Eleni Christodulelis | Ohio State University (original) (raw)

Drafts by Eleni Christodulelis

Language Variation and Change, 2019

Some Portuguese verbs have two past participle forms- one regular, stem + -do, and the other irre... more Some Portuguese verbs have two past participle forms- one regular, stem + -do, and the other irregular, often identical to the 1SG present indicative. Per grammars, perfect auxiliaries ter/haver take regulars, irregulars appear with passive/adjectival ser/estar. To test these claims, we analyzed naturally-occurring data from Brazil (20th century) and Portugal (19th and 20th). We coded 1077 tokens from 21 verbs for ten predictors and performed mixed-effects logistic regressions in R. Irregulars appear with ter/haver 54% overall and in 68% of cases from Portugal. Our results demonstrate that past participle choice is determined by the interaction of several linguistic factors. Lexical verb is the most significant predictor of participle selection; verbs with irregular participles identical to 1SG occur in the irregular significantly more than other verbs. We conclude that the “doctrine of form-function symmetry” (cf. Poplack et al. 2013) for Portuguese participle choice as proclaimed in grammars is empirically indefensible.

Conference Presentations by Eleni Christodulelis

Definite Articles with Anthroponyms Across Varieties of Spanish: Quantitative Support of Conventi... more Definite Articles with Anthroponyms Across Varieties of Spanish:
Quantitative Support of Conventional & Conversational Implicature

Within definiteness theory, research devoted to definite determiners with anthroponyms (i.e. people’s names) has focused solely on languages that require definite determiners to accompany anthroponyms by the syntax (e.g. Modern Greek) (Matushansky 2006, 2008). Prior to Christodulelis (2014; 2016), no empirical account had been given for languages in which definite determiners optionally accompanied names (e.g. across varieties of Spanish and Portuguese). In work on Mexican Spanish, Christodulelis (2016) showed qualitatively that two types of implicature convey the array of meanings contributed by the determiner. Conventional Implicature (CI) (following Potts (2007)) conveys a non-neutral attitude on the part of the speaker toward the referent whose name is accompanied by the definite article, while conversational implicature conveys the precise type of attitude (e.g. negative, angry, hurt, contempt, etc. as in (1); or affection, love, warmth, etc. as in (2)).
The present study provides quantitative support for these theoretical claims by submitting the results of a large-scale online survey wherein participants from a variety of dialect areas in the Spanish-speaking world rated the use of a definite article with an anthroponym on a five-point Likert scale (as a proxy for a pragmatic felicity judgment) with a 1 indicating that the participant could not imagine the speaker saying such a thing and a 5 indicating that they could certainly imagine the speaker saying such a thing. Each participant randomly received one of two sets of questions, each of which paired particular utterances with particular contexts, but the name of the referent in question in a given item either appeared with the definite article or without it (no participants received both versions of a given item). Linear regression analysis of the normalized rating scores in RStudio (RStudio Team 2015) showed that items containing a definite article with a name were rated as significantly less felicitous (lower in the rating scale) in contexts conveying a “neutral” attitude on the part of the speaker toward the referent, a result which corroborates Christodulelis’s (2016) qualitative finding of article infelicity in the absence of an emotionally charged context. Through regression analysis, this portion of the study revealed that when participants were given an utterance containing a definite determiner with a name, they rated the speakers as feeling significantly stronger toward the referent (either positively or negatively so depending on the context) than in cases with bare anthroponyms.
This study’s findings corroborate the qualitative work of Christodulelis (2016), but also more broadly expand the body of work on determiners and their context-sensitive meanings. It also contributes to our understanding of conventional implicature and, in utilizing an online survey with numerical value responses, allows for the study of a low-frequency pragmatic phenomenon with ample amounts of data.

Examples

(1) Maria believes that Memo cheated on her with Lupe and says to her friend:

Vi a Memo con la Lupe.
‘I saw Memo with Lupe.’

(2) You’re at your grandma’s house and you see a picture of a child that you don’t know. Your grandma sees you looking at the picture and says:

Esa foto fue sacada hace muchos años. El Rafa ya tiene quince años.”
‘That picture was taken years ago. Rafa is already fifteen.’

References

Christodulelis, Eleni. 2014. Pragmatic constraints on definite article use with anthroponyms in Brazilian Portuguese. Ms., The Ohio State University.
Christodulelis, Eleni. 2016. Definite determiners with anthroponyms in Mexican Spanish as (Anti-) Honorifics. Paper presented at AMPRA 2016, Bloomington, Indiana.
Matushansky, Ora. 2006. Why Rose is the Rose: on the use of definite articles in proper names. In Olivier Bonami and Patricia Cabredo Hofherr (eds.), Empirical Issues in Formal Syntax and Semantics 6, 285-308. http://www.cssp.cnrs.fr/eiss6/index_en.html
---------. 2008. On the linguistic complexity of proper names. Linguistics and Philosophy 31.573-637.
Potts, Chris. 2007. Into the conventional-implicature dimension. Philosophy Compass 2.665-679.
RStudio Team. 2015. RStudio: Integrated Development for R. RStudio, Inc., Boston, MA URL http://www.rstudio.com/.

Papers by Eleni Christodulelis