PreprintPDF Available

Linguistic variation and age of speakers in Namibian German: loan word usage in “Wenker sentences”

Authors:
Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

In Namibia, German is spoken as a minority language in a speech community that has its roots in immigration from Europe in the context of colonialism. In contrast to what is happening in most varieties of German outside Europe, Namibian German is consistently passed on to the younger generations, thus offering a special opportunity to investigate the age of speakers as a factor in linguistic variation. To study this, I used data from an online survey eliciting translations of the classical "Wenker sentences" into Namibian German, with over 100 participants from 12 to 78 years of age. My paper presents results from quantitative and qualitative analyses of loan words from the major contact languages Afrikaans and English. Findings indicate that the quantity of loan words is interrelated with the age of speakers, while their grammatical integration does not show cross-generational differences. Comparing my findings with those from other varieties, I discuss the vitality of Namibian German and examine how the concepts of age grading and apparent-time change can contribute to our understanding of linguistic variation in language islands.
Content may be subject to copyright.
1
pre-final version; will be published in:
Hans C. Boas (ed.), German Abroad: Comparative Perspectives on Language Contact.
Leiden: Brill.
Christian Zimmer
FU Berlin
Linguistic variation and age of speakers in Namibian German:
loan word usage in “Wenker sentences”
Abstract
In Namibia, German is spoken as a minority language in a speech community that has its
roots in immigration from Europe in the context of colonialism. In contrast to what is
happening in most varieties of German outside Europe, Namibian German is consistently
passed on to the younger generations, thus offering a special opportunity to investigate the age
of speakers as a factor in linguistic variation. To study this, I used data from an online survey
eliciting translations of the classical “Wenker sentences” into Namibian German, with over
100 participants from 12 to 78 years of age. My paper presents results from quantitative and
qualitative analyses of loan words from the major contact languages Afrikaans and English.
Findings indicate that the quantity of loan words is interrelated with the age of speakers, while
their grammatical integration does not show cross-generational differences. Comparing my
findings with those from other varieties, I discuss the vitality of Namibian German and
examine how the concepts of age grading and apparent-time change can contribute to our
understanding of linguistic variation in language islands.
1
1. The special status of German in Namibia
Namibian German has a special status among varieties of German outside Europe (cf. Wiese
et al. 2014). The vast majority of these varieties declined significantly over the past decades,
and language death is imminent in many cases now. In some cases, the variety will become
extinct within one generation, since German is not passed on to younger speakers anymore
(cf., e.g., Boas 2003, 2009 on Texas German).
2
In contrast to these moribund varieties,
1
The research on which this paper is based is conducted within the project Namdeutsch: Die Dynamik
des Deutschen im mehrsprachigen Kontext Namibias”. This project is funded by the Deutsche
Forschungsgemeinschaft (DFG, German Research Foundation) WI 2155/9-1; SI 750/4-1. I would like to thank
Sheena Shah, Horst Simon, Heike Wiese, and two anonymous reviewers for their useful comments on earlier
versions of this paper. Furthermore, I would like to thank Heike Wiese for the great opportunity to use the
translations of the “Wenker sentences”, which she collected in 2013 and 2014. Finally, Semra Kizilkaya and
Britta Stuhl are to be thanked for their help with the data preparation.
2
See the contributions in Johannessen and Salmons (2015b) on German and other Germanic languages in
North America and Eichinger, Plewina, and Riehl (2008) on Central/Eastern Europe. Cf. also Behrend & Knipf-
2
German is consistently passed on to the younger generations in Namibia.
3
Moreover, although
being a minority language, it has strong institutional support as it is used in kindergartens,
schools,
4
the media, some areas of business, in worship, and in tourism (cf. Ammon 2014,
35969).
5
Consequently, there are Namibian German-speakers of every age, and Namibian
German thus offers a special opportunity to investigate the age of speakers as a factor in
linguistic variation within a language island.
6
In this paper, I examine how the age of speakers and the use of loan words are
interrelated in Namibian German. With this study, I aim at filling several research gaps and
intend to contribute to a detailed description of the situation of German in Namibia. For
example, the organisation of the varieties of Namibian German still remains unclear (cf.
Kellermeier-Rehbein 2016). In the literature, it has been claimed that there is a youth variety
(sometimes referred to as “Nam-Släng”) that differs strongly from the language use of adults,
which is said to very much resemble the German German standard variety (cf. Kellermeier-
Rehbein 2015 for an overview). The term “Nam-Släng” was promoted by the German-
Namibian musician “EES” (Sell 2011). Older generations typically use(d) the term
Südwesterdeutsch (‘South-Westerners’ German’), which relates to Namibia’s former
colonial name (cf. Gretschel 1995, Pütz 1991). However, as I will show in the following
sections, the data do not support the idea of a clear-cut age-related dichotomy. This is why I
prefer the name “Namdeutsch” (‘Nam-German’) to refer to German with Namibia-specific
features. This term is also used by Namibians as a more neutral, overarching term (cf. Zimmer
in press a for more details).
7
Furthermore, contrasting older and younger speakers systematically will provide new
insights into the vitality of German in Namibia. As will be shown in the following sections,
there is no evidence for an increasing degree of relexification in younger speakers. Quite the
Komlosi (2006), Keel and Mattheier (2003), and Keel, ##this volume, for a discussion of the reasons for the
decline of German language islands.
3
Today, ca. 20,000 Namibians belong to the German speaking minority. For a detailed overview of the
present situation cf. Ammon (2014, 35969), Shah and Zappen-Thomson (2018), and Zimmer (in press a). Note
that German is consistently acquired only in the speech community that has its roots in immigration from
Europe. In contrast, Kiche Duits (‘Kitchen German’) a German-based contact variety that was mainly used for
inter-ethnic communication is a dying variety. Nowadays, Afrikaans and English are usually used as lingua
francas instead of German (cf. Deumert 2009).
4
Some remarks regarding norms imparted by school education can be found in Section 3.2.2.
5
The institutional support is one important difference to language island varieties spoken in Sectarian
communities in the Americas (e.g. Pennsylvania Dutch, cf. Louden 2016). Furthermore, in these communities
German is passed on to the younger generations, but children are schooled in the majority language, for example
English (cf. Johannessen and Salmons 2015a).
6
For want of a better term, I use language island here although the metaphor does not work very well
with regard to Namibian German because of its strong contact with other Namibian languages. The major
problem with regard to obvious alternatives, such as external variety or extraterritorial variety, is the term
variety as the number and organization of varieties of German in Namibia has not yet been determined
conclusively.
7
In addition, “Näm-Slang” is in particular used to refer to mediated language use in online media, such as
Youtube. As it is not clear how representative this kind of language use is for everyday language, I prefer the
neutral overarching term “Namdeutsch”.
3
contrary: the sociolinguistic variability of German in Namibia can be interpreted as a sign of
linguistic vitality.
Comparing my findings on Namdeutsch with both external varieties and internal
varieties, I discuss to which extent these findings can be explained by cross-linguistically
valid principles. These comparisons will also help us to assess the vitality of Namdeutsch.
I have chosen to study lexical borrowings, which are the most obvious result of
language contact, in connection with the age of speakers. Lexical borrowings are numerous in
Namdeutsch, which is probably also true for every other contact variety. The high number of
loan words enables us to conduct the kind of quantitative analysis that is necessary to answer
my research questions outlined above. Furthermore, the grammatical integration of loan
words has been considered an important aspect when assessing the vitality of a language
island (cf. Boas 2009, 23234).
In the following, I first describe the data collection (Section 2). Based on this, I analyse
the use of loan words (Section 3). I provide information on the number and origin of the loan
words used in our data, and on the grammatical integration of borrowed lexical items (Section
3.1). The results will be contrasted with comparable data from Texas German. In Section 3.2,
I look into the interrelation of speakers’ age and the amount of loan words used. As will be
shown, the pattern we find resembles those in many internal varieties. Finally, I will discuss
the implications of my study in a short conclusion (Section 4).
2. Data collection: The “Wenker sentences
The empirical basis of my study is a corpus of “Wenker sentence” translations into Namibian
German that was compiled by Heike Wiese in 2014 within the context of a pilot study relating
to the project mentioned in the acknowledgements (Wiese 2014). Participants were asked to
translate the so-called “Wenker sentences” into their local vernacular, that is, to reformulate
them in accordance with their way of talking to their friends or family.
8
These sentences,
which are a classical tool of German dialectology, were designed by Georg Wenker in the 19th
century for an early large-scaled study on language variation.
9
Subsequently, they were used
not only by Wenker himself but also by numerous other scholars and are still used by many
linguists today. Thus, my results are highly comparable with numerous other studies on
German varieties, both recent and older ones. This great advantage outweighs some
disadvantages with regard to the contents. First, a few sentences are not well-suited for the
Namibian climate, as can be seen in (1).
8
The complete data set including the instruction for the participants and the original “Wenker sentences“
are accessible online: <https://www.uni-potsdam.de/de/dspdg/projekte/namdeutsch/korpus.html> (20th June
2018).
9
For more information on Wenker’s research cf., e.g., Fleischer (2017) and Lameli (2008, 25659).
4
(1) Der gute alte Mann ist mit dem Pferd auf dem Eis eingebrochen und in das
kalte Wasser gefallen. (Wenker sentence 4)
The good old man broke through the ice with his horse and fell into the cold
water.
10
A few participants refused to translate this sentence, pointing out that water never freezes in
Namibia. However, since most participants did translate this sentence (as well as two further
ones, which were similarly problematic with respect to their content), there was only a
marginal reduction of data points. Furthermore, on the basis of this translation task it is only
possible to examine “core borrowings, which duplicate meanings for which a native word
already exists” (Haspelmath 2009, 46). Thus, cultural borrowings, e.g. words for technical
innovations are not part of my investigation (for an overview concerning this distinction see
Haspelmath 2009, 4649).
Second, the content of some sentences is somewhat archaic from a present-day
perspective (cf., e.g., (2)).
(2) Geh, sei so gut und sag deiner Schwester, sie soll die Kleider für eure Mutter
fertig nähen und mit der Bürste rein machen. (Wenker sentence 17)
Go, be so good and tell your sister she should finish sewing the clothes for your
mother and clean them with a brush.’
Nonetheless, the participants were able to understand the original sentences as they produced
meaningful translations. No one indicated that s/he did not understand the originals, which is
not very surprising given that most German-speaking Namibians have at least a thorough
passive knowledge of Standard German.
Another methodological aspect concerns the type of data that is produced by translation
tasks. Translations such as Wenker elicitations do not yield authentic language data
comparable to speech corpora. However, they provide an important type of evidence: if
similar deviations from the original show up frequently and even across speakers, then this
will indicate that speakers consider the deviating structure as typical for their language use.
11
Table 1 comprises a small set of representative translations of a “Wenker sentence. The
first line gives the original sentence.
10
Most translations of the “Wenker sentences“ into English are taken from
< http://www2.ku.edu/~germanic/LAKGD/wenkersaetze.shtml> (26th June 2017).
11
Fleischer, Hinterhölzl, and Solf (2008) provide a detailed discussion and application of this differential
method for studies on historical syntax; Fleischer (2015) presents comparable studies on dialect syntax using the
original “Wenker data”. See also Wiese et al. (2014).
5
Der gute alte Mann ist mit dem Pferd auf dem Eis eingebrochen
und in das kalte Wasser gefallen.
The good old man broke through the ice with his horse and fell into the cold water.
Der Outoppie ist mit dem Gaul durchs Eis gesmashed und ins Wasser gedroppt
Der toppie is mit dem pferd durchs eis gedonnert in das kalte wassa
Der oulike ou toppie ist mit n' Pferd ins Eis eingebrochen und ins freezing Wasser gemoert.
Der gute alte Mann is mit'm Pferd auf'm Eis eingebrochen und ins kalte Wasser gefalln.
Der gute ou top ist mit dem Pferd auf dem Eis eingebrochen und ins kalte Wasser geflogen.
Table 1: Translations of “Wenker sentence 4
In order to reach as many speakers as possible, the data collection was carried out via an
online questionnaire. On the website, participants were given some general information about
Georg Wenker and the “Wenker sentences”, together with examples from earlier studies on
dialects in Germany. To recruit participants, the study was advertised through interviews with
the local German-language radio (Hitradio Namibia), the German-language Namibian
newspaper Allgemeine Zeitung, and through a cooperation with the German Lutheran bishop
in Namibia, resulting in a sample size of of 216 participants. In addition to providing
translations of the 40 sentences, speakers were asked to fill in a questionnaire on biographical,
social, and sociolinguistic data, such as age, gender, birthplace, residence, occupation, and
languages spoken at home and at work. For my study, I excluded data sets of participants who
reported that they were not born or did not grow up in Namibia and of those who did not
provide information about their residence and/or birthplace. The remaining 116 data sets
cover a broad range of German-background speakers from different regions of Namibia, aged
12 to 78 (mean age: 44.9 years with a standard deviation of 17.4; 74 female, 41 male
participants and one data set without information on speaker’s gender).
12
3. Loan words
The participants made massive use of loan words.
13
99 lexemes of the original “Wenker
sentences, which consist of 470 word forms and 238 lexemes, were replaced by a loan word
12
All but two of these participants indicated that they speak German at home (possibly in addition to
other languages). The translations of the two other participants (one speaking only English at home, the other
one Oshivambo) did not deviate strikingly from the rest of the translations. For instance, they are not
characterised by a particularly high amount of loan words.
13
I use the term loan word according to its definition as a „word that at some point in the history of a
language entered its lexicon as a result of borrowing“ (Haspelmath 2009, 36). Loan words have to be
distinguished from code switches. “Code-switches […] by definition are not integrated into a recipient language,
unlike established loanwords” (Poplack 2012, 645). Since graphematic integration is not a necessary condition
for the classification as a loan word, and phonological adaption cannot easily be assessed on the basis of written
data, and moreover only six tokens in the whole data set are morphologically unintegrated, I generally use the
term loan word to refer to non-native lexical items in the translations. The few instances of morphologically
unintegrated words which could possibly be analysed as code-switches are commented on in Section 3.1.
6
by at least one participant. Loans from Afrikaans were used most often by a wide margin
(1096 tokens, 158 types),
14
followed by loans from English (290 tokens, 77 types).
15
Apart
from these lexical transfers from the two major contact languages of German in Namibia,
there are also some single words borrowed from Bantu and Khoisan languages, for example,
ge!xinat (‘stolen’), which appears to be borrowed from Khoekhoegowab (cf. (3)).
(3) Wer hat mein Korb mit Fleisch ge!xinat??
Who stole my basket of meat?
Another interesting word which is neither inherited from German German nor borrowed from
Afrikaans or English is nxa (‘good’, cf. (4)).
(4) Es hört now auf zu schneien, dann wird das wieder nxa
It will soon stop snowing, then the weather will be good again.
Probably, the <x> represents a click here. During our field research, we noticed that nxa is
used in Namdeutsch on a relatively regular basis and that a click is articulated. However, it is
possible that this word is actually not a loan word but a somewhat artificial creation. The
word is comparably easy to pronounce for speakers who are not familiar with clicks and
might be used as a means to express openness towards non-Germanic languages.
16
While the majority of the original lexical items was replaced by at least one participant,
some content words were never substituted by a loan word, e.g., Durst (‘thirst’). The word
that was replaced most often is müde (‘tired’). Only 18 out of 116 participants used this
adjective. Commonly, it was substituted by its Afrikaans equivalent moeg (63 times).
Furthermore, I do not distinguish between nonce borrowings and established loan words in this paper, since such
a distinction should be based on solid data on their respective frequencies (cf., e.g., Poplack 2012). If an item
occurs only once in our data set, one cannot conclude that this item is generally used very rarely by the speech
community at hand, which would be the defining criterion for the classification as a nonce borrowing.
14
In some cases it is not possible to decide whether a word was borrowed from Afrikaans or English due
to similarities in the lexicons of these languages. These words were excluded from the overview given above.
15
Possibly, some of the first settlers were bilinguals and spoke next to High German also Low German,
which is in some respects quite similar to Afrikaans, besides High German. Thus, some loan words, which are
classified as Afrikaans here, might also have been transferred to Namibian (High) German from Low German.
For example, mooi (‘fine’) is used both in Afrikaans and in some Low German varieties (cf. Kellermeier-
Rehbein 2018). However, words such as mooi are generally perceived as originating from Afrikaans by the
Namibian German speech community, cf. the following statement: “Wenn wir sagen […] ‘war die Pad heute
mooi oder war die schlecht’, dann wisst ihr gar nicht, was wir meinen damit. Das ist ein afrikaanser Begriff” ‘If
we say ‘how was the road? Mooi or bad?’ you don’t know what we mean. It’s an Afrikaans term’). This excerpt
is taken from an interview with a Namibian which will be accessible in the near future. From a sociolinguistic
perspective, especially the perceived origin of these words is crucial with regard to age-grading (cf. Section 3.2).
Nowadays, Low German does not play any role in Namibia anymore while almost all speakers of German are
also fluent in Afrikaans.
16
Tom Güldemann and Sylvanus Job are to be thanked for pointing at this interpretation and for a fruitful
discussion on other aspects of this paper.
7
The loan words in our data cover a broad range of parts of speech. This is illustrated in
Table 2.
example
origin
moeg (‘tired’)
Afrikaans
Story
English
gooi (‘give’)
Afrikaans
mos (~‘just’)
Afrikaans
Table 2: Loan words in Namdeutsch: parts of speech
Interestingly, the participants also substituted German words belonging to basic vocabulary.
For example, German kinship terms for close relatives were used comparatively infrequently
(cf. Table 3).
original lexeme
original retained
substitutions
proportion
of substitutions
Mutter (‘mother’)
37
44 (Ma)
54.3 %
Bruder (‘brother’)
31
27 (broer)
4 (boet)
50.0 %
Schwester (‘sister’)
59
19 (sister or variant thereof)
10 (sus or variant thereof)
33.0 %
Table 3: The substitution of German kinship terms
3.1 Grammatical integration
Borrowings are generally integrated into Namdeutsch morphosyntactically. In our data, this
can be observed for speakers of all generations. There are no age-related differences in our
data with regard to this aspect. In the following, this will be shown for adjectives, verbs and
nouns. Borrowings belonging to other inflecting parts of speech (pronouns and determiners)
are not attested in our data.
The integration of adjectives into the grammatical system of German can easily be
shown by example of adjectival loan words from Afrikaans. Afrikaans has neither
grammatical gender nor overt case marking on adjectives. At most, there are two word forms
per lexeme: the bare stem (e.g. mooi ‘nice’, cf. (5) and a word form with the suffix -e, e.g.
mooi-e, cf. (6) (cf. Donaldson 1993, 16390 for an overview of the intricate system of
adjectival inflection in Afrikaans).
8
(5) Sy is ’n mooie mens.
17
She is a very nice person.’
(6) Ek soek n mooi kar.
‘I’m looking for a nice car.’
In Namdeutsch, however, adjectives borrowed from Afrikaans are used according to the
complex inflection rules of German (cf. also Kellermeier-Rehbein 2015, 49, Shah 2007, 23
24). The presence/absence of determiners as well as case, number, and gender of the head
noun determine the grammatical properties of the adjective in attributive function. For
example, kwaai (‘evil’) in (7) conforms to these rules, yielding a word form that does not
exist in Afrikaans. All ten participants who used kwaai in the context shown in (7) used the
word form kwaaien or a graphemic variant thereof.
18
(7) Hey laitie, bleib net da unten, sonst beißen dich die kwaaien Gänse.
‘Hey kid, stay down here. Otherwise the mean geese will bite you.
Also other adjectives were almost always inflected according to the grammatical system of
(Standard) German including borrowings from English (cf. (8), which contains the
graphematically and morphosyntactically integrated word form neiß-e ‘nice-NOM.SG’). All in
all, 97 adjectival loan word tokens were used in the translations (not counting predicatively
used and hence not inflected adjectives). Only four of them (4.3%) are not
morphosyntactically integrated (cf. (9)).
19
(8) Der neiße Outoppie ist mit dem Gaul im Eis eingebrochen und ins kalte Wasser
gefallen.
The nice old man broke through the ice with his horse and fell into the cold
water.’
(9) Liebes Kind, bleib net hier unten stehen, die bleddie Gänse beißen dich vreck.
‘Dear child, stay down here, the bloody geese will bite you to death.’
17
This example is taken from Donaldson (1993, 169).
18
Note that net in the example is borrowed from Afrikaans (‚just‘) and is not to be confused with the
dialectal German German negation net (not).
19
Note the difference between kwaai-en (evil-PL) in (7) vs. bleddie-Ø (‚bloody‘) in (9). This difference
might be motivated phonologically: Using the Standard German suffix -en with bleddie results in a hiatus, which
is generally dispreferred in German.
9
Likewise, borrowed verbs are usually adapted to the (Standard) German conjugation. For
example, chau-t (‘eat-3SG’) in (10) contains the third person singular suffix -t, which does
neither exist in Afrikaans nor in English. Only two of 123 verbal loan word tokens (1.6%) are
not integrated morphosyntactically (cf. (11)). collapsed in (11), for example, is inflected
according to the English grammatical system.
(10) Er chaut die Eier immer ohne Salz und Pfeffer.
He always eats eggs without salt and pepper.
(11) Der gute ou toppie ist mit dem Gaul auf dem Eis collapsed und in das kalte
Wasser gefallen.
The good old man collapsed with his horse on the ice and fell into the cold
water.’
In contrast to adjectives and verbs, the morphological behaviour of nouns is less informative
with regard to morphological integration because all plural markers that were used in the
translations and that could be interpreted as originating from Afrikaans (e.g. -e) or English
(e.g. -s) also exist in German. However, some borrowings were used in a way that does not
comply with the noun’s morphological behaviour in the donor language. For example, the
originally English noun Farmer is consistently used without a plural marker (cf. (12)). In all
of the 47 occurrences in our data Farmer is used with a zero plural ending.
(12) Die Farmer hatten 14 Beester und 12 Schafe zum Verkaufen
zusammengetrieben.
The farmers had mustered 14 oxes and 12 sheep together in order to sell them.’
This complies with a rule of the (German) German plural system according to which stems
ending in -er ([ɐ]) do not take a plural marker. In contrast to this, English would require the
plural marker -s here.
This is an interesting observation because it has been claimed that the plural marker -s
spreads in German German due to language contact with English (cf. Bornschein and Butt
1987, 14344). However, English loan words like Farmer (cf. also Computer, Printer,
Synthesizer etc.) that are used without a plural marker in German but are used with -s in
English show that the -s is not simply taken over along with lexical borrowings. The English
loan words are adapted to the German grammatical system: depending on the phonological
10
structure of the loan, it is used either with -s or without a plural marker.
20
As shown above,
this is even the case for Namdeutsch, where we have much more intense language contact
with English than in German German. The linguistic repertoires are kept distinct in this
respect (farmers in English vs. Farmer in German).
21
Thus, we do not find these
developments in Namibian German that Boas (2009, 232) considers to be characteristic of
language death”: there is no evidence for a breakdown of Namdeutsch morphology. Even loan
words (adjectives, verbs, and nouns) are used according to the grammatical system of
German.
In this respect, Namibian German differs from other German language islands, e.g.,
Texas German, where we can observe morphological restructuring. There, plural forms
without a plural suffix are particularly prone to change (a tendency we cannot observe at all
for Farmer in Namdeutsch). These changes are not limited to the inflection of loan words: We
generally find an increase in productivity of two plural morphemes, namely -s and -n, a
characteristic of dying languages and dialects” (Boas 2009, 234). For example, -s is quite
often used with Garten (‘garden’), which then deviates from Standard German Gärten (cf.
Table 4; adapted from Boas 2009, 231).
plural form
frequency
Garten
57% (27)
Gärten
7% (3)
Gartens
36% (17)
Table 4: Plural forms of Garten (‘garden’) in the Texas German Dialect Project (TGDP, Boas 2009,
231)
Gilbert (1972) reports only two instances of Gartens in his data, which accounts for 14% of
all Garten plural forms. Hence, there seems to be an increase in the use of the plural marker -s
in Texas German. Also for Wagen (‘waggon’) there is a considerable amount of s-plurals that
deviate from Standard German (cf. Table 5; adapted from Boas 2009, 229).
22
20
The -s is the default plural marker for loan words in German. It is not only used with English loan
words, but also with those from other languages, cf., e.g., German Sauna-s vs. Finnish sauna-t (‘saunas’). It has
been suggested that this marker is a reanalysed native genitive suffix (cf. Nübling and Schmuck 2010). For more
information on the German s-plural, which can be replaced by other markers if a lexeme is used frequently (e.g.,
Saunas > Saunen; Zimmer in press b) cf., e.g., Clahsen (1999), Gaeta (2008), Wegener (2004), and Zimmer
(2018, 10318).
21
The word Farmer could possibly have been part of the lexicon of the German-speaking emigrants who
settled in what is Namibia today in the 19th century. However, the question whether it was borrowed in Europe or
in Africa is not particularly relevant for the argument. The intensity of language contact today is crucial here
22
Erroneously, the instances of Wagens are classified as containing an n-plural in Boas (2009, 229). This
seems to be just a typo since all listed speakers doubtlessly produced an s-plural. Accordingly, these instances
are classified as containing an s-plural here.
11
plural form
frequency
Wagen
78% (38)
Wagens
20% (10)
Wage
2% (1)
Table 5: Plural forms of Wagen (‘waggon’) in the TGDP (Boas 2009, 229)
Note, however, that this does not hold for all nouns in Texas German. For instance, Teller and
Zimmer are used with -s only once in Boas’s data (2% respectively). Instead, -n is used quite
often (cf. Table 6 and Table 7).
plural form
frequency
Teller
62% (29)
Tellern
36% (17)
Tellers
2% (1)
Table 6: Plural forms of Teller (‘plate’) in the TGDP (Boas 2009, 228)
plural form
frequency
Zimmer
43% (20)
Zimmern
55% (26)
Zimmers
2% (1)
Table 7: Plural forms of Zimmer (‘room’) in the TGDP (Boas 2009, 228)
Both forms (-s and -n) seem to be used to ‘repair’ Standard German zero-forms, which do not
conform with one major morphological principle: the relevance hierarchy (cf. Bybee 1985,
199; Dammel and Gillmann 2014; Zimmer 2018, 16369). According to this principle, it is
functionally motivated to mark number (in contrast, e.g., to case) close to the stem of a noun
and not (only) on determiners. Numerous instances of change support this claim in general
(cf., e.g., Dammel and Gillmann 2014). The spread of -s and -n in Texas German seems to be
explicable on the basis of this language-overarching principle, too.
23
This indicates once again
that language contact in scenarios like those of German in Texas leads to (accelerated) change
and that these changes can be explained by basic principles but do not lead to a linguistic
mess.
24
If both upcoming plural markers in Texas German are phonologically possible (i.e. if a
singular form does not end in n, as in Wagen or Garten), the use of -n significantly outweighs
23
Note that this observation is relevant for Garten, too. In Standard German, the plural form is Gärten
(with Umlaut). In Texas German, though, this form is very rare (cf. Table 4). Usually, the zero form is used.
Also, the plural form with Umlaut of Wagen (Wägen), which is possible in some German dialects, does not
appear at all in the Texas German data (cf. Table 5).
24
This conforms with Yager et al. (2015), who emphasize that variation in language islands such as Texas
German cannot be seen as the result of attrition or incomplete acquisition but rather as reflecting the emergence
of a new system.
12
the use of -s. Zimmers and Zimmern are equally possible, but Zimmern dominates strongly (cf.
Table 7). This runs counter to an explanation of these changes in the morphological system of
Texas German as mere structural transferences from English because the English plural suffix
is not used dominantly despite being a possible alternative. The increase in the use of -n might
result from a reanalysis of the dative plural -n as a plural marker (cf. (13)) a development
that might be supported by the ongoing loss of the dative case in Texas German (cf., e.g.,
Boas 2009, 175218).
25
(13) mit zwei Zimmer-n > Zimmer-n-Ø
mit two room-PL-DAT room-PL
In sum, Namdeutsch seems to behave more conservatively with regard to the morphological
system than Texas German does.
26
In Texas German there is a lot of variation in plural
markers, which Boas (2009) interprets as manifestations of the imminent death of this
language island. Changes in the morphological system cannot only be explained by
transferences from English. Also, the Namdeutsch morphological system does not seem to be
fundamentally influenced by transfer from its contact languages, as has been shown for the
plural forms of loan words. Adjectives and verbs are generally integrated
morphosyntactically. Furthermore, we did not observe much morphological variability in
Namdeutsch the vitality of Namibian German is reflected in its morphological stability.
3.2 Age of speakers and the amount of loan words used
In the following, I analyse the interplay between age and the use of loan words. Per group, the
number of loan words was compared to the number of their native German counterparts.
Added words were ignored, e.g. the intensifier bläddie in (14)b), which does not have an
equivalent in the original sentence (14)a). Such added intensifiers were studied separately (see
Section 3.2.3). In addition, those parts of the translations where the whole structure of the
original sentence had been changed were not included into my analysis, since these cases are
25
Maybe the design of the data collection also contributed to the high proportion of -n: The participants
were asked to translate two rooms into Texas German and were given no further context. Depending on the
context one would have to translate this phrase as containing a dative case (e.g. in a flat with two rooms: eine
Wohnung mit zwei ZimmernDAT.PL). If the participants had an example like this in mind, the use of zwei Zimmern
would be in accordance with Standard German. However, it seems not very likely that the participants thought of
an example with dative case, because dative case is quite marked and has furthermore declined significantly in
Texas German (cf., e.g. Boas 2009, 175218).
26
One major explanation for this difference might be the use of Standard German in those Namibian
schools attended by most speakers of Namdeutsch.
13
less relevant for my research question, cf., e.g., (15) where neither the original word
Geschichte (‘story’) has been kept nor an equivalent loan word has been used.
27
(14) a. Das Feuer war zu heiß, die Kuchen sind ja unten ganz schwarz gebrannt.
(“Wenker sentence” 6)
b. Das Feuer war zu bläddie heiss, die Kuchen sind von unten schwarz gebrannt.
The fire was too (bloody) hot. The cakes are burned black on the bottom.
(15) a. Wem hat er denn die neue Geschichte erzählt? (“Wenker sentence” 21)
Who did he tell the new story to?
b. sag mir wem er das gesagt hat
‘tell me to whom he said that’
In order to investigate the age of speakers as a factor in linguistic variation, I defined six age
groups. In accordance with the standard literature on age-grading and apparent-time change, I
used the sociolinguistically crucial stages of school and adolescence (< 20), work (20 60),
and retirement (60 +) as a general frame.
28
Proceeding from this general frame I further
divided the largest, work phase into four equal sections, resulting in the six age groups
depicted in Table 8. Note that the participants in our corpus are more or less evenly
distributed over the six groups with only a single outlier in the 50-59 section (cf. Table 8),
which allows for this detailed segmentation.
age
years of birth
n (persons)
< 20
1994 − 2001
16
20 29
1984 − 1993
10
30 39
1974 − 1983
14
40 49
1964 − 1973
17
50 59
1954 − 1963
32
60 +
1935 − 1953
27
Table 8: Age groups
27
Needless to say, these deviations as such are very revealing. However, strong structural deviations from
the original sentences are relatively small in number and they cannot simply be compared with 1:1 substitutions
of single words.
28
60 is the default retirement age in Namibia.
14
3.2.1 Loan word use and age of speakers in the “Wenker sentences”
To investigate the amount of loan words used in different age brackets, I considered all loan
words and their native German counterparts. Figure 1 gives an overview of the proportion of
loan words per age group.
Figure 1: Age groups and proportion of loan words
Table 9 provides the results of χ2 tests, which have been applied to assess the significance of
the differences between two adjacent age groups, each.
p
χ2
Φ
<20 vs. 20-29
> 0.05
2.4042
-
20-29 vs. 30-39
0.01019
6.6005
0.07721732
30-39 vs. 40-49
< 0.001
146.06
0.280981
40-49 vs. 50-59
< 0.001
66.665
0.1426299
50-59 vs. 60 +
0.001226
10.45
0.05421087
Table 9: Differences between age groups (use of loan words)
Obviously, the age group of the 40 to 49 year old participants differs most strongly (and
highly significantly) from its adjacent age groups. There are also significant differences
between the 20 to 29 and the 30 to 39 year old participants as well as between the 50 to 59
year old speakers and the speakers older than 60. However, the differences between these age
groups can be regarded as rather marginal (Φ < 0.1).
Interestingly, there is no significant difference between the youngest and the oldest age
group with regard to their willingness to use loan words (p > 0.05). The overall pattern
observable in Figure 1 can roughly be described as a U-shape: The oldest and the youngest
< 20 20-29 30-39 40-49 50-59 60 +
0%
5%
10%
15%
20%
25%
30%
35%
Proportion of loan words
15
age group use loan words with a comparably high frequency, while especially the group of 40
to 49 year old participants does not use loan words frequently at all.
This not only holds true for the sum of all words but also for single words, which can be
seen in Figure 2Figure 3. Here, the interrelation between age and willingness to use a
specific loan word is depicted. The two most frequently used loan words moeg (‘tired’) and
mooi (‘nice’) have been selected for this purpose. All other loan words have not been used
frequently enough to be suitable for this kind of analysis.
Figure 2: Age groups and proportion of moeg vs. müde
Figure 3: Age groups and proportion of mooi vs. schön
The graphs are very similar in all three figures (although, of course, on different levels),
hinting at a robust pattern in the data. Seemingly, the U-shape in Figure 1 is not a random
< 20 20-29 30-39 40-49 50-59 60 +
40%
45%
50%
55%
60%
65%
70%
75%
80%
85%
90%
moeg vs. müde (tired)
< 20 20-29 30-39 40-49 50-59 60 +
25%
30%
35%
40%
45%
50%
55%
60%
65%
70%
75%
80%
85%
mooi vs. schön (nice’)
16
result, but rather a pattern that is also detectable in the use of single words. Furthermore, the
similarities between the three figures at hand indicate that there are certain words that are used
by the youngest and the oldest age group equally frequently but not by the 40 49 year old
participants. This rules out a thinkable explanation of Figure 1, according to which young and
old people both use many, but different loan words. This finding is supported by the
distribution of borrowed types across generations, which is depicted in Table 10. 121 types
were used only once in our data set.
29
Logically, these types occur either in the translations of
younger (<40 years) or older (>40 years) participants respectively. But among those types that
were chosen by at least two participants, the majority was used both by older and by younger
participants (69.4%). 90.9% of the tokens in this category belong to one of these cross-
generationally used types. Looked at from the opposite perspective, this shows that there
might be some generation-specific words. chauen (from English ‘to chow’), for example, is
used quite frequently by the younger participants (14 occurrences) but at all not by older
participants. However, this example is definitely an exception. The vast majority of loan
words (especially in terms of tokens) is used cross-generationally.
type
frequency
number of
types
thereof cross gene-
rationally used
number of
tokens
thereof belonging to cross
generationally used types
1
121
0%
121
0%
>1
134
69.4%
1345
90.9%
Table 10: Distribution of borrowed types
3.2.2 Comparison and interpretation
The observed patterns of age-related variation in Namdeutsch differ quite strongly from what
one might have expected. In German language islands, young speakers typically deviate more
strongly from the standard variety than older speakers (cf., e.g., the pattern of age related
variation in the use of accusative forms in Texas German where dative case is used in
Standard German depicted in Figure 4; data taken from Salmons (1994, 61).
29
One might speculate that we are dealing with code switches here, which indeed cannot be ruled out for
every word. However, in many cases the words at hand are definitely not to classify as code switches as they are
established loan words and are, for example, also used frequently in the German language newspaper in
Namibia, the Allgemeine Zeitung. E.g., this holds true for Farm, which was used only once in the translations.
Many of those single occurrences are due to rather free translations of the “Wenker sentences”.
17
Figure 4: Year of birth and the proportion of non-standard accusative forms in Texas German (data
taken from Salmons 1994, 61)
Data like those depicted in Figure 4 are indicative of apparent-time change: age differences
are assumed to be temporal analogues, reflecting historical stages in the progress of the
change(Tagliamonte and D’Arcy 2009, 61). Consequently, the graph in Figure 4 seems to be
interpretable as reflecting the diachronic divergence of Texas German from its European
source varieties. The patterns revealed for Namdeutsch loan words, however, do not match
such a development. Instead, the graphs in Figure 1 to Figure 3 strikingly resemble what
Downes (1984, 190) calls the “normal patterns of age differentiation, when a variable is not in
the process of change” (cf. Figure 5), suggesting that we are not dealing with apparent-time
change here, but rather with more or less stable sociolinguistic variation.
Furthermore, an interpretation in terms of apparent-time change would be less advisable
because we are concerned with loan words. The apparent-time approach “relies on the
assumption that in most cases individual vernaculars remain stable throughout the course of
an adult lifetime” (Bailey 2002, 320), but especially the lexicon of an individual is capable of
change in every stage of life (cf., e.g., Kerswill 1996). This also runs counter to another
conceivable explanation according to which speakers born in the period of 1964 − 1973
behave differently: Conceivably, there has been a historical event which influenced the
language attitudes of this specific generation. For example, the independence of Namibia in
1990 or associated developments might explain why particularly this group behaves
differently. However, one would expect that not only one specific generation has been
influenced by such an event. To evaluate the possible explanations for the observed pattern
conclusively one would need to analyse longitudinal data, which are not available so far.
30%
35%
40%
45%
50%
55%
60%
65%
70%
75%
1912 and later 1900 ‒ 1911 1899 and before
Proportion of non-standard accusative forms in
Texas German
18
Figure 5: Normal distribution of age differentiation, when a variable is not in the process of change
(adapted from Downes 1984, 19091)
In Downes’s figure, the y-axis depicts the prestige attached to a variant, locating variants with
low prestige close to the x-axis. According to this graph, people between 40 and 50 years of
age use the fewest low prestige variants.
30
This fits quite nicely with my observations on the use of loan words in Namdeutsch
provided that loan words are considered low prestige variants. This is generally the case.
Although Namibian-specific patterns of German seem to play an important role in
constructing a German-Namibian identity, the German German standard variety functions as a
kind of normative ideal and is taught in schools. Thus, obvious deviations from this ideal are
generally considered non-standard.
31
This point of view is detectable in many statements of
German-speaking Namibians. For example, in her answer to the question
32
of which language
she usually speaks at home in the family, one participant emphasised that she attaches great
importance to Hochdeutsch (‘Standard German’), cf. (16).
33
30
Downes’s figure is not based on a specific phenomenon. It is rather an abstract summary of recurrent
findings on age-grading patterns. For single phenomena cf., e.g., Chambers and Trudgill (1998, 7879).
31
Note that language attitudes are one major research topic within our project (cf. Wiese et al. 2017). In
the near future, this topic will be treated more thoroughly and in much more detail.
32
This question was part of the questionnaire on biographical, social, and sociolinguistic data, which was
filled out by the participants who translated the “Wenker sentences”.
33
Although originally being a geographical designation in dialectology, Hoch- (‘high’) in Hochdeutsch is
usually interpreted as an evaluative term by linguistic lay people, meaning ‘valuable’.
19
(16) Question: Welche Sprache sprechen Sie meistens zu Hause, in der Familie?
(Which language do you usually speak at home, in the family?)
Answer: deutsch, achten sehr auf Hochdeutsch
(German, we attach great importance to Standard German)
In several interviews it was also evident that the German German standard variety is at least
traditionally considered a prestige variety by German-speaking Namibians. Cf., e.g., the
following statement by a German-speaking Namibian who I interviewed during her stay in
Germany in 2017:
34
(17) Interviewer: War das schon immer so, dass Sie Standarddeutsch und
Südwesterdeutsch sprechen konnten und dann hin und her wechseln konnten,
wie Sie wollten?
Interviewee: Jaja, wenn ich mir Mühe gebe, spreche ich schon anständiges
Deutsch […] Aber wenn meine Kinder jetzt hier sind, dann sprechen wir schon
Südwesterdeutsch.
Interviewer: ‘Has it always been the case that you were able to speak both
Standard German and Südwesterdeutsch and switch back and forth as you
wanted to?’
Interviewee: ‘Yes, if I make an effort I speak decent German […] But when my
children are here we speak Südwesterdeutsch.’
Interestingly, the interviewee paraphrased Standarddeutsch (‘Standard German’) as
anständiges Deutsch (‘decent German’). Furthermore she says that she is able to speak
Standard German if she makes an effort. Both aspects illustrate the prestige of the standard
variety. The interviewee also reports that her father demanded her and her siblings to speak in
such a way:
(18) Bei dem [ihrem Vater] mussten wir Hochdeutsch reden, ne. Der war richtig
giftig, wenn wir dann Südwester [sprachen]. Der hat schon drauf geachtet, ne.
Aber bei dem waren wir dann, weil meine Eltern geschieden waren wirklich
einmal im Jahr in den Ferien bei denen. Naja, hat nicht viel genützt.
34
Because the interviewees used the term “Südwesterdeutsch” throughout the interview I adapted this
term during the conversation with them.
20
‘We had to speak Hochdeutsch when we’ve been with him [her father]. He was
really resentful when we [spoke] Südwester. He attached importance to that. But
we were with him only once a year in the holidays because my parents were
divorced. Well, that didn’t help a lot.’
Apart from the narration of the normative pressure exerted by the father of the children, the
quote also contains an implicit evaluation of Hochdeutsch and Südwesterdeutsch. Claiming
that the normative pressure of the father did not help a lot (hat nicht viel genützt) because
father and children did not spend much time together implies that normative pressure from his
side, if exerted more regularly, would have helped them (to speak better).
These two quotes are quite representative with regard to all statements on this topic
made by the German-speaking Namibians we interviewed so far. Therefore, I assume that
generally − Standard German German can be considered a prestige variety within the German
speech community in Namibia. Given this, Downes’s graph (cf. Figure 5) and the data in the
Figure 1 Figure 3 strikingly resemble each other.
3.2.3 Possible explanations
The interrelation of age and the individually perceived importance of norms can be considered
as a basic explanation of this recurrent pattern:
“It is suggested that when speakers are young, the influence of the overt norms is likely
to be relatively weak. […] In their middle years, people’s lives tend to become more
public, and they have to adapt to the norms and values of the mainstream society. These
may be experienced as a result of pressures of work, personal independence or
geographic and/or social mobility, all of which lead to greater variability in social
relationships. For older, retired people, on the other hand, the pressures to conform to
societal norms may weaken once more” (Cheshire 2005, 1556)
The importance of (linguistic) norms in the professional life has repeatedly been identified as
one major reason for age-grading (cf. also, e.g., Mattheier 1980, 5255). In the special case of
German in Namibia, one has to take a closer look at the role of the language at the work place,
since German is usually not the (only) language used during working time. 43 speakers in our
data indicated that they use more than one language at work (= 47.8% of all participants who
pursue a profession), above all English, Afrikaans, and German. 56 participants do usually not
use German at all, 8 participants usually use only German at work. To examine whether
language use at the workplace influences the speaker’s closeness to Standard German, the
frequencies of loan words of the latter two groups were compared (only German vs. never
German at work, cf. Table 11).
21
German at work?
number of persons
number loan words
used
proportion of loan words
(vs. native German words)
only German
8
104
19.5%
no German
56
695
20.0%
Table 11: Interrelation of language at work and use of loan words
Even between these two groups, which represent the extreme positions with regard to German
at work, there is no significant difference concerning the use of loan words (p > 0.05). Hence,
it can be ruled out that German as used as a work place language directly influences the
willingness to use loan words. There is no direct link with the proportion of loan words in the
“Wenker translations: people who use only German at the work place and people who never
use German at work use loan words equally often in our data.
Another possible influence of work life on language use as an explanation for the
U-shape can be ruled out on the basis of the metadata as well. Since many German-speaking
Namibians have a tourism related occupation and Namibia is a popular destination for tourists
from Germany, there is a considerable amount of (language) contact between German
speaking Europeans and (some) Namibians. However, no direct influence can be attested in
our data, which is shown in Table 12. While speakers with a tourism related occupation do
use loan words a bit more often, this difference is not significant (p > 0.05).
Occupation
number of persons
number of loan
words used
proportion of loan words (vs.
native German words)
tourism related
12
124
18.0%
not tourism related
/ no occupation
114
1392
21.2%
Table 12: Interrelation of tourism related occupation and use of loan words
Hence, there is no obvious and direct impact of contact with German-speaking Europeans or
of German language use at work on the proportion of loan words used in the “Wenker
translations. Instead, age specific (maybe subconscious) language attitudes might determine
the way Namibian German is used. In particular, the group of 40-49 year old speakers seems
to be more conservative not only within the German-speaking community in Namibia, but
also in other speech communities worldwide (cf. Downes 1984, 19091, Cheshire 2005,
155556).
35
35
Of course, one has to keep in mind that we are dealing with translations and not with natural language
use. Therefore, it seems to be reasonable to compare the translations with natural language data as soon as our
corpus is available (cf. fn. 42). For example, it cannot be completely ruled out that single participants
exaggerated the use of Namibia specific words or structures to some extent. But even if that is the case, age
related differences are highly interesting from a sociolinguistic perspective: how do speakers of different ages
want their natural every-day language to be perceived?
22
Besides work-related aspects, of course also private developments come into play with
regard to the explanation of age-grading. Especially the birth of children can have an effect on
the individually perceived importance of linguistic norms. This was mentioned in several
interviews we conducted in the context of our data collection in 2017. Cf. (19), which is part
of an interview with a young couple.
(19) Interviewee I (female): Ich denk, die Einstellung ändert sich. Ich denk, als wir
in der Schule waren, wars einfach cool, anders zu reden und so viele Wörter und
Südwesterdeutsch wie möglich mit reinzuhauen und das ändert sich einfach.
Jetzt, wo ich mein Kind in der Schule hab und drauf achte, dass wir ein
saubereres Deutsch sprechen weil, wenn er es nicht von mir hört oder in der
Schule, wo hört er es dann? Das geht dann doch verloren. […]
Interviewer: Das heißt, Du hast früher anders gesprochen?
Interviewee I (female): Ja. [...] Meine Einstellung hat sich geändert und ich
denke, bei vielen ist es wahrscheinlich auch so. […]
Interviewer: Wie ist es bei Dir, [Interviewee II]? Hast Du auch, hat sichs auch
geändert im Vergleich zu Deiner Schulzeit zum Beispiel?
Interviewee II (male): Nee, ich glaub weniger.
Interviewee I (female): I think the attitude changes. I think it was cool to speak
differently, to use as many words as possible, and South-Westerners German
when we went to school. But this simply changes. Now my child goes to school
and it is important to me that we speak a cleaner German. If he [her son] does
not hear it from me or in school where does he hear it elsewhere? It will be lost
otherwise.
Interviewer: This means, you spoke differently in the past?
Interviewee I (female): Yes. […] My attitudes have changed and I think this
holds probably also true for many others.
Interviewer: What about you, [Interviewee II]? Did your speaking also change,
for example in contrast to your school days?
IntervieweeII (male): No, I don’t think so.’
Interestingly, the woman’s answers differ quite strongly from her husband’s statement: While
her attitudes and her speech style have changed, there has apparently not been such a change
in his way of thinking or speaking. This kind of (age related) gender difference seems to be
reflected in our data. Figure 6 depicts the interplay of age and gender with regard to the
amount of loan words used in the translations of the “Wenker sentences”.
23
Figure 6: Proportion of loan words in the translations of the “Wenker sentences”: The interplay of age
and gender
While the differences between male and female participants are marginal in most age groups,
there are strong differences in the group of 20‒29 year old participants and, even more
pronounced, in the bracket of 30‒39 year old informants. Obviously, this has to be seen in the
context of parenting. Typically, children are mainly raised by their mothers in the speech
community at hand and parenting can have the effect described by the female interviewee in
(19). Simultaneously, males in the relevant age group do not tend to speak “clearer” German
than in their school days. Quite to the contrary: in our data they even use significantly more
loan words. This might be explained as some kind of countermovement: Males might want to
demarcate themselves from females, who change their linguistic attitudes and who approach
standard (German) German. This fits quite nicely with some remarks of male interviewees.
For example, one 31 year old informant remarked that it sounds “verweiblicht (‘effeminate’)
if someone uses the German German word grillen (‘to barbecue’) instead of the Namibian
German braai, which is an Afrikaans loan word.
3.2.3 Further evidence: intensifiers and Afrikaans vs. English loan words
As I have shown above, older and younger participants, in general, translated the “Wenker
sentences in a similar way. This outcome is based on the study of 1:1 substitutions of lexical
items. In addition, words that were added in the translation and do not have a counterpart in
the original sentence also suggest this general claim. Especially intensifiers were often added
to the original linguistic material, as illustrated in (20) (cf. also (14) above).
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
55%
60%
65%
<20 20-29 30-39 40-49 50-59 60 +
Proportion of loan words:
The interplay of age and gender
m
f
24
(20) Wir sind müde und haben Durst. (“Wenker sentence” 23)
Wir sind vrek moeg und haben Durst.
‘We are (bloody) tired and thirsty.
The intensifiers bleddie (‘bloody’), flippen (probably stemming from flippant), focken
(probably stemming from fuck), heavy, kak (‘shit’), and vrek (literally perished’) or
graphemic variants thereof, which have to be considered non-standard, were used cross-
generationally in the “Wenker translations as is depicted in Table 13. Thus, they cannot be
regarded as a feature of Namibian German youth language.
36
age group
number of intensifiers
(tokens)
number of persons who used
at least one intensifier
< 20
8
7
20-29
4
2
30-39
8
4
40-49
2
1
50-59
5
4
60 +
9
7
Table 13: Intensifiers and age
Apart from the use of intensifiers, I checked another conceivable characteristic of Namibian
German as used by younger speakers: several informants told us that one very significant
difference between the Namdeutsch of older and younger speakers is the amount of English
vs. Afrikaans loan words. This is also what one could expect because of recent developments
in the country: in the wake of Namibian independence from the Republic of South Africa in
1990, English has increasingly gained importance. It has been chosen as the official language
by the independent Namibian government resulting in strong institutional support of this
language (cf. Shah and Zappen-Thomson 2018). Ever since, the importance of Afrikaans as
the major inter-ethnic lingua franca declines. Therefore, it seems to be quite plausible that
younger speakers use more loan words from English than the older generations do (cf. Shah
2007, 43, Kellermeier-Rehbein 2016, 230). However, at least in the corpus data, there is no
evidence for such a development, as can be seen in Figure 7Figure 8.
37
36
The absolute numbers in Table 13 are informative only to a limited degree since there are not equally
many speakers per age group. However, the lowest number of intensifiers in the age group of 40 to 49 year old
participants fits very well with what has been observed concerning the conservativeness of this group.
37
Words which could be borrowed either from Afrikaans or from English (cf. also fn. 15) were excluded
from these overviews.
25
Figure 7: English vs. Afrikaans loan words and age (tokens)
Figure 8: English vs. Afrikaans loan words and age (types)
Why can’t we find the assumed proportional increase of English loan words in our data?
Maybe both older and younger speakers use fewer Afrikaans loan words nowadays due to a
conscious decision in connection with (changed) language attitudes, while there is still a high
amount of Afrikaans loan words in every generation. Another possible explanation for the
constancy of Afrikaans loan words in all generations might be that we are dealing with an
artefact of the design: maybe Afrikaans words are considered particularly salient and are used
in the translations to emphasise the deviance of Namdeutsch from European German. Or,
there is indeed no decrease of Afrikaans loan words in relation to borrowings from English in
the younger generations, and language production and language perception diverge in this
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
< 20 20-29 30-39 40-49 50-59 60 +
Afrikaans vs. English loan words (tokens)
Afrikaans English
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
< 20 20-29 30-39 40-49 50-59 60+
Afrikaans vs. English loan words (types)
Afrikaans English
26
respect. To date, there is no way to answer this interesting question conclusively, highlighting
the need for a comprehensive corpus of Namdeutsch.
38
In sum, we did not find any evidence for a stronger manifestation of relexification in the
younger generations.
39
In this respect, Namdeutsch and Texas German resemble each other as
there have been comparatively few changes in the Texas German lexicon over the past four
decades (Boas and Pierce 2011, 147). Both Namdeutsch and Texas German are not
characterised by imperfect acquisition (cf. also Yager et al. 2015), which might explain why
neither lexicon seems to be affected by lexical erosion (cf. Boas and Pierce 2011, 14547).
Beyond this general picture, it is difficult to compare Texas German and Namdeutsch.
Today, younger generations do not have any competence in Texas German. Hence, a fine-
grained analysis of age-grading features as presented here for Namdeutsch can nowadays
hardly be done for Texas German (and many other German language islands). But, as I have
shown, comparisons with internal varieties can be promising and reveal remarkable parallels
with language islands, such as the U-shape pattern characteristic of age variation.
4. Conclusion
In the corpus data I analysed here, there are several specifics which set Namdeutsch apart
from other varieties of German, such as words with clicks (e.g. nxa, cf. p. 6) and the high
amount of lexical borrowings from Afrikaans. Nevertheless, I have identified patterns and
properties that are also detectable in other varieties, both language islands and internal
varieties. First of all, the pattern revealed for the interrelation of age and loan word use
strikingly resembles the normal distribution of stable sociolinguistic variation in internal
varieties (Downes 1984, 190). At first sight, it might not seem to be very exciting that a
speech community follows the normal pattern. However, the observation that a language
island, especially a rather small one, follows this pattern is remarkable. The small speech
community obviously behaves similarly to others. What we find in Namdeutsch is not merely
random variation caused by intense language contact, but rather variation structured according
to cross-linguistic principles. In addition, it is noteworthy that there is sociolinguistic variation
at all, which I interpret as evidence for the vitality of the language island. That there is no
38
Within our research project on Namibian German, we will compile a systematic corpus of production
data. This corpus will contain free conversations, elicited language data from different registers, and
sociolinguistic interviews. Once completed, the corpus will be freely available (cf. Wiese et al. 2017).
39
Note that our participants translated sentences from Standard German into their vernacular usage.
Hence, the participants adapted the original sentence to their active language use in informal situations. This
means that they had to know the Standard German word to be able to select the Namdeutsch counterpart. In
contrast to this, unknown words might have been taken over from the originals, and there might hence have been
Standard German words that the participants did not know at all. But, as most members of the community
possess a thorough knowledge of Standard German, I assume that all or the vast majority of words were known.
Note also that no participant pointed out that s/he does not know a word in the original sentences, while they did
comment on contents they found unsuitable for a Namibian context (see above). This is also supported by the
translations of the participants, all of which make sense in terms of contents.
27
evidence for a strong tendency to relexification also shows that we are dealing with a
language island that is small but not moribund. In my study, this finding was further
supported by a comparison of Texas German and Namdeutsch, which revealed that the
morphological system of Namdeutsch is much more stable than its Texas German counterpart.
While the developments in Texas German have been interpreted as “characteristic of language
death by Boas (2009, 232), there is no evidence for an increase of non-standard plural
markers in Namdeutsch, and even loan words such as Farmer are used according to the
German plural system.
Moreover, the loan word use in our data does not support the idea of a clear-cut
dichotomy between a youth language (“Nam-Släng”) and a standard variety closer to German
German that is used by older speakers. Instead, we find evidence for a cross-generational use
of Namibia-specific loan words. This corresponds nicely with what many interviewees said
during the group interviews on metalinguistic topics conducted by Heike Wiese in connection
with fieldwork done in Namibia in 2013. The adolescent participants were asked what their
parents think of Namlish/Nam-Släng/slang.
40
(21) Meine Ma, die redet auch so. Also wenn die mit uns Jugendlichen hier
irgendwo ist, dann redet die auch mit uns im Slang.
‘My mum also talks like this. Well, when she is with us adolescents here
anywhere, then she talks slang with us.’
(22) Interviewer: Wie finden eure Eltern Namlish?
Interviewee I: Meine Mutter spricht das auch.
Interviewee II: Meine Eltern auch.
‘What do your parents think of Namlish?
My mother speaks it as well.
My parents, too.’
(23) Die [her mother] gebraucht nicht hochdeutsche Wörter, […] die gebraucht
auch nicht Namlish, sondern normal Deutsch.
‘She [her mother] doesn’t use Standard German words, […] she also does not
use Namlish, but normal German.’
40
Note that Namlish is the term many participants used to refer to German with many Namibia-specific
characteristics. Therefore, H. Wiese used it in the interviews, too. Besides this usage, Namlish is often used to
refer to English with Namibia-specific characteristics (Namlish is a blend consisting of Namibian and English).
The participants seem to have generalised the term and use it with the meaning ‘originally European language
with many Namibia-specific characteristics’). The names used to refer to Namibian-style language use will be
studied in detail within our project.
28
Especially the statement in (23) is remarkable with regard to the organisation of varieties of
German as used in Namibia because neither Standard German nor slangare considered to
be “normal German” but obviously something else (something in between?). This supports
the idea that a simple dichotomy of Standard German vs. Nam-släng would force Namibian
German language use into two rather inadequate categories. By analysing in detail the age of
speakers as a factor determining variation, I aimed to contribute to a more fine-grained
description of Namibian German, which is necessary to understand this particular language
island and sheds new light on the interplay of language contact, variation, and change.
Still, we are only beginning to understand the nature of Namibian German. Our method
enabled us to collect enough data to analyse the interrelation of loan word use and age of
speakers quantitatively, which revealed some interesting insights. At the same time, this
method also has some obvious limitations. For example, I analysed translations and not
natural language use and we were able to elicit the data only in writing, highlighting the need
for further research in the field.
References
Ammon, Ulrich. 2014. Die Stellung der deutschen Sprache in der Welt. Berlin: de Gruyter.
Bailey, Guy. 2002. ‘Real and Apparent Time’. In The Handbook of Language Variation and
Change, edited by J. K. Chambers, Peter Trudgill, and Natalie Schilling-Estes, 312
32. Malden, MA: Blackwell.
Berend, Nina, and Elisabeth Knipf-Komlosi, eds. 2006. Sprachinselwelten The World of
Language Islands. Frankfurt: Lang.
Boas, Hans C. 2003. ‘Tracing Dialect Death: The Texas German Dialect Project’. In
Proceedings of the 28th Annual Meeting of the Berkeley Linguistics Society, edited by
Julie Larson and Mary Paster, 38798. Berkeley, California: Berkeley Linguistics
Society.
———. 2009. The Life and Death of Texas German. Publications of the American Dialect
Society 93. Durham, NC: Duke University Press.
Boas, Hans C., and Marc Pierce. 2011. ‘Lexical Developments in Texas German’. In Studies
on German Language Islands, edited by Michael Putnam, 12950. Studies in
Language Companion Series 123. Amsterdam & Philadelphia: Benjamins.
Bornschein, Matthias, and Matthias Butt. 1987. ‘Zum Status des s-Plurals im gegenwärtigen
Deutsch’. In Linguistik in Deutschland: Akten Des 21. Linguistischen Kolloquiums,
Groningen 1986, edited by Werner Abraham and Ritva Århammar, 13553.
Linguistische Arbeiten 182. Tübingen: Niemeyer.
Bybee, Joan L. 1985. Morphology: A Study of the Relation between Meaning and Form.
Typological Studies in Language 9. Amsterdam & Philadelphia: Benjamins.
———. 1994. ‘Morphological Universals and Change’. In The Encyclopedia of Language
and Linguistics, edited by Ron E. Asher, 255762. Oxford et al.: Pergamon Press.
Chambers, J. K., and Peter Trudgill. 1998. Dialectology. 2nd ed. Cambridge & New York:
Cambridge University Press.
29
Cheshire, Jenny. 2005. ‘Age- and Generation-Specific Use of Language / Alters- Und
Generationsspezifischer Sprachgebrauch’. In Sociolinguistics: An International
Handbook of the Science of Language and Society / Soziolinguistik: Ein
Internationales Handbuch Zur Wissenschaft von Sprache Und Gesellschaft, edited by
Ulrich Ammon, Norbert Dittmar, Klaus J. Mattheier, and Peter Trudgill, 2nd ed.,
2:155263. Handbücher Zur Sprach- Und Kommunikationswissenschaft / Handbooks
of Linguistics and Communication Science (HSK) 3. Berlin & New York: de Gruyter.
Clahsen, Harald. 1999. ‘Lexical Entries and Rules of Language: A Multidisciplinary Study of
German Inflection’. Behavioral and Brain Sciences 22: 991060.
Dammel, Antje, and Melitta Gillmann. 2014. ‘Relevanzgesteuerter Umbau der
Substantivflexion im Deutschen’. Beiträge zur Geschichte der deutschen Sprache und
Literatur (PBB) 136: 173229.
Deumert, Ana. 2009. ‘Namibian Kiche Duits: The Making (and Decline) of a Neo-African
Language’. Journal of Germanic Linguistics 21: 349417.
Donaldson, Bruce C. 1993. A Grammar of Afrikaans. Mouton Grammar Library 8. Berlin &
New York: Mouton de Gruyter.
Downes, William. 1984. Language and Society. London: Fontana.
Eichinger, Ludwig, Albrecht Plewina, and Claudia Maria Riehl, eds. 2008. Handbuch der
deutschen Sprachminderheiten in Mittel- und Osteuropa. Tübingen: Gunter Narr.
Fleischer, Jürg. 2015. ‘Pro-Drop und Pronominalenklise in den Dialekten des Deutschen: Eine
Auswertung von Wenkersatz 12’. In Deutsche Dialekte: Konzepte, Probleme,
Handlungsfelder. Akten des 4. Kongresses der Internationalen Gesellschaft für
Dialektologie des Deutschen (IGDD), edited by Michael Elmentaler, Markus Hundt,
and Jürgen Erich Schmidt, 191209, 504505 [maps]. Zeitschrift für Dialektologie
und Linguistik Beihefte 158. Stuttgart: Steiner.
———. 2017. Geschichte, Anlage und Durchführung der Fragebogen-Erhebungen von
Georg Wenkers 40 Sätzen: Dokumentation, Entdeckungen und Neubewertungen.
Deutsche Dialektgeographie 123. Hildesheim, Zürich & New York: Olms.
Fleischer, Jürg, Roland Hinterhölzl, and Michael Solf. 2008. ‘Zum Quellenwert des
Althochdeutschen Tatian für die Syntaxforschung: Überlegungen auf der Basis von
Wortstellungsphänomenen’. Zeitschrift für germanistische Linguistik 36: 21039.
Gaeta, Livio. 2008. ‘Die Deutsche Pluralbildung zwischen deskriptiver Angemessenheit und
Sprachtheorie’. Zeitschrift für germanistische Linguistik 36: 74108.
Gilbert, Glenn G. 1972. Linguistic Atlas of Texas German. Austin: Univ. of Texas Press.
Gretschel, Hans-Volker. 1995. ‘The Status and Use of the German Language in Independent
Namibia: Can German Survive the Transition?’ In Discrimination through Language
in Africa? Perspectives on the Namibian Experience, edited by Martin Pütz, 299313.
Berlin & New York: de Gruyter.
Haspelmath, Martin. 2009. ‘Lexical Borrowing: Concepts and Issues’. In Loanwords in the
World’s Languages. A Comparative Handbook, edited by Martin Haspelmath and Uri
Tadmor, 3554. Berlin: De Gruyter Mouton.
Johannessen, Janne Bondi, and Joseph C. Salmons, eds. 2015a. Germanic Heritage
Languages in North America: Acquisition, Attrition and Change. Studies in Language
Variation 18. Amsterdam & Philadelphia: Benjamins. https://doi.org/10.1075/silv.18.
———. 2015b. ‘The Study of Germanic Heritage Languages in the Americas’. In Germanic
Heritage Languages in North America: Acquisition, Attrition, and Change, edited by
Janne Biondi Johannessen and Joseph C. Salmons, 117. Studies in Language
Variation 18. Amsterdam & Philadelphia: Benjamins.
Keel, William, and Klaus Mattheier, eds. 2003. German Language Varieties Worldwide:
Internal and External Perspectives. Frankfurt: Peter Lang.
30
Kellermeier-Rehbein, Birte. 2015. ‘Namslang ‒ Deutsche Jugendsprache in Namibia?’ In
Sprachvariation und Sprachreflexion in interkulturellen Kontexten, edited by Corinna
Peschel and Kerstin Runschke, 4162. Frankfurt am Main: Lang.
———. 2016. ‘Sprache in postkolonialen Kontexten. Varietäten der deutschen Sprache in
Namibia’. In Sprache und Kolonialismus, edited by Thomas Stolz, Ingo H. Warnke,
and Daniel Schmidt-Brücken, 21334. Berlin & Boston: de Gruyter.
———. 2018. Deutsch in Namibia Ergebnis von Sprach- und Varietätenkontakt.
Presentation given at the conference Variationist Linguistics meets Contact linguistics,
Ascona, May 22, 2018.
Kerswill, Paul. 1996. ‘Children, Adolescents, and Language Change’. Language Variation
and Change 8: 177202.
Lameli, Alfred. 2008. ‘Was Wenker noch zu sagen hatte... Die unbekannten Teile des
“Sprachatlas des Deutschen Reichs”’. Zeitschrift für Dialektologie und Linguistik 75:
25581.
Louden, Mark L. 2016. Pennsylvania Dutch: The Story of an American Language. Baltimore:
Johns Hopkins University Press.
Mattheier, Klaus J. 1980. Pragmatik und Soziologie der Dialekte. Einführung in die
kommunikative Dialektologie des Deutschen. Uni-Taschenbücher 994. Heidelberg:
Quelle und Meyer.
Nübling, Damaris, and Mirjam Schmuck. 2010. ‘Die Entstehung des s-Plurals bei
Eigennamen als Reanalyse vom Kasus- zum Numerusmarker. Evidenzen aus der
deutschen und niederländischen Dialektologie’. Zeitschrift für Dialektologie und
Linguistik 77: 145182.
Poplack, Shana. 2012. ‘What Does the Nonce Borrowing Hypothesis Hypothesize?’
Bilingualism: Language and Cognition 15: 64448.
Pütz, Martin. 1991. ‘“Südwesterdeutsch” in Namibia: Sprachpolitik, Sprachplanung und
Spracherhalt’. Linguistische Berichte 136: 45576.
Salmons, Joseph. 1994. ‘Naturalness & Morphological Change in Texas German’. In
Sprachinselforschung. Eine Gedenkschrift Für Hugo Jedig, edited by Nina Berend and
Klaus J. Mattheier, 5972. Frankfurt am Main [et al.]: Lang.
Sell, Eric E. 2011. Esisallesoreidt. NAM Släng - Deutsch. Vol.2. Windhoek: EES Records.
Shah, Sheena. 2007. ‘German in a Contact Situation: The Case of Namibian German’. eDUSA
2: 2045.
Shah, Sheena, and Marianne Zappen-Thomson. 2018. ‘German in Namibia’. In Heritage
Language Policies around the World, edited by Corinne A. Seals and Sheena Shah,
12847. Abingdon & New York: Routledge.
Tagliamonte, Sali A., and Alexandra D’Arcy. 2009. ‘Peaks beyond Phonology: Adolescence,
Incrementation, and Language Change’. Language 85: 58108.
Wegener, Heide. 2004. Pizzas und Pizzen Die Pluralformen (un)assimilierter Fremdwörter
im Deutschen’. Zeitschrift für Sprachwissenschaft 23: 47112.
Wiese, Heike, Horst J. Simon, Marianne Zappen-Thomson, and Kathleen Schumann. 2014.
Deutsch im mehrsprachigen Kontext: Beobachtungen zu lexikalisch-grammatischen
Entwicklungen im Namdeutschen und im Kiezdeutschen’. Zeitschrift Für
Dialektologie und Linguistik 81: 274307.
Wiese, Heike, Horst J. Simon, Christian Zimmer, and Kathleen Schumann. 2017. ‘German in
Namibia: A Vital Speech Community and Its Multilingual Dynamics’. Language &
Linguistics in Melanesia, 221245.
Yager, Lisa, Nora Hellmold, Hyoun-A Joo, Michael Putnam, Eleonora Rossi, Catherin
Stafford, and Joseph C. Salmons. 2015. ‘New Structural Patterns in Moribund
Grammar: Case Marking in Heritage German’. Frontiers in Psychology 6: 1716.
31
Zimmer, Christian. in press a. ‘Deutsch als Minderheitensprache in Afrika’. In Sprache und
Raum Deutsch. Ein internationales Handbuch der Sprachvariation, edited by
Joachim Herrgen and Jürgen Erich Schmidt. Handbücher zur Sprach- und
Kommunikationswissenschaft, 30.4. Berlin: De Gruyter Mouton.
———. in press b. ‘Zweifel bei der Flexion von Fremdwörtern: Morphologische Integration
und Variation’. In Zweifelsfälle: Definition, Erforschung, Implementierung, edited by
Renata Szczepaniak, Eleonore Schmitt, and Annika Vieregge. Germanistische
Linguistik. Hildesheim, Zürich & New York: Olms.
———. 2018. Die Markierung Des Genitiv(s). Empirie und theoretische Implikationen von
morphologischer Variation. Reihe Germanistische Linguistik 315. Berlin & Boston:
de Gruyter.
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Among (post-)colonial varieties of German, Namibian German is a particularly interesting case. It has a unique status compared to the other extra-territorial varieties as well as to those in the German-speaking area in Europe. First, it is based on a speech community with German ancestry who still live in Namibia today, which distinguishes it from such colonial varieties as Unserdeutsch in the South Pacific and makes it more similar to such German “language island” varieties as, e.g., Texas German in the United States or the German varieties still spoken in Brazil. Second, though, unlike language island varieties as well as other postcolonial varieties and more similar to those in Germany, Namibian German is linguistically vital. It is passed on to younger generations and is also used in public domains, supporting, e.g., register differentiation. Third, unlike most varieties in Germany, however, it is integrated in a setting of societal multilingualism, with speakers who routinely use two or more languages in addition to German in their daily lives, and with a broader context of high linguistic diversity, offering a wealth of language contact opportunities. In this paper, we describe this special status of Namibian German and present first results from a project that capitalises on this to investigate the (socio-)linguistic dynamics that this setting supports, affording us a spotlight on tendencies of language attitudes and language variation in contact situations of German.
Article
During the last decades, “natural” has often been used by linguists in an inductive or even anecdotal way as a synonym of “intuitively plausible” or of “cross‐linguistically frequent,” in reference to both synchrony and diachronic change. In more theoretical views, it often overlaps with cognitively simple (cf. Anttila, this volume), elementary and therefore universally preferred, and with Praguian (especially Jakobson's) notions of markedness (where unmarked loosely corresponds to natural).
Book
While most world languages spoken by minority populations are in serious danger of becoming extinct, Pennsylvania Dutch is thriving. In fact, the number of Pennsylvania Dutch speakers is growing exponentially, although it is spoken by less than one-tenth of one percent of the United States population and has remained for the most part an oral vernacular without official recognition or support. A true sociolinguistic wonder, Pennsylvania Dutch has been spoken continuously since the late eighteenth century, even though it has never been “refreshed” by later waves of immigration from abroad. In this probing study, Mark L. Louden, himself a fluent speaker of Pennsylvania Dutch, provides readers with a close look at the place of the language in the life and culture of two major subgroups of speakers: the “Fancy Dutch," whose ancestors were affiliated mainly with Lutheran and German Reformed churches, and conservative Anabaptist sectarians known as the “Plain people”-the Old Order Amish and Mennonites. Drawing on scholarly literature, three decades of fieldwork, and ample historical documents-most of which have never before been made accessible to English-speaking readers-this is the first book to offer a comprehensive look at this unlikely linguistic success story.
Article
Based on corpus data and the results of a linguistic survey, we will examine grammatical and lexical developments in two relatively new varieties of Modern German: Kiezdeutsch and Nam- deutsch. Both varieties are spoken in multilingual speech communities in Germany and Namibia respectively. In spite of considerable differences in the contact languages involved, the social backgrounds of the speakers and other differences, our comparison shows that the inovations have a series of conspicuous similarities. For example, gibs has developed into an element similar to a particle indicating existance and so functions as a focus marker. New particles are also being borrowed. Furthermore, in the course of a semantic-pragmatic development, the article in (ein) bisschen or bietje/bikkie has been gradually disappearing. In comparison to the German spoken in monolingual contexts, the relatively strong changes in Kiezdeutsch and Namdeutsch bear witness to the highly innovative power of these varieties. In addition, the parallelism of the developments in the domains studied here indicates fundamental tendencies of the internal structure of German.