ArticlePDF Available

Coherence in new urban dialects: A case study

Authors:

Abstract and Figures

This paper investigates evidence for linguistic coherence in new urban dialects that evolved in multiethnic and multilingual urban neighbourhoods. We propose a view of coherence as an interpretation of empirical observations rather than something that would be "out there in the data", and argue that this interpretation should be based on evidence of systematic links between linguistic phenomena, as established by patterns of covariation between phenomena that can be shown to be related at linguistic levels. In a case study, we present results from qualitative and quantitative analyses for a set of phenomena that have been described for Kiezdeutsch, a new dialect from multilingual urban Germany. Qualitative analyses point to linguistic relationships between different phenomena and between pragmatic and linguistic levels. Quantitative analyses, based on corpus data from KiDKo (www.kiezdeutschkorpus.de), point to systematic advantages for the Kiezdeutsch data from a multiethnic and multilingual context provided by the main corpus (KiDKo/Mu), compared to complementary corpus data from a mostly monoethnic and monolingual (German) context (KiDKo/Mo). Taken together, this indicates patterns of covariation that support an interpretation of coherence for this new dialect: ourfindings point to an interconnected linguistic system, rather than to a mere accumulation of individual features. In addition to this internal coherence, the data also points to external coherence: Kiezdeutsch is not disconnected on the outside either, but fully integrated within the general domain of German, an integration that defies a distinction of "autochthonous" and "allochthonous" German, not only at the level of speakers, but also at the level of linguistic systems.
Content may be subject to copyright.
1
Coherence in new urban dialects: a case study
Heike Wiese, Ines Rehbein
1. Introduction
Coherence is a concept that is relevant for a lot of different disciplines, among them
philosophy (Lewis 1946), Bayesian epistemology (Shimony 1955), quantum physics (Glauber
1963), epidemiology (Bradford Hill 1965), higher algebra (MacLane 1965), and cybernetics
(Wolkowski 2007). What the different notions of coherence in the different fields have in
common is reference, in one way or another, to systems and supportive relations between
their elements. In the sense we are interested in here, coherence is a core property of linguistic
systems.
Against this background, a domain of particular interest in a discussion of coherence is
that of multilingual urban contexts. In present-day Europe (as in a number of other regions),
these contexts are characterised by a high linguistic diversity – sometimes called
‘superdiversity’ (cf. e.g., Vertovec 2007) – that supports a large range of linguistic repertoires
and ample opportunities for language contact. Among others, this has led to the emergence of
new variants of the respective majority languages that reflect the special dynamics of this
diverse setting.1 For a discussion of coherence, these new variants are particularly interesting
because their status as proper, identifiable systems is still controversial, not only in linguistics,
but also in public discussion, where they are subject to intense, sometimes heated discussions.
In linguistics, the controversy focuses on such questions as whether these new variants
represent full-blown linguistic systems in their own right, whether they can be described as
dialects or varieties, and whether they are spoken consistently (cf. e.g., Auer 2013 for a
discussion of Kiezdeutsch in Germany, lit. ‘(neighbour-)hood German’, a German example
for such a new way of speaking in multilingual urban neighbourhoods). In contrast to this,
public discussion approaches the status of such variants from a quite different angle. Here, the
issue is predominantly framed within a language ideology that excludes them from the
domain of “proper” or “correct language”. In this context, denying them the status of a

1For an overview, cf. contributions in Svendsen & Quist (Eds.) (2010), Kern & Selting (Eds.) (2011),
Nortier & Svendson (Eds.) (2014).
Authors‘copy;finalversionin:Lingua172/173:4561[Special
IssueonCoherence,covariationandbricolage,ed.GregoryGuy
&FransHinskens]
2
systematic variety or dialect is in line with devaluating them as “faulty” speech rather than a
legitimate language variant, and thus ultimately supports a devaluation and Othering of
speakers (cf. Wiese to appear for a more detailed analysis of these patterns). Evidence for
linguistic systematicity and interpretations of coherence are thus interesting not only from the
point of view of linguistic theory, but also for public dissemination, where they can contribute
to counteracting tendencies of social exclusion that is played out via linguistic proxies.
In what follows, we are first going to further develop our notion of linguistic
coherence (Section 2) and then present a case study that fleshes out our conceptual
suggestions. The subject of our case study will be data from Kiezdeutsch,. We will first
describe the corpus data that provides the empirical basis for our study, the KiezDeutsch
Korpus (Section 3), and then investigate the question of linguistic coherence for this data
using qualitative and quantitative analyses (Section 4). The final section brings our results
together and discusses the implications of these findings for our understanding of new
linguistic patterns in contexts of language variation and change (Section 5).
2. Coherence as an interpretation of data
A general perspective shared among the contributions to this volume is that of coherence as a
relationship between different linguistic variables. Given this basic assumption, what is the
nature of this relationship, and at what level or levels does it hold? How does it fit in with the
other two concepts focussed in this volume: covariation and bricolage? Let us have a closer
look at these key questions to set up the conceptual foundations for our case study on
Kiezdeutsch.
A crucial point for any empirical investigation into the coherence of linguistic
phenomena (varieties, styles, ways of speaking) is to acknowledge that coherence is not
something we can observe in the world, but rather is an interpretation of our observations, our
way to make sense of the data. This is similar to what has been noted, for instance, for
causality, a concept that some disciplines tie closely to coherence:2 while we engage in causal
reasoning for a broad range of social, psychological, and physical domains and habitually
ascribe cause-effect relationships to phenomena, causality is not something empirically
accessible to us itself, but is rather an interpretation we develop for observed co-occurrences.
We cannot directly observe that A causes B, but only that A and B will occur together – or, as
Hume (1740) put it in his discussion of causation, that Aand B are “constantly conjoined”.

2For example in signal processing, coherence entails causation provided certain supporting conditions hold; cf.
Brillinger (1975); similar models are used in economics (Granger 1977).
3
Similarly, we cannot directly observe that a linguistic phenomenon P is coherent, but only
derive this as an interpretation of our empirical and theoretical findings for P. From this
perspective, then, the investigation of coherence does not necessarily clash with
sociolinguistic approaches such as Rampton et al. (2014), who state that
“With the emergence of post-structuralism, burgeoning interest in agency, fragmentation
and contingency have weakened the linguist’s traditional assumption that system and
coherence were there in their data, just waiting to be discovered” (Rampton et al. 2014:5)
What is “there in the data” are co-occurrences of phenomena and, more specifically, co-
occurrences and possible variations of linguistic patterns, which is where the second concept
mentioned in the title of this volume, “covariation”, comes in. However, if we want to
investigate coherence, we are not just looking for covariations of more or less random
elements A and B, but for those of related variables. In order for us to interpret these patterns
as coherence, the nature of those relationships should be such that A and B have something in
common, for instance that they share particular elements or are based on overlapping sources.
Such commonalities, together with the observed covariations, might suggest that A influences
B, that A is a driving force behind the emergence of B, that B depends on A, and/or vice
versa, or that they are influenced by the same source. What is important is that A and B are
linked up with each other in a way that makes sense from the point of view of linguistic
analysis, suggesting a system rather than isolated individual elements.
A possible pattern might then be, for instance, one where some elements are focal
points connecting several others, while others have a more peripheral position. For an
interpretation of coherence, we do not need to make any assumptions about the specific
patterns of links: what is crucial is the existence of a linked-up network. This, then, is a
somewhat weaker view of coherence than a prominent one from cybernetics (cf. Wolkowski
2007), where homomorphisms between different subsystems are a precondition for coherence.
Figure 1 gives a graphic illustration for the view of coherence we suggest here.
Figure 1: Coherence as a pattern of interconnecting relationships
4
There are two important aspects of this view of coherence: (1) coherence is characterised as a
particular interpretation of the data, as noted above, but crucially one that is grounded in
linguistic analysis, and (2) this analysis is initially directed at the level of the linguistic
system.3 The second touches on a central point raised in the Rampton et al. quote above, and
also on the last term put forward in the title of this volume, namely “bricolage”. While we
might find evidence supporting an interpretation of coherence if we look at our data from the
point of view of the linguistic system, the picture changes if we look at the actual linguistic
behaviour of speakers and the different choices they make in different speech situations.
Obviously, as agents of their linguistic behaviour, speakers are not restricted to using
linguistic elements from one source only, but have a larger pool from which to choose in any
given communicative setting. In particular in contexts of high cultural and linguistic diversity
as we find them in present-day urban Europe, speakers’ repertoires can encompass elements
from a large range of different codes, including not only different styles and registers, but also
different dialects and languages. They can make use of these resources in diverse and creative
ways, leading to the emergence of new ways of speaking in heritage and majority languages
and to patterns of code switching, bricolage, and crossing that defy fixed “language-body-
place” ties (cf. Quist 2010).
This high linguistic diversity and its inherent dynamics have provoked controversy as
to whether we are dealing with distinct varieties here at all, or rather with fluid styles or
stylistic practices.4 However, while we agree that speakers’ linguistic behaviour is too
heterogeneous and variable to support a single, distinct variety that would represent their
repertoire, this does not mean that there aren’t, at the level of language, any identifiable
linguistic systems that form part of their resources. It is these systems that we believe our
interpretations of coherence should target: while speakers are free to choose from different
resources, and to mix, merge, and combine them in their communicative behaviour,
supporting different acts of social identity, the linguistic elements themselves will interact
with each other, depend on each other and influence each other within different linguistic
patterns, thus forming, at the level of language, identifiable systems in their own right. Figure
2 illustrates this, with a speaker who can access and combine, depending on the speech

3This underlines, again, that we are investigating coherence at the level of linguistic systems here. This focus of
our paper does not exclude, though, that coherence can also be analysed from sociolinguistic perspectives on
speech communities, e.g. for the use of particular features as markers of group identity.
4For a discussion of the different perspectives cf. e.g., Hinskens (2007), Quist (2008), Auer (2013), Wiese
(2013).
5
situation, elements from a heterogeneous set of different, possibly overlapping linguistic
systems that each lend themselves, at the language level, to interpretations of coherence.
Figure 2: Diversity of speakers’ resources vs. coherence at the level of linguistic systems
When looking for such systems, we should then target linguistic data not from a preconceived
notion of particular languages or varieties, but from a perspective of typical speech situations,
thus identifying the relevant domain of interaction for particular linguistic elements. This,
then, moves beyond both the concept of the “ideal speaker-hearer” in the generative tradition
and that of the “authentic speaker” in older dialectological and sociolinguistic approaches (cf.
Eckert 2003 for a critique): we are not looking for speakers who use a particular linguistic
code consistently, to its full extent, and/or invariably, but for a linguistic system that might
emerge out of recurring usages in the communicative behaviour found in certain speech
situations.
In order to to support an interpretation of “coherence”, then, we should look for
patterns in our linguistic data that suggest systematic links between different phenomena
and/or between different linguistic levels, and we should look at specific speech situations for
such patterns.
The relevant speech situation for our case study is that of informal peer-group
conversations in mixed groups of adolescents from different linguistic backgrounds (including
monolingual speakers of the majority language, in this case German), which has been shown
to support the kind of new ways of speaking that we are interested in.
6
3. KiDKo, a corpus for Kiezdeutsch
As the empirical basis for our investigation, we use a corpus of spontaneous speech in urban
neighbourhoods in Berlin, the KiezDeutsch Korpus, short KiDKo.5 KiDKo provides data from
self-recordings in conversations, conducted mostly in German, among adolescents in informal
peer-group situations. The corpus gives detailed information on anchor speakers (age, gender,
linguistic background) and, where possible, on the other speakers involved in the
conversations. It consists of audio files aligned with annotated transcriptions and allows
linguistic queries at different levels. To support word searches, it includes a “normalisation”
layer where non-canonical pronunciations, punctuation, and capitalisation are transferred to
standard German spelling. The normalisation also served as the input for automatic part-of-
speech (POS) tagging, which considerably reduces the number of unknown words caused by
the high number of pronunciation variants for each canonical form, and thus increases tagging
accuracy considerably.
The main part of the corpus, “KiDKo/Mu, with approximately 355,000 tokens, is
from a multiethnic neighbourhood, Berlin-Kreuzberg, with speakers of different multilingual
backgrounds as well as monolingual Germans. In addition, KiDKo encompasses a smaller,
complementary part, “KiDKo/Mo”, of approximately 151,000 tokens, that supports
comparisons of the Kiezdeutsch data with a mostly monolingual speech community: this part
is from Berlin-Hellersdorf, a predominantly monoethnic German neighbourhood with
comparable socioeconomic indicators, and all speakers in this part have a monolingual
German background.
KiDKo is accessible via a corpus website www.kiezdeutschkorpus.de that also
provides more detailed information on the corpus data.
4. Empirical foundations for coherence interpretations: The case of Kiezdeutsch
In what follows, we are going to investigate the data we find in KiDKo from a perspective of
coherence, thus applying the conceptual background we developed in section 2. In order to do
so, we will concentrate on a number of phenomena in Kiezdeutsch that have been
linguistically described elsewhere, which allows us to focus on interactions between different
domains and patterns of covariation here. We will investigate links between different, related
phenomena at the same linguistic levels as well as links between different linguistic levels
involved in the emergence of specific phenomena. In addition to such qualitative analyses, our

5Cf. Rehbein & Schalowski (2012; to appear).
7
investigation will involve quantitative assessments of phenomena and comparisons between
the Kiezdeutsch data from KiDKo/Mu and the corresponding data from KiDKo/Mo.
4.1 Links between phenomena
As an example for links between different, related phenomena, let us look at the domain of
determiner use in Kiezdeutsch. A phenomenon that has often been described here is the
absence of overt determiners (such as articles and pronouns) in places where they would be
present in standard German. (1) and (2) give a range of examples, covering the absence of
definite (1ab), (2a) and indefinite articles (1cde) and personal or demonstrative pronouns (2),
from KiDKo/Mu and from other studies on language use among adolescents in multilingual
urban neighbourhoods [in the KiDKo data, capitalisation indicates main stress; labels identify
multi- vs. monoethnic contexts (Mu vs. Mo), anchor speaker code (e.g., H27), male vs. female
sex (M vs. W), and family/heritage language (e.g., A: Arabic, D: German, K: Kurdish, T:
Turkish)].
(1a) wo kuckst du heute FUßballspiel (KiDKo, MuH27WT)
where watch you today football.game
‘Where will you watch [the] football game today?’
(1b) ich muss KUgelschreiber zurückpacken (KiDKo, MuH11MD)
I must ballpen back.put
‘I have to put [the] ballpen back.’
(1c) wenn wir portemonNAIE finden (KiDKo, MuH25MA)
when we wallet find
‘when we find [a] wallet’
(1d) er hat verLETzung (KiDKo, MuP1MK)
he has injury
‘He has [an] injury.’
(1e) daNACH vor meinem FENster is so BRIEF (Kern & Selting 2006: 245)
then in.front.of my window is PTCL letter
‘Afterwards, there was [a] letter in front of my window.’
(2a) hast comPUter angemacht (KiDKo, MuH23MT)
have computer on.switched
‘Have [you] switched [the] computer on?’
(2b) dann LÖSCH das kurz und dann MACHST du nochmal (KiDKo, MuH1WD)
then delete that briefly and then make you again
‘Then you delete that briefly, and then you do [that] again.’
(2c) vielleicht WAR schon mal kaputt gegangen (KiDKo, MuP1WK)
maybe had already once broken gone
‘Maybe [it] had already broken before.’
(2d) wenn ich sehr liebe (Dirim & Auer 2004: 209)
when I very love
‘when I love [her] very much’
8
As the examples show, this data can be found across linguistic backgrounds, including
monolingual German speakers (1b), (2b). Such determiner-less constructions can also be
found in KiDKo/Mo, the corpus part covering conversations from a mostly monolingual
German neighbourhood:
(3a) du bist kleines DROgenopfer (KiDKo, Mo01MD)
you are little drug.victim
‘You are [a] little drug victim.’
(3b) weil fenster war ja OFfen (KiDKo, Mo18MD)
because window was PTCL open
‘Because [the] window was open after all.’
(3b) hast =e wiedergefunden (KiDKo, Mo04MD)
have you again.found
‘Have you found [it] again?’
(3b) und VORgestern war auch schon da (KiDKo, Mo05WD)
and before.yesterday was also already there
‘And the day before yesterday, [he] was there already, too.’
However, while both parts of the corpus (Mu and Mo) contain utterances without determiners
in places where we would have them in standard German, a comparison between the two parts
shows that this pattern is considerably more frequent in KiDKo/Mu, for articles and pronouns
alike, with 685 instances in KiDKo/Mu and only 165 in KiDKo/Mo. As a comparison of raw
frequencies is not meaningful, we performed a chi-square6 test on the raw corpus counts to
find out whether the difference between the two samples is statistically significant. In order to
do so, we also retrieved the number of constructions with an overt article or pronoun in both
parts of the corpus, with the help of the POS annotation layer.7 While the figures for
determinerless constructions were obtained by a manual search of the whole corpus, those for
overt determiners were identified via automatic searches. We obtained the number of
pronouns realised in the two parts of the corpus by searching for the POS tags PDS
(demonstrative pronoun) and PPER (personal pronoun). For constructions with an overt
article, we conducted a corpus search for sequences of words starting with either an (ART), a
merged preposition with an article (APPRART), or an attributive possessive pronoun
(PPOSAT), followed by a noun. Between the determiner and the noun, we allowed for
optional attributive adjectives (ADJA), potentially preceded by one or more intensifying
particles (ADV) or adverbially used adjectives (ADJD), and also for any number of fillers
(e.g. ‘uh’, ‘uhm’) and silent pauses (SPFILL, PAUSE). Since there is no overt article for

6WeusedthestatisticalsoftwareRtoperformaPearson’sChisquaredtestwithYates‘continuitycorrection.
7The annotation scheme we used is an extended version of the Stuttgart-Tübingen Tagset (STTS; Schiller et al.
1999) with additional tags for spoken language phenomena such as fillers, backchannel signals, or discourse
markers.
9
indefinite plural NPs in German, we excluded all plural forms from the counts. This had to be
done manually as the corpus does not include morphological annotations. First, we extracted
all NPs, sorted them and removed duplicate instances so that we only had one entry for each
NP type (but kept track of the type frequencies). This allowed us to remove all plural forms in
an efficient way. For entries which could either be a singular or a plural form (e.g. den Türken
‘the Turks’, which can either be accusative singular or dative plural), we referred to the
context of the utterance or decided in favour of the more plausible analysis.
Table 1 shows the frequencies for instances with determiner omission and with overt
determiners for KiDKo/Mu and KiDKo/Mo. The ratio of omitted to overt determiners is
higher for both articles and pronouns in the multiethnic part of the corpus, and the difference
is highly significant in the case of articles (p<0.001), and significant in the case of pronouns
(p=0.01756).
KiDKo/Mu KiDKo/Mo
NP with determiner omission 360 65
NPs with overt determiner 8,158 4,106
% omitted 4.23 1.56
pronoun omission 325 100
overt pronoun 32,849 13,346
% omitted 0.98 0.74
Table 1: Frequencies for determiner omission and overt determiners for KiDKo/Mu and
KiDKo/Mo
This indicates that the option to omit determiners is significantly further developed in
Kiezdeutsch than in comparable language use in a more monolingual context, pointing to a
characteristic feature of Kiezdeutsch. For the issue of coherence, the question is now whether
it is linked to other characteristics of Kiezdeutsch in a meaningful way: does it covary with
other phenomena that are related to it on linguistic levels? In the following paragraphs, we
are going to show that this is indeed the case: the pronounced optionality of determiners we
find in Kiezdeutsch is not an isolated phenomenon, but forms the core of a network of related
phenomena contributing to this variety.
In the following paragraphs, we present three examples of this: bare local NPs, new
directive particles, and new light verb constructions. In each case, we briefly describe the
observed phenomenon, focussing on its linguistic systematicity and its relation to the
omission of determiners, and then compare its frequency with the data from KiDKo/Mo. Note
that the respective frequencies for these three phenomena were not included in the overall
figures for omissions of determiners, so that they are not counted twice.
10
Bare local NPs
Bare local NPs in Kiezdeutsch are frequently mentioned in (socio-)linguistic studies of it, and
in the public perception they seem to be so salient as characteristics of this dialect that they
have become something like a shibboleth for it. (4) gives some examples from KiDKo/Mu
and from other studies:
(4a) als ob du gleich HOCHzeit gehst (KiDKo, MuH3WT)
as if you shortly wedding go
‘as if you are about to go [to a] wedding’
(4b) was machst du WESTdeutschland (KiDKo, MuH27WT)
what make you West.Germany
‘What are you going to do [in] West Germany?’
(4c) KLAR warst du schule (KiDKo, MuH12MD)
sure were you school
‘Of course you were [at] school!
(4d) wärst du REno gekomm (KiDKo, MuP1MK)
had you Reno come
‘You should have come [to] Reno [= name of a shop].’
(4e) die muss bahnhof gehn (Kallmeyer & Keim 2003: 42)
she must train.station go
‘She must go [to the] train station.’
(4f) gehen Sie andere Stadtteil (Auer 2013: 33)
go you other town.part
‘Go [to a] different part of the town.’
In cases such as the above, a bare NP functions as a local expression by itself, whereas in
standard German it would be inside the DP complement of a PP. In some studies, this
phenomenon is described as a “lack of prepositions”. Note, though, that this suggests a
perspective that takes standard German as a yardstick and assesses other varieties primarily
against this. In order to account for such phenomena in Kiezdeutsch as grammatical patterns
in their own right, we therefore talk about “bare local NPs”.
Bare local NPs present a highly systematic linguistic phenomenon. In short, one can
identify the following semantic and syntactic rules for the absence of prepositions:8
Only lexical prepositions can be absent, while functional ones are overt.
A central semantic domain for this is the local domain and, to a somewhat lesser
degree, the temporal domain.
In the local domain, this is restricted to reference to (a) place and (b) direction with the
NP-referent as a goal (rather than a source).

8For detailed discussions cf. Wiese (2012, 2013), Auer (2013), Pohle (2013), Siegel (2014).
11
Prepositions are usually only absent if there is also no determiner.
The last rule makes the relation to the omission of determiners clear: the absence of a
determiner supports the absence of a preposition, that is, an absent determiner makes it
possible to leave out the preposition as well, pointing to an implicational relationship. This
means that prepositions are overt in constructions with pronouns, and usually also in
constructions with full DPs: of the 158 instances we found for the omission of local
prepositions in KiDKo/Mu, only 8 contain a determiner. However, this does not rule out the
reverse, namely constructions with a preposition, but no determiner, cf. (5):
(5a) der fliegt auf BOden (KiDKo, MuH25MA)
he flies onto floor
‘He flies onto [the] floor.’
(5b) da in KÜche (KiDKo, MuH12MD)
there in kitchen
‘there, in [the] kitchen’
Taken together, this indicates that the two phenomena are linked up in a one-way relationship:
the option to omit determiners that we found to be pronounced in Kiezdeutsch supports the
absence of local prepositions.9 Accordingly, bare local NPs can occur not only in contexts
where we would not have a determiner in standard German either, e.g., with proper names
such as in (4b) and (4d) above,10 but also systematically include common nouns.
To sum up, bare local NPs represent a systematic linguistic phenomenon that is linked
up with the Kiezdeutsch characteristic we took as a starting point, the omission of
determiners, in a meaningful way, involving a unidirectional relationship. Given this result,
let us have a look at possible patterns of co-occurrence, comparing our findings with that for
KiDKo/Mo.
Bare local NPs do occur in KiDKo/Mo, and they seem to follow the same patterns as
analysed for Kiezdeutsch, but they are not as common. (6) gives two examples:
(6a) ich war so GEStern internetcafé (KiDKo, Mo05WD)
I was PTCL yesterday internet.cafe
‘I was [in an] internet cafe yesterday.’
(6b) wir jehn doch auch HALle noch (KiDKo, Mo05WD)
we go PTCL also hall still
‘We will also go [to the] hall after all.’

9An anonymous reviewer suggests that it might also be the other way round, that is, that the absence of
prepositions supports the omission of determiners. We do not believe that this is the underlying pattern, given
that there are many instances of determiner omission in the presence of prepositions, while we find hardly any
cases with preposition omission in the presence of determiners. This strongly suggests that the implicational
relationship here goes from the omission of determiners to the omission of prepositions, not the other way round.
10Note that in standard German, (4c) would have an article (unlike its English counterpart).
12
While in KiDKo/Mu, there are 158 instances of constructions (of which 9 have an overt
determiner, see above), we count only 49 instances of this construction in KiDKo/Mo (all of
them without a determiner), and 31 instances are by the same speaker (MoH5WD). In order to
compare ratios of omitted versus overt prepositions, we focus on the prepositions in ‘in’ and
zu ‘to’, which account for the majority of all instances of this construction in the corpus.
Again, we made use of the POS annotations to retrieve all instances of prepositional phrases
with either zu or in, including all possible variants of the two forms with merged determiners
(im, inne, innen, inner, zum, zur and so forth). Then we manually checked these instances and
excluded all cases of non-local PPs from the counts. Table 2 shows the number of bare local
NPs which, in standard German, would have been realised with either in or zu, and compares
them to the number of local NPs where preposition (in/zu) and determiner have been realised.
Again the rate of omission is higher in the Kiezdeutsch data than in the more monolingual
speech community, and the chi-square test shows that this difference is statistically significant
(p=0.0028).
KiDKo/Mu KiDKo/Mo
bare local NPs (all preposition
omissions) 158 49
of those: bare local NPs (only in /
zu) 109 40
full local PPs (headed by in / zu) 661 439
% in/zu. (out of all in/zu) 14.16 8.35
Table 2: Frequency of bare local NPs vs. local PPs for KiDKo/Mu and KiDKo/Mo
This indicates that the option to use bare local NPs without prepositions is further developed
in Kiezdeutsch. As our data shows, this is not an isolated phenomenon, but cooccurs with a
grammatically related phenomenon, namely the omission of determiners in KiDKo/Mu. thus
establishing a first link within Kiezdeutsch.11
New directive particles
Another phenomenon that has been described for Kiezdeutsch, which is related to the
omission of pronouns, is the emergence of a new particle lassma, sometimes also just lass, in
speaker-inclusive directive (adhortative) speech acts as illustrated in (7):

11In the current investigation, we look at such grammatical relations between phenomena and focus on speech
communities, that is, we do not investigate correlations between phenomena for individual speakers (which
would also be possible in KiDKo), but rather cooccurrences within the multi- versus monolingual community of
phenomena that we find to be linked to each other by grammatical relations.
13
(7a) lass ma irgendwas MAchen (KiDKo, MuP1MK)
LASS MA something DOINF
‘Let’s do something.’
(7b) lass ma morgen SAUfen gehen (KiDKo, MuH11MD)
LASS MA tomorrow drink goINF
‘Let’s go out and drink tomorrow.’
(7c) lass ma CHILlen (KiDKo, MuH9WT)
LASS MA chillINF
‘Let’s chill.’
(7d) lass HINten hochgehn [directed at a group including the speaker]
LASS back up.goINF (KiDKo, MuP6MD)
‘Let’s go up back there.’
(7e) lass doch GLEISCH rugby spielen (KiDKo, MuH13MT)
LASS PTCL shortly rugby playINF
‘Let’s play rugby soon.’
These usages presumably go back to lass uns, where the imperative lass ‘let’ is combined
with a pronominal accusative object uns ‘us’ similar to English let us / let’s. In the examples
in (7), lass is not combined with such an object anymore, but appears as a bare form that
indicates a speaker-inclusive directive by itself. This bare form is frequently combined with a
particle mal, often reduced to ma in spoken language, that is commonly used to tone down
commands and thus often marks directive speech acts. Interestingly, lassma is not the only
one of its kind in Kiezdeutsch: it has been shown to be part of a new subsystem of directive
particles in Kiezdeutsch that involves another, speaker-exclusive element musstu (cf. Wiese
2009). Musstu is derived from a 2nd person singular form musst of the verb müssen , ‘must’,
followed by a cliticised personal pronoun (d)u, ‘you’. The combined form musst du / musstu,
and also just musst alone, can initiate speaker-exclusive directives, as illustrated in (8):
(8a) musst du ihn FRAgen (KiDKo, MuH17MA)
MUSST DU him askINF
‘Ask him / you have to ask him.’
(8b) musst du GAS geben (KiDKo, speaker SPK2 in transcript MuH12MD)
MUSST DU gas giveINF
‘Speed up / You have to speed up.’
(8c) musst RIEschen (KiDKo, speaker SPK3 in transcript MuH9WT)
MUSST smellINF
‘Smell it / You have to smell it.’
Together, the two patterns illustrated in (7) and (8) support an emerging subsystem of
directive particles with two elements that complement each other at the pragmatic level as
speaker-inclusive vs. -exclusive, while converging at grammatical levels:
14
They both form bisyllabic trochaic (lassma / musstu) or monosyllabic elements (lass /
musst).
They both occur in sentence-initial position.
They are both combined with infinitives.
While ending up as grammatical matches, the two elements do not follow identical
developmental pathways to get there. Speaker-inclusive lass(ma) is linked to the second
aspect of our primary phenomenon (the omission of determiners), namely the optionality of
pronouns, in this case uns. Speaker-exclusive musst(du), on the other hand, keeps its pronoun
(du) in the bisyllabic variant, but is based on a nonstandard option of verb-first (V1) order for
declaratives.12 While V1 occurs in other varieties of spoken German as well (cf. Önnerfors
1997 for an overview), in Kiezdeutsch this seems to be elaborated, making V1 a salient
characteristic, with broader options than those described for other contexts of informal spoken
German (cf. Dirim & Auer 2004:Ch.6, Wiese 2009). This V1 option allows the finite verb,
musst, to stand in first position and thus precede rather than follow its subject, du, which can
then be cliticised or even omitted. The V1 pattern is part of a broader range of options for the
German forefield, the domain in front of the finite verb:13 in standard German declaratives,
this domain is usually occupied by exactly one constituent (verb-second, V2), but in
Kiezdeutsch it permits, in addition to this, either none (V1) or two constituents (verb-third,
V3) as well. We will come back to these options in more detail in section 4.2.
For our current discussion, the relevant point is that we find an element lass(ma) in
Kiezdeutsch that, again, is systematically related to the option of omitting determiners (in this
case, pronouns), and forms in turn part of a new subsystem of directive particles. This
subsystem includes another element musst(du) which depends on a different Kiezdeutsch
characteristic, namely a broader range of word order options for declaratives. Taken together,
these different relationships support a further network of linked-up elements.
Both patterns occur in the more monolingual part of the corpus, KiDKo/Mo, as well.
Table 3 gives the numbers for both corpora, for the new directives lass(ma) and musst(du)

12Note that in principle, one could assume that such sentences as in (8) are elliptic, with a sentence-initial
adverb such as dann ‘then’ deleted. Since there is no evidence for such adverbs, though, we follow other
discussions of V1 in German (e.g., Önnerfors 1997, Simon 1998) in regarding only those sentences as elliptic
where an obligatory constituent is missing..
13In a standard topological description of German sentences going back to Drach (1937), Bech (1955), a
sentence can be divided into three main fields deliminated by the “verbal brackets”, that is, the positions for the
finite verb in verb-second main clauses (left verbal bracket) and for nonfinite verb forms in main clauses or the
finite verb in SOV subordinate clauses (right verbal bracket). The “forefield” is the domain preceding the left
verbal bracket, the “middle” field is between the two brackets, and the “post-field” is the slot for extraposed
material following the right bracket.
15
compared to the standard forms lass uns (ma) with an overt pronoun uns, and du musst with
V2 order.
KiDKo/Mu KiDKo/Mo
lass(ma(l)) 118 37
lass uns (ma(l)) 15 4
% lass(ma(l)) 88.72 90.24
musst ((d)u) 44 8
du musst 75 56
% must ((d)u) 36.97 12.50
Table 3: Corpus frequencies for instances of directive lass(ma(l)) / lass uns (ma(l)) and
musst((d)u) / du musst, and% of occurrences for each pattern
As the table shows, the number of hits in the corpus for these patterns is much smaller, in
particular in the case of KiDKo/Mo. If we nevertheless compare the frequencies statistically,
we do not find a difference for lassma between KiDKo/Mu and KiDKo/Mo, but a highly
significant difference for musstu (p<0.001), again indicating an advantage for the multiethnic
speech community. This suggests that at least for this area, the novel subsystem of directive
particles we find here is further developed in Kiezdeutsch, thus adding further connections to
the network of phenomena we identified.
New light verb constructions
The third phenomenon we are going to investigate here is that of new light verb constructions
(short: LVCs). In these constructions, a light verb is combined with a bare NP, forming a
constituent that can often be paraphrased by a single verb that is morphologically related to
the noun, as illustrated in (9):
(9a) der hat doch zu viel STREIT gemacht (KiDKo, MuH2WT)
he has PTCL too much quarrel made
‘He quarreled too much.’
(9b) hab isch SCHOCK bekomm (KiDKo, MuH19WT)
have I shock received
‘I got a shock / I got shocked.’
(9c) wir hätten ARSCHkarte gehabt (KiDKo, MuH11MD)
we had arse.card had
‘we would have gotten the short end of the stick’
(9d) macht ihr GLEISCH farbe (KiDKo, SPK2 in MuH9WT)
make you immediately colour
‘Do you colour it [hair] right away?’
LVCs are also known from other varieties of German, including standard German. In standard
German, LVCs are lexicalised and form a fairly closed class, while in Kiezdeutsch, we find
novel instances indicating that this pattern has become productive here. The occurrence of
16
bare NPs here suggests another link to our initial phenomenon, the omission of determiners:
in Kiezdeutsch this omission seems to support a pattern in another related domain, in this
case, that of light verbs.
LVCs in general are characterised by a morpho-syntactic reduction of the NP. This can
mean the omission of determiners we find here, but it can also restrict a DP to singular or
plural, definite or indefinite occurrences, as in some LVCs in standard German. This morpho-
syntactic reduction of the NP is complemented by a semantic reduction of the verb, which
gets bleached as a light verb and basically reduced to indicating aktionsart.
Morphosyntactically, the verb provides the grammatical integration of the NP, while the NP
provides the main lexical content for the construction. It is this division of labour at
morphosyntactic and semantic levels that is at the core of LVCs and that we also see in the
Kiezdeutsch examples. The option to omit determiners hence supports the further
development of such constructions, but it is only part of the story: bare NPs enter into a
pattern that provides its own grammatical dynamics and involves such additional features as
their interaction with semantically bleached verbs.
As in the other cases, we also find instances in the more monolingual part of the
corpus, KiDKo/Mo. (10) gives two examples:
(10a) nachher (krieg) ICH wieder ANschiss (KiDKo, Mo01MD)
afterwards get I again dressing-down
‘Afterwards, I will get a dressing down / will be dressed down.’
(10b) und dann hab ich ZEUgenaussage gemacht gehabt (KiDKo, Mo12MD)
and then have I witness.statement made had
‘And then I made a witness statement.’
The case numbers here are too small to support a proper comparison: Table 4 gives absolute
and normalised14 numbers for the two corpus parts, both for tokens and for different types of
new LVCs.
KiDKo/Mu
(absolute) KiDKo/Mu
(normalised) KiDKo/Mo
(absolute) KiDKo/Mo
(normalised)
tokens 24 1.05 10 1.02
types 14 9
Table 4: Corpus frequencies for instances of new LVCs (absolute numbers and counts
normalised by corpus size)

14We report normalised frequencies per 10,000 tokens. When dividing the raw frequency counts by corpus size,
we excluded pauses, punctuation, and all passages with uninterpretable material or language material other than
German, yielding corpus sizes of approx. 229,000 tokens for KiDKo/Mu, and approx. 98,000 tokens for
KiDKo/Mo.
17
If we look at type and token numbers, though, this points to a potential difference between the
two corpora that would be interesting to investigate further for data with higher frequencies:
in our data, we find nearly the same number of types and tokens in the more monolingual
context, KiDKo/Mo, but nearly twice as many tokens as types in the multilingual context of
KiDKo/Mu. If this is borne out by more data, it could indicate a tendency towards
lexicalisations, that is, the emergence of new, more established and lexicalised LVCs in
Kiezdeutsch.
One promising candidate for this is the combination of Schock ‘shock’ as a bare noun
with a light verb bekommen or kriegen ‘to get/receive’, as illustrated in (9b) above.15 This
combination appears four times in KiDKo/Mu (no hits in KiDKo/Mo). Two of the hits follow
the same intonational pattern as indicated in (9b), with a strong emphasis on Schock. This is a
pattern that we also found in field notes from conversations between young people in
multiethnic inner-city neighbourhoods of Berlin in parks, in playgrounds, and in the street; cf.
(11) [(11a) is from a conversation between two girls, the speaker is 11 years old and of
monolingual German background; (11b) is from a conversation in a group of four young
women, the speaker is 15 years old and of bilingual Turkish-German background]:
(11a) hab=ich voll SCHOCK bekomm (field notes, Kreuzberg 2014)
have I PTCL shock receiv(ed)
‘I really got a shock / I was really shocked.’
(11b) mann, ich hab SCHOCK bekomm (field notes, Neukölln 2013)
man I PTCL shock receiv(ed)
‘Man, I got a shock / I was shocked.’
At present, there is not enough data to draw well-substantiated conclusions from this. If this
pattern gets confirmed by subsequent empirical investigations, it would indicate a further
development for novel LVCs in Kiezdeutsch, from a pattern that is initially spontaneously
produced, to one that gets established and more fixed by frequent usage and might then be
lexicalised. These lexicalised forms can then be identified as an additional point in the rich
network of linguistic phenomena that sets forth from the core pattern of determiner omission
we observe for Kiezdeutsch.
In the following section, we are going to have a look at a phenomenon from a different
area, namely that of word order. As mentioned above, this complements our findings on links
between different phenomena within Kiezdeutsch with evidence from a single phenomenon
where we find links between different linguistic levels.

15The reduction of the full infinitive form bekommen to bekomm is phonologically motivated (shwa-reduction
and assimilation of two nasals) and common in other variants of informal spoken German as well.
18
4.2 Links between different linguistic levels
As briefly mentioned in connection with directive musst(du) above, we find nonstandard word
order options in Kiezdeutsch that concern the forefield in declarative sentences. In German
declaratives, the finite verb is in a position forming what is commonly called the ‘left
sentence bracket’, marking the boundary of a left-peripheral domain, the ‘forefield’, that can
be used to indicate information-structural aspects. In standard German declaratives, the
forefield is usually occupied by exactly one constituent (XVfin), leading to the well-known
verb-second order (V2) for the finite verb. In Kiezdeutsch, this constraint seems to be weaker,
such that the verb can be in first position, with no element in the forefield (V1), as mentioned
above, and also in third position, with two elements occupying the forefield (V3). In what
follows, we are going to concentrate on V3 options for declaratives, since for this pattern we
can draw on analyses from previous research on Kiezdeutsch.
In contrast to what is sometimes implied in descriptions of this pattern (e.g., Dirim &
Auer 2004), it does not constitute a replacement of the German XVfin order with (Adv) SV
which would be similar to, for instance, English. Rather, this phenomenon concerns
specifically the forefield while keeping the verb bracket intact, and thus still follows the
general pattern of German sentences (cf. Wiese 2013). Crucially, only the finite verb stands in
the left sentence bracket: non-finite parts of the predicate and separable verb particles remain
in the right sentence bracket after the ‘middle field’ which includes possible objects, just as in
standard German. This, then, constitutes the German verb bracket (with its SOV basis), rather
than a typologically different SVO.
(12) gives some data from KiDKo/Mu that illustrates this, and some further examples
from other studies (the latter do not involve an element that could go in the right verb bracket,
e.g., infinitives or participles, and thus do not provide grounds for a distinction of SVO from
V3):
(12a) danach sie hat misch AUCH geblockt (KiDKo, MuH9WT)
afterwards she has me also blocked
‘After that, she blocked me, too [in an internet-based social network].’
(12b) da man kann se fast ein JAHR behalten (KiDKo, speaker Rxxx
there one can them nearly a year keep in MuH1WD)
‘There, one can keep them for nearly a year [= books etc., from a library].’
(12c) HEUte ich werd meine zigaRETten mitbringen und frag ihr (KiDKo, MuH11MD)
today I will my cigarettes with.bring and ask her
‘Today, I will bring my cigarettes and will ask her.’
(12d) SIEBte man muss (-) viel äh (unverständlich) machen (KiDKo, MuH25MA)
seventh one must a.lot ahm (INCOMPREHENSIBLE) make
19
‘In seventh [grade] one has to, ahm [], do a lot.’
(12e) wenn die frau keine JUNGfrau is die ziehen diese frauen nackt AUS (KiDKo,
when the woman no virgin is they dress these women naked un- MuH21MT)
‘If the woman is not a virgin, they undress these women altogether.’
(12f) jetz ich bin 18 (Auer 2003: 259)
now I am 18
‘Now I am 18.’
(12g) danAch wir warn auf KLO (Kern & Selting 2006: 248)
afterwards we were on loo.’
‘Afterwards, we went to the loo.’
The V3 option is not restricted to Kiezdeutsch. Schalowski (to appear) provides evidence for
this pattern from spoken German, coming from a range of different contexts in predominantly
monolingual German settings, among them TV discussions, conference talks, and
conversations among friends (cf. also Wiese 2012). (13) gives an example from KiDKo/Mo:
(13) EY vorhin ick bin so na= HAUse jelaufen und … (KiDKo, Mo05WD)
ey earlier I have PTCL to=home went and
‘Ey, earlier I was going home, and …’
However, again, this option seems to be more readily available in multilingual contexts: (13)
is one of only two instances of V3 in KiDKo/Mo, compared to over 100 instances in
KiDKo/Mu. Table 5 sets the number of declaratives with V3 order in relation to the overall
number of declarative utterances in the two corpus parts (For this, ‘declarative’ has been
operationalised as an utterance that includes at least one finite verb and is represented in the
corpus’ normalisation layer by a unit that is segmented by a single dot. One-word utterances
and unfinished utterances have been excluded.)
KiDKo/Mu KiDKo/Mo
no. of declaratives 19,324 8,065
no. of V3 declaratives 126 2
% V3 declaratives 0.65 0.02
Table 5: Corpus frequencies for declaratives
Again, then, this indicates an advantage for the multilingual speech community.16 This time,
the development is not related to options at the DP level – such as the omission of determiners
that we found to play a central role for the phenomena discussed in the previous sections.
Instead, we find variation at a higher functional layer, namely that of the sentence, where we
now have a broader range of options for the forefield.

16Note, though, that we did not test for significance here, given the extremely low number of occurrences (only 2
findings) in the Mo corpus.
20
And again, these options are not random, but support systematic patterns. An
interesting aspect of the V3 phenomenon is that the patterns we can identify for the forefield
interact at different, distinct but related linguistic levels. Let us sketch this briefly.
Our data suggests that at the core there is a pattern at the level of syntactic functions.
We find here a strong tendency towards an ordering of ‘adverbial>subject’ in the forefield: in
92% of the V3 instances identified in KiDKo/Mu, the first element is an adverbial, in 94% of
the instances, the second one is a subject, and in 90% of all V3 instances, we find the
combined order ‘adverbial>subject’ before the finite verb. This order shows a surface
similarity to Adv SVO in all cases that lack a candidate for a position in the right sentence
periphery, a similarity that might further support this pattern in Kiezdeutsch, where speakers
are usually familiar with Adv SVO from the German of first-generation immigrants (often the
grandparents or parents of the speakers or their friends). The strong tendency towards
adverbials in the left-most position corresponds, in the domain of syntactic categories, to a
tendency, albeit less strong, towards phrases headed by an adverb or a preposition in the first
position (in 79% of the cases).
The grammatical pattern we find for syntactic functions (and syntactic categories) is
linked to preferences at information- and discourse-structural levels (cf. Wiese 2013,
Schalowski to appear): the first element is most often a framesetter (as, for instance, in (12d)
above) or a discourse linker such as dann ‘then’ or danach ‘afterwards’ (cf. for instance, (12a)
above), while the second is a topic, that is, refers to the element that the sentence is about (cf.,
e.g., Krifka 2006). ‘Framesetter/linker > topic’ is an ordering we find in 94% of the V3 cases
in KiDKo/Mu where the information-/discourse-structural status could be identified
unambiguously. If we take into account all V3 hits in KiDKo/Mu, we observe a framesetter or
linking element as a first element in 63% of the cases, and a topic as a second element in 81%.
At this level, the ordering is supported by a general preference, independently of particular
language-specific grammars, to place topical elements in the left periphery.17 As our data
shows, this preference can be realised more directly in the multilingual context that
Kiezdeutsch draws on, a context that, as argued elsewhere (e.g., Wiese 2009, 2013), provides
a more liberal grammatical system to support such options: in a speech community with a
high incidence of multilingualism, we can expect more openness towards new developments
compared to more monolingual contexts, given that speakers are familiar with more diverse
repertoires and higher degrees of linguistic variation.

17Cf. Jacobs (2001), Krifka (2006).
21
The preference for topical subjects in second position in turn supports, at the level of
syntactic categories, pronominal DPs in that place (in 72%). At the level of semantics, the
trend towards framesetters or discourse linkers followed by topics corresponds to a majority
of temporal expressions in the first position (57% of all V3 cases) and reference to human
being in the second position (93% of all V3 instances).
The different findings at syntactic, information- and discourse-structural, and semantic
levels, then point to a network that links up different grammatical and pragmatic domains in a
systematic way. Figure 3 illustrates this.
Figure 3: Links between grammatical and pragmatic domains for V3 forefields
The evidence we found for V3 hence complements that from the previous sections, for
developments connected to determiner omissions in Kiezdeutsch: taken together, our findings
indicate links between different grammatical and pragmatic levels for the same phenomenon
(as in the case of V3) as well as links between different linguistic phenomena (as for the
absence of determiners, new particles, and bare local expressions), suggesting a rich ecology
of networks that supports an interpretation of coherence for this new German dialect.
5. Coherence, language variation and change: the ecology of new urban dialects
In the present paper, we have combined qualitative and quantitative analyses in an
investigation of linguistic coherence. As our starting point, we proposed a view of coherence
as an interpretation of empirical observations rather than something that would be “out there
in the data”. We argued that this interpretation should be based on evidence of systematic
links between linguistic phenomena, as established by patterns of covariation between
phenomena that can be shown to be related at linguistic levels.
SYN – syntax; IS – information
structure
;
SEM
semantics
22
Against this background, we identified as a particularly interesting domain for
investigations into coherence the new urban dialects that have emerged in multiethnic and
multilingual Europe over the last decades. These dialects emerge in highly diverse contexts
characterised by a large range of contact situations, diverse repertoires, and the fast pace of
language variation and change this brings with it, thus providing an interesting challenge for
assumptions of coherence.
For our case study, we chose a set of phenomena that have been described for
Kiezdeutsch, a new dialect from multilingual urban Germany. Qualitative analyses pointed to
linguistic relationships between different phenomena and between pragmatic and linguistic
levels. Quantitative analyses, based on corpus data from KiDKo, pointed to systematic
differences between the main corpus KiDKo/Mu, with data from a multiethnic and
multilingual context, and the complementary corpus KiDKo/Mo, with data from a mostly
monoethnic and monolingual (German) context, showing advantages for the Kiezdeutsch data
provided by KiDKo/Mu.
Taken together, this indicates patterns of covariation that support an interpretation of
coherence for this new dialect: our findings point to an interconnected linguistic system,
rather than to a mere accumulation of individual features. Figure 4 brings together our results
for the different links in a complex network, relating to the depiction of coherence we
suggested in section 2 (Figure 1 above):
Figure 4: A network of linked phenomena in Kiezdeutsch
The graphic renders as a central node the option to omit determiners that we took as our point
of departure, and which we found to be significantly more pronounced in the Kiezdeutsch
data provided by KiDKo/Mu than in KiDKo/Mo. As the graphic illustrates, this option
supports three different phenomena we discused here: (1) the occurrence of bare, determiner-
omission of
determiners
bare NPs in
local expressions
omission of
lexical prepositions
bare nouns
in LVCs
semantic bleaching
of verbs
poss. LVC
lexicalisations
lassma
musstu
V1
declaratives
V3
declaratives
more liberal
forefiel
d
23
less NPs in local expressions, (2) the use of bare nouns in light verb constructions (LVCs),
and (3) the emergence of a directive particle lassma as a merged form without pronoun (from
lass uns mal ‘let us PTCL’). Each of these phenomena were in turn shown to be connected with
other developments: (1) bare NPs in local expressions support the omission of lexical
prepositions; (2) the use of bare nouns in LVCs is linked up with the semantic bleaching of
the verbs that enter these constructions and, together with these, can support lexicalisations of
LVCs; (3) lassma forms a subsystem of speaker-inclusive vs. -exclusive directive particles
with another element musstu (from musst du ‘must you’), whose emergence is supported by
the option to use V1 declaratives; this option in turn draws on a more liberal forefield in
Kiezdeutsch, which also supports the use of V3 declaratives.
The rich network we identified here suggests that when accessing Kiezdeutsch
characteristics in their linguistic behaviour, speakers do not only draw on a feature pool in the
sense of Mufwene (2001) or Cheshire et al (2011), that is, a pool of competing features
coming from different sources in language contact situations. Kiezdeutsch is a resource that
goes beyond this, it points to the formation of something we can describe as a feature pond: a
rich ecology that brings forth interconnected patterns at different levels.
As emphasised above, this does not mean that speakers are forced to use Kiezdeutsch
in an “everything or nothing”-arrangement: a feature pond still leaves it open for speakers to
pick and choose from its elements and patterns. As shown in more detail elsewhere (cf. Wiese
2013), the feature pond perspective thus allows us to acknowledge systematic structural
relations within and between linguistic patterns without neglecting speakers’ sociolinguistic
repertoires and their selective choices from a broader range of linguistic resources in different
communicative situations.
Note that accounting for Kiezdeutsch as a systematic variety along these lines does not
entail separating it from other variants of German. Rather, its characteristics are embedded in
the general range of options in contemporary German, pointing not only to internal coherence,
but also to something we could call external coherence, that is, an integration with other
variants of German. Taking an opposite view, Auer (2013) claims that such Kiezdeutsch
characteristics as discussed here, that is, local expressions without prepositions, the omission
of determiners, word order variations, and others, “intervene deeply in the structures of
autochthonous German in its standard and nonstandard forms” and “would have the potential
to constitute a new variety that would differ substantially from autochthonous German” (Auer
2013: 37f; German original, our translation).
24
We believe that the internal coherence that supports a new variety here does not set
Kiezdeutsch apart from “autochthonous German” as an allochthonous element in this way.
Rather, if one takes a closer look at the data, one typically finds that Kiezdeutsch
characteristics point to an ongoing change that one also sees in other German varieties,
although the phenomena in question might not (yet?) be developed as far, that is, they might
not be as systematic, as widespread or as frequent as in Kiezdeutsch.
This is certainly the case for the reduction of determiners, which reflects a general
tendency in modern German, from phonological reduction and cliticisation to full omission.18
And as became evident in the previous sections, the Kiezdeutsch phenomena we investigated
for KiDKo/Mu were, in each case, also attested in the complementary corpus KiDKo/Mo with
data from a more monolingual speech community, indicating a relative rather than absolute
difference. Kiezdeutsch characteristics are not alien to German, nor are they necessarily
novel. To take two examples from our analyses that are frequently mentioned in connection
with Kiezdeutsch (and also in Auer 2013): Bare local NPs, at least for public transport stops,
were already the subject of language purists’ dismay nearly a century ago – and possibly
much earlier, cf. (14) below. And as far back as in the 15th century, we find sentences that
show a strong similarity to the V3 pattern we described for Kiezdeutsch (cf. (15)),19
suggesting that the move to V2 in New High German might not have been as inflexible as
usually assumed, but that word order options that were more widely available earlier might
still have been maintained in informal spoken language:20
(14) “Today, we are nearly at a point to say: wherever it is possible, the preposition will be
left out, whatever relations we can figure out just so, will not be expressed. This means
nothing other than: we, that is, our language become silenced. At the train station one
hears: Does the train go V Berlin? No, the train goes V Hamburg; on the tram: I want to
go V churchyard.” (Briegleb 1932:16)
(15) Dar nach die edel kungin fuer enhalb Ofen auf des Laslaes Wans gueter …
there after the noble queen went from Ofen on the Laslaes Wans estates
“Afterwards the noble queen went from Ofen to Laslaes Wans’ estates …”
This suggests that Kiezdeutsch, while supporting an interpretation of internal coherence, is
not disconnected on the outside either, but fully integrated within the general domain of
German: an integration with other variants of German that defies a distinction of
“autochthonous” and “allochthonous” German, not only at the level of speakers, but also at

18Cf. e.g., Leiss (2010) on the decline of article usage in German.
19Cf. Speyer (2008); from: Helene Kottaner: Die Denkwürdigkeiten der Helene Kottanerin, Wien 1445-1452.
20Cf. Wiese (2012, 2013) for a more detailed discussion. We thank Barbara Schlücker for pointing out the
Briegleb (1932) reference to us; the quote is in German, our translation; “V” as in the original, indicating
omissions.
25
the level of linguistic systems. This, then, provides us with a special vantage point into
language variation and change in modern German, offered by a dynamic new dialect that is
“in full development, in contrast to the relatively static traditional dialects” (Hinskens 2011).
It is these dynamics that make Kiezdeutsch a particularly interesting object of research, while
the evidence we find for internal and external coherence suggests that what Kiezdeutsch has
to offer is relevant over and above this young dialect proper.
References
Auer, Peter (2003). 'Türkenslang': Ein jugendsprachlicher Ethnolekt des Deutschen und seine
Transformationen. In: Annelies Häcki Buhofer (Ed.), Spracherwerb und Lebensalter.
Tübingen: Francke. 255-264
Auer, Peter (2013). Ethnische Marker im Deutschen zwischen Varietät und Stil. In: Arnulf
Deppermann (Ed.), Das Deutsch der Migranten [IDS Jahrbuch 2012]. Berlin, New
York: de Gruyter. 9-40.
Bech, Gunnar (1955). Studien über das deutsche Verbum infinitum. Tübingen: Niemeyer.
Bradford Hill, Austin (1965). The environment and disease: Association and causation. Proc.
Royal Society of Medicine 58;5: 295-300.
Briegleb, Otto (1932). Wider die Entartung der Sprache / durch Erstarrung - Falsche
Zusammensetzung - Wortverstümmelung. Leipzig: Verlag von Friedrich Brandstetter.
Brillinger, David Ross (1975). Times Series: Data Analysis and Theory. New York: Holt-
Rinehart.
Cheshire, Jenny; Kerswill, Paul; Fox, Susan, & Torgersen, Eivind (2011). Contact, the feature
pool and the speech community: The emergence of Multicultural London English.
Journal of Sociolinguistics 15;2: 151-196
Dirim, İnci, & Auer, Peter (2004). Türkisch sprechen nicht nur die Türken. Über die
Unschärfebeziehungen zwischen Sprache und Ethnie in Deutschland. Berlin: Walter de
Gruyter.
Drach, Erich (1937). Grundgedanken der deutschen Satzlehre. Frankfurt: Diesterweg.
Eckert, Penelope (2003). Elephants in the room. Journal of Sociolinguistics 7;3: 392-397.
Glauber, Roy J. (1963). The quantum theory of optical coherence. Physical Review 130;6:
2529-2539.
Granger, Clive William John (1977). Forecasting Eonomic Time Series. New York:
Academic Press.
Hinskens, Frans (2007). New types of nonstandard Dutch. In: Christian Fandrych & Reinier
Salverda (Hg.), Standard, Variation und Sprachwandel in den germanischen Sprachen.
Tübingen: Narr. 281-300.
Hinskens, Frans (2011). Emerging Moroccan and Turkish varieties of Dutch: ethnolects or
ethnic styles? In: Kern & Selting (Eds.), 101-139.
Hume, David (1740/1975). A Treatise of Human Nature: Being an Attempt to introduce the
Experimental Method of Reasoning into Moral Subjects. Ed. L. A. Selby-Bigge.
Oxford: Clarendon Press.
Jacobs, Joachim (2001). The dimensions of topic-comment. Linguistics 39: 641-681.
Jacomine Nortier & Bente A. Svendson (Eds.) (2014). Language, Youth, and Identity in the
21st Century. Oxford.
Kallmeyer, Werner, & Keim, Inken (2003). Linguistic variation and the construction of social
identity in a German-Turkish setting. A case study of an immigrant youth-group in
26
Mannheim, Germany. In: Jannis Androutsopoulos & Alexandra Georgakopoulou (Eds.),
Discourse constructions of youth identities. Amsterdam: Benjamins. 29-46.
Kern, Friederike & Selting, Margret (2006): Einheitenkonstruktion im Türkendeutschen:
Grammatische und prosodische Aspekte. Zeitschrift für Sprachwissenschaft 25: 239-
272.
Kern, Friederike, & Selting Margret (Eds.) (2011). Ethnic Styles of Speaking in European
Metropolitan Areas. Amsterdam: Benjamins.
Krifka, Manfred (2007). Basic notions of informations structure. Interdisciplinary Studies on
Information Structure 6: 12-49.
Leiss, Elisabeth (2010). Koverter Abbau des Artikels im Gegenwartsdeutschen. In: Dagmar
Bittner & Livio Gaeta (Hg.), Kodierungstechniken im Wandel. Das Zusammenspiel von
Analytik und Synthese im Gegenwartsdeutschen. Berlin, New York: de Gruyter. 137-
158.
Lewis, Clarence Iriving (1946). An Analysis of Knowledge and Valuation. La Salle, IL: Open
Court.
MacLane, Saunders (1965). Categorical algebra. Bulletin of the American Mathematical
Society 71;1: 40-106.
Mufwene, Salikoko. S. (2001). The Ecology of Language Evolution. Cambridge University
Press.
Nortier, Jacomine, & Svendsen, Bente A. (Eds.) (2014), Language, Youth, and Identity in the
21st Century. Oxford University Press.
Pohle, Maria (2013). Registerdifferenzierung bei Kreuzberger Jugendlichen. MA thesis,
University of Potsdam.
Quist, Pia (2008). Sociolinguistic approaches to multiethnolect: Language variety and stylistic
practice. International Journal of Bilingualism 12: 43-61.
Quist, Pia (2010). Untying the language-body-place connection: a study on linguistic variation
and social style in a Copenhagen community of practice. In: Peter Auer & Jürgen Erich
Schmidt (Eds.), Language and Space: Theories and Methods. Berlin, New York: de
Gruyter [HSK 30.1]. 632-648.
Rampton, Ben; Maybin, Janet, & Rob, Celia (2014). Methodological foundations in linguistic
ethnography. Working Papers in Urban Language & Literacies 125. King’s College
London.
Rehbein, Ines, & Schalowski, Sören (to appear). „STTS goes Kiez.“ Experiments on
Annotating and Tagging Urban Youth Language. Journal for Language Technology and
Computational Linguistics.
Schalowski, Sören (to appear). From adverbial to discourse connective: the function of
German 'dann' and 'danach' in non-canonical prefields. To appear in: Mirjam Fried &
Eva Leheckova (Eds.), Connectives as a Functional Category: Between Clauses and
Discourse Units. Amsterdam: Benjamins.
Schiller, Anne; Teufel, Simone; Stöckert, Christine, & Thielen, Christine (1999). Guidelines
für das Tagging deutscher Textcorpora mit STTS. Technical report, University of
Stuttgart and University of Tübingen.
Shimony (1955). Coherence and the axioms of confirmation. Journal of Symbolic Logic 20:1-
13.
Siegel, Vanessa (2014). Präpositionalphrasen ohne Präpositionen? Zur syntaktischen
Reduktion im ‚Türkendeutschen‘. In: Helga Kotthoff & Christine Mertzlufft (Eds.),
Jugendsprachen. Stilisierungen, Identitäten, mediale Ressourcen. Frankfurt am Main:
Peter Lang. 67-94.
Speyer, Augustin (2008). Doppelte Vorfeldbesetzung im heutigen Deutsch und im
Frühneuhochdeutschen. Linguistische Berichte 216: 455-485.
27
Svendsen, Bente Ailin, & Quist, Pia (Eds.) (2010), Multilingual Urban Scandinavia: New
Linguistic Practices. Clevedon: Multilingual Matters.
Vertovec, Steven (2007). Super-diversity and its implications. Ethnic and Racial Studies 30;6:
1024-1054.
Wiese, Heike (2009). Grammatical innovation in multiethnic urban Europe: new linguistic
practices among adolescents. Lingua 119: 782-806.
Wiese, Heike (2012). Kiezdeutsch. Ein neuer Dialekt entsteht. München: C.H. Beck.
Wiese, Heike (2013). What can new urban dialects tell us about internal language dynamics?
The power of language diversity. Linguistische Berichte, Special Issue 19: 208-245.
Wiese, Heike (to appear). “This migrants’ babble is not a German dialect!” – The interaction
of standard language ideology and ‘us’/‘them’-dichotomies in the public discourse on a
multiethnolect. To appear in Language in Society 2015;3.
Wiese, Heike; Freywald, Ulrike; Schalowski, Sören, & Mayr, Katharina (2012). Das
KiezDeutsch-Korpus. Spontansprachliche Daten Jugendlicher aus urbanen
Wohngebieten. Deutsche Sprache 2: 97-123.
Wolkowski, Zbigniew William (2007).The concept of coherence considered as a systems
isomorphism. Systems Research 9;4:61-66.
... Kiezdeutsch, an urban vernacular of German characterised by a multiethnic and multilingual speaker base (Wiese 2006(Wiese , 2009(Wiese , 2013, has received repeated attention for a verb-third (henceforth V3) linearisation involving an initial adverb and a preverbal nominative subject DP (1) (Alexiadou & Lohndal 2018;Freywald et al. 2015;Hinterhölzl 2017;Schalowski 2015Schalowski , 2017Sluckin & Bunk 2023;te Velde 2017;Walkden 2017;Wiese 2006Wiese , 2009Wiese , 2012Wiese , 2013Wiese et al. 2009;Wiese & Rehbein 2016;Wiese & Müller 2018;Wiese et al. 2020). 2 (1) jetzt now er he explodiert explodes oder or so so 'now he'll explode or something.' (KiDKo,MuH25MA_11) V3 is remarkable because Standard German (henceforth SG) has a stricter verb-second (V2) constraint (see den Besten 1983;Haider 2010;Holmberg 2015) in which only one constituent may precede the finite verb (VFIN) in matrix contexts. ...
... As introduced, Kiezdeutsch is a vernacular characterized by multilingual and multiethnic speakers in urban areas of Berlin (Wiese 2006(Wiese , 2009), a dynamic arguably facilitates innovative language use (Wiese 2006(Wiese , 2009(Wiese , 2013Wiese et al. 2009;Wiese & Rehbein 2016). Thus, we first ground this study in the literature on language contact and sociolinguistic diversity. ...
... However, it is not clear what acquisitional mechanism is responsible, aside from less homogeneous input; the endogenous interaction between acquisition mechanisms and multilingualism requires further exploration. Furthermore, while Wiese (2009Wiese ( , 2013, Wiese & Rehbein (2016), and Wiese & Müller (2018) view an increased frequency of V3 as novel, Wiese and her co-authors do not consider it an innovation but rather the rediscovery of a minor-use pattern in spoken German. We will revisit this particular claim critically in Section 7. ...
Preprint
Full-text available
This study revisits the V3 linearisation AdvP>Subject>finite verb in Kiezdeutsch, comparing it to resumptive verb-third Left Dislocation and Hanging Topic Left Dislocation. Using corpus data, preverbal object DPs are shown to almost never occur across verb-third distributions, yet preverbal nominative subjects and spatio-temporal elements are unproblematic. This behaviour is argued to involve a low C-domain position encoding a Subject of Predication requirement (cf. Cardinaletti 2004) tied to aboutness and nominative Case-assigning features, but not a strict D-related subject EPP. Based on comparison with other corpora and analysis of metadata, speakers from non-German speaking homes, i.e. successive bilinguals are argued to have innovated this property. A novel account is suggested for the emergence of V3 based on claims that it results from a natural informational order (Wiese et al. 2020), which is formalized as a Minimal Default Grammar (Roeper 1999) available to children before they fully acquire CP and TP. Children acquiring a V2 language must either reject V3 or incorporate it into a V2 syntax. Lacking adequate counter-evidence in their input, Kiezdeutsch speakers do the latter. (Forthcoming Journal of Germanic Linguistics)
... In the second instance, I examine non-canonical verb-third linearisations in Kiezdeutsch (Wiese 2006(Wiese , 2009Wiese et al. 2009;Wiese and Rehbein 2016;Wiese and Müller 2018;Freywald et al. 2015;Te Velde 2017;Hinterhölzl 2017;Walkden 2017;Alexiadou and Lohndal 2018;Westergaard et al. 2019), a contact variety of German, which violate the verb-second (henceforth V2) constraint, i.e., only one XP (of any type) may precede the finite verb (den Besten 1983). This verb-third distribution is remarkable because it appears to exclusively involve a distribution involving a preverbal subject and a clause initial frame/scene-setting adverbial adjunct. ...
... In Part II of this dissertation, I discuss the apparent semi-optional subject position in the German contact variety, or rather multiethnolect Kiezdeutsch (Wiese 2006(Wiese , 2009Wiese et al. 2009Wiese et al. , 2012Wiese 2012;Wiese and Rehbein 2016;Wiese and Müller 2018;Wiese et al. 2020;Freywald et al. 2011Freywald et al. , 2015Te Velde 2017;Walkden 2017;Hinterhölzl 2017;Alexiadou and Lohndal 2018;Westergaard et al. 2019). As briefly introduced above, Kiezdeutsch allows verb-third orders involving clause initial adverbs only in the presence of preverbal subjects and not typically other XPs: This behaviour resembles orders in SVO and non V2 languages. ...
... Walkden (2017) finds that young monolingual speakers produce significantly fewer and more unreliable instances of %V3; it was not clear that these instances were not errors. However, limited instances of %V3 have also been attested for adult spoken German (Schalowski 2015(Schalowski , 2017Wiese and Rehbein 2016;Wiese and Müller 2018;Bunk 2020;Wiese et al. 2020). Wiese and Müller (2018) and Bunk (2020) consider %V3 a core but overlooked feature of the spoken German repertoire, beyond its notoriety in Kiezdeutsch. ...
Thesis
Full-text available
This dissertation approaches the syntax of structures involving non-canonical subjects and non-canonical subject positions. In particular, I investigate two phenomena: Locative Inversion (LI) in English, French, Italian, and Hebrew; and well-known verb-third violations of the verb-second (V2) rule in Kiezdeutsch, an urban contact variety of German. In LI a spatio-deictic XP appears to occupy the preverbal canonical subject position, while the canonical nominative subject DP surfaces in a postverbal inverted position. From a theoretical and empirical perspective, I compare the distribution of different covert and overt arguments participating in LI and the availability of LI in embedded and matrix contexts crosslinguistically. The second case study concentrates on remarkable Kiezdeutsch V2 violations, as they appear to follow a regular order of [frame-setting adverb → Subject → finite verb]; this is remarkable not only on account of it violating an otherwise strict V2 requirement, but it also indicates the innovation of a subject position that Standard German does not have. I carry out a corpus study and find that an apparent subject requirement extends to other verb-third resumptive- dislocation phenomena, yet it is shown that we cannot understand this requirement in the sense of an EPP position associated purely with nominative DP subjects. From a theoretical perspective, this dissertation develops and applies a theory of subject requirements, which is able to account for the breadth of investigated crosslinguistic variation in LI and the presence or absence of a high clausal subject requirement in verb-third V2-violations in Kiezdeutsch and more standard varieties of German. Ultimately, I make use of finite differences across C and T in the distribution of D, φ, and discourse-related δ-features (cf. Miyagawa 2017) features via different inheritance options from the phase head. I demonstrate that the presence of non-canonical subjects in LI and the presence of canonical subjects in a seemingly non-canonical subject position in Kiezdeutsch can both be derived via variation in the placement of a δ-feature with a specification for Subject of Predication orthogonal to typical EPP requirements related to D and φ.
... (3) Novel patterns are generally only one option, which is realised considerably less frequently than standard-close variants (Wiese & Rehbein 2015). Furthermore, this option can often also be found in more monolingual contexts, if comparatively less developed there, indicating a quantitative advantage for the multilingual community (Wiese 2013;Schalowski 2015;). ...
Chapter
Full-text available
Using the cover term "Kiezdeutsch", we discuss urban contact dialects in Germany, drawing on different lines of research with different conceptual/theoretical backgrounds. We look at the setting of Kiezdeutsch, a society whose strongly dominant monolingual habitus contrasts with its linguistically highly diverse makeup (Section 1); give pointers to pertinent corpora that are available through open access and to the different research foci for Kiezdeutsch so far (Section 2); and provide an overview of findings at grammatical, pragmatic, lexical, and prosodic levels (Section 3). Finally, we summarise sociolinguistic findings from the Kiezdeutsch corpora, including domains of usage and the attitudes and perceptions evident in the macro context (Section 4).
... In line with Wiese and Rehbein (2015), we understand coherence as an interpretation of (socio-)linguistic data that focuses on links between different variables suggesting a system rather than isolated individual elements. In this paper, we investigate coherence from the point of view of registers, that is, socially recurring intraindividual linguistic variation that is influenced by different communicative (social and functional) settings (cf. ...
Chapter
Full-text available
The paper investigates sociolinguistic coherence and differentiation for the example of Namibian German (‘Namdeutsch’), based on corpus data and a copy-editing task. The Namdeutsch speech community draws on a local Namibian identity as well as an ethnic German identity. At the linguistic level, this leads to a tension between a tendency for Namdeutsch to develop distinctive local features on the one hand, and to remain close to standard German in Germany on the other hand, and this can interact with register distinctions. Data from the DNam corpus of German in Namibia shows that noncanonical local variants are primarily associated with informal registers, but that some are also used in formal language. We hypothesised that particularly variants with weaker overt reflexes, which we assumed to be of lower social salience, can enter formal registers. This was confirmed in a copy-editing task where Namdeutsch speakers were asked to correct a newspaper article. Taken together, our findings point to a broader Namdeutsch dialect that encompasses informal and formal settings in an orderly heterogeneity that is modulated by social meaning linked to local and ethnic identities and a hierarchy of sociolinguistic salience reflecting the overt manifestation of linguistic variables.
... Kiezdeutsch reflects the internal tendencies of German, pointing to an acceleration of internal language change and making it a 'pioneer dialect.' Further, comparing to language use in a more monolingually German neighborhood, Kiezdeutsch shows a tendency towards more nonstandard patterns (Wiese & Rehbein, 2015). In short, the assumption that HSs do not differ from monolingual native speakers in their dominant language has become more questionable, thus increasing the impetus to directly test both languages in HSs. ...
Article
Full-text available
There has been a substantial amount of research on heritage language acquisition and diachronic change. Although recent work has increasingly pointed to parallels between those two areas, it remains unclear how systematic these are. In this paper, we provide a bird's eye view, illustrating how patterns of diachronic change are mirrored in heritage language grammars. In doing so, we focus on one of the best-described grammaticalization processes – namely, the formation of articles from demonstratives and numerals, reviewing studies on heritage varieties which mirror those processes. Based on this review, we make two main points: that change in heritage language can be predicted based on established diachronic scenarios, and that heritage languages often amplify incipient changes in the baseline. After discussing a number of attested changes in a bilingual context, we identify directions for future research in the domain of determiners in heritage languages.
Article
Full-text available
This thesis is a corpus linguistic investigation of the language used by young German speakers online, examining lexical, morphological, orthographic, and syntactic features and changes in language use over time. The study analyses the language in the Nottinghamer Korpus deutscher YouTube‐Sprache ("Nottingham corpus of German YouTube language", or NottDeuYTSch corpus), one of the first large corpora of German‐language comments taken from the videosharing website YouTube, and built specifically for this project. The metadatarich corpus comprises c.33 million tokens from more than 3 million comments posted underneath videos uploaded by mainstream German‐language youthorientated YouTube channels from 2008‐2018. The NottDeuYTSch corpus was created to enable corpus linguistic approaches to studying digital German youth language (Jugendsprache), having identified the need for more specialised web corpora (see Barbaresi 2019). The methodology for compiling the corpus is described in detail in the thesis to facilitate future construction of web corpora. The thesis is situated at the intersection of Computer‐Mediated Communication (CMC) and youth language, which have been important areas of sociolinguistic scholarship since the 1980s, and explores what we can learn from a corpus‐driven, longitudinal approach to (online) youth language. To do so, the thesis uses corpus linguistic methods to analyse three main areas: 1. Lexical trends and the morphology of polysemous lexical items. For this purpose, the analysis focuses on geil, one of the most iconic and productive words in youth language, and presents a longitudinal analysis, demonstrating that usage of geil has decreased, and identifies lexical items that have emerged as potential replacements. Additionally, geil is used to analyse innovative morphological productiveness, demonstrating how different senses of geil are used as a base lexeme or affixoid in compounding and derivation. 2. Syntactic developments. The novel grammaticalization of several subordinating conjunctions into both coordinating conjunctions and discourse markers is examined. The investigation is supported by statistical analyses that demonstrate an increase in the use of non‐standard syntax over the timeframe of the corpus and compares the results with other corpora of written language. 3. Orthography and the metacommunicative features of digital writing. This iii iv analysis identifies orthographic features and strategies in the corpus, e.g. the repetition of certain emoji, and develops a holistic framework to study metacommunicative functions, such as the communication of illocutionary force, information structure, or the expression of identities. The framework unifies previous research that had focused on individual features, integrating a wide range of metacommunicative strategies within a single, robust system of analysis. By using qualitative and computational analytical frameworks within corpus linguistic methods, the thesis identifies emergent linguistic features in digital youth language in German and sheds further light on lexical and morphosyntactic changes and trends in the language of young people over the period 2008‐2018. The study has also further developed and augmented existing analytical frameworks to widen the scope of their application to orthographic features associated with digital writing.
Article
This special issue provides new perspectives on how standard language ideology (SLI) informs and influences attitudes towards ‘non-standard’ language varieties. Standard languages are commonly viewed as uniform, minimally varying (homogeneous) forms of language [Haugen, E. 1966. “Dialect, Language, Nation.” American Anthropologist 68: 922–935.], and existing research in the area of language standardisation has largely focussed on standard languages in (perceived) monolingual nations, mainly restricted to those in Western Europe. The theoretical frameworks used to discuss language standardisation have often assumed that speakers of standard languages are monolingual (even if tacitly acknowledging varieties related to the standard, which can differ in terms of their ‘distance’ from that standard). Yet standard languages always exist alongside other languages and language varieties (including ‘non-standard’ varieties of the standard itself), and many speakers use the standard as an acquired variety alongside at least one other variety or language. This special issue challenges the assumption of monolingual standard languages, by presenting evidence of the impact – linguistic, social and cultural – of standard language ideologies on ‘non-standard’ language varieties that exist alongside standard languages in various political and cultural contexts within and outside Europe.
Article
Full-text available
This article investigates a public debate in Germany that put a special spotlight on the interaction of standard language ideologies with social dichotomies, centering on the question of whether Kiezdeutsch , a new way of speaking in multilingual urban neighbourhoods, is a legitimate German dialect. Based on a corpus of emails and postings to media websites, I analyse central topoi in this debate and an underlying narrative on language and identity. Central elements of this narrative are claims of cultural elevation and cultural unity for an idealised standard language ‘High German’, a view of German dialects as part of a national folk culture, and the construction of an exclusive in-group of ‘German’ speakers who own this language and its dialects. The narrative provides a potent conceptual frame for the Othering of Kiezdeutsch and its speakers, and for the projection of social and sometimes racist deliminations onto the linguistic plane. (Standard language ideology, Kiezdeutsch, dialect, public discourse, Othering, racism by proxy)* Open Access: doi:10.1017/S0047404515000226
Article
Full-text available
This paper argues that the common denominator of topic-comment (TC) constructions in natural languages is not a single functional feature (e.g. aboutness) but rather the fact that they share some salient semantic attributes with prototypical examples of TC. The paper tries to disentangle these prototypical semantic attributes of TC and then shows, mainly referring to examples from German, that different TC constructions are characterized by different combinations of these attributes and therefore cannot be analyzed properly by unitary theories of TC function.