Tables
2.1 A comparison of thirty-three variables in a recital of the Parable of the Prodigal Son by John Townend with corresponding SED data from Skelmanthorpe, Yorkshire.
2.2 A description of selected phonological variables in John Townend's recital of the Parable of the Prodigal Son.
2.3 A description of selected lexical and morphonological variables in John Townend's rendering of the Parable of the Prodigal Son.
3.4 Mixed methods logistic regression modelling for tapped and trilled /r/ (omitting speaker OM4), N = 2289.
3.5 Mixed methods logistic regression modelling for labiodental and labialized /r/, N = 2511.
4.1 Distribution of seven key phonetic features for early twentieth-century Received Pronunciation.
4.2 Parameters influencing the values of VOT for any speaker.
4.6 Comparative ranges of VOT for /k/ with Virginia Woolf, T. S. Eliot and Elizabeth Bowen.
4.10 VOT values for broadcast by Elizabeth Windsor on the occasion of her 21st birthday from Cape Town, South Africa in 1947.
4.11 VOT values for Queen Elizabeth II (1953–) in Christmas broadcast from 1953.
4.12 VOT values for Queen Elizabeth II (1953–) in Christmas broadcast from 1984.
4.13 VOT values for Queen Elizabeth II (1953–) in Christmas broadcast from 2014.
4.14 Average VOT (ms) values for Christmas broadcasts by Queen Elizabeth II from 1953 to 2014 and 21st birthday broadcast in 1947.
6.1 The population of Liverpool between 1801 and 1911, from census returns. Taken from Honeybone (2007: 115, via Neal 1988: 2).
6.2 The proportion of Liverpool population born outside of England, from Honeybone (2007: 116).
6.3 Numbers of Archive vowel tokens analysed across each lexical set, arranged by sex.
7.1 Information about the Berliner Lautarchiv (BLA) speakers.
8.1 Provenance, age and details of the recordings for the eight Central Belt Scottish English men, whose speech was recorded in 1916 and 1917.
8.2 Education, occupation, stated dialect and provenance of parents, for the overall sample of eight Central Belt Scottish men, from whom recordings were made in 1916 and 1917.
8.3 Mean percentage of variants used in syllable-onset and coda position (onset N = 139; coda N = 502).
8.4 Seven phonotactic environments conditioning weak /r/ in the Berliner Lautarchiv full sample. v indicates an unstressed syllable, V indicates a stressed syllable, C indicates any consonant.
8.5 Seven phonotactic environments conditioning weak /r/ in the WL07 corpus. v indicates an unstressed syllable, V indicates a stressed syllable, C indicates any consonant.
9.1 Distribution of seven key phonetic features for early twentieth-century supraregional Irish English. A dash ‘–’ indicates that the recording(s) in question did not provide any context where the feature could be assessed.
9.2 STRUT vowel realisations for five speakers in pre-WWII recordings and two speakers in post-WWII recordings (median values of F1 and F2), arranged (i) chronologically, (ii) by value of F1 and (iii) by value of F2.
9.3 Features of pre-WWII recordings (present to differing degrees with different speakers).
11.1 Regional and social characteristics of the ten speakers.
11.2 Percent of rhoticity by speaker and by speaker's geographic location.
11.3 F1, F2, and duration differences between LOT and THOUGHT, by speaker (*** p < .001, ** p < .01, * p < .05, italics: p > .05).
11.4 F1, F2, and duration differences between LOT and PALM, by speaker (*** p < .001, ** p < .01, * p < .05, italics: p > .05).
11.5 F1, F2, and duration differences between NORTH and FORCE, by speaker (*** p < .001, ** p < .01, * p < .05, italics: p > .05).
12.1 Upper Midwestern datasets. “Transcripts” refers to manual transcriptions of lexical items, not free conversation.
12.2 Distribution of 277 UME speakers by birth date and gender.
12.3 Order of selected conditioning effects of following consonants on /æ/ raising.
12.4 Comparison of TRAP variants in UME. Light shading identifies Wisconsin preference for raising before fortis fricatives; dark shading identifies Minnesota preference for raising before lenis velar plosives.
12.5 Words from Arthur used in the present study. Some words appear more than once (e.g., back, that, etc.).
12.6 DARE subjects from southeastern Wisconsin used in the present study. Subjects are listed by general acoustic pattern.
15.1 Auditory-impressionistic coding of seven Canadian-sounding World War I veterans for eight phonological variables (see key for explanation of variables; nd = no data).
15.2 Key to phonemic symbols used in vowel charts (Figures 15.1–15.3). Standard vowel class keywords from Wells (1982) are given in SMALL CAPS. Where these are inappropriate, other keywords are given in italics.
16.2 Mean F1 nuclear midpoint values, in Bark units, per Age/Origin group.
17.1 Summary of annotation conventions for Trinidadian and Jamaican intonation.
17.2 Summary of pitch accent and boundary tones in contemporary Jamaican.
18.1 Speakers and words in the corpus of early educated spoken GhE.
18.2 Sociobiography of speakers of present-day educated spoken GhE.
20.1 Features of Tristan da Cunha English relevant to historical reconstruction.
20.2 /h/ insertion in four TdCE speakers by place of interview (Southampton vs. TdC).
21.1 Details of the speakers selected for the historical analysis.
21.2 The number of tokens analysed per vowel per test speaker.
22.2 Coding scheme for the closing diphthongs. The lowest-score for glide weakening, i.e. (0), indicates no off-glide resulting in a monophthongal token.
22.3 The counts and the relative proportions of glide weakening variants for the four different lexical sets.
22.4 The numbers of (0) offglide variants versus other variants for FACE and GOAT in speakers with at least one Scottish parent versus speakers with no Scottish parents.
