Convergence vs. divergence from a diasystematic perspective

Convergence vs. divergence from a diasystematic perspective* Stefen Höder Convergence and divergence are usually deined as changes in opposite directions – convergence increases, divergence decreases interlingual similarities between two given languages or varieties. Additionally, convergence is oten explained as the ‘natural’, expectable process in language contact, whereas divergence is associated with psychosocial mechanisms. Based on observations from the recent development of Low German in its present intense contact with High German, this contribution argues that the distinction between convergence and divergence is not as straightforward as it seems and that it is not convergence as such that can be explained without the involvement of any extralinguistic factors, but rather pro-diasystematic change (as opposed to counter-diasystematic change) – i.e. innovations that facilitate the establishment of languageunspeciic structures in a common constructional system.

1.

Introduction

While everyone agrees that ‘convergence’ and ‘divergence’ constitute a central conceptual pair in contact linguistics, deinitions given for the terms vary, and both are oten used without reference to any particular deinition.1 It is, however, commonly assumed.

* I wish to thank Karoline Kühl and Kurt Braunmüller for their valuable comments on an earlier version of this paper.

S F O NY A O P M R P G CO

1. Unfortunately, ‘convergence’ (then without ‘divergence’ as a counterpart) is also used by some scholars with the general meaning of ‘transfer’, ‘interference’, ‘interlingual element’, or a particular type of transfer phenomenon (cf. the usage in Myers-Scotton’s [2002: 101f.] terminology). his usage of ‘convergence’ is unrelated to the issue addressed here.

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

40

Stefen Höder

that convergence and divergence are epiphenomena, since languages2 or speciic structures converge or diverge as a result of the underlying basic mechanisms of language contact and change, caused by a range of interacting psychological, social, and cultural factors (cf. Hinskens, Auer & Kerswill 2005: 17f.), and ii. that convergence and divergence are antonyms, meaning that an innovation represents either convergence or divergence in relation to a speciic pair (or set) of languages (cf. Weinreich’s [1954: 395] classic deinition). i.

his is not so say that convergence and divergence are mutually exclusive: It goes without saying that if there is convergence between (certain structures in) languages A and B, the same process may also entail divergence in relation to other language pairs, such as AC or BC. In this regard, convergence and divergence can certainly be “two sides of the same coin” (Hinskens, Auer & Kerswill 2005: 2). We would expect, though, that iii. it is impossible for a given process to be both convergence and divergence in relation to the same pair of languages. he focus of the present chapter is theoretical. Its principal aim is to demonstrate that the expectation in (iii) is not as self-evident as it might seem, and that there are indeed cases in which convergence at the same time is divergence, as exempliied on the basis of observations from the recent development of Low German in its present intense contact with High German varieties. If the distinction between convergence and divergence thus turns out to be blurrier that usually assumed, then this also afects the possible association of either process with speciic factors in language change. Consequently, this chapter follows up with the question whether or not it is possible to maintain the (more or less standard) view that convergence represents the natural outcome of language contact, whereas divergence is the result of social rather than cognitive processes in language contact situations (Kühl & Braunmüller, this volume). Recent theoretical work on the organisation of linguistic knowledge in language contact situations suggests that multilinguals in language contact situations do not keep the original language systems separate, but rather develop a common system including language-speciic and language-unspeciic elements (cf. the work on Diasystematic Construction Grammar [DCxG] by Höder [2011, 2012,

S F O NY A O P M R P G CO

2. Henceforth, this contribution does not distinguish categorically between languages and varieties unless explicitly noted. Similarly, no distinction is made between bi- or multilingualism and bi- or multi(dia)lectalism. he more general terms are meant also to encompass the more speciic ones.

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

Convergence vs. divergence from a diasystematic perspective

41

2014], which is based on the cognitively and typologically oriented constructionist approaches by Goldberg [1995, 2006] and Crot [2001]). From this perspective, we should expect that systematic – i.e. primarily cognitively motivated – mechanisms indeed explain innovations that facilitate the establishment of languageunspeciic structures and simplify such structures within a common system. While such innovations oten lead to convergence, they may also involve certain types of divergence, as the discussion of some examples will show. his chapter is organised as follows: Section 2 provides a brief discussion of the deinition and scope of the terms ‘convergence’ and ‘divergence’. Section 3 introduces the diasystematic perspective and elaborates on the concept of ‘prodiasystematic change’. Section 4 illustrates this concept and its relation to convergence/divergence on the basis of example cases from recent Low German. he chapter concludes with a short summary in Section 5.

2. Deinition and types of convergence and divergence In terms of a broad deinition, convergence and divergence can be said to be types of relational diachronic language change – as opposed to diachronic stability – in which two or more given languages become structurally more similar or dissimilar, respectively (cf. Weinreich 1954: 395; Hinskens, Auer & Kerswill 2005: 1f.; Harnisch 2010: 275; Kühl & Braunmüller, this volume). As to the exact scope of the two terms, some additional remarks may be in place. Firstly, convergence and divergence as deined above do not imply bidirectionality. For example, Mattheier (1996: 34) argues in favour of a terminological distinction between bidirectional ‘convergence’, then deined as a type of diachronic change by which varieties A and B become more similar to each other by adopting intermediate forms, and unidirectional ‘advergence’, which he deines as the adoption by variety A of elements pre-existing in B (or similar to B), with B remaining unchanged.3 3. While it is evident that such a more ine-grained terminology is useful, there are, in fact, more logical possibilities. For example, A and B may converge by moving in the direction of a third language C without picking up features from each other or adopting intermediate forms. his is what happens typically in cases of vertical dialect-standard convergence, with two (possibly non-contiguous) dialects A and B becoming more similar to each other as a consequence of their advergence to a rooing standard variety C. (For example, the syntax of Bavarian dialects may become more similar to Ripuarian as a result of Standard German inluence.) his is diferent from processes of dialect levelling or koineisation, when there is uni- or bidirectional convergence between A and B as a result of direct dialect contact (as, for instance, between Bavarian and East Franconian dialects).

S F O NY A O P M R P G CO

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

42

Stefen Höder

Secondly, convergence and divergence do not have to afect whole languages, but can also cause the emergence of new varieties. For example, the contact between Standard (High) German and Low German in Northern Germany, particularly intense during the 20th century, resulted in the establishment of a new regional dialect (regiolect) of High German (North High German; see Section 4.1), a converged variety, while the remainder of the High German varieties did not undergo similar changes. hirdly, deined as changes relating to the structure of two or more languages, convergence and divergence can be restricted to speciic parts of the language system (e.g. sentence intonation or verbal syntax) and apply to diferent aspects of linguistic structure. he perhaps most useful distinction is the one between formal and functional change. If two languages A and B possess an equivalent lexical or grammatical concept but employ diferent forms to express it (‘one function, two forms’), formal convergence/divergence leads to an increase/decrease in formal similarity; eventually, formal convergence may result in total congruence between the equivalents in A and B (‘one function, one form’).4 For example, the lexical concept ‘Saturday’ is traditionally expressed diferently in South German and North German regiolects. Recently, though, North German speakers tend to take over the southern form Samstag at the expense of the traditional form Sonnabend (cf. Low German Sünnavend). On the other hand, if two languages A and B employ diferent concepts (‘two functions, one/two forms’), there is a possibility of functional convergence/divergence; functional congruence between the equivalents in A and B (‘one function, one/two forms’) is a potential endpoint of functional convergence. For example, Standard German Konzept is today used to express the meaning ‘idea, term’, in addition to the original ‘drat, plan’, in parallel to English concept, resulting in functional equivalence and an increase in formal similarity between the two languages. here is no a priori reason why functional convergence should always be accompanied by formal convergence or vice versa: in principle, formal and functional changes can be antagonistic, and either type of formal or functional change can also co-occur with formal or functional stability. A fourth, and perhaps trivial, remark concerns the role of language contact. Although ‘convergence’ and ‘divergence’ are mostly used in contexts where language contact does play a role, both concepts are basically descriptive in nature and do not necessarily refer to contact-related phenomena. Descriptively

S F O NY A O P M R P G CO

4. Whether interlingual congruence, in this sense, is tantamount to identity, i.e. whether two congruent elements in A and B are ‘the same’, depends on the theoretical perspective. From a structuralist point of view, for example, they can still have diferent values in relation to the linguistic systems of A and B in their entirety. From a diasystematic point of view, in contrast, congruence implies a loss of language-speciicity (cf. the discussion in Section 3).

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

Convergence vs. divergence from a diasystematic perspective

43

speaking, isolated languages may converge, too (and we would expect them to do so in certain cases, for example when typologically salient features of a language are lost). Furthermore, it is conceivable that convergence in language contact is nevertheless independent of contact (for example when typologically frequent features emerge in two languages which happen to be spoken in adjacent areas) or that divergence in contact is caused by contact-independent factors (see Kühl & Braunmüller, this volume). Figure 1 summarises the diferent criteria in a tentative classiication. diachronic development

stability change

type

convergence divergence

role of contact

none contact-induced

directionality

bidirectional unidirectional whole language emerging variety

part of system

verbal syntax …

structural aspect

form function

Figure 1. Tentative classiication of convergent/divergent changes

3. Convergence vs. pro-diasystematic change Convergence and divergence in language contact situations obviously presuppose some degree of individual and social multilingualism. Traditionally, however, linguistic theory has viewed and treated languages in contact as coexisting, but distinct systems (or, as the generative tradition has it, distinct grammars), and the inevitable contact phenomena have been analysed in terms of transfer from one system to another. More recent research, however, proceeds on the assumption that there is no such strict separation in the way multilinguals handle their languages cognitively (cf. Grosjean 2008: 13f.), and some scholars even argue in

S F O NY A O P M R P G CO

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

44

Stefen Höder

favour of overall ‘linguistic repertoires’ (Matras 2009: 308f.), i.e. sets of elements from which multilinguals choose the ones that are appropriate in a particular communicative situation, but which are not essentially part of a speciic language system. his section discusses how this insight can afect our view on convergence and divergence phenomena. As Matras (2009: 3) puts it, the locus of language contact is “the language processing apparatus of the individual multilingual speaker and the employment of this apparatus in communicative interaction”. Language contact and contactinduced change have thus at least two dimensions, namely a cognitive and a social one, and involve both individual speakers and speaker groups. Firstly, convergence and divergence presuppose innovations that are initiated, spread, and potentially accepted by individual speakers with some (if only partial and receptive) knowledge of the two or more languages involved. To put it simply, speakers of A need to know enough of B to be able to adopt or avoid particular elements in B successfully. More speciically, a necessary precondition for convergence and divergence is that multilingual speakers are able to process and access the relevant structures in A and B as interlingual equivalents, i.e. to identify mutual counterparts in the two languages. his mechanism, known in contact linguistics as ‘interlingual identiication’ (Weinreich 1964: 7f.), is not self-evident, but arises from a creative and in part arbitrary association by multilingual speakers, motivated by formal or functional similarities between A and B, including phonic, semantic, morphological, syntactic, pragmatic, and frequential features (Höder 2012: 245f.). Secondly, convergence and divergence require communities (however small, unstable, or luctuating they may be) to adopt the respective innovations as part of their linguistic system. he arbitrariness of interlingual identiication makes the social dimension even more important: Multilingual individuals can, of course, link elements in diferent languages in an ad hoc fashion, whereas for these innovations to become permanent changes, interlingual equivalences have to be conventionalised within the relevant speaker group. Such interlingual relations (cf. Figure 2) can be labelled diasystematic links, following Weinreich’s (1954) original concept of ‘diasystem’, deined by similarities between two or more systems and embedded in multilingual speakers’ cognition and language use (Weinreich 1954: 390).5 Note that, despite the rather unfortunate practice in structuralist dialectology (cf. Moulton 1960), diasystems in this sense cannot be constructed at will by linguists, but represent parts of speakers’ actual linguistic knowledge.

S F O NY A O P M R P G CO

5. he term ‘diasystem’ is sometimes used to refer to the set of varieties belonging to one standard language (or under the roof of one standard variety), classiied along the diatopic, diastratic, diaphasic and possibly diachronic or diamesic dimensions. his is conceptually unrelated to the structural deinition of a diasystem as posited by Weinreich.

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

Convergence vs. divergence from a diasystematic perspective

45

interlingual links

A

B

monolingual elements

Figure 2. Monolingual elements and interlingual links

Instead of approaching diasystematic links as an extra layer of sui generis elements on top of the language-speciic ones, the framework of Diasystematic Construction Grammar (DCxG)6 provides an alternative view. he basic principle of DCxG is cross-linguistic generalisation in multilingual environments: Multilingual speakers organise their linguistic knowledge by means of abstraction and generalisation on the basis of the available input, and this generalisation goes as far as possible, regardless of any pre-existing boundaries between languages A and B. he speakers’ linguistic knowledge thus constitutes one combined system, in which some elements – in DCxG terminology, idioconstructions – are language-speciic, i.e. they are used for a particular set of communicative purposes and contexts conventionally associated with language A or B in the multilingual community, while others – diaconstructions – lack such pragmatic restrictions and are accordingly unspeciied for language (cf. Figure 3). Diaconstructions and idioconstructions are connected by inheritance links and thus form a constructional network.7 he main advantage of the DCxG approach is that it regards multilingualism as an inherent part of the language system. hus, it not only does away with the problems caused by the assumption of separate linguistic systems in linguistic theory (cf. the general critique by, for example, Durrell [2004: 198f.]), but also ofers a more realistic view of the way multilinguals achieve an economic reduction of

6. For reasons of space, the theoretical characteristics of DCxG or construction grammar in general cannot be discussed in more detail here. For an extensive survey of DCxG, see Höder (2012, 2014); Hofmann & Trousdale’s (2012) handbook provides an up-to-date overview on diferent approaches and current developments in construction grammar.

S F O NY A O P M R P G CO

E S PUBLISHIN PANG JAMIN

7. he general principle is also transferable into other theoretic frameworks. Peterson (forthc.) takes a similar approach in his work on inter- and intralingual variation in Role and Reference Grammar.

© JOHN

BE

2nd proofs

46

Stefen Höder

diaconstructions

inheritance link

B

A

idiosyncratic constructions

Figure 3. Idiosyncratic constructions and diaconstructions

their ‘linguistic burden’ (Weinreich 1964: 8), i.e. the cognitive cost of maintaining two separate language systems: For multilingual speakers, the establishment of diaconstructions is a simpliication of their linguistic knowledge, in the same way as the establishment of any type of (monolingual) abstract pattern is a simpliication. Diachronically speaking, the expectable consequence of this type of simpliication is ‘pro-diasystematic change’: a reduction in the overall number of idiosyncrasies and a corresponding increase in the number of diasystematic elements across the relevant speaker group. Every instance of pro-diasystematic change in a particular language contact situation can thus be interpreted and understood as a simpliication of the multilingual system, while in principle being neutral as to the complexity of the individual language systems involved. he reverse process, ‘counter-diasystematic change’, leads to an increase in idiosyncrasies and a reduction of common structures. While counter-diasystematic change is, of course, not in the least ruled out, it cannot be explained by the same system-internal mechanisms as pro-diasystematic change, but has instead to be the result of intervening, non-cognitive, presumably sociolinguistic factors – such as strategies for multilinguals to explicitly mark the diferences between communicative contexts or diferent groups, ethnic loyalty, or local identity (see, for example, the remarks on naboopposisjon in Kühl & Braumüller, this volume). One of the most typical scenarios for contact-induced change is a gradual development towards functional congruence, i.e. the establishment of one-to-one correspondences between functionally equivalent structures, while formal similarity is not necessarily a part of that process. To mention but a few examples, Aikhenvald (2007: 28f.) mentions ‘morpheme-per-morpheme intertranslatability’ as a characteristic areal pattern resulting from long-term language contact. Heine & Kuteva (2005: 179f.) point out that ‘exact structural equivalence’ is oten to be found as a consequence of contact-induced grammaticalisation. Ross’s (2007) concept of ‘metatypy’, elaborated on the basis of contact-induced syntactic restructuring in Oceanic languages, can easily be related to the idea of diasystematicity.

S F O NY A O P M R P G CO

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

Convergence vs. divergence from a diasystematic perspective

47

From a diasystematic perspective, these types of change can be characterised as the establishment of multilingual elements with one function and diferent language-speciic forms within one common system (see Figure 4). If we return to the earlier example of English concept vs. German Konzept ‘concept, idea [in addition to original ‘drat, plan’]’ (Section 2), we can conclude that the functional convergence between the two lexemes constitutes an instance of pro-diasystematic change for any relevant bilingual speaker group such as, say, German-speaking linguists who use both languages in academic contexts. (form) functionA/B

formA

formB

functionA

functionB formA functionA/B

formB functionA/B

Figure 4. Functional convergence as pro-diasystematic change

At irst sight, then, one might leap to the conclusion that ‘pro-diasystematic change’ is synonymous with ‘convergence’ and ‘counter-diasystematic change’ is just another term for ‘divergence’. hat would, however, be misleading. For one thing, the perspective on language systems in contact is essentially diferent: he idea of diasystematicity and pro- vs. counter-diasystematic change relates languagespeciic structures to common ones within the same system, while ‘convergence’ and ‘divergence’ are relational to language-speciic systems. For another thing, convergence and pro-diasystematic change are not coextensive. he evidence discussed in Section 4 shows (a) that diferent types of convergence amount to different degrees of diasystematicity, and (b) that pro-diasystematic change can even consist in divergence phenomena.

4. Exemples: Recent Low German 4.1

Background

Low German (LG) is the traditional regional language of Northern Germany, consisting of a dialect continuum under the roof of Standard German. hroughout its history, LG has been in contact with High German (HG) dialects and the developing HG standard variety, and has gradually lost ground to HG from Early Modern Times onwards, as a result of a successive and still ongoing language shit. Today, LG is used as a language of everyday communication in informal

S F O NY A O P M R P G CO

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

48

Stefen Höder

domains by a relatively small minority (maximally 14%, probably far less) of the population in the traditionally LG-speaking part of the country, mostly in rural regions in the North LG and Mecklenburg dialect areas (i.e. in the federal states of Schleswig-Holstein, Hamburg, northern Lower Saxony, Bremen, and Mecklenburg-Vorpommern; Möller 2008: 33).8 All speakers of present-day LG are bilingual in HG, normally a regional variety, North HG. North HG emerged as a result of the shit towards HG by the majority of the population during the 20th century and still possesses a wide range of features that are traces of the original HG-LG bilingualism, some of which are typically associated with low socio-economic status (cf. Höder 2011: 116f.). In terms of the tentative classiication developed in Section 2, it thus qualiies as a newly emerged variety characterised by long-term variety-speciic unidirectional convergence. North HG can be considered the L1 variety (or one of the L1 varieties) of most North Germans, including the non-LG speaking majority. here is, however, a growing individual and public interest in a (re-)vitalisation of LG, resulting in the use of LG in the literature, in the media and on the internet as well as the acquisition of LG as an L2 in educational institutions for adults and the implementation of LG as a subject in schools in some federal states. he current situation is thus characterised by two contrary sociolinguistic developments: On the one hand, the traditional LG speaker group is declining, and there is a continual loss of communicative domains to HG, with the traditional diglossic distribution gradually disappearing. On the other hand, there is today a growing group of secondary LG speakers, who are oten quite enthusiastic L2 learners or speakers with a native receptive or heritage competence in LG, and a considerable domain gain within this secondary group. Furthermore, reports on recent LG show a massive inluence from (North) HG (cf. Elmentaler 2008, 2009; Berg 2013, this volume; Schröder & Elmentaler 2009; Schröder & Vorberger 2011; Schröder 2012; Hansen-Jaax 1995; Höder 2011). While many of these changes are still conined to LG as used by language enthusiasts and, therefore, have to be classiied as pertaining to an emerging new variety, it is not unlikely that they will spread to LG as a whole in the near future. From a contact linguistic perspective, the current situation suggests two possible types of development. Firstly, one should expect that the intensive contact between LG and HG continues to cause or facilitate pro-diasystematic change. Secondly, one should expect that the secondary speaker group in particular develops a tendency

S F O NY A O P M R P G CO

8. All LG material in this chapter is taken from North LG dialects in the vicinity of Hamburg. As there is no standardised LG orthography, all examples are harmonised orthographically. For the phonetic transcription conventions used, see Höder (forthc.).

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

Convergence vs. divergence from a diasystematic perspective

49

to explicitly mark their (newly acquired) linguistic identity by choosing more idiosyncratic LG forms, thus causing counter-diasystematic change. With this in view, the following sections discuss some exemplary cases of lexical, grammatical and phonological convergence and divergence in recent LG and investigate whether they have to be classiied as pro- or counter-diasystematic.9 4.2

Formal and functional convergence

he combination of formal and functional convergence is, in a way, the prototypical type of convergence. It can be exempliied by some instances of lexical change in recent LG. here are, of course, rather obvious cases such as the frequent LG loan-formations on the basis of HG compounds in, for example, technical domains such as Muuswieser ‘mouse pointer’ (cf. HG Mauszeiger) in (1) and Bedriefssysteem ‘operating system’ (cf. HG Betriebssystem) in (2): (1) en lütt Pinsel is nu an’n indef small paintbrush be.prs.3sg now at=def.sg.m.obl Muuswieser to sehn mouse_pointer infmarker see-inf ‘Now a small paintbrush appears at the mouse pointer.’10 (2) Linux is en Bedriefssysteem för Reekners. Linux be.prs.3sg indef operation_system for computer-pl ‘Linux is an operation system for computers.’11

Such lexical innovations (cf. traditional LG Muus ‘mouse [animal]’, Bedrief ‘enterprise, irm’) constitute both functional and formal unilateral convergence towards HG. From a diasystematic point of view, they are based on the language-unspeciic generalisation of metaphorical concepts in technical terminology and the speciication of schematic forms.12 For instance, an emerging diaconstruction associates the lexical concept ‘mouse [animal]’ and the technical meaning ‘mouse [computer device]’ with the phonological schema /m_(aʊ̯~u)s/, from which the languagespeciic forms are derivable by means of diferent stem vowels. his is, of course, a plain example of pro-diasystematic change (cf. Figure 5). 9. Berg’s (2013, this volume) analyses of the (re-)emergence of dative case marking in Westphalian LG can also be read as complementary case studies in the ield of case morphology.

S F O NY A O P M R P G CO

10. Source: www.plattpartu.de/comp/download/wordtricks/format_oeverdragen.pdf (21 October 2013).

E S PUBLISHIN PANG JAMIN

11. Source: nds.wikipedia.org/wiki/Linux (21 October 2013).

12. Of course, HG Maus as a technical term is, in turn, a loan translation from English mouse.

© JOHN

BE

2nd proofs

50

Stefen Höder

/maʊ̯s/ ‘mouse [animal, computer device]’HG /mus/ ‘mouse [animal]’LG

/m_(aʊ̯~u)s/ ‘mouse [animal, computer device]’

/u/LG

/aʊ̯/HG

Figure 5. Emerging diaconstruction /m_(aʊ̯~u)s/ ‘mouse’

A more interesting case of formal and functional convergence is the levelling of incongruent lexical concepts (Höder 2011: 127; Hansen-Jaax 1995: 85f.). Traditional LG has some lexical contrasts that lack an equivalent in HG, e.g. the contrast between Farv ‘paint’ and Klöör ‘colour’ (cf. HG Farbe ‘paint, colour’) as in (3) or the one between twüschen ‘between’, mank ‘among’, bi ‘at, next to’, and ünner ‘under’ (cf. HG zwischen ‘between, among’, bei ‘at, next to, among’, unter ‘under, among’) as in (4) (cf. the corresponding entries in Mensing 1927–1935): (3) a.

Hier hett een in’e Farv peddt. here have.prs.3sg one in=def.sg.f paint step-pst.ptcp ‘Someone stepped into the paint here.’ b. De hebbt all ’n annere Klöör. those have.prs-pl all indef other-strong.f.sg colour ‘Each of them has a diferent colour.’

(4) a.

Ik seet twüschen mien Vadder un mien Süster. 1sg sit.pst-1sg between my father and my sister ‘I was sitting between my father and my sister.’ b. Dor is ’n swatte Art mank there be.prs.3sg indef black-strong.f.sg pea among de gelen. def.pl yellow-weak.pl ‘here is a black pea among the yellow ones.’ c. De Wöddeln liggt bi de Arten. def.pl carrot-pl lie.prs-pl at def pea-pl ‘he carrots are next to the peas.’ d. Ünner de olen Eken steiht ’n under def.pl old-weak.pl oak-pl stand.prs-3sg indef lütten Boom. small-strong.m.sg tree ‘Under the old oaks there is a little tree.’

S F O NY A O P M R P G CO

E S PUBLISHIN PANG JAMIN

In recent LG, these contrasts tend to be neutralised in favour of constructions that are functionally congruent and formally related (‘diamorphic’; Haugen 1956: 46f.)

© JOHN

BE

2nd proofs

Convergence vs. divergence from a diasystematic perspective

51

to the HG lexemes. LG Farv is thus used to express both ‘paint’ and ‘colour’ (cf. (5)), Klöör is disappearing. Likewise, mank ‘among’ is used less frequently, and semantically related prepositions are used instead (cf. (6)). (5) In ole Tieden hett de Snee gor in old-weak.pl time-pl have.prs.3sg def.sg.m.nom snow at_all keen Farv hatt. no paint have.pst.ptcp ‘In olden days, snow did not have a colour at all.’13 (6) […] twüschen all de Minschen an Gleis een. between all def.pl human-pl at track one ‘[…] among all the people on platform one.’14

Diasystematically speaking, this process of formal and functional convergence can be modelled as the emergence of diaconstructions that replace functionally incongruent idiosyncratic elements (cf. Figure 6), hence as clearly pro- diasystematic change. On the formal side, the preference for diamorphic elements results in phonologically schematic forms which are largely unspeciied for language. he missing language-speciic sounds are mostly derivable on the basis of regular correspondences between HG and LG. For example, the language-unspeciic form expressing the lexical concepts ‘paint, colour’ can be approximately given as /ˈfaː_(bə~f)/: HG word-inal [bə] as in Farbe regularly corresponds to LG [f] as in Farv, as a result of diachronic apocope and a morphophonological inal devoicing and fricativisation rule in LG; both LG and the regional variety North HG have [aː] as the stem vowel (diachronically relecting earlier [ar] as in Standard HG).15 /fabə/ ‘paint, colour’HG /faf/ ‘paint’LG

/fa_(bə~f)/ ‘paint, colour’

/-bə/HG

/-f/LG

Figure 6. Emerging diaconstruction Farbe/Farv ‘paint, colour’

13. Source: weicherriese.beepworld.de/plattfoerkinners.htm (21 October 2013).

S F O NY A O P M R P G CO

14. Source: www.st-marien-neukloster.de/uploads/tx_rkgembrief/gb-2013-4-aug-sep.pdf (21 October 2013).

E S PUBLISHIN PANG JAMIN

15. Under certain circumstances, apocopated (as well as syncopated) words are additionally marked by a suprasegmental phoneme (‘Knick’). his feature is neither transcribed nor discussed here; for an overview cf. Höder (forthc.).

© JOHN

BE

2nd proofs

52

Stefen Höder

In a similar manner, pro-diasystematic change also leads to the loss of grammatical idiosyncrasies. A good example is the emergence of an analytic future in recent LG. Traditional LG has a four-tense system, with two synthetic tenses (present – or, more accurately, nonpast – and past) and two analytic tenses (present/ nonpast perfect, past perfect) that express anteriority vis-à-vis events narrated in the nonpast or the past tense, respectively (cf. (7)). (7) a.

He drinkt Kafe. 3sg.m.nom drink.prs-3sg cofee ‘He drinks/is drinking cofee [nonpast].’ b. He hett Kafe drunken. 3sg.m.nom have.prs-3sg cofee drink-pst.ptcp ‘He has drunk/has been drinking cofee [nonpast perfect].’ c. He drünk Kafe. 3sg.m.nom drink.pst-3sg cofee ‘He drank/was drinking cofee [past].’ d. He harr Kafe drunken. 3sg.m.nom have.pst-3sg cofee drink-pst.ptcp ‘He had drunk/had been drinking cofee [past perfect].’

Posteriority, in contrast, cannot be marked explicitly by means of tense forms; generally, nonpast forms are used to refer to future events. his is in line with many dialectal or colloquial varieties of German, but unlike Standard HG, which optionally uses an analytic future construction [werden ‘become’ + ininitive]. Additionally, [werden + ininitive] can also function as a marker of epistemic modality, expressing a relatively high degree of conidence (cf. (8)): (8) Er wird Kafee trinken. 3sg.m.nom become-ind.prs.3sg cofee drink-inf ‘He will drink/will be drinking/is going to drink cofee [future].’ ‘He is probably drinking cofee [epistemic].’

he formal equivalent of this construction in traditional LG, however, has only the epistemic meaning as in (9): (9) He ward Kafe drinken. 3sg.m.nom become-prs.3sg cofee drink-inf ‘He is probably drinking cofee.’

S F O NY A O P M R P G CO

Recent LG speakers, in contrast, use [warrn + ininitive] also to refer to future events, i.e. in a similar manner to the HG construction. his is illustrated in (10):

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

Convergence vs. divergence from a diasystematic perspective

53

(10) Ik warr denn bi mienen neegsten Besöök 1sg become.prs-1sg then at my-sg.m.obl next-sg.m.obl visit in Büsum kieken in Büsum look-inf ‘hen I’m going to look during my next visit to Büsum.’16

his innovation constitutes formal convergence, since the ways of expressing anteriority become more similar in both languages. It is also functional convergence, since the warrn construction becomes functionally equivalent to its formal counterpart in HG. From a diasystematic perspective, this process can be seen as the emergence of a language-unspeciic diaconstruction [become + ininitive] by means of a generalisation of the HG construction. his syntactic diaconstruction is lexically schematic in that it does not specify a particular modal verb but rather a set of interlingually corresponding verbs sharing the same meaning (cf. Figure 7).

BECOME + infinitive

‘future’ werden + infinitive ‘future’HG warrnLG

werdenHG

Figure 7. Emerging diaconstruction [become + ininitive] as a future marker

4.3

Functional convergence, formal con- or divergence

Functional convergence does not always co-occur with formal convergence. A simple, but telling example is the fate of LG Schrank and Schapp, both meaning ‘cupboard’. Schapp is the older variant, but has coexisted along with the synonymic HG loanword Schrank for quite a long time; both variants are established in traditional LG and can be used interchangeably by the same speakers under the same communicative circumstances. In recent LG, speakers appear to settle for one of the two forms, while the other form is avoided, thus reducing pre-existing variation in favour of a single equivalent to the HG lexeme Schrank. In either case,

S F O NY A O P M R P G CO

16. Source: www.kuestenforum.de/showthread.php?p=8993 (21 October 2013).

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

54

Stefen Höder

the result is a one-to-one correspondence between LG and HG on the functional side, whereas it can be accompanied by formal convergence (Schrank) or divergence (Schapp). From a diasystematic point of view, the establishment of Schrank as the only variant by some speakers has to be interpreted as maximally prodiasystematic, while the choice of Schapp has to be explained by system-external factors, e.g. a preference for explicitly marked LG forms as a means of expressing linguistic identity. he case of Schrank/Schapp is somewhat parallel to a more complex development in phonology, namely an ongoing merger of close-mid vowels and diphthongs in recent LG (cf. Höder 2011: 131f.).17 LG

North HG

i()

y()

u()

i()

y()

u()

i

y



i

y



e()

ø()

o()

e()

ø()

o()



œ

, 

o i i 



œ

, 

i

()

au i 

a, a

()

au i 

a, a

Figure 8. Vowel systems of LG and North HG

Figure 8 shows the vowel systems of LG and North HG. Traditional LG has a irmly established phonemic contrast between (in some contexts long) close-mid vowels such as /e(ː) o(ː)/ and mid-closing diphthongs such as /ɛɪ̯ ɛo̯/ (cf. (11–12)); there are even minimal pairs (as in (13)).18 (11) a. geven [ˈɡeːbm̩] ‘give’, negen [ˈneːɡŋ̩] ‘nine’ b. een [ɛɪ̯n] ‘one’, he [hɛɪ̯] ‘he’ (12) a. maken [ˈmokŋ̩] ‘make’, Vagel [ˈfoːɡl̩] ‘bird’ b. hoog [hɛo̯x] ‘high’, snopen [ˈʃnɛo̯pm̩] ‘eat sweets’ (13) a. leev [lɛɪ̯f] ‘dear, kind’, leev [lef] ‘live-3sg.pst’ b. groot [ɡrɛo̯t] ‘big’, Graad [ɡrot] ‘degree’

S F O NY A O P M R P G CO

17. he phonetic examples are taken from a small informal corpus collected in 2010/11.

18. Diachronically, the paradigmatic opposition between mid-close monophthongs and midclosing diphthongs oten – though not always – relects a syntagmatic contrast in Middle LG, viz. the distinction between originally short vowels in open syllables and originally long vowels.

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

Convergence vs. divergence from a diasystematic perspective

55

hese vowels are as a rule identiied conventionally with HG long close-mid vowels /eː/ and /oː/. his identiication is partly due to the frequent, but by no means general correspondence between these vowels in cognates as in (14). A more important factor, however, is that they are established as conventional equivalents in, for example, the phonological integration of established or ad hoc loanwords in either language as in (15), meaning that HG /eː oː/ regularly appear as /ɛɪ̯ ɛo̯/ in LG, while original LG /e(ː) ɛɪ̯/ both are rendered as /eː/ in HG and original LG /o(ː) ɛo̯/ both are rendered as /oː/ in HG. (14) a. b. c. d.

LG hekeln [ˈhekl̩n] ‘crochet’ ←→ North HG häkeln [ˈheːkl̩n] ‘crochet’ LG See [zɛɪ̯] ‘sea’ ←→ HG See [zeː] ‘sea’ LG Vagel [ˈfoːɡl̩] ‘bird’ ←→ HG Vogel [ˈfoːɡl̩] ‘bird’ LG groot [ɡrɛo̯t] ‘big’ ←→ HG groß [ɡroːs] ‘big’

(15) a. LG Reep [rɛɪ̯p] ‘rope’ > HG Reep [reːp] ‘rope [nautical term]’ b. LG snopen [ˈʃnɛo̯pm̩ ] ‘eat sweets’ > North HG schnopen [ˈʃnoːpm̩ ] ‘eat sweets’ c. HG System [zʏˈsteːm] ‘system’ > LG Systeem [zɪˈstɛɪ̯m] ‘system’ d. HG Fraktion [frakˈtsi̯oːn] ‘parliamentary group’ > LG Fraktschoon [frakˈtʃɛo̯n] ‘parliamentary group’

his set of interlingual correspondences is thus ambiguous, as two vowels in LG correspond to only one vowel in HG.19 In recent LG, however, speakers tend to resolve this ambiguity by merging original monophthongs and diphthongs into one common set, with diphthongisation being treated as a merely allophonic feature as in (16). (16) a. b. c. d.

geven [ˈɡeːbm̩, ˈɡɛɪ̯bm̩] ‘give’, negen [ˈneːɡŋ̩, ˈnɛɪ̯ɡŋ̩] ‘nine’ een [ɛɪ̯n, eːn] ‘one’, he [hɛɪ̯, heː] ‘he’ maken [ˈmokŋ̩, ˈmɛo̯kŋ̩] ‘make’, Vagel [ˈfoːɡl̩, ˈfɛo̯ɡl̩] ‘bird’ hoog [hɛo̯x, hox] ‘high’, snopen [ˈʃnɛo̯pm̩, ˈʃnopm̩] ‘eat sweets’

On the functional side, this merger is a clear case of convergence: he number of phonemes and distinctive qualitative contrasts becomes more similar in HG and recent LG. On the formal side, however, this change can be counted as either convergence or divergence, depending on the phonetic realisation of the merged phoneme in particular lexemes, which in turn appears to be depending on individual

S F O NY A O P M R P G CO

19. In many dialects, the same is also true for a third pair of LG vowels, namely the monophthong /ø(ː)/ and a diphthong /œi ̯/ (as in prökeln [ˈprøkl̩n] ‘rake, pick, poke’ vs. schöön [ʃœin̯ ] ‘beautiful, nice’), which both correspond to HG /øː/. In Hamburg and vicinity, however, LG /œi /̯ has merged with another diphthong /ɔi/̯ , which adds even more complexity to the interlingual correspondences. his special case is not discussed in detail here.

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

56

Stefen Höder

preferences or pragmatic conditions. A tempting explanation would be that the divergent diphthongal realisation is used to mark a lexeme as Low German, i.e. as a kind of socially motivated hyperdialectalism (this remains to be investigated), while monophthongal realisation represents formal convergence and, thus, the more expectable outcome from a purely system-internal perspective. From a diasystematic point of view, though, the diphthong–monophthong merger is an instance of pro-diasystematic change in any case. By neutralising an existing opposition in LG, speakers not only simplify the monolingual system of LG, but also the bilingual LG-HG system by increasing interlingual congruence and reducing the individual languages’ idiosyncrasies. Formal convergence resulting in interlingually equivalent monophthongs, when occurring, represents an additional pro-diasystematic development, which is in principle independent from the primary process of functional convergence. 4.4

Functional convergence, formal divergence

he last example to be discussed here is the combination of functional convergence with formal divergence, also related to the vowel systems of LG and HG, but restricted to the distribution of interlingually corresponding vowels in cognates. As a result of regular sound changes in the past, it is possible to establish pairs of interlingually equivalent vowels that frequently correspond to each other. For example, a regular diphthongisation process afected long close vowels in Early Modern HG (/iː > ɑi̯, yː > ɔi̯, uː > au̯/, North HG values given for the modern forms), while LG retained the old monophthongal qualities; this is relected in the very frequent correspondences between HG /ɑi̯/ and LG /i(ː)/ and so forth (as in (17)). Similarly, original closing diphthongs were monophthongised in Old Saxon (/ai̯ > eː, au̯ > oː/), and the resulting mid-close monophthongs were again diphthongised in Modern LG (/eː > ɛɪ̯, oː > ɛo̯/), resulting in frequent correspondences such as in (18). Correspondences between phonetically identical vowels are also frequent, especially with mid vowels (as in (19)). (17) a. LG mien [miːn] ‘my’ ←→ HG mein [mɑi̯n] ‘my’ b. LG düütsch [dytʃ] ‘German’ ←→ HG deutsch [dɔi̯tʃ] ‘German’ c. LG Huus [hus] ‘house’ ←→ HG Haus [hau̯s] ‘house’ (18) a. LG Been [bɛɪ̯n] ‘leg’ ←→ North HG Bein [bɑi̯n] ‘leg’ b. LG ok [ɛo̯k] ‘also’ ←→ HG auch [au̯x] ‘also’

S F O NY A O P M R P G CO

(19) a. LG beden [ˈbeːdn̩] ‘pray’ ←→ HG beten [ˈbeːtn̩] ‘pray’ b. LG Trepp [trɛp] ‘stairs’ ←→ HG Treppe [ˈtrɛpə] ‘stairs’

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

Convergence vs. divergence from a diasystematic perspective

57

Apart from such salient correspondences, there are also more marginal ones which can be traced back to sound changes that were irregular or restricted to particular phonotactic contexts. For example, LG /ɑi̯/ corresponds to North HG /eː/ in a small set of words such as LG meihen [mɑi̯n] ‘mow’ and HG mähen [meːn] ‘mow’, and LG /ɪ/ corresponds to HG /e/ in relatively few words such as LG Finster [ˈfɪnstɐ] ‘window’ and HG Fenster [ˈfɛnstɐ] ‘window’. One of the most infrequent correspondences is, paradoxically, the one between LG /ɑi̯/ and North HG /ɑi̯/, although these phonemes sound the same in both languages (cf. the examples in (20)). (20) a. LG Arbeit [ˈaːbɑɪ̯t] ‘work’ ←→ North HG Arbeit [ˈaːbɑɪ̯t] ‘work’ b. LG Ei [ɑi̯] ‘egg’ ←→ North HG Ei [ɑi̯] ‘egg’ c. LG beid [bɑi̯t] ‘both’ ←→ North HG beide [ˈbɑi̯də] ‘both’

In recent LG, marginal correspondence relations tend to be overridden by means of a generalisation of more frequent ones, leading to LG forms like [mɛɪ̯n, meːn] ‘mow’ (consistent with the correspondence illustrated in (19a) and the diphthong– monophthong merger discussed above) or [ˈfɛnstɐ] ‘window’ (consistent with the correspondence in (19b)). What is, at a irst glance, more surprising is that even in the case of interlingual homophones such as the ones in (20), the original LG vowels are oten substituted by ones that are part of more salient correspondence relations. his substitution results in innovative variants such as [ˈaːbɛɪ̯t] ‘work’, [ɛɪ̯] ‘egg’ or [bɛɪ̯t] ‘both’. his is an instance of formal divergence, in which phonetically identical vowels in two languages (or even whole homophone words) are replaced by language-speciic, diverging forms. From a diasystematic perspective, however, this type of divergence can easily be analysed as pro-diasystematic, as divergence in form is accompanied by convergence in function: he set of correspondence relations between LG and HG, and hence the common bilingual system, is being regularised and simpliied. his type of divergence is thus explainable (and predictable) by language-internal mechanisms alone, without invoking any sociolinguistic factor. Of course, it is also perfectly possible that the production of innovative forms like [ˈaːbɛɪ̯t] ‘work’ is additionally motivated by a desire to explicitly mark a word as LG which would otherwise be language-unspeciic, and it is quite expectable that such an innovation, once it is established, can be utilised as a marker of language-speciicity.

S F O NY A O P M R P G CO

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

58

Stefen Höder

5. Conclusion It goes without saying that the mechanisms leading to convergence and divergence in language contact situations are the same that lead to language change in general, including structural and sociolinguistic factors as well as the communicative and social interplay between individuals and speaker groups. he discussion of example cases from the current development in presentday LG shows, however, that structural mechanisms are not necessarily restricted to language-internal (monolingual) structures, but can extend to two or more languages. From a diasystematic angle, a contact-induced change is irst and foremost a change within one multilingual system, not a change in a monolingual system in contact with a diferent – and separate – system. Pro-diasystematic change can be explained as a simpliication within a combined system shared by two or more languages. As the discussion of LG examples illustrates, this type of change can lead to diferent degrees of functional convergence, formal convergence and, under certain conditions, even divergence. Extralinguistic factors, in contrast, come into play when the multilingual system is retaining complexity or being complexiied instead of becoming simpler – in such cases, we should expect that multilingual speakers beneit from this instance of change in some other way, e.g. because a particular form has some indexical value within the relevant speaker group. Of course, it is possible that extralinguistic factors reinforce prodiasystematic change. It is, however, crucial to note that an impact of social factors is not a necessary condition in such cases.

References Aikhenvald, A. Y. 2007. Grammars in contact. A cross-linguistic perspective. In Grammars in Contact. A Cross-linguistic Typology [Explorations in linguistic typology 4], A. Y. Aikhenvald & R. M. W. Dixon (eds), 1–66. Oxford etc.: Oxford University Press. Berg, K. 2013. Morphosyntax nominaler Einheiten im Niederdeutschen [Germanistische Bibliothek 49]. Heidelberg: Winter. Crot, W. 2001. Radical Construction Grammar. Syntactic heory in Typological Perspective. Oxford etc.: Oxford University Press. DOI: 10.1093/acprof:oso/9780198299554.001.0001 Durrell, M. 2004. Linguistic variable – linguistic variant. In Sociolinguistics. An International Handbook of the Science of Language and Society, Vol. 1, 2nd edn [Handbooks of linguistics and communication science 3.1], U. Ammon et al. (eds), 195–200. Berlin/New York: de Gruyter. Elmentaler, M. 2008. Varietätendynamik in Norddeutschland. Sociolinguistica 22: 66–86. Elmentaler, M. 2009. Modernes Niedersächsisch – Dialektwandel im nordniederdeutschen Raum. In Low Saxon Dialects Across Borders [Zeitschrit für Dialektologie und Linguistik. Beihete 138], A. N. Lenz, C. Gooskens & S. Reker (eds), 339–363. Stuttgart: Steiner.

S F O NY A O P M R P G CO

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

Convergence vs. divergence from a diasystematic perspective

Goldberg, A. E. 1995. Constructions. A Construction Grammar Approach to Argument Structure. Chicago: University of Chicago Press. Goldberg, A. E. 2006. Constructions at Work. he Nature of Generalization in Language. Oxford etc.: Oxford University Press. Grosjean, F. 2008. Studying Bilinguals. Oxford etc.: Oxford University Press. Hansen-Jaax, D. 1995. Transfer bei Diglossie. Synchrone Sprachkontaktphänomene im Niederdeutschen. Hamburg: Kovač. Harnisch, R. 2010. Divergence of linguistic varieties in a language space. In Language and Space. An International Handbook of Linguistic Variation, Vol. 1: heories and Methods [Handbooks of linguistics and communication science 30.1], P. Auer & J. E. Schmidt (eds), 275–295. Berlin/New York: De Gruyter Mouton. Haugen, E. 1956. Bilingualism in the Americas. A Bibliography and Research Guide [Publications of the American Dialect Society 26]. [s. l.]: University of Alabama Press. Heine, B. & Kuteva, T. 2005. Language Contact and Grammatical Change. Cambridge etc.: Cambridge University Press. DOI: 10.1017/CBO9780511614132 Hinskens, F., Auer, P., & Kerswill, P. 2005. he study of dialect convergence and divergence. Conceptual and methodological considerations. In Dialect Change. Convergence and Divergence in European Languages, P. Auer, F. Hinskens & P. Kerswill (eds), 1–48. Cambridge etc.: Cambridge University Press. Höder, S. 2011. Niederdeutsch und Norddeutsch – ein Fall von Diasystematisierung. Niederdeutsches Jahrbuch 134: 113–136. Höder, S. 2012. Multilingual constructions: A diasystematic approach to common structures. In Multilingual Individuals and Multilingual Societies [Hamburg studies on multilingualism 13], K. Braunmüller & C. Gabriel (eds), 241–257. Amsterdam: John Benjamins. DOI: 10.1075/hsm.13.17hod Höder, S. 2014. Constructing diasystems: Grammatical organisation in bilingual groups. In he Sociolinguistics of Grammar [Studies in language companion series 154], T. Åfarli & B. Mæhlum (eds), 137–152. Amsterdam: John Benjamins. DOI: 10.1075/slcs.154.07hod Höder, S. forthcoming. Low German: A proile of a word language. In Phonology of syllable and word languages in theory and practice [Linguae & litterae], J. Caro Reina & R. Szczepaniak (eds). Berlin/New York: de Gruyter. Hofmann, T. & Trousdale, G. (eds). 2012. he Oxford Handbook of Construction Grammar. Oxford/New York: Oxford University Press. Matras, Y. 2009. Language contact. Cambridge etc.: Cambridge University Press. DOI: 10.1017/CBO9780511809873 Mattheier, K. J. 1996. Varietätenkonvergenz: Überlegungen zu einem Baustein einer heorie der Sprachvariation. Sociolinguistica 10: 31–52. Mensing, O. (ed.). 1927–1935. Schleswig-Holsteinisches Wörterbuch. Volksausgabe. Neumünster: Wachholtz. Möller, F. 2008. Plattdeutsch im 21. Jahrhundert. Bestandsaufnahme und Perspektiven. Leer: Schuster. Moulton, W. G. 1960. he short vowel systems of Northern Switzerland. A study in structural dialectology. Word 16: 155–182. Myers-Scotton, C. 2002. Contact Linguistics. Bilingual Encounters and Grammatical Outcomes. Oxford etc.: Oxford University Press.

59

S F O NY A O P M R P G CO

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs

60

Stefen Höder

Peterson, J. Forthcoming. Multilingualism, multilectalism and register variation in linguistic theory. Extending the diasystematic approach. In he Syntax-Semantics Interface [Studies in language and cognition], A. Latroite & R. Osswald (eds). Düsseldorf: Düsseldorf University Press. Ross, M. 2007. Calquing and metatypy. Journal of Language Contact 1: 116–143. DOI: 10.1163/000000007792548341 Schröder, I. 2012. ‘Da nicht für.’ Grammatische Kontaktphänomene HochdeutschNiederdeutsch. Sociolinguistica 26: 18–29. Schröder, I. & Elmentaler, M. 2009. Sprachvariation in Norddeutschland (SiN). Niederdeutsches Jahrbuch 132: 41–68. Schröder, I. & Vorberger, L. 2011. Standardadvergenz im nördlichen Niedersachsen? Niederdeutsches Jahrbuch 134: 137–147. Weinreich, U. 1954. Is a structural dialectology possible? Word 10: 388–400. Weinreich, U. 1964 [1953]. Languages in Contact. Findings and Problems, 3rd edn. London etc.: Mouton.

S F O NY A O P M R P G CO

E S PUBLISHIN PANG JAMIN

© JOHN

BE

2nd proofs