Main Content

Project A3 - Linguistic representations and Standard oriented speech

PI: Alfred Lameli
Ph.D. student: Nadja Spina

Research context

PhD project A3 investigates German regional prosody, a research topic which has long been neglected by German linguistics. Only few studies have so far tackled prosodic topics, such as speech tempo and reduction (e.g. Hahn & Siebenhaar 2019, Hahn in print), regional intonation (Peters 2006), Middle Franconian tonal accents (Werth 2011, Schmidt 1986) or speech rhythm and timing in Swiss German varieties (e.g. Leemann & Siebenhaar 2007). Very little attention has thereby been paid to prosodic phrasing. Thus, the latter authors observe different degrees of pre-boundary lengthening in different Swiss German varieties while Streck (2004) makes the same observation for the German varieties of Mannheim and Hamburg.

Pre-boundary (PBL) lengthening forms the object of research of the present PhD project. Found for a variety of different languages across the world, PBL constitutes a universal acoustic correlate of prosodic phrasing and has the potential to signal intonation phrase boundaries (IP-boundaries) and the information structure of an utterance (for example, in the case of prosodic disambiguation, (see Schubö & Zerbian 2020 for Standard German). Across languages, PBL differs in its general degree, its implementation pattern and its perceptive relevance depending on the influence of specific phenomena in the different word prosodic and phonological systems of the languages. For examples, it used to be assumed that quantity languages, which exhibit lexically-distinctive pairs of long and short
phonemes, show PBL to a lesser degree and that it was hence perceptively less relevant (e.g. Vaissière 1983). More recent studies show that quantity languages exhibit PBL to a degree that is comparable to other languages but that PBL, in quantity languages, is
implemented in language specific ways in that short vowels are lengthened in a restricted manner so as to maintain the quantity distinctions in the phonological system (see Spina & Schubö 2021 for Czech). Studies investigating the perceptive relevance of PBL have not yet been conducted so far, but Yang et al. (2014) observed that native speakers of Mandarin Chinese show greater variation in their production of PBL as they are more sensitive to F0 because of the lexically-distinctive tones in their language system. This leaves more room for variation of duration than for variation of F0 as a boundary signal. Simultaneously, the perceptive relevance of F0 for boundary detection is weaker than that of PBL.

Departing these findings for PBL in a cross-linguistic perspective, the present PhD project will investigate differences in the general degree, the implementation patterns and the perceptive relevance of PBL in German dialects that exhibit specific phenomena which are linked to variation in duration. These differences will be investigated both horizontally, across the dialects and, vertically, along the axes between the respective dialects and the Standard German (see Kehrein 2012). In this way, the present PhD project contributes to filling the lack of research on regional prosody in German forms the basis for studying PBL language varieties. Central to the present PhD project is the question of the existence, the stability and the dynamics of a mental representation of PBL. After all, PBL is a phenomenon which is learnt by speakers. While infants as young as 6-month-olds are able to perceive PBL (see Holzgrefe-Lang et al. 2018 for Standard German), productions of PBL cannot or only minimally be found in their speech (Oller & Smith 1977 for English). Speakers hence learn the specific patterns and cue weighting of PBL in their native language. Do they also form mental representations of PBL during this? Does the sensitivity for PBL lie in these representations? How dynamic and how stable are these representations? Will perception experiments across different German dialects and along different axes between the dialects and Standard German give us the answer? The present PhD project will show.

Current dissertation project


The specific aims of the present PhD project are to answer the following research questions:

1. To what degree do the dialects investigated exhibit PBL and what implementation patterns can be found? Which dialect-specific phenomena may be responsible for these patterns?

2. Does the respective degree or the implementation pattern of PBL change along the axes between the dialects and Standard German? What implications does this have for the existence, stability and dynamics of mental representations of PBL?

3. Is the perceptive relevance as a cue to IP-boundaries stronger or weaker for listeners who were socialized in the dialects investigated than for speakers who were socialized in other dialects?

4. Does the sensitivity for PBL of listeners who were socialized in the dialects investigated change along the dialect-standard-axis or along the continuum between their own and another dialect? What implications does this have for the existence, stability and dynamics of mental representations of PBL?


Research questions 1 and 2 will be investigated by an analysis of the REDE-corpus, a corpus of recordings from about 150 places in Germany comprising spoken translations of the “Wenker sentences” by the speakers into their individually “broadest dialect” and “best Standard German”, readings of the text “The North Wind and the Sun” and conversations between the speakers and familiar persons, such as their friends (Freundesgespräch), as well as interviews between the speakers and the experimenter (cf. Kehrein 2012). The readings of “The North Wind and the Sun”, of the Freundesgespräche and the experimenter interviews offer the opportunity to compare identical words in IP-medial and IP-final position along the vertical axis between the individual dialects and Standard German both in read and in spontaneous speech as speakers adapt their speech style to the different situations. The speakers’ translations of the Wenker sentences into their dialects make it possible to compare PBL horizontally across the dialects investigated. In each case, the boundaries between the individual segments of words immediately preceding a prosodic boundary (and of identical words appearing IP-medially) will be annotated in order to measure and compare their duration.

Research questions 3 and 4 will be tested in a following perception study applying EEGmethods (CPS or P600). For this purpose, a speaker (disregarding their dialect or the degree of their dialect) could be recorded reading a set of sentences with different syntacially ambiguous structures, such as „[He saw the man] [with the binoculars]” versus „[He saw] [the man with the binoculars]”. The recordings could be played to listeners whose reactions to the presence or absence of the disambiguating prosodic boundary will be measured. The listeners could be asked to distinguish between the two different meanings of the sentences. The recorded material will be manipulated so that F0 and pauses will be neutralised, leaving PBL as the only remaining boundary cue. The PBL values will be synthesised so that they will equal the values measured for the dialects in the corpus study. If speakers of the dialects investigated perceive IP-boundaries more easily than speakers of another dialect, the perceptual relevance of PBL will be higher for them. If they have more difficulties to perceive IP-boundaries, then the perceptual relevance of PBL will be lower (and that of F0 higher). The latter possibility can be tested by running the same experiment again, this time manipulating PBL instead of F0. In order to test research question 4, the PBL values would be manipulated so that they form a continuum between the values measured for the dialects in the corpus study and the values measured for Standard German in previous studies (e.g. Schubö & Zerbian 2020) or a continuum between the values measured for the dialects investigated and another dialect. If speakers of the dialects investigated have increasing difficulties to perceive IP-boundaries approaching Standard German or another dialect, their sensitivity for PBL changes along the dialect-standard-axis or towards other German varieties.

Preliminary work

A preliminary analysis of the data of the IDS-corpus “Deutsch heute” as measured in the project Sprechtempo and Reduktion im Deutschen (SpuRD, see Hahn in print), has shown that Alemannic, Bavarian, Middle and Rhenish Franconian dialects exhibit PBL to a greater degree than other dialects such a Northern Low German. “Deutsch heute” is a corpus of standard-targeted readings of the text “The North Wind and the Sun”. In this text, the word “sun” (German: Sonne) appears twice in IP-final and twice in IP-medial position and was thus suitable for a preliminary comparison of the durations of its individual segments between both positions and hence for the degree of PBL in the entire German language area.

All dialect areas in which, according to this preliminary analysis, show stronger PBL exhibit dialect-specific phenomena which are linked to duration variation. High Alemannic exhibits lengthening of monosyllables (Seiler 2010) and lexically distinctive pairs of long and short phonemes (Krähenmann 2003), similar to quantity languages. Northeastern Swabian exhibits lengthening of vowels before former geminates (Auer 1989, Seiler 2010). Middle Bavarian, the syllable structure rule “long vowel + lenis consonant, short vowel + fortis consonant” applies (Hinderling 1980, cf. Moosmüller & Brandstätter for Viennese). Combining the two dialect areas, Lameli (2021) observes systematic lengthening of vowels depending on the consonantal context on Upper German.

Stronger PBL also delineates a Rhenish Franconian-Central Hessian continuum with a particularly strong lengthening effect in the area of Mannheim. This is in line with Streck (2004), who observes stronger PBL in the urban variety of Mannheim compared to the urban variety of Hamburg. Also for Mannheim, Peters (2006) discusses the lengthening of accented syllables, which are produced with a typical intonation pattern. Lengthening in this case could only be a “by-product” of intonation as it takes more time to produce certain
intonation contours. This seems also to be the case with the Middle Franconian tonal accents (Werth 2011). In this case, the difference in duration between tonal accent 1 and 2 results from the need to lengthen tonal accent 2 in order to produce its distinctive intonation contour. Nonetheless, it should not be disregarded that duration may have the function to enhance the contrast (Werth 2011). Streck (2004) also sees the possibility that the contrast between accented and unaccented syllable in the urban variety of Mannheim is expressed by duration. In accordance, Peters (2006) mentions extreme lengthening of IPfinal accented syllables.

Relation to other projects

Since PBL is a phenomena which is learnt by speakers, it is directly related to projects B1, B2 and B3. Project B1 investigates word prosodic processes, such as omitting unaccented syllables, during language acquisition and their implications for the revision of mental representations. How do children deal with syllables affected by PBL? Do they also have to revise their mental representation of a segment if they hear it phrase-medially and phrasefinally? Moreover, the results of the present PhD project may support the findings of (Werth 2011) regarding the relevance of duration for the distinction of the Middle Franconian tonal accents, which will have implications for their acquisition investigated in project B2. The present PhD project will also have implications for language contact in L1-acquisition (project B3). Children who grow up listening to the PBL patterns of more than one language
could either form a “mixed” PBL pattern comprising properties of either language or form their own PBL pattern, comprising new properties (cf. Atterer & Ladd 2004, Queen 2006 for intonation). Furthermore, the present PhD project is related to project C1 since dialectspecific PBL, which may prevail in standard-oriented speech registers could contribute to the perception of regiolect boundaries. Applied to project A2, different mental representations of PBL may also lead to special communicative situations or synchronization processes. In this sense, the present PhD project constitutes a linking element of the research training group (RTG) 2700.


Hahn, Matthias und Beat Siebenhaar (2019): "Schwa unbreakable – Reduktion von Schwa im Gebrauchsstandard und die Sonderposition des ostoberdeutschen Sprachraums". In: Kürschner, Sebastian/Mechthild Habermann/Peter O. Müller (Hg.), Methodik moderner Dialektforschung: Erhebung, Aufbereitung und Auswertung von Daten am Beispiel des Oberdeutschen. Georg Olms: 215–236.

Hahn, Matthias (i. Dr.): Sprechgeschwindigkeit und Reduktion im deutschen Sprachraum. Eine Untersuchung zur diatopischen Variation standard- intendierter Vorleseaussprache. Dissertation Universität Leipzig.

Werth, Alexander (2011): Perzeptionsphonologische Grundlagen der Prosodie. Eine Analyse der mittelfränkischen Tonakzentdistinktion. Steiner. (Zeitschrift für Dialektologie und Linguistik. Beihefte. 143).

Schmidt, Jürgen Erich (1986): Die mittelfränkischen Tonakzente (Rheinische Akzentuierung). Mainzer Studien zur Sprach- und Volksforschung. 8. Steiner.

Leeman, Adrian & Beat Siebenhaar (2007): Intonational and temporal features of Swiss German. In: Proceedings of the 16th International Congress of Phonetic Sciences (ICPhS), Saarbrücken, Germany, 2007:957−960.

Schubö, Fabian & Sabine Zerbian (2020): Phonetic content and phonological structure affect pre-boundary lengthening in German. In: Proceedings of the 10th International Conference on Speech Prosody, Tokyo, Japan 2020: 111-115.

Vaissière, Jacqueline (1983): Language-independent prosodic features. In: Cutler, Anne/Robert D. Ladd (Hg.), Prosody: Models and measurements. Springer Series in Language and Communication, 4: 53-66.

Spina, Nadja & Fabian Schubö (2021): Vowel length affects pre-boundary lengthening in Czech. In: Proceedings of the 1st International Conference on Tone and Intonation (TAI), Sonderborg, Denmark 2021: 166-170.

Kehrein, Roland (2012): Regionalsprachliche Spektren im Raum: Zur linguistischen Struktur der Vertikale. Steiner (Zeitschrift für Dialektologie und Linguistik. Beihefte. 152).

Yang, Xiaohong, Xiangrong Shen, Weijun Li & Yufang Yang (2014): How Listeners Weight Acoustic Cues to Intonational Phrase Boundaries. PloS ONE 9(7): 1-9.

Holzgrefe-Lang, Julia, Caroline Wellmann, Barbara Höhle & Isabell Wartenburger (2018): Infants’ Processing of Prosodic Cues: Electrophysiological Evidence for Boundary Perception beyond Pause Detection. Language and Speech, 61(1): 153-169.

Oller, Kimbrough D. & Smith (1977): Effect of final-syllable position on vowel duration in infant babbling. Journal of the Acoustical Society of America, 62(4): 994-997.

Seiler, Guido (2010): Einsilblerdehnung: Neue Lösungen für ein altes Problem. In: Christen, Helen/Sibylle
Germann/Walter Haas/Nadia Montefiori/Hans Ruef (Hg.), Alemannische Dialektologie: Wege in die Zukunft.
Steiner. (Zeitschrift für Dialektologie und Linguistik. Beihefte. 141).

Krähenmann, Astrid (2003): Quantity and Prosodic Assymetries in Alemannic. De Gruyter.

Auer, Peter (1989): Zur Dehnung im Alemannischen (am Beispiel des Konstanzer Stadtdialekts). In: Steiner. Zeitschrift für Dialektologie und Linguistik, 51: 32-57.

Hinderling, Robert (1980): Lenis und Fortis im Bairischen. Versuch einer morphophonemischen Interpretation In: Steiner. Zeitschrift für Dialektologie und Linguistik, 47: 25-51.

Moosmüller, Silvia & Julia Brandstätter (2014): Phonotactic information in the temporal organization of Standard Austrian German and the Viennese dialect. Language Sciences, 46: 84-95.

Lameli, Alfred (2022): Syllable Structure Spatially Distributed: Patterns of Monosyllables in German Dialects. Journal of Germanic Linguistics, 34(3): 241–287

Peters, Jörg (2006): Intonation deutscher Regionalsprachen. De Gruyter.

Streck, Tobias (2004): Finale Dehnung im Deutschen: Eine kontrastive Untersuchung zu den städtischen Varietäten von Mannheim und Hamburg. Staatsexamensarbeit Universität Freiburg.

Atterer Michaela & Robert D. Ladd (2004): On the phonetics and phonology of “segmental anchoring” of F0: evidence from German. Journal of Phonetics. 32(2): 177-197.

Queen, Robin M. (2006): Phrase-final intonation in narratives told by Turkish-German bilinguals. International journal of bilingualism, 10(2): 153-178.