Sorbian is a West Slavic language spoken in eastern Germany, in Saxony and Brandenburg near the borders of Poland and the Czech Republic, and is recognized as an endangered language by UNESCO (Moseley Reference Moseley2012). It is commonly referred to as Sorbian in English, but has historically been referred to as both Wendish and Lusatian. The Sorbian speech area used to expand from its northernmost point approximately 50 km south-east of Berlin to its southernmost point approximately 8 km from the borders of the Czech Republic (Stone Reference Stone, Comrie and Corbett1993). This area is also referred to as Lusatia (Figure 1). However, the Sorbian-speaking area continues to shrink every year and is currently much smaller than Stone (Reference Stone, Comrie and Corbett1993) describes. Upper Sorbian is currently only used in daily communication in and immediately around Budyšin (personal communication, Lechosław Jocz).
Location of the Upper Sorbian and intermediate dialect speaking area. The thick black line marks the dialect border between the intermediate dialects (the northern area) and the Upper Sorbian speaking area (the southern area). The dotted box represents the Lower Sorbian speaking area.
Sorbian is divided into two main branches, Upper and Lower Sorbian, and each is recognized as a different and distinct language in the European Charter for Regional or Minority Languages (Dołowy-Rybińska Reference Dołowy-Rybińska2011). However, they are often considered dialects of the same language for social and political reasons (Personal Communication, Lechosław Jocz). Upper Sorbian is spoken in the south and is more closely related to Czech. A generous estimation of the number of Upper Sorbian speakers rests at approximately 20,000–25,000. However, less than half use the language daily (Šatava Reference Šatava2005). There is also an area north of the Upper Sorbian-speaking area, which has a number of ‘intermediate’ dialects which are hard to classify as either Upper or Lower Sorbian. This area covers the area between Grodk (German: Spremberg) in the north and Wojerecy (German: Hoyerswerda) in the south (Sussex & Cubberley 2006). Slepo Sorbian belongs to this area and is spoken in and around the area of the Slepo village. There are currently approximately 50 speakers of Slepo Sorbian. The vast majority of Slepo Sorbian speakers are elderly, aged 80 and above, while there is a much larger number of young native Upper Sorbian speakers.
The Sorbian linguistic speech area is, however, not homogeneous. The majority of the Sorbian-speaking area is predominantly populated by German-speaking people, making Sorbian speakers a minority even in Lusatia (Marti Reference Marti2007). Only approximately 1% of the inhabitants of Lusatia are Sorbian speakers (Elle Reference Elle, Huss, Camilleria Gima and King2003). This has made the maintenance of the Sorbian language difficult, even though it is protected under the European Charter for Regional or Minority Languages (First Report 2000: Annex 3). This problem has been compounded by the development of the lignite mining industry which has caused the displacement of many Sorbian people (Mellor Reference Mellor2008).
There are a number of Upper Sorbian schools in Budyšin (German: Bautzen) where students go to learn literature, history and civil studies. The primary Sorbian schools teach all subjects in Sorbian, and German is taught as a ‘second mother tongue’; in the larger bilingual area of Saxony, there are schools where Sorbian is taught as a language from grade one (Elle Reference Elle, Huss, Camilleria Gima and King2003).
The present description is based on my fieldwork. The word list and recorded text provided with this paper were recited by a twenty-two-year-old female speaker of Upper Sorbian, who was born in Worklecy (German: Räckelwitz). However, variations of some pronunciations are produced by three other speakers of Upper Sorbian (ages 20, 23 and 35 years). Data for the Upper Sorbian speakers was recorded at the Max Planck Institute for Evolutionary Anthropology in a sound attenuated booth with an Audix HT5P. ‘The North Wind and the Sun’ was translated by Expert Language Translations.
Consonants
The consonant inventory of Upper Sorbian is comprised of 29 consonants. The labels ‘hard’ and ‘soft’ are convenient shorthand for unpalatalized and palatalized, respectively.
There is a two-way contrast in stops: voiced and voiceless. The voiced stops typically have voicing during the closure, but this is not always the case (e.g. baba [
a
b
a]; [b
a
b
a] ‘midwife, pound cake’). Both voiced and voiceless stops may have aspiration (e.g. chabłak [k
h
a
b
h
w
a
k
h
] ‘waverer’). It should also be noted that in Upper Sorbian, the phonemic status of /ɡ/ is marginal, only present in loans and onomatopoeia. Jocz (Reference Jocz2011) also describes /f
v/ as problematic with respect to their phonemic status. This is largely based on the fact that they typically only appear in loan words or in complementary distribution with opposing phonemes. The uvular trill /ʀ/ (Figure 2) does have a phonemically contrastive palatalized counterpart /ʀ
j
/ (Figure 3). Where /ʀ/ and /ʀ
j
/ are realized as a trill, they both typically have 1–2 contacts between the tongue and the uvula. However, both are often produced as an approximant [ʁ] (e.g. rat [ʁa
t] ‘good’) and [ʁ
j
] (e.g. rjadka [ʁ
j
a
t
k
a] ‘line, verse’). /ʀ/ is also frequently realized as either a voiced or a voiceless uvular fricative, [ʁ] (e.g. rat [ʁa
t] ‘gladly’) and [χ] (e.g. rat [χa
t] ‘gladly’), respectively. Jocz (Reference Jocz2013) notes that /ʀ/ and /ʀ
j
/ are more likely to be realized as a trill in clear and careful pronunciation. The rhotic is also occasionally dropped from speech completely, leaving only the glide (e.g. rjadka [j
a
t
k
a] ‘line, verse’). The phonemic status of /ʀ
j
/, however, is somewhat contentious among Sorbian linguists; it has been described as a phonemically palatalized uvular trill by some researchers (Šewc Reference Šewc1984), while others have stated that it has undergone depalatalization and is now biphonemic, /ʀj/ (Wornar Reference Wornar and Sawicka2007, Jocz Reference Jocz2013). The phoneme /x/ is often realized as a voiceless trill with noticeable frication, which typically has 2–3 contacts between the tongue and the uvula (e.g. brach [bʁa
] ‘fault, defect’).
A spectrogram of an US speaker's articulation of /ʀ/ from rad ‘gladly’.
A spectrogram of an US speaker's articulation of /ʀ j / from rjadka 'line, verse’.
F1 and F2 values for the Upper Sorbian monophthongs. Measurements were taken from six speakers of US (three male and three female). The data was normalized using a Bark correction (Traunmüller Reference Traunmüller1990). Plots were made in R (R Development Core Team 2013) using the phonR package (McCloy Reference McCloy2012).
Vowels
The vowel system in Upper Sorbian is often described as having eight vowels, /i ɪ e ɨ a
u ʊ o/ (Pianka & Tokarz Reference Pianka and Tokarz2000), but it has also been described as having seven vowels, /i ɪ e
a
u ʊ o/ (Šewc Reference Šewc1966, Reference Šewc1984; Faßke Reference Faßke1990), and ten vowels, /i ɪ ɛ e ɨ a
u ʊ o ɔ/ (Jocz Reference Jocz2011). The monopthongs /ɪ ʊ/ are perhaps better thought of as allophones of the diphthongs /i
ɪ/ (e.g. pěc [p
j
i
ɪ
t
s] ‘stove’) and /u
ʊ/ (e.g. ból [b
u
ʊ
l] ‘pain, hurt’). Jocz (Reference Jocz2011) notes that the former typically only surfaces as a diphthong under strong sentential stress in monosyllabic words. However, /u
ʊ/ is commonly realized as a diphthong with a characteristic rise in F2 throughout articulation. The F2 increase is relatively small. /u
ʊ/ can be thought of as a diphthongoid. Upper Sorbian has both onglides and offglides /j/ (e.g. beja [b
e
j
a] ‘bitch’) and /w/ (e.g. błazny [b
w
a
z
nɨ] ‘fool’), which are often transcribed by Sorbian linguists as [
] and [
], respectively.
The discrepancy in the interpretation of the vowel inventory arises largely from the phonological distribution of each vowel. The palatalization contrast yields a complementary distribution for /i ɨ e ɛ/. The vowels /i e/ are realized following palatalized consonants, while /ɨ ɛ/ are realized following unpalatalized consonants. /o ɔ/ are also problematic because they have a complementary distribution in most traditional descriptions. /o/ is typically realized before /w/ and in some dialects, before labials and velars. However, Jocz (Reference Jocz2011) notes that in normal speech, /o/ is realized without any /w/ (e.g. bowa [b o a] ‘bucket (gen)’).
Suprasegmentals
In general, primary stress always falls on the first syllable (e.g. wysoke [ˈwɨs
o
kə] ‘high neut’.) and its phonetic realization is a rising or higher pitch on the stressed syllable. There are a few exceptions to stress placement in Sorbian: stress can fall on a syllable besides the first in loan words (e.g. uniwersita [u
n
iˈʋeɐs
i
t
a] ‘University’); there are also a small number of indigenous compounds which bear stress on a different syllable (e.g. bohužel [b
o
h
uˈʒe
l] ‘unfortunately’), although stress often appears on the first syllable in these compounds as well (e.g. bohužel [ˈb
o
h
uʒe
l] ‘unfortunately’). Secondary stress is also assigned to the penultimate syllable in words with four or more syllables (e.g. beletrija [ˈb
e
l
eˌt
h
i
a] ‘fiction’). Secondary stress is realized as a lengthening of the vowel (Mucke Reference Mucke1965).
Palatalization
Upper Sorbian has undergone a palatal nasal decomposition (Rubach Reference Rubach2008) of /n j / into /j n/ in word-final position (e.g. kóń [k u j n] ‘horse’) and has lost the palatalization contrast in word-final and preconsonantal positions. The palatalization opposition now only occurs before the vowels /a o u/. Consonants are palatalized before /i/ (e.g. ličak [l j iʧa k h ] ‘calculator’), /e/ (e.g. wón so lehnje [w u ʊ n s o l j e n j ə] ‘(he) lies’), and /ɪ/ (e.g. bělić so [b j ɪl j iʧ s o] ‘to peel’), while unpalatalized consonants appear before /ɨ ɛ/. However, Jocz (Reference Jocz2013) notes that palatalization before /i e ɪ/ is now optional.
The phonemic status of palatalization in Upper Sorbian is an area of debate among modern Sorbian linguists. Jocz (Reference Jocz2015) argues that phonemic palatalization is no longer present in Upper Sorbian. He notes that there is a clear glide following palatalized consonants and that the consonant does not have to be palatalized at all in these situations. There are also strong distributional restrictions on the palatalized consonants: the contrast is neutralized before the vowels /i ɪ ɨ ɛ/ and there is also no contrast before the glide /j/, as in Russian. Furthermore, the palatalization contrast would make Upper Sorbian a typologically unusual language because of the presence of /ʀ j /, yet not /s j / (/ʀ j / is undescribed in other languages as a phoneme, but /s j / is cross-linguistically common). Finally, palatalized phonemes are rare in word roots, but are quite common at morpheme and word boundaries.
Transcription of the recorded passage
Orthographic transcription
Něhdy raz so sewjerny wětr a słónco wadźeštej, štó z njeju je sylniši. Po puću dźěše pućowar, kotryž bě do ćopłeho płašća zababjeny. Wětr a słónco sej praještaj: Sylniši a dobył je tón, kotryž dočini, zo so pućowar swój płašć sleče. Sewjerny wětr ze wšej mocu duješe. Ale čim bóle duješe, ćim krućišo so pućowar do swojeho płašća zababi. Bórze bě sewjerny wětr bjez dycha. A nětko słónco ze swojimi přećelnymi pruhami powětr hrěješe. A hlej, hižo bórze so pućowar swój płašć sleče. Sewjerny wětr nětko widźeše, zo bě słónco sylniše hač wón.
Semi-narrow transcription
Acknowledgements
I would like to thank Dr. Kochetov for being helpful and supportive throughout the planning stages of this project, his advice made this possible. I would also like to thank Prof. Dr. Wornar, Dr. Wölke and the Sorbian Institute in Budyšin for helping me find native speakers of Sorbian and accommodating my needs. Prof. Dr. Jocz was also helpful in finding participants and directing me to useful articles about Sorbian. I would also like to thank the director at the Max Planck Institute for Evolutionary Anthropology (MPI: EVA), Prof. Dr. Comrie and the researchers at MPI: EVA, Dr. Lancia and Dr. Grawunder, for making the Phonetics Laboratory available during my stay there. I would like to thank all the participants for helping make this project a reality. Finally, I would like to thank the reviewers for their helpful comments. This work was funded by the Germany/Europe Fund granted by the University of Toronto.
