms of grammatical roles. Consequently, these languages appear to make use of case after all. This raises the question as to why the distribution of overt DP’s is not regulated by case.

In this talk, I develop a novel theory of case, which addresses this question. In particular, I argue that DP’s can be licensed in (at least) two ways: i) via deictic anchoring (usually via of demonstrative determiners) ii) via anchoring to a grammatical category (this is what we think of as case)

According to this proposal case-marked DP’s are to (caseless) deictic DP’s what infinitives are to matrix root clauses. Specifically, both D and INFL serve as anchoring categories. In root clauses, the event is anchored relative to the utterance situation, in deictic DP’s it is the nominal referent, which is anchored to the utterance situation. In contrast, infinitive clauses are anchored relative to the embedding predicate (Enç 1987) while case-marked nominals are anchored relative to the case-assigning predicate.

This predicts that within and across languages, DP’s whose distribution is not regulated by case, must have deictic force. Conversely, if a given DP does not have deictic force, it must be licensed by Case. I show that this prediction is borne out in both Blackfoot and Halkomelem. Time permitting I will explore the cross-linguistic validity of this claim. Note that the proposal does not predict demonstrative determiners to be in complementary distribution with case. In particular, I follow Wiltschko 2009) in assuming that deictic features may be modificational, in which case they do not serve the anchoring function. References Baker, M. C. (1996). The Polysynthesis Parameter. Oxford University Press, New York. Enç, M. 1987. Anchoring Conditions for Tense. Linguistic Inquiry 18:633-657.Kiss 1995 Frantz, D. G. 1991. Blackfoot Grammar. Toronto: University of Toronto Press.

Galloway, B. 1993. A Grammar of Upriver Halkomelem. Berkeley, University of California Press.

Jelinek, E. and Demers, R. A. 1994. Predicate and pronominal argumentsin Straits Salish. Language, 70(4):697–736.

Kiss, K.E. 1995. (ed) Discourse Configurational Languages. Oxford University Press. Kroeber, P. D. 1999. The Salish language family: reconstructing syntax. Lincoln, NE:

University of Nebraska Press Vergnaud, J.-R. 2006. Letter to Noam Chomsky and Howard Lasnik (1976). In Robert

Freidin, and Howard Lasnik, eds. Syntax: Critical concepts in linguistics. Vol. 5. 21-34. London: Routledge.

Wiltschko, M. 2009. ‘What's in a determiner and how did it get there?’ In: J.Ghomeshi, I. Paul, M. Wiltschko (eds). Determiners: universals and variation. Amsterdam: John Benjamins. 25-66.

Wiltschko, M. 2010. ‘Varieties of INFL: TENSE, LOCATION, and PERSON.’ In: J. Craenenbroeck, (ed.) Alternatives to Cartography. Berlin, New York: Mouton de Gruyter 153-202.

La description linguistique des propriétés spatiales : les notions de la cible et du site Ki-Chan Yune (Université du Québec à Montréal)

La présente recherche a pour objectif de décrire les notions de la « cible (figure) » et du « site (ground) », qui sont considérées comme des notions-clés pour la description linguistique des propriétés spatiales.

Claude Vandeloise (1986, 1987), Ronald W. Langacker (1987), Andrée Borillo (1998) et Leonard Talmy (2003) décrivent ces deux notions comme suit : la cible est plus petite que le site; la cible est difficile à repérer, alors que le site est facile à repérer; en général, la cible est mobile, tandis que le site est immobile ou stable; et la position de la cible est une information nouvelle, alors que la position du site est une information ancienne.

Selon ces auteurs, l’expression « la bicyclette est près de la maison » est correcte parce que la cible (la bicyclette) et le site (la maison) satisfont aux notions ci-dessus. Par contre, l’expression « ? la maison est près de la bicyclette » est bizarre, car les propriétés de la cible (la maison) et du site (la bicyclette) ne correspondent pas aux notions ci-dessus.

Pourtant, les descriptions de ces linguistes sont contestables, puisqu’on peut trouver de nombreux contre-exemples dans lesquels les caractéristiques de la cible et du site ne correspondent pas à ces notions. Par exemple, dans l’expression « L’Arc de triomphe est près de mon studio », la cible (l’Arc de triomphe) est plus grande que le site (mon studio); la cible est aussi facile è repérer que le site; la cible et le site sont des entités immobiles; et la cible est plus connue que le site. Ensuite, dans l’expression « Fernando Alonso est tout près de Lewis Hamilton », dans une course de Formule 1, la cible et le site sont de même grandeur; les propriétés spatiales de la cible sont aussi connues que celles du site; la cible est aussi familière que le site; ces deux entités sont toutes deux immédiatement perceptibles, etc. Finalement, dans l’expression « Regarde le ciel près du soleil, sa couleur est magnifique », dans un contexte du coucher du soleil, la cible est statique de manière permanente, alors que le site est mobile; la cible est plus grande que le site; ces deux entités sont autant présentes, perçues et familières, etc. De plus, le site peut être non seulement une simple entité de référence, comme dans les expressions ci-dessus, mais aussi le lieu, l’origine ou la destination par rapport à la localisation de la cible, respectivement comme dans les expressions suivantes : « un escaladeur grimpe ce mur », « cet homme tombe de l’arbre » et « cet étudiant monte au deuxième étage ».

Nous croyons que la cible et le site sont respectivement des entités du thème et du rhème et que ces deux entités peuvent avoir diverses caractéristiques spatiales selon le contexte. Références Borillo, Andrée (1998). L’espace et son expression en français, collection l’essentiel

français, Éditions Ophrys, Gap/Paris, 170 p. Langacker, R.W. (1987). Foundations of Cognitive Grammar, vol. 1 : Theoretical

prerequisites, Stanford University Press, Stanford. Talmy, Leonard (2000). Toward a Cognitive Semantics, Volume I : Concept Structuring

Systems, The MIT Press, Cambridge (Massachusetts), 565 p.

Vandeloise, Claude (1986). L’espace en français, Editions du Seuil, Paris, 244 p. ---------------------- (1987). ‘La préposition À et le principe d’anticipation’, Langue

française, Vol.76, No.1, Larousse, Paris, pp.77-111.