More than a hundred years ago a phonetic observation was made that in the dialect of the Marchfeld (Central Bavarian variety) only two combinatory patterns of vowels and consonants occur: Long vowels are followed by lenis consonants and short vowels are followed by fortis consonants (Pfalz 1913). This became widely known as Pfalz' law. Since then, this distribution has been confirmed in many varieties (i.e., Bannert 1976, Hinderling 1980), however studies also showed that Pfalz's law cannot be used to describe the timing in all Central Bavarian varieties. For example, in the study of Moosmüller & Brandstätter (2014), they presented evidence that not only tokens with long vowels and lenis consonants or short vowels with fortis consonants occur but also a third combination, namely tokens with long vowels and fortis consonants occur in two varieties of Vienna and are not adapted to fit the above mentioned pattern (i.e., by lenition of the fortis consonant or by shortening of the long vowel). They argue that these sequences are introduced via an interaction with the Standard variety as spoken in Germany, because there all possible combinations between long and short vowels and fortis and lenis consonants occur.
The data recorded by Pfalz was analysed via auditory impression and was never measured using phonetic standards. In this submission we use the original data of Pfalz, and will analyse the digitized sound files based on phonetic standards using Praat (Boersma & Weenik 2020). As concerns the quality of the data, deterioration of the signal-to-noise-ratio affects temporal organisation less than, i. e., spectral information.
The data includes recordings of three male speakers of the dialect of Deutsch-Wagram (18 years, 38 years, and 54 years). For each speaker so-called 'Wenker Sätze' and a individual narrative were recorded. The data was transcribed by Pfalz using his own transcription system as described in Pfalz (1913).
For the present submission (1) the transcriptions of Pfalz will be evaluated and (2) the sound files will be phonetically analysed. From the transcribed data the tokens which were relevant for the postulated observation of Pfalz, with sequences of long vowel and lenis consonants or short vowel and fortis consonants, will be identified. Then, these tokens will also be identified and manually segmented in the sound file, and the duration measures for vowels and consonants will be extracted. The analysis will give the first phonetic measurements of the historical data of Pfalz and will broaden the insight in the timing system of the dialect of the Marchfeld at this time.