The role of verb polysemy in constructional profiling
A multifactorial feature analysis of ‘give’ in the dative alternation
Karolina Krawczak
Introduction: The present study focuses on a problem in lexico-grammatical semantics and in doing so it provides an illustration of how Cognitive Linguistics – a usage-based approach to language – addresses the issues of hypothesis testing and result falsifiability. More specifically, the study employs quantitative corpus methods to investigate the interaction between the semasiological structure of a single verb and the onomasiological structuring of the dative alternation. The verbal category examined here is give in English and dać/dawać in Polish. The specific method used is known as the profile-based approach – a combination of multifactorial qualitative analysis with multivariate statistical modeling (Geeraerts et al. 1994; Gries 2003; Heylen 2005; Bresnan et al. 2006; Gries & Stefanowitsch 2006; Glynn 2007; Divjak 2010; Glynn & Fischer 2010; Glynn & Robinson 2014).
Objectives: The primary objective is to examine the relationship between morpho-syntactic variation and lexical semantic variation. More specifically, the study addresses the importance of accounting for variation in semasiological lexical structure while modeling morpho-syntactic structure. It is argued here that the polysemous nature of lexemes licensed by constructions has an impact on the choice of alternate constructions. In other words, some meanings of a given lexeme are likely to be more distinctly associated with one construction than the other. The study employs quantitative modelling to test this hypothesis.
Object of study: The alternation under investigation obtains between two dative constructions associated with the verb give in English and with its perfective and imperfective equivalents in Polish dać/dawać. The alternation is illustrated in (1) and (2):
(1) [‘give’ + RECIPIENT+ THEME]
a. She gave [Peter NP RECIP] [the keys NP THEME].
b. Dała [Piotrowi DAT NP RECIP] [klucze ACC NP THEME].
(2) [‘give’ + THEME + RECIPIENT]
a. She gave [the keys NP THEME] [to [Peter NP RECIP]PP].
b. Dała [klucze ACC NP THEME] [Piotrowi DAT NP RECIP].
The difference between the two constructions lies in their word order. In English, the construction in (1a), where the proper noun designating the recipient precedes the theme, is referred to as the double object construction. The other construction in (2a), where the order of participants is reversed, is known as the prepositional dative. In Polish, both alternates are double object constructions and case marking is used to differentiate between the two objects in the argument structure. In addition, in Polish we are, in fact, dealing with four possible constructions, as the verb slot can be instantiated by either the perfective or imperfective form of the lexical category ‘give’.
Descriptive goals: The present study has a number of descriptive goals. In most general terms, the objective is to test the findings of Bresnan et al. (2006) for the dative alternation in English. However, there are a number of important differences. Firstly and most importantly, based on the assumption that semasiological variation (polysemy) of the verb contributes crucially to the onomasiological structuring of constructions, this study includes lexical semantic contribution of the lexeme to the constructional profiling. Bresnan et al. (2006) exclude polysemy from their logistic regression model by treating it as a random variable. To make this inclusion feasible, the analysis here is limited to only one lexeme. Secondly, in order to verify the results for English, the study employs a different type of data: interactive and web-based. Finally, the analytical tools will also be applied to another language, Polish.
Hypotheses: There are three corresponding hypotheses that will be tested. Firstly, it is expected that the results obtained in Bresnan et al. (2006) will be confirmed for the new data set for English. Secondly, it is assumed that the integration of the semasiological variation of the verb into the model will improve the descriptive and predictive accuracy of the analysis. Finally, it is also hypothesized that the results will extend to Polish, thus explaining in a statistically significant and predictively accurate model the choice between the two constructions in this language.
Data & results: In this study, the data amount to over 600 occurrences of the two constructions in Polish and American English. The observations were extracted from the blog-based components of the TenTen corpus (Sketch Engine, Kilgarriff et al. 2014). The extraction was based on regular expressions and followed by manual cleaning of the data. All the contextualized examples were then manually annotated for a set of features. In addition to the variables found significant in Bresnan et al. (2006), the data were also tagged for verb sense, which is crucial here as it relates to the central claim of the paper, i.e., the impact of the verb polysemy on onomasiological choice between constructions. The findings clearly demonstrate that the semasiological variation contributed by the lexeme is an important predictor of the use of the dative constructions. This is true both for English and for Polish.
References
Bresnan, J., A. Cueni, T. Nikitina & R. H. Baayen. 2006. Predicting the dative alternation. In G. Bouma, I. Kraemer & J. Zwarts. (Eds.), Cognitive foundations of interpretation, 69-94. Amsterdam: Royal Netherlands Academy of Sciences.
Divjak, D. 2010. Structuring the lexicon: A clustered model for near-synonymy. Berlin: Mouton de Gruyter.
Geeraerts, D., S. Grondelaers & P. Bakema. 1994. The structure of lexical variation: Meaning, naming, and context. Berlin: Mouton de Gruyter.
Glynn, D. 2007. Mapping meaning: Toward a usage-based methodology in Cognitive Semantics. PhD dissertation, University of Leuven.
Glynn, D. & K. Fischer. (Eds.). 2010. Quantitative Cognitive Semantics: Corpus-driven approaches. Berlin: Mouton de Gruyter.
Glynn, D. & J. Robinson. (Eds.) 2014. Corpus Methods for Semantics. Quantitative studies in polysemy and synonymy. Amsterdam: John Benjamins.
Gries, St. Th. 2003. Multifactorial analysis in Corpus Linguistics: A study of particle placement. London: Continuum Press.
Gries, St. Th. & A. Stefanowitsch. (Eds.). 2006. Corpora in Cognitive Linguistics: Corpus-based approaches to syntax and lexis. Berlin: Mouton de Gruyter.
Heylen, K. 2005. A quantitative corpus study of German word order variation. In St. Kepser & M. Reis (Eds.), Linguistic evidence: Empirical, theoretical and computational perspectives, 241-264. Berlin: Mouton de Gruyter.
Kilgarriff, A., V. Baisa, J. Buta, M. Jakubek, V. Kov, J. Michelfeit, P. Rychly & V. Suchomel. 2014. The Sketch Engine: Ten years on. Lexicography 1: 7-36.
Kamil Malarski
Receding rhoticity in South-West England: a note from fieldwork
By default, the southwestern varieties of English English have been treated as rhotic (Cruttenden 2014, Collins and Mees 2013). Judging, however, by the pace with which rhoticity has been receding in this region (Orton 1962, Trudgill 1999), it seems reasonable to argue that this feature should be there much less spread today than it was a few decades ago. Indeed, Piercy (2011, 2012) has found a very dramatic loss of non-prevocalic /r/ in young people in Dorchester, Dorset. The aim of this project is to check whether these results will hold for a larger sample of the Dorset population, and whether they extend over to Devon and Cornwall to the west.
I am presenting here my ongoing PhD project which comprises three sets of data from the South-West: a) informal notes from the field, b) rapid anonymous surveys in supermarkets, and c) sociolinguistic interviews. Done in different contexts and across different strata of the population, this data will address the following research questions: 1. Is rhoticity receding among young speakers in Cornwall, Devon and Dorset?, 2. How fast is this process?, 3. How much do rhoticity rates differ for different socioeconomic groups?, 4. Are there significant differences between the researched counties?
WA Friday Lectures (now called WA Friday Lunch Talks) are monthly meetings with presentations of current research results or research in progress by WA faculty, staff, or PhD students. They are usually held in room 316A on Fridays at 1:30 PM. Each talk is of 30 minutes (+ 10 minutes for discussion).
The Department of the History of English has the pleasure to announce that the 6th meeting of the Advanced Old English Study Group will take place on Tuesday, December the 19th, at 6:15 p.m. in Room 229 Collegium Chemicum.
The Department of the History of English has the pleasure to announce that the 6th meeting of the HELL Reading Club this academic year will take place on Tuesday January the 9th at 4.45p.m, in Room 229 Collegium Chemicum. 
W piątek 12 stycznia 2018 w godzinach 13:15-14:45 w sali C1 Collegium Novum odbędzie się spotkanie szkoleniowe i dyskusja na temat oferty Narodowego Centrum Nauki, nowości w konkursach grantowych i kwestii związanych ze składaniem wniosków. Spotkanie poprowadzi dr hab. Wojciech Sowa, koordynator dyscyplin w dziale Nauk Humanistycznych, Społecznych i o Sztuce (HS). Jego wiedza i doświadczenie nabyte w trakcie wieloletniego kierowania pracą panelów zespołu eskpertów NCN są cenne dla pracowników i doktorantów składających wnioski grantowe. Gorąco zachęcamy do udziału!
