LINGUIST List 21.3385

Tue Aug 24 2010

Review: Sociolinguistics: Norde et al. (2010)

Editor for this issue: Joseph Salmons <>

        1.    Jessamyn Schertz, Language Contact

Message 1: Language Contact
Date: 24-Aug-2010
From: Jessamyn Schertz <>
Subject: Language Contact
E-mail this message to a friend

Discuss this message

Editor's note: This issue contains non-ISO-8859-1 characters. To view the correct characters, go to

Announced at

EDITORS: Muriel Norde, Bob de Jonge, Cornelius Hasselblatt TITLE: Language Contact SUBTITLE: New perspectives SERIES TITLE: IMPACT: Studies in Language and Society 28 PUBLISHER: John Benjamins YEAR: 2010

Jessamyn Schertz, Department of Linguistics, University of Arizona


The articles in this volume are based on presentations at ''Language Contact in Times of Globalization'', held at the University of Groningen in September 2006. In their introduction, the editors emphasize the widening scope of language contact studies, which ranges from the detailed investigation of a single contact situation to cross-linguistic comparison of contact phenomena, and which extends to the social effects that language contact can trigger in communities. These ten articles, which span a variety of language families, research methodologies, and subfields of linguistics, give the reader a taste of the diversity of the work being done on contact phenomena and at the same time raise many questions for future research.

1. Pieter Muysken: Ethnolects as a multidimensional phenomenon

In the opening paper, Muysken proposes a broad framework for the 'multidimensional' study of ethnolects, varieties of a language spoken by particular ethnic groups. Using primarily examples from the Dutch 'Roots of Ethnolects' project, he provides examples demonstrating that ethnolects are affected by properties of second language acquisition, linguistic marking of identity, language mixing, and properties of the ''original'' language of the ethnic group, but that ethnolects cannot be sufficiently defined by any of these individually. While much work on ethnolects focuses on direct comparison of the ethnic variety with the 'standard' variety of the dominant language, Muysken proposes complementing these studies with a framework that takes into consideration the following factors: the local variety of the 'original' language of an ethnic group (L1); crossing or mixing between the original and the ambient language (L1/L2); 'universal principles' of simplification of grammatical elements (UP); and approximation to the ambient language (L2). It is not clear where sociolinguistic factors might fit into this framework, and in particular which dimensions would be relevant for speakers of ethnolects who are not necessarily speakers of the ''original'' language but who still are speakers of a pronounced ethnic dialect, and whose stylistic choices may even drive dialect change (cf. Mendoza-Denton 1997 on 'Chicano English'). Nevertheless, Muysken's framework definitely allows for more nuance than a single-dimension comparison with the ambient language and can be extended beyond ethnolects to practically any contact language situation.

2. John Nerbonne, Timo Lauttamus, Wybo Wiersma, Lisa Lena Opas-Hänninen: Applying language technology to detect shift effects

Nerbonne et al. present the results of a computational project comparing the surface syntax of first and second generation Finnish Australian English speakers. Using automated coding of parts of speech of transcribed conversational corpora, the program extracted groups of three words and their corresponding parts of speech for the two groups, allowing the authors to identify and quantify patterns of deviation in the distribution of parts of speech. The automated technique has the additional advantage that it not only reveals the speech errors of L2 speakers, but also identifies and gives a reliable measure of overuse of certain (correct) constructions disproportionately favored by L2 speakers, a phenomenon not easily addressed using traditional linguistic analysis. The authors go on to discuss the results in the context of language contact, showing that some, but not all, of the discrepancies are attributable to the speakers' L1 (Finnish). The presentation is very non-technical, the focus being to present discussion of the results to a linguistic audience and, perhaps more importantly, to show how techniques from speech technology can be used to answer linguistic questions not usually addressed with computational methodologies.

3. Ricardo Otheguy, Ana Celia Zentella and David Livert: Generational differences in pronominal usage in Spanish reflecting language and dialect contact in a bilingual setting

The volume continues with another article comparing syntactic structures in first and second generation immigrants, this time focusing on null vs. overt subject pronoun use in the Spanish spoken by newcomers to New York City. Using as a data source the Otheguy-Zentella corpus of NYC conversational Spanish, the authors find that overt pronoun use is higher among second-generation immigrants than it is among newcomers. Furthermore, speakers from the Caribbean showed slightly different manifestations of this pattern than those from the mainland: while for Caribbean Spanish speakers there is a larger increase in 1st and 3rd person than there is for second person, for mainland speakers it is reversed: there is a larger increase for 2nd person than there is for 1st and 3rd person. The authors attribute the global increase in pronouns to influence from English, which almost always require overt subject pronouns, but which otherwise shows general surface syntactic similarity to Spanish pronoun/verb use. This idea is supported by the fact that time spent in the city and knowledge of English correlate positively with higher overt pronoun use, while age of arrival and knowledge of Spanish correlate positively with higher null pronoun use. The 'mirror image' effect of second-person pronouns is attributed to leveling of the Caribbean and Mainland dialects. Thus Spanish speakers in New York are adapting not only to English, but also to other dialects of Spanish.

4. Piibi-Kai Kivik: Personal pronoun variation in language contact: Estonian in the United States

This paper again focuses on variation of pronoun use by immigrants to the United States, this time looking at Estonian, which has two sets of personal pronouns, 'long' and 'short,' and allows null subjects in the first and second persons. Data was collected via sociolinguistic interviews from three groups: two groups of late bilinguals (World War II refugees who came over after secondary school and new immigrants), and ''early bilinguals,'' WWII refugees who had typically arrived in the United States before secondary school. Reasons for variation between long and short pronouns were globally the same for American Estonian speakers as for monolingual Estonian speakers; that is, determined by prominence (focus or contrast) and case. However, specific patterns of variation differed between the groups. As in Otheguy et al.'s study, early bilinguals showed greater use of overt subject pronouns, perhaps again because of greater accommodation to English. Some other interesting results were noted; for example, group members who diverged from the pattern of their group often had speech closer to that of another group with whom they were in close contact.

5. A. Seza Doğruöz and Ad Backus: Turkish in the Netherlands: Development of a new variety?

This article focuses on potential effects of contact on conversational Turkish spoken by second-generation immigrants to the Netherlands. Working in the framework of Construction Grammar (Goldberg 2005), the authors present examples from conversations that are considered to be unconventional and show how some of the deviations from the standard variety appear to be able to be attributable to structural influence from Dutch. For example, Turkish speakers in the Netherlands sometimes omitted case-markers on nouns that are case-marked in standard Turkish, which could be explained by the fact that Dutch does not mark case. However, when conversations with native Turkish speakers from Turkey were analyzed, similar unconventional constructions occurred, albeit less frequently, suggesting that the unconventional structures cannot be completely attributed to Dutch influence. There are several types of structures that only occur in Netherlands Turkish speakers (word order patterns, addition of a lexical item, omission of possessives, and unconventional derivational morphemes), but as these do not appear to be systematic, the authors conclude that there is not sufficient evidence to posit Dutch-influenced structural change in Netherlands Turkish. Instead, they claim that the influence of Dutch is item-based, with unconventional constructions resulting from translations of equivalent Dutch expressions.

6. Charlotte Gooskens, Renée van Bezooijen and Sebastian Kürschner: The reflection of historical language contact in present-day Dutch and Swedish

The authors of this paper discuss how the distribution of loanwords in the Dutch and Swedish lexicons reflects the similarities and differences in the linguistic contact situations of the languages throughout history. Based on carefully controlled corpora of speeches made in the European Parliament, words were coded for loanword status and language of origin. The authors then analyze the data based on a combination of linguistic and historical facts: for example, the prevalence of French loanwords in Dutch reflects the greater extent of contact between Dutch and French populations than Swedish and French. On the other hand, Swedish has a large number of loanwords from Low German, while Dutch has almost none, even though the two language communities had a similar amount of contact with Low German-speaking communities in the Middle Ages. The authors argue that the reason for this is that the lexicons of Low German and Dutch were very similar, leaving little room for borrowing, while those of Low German and Swedish were considerably different.

7. Hélène B. Brijnen: The impact of German on Schleife Sorbian: the use of 'gor' in the Eastern Sorbian border dialect

This short article focuses on the use of the particle 'gor' in the Schleife dialect of Sorbian, a Slavic language which is spoken in a small patch of Eastern Germany and which has been surrounded by the German speakers since the Middle Ages. After a summary of the sociolinguistic context of the language and its speakers, the author discusses the geographical distribution of 'gor,' a borrowing from the German 'gar,' pointing out that the particle occurs in Lower Sorbian but is rare in Upper Sorbian. The fact that there are other German borrowings in Upper Sorbian dialects leads her to suggest that 'gor' was not borrowed because it did not fit with the phonological system of Upper Sorbian, which had no /g/. She also notes that while the German equivalent 'gar' can be used in both positive and negative contexts, in this dialect of Sorbian it is almost always used in a negative context. Brijnen goes on to compare the contemporary dialect to that of a nineteenth century writer, claiming that comparative study can shed light on the history of German as well as Sorbian, but leaves the details for future work.

8. Wilbert Heeringa, John Nerbonne and Petya Osenova: Detecting contact effects in pronunciation

Taking a quantitative approach to the relationship between linguistic and geographic proximity, this article examines whether the dialects of Bulgarian which are closer to neighboring countries (Macedonia, Serbia, Romania, Greece, and Turkey) are phonologically more similar to the languages of these countries. The article is valuable not only for its analysis of the data, but also for its comparison of three computational techniques for measuring phonological similarity. Each step of the analysis was carried out using 1) Levenshtein distance (the number of modifications necessary to change one string of sounds to another); 2) phone frequency method (PFM), which compares the frequency of each phone in comparable corpora (Hoppenbrouwers & Hoppenbrouwers 2001), and 3) feature frequency method (FFM), which compares the frequency of each phonological feature in comparable corpora (Hoppenbrouwers & Hoppenbrouwers 2001). Although the three techniques produced slightly different results, indicating the need for further examination, they generally concurred, showing a correlation between linguistic similarity and geographic proximity with respect to Macedonian, Serbian and Romanian, while for Greek and Turkish there were negative trends and correlations. The authors offer possible historical and sociolinguistic explanations for the lack of correlation for Greek and Turkish.

9. Jason Shaw and Rahul Balusu: Language contact and phonological contrast: the case of coronal affricates in Japanese loans

Shaw and Balusu present an in-depth report on the results of a study on the [ti]-[tʃi] distinction in Japanese loanwords in two generations of speakers. Analysis of target loanwords including the two sequences found that both younger (age 20-23) and older (50-56) speakers distinguish orthographic 'chi' from 'ti' (which occurs only in loanwords), with longer frication on 'chi' tokens and shorter frication on 'ti' tokens. The authors also found that the duration of frication in both categories is conditioned by prosody: for instance, accented syllables have longer frication than unaccented syllables. Although there is a contrast between [ti] and [tʃi] in each prosodic position, the older speakers show overlap between the two categories when collapsed over all prosodic positions: that is to say that a stressed [ti] could have more frication than an unstressed [tʃi]. However, for younger speakers, although prosodic effects are still present in each category, [ti] and [tʃi] do not overlap. Interestingly, in the older speakers, two [ti] categories seemed to emerge: loanwords which had been introduced after the speaker had reached adulthood were more [ti]-like (less frication). The authors conclude that older speakers had acquired the [ti]-[tʃi] contrast later in life, while younger speakers learned it while learning their first language. Finally, the authors address the question of why this particular non-native contrast might be preserved, while others (i.e. [r] vs. [l]) are not. They propose that there must be non-contrastive variation within a native category, such as the prosodic conditioning that occurs in this case, in order for non-native categories to form.

10. Nicola Borrelli: Translating cultures within the EU

Turning to a very different perspective on language contact, the final article in the collection deals with the translation of European Union documents into different member languages, in an attempt to understand ''to what extent the translations of Brussels' official documents mirror the specific national perspectives of their translators and how these localising spurs interact with the general policies of the European Union'' (182). Borrelli focuses on the English and Italian translations of an originally French video meant to promote the European Constitution to the public. Citing specific examples of non-literal translations, the author concludes that the contrast between Italian 'Euro-optimism' and UK 'Euro-scepticism,' as well as national cultural values (based on Hofstede's (2001) 'dimensions of cultural differences') is reflected in the choices made by the translators. Some of the discrepancies are indeed interesting; using texts from more than one translator in each language would provide even more convincing evidence that the choices of wording are an indication of the cultural values of a nation, as claimed by the author, rather than the stylistic choices of an individual writer.


In his opening article, Muysken proposes that ethnolects need to be analyzed from many perspectives, and the findings in this book attest to the need for multidimensional perspectives on language contact in general. The main strengths of the volume are its broad coverage of many different facets of language contact, demonstrating effects that a first language can have on a second (Nerbonne et al.), a second language on a first (Otheguy et al., Kivik et al., Doğruöz and Backus), areal effects that are not necessarily attributable to knowledge of a second language (Brijnen, Heeringa et al.), loanword phenomena (Gooskens et al., Shaw and Balusu), and translation studies (Borrelli). This diversity gives the reader an idea of the many paths that the study of contact linguistics can take (and points towards many more yet untraveled), even though the lack of discourse between the papers causes it to read more as a sampling of the field than as a unified volume.

Although most of the studies are corpus-based, there is a nice breadth of methodology as well; for example, the two computational studies demonstrate that automated quantitative techniques can reveal previously undiscovered patterns about both the syntax (Nerbonne et al.) and the sound systems (Heeringa et al.) of languages in contact, while detailed examinations of specific constructions uncover processes not apparent on the surface (such as the sub-categorization of Japanese /ti/ presented by Shaw and Balusu). Although some work is set within a theoretical framework (i.e. Construction Grammar), the focus of the majority of the articles is largely descriptive, followed by possible implications of the findings for the language contact processes. As most of these articles are quite self-contained, however, the reader is often left without sufficient evidence to evaluate the larger-scale claims. This is especially striking in light of the work of Doğruöz and Backus, who highlight the importance of control groups in corpus work: after presenting seemingly plausible speculation about contact effects, they then show how these ideas are refuted by the fact that speakers without exposure to the ''contact'' language produce the same patterns.

The articles in this volume provide a rich source of interesting contact phenomena, too specific and nuanced to be encompassed by the blanket term ''accommodation.'' The next step is to explore whether these patterns have parallels across languages. In a field with such extraordinary breadth and diversity, it is only by cross-linguistic comparison and extensive dialogue between researchers that it will be possible to begin to answer some of the big-picture questions posed by many of the authors as areas for future research: for instance, which sorts of patterns or constructions are most susceptible to change, and which are more resistant to contact influence. Addressing these questions will lead to a better understanding not only of language contact, but also of first and second language acquisition, and this volume provides an excellent starting point for exploring them in a systematic manner.


Goldberg, A. 2005. Constructions at work. The nature of generalization in language. Oxford: OUP.

Hofstede, G. 2001. Culture's consequences: comparing values, behaviors, institutions, and organisations across nations. Thousand Oaks CA: Sage.

Hoppenbrouwers, C. and G. Hoppenbrouwers. 2001. De Indeling van de Nederlandse Streektalen. Dialecten van 156 Steden en Dorpen Geklasseerd Volgens de FFM. Assen: Koninklijke van Gorkum.

Mendoza-Denton, Norma. 1997. Chicana/Mexicana identity and linguistic variation: an ethnographic and sociolinguistic study of gang affiliation in an urban high school. PhD dissertation, Stanford University.


Jessamyn Schertz is a graduate student at the University of Arizona, on leave this year for a research position with the European Union Sound to Sense project at Radboud University in Nijmegen, the Netherlands. Her main research interests lie in sociophonetics, the phonetics/phonology interface, and the effect of language contact on sound systems.

Page Updated: 24-Aug-2010