Language change in progress

The permanent source of RP development, or innovations, has always been the so-called Popular London speech or Cockney. The hybrid of Es­tuary English is only a sign of its spreading through different, in particular more mobile, strata of the society. We will follow Alan Crattenden, the new editor of "The Introduction into the Pronunciation of English" by A.C. Gimson, in reviewing current changes in RP. They are grouped ac­cording to the degree of process completion: processes almost complete, changes well-established, recent innovations and innovations on the verge of RP (Cruttenden 2001: 82-83).

In the group of "processes almost complete" we find pronunciations which have been introduced into teaching English phonetics practice since the 60s. They are:

the diphthong /зо/ replaced the older realization as [ou], as in over, boat, comb;

a) /tj, dj/ in unaccented positions are regularly changed to /tj, d$/, as in culture, soldier,

b) the distinction between /эз/ and /o:/ is lost, both are now /o:/ inpaw, pour (or pore).

c) Another trend is followed less consistently but since it has already been registered in the two major pronouncing dictionaries (Jones 1997; Wells 2000) it will gradually find a way into teaching manuals and teaching practice too:

d) /j/ is lost following /I, s, z/, e.g. luminous ['lummas], suit [suit], exhume [ig'zu:m].

e) The diphthong /еэ/ is realized as a monophthong: in fare [h], tear [ts:].

f) A more subtle realizational feature is /r/ pronounced as a post-alveolar approximant in all positions and not, as formerly, as a tap in intervocalic positions following an accented syllable, e.g. very, error.

Among the changes well-established there are:

a) /i/ in many (but not all!) unaccented syllables replaced by /э/, e.g. quality ['kwnbti] but palace ['paehs];

b) /u/ used in place of/иэ/ in some, particularly monosyllabic, words, e.g. in sure, poor, cure, moor, tour but less likely mpure and impossible in dour, lure, Ruhr and in doer, fewer, newer, viewer,

c) final /i/ replaced by /i:/ in words like city, pretty, dirty;

d) the quality of/ж/ becomes more open, i.e. close to [a], e.g. mad, rat, cap;

e) pre-consonantal /t/ becomes a glottal stop, e.g. not very [nttfven] but glottalization is not acceptable before /1/, e.g. little [h?l] is considered substandard;

f) /j/ is lost after /n/, e.g. news [nu:z];

g) fronting of/u:/ to /u/, e.g. soon [sun];

h) accented /tj, dj/become /tj, йъ/, e.g. tune [tju:n], endure [тУзиэ] (See the commentary on this case below based on (Wells 1999.).

In the section of recent innovations the author includes pronunciations which are now heard in General RP but not yet typical of a majority of speakers:

a) /i3/ and /иэ/ are realized as [i:] and [o:], e.g. beer [bi:], sure [Jo:], the latter competing with [Jo:];

b) unrounding as well as fronting of/u,u:/as [i] and [i:], e.g. good [gid], soon [si:n]; fronting on its own of/u:/ has been common in RP rather longer;

c) the realization of /r/ without a tongue tip contact as [u] in red has been described as one of the features of Estuary English but, it is argued in the book, may be a general tendency within RP;

d) there is one intonation feature: the over-frequent use of a "checking" high rise on declarative sentences in conversational narratives, like, for example, Iwas at Heath'row yesterday. They've got a new duty-free shop.

There is also a group called innovations on the verge of RP which presents two features first described for Estuary English. The following features are on the verge of being acceptable as part of General RP: vocalization of dark [1] and glottalization of/t/ before an accented vowel and before a pause (see examples above).

Methods of collecting data in phonetic research are of great interest, as they may reveal different aspects of sound change in progress.

First of all, the direct observation method which, we can assume, has been most commonly used by the British scholars, starting from Daniel Jones, to attest the current trends in the speech of their contemporaries (see above).

A more objective method, though actually registering not the linguistic facts, but only speakers' attitudes to them is Pronunciation Preference Survey (originally called Opinion Polls). It was carried out by J.С. Wells in 1998 and continued through the next five years at least. About 2000 people, native speakers of British English, answered a questionnaire covering about a hundred items of uncertain or disputed pronunciation. The results showed people's preferences, although not their actual pronunciations. They provided a basis for the ordering of variants in dictionaries and ex­hibited strong associations between the variant preferred and the respon­dents' age which, as is well known, may serve as good evidence of language change in progress (Wells 1999:1245-1248).

J.C. Wells accounted for the choice by the fact that a number of words investigated in the survey were the subject of much popular discussion. For example, in RP garage is traditionally called ['дэега:з] or ['дэегэёз]; popularly, though, it is ['дэепс1з]. In the survey, the proportion of those preferring initial stress plus strong second vowel was 56%, a weak second vowel 39%, and final stress 5% (normally associated with American pro­nunciation). The ['gaendj] variant was particularly favoured by the young (66%) and particularly disfavoured by the old (13%).

The findings of the survey suggest that the trend may be showing itself in different phonetic contexts in varying proportions, and is more readily ac­cepted in familiar words than in more learned ones. For example, the /tj/ change to /tj/ is more readily accepted in tune [tju:n] than in perpetual and situation. Thus the process lasting for centuries makes progress quicker before an accented vowel in a more common word, and was, therefore, voted for by the young. (In his oral presentation at the conference J.C. Wells commented that for himself he would have selected the unaccented variants of affricates in perpetual, situation, whereas the accented position, and tune in particular, sounded vulgar to the professor).

The results of the Survey illustrated by graphs and figures may be found in the Longman Pronunciation Dictionary (LPD) (Wells 2000). Thus, for instance, we can find that the word graph, in the opinion of 60% British speakers should be pronounced with the vowel /a:/ but another 40 % of speakers prefer /ae/ in that position. Young people's voting for /ae/ in the word chance, instead of/a:/, and for /t>/ in won instead of/л/ is viewed by the author as a decline in deference towards RP as a model, both in the South of England and in the North.

Other factors to explain the change were the influence of morphology, when the change in pronunciation is considered to be an attempt to regularize it, to bring into line with other cases, e.g. youths [ju:3z] become voiced as paths, booths; the influence of spelling may be illustrated by the wordsybre- head [f o:hed] selected by the young, and restaurant ['restrtmt] in general use. Nevertheless, there are plenty of cases which are unaccountable, and American influence may appear to be a way of old forms restoration, not an advanced forms introduction. This may be the case of stress shift to the suffix -ary, which used to have a strong form in earlier periods of British English history, and has been preserved as such in American usage: young British speakers choose a secondary (tertiary) stress in 'necessary, 'ordinary in adjectives and a full stress on the suffix in adverbs yoluritarily, prdi'narily.

Some data on the American usage from the American English 1993 poll panel preference are included in LPD: greasy is pronounced with /s/ by 86% and with /z/by 14% of respondents. Incidentally, the British usage is not commented upon: /s/ variant also comes first, and /z/ second.

Word stress placement is another variable: in the word hospitable in British 1988 survey 81% prefer the [rms'pitabl] with the stress on the second syllable and 19 % on the initial one ['hrjspit9bl].A similar case of stress shift: in applicable British'84% preference is on the side of [э'рккэЫ] and 16% on the side of ['эерЬкэЫ]; the American selection is just the opposite: 64% for ['aeplik-] and 36% [a'plik-].

The above-adduced examples show that the two varieties may be converging in one feature and diverging in another, with either the American or the British leading. For instance, the British are leading in the shift of stress from 'comparable to com'parable, the Americans are in the lead for 'harass changing to ha'rass (only the young people seem to prefer it in U.K.).

There maybe a still more objective method of evaluating people's actual pronunciation variants through recording their speech and acoustic analysis, i.e. processing data with a computer program. The Telsur (Telephone Survey) project carried out at the Linguistics Laboratory of Pennsylvania University was designed to investigate vowel change in progress. It resulted in mapping major dialect regions of the U.S.A. and describing their defining features for "The Phonological Atlas of North America". It provided ample data on vowel change processes going on in different directions in the North and the South of the country.

There were 607 speakers, with 2 subjects for each area, and 4 to 6 for larger metropolises, who had agreed to answer a set of questions over the phone or in a face-to-face interview. The basis for dialect divisions was the impressionistic coding and acoustic analysis of 238 vowel systems. Age groups data served as a source of demonstrating the direction and the stage of sound change in progress.

Several analytical tools are instrumental in dealing with such complex systems; there are programs, for instance, for plotting, analyzing and superimposing vowel systems in a two-dimensional, computerized display. The two dimensions correspond to the two formants of vowel spectrum: Fl — the vertical dimension, F2 — the horizontal dimension. The plot of an individual vowel system looks like a vowel chart. Age, sex and class divisions may be grouped for comparison between the mean (averaged) point where a vowel is realized, and the deviations from that point made by speakers of different age groups. As a result, we can see a vector with the beginning point showing the old group data and the ending point — the young people's data (the average point is somewhere in the middle). It's a picture of sound change in apparent time.

The analysis of vowel /эе/ fronting and raising in the speech of the native speakers in Philadelphia has shown that among the internal factors which affect the progress of change is the nature of the consonant which follows (voiced or voiceless, i.e. lenis or fortis) and the lexical occurrence of words: the words bad, mad, glad are treated separately as the most advanced group.

A more expensive and time-consuming method is following the same people in one or, more impressive, two decades time, and repeating the same questions in an interview with an aim of finding individual and group changes. With the British corpus it was done by Peter Trudgill, a well-known socio-linguist, in a small city of Norwich and by Australian research workers with Xmas greetings by Queen Elizabeth II recorded over 40 years. Even the Queen, it turned out, is liable to change her accent from the highly conservative U-RP closer to a more democratic BBC announcers' pattern.

Дата добавления: 2014-01-11