Taal & Tongval Borrowing. Pragmatic and Variational Linguistic Approaches. Gent, KANTL, 27 November

Transcription

1 Taal & Tongval 2015 Borrowing Pragmatic and Variational Linguistic Approaches Gent, KANTL, 27 November

2

3 Organizing Committee Reinhild Vandekerckhove Eline Zenner Gunther De Vogelaer Dirk Geeraerts Ben Hermans Jan Nijlen Twilhaar Wim Vandenbussche Roeland Van Hout Jacques Van Keymeulen Scientific Committee Gunther De Vogelaer Dirk Geeraerts Ben Hermans Jan Nijlen Twilhaar Reinhild Vandekerckhove Wim Vandenbussche Roeland Van Hout Jacques Van Keymeulen Eline Zenner Funding FWO Flanders KANTL Special thanks to Valerie Bouckaert

4 Program Registration & coffee Grote zaal Eline Zenner & Reinhild Vandekerckhove Welcome and introduction Grote zaal Gisle Andersen The pragmatic perspective: The role of discourse context in (incipient) borrowing Coffee break Grote zaal Chair: Eline Zenner Ad Backus & Derya Demirçay Beyond asymmetry: Modeling lexical borrowing when there is no base language Henning Radke Meertaligheid en Codewisseling. De interactie tussen het Nederlands en het Sranan Tongo in Surinaamse online-communicatie Hielke Vriesendorp Omg zo fashionably English Tapijtenzaal Chair: Karlien Franco Jocelyne Daems, Dirk Geeraerts & Kris Heylen The success of English loans in IT terminology in Belgian Dutch and Netherlandic Dutch Isabelle Delaere & Gert De Sutter R&D of O&O? Een multivariate corpusstudie naar de keuze tussen leenwoorden en endogene alternatieven in vertalingen Costanza Asnaghi "La fiction sulla love story del boss ha fatto flop": The invasion of Anglicisms and their geographical distribution in Italian Lunch break

5 Grote zaal Esme Winter-Froemel Exploring the pragmatic necessity of borrowing: Expressivity, playfulness, euphemism / dysphemism, naming Grote zaal Chair: Laura Rosseel Eric Hoekstra & Frits Van der Kuip Peanut butter and compositionality Karlien Franco, Eline Zenner & Dirk Speelman Gender assignment to anglicisms in Dutch: A variationist perspective Stef Grondelaers Do the Dutch use English to sound superior, dynamic, or both? Evidence from speaker evaluation and free response experiments. Tapijtenzaal Chair: Jocelyne Daems Freek Van de Velde & Eline Zenner Pimp my Lexis: Constructional borrowing. Timothy Colleman & Sam Beke On potentially contact-related shifts in Dutch and Afrikaans argument structure constructions Aafke Lettinga, Peter Broeder, Carel van Wijk Code-switching in textspeak: domineert het Engels de digitale communicatie van Nederlandse jongeren? Coffee break Grote zaal Frank van Meurs Attitudes to English and Dutch job titles among Dutch and Flemish students: An experimental study Conclusion and reception

6 Abstracts Plenaries Gisle Andersen... 1 Frank van Meurs, Berna Hendriks & Dirk Sanders... 3 Esme Winter-Froemel... 5 Parallel sessions Costanza Asnaghi... 7 Ad Backus & Derya Demirçay Timothy Colleman & Sam Beke Jocelyne Daems, Dirk Geeraerts & Kris Heylen Isabelle Delaere & Gert De Sutter Karlien Franco, Eline Zenner & Dirk Speelman Stef Grondelaers Eric Hoekstra & Frits Van der Kuip Aafke Lettinga, Peter Broeder & Carel van Wijk Henning Radke Freek Van de Velde & Eline Zenner Hielke Vriesendorp... 28

7 Gisle Andersen Norges Handelshøyskole The pragmatic perspective: The role of discourse context in (incipient) borrowing In many respects, it makes sense to talk of a recent shift in focus a pragmatic turn in research on borrowing. An increasing number of studies are exploring the role of socio-pragmatic context in lexical and terminological borrowing, and an increasing number of studies are concerned not principally with lexis and terminology, but with how discourse-pragmatic items are copied from one language to another. A third aspect of this shift is increased attention on phraseological units and the co-text in which individual borrowed forms are used. Thus, the term pragmatic borrowing has come to be used about the scientific meeting ground between contact linguistics and pragmatics, and in a more specific sense about the integration of discourse-pragmatic features (e.g. interjections, discourse markers or politeness formulae) from a source language to a recipient language. The pragmatic aspects of borrowing incorporate a range of cultural, social and cognitive factors that may be seen to motivate borrowing, such as the emblematic nature of individual forms and their potential for expressing such notions as coolness, urbanism, youth, globalisation, etc., reflecting common associations with particular usage groups. Pragmatic borrowing may also be motivated by the more general social prestige of the source language culture, or factors relating to mental processing, precision, creativity, markedness, or the like. The paper first gives a brief overview of these developments, exemplified by studies that illustrate the various stages of borrowing, from idiosyncratic single-word switches via more regularised switches and incipient borrowing to more persistent borrowing. Next, the paper highlights recent cases of borrowing of certain discourse-pragmatic items from English into Norwegian (discourse-topic markers, interjections and backchannel items). The case studies are meant to illustrate how a set of methodologies and theoretical backgrounds (speech act theory, relevance theory, Sinclair s extended units of meaning, variational pragmatics) may fruitfully contribute to 1

8 adequate accounts of pragmatic borrowing phenomena. It is argued that, in order for this fledging field to mature, we need to anchor our studies more firmly in pragmatic theory in more systematic ways than what has been done hitherto. 2

9 Frank van Meurs, Berna Hendriks & Dirk Sanders Radboud Universiteit Nijmegen Attitudes to English and Dutch job titles among Dutch and Flemish students: An experimental study Linguistic borrowing has been said to be motivated, among other reasons, by prestige considerations (Hock, 1986). However, the use of loanwords can also evoke negative feelings motivated by feelings of linguistic purism because speakers feel their mother tongue is under threat (Grezel, 2007). It has been shown that degrees of linguistic purism, specifically opposition to English loanwords, can vary across regions (Thøgersen, 2004). These considerations lead to the expectation that English loanwords would either be regarded as prestigious or would be rejected by non-native speakers of English and that such reactions may be different in regions with different sociolinguistic circumstances. The purpose of this study was to investigate if in two regions with a shared official language (Dutch) but different sociolinguistic background and history, Flanders and the Netherlands, English loanwords were indeed perceived differently. It has been argued that because of French dominance over Flemish, attitudes towards loanwords in Flanders are negative, whereas in the Netherlands attitudes are more positive because Dutch has not been threatened by another language (Geeraerts & Grondelaers, 2000; Zenner et al., 2013). In an experiment with a between-subject design, 155 Dutch and Flemish university students evaluated equivalent Dutch and English job titles (e.g. hoofredacteur/ editor-in-chief) with regard to associations, comprehensibility, attractiveness, naturalness, and intention to apply for the job indicated by the title. In addition, general attitudes towards English loanwords were measured. Findings did not reveal differences between the Dutch and Flemish participants in evaluation of the English versus Dutch job titles nor in general attitude towards English loanwords. Both groups displayed more positive attitudes towards Dutch job titles than the English equivalents, while their general attitude to English loanwords was positive. 3

10 It can be concluded that English was not perceived as more prestigious than Dutch, contrary to claims in the literature (Kahane, 1992; Latomaa, 1998; Takashi, 1990). Moreover, there was no difference in language purism between participants from regions that might be expected to differ in this respect on the basis of their sociolinguistic backgrounds (contra Geeraerts & Grondelaers, 2000; Zenner et al., 2013). The finding of the current study that Dutch job titles were considered more prestigious than English equivalents contrasts with findings from earlier studies among similar Netherlandic target groups, which showed 1) no differences in evaluations of job ads with English versus Dutch words (Van Meurs et al., 2004; 2006) and 2) a generally better evaluation of English job titles than of their Dutch equivalents (Van Meurs et al., 2007). These contrasting findings challenge the notion of consistency of language attitudes among people with similar sociolinguistic backgrounds. Possible explanations for the findings will be offered, including homogeneity of Dutch and Flemish participants in age and education and effects of content and text length. 4

11 Esme Winter-Froemel Universität Trier Exploring the pragmatic necessity of borrowing: Expressivity, playfulness, euphemism / dysphemism, naming In traditional research, necessary and luxury loanwords have often been opposed to each other as representing two fundamentally different types of borrowing. However, the terms of necessity and luxury have been shown to be problematic, as necessary loanwords can be avoided by choosing alternative types of contact-induced innovations in the recipient language, and from a usage-based perspective, speakers also perceive a certain need for luxury borrowings. In my paper, I will thus rephrase the distinction in terms of catachrestic (~necessary) vs. noncatachrestic (~luxury) borrowings (Onysko & Winter-Froemel 2011) and focus on the latter group in order to investigate basic principles motivating the introduction of the loanwords in the recipient language. More specifically, I will discuss expressivity, playfulness, and euphemism / dysphemism, which have been proposed in previous research as basic functions of borrowings (see, e.g. Carstensen 1965, Galinsky 1967), and I will explore how these concepts can be refined from a usage-based perspective. Finally, I will turn to the function of naming, which is traditionally considered to be fundamentally different from the pragmatically based factors listed above. Contrary to this belief, I will argue that naming also has a pragmatic dimension by virtue of speakers intentions to (re)conceptualize the objects in question. 5

12

13 Costanza Asnaghi UGent La fiction sulla love story del boss ha fatto flop : the invasion of Anglicisms and their geographical distribution in Italian Anglicisms enter in Italian primarily at the lexical level (e.g. Berruto 2012) as luxury loans that coexist with a pre-existing Italian word. Although geographical distribution in the lexicon in a quantitative manner has already been studied in a specific area of Italy for other linguistic features (i.e. using generalized additive mixed-effects regression model as a predictor for lexical variation in Tuscany dialects, Wieling et al. 2014), to our knowledge there are no quantitative studies on which areas of Italy are more receptive to Anglicisms. The aim of this study is to verify whether a specific list of Anglicisms are penetrating the Italian language, and in particular the Italian language of newspapers, uniformly on a geographical level. This study surveys the patterns of regional diffusion of a selection of English borrowings in about 500 Italian newspaper websites based in over 100 Italian provinces. Newspapers are particularly receptive to linguistic innovations (Dardano 2002, Bonomi 2002, Crocco 2015). Each newspaper is geographically bound, therefore the language preferences of each newspaper are presumably influenced by the local language. The frequency of the English borrowings is automatically counted through site-restricted web searches (Grieve et al. 2013; Author 2013; Author et al. 2014) and then compared to the frequencies of onomasiologically related Italian expressions; for instance the frequencies of break and pausa in each newspaper website. The collected data is analyzed through local spatial autocorrelation (Ord & Getis 1995; Grieve et al. 2011), a statistical technique that allows to visually represent the preference for the use of an English word or the corresponding Italian word in a manner that identifies regional linguistic areas that are comparable to traditional isoglosses. Multivariate statistics allows for an overall interpretation of the data. Unexpectedly, when compared to pre-existing Italian dialect maps, the latent factors of geographical distribution of the Anglicisms in Italian align almost completely (see Figure 1). 7

14 This work shares its theoretical Cognitive Sociolinguistics background with previous studies on lectal differences in Dutch (e.g. Geeraerts et al. 1999; Speelman et al. 2003; Ruette et al. 2014) and on Anglicisms in Dutch (Zenner et al. 2012). References Author PhD dissertation. Author et al Berruto, Gaetano Sociolinguistica dell italiano contemporaneo. Nuova edizione. Roma: Carocci. Bonomi, Ilaria L italiano giornalistico. Dall inizio del 900 ai quotidiani on line. Firenze: Franco Cesati. Crocco, Claudia [in press]. Corpora e testi di italiano contemporaneo. In Maria Iliescu, Eugeen Roegiest (eds.), Manuel des anthologies, corpus et textes romans (Manuals of Romance Linguistics, MLR, 7), Cap Berlin: De Gruyter. Dardano, Maurizio La lingua dei media. In Valerio Castronovo & Nicola Tranfaglia, La stampa italiana nell età delle TV. Dagli anni Settanta a oggi, Roma & Bari: Laterza. Geeraerts, Dirk, Stefan Grondelaers & Dirk Speelman Convergentie en divergentie in de Nederlandse woordenschat. Een onderzoek naar kleding- en voetbaltermen. Amsterdam: Meertens Instituut. Grieve, Jack A regional analysis of contraction rate in written standard American English. International Journal of Corpus Linguistics, 16(4) Grieve, Jack, Dirk Speelman & Dirk Geeraerts A statistical method for the identification and aggregation of regional linguistic variation. Language Variation and Change Grieve, Jack, Costanza Asnaghi & Tom Ruette Site-restricted web searches for data collection in regional dialectology. American Speech, Ord, J.K. & Arthur Getis Local spatial autocorrelation statistics: distributional issues and an application. Geographical Analysis, 27(4) Ruette, Tom, Dirk Geeraerts, Yves Peirsman, Dirk Speelman Semantic weighting mechanisms in scalable lexical sociolectometry. In Benedikt Szmrecsanyi & Bernhard Wälchli (eds.), Aggregating Dialectology, Typology, and Register Analysis: Linguistic Variation in Text and Speech, Berlin & New York: de Gruyter. Speelman, Dirk, Stefan Grondelaers & Dirk Geeraerts Profile-based linguistic uniformity as a generic method for comparing language varieties. Computer and the Humanities 37: Wieling, Martijn, Simonetta Montemagni, John Nerbonne & R. Harald Baayen Lexical differences between Tuscan dialects and standard Italian: Accounting for geographic and socio-demographic variation using generalized additive mixed modeling. Language 90(3)

15 Zenner, Eline, Dirk Speelman & Dirk Geeraerts Cognitive Sociolinguistics meets loanword research: Measuring variation in the success of anglicisms in Dutch. Cognitive Linguistics

16 Ad Backus & Derya Demirçay Tilburg University Beyond asymmetry: Modeling lexical borrowing when there is no base language In recent codeswitching data from informal conversations within the Turkish-Dutch bilingual community in The Netherlands, we frequently witness large stretches of discourse in which there is so much alternation that it becomes impossible to say whether the speakers are using Dutch or Turkish. In a situation such as this, it becomes difficult to say which Dutch words have been borrowed into Turkish and which Turkish words into Dutch. The base language itself has become a mixture, and this has implications for how we model lexical borrowing. In our presentation, we will introduce these data, taken from second generation bilinguals. In the descriptive first part of the talk, we will provide a quantitative and qualitative analysis. Language use is shown to cover a whole range of phenomena. On the one hand, we find the traditional patterns of bilingual speech, with sometimes Turkish and sometimes Dutch clearly functioning as the base language into which words or larger chunks from the other language are inserted. On the other hand, there are many examples of clean alternation, in which a large chunk in one language is followed by a large chunk in the other. However, since these chunks are often quite small, the predominant type of codeswitching is often rapid alternation between chunks of Turkish and of Dutch material. The question is how to describe this pattern, and that s what the second, theoretical, part of the paper will focus on. Our suggestion will be that daily experience with this rapid back-and-forth switching has created a bilingual register which represents a next step on a continuum from language separation to language fusion. Most language contact settings stop well short of fusion, either because separation stabilizes in a robust language choice pattern, or because the community undergoes language shift. Though not the focus of the paper, we will suggest an explanation for why the Turkish immigrant community does seem to move further on this continuum than many other communities. 10

17 Timothy Colleman & Sam Beke Ghent University On potentially contact-related shifts in Dutch and Afrikaans argument structure constructions This paper explores the potential effects of language contact on the syntax and/or semantics of (partially or completely) schematic argument structure constructions, and is as such part of the small but growing body of work at the intersection between language contact research and construction grammar (see, e.g., Doğruöz & Backus 2009, Pietsch 2010, Van de Velde & Zenner 2010, Höder 2012, Colleman 2014, Colleman & Noël 2014). If we define constructional change, following Hilpert (2013: 16) as change which selectively seizes a conventionalized form-meaning pairing of a language, altering it in terms of its form, its function, its frequency, its distribution in the linguistic community, or any combination of these, this raises the question of whether all of these dimensions are subject to change under the influence of language contact, and, if so, whether they are equally susceptible to such changes. As a first step towards answering this question, the present paper briefly looks into three potentially contact-induced cases of ongoing constructional change: (i) the emergence, in present-day informal varieties of Dutch, of a new schematic pattern modeled on the English time -away construction (see Jackendoff 1997) as in We dansten de avond weg We danced the night away (ii) the extension of ditransitive and prepositional-dative constructions in Afrikaans to verbs that could formerly not be used in the construction(s) in question under the probable influence of English, which we will argue to constitute a case of distributional assimilation (Gast & van der Auwera 2012) in the domain of argument structure semantics an example is the use of ontneem in a secundative pattern, as in Dit het hom van sy krag ontneem This deprived him of his strength (iii) the occasional use, in present-day Dutch, of a possessive pronoun (e.g. zijn weg his way ) rather than a combination of a reflexive with an indefinite article (e.g. zich een weg him(self) a way ) in the 11

18 Dutch equivalent of the English way-construction (on this Dutch construction, see, e.g., Verhagen 2003) Each of these cases of ongoing change will be illustrated with corpus data, and we will address the question how sure we can be, on the basis of present knowledge, that English influence is involved. References Colleman, T On contact-induced constructional extension in present-day Afrikaans. Paper presented at the 8th International Conference on Construction Grammar, University of Osnabrück, 03-06/09/2014. Colleman, T. & D. Noël Tracing the history of deontic NCI patterns in Dutch. In: I. Taavitsainen et al. (Eds.), Diachronic Corpus Pragmatics. Amsterdam: Benjamins, Doğruöz, A. S. & A. Backus Innovative constructions in Dutch Turkish: An assessment of ongoing contact-induced change. Bilingualism: Language and Cognition 12: Gast, V. & J. van der Auwera What is contact-induced grammaticalization? Evidence from Mayan and Mixe-Zoquean languages. In: B. Wiemer et al. (Eds.), Grammatical Replication and Borrowability in Language Contact. Berlin: Mouton de Gruyter, Hilpert, M Constructional change in English. Cambridge: CUP. Höder, S Multilingual constructions: a diasystematic approach to common structures. In: K. Braunmüller & Chr. Gabriel (eds.), Multilingual individuals and multilingual societies. Amsterdam: Benjamins, Jackendoff, R Twistin' the night away. Language 73: Pietsch, L What has changed in Hiberno-English: Constructions and their role in contact-induced change. Sprachtypologie und Universalienforschung 63: Van de Velde, F. & E. Zenner Pimp my Lexis: het nut van corpusonderzoek in normatief taaladvies. In: Hendrickx E., Hendrickx K., Martin W., Smessaert H., Van Belle W., Van der Horst J. (Eds.), Liever meer of juist minder? Over normen en variatie in taal. Gent: Academia press, Verhagen, A Hoe het Nederlands zich een eigen weg baant. Vergelijkende en historische observaties vanuit een constructie-perspectief. Nederlandse Taalkunde 8:

19 Jocelyne Daems, Dirk Geeraerts & Kris Heylen KU Leuven The success of English loans in IT terminology in Belgian Dutch and Netherlandic Dutch This paper sets out to empirically test variation in the success rates of English loanwords in the two main varieties of Dutch, Belgian Dutch and Netherlandic Dutch. Crucially, we adopt a usage-based onomasiological perspective, taking the concept expressed by the loanword as starting point. This allows us to overcome a number of methodological issues present in existing loanword research (such as topic specificity; Zenner & Kristiansen 2013). Specifically, we calculate the relative preference for the loanword vis-à-vis alternative lexicalisations for a given concept (Zenner et al. 2012). For the concept bijlage attachment in Table 1, this means that the English loanword attachment is chosen approximately one out of two times in both the Netherlands (48%) and in Dutchspeaking Belgium (56%). bijlage Neth.Dutch % Belg.Dutch % attachment bijlage bijvoegsel Table 1 - Lexicalisation preferences for bijlage in quality newspapers Our analysis zooms in on differences in success rates for English loanwords in both varieties of Dutch and in different registers. Recent research on the use of English for person reference nouns (Zenner et al. 2012: 780) and in job ads and job ad titles (Zenner et al. 2013: 1050) has shown that in comparison to other lexical fields, the domain of IT is typically very successful in this respect, i.e. most English terms occur in this field. Focusing on several variables, such as length of the variant, concept frequency, age of the borrowed item, necessity of the loan (i.e. filling a lexical gap), we try to map out possible factors determining the success of 20 key concepts from within the IT domain. As such, the profile-based method (cf. Table 1), introduced in cognitive-linguistic 13

20 lexicology in Geeraerts, Grondelaers and Speelman (1999), is applied to a multi-layered dataset of over one billion words. References Geeraerts, Dirk, Stefan Grondelaers & Dirk Speelman Convergentie en divergentie in de Nederlandse woordenschat. Een onderzoek naar kleding- en voetbaltermen. Amsterdam: Meertens Instituut. Zenner, Eline & Gitte Kristiansen New Perspectives on Lexical Borrowing. Berlin: de Gruyter. Zenner, Eline, Dirk Speelman & Dirk Geeraerts Macro and micro perspectives on the distribution of English in Dutch. A quantitative usage-based analysis of job ads. Linguistics 51(5): Zenner, Eline, Dirk Speelman & Dirk Geeraerts Cognitive Sociolinguistics meets loanword research: Measuring variation in the success of anglicisms in Dutch. Cognitive Linguistics 23(4):

21 Isabelle Delaere (*) & Gert De Sutter ( ) KU Leuven(*), UGent ( ) R&D of O&O? Een multivariate corpusstudie naar de keuze tussen leenwoorden en endogene alternatieven in vertalingen. In deze bijdrage staat de vraag centraal hoe professionele vertalers omgaan met leenwoorden wanneer er ook een endogeen alternatief voorhanden is. Binnen de groep van professionele taalgebruikers is deze vraag het meest prangend voor vertalers, aangezien zij aan de ene kant vertalingen moeten afleveren die zich moeten inpassen in de doeltaal(cultuur) (en dus niet te veel mogen opvallen), terwijl ze aan de andere kant voortdurend blootstaan aan een anderstalige brontaaltekst die in min of meerdere mate syntactische en lexicale gelijkenissen vertoont met de doeltaal (waardoor de kans op syntactische en lexicale transfer groter wordt). In de vertaalwetenschap wordt dit spanningsveld aangeduid met de termen normalisatie en interferentie of shining through (Teich 2003). In dit onderzoek willen we nagaan of vertalers meer geneigd zijn te normaliseren dan niet-vertalers en in welke mate de brontaal, het register, en de aanwezigheid van een revisor daarbij een rol spelen. Daartoe gebruiken we het Dutch Parallel Corpus, een parallel, bidirectioneel corpus voor de talenparen Nederlands-Engels en Nederlands-Frans (Macken 2011). Hieraan onttrekken we 7 profielen, d.w.z. sets van onomasiologische varianten die telkens uit een Engels leenwoord (bv. unit, service) en een endogene variant (bv. afdeling, dienstverlening) bestaan. Voor de selectie van onze profielen maken we gebruik van de criteria beschreven door Zenner In een eerste stap gebruiken we profielgebaseerde correspondentieanalyse (Plevoets 2008) om de datastructuur te exploreren, daarna voeren we een binaire logistische regressieanalyse uit om het precieze effect van de genoemde factoren na te gaan. De analyses tonen aan dat vooral register een erg grote invloed heeft op de keuze tussen een leenwoord en een endogeen alternatief, zelfs in die mate dat het intuïtief verwachte brontaaleffect er helemaal door verdwijnt (het gebruik van Engelse leenwoorden in vertalingen uit het Engels ligt dus niet hoger dan in vertalingen uit het Frans). Opvallend is ook dat register enkel een rol speelt in vertaalde 15

22 Nederlandse teksten, en dus niet in origineel Nederlandse teksten. Dit zou erop kunnen wijzen dat vertalers bewuster omgaan met de keuze tussen leenwoorden en endogene varianten doordat ze voortdurend geconfronteerd worden met bronteksten in andere talen. Tot slot heeft ook de aanwezigheid van een revisor een effect op de keuze tussen een leenwoord en een endogeen alternatief, in die zin dat de aanwezigheid van een revisor een significante toename van endogene alternatieven veroorzaakt. Ook eerder onderzoek (Kruger 2012) heeft aangetoond dat de aanwezigheid van een revisor vaak leidt tot conservatievere taalkeuzes. Hoewel we geen sluitende verklaring kunnen bieden voor de geobserveerde tendensen, leiden deze wel tot een potentieel verklarende hypothese. Zo lijkt het niet onlogisch dat er binnen bepaalde registers impliciete, maar toch sterk aanwezige normen heersen die resulteren in registerspecifiek taalgebruik. Verder (kwalitatief) onderzoek zal moeten uitwijzen in welke mate dergelijke impliciete normen daadwerkelijk een rol spelen bij de keuze tussen een leenwoord en een endogeen alternatief. Referenties Kruger, H. (2012). A corpus-based study of the mediation effect in translated and edited language. Target. International Journal of Translation Studies., 24(2), Macken, L., De Clercq, O., & Paulussen, H. (2011). Dutch Parallel Corpus: a Balanced Copyright-Cleared Parallel Corpus. Meta, 56(2), Plevoets, K. (2008). Tussen spreek- en standaardtaal. Een corpusgebaseerd onderzoek naar de situationele, regionale en sociale verspreiding van enkele morfosyntactische verschijnselen uit het gesproken Belgisch-Nederlands. (Doctoral dissertation), Katholieke Universiteit Leuven, Leuven. Teich, E. (2003). Cross-linguistic variation in system and text. A methodology for the investigation of translations and comparable texts. Berlin/New York: Mouton De Gruyter. Zenner, E. (2013). Cognitive Contact Linguistics. The macro, meso and micro influence of English on Dutch. (PhD dissertation), KU Leuven, Leuven. 16

23 Karlien Franco (*), Eline Zenner (* ) & Dirk Speelman (*) KU Leuven (*), FWO Flanders ( ) Gender assignment to anglicisms in Dutch: a variationist perspective In this paper, we investigate the adaptation of anglicisms to the recipient language in a weak contact situation, focusing on actual usage. More specifically, we investigate gender assignment to recently borrowed English loanwords in Dutch. Studies on gender assignment to anglicisms have so far mostly relied on raw frequencies or simple proportions in dictionary or corpus data. As a result, gender assignment to loanwords is often conceived as a categorical process: the gender that is listed in a dictionary or found in a corpus is said to be determined by a clear set of (receptor) language-related agreement rules (see for instance Corbett 1991; Onysko 2007; Poplack, Pousada & Sankoff 1982). However, Callies, Onysko & Ogiermann (2012) show that, by solely focusing on frequencies and proportions in corpus or dictionary data, the amount of gender variation is underestimated. In line with this result, we focus on reported language usage in context, elicited by means of a questionnaire. On the one hand, we inquire into the factors that underlie gender assignment to established and unestablished anglicisms in Dutch. On the other hand, we explicitly take congruency between speakers in the language community into account. More specifically, we introduce a measure of homogeneity, based on the measure of internal uniformity (Geeraerts, Grondelaers & Speelman 1999). We investigate the language and speaker-related determinants of the degree of homogeneity between members of the language community in their choice for a definite article for an anglicism. Our data were collected by means of a forced choice task with 45 participants from Flanders. We elicited the Dutch definite article for 175 established and unestablished anglicisms. We use a range of statistical techniques, including mixed-effects logistic regression, conditional inference trees and random forest analyses, to inquire into the factors that underlie gender assignment to anglicisms in Dutch. We also make use of heatmaps and linear regression analysis to uncover which factors 17

24 determine when and why respondents disagree with each other regarding the gender of an anglicism. Our analyses show that common gender de is the default article for nominal anglicisms in Dutch. This article is used for any anglicism, unless there is some reason for choosing neuter gender het. Furthermore, our results indicate that gender assignment to anglicisms in Dutch is prone to inter and intra-speaker variation. The amount of gender variation increases when respondents are faced with a conflict between the default article de and some factor that favors its neuter counterpart, het. References Callies, M., Onysko, A. & Ogiermann, E Investigating gender variation of English loanwords in German. In: Furiassi, C., Pulcini, V. & Rodríguez González, F. (eds.). The Anglicization of European Lexis, Amsterdam/Philadelphia: John Benjamins Publishing Company. Corbett. G.G Gender. Cambridge: Cambridge University Press. Geeraerts, D., Grondelaers, S., & Speelman, D. (1999). Convergentie en divergentie in de Nederlandse woordenschat: een onderzoek naar kleding- en voetbaltermen. P.J. Meertens-Instituut. Onysko, A Anglicisms in German. Borrowing, Lexical Productivity, and Written Codeswitching. Berlin: Walter de Gruyter. Poplack, S., Pousada, A. & Sankoff, D Competing influences on gender assignment: variable process, stable outcome. Lingua

25 Stef Grondelaers Radboud Universiteit Nijmegen Do the Dutch use English to sound superior, dynamic, or both? Evidence from speaker evaluation and free response experiments Although the prestige of English is often invoked to explain the popularity of English loanwords in Dutch, there is comparatively little evidence to support the idea that English borrowings index some sort of superiority. Moreover, what little evidence there is does not seem to differentiate between two different sorts of prestige. Zenner et al. (2014: 11) demonstrated that the use of English is typical for more highly educated males which testifies to overt superiority, but also that it underlines their identity as young, modern individuals (see also Zenner & Van de Mieroop: in press), which seems to be more indicative of the covert dynamic prestige of media personalities, for whom it is more important to radiate coolness and street credibility than traditional superiority. Dynamic prestige has been shown (Grondelaers & Kristiansen 2013; Grondelaers & Speelman 2013) to motor the vitality of a number of variables and varieties which have no overt prestige (such as Flemish Tussentaal). Hence, the two prestige types appear to be in complementary distribution. In this paper, we report two perception experiments designed to investigate whether the use of English loan words in Dutch can flag both superiority and dynamism, as the available evidence seems to suggest. In a Speaker Evaluation experiment, participants from the Nijmegen and Amsterdam areas evaluated 8 speech clips representing stimuli produced with either an English loan or its Dutch alternative; stimuli were rated on scaled measures pertaining to superiority and dynamism. Although luxury loanwords had been selected for which evident Dutch alternatives are available such as victim for slachtoffer, participants did not evaluate the Dutch and English guises any differently on the two dimensions. In a design which has been shown to be sensitive to tiny cues, an absence of perceptual effects for comparatively conspicuous variants signals that these variants do not index the hypothesized meanings, or that we have failed to trigger these meanings because the manipulated variants were (still) too inconspicuous for participants to notice. 19

26 While we could design a speaker evaluation experiment in which the visibility of the critical variable is boosted by including even more noticeable loans, we also carried out a Free Response experiment (both in 2014 and 2015), in which participants were asked to return the three adjectives which first came to mind when confronted with a number of labels including Dutch which contains a lot of English loan words as in the sentence Mijn enige concern is of we de job kunnen handlen. We used distributional semantic analysis on the basis of the 0.5 billion words Twente News Corpus to cluster the adjectives returned in the Free Response-tasks into 35 evaluative dimensions; in both the and the 2025-data, the dimension Modernity (on which a number of youth and dynamism qualifications clustered) was by far the most important perceptual determinant. Possible confounds in these data, however, are that unsupervised evaluation naturally gravitates towards the negative (Kanouse 1984), and that traditional prestige correlates are not often spontaneously foregrounded (Grondelaers & Speelman: forthcoming). Before we infer from the experiments reported here that English loans in Dutch elicit modern dynamism rather than traditional superiority, an urgent concern to be addressed first is a better alignment between the experimental tools, and a more accurate view of their respective (dis)advantages. An alternative conclusion could be that for young people, the concept of prestige is shifting from top-down superiority to peer-evaluated dynamism 20

27 Eric Hoekstra & Frits Van der Kuip Fryske Akademy Peanut butter and compositionality 1. Nobody has been able to say anything deep about peanut butter, Prof. Koster claimed in an interview (Koenen 1990:37). While this is questionable enough in itself (see Donselaar 2005), it is certainly false in the field of borrowing (see Zenner, Speelman & Geeraerts 2014). With respect to borrowing, the question arises why the Dutch word pindakaas (literally peanut cheese > peanut butter ) was borrowed wholesale into Frisian as pindakaas, whereas similar collocations with kaas cheese are Frisianised to tsiis, such as Fricotsiis Frico cheese ). 2. The Dutch expression zijn leven his life > ever ) was borrowed into Frisian in the 17th century as syn leven, where leven is recognisable as a Dutch loan due to its lexical form and phonology. The fully Frisian form would be syn libben (which is also attested). Interestingly, the noun leven only appears as part of the expression mentioned above, in which the literal meaning of life has been nullified or been given a metaphorical interpretation which involves a quantification over time. Wherever the literal meaning of life is involved, Frisian consistently has libben (Anonymous). 3. What these examples have in common is that they testify to the importance of the distinction between a word s literal meaning and its metaphorical meaning (compare Onysko & Winter-Froemel 2011). To exemplify, pindakaas (literally peanut cheese ) is not some form of cheese. The examples involving leven show that this item was borrowed in a collocation in which its literal meaning was modified. 4. The example of peanet butter suggests that the principle of compositionality is at stake. Pindakaas does not easily become pindatsiis because it is not some kind of cheese. Syn leven was borrowed in its Dutch form because it does not refer to somebody s life (instead it functions as a temporal quantifier). It is generally believed that compositionality is a semantic principle, but at the same time it is generally recognised that there are many expressions in natural language which violate compositionality (such as poisoned coffee, which is not some form of coffee). This leads us to postulate that compositionality is not a semantic principle, but a pragmatic principle 21

28 that is closely related to the pragmatic principle proposed in Levinson (2000). Differences between our conceptualisation and Levinson s will be discussed. Thus borrowing is sensitive to compositionality in that words are more easily borrowed when they do not entertain a subset or superset relation to existing word fields in the borrowing language. References Anonymous anonymized reference. Donselaar, J. (2005) Pindakaas, een oud woord uit Suriname. Trefwoord 2005 oktober. Koenen, L. (1990) Het vermogen te verlangen. De theoretische taalkunde. Prof. Dr. Jan Koster Nog steeds weet niemand iets diepzinnigs over pindakaas te zeggen, Nijgh & Van Ditmar, Amsterdam. Onysko, A. and E. Winter-Froemel (2011) Necessary loans luxury loans? Exploring the pragmatic dimension of borrowing. Journal of Pragmatics 43, Zenner, E., D. Speelman and D. Geeraerts (2014) Core vocabulary, borrowability and entrenchment. A usage-based onomasiological approach. Diachronica 31 (1),

29 Aafke Lettinga, Peter Broeder & Carel van Wijk Tilburg University Digital language use: language variation in text messages The study of language variation in digital language use has received increasing attention with the advent of the text message, often accompanied by dystopic comments about the corruption of written language (e.g., Androutsopoulos 2011). However, what shouldn t be overlooked is that the genre of the text message, with its short content and its possibility of immediate response is by nature more dialogic than most other writing. This way, the register used in the messages can be expected to differ from the standard language. Text messages are often characterized by features such as abbreviations, smiley s, misspellings, rude and informal language, and ellipsis. One aspect of informal language is the abundant use of English words. The degree in which these characteristics are used depends on the social hierarchy between sender and receiver. This paper concerns an analysis of empirical data collected among Dutch students aged between 18 and 23. The students were asked to write two text messages to a peer and two to a teacher. The messages varied on two independent factors, the character of the message (a cancellation and an invitation) and the social relationship between sender and receiver. The aim of the study is to investigate how teenagers use certain loanwords or other characteristics supposedly typical of text messages to achieve social or pragmatic effects in the different social contexts. Against this background, the main research question addressed in this paper is: To what extent does language use vary in text messages depending on the content and the type of recipient? The results show that the participants are capable of switching their register when adapting to recipients from different places on the social hierarchy. In the analysis, the particular focus will be on relational, referential, and expressive aspects of the messages (cf. Schultz von Thun (2010). 23

30 References Androutsopoulos, J. (2011), Language change and digital media: a review of conceptions and evidence, In: Kristiansen, T. & N. Coupland & (Eds.), Language standardization in Europe, Oslo, Novus Press. Schultz von Thun, F. (2010), Hoe bedoelt u? Een psychologische analyse van menselijke communicatie, Groningen: Noordhoff. 24

31 Henning Radke Duitsland Instituut Amsterdam, Universiteit Amsterdam Meertaligheid en Codewisseling: De interactie tussen het Nederlands en het Sranan Tongo in Surinaamse online-communicatie Meertaligheid is een centraal element van de Surinaamse samenleving. In het kleine Zuid-Amerikaanse land worden Indo-Iraanse, Austronesische, Chinese, inheemse en Germaanse talen gesproken. Daarnaast telt de Surinaamse gemeenschap minstens zes Creoolse talen. Er bestaan verschillende meningen over het totale aantal talen in Suriname: sommigen spreken van meer dan 15 talen in dit land met ongeveer inwoners een unieke samenstelling die aanleiding geeft tot onderzoek naar taalcontact en codewisseling. In de voorgestelde bijdrage wordt de interactie tussen de twee belangrijkste talen van Suriname onderzocht, te weten het Surinaams Nederlands (SN) en het Sranan Tongo (ST), de meest gesproken Creoolse taal van het land. Daarbij ligt de nadruk op de sociolinguïstische functies en structurele invloeden die veroorzaakt worden door codewisseling (CW). Een beschrijving van lexicale en syntactische veranderingen staat centraal. Door middel van de Leipzig Glossing Rules wordt de morfosyntactische structuur van een aantal voorbeeldzinnen blootgelegd en de interactie tussen het Sranan Tongo en het Nederlands zichtbaar. Welke invloed heeft CW op de positie van de werkwoorden in het Nederlands/Sranan Tongo? En welke categorie vormen trigger- en leenwoorden in meertalige Surinaamse uitingen? Deze vragen worden onderzocht aan de hand van een corpus dat tijdens een verblijf van 3 maanden in Paramaribo werd samengesteld. Dit corpus bevat uitsluitend data uit computer-mediated communication (CMC). Voor een deel van de Surinaamse samenleving behoort CMC tot het dagelijkse leven. Door deze ontwikkeling is het mogelijk om de Surinaamse meertaligheid vanuit een nieuw perspectief te onderzoeken een perspectief dat kenmerken van mondelinge én schriftelijke communicatie bevat (zie Dorleijn 2008: 127). Tot de mondelinge kenmerken behoort onder meer het net genoemde verschijnsel codewisseling. 25

32 De voorgestelde presentatie wil een bijdrage leveren aan een relatief nieuw gebied van de taalkunde waarvan CMC de basis vormt. Daarnaast wordt de studie van kleine talen en de Surinaamse variëteit van het Nederlands gesteund. Belangrijke referentiekaders zijn o.m.: - het Woordenboek Surinaams-Nederlands (2009) van Renata de Bies - de theorieën van Gumperz (1982) - het Matrix-Language Model van Carol Myers-Scotton (1993; 2006). - A typology of Code-Mixing, Pieter Mysken (2000) Referenties Margreet Dorleijn & Jacomine Nortier (2008): Code-switching and the internet. In: Barbara E. Bullock & Almeida Jacqueline Toribio (eds.): The Cambridge Handbook of Linguistic Code-switching. Cambridge: Cambridge University Press, pp

33 Freek Van de Velde (*) & Eline Zenner (* ) KU Leuven (*), FWO Flanders (* ) Pimp my Lexis: Constructional borrowing. Contact linguistic research has repeatedly debated the distinction between borrowing and codeswitching, particularly concerning the status of single-word switches. More recently, it has been proposed to let go of the idea of a sharp dichotomy between both phenomena and instead consider them as endpoints on a continuum (e.g. Matras 2009). Single-word switches should then be considered as a grey-zone on this continuum: depending on the context of use and the profile of the user, they can be more like borrowing or more like codeswitching. This paper draws attention to another grey-zone on this continuum from borrowing to codeswitching, namely (semi-)fixed foreign expressions and constructions (e.g. English as good as it gets occurring in Dutch) (cp. Doğruöz & Backus 2009). More specifically, we study recipient-language creativity in moulding fixed and conventionalized source language expressions. We zoom in on one specific expression: in 2005 the multi-word unit Pimp my Ride was introduced in Dutch as a proper name. Through an analysis of corpus extracts from newspaper language, we demonstrate how the expression went through a period of increasing constructional flexibility (from [pimp my ride] over [pimp my N] to [pimp POSS N]) and semantic extension (from [pimp my CAR] over [pimp my VEHICLE] to [pimp my N]). Eventually, this evolution led to the introduction of a new verb (pimpen opleuken ) in Dutch. As we will discuss, the rapid spread of both the construction and the verb are surprising for a number of reasons (for instance the rather salacious nature of the original meaning of the word pimp, the proper name status of the original expression, and the existence of a valid Dutch alternative (opleuken)). At the same time, the propagation of the term can be framed against the strong position of Anglo-American pop culture in the Low Countries. As such, we will open up the debate of the influence of TV and media on language change (see Labov 2001). 27

34 Hielke Vriesendorp Universiteit Leiden Omg zo fashionably English. Codeswitching to English as an Identity Practice for Gay Adolescents In the research of De Decker and Vandekerckhove (2012) the presence of English in the chatspeak of Dutch-speaking youths was found to be considerable: 13.3% of chat messages in their corpus contained at least one English word. These occurrences were almost exclusively cases of borrowing. However, in a similar corpus with myself as the central chatter not only many English borrowings occur, but also a high number of multi-word and multi-sentence code-switches. A quantitative analysis shows that in the conversations between adolescents who both construct a gay identity code-switches to English are used far more often than in other conversations. A queer linguistic, third-wave sociolinguistic analysis of this is that these code-switches can be accounted for as an in-group identity practice for these adolescents constructing a gay identity. A discourse analysis shows that chatters who construct a gay identity associate English with gay coolness and a conversation-analytical approach shows that multi-word code-switches to English are used to activate gay-cool connotations. 28

35 29

36