|Title:||Causation in Sound Change: Perception-based Monophthongization in Old High German||Add Dissertation|
|Author:||James Ritchie||Update Dissertation|
|Email:||click here to access email|
|Institution:||University of California, Berkeley, Department of German|
|Linguistic Subfield(s):||Historical Linguistics; Phonetics;|
German, Old High
|Abstract:||Past treatments of the Old High German (OHG) monophthongization of Germanic *ai and *au mainly before given consonants can be characterized largely as searches for a single conditioning articulatory feature. There is currently no agreement in the literature, however, about what that feature is. Conventional articulatory definitions offer only an incomplete account of the potential effect sof the coarticulatory interaction between the conditioning environments and the original diphthongs. Although the specific details of the OHG speech sounds remain uncertain, we can use laboratory phonetics and our knowledge of modern speech sounds to create a plausible model of the OHG monophthongization.
This dissertation argues that the OHG monophthongization was a perception-based sound change. AN experiment was conducted to test the perceptual identification of [aI] and [aU] under conditions of increased speaking rate. Conditions of increased speaking rate were artificially created by truncating 40% of the onset from [aI] and [aU] before a variety of medial consonants in a set of nonsense tokens. Seventeen subjects were tested for their perceptions of these truncated diphthongs. Given the perceptual salience of the spectral dynamic portions of the diphthong off-glides, the null hypothesis that there would be no significant rates of error in the correct identification of the truncated diphthongs was tested and disproved.
Statistically significant rates of errors occurred before medial pharyngeals and medial dentals. There was no significant rate of error in diphthong perception before other medial consonants. These former environments reflect the same environments which condition the OHG monophthongization. The results of the experiment are particularly interesting in view of the fact that the spectral dynamic portions of the signal were left intact, yet not perceived. The perceptual interactions of the conditioners of the OHG monophthongization with the Germanic diphthongs *ai and *au appear to have introduced systematic ambiguities in the speech signal which motivated listener reanalyses of the diphthongs as monophthongs.