Content uploaded by Heike Wiese
Author content
All content in this area was uploaded by Heike Wiese on Jun 01, 2015
Content may be subject to copyright.
1
Pejoration in contact:
m-reduplication and other examples from urban German
Heike Wiese, University of Potsdam,
& Nilgin Tanis Polat, Ege University Izmir
Abstract
This paper analyses the expressions with pejorative functions in the
dynamic multilinguals setting of urban German, and their possible Turkish
sources. The focus of our investigation is on pejorative functions of m-
reduplication (“Cola Mola”). In addition, we discuss usages of “Scherz” /
“Spaß” ‘joke, fun’ in urban German, as well as their Turkish counterpart
“şaka”, as a marker that cancels the performative force of a preceding
utterance and can thus bring about depejoration, the cancellation of initial
pejoration.
We show that the pejoration involved in our examples is pattern-
based rather than linked to individual evaluative elements, and account for
this as “constructional pejoration”. Interestingly, the patterns we find here
are not exclusively pejorative, but can also support such concepts as
amplification, ‘coolness’, and ludic aspects, putting a spotlight on links
between pejoration and other cognitive domains. We model these links in a
network of systematic conceptual relationships and pragmatic inferences.
Preliminary version, author’s manuscript;
final version to appear in:
Finkbeiner, Rita; Meibauer, Jörg, & Wiese, Heike (eds.),
Pejoration. Amsterdam, Philadelphia: John Benjamins
[Linguistics Today].
2
Keywords: language contact, pejoration and amplification, pejoration and
ludic functions, constructional pejoration, depejoration, echo word
formation, m-reduplication, Kiezdeutsch, Turkish.
1 Introduction: Pejoration and structural borrowing
The phrase “pejoration in contact” in our title is to be understood first of all
as referring to language contact settings: we will investigate pejorative
patterns and their development in settings of Turkish-German language
contact in urban Germany. However, there is a second sense of this phrase
that will be also central to our study: we will take a closer look at the
conceptual networks that will become evident in such developments, that is,
we will investigate pejoration in contact with other cognitive domains.
Language contact settings are a favourable context for new
developments and cross-fertilisations between linguistic systems.
Multilingual speakers are more tolerant of language variation and change
and can thus play a key role in new lexical and grammatical developments,
including the transmission of elements and structures from one language to
another. As Matras (2011) showed in a review of borrowing hierarchies, a
domain that particularly facilitates such transmissions is the discourse level,
which is less subject to analytical control, supporting the borrowing of
“those components of speech that are designed to grab the hearer’s attention
3
and direct the hearer emotionally through the discourse” (Matras 2011:
228).
This makes multilingual settings a promising domain for the
investigation of expressive elements such as pejoratives. As emotional
elements, they get higher attention and, related to that, are better
remembered later (cf. Jay & Janschewitz 2007), while the comparative lack
of analytical control they underlie is associated with a central property that
Potts (2007a,b) called their ‘immediacy’: expressives “do not offer content
so much as inflict it” (Potts 2007b: 167).
In this paper, we are going to present such a study on pejoratives in
language contact, choosing urban Germany as our domain, which
constitutes a particularly interesting case since we find new developments
not only in minority languages, but also ample evidence for borrowing into
the majority language, German.
Present-day urban Germany, like other European countries, is
characterised by neighbourhoods with a high social and linguistic diversity,
with widespread multilingualism and rich opportunities for language
contact. While we find a large range of different heritage languages, a
background language that has gained particular peer-group prestige among
adolescents is Turkish. As a result, in the group of young, adolescent
speakers, Turkish competence is not restricted to those coming from
Turkish-background families: speakers with other heritage languages (such
as, e.g., Arabic, Croatian, or Twi), as well as those with a monolingual
4
German background, will often master at least some short routines and
occasionally be even fluent in colloquial Turkish to some degree (cf. Dirim
& Auer 2004, Wiese 2013). Hence, unlike in the borrowing scenarios
commonly investigated, in this context bilingualism is not restricted to
minority speakers (“unidirectional bilingualism” in Matras’ 2011 terms),
which sets the stage for rich transmission opportunities.
Accordingly, we observe new developments not only in the different
heritage languages involved, but also in German (for overviews cf. e.g.,
Keim 2010, Kern & Selting eds. 2012). However, the evidence found for
direct borrowing into German so far has been restricted to the lexical level,
while structural developments in urban German, or Kiezdeutsch
‘(neighbour-)hood German’, have often been shown to take up general
developmental tendencies within German (cf. Wiese 2013).
1
This is not surprising given that speakers are usually born in
Germany and, even if their family maintains a different heritage language,
grow up with German as well. Thus Kiezdeutsch is spoken by speakers with
German as a first or very early second language, and, accordingly, the
widespread multilingualism in its speech community is reflected in its
1
For an overview of findings on Kiezdeutsch cf. also Wiese (2012), who describes this language use
as a new, multiethnic dialect of German. Converging findings from different urban areas suggest the
development of a new variety in multiethnic and multilingual speech communities across Germany
(which will, at the level of individual speakers, also involve monolingual Germans). As such,
Kiezdeutsch can draw on a large pool of speakers: according to the 2010 census data for Germany,
in cities over 500,000 inhabitants, every second minor grew up in a family with migrant
background, pointing to widespread multilingualism in those areas.
5
openness to new developments more than in grammatical transfers from a
particular heritage language.
However, at least for Turkish as a donor language, structural
borrowings into German should be at least an option, given its
sociolinguistic salience in Kiezdeutsch contexts. In the following, we
present two candidates for such borrowings, that is, patterns in Kiezdeutsch
with strong parallels in Turkish that might be the source for their
development in German: m-reduplication, and constructions with
“Scherz”/“Spaß” ‘joke, fun’ as a pragmatic marker. For our discussion of
pejoration, the main focus of this article will be on pejorative aspects of m-
reduplication, followed by a short exploration into usages of
“Scherz”/“Spaß” for depejoration, that is, a cancellation of initial pejoration.
As we will show, in both cases pejoration is only one of several functions
for this construction, making these developments particularly interesting for
a conceptual discussion of pejoration since they highlight relations and
interactions with other cognitive domains. In our study, we will analyse
such links and the reorganisation they might undergo in transmission.
2 Pejoration through m-reduplication
M-reduplication is an instance of echo word formation, where a word is
repeated in a phonologically modified form, with the modification targeting
left-peripheral positions in the phonological string. A first impression of this
6
pattern in urban German is given by the following examples from
spontaneous speech among adolescents in Kreuzberg, a multiethnic and
multilingual inner-city neighbourhood of Berlin. (1) is from an utterance
overheard in the street; (2) is from the KiezDeutsch-Corpus (‘KiDKo’,
Wiese et al. 2012), uttered by a Turkish-German bilingual speaker
(capitalisation indicates main stress).
(1) er kommt schon wieder mit FAHRrad-MAHRrad.
he comes yet again with bike -mike
‘He comes by bike yet again.’
(2) die sind immer mit der letzten miNUte. ey diese schisser-misser.
they are always with the last minute man these shitter-mitter
‘They always come in the last minute. Man, these scaredycats.’
As these examples already indicate, m-reduplication can carry a pejorative,
dismissive meaning (cf. also Wiese 2013). This interacts, as we will show,
with an amplifying aspect that has been described as a general contribution
of m-reduplication. In what follows, we will first describe m-reduplication
in Turkish, which we regard as a primary source for the development of this
pattern in German (2.1), then have a look at a similar development in
Yiddish and, based on it, American English (2.2), and the additional support
that m-reduplication in Kiezdeutsch might gain from existing patterns of
echo formation in German (2.3). Against this background, we will
investigate such data as illustrated in these examples in more detail and
analyse its pejorative aspects (2.4). Taken together, this will then allow us to
7
model the links that these developments suggest between pejoration and
other domains (2.5).
2.1 A possible source: Turkish m-reduplication
In Turkish, m-reduplication (called “mühmele” in traditional grammars) is
well attested in colloquial usage. (3) and (4) are two examples taken from
Turkish grammars:
(3) Bahçede ağaç mağaç yok. (Lewis 2000: 235)
garden.LOK tree mee is.not
‘There are not trees or shrubs or bushes in the garden.’
(4) Doktor hastanın gözüne mözüne baktı. (Göksel & Kerslake
doctor patient.GEN eye.POSS.DAT meye checked 2005:99)
‘The doctor checked the patient’s eyes etc.’
This pattern involves an echo word as the second element that is derived
from the base either by adding an initial /m/, if the base starts with a vowel
(as in (3)), or replacing the initial consonant by /m/ (as in (4)). Turkish
generally does not have consonant clusters at the beginning of native words,
but they can occur in loan words. If that is the case, only the leftmost
element tends to be replaced by /m/, as in the following quote by Turkish
president Recep Erdoğan, from an election speech held at Bursa on March
20
th
, 2014:
2
(5) Twitter, Mwitter hepsinin kökünü kazıyacağız
Twitter Mwitter all.GEN root.POSS.ACC chip.off.FUT.1PL
‘We will eradicate Twitter and so on.’
2
Cf. also Stamer (2014) who confirms this tendency with data from an internet search for m-doublets
in Turkish that are based on loan words with initial consonant clusters. Thanks to Ad Foolen for
pointing out the “Mwitter” example, and its discussion on langlog (see below) to us.
8
The second element in this pattern is an occasional, ad-hoc form that is not
lexicalised and thus carries no lexical meaning.
3
The base is most frequently
a noun, but can also come from other syntactic categories, such as
adjectives, adverbs, and verbs. The pattern of m-reduplication is highly
productive in Turkish and can be applied broadly, since words that already
start with /m/ and would hence not be suitable as a base, are rare in Turkish
and mostly restricted to proper names and loan words (cf. Yastı 2007).
In general, Turkish grammars give amplification or generalisation
(‘and so on’) as the meaning for this patterns, and this is also the case for
descriptions offered by publications of the Institute for Turkish Language,
Türk Dil Kurumu, a state institution (e.g., Korkmaz 2007, Karaağaç 2013).
This seems to be the most basic usage for such m-reduplications,
presumably with an iconic foundation where a phonologically modified
repetition of the base word indicates “more of the same / similar” at the
level of meaning (Demircan 1987:25 talks of “blurred similarity”).
However, as the Erdoğan quote already suggests, there can also be
an expressive component to it and in particular a pejorative one.
Accordingly, studies on reduplication and echo words count Turkish m-
reduplication as part of expressive language (e.g., Samarin 1970, Schroeder
1989, Müller 2004). Schroeder (1989:40) subsumes devaluation and
3
Eren (1949) suggests, however, that frequently recurring instances of particular m-reduplications
might, in some cases, diachronically lead to lexicalisations.
9
vagueness as two specifications for a general meaning “and so on”, hence
amplification, of Turkish m-reduplication.
4
The pejoration involved in Turkish m-reduplication could be
specified as dismissal, that is, a form of pejoration that is less strong and not
as direct as, for instance, that involved in expletives and curses. In a
contribution to a discussion of the “Mwitter” quote on languagelog, Elif
Batuman describes this kind of pejoration, and gives an example for
learning it in her childhood:
“although Turkish reduplication can be used for lots of things
(vagueness/ euphemism, shortening a list, etc.), it often *does* have a
dismissive/ distancing connotation – something like "or whatever," "or
whatnot." […] A reduplication anecdote from my childhood: once, a
cousin and I were dragged on a long boring shopping expedition with our
mothers. On the way back, we stopped at a shop where my mom bought
me a doll [bebek]. My cousin was a bit older, and I don't think he got a
toy. When we got back, an uncle asked where we had been. My cousin
said, "We were coming home, but then we stopped to get bebek mebek
for Elif." His parents reprimanded him, + that's how I understood that he
had said something dismissive!” (Elif Batuman, languagelog)
5
4
In addition, he lists two more specifications, namely (1) adding possible variants and (2) adding a
similar entity. If we use “vagueness” in a broader sense, those might be subsumed under it. Cf. also
Akaslan (2006) for the pejorative meaning of this phenomenon.
5
http://languagelog.ldc.upenn.edu/nll/?p=11255, last accessed Sept 12
th
, 2014.
10
This already points to an interaction of amplification and pejoration; we will
come back to this in 2.5 when analysing in more detail something we will
call the “whatever” effect that we believe is involved here.
The pattern of dismissal we see in m-reduplication here is also part
of the Turkish usage brought to Germany by the initial ‘guest worker’
generation of Turkish immigrants in the 1960s/70s. The example in (6) is a
quote from the movie “Almanya” from 2011, by Yasemin und Nesrin
Şamdereli, which portrays the experiences of a family with Turkish roots in
Germany. In this scene, the grandfather, who came to Germany first,
comments on his wife’s suggestion to apply for German citizenship after
living in the country for 45 years. He expresses his opposition to this
suggestion by saying:
6
(6) Alman pasaport malman pasaport istemiyorum.
German passport Merman passport want.NEG.PRES.1SG
‘I don’t want a German passport.’
This makes Turkish m-reduplication a possible basis for the pattern we find
in contemporary German in multilingual settings where Turkish is a salient
heritage language. And, as Southern (2005) points out, it generally seems to
travel well: in his investigation of this pattern, he talks of “contagious
couplings”, emphasising the ease with which it gets transferred, and
identifies Turk languages as a source from which such reduplication spread
into typologically diverse languages, including Iranian, South Asian, Slavic,
and Germanic languages. In a similar vein, if restricted to a smaller
6
In the movie, this scene is in Turkish, with German subtitles.
11
geographical area, Inkelas & Zoll (2005: 42) speak of a “virtually pan-Asian
‘X and the like’ construction”. Stolz (2008) discusses the areal diffusion of
Turkish m-reduplication within the area of the former Ottoman Empire and
talks about its “easy borrowability” (Stolz 2008: 122), which he links to the
cognitive salience of, phonologically, the word-initial position and,
pragmatically, the expression of speakers’ attitudes involved in m-
reduplication.
Another contribution to the languagelog discussion, from a user
tuncay, further supports a transmission of Turkish m-reduplication by
bilingual speakers, pointing to a tendency to spontaneously integrate this
expressive pattern into a second language:
“As someone who almost never "errrr"s while speaking, trying to spit out
random nouns instead of m-reduplicating one is one of the few things
that make me "errr" in my non-native English.” (user tuncay,
languagelog)
7
Taken together, this further supports a view of Turkish m-reduplication as a
likely source for this pattern in Kiezdeutsch. Before we analyse the relevant
Kiezdeutsch evidence, let us have a look at a similar and better known
development in American English from a Yiddish basis, which might
ultimately also go back to Turkish and shows an even more pronounced
‘dismissal’ pejoration.
7
http://languagelog.ldc.upenn.edu/nll/?p=11255, last accessed Sept 12
th
, 2014.
12
2.2 Similar developments in American English from a Yiddish source
Productive echo-word formations as exemplified in the Kiezdeutsch m-
reduplication are not a novel phenomenon in Germanic languages. A well-
known earlier example is Yiddish ʃm-reduplication, which has been further
transmitted into American English and also into colloquial Russian.
8
(7) and
(8) give examples from Yiddish and English, respectively:
(7) a. Libe-ʃmibe
love-shmove
“said when one wishes to jeer at love (Liebe)” (Spitzer 1952: 228)
b. gelt-ʃmelt
money-shmoney
“money [Geld] is not worth anything” (Spitzer 1952: 228)
(8) a. Gadgets, Schmadgets … As Long As It Takes Pictures!
(Feinsilver 1961: 302, citing a Konica advertisment)
b. messy, shmessy
“messy: who cares!” (Southern 2005:302)
ʃm-reduplication is phonologically somewhat more complex than its Turkish
counterpart,
9
and syntactically more restricted: unlike in Turkish, the echo
pairs are syntactically not integrated, but occur by themselves, outside
sentential contexts, and there also seems to be a stronger tendency to use
nouns as a base, in addition to some usages with adjectives.
Like Turkish m-reduplication, Yiddish and English ʃm-/shm-
reduplication is restricted to colloquial language. Unlike in Turkish, this
construction always implies pejoration, which generally carries an ironic
8
Cf. e.g., Spitzer (1952), Weinreich (1953), Southern (2005), Stolz (2008).
9
Cf. Nevins & Vaux (2003) for a detailed analysis of the phonological processes involved.
13
tone, described as ironical rebuke (e.g., Spitzer 1952) or ironic deprecation
(e.g., Southern 2005).
Diachronically, according to Southern (2005), Turkish m-
reduplication (together with East Slavic sh-) might have been one source for
Yiddish ʃm-reduplication. Another source might have lain in Yiddish itself,
namely, as Spitzer (1952) argues, in Eastern Yiddish words beginning with
ʃm- that have the connotation ‘rebuke’, such as ʃmu(e)s ‘superfluous talk’ or
ʃmir in “nit a ʃmir” ‘not a jot’ (Spitzer 1952: 232). Figure 1 depicts this
developmental path.
Figure 1: Development of Yiddish ʃm- and English shm-reduplication
This suggests a two-fold source for the development of this reduplication:
an external motivation from language contact, which gets further support
from language-internal patterns. If we have a look at the basis for m-
reduplication in Kiezdeutsch, which, like Yiddish, emerged in a language-
contact situation, we can also identify such language-internal support here:
in addition to the motivation we find in Turkish, there are some existing,
Yiddish
pejorative lexemes
b
e
g
innin
g
with
ʃ
m-
Turkish
amplifying/pejorative
m-reduplication
Yiddish
pejorative-ironic
ʃ
m-redu
p
lication
English
pejorative-ironic
s
hm-redu
p
lication
14
similar patterns in German, which also relate to the pejoration involved
here.
2.3 Support from existing patterns in German
German has a number of lexicalised elements that are based on rhyme
reduplication and which, even if this is not productive, provide an existing,
language-internal pattern that might further support the incorporation of m-
reduplication.
Such lexicalised rhymes often involve /m/ substitution for the onset
of the second element, parallel to that in m-reduplication, e.g., in
“Kuddelmuddel” ‘jumble bumble’, “Schickimicki” ‘fancy-pancy’, or
“Heckmeck” ‘hassle’, or Techtelmechtel ‘hanky-panky’. As several of the
English translations indicate, similar constructions can also be found in
English, providing a further motivation for the integration of shm-
reduplication there (cf. also Southern 2005). In German, the existence of a
group of /m/-rhymes makes the parallel to Turkish m-reduplication even
closer.
In addition, one also finds rhymes with other consonants, e.g.,
“Remmidemmi” ‘racket’, “Larifari” ‘airy-fairy’, or “ratzfatz” ‘in a jiffy’, as
well as a related, but somewhat different pattern of ‘ablaut reduplication’,
which involves vowel (rather than consonant) replacement, frequently /ı/ →
/a/, such as in “Singsang” ‘singsong’, “Wirrwarr” ‘hurly-burly’,
15
“Mischmasch” ‘mishmash’, “Krimskrams” ‘hotchpotch’, “Wischiwaschi”
‘wishy-washy’, or “Tingeltangel” ‘honky-tonk’.
What these reduplications have in common is that they are not
productive, but lexicalised as one unit. Synchronically, the second element
is usually a nonsense word – even though it might have a lexical basis
diachronically (cf. e.g., Müller 2004). More often than not, the same is true
for the first element, in particular in lexicalised m-rhymes, so that both
constituents by themselves are semantically void.
10
While the option to have
two semantically empty components is specific to these lexicalised echo
formations, the fact that the construction is semantically non-compositional
is something they share with productive m-reduplication.
Similar to Turkish m-reduplications, German rhyme lexicalisations
are part of colloquial speech. And, interestingly for our discussion here, they
are mostly pejorative. Bzdęga (1965:38), in his comprehensive overview of
word formation by reduplication in German, calls “pejorative” the second-
most “accompanying feeling”, right after “jocular”, and, as the examples
above illustrate, the tendency towards pejoration is particularly pronounced
in lexicalisations involving /m/ substitution.
Further language-internal support for a pejorative aspect in m-
reduplication might come from a repetitive mocking/teasing pattern based
10
With exceptions such as “Schickimicki”, where the left constituent points to an adjective “schick”
‘chic’
16
on proper names that is used among children.
11
In this pattern, the speaker
mocks somebody by chanting his/her name together with a rhyme twin. The
twin is derived from replacing the name’s onset with another consonant,
again often bilabials such as /p/ or /m/, e.g. “Karsten – Parsten!” or “Petra –
Metra!”
Pejoration is here iconically supported by the deformation that the
name undergoes, which can serve as a symbolic attack and derision of the
name’s bearer. In addition, it is supported at the level of intonation and
rhythm: these reduplications are usually sung in a repetitive and ritualised
style characterised by prominent trochees that is associated with mocking
and derision.
Taken together, this points to at least two areas where German might
provide language-internal support for the integration of m-reduplications
with pejorative usages: lexicalised rhymes and name rhyme mocking. While
neither of these provide an exact model for the productive m-reduplications
we found in Kiezdeutsch, they show enough parallels to highlight the
principal possibility of such a pattern in German, and we should thus keep
them in mind when we now analyse those reduplications in more detail.
11
We are grateful for the audience at the DGfS 2014 workshop on Pejoration for pointing this area
out to us.
17
2.4 m-reduplication in urban German
The two examples for Kiezdeutsch m-reduplication we presented in (1) and
(2) above are here repeated as (9) and (10), together with an example from
Stamer (2014:37), who investigated m-reduplication in interviews with
young people from Bremen and Frankfurt (11), and two examples
volunteered by speakers in a focus group discussion in Berlin ((12) and
(13)).
(9) er kommt schon wieder mit FAHRrad-MAHRrad.
he comes yet again with bike -mike
‘He comes by bike yet again.’
(10) die sind immer mit der letzten miNUte. ey diese schisser-misser.
they are always with the last minute man these shitter-mitter
‘They always come in the last minute. Man, these scaredy-cats.’
(11) Wir gucken immer spontan,
we look always spontaneously
nicht so wie ihr so planen-manen und so …
not PTCL like you PTCL
12
plan-man and such
‘We always look spontaneously, not like you guys, with plannning
and so on.’
(12) Die essen Chips Mips.
they eat crisps misps
‘They eat crisps.’
(13) Lassma U-Bahn Mu-Bahn!
let.PTCL subway mubway
‘Let’s take the subway!’
12
The particle “so” lit. ‘so/such/like’ is probably used as a focus marker here, similar to “like” in
colloquial English, but with two instances of “so” forming a bracket around the focussed
coconstituent, which is an option in German (cf. Wiese 2011 on “so” as a German focus marker).
18
In (9) and (10), pejoration is further supported by the context: in (9), the
adverbial “schon wieder” ‘yet again’ indicates a negative evaluation, while
in (10), the base for the reduplication, “Schisser” ‘scaredy-cat’ (lit. ‘shitter’),
is a derogative term for people who are afraid (in this case, this was directed
at the Turkish team in a football Euro Cup). Contexts like this might further
support the development of pejorative elements, and frequent occurrences of
negative bases might also play a similar role in Turkish (cf. also Dammel &
Quindt, this volume, for similar effects in the diachronic development of
pejorative affixes in German).
However, as the other examples illustrate, m echoes also occur in
neutral contexts. In order to find out what the status of m-reduplication in
German is in such contexts, we conducted a focus group discussion (as
mentioned above). Participants in this discussion were 19 speakers, 15 to 17
years old, from the neighbouring Kreuzberg and Neukölln districts of
Berlin, which are similarly multiethnic and multilingual. The majority of
participants (14 out of 19) came from families with a Turkish background,
in addition to Arabic, Kurdish, and monolingual German backgrounds. All
participants were familiar with m-reduplication in German and said that they
regularly used it themselves.
We presented participants with the quote in (12) above and initiated
a discussion of the presumed speaker’s evaluation for the proposition and
the association the reduplication awoke (“Does this mean that the speaker
likes that they bring crisps, or not, or does it leave it open? What does this
19
pattern of saying ‘Chips Mips’ sound like for you?”) Results indicate two
different elements: (1) reduplication can signal pejoration, specifically in the
form of dismissal or devaluation, and (2) it can mark the speaker as casual,
with participants characterising him by such attributes as “cool”, “chilled”,
or “easy”. The quote in (14) gives an example for the kind of assessments
participants offered:
(14) “Could be he does not like it that they eat crisps, or he’s just cool.”
In tune with the evaluation as casual, participants categorised the utterance
as colloquial and associated it specifically with peer group situations among
young people, as illustrated by the quotes in (15):
(15) “sounds youthful”, “vernacular that young people use”,
“more for young people”, “when we are among ourselves”
This suggests a further development of m-reduplication here, compared to
the Turkish pattern that is presumably its source: in the linguistically diverse
context of urban Germany, it is particularly the group of young, adolescent
speakers that is at the forefront of linguistic innovation,
13
and in this group
the adoption of a colloquial pattern from Turkish in German might take on
an additional function as an in-group marker of ‘coolness’. This ‘coolness’
feature can then co-occur with pejoration, while the amplification that is at
the core of the Turkish m-reduplication seems to lose its significance.
13
In general, adolescents are often motors of linguistic changes since they are in the process of
forging new identities for themselves, leading to new styles and innovations, which is also reflected
in their linguistic practices. Cf. also Kerswill (1996), Eckert (2000) on adolescents as a core group
in linguistic innovation.
20
Before we look into this interaction at the meaning level in more
detail, let us first have a look at the phonological and syntactic behaviour of
m-reduplication in German, in order to account for its characteristic features
compared to its Turkish kin.
Phonologically, the pattern follows essentially the Turkish example,
with two adjustments to the different German phonology. First, /m/ is
always a replacement, never an addition, since German words generally do
not start with a vowel, but will have a glottal stop in front if otherwise the
left-most element in the phonological string would be a vowel. Hence,
/mu:bɑ:n/ “Mu-Bahn” in (13) does not require a special rule that introduces
/m/ as an additional fronted element, but can be regularly derived from
replacing the consonant, ʔ, in the onset of /ʔu:bɑ:n/ “U-Bahn”. Second, it is
always the whole onset that is replaced, not just the left-most consonant, as
illustrated in (11). Hence, in German m-reduplication, “Twitter” would not
be rendered as “Mwitter” (cf. the Erdoğan quote in (5) above), but rather as
“Mitter”. This is related to the fact that German routinely has consonant
clusters in the onset, unlike Turkish where this is mostly restricted to non-
native words and treated as an exception. Taken together, this means that we
can account for the formation of the echo twin in German by a simpler
phonological rule than in Turkish, namely by a unified replacement of the
onset by /m/.
The occurrence of m-reduplications in sentential embeddings in the
examples above suggests that, like those in Turkish, they are syntactically
21
more integrated than shm-reduplications in Yiddish and English. This is
further supported by results from the focus group discussion, which at the
same time also indicate that the integration in German might not go as far as
in Turkish.
According to evaluations offered in that discussion, m-reduplications
are possible in different sentence types and speech acts, such as
declaratives/statements, imperatives/commands, and
interrogatives/questions. Furthermore, while they frequently seem to occur
at the end of a sentence (cf. (9), (10), (12) and (13) above), this does not
seem to be a necessary requirement, so that m-reduplications might also be
followed by additional material in the right sentence periphery.
There seems to be a strong tendency to use nouns as a base in m-
reduplications, rather than elements of other syntactic categories (note,
though, that this is a tendency, not a categorical restriction). However,
speakers did not allow nominal modifiers, e.g., adjectives, nor articles with
nominal m-reduplications. Note, though, that in (10), which is from a corpus
of spontaneous conversations, the (nominal) reduplication is preceded by a
determiner, namely a demonstrative pronoun “diese” ‘these’. This might
either indicate that we are not facing a strict rule, but rather a tendency here,
or it might point to a different grammatical status of the echo pair in this
construction, namely as a syntactically simple element, that is, a nominal
compound “Schittermitter” (lit. ‘shitter.mitter’).
22
This latter option would mean that m-reduplication can work at the
morphological level as well, supporting word formation. Such a pattern
might be reinforced by the lexicalised m-rhymes that already exist in
German (and which also typically occur in the nominal domain, see the
examples in 2.3 above), thus providing a productive counterpart to these
lexicalisations.
In general, we could regard the locus of m-reduplication as
somewhat at the border between the word and sentence levels since it
operates on single words and does not support full syntactic integration. In a
broader approach to the lexicon that understands it as a generative domain
standing on its own (cf. e.g., Jackendoff 2002), m-reduplication can then be
described as part of the lexicon, captured by a complex entry that combines
phonological, syntactic, and semantic representations. Crucially, this entry
as a whole supports the ‘dismissal’ pejoration we found here: the pejoration
is not contributed by a specific word or affix with a deprecatory semantics,
but is associated with the pattern of m-reduplication. Neither of the two
elements involved in this pattern by itself can provide the pejoration: The
first element, the base word, retains its regular semantics, and the second
element, its echo twin, is a nonword that is produced ad hoc and thus does
not contribute any meaning, pejorative or otherwise. It is only the specific
combination that can be pejorative. What we find here can hence be
described as “constructional pejoration” (using the term “construction” in a
broad, pretheoretic sense for a linguistic pattern involving phonology,
23
syntax, and semantics).
14
Figure 2 brings our results for this construction
together in a way that accounts for the characteristics of pejorative usages of
German m-reduplication we found, and captures our understanding that the
pejoration is associated with a pattern, rather than an individual lexical
element.
Figure 2: Constructional pejoration in German m-reduplication
At the PHON level, “O” is a phonological representation that is copied, with
a modification that replaces the onset by /m/. At the SYN level, a primitive
syntactic representation, represented by a syntactic category, is followed by
itself (we chose “N” for this category here to highlight the dominance of
nouns, but this is not meant to be read as exclusionary). The two elements
are combined by simple juxtaposition, lacking an appropriate syntactic link,
which places them somewhat outside the realm of regular syntax and thus
accounts for the weak phrasal integration we found. At the SEM level, only
the first element has a semantic counterpart, here rendered as e, indicating
an entity. The second element does not show up in the semantic
14
Cf. also Stolz (2008:108), who assumes a “constructional meaning” for reduplications. D’Avis &
Meibauer (2013) discuss constructional pejoration for the case of pseudo-vocatives such as German
“Du Idiot” ‘You idiot!’, where they show that the insulting meaning is also due to the construction.
PHON: O O
[ONSET := /m/]
SYN: N N
SEM: e
PRAG: p . p ϵ {
< speaker I
n
e >, … }
24
representation: the echo twin, as an ad hoc nonsense word, remains
semantically void. In the pragmatic domain, this grammatical representation
as a whole is associated with a pejorative triple.
15
With this triple, we follow
Potts’ (2007a:177) definition of expressive indices: an expressive index <a I
b> is a triple where a and b are individuals
16
and I is an expressive level
indicating an interval that conveys negative, neutral, or positive
relationships between a and b. In our case, the relationship is between the
speaker and the entity e (from SEM), and the interval is marked as negative
(in our representation, this is indicated by a superscript “n”). Since
pejoration is not the only function of m-reduplication, we defined this triple
as an element of a larger set. In Figure 2, we have left the other elements of
this set underspecified (indicated by “…”). However, this is not meant to
indicate randomness, but rather as an abbreviation for the particular
additional functions m-reduplication can have (and which we will treat in
the next section).
As Figure 2 shows, we do not treat the pejorative aspect here as a
part of semantics proper, in contrast, for instance, to Hom (2008), who
analyses pejorative aspects of slurs as a combinatorial part of truth-
conditional semantics. At the same time, we do account for pejoration as
15
The pragmatic level ‘PRAG’ we added to our representation here further exploits the option in
Tripartite Parallel Architecture to identify further conceptual tiers (along those lines, Jackendoff
2002:Ch.12.5 for instance proposes a separate tier for Information Structure, whose elements can
then be coindexed with representations from the other domains).
16
In Potts (2007b), a and b are defined as situations in a broad sense, taking into account comments
by Zimmermann (2007). In our case, such an extension can also capture pejoration for bases of
other syntactic categories than nouns, for instance verbs.
25
part of a lexical entry, following the broader understanding of lexicon
described above. By doing so, we capture the fact that pejoration is – as one
of several pragmatic options – conventionally associated with m-
reduplication. This is in accordance with a proposal put forward by Camp
(2013), who argues that slurs “conventionally signal a speaker’s allegiance
to a derogating perspective on the group identified by the slur’s extension-
determining core” (Camp 2013:331). Along the same lines, we believe that
m-reduplication can conventionally signal the speaker’s allegiance to a
derogating perspective on the kind identified by the semantics of the first
element, the base word for the reduplication.
Similarly as noted for slurs, the pejorative component projects out of
such complex constructions as negation, questions, modals, and others (cf.
Potts 2007, Camp 2013). Following Whiting’s (2013) analysis for slurs
contra Hom (2008), we believe that this is also true for ‘didactic uses’ in the
sense of Anderson & Lepore (2013). These are cases where the derogative
aspect of a slur is rejected, while the group membership identified by its
descriptive content is confirmed. As a pertinent example imagine that one of
us, H. Wiese, who is German, defends herself against being called a
“Kraut”. The question is then whether (16) would reflect a statement where
she uses a slur without actually embracing its pejorative content:
(16) I am a German, not a Kraut.
26
However, as Whiting (2013), showed, something like (16) is actually not
felicitous as it stands: the slur in this statement should be rendered in scare
quotes indicating that the speaker does not use the term straightforwardly,
but only in quotation, thus distancing herself from how group membership
to Germans is expressed by “Kraut”:
(16)’ I am a German, not a “Kraut”.
It is only this quotational usage as in (16), but not the straightforward usage
in (16)’ that allows the pejorative content to be cancelled under negation.
Similarly for m-reduplication: here, too, pejoration can project out of
embedded usages if the construction is used straightforwardly, but can be
cancelled if it is quotational in the way described for (16)’. (17) vs. (17)’,
taking up our first example from the beginning, illustrate this for m-
reduplication:
(17) Er sagt, er kommt mit Fahrrad-Mahrrad.
‘He says he is going to come with bike-mike.
(17)’ Er sagt, er kommt mit “Fahrrad-Mahrrad”.
‘He says he is going to come with “bike-mike”.
In the first case (17), pejoration can project out such that the speaker herself
expresses the pejoration involved. In the second case (17)’, this gets
embedded such that the pejoration can be associated with the third party
(“he”) whose statement is reported here.
Note that the same applies to the other elements of the pragmatic set
we sketched for m-reduplication in Figure 2, e.g. speakers’ coolness: while
in (17), the speaker herself can express being ‘cool’ or ‘chilled’ by using m-
27
reduplication, in (17)’, this gest associated with the person “he” who said he
would come by bike.
In the following section, we are going to have a close look at the way such
other elements as coolness, amplification and fun, which we only
represented by a place-holder “…” in Figure 2, come together with
pejoration in this set: what brings the elements of this set together? How
does pejoration come about here, and how does it interact with the other
conceptual domains observed for m-reduplication in German and in
Turkish?
2.5 Pejoration, amplification, ‘coolness’, and fun: From echo word
formation to pejoration
Probably the domain of echo words generally combines pejoration and
amplification. McCarthy & Prince (1986), for instance, who assume that
echo words are a nearly universal phenomenon, give the following general
description of their meaning: “Echo words may have either the pejorative
meaning typical of English or a loose kind of plurality (‘X and such’)
characteristic of the languages of the Indian subcontinent” (McCarthy &
Prince 1986:67). Against the background of our findings from m- and ʃm-
reduplication, we can connect these two meaning domains with each other
and further link them to two other domains that can also play a role in echo
words, namely ‘coolness’ and fun. These four domains are embedded in a
cognitive network for echo word formation that can be described as follows.
28
At the form side of echo word formation, we can identify two
processes, copy and modify. Copying yields the echo word as a reproduction
of the base, and modifying leads to partial rather than full reduplication,
e.g., by /m/ addition or substitution. Both processes lend themselves to
iconic interpretations: copying provides “more of the same” at the form
level, which can be iconically associated with amplification at the meaning
level, and modification leads to a change in form that can be interpreted as
deformation.
Amplification makes the message less specific and thus brings in
vagueness, whereas deformation leads to a deficient, meaningless element.
In both cases, the hearer faces an apparent violation of the Maxim of
Manner, specifically “be brief” and “avoid obscurity”, which should
proscribe meaningless repetition and nonsense words. This supports a
resolution via pragmatic inferencing, comparable to what Meibauer (2013)
suggests for the derivation of pejoration on the morphological level, in his
case for pejorative meanings that emerge through metaphorical
interpretations targeting negative evaluations of source domains. In our
case, pragmatic inferencing leads to a reassessment of vagueness and
deformation that opens a route to pejoration, organised by two main
processes that we capture as the “Whatever” effect and the “Nonsense”
effect.
The “Whatever” effect takes its name from a common use of this
term, in particular by young people, signalling that one does not care about
29
specifics.
17
If the source of this negligence is in the referent, this will then
signal dismissal and devaluation and thus lead to pejoration: the referent is
not worth caring about. This accords, for example, with what Koo & Rhee
(this volume) find for Korean, where a lack of specification signals that the
referent is “not deserving fine-tuned attention”, and similarly, Spitzer (1952:
230) described that m-reduplication can be used by a speaker “in order to
express variants of a phenomenon which he does not care, or is not able, to
name for the moment”. For the case of German, Finkbeiner (this volume)
suggests that the use of “bla bla bla” as a place-holder can indicate that the
exact content of an utterance is irrelevant and thus trigger a pejorative
meaning.The inattentiveness to detail, however, can also have a source in
the speaker herself, and this is where the expression of ‘coolness’ that we
found for m-reduplication in Kiezdeutsch comes in: by adding vagueness,
the speaker can also indicate that she does not need to care, thus presenting
herself as ‘cool’ or ‘chilled’.
The “Nonsense” effect links up pejoration with ludic aspects linked
to the other route we found, starting from deformation. In this case, the
occurrence of a meaningless element, the echo word, is interpreted as a
deliberate production of a pseudo-label by the speaker, targeted either at the
referent or directly at language itself. In the first case, this leads to
pejoration: the referent is ridiculed similarly as in the name rhyming chants
17
Readers familiar with “Bollywood” movies might remember the brilliant rendition of such
“Whatever!”s as a signature expression by the character Pooja, a cool young Londoner of Indian
background in “Kabhi Khushi Kabhie Gham” (director Karan Joha, 2001).
30
we found in German. While this further supports aspects of dismissal and
devaluation, the second case, targeting the language level, sets up a
connection with ludic aspects: deliberately deforming a lexical item,
together with the rhyming pattern, points to playful word game – the
speaker is just having fun, rather than making fun of somebody or
something.
18
This indicates a general ludic potential of echo word formation
(Spitzer 1952: 226 talks of “jocular repetition” for ʃm-reduplication), and
because of this, m-reduplication can also be used to defuse a conflict
situation, since the pejoration tends to be flippant rather than overtly
aggressive.
Figure 3 depicts the network we can thus identify for pejoration.
18
For informants Stamer (2014) interviewed, the “fun” aspect of m-reduplication phenomenon was
the only description they offered. Note, though, that this might be related to what Potts (2007a) calls
the “descriptive ineffability” of expressives: speakers are often unable to describe the meaning of
expressives, and rather refer to their usage conditions.
31
Figure 3: Links to pejoration in m-reduplication
What we find here, then, is a route from echo word formation to such
pejorative aspects as dismissal and devaluation, associated with ‘coolness’
on the one hand and with ludic aspects on the other hand. As we saw in the
different cases we looked at, echo words in different languages can activate
these domains in different patterns. While Kiezdeutsch m-reduplication
seems to employ the whole range on the right side, Yiddish and Turkish
reduplications might only link ludic aspects, but not ‘coolness’ with
pejoration. And while in Turkish amplification maintains a central role, the
ludic aspect seems to play only a minor one, suggesting that this is a domain
further activated in transmission, where the creation of nonsense words in
this echo word formation might be highlighted.
Amplification
‘and so on’
less specific:
vagueness
W
H
A
T
E
V
E
R
speaker
keeps
information
vague because
s/he does not
care about
specifics
Inference
(Maxim of Manner)
speaker does not
need to care:
‘chilled’, ‘cool’
referent is not
worth caring:
dismissal,
devaluation
Deformation
‘deficient’
meaningless
element
N
O
N
S
E
N
S
E
speaker
deliberately
produces
pseudo-label
for referent
referent is ridiculed:
dismissal,
devaluation
word play:
ludic
Echo
words
32
The intersection of pejoration with ludic aspects is even more
pronounced in another development suggesting a Turkish-based pattern in
contemporary German that we briefly want to discuss now, namely
depejoration through a specific use of the nouns “Scherz” and “Spaß” .
3 Depejoration through “Scherz/Spaß” ‘just kidding’
While “Spaß” is close to English “fun”, “Scherz” does not have a direct
counterpart in an English noun. The closest translation would be ‘joke’.
However, unlike ‘joke’, “Scherz” does not cover conventionalised jocular
narratives (as in ‘telling a joke’), and it crucially involves a playful,
nonserious and non-threatening aspect, rather than wittiness and unexpected
combinations or turns. Those functions of English “joke” are covered by the
noun “Witz” in German.
In Kiezdeutsch contexts, “Scherz” and “Spaß” are not only used as
regular count nouns, but can also be attached to the end of utterances in their
bare form, as illustrated in (18) through (20) (examples from KiDKo, see
above; capitalisation indicates main stress, (-) indicates short pauses, right-
most two letters in labels identify male vs. female sex (M vs. W), and
family/heritage language (e.g., T: Turkish, K: Kurdish, D: German))
(18) du kleiner hund SPAß mann [MuH3WT]
you little dog Spaß man
‘You little dog – just kidding, man.’
(19) ey du hast da was verLORN (-) SPAß mann. [MuP1MK]
ey you have there something lost Spaß man
33
‘Ey, you lost something there – just kidding, man.’
(20) ich mach BOxerschnitt (-) SCHERZ PASST mir nich [MuH12MD]
I make boxer.cut Scherz suits me no’
‘I’l get a boxer haircut – just kidding, doesn’t suit me.’
Similarly to German m-reduplication, this construction is common in peer-
group situations in such multilingual neighbourhoods as Berlin-Kreuzberg,
and again, there might be a Turkish source for it.
3.1 “Scherz/Spaß” and “şaka” in German and Turkish
The function of “Scherz” and “Spaß” in such examples is similar to that of
“just kidding” in English. While “Scherz” and “Spaß” can be embedded in
larger statements fulfilling this function, e.g. “Das war nur ein Scherz/Spaß”
‘That was only a joke’, the usage illustrated here has, to our knowledge, not
been described for German yet. It has a close counterpart in Turkish “şaka”
‘joke, jest’, though, which might be a source for this development (and,
similarly as “Scherz”, does not cover the “Witz” aspects of English “joke”
mentioned above). (21) gives an example for the usage of “şaka” in a
Turkish cartoon (Cover of the weekly magazine Penguen, 2004/43;109):
(21)
34
(‘Wake up, wake up, we are in the EU!...’)
Şaka lan şaka
joke man joke
‘Just kidding, guys’
(‘We are not….Sleep on!’)
Interestingly, both şaka and Scherz/Spaß in this usage are also found in
Turkish-German code switching in urban Germany, thus providing a
suitable context for transfers. (22) is an utterance in a conversation between
two young men, Turkish-German bilinguals, in Berlin-Kreuzberg, where
şaka follows a longer pejorative passage in Turkish,
19
(23) is from the
KiDKo corpus,
20
the speaker is a young, Turkish-German bilingual woman,
who uses Scherz after a Turkish passage defaming a mutual acquaintance (in
the examples, Turkish is marked by continuous underlining, German by
dotted lines):
(22) Oğlum siktir git, geh ma weg,
boy.POSS1SG sod.off go
IMP
go
IMP
MP away
verpiss dich du Hurensohn. Şaka lan.
piss.off yourself
ACC
you whore.son şaka man/guy
‘Boy, sod off, go away, you son of a bitch. Just kidding, man.’
(23) ta kıçını boyamış SCHERZ mann
also butt.POSS paint.PERF Scherz man
‘and apparently painted her butt – just kidding, man.’
In Turkish, şaka is intonationally isolated, and usually separated from the
preceding sentence by a short break. In contrast to this, German Scherz/Spaß
19
Data from self-recordings in Berlin-Kreuzberg, Denkwerk project “Let’s do language” (C-
Schroeder & H. Wiese, 2011-2014), funded by Robert-Bosch-Foundation.
20
KiDKo, SPK2 in transcript MuP6MD.
35
can be intonationally – if not syntactically – integrated into the sentence and
attached right to the end of it (cf., e.g., (18) and (23) above). This might
point to a stronger grammaticalisation of this pattern in German.
The general effect of both Scherz/Spaß and şaka is to take back the
force of the preceding utterance: the speaker first makes the hearer believe
something, which s/he then revokes. This can serve two functions: (1)
Joking – The speaker says something, which s/he then resolves, indicating
that s/he has led the hearer on, as in the cartoon above. (2) Defusing
Criticism – The speaker says something unfriendly to the hearer, which s/he
then annuls.
In both cases, Scherz/Spaß targets the speech act: it allows the
speaker to take back her/his commitment to the preceding speech act.
Accordingly, Scherz/Spaß is not restricted to verbal messages, but can also
be used for non-verbal acts, for instance after rude gestures. In jocular
contexts, Scherz/Spaß supports a funny effect, while in contexts of criticism,
it enables the speaker to perform a pejorative or aggressive speech act that
s/he can then mark as ‘not serious’. The playful meaning aspects that
distinguish Scherz from the related noun Witz ‘joke’, as mentioned above,
make it particularly suitable for such usage, and can account for its
occurrence in the urban youth context of Kiezdeutsch, which might
36
otherwise be surprising given its slightly old-fashioned, more marked status
in present-day German.
21
The intonational integration of Scherz/Spaß in Kiezdeutsch makes it
possible to take back criticism and thus depejorate quickly, supporting a
commitment-free speech act. The following quote illustrates this: this is a
meta-linguistic explanation offered by a young speaker in Kreuzberg (12
years old, French-German bilingual) who had used Scherz after criticising
her friend’s new pants as ugly.
(24) “Because she was boasting too much, I wanted to say something bad
against her, but I did not want to hurt her or make the mood sad, not
start a fight.”
In this function, Scherz and Spaß occur in assertions as well as in commands
and questions (cf. (25) through (27)), and they can be recurring elements in
playful banter (as in (28)):
22
(25) halt die klappe SPAß mann [KiDKo, MuH3WT]
hold the trap Spaß man
‘Shut up – just kidding, man.’
(26) Lass mich in Ruhe Scherz!
let me in calmness Scherz
‘Leave me alone – just kidding.’
(27) Was’n das für’n Scheiß? Scherz.
what.MP
CLITIC
that for.a
CLITIC
shit Scherz
21
In the DWDS corpus of 20th century German (Digitales Wörterbuch der Deutschen Sprache des
20. Jahrhunderts, Berlin-Brandenburg Academy of Science, Kernkorpus 20.Jh., about 100 million
tokes), the use of “Scherz” in the second half of the 20
th
century is less than half of that in the first
half (1900-1949: 929 hits; 1950 -1999: 432 hits), whereas the usage of “Spaß” increases (1900-
1949: 1182 hits; 1950 -1999: 1425 hits).
22
Data in the first three examples from fieldnotes in Berlin-Kreuzberg (H.W.).
37
‘What rubbish is that, then? Just kidding.’
(28) Gib nicht an Scherz.
boast not PART Scherz
Doch, ich gebe an Scherz.
yes/however I boast PART Scherz
‘Stop boasting – just kidding.’ – ‘No, I will boast – just kidding.’
Such examples point to another interaction of pejoration with ludic aspects:
the jocular marking that creates commitment-free speech acts brings
together joking and (de-)pejoration.
3.2 Depejoration and joking: words said in jest
Scherz/Spaß and şaka explicitly characterise an utterance as a “joke” and
thus mark it as not serious, as something not said with its performative
force, but just ‘in jest’. This ludic categorisation signals that the hearer has
been led on and thus cancels the aggression involved in prior pejoration.
This is in accordance with general features of jokes (cf. e.g. Raskin 1985,
Attardo 2003): a joke sets up a schema, leading to a conflict (usually the
detection of incongruity) and its resolution.
In the case of depejorative Scherz/Spaß, the schema is set up by a
speech act that involves criticism or pejoration, usually targeted at the
hearer, and will thus be perceived negatively by the hearer, leading to a
conflict at the level of social relationships: the aggression involved clashes
with the expected behaviour in the kind of situation where depejorative
Scherz/Spaß typically occurs, namely peer-group conversations among
friends. This incongruity is then resolved by the speaker who robs his
38
speech act of its performative force and marks it as ludic. Hence, the “comic
release” in this case is constituted by the relief that the social relationship is
not threatened.
By marking the speech act as a joke, Scherz/Spaß annuls a pejoration
altogether, indicating that it was not meant seriously by the speaker right
from the beginning. It is thus more powerful than an apology, which can
also retract a pejoration, but keeps the initial speech act intact. A
construction with Scherz/Spaß creates something like a second-order joke:
the joke is that the preceding was only a joke.
This is, then, similar to bald-faced lies in that in the end, it is part of
the common ground for both speaker and hearer that the speaker’s utterance
was intentionally false (cf. Meibauer 2014).
23
Nevertheless, unlike in bald-
faced lies, a construction with Scherz/Spaß is not an act of verbal
aggression, but rather the opposite, due to its jocular marking: in jokes it is
acceptable to be fooled by the speaker; as Mey (1991:236) put it: “the hearer
is left with the feeling of having been taken in—but not necessarily in an
unpleasant way, since it was, after all, a joke’’. Taken together this
construction, then, combines ludic and pejorative aspects in a way that
allows speakers to criticise the hearer and use pejoration without causing
social conflicts.
23
Note, though, that there can also be cases where the speaker did intend to perform a pejorative
speech act at the time, but afterwards wants to take it back, and therefore adds “Scherz” (We thank
Rita Finkbeiner for pointing this out to us). We believe that such cases constitute a kind of “second-
order deception” where the speaker pretends to have uttered something intentionally wrong which
in fact was initially meant seriously.
39
4 Conclusion: Developmental paths and constructional pejoration
In this paper, we investigated the development of pejorative patterns in
urban German together with their Turkish counterparts, which we identified
as a plausible source for these developments, and some parallels in Yiddish
and American English. For German, our main focus was on Kiezdeutsch, a
new urban dialect characteristic of peer-group conversations among the
particularly dynamic group of young people in multilingual inner-city
neighbourhoods. We discussed two patterns. The major part of our study
was on echo word formation, that is, constructions where a word is followed
by an echo twin, in the case of urban German and Turkish, involving a left-
peripheral /m/ (“Cola Mola”), in Yiddish and English, /ʃm/ (“messy,
shmessy”). On top of this, we investigated the use of German
“Scherz”/“Spaß” and Turkish “şaka” ‘joke, fun’ as markers that cancel the
performative force of a preceding utterance and can thus induce
depejoration, the annulment of pejoration.
For echo formation, we found, in addition to Turkish m-reduplication
as a likely source, language-internal parallels in lexicalised rhymes and
name rhyme mocking from German. Together, this supports the
development of a pejorative pattern: as we showed, pejoration does not
emerge from individual evaluative elements in this case, but is associated
with the construction as a whole, something we analysed as “constructional
pejoration”.
40
Such constructional, pattern-based pejoration can highlight links
with other domains: for both m-reduplication and “Scherz”/“Spaß” as
performative markers, we found that (de-)pejoration is only one of the
possible functions for the construction, alongside others such as
amplification, marking of ‘coolness’, and ludic aspects. Our analysis
indicated that these liaisons are not random, but can be captured by
systematic conceptual relationships and pragmatic inferences, integrating
pejoration into a rich network of cognitive domains.
5 References
Akaslan, Talat. 2006. Yaşlı maşlı ama dipdiri: Yarı-ikilemenin anlam-
sözdizimi. In: Semiramis Yağcıoğlu & Ayşen Cem Değer (eds.):
Advances in Turkish Linguistics. Proceedings of 12th International
Conference on Turkish Linguistics, 11-13 August, 2004, Izmir: Dokuz
Eylül Yayıncılık. 53-64.
Anderson, Luvell & Lepore, Ernie. 2013. Slurring Words. Noûs 47: 25-48.
Attardo, Salvatore. 2003. The pragmatics of humor. Journal of Pragmatics
35; 9: 1287-1294.
Bzdęga, Andrzej. 1965. Reduplizierte Wortbildung im Deutschen. Poznan:
Praca Wydana z Zasiłku Polskiej Akademii Nauk.
Camp, Elisabeth. 2013. Slurring perspectives. Analytic Philosophy 54;3:
330-349.
41
d’Avis, Franz & Meibauer, Jörg. 2013. Du Idiot/Din Idiot! Pseudo-vocative
constructions and insults in German (and Swedish). In: Barbara
Sonnenhauser & Patrizia Noel Aziz Hanna (eds.): Vocative!
Addressing Between System and Performance. Berlin, New York:
Mouton de Gruyter, 113-141.
Dammel, Antje, & Quindt, Olga. this volume. How do pejorative
derivational meanings emerge? A bit of /Geforsche /or /Forscherei/.
Dirim, İnci, & Auer, Peter. 2004. Türkisch sprechen nicht nur die Türken.
Über die Unschärfebeziehungen zwischen Sprache und Ethnie in
Deutschland. Berlin: Walter de Gruyter [Linguistik – Impulse &
Tendenzen 4].
Demircan, Özgür. 1987. Emphatic Reduplication in Turkish Studies on
modern Turkish. In: Hendrik E. Boeschoten & Ludo Th. Verhoeven
(eds.): Studies on modern Turkish: Proceedings of the 3rd Conference
on Turkish Linguistics. Tilburg: Tilburg University Press. 24-
41.Eckert, Penny. 2000. Linguistic Variation as Social Practice.
Oxford: Blackwell.
Eren, Hasan. 1949. İkiz Kelimelerin Tarihi Hakkında. Ankara Üniversitesi
Dil ve Tarih-Coğrafya Fakültesi Dergisi 2;7: 283-286.
Feinsilver, Lillian. 1961. On Yiddish shm-. American Speech 36. 302-303.
Finkbeiner, Rita. this volume. Bla bla bla in German.
Göksel, Aslı & Kerslake, Celia. 2005. Turkish. A Comprehensive Grammar.
London, New York: Routledge.
42
Hom, Christopher. 2008. The semantics of racial epithets. Journal of
Philosophy 105:8, 416–440.
Inkelas, Sharon & Zoll, Cheryl. 2005. Reduplication. Cambridge:
Cambridge University Press.
Jackendoff, Ray S. 2002. Foundations of Language. Brain, Meaning,
Grammar, Evolution. Oxford: Oxford University Press.
Jay, Timothy & Janschewitz, Kristin. 2007. Filling the emotion gap in
linguistic theory: Commentary on Potts’ expressive dimension.
Theoretical Linguistics. 33, 215-221.
Karaağaç, Günay. 2013. Dilbilim Terimleri Sözlüğü. Ankara: TDK.
Keim, Inken. 2010. Sprachkontakt: Ethnische Varietäten. In: Hans-Jürgen
Krumm, Christian Fandrych, Britta Hufeisen & Claudia Riemer (eds.):
Deutsch als Fremd- und Zweitsprache. Berlin, New York: de Gruyter
[HKS 35;1]. 447-457.
Kern, Friederike, & Selting, Margret (eds.). 2012. Ethnic Styles of Speaking
in European Metropolitan Areas. Amsterdam: Benjamins [Studies in
Language Variation 8].
Kerswill, Paul. 1996. Children, adolescents, and language change.
Language Variation and Change 8: 177-202.
Koo, Hyun Jung & Rhee, Seongha. this volume. Pejoratives in Korean.
Korkmaz, Zeynep. 2007. Gramer Terimleri Sözlüğü. Ankara: TDK.
Lewis, Geoffrey. 2000. Turkish Grammar. Second edition. Oxford: Oxford
University Press.
43
Matras, Yaron. 2011. Universals of structural borrowing. In: Peter Siemund
(eds.): Linguistic Universals and Language Variation. Berlin:
Mouton. 204-233.
McCarthy, John & Prince, Allan. 1986. Prosodic Morphology. Working
Paper [published 1996], University of Mass. at Amherst Linguistic
Department Faculty Publication Series.
Meibauer, Jörg. 2013. Expressive compounds in German. Word Structure
6;1: 21-42.
Meibauer, Jörg. 2014. Bald-faced lies as acts of verbal aggression. Journal
of Language Aggression and Conflict 2;1: 127-150.
Mey, Jacob L. 1991. Pragmatic gardens and their magic. Poetics 20: 233-
245.
Müller, Hans-Georg. 2004. Reduplikationen im Türkischen.
Morphophonologische Untersuchungen. Wiesbaden: Harrassowitz
[Turcologica 54].
Nevins, Andrew, & Vaux, Bert. 2003. Metalinguistic, shmetalinguistic: the
phonology of shm-reduplication. Proceedings of the Chicago
Linguistics Society Annual Meeting 39: 702-721.
Potts, Christopher. 2007a. The centrality of expressive indices. Theoretical
Linguistics 33(2), 255–268.
Potts, Christopher. 2007b. The expressive dimension. Theoretical
Linguistics 33(2), 165–198.
Raskin, Victor. 1985. Semantic Mechanisms of Humor. Dordrecht: Kluwer.
44
Samarin, William J. 1970. Inventory and Choice in Expressive Language.
Word 26: 153-169.
Schroeder, Christoph. 1989. Sprachlicher Ikonismus: Theoretische
Grundlagen und Beispiele aus dem Türkischen. Papiere zur Linguistik
41: 3-76.
Southern, Mark. 2005. Contagious Couplings: Transmission of Expressives
in Yiddish Echo Phrases. Westport, Conn.: Greenwood.
Spitzer, Leo. 1952. Confusion shmooshun. Journal of English and
Germanic Philology 51: 226-233.
Stamer, Heiko. 2014. Frischer Wind-Mind in der deutschen Sprache:
Gebrauchskontexte und Spracheinstellungen der m-Reduplikation
nach türkischem Vorbild in jugendsprachlichen Varietäten des
Deutschen. MA-Thesis, University of Bremen [Advisors: Christel
Stolz and Thomas Stolz].
Stolz, Thomas. 2008. Total reduplication vs. echo-word formation in
language contact situations. In: Peter Siemund & Noemi Kintana
(eds.): Language Contact and Contact Languages. Amsterdam:
Benjamins. 107-132.
Weinreich, Uriel. 1953. Languages in Contact: Findings and Problems. Den
Hague, Paris: Mouton.
Whiting, Daniel. 2013. It’s not what you said, it’s the way you said it: slurs
and conventional implicature. Analytic Philosophy 54;3: 364-377.
45
Wiese, Heike. 2011. so as a focus marker in German. Linguistics 49;5: 991-
1039.
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: 207-245 [“Dialektologie in neuem
Gewand. Zu Mikro-/Varietätenlinguistik, Sprachenvergleich und
Universalgrammatik”, ed. Werner Abraham & Elisabeth Leiss].
Wiese, Heike; Freywald, Ulrike; Schalowski, Sören, & Mayr, Katharina.
2012. Das KiezDeutsch-Korpus. Spontansprachliche Daten
Jugendlicher aus urbanen Wohngebieten. Deutsche Sprache 2: 97-
123.
Yastı, Mehmet. 2007. Türkçe Deyimlerde Geçen İkilemelerin Ses ve Şekil
Özellikleri. Türkiyat Araştırmaları Dergisi 21: 51-87.
Zimmermann, Malte. 2007. I like that damn paper – Three comments on
Christopher Potts’ The expressive dimension. Theoretical Linguistics
33;2: 247–254.
Acknowledgements
Part of the research presented here was supported by a German Research
Foundation (DFG) grant to the Collaborative Research Centre (SFB)
‘Information Structure’ 632 of Potsdam University, Humboldt-University
46
Berlin, and Free University Berlin (project B6 ‘Kiezdeutsch’, PI: Heike
Wiese). For feedback and comments on different aspects of the material and
analyses, we would like to thank two reviewers, participants of the DGfS
2014 workshop on Pejoration organised by Jörg Meibauer and Rita
Finkbeiner, audiences at the ‘Turkish Week’ (University of Potsdam,
November 2014) and at invited talks at Humboldt-University Berlin,
University of Cologne, and Meertens Institute Amsterdam.