1 Oxford Graduate Seminar, 12 th November 2007 Phonological innovation in London teenage speech: ethnicity as the driver of change in a metropolis Paul.

1

Oxford Graduate Seminar, 12th November 2007

Phonological innovation in London teenage speech: ethnicity as the driver

of change in a metropolis

Paul Kerswill†, Eivind Torgersen† and Sue Fox‡†Lancaster University, ‡Queen Mary, University of London

2

Or …

“New contact varieties as the source of innovation in a highly levelled, and still levelling, dialect area”

3

Innovation, levelling and diffusion These are three basic mechanisms of change.

• Innovation:– not predicated on contact – endogenous in the

sense of ‘generated from within the speech community’

• Levelling:– “… dialect levelling and by extension accent levelling, a process

whereby differences between regional varieties are reduced, features which make varieties distinctive disappear, and new features emerge and are adopted by speakers over a wide geographical area” (Williams & Kerswill, 1999:149)

– by definition non-directional– predicated on face-to-face contact (but not always)

• Diffusion– the directional spread of a feature– similarly predicated on face-to-face contact (again

not always)

4

Interaction of ‘internal’ and ‘external’ factors

Neogrammarian change: slow, subconscious, in principle governed by internal factors

Labov’s Principles of Vowel Shifting are intended as universal, and govern Neogrammarian change for vowels:

Principle I• In chain shifts, long vowels rise.Principle II• In chain shifts, short vowels fall.Principle IIa• In chain shifts, the nuclei of upgliding diphthongs

fall.Principle III• In chain shifts, back vowels move to the front. (Labov, 1994:116)

5

Drift

• We’ll look at an example of a set of Neogrammarian vowel shifts

• Such shifts seem to be susceptible to drift-like behaviour– a shift process, once started, can

continue in a new speech community even after separation

►What effect do non-internal (contact and non-linguistic) factors have on drift-like changes?

6

Finding a testing ground for the interaction of internal principles

and external factors• Insight from dialectology: a metropolis is the

supposed origin of change• A Western metropolis is usually the location with

most immigration and in-migration in its region

• Influence of non-internal effects likely to be high due to (i) language contact and (ii) complex intergroup relations

• Related to this is the likelihood of finding new L1 varieties of the language following contact with L2 varieties through individual bilingualism. These new varieties are contact dialects

• Possibility of innovation resulting from contact with these varieties

7

Dialect levelling (“supralocalisation”) in the

south-east of England• Reports of widespread homogenisation

in the south-east (Kerswill & Williams 2000; Britain 2002)

• New features are assumed to originate in London, based on gravity model (diffusion)– cf Wells (1982: 302): ‘its working-class accent is

today the most influential source of phonological innovation in England and perhaps in the whole English-speaking world.’

►Hypothesis: the new, ‘levelled’ features spread out from London

8

A problem with the gravity model

• the gravity model assumes spread by diffusion, not levelling

• if we observe gradually increasing homogenisation with no directionality, then this can’t be the result of diffusion

• (the partial exception would be where diffusion has run its course, leading to complete replacement – but directionality should be visible while the diffusion is ongoing)

9

London and three “South-east periphery”

towns

10

Regional dialect levelling (“supralocalisation”) in the south-

east of England• Reduced amount of H-dropping (’ouse)• Increased amount of TH-fronting (fing,

bruvver)• GOAT-fronting to []• “RP” variant in MOUTH []• Low-back onset of PRICE [],

lowered/unrounded from [ɪ], [ɔɪ] or [ɒɪ]• Raising of onset of FACE to [ɛ ̝̝̝̝ɪ]• Fronting of GOOSE to []• Fronting of FOOT to [] or []• Lowering and backing of TRAP to []• Backing of STRUT to []

11

We will focus on …

• Reduced amount of H-dropping (’ouse)• Increased amount of TH-fronting (fing)• GOAT-fronting to []• “RP” variant in MOUTH []• Low-back onset of PRICE [],

lowered/unrounded from [ɪ], [ɔɪ] or [ɒɪ]• Raising of onset of FACE to [ɛ ̝̝̝̝ɪ]• Fronting of GOOSE to []• Fronting of FOOT to [] or []• Lowering and backing of TRAP to []• Backing of STRUT to []

12

… four “diphthong-shift” vowels

• Reduced amount of H-dropping (’ouse)• Increased amount of TH-fronting (fing)• GOAT-fronting to []• “RP” variant in MOUTH []• Low-back onset of PRICE [],

lowered/unrounded from [ɪ], [ɔɪ] or [ɒɪ]• Raising of onset of FACE to [ɛ ̝̝̝̝ɪ]• Fronting of GOOSE to []• Fronting of FOOT to [] or []• Lowering and backing of TRAP to []• Backing of STRUT to []

13

… and two monophthongs undergoing change

• Reduced amount of H-dropping (’ouse)• Increased amount of TH-fronting (fing)• GOAT-fronting to []• “RP” variant in MOUTH []• Low-back onset of PRICE [],

lowered/unrounded from [ɪ], [ɔɪ] or [ɒɪ]• Raising of onset of FACE to [ɛ ̝̝̝̝ɪ]• Fronting of GOOSE to []• Fronting of FOOT to [] or []• Lowering and backing of TRAP to []• Backing of STRUT to []

14

Diphthong shift (Wells 1982)

But note that /u:/,

or GOOSE, now falls outside

the Diphthong Shift set …

… and this is

allowed for by Wells

15

Drift in the diphthongs of early New Zealand English (Trudgill

2004)• NZE has Cockney-like diphthongs today, but with

more extreme shifts in MOUTH• Trudgill finds evidence that diphthong shift got

greater during the 19th century, and concludes that this is due to drift.

• Britain (2005) argues that the evidence for continued shifting is only likely for FACE

• Either way, diphthong shift clearly thrived and then stabilised, in the absence of the strong social sanctions against it in south-east England at the same time

► Research question: what is happening to drift in London today, a typologically very similar variety of English, but where the sociolinguistic set-up is extremely different from early and current NZE?

16

Reduced H-dropping in the South-east periphery and a northern

English city

0

20

40

60

80

100

Elderly Boys Girls

Milton Keynes

Reading

Hull

17

Changes in MOUTH and PRICE

18

Percentage use of variants of /au/ (MOUTH), Reading Working Class,

interview style (1995) (from Kerswill & Williams 2005).

[a

]

Survey of English Dialects (SED) informants, 1950-60s

Elderly age 70-80 (2f, 2m) 53.5 38.1 3.3 0 4.1 0.7

Girls age 14 (n=8) 0 2.3 0 8.0 0 90.4

Boys age 14 (n=8) 3.8 3.2 0 5.7 0 87.1

19

Percentage use of variants of // (MOUTH), Milton Keynes Working Class,

interview style (1995)

SED informants, 1950-60s

Elderly age 70-80 (2f, 2m) 63.2 25.6 9.8 0 1.2 0

Girls age 14 (n=8) 0 0 0 5.9 4.7 88.8

Boys age 14 (n=8) 0 0 0 12.3 3.8 83.1

20

Percentage use of variants of /ai/ (PRICE), Reading Working Class,

interview style

Elderly age 70-80 (2f, 2m) 0 12.4 47.8 21.8 1.7 15.7

Girls age 14/15 (n=8) 2.8 21.2 45.1 21.1 4.3 5.1

Boys age 14/15 (n=8) 0.6 19.1 63.7 13.7 2.7 0

21

Percentage use of variants of (a) (PRICE), Milton Keynes Working

Class, interview style (1995)

Elderly age 70-80 (2f, 2m) 0 0 24.4 56.6 15.3 3.4

Girls age 14/15 (n=8) 25.4 44.6 29.2 0.5 0 0

Boys age 14/15 (n=8) 1.0 38.0 60.0 0 0 0

22

MOUTH and PRICE in the South-east

• MOUTH: simultaneous replacement of various regional forms through the south-east, both rural and urban, by [aʊ]– very rare in south-eastern vernacular varieties– very similar to traditional Received Pronunciation– not a phonetically levelled form, i.e. not arrived at as

either the survival of a majority form or the appearance of a phonetically intermediate form

• PRICE: the rise of [ɑɪ], which is not RP, but is a phonetically intermediate variant– good candidate for phonetic levelling – and also

geographical (non-directional) dialect levelling

23

GOAT: Male born 1915, Reading (r. 1996).

200

300

400

500

600

700

800

900

10001200140016001800200022002400

F2

F1

START

TRAP

GOAT

GOOSE

FLEECE

24

GOAT: Male born 1981, Reading (r. 1996).

200

300

400

500

600

700

800

900

10001200140016001800200022002400

F2

F1

GOOSE

FLEECE

GOAT

TRAP START

25

Phonological/phonetic change in London

• the fate of h-dropping• MOUTH• PRICE• GOAT• FACE

26

Research question: Is this city the origin of all these changes?

27

Are these the innovators?

Roll Deep Crew (East London hip-hop crew)

28

Investigators: Paul Kerswill (Lancaster University)Jenny Cheshire (Queen Mary, University

of London)

Research Associates: Sue Fox, Arfaan Khan, (Queen Mary,

University of London)Eivind Torgersen (Lancaster University)

LinguisticLinguistic innovators: innovators: the English of adolescents the English of adolescents in Londonin London (2004–7) (2004–7) Multicultural London Multicultural London English: the emergence, English: the emergence, acquisition and diffusion of acquisition and diffusion of a new variety a new variety (2007(2007–10)–10)

E· S· R· C

ECONOMIC

& S O C I A L

RESEARCH

C O U N C I

L

Funded by the Economic and Social Research Councilwww.ling.lancs.ac.uk/activities/278/www.ling.lancs.ac.uk/activities/539/

29

Research question 1: innovation

•What evidence is there that phonological and grammatical innovations start in London and spread out from there?

30

Research question 2: multilingualism

• One-third of London’s primary school children in 2001 had a first language other than English. Does this degree of multilingualism have any long-term impact on ‘mainstream’ English?

Reinterpreted in terms of the current spoken English of the capital, this becomes:

• Does the use of a putative Multicultural London English by adolescents lead to language change?

31

Research question 3: the innovators

•Which types of Londoners, socially (including ethnically) defined, innovate linguistically?

32

Research question 4: inner vs. outer London as

sources of change• Inner and outer London boroughs differ in: – ethnic profile– proportion of recent migrants– non-first language English speakers– socio-economic class

• Is there evidence that different linguistic features, including innovations, are characteristic of inner London vs. outer London?

33

Research question 5: social factors

• What social mechanisms facilitate (1) innovation and (2) diffusion?• social network• ethnicity• gender• identity

• Operationalisation of these social factors

34

HaveringHackney

35

36

37

Languages spoken

Hackney Turkish 10.61%

Yoruba 6.79%

Bengali + Sylheti 5.41%

Havering Panjabi 0.36%

Hindi/Urdu 0.32%

Gujarati 0.09%

38

Population

•Hackney: 208,365

•Havering: 224,248

39

Project design

• 16 elderly Londoners• 105 17 year old Londoners• from inner London (Hackney) and

outer London (Havering)• female, male• “Anglo” and “non-Anglo”• Free interviews in pairs• 1.4m words transcribed

orthographically, stored in a database time-aligned at turn level

40

H-dropping

MK &

Reading elderly (1995)

MK

14 year olds

(1995)

Reading

14 year olds

(1995)

Hackney

17 year olds

(2005)

Havering

17 year olds

(2005)

92% 14% 35% 9% 32%

Percent ‘dropped’ H in lexical words (interviews)

1. Correspondence between MK and Hackney is very surprising, because MK is highly mobile with a very ‘levelled’ accent, while Hackney is not mobile with an accent with many innovations.

2. Correspondence between Reading and Havering less surprising: both are areas with fairly mobile populations and somewhat levelled accents

41

Monophthongs in Hackney – anticlockwise chain shift

Elderly speakers (circles), Young speakers (diamonds)

2500 2000 1500 1000 500800

650

500

350

200

F2 (Hz)

æ

e

u:

:

æ

e

u:

:

42

Monophthongs: groups of speakers in Hackney

Elderly speakers (circles), non-Anglo speakers (inverted triangles), Anglo speakers with non-Anglo networks (triangles), Anglo speakers with Anglo networks (squares)

2500 2000 1500 1000 500800

650

500

350

200

F2 (Hz)

æ

e

u:

:

æ

e

u:

:

æ

e

u:

:

æ

e

u:

:

Anglos with Anglo

network

Anglos with non-

Anglo network

Non-Anglos

FOOT is relatively

back compared

to Havering – see next

slide!

43

Monophthongs in Hackney and Havering: the extremes

æ

Non-Anglo Youth, Hackney

æ

Anglo Youth, Havering

FOOTFOOT

GOOSE

GOOSE

44

Working-class white Londonerborn 1938 (Hackney)

300

400

500

600

700

800

5007009001100130015001700190021002300

F2

F1

CHOICE

GOAT

PRICE

START

STRUTTRAP

MOUTH

FACE

45

Young speakers in Hackney

300

400

500

600

700

800

5007009001100130015001700190021002300

300

400

500

600

700

800

900

1000

500700900110013001500170019002100230025002700

F2

F1

Laura, Anglo300

400

500

600

700

800

5007009001100130015001700190021002300

F2

F1

Alan, Kuwait

300

400

500

600

700

800

900

1000

500700900110013001500170019002100230025002700

F2

F1

Grace, Nigeria

Jack, Anglo

46

Young Havering Anglo speakers

300

400

500

600

700

800

900

1000

500700900110013001500170019002100230025002700

F2

F1

300

400

500

600

700

800

5007009001100130015001700190021002300

F2

F1

Donna Ian

47

Innovation, diffusion and levelling revisited

Loss of H-dropping• London matches London periphery in

loss of H-dropping– unexpected match between inner-city non-

Anglos and high-contact south-east periphery Anglos in Milton Keynes (a New Town)

– same feature – different social embedding– in south-east periphery, high mobility may

lead to susceptibility to overt norms (h-fulness)

– in London, may be a result of high contact with L2 varieties of English (which may be h-ful)

48

Fronting of GOOSE• Advanced in London, matching periphery

– GOOSE in London is rarely diphthongal in our data, so falls outside Diphthong Shift

– unexpectedly, most advanced among non-Anglo Londoners and Anglos with non-Anglo networks

– as with loss of H-dropping, the same feature has different social embedding in inner London and south-east periphery

– extreme fronting among inner city non-Anglos is innovatory

– levelling in periphery

Fronting of FOOT• Less advanced in London than in periphery

– in London, more advanced in Havering (outer city), in line with the Anglos in the periphery

– lack of fronting in inner city is conservative, matching Caribbean Englishes

– levelling in periphery

49

GOAT• (1) GOAT-fronting

– Prevalent among south-east periphery speakers – levelling (shared innovation). Agnostic as to Diphthong Shift reversal

– Absent in most inner-London speakers of both sexes and all ethnicities, present in outer-city girls

• Instead, (2) GOAT-monophthongisation– highly correlated with ethnicity (Afro-Caribbean,

Black African) and multi-ethnic network (for Anglos)– monophthongisation: a result of innovation in the

inner city, resulting from contact with British Caribbean English and L2 Englishes. No general diffusion except to minority ethnic speakers outside the inner city

– looks like Diphthong Shift reversal

50

PRICE• Lowering across region – Diphthong Shift

reversal• But added fronting is greater in London than

south-east periphery– fronting and monophthongisation correlated with

ethnicity – strongest among non-Anglos– seems to be a geographically directional and

diachronically gradual process– The change (from approximately [ɔɪ]) involves

lowering of the onset – and as such is a reversal of Diphthong Shift

– interpretable as a London innovation with diffusion to periphery

51

• Monophthongisation of FACE, PRICE and GOAT is correlated with four interacting scales:

1. Non-Anglo > Anglo2. Non-Anglo network > Anglo network3. Male > female4. Inner London > outer London > South-east

periphery (Milton Keynes, Reading, Ashford)

The nature of the interaction is not yet clear

52

MOUTH

• In the south-east periphery, the RP-like realisation [aʊ] has made inroads

• In London, [a:] is the norm

• Additionally, [ɑʊ] is used by some non-Anglos, especially girls, in the inner city

• RP-like [aʊ] is not the result of ‘levelling’ in the sense of the selection of a majority or phonetically intermediate form, but may be seen as socially more unmarked

• But the outcomes suggest three different changes:

– (1) south-east periphery [aʊ]

– (2) inner-city [a:]

– (3) inner-city non-Anglo [ɑʊ]

53

Contact, innovation, diffusion and levelling in dialectology

(1) Overall patterns:– divergence/innovation in inner London– non-Anglos and Anglos with non-Anglo

networks in the lead in innovation– some evidence of diffusion to south-east

periphery– but also levelling in periphery, without

involvement of inner London– Havering lies between inner London and

periphery

54

(2) Locus of contact in dialectology

– In modern metropolises new contact varieties result from language contact following large-scale concentrated immigration

– Transmission of innovations through social networks can be demonstrated quantitatively (harder to show in individual cases!)

– Contact varieties have the potential to spearhead language change, given the right social relations and favourable identity factors

55

(3) Where does contact not count?

– Transmission is said to be dependent on face-to-face contact

– But there is evidence that this is not necessary:

• th-fronting in Great Britain (θ f; ð v) up to about 1980 was geographically gradual and very slow (250+ years)

• Since then it has spread in a manner that cannot be explained by face-to-face contact and is no longer geographically gradual

– becoming increasingly mainstream in North of England and Scotland simultaneously in about 1980 (Kerswill 2003)

– spreading to low-contact working-class speakers first (Stuart-Smith et al. 2007)

• the spread of [aʊ] in the south-east periphery is rapid and simultaneous, and is not a typical automatic result of levelling as predicted by Trudgill (majority and/or intermediate form wins out)

56

(4) We need to account for the spread of features by face-to-face contact and absence of contact

– Milroy (2004; 2007) suggests an accessibility hierarchy, with a number of features being available ‘off the shelf’. th-fronting is one of them

– Observation suggests that some of the new vowel features are adopted outside London, but mainly by minority ethnic speakers – is this because of Trudgill-style levelling, or are the identities they signal not (yet) available to Anglo youth outside London?

57

Contact, levelling and diffusion in relation to Neogrammarian

change• Briefly: taking the long view, we can see

that the Diphthong Shift reversal we have observed is consistent and ‘regular’, even partly mirroring the order in which it is thought to have progressed in the first place

• But the social and phonetic detail is extremely messy

58

Innovation, levelling and diffusion revisited

• Little that we have discovered flatly contradicts the predictions of the gravity model, provided that:

• We recognise that different features have different social values (social indexation)

• We recognise some salience-like concept (not discussed here!)

• We recognise that ideology and identity must be added to face-to-face contact

59

Consequences for dialectology

• Sources of innovation must today be sought in minority-ethnic metropolitan varieties

and:• need to recognise a more complex

diffusion and levelling model

60

BibliographyBritain, David (2002). Phoenix from the ashes?: The death, contact,

and birth of dialects in England. Essex Research Reports in Linguistics 41: 42-73

Britain, David (2005). Where did New Zealand English come from? In A. Bell, R. Harlow & D. Starks (eds.), Languages of New Zealand. Wellington: Victoria University Press. 156-193.

Cheshire, Jenny, Fox, Sue, Kerswill, Paul & Torgersen, Eivind (in press) Ethnicity, friendship network and social practices as the motor of dialect change: linguistic innovation in London. Sociolinguistica 22, Special Issue on Dialect Sociology, edited by Alexandra N. Lenz and Klaus J. Mattheier.

Kerswill, Paul (2003). Dialect levelling and geographical diffusion in British English. In D. Britain & J. Cheshire (eds.), Social dialectology. In honour of Peter Trudgill. Amsterdam: Benjamins. 223-243.

Kerswill, Paul, Torgersen, Eivind & Fox, Sue (2008fc) Reversing ‘drift’: Innovation and diffusion in the London diphthong system. Language Variation and Change 8(3).

61

Kerswill, Paul, & Williams, Ann (2000). Creating a new town koine: Children and language change in Milton Keynes. Language in Society 29:65-115.

Kerswill, Paul, & Williams, Ann (2005). New towns and koineization: Linguistic and social correlates. Linguistics 43:1023-1048.

Meyerhoff, M. & Niedzielski, N. (2003). ‘The globalisation of vernacular variation’, Journal of Sociolinguistics 7(4): 534-555.

Milroy, L. (2004). ‘The accents of the valiant. Why are some sound changes more accessible than others?’ Plenary lecture given at Sociolinguistics Symposium 15, University of Newcastle upon Tyne.

Stuart-Smith, Jane, Timmins, Claire & Tweedie, Fiona (2007). ‘Talkin' Jockney’? Variation and change in Glaswegian accent. Journal of Sociolinguistics 11 (2), 221–260.

Torgersen, Eivind, & Kerswill, Paul (2004). Internal and external motivation in phonetic change: Dialect levelling outcomes for an English vowel shift. Journal of Sociolinguistics 8:23-53.

Trudgill, Peter (2004). New-dialect formation: The inevitability of colonial Englishes. Edinburgh: Edinburgh University Press.

Wells, John C. (1982). Accents of English. Cambridge: Cambridge University Press.