'/L/-Vocalisation in Lucerne Swiss German Dialects: A
Total Page：16
File Type：pdf, Size：1020Kb
Zurich Open Repository and Archive University of Zurich Main Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch Year: 2017 /l/-vocalisation in Lucerne Swiss German dialects: a sociophonetic analysis using big data Zihlmann, Urban B ; Leemann, Adrian Abstract: This study compares the Lucerne German allophones of Middle High German <l> using data from the crowdsourcing smartphone app Dialäkt Äpp, to those of the Linguistic Atlas of German-Speaking Switzerland. In doing so we examine language change in real-time across c. 70 years. In the north of the Canton of Lucerne, use of /l/-vocalisation has receded, whereas in the west, and in and around the city of Lucerne, /l/-vocalisation occurs more frequently than at the time of the historic atlas. Both age and sex differences were found to be statistically non-significant, though older speakers tended to vocalise more. Generally, however, we report substantial areal heterogeneity which we suspect is due to the Canton’s geographical location in a dialectal transition zone, increased personal mobility, and the growing presence of broadcast media in dialect. DOI: https://doi.org/10.18452/18805 Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-150106 Conference or Workshop Item Published Version Originally published at: Zihlmann, Urban B; Leemann, Adrian (2017). /l/-vocalisation in Lucerne Swiss German dialects: a sociophonetic analysis using big data. In: Phonetics Phonology in German-speaking countries (PP 13), Berlin, 27 September 2017 - 29 September 2017, 201-204. DOI: https://doi.org/10.18452/18805 Proceedings P&P13 /l/-vocalisation in Lucerne Swiss German dialects: a sociophonetic analysis using big data Urban Zihlmann1, 3, Adrian Leemann2, 3 1Phʀnɶtics Labʀʃatʀʃy, ɔɶʁaʃtmɶnt ʀf Cʀmʁaʃativɶ Linguistics, Univɶʃsity ʀf ɫuʃich, 2ɔɶʁaʃtmɶnt ʀf Linguistics anɵ ɕnglish Languagɶ, Lancastɶʃ Univɶʃsity 3Phʀnɶtics Labʀʃatʀʃy, ɔɶʁaʃtmɶnt ʀf Thɶʀʃɶtical anɵ ɑʁʁliɶɵ Linguistics, Univɶʃsity ʀf Cambʃiɵgɶ uʃban.гihlmannɐuгh.ch, a.lɶɶmannɐlancastɶʃ.ac.uk Abstract ʁʃɶstigiʀus nʀn-vʀcalising ɵialɶct, /l/-VOC nʀt ʀnly bʀʃɶ thɶ stigma ʀf bɶing assʀciatɶɵ with a This study compares the Lucerne German lʀw sʀciʀ-ɶcʀnʀmic status, but was alsʀ suʁ- allophones oɐ Middle Hiɑh German <l> usinɑ ʁʀsɶɵly unʁlɶasant tʀ listɶn tʀ. Uʃban uʁʁɶʃ- data ɐrom the crowdsourcinɑ smartphone app class sʁɶakɶʃs thus allɶgɶɵly avʀiɵɶɵ vʀcalising Dialäɕt Äpp, to those oɐ the Linɑuistic Atlas oɐ /l/s (Chʃistɶn 1988: 203). It is thɶn ʃathɶʃ cʀun- German-Speaɕinɑ Switzerland. In doinɑ so we tɶʃintuitivɶ that a ʃɶcɶnt stuɵy fʀunɵ /l/-VOC tʀ examine lanɑuaɑe chanɑe in real-time across c. havɶ sʁʃɶaɵ acʃʀss CH (Lɶɶmann ɶt al. 2014: 70 years. In the north oɐ the Canton oɐ Lucerne, 192), imʁlying that in sʀmɶ ʃɶgiʀns, /l/-VOC use oɐ /l/-vocalisation has receded, whereas in must havɶ lʀst its stigma. Wɶ will ɵiscuss this the west, and in and around the city oɐ Lucerne, latɶʃ as this stuɵy ʃɶʁʀʃts similaʃ ʃɶsults. /l/-vocalisation occurs more ɐrequently than at Whɶthɶʃ sʀcial class is still a ɵistinguishing the time oɐ the historic atlas. Both aɑe and sex factʀʃ fʀʃ /l/-VOC tʀɵay must bɶ ɶstablishɶɵ in diɐɐerences were ɐound to be statistically non- futuʃɶ ʃɶsɶaʃch. Thɶʃɶ is, hʀwɶvɶʃ, ɶviɵɶncɶ fʀʃ siɑniɐicant, thouɑh older speaɕers tended to vo- ʀthɶʃ ɶxtʃalinguistic factʀʃs tʀ bɶ cʀʃʃɶlatɶɵ calise more. Generally, however, we report sub- with /l/-VOC: Thɶ Linɑuistic Atlas oɐ German- stantial areal heteroɑeneity which we suspect is Speaɕinɑ Switzerland (SDS) ʃɶʁʀʃtɶɵ highɶʃ /l/- due to the Canton’s ɑeoɑraphical location in a VOC ʃatɶs amʀngst yʀungɶʃ sʁɶakɶʃs fʀʃ sʀmɶ dialectal transition zone, increased personal ʃɶgiʀns, anɵ amʀngst ʀlɵɶʃ sʁɶakɶʃs fʀʃ ʀthɶʃs mobility, and the ɑrowinɑ presence oɐ broad- (154). Chʃistɶn fʀunɵ yʀungɶʃ sʁɶakɶʃs tʀ bɶ cast media in dialect. lɶss likɶly tʀ vʀcalisɶ (1988: 203), anɵ Lɶɶmann ɶt al. camɶ tʀ a similaʃ cʀnclusiʀn as thɶ yʀung- Introduction ɶst sʁɶakɶʃs ʀf thɶiʃ stuɵy vʀcalisɶɵ lɶss than Fʀʃ Gɶʃman-sʁɶaking Switгɶʃlanɵ (CH), /l/- thɶ ʀlɵɶʃ ʀnɶs (2014: 206). ɑs fʀʃ sɶx, nʀ infʀʃ- vʀcalisatiʀn (/l/-VOC) ʃɶfɶʃs tʀ thɶ usɶ ʀf [u̯ ] matiʀn has bɶɶn fʀunɵ, which is why wɶ in- ʃathɶʃ than a cʀnsʀnantal allʀʁhʀnɶ (ɶ.g. [l] ʀʃ cluɵɶ this factʀʃ in thɶ cuʃʃɶnt analysis. []) fʀʃ /l/. It is bɶliɶvɶɵ tʀ havɶ ʀʃiginatɶɵ in thɶ ɕmmɶntal, a ʃuʃal aʃɶa in thɶ cantʀn ʀf ɒɶʃn Methodology (ɒɕ; Chʃistɶn 1988: 15), anɵ ʀccuʃs mainly 243 ʃɶcʀʃɵings ʀf <Chällɶ>, i.ɶ. VLLV (‘la- ʀʃally. Thus, it ʃɶmains unclɶaʃ whɶn it fiʃst aʁ- ɵlɶ’ Kelle; 111 malɶ; 132 fɶmalɶ; agɶ 10-77; ʁɶaʃɶɵ in CH, thʀugh histʀʃical sʀuʃcɶs fʃʀm mɶan=30.1; mɶɵian=26.5; Sɔ=15.0) wɶʃɶ ɶx- thɶ 18th cɶnt. suggɶst that it haɵ alʃɶaɵy bɶɶn tʃactɶɵ fʃʀm thɶ Dialäɕt Äpp (DÄ) cʀʃʁus. ɑll aʃʀunɵ in that timɶ (ɒaumgaʃtnɶʃ 1940: 73-4). tʀkɶns wɶʃɶ ʃɶcʀʃɵɶɵ at a samʁling ʃatɶ ʀf uʁ tʀ Histʀʃically, /l/-VOC has bɶɶn ʁɶʃcɶivɶɵ ʃa- 48 kHг (iPhʀnɶ Sʁɶcificatiʀns 2016) anɵ stʀʃɶɵ thɶʃ nɶgativɶly. In L1 acʂuisitiʀn, /l/ is ʀftɶn ʃɶ- as .WɑV fʀʃmat. Fʀʃ fuʃthɶʃ infʀʃmatiʀn ʀn DÄ ʁlacɶɵ by vʀcalisatiʀns (Lin & ɔɶmuth 2015: sɶɶ Lɶɶmann ɶt al. 2016. 13) ɵuɶ tʀ its aʃticulatʀʃy cʀmʁlɶxity “invʀlv- ɑll tʀkɶns wɶʃɶ auɵitʀʃily chɶckɶɵ fʀʃ valiɵ- ing cʀʃʀnal, latɶʃal, anɵ vɶlaʃ cʀntact” (Hall- ity (backgʃʀunɵ nʀisɶ, cliʁʁing ɶtc.). Thɶ ʃɶ- Lɶw & Fix 2012: 795). This ʁʀtɶntially lɶaɵs tʀ maining 219 tʀkɶns wɶʃɶ thɶn ɶxaminɶɵ fʀʃ /l/- an assʀciatiʀn ʀf /l/-VOC with chilɵʃɶn’s VOC auɵitʀʃily, anɵ fʀʃ vɶlaʃiгatiʀn (VɕL) sʁɶɶch (Stalɵɶʃ 1819: 64). Fuʃthɶʃmʀʃɶ, Haas with Praat. [l] has a ʃɶlativɶly lʀw ɐ1, anɵ a ʃɶ- fʀunɵ that sʁɶakɶʃs ʀf uʃban ɵialɶcts cʀnsiɵɶʃɶɵ lativɶly high ɐ2, whɶʃɶas fʀʃ [], thɶ ʀʁʁʀsitɶ aʁ- /l/-VOC amʀngst ʀthɶʃ attʃibutɶs as ʁɶasant anɵ ʁliɶs ɵuɶ tʀ thɶ ʃɶtʃactiʀn ʀf thɶ tʀnguɶ bʀɵy, unaɶsthɶtic (1973: 64), i.ɶ. fʀʃ sʁɶakɶʃs with a which ʃɶsɶmblɶs thɶ acʀustic ʁʃʀʁɶʃtiɶs ʀf a 201 Proceedings P&P13 high back unʃʀunɵɶɵ vʀwɶl (Sʁʃʀat & Fujimuʃa ɑlthʀugh it sʁʃɶaɵ tʀ thɶ sʀuthɶʃn ML ʃɶgiʀns 1993: 292). ɒasɶɵ ʀn this, thɶ mɶan ɐ1 anɵ ɐ2 tʀ sʀmɶ ɶxtɶnt, [l] has gainɶɵ gʃʀunɵ in 2016, valuɶs fʀʃ ɶach tʀkɶn wɶʃɶ mɶasuʃɶɵ. Thɶ tʀ- mʀst nʀtably alʀng thɶ ɶastɶʃn bʀʃɵɶʃ (sɶɶ Fig- kɶns wɶʃɶ labɶllɶɵ as [l] if thɶ ɐ1-ɐ2 ɵiffɶʃɶncɶ uʃɶ 2). was >1000 Hг, anɵ as [] if <1000 Hг. Thɶ num- bɶʃ ʀf [l], [] anɵ [u̯ ] wɶʃɶ thɶn cʀuntɶɵ fʀʃ ɶach lʀcatiʀn, althʀugh ɵuɶ tʀ limitatiʀns ʀf sʁacɶ, thɶ ɵiscussiʀn ʀf VɕL has bɶɶn lɶft ʀut. Wɶ cʃɶ- atɶɵ maʁs using QGIS (ɔɶvɶlʀʁmɶnt Tɶam 2016). Fʀʃ statistical analysɶs, RStudio was usɶɵ aʁʁlying chi-sʂuaʃɶ tɶsts. Mʀʃɶ nuancɶɵ statistics such as lʀgic ʃɶgʃɶssiʀns ʀʃ linɶaʃ mixɶɵ ɶffɶcts mʀɵɶls will bɶ ʃun in thɶ futuʃɶ. Results Fiɑure 1. /l/-VOC in LU in the 1950s (leɐt; SɔS) and In this sɶctiʀn, wɶ ʁʃɶsɶnt /l/-VOC ɵistʃibu- 2016 (riɑht; ɔÄ). Base map: Swisstʀʁʀ. tiʀn, thɶ changɶ ʀbsɶʃvɶɵ whɶn SDS anɵ DÄ aʃɶ cʀmʁaʃɶɵ, anɵ thɶ ʃɶsults ʀn sʀcial infʀʃmatiʀn Chanɑe by locality that may hɶlʁ ɶxʁlain thɶ ʃɶtʃɶatmɶnt anɵ ɵiffu- Figuʃɶ 2 shʀws thɶ changɶs that havɶ takɶn siʀn tɶnɵɶnciɶs ʀbsɶʃvɶɵ. Fʀʃ ʃɶfɶʃɶncɶ, Tablɶ ʁlacɶ in Lucɶʃnɶ (LU): hɶʃɶ, wɶ subtʃactɶɵ thɶ 1 ʁʃʀviɵɶs an ʀvɶʃviɶw ʀf thɶ 32 lʀcalitiɶs as SDS ʁɶʃcɶntagɶs fʃʀm thɶ DÄ ʀnɶs, which is usɶɵ in thɶ cuʃʃɶnt stuɵy. nɶgativɶ if /l/-VOC has ʃɶtʃɶatɶɵ, anɵ ʁʀsitivɶ Table 1. Names oɐ the reɑions used in SɔS and ɔÄ. if it is mʀʃɶ fʃɶʂuɶnt in 2016. Wɶ fuʃthɶʃ gʃʀuʁɶɵ thɶ DÄ ɵata by sʁɶakɶʃ, sɶx, anɵ agɶ Area Regions (1-3 political communities per region) (<20; 21-35; 36<: ɶach gʃʀuʁ cʀntaining aʁ- Entlebuch (EB) 1. Entlebuch; 2. Escholzmatt; 3. Marbach; ʁʃʀximatɶly thɶ samɶ numbɶʃ ʀf sʁɶakɶʃs). On 4. Schüpfheim balancɶ, /l/-VOC has unɵɶʃgʀnɶ mʀʃɶ ʃɶtʃɶat- Hinterland (HL) 5. Altbüron; 6. Luthern; 7. Pfaffnau; 8. Zell mɶnt than ɵiffusiʀn – ɶsʁɶcially in thɶ ɶastɶʃn Lucerne- 9. Ebikon; 10. Eschenbach; 11. Hitzkirch; ʁaʃts ʀf LU (ʀʃangɶ). Hochdorf (L-H) 12. Hohenrain; 13. Horw; 14. Lucerne Midland (ML) 15. Beromünster; 16. Dagmersellen; 17. Grosswangen; 18. Malters; 19. Menznau; 20. Neudorf; 21. Neuenkirch; 22. Nottwil; 23. Rothenburg; 24. Ruswil; 25. Schötz; 26. Sempach; 27. Sursee; 28. Triengen; 29. Willisau; 30. Wolhusen Mount Rigi (RG) 31. Weggis Schongau (SC) 32. Aesch Comparinɑ SɔS and ɔÄ distributions Fiɑure 2. /l/-VOC chanɑe ɐrom the 1950s to 2016. Figuʃɶ 1 shʀws thɶ usɶ ʀf [u̯ ] basɶɵ ʀn SDS Neɑative numbers indicate a decrease in /l/-VOC, anɵ DÄ ɵata in %. Thɶ ʁɶʃcɶntagɶ ʀf thɶ SDS positive numbers an increase. Base map: Swisstʀʁʀ. ɵata is basɶɵ ʀn thɶ numbɶʃ ʀf [u̯ ] ʁɶʃ ʃɶgiʀn: if ʀnly [l] ʀccuʃʃɶɵ (as ɶ.g. in 6), it was cʀɵɶɵ as Eɐɐect oɐ social cateɑories 0%, [u̯ ] anɵ [] (as ɶ.g. in 2 anɵ 8) was cʀɵɶɵ as 50%, anɵ ʀnly [u̯ ] as 100%. In thɶ 1950s, mʀst In all ʀf LU in 2016, fɶmalɶ sʁɶakɶʃs ʁʃʀ- ʃɶgiʀns in ɕntlɶbuch (ɕɒ; 1-4), Lucɶʃnɶ-Hʀch- ɵucɶɵ 23.7% [u̯ ], whilɶ malɶ sʁɶakɶʃs ɵiɵ sʀ at ɵʀʃf (9-14), anɵ in thɶ nʀʃth ʀf Miɵlanɵ (ML; a ʃatɶ ʀf 23.8%. Nʀ statistically significant ɵif- 15-30) vʀcalisɶɵ, whilɶ thɶ sʀuthɶʃn ML ʃɶ- fɶʃɶncɶ in sɶx wɶʃɶ fʀunɵ whɶn all sʁɶakɶʃs aʃɶ giʀns ʁʃɶɵʀminantly usɶɵ [l]. Suʃʁʃisingly, cʀnsiɵɶʃɶɵ, χ2(1, N = 219) < .001, p = 1.00. Wɶ whilst nʀwaɵays /l/-VOC is usɶɵ unanimʀusly fuʃthɶʃ fʀunɵ that 36< sʁɶakɶʃs usɶɵ [u̯ ] 27.3%, in ɕɒ (1-4), in thɶ 1950s ʀnɶ ʃɶgiʀn (2) is ʃɶ- 20-35 ʀnɶs 23.2%, anɵ <20 ʀnɶs 21.4% (nʀt sta- ʁʀʃtɶɵ tʀ havɶ fʃɶɶ vaʃiatiʀn bɶtwɶɶn [] anɵ [u̯ ]. tistically significant, χ2(2, N = 219) = 0.71, p = In thɶ ʃɶgiʀns that aʃɶ histʀʃically vʀcalising, .70). ɒʃɶaking uʁ thɶ ɵata (sɶɶ Tablɶ 2), fʀʃ thɶ /l/-VOC sɶɶms tʀ havɶ bɶcʀmɶ lɶss fʃɶʂuɶnt.