THE PHONOLOGY OF NOUNS AND VERBS

Nouns are known to be phonologically privileged (Smith 2011). Compared to other parts of speech, they support a greater array of phonological contrasts, a larger number of underlying distinctions, more variety in surface patterns, and a greater resistance to assimilations. Smith (2011) posits a universal privilege scale Noun > Adjective > Verb, pointing out that the evidence mostly involves prosodic phenomena. Why would such generalizations hold?  In this talk, I will suggest that the answer is found in sentence stress.  A preliminary study of sentential prominence in the inaugural addresses of six U.S. presidents (Anttila, Dozat, Galbraith, and Shapiro, to appear) shows that noun and adjective stresses are loud and mechanical, whereas verb and function word stresses are soft and meaningful. Sentence stress impacts different parts of speech differently, providing a rationale for systematic differences among nouns, adjectives, and verbs in their word phonology. I will conclude by showing how this view explains several puzzles in the lexical phonology of Finnish.

It is largely accepted that in bilinguals and second-language (L2) leaners the native (L1) and L2 sounds coexist in a common phonological space and constantly co-influence each other. The most known phenomenon is the L1 influences on L2 production, called having a ‘foreign’ accent in L2 speech. Generally, it happens when an L1 sound is used to produce a similar L2 sound(s) (e.g., the Japanese /r/ is used to produce the English /r/ and /l/ sounds). Accents is the most salient feature of L2 speech and might persist despite years of L2 experience and learning. While the effects of L1 on L2 production are well established, less is known about the impact of L2 learning and use on the production of native sounds. In the first part of the talk, I will present some recent techniques that have been developed to help second-language learners to get rid of their accents and will discuss the factors that contribute to better L2 sound learning. In the second part of the talk, I will present the results of recent studies on the influence of L2 learning and use on native sound production in novice learners and experienced L2 speakers immersed into an L2-speaking country. The results suggest position-sensitive and global interactions between native and non-native vowel systems and point out to malleability of native and foreign speech in adolescences and adults.

In this talk, I will present novel Catalan data from varying types of child and adult Catalan-Spanish bilinguals whose onset of acquisition to Catalan has been by age 3.  This cross-sectional study allows us to follow the development of several linguistic structures along childhood and observe the outcomes of prolonged language immersion in adulthood. By taking into account language dominance, measured through quantity and quality of input, onset of exposure, linguistic environment, and language identity, we want to contribute to the debate on the factors that characterize the outcomes of early bilingualism (Cornips & Hulk, 2008; Meisel, 2009, 2011; Perpiñan, 2017; Tsimpli, 2014; Unsworth, 2013, Unsworth, Argyri, Cornips, Hulk, Sorace & Tsimpli, 2014).

In particular, I will focus on the acquisition of Catalan non-personal clitics in three types of adult and child bilinguals: Spanish-dominant, Catalan-dominant and Balanced Bilinguals, from two different linguistic environments (Metropolitan area of Barcelona vs. Central Catalonia). Non-personal (or so-called ‘adverbial’) Catalan clitics enand hiare obligatory in Catalan to refer to partitive, locative, or oblique arguments. I will illustrate the phenomena with the partitive clitic.

The partitive clitic in Catalan can be the direct object of a transitive verb or the subject of an unaccusative verb. Focusing on transitive verbs, I investigate the use of partitive clitics for indefinite (1a), and quantified (1b) direct objects:

(1) a. La Clara és al.lèrgica als fruits secs, per això no *(en) menja.
The Clara is allergic to fruits dry,     that’s why no EN eats.
‘Clara is allergic to nuts, that’s why she doesn’t eat them’

b. Quants jugadors té un equip de futbol? *(En) té onze.
How-many players has a team of football? EN has eleven.
‘How many players does a football team have? It has eleven.’

Spanish does not have overt non-personal clitics, and this absence can be accounted for in two ways: 1- these forms are never generated in the syntax; 2- these forms are generated in the syntax but are phonologically lost in the spellout. These theoretical assumptions make different predictions as for the acquisition of these categories: do bilinguals need to create a new functional category or do they need to morphologically fill an already existing syntactic category? More generally speaking, we are interested in the mental representation of these categories in bilinguals whose languages vary in this respect.

 The results from an oral production task and an acceptability judgment task in over 300 bilinguals (children and adults) show delayed and partial acquisition, even in bilinguals who live in predominantly Catalan speaking areas. We find quantitative and qualitative differences depending on the clitic, and certain simplification of the system in the bilingual grammar. These results corroborate what has been found in other stable bilingual societies (Spanish-Basque by Ezeizabarrena, 2012; English-Welsh by Gathercole & Thomas, 2009) and raise questions regarding critical period effects in language development, and highlight the importance of quantity and quality of input received.

Sociolinguistics with Smartphones in Minority Language Areas: ‘Stimmen’

This paper presents a tool for citizen science and sociolinguistic research: the application ‘Stimmen’ (Voices in Frisian). Citizen science is an umbrella term for projects in which the public partakes as data collectors, data processors, analysts, educators or formulators of research questions (cf. Bonney et al. 2016). While sociolinguists have made use of citizen science for a long time, recent technological developments, such as smartphones, have made it easier than ever to involve the public in our research.

‘Stimmen’ is inspired by language documentation efforts that rely on the public as collectors of speech recordings and translations, but with the inclusion of gamified components.  To give users an incentive to use the app it contains a perceptual dialectology task that guesses where the user is from (within the Netherlands). Additionally, a picture naming task is available, designed specifically to collect speech data from lesser-used and oral languages. The picture-naming task consists of 88 different pictures (without text) that must be named by the user (in the language of their choice).

‘Stimmen’ was launched in late 2017.  More than 15,000 users have provided data so far, and more than 46,000 speech recordings have been made. This paper presents results from the perceptual dialectology task as well as the picture naming task. The discussion focusses on the Frisian data, and some changes within this minority language in the Netherlands. Two main findings are discussed: the fact that stereotypical regional features in Frisian have stable isoglosses; and that, in the ‘Stimmen’-data, language loss and regional dialect levelling cannot be sufficiently teased apart. The latter finding indicates that crowd-sourced data, e.g. from citizen science projects, must be enriched with additional qualitative methods, emphasising the importance of triangulation for investigations language change. I end with some comments about the value and future outlook of citizen science for studies of language variation and change.

Since the introduction of formal experimental methods to syntactic theory (Cowart, 1997; Schütze, 1996) implemented in experimental syntax (Myers, 2009; Sprouse et al., 2016), there is an ongoing debate about what experimental methods can tell us about syntactic theory. On the one hand, informal grammaticality judgments traditionally used in theoretical syntax are the necessary starting point for a systematic reflection on linguistic phenomena (Phillips, 2010). On the other, formal methods may be necessary to give us more precise and stable tools for developing the empirical basis of theories and thus significantly contribute to establishing these theories. However, a complete switch from informal judgments to formal experiments is costly in terms of time and money.

            In this talk, I will explore a potential contribution of formal experimental data from adult participants to morphosyntactic theories by applying Sprouse’s (2016) experimental syntax framework to Russian. In Study 1, Wh-Movement vs. Scrambling, I will compare a syntactic explanation of filler-gap dependencies with an explanation from another cognitive system. In Study 2, Genitive of Negation, I will discuss preliminary data from a factorial design experiment with a large sample that investigates constraints on usage of the genitive of negation in modern Russian.

CASTL (Center for Advanced Study in Theoretical Linguistics) at UiT The Arctic University of Norway is pleased to announce the conference Baltic Linguistics: New Perspectives and Methods to be held in Tromsø on November 21-22, 2018. The main goal of the conference is to bring together researchers from around the world working on the Baltic languages within various theoretical frameworks and methodological perspectives. We welcome submissions in the following areas:

•  Syntax
• Phonology
•  Morphology
• L1/L2/Ln Acquisition
• Historical linguistics
• Experimental linguistics
• Development of language resources

Keynote speakers:

Ineta Dabašinskienė (Vytautas Magnus University)
Juris Grigorjevs (Latvian Language Institute, University of Latvia)
Axel Holvoet (Vytautas Magnus University)

Many Germanic languages use demonstrative pronouns as (what appears to be) verbal anaphora, such as German das (López & Winkler 2000 a.o.) and mainland Scandinavian det (Lødrup 1994, 2012, Houser et al. 2007, Bentzen et al. 2013 a.o.): Har du sunget? – Jeg har det ‘Have you sung? I have that’. English seems like the odd man out in Germanic in not making such free use of demonstratives as verbal anaphora; the closest equivalent to e.g. Norwegian det in English seems to be verb phrase ellipsis (‘Have you sung? I have.’).

This talk places a spotlight on data from English varieties which do appear to have verbal that-anaphora. Many speakers allow what appears to be a verbal-anaphoric use of that – but only if that is fronted: John can make curry really well – That he can. And a few dialects of English allow that to appear in post-auxiliary position: John can make curry really well. – He can that! These structures make English look like less of an odd man out among Germanic varieties; however, it turns out that they show subtle differences from each other, and from Norwegian det, in terms of syntax, semantics, and prosody. The talk aims to give an overview of how variation between the different English thats, and between English that and Norwegian det, might be accounted for – in the process trying to answer some mysterious lingering questions about the distribution and interpretation of det.

In this talk, I propose a new approach to exclamative sentences (i.e., sentences such as “What a nice city Tromsø is!”). I claim that exclamatives can denote both truth-conditional and expressive content, and that they should therefore be analyzed as multidimensional semantic objects. According to many current approaches, the descriptive content that ‘Tromsø is a nice city’ is presupposed and not asserted; the exclamative thus contributes merely an emotive stance and lacks assertive force. I will first present and discuss recent experimental evidence contradicting this view and supporting the idea that the descriptive content of exclamatives is ‘at-issue’. In particular, I will demonstrate that the often-cited infelicity of certain reactions to particular exclamation forms (e.g., strong denial in the context of wh-exclamatives) is actually a very subtle matter. I will then turn to the issue to what extent exclamatives can serve as responses to questions, and I will discuss cases that suggest that some of the response uses can in fact count as assertion speech acts.

In this talk I will first present the notion of “prominence” as an organizational principle of language (Himmelmann & Primus 2015) and then illustrate how this applies to reference. In the following the discourse functions of personal and demonstrative pronouns will be investigated. These pronouns are used to convey different discourse functions and thus shape the ‘prominence profile’ that organizes referents in discourse representation in particular ways. First, they refer to different entities available in discourse; second, they contribute to discourse progression by indicating maintenance or shifting of the prominence profile. The talk presents experimental data on the comprehension of personal pronouns and two types of demonstrative pronouns in German that illustrate how prominence profiles are dynamically constructed.

In this talk, I explore the productive ‘out-PRED’ phenomenon in English (e.g. out-sing, out-do, out-run, out-smart), in which a prefix ‘out-‘ can affix to (a sub-class of) predicates (‘PRED’s), interacting with PRED’s argument structure in a surprising way.  I draw the following novel generalizations about this phenomenon:

(1) out-PRED inherits all of PRED’s morphophonological irregularities
(2) none of PRED’s internal arguments can surface in out-PRED
(3) if the interpretation of PRED depends on its internal argument, out-PRED is impossible
(4) out-PRED can always be passivized, even when PRED cannot be

Together, these indicate that out– prefixation creates an argument structure that is syntactically distinct from that of PRED. I pursue an analysis in which out– merges with PRED before any of its internal argument(s) can merge, and the newly formed out-PRED projects its own argument structure with its own Voice-related properties such as passivizability. A core component of this analysis is that internal arguments of PRED-type predicates must be severed from the lexical predicate (i.e., introduced in a syntactic position outside of √P/VP). This adds to a growing literature that converges on the result that syntax transparently encodes Neo-Davidsonian semantic argument structure; i.e., all arguments in a predicate are introduced by individual semantic functions, which each in turn map onto individual functional projections in the syntax.