Pre-aspiration in Welsh English: A case study of Aberystwyth

Access to files

•   FULL-TEXT.PDF (pdf)

Abstract

This thesis examines the phonetic, phonological, and social aspects of pre-aspiration in English spoken in Aberystwyth, mid Wales. Pre-aspiration refers to a period of voiceless (primarily) glottal friction occurring in the sequences of sonorants and phonetically voiceless obstruents (e.g. in mat [mahts] or mass [mahs]). Chapter 1 summarises the objectives of this thesis and where the thesis is positioned with respect to our current knowledge of the phenomenon and the relevant theoretical issues. Chapter 2 introduces the data used to address these objectives.Pre-aspiration is usually considered as consisting of a voiced glottal component, or breathiness, and a voiceless glottal component, or voiceless pre-aspiration, and these are treated as a single unit in a number of analyses (Helgason 2003; Helgason & Ringen 2008; Karlsson & Svantesson 2011; Morris 2010; Ringen & van Dommelen 2013; Stevens & Hajek 2004b, 2004c; Stevens 2010, 2011). Chapter 3 shows that this is not adequate because distinguishing the two enables us to discover patterns that would remain obscured otherwise – such as breathiness being a possible precursor to pre-aspiration. This is demonstrated through the segmental and prosodic conditioning of pre-aspiration and breathiness.Chapter 4 shows that although pre-aspiration is not an obligatory feature of Aberystwyth English (in the sense that it would occur in 100% of time where it can), it nevertheless forms two clear categories sensitive to phonological rather than phonetic vowel height. However, phonological vowel height on its own cannot explain these two categories and interacts with a number of other conditioning factors.Whilst Chapter 3 investigates the relationship between pre-aspiration and breathiness, Chapter 5 looks into that of pre-aspiration and glottalisation and demonstrates that the two can occur in the same environment, which enlightens the debates related to the historical connections between pre-aspiration and glottalisation in particular (e.g. Kortland 1988). It furthermore reveals that although it is not known why they are co-occurring for some speakers and mutually exclusive or allophonic for others, their relationship is conditioned prosodically and not segmentally.Chapter 6 illustrates that pre-aspiration is an acoustic correlate of the fortis-lenis contrast in plosives in production at least equally well as breathiness, voicing, release duration, or the duration of the preceding vowel, and better than voiceless closure duration, glottalisation, or f0 before or after the plosive in question in the word-medial (cotter [khɒhtsə] ~ codder [khɒdə]) and the word-final positions (cot ~ cod). It is therefore at least as important as the other four correlates.Chapter 7 finds that pre-aspiration also exhibits social conditioning. Females pre-aspirate more frequently than males, which is often found in pre-aspiration studies, but this difference disappears as the age decreases. Furthermore, the frequency of breathiness, and the duration of pre-aspiration and breathiness are not conditioned by gender. However, all four variables are affected by age. Pre-aspiration thus seems to be undergoing an advancing sound change according to Labov’s Principle II (2001: 292) and breathiness seems to be its precursor. Chapter 8 summarises the results and outlines questions for further research.

Layman's Abstract

This thesis examines pre-aspiration in English spoken in Aberystwyth, mid Wales. Pre-aspiration is a sound very similar to h (as in honey), and it can be found mainly in the sequences of vowels (a, e, i, o, u, etc.) and certain consonants (p, t, k, f, th, s, sh). In English, pre-aspiration would be manifested as a h sound inserted in the following words as indicated: paht, peht, piht, poht, puht, mahp, mahtt, mahck, ohff, mohth, mahss, mahsh. Because the sound is understudied in English, the work focuses on what other sounds condition the presence of the phenomenon and its duration, if it is sensitive to social factors such as age or gender as well, and it suggests some implications for sound changes including pre-aspiration.Chapter 1 summarises the questions addressed in this thesis and outlines where this work is positioned with respect to our current knowledge of the phenomenon and the relevant theoretical issues. Chapter 2 describes the data used to answer these questions.Pre-aspiration has been usually seen as consisting of two sounds: a voiced h and a voiceless h, i.e. a h sound during which the vocal folds either vibrate or not. The voiced pronunciation of h is known as breathiness. The two h’s are usually treated as a single unit. Chapter 3 shows that this is not adequate because distinguishing the two enables us to discover patterns that would remain obscure otherwise.Chapter 4 suggests that pre-aspiration is conditioned by a mixture of phonetic effects, i.e. effects that should be explainable articulatorily or aerodynamically, and phonological effects, i.e. effects that are of a cognitive character and cannot be explained by articulatory and aerodynamic factors.Whilst Chapter 3 investigates the relationship between pre-aspiration and breathiness, Chapter 5 looks into that of pre-aspiration and glottalisation. Glottalisation is frequently found in today’s English in words such as water, in which the t sound is missing and replaced by a glottal catch. At first blush, pre-aspiration and glottalisation would seem incompatible as the former involves spreading of the vocal folds and the latter their closing. This chapter nonetheless demonstrates that pre-aspiration and glottalisation can co-occur in the same sound. This finding enlightens the debates related to the historical connections between pre-aspiration and glottalisation in particular (e.g. Kortland 1988).Chapter 6 illustrates that pre-aspiration helps to distinguish /p/ from /b/ (lap vs lab), /t/ from /d/ (pat vs pad), and /k/ from /g/ (lack vs lag) as it only occurs in /p/, /t/, and /k/. It is therefore at least as helpful in this regard as other such correlates of the /p/, /t/, /k/ vs /b/, /d/, /g/ distinction that are traditionally thought of as the most important. Chapter 7 finds that women pre-aspirate more frequently than men, which is in accordance with many pre-aspiration studies, but this difference decreases as the age decreases and results in no gender difference in the youngest speakers. Breathiness is not affected by gender. The frequency and the duration of pre-aspiration and breathiness are nevertheless conditioned by age: on the whole, pre-aspiration and breathiness are becoming more frequent and longer in duration. This chapter also supports the suggestion that breathiness is a precursor to pre-aspiration. Chapter 8 summarises the results and outlines questions for further research.

Additional content not available electronically

I obtained recorded speech from 18 speakers from Aberystwyth to write the thesis.

Keyword(s)

Welsh English; breathiness; glottalisation; pre-aspiration

Bibliographic metadata

Institutional metadata

Record metadata