Posters PLIN Day 2023

Metaphorical production in L2: the impact of genre on creativity and cultural variation 

Rocío Cuberos Vicente , Barbara De Cock and Ferran Suñer Muñoz Universitat de Barcelona1 , Université catholique de Louvain (UCLouvain) 

This poster details a corpus-based study into the written production of linguistic metaphors in L2 Spanish in two different genres: narrative texts and argumentative essays. As an essential component of learners’ overall communicative competence, metaphorical competence has received much attention over the past few years especially because it has been found to favour vocabulary learning as well as grammar acquisition and pragmatic competence (Littlemore & Low, 2006). To date, language learning research has primarily focused on metaphor comprehension, while much less work has been devoted to investigating the production of metaphors in L2. These studies have observed that the ability to produce metaphors develops steadily as L2 proficiency increases (Author, 2019; Nacey, 2019), that learners, just as native speakers, tend to produce metaphors in clusters, especially at advanced stages of learning (Littlemore et al., 2014), and that these clusters serve different communicative functions at each level of competence (Littlemore et al., 2014). However, little is known about the influence of genre on L2 metaphorical production. By comparing metaphorical production across narrative and academic genres, this study seeks to reveal differences in writing performance regarding the use of metaphors that result from genre-specific language demands faced by learners. One main aim is thus to measure how overall metaphor density and metaphor clusters varies per genre. A second aim is to assess the ways in which the learners’ use of metaphors contributes to perform (different) communicative functions in the two focal genres. This research examines the deliberate and non-deliberate production of conventional, creative and transferred metaphors in narrative and argumentative essays produced by 13 advanced French-speaking learners of Spanish. The corpus was collected in a guidedlearning context at the Université catholique de Louvain. Participants were asked to write an argumentative text about workplace friendships and a narrative text about how they discovered their professional vocation. Metaphors were identified using the Metaphor Identification Procedure Vrije Universiteit (Steen et al., 2010) and deliberate metaphors by applying the Deliberate Metaphor Identification Procedure (Reijnierse et al., 2018). Guidelines of both procedures were adapted to L2 production to examine creativity and transfer (Author, 2019). Discrimination between novel and L1-based metaphors was performed by native speakers of French. Those cases where the cross-domain mapping was impossible to follow were not marked as L1-based or creative metaphors but as errors, considering that communication was lost. This analysis led us to include three post-tags to the main categories (direct, indirect and implicit metaphors): (1) Spanishbased metaphors, (2) L1-based metaphors, and (3) creative metaphors. Metaphor clusters were identified through a series of time analyses following Littlemore et al. (2014). Preliminary results offer valuable insights into how L2 learners have the ability to use metaphors to create textual cohesion and how creativity and transfer have an impact on the written production of metaphors in both genres. Also, tentative results suggest that the use of metaphors in the two genre differs not in the variety of metaphors but in the metaphorical density. Finally, pedagogical implications for L2 learning will be discussed. 

References: Author (2019) Littlemore, J., Krennmayr, T., Turner, J., & Turner, S. (2014). An Investigation into Metaphor Use at Different Levels of Second Language Writing. Applied Linguistics, 35(2), 117–144. https://doi.org/10.1093/applin/amt004 Littlemore, J., & Low, G. (2006). Metaphoric Competence, Second Language Learning, and Communicative Language Ability. Applied Linguistics, 27(2), 268–294. https://doi.org/10.1093/applin/aml004 Nacey, S. (2019). Development of L2 Metaphorical Production. In Metaphor in Foreign Language Instruction (pp. 173–198). De Gruyter Mouton. https://doi.org/10.1515/9783110630367-009 Reijnierse, W. G., Burgers, C., Krennmayr, T., & Steen, G. J. (2018). DMIP: A Method for Identifying Potentially Deliberate Metaphor in Language Use. Corpus Pragmatics, 2(2), 129–147. https://doi.org/10.1007/s41701-017-0026-7 Steen Gerard, Dorst, Aletta, Herrmann, J. Berenike, Kaal, Anna, Krennmayr, T., & Pasma, Tryntje. (2010). A Method for Linguistic Metaphor Identification. From MIP to MIPVU. John Benjamins Publishing Company. http://benjamins.com/catalog/celcr.14

Sound correspondence patterns as cross-linguistic constructions 

Anna Hagel Christian-Albrechts University of Kiel 

When speakers listen to input from a new language, they will, as with any other input, compare it to structures they have already encountered. As the speakers already have a history of language processing at that point, their constructicons are not blank slates, but offer plenty of material for comparison. Depending on the languages involved, overlaps between what is already stored and what is newly processed can be rather extensive, especially when one is dealing with structures from closely related languages or from different varieties of one and the same language. Cognates, e.g., can often be successfully decoded based on formal similarities even when speakers have not encountered them before. At the same time, however, even cognates from closely related languages can feature more or less extreme formal differences, which might or might not jeopardize their recognizability, but which will not go unnoticed, especially when the differences form systematic correspondences and occur across a wide variety of words (Hagel 2020). But how exactly do speakers handle such partial overlaps when they process input like that? In CxG, it is assumed that speakers regularly identify partial formal and/or functional overlaps between structures, generalize them and store them as partially schematic constructions. Diasystematic Construction Grammar (cf, e.g., Höder 2019), a CxG approach dedicated to language contact and multilingual knowledge organization, claims that there is no convincing argument why speakers should only do this with structures from one and the same language. Instead, DCxG suggests that all linguistic knowledge is represented and captured in one integrated system, allowing for language-specific patterns (idioconstructions) as well as for cross-linguistic generalizations (diaconstructions). Based on these assumptions, sound correspondences could be analysed as a form of cross-linguistic constructions. This poster presents a project that investigates what kind of input speakers need in order to identify and generalize such cross-linguistic phonological patterns. It explores this question based on DanishSwedish correspondences. Danish and Swedish are said to be mutually intelligible, but a number of studies have shown that speakers of Swedish have great difficulties in understanding spoken Danish due to special pronunciation features of Danish (e.g. Gooskens & van Bezooijen 2013, Delsing & Lundin Åkesson 2005). These result in a number of systematic phonological correspondences between Danish and Swedish. Furthermore, since there are several correspondences between Danish and Swedish that have certain features in common (e.g., a lenis-fortis difference in the pronunciation of all ‘voiceless’ plosives /p/, /k/, /t/), they also offer some potential for abstraction beyond single correspondences. In order to investigate the acquisition of specific correspondences as well as possible generalizations beyond them, a series of psycholinguistic experiments is conducted in which speakers of Swedish are exposed to different sets of spoken Danish words during a priming phase and afterwards tested in a translation task. The poster explains the experimental design against the background of the respective assumptions from DCxG and presents first results from the psycholinguistic experiments. 

References: Delsing, Lars-Olof & Katarina Lundin Åkesson. 2005. Håller språket ihop Norden? En forskningsrapport om ungdomars förståelse av danska, svenska och norska. Kopenhagen: Nordiska Ministerrådet. Gooskens, Charlotte & Renée van Bezooijen. 2013. Explaining Danish-Swedish asymmetric word intelligibility. An error analysis. In Charlotte Gooskens & Renée van Bezooijen (eds.), Phonetics in Europe: Perception and Production, 59–82. Frankfurt am Main: Peter Lang. Hagel, Anna. 2020. Strange sounds, familiar words. Interlingual decoding from a CxG perspective. Belgian Journal of Linguistics 34, 122–134. Höder, Steffen. 2019. Phonological schematicity in multilingual constructions: A diasystematic perspective on lexical form. Word structure 12.3, 334–352.

Interrogative constructions in Mandarin and French: A contrastive study based on web corpus data 

Hung-Hsin Hsu Université catholique de Louvain (UCLouvain) 

Interrogative construction is underexplored in contrastive studies (Coveney 2011, Curry & Chambers 2017, Axelsson 2020), especially with regard to languages other than English. The present study aims to contribute to filling this gap by examining questions in French and Mandarin. In particular, I examine question types (i.e. polar, disjunctive and whquestions) and their corresponding structures (declarative forms, subject-verb inversion, use of question particles such as French est-ce que or Mandarin ma and ne, A-not-A constructions, fragments, etc.) Relying on web corpus data, I set out to test two main hypotheses: (1) based on the theory of social economics of questions (Lavinson 2012), polar questions are the most frequent question type in the two languages; and (2) since French has a wider array wh-question structures at its disposal (Coveny 2011, Guryev 2017:1), wh-questions are to be more widespread in French than in Mandarin. In addition, I aim to map the structures available in the two languages to express the main question types. The study relies on data extracted from two monolingual web subcorpora: the 1-billionword Taiwanese subcorpora of the Chinese Web 2017 (zhTenTen17) and the 3-billion-word France subcorpora of the French Web 2017 (frTenTen17; Jakubíček et al. 2013), available on Sketch Engine (Kilgarriff et al. 2004). Interrogative constructions were automatically extracted on the basis of question marks (see Biber et al. 1999:211 and Axelsson 2020) and 500 randomly selected occurrences were coded for question type and structure in each language. Results show that the distribution of question types differs significantly in the two languages (χ2 = 19.312, df=2, p < .0001) (see Figure 1). Figure 1 Distribution of question types across languages (MD = Mandarin, FR = French) The prevalence of polar questions in the two languages corroborates the first hypothesis. In addition, in line with the second hypothesis, wh-questions are found to be more frequent in French, possibly due to the high number of different structures available to express whquestions, namely inversion, declarative forms, est-ce que particle, wh-in-situ, and fragments (see Figure 2). Results further show that in French inversion is the dominant structure in polar (e.g. Êtesvous prête?) and wh-questions (e.g. Qui étaient-ils, ces hommes?); while the use of the estce que particle remains marginal. In Mandarin, on the other hand, canonical polar questions require the use of the ma particle (71% of cases, n = 180) (e.g. Kěyǐ kǎoròu ma?). However, almost 20% (n = 47) of Mandarin polar questions in our sample are expressed as declarative sentences without any particle (e.g. Zhè zhǒng rén shuō yào pīn jīnjì?), a structure that is attested in nearly all languages (Ultan 1978, Siemund 2001). Finally, fragment questions appear to be much more frequent in French than in Mandarin (27.8% vs. 7.4%), which might also be due to the many options available in French to form fragment questions. Figure 2 Distribution of question structures across languages 

References: Axelsson, K. 2020. Questions in English and Swedish fiction texts: A study based on parallel and comparable corpus data. Languages in Contrast 20(2), 235-262. Biber, D., Johansson, S., Leech, G., Conrad, S. and Finegan, E. 1999. Longman Grammar of Spoken and Written English. Harlow: Longman. Coveney, A. 2011. L’interrogation directe. Travaux de linguistique 63(2), 112-145. Curry, N. and Chambers, A. 2017. Questions in English and French Research Articles in Linguistics: A Corpus-Based Contrastive Analysis. Corpus Pragmatics 1(4), 327-350. Guryev, A. 2017. La forme des interrogatives dans le Corpus suisse de SMS en français : étude multidimensionnelle. PhD. Dissertation. Université de la Sorbonne Nouvelle (France) & Université de Neuchâtel (Switzerland). Jakubíček, M., Kilgarriff, A., Kovář, V., Rychlý, P., & Suchomel, V. 2013. The TenTen corpus family. In 7th International Corpus Linguistics Conference CL, 125-127. Kilgarriff, A., Rychlý, P., Smrž, P., Tugwell, D. 2004. The sketch engine. Proceedings of the 11th EURALEX International Congress, 105-116. Levinson, S. 2012. Interrogative intimations: On a possible social economics of interrogatives. In Jan de Ruiter (ed.): Questions: Formal, Functional and Interactional Perspectives. New York: Cambridge University Press, 11–32. Siemund, P. 2001. Interrogative constructions. In Haspelmath, M., König, E., Oesterreicher, W., and Raible, W. (eds.): Language Typology and Language Universals. Berlin: Walter de Gruyter, 1010–1028. Ultan, R. 1978. Some general characteristics of interrogative systems. In Universals of human language, vol. 4, ed. by Joseph H. Greenberg, Stanford, Calif.: Stanford University Press, 211-248.

Verbal inflection in agrammatism: a contrastive study of German and French 

Mélanie Mottin Université catholique de Louvain (UCLouvain) 

This poster provides an overview of a research project on agrammatism in German and French. Agrammatism is defined as a syndrome associated with non-fluent aphasia speech disorder. It is characterized by (i) a breakdown of functional elements, (ii) a verbal processing impairment and, (iii) a reduction of length and complexity of syntactic structures (Thompson & Bastiaanse, 2012). Despite these attested recurrent manifestations, the greatest challenge regarding agrammatism is to account for the high symptom variability among patients but also across languages and modalities. Due to this variability, the very existence of agrammatism as a syndrome as such is a debated topic in aphasiology and linguistics (Penke, 1998, Pillon, 1987). In this context, crosslinguistic investigations of agrammatic impairment are essential (Beveridge & Bak, 2011, Thompson & Bastiaanse, 2012). So far, French and German have never been the starting point for a cross-linguistic analysis of agrammatism, although both languages show form-meaning similarities at morphosyntactic and syntactic levels. The present research aims to fill this gap by comparing different morphosyntactic constructions considered problematic in agrammatism in German and French. To this aim, seven psycholinguistic experimental tasks were designed. The poster first introduces the general research topic, with a focus on the definition of agrammatism, the crosslinguistic constructions under study, and the research questions. Second, the methodology used for this study, namely an experimental method, is explained. In this regard, special attention is given to the experimental tasks aiming to test verbal inflection, especially verbal temporal reference and conditional. These two factors have both crosslinguistic relevance and may be problematic for agrammatic people (see for instance Wenzlaff & Clahsen, 2005, Bastiaanse, 2013). The production of temporal reference and conditional in verbs was tested with agrammatic people and control persons with three elicitation tasks. Then, the preliminary results of these tests comparing the production of agrammatic people in German and French are displayed. Whereas there are contrastive language-dependent tendencies as well as a high influence of inter-subject variability, similarities are also present across subjects and languages. These observations are briefly compared with existing linguistic hypotheses about agrammatism in the literature (for instance Friedmann & Grodzinsky, 1997, Wenzlaff & Clahsen, 2005, Kok, Kolk, & Haverkort, 2006, Bastiaanse, 2013). Finally, the crosslinguistic relevance of these hypotheses is evaluated. 

References: Bastiaanse, R. (2013). Why reference to the past is difficult for agrammatic speakers. Clinical Linguistics & Phonetics, 27(4), 244–263. Beveridge, M. E. L., & Bak, T. H. (2011). The languages of aphasia research: Bias and diversity. Aphasiology, 25(12), 1451–1468. Friedmann, N., & Grodzinsky, Y. (1997). Tense and agreement in agrammatic production: Pruning the syntactic tree. Brain and Language, 56, 397–425. Kok, P., Kolk, H., & Haverkort, M. (2006). Agrammatic sentence production: Is verb second impaired in Dutch? Brain and Language, 96(3), 243–254. Penke, M. (1998). Die Grammatik des Agrammatismus: Eine linguistische Untersuchung zu Wortstellung und Flexion bei Broca-Aphasie. Tübingen: Niemeyer. Pillon, A. (1987). L’agrammatisme dans tous ses états. Cahiers de Psychologie Cognitive, 7(4), 335–369. Thompson, C. K., & Bastiaanse, R. (2012). Introduction to agrammatism. In C. K. Thompson & R. Bastiaanse (Eds.), Perspectives on agrammatism (pp. 1–16). New-York/London: Psychology Press. Wenzlaff, M., & Clahsen, H. (2005). Finiteness and verb-second in German agrammatism. Brain and Language, 92(1), 33–44.

Entrenchment of semi-schematic constructions in AL-learners of Swedish 

Sarah Paetzke Christian-Albrechts University of Kiel 

We conducted a pilot study experiment to investigate the entrenchment of two Swedish semischematic constructions in AL-learners of Swedish, compared to L1-speakers. It is generally assumed in usage-based linguistics that more exposure to individual constructions leads to a higher degree of entrenchment (Schmid 2017). Hence, frequency is also assumed to be a key factor in language acquisition (Bybee 2010, Schmid 2017). Moreover, highly frequent constructions are recognized, accessed, and retrieved more easily than frequent ones (Schmid 2017: 18). Following Olofsson & Prentice (2020), we built a phrasal decision experiment in OpenSesame (Mathôt, Schreij & Theeuwes 2012; Mathôt & March 2022), in which the participants had to decide as quickly and as accurately as possible whether the stimulus expression is a legitimate way to express whether someone (1) not so clever or (2) a little crazy. We used two constructions: (1a) [SUBJ, inte, VARA DEF ADJsup SUBSTdef i SUBSTdef; 〈SUBJ is not so clever〉] (1b) X är inte den vassaste kniven i lådan ‘X is not the sharpest knife in the drawer’ (1c) X är inte den krispigaste chipset i påsen ‘X is not the crispiest chips in the bag’ (2a) [SUBJ, inte, HA alla SUBSTindef,plural i SUBSTdef; 〈SUBJ is a little crazy〉] (2b) X har inte alla hästar i stallet ‘X does not have all horses in the stable’ (2c) X har inte alla cornflakes i paketet ‘X does not have all cornflakes in the packet’ These semi-schematic constructions contain slots which can be filled more or less productively (Goldberg 2006: 5). We tested four types of instantiations of these two constructions: (a) frequent, (b) infrequent, (c) similar in structure and (d) ungrammatical sentences. Frequent and infrequent instantiations were identified using corpus data from the corpus infrastructure Korp (Borin, Forsberg & Roxendal 2012). While highly frequent expressions such as (1b) and (2b) are likely to be accessed as lexicalized constructions, infrequent ones such as (1c) and (2c) are expected to be processed using a schema (Bybee & Thompson 2007: 271). First results show differences in the response time and accuracy between the groups, as well as between the four types of instantiations. 

References: Bybee, Joan. 2010. Language, usage and cognition. Cambridge: Cambridge University Press. Bybee, Joan & Sandra Thompson. 2007. Three frequency effects in syntax. In Joan Bybee (ed.), Frequency of use and the organization of language, 269–278. Oxford/New York: Oxford University Press. Goldberg, Adele. 2006. Constructions at work. The nature of generalization in language. Oxford [u.a.]: Oxford University Press. Mathôt, Sebastiaan, Daniel Schreij & Jan Theeuwes. 2012. OpenSesame: An open-source, graphical experiment builder for the social sciences. Behavior Research Methods 44(2), 314–324. Mathôt, Sebastiaan & Jennifer March. 2022. Conducting linguistic experiments online with OpenSesame and OSWeb. Language Learning 72 (4), 1017–1048. Schmid, Hans-Jörg. 2017. A framework for understanding linguistic entrenchment and its psychological foundations. In Hans-Jörg Schmid (ed.), Entrenchment and the psychology of language learning. How we reorganize and adapt linguistic knowledge. Washington, DC: De Gruyter Mouton. Olofsson, Joel & Julia Prentice. 2020. För tre enorma öl sedan. Befästning av semi-schematiska konstruktioner i L2-svenska. Språk och stil 30, 91–116.

Multiple translations as evidence for cognitive and linguistic factors influencing the French translation of English [N+N(+N)…] constructions 

Thomas Prinzie Université catholique de Louvain (UCLouvain) 

Research in Translation Studies naturally relies on constructs and methods from various subfields of linguistics. Still, there remains a need for a unified theoretical framework and especially for improved interaction with cognitive linguistics and psycholinguistics (De Sutter & Lefer, 2020; Halverson & Kotze, 2021). In this context, Halverson’s (2017) revised “gravitational pull” (RGP) model aims to explain translators’ behavior in terms of usage-based factors such as the salience of source and target items and the entrenchment of translation pairs. However, issues regarding how these factors are best operationalized remain unresolved (e.g. Boswijk & Coler, 2020; Schmid & Günther, 2016; Gilquin, 2008; see also Hartsuiker et al., 2016). Previous RGP studies have explored the under- or overuse of particular lexical or grammatical forms in translation. Here, I propose a complementary approach to elucidate the cognitive and linguistic factors underlying translator decisions, namely based on the variability of multiple translations, i.e., the number of different solutions for a given source instance. This approach, already championed by Malmkjær (1998) and Campbell (2000), has remained underexplored for lack of data (but see Castagnoli, 2020). Recent compilation of the Multilingual Student Translation corpus (MUST; Granger & Lefer, 2020) now provides the necessary access to multiple translators’ interpretations and realizations. As a test case, I present a study of multiple French translations of English texts in MUST, focusing on decisions made for noun sequences (e.g., disaster relief program coordinator). The bare juxtaposition of nouns in these [N+N(+N)…] constructions provides efficient information packing in English but poses specific challenges for translation into French (Lefer & De Clerck, 2021). At the formal level, French tends to use prepositional post-modification [N+Prep+N(+Prep+N)…], which can impose a significant cognitive load for longer sequences (e.g., disaster relief program coordinator → coordinateur du programme d'aide en cas de catastrophe). At the semantic level, potentially ambiguous relationships between constituents, left implicit in English, often need to be stated explicitly in French. An additional motivation to study these constructions is that they represent an extension of the case of noun compounds, whose processing and linguistic representation have been investigated by psycholinguists (Baayen et al., 2010; Gagné, 2011) and in NLP (Nakov, 2013; Dima, 2016). The nature of these noun sequence constructions requires analysis of factors and translator decisions at multiple hierarchical levels. This appears to be both a methodological challenge and an opportunity for fresh insights. I present results from two analyses: (1) a qualitative description of how translation solutions vary at three structural levels (main constituents, linking prepositions, and surface grammar), which can reveal different aspects of how translators interpret source ambiguity; and (2) quantitative multifactorial models of translation variability as a function of sequence length, lexicalization, frequency of use, and structural ambiguity. 

References: Baayen, H. R., Kuperman, V. & Bertram, R. 2010. Frequency effects in compound processing. In S. Scalise & I. Vogel (eds.), Cross-disciplinary issues in compounding. Amsterdam: John Benjamins. 257-270. Boswijk, V., & Coler, M. 2020. What is salience? Open Linguistics, 6(1). 713-722. Campbell, S. 2000. Choice Network Analysis in Translation Studies. In: Olahan, M. (ed.) Intercultural Faultlines. Research Models in Translation Studies 1. Textual and Cognitive Aspects. Manchester: St. Jerome. 29–42. Castagnoli, S. 2020. Translation choices compared: Investigating variation in a learner translation corpus. In Granger, S. & Lefer, M.-A. (eds.). Translating and comparing languages: Corpus-based insights. Selected Proceedings of the Fifth Using Corpora in Contrastive and Translation Studies Conference. Louvain-la-Neuve: Presses Universitaires de Louvain. 25-44. De Sutter, G., & Lefer, M.-A. 2020. On the need for a new research agenda for corpus-based translation studies: A multi-methodological, multifactorial and interdisciplinary approach. Perspectives, 28(1). 1-23. Dima, C. 2016. On the compositionality and semantic interpretation of English noun compounds. Proceedings of the 1st Workshop on Representation Learning for NLP. 27-39. Gagné, C. L. 2011. Psycholinguistic perspectives. In Lieber, R., & Štekauer, P. (eds.). The Oxford handbook of compounding. Oxford University Press. 255-271. Gilquin, G. 2008. What you think ain’t what you get: Highly polysemous verbs in mind and language. Du fait grammatical au fait cognitif. From gram to mind: Grammar as cognition, 2. 235-255. Granger, S., & Lefer, M.-A. 2020. The Multilingual Student Translation corpus: A resource for translation teaching and research. Language Resources and Evaluation, 54(4). 1183- 1199. Halverson, S. L. & Kotze, H. 2021. Sociocognitive constructs in Translation and Interpreting Studies (TIS): Do we really need concepts like norms and risk when we have a comprehensive usage-based theory of language? In S. L. Halverson & Á. Marín García (eds.). Contesting epistemologies in Translation and Interpreting Studies. London: Routledge. 51-79. Halverson, S. L. 2017. Gravitational pull in translation: Testing a revised model. In De Sutter, G., Lefer, M. A., & Delaere, I. (eds.). Empirical translation studies: New methodological and theoretical traditions. Walter de Gruyter. 9-46. Hartsuiker, R. J., Beerts, S., Loncke, M., Desmet, T., & Bernolet, S. 2016. Cross-linguistic structural priming in multilinguals: Further evidence for shared syntax. Journal of Memory and Language, 90. 14-30. Lefer, M.-A. & De Clerck, M. 2021. L’apport des corpus intermodaux en lexicologie contrastive: Étude comparative de la traduction écrite et de l’interprétation simultanée des séquences de noms. In S. Hanote & R. Nita (eds.). Morphophonologie, lexicologie et langue de spécialité. Rennes: Presses Universitaires de Rennes. 145-162. Malmkjær, K. 1998. Love thy neighbour: Will parallel corpora endear linguists to translators? Meta: Translators' Journal, 43(4). 534-541. Nakov, P. 2013. On the interpretation of noun compounds: Syntax, semantics, and entailment. Natural Language Engineering, 19(3). 291-330. Schmid, H. J., & Günther, F. 2016. Toward a unified socio-cognitive framework for salience in language. Frontiers in psychology, 7. 1110.

The prosody of insubordination in French and Spanish 

Víctor Royo Viñuales ULiège, Université catholique de Louvain (UCLouvain) 

Insubordination –defined by Evans (2007:367) as the “conventionalized main clause use of what […] appear to be formally subordinate clauses– has received an increasing amount of scholarly attention exploring both its syntactic status and its discourse functions. However, as D’Hertefelt (2018:220) points out, “prosody remains […] one of the […] least studied features of insubordination”. Kaltenböck’s (2016) and Elvira-García et al.’s (2017) studies form notable exceptions in this respect. This paper aims to further fill this gap by providing an exploration of the prosodic behaviour of hypothetical manner constructions in French and Spanish, introduced by comme si and como si (both meaning ‘as if’) respectively. More precisely, it seeks to answer two specific research questions: (i) can different degrees of (in)dependence be correlated with distinct prosodic configurations? and (ii) can prosody help us investigate analytical ambiguity that may give rise to insubordination in discourse? To these ends, this contrastive study relies on the prosodic analysis of (in)subordinate ‘as if’ clauses, following the Autosegmental Metrical (AM) model (Pierrehumbert 1980) and specifically using both Fr_ToBI (Delais-Roussarie et al. 2015) and Sp_ToBI (Prieto & Roseano 2010, Hualde & Prieto 2015) transcription systems. Following previous applications of these systems (Delais-Roussarie et al. 2015; Elvira-García et al. 2017), the analysis focuses on nuclear configurations (pitch accent + boundary tone), as these are usually found to convey the most relevant intonational information. These were applied, for each language, to a 100-hit sample of spoken corpus data, providing a preliminary overview of the different prosodic patterns, as well as to an ad hoc 400-hit sample of semi-spontaneous spoken data elicited through a Discourse Completion Task (Vanrell et al. 2018). First, the results show that different prosodic configurations do correlate with distinct degrees of (in)dependence. Syntactically-independent clauses generally show prosodically-independent behaviour, i.e. they constitute an independent intonational phrase as they are preceded by a main clause displaying a falling tone. By contrast, prototypical subordinate clauses are generally preceded in discourse by a main clause showing a continuation rise (Elvira-García et al. 2017:32), which implies prosodic integration of the subclause into the main clause. Second, this study sheds clear light on how insubordinate constructions arise in discourse. Crucial in this respect are bridging contexts (Evans and Wilkins 2000:550) which contextually support both a manner and a denial reading, illustrated in (1) and (2) (see also Brinton 2014:104). The analysis shows that a significant number of subordinate clauses like (1) and (2) display a risingfalling pitch contour, which is also widely attested in insubordinate clauses conveying denial of an assumption in the two languages. This finding thus supports Van linden & Van de Velde’s (2014) hypothesis of insubordination arising through a process of hypoanalysis (Croft 2000) and shows that speakers do reanalyse this contextual meaning as an inherent semantic property of the subordinating conjunction, allowing insubordinate uses to appear in discourse. (1) Hablas como si me conocieses. (2) J’ai dit “félicitations” comme si ça m’intéressait. You speak as if you knew me. I said “congratulations” as if I cared 

References: Brinton, L.J. (2014). The Extremes of Insubordination: Exclamatory as if! Journal of English Linguistics, 42(2), 93-113. Croft, W. (2000). Explaining Language Change. An Evolutionary Approach. London: Longman. Delais-Roussarie, E., Post, B., Avanzi, M., Buthke, C., Di Cristo, A., Feldhausen, I., Jun, S.- A., Martin, P., Meisenburg, T., Rialland, A., Sichel-Bazin, R. & Yoo, H. (2015). Intonational phonology of French: Developing a ToBI system for French. In S. Frota and P. Prieto (eds.), Intonation in Romance, 63-100. Oxford: Oxford University Press. D’Hertefelt, S. (2018). Insubordination in Germanic. The Gruyter Mouton: Berlin/Boston. Elvira-García, W., Roseano, P. & Fernández-Planas, A.-M. (2017). Prosody as a cue for syntactic dependency. Evidence from dependent and independent clauses with subordination marks in Spanish. Journal of Pragmatics, 109, 29-46. Evans, N. (2007). Insubordination and its uses. In I. Nikolaeva (ed.), Finiteness: Theoretical and empirical foundations, 366-431. Oxford: Oxford University Press. Evans, N. & Wilkins, D. (2000). In the mind’s ear: The semantic extensions of perception verbs in Australian languages. Language 76: 546–592. Hualde, J.-I. & Prieto, P. (2015). Intonational variation in Spanish: European and American varieties. In S. Frota & P. Prieto (eds.), Intonational Variation in Romance, 350-391. Oxford: Oxford University Press. Kaltenböck, G. (2016). On the grammatical status of insubordinate if-clauses. In G. Kaltenböck, E. Keizer & A. Lohman (eds.), Outside the clause: Form and function of extra-clausal constituents, 341-377. Amsterdam, Philadelphia: John Benjamins. Pierrehumbert, J. (1980). The Phonology and Phonetics of English Intonation (thesis). MIT, Cambridge, MA. Prieto, P. & Roseano, P. (2010). Transcription of Intonation of the Spanish Language. Munich: Lincom Europa. Van linden, A. et Van de Velde, F. (2014). (Semi-)autonomous subordination in Dutch: Structures and semantic–pragmatic values. Journal of Pragmatics, 60, 226-250. Vanrell, MdM., Feldhausen, I. & Astruc, L. (2018). The discourse completion task in romance prosody research: status quo and outlook. In I. Feldhausen, J. Fliessbach & MdM. Vanrell (eds.), Methods in Prosody: A Romance Language perspective, 191-227. Berlin: Language Science Press.

Predictable Ways of Being Unpredictable: Unconventional Uses of Verbal Constructions 

Margot Vancauwenbergh University of Antwerp 

This study sets out to unveil cross-linguistic systematicity in the exploitation of certain verbal constructions to express unconventionality, i.e. to indicate that (1) aspects of the speech event deviate from the convention, and/or that (2) the combination of a construction and its linguistic context is non-canonical (see also Haspelmath (1999), Schmid & Günther (2016) or De Wit et al. (2020) for definitions of related notions, such as extravagance and salience). Three particular constructions are examined: the progressive, GO-constructions and COME-constructions. The main objective of this study is to provide cross-linguistic evidence for the idea that these constructions are recruited significantly more often to express unconventionality than their simple tense alternatives. The tendency of these constructions to convey unconventionality is deemed inherent to their semantic properties: each of these constructions indicates that the situation they report is not seen as fully and regularly integrated in what the speaker conceives of as the ‘ground’. The sense of unconventionality is therefore expected to be systematic and stable across languages and across different stages of the constructions’ grammaticalization processes. A first data collection samples progressives, GO- and COME-constructions in English, Dutch and French and pairs them with minimally different instances of their simple tense alternatives within translation corpora (Europarl (Koehn 2005) and Harry Potter (van der Klis et al. 2020)). All contexts of these attestations are subsequently measured for unconventionality using a comprehensive set of indices (Petré 2017). This will then serve as the source data to expand the dataset to 19 languages, ensuring a cross-linguistic analysis across Germanic, Romance, Slavic, Baltic and Finno-Ugric families. On the basis of this dataset, the hypothesis that progressives, GO- and COME-constructions express unconventionality significantly more often is tested for all languages and constructions under examination. To additionally verify the hypothesis that these constructions convey unconventionality irrespective of their grammaticalization states, a further analysis takes the form of a quantitative comparison of the grammaticalization degrees of the constructions’ language-specific instantiations in English, Dutch and French. 

References: De Wit, A., P. Petré & F. Brisard. 2020. Standing out with the progressive. Journal of Linguistics 56. 479-514. Haspelmath, M. 1999. Why is grammaticalization irreversible? Linguistics 37. 1043–1068. Koehn, P. 2005. Europarl: A Parallel Corpus for Statistical Machine Translation, MT Summit. Petré, P. 2017. The extravagant and the progressive: An experimental corpus study on the grammaticalization history of [be Ving]. English Language and Linguistics 21. 227– 250. Schmid, H-J & F. Günther. 2016. Toward a unified socio-cognitive framework for salience in language. Frontiers in Psychology 7. 1110. van der Klis, M., B. Le Bruyn & H. de Swart. 2020. De la sémantique des temps verbaux à la traductologie: une comparaison multilingue de L’Étranger de Camus. In E. Corre et al. (eds.) Linguistic approaches to Tense, Aspect, Modality, Evidentiality, based on the Novel L’Etranger ("The Stranger") by Albert Camus, and its Translations. Benjamins.

As if grammar, discourse and prosody don’t interact: Comparative study of the prosodic profiles of English as if-clauses and Dutch alsof-clauses 

Wout Van Praet , Liesbeth Degand and An Van Linden Université catholique de Louvain (UCLouvain)1 , ULiège 

This study examines the production of clauses introduced by the conjunction as if in spoken English and its equivalent alsof in spoken Dutch. In both languages, as if- and alsof-clauses can have different levels of grammatical and discursive (in)dependence, e.g. (1)-(4). Our aim is to provide a comparative analysis of the prosodic profiles of these clauses and examine if their different dependency statuses are signalled prosodically. (1) Het is alsof ik niet besta. (CGN) ‘It’s as if I don’t exist.’ (2) He walks around as if he owns the place. (BNC) (3) Alles prikt gewoon. Alsof je allemaal glaswol over je heen krijgt. (CGN) ‘Everything just stings. As if you’re getting glass wool all over you.’ (4) S1: Yet another flower. S2: Hm as if we haven’t got enough. (BNC) The study is usage-based and grounded in analyses of spoken data randomly extracted from the ‘British National Corpus’ (BNC) for English and the ‘Corpus Gesproken Nederlands’ (CGN) for Dutch. For each language, 250 examples were coded, to arrive at sufficiently large sets of data for the various grammatical and discursive subtypes. To investigate the interplay between grammar, discourse, and prosody, three coding schemes are used. The grammatical scheme assesses the clauses’ level of grammatical (in)dependence, based on: clefting, pronominal proportionality, and fronting [1]. The degree of discourse dependence is established based on discourse-oriented features like speech-functional value, modality, turn-taking, and co-referentiality [2-4]. The prosodic scheme, finally, probes for prosodic features potentially correlating with the different levels of grammatical and/or discursive dependence: (i) placement of (major) intonation boundaries, (ii) pitch contour of the intonation unit (IU), (iii) key (i.e. pitch level of the IU onset: Low, Mid, or High), and (iv) amount of pitch variation. Preliminary findings confirm that the different grammatical statuses and discourse uses that as if- and alsof-clauses can have correlate with different prosodic profiles. First, the level of grammatical dependence of the clause is reflected in the segmentation into IUs, with grammatically fully integrated clauses being also prosodically integrated in the same IU with their matrix. A cline is observed here from complements (1) to non-obligatory adverbials (2) to independent clause uses (i.e. dependency shifts (3) and insubordination (4)). Moreover, while the amount of pitch variation throughout the production of the clause appears to be a minor factor, key does correlate with the dependency status of the clause. Here too, a cline is found from non-obligatory adverbials (mid-low key), to dependency shifts (mid to mid-high key) to insubordination (high key). Finally, while we found no fixed intonation contour for the different clause uses in either language, Dutch insubordinate clauses were frequently uttered on a risefall tone, which has been associated with ‘challenging’ meanings [5]. English insubordinate clauses, by contrast, tend to be produced with a fall, especially ‘high falls’, which signal speaker engagement. Both meanings, we argue, can be linked to the ‘assumption-denying’ function that these clauses serve. 

References: [1]Smessaert, H., B. Cornillie, D. Divjak, and K. van den Eynde. 2005. Degrees of Clause Integration. Linguistics 43: 471–529. [2]Crible, L. and L. Degand (2019). Domains and functions: A two-dimensional account of discourse markers. Discours 24: 3-35. [3]Verstraete, J-C. 2007. Rethinking the Coordinate-Subordinate Dichotomy. Interpersonal Grammar and the Analysis of Adverbial Clauses in English. Berlin: Walter de Gruyter. [4]Debaisieux, J-M, ed. 2013. Analyses linguistiques sur corpus: Subordination et insubordination en français. Paris: Hermes Science publications. [5]Cruttenden, A. 1997. Intonation. Cambridge: CUP.

Pimp my ride: The trajectory from English construction to productive Dutch verb Eline Zenner , Stefano De Pascale, Dirk Pijpops and Freek Van de Velde KU Leuven , Vrije Universiteit Brussel (VUB) , ULiège 

The research presented in this poster empirically demonstrates how the English construction ‘Pimp my ride’ led to the productive verb pimpen ‘to fancify’ in Dutch, despite (i) the original taboo meaning of pimp, (ii) the verb’s introduction in the Low Countries as part of a name for an English TV show and (iii) the existence of the Dutch near-synonym opleuken. The studies, which add to a pilot conducted on newspaper data published between 1998 and 2009 (AUTHORS 2010), draw from research in (Diasystematic) Construction Grammar (Höder 2012) and usage-based contact linguistics (Backus 2020). Data consists of over 160,000 tweets posted between 2007 and 2020 containing a string of pimp. First, a subset of the tweets are analyzed manually for deconstructionalization patterns, defined here as the gradual or stepwise deviation from the formal and semantic properties of the original expression Pimp my Ride. A holistic deconstructionalization score for each instance of pimp occurring in the constructional template [pimp POSS N] (N=965) is calculated based on semantic and formal properties of the open slots. Modeling this score in relation to year of attestation reveals a significant increasing deviation from the original construction, however only when also including the original Pimp my ride construction in the analysis. Second, three substudies aim to pattern the further evolution of Dutch pimpen outside of the original [pimp POSS N] template. A first substudy models progressive deconstructionalization by annotating semantic features of the pimped object and morphological properties of the verb in 3,596 instances of pimp not part of the constructional template [pimp POSS N]. Results mainly reveal a fairly stable aggregate deconstructionalization score over time. The second substudy focuses on the morphological derivations of the verb (e.g. overpimpen ‘to overpimp’) through a big data approach that includes all 160,000 tweets. Plotting the diversity of these forms through time reveals a decline in the number of derivations after 2012. Adopting an onomasiological perspective, the third substudy compares the use of pimpen (N=3736) to the morphologically enriched oppimpen (N=802) and to the Dutch near-synonymous lexicalization opleuken (N=1032). Results reveal a decline in the use of pimpen, particularly in the Netherlands and mainly in favor of opleuken. The multifaceted approach presented in these studies, combining semantic and formal characteristics, semasiology and onomasiology, manual annotation and big data approaches, help us chart verbal pimp’s trajectory in Dutch profoundly. As such, the studies presented here more generally contribute to our understanding of the borrowing and nativization of constructions in contact.

References: Backus, A. (2020). Usage-based approaches. In The Routledge Handbook of Language Contact, eds. Evangelia Adamou & Yaron Matras (Routledge). Höder, S. (2012). “Multilingual constructions: a diasystematic approach to common structures,” in Multilingual Individuals and Multilingual Societies, eds K.Braunmüller and C. Gabriel, (Amsterdam: Benjamins), 241–257. doi: 10.1075/hsm.13.17hod AUTHORS (2010)