2017-2020 NSF BCS 1656133 - Effects of variation and variability in the acquisition of two dialects of Spanish. (with Alicia Avellana and Lucia Brandani, Universidad de Buenos Aires)
2016-2018 MSU HARP – Development Intramural Research Grant Effects of variation and variability in the acquisition of Rioplatense and Paraguayan Spanish spoken in Buenos Aires Co-PIs: Alicia Avellana and Lucia Brandani Universidad de Buenos Aires
NSF BCS-1656133 2017-2020
This project investigates children’s acquisition of Spanish under conditions of language contact. Language contact involving closely related varieties poses an interesting problem for language acquisition. The child is exposed to input data which may not be consistent with a single grammar, despite the fact that overall similarity between the two varieties would potentially support a single grammar analysis. The principle activities will be to build two corpora of child and adult natural speech of the two Spanish varieties and analyze children’s acquisition of three linguistic properties that exhibit variability across varieties. The corpora will be made available to researchers worldwide via the Child Language Data Exchange System. The corpus analysis will be supplemented with experimental data. Qualitative and quantitative results will shed light on the nature of the language acquisition process and the effect that variability has on it. The project tests two hypothesis: first, that the there is tension between the child’s drive to make generalizations as quickly as possible and the drive to be faithful to the input, which resists simplification. The second hypothesis is that in situations of contact children are not “choosing” a particular dialect in all of its properties, but rather, that generalizations are made on a property-by-property basis.
Pronouns can only be interpreted by combining information from many levels of representation, from low-level grammatical features (person, number, gender) to high-level discourse conditions. How do children integrate these different pieces of information? This project uses multiple experimental methods, including picture selection, Truth-value judgment, and act-out tasks to probe children’s sensitivity to discourse and grammatical features in pronoun resolution.
Light verbs are characterized by their syntactic and semantic flexibility. The claim has been made that these verbs are composed of a small number of semantic “atoms”. For example have is composed of a state, [BE], plus a prepositional element, [AT]. And get is composed of the same atoms plus a transition [CHANGE], which can either be initiated by a volitional agent or not. This project uses TVJ tasks to ask two questions: 1) Do children have access to both agentive and non-agentive interpretations of the [CHANGE] atom? and 2) How do children anchor the [BE]+[AT] state in time?
This project tries to extract morphological information from written language transcriptions combining results from automated tools (CHILDES) with judgements from informants. They want to use the knowledge of native speakers to improve the performance of the automatic systems.
This project replicates an act-out task from Munn et al. (2006) testing comprehension of plural and singular definites by Vietnamese children ages 3 to 7. Contra results from English and Spanish, this experiment shows Vietnamese children make few definiteness errors, instead struggling with number. It argues that this difference arises from how children integrate number and definiteness when the task is difficult. Children acquiring languages with definite determiners and grammatical number (English or Spanish) sacrifice definiteness in favor of number, while those acquiring languages like Vietnamese prioritize definiteness, resulting in number errors.
In many dialects of Spanish, syllable-final [s] undergoes a weakening process and therefore is sometimes reduced to an aspiration, a creaky vowel, or is completely absent. In some dialects it seems that lenition happens predominantly before particular types of consonants and in some dialects it seems to be quite widespread. Studies have also shown that speech rate impacts lenition in some dialects but not as much in others (faster speech rate=more lenition). In this project we compare quantitative and qualitative properties of [s] lenition in three dialects of Spanish: Argentinian, Paraguayan and Chilean. However, to improve the quality and comparability of the recordings, we use a reading task.
Unlike English, Spanish allows for sentences without an overt subject, as in “Voy a la tienda” (“Am going to the store”), where “I” is not pronounced. Although both overtly pronounced and null subject pronouns are grammatical, they don’t have identical distributions of use. How do children learn their distribution? In Mexican Spanish, when a subject refers to the same entity as the previous subject, pronoun omission is preferred; when the referent of a given subject is different than the previous one, an overt subject is preferred. Previous studies hypothesized that this contrast is learnable from the language input that children receive in two steps: (1) children track that overt 1/2 person pronouns (“yo”, ‘I’ and “tú,”‘you’), whose reference is unambiguous, correlate with a change in who the pronoun refers to, and null indicates the same subject as before within a stretch of discourse; (2) children then extend this knowledge to 3rd person pronouns. In this project, we ask if an even narrower learning path is possible, namely one in which children track 1st and 2nd person pronouns only in clauses embedded within other clauses. We examine a corpus of interactions between 6 Mexican mother-child pairs to determine whether it is possible to learn what the null vs. overt distinction indicates with respect to the reference of the pronominal subject using only embedded clauses, or whether children need a longer or wider discourse window.
This study focuses on children’s production of the participial forms (eat-en, giv-en) in Perfect tenses where these forms alternate with simple past forms (ate, gave). The Perfect is formed by [have + verb-en], as in “I have gone/eaten”. In modal contexts such as “he should’ve driven,” many speakers say “he should’ve drove.” Instead of the participle “driven”, they use the simple past form “drove”. This phenomenon is called participial leveling. Previous work has shown that Michigan speakers accept both forms, but it is unclear how often they produce these forms. We hypothesize that these forms are both part of the mental lexicon of adult and child speakers of English in Michigan. We designed a board game activity to elicit the modal Perfect form (where the character did silly things but should have done non-silly things), in order to compare children’s (ages 4-5) and adults’ production of the modal Perfect and simple Past forms. The results will provide us information about the use of these forms by both adults and children and will help us determine to what extent children and adults have or not the same leveling patterns.
In many Latin-American dialects of Spanish there is a process that weakens syllable-final [s] and this affects the realization of plural morphology. So, Las patas (the ducks) can have the -s produced as [s], [h] (an aspiration), or nothing at all, making it identical to the singular form “la pata”. Different dialects have different constraints on where and when lenition mostly occurs and they also vary in the overall rates of lenition. We ask how the variation in lenition in Chilean, Paraguayan and Argentinian Spanish impacts speakers use of “quantity” words such as cardinal numbers, quantifiers, and determiners in the noun phrases across dialects. We have two goals: first, we examine qualitative and quantitative properties of lenition between the different dialects in order to determine whether the described differences across dialects are also present in Child Directed Speech and in Child speech (4-5 year-olds); and second, we test if there is a correlation between the rates and type of lenition and the use of different types of quantity words by mothers and children. The idea is to determine whether lenition changes how the information about plurality is encoded in child- directed-speech and child speech.
Any conversation requires attention from the participants to both what is said and to what is considered to be shared knowledge (the common ground). Children and adults often have different strategies to determine the information that can be assumed to be in the common ground; for example, a child might say, “They ate lots of carrots,” without realizing that the person they’re talking to might not know who “they” refers to. In this study we are interested in how mothers reading to their children handle potential pronoun ambiguities. We recorded videos of mothers reading a book that we designed to a child. We aim to determine whether there is any awareness of the ambiguous pronouns, and if there is any correlation between the attentiveness of a child and the attention given by the mother to the textual ambiguities. Recent work (Hirsh-Pasek et al. 2015) suggests that the quality more than the quantity of speech children hear is important for language development and later on literacy development. However, the properties measured are mostly related to how “good” the interaction is in terms of engagement and attention. In this project we examine how the quality of interaction affects pronoun resolution. Results of this research may inform teacher training and early literacy guidelines.
The Present Perfect (“I have jumped”) competes with the Simple Past (“I jumped”) in many of its uses. Previous work suggests that the Simple Past is taking on more uses than before. For example, the Present Perfect and Simple Past are interchangeable in sentences like (“I’ve already eaten” vs. “I already ate”). However, some speakers can use both tenses interchangeably , while other speakers reject the past tense in this context. In this experiment, our goal is to determine whether Present Perfect in narratives will elicit speakers to use the perfect more often, so we can determine when the present perfect is necessary. We hypothesize that in a narrative context where the perfect is used extensively, participants will be more likely to use the perfect to retell the story than the story that uses simple past and present. The results will tell us in what contexts they find it acceptable to use the perfect instead of simple past tense and vice versa.
The disjunction ‘or’ has multiple meanings depending on its context. It can have an ‘exclusive’ meaning, where it means “one or the other, but not both”, as in “You can have coffee or tea”. It can also have an ‘inclusive’ meaning, as in “I didn’t have tea or coffee” meaning “I didn’t have tea and I didn’t have coffee”. Linguists agree that the basic meaning of ‘or’ is inclusive, and the exclusive meaning is derived from an implicature. This study will lay new ground work by offering a comprehensive corpus study that investigates the percentage of usage of inclusive and exclusive ‘or’ in child-directed speech, and what contexts lend itself to one interpretation and/or an other. The use of OR will be compared with the use of SOME, whose interpretation is also derived by an implicature (some and not all). The results will give us a better understanding of whether children receive input data that supports contexts that facilitate calculate scalar implicature and will inform us whether children’s difficulties with implicatures arise from impoverished input.
In some dialects of English, “have gotten” and “have got” have distinct meanings. “Floyd has got food” means that Floyd possesses food, whereas “Floyd has gotten food” means that Floyd has obtained food, whether or not he has it when the sentence is uttered. Additionally, changing the tense makes the “have got” sentence fail, as in “Floyd had got food.” Furthermore, “Have you got food?” and “I haven’t got food” are ungrammatical for some speakers, who disallow “have” and “got” to be split. These facts pose three main problems: why does “get” have more than one meaning in the present perfect, how can we explain the lack of tense interaction, and does the explanation change if we can split “have got” into parts? This research proposes two solutions, one for each answer to the last question. The first solution proposes that “have got” is an inseparable unit in the grammar. For the other solution, “have got” is a separable unit in the grammar. Both analyses look at the account for the possessive meaning of “have got” in past research of Pylkkänen (2002), who has proposed a silent word called the ‘Low Applicative’ which usually marks a transfer of possession, but occasionally has another meaning. This research proposes that one of the alternate meanings of the Low Applicative is a simple possessive. Lastly, these analyses both propose that part of “have got” encodes the present tense; the first solution proposes the whole unit does so, whereas the second only requires “has”.
Spanish allows for sentences without an overt subject. Null and overt pronouns are preferred in different contexts, but both are grammatical. This leaves children with the task of learning when to use which form. In Mexican Spanish, overt subject pronouns generally refer to a different person than the previous subject, while null subject pronouns are preferred when maintaining the same reference as the previous subject. Previous studies hypothesized that this contrast can be acquired if children first track null and overt pronouns in the domain of first- and second-person pronouns (“yo”, ‘I’ and “tú”, ‘you’), then use this knowledge from the 1/2-person domain to help them interpret 3rd person pronouns. However, a potential problem lies in the fact that dialects (such Paraguayan Spanish) vary in the acceptability of overt subjects in different contexts, allowing overt pronouns in many more contexts than other varieties of Spanish. Using an Argentinian corpus of Paraguayan mother-child interactions we examine the distribution of null/overt subject pronouns across 1st, 2nd, and 3rd person contexts, and also across contexts with and without a switch in pronoun reference. Our goal is to show how children who speak this dialect end up learning to interpret pronoun reference.
Children are able to form categories during word learning in order to understand and create structure. Based on the work on Schuler et. al. This project will be looking how adults and children learn category information using artificial language learning. Specifically, it will look at how word frequency and structural combination have an influence on judgements.
This project uses distributed morphology to explore the underlying syntax of case morphology in German, and in particular exploring variation and/or change in the genitive case.
The project looks at various nonstandard English variables, as well as African American nonstandard variables judgments. Participants will be presented with both sets of variables, to see how adept people are at correctly identifying African American English variables. A second group of native AAE speakers will be asked if they personally have the variables presented to them. From there the comparison of their SE/AAE performance will be evaluated in conjunction with school performance. Additionally, the results from the first group will be evaluated to see if there is an over generalization that all nonstandard variables will be assigned to AAE.
Rob is looking at lenition in Paraguayan and Argentinian Spanish and its relation to language acquisition of number.
This project focuses on how or patterns compared to other scalar implicatures, and is investigating whether it truly behaves like an implicature.