Research Committee | Previous Projects
Compilation of a phonological database of contemporary Scottish English
Dominic Watt, Centre for Linguistic Research, University of Aberdeen
The proposed project aims to compile a systematically collected database of contemporary urban Scottish English from the four largest urban centres in Scotland, viz., Aberdeen, Dundee, Edinburgh and Glasgow. The focus will be on phonological variation within and between the accents of English spoken in each locale, with an initial emphasis on establishing the nature and range of phonetic variation in the following variables, some (but not all) of which have been described for some (but not all) of the Scottish urban centres in question: Voice Onset Time (VOT) in /p t k b d g/; pre- and post-vocalic rhoticity (presence/absence of post-vocalic rhoticity, and type of articulations used); long-term characteristic F0.
The data collection stage of the study will involve recording 80 adult speakers of Scottish English (20 per locale; 10 male, 10 female, divided into two age groups: 16-25; 40-65) using high-fidelity digital recording equipment in quiet settings. The stimulus materials will be comprised of a standardised word list (to be read twice) designed to incorporate frequent words exemplifying these variables in a representative range of phonological contexts, alongside two reading passages likewise designed to contain a large number of the target consonants in a wide range of contexts.
The resultant recordings will be downloaded to a Pentium IV PC via Kay Elemetrics CSL4500 signal capture equipment, and will be analysed using Kay MultiSpeech software used in conjunction with other appropriate signal editing and analysis software (Adobe Audition, Sensimetrics SpeechStation 2). VOT measurements will be expressed in milliseconds, while rhoticity will be assessed qualitatively via narrow IPA transcriptions and visual inspection of wide-band spectrograms, and quantitatively via subsequent coding of the phonetic transcriptions and measurements of appropriate acoustic parameters (stop gap duration in taps, F3 minima in approximants, etc.).
Characteristic F0 values will be be derived for each individual speaker on the basis of averaged readings from recordings of all read materials calculated using MultiSpeech, and will be pooled for speaker groups (speaker sex, age-group, locale) if this is deemed appropriate.
The results of this investigation will represent a representative set of population statistics on key aspects of the phonology of contemporary Scottish English, and as such will be of value to forensic investigations carried out by law enforcement agencies within Scotland and the United Kingdom. The database of recordings will also provide important information on other important consonantal and vocalic variables for later analysis.