lllllllllllill. I -l I I I

Wm

:wrWS

)(l
\(Qv \

II
‘9

. .' #147?“

V1
1.1

This is to certify that the

dissertation entitled

African Americans in Lansing and the Northern
Cities Vowel Shift: Language Contact and
Accommodation

presented by
Jamila Jones

has been accepted towards fulfillment
of the requirements for

Ph.D. , Linguistics
_ degree in

 

 

 

 

V Mafiprofessor

Dme May 12, 2003

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

LlBFiARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

A850 82’”

 

T "L— 7"; :r E; 7:. m r q
‘ ‘ {J k

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DateDue.p65-p.15

 

AFRICAN AMERICANS IN LANSING AND THE NORTHERN CITIES
VOWEL SHIFT: LANGUAGE CONTACT AND ACCOMMODATION
By

Jamila Jones

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Linguistics and Germanic, Slavic, Asian, and African
Languages

2003

ABSTRACT

AFRICAN AMERICANS IN LANSING AND THE NORTHERN CITIES
CHAIN SHIFT

By
Jamila Jones
This dissertation examines the nature of dialect contact and the
role that pronunciation plays in the preservation and evolution of an
African American identity. The primary focus involves the acoustic
analysis of audio-recorded speech samples to determine the degree of
accommodation of a representative sample of African American

respondents to step one (/a:/ raising) of the Northern Cities Chain Shift

(NCCS). Linguists have asserted that African Americans generally do
not participate in the vowel changes that affect the White English
vernaculars in the United States. However, most studies have been
conducted in large urban areas or Southern rural areas while vowel
studies of medium sized communities have been neglected. This study
closes this vacuum by examining the vowel systems of African
Americans in a mid-sized city in the inland North, one of the purported
NCCS dialect areas.

Acoustic analysis was conducted on vowels read in citation form
to explore the effect of gender, age, social status, and network

relations on the height and duration of /a/. Results indicate a strong

correlation between vowel height and gender and duration and gender.

The use of Northern influenced front vowels and Southern influenced

back vowels indicates that social history and social cultural choices

have affected pronunciation.

Copyright
JAMILA JONES
2003

DEDICATION

To my son Ahmad, who was patient through it all.

"ACKNOWLEDGMENT"

I would like to express gratitude to my advisor, Dennis R.
Preston, for the support, expertise, and assistance, which he granted to
me over the years. Geneva Smitherman, Denise Troutman, and David
Dwyer's courses helped me to understand the issue of hegemony as
well as some of its linguistic ramifications and I thank you all. I
would also like to thank the Department of Linguistics and German,
Slavic, Asian, and African Languages. I must also extend appreciation
to Michigan State University's Statistical consulting office for help
with the statistical models used to calculate correlation of linguistic

and social effects.

vi

TABLE OF CONTENTS

List of Tables

List of Figures

Chapter One

Introduction

1.0 Introduction
1.1 Lansing, Michigan
1.2 The Problem Statement
1.3 Purpose of the Study

1.4 Boundaries of the Study

1.5 Hypotheses

Chapter Two

Background to the Study

2.0 Introduction
2.1 AAE Vowel Systems
2.2 The Northern Cities Chain Shift
2.3 Accommodation Theory

2.4 Speech Accommodation

2.5 Sociolinguistic Studies of Accommodation in

Production

vii

xi

xiii

11

12

I4

15

15

22

24

26

27

2.6 The Philadelphia short (a) Pattern
2.6.1 Tense and Lax Vowels
2.6.2 King of Prussia
2.6.3 African Americans in Philadelphia ..

2.7 First Dialect Acquisition in Detroit

2.8 Related Studies

2.9 Race

2.10 Chapter Summary .........

Chapter Three
Methodology

3.0 Introduction

3.1 Data Collection Procedures
3.2 The Gating Experiment
3.3 Data Analysis
3.4 Original Dialects
3.5 The NCCS
3.6 Duration

3.7 Summary

Chapter Four

Results

4.0 Introduction

4.1 Raising and Fronting

viii

30

30

32

33

36

37

39

46

47

48

54

55

60

67

72

72

74

74

4.2 Social Factors
4.3 Age and Class
4.3.1 Network Relations
4.4 Adjacent Segments
4.4.1 Following Manner of Articulation
4.4.2 Preceding Place of Articulation
4.4.3 Following Place of Articulation
4.5 Duration
4.5.1 Social Effects
4.5.2 Gender
4.5.3 Age and Class

4.5.4 The Effects of Adjacent Segments

4.6 Summary ................
Chapter Five

Qualitative Discussion

5.0 Introduction
5.1 Regional Accommodation
5.2 Social Factors
5.3 NetE Scores
5.4 NetS Scores ..
5.5 Initial Hypotheses...........................

5.6 Final Comments

Appendices

ix

78

81

84

88

89

96

98

98

100

100

101

102

104

105

105

106

117

118

126

131

Appendix A:
Appendix B:
Appendix C:
Appendix D:

Appendix E:

Appendix F:

Appendix G:

References

Indices of Social Class
The Questionnaire
The Linguistic Environments
Debriefing Script

F1 and F2 /2e/ Scores

ANOVA Summary ......

Individual Vowel Plots

134

137

139

142

143

174

175

207

LIST OF TABLES

Table 2.0 similarities between SAAE and SWE
Table 2.1 Differences between SAAE and SWE
Table 3.0 Distributions of Respondents ..............
Table 3.] F2 lae/ Fronting
Table 3.2 F1 /m/ Raising

Table 3.3 F2 /0/ Fronting

Table 4.0 Respondents
Table 4.1 Respondents with Index scores of "2"
Table 4.2 Raisers according to Gender ..............

Table 4.3 Non-raisers according to Gender ..

Table 4.4 Raising according to Age ..

Table 4.5 Non-raising according to Age...............
Table 4.6 Raising according to Class
Table 4.7 Non-raising according to Class
Table 4.8 Ethnic Network.................................
Table 4.9 Distribution of NetE Scores
Table 4.10 Distribution of NetS Scores........

Table 4.11 Coding for ANOVA

Table 4.12 Thomas's ANOVA Results .

Table 4.13 Mean Scores for Thomas's Following

xi

21

21

50

66

66

66

76

77

79

79

82

82

84

84

85

85

87

89

92

92

Manner of Articulation

Table 4.14 Ranking of Mean F1 /&/ formant values....

Table 4.15 Rank Order for Following Manner of

Articulation

Table 4.16 Coding: Preceding Place of Articulation

Table 4.17 Ann's ANOVA
Table 4.18 Duration for Michigan speakers

Table 4.19 /a=:/ Tokens

Table 4.20 Coding: Place of Articulation

xii

94

95

96

98

99

100

103

LIST OF FIGURES

Figure 3.0 Rachel's digitized signal and spectrogram ......... 57
Figure 3.1 Vowel Plot for Rachel 58
Figure 3.2 Mean Vowel Formants for Rachel 59
Figure 3.3 Normalized DARE 63
Figure 3.4 Peterson and Barney (1952) Vowel Plot men 64

Figure 3.5 Peterson and Barney (1952) Vowel Plot Women.. 65
Figure 3.6 Vowel Formants for Advanced NCCS Speaker .. 68

Figure 3.7 Mean Vowel Plot for Advanced NCCS Speaker .. 69

Figure 3.8 Average Formants for Michigan Women ........... 70
Figure 3.9 Average Formants for Michigan Men 71
Figure 4.0 Percent of Total Respondents.......................... 75
Figure 4.1 Gender 78
Figure 4.2 Age 81
Figure 4.3 Class 83
Figure 4.4 Thomas's /2e/ Distribution 91
Figure 4.5 Following Mannner of Articulation for Thomas... 93
Figure 4.6 Ann's /ae/ Distribution 97
Figure 5.0 Mean Formants for Henry 121
Figure 5.1 Mean Formants for Mandy 122
Figure 5.2 Mean Formants for Curtis 123

xiii

Chapter One

1.0 Introduction

This dissertation reports on research on the nature of dialect
contact among 31 African American residents of the Inland North. The
study examines their accommodation to the pronunciation of the
Northern Cities Chain Shift (NCCS), an ongoing vowel rotation taking
place in such cities as Buffalo, Detroit, Chicago, and Cleveland.
Linguists (e.g., Thomas 1997, 2001, Labov 1994, Wolfram and
Schilling-Estes 1998, Henderson 1996) have asserted that African
Americans are not participating in this vernacular change; however,
most such studies have been skewed, concentrating mainly on male,
working class African Americans in segregated portions of larger inner
cities. These studies have also tended to present African American
Vernacular English (AAVE) as the only representative of African
American English (AAE), ignoring the fact that there is considerable
linguistic variation among African Americans in terms of region, social
class, sex, and age (Stockman and Newkirk 2000). Moreover, studies of
the phonology of African Americans in mid-Sized cities are rare.

The first chapter of this study includes a general introduction, a
brief history of Lansing, an outline of the problem, purposes of the
study, an overview of the methodology, the limitations of the study,

and the hypotheses. The second chapter surveys the previous

sociolinguistic literature on low front vowel raising, accommodation
theory, and linguistic subjugation, including an elaboration of the
hierarchical, hegemonic, nature of race in the United States. Chapter
Three delineates the methodology for the investigation and describes
the procedures used to conduct the acoustic and statistical analyses.
Chapter Four presents the results, and Chapter Five summarizes the
conclusions and discusses qualitative implications.
1.1 Lansing, Michigan

Lansing, Michigan was founded in 1829 and settled in the late
18305, largely by White settlers from New York State. The fifth census
of the United States indicates that there were thirty-two slaves in
Michigan in 1830. However, the sixth census does not indicate whether
slavery existed in Lansing, Michigan or even in Ingham County in
1840. In fact only one free Black man between the age of 10-24 was
recorded in Ingham County by the 1849 census. The 1870 census
reports 77 African Americans residing in incorporated Lansing and also
indicates that Michigan African Americans were originally from a
variety of states: Michigan, Virginia, Ohio, Kentucky, and Indiana. The
African American population grew gradually, and, according to Meyer
(1970), prior to World War I, the majority of Lansing's Black
population originated from Michigan, nearby Northern states, Canada,

Kentucky, and Virginia. After 1915, the majority of African American

in-migration to Lansing came from the Deep South states of South
Carolina, Georgia, Alabama, Mississippi, Arkansas, and Tennessee.

Meyer (1970) states that early African Americans were dispersed
throughout Lansing neighborhoods until the West Side Black residential
area crystallized between 1915-1939, becoming the recognized "Negro
area" in west central Lansing. By 1950, the Black area had expanded
westward toward West Street and eastward toward Pine Street. There
was also a cluster of Blacks living south of Kalamazoo Street on the
east Side of the city. By 1960 (112) “the major spatial expansion of the
Black core area occurred northward from Lenawee to Michigan Avenue
and westward to Jenison Street in the north and to the city limits in the
south.” The Oldsmobile plant and the Grand River formed the boundary
for the Black area. Some decentralization of the Black community
occurred as a result of the [-496 highway corridor through Lansing in
1961-1969, Oldsmobile expansion in 1961-1970, and the state Capitol
project in 1966 (Meyer 1970, Hawkins 1979). Although many Blacks
chose to relocate in Black neighborhoods, there was some movement
beyond the traditional Black neighborhoods.

In 1990, Blacks formed 19% of the reported population (127,321)
for the city of Lansing and 7% of the reported population (50,677) for
the city of East Lansing. In the 2000 census, Blacks comprised 21.9%
of the reported population (119,128) for the city of Lansing and 7.4%

(46,525) for the city of East Lansing. The present—day Lansing African

American community is progressive enough to have two Afro-centric
charter schools, a street named after Martin Luther King Junior, and
Black leadership at various levels in government and educational
institutions.

The linguistic environment in which Lansing African Americans
find themselves is, from a traditional dialect point of view, part of the
Inland North (e.g., Labov 1996), but this study will concern itself with
the accommodation (or lack thereof) of African American speakers to
one feature of the so-called Northern Cities Chain Shift, a rather recent
development in the urban areas of the Inland North.

1.2 The Problem Statement

Before 1990, the concept of race as used by the United States
Census Bureau was based on the extent of "Negro" ancestry. The
concept used by the Bureau in the 1990 and 2000 censuses reflects self-

identification. The Census Bureau states:

The concept of race as used by the Census Bureau reflects self-identification;
it does not denote any clear-cut scientific definition of biological stock. The
data for race represents self-classification by people according to the race with
which they most closely identify. Furthermore, it is recognized that the
categories of the race item include both racial and national origin or socio-

cultural groups (Census of Population 199028-11). During interviews

conducted by census enumerators, if the person could not provide a single race

response, the race of the mother was used (l990:B-11).

The present era, sometimes referred to as ‘post-modern,’ has
precipitated changes in racial outlook and racial counting as a result of
the civil rights movement, affirmative action, and responses to both.
We live in a reputedly color-blind society in which African Americans
are no longer necessarily confined to delimited areas of cities and equal
opportunity is the law. In spite of the legality of movement, which
seemingly grants individuals the right to choose their personal
lifestyles and make individual decisions with regard to self-
identification based on professional, regional, and ideological
connections, most cities have Black areas and White areas, and studies
by sociologists reveal that "Black and White Americans live in separate
worlds and often do not speak the same language" (Feagin and Sikes
1994:320, Benjamin 1991). Black conservatives and progressives agree
that institutional racism exists but differ in terms of responses to it.
Some, for example, believe that race should be ignored while others
believe that racism has to be aggressively opposed.

Lois Benjamin (1991) studied the coping strategies of a group of
elite African Americans and reported that even the Black elite tended to
socialize primarily with other Blacks in the 19905. Is it indeed the case
that hegemonic separation of identity covers all aspects of life,

including language? Blacks and Whites expect certain identity patterns

and are surprised and nonplused when behavior patterns run contrary to
"color." One cannot deny that there have been many societal changes
in the past thirty years that point to changed conditions for a sizeable
portion of the Black population, although the number of incarcerated
Black men and continued deprivation deep in inner cities continue to
plague society. The question asked here is whether enough has changed
to alter the caste like apartheid system which influences how one
speaks.

Stereotypes have an unquestionably deep and abiding staying
power, and studies of regional standards have yet to fully confront the
myth of General American English (Hagiwara 1997, Preston 1993,
Niedzielski and Preston 2000) and the question of the homogeneity of
the phonological system of AAE. I believe that regional studies of the
entire range of AAE will help in exploding or confirming the myth of a
homogeneous AAE phonological system.

Whether linguists choose to view linguistic variation from the
perspective of communities of practice (Eckert 2000), social network
(Milroy 1980), inherent polarity between prestige and stigmatization
(Labov 1972, Trudgill 1986, Preston 1993), or linguistic acts of
identity (Tabouret-Keller 1985), there is a great deal of hegemonic
maneuvering within the United States hierarchical social structure, and
there is apparently more reflection among some people who were once

not allowed choice in foregrounding particular aspects of regional

identity. The challenge is to ascertain whether changes in racial
classification, living space, and the law have affected network
relations, merely camouflaged the status quo privilege of
institutionalized privileged groups, or indicated that power and
privilege are shared in demographics of cities in the United States.
Therefore, my research question concerns whether accommodating to
the regional standard on the part of African Americans in Lansing is
among the things that have changed. Is it possible that African
Americans in Lansing are seeking a more local identity by
accommodating to the NCCS?

Irvine and Gal (2000:37) remark that "it has become a
commonplace in sociolinguistics that linguistic forms including whole
languages can index social groups." This has certainly been the case
with African American Vernacular English (AAVE) formerly known as
Black English Vernacular (BEV), used by Labov (1972:xiii) to refer to
"the relatively uniform dialect spoken by the majority of Black youth in
most parts of the United States today, especially in the inner city areas
of New York, Boston, Detroit, Philadelphia, Washington, Cleveland,
Chicago, St. Louis, San Francisco, Los Angeles, and other urban
centers. It is also used in rural areas and in the casual intimate speech
of many adults." Labov goes on to say that the term "Black English" is
not suitable for this dialect since the phrase implies a dichotomy

between Standard English on the one hand and Black English on the

other. He says that "Black English" might be best used for the whole
range of language forms used by Black people in the United States: a
very large range indeed, extending from the Creole grammar of Gullah
spoken in the Sea islands of South Carolina to the most formal and
accomplished literary style.

I use the term African American English (AAE) in this extended
sense to include the entire range of the repertoire of African American
ex-slave descendants. This means that a speaker of AAE who may use
the syntax of the language of wider communication, but the phonology,
lexicon, and prosody of AAE will still be an important part of the range
of ex-slave descendants raised by Black people in myriad Black
communities in the United States. This means that syntax is not the
only basis for inclusion or exclusion from the ranks of Black linguistic
authenticity, but pronunciation and vocal quality can be, under the right
contexts, as important as syntax as identity markers from my
perspective. Lippi-Greene (1997:177) states that "upper-middle-class
Blacks may seldom or never use grammatical features of AAVE, but
such persons are often heard marking their language in a variety of
ways to signal solidarity with the greater African American community.
This may mean the use of AAVE intonation, tag questions, and address
systems, or more subtly, rhetorical features and discourse strategies."
This definition draws on Smitherman's (2000: 251) explanation of

African American Verbal Tradition (AVT) and Spears' (2001: 240)

elucidation of Standard African American English (SAAE). Baugh's
(1999) discussion of pronunciation as a salient social marker that
realtors use to direct the flow of housing in cities proves that fine
phonetic differences are indeed social markers, which can include and
exclude people from in and out-groups. In spite of the security that the
language that Black Americans speak provides them, what Irvine and

1

Gal (2000) term ‘erasure’ has rendered AAE speakers as "others,’ non-
participators in the dialects of wider communication in the regions
where African Americans live. "Erasure is the process in which
ideology, in simplifying the sociolinguistics field, renders some
persons or activities invisible. Facts that are inconsistent with
ideological schema either go unnoticed or get explained away. So, for
example a social group or language may be imagined as homogeneous,
its internal variation disregarded” (Irvine and Gal 2000:38). This is
surely the case with regard to statements that African Americans are
not participating in the vernacular changes affecting the White
vernaculars in the same region. It has become expedient to accept
certain aspects of AAE as more authentic than some others. Morgan
(1998, 1994) provides a detailed discussion from the perspective of
members of the Black language community concerning the issue of

authenticity versus slavishness in the use of AAVE. In the end, the

speech community itself dictates its forms based on its social history.

In the past, African Americans living in the South sounded a
great deal like their Southern White neighbors, although there were
differences between them. Therefore, it is certainly conceivable in
modem (or post-modern) times that Black residents of medium-sized
cities such as Lansing and small cities such as East Lansing may adapt
some of the features of their neighbors, and that this adaptation may
have its own internal variation.

Irvine and Gal (2000:37) also add, "linguistic features that index
social groups or activities appear to be iconic representations of them,
as if a linguistic feature somehow depicted or displayed a social group's
inherent nature or essence." Spears (2001:242) and Baugh (1999) note
that the language of Black people is stigmatized because Black people
are stigmatized in American society. Lippi-Greene (1997:178) contends
that African Americans' language is "tangible and irrefutable evidence
that there is a distinct, healthy, functioning African American culture
which is not white and which does not want to be white." In spite of the
health of the Black community and its institutions, decentralized Black
communities may be more heterogeneous than in previous centuries as
African Americans navigate a hierarchical system that is apparently no
longer Black versus White, but also includes dimensions of shades of
brown and yellow which promise possible alliances or restructuring,
and this decentralization challenges assumptions of authenticity. Does

movement away from self-contained, geographically located Black

10

communities into self maintained Black alliances encourage adaptation
to the phonology of the regional standard? Do African Americans
maintain a Black linguistic identity when they no longer reside in the
geographical space designated as African American?

These concerns are related to the specific problem in this
research: whether African American in-migrants to the inland North
from the South after World War II or their descendants, differing from
the majority population in terms of ethnicity and original dialect, have
adapted to the vernacular of the region (NCCS). This study focuses on
step one of the NCCS — low front vowel raising.

1.3 Purpose of the Study

The purpose of the study is to describe the degree of low front
vowel raising or its absence among 31 African Americans residents of
the Greater Lansing area. The description will include an analysis of

the status of lae/ raising as it pertains to social and linguistic factors.

The study will also explore possible qualitative explanations for
variation within the sample studied here. The most immediate purpose
of the study is to seek to ascertain whether African Americans in this
section of the Midwest have adapted to the emerging vernacular pattern
of the area (e.g. the NCCS). The present study takes a 31-member

sample and measures the height and duration of /m/ (the low front

vowel) based on acoustic analysis. Vowel height and duration patterns

are correlated to social status, age, gender, and network and to place,

11

and manner of articulation to see if there is an effect for any of these
social and linguistic factors.
1.4 Boundaries of the Study

The study will be limited to low front vowel raising because it
has been posited as the first step of the NCCS (Labov 1994; 1996,
Eckert 2000) for the majority of the White population involved in the
shift. LPC analysis was chosen over impressionistic analysis, and the
investigator has analyzed the respondents' vowels within the context of
the other vowels in the system. An acoustic analysis also makes it
possible to compare vowel charts of different respondents within a
study and between studies. The motivation for investigating duration
lies in the fact that accommodation to one feature does not preclude the
presence of other acoustic characteristics, which might differentiate
AAE speech from the wider vernacular. Other acoustic characteristics,
which might cause differences in vowel quality, (e.g., direction and
extent of diphthongization) are left for future study. Duration was
chosen because most African Americans are from the South, and longer
duration is a well-known feature of Southern vowel pronunciation (e.g.
Thomas 2001).

The respondents were all born or raised in the Greater Lansing
area. They were obtained through a variety of means, but, essentially, I
followed Milroy's (1980) friend of a friend methodology. An African

American professor at Michigan State University introduced me to

12

some of her fellow church-members. An African American colleague
introduced me to a Black barber who in turn introduced me to people
who introduced me to other individuals who introduced me to family
members. Additionally, I was able to contact respondents at several
houses of worship. However, final selection of the persons actually
studied was based on the quality of the spectrograms that the tapes and
mini-discs produced.

Linguistic differentiation and engagement with the regional
standard are complex phenomena, which are tied to an individual's
social history (Labov 1980), linguistic ideology, social network (Milroy
1980), and reaction to the hegemonic forces which are arrayed against
him or her (Morgan 1998). It is not possible to uncover all the factors,
which cause an individual to choose one particular pattern over another
pattern that he or she has been exposed to at home, in school and in the
wider society. A sociolinguistic interview by its very nature can only
provide a snapshot of a particular moment in time when a respondent
reacted to the stimulus (i.e. interviewer and tape recorder) in a
particular manner to produce the data which drive a study such as this

one.

13

1.5 Hypotheses

The major hypothesis of this study is that African Americans in
the Greater Lansing area have engaged in institutions in the wider
community and had sufficient networking opportunities in the broader
community to accommodate to the early stages of NCCS (i.e. low front
vowel raising); therefore, their accommodation will mirror that of
majority speakers in terms of social and linguistic influences. It is also
posited that individuals with open personal networks will accommodate
more than individuals with dense closed personal networks (Milroy
1980:20). On the other hand, it is also hypothesized that African
Americans will retain some Southern vowel features such as greater
duration, which will serve to differentiate African American speech
from Northern White speech, and respondents who show local (i.e.,
neighborhood and ethnic) loyalty will accommodate less than
individuals who do not (Preston and Ito 1998).

The remaining chapters characterize the collection of low-front
vowel data from these respondents and the acoustic and statistical

analyses which seek to establish these predicted patterns of behavior.

14

Chapter Two
Background to the Study

2.0 Introduction

Chapter two includes a review of the literature regarding African
American accommodation to the NCCS, issues concerning
accommodation theory, dialect contact, language ideology, and the
filtering effect of race and culture on pronunciation.
2.1 AAE Vowel Systems

Although there is a certain amount of avid fascination (Morgan
1998, Van Keulen et al. 1997) associated with Black language, on the
part of Whites at least, European Americans and African Americans
appear to regard the different linguistic levels of AAE somewhat
differently. Whites are critical of the phonology, grammar, lexicon, and
rhetorical practices of AAE, while most African American criticism is
directed toward grammar (Niedzielski and Preston 1999). For example,
if such elements such as subject-verb agreement are intact, speech is
generally considered standard in the Black community. Well-known
celebrities such as Oprah Winfrey appear to base Standard English on
just such features as subject-verb agreement (Lippi-Green 1997).
Phonology, and what Smitherman (2000:251-67) terms African
American Verbal Tradition (AVT), which involves African American

prosody and ways of speaking, are not areas that African American

15

speakers criticize in the same manner that European Americans criticize
the entire range of AAE features and practices.

On the other hand, if an individual sounds Black, based on any
phonetic or grammatical cue, that may be sufficient cause for
stigmatization on the part of some European Americans (e.g., Baugh
1999). In spite of that, the continued use of AAE phonology and
rhetorical style in the public forum by prominent African Americans
such as Clarence Thomas, Jesse Jackson, the late Malcolm X, the late
Martin Luther King Jr., Oprah Winfrey, Cornel West, Spike Lee, and
other African Americans of stature proves that African Americans view
at least AAE phonology and ways of talking differently from European
Americans. In fact, the deliberate public use of Black phonology has
implications for language ideology, suggesting that Black phonology is
a characteristic of Black identity that is not easily (or perhaps
willingly) abandoned.

Ash and Myhill (1986) have demonstrated that African
Americans who move in White circles show a major shift in their
grammar in the direction of White norms and a lesser shift in
phonology and lexicon. However, Whites who move in Black circles
show a greater shift in the Black direction in terms of phonology and a
lesser shift in terms of syntax. Wolfram et a1. (1997) focused on the
speech of a single member of the only African American family on the

isolated island community of Ocracoke off the outer banks of North

16

Carolina and demonstrated that this solitary speaker showed greater
alignment to AAE in terms of phonology and lexicon and lesser
alignment in terms of syntax and morphology. These studies serve to
reinforce the salience of pronunciation as an identity marker among
Black people.

Thomas (2001:161) and Bailey and Thomas (1998292) note that
little work has been done on the vowel systems of African American
English (AAE) speakers, making it a seriously neglected area of

research. The few exceptions include reports on glide shortening in /aI/
(so that “time” sounds like “Tom”) and /01/ (so that “soy” sounds like
“saw”), the merger of /8/ and /1/ before nasals (so that “pin” and “pen”

are both pronounced “pin”), and the reversal of tense and lax front
vowels (so that, for example, the vowel of “it” sounds like “eat” and
that of “bet” sounds like “bait”). But, as I will show later, these
features are general in much of Southern American English and not at
all exclusively African American. In fact, and more important to this
study, too few studies have investigated the linguistic and social factors
influencing vowel choice or changing vowel patterns in urban Black
Northern communities to be able to truly say that African Americans
are not participating in Northern vowel changes. In addition, research
concerning vowel usage by other than working class AAE speakers has

received slight attention.

17

I review here the few studies of Southern African American
vowel systems. Juanita Williamson's (1968) description of the speech
of 24 African Americans in Memphis, Tennessee covers sixteen vowels.
Words were transcribed phonetically based on an impressionistic
analysis. Since the focus of my study is on the first step of the NCCS, I

will examine only her characterization of /d/ and /a=./. She said that the
phoneme /o/ of pot, rock, college occurs most frequently as a low-
central vowel, which frequently has a short upglide [0“] and an even

more diphthongal allophone before voiced consonants. She found that

in disyllabic words (e.g., college, cottage (cheese), vomit), /a/ usually
occurs as a monophthong, although it is sometimes lengthened [0']. She
found that /2B/ always occurred as a diphthong [ae'3 , m' ] before /k, g/ as

in sack, bag. The distribution of these diphthongs appears to be

sensitive to social factors as well. [we] is used more often by well-
educated, highly cultured people, while the other respondents use [m'].
/a°:/ may also occur as this upgliding diphthong before /s, j, tj, and n/
(e.g., ask, ashes, catch, dance). /a:/ usually occurs as a monophthong
in polysyllabics (e.g., casket, January, pasture). laB/ before intervocalic

/r/ (e.g., marry, carried, parents) is usually a monophthong, although

sometimes lengthened. A diphthong [2e' , ma] also occurs sporadically

l8

in these words. In short, Williamson has said that /2e/ is sometimes

diphthongal in some environments and for some social groups.

Bailey and Thomas (1998), Thomas and Bailey (1998), and
Thomas (2001) have also examined Southern African American English
and have assembled a database of mechanical recordings from African
Americans whose dates of birth range from 1844 to 1984, including
Caribbean speakers and ex-slave recordings. Unfortunately, their
sample does not include areas of the South where most Lansing
speakers' parents might have resided, and only one of Thomas' (2001)
respondents comes from a Northern state. Most of their speakers come
from Texas and North Carolina with few from other Southern states.
Nonetheless, their work shows vowels analyzed acoustically and plotted
on F1, F2 axes as in the present study, a technique which allows
researchers to compare formants and vowel system configurations for
different individuals and groups of speakers. They concluded that by
the end of the nineteenth century some of the distinctive features of the

African American vowel system were as follows:
1) Fully backed (or “nonfronted”) /u/(cooed), /U/ (foot) and /o/

(boat)

2) non-fronted onset of /au/ (house)

3) shortening of the offglide of /a1/ (ride) before voiced

consonants

l9

4) raising of /ae/.

Thomas (1997) also notes that /ae/ is usually raised to /e/ and that /e/
and /I/ correspondingly tend to be shifted upward and fronted as well.
The merger of /(l/ and /o/ (the vowels of cat and caught) is rare among

African Americans.

If in fact the raising of /2e/ is a characteristic of Southern African
American speech, it will do us little good to see if that feature has been
adopted by African Americans in the Lansing (or other NCCS areas)
since they already have it in their parent system. A comparison of the
Thomas and Bailey (1998), Bailey and Thomas (1998), and Thomas
(2001) vowel plots with NCCS plots, however, clearly shows that the

/a:/ raising pattern of the NCCS (see section 2.4) is distinct from

Thomas' Texas, North Carolina, and other Southern vowel patterns.

Although /m/ fronts in both patterns, /8/ and /I/ back and lower in the
NCCS, but not in the SS. It also appears that /ae/ is fronted and not

raised in most of Thomas's plots. As will be clear later the
configurations are quite different. Most importantly, I will Show from
vowel plots of older African American speakers from those areas of the
South which were the primary inputs to the Lansing population are not
characterized by the /a3/ raising other investigators have suggested.

This position will be outlined in detail below.

20

The following two tables adapted from Bailey and Thomas (1998)

point to differences and similarities between Southern White English

(SWE) and AAE vowel systems.

Table 2.0: Similarities between the vowel systems of Southern

 

 

AAE and SWE

Feature AAE SWE Emergence
Merger of /8/ and /1/ before nasals + + 1875-1940
Glide shortened /aI/ before voiced + + 1875-1940
obstruents

Merger of tense and lax vowels before /1/ + + 1900-1940
Merger of /o/ and /o/ before /r/ + + 1900-1940

 

Adapted from Bailey and Thomas (1998:105)

Table 2.0 indicates that AAE and SWE share a series of conditioned

vowel mergers such as /8/ and /1/ before nasals (pen and pin), tense and

lax front vowels before /1/ (fill/feel), and /o/ and /o/ before /r/

(horse/hoarse). The differences are explored in table 2.1.

Table 2.1: Differences between the Southern AAE vowel system and

SWE vowel system

 

 

Feature AAE SWE Emergence
Non-front onsets of /au/ + - Before
1860
“Back” back vowels + - Before
1860
Onset of /e/ as low as /m/ - + After 1900
Glide shortened /ai/ before voiceless - + After 1900
consonants
Merger of /o/ and /(l/ - + After 1900
Onset of /o/ lowered and fronted - + After 1900

 

Adapted from Bailey and Thomas (1998:105)

21

Some of the SWE features, which are posited as different from AAE
(Table 2.1), are parts of what has come to be known as the Southern

Shift (SS). In the SS, the lax front vowels /I/ and /e/ are moving

upward and taking on the gliding quality of tense vowels. For example,

/8/ takes on a glide and becomes more like [81]. The front tense vowels,

/i/ and /e/ move back and downward, retaining their diphthongal quality

so that /i/ sounds like [e1] and /e/ sounds like [a1]. The back vowels,
/u/, /U/ and /o/ are moving forward, and the onset of /dI/, which is

monophthongized in some environments, is also moving forward
(Wolfram and Schilling-Estes 1998, Labov, Yaeger and Steiner 1972,
Feagin 1986, Fridland1999).

2.2 The Northern Cities Chain Shift

The raising of /2e/ is considered the first step in a series of vowel

changes characteristic of the Northern cities of Detroit, Buffalo,
Chicago, Cleveland, Milwaukee, and Rochester. Lansing, Michigan is
also participating in these changes (Labov 1996). Although the raising

of the nucleus of /a:/ in words such as bad, ask and dance in New York

City was studied in Labov (1966), the NCCS was first explicitly
identified by Fasold (1969) in an unpublished paper which investigated

the raising of /a':h/, the fronting of /d/, and the fronting of /o/ among 24

speakers from the Detroit survey of Shuy, Wolfram and Riley (1966,

cited in Labov 1994:178). Callary (1975) found that the height of /2e/

22

is directly correlated with the size of the community; the larger the
community, the more raising exists. Labov (1994) has described the
NCCS as a chain shift based on a concept concerning causal movement
of the phones of particular phonemes first explicated by Martinet
(1952). A shift occurs when phonetic properties of a phoneme change,
causing the phoneme or its allophones to enter the vowel space (i.e.,
F1, F2 position) of another phoneme, thereby, leaving “vacant” vowel
space which may then be entered by another nearby phoneme. Wolfram
and Schilling-Estes (1998:138) summarize the NCCS succinctly, if not

in order. "For example, a vowel like the /o/ in coffee is moving forward
toward the /G/ offather. The low vowel in a word like pop or lock, in
turn moves toward the /m/ of bat, which in turn, moves upward toward
the vowel [e] of bet. At the same time, another rotation moves the short
vowel [I] of bit toward the [e] of bet. The [8], in turn, moves
backward toward the [A] vowel of but, which is then pushed backward."
Although Wolfram and Schilling-Estes begin their description with /o/.

Labov (1994) and Gordon (1997: 24) order the movement as follows:
Changes nearing completion:

1. /a=:/ (bad) is fronted (tensed) and raised to /e/, bed, or even
[I] (bid). It is sometimes accompanied by an inglide, [8°] or [1°].
/zc/ raising is characterized as the oldest change.

Midrange changes:

23

2. /o/ (pot) is sometimes fronted as far as [a]

3. /o/ (bought) is lowered, fronted and unrounded to approach

[a]
New and vigorous changes:

4. /I/ is sometimes lowered to the position of [e].
5. /8/ is backed to [A]
6. /A/ is backed and often rounded, resulting in variants near [0].

These vernacular changes are occurring in the European American
speech community, and if Speech Accommodation Theory (SAT) or
Communication Accommodation Theory (CAT) extends to members of
other ethnic groups, African Americans residing among European
Americans in Lansing may have adapted some of the vowel changes,
which are part of the emerging speech patterns of residents of the urban
Inland North dialect area.
2.3 Accommodation Theory

In this section, I will briefly discuss accommodation theory and
support the discussion with a number of empirical sociolinguistics
studies involving speech accommodation.

African American pronunciation is sometimes studied in a
vacuum, meticulously comparing subordinated and segregated African
Americans to assimilated European American descendants of even

recent immigrants, who both desire and are permitted and expected to

24

assimilate, and never realize that accommodation for one group could
be construed as subordination for another. No doubt aspects of
subordination involve accommodation, just as speech differentiation
sometimes involves stigmatization. For example, regional and ethnic
speech differentiation exists among the Jewish, Italian, and Irish speech
communities in Boston (LaFerriere 1979); however, these differences
have not always led to social stigmatization. It is also the case that the
regional differences of Appalachian speakers are sometimes cause for
ridicule (Evans 2001). Nonetheless, differences in pronunciation among
African Americans are associated with institutionalized discrimination
due to the fact that race is such a central part of social categorization
and social identity in the United States, and race cuts across ethnicity.
An ethnicity may be dropped or trivialized after two or three
generations in America, but unless a Black individual is able and
chooses to pass for White, his or her race is his or her ethnicity. In
short, accommodation does not make an individual White.

In any case, accommodation connotes adjustment that has not
been forced; subordination explicitly involves force, whether overt or
covert, denigration, and enforced change by the dominant group,
whether from schoolteachers or other elements of the society (e.g.,
employers, see Lippi-Green 1997). The forceful nature of subordination
is evidenced in the treatment that Black children and other subordinated

minority groups receive in schools in the United States, including the

25

high percentage of Black children in special education classes
throughout the school systems in the nation. (Lippi-Greene 1997:
Chapter 6, Van Keulen et al. 1997:138).
2.4 Speech Accommodation Theory

Accommodation theory, first known as Speech Accommodation
Theory (SAT) and subsequently expanded to Communication
Accommodation Theory (CAT), deals with the issue of accommodation
from the perspective of social psychology (Giles 1973, Giles et al.,
1991). Giles (1973) first used the terms convergence and divergence to
define a speaker's accent orientation towards an interlocutor. Giles
defined accent accommodation or change in these two directions.
Convergence occurs when a speaker, desiring to gain the social
approval of the receiver, reduces pronunciation differences and
attempts to adapt his or her accent to that of the receiver. However, if
the speaker chooses to disassociate himself or herself from the
interlocutor, he or she may diverge linguistically by emphasizing
pronunciation differences. This concept was later expanded to deal with
accommodation in terms of speech rates, pause phenomena, utterance
length, smiling, gaze and so on (Giles et al. l991:7). The terms
convergence and divergence also gained prominence in linguistic
circles when Labov and Harris (1986) and Bailey and Maynor (1987)
announced that Black vernaculars were diverging from White

vernaculars. Black linguists in particular asked for clarification of this

26

position, and American Speech (1987) printed a panel discussion held at
a national linguistics meeting concerning it. The issue of convergence
and divergence has ramifications for the United States as a nation and
for African Americans as a "nation within a nation." Although African
Americans are criticized for not accommodating enough to the language
of wider communication, there are other social groups that do not
readily accommodate albeit for slightly different reasons. An overview
of a few sociolinguistics studies related to accommodation in
pronunciation will illustrate this.
2.5 Sociolinguistic Studies of Accommodation in Pronunciation
Trudgill (198611-11) dealt with the way in which dialect contact
can lead to dialect change in speech communities. He looked at contact
between speakers of British English in the United States and US
English speakers and also suggested that it might be useful to study the
way in which British pop musicians imitated aspects of American
English pronunciation in order to determine salient aspects of American

pronunciation. Trudgill found that pop singers pronounced /aI/

monophthongally, used rhotic /r/, and pronounced vvords like body and

top with an unrounded [a] rather than the British [0]. /?/, which is the

pronunciation of intervocalic /t/ (better) that most working class British

speakers use, was avoided and the American [r~d], a voiced alveolar

flap, was used. For this study, the most interesting aspect of pop

singers' imitation of American English was the way they dealt with /2e/.

27

Pop singers disregarded their dialect areas and pronounced American

[an] as [22] even when their dialect dictated [a:] pronunciation.

Trudgill compared the imitation of British rock stars with the
accommodation of British expatriates residing in the United States. His
data was based on notes made on the segmental phonology of native
speakers of British English, informal observations at conferences and
lectures, and his own speech during a year's stay in the United States.
He found that British expatriates did not acquire the monophthongal

pronunciation of /a1/ because this feature is an imitation of Black or

Southern singers, not a general feature of American pronunciation.

On the other hand, Trudgill found that, although /r/ is salient for
British speakers, it was not easily accommodated to because of a
phonotactic constraint in British English which permits /r/ to occur

only before vowels. Trudgill also did not find the substitution of /d/ for
/o/. He says that adoption of this change would have led to the loss of

contrast between pairs such as: hot ~ heart, pot ~ part and cod ~ card,
etc. in his British dialect. He also felt that the relationship between

British English /I)/ and US English /0/ is not entirely direct. Some
words which in British English have /D/ have /o/ in US English (e.g.,
lost, long, off). Other words which have /0/ in British English have /A/
in US English (e.g., of, what, was). However, the use of /ze/, as in

dance, last, etc., is a change the British English speakers made early, if

28

they were going to accommodate to US English at all. British English

has /a=:/ in, e.g., ant, romance, so it is straightforward to substitute
/da=:ns/ for /da:ns/. Nevertheless, not all British speakers accommodate
to the US /2e/. For example, Trudgill remarks that it is too salient for

him, and I take that to suggest that it is such a strong marker of identity
that he felt compelled to resist it. In short, in spite of similarities in

race and class, Trudgill was unable to or unwilling to add US /a=./ to his

repertoire.

The realization of intervocalic /t/ as [d] is also accommodated to
early on by British speakers in North America. Neither the ethnicity nor
language of British speakers residing in the United States is subject to
stigmatization; therefore, the decision to accommodate may be made to
facilitate understanding, but not for any latent prestige issues.

There have been studies dealing directly with accommodation to
the vernacular form of AAE by people of color from other ethnic and
language groups, such as that of Puerto Rican accommodation to
features of Black English in New York City (Wolfram, 1973). First
generation Puerto Ricans accommodated to features of AAE based on
their extent of social contact with Blacks and the color of their skin,
which reflected the way the majority population treated dark Puerto
Ricans. In a similar manner, African French-speaking immigrants to
Ontario and refugee continental Africans identify linguistically with

Black Americans due to institutionalized hegemonic practices of the

29

dominant social group in Canada (Ibrahim 1999). These youths actively
seek to acquire Black English as a second language (BESL).
2.6 The Philadelphia short (a) pattern
2.6.1 Tense and lax vowels

One of the primary concerns of this dissertation is African
American accommodation to step one of the NCCS, which involves low

front vowel raising /a3/ raising), therefore, it is worthwhile to see how a
variety of social groups have accommodated to the pronunciation of /a=:/

in their region. Labov, Yaeger and Steiner (1972:41) describe the use of
[+tense] "as a classificatory feature in the abstract phonological rule
which selects certain short or lax vowels and differentiates them from

others by a variety of phonetic features." Lax vowels /I, 8, a, A, U/

(Ladefoged 1993:86) occur in stressed closed or CVC syllable final
position in Standard English. Tense vowels may occur in both closed
and open syllables. Stevens (1998:294-6) describes the vowels that are
intermediate between peripheral and the central schwa vowel as lax. He
says "these intermediate configurations are achieved by positions of the
tongue body and lips that are less extreme." The most extreme positions
are tense. Labov (1994: 175) differentiates tense vowels on the basis
of their being located closer to the periphery of the two or three
formant vowel space and by their relatively greater length and
amplitude. Ladefoged (1993:86) points out that lax vowels are paired

with tense vowels in tense lax pairs respectively: [i,I] as in beat, bit;

30

[e1, 8] as in bait, bet; and [u, u] as in boot, foot. He characterizes lax

vowels as shorter, lower, and slightly more centralized than the
corresponding tense vowels. Ladefoged (86) also says that there are no

vowels similar in quality to [m] and [A] in most forms of (Standard)

American English. However, the NCCS involves the tensing and raising

of [a3], and Labov, Yaeger and Steiner (1972241) have defined

[+tense] in their studies as vowels, which appear regularly with
extreme formant positions (relative to neighboring vowels). Although

Ladefoged (1996:) and Labov (1994) have referred to /m/ as a short or
traditionally lax vowel, Strange et al. (1983:698) categorize /ae/
acoustically as an intrinsically long vowel and group /ze/ with /e, (1, o/
in terms of length as opposed to /I, e, A, u /, which have been

characterized as the intrinsically short vowels.

In New York City and Philadelphia, studies of /ae/-raising show
that only tense or long /m/ is raised; in the NCCS area, however, Iae/
appears to be ubiquitously tense. Nevertheless, since studies in New
York and Philadelphia touch on accommodation, 1 will review some of

them.

31

2.6.2 King of Prussia

Avilla Payne's (1980) research offers a unique perspective on
second dialect acquisition by showing that certain complex
phonological patterns can be learned only from parents who are native
to a particular area. In spite of the fact that some children who had
moved to King of Prussia, Pennsylvania, a suburb of Philadelphia, had
accommodated to most of Philadelphia’s phonological system, they did
not acquire the short (a) pattern (i.e., /a:/) if both their parents were not
native to the area (Payne 1980). Payne's research was framed in
response to whether a child will learn to speak like peers or retain the
system learned from parents. The bare outline of the Philadelphia rule

is as follows: /2e/ is tensed before front nasals, voiceless fricatives, and

voiced stops when these are followed by an inflectional boundary or
another consonant. Children from another dialect area have to learn that

/m/ is lax before /j' /and /d/ except in the case of the "affective”

adjectives mad, glad, bad, which have to be learned as lexical
exceptions. For example, when children from the NCCS area moved
into King of Prussia, they had to learn to block application of the

tensing rule for part of a class /j, d/ and three lexical items because, as
noted above, the NCCS rule tenses /m/ in all linguistic environments

without lexical conditioning. Payne discovered that children who
moved to King of Prussia at an early age and lived between 5-8 years

there had more success than children who moved in at a later age.

32

Although all of the children in the study acquired the Philadelphia
pattern to some extent, unless a child's parents were locally born and
raised, the possibility of acquiring the Philadelphia /a=:/ was extremely
slight. To some extent the age result coincides with what researchers
know about foreign language acquisition, "the later in life subjects
begin learning English, the more strongly accented their sentences will
be judged to be by native English speakers (Flege l995:23)." Werker
(1995) and Kuhl (1995) suggest that infants' phonetic perception may
be tuned to properties of the native language as early as six months.
Since children learn pronunciation from primary care-givers and focus

on the point vowels, the three extreme vowels corresponding to [11],
most back —— tongue/ lips rounded; [i], most front — tongue/ jaw
closed/ lips spread; and [a] most pharyngeal/ jaw open (Pickett 1999:

43), the role of the primary care-giver should be particularly
significant.
2.6.3 African Americans in Philadelphia

Henderson (1996) also examined the Philadelphia Im/ pattern;

moreover, her study concentrated on 30 educated African Americans
adults who were well integrated into the Philadelphia White
community. Her respondents grew up and lived in predominately White
neighborhoods, attended White schools, had White friends and
socialized and worked with Whites; however, only six out of 30

respondents had acquired the Philadelphia /m/ pattern exactly. Ten

33

suburban respondents, generally men in the over- 40 category, showed a
great deal of similarity to the Philadelphia pattern; otherwise, the rest
of the sample had not acquired the pattern. The rest of the sample
tended to tense in the environments that were normally lax for most
Philadelphia White speakers and lax in the environments that were
normally tense. There was also more laxing than expected for the mad,
bad, glad adjectives because these words are categorically tense in the
White Philadelphia dialect. Three of the six who acquired the
Philadelphia pattern had grown up in the same suburb and attended the
same schools. Five of the six used to reside in the same suburb. Three
of the six were related, and one of the six, 3 well-educated attorney,
grew up in a nearby suburb and was the only Black in her graduating
class of one hundred and sixty-five students. Labov and Harris have the
following to say about the White vernacular sound changes in major
cities.

Research on the social origins of these sound changes in

the local White community indicate that the most advanced

patterns are to be found among the people with the highest

prestige; draftsmen, bank tellers, school teachers,

politicians, block captains and local influentials (1980). A

comparison of sound changes in many cities leads us to the

conclusion that they serve as symbolic claims to local right
and privileges. (1986:18).

34

Perhaps living in the suburbs for more than one generation also imbues
African Americans with an unconscious sense of belonging to that
particular community, and this sense is evidenced in their speech.
Although Henderson (1996) documented the social and psychological
isolation that her respondents experienced in those suburbs, six
acquired the complex Philadelphia pattern for reasons that Henderson
did not adequately explore. Labov and Harris (1986: 21) argue strongly
that social history, the kinds of social experiences that people have had
in dealing with members of other groups, and the way that they have
used language in their life, are actually more reliable predictors than

social network.

35

2.7 First Dialect Acquisition in Detroit
Toni Deser's study (1991) also tackled the question of the
relative weight of parental and peer influence on native dialect

acquisition. She chose to examine /2e/, /a1/, and duration among six

Black Detroit families taken from the Shuy, Wolfram, and Riley's

larger Detroit Dialect study. Raised /ae/ is, of course, the first vowel
involved in the NCCS, while /aI/ monophthongization and duration are

Southern dialect markers. In order to categorize the six families that
she studied into Northern, Southern or mixed dialect families, three
speech-language pathologists with phonetic training evaluated Deser's
speaker sample. According to Deser, the Northern dialect families

generally showed a raised pattern in which /a:/ is at the same level as
/e/. Although the Southern dialect families did not differ appreciably
from the Northern families, /2e/ raising was observed when young

females from the two dialect groups were compared on their /2e/ mean

scores for a reading task. Deser's results indicate that a small degree of

measurable /2e/ raising was produced by the Northern dialect girls. The

Southern speaking children used smaller vocal track configurations to

produce their /m/ and /8/ than the Northern dialect children. The three

Northern dialect families showed a pattern in which the youngest

family members produced the most Northern-like /2e/ targets, and

36

children with one Southern dialect speaking parent tended to produce
more Southern-like vowels than children with no Southern parent.

Deser concluded that monophthongal or diphthongal /dI/ was a

reliable differentiator of Northern and Southern dialect speakers.

Moreover, older siblings tend toward a more monophthongal /OI/ than

younger siblings. Adolescent girls tend to produce more monophthongal
variants than the rest of the sample. Deser also considered duration a
robust predictor of dialect affiliation. Vocalic duration was greater on
average for Southern dialect speakers, and, although differences were
small, they were consistent across the voiced/voiceless consonant
environment and across speaking styles (123). Deser argues that
although age was a factor (preadolescent versus adolescent), children
continue to be affected by their parent's dialect. Therefore, Deser's
study seems to support the notion that the input of the caregiver is
important in native dialect acquisition.
2.8 Related Studies

In a pilot study, Jones (1996) examined the NCCS participation
of adolescents in two mixed-race families (White mothers/ Black

fathers), specifically focusing on /2£/ fronting and raising. An

assumption of the study was that first generation mixed children might
have more networking opportunities within the White community than
other African Americans and might show greater alignment to the

NCCS — the local White vernacular. The results indicated that the girl

37

with the greater network in the White community raised /m/ higher than
/8/ and the other two mixed race children who resided in the Black
community raised only to the level of /e/. However, all of the mixed

race children showed Southern dialect influence among the back
vowels.

Although Edwards (1992) did not examine similar vowels in his
Eastside Detroit study, results of the Black English and White
vernacular comparison indicate that older informants (40+) are more
prone to choose BE variants than informants under forty.

Finally, Denning’s (1989) study of final /i/ in East Palo Alto
emphasizes that all the facts with regard to convergence or divergence
are not in yet, and that there are no last words on this issue. His study
indicates that younger Black speakers in East Palo Alto are
pronouncing the following vowels according to the local White

vernacular. /i/ is pronounced with greater height and frontness, /o/ is

pronounced more like a monophthong rather than the upgliding
diphthong of older Blacks, and there is neutralization of the differences

between /0/ and /(1/ as in hawk/hock. Denning's work is significant in

that it verifies accommodation to aspects of surrounding White

community speech.

38

2.9 Race

Evelyn Higginbotham (1992) presents an argument that race is
the metalanguage through which one must discuss the social constructs
of class and gender in the United States. Although the White founding
fathers eschewed the idea of a governmental monarchy and titled
aristocracy, America has always been troubled by the subordination of
people according to color, class, and caste. Race has been defined
scientifically and socially by a number of theorists and social
scientists. Zack (1998:74) states that according to biological
anthropologists, the racial unit is not an individual but a population
that has more of some physical trait than other populations. Although
physical anthropologists unite in their insistence that race is only a
social construct, and biologists agree that human variability between
the populations of Africa, Asia and Europe is no greater than variability
within those populations (Appiah 1998228, Smedley 2001), hair texture,
bone, and color still define race in America. Or as Zack, who pushes
for a mixed-race social category, (1998:75) says, "Race is what cultures
take it to be," but in America a drop of Black African ancestry used to
make a person Black. In spite of this fact there are individuals who
have chosen to evade this classification culturally (e.g., Tiger Woods),
and opt for a mixed race category. Cornel West (2001: 39) has defined

Blackness in America as a political and ethical construct. He says:

39

First, blackness has no meaning outside of a system of race-
conscious people and practices. After centuries of racist
degradation, exploitation, and oppression in America, being black
means being minimally subject to white supremacist abuse and
being part of a rich culture and community that has struggled
against such abuse. All people with black skin and African
phenotype are subject to potential white supremacist abuse.
Hence, all black Americans have some interest in resisting racism
—even if their interest is confined solely to themselves as
individuals rather than to larger black communities.
Defining race as a mere social construct does not lessen its effects,
although the term has bearing on another social construct —— ethnicity
— which is often confused with race or used as a synonym for race.
Race must be considered separately from ethnicity because
ethnicity can be optional or symbolic, perhaps adopted for certain
holidays by White Americans, but never a slot on an employment or
loan application, nor would an individual's choice of spouse, housing,
social group or success in the world likely be affected by his ethnicity
alone excepting dark-skinned Hispanics and to a certain extent Jews,
although in that case religion is a complicating factor (Watersl998:
403). Polish Americans, Italian Americans, Swedish Americans, and
Appalachians are all subsumed under one superordinate ethnicity ——
White. Perhaps one could argue in a similar fashion for Blacks if the
Somalis, Sudanese, West Indians, and Nigerians decide to pass on their
ethnic culture beyond the first generation; however, the thrust of the
argument remains the same — the Black-White divide is a racial divide

and has been an integral part of American culture as Higginbotham

(19922253) aptly characterized it, "arbitrarily contrived to produce and

40

maintain relations of power and subordination" —— Black subordination.
The other extreme of subordination is resistance -—- both conscious and
unconscious. This resistance is captured in another meaning of
ethnicity as defined by Van Keulen et al. (1997), who describe
ethnicity as cultural traits that are heightened by Black people to define
their ethnicity as African Americans. Black hairstyles, dress, food,
religion, and language are ways in which difference is heightened by
Blacks in America to celebrate and take pride in the culture that Black
people have made.

Historical examples of Black women who were denied the right
of womanhood and ladyship, and examples of Black US. senators who
were denied manhood outside their offices in Washington DC.
(Higginbothaml992) are numerous in historical documents and continue
to color class and gender experiences in the United States. It has been
sufficiently shown and continues to be documented that race, whether
biological or socially conceived, functions to make a Black person's
experience stereotypically and compellingly different from that of a
White person even in the let century.

Martin Robinson Delany was the first to coin the term "nation
within a nation" to describe Black America (Higginbothaml992),which
implies opposition and resistance as opposed to integration and
accommodation. For example, African Americans in the South under

Jim Crow were well aware of the way Southern Whites wanted them to

41

behave. Sometimes the decision to use something other than the
language of wider communication was actually accommodation to the
White perception of what the resisting party’s place should be. Morgan
(1998: 254) writes that Southern Blacks were penalized for educated
speech and were constrained to speak the way that Southern Whites
thought that they should speak. In other words, an African American
speaking Standard English could have affronted the sensibilities of a
Southern White causing the White to perceive the Black speaker as
engaging in a form of resistance to linguistic codes imposed by
Southern segregationists in the post plantation society. This Black
would be punished for acting (talking) "uppity" in opposition to the
Southern White linguistic code. Under different circumstances,
speaking vernacular AAE or other than the "Standard" approved
language could also be perceived as resistance to teachers, elders, and
societal expectations in the North. Kohl and Hinton (1972:119-120)
narrate a story in which a teacher was assaulted for insisting on calling
students by their legal names and refusing to call them by their
assumed nicknames of Akmir, Arkbar, and Rabu. Although European
Americans might give up the ethnic names which might differentiate
them from majority society, these African American youth moved in a
different direction and assumed names which were "intentionally
bizarre" in the context of the majority society (1972:120) to

differentiate themselves from it. This act was openly resistant to a

42

society which they deemed illegal (120). This intentional cultural
differentiation still exists today among hip-hop generation and youth
culture, which relates Black culture to Black ideology as artists assume
names like Boo-Yaa, Kurupt, Wu-Tang Clan and Sister Souljah to name
a few.

Linguistic resistance may also take a form characterized as
"inversion" in which Blacks appear to submit to White forms but do not
(Holt 1972:154). Baugh (2000:8) presents an illustrative example of
this when he narrates a boyhood incident in which he verbally insulted
a Hispanic student, and the teacher overheard it.

Teacher: John: Stop it.

JB: Hey man! He's hitting me. I ain't doing nothing.

Teacher: You're making fun of him.

JB: Yeah, but he's hitting me, I'm just talking.

Teacher: But you are making fun of the way he talks, so stop it.

JB: (shucking and jiving in my best rendition of exaggerated

Standard English) I'm very sorry, I didn't realize I was doing

anything wrong.

Teacher: Now, John, why don't you speak that way all of the

time and improve yourself?

Baugh (2000:8) states that his teacher failed to realize what his Black
peers sensed immediately; namely, his "overt attempt to mock the

teacher and Standard English with one blow." It is extremely difficult

43

to characterize the relationship between African American identity and
language because context is so pertinent to the issue of language
choice.

In fact, during different periods, Blacks have made various
aspects of difference salient, such as the natural hairstyle, reclaimed as
an emblem of beauty and political resistance during the 19605, and
Black language, made salient by hip hop artists in present day popular
culture. All of these identity markers are engagements in a hegemonic
ritual, which flaunts difference in acknowledgement that difference is
the symbolic battleground for the oppression of Blacks in America. If
difference is flaunted enough maybe it will be recognized and accepted
on its own terms. Cornel West (2001:6) characterized White America's
perspective on the Black "problem" in the following manner:

Hence, for liberals, black people are to be "included" and

"integrated" into "our" society and culture, while for

conservatives they are to be "well behaved" and "worthy of

acceptance" by "our" way of life. Both fail to see that the
presence and predicaments of black people are neither

addition to nor deflections from American life, but rather

constitutive elements of that life.

One assumes that West means that Black Americans are as
American as White Americans, and that this nationality is theirs
to no small degree. Blacks are not visitors or new immigrants, but
part of the land of their birth, and the land of sometimes

unmitigated toil on the part of their ancestors. African Americans'

identity irrespective of personal ideology is a product of a contact

situation and their personal response to social history. The flaws
are in the society: "rooted in historical inequities and
longstanding cultural stereotypes” (West 2001 :6). Nonetheless,
nationality, culture, and language continue to be viewed through
the prism of race, although aspects of Black American culture are
essential to American culture. Blacks are "other"; the other which
happens to be part of American culture. Yet one questions the
attitude of the individual people who make up the group about
race, language and culture. Surely their outlook is not
monolithic. Just as "Black leaders" do not share a uniform
ideology, it is not realistic to assume that Black people share an
unequivocally fixed worldview once one moves beyond the issues
of past wrongs, subjugation endured, and continuation of
hierarchical social structures. Does the use of language reflect
actual social beliefs? Does race still matter in a city such as

Lansing, Michigan?

45

2.10 Chapter Summary

Chapter Two focuses on the African American vowel
systems from the perspective of the paucity of acoustic analysis
studies generating vowel plots of total vowel systems. The issues
of language accommodation and perhaps language subordination
have been touched upon. In spite of the equivocal nature of the
research on the subject, the caretaker's role in dialect transmission
has been shown to be a viable one based on the results of a
number of accommodation studies. Acoustic phoneticians and
linguists view "earlier as better" with regard to language or
dialect acquisition. The studies regarding African American
accommodation to the NCCS in other cities confirm that
statements concerning lack of accommodation by AAE speakers to
NCCS are not comprehensive ones. Finally, race may be the
metalanguage or filter through which to examine language, culture
and ethnicity, and that filter will be used in the interpretation of

the results of this study.

46

Chapter Three
METHODOLOGY

3.0 Introduction

The methodology for this study is based on the sociolinguistic
truism that variation in the language of a given speech community is
not random but ordered (Weinrich, Labov, and Herzog 1968) and
correlated to the social variables of age, sex, social class, and ethnicity,
among others. The study also explores the way in which the density or
multiplexity of a social network may influence a group’s projections of
its social identities (Milroy 1980, 1992). Consideration was also given
to communication accommodation theory (CAT) which “proposes that
speech convergence reflects, in the unmarked case, a speaker’s or
group’s need (often unconscious) for social integration or identification
with another” (Giles et al., 1991: 18). Therefore, short-term
accommodation can [signal individual solidarity with a conversational
partner or, in the long-term, realignment of a group’s entire code,
possibly reflecting realignment of underlying beliefs, attitudes and
social structure (2). Nonetheless, subjugated groups, women, youths,
and ethnic minorities often openly resist linguistic domination by the
majority culture without divesting themselves of the ability to code-
switch and accommodate to the wider society’s public face if need be
(see Gal 1995, Abu Lughud 1989). However, this accommodation is

often only instrumental and does not imply a realignment of group

47

loyalty or personal identity. In this study I have presented the idea that
phonology is the level which speakers who are involved in acquiring
wider community norms hold on to as expressions of group identity.
Hudson (1996) suggests that “pronunciation reflects the permanent
social group with which the speaker identifies.” Most African
Americans presently residing in the Inland North differ from European
Americans in terms of original dialect, length of time in the area, and
of course racial ethnicity. Needless to say, these differences may have
affected African American's accommodation to the vernacular of the
dominant culture.

This chapter will describe the methodology used to explore
African American speakers' accommodation to the NCCS in Lansing,
Michigan. I will discuss data collection procedures and analysis,
selection of respondents, social demographics, and perception tests
undertaken to situate Lansing African Americans’ involvement or lack
thereof in the NCCS, a vowel shift nearly completed for most European _
American, long-term residents of urban, southeastern Michigan.

3.1 Data Collection Procedures

Sociolinguistic interviews were conducted for thirty-one male and
female African American respondents who read a one hundred and six
vocabulary word list and a reading passage (in appendix C) structured

to represent the vowels which are involved in the NCCS.

48

Respondents were born or raised in the Greater Lansing area, and
ranged in age between 19-74. They were contacted through "the friend
of a friend" method and recruited by way of friends from local
churches, mosques, and barbershops. Respondents were often asked to
recommend other respondents. Pseudonyms are used for all respondents
in this study.

The respondents selected for this study were divided into
“younger” (under 40) or “older” (over 40) age groups. Using a
modified scale based on Warner et al. (1960), the respondents were
divided into working and middle class groups. Most respondents were
interviewed in their homes so that the researcher could assess their
standard of living and neighborhood type, and, based partly on these
factors, divide them into working or middle class groups (see appendix
A for the worksheet). Scores 1-7 were assigned to the respondent’s
occupation, housing, and neighborhood with the lowest number
indicating the highest status, housing, education, and neighborhood.
Occupation was multiplied by four, education and housing by three, and
neighborhood by two. In this study, respondents who scored 20-50
were called middle class and those who scored 51-70 working class.
Respondents outside of those ranks were not considered. High school
students and non-working spouses received the same score as the
principal working member of the family. Table 3.0 shows the

distribution of the respondents according to sex, age, and social status.

49

Table 3.0 Distribution of Respondents

 

 

 

 

 

 

 

 

 

 

 

MIDDLE
OLDER
MEN WOMEN MEN WOMEN
3 6 4 3
WORKING
YOUNGER OLDER
MEN WOMEN MEN WOMEN
2 6 3 4

 

 

 

 

 

 

A respondent’s social network was determined by means of a scale
developed by Milroy (1980:20). Milroy describes closed, high density
social networks as those in which a given person’s contacts all know
each other or may even be linked to one another in more than one
capacity — as a co-employee, a kinsman, or friend. In contrast, an open,
low-density network is characterized by associations which are single
stranded. This type of individual associates with people in a single
capacity only, and a given person’s contacts will not know one another.

Scores were computed based on the following characteristic:

50

A: Membership in high-density territorially based network
B: Substantial kinship ties in neighborhood (more than one
household in addition to respondent’s own)

C: Work at the same place with at least two people from
neighborhood

D: Work at same place with at least two people from
neighborhood of the same sex as respondent

E: Associates extensively with people from place of work in

leisure time activities

Each respondent was assigned a score of one for each of the above
descriptions that held true. A respondent with a total score of zero
would be one with the loosest or weakest network relations; one with a
score of five would be one with the strongest network ties. This score
was designated NetS.

The purpose of the other portion of the social network
investigation was to determine the degree to which a speaker was
embedded in his or her ethnic networks (NetE). In this test, speakers
were asked to characterize the percentage of their close acquaintances
that were members of the same ethnic group. A score of zero indicated
that the speaker had no close association with members of his own

ethnic group, and a score of five indicated that a speaker had exclusive

51

 

association with individuals from his ethnic classification (see
appendix B question 7).

The association between social network and /m/ raising was

correlated to see if statistical significance could be determined. If tight
neighborhood network association suggests less correlation with wider
community linguistic norms, there should be a negative correlation
between the social network score (the higher the score, the denser and
more multiplex the respondent’s neighborhood network) and the lee/

6‘19,

raising index score ( = not raised, “2” = raised). A high network

score should result in a low /ae/ index score.

Finally, respondents were asked questions designed to obtain
details about the respondent's childhood, schools, hobbies, and
occupation in order to gauge speakers' attitudes toward Lansing and
their particular neighborhood. Pseudonyms are used in the following

sample questions and responses:

Sample # 1

Interviewer: Do you have substantial kinship ties in your
neighborhood? Meaning do you have relatives who live in the same
neighborhood that you do?

Jane: Umhum (yes).

Interviewer: Do you work at the same place with at least two people

from your neighborhood?

52

Jane: Yes

Interviewer: And do you work at the same place with at least two
people from your neighborhood who are of the same sex as you are?
Jane: Yes.

Interviewer: Do you associate extensively with people from work in
leisure time activities?

Jane: No.

Interviewer: Okay. I have one more question to ask you. So in your
neighborhood, do most of your friends know other friends? Are most
of your friends, friends of friends?

Jane: In my neighborhood or my friends?

Interviewer: Your friends.

Jane: Do most of my friends know my other friends? =

Interviewer: = Know your other friends?

Jane: Yes.

Interviewer: Okay.

Interviewer: Do you think —uh - this is a good place to grow up?
Lansing? Do you think that it’s a good place to grow up?

Jane: Um. I’d say yes.

Interviewer: Why?

Jane: Uh. It’s you know — if you look for ...There’s uh. There’s a lot
of... There’s recreational activities and there’s a lot of neighborhoods

where there’s a lot of other children growing up. You can grow up with

53

other children. It’s a (pause) family, lots of families uh, uh. What else
can I say?

Each interview was digitally recorded on a Sony MZ-R30
portable mini-disc recorder with a clip on external microphone or a
Sony TCM-SOOOV tape-recorder. Each session took approximately 30
to 60 minutes.

3.2 The Gating Experiment

The interview also included a perceptual experiment of the sort
referred to as "Gating Experiments." The one used here is a perception
experiment, which involved identification of the major vowels that are
shifting in the NCCS. For the particular purposes of my study, it
involved the degree to which comprehension of the NCCS is advanced
among African American respondents. In such experiments (Labov
1994:194-5), respondents guess the identity of a word, which may be
interpreted as an unshifted (e.g. socks) or the shifted form (e.g. sacks)
in the NCCS. In order to determine whether rural, Appalachian, and
African American speakers were able to identify the phonemic
categories of the vowels involved in the NCCS, my colleagues and I
asked students in an introductory class in the Integrative Studies in
Arts and Humanities Program at Michigan State University to listen to
a word list read by advanced NCCS respondents. Young, southeastern
Michigan urban respondents were teased out from the class and used as

the control group for the tape. Students wrote down the word that they

54

heard. They listened to the Gating tape on a one by one basis and were
graded according to how many words they recognized as the containing
the correct phoneme. Distractors were not included. All the phonemes
on Gating tape are involved in the NCCS. I asked the Lansing
respondents to listen to the same recording that the IAH class had
listened to and write the words that they thought they heard the
speakers read (see appendix F). The Lansing results were compared to
the Michigan State University southeastern Michigan students' results
and the results of the African American students in the IAH class.
There was no statistically significant difference between Lansing AA
speakers and the AA speakers in the IAH class, but there was
significant difference between the responses of the urban southeastern
Michigan White students and the Lansing AA speakers.
3.3 Data Analysis

The vowel tokens used to analyze the respondents' vowel systems
came primarily from the word list, with tokens occasionally
supplemented from the reading passage or conversation. Vowel
measurements were taken at the center or steady state of the vowel
nucleus and analyzed by means of Linear Predictive Coding (LPC) on
Computer Speech Lab (CSL) model 43008 to obtain the first two
formants (F1 and F2). A few sounds were analyzed on Pratt 3.9.5. in

order to copy pictures of formants.

55

F1 and F2 refer to formant frequencies and are related to the
length of the vocal tract and the constrictions formed by the pharynx,
lips and tongue when speaking (Pickett 1999:35). Ladefoged (1993:
192) compares the vocal tract to a bottle: "Any body of air, such as that
in the vocal tract or that in a bottle, will vibrate in a way that depends
on its size and shape." The vocal tract acts like a resonator. As air
passes through it, the sound of the source is filtered; every
configuration of the vocal tract has a set of characteristic resonances or
formant frequencies. Each vowel sound has a characteristic formant
frequency locations. These formant frequencies show up on sound

spectrograms as dark horizontal bars (Figure 3.0).

56

Figure 3.0: Digitized signal and spectrograms of Rachel's ‘jazz,’ ‘rack’

and ‘brag’

 

 

 

 

 

, . a. his ‘ i453" .4 -
jazz rack brag

Rachel's formants for jazz are F1/ 712; F 2/ 1783, rack F1/ 772; F2/

1611 and brag F1/ 720; F2/ 1896.

57

 

3C
200

E1]

 

300-

400-

500—

600-

700-

800-

900«

1000-

00 2q00 2§00 2100 2200 2900 1?00 quo 1100 lgoo 1900 890 600
fiobou!
'bhate Okid
0% .33) 0" 0"! gholo
ONO . fwholo
.'
OP“! .5 i.
'bmakc K2630]: B
3
Aducks
”mam. <><>bet A 7' v
sung (p D ‘
DUhand 0 “Sign ED 0 syfilost
A '
1:15am 33 D 3
Clplant Cl
<Z>ben g A» I.“ Voloset
v. Vthowht
‘-
' V1335
-

 

1100

 

 

Rachel. 48, working. Iansing

Figure 3.1: Vowel Plot for Rachel

Such F1 and F2 vowel formant readings were entered into a text file and

fed into Plotnik, a program developed by William Labov at the

University of Pennsylvania, to display, analyze, and compare vowel

systems. Figure 3.1 shows the vowel plot for Rachel. The higher the

vowel, the lower the F1 value; the fronter the vowel the greater the F2

value. The F1 axis Shows formant values along the height dimension,

and the F2 axis shows these values along he horizontal dimension. The

58

Plotnik program was used to plot vowels and calculate mean formant

scores. Vowels were then displayed on Excel (Figure 3.2).

Rachel's vowel plot

F2

2500 2300 2100 1900 1700 1500 1300 1100 900 700 500
' ‘ ‘ 200

 

300
o 400
I . 500

° 600

F1

700

.0:

ae’ 0
CI. 0
0 900

 

 

 

1000

Figure 3.2: Rachel's mean vowel formants

59

Once respondents' vowels were plotted and coded and means calculated,
I was able to compare respondents' vowel charts to four background
systems. (1) Peterson and Barney's (1952) classic study of the vowel
acoustics of American English exemplified the unshifted or pre-shifted
European American system. (2) The unshifted African American system
was exemplified by a normalization of the vowel systems of several
respondents from Alabama and Mississippi, which I analyzed for this
study. The data were taken from recordings made for the Dictionary of
Regional English (DARE). (3) The NCCS, an emerging vowel change
affecting European American residents of the Inland North, is the
model for the shifted Northern system as exemplified by Hillenbrand et
al. 1995 and Labov (1996) (4) The Southern Shift (SS), a vowel change
affecting European American residents of the Southern United States,
exemplified the shifted Southern states vowel system Labov (1996).
3.4 Original Dialects

I needed to establish that African Americans would not have been

involved in any /2e/ raising or /o/ fronting before moving North. To do

this I referred to Thomas's (2001:166) only representation of an African
American speaker from the source states for the majority of
southeastern Michigan African Americans — a male, Dallas County,
Alabama speaker born in 1856 and recorded in 1941. Thomas’ plot
shows that these two vowels are in no position which could be confused

with NCCS influences. Thomas’s much younger Columbus, Ohio

60

speaker (an area not influenced by the NCCS) also shows a very similar

positioning of both /m/ and /a/ (177) to that of the southern DARE

respondents that I analyzed for this study. I also conducted acoustic
analyses (through LPC analyses of F1 and F2 characteristics of vowels
on Computer Speech Lab) of one Mississippi and six Alabama speakers
whose voices were recorded during the progress of the fieldwork for the
Dictionary of American Regional English (DARE) — speakers born in
the late 19th and early 20th Century. Their vowel systems were
analyzed to determine what the Lansing parent vowel system might
have looked like. I calculated the steady state vowel formants of the six
Alabama DARE respondents and one Mississippi respondent. These
respondents were between the ages of 24 and 64 when they were

recorded from 1966 to 1970.

Researchers have suggested that each vowel has a particular ratio
of formant frequencies no matter who produces it. For instance, the F2
of /i/ is about ten times the F1 in both the child’s and the adult’s
pronunciation. In fact, researchers found that the F2 to F1 ratio for /i/
in the 1952 Petersen and Barney study clustered around 8.71 (Pickett
1999). The relationship between F1 and F2 is similar for a particular
vowel but differs across vowels. Nevertheless, it is difficult to compare
raw measurements for different speakers because vowels heard as the
same will have different physical realizations because of the

differences in vocal tract length (Peterson and Barney 1952).

61

Normalization transforms all measurements into a single reference grid
so those vowels that sound the same will have the same formant values.
Acoustic analysis shows that these speakers' vowels are very different
from those involved in the NCCS. The normalized DARE results
suggest that the parent system for Lansing AAE speakers resembles

figure 3.3. The normalized vowel plot indicates that /2e/ is not raised
(although fronted); /m/ is quite lower than /e/. The onset of /au/ is
non-front; in fact, it nearly covers /a/. /a/ and /o/ are distinct; the onset
of /e/ is not lowered; /o/ is not fronted; /o/ is back; /u/ and /u/ are not

completely back, but they are not as fronted as in the SS. Therefore,

any raising of /2e/ or fronting of /0/ among AA respondents in Lansing

will have resulted from contact with the emerging NCCS system.

62

 

2300

2100

1900

Normalized DARE

F2
1700 1500 1300 1100 900 700

500

 

 

088

U
0
U
O
O
O
oA 9y
° 0
cm
8” .0
o

200

300

400

500

600

700

 

800

 

Figure 3.3: Normalized DARE (Jones 2000)

F1

The 1952 Peterson and Barney vowel chart represents the original

of these analyses.

63

this vowel system to determine the degree of African American

of the formants of the vowels of the Lansing respondents were

European American Michigan dialect before the NCCS, and I also used

accommodation to the NCCS. The relative position of the mean scores

compared to the positions of the same vowels in the Peterson and
Barney system which is the 'pre-shifted' system posited to be the
Michigan system before the NCCS. Again, acoustic measurements were
conducted to determine the position of these respondents' vowels. Index

scores were assigned to the vowel positions of each speaker as a result

IE
1800

3000 2am 26100 2100 2goo 2900

1 600 1 400 l 200 1 000 800 600
2m 1 l J I l

 

Kelm" .fiboot

300-

400 . 0 bit " boat

"b bait 0 put

0 bed V bought
600 - A bus

7001 Dbad Ipot

BOO-I

 

 

Figure 3.4: Adapted from Peterson and Barney 1952; /e/ and /0/ adapted

from Stevens 1998 (men)

64

1E1

3000 28100 26100 2100 22100 2000 1 8100 1 6:00 1 100 12:00 1 000 800

 

300 a "0 beat

.9‘ boot
400 -

0 bit
0 put

Q) bait

V bou ht
60° ‘ 0 bet 4‘ "m g

A but
800 -

D bat ' P°t

ICED-i

 

Figure 3.5 Adapted from Peterson and Barney 1952; /e/ and /o/ from Stevens 1998
(women)

For example, in the Peterson and Barney system, /a3/ (bat) is
significantly lower than /8/ (bet), and not as fronted as /£/; the DARE

results (figure 3.0) indicate that Blacks who moved into Lansing from

the South had a fronted but not raised /a=./; therefore, raising to the
level of /e/ is considered accommodation to NCCS because it is

indicative of movement that was not present in the original dialects.

The following tables show index scores assigned to /a:/ and /a/. Since

only fronting was calculated for /a/ only F2 scores were indexed.

65

Table 3.1

 

F2 INDEX 1 /ae/ is significantly back of /e/

 

“FRONTING” 2 /m/ is not significantly different from /£/

 

3 /m/ is significantly front of /8/ but closer
to /8/ than /i/

 

4 /2e/ is significantly front of /8/ but closer
to /i/

 

 

 

5 /m/ is not significantly different from /i/

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 3.2
F1 INDEX 1 /ze/ is significantly lower than /8/
“RAISING” 2 /2e/ is not significantly different from /8/
3 /2B/ is significantly higher than /8/
4 /2e/ is significantly higher than /8/ but
closer to /I/
5 /a=:/ is not significantly different from /I/
Table 3.3
F2 0 /a/ is significantly back of /A/
{NDEX l /a/ is not significantly different from /A/
T112533] 2 / a/ is significantly front of /A/ but closer to /A/

than /8/

 

 

3 /d/ is significantly front of /A/ but closer to /8/

than
/A/

 

 

4 / d/ is not significantly different from /8/

 

Each respondent's vowel system was judged based on the relative

positions of the vowels in the system with regard to other vowels in

same system, and evaluated based on the type of index score that the

vowel received. Fourteen phonemes (see appendix C) of the American

66

 

 

English vowel system were analyzed for each respondent's system and
compared to Peterson & Barney, normalized DARE, NCCS, or SS

patterns, although only /m/ raising and fronting and /d/ fronting are
focused on for this study. Respondents' /2e/ vowels were considered
raised and their /0/ vowels fronted based on acoustic measurements

with appropriate t-tests which determined the position of the vowels in
question. T-tests conducted between mean vowels easily identify
whether a particular vowel has invaded the phonemic space of another

vowel. If t-tests reveal that /2e/ is significantly lower than /8/
(significance at .05), /a:/ receives an index score of "1.” If /ae/ is not
significantly lower than /8/ based on t-test results, /ae/ receives a score

of "2.”
3.5 The NCCS

Advanced NCCS speakers do more than simply shift /2e/ to the
level of /e/ or /I/; advanced NCCS vowels systems also include the
falling of /e/, /I/, and the fronting of /d/. Labov (1994:196) discusses
the attraction of /8/ and /CI/ to the hole resulting from the tensing and
raising of /ae/. In advanced stages of NCCS /8/ may also shift back
toward /A/. Therefore, the positions of /m/, /8/, and /a/ relative to each

other and to other sounds in an individual's vowel system indicate

whether a person is an advanced NCCS speaker or not. Using the system

67

of index scores to compare Lansing AA speakers to vowel systems such

as that of Brenda Einhorn's, an advanced NCCS speaker, (Labov, 2000)

figure 3.6 and 3.7 and figures 3.8 and 3.9 (Hillenbrand et al. 1995) help

to situate Lansing AA speakers’ accommodation to the NCCS. The SS

figure 2.1 of chapter two was also referenced to determine if there were

SS influences in the Lansing AA vowel system.

111

3000 28100 26550 2100 2200 2900 1 8.00 1 6100 1 100 1 200

1900 800

600

 

3m .
70 speech 0" shoes
70 heel K: .5 to
400 .1 0 big e .7' Tuesday 0” coop
El Dan 0" 8'
wlt>o ‘0‘!) 000 .3 boots
500 4 ha'b Haven J

D
j~%o E1 bed 0 E circles i O”
600- El ham fink d‘m

‘Ee
E] %°%‘° 8mc¢:§d:::m

7004 E1

"0
Dash bz'bg:

 

811% 983 fiAbucketE“
Oshed '37»-
I
Icot2 hog3
ICIJO~
I Ihog
Ihot2 I I»
1100

gfua'

90.6 Janet!

 

 

Brenda Einhorn,33. 1‘. Grand Rapids, MI TS 112

Figure 3.6 Vowel tokens for an advanced NCCS speaker (Labov

2000)

68

Brenda Elnhorn, F, 33, Grand Raplde

F2
2400 2200 2000 1800 1600 1400 1200 1000

 

 

[u ] two

9::
’c

0n:

 

 

800

200

300

400

500

600

700

800

900

1000

1100

Figure 3.7: Mean vowel plot for an advanced NCCS speaker: adapted

from Labov's (2000) Plotnik 5

69

3000 2500

Hillenbrand at

F2
2000

al. women

1500

1000

500

 

0P8

 

°>

.C:

100

200

300

400

500

600

700

800

900

 

 

Figure 3.8 Average formants for Michigan women; adapted from

Hillenbrand et al. 1995

70

1000

Hillenbrand et al. men

F2

2500 2300 2100 1900 1700 1500 1300 1100 900 700 500

 

 

100
200
I 300

O u

' O
400

e
C U
. e o
e 500
re 6
0

° A 600

e a

O
700

(l
e

800
900

 

 

Figure 3.9 Average formants for Michigan men; adapted from
Hillenbrand et al. 1995
A separate study of Michigan women and men (Hillenbrand et al. 1995)

reinforces Labov's characterization of the NCCS. /2e/ is at the height of
/8/ for men and higher for women. /8/ backing is also characteristic of

the men and women in the Hillenbrand et al. study. It is important to

notice that /m/ is lower in the pre-NCCS as characterized by Peterson

71

and Barney (1952) and for the normalized DARE respondents (Jones
2000)
3.6 Duration

Southerners have been characterized as pronouncing vowels
longer than Northerners (Deser 1991). Even though African Americans

may accommodate to the NCCS and raise /ze/, there may be other

acoustic features which also differentiate their low front vowel
phonetically from the raised low front vowel of the Inland North. The
duration of the vowel beginning with the first pitch period was

measured for 31 /EB/ token times thirty respondents. Calculations were

run on SAS using a proc mixed procedure, which is a repeated measure
procedure. The maximum algorithm was used to estimate the unknown
covariance parameters for gender, age, class, following manner of
articulation, and following place of articulation. Preceding place of
articulation was not measured in relation to duration.
3.7 Summary

The methods used for this study have been explained. Section
3.2 dealt with data collection procedures, demographic information and
aspects of the sociolinguistic interview. Section 3.2.1 discussed a
perceptual test known as the Gating Experiment. Section 3.4 presented
data analysis procedures and 3.4.1 centered around background vowel
systems, which may have contributed to AA Lansing pronunciation, and

the index scores used to ascertain accommodation to NCCS. Methods

72

used to analyze the effects of social and linguistic factors on raising
and duration were explained. The results of the study will be covered in

Chapter four.

73

Chapter Four
Results
4.0 Introduction
Chapter four examines the extent of accommodation to step one

(/2e/ fronting and raising) of the NCCS by thirty-one African Americans

indigenous to the Greater Lansing area. The relationship of linguistic

and social factors to the raising of /m/ and the mean duration for the
pronunciation of the /33/ will also be discussed in terms of what is

characteristic for the respondents in this sample.
4.1 Raising and Fronting

All of the African Americans who participated in this study and
the African Americans who were sampled from DARE had a fronted

/ae/, which happens to be characteristic of the SS (Labov 1994: 215;

Evans 2001); therefore, raising (F1) and not fronting (F2) is the
primary predicator of accommodation to the NCCS among Black
respondents in the Greater Lansing area. Results for raising (F1) and
not fronting (F2) will be the focus of this section.

The results of acoustic analysis conducted on each respondent's
vowels were studied relative to adjacent vowels. T-tests were run to
determine if there was a significant difference between the steady state

F1 of /ae/ and /e/, /ae/ and /I/ and /2e/ and /i/ to determine the degree of

74

N = 31 African Americans in Lansing
‘ 64% 36%
: 70°/o
3 60% «
g 50%«
i 40%
i
l
l
!

 

7 30%4

20%«
; 10%~
‘ 0%«

 

 

 

 

raised non-raised

Figure 4.0 Percent of Respondent with F1 /m/ 2 index score

/2c/ raising. An index score was then assigned (see Chapter three).
Twenty of the thirty-one respondents had a F1 /ae/ 2 index score.

Although the investigator interviewed more than forty respondents,
thirty-one of the best quality tapes were chosen for acoustic analysis.
The respondents are listed in table 4.0. The coding, OO, refers to very
old. Two of the respondents were much older than those categorized as

older.

75

Twenty of the respondents raised /ae/ to the level of /£/, which is

the regional norm for the European American community (Labov 1996,
Hillenbrand et al. 1995: 3103, Hagiwara 1997:657). The respondents
who obtained an index score of two are in table 4.1 with corresponding
status, age, network relations, and perception (Gating) data.

Table 4.0 Respondents

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Pseudonym SAS SAS Sex Age Age Relationship

1. Bertie W 61 F 42 O Paul's wife

2. Rhonda W 56 F 38 Y Paul's sister

3. Nubia W 53 F 27 Y Rachel's
daughter

4. Rachel W 53 F 48 O Nubia's mother

5. Norma W 53 F 42 O Rachel's sister

6. Lorna W 57 F 28 Y Olive's daugfler

7. May W 52 F 19 Y Dave's girlfriend

8. Alana W 56 F 28 Y

9. Veronica W 51 F 43 O

10. Nancy W 56 F 27 Y

11. Winston W 52 M 74 OO Gregory's brother
& Kendra's uncle

12. Alton W 53 M 54 O

13. Chuck W 56 M 28 Y

14. Qasim W 53 M 62 O Ibrahim's cousin

15. Paul W M 41 O Bertie's husband
& Rhonda's
brother

16. Rana M 26 F 50 O Ibrahim's wife &
Jane's stepmother

l7. Debbie M 28 F 24 Y Ann's niece

18. Mandy M 44 F 39 Y

19. Cassy M 45 F 21 Y Ibrahim's niece
&
Jane's cousin

20. Ann M 46 F 42 O Debbie's aunt

21. Jane M 26 F 25 Y Ibrahim's
daughter

22. Olive M 48 F 48 O Lorna's mother

 

 

 

 

 

 

 

 

76

Table 4.0 (cont'd).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

23. Dolly M 42 F 39 Y

24. Mali M 29 F 28 Y Gregory's
daughter

25. HenryI M 41 M 56 O

26. Thomas M 32 M 27 Y

27. Crawford M 49 M 45 O

28. Gregory M 27 M 66 O Mali's father &
Winston's
brother

29. Ibrahim M 26 M 52 O Jane's father

30. Curtis M 29 M 28 Y

31. Dave M 43 M 21 Y May's boyfriend

 

 

 

Table 4.1 Respondents with an F1 index score of /2e/ "2"

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

pseudonym SAS Sex Age NetE NetS Gating F1
/2e/
1. Bertie W F O 4 3 13 2
2 Rachel W F O 4 2 15 2
3. Norma W F O 3 1 l3 2
4. Veronica W F O 2 2 15 2
5. Nubia W F Y 4 1 l3 2
6. Alana W F Y 4 0 2
7. Nancy W F Y 4 5 14 2
8. Alton W M O 4 3 10 2
9. Paul W M O 3 2 l4 2
10. Rana M F O 3 2 16 2
11. Ann M F O 2 1 14 2
12. Olive M F O 4 4 l4 2
13. Debbie M F Y 2 l 10 2
14. Cassy M F Y 3 l 14 2
15. Jane M F Y 4 4 16 2
16. Mali M F Y 3 4 12 2
17. Dolly M F Y 3 0 14 2
18. Henry M M O 4 3 13 2
19. Crawford M M O 4 3 l4 2
20. Thomas M M Y 3 3 13 2

 

77

 

 

The other eleven respondents have F1 index at /ee/ 1 (/2e/ lower than
/e/) or a non-raised score.

4.2 Social Factors

In order to study the effect of social factors on raising, a variety
of statistical tests were run. Chi-square (a non-parametric test) results
for gender, age, and status reveal that status and age are not significant

although gender is (chi-square = 4.465, DF = 1, P-Value =0.035).

F1 Raising Gender N = 31
79%

 

I 90%
i 8070‘

l

* 70% 58%

| 50%' 42% [amen n= 12

t 40% :lwomen n= 19
; 30%4 ‘ ' ** *
i 2070

10% ‘
I 0%

 

 

 

 

 

 

 

 

Index scores

Figure 4.1 Gender

78

Table 4.2 Raisers according to gender

 

Gender N = 31 /m/ 2 index scores Percent

 

Men 11 =12 n = 5 42%

 

19 n 15 79%

Women 11

 

 

 

 

 

 

Table 4.3 Non-raisers according to gender

 

Gender N = 31 lzel 1 index scores Percent

 

Men n =12 n = 7 58%

 

Women 11 19 n=4 21%

 

 

 

 

 

 

With regard to gender, 42% of the men in this sample were at /a:/ 2,

whereas 79% women in the sample had a raised index score. Only 21%

of the women were at /ze/ 1 in contrast to 58% of the men. The

phenomenon of gender consistently influencing speech beyond the
differences in vocal tract size has received elaboration in a number of
studies. Sachs et al. (1973) and Goldstein (1980) suggest that the
gender differences that are seen in acoustic output are the result of
more than mere physiological differences in the length of the vocal
track, but may also reflect the effect of social expectations. Sachs et al.
(1973) examined the formants of pre-adolescent boys and girls of
similar weight, height and vocal track length and found that the sex of

the children was still perceivable by judges. The researchers concluded

79

that social factors extenuate the admittedly physical differences due to
vocal tract length. Their study suggests that linguistic sexual
dimorphism begins in early childhood when children are socialized to
fulfill expectations for gender roles. They also found that pre-
adolescent boys tended to pronounce their vowels with lower pitch than
girls, although there were two girls in the sample who were consistently
identified as boys because their pitch and formant patterns matched
those of boys. Goldstein (1980) examined the effects of anatomy on the
production of vowels by men, women, and children and also concluded
that anatomy is not the only factor that affects male female differences,
but that women tend to pronounce vowels in a way that utilizes more
vowel space than men use. Goldstein indicates that women's vowels are
more peripheral than those of men (230). She ascribes this to a
tendency for women to speak more clearly, although she also notes that
this is perhaps especially characteristic of western culture (234).
Sociolinguists have also examined and held gender accountable
for many of the linguistic differences between men and women. In
Evans' (2001) work on Appalachian speakers in Michigan and Ito's
(1999) work among rural Michigan speakers, a higher percentage of
women received index scores of 2 than men. Labov (1994:156) stated
that "in most of the vowel shifts that we will look at, women are
considerably more advanced than men." Herndobler (1993: 139)

described women as standard culture bearers and suggested that /ae/

80

raising among the working class White women in her sample had been
taken as "citified and sophisticated in the psyches of urban women."
One must also emphasize that within the field of Black women's
language (BWL), Houston Stanback (1985) has consistently focused on
the view that Black women's speech is not the same as Black men's
speech or White women's speech. However, Eckert (1989:247) put
forward the notion that there is no "constant relationship between
gender and variables and that gender based variation appears within as
well as between groups." The results of the Lansing AAE sample also
contribute to the literature of differentiation between the speech of men

and women.

4.3 Age and Class

F1 Raising Age N=31

73%
80%
60%
40%
20%
0% *4

 

[Each—=15
Mouse n_= 16:,

 

 

   

 

 

1 2
Index Scores

Figure 4.2 Age

81

Neither status nor age were statistically significant in this

sample; however, a greater number of older rather than younger

respondents raised /a/ to an index score of 2 (Figure 4.1), and more

middle-class than working class had an index score of 2 (Figure 4.2).

Table 4.4 Raisers according to age

 

 

 

Age Total N = 31 /m/ 2 index scores Percent
Young 11 = 16 n = 9 56%
Older 11 =15 n =11 73%

 

 

 

 

 

 

Table 4.5 Non-raisers according to age

 

 

 

 

Age Total N = 31 lae/ 1 index score Percent
Younger n = 16 n = 7 44%
Older n = 15 n = 4 27%

 

 

 

 

 

When the intersection of gender, age, and class are considered, middle-

class women lead in /m/ raising with older speakers slightly in

ascendancy over younger. This phenomenon lends itself to an
instrumental interpretation. These women are employed in a work
force, which is predominately white. The fact that older women raise,
is an indication that accommodation to NCCS is a reality in the
community although this reality must be positioned against the fact that

a sizable number of speakers do not raise /2e/. Which means there is

82

the element of choice, albeit unconscious. Wolfram (1969) noted that
African American women, older speakers, and middle-class speakers
are more sensitive to socially diagnostic features than men and youth
(117-8). This sensitivity may transfer to the vernacular of the region, as
exemplified in the phonological system of the Inland North. This is the

system, which most overtly states that a speaker belongs to this region.

F1 Raising Class N = 31
69%

 

60%

    

[lworking n = 15'
i .
[Elmiddle n] = 16

 

 

 

 

 

Figure 4.3 Class

Sixty percent (n = 15) of the working class respondents received an
index score of 2 (a raised score), while forty percent received an index
score of 1 (not raised). Although not significantly different, sixty-nine
percent of middle-class speakers (n = 16) received an index score of 2,

and only thirty-one percent received an index score of 1 or not raised.

83

Table 4.6 Raisers according to class

 

 

 

Class Total N = /a3/ 2 index scores Percent
31

Working n = 15 n = 9 60%

Middle n = 16 n = 11 69%

 

 

 

 

 

 

Table 4.7 Non-raisers accrding to class

 

 

 

Class Total N = 31 lae/ 1 index scores Percent
Working 11 = 15 n = 6 40%
Middle n=16 n=5 31%

 

 

 

 

 

 

There is a tendency for middle-class speakers to acquire features of the
wider community before working class speakers. For example, Wolfram
(1969:60) found less consonant cluster deletion among middle-class
speakers.
4.3.1 Network Relations

Network Ethnicity (NetE), which is a measure of the proportion
of close friends and associates from a respondent's same ethnic group,
also had no significant effect on raising. Pearson product moment
correlation measures were run for NetE and NetS, and no association
was found between either NetE or NetS and raising in this sample.

NetS, which is an evaluation of the density of a person's social network

84

based on the amount of relatives and colleagues in the neighborhood,
and the type of social relations the person has with co-workers. In order
to obtain a perfect (5) NetE score, respondents had to self-report that
100% of their close friends and associates were African Americans. The
scale was constructed as follows:

Table 4.8 Ethnic Network

 

Percentage of Friends and Score

associates from same ethnicity

 

100% African American 5
75% - 99% 4
50% - 74% 3
25% -49% 2
1% - 24% 1
0% 0

 

NetE scores for the 31 respondents were distributed in the following
manner:

Table 4.9 Distribution of NetE Scores

 

NetE Responses (N = 31)

Scores

0

l

2 XXXX(n=4)

3 XXXXXXXXX(n=9)

4 XXXXXXXXXXXXXXXXX(n=17)
5 X(n=1)

 

85

55% of the respondents identified their close friends and associates as
members of the same race. The four respondents obtaining a score of 2
are female. Three received a raised index score, but this number is not
large enough to provide statistical significance and merely suggests
that these individual are ideologically open to ethnic others.

NetS, which measured the density of a respondent's social
network, received a wider distribution in this sample, which indicates a
more open social network. In order to obtain a perfect NetS score, a

respondent needed to have the following characteristics:

1. Membership in a high-density territorially based network

2. Substantial kinship ties in neighborhood

3. Work at the same place with at least two people from neighborhood
4. Associates extensively with people from work in leisure time
activities

The respondents' responses were distributed in the following manner:

86

Table 4.10 Distribution of Net S Scores

 

NETS Distribution of total scores
Scores N = 31

0 XXX (n = 3)

1 XXXXXXXX (n = 8)

2 XXXXXXX(=7)

3 XXXXXX (n = 6)

4 XXXXX (n = 5)

5 XX (n = 2)

The NetS scores are unlike the NetE scores, which were skewed to the
high end of the scale because the majority of the population reported
that their close friends were ethnic sames. If one considers the
networks of 58% (n = 18) of the African Americans in category 0 - 2 as
open and categorize those obtaining scores of 3 - 5 as possessors of
dense networks, then it is obvious that the majority of African
Americans in this sample have open social networks. It is also the case
that the openness of the network does not necessarily extend to
incorporating other ethnicities into it. It appears to be the case that the

networks can be open, but populated principally or even uniquely by

 

other African Americans.

87

Similarly, the Gating score, which tested a respondent's ability to
perceive the NCCS, did not seem to be correlated to raising, although a
previous study indicated that there were significant perceptual
differences between African Americans in Lansing and young, White,
urban, southern Michigan speakers (Preston 2000, Jones 2001).

4.4 Adjacent Segments

Twenty out of thirty-one (65%) in this sample received raised

index scores, and one-way ANOVAs were run for each of the twenty

respondents with a raised index score to examine the effects of the /33/

mean scores for the various linguistic subgroups. Following manner of
articulation, preceding place of articulation, and following place of
articulation were examined. Eleven of the twenty (55 %) raisers
exhibited significant differences among the means for following manner

of articulation. One out of twenty showed that the subgroup /ze/ means

for preceding place of articulation were not equal, and none of the

twenty raised respondents showed differences among the /2e/ means for

following place of articulation. The dependent or response variables for

the ANOVA runs were the F1 formant scores for /ae/. The independent

variables were following manner of articulation, preceding place of
articulation and following place of articulation respectively. The
coding system from Plotnik was adapted into a model, which was used
to run one-way ANOVAs on SAS statistical program. The Tukey-

Kramer method for multiple comparisons was also incorporated in the

88

statistical model to help the investigator study the differences among
the subgroups.
4.4.1 Following Manner of Articulation

The coding system in Plotnik combines manner with voicing, so
the following coding system was used to run one-way ANOVAS.

Table 4.11 Coding for ANOVA

 

Code Segment Token

 

11 Voiceless Stop (VLS) apple, nap, zap, pat,

mattress, rack black

 

12 Voiced Stop (VDS) tab, cabin, dad,

Saginaw, brag, rag

 

 

 

 

31 Voiceless Fricative (VLF) laugh, bath, ask, past,
cash, mash

32 Voiced Fricatives (VDF) have, has, jazz

4 Nasals gamble, Sam, Lansing,

thank, gang, banker,

 

 

plant

 

The Tukey-Kramer method was used to make multiple comparisons
among mean scores for individual speakers after running the one-way

ANOVAS. Eleven out of twenty respondents Show an /m/ mean score

raising effect. Nasals showed the greatest effect for all eleven
respondents. The ranking for following manner of articulation is as
follows: Nasals>Voiced Stops>Voiced Fricatives>Voiceless Fricatives

>Voiceless Stops. For 91% (n= 10) of the respondents, nasals are most

89

 

different from voiceless stops. For 9% (n = 1) of the respondents,
nasals are most different from voiceless fricatives. This occurs in the
vowel system of Alton, a working class older respondent, whose
ranking order is N>VDF> VLS> VDS> VLF. Nasals always received
the most raised formant mean scores among those eleven individuals
who showed a raising effect for following manner of articulation (see
appendix F). Thomas will be used as an exemplar. The following vowel

plot shows the /ae/ distribution for Thomas. /2e/ is represented by the

clear squares on the vowel plot.

90

[El

600

 

 

3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800
l l l l l I J l L l l
300- Tow“ KO
"0 Kosmp e" 0"
0910*“
4004 '8" t & .fi
0 C
v
o "c O
'bmake O 8 O O . O
5004 "o o '5 A O" “O
c: A
600« u A v
E 0o ”%°n’% ‘ 7.?
700 d: an '3 ”"9 I
‘ Cb - W" lost
'3 3’0 Dhat Sf . '
900. g rock n . b I lot
:1 black2 ' I '
0 apple I I possible
9004 I hot
10004
1100

 

 

Thomas. Lansing. 28. Middle
Figure 4.4 /33/ distribution for Thomas

91

Thomas's /m/ distribution is typical for the sample. Subgroup mean

scores are ranked for Thomas as follows: N > VDS > VDF > VLF >

VLS. Thomas's ANOVA results are as follows:

Table 4.12 Thomas's ANOVA Results

 

Source DF SS MS F Value P

 

 

Manner 4 149318.2902 37329.5725 18.68 0.0001

 

 

 

 

 

 

 

Table 4.13 Mean scores for Thomas' following manner of articulation

 

 

 

 

 

 

Following segment Mean scores
Nasals 577
VDS 676
VDF 681
VLF 705
VLS 773

 

 

 

 

Thomas's nasals were different from all other subgroups. Voiceless
stops were different from nasals and voiced stops. Thomas's nasals
(hand, thank, banker, gamble, Lansing, and Sam) were higher than the

mean formant score for /s/. Thomas’ lowest formant scores for

following manner of articulation were apple, black, pat, and rack

(ANOVAs for the remaining respondents are in appendix F).

92

100‘

200

300

400

500

600

700

800

900

Tukey-Kramer Test for ANOVA Rune: Thomas

Following Manner of Articulation
1 2 3 4 5

L

 

 

Nasals VLS VD VLF VDF

 

 

Figure 4.5 Following manner of articulation for Thomas

articulation, nasals are different from everything else. Voiceless stops
are different from everything if the marginal difference between

voiceless stops and voiceless fricatives (P = .07) is acceptable. Voiced
stops are different from nasals and voiceless stops, but are the same as
voiceless fricatives and voiced fricatives.
same as voiced stops and voiced fricatives but different from nasals and
marginally different from voiceless stops. Voiced fricatives are

different from nasals and voiceless stops, but the same as voiced stops

Figure 4.4 shows that with regard to following manner of

and voiceless fricatives.

93

 

Voiceless fricatives are the

Nasals promote raising for the eleven respondents who showed an
effect for following manner of articulation. Voiceless stops demote for
99% of these individuals (N = 11). There is variation with regard to
the other phonetic environments based on the ranking of mean

/ze/ scores. Table 4.14 shows rank order for the eleven respondents

Table 4.14 Ranking of mean Fl/ae/ formant values

 

 

 

 

 

 

 

 

 

 

 

 

 

Pseudonym Ordered environments
Promoting Demoting

Alana Nasal VDS VLF VDF VLS
Alton Nasal VDF VLS VDS VLF
Bertie Nasal VDS VDF VLF VLS
Crawford Nasal VDF VDS VLF VLS
Jane Nasal VLF VDS VDF VLS
Mali Nasal VLF VDS VDF VLS
Nancy Nasal VDS VDF VLF VLS
Norma Nasal VDS VLF VDF VLS
Olive Nasal VDS VDF VLF VLS
Thomas Nasal VDS VDF VLF VLS
Veronica Nasal VDF VDS VLF VLS

 

 

 

 

 

 

who showed differentiation among mean Fl /2e/ formant values for

following manner of articulation. Six out of eleven (55%) rank voiced

94

 

stops after nasals as promoters of low front vowel raising. Voiced

fricatives rank third, voiceless fricatives fourth, and voiceless stops

obviously demote low front vowel raising among African Americans in

Lansing. The following ranking comparison is based on majority

percentages for the Lansing sample.

A comparison of Lansing African American results with Evans's

Appalachian results (2001), Ito's rural mid-Michigan results (1999),

and Labov's (1994:100) NCS Detroit results reveal similarities as well

as differences.

Table 4.15 Rank order for following manner of articulation

 

 

 

 

 

 

 

 

 

 

Lansing African Ypsilanti Rural mid- Northern
Americans Appalachian Michigan Cities Shift
LDetroit)

Nasals (Promote) Nasals Nasal Nasals
Voiced Stops Voiced Voiced Fricative Voiceless

Affricates Fricative
Voiced Voiceless Voiced Stop Voiced
Fricatives Affricates Stops
Voiceless Voiced Stops Voiceless Stops Voiced
Fricatives Fricatives
Voiceless Stops Voiced Voiceless Voiceless

Fricatives Fricatives Stops

Lateral

Voiceless Stop

Voiceless

Fricative

 

 

 

 

For all studies, the most advanced /ze/ tokens appear before nasals,

whereas, voiceless stops clearly demote raising. Results for fricatives

are mixed.

95

 

4.4.2 Preceding Place of Articulation

With regard to preceding place of articulation, one-way ANOVAs
were also run for the twenty respondents who had obtained an index
score of 2. The following coding system was adapted from Plotnik to

run one-way ANOVAs for preceding place of articulation.

Table 4.16 Coding: Preceding Place of Articulation

 

 

 

 

 

 

 

 

 

 

 

Code Preceding Place of Tokens
Articulation

1 Labial Pat, past, pal, badge,
bath, banker

2 Apical tab, dad, Sam,
Saginaw, zap, thank

4 Velar cabin, cash, gamble,
gang

5 Nasal nap, mattress, mash

6 Liquid laugh, Lansing, rack,
rag

7 Obstruent + liquid black, brag, plant

 

Those tokens that did not have a preceding segment were eliminated

from consideration for this portion of the analysis. /0/ was grouped

with apical in this classification. With the exception of one respondent,
there were no significant differences among the means of preceding
place of articulation.

Although Ann, a 42 year old middle-class East Lansing resident,

did not exhibit any differences among mean subgroups for following

96

manner of articulation, she is the only respondent in this sample who
shows a significant difference among mean scores for preceding place
of articulation. The Tukey-Kramer test shows that preceding obstruents

+ liquids have a significant demoting effect in Ann's system.

E

 

 

 

3000 2800 2600 2400 2200 2000 l 800 l 600 l 400 1 200 l 000 800
200 J l l 1 l l l l l 1 i
300 4 "0 9“"

R0 '50 bead
400 - 0 fish
a 6’
O
s
O 0 0K0 1099
500 4 "O 0 e
3> D . o pill 5." 4‘"
0 hit
[21 banker 1"
600 - I
. El badge V
111:] ammo, .
Cl
-’w_i O 0 past
D D 7%] % El rag W
0 Do a Dr A v
300 J r: (I: ‘9 l
I] D U
'17 bite d‘ A A v'
‘3 O I ' v
900 - . v'
D plant V '
V
El mattress E1 b | ack . V I
I
1 000 - 9°.”
4 4
I
I

l 100 ‘7

Ann, East Lansing, 42, Middle

Figure 4.6 /ae / distribution for Ann, East Lansing

97

Table 4.17 Ann's ANOVA

 

Source DF SS Mean Square F Value P

 

 

Preceding 5 117415.0225 23483.0045 2.69 0.0497

 

 

 

 

 

 

 

4.4.3 Following Place of Articulation

None of the twenty respondents showed significant differences
for following place of articulation. Ito (1999:83-4) also reported that
following place of articulation had no effect on low front vowels among
White rural Michigan speakers. Evans's (2001: 50) results for
Appalachian speakers indicate that voiced labials promote low front
vowel raising and following velars demote. Labov (1994:100) ranks
following place of articulation as follows: palatal > apical > labial >
velar and emphasizes that " black,” which is the token that is (usually)
the lowest and farthest back has a voiceless velar stop and a preceding

obstruent + liquid.

4.5 Duration
Although Ladefoged (1993) and Labov (1994) have referred to

/m/ as a short or traditionally lax vowel, Strange et al. (1983:698)
categorize /m/ acoustically as an intrinsically long vowel and group /m/
with /e, a, o/ in terms of length as opposed to /I, 8, A, u/ which have

been characterized as the intrinsically Short vowels. The average mean

duration of vowels in Hillenbrand's study supports the contention that

98

/a=:/ is an intrinsically long vowel and not an acoustically short vowel.

The mean duration of Hillenbrand et al. (1995) substantiates the claim

that /a‘:/ is acoustically longer than /I, e, A, u/.

Table 4.18 Duration for Michigan speakers adapted from Hillenbrand et

al. (1995:3103)

 

/i/ /I/ /e/ /8/ /aa/ /(1/ /o/ /0/ /U/ /u/ /A/

 

Men 243 192 267 189 278 267 283 265 192 237 188

 

 

Women 306 237 320 254 332 323 353 326 249 303 226

 

 

 

 

 

 

 

 

 

 

 

 

In this case the mean duration for /2e, e, i, a, o, o/ is longer than those

vowels characterized by Strange et al. (1983) as intrinsically short
vowels.

Pronouncing vowels with longer duration has been reported as a
feature of Southern speech. Toni Deser (1991:112), in her analysis of
Northern and Southern dialect children in Detroit, concluded that vowel
duration is an aspect of dialect. In order to examine the effects of
adjacent segments on vowel length among AAE speakers, the duration
of the vowel nucleus minus consonant onset and offset was measured
for thirty-one tokens times thirty-one respondents. The tokens are given

in 4.10:

99

 

Table 4.19 /2e/ Tokens

 

 

 

 

 

 

 

 

 

 

 

1. apple 7. black 13. mattress 19. rack 25. laugh

2. pat 8. nap 14. cash 20. brag 26.plant

3. pal 9. tab 15. gang 21. has 27. rag

4. Saginaw 10. dad 16. zap 22. badge 28. ask

5. gamble 11. mash 17. bath 23. thank 29. have

6. Lansing 12. banker 18. jazz 24. cabin 30. past 31.
Sam

 

 

Calculations were run on SAS using the proc mixed repeated measures
procedure. The maximum algorithm was used to estimate the unknown
covariance parameters for gender, age, and class, following manner of
articulation, and following place of articulation. Preceding place of

articulation was not examined in relation to duration.

4.5.1 Social Effects

4.5.2 Gender

Results indicate that there are certain social effects related to duration.
Women had a greater mean vowel duration than men (p = 0.0062), but
these results parallel the results of Hillenbrand et al. (1995), who found
significantly shorter duration for men than for women and children in
their study of respondents from Michigan's Lower Peninsula. The mean

duration for /hmVD/ spoken by Hillenbrand's speakers were:

100

 

Men 278 ms
Women 332ms

The Lansing AAE speakers mean /2e/ duration as measured for have,
which most approximates the /h2eVD/ utterances spoken by the

respondents in Hillenbrand's examination, is as follows:
AAE speakers in Lansing:

Men 327ms

Women 369ms

It may be that Lansing African Americans pronounce /m/ with greater

length than White speakers in Lower Michigan, and that, of course, may
reflect their Southern heritage, although, as the next section shows,
those speakers who might be expected to have preserved original

dialect forms best are not those who have the longest duration.

4.5.3 Age and Class
There was no effect for age except that the two older men in the

study pronounced /2e/ significantly shorter than everyone else. Winston

(74) and Gregory (66) were coded 00 since they were quite a bit older
than those in the 40-60 category. They were coded differently from the
other respondents to see if the results might be significant. The results
show that the duration for 00 is shorter than 0 (p = 0.0248). There is
no effect between 0 and Y (p = 0.9620); however, duration for the 00

respondents is also significantly shorter than Y (p = 0.0211). This

101

comparison may not be very important because there were only two men
who were much older than everyone else. Yet, the results are
suggestive and bear further study. There was no effect for class in this
sample.
4.5.4 The Effects of Adjacent Segments

The effects of following manner of articulation and following
place of articulation were examined in relationship to duration.
Preceding place of articulation, laterals and affricates were not studied.
Manner of articulation was divided into five levels and place of
articulation was divided into four. Table 4.11 illustrates the coding
system. There were five levels for manner: VLS, VDS, VLF, VDF and
N. The repeated measures procedure indicates that VDS and VLF are
similar to each other but different from the other levels. The five levels
are ranked in terms of greater to lesser duration as follows: VDF > VLF

> VDS > N >VLS. /a=:/ is pronounced longer when followed by voiced

fricatives in this sample. Voiceless fricatives and voiced stops are not
significantly different from one another, but are significantly different

from nasals, voiceless stops and voiced stops.

102

Table 4.20 Coding: Place of Articulation

 

 

 

 

 

 

 

 

Place _> Labial/Labio- Apical Palatal Velar
dental/Interdental
151.333" 5.1.123 1 2 3 4
l Voiceless nap, apple , zap pat, rack,
Stops mattress black,
2 Voiced tab, cabin dad Saginaw,
Stops rag, brag
3 Voiceless bath, laugh ask, past cash,
Fricatives mash,
4 Voiced have has, jazz
Fricatives
5 Nasals gamble, Sam Lansing, gang,
plant, banker
thank

 

 

 

 

 

 

Following place of articulation also seemed to affect vowel

duration. The mean duration for /a°./ followed by a palatal is longer than

when the vowel is followed by an apical (P = 0.0001), labial (P =

0.0001) or velar (P = 0.0001). Mean duration for /2e/ with a following

labial, labio-dental, interdental, and apical have a similar effect. The

labial group is also similar to the velar group; however, following

103

 

labials and palatals significantly differ from one another in their effect
on the vowel; moreover, palatals and velars have a significantly

different effect on /ae/ duration. Following place of articulation may be

ranked as follows from greater to lesser duration: palatal
>apical>labial>velar. The mean length for apical, labial and velar are
almost the same.
4.6 Summary

Chapter Four contains an analysis of features of the low front
vowel among 31 African Americans. Characteristics of the low front
vowel are examined within the context of the social and linguistic

factors which may effect /ae/ raising. Gender is considered a significant

contributor to vowel differentiation whether the analysis involved
raising or duration. Net-work relations did not show an effect for the
way the speakers in this sample pronounce the low front vowel. Age
and status were not significant but suggested that certain trends were at
work which parallel normal sociolinguistic behavior. Following manner
of articulation had the greatest effect on raising with nasals clearly
promoting and voiceless stops demoting. Chapter Five centers around a
discussion of the effect that region has on the pronunciation of the low

front vowel among AAE speakers.

104

Chapter Five
Qualitative Discussion

5.0 Introduction
Chapter five summarizes the results and discusses some qualitative
implications. This chapter also deals with sections of the interviews in
which respondents expound on their ideology and thoughts about
residing in the Greater Lansing area. I initially hypothesized that
African Americans in the Greater Lansing area had engaged in
institutions in the wider community and had sufficient networking
experiences to accommodate to the early stages of NCCS; therefore, it
was assumed that accommodation patterns would be similar to those of
majority speakers. Indeed, it was posited that young, middle-class
women, and individuals with open personal networks would lead in
accommodation in the African American community. It was also
believed that individuals who show strong neighborhood and ethnic
loyalty would not accommodate to the White vernacular of the region.
5.1 Regional Accommodation

I have shown that twenty (64%) of the thirty-one respondents in
this sample exemplify the first step of the NCCS pattern that is typical
for the Inland North. I have plotted the average formant frequencies of
vowels produced by the 45 men and 48 women in the Hillenbrand et al.

(1995) study of respondents from the southeastern and southwestern

105

/

figures 3.8 and 3.9, reinforce the research by Labov (1994) and Eckert

portions of Michigan. The vowel plots, which were represented in

(1989, 2000) concerning the NCCS. These Michigan vowel plots show

that /ae/ is higher than /8/ for women and equal to /e/ for men. A similar
pattern of /2e/ at the height of /e/ emerged for 64% of the sample of

African Americans in Lansing. Blacks can and do adopt a regional
pronunciation, but I will show here that their overall vowel system
performance is still characteristically Black.

The African Americans, who raise lae/ in this sample, tend to

accommodate to Northern speech in terms of front vowel pronunciation
and retain Southern AA pronunciation for the back vowels. Vowel
duration may also be greater for Blacks than for White Michigan
speakers, which is also indicative of Southern pronunciation. These
acoustic differences and others differentiate Northern AA speech from
Northern EA Speech.
5.2 Social Factors

There seems to be very little correlation between network and
raising. Why is that? The feelings that African Americans project in the
interviews conducted for this study are complicated. They believe that
Lansing is a healthy, family-oriented city that is not crime ridden like
Detroit or other large urban areas. They seem to perceive that Blacks in
Lansing, although relatively well off, are stifled culturally and perhaps

uninterested in the activities of the larger community. They do not

106

appear to initiate extensive contact with Whites, based on their self-
reported social network scores. Moreover, their ethnic network
relations seem to be carefully hemmed in, which could be universal for
Blacks in America. Feagin and Sikes (1994) interviewed 209 middle-
class African Americans from various parts of the nation concerning
their social experience in the United States, and many of these African
Americans related that modern racism, subtle or otherwise, is an
inescapable part of the fabric of the United States. Avoidance of Whites
beyond necessary interaction, although not the only strategy, is one of
the strategies that even middle-class Blacks use to deal with White
racism or even the possibility of racism. Feagin and Sikes (l994:4)
assert that "almost any encounter with Whites, in workplaces, schools,
neighborhoods, and public places, can mean a confrontation with
racism." They (275) add:
One way to deal with discrimination is to try to avoid situations
where it might occur, even at some personal cost. A physician in
a southwestern city responded to a question about dealing with
discrimination this way: "It just depends on what the situation
is, whether or not it's personal, business; it just kind of depends
on what, you know, exactly what it is. I usually don't go places
where I'm not wanted, so I'm not the kind of person that
trailblazes —— where people tell you that they don't want you in a

certain situation and you persist. It's kind of a hard question to
answer.

107

Consequently, many African Americans avoid social situations where
racism might occur or prepare psychological shields, guards and eternal
vigilance to escape physical or psychological racist attacks, which are
part of American life.

Lois Benjamin (1991: 278), in her study of 63 men and 37
women, relates that the racial composition of the social contacts of her
participants was 62 percent Black, 4 percent White, and 34 percent
mixed. African Americans in the Greater Lansing area are not very
different from the respondents in the Feagin and Sikes study (1994) or
the Benjamin (1991) study when they report that most of their social
contacts are African American. Although they admit that they like the
safety net that Lansing gives them, this sample does not admit to
significant networking in the White community. Their comments
concerning the dismemberment of the Black community, which have
been echoed by Meyers (1970) and Hawkins (1979), may also have
relevance to the lack of effect for network. In the following excerpt,
Rachel, a 48-year-old working class respondent, discusses life in
Lansing when she was growing up. There is an element of compromise
in African Americans’ reflections on their relationship with Lansing.
There is a feeling that things could have been much worse although
they were not perfect. There is wariness in their reflections, which

causes me to wonder if this wariness is symptomatic of small cities

108

such as Lansing. Are African Americans wary of saying too much, or

are they complacent, as one of my other respondents contends?

Interviewer: Can you tell me a little bit about Lansing when you were
growing up? What was it like?

Rachel: It was a nice place to raise your kids. It was uh. I used to
remember the neighbors. It was like more like a family type thing.
'Cause if we did anything, you know, they like told on my mother. It
was like we couldn't get away with anything and nowadays when people
kinda don't do that like if we fell and hurt our leg or something; the
neighbor. If my mom was busy or something, the neighbor would help
you so it was a nice place like that. Now me, myself, now growing up I
knew there was a lot of racial tension during that time, but at the same
time we knew people that wasn't like that. 'Cause like White people. I
had White friends when I was growing up that came over our house and
play so uh I thought it was a nice place to live. It was quiet. It was
low key. It wasn't busy you know compared to Detroit. I couldn't

imagine being raised there.

In the previous section, Rachel demonstrates her loyalty to

Lansing and her general openness toward society at large. Although

she stated in another section of the interview that most of her close

109

friends and associates are African American, she points out that She had
White friends growing up.

Henry, a 56 year old personnel management specialist for the
state of Michigan articulates his take on African Americans life in
Lansing.

Interviewer: What do you plan on doing when you retire?

Henry: Move to Brazil or Cuba.

Interviewer: Why Brazil or Cuba?

Henry: I love the people; I love the culture and I love the fact that they
are still connected to their African roots.

Interviewer: Do your best or closest friends live in your neighborhood?
Henry: Uh. No more.

Interviewer: They used to when you were little?

Henry: We used to be in a very small area here in Lansing, which is
probably uh from the Grand River uh Where the Grand River north to
St. Joe.

Interviewer: Grand River — do you mean like coming out of East
Lansing?

Henry: No. We are talking about in Lansing — in Lansing.
Interviewer: Uhum. The Grand River — the actual river in Lansing.
Henry: Yeah. There's, you know, the bridge that goes over Martin

Luther King that used to be on the other side on North River Drive. On

110

this side on the north side was where Black folks lived at. We lived
from there over to St. Joe Street.

Interviewer: Um.

Henry: Sometimes. Let's say St. Joe, maybe Hillsdale.

Interviewer: Right.

Henry: And we went from. Probably from Pine Street on further than
— Not even this far down. From about. (I'll try to get this) Probably
from Pine Street back to uh past Main Street School. We lived in a
very small contained area. 90% of all the Black people in Lansing.
Interviewer: And what changed that?

Henry: I would say that when the time the Military Highway came
through. It's called 96. That's the Military Highway. After the riots,
the American government realized that it couldn't get a lot of the tanks
and what have you. That was a major project for all over the country.
They can get from one end of the country from Boston to L.A. It went
through every major community — Black community in the United

States.

Henry tells the interviewer that the highway came right through the

Black community and displaced 80% of the people in Lansing's Black

community.

Interviewer: Where did people move? The south side?

lll

Henry: Yeah. As a matter of fact during that time there were no Black
people living in East Lansing; no Okemos; no Waverly; no south side of
Lansing. Didn't no one live on the south side of Lansing (rising
intonation). Didn't live across the bridge. There were a few people
across the bridge.

Interviewer: Do you spend time with your co-workers after work?
Henry: Most of my co-workers are White and I don't socialize with

them. No.

Henry does not project local loyalty although there is loyalty
expressed toward his ethnicity in his comments. Henry's discussion
indicates that the character of the Black community has changed over
the years. Henry implies that the Black community as a geographical
entity may not exist anymore. He is saddened by the break up of the

geographical Black community and has decided where to assign blame.

Interviewer: Is this a good place to grow up?

Henry: Uh. It's like a coin with two sides to it. One side is uh good
place to grow up because you don't have the uh lot of the uh big city
crime and what have you, but uh on the other side — the negative side.
It's like culturally deprived for African Americans. It's a very ultra-
conservative city, a very racist city. Uh a city where you can make

money —live well to a certain extent, but you cannot. It's culturally

112

deprived. Uh. You have uh. The city is uh. Now you don't have
communities no more. Everybody is spread out everywhere; therefore,
you don't have that cohesiveness no more — that family
(unintelligible). You don't have that. People are very backward, very
disenfranchised here because they have done everything to keep us for
just being consumers, but not being producers or owning anything.
They destroyed all the Black community businesses where they came
through with the Military Highway and they never went anywhere. We
had Black cab companies — all kinds of stuff here. It's all gone. Uh.
Uh. Black folks here are just complacent. You. They just kinda like.
They live... And relationships and they see things in their
understanding. Uh. They're just living here and uh. I got an Escort
and I got a house and I make 60/70 thousand a year so what I got to say
anything about anything what's going on. And they do not get involved
in any social issues — very few if any social issues here. I could go

on, but...

Bertie, who is a 42-year-old cashier, also echoes some of Henry's
complaints about the lack of progressiveness of life in Lansing. She
subtly indicates that the cultural part of community is sadly lacking in

Lansing.

113

Interviewer: So what is the best and worst thing about living in this
area?

Bertie: In this area or in Lansing?

Interviewer: Lansing in general and then you can talk a little bit about
your neighborhood.

Bertie: Lansing, in general, like I said, the best would be because it's
not so fast paced and it's a good place to raise children uh, but you
know there's not a lot of prominent Black people in Lansing that '5
doing too much of anything you know. 'Cause it's like they are stifled
you know here.

Interviewer: Has it always been like that?

Bertie: Yes! Yes!

African Americans do not say that they dislike the fact that they are
socially separated from the Whites around them. Ann, 42 year old

middle-class resident of East Lansing, discusses the situation.

Interviewer: Do you spend a lot of time with your co-workers after
work?

Ann: No.

Interviewer: You don't?

Ann: (Indicates that she does not.)

Interviewer: Why not?

114

Ann: I don't know —just don't — separate lives.

Interviewer: What do you do when you have spare time?

Ann: I go to school. Wash.

Interviewer: Wash clothes?

Ann: Uhhuh.

Interviewer: Do you think this is a good place to grow up?

Ann: Yes I do.

Interviewer: Why?

Ann: It's versatile. It's quiet. Uh - safe, mostly safe. Uh access to a
lot of different things.

Interviewer: Have you ever wanted to live anywhere else?

Ann: Yes.

Interviewer: Where? Some place quiet?

Ann: Uh no. Like uh. Actually, no. IfI wanted to live somewhere
else, I would like a place that uh where you'd have a lot of things to do.
Museums take your kids to museums - different uh fun things to do.
You know you're pretty limited here.

Interviewer: Yeah, Michigan State.

Ann: Uhhum.

Interviewer: It takes up most of the town.

Ann: If they wanted to build something besides a water park, they
could, but..

Interviewer: = They have that new mall.

115

Ann: =It's not that much. There's not really no real culture here except
for the Capital building.

Interviewer: So, what's the best and worst thing about living in the
area?

Ann: Uh. I would say the best thing is the quietness. Uh the people,
uh.

Interviewer: What's good about the people?

Ann: They don't bother you.

Interviewer: Oh. Okay.

Ann: They go about their business. They don't say anything to you.
It's safe...and I would say the worst thing is it's really not that much to
do. Go to work. You have your few places you can go for activities,
but there is nothing exciting really around in this town. It's pretty

much. It is kind of like a farm town.

Ann is somewhat neutral about the Lansing area. She accepts
social segregation as a fact that does not need to be contended with.
She is not closed to ethnic others, but enjoys the tranquility of being
left alone. It is difficult to isolate a single monolithic view based on
the interviews. What ably comes across is the continued separateness
of Black Americans; in spite of, societal changes. Each individual has
a slightly different stance with regard to residing in the Lansing area;

however, each one is well aware of themselves in relation to the wider

116

community by virtue of their ethnicity. Ann states that we live
separate lives. Henry says that Blacks in Lansing do not participate in
any social issues in Lansing. Bertie says that Lansing is not
progressive enough, and Blacks do not have a prominent place in its
social structure. Yet all acknowledge that this is a good place to grow
up and raise children.
5.3 NetE Scores

Rachel reported that 90% of her close friends and associates are
African Americans. Based on her interview, it is possible to suppose
that a NetE score of 4 may not adequately express Rachel's social
network because she said that She had White friends as a child, and it is
conceivable that she may have White friends now or perhaps she
associates with a wide range of people who are not classified as
friends. The respondents who obtained NetE scores of 3 (those who
indicated that 50 to 74% of their close friends and associates belong to
the same ethnic group) have careers which place them in the culture of
wider communication. Rana, for example, works for a major airline;
Norma does landscaping for the State of Michigan; Thomas is a music
teacher who plays for a White church on Sundays; Chuck works as a
cook at the Kellogg Center; Dolly is a secretary for the State of
Michigan; Mali works for the Michigan Supreme Court, and Curtis is a

Ph.D. candidate at Michigan State University. All of these individuals

117

have diverse associations in the world of wider communication, yet
they state that the majority of their friends are African Americans.
5.4 NetS Score

In terms of the NetS score, which represents the density of a
respondent's network, one finds that the respondents with the least
amount of open social contact (those who obtained NetS scores of "0")
are eclectic with regard to raising. Qasim, a disabled laborer, who

obtained an /a:/ raising index score of 1, does not work and has not

worked for over twenty years. Dolly, who works for the state, received
a raising score of 2 and Alana, who is a child care worker who works

from her home, received an /m/ raising score of 2. These individuals

have the least dense networks of all the respondents. What is it they
have in common with regard to NetS? One surmises that it must be
lack of commitment to the world of wider communication.

The results indicate that racism and the effects of racism on an
individual or personal ideology must be taken into consideration in
order to understand the nature of social networks in this sample. Jane is
a 25 year old middle-class raiser, who obtained a NetE score of 4 (90%
of her friends are African Americans) and a NetS score of 4 which is
indicative of a dense social network. Yet, she did not espouse
dissatisfaction with Lansing during the interview. She considers
African American culture part of American culture and believes that

Lansing is a good family oriented city.

118

Interviewer: Do you think uh this is a good place to grow up? Is
Lansing? Do you think it’s a good place to grow up?

Interviewer: Um. I’d say yes.

Interviewer: Why?

Jane: Uh. It’s you know — if you look for ...There’s uh. There’s a lot
of... There’s recreational activities and a there’s a lot of neighborhoods
where there’s a lot of other children growing up. You can grow up with
other children. It’s a (pauses) families, lots of families uh, uh. What

else can I say?

Jane is loyal toward her neighborhood and toward the city of
Lansing. She does not express dissatisfaction with the city of Lansing
or the state of Blacks in the city. In a section of the interview that is
not recorded here, she states that African American culture part of
American culture.

In spite of the few exceptions, there is a pattern or tenor to most
interviews which reveals that social network does not sufficiently
explain accommodation patterns among the Blacks in this sample.

A comparison of Mandy, a 39-year-old middle-class woman, and
Henry, a 56-year-old middle-class man, exemplifies the complexity
encountered when attempting to change social history, network

relations and worldview into categorical data. Mandy's father was a

119

professor at Michigan State University; she attended Red Cedar and
East Lansing High School. She self-reports a NetE score of 2 and a
Net S score of 2 which means that she admits ethnic others into her
open (low NetS score) social network. Mandy asked me not to record
her personal narrative, but she confided that she dates White men and
admits that she was very insecure about color when she was a child.
Nevertheless, she has a raising index of 1. Neither social network nor
ideology can explain Mandy's linguistic choices. Mandy's grammar is
very standard, but she does not raise. Henry freely uses some
nonstandard forms in the interview, follows an Afrocentric lifestyle,
but raises. Thus, attention to the pronunciation of the regional
vernacular does not necessarily entail attention to the syntax of
"Standard English" and the reverse. One could attribute complexity of
this sort to the "push pull" discussed by Smitherman (2000) or view
this complexity as the way in which individual Blacks voice resistance.
Morgan (1994:129) presents the dual nature of AAE as:

1. an expression of African character

2. a symbol of resistance to slavery and oppression

3. an indicator of a slave mentality or consciousness
There are African Americans who consider vernacular AAE, for
example, to be a slave dialect spoken by those who still retain a slave
mentality (Morgan 1994). Others however consider the vernacular

version of AAE the language variety of choice. The features that a

120

speaker chooses to accommodate to or resist depend on his or her
perspective concerning Black authenticity.

Henry, who states that most of his friends are African Americans
(NetE 4), travels to Cuba, Brazil and Egypt to support his Afrocentric
identity, and has a NetS score of 3. This is an average network score,
which places him in the center of the density scale. However, Mandy
who has a more open social network and is open to ethnic others has

a lower /m/ score than Henry.

Henry's vowel plot

F2

2500 2300 2100 1900 1700 1500 1300 1100 900 700 500
200

 

300
400
500
.3 600

700
(11.1

0
$0

0 800

 

 

900

 

Figure 5.0 Mean formants for Henry, 56, Middle, Lansing

121

Although Henry raises the low front vowel and Henry's /o/ is also

fronter than Mandy's, there are many other vowel features such back

back vowels (/o/, /U/ and /u/) which authenticate his African American
identity, while the raising of /2e/ may contribute to his regional

identity. Mandy has a NetE score of 2, and associates with Whites;

however, she does not raise the low front vowel or front /a/ to the same

extent that Henry does. Nonetheless, she, like Henry, has the extreme

back vowels, /o/, /u/, /u/, and /d/, which have been documented (Bailey

and Thomas 1998) as Southern AA vowel features.

Mandy's vowels

F2

2500 2000 1500 1000 500
200

 

300

400

C

’ ° ’ 500
.0

600

098

 

.0

O
.0

700

800

900

 

1000

 

Figure 5.1 Mandy

122

Curtia' vowel plot

F2

 

2100 1900 1700 1500 1300 1100 900 700 500
‘ ‘ 200
u M 300
e
e
I
. i .U 400
0 0y
A o O :
8 e e
06 . 500
CI
a e
e
.au .3 600
.0
700

 

 

 

Figure 5.2 Curtis

Henry, Mandy, and the other respondents embody the interaction
between variation and ethnicity among African American speakers.
However, Mandy who has a more open social network and is open to

ethnic others has a lower Ize/ raising score than Henry.

For instance, Curtis's vowel plot exemplifies variation of a different

sort. Curtis obtained an F1 /a:/ index of 1,a non-raised score, and his
front vowels pattern like Southern Shift vowels. The /i/ and /I/ reversal

pattern of the SS is part of his vowel system. His front vowels as well
as his back vowel have SS influence due to the fact that his parents are

from the South and he visited the South as a youth.

123

The notion of what it means to be an African American can not
be monolithic, but social history presupposes certain influences, which
used, to include socialization in the "Black community." Black people
come in different sizes, shapes and ideologies — even regional
ideologies. This notion of authenticity does not imply that the only true
"Black " speakers are the vernacular speakers (see Morgan 1994), but
views African American language as a subtle language and meaning
system that is a product of social history. W.E.B. Dubois, Jesse
Jackson, James Baldwin, Tupac Shukur, Malcolm X, Martin Luther
King Jr., Shelby Steele, J. California Cooper, Maya Angelo, and Toni
Morrison were and are erudite authentic speakers who possess a
"Black" voice and language which includes and reaches beyond "core"
grammar to capture language and use it as a vehicle to express
experience. Yet, these "leaders" and scholars espouse different
ideologies and regional voices. There is a supposition in society that
Blacks who associate with Whites extensively will use the English of
wider communication. Baugh (1999:75) states that his work, which
concentrates on Black adults of all social backgrounds, indicates that
there is movement both toward the linguistic usage of the mass media
and movement away from it. Baugh adds that this point is critical in
regard to Labov's (1985) observation that Black and White dialects

continue to change independently because many readers who are

124

unfamiliar with the diversity of Black American culture might falsely
assume that this research applies to the majority of Black Americans.
However, African Americans who operate on more than one social level
in the Black community realize that linguistic variation exists within
the Black community. This notion of authenticity colors the Lansing
interviews because most individuals locate their experience in "Black"
space. Social networks, close friends and ideologies are Black, and
their experience is that of Black people living in a world which has
historically disrespected them, their experiences and their
contributions, and in a community which leaves them alone socially for
the most part. Perhaps it is for this reason that most respondents have
expressed a certain amount of satisfaction with Lansing as a community
even as they deny deep connections with the White Lansing community.
If one considers the fact that AAE back vowels have remained
relatively stable in Lansing, the general lack of participation in the SS

front vowel movement, and the fact that he/ is raised and /e/ is

sometimes backed, one clearly realizes that Northern regional
alignment on the part of some African Americans, and the retention of a
strong African American identity in pronunciation are in interaction.

Of course, identity evolves over time as social history changes. Social
network theory predicts that individuals with open networks (lower
NetS scores) are more likely to be attuned to the language of wider

communication (Milroy 1980). Network in the traditional Milroy

125

(1980) sense has been used to measure accommodation among
homogeneous racial groups who spoke a different dialect, but who did
not belong to a different race. Milroy (1980), Evans (2001), and lto
(1999) researched homogeneous racial groups. These linguists found a
strong correlation between social network and pronunciation. Childs
and Mallinson (2003) also observed correlation between strong social
cultural ties to the regional community and accommodation in
pronunciation and grammar among African Americans in Appalachia.
However, strong network ties with ethnic others were not reported in
the Lansing sample.

5.5 Initial Hypotheses

Raised and non-raised /2e/ exist in tandem in the Lansing

community. In fact, members of the same family differed with regard
to the pronunciation of the low front vowel. Ibrahim does not raise, but
his daughter, Jane, and his wife, Rana, do. Olive raises, but her
daughter, Lorna does not. Paul and his wife, Bertie, raise, but his
sister, Rhonda, does not. Although language indexes social groups,
internal variation should not be ignored. Wolfram and Schilling-Estes
(1998: 162) contend the NCCS functions as a social marker. They
state:

There is clear-cut social stratification of the linguistic variants

and participants in the community may even recognize this

distribution, but the structure does not evoke the kind of overt

commentary and strong value judgements that the social
stereotype does.

126

The fact that middle-class Blacks have moved away from contained
neighborhoods set aside for Black people, and are now spread out
throughout the city has encouraged adaptation to aspects of the
phonology of the regional standard since adaptation does not threaten

community identity. In addition to /m/ raising, there are cases of /e/
backing and lowering. Although /e/ lowering is not directly within the

boundary of this study, it provides further evidence of accommodation;

nonetheless, other vowels appear to be stable. /0/ is not fronting, and
though /u/ and /u/ may not be as back as Bailey and Thomas (1998)

portray, they are still back and not fronted as in the White vernaculars.
In my opinion accommodation should not be viewed as imitative
assimilation but considered the adjustment of certain features to fit into
the regional social unit.

There are aspects of the African American phonological system
that have yet to be explored. Dorill (19862151) asserts that African
American vowels are monophthongal, whereas, Whites pronounce
vowels with an upglide. Vocal quality and intonation patterns differ
from the majority's vernacular dialect. African Americans like other
Americans are members of many subcultures as defined by occupation,
class, region, gender, religion, color, and social experience. All of
these subcultures intersect within the broader American economic
culture so that accommodation in the linguistic marketplace becomes

akin to the air that individuals breathe.

127

We posit that whatever raising that occurs among African
Americans may be due to instrumental accommodation with women
leading and men holding on to the covert prestige of masculinity
associated with lower low front vowels. There are phonetic and

regional reasons for the advancement of /ae/ and social reasons for

women's overwhelming participation.

The strongest association exists between gender and raising and
gender and duration. This fact of language differentiation has been
found in many studies (Ibrahim 1986; Goldstein 1980; Sachs et al.
1973; Eckert 1989, 2000; Hillenbrand et al. 1995; Labov 2001, Davis
1998) based on physiological as well as social data.

There is support for the hypothesis that African Americans in
Lansing have engaged in institutions in the wider community and have
had sufficient networking experiences in the broader community to
accommodate to the early stages of NCCS based on the percentage of

respondents (64%) participating in /a:/ raising as exemplified by
/2e/ 2 index scores. It was also assumed that African American

accommodation would be similar to the accommodation patterns of
majority speakers in terms of both social and linguistic influences.
Michigan research of rural speakers (Ito 1999) and Appalachian
speakers in Ypsilanti (Evans 2001) indicate that women lead in

adaptation to NCCS. Women lead in the Lansing research as well.

128

In order to understand age and the linguistic marketplace, I refer
to Wolfram (1969) who avowed that African American adults are more
sensitive to socially diagnostic variables than younger people. Eckert
(1997) refers to the linguistic life course and presents evidence that
adults have regularly been shown to be more conservative and
normative in their use of variables than younger people. Often research
into the lives of African Americans does not take into account the
weight of double consciousness and two knowledge levels —— one
learned at an early age and the other acquired in school and through
transactions with the broader community. All of the older raisers are
employed in some aspect of the public sphere, which might increase
their sensitivity to the regional standard or NCCS. More research into
language of African American adults involved in the public in various
capacities is needed.

The prediction that African Americans in Lansing would not
discard all features which have served to differentiate African
American speech form Northern White speech bore fruit in the duration
results which indicated that African Americans in Lansing may

pronounce /2e/ with greater duration than EA speakers. I believe that

this Southern dialect feature may have been retained as an ethnic
marker, which should be studied in conjunction with the African
American prosodic system and not only within the confines of the

African American vowel system.

129

The question of whether local loyalty (e.g., ethnic and
neighborhood) would inhibit accommodation did not bear fruit in this
sample. First of all neighborhoods have changed and there is more than
one location for the Black community today, and a strong or clear
correlation between ethnic loyalty and a raised or lowered low front
vowel could not be established.

The goal of ascertaining the degree to which African Americans
in Lansing had bought into the idea of a regional identity, which in this
case is the identity of the Inland North, was implicit in the research.
The existence of raised patterns juxtaposed against non-raised patterns
would suggest that the raised pattern is a marker of some sort of
affiliation. However, I have not identified this affiliation beyond
regional affiliation. A more finely grained study might capture the
social subcultures that these internal differences mark. It has been
established that the raising pattern is presently intrinsic to the identity
of the Inland North (Labov 1996, Hillenbrand et al. 1995 and Eckert
2000), and could possibly be construed as a marker of Northern
identity. I have posited that Blacks also mark Northern identity with
features of the Inland North.

In addition to questioning whether vowel height and vowel
duration are tied to gender, age, social status, and social network
practices, linguistic factors such a place and manner of articulation

were considered. Fifty-five percent of the twenty raisers show

130

significant differences among the means for following manner of
articulation. Nasals promote and voiceless stops demote with regard to
following manner of articulation. The most predominate ranking in this
research is as follows: N>VDS>VLF>VDF>VLS. Ranking of the mean
scores is similar to other social groups.

There was no effect for following place of articulation. An effect
for preceding place of articulation showed up in the system of one
respondent. In this case obstruents plus liquids demoted low front
vowel raising. Thus, the general conclusions are that the linguistic
factors are universal and function similarly across ethnic groups.
Conclusions have been reached concerning some linguistic and social
variables. This study was limited to the investigation of low front
vowel raising among African Americans because it is considered the
oldest change (Eckert 2000). Accommodation was affected by phonetic
and social facts (women do more of it than men). Style shifting was
not investigated but could be the subject of further research. Research
beyond the formants should also prove productive in isolating exactly
what it is that people do to accommodate and differentiate.

5.6 Final Comments

Individuals are affected by their caregivers and participation in
the local speech community, and then are open to the mitigating
influence of education, personal history, local loyalty, and network.

The social roles, gender roles, and physiques that people have affect

131

speech, but the social environment determines what can be selected.
Thus, with regard to African Americans in Lansing, the choice of
Northern front vowels, the retention of Southern influenced back
vowels and the use of duration present an interesting map of the way an

African American regional identity has been constructed.

132

APPENDICES

133

Occupation:

HOUSING

fl

APPENDIX A: Indices of Social Class
Lawyers, doctors, engineers, judges, architects, and managers of
large businesses
High school teachers, trained nurses, librarians, small business
owners, accountants, large farm owners
Social workers, elementary school teachers, optometrists, and
minor officials of business, bank clerks, auto sales, contractors
Small business managers, typists, mail clerks, most store clerks,
factory foremen, private repairmen (i.e. plumbers)
Beauticians, carpenters, plumbers ( employed by others),
barbers,
firemen, bartenders, restaurant cooks, tenant farmers
Semi-skilled workers, skilled worker assistants, watchmen,
truck drivers, waitpersons (in small restaurants)
Heavy laborers, janitors, newspaper delivery, odd job persons,

migrant workers

Grand, ostentatious
Very good, attractive, roomy, landscaped
Good, only slightly larger than utilitarian demands, more

conventional and less showy than the first two categories

134

Neighborhood

Average, private, one and a half to two story, nice lawns,

some extra room, small, well cared for lawns

Fair, just enough room for needs, well kept up but no

extra

Poor run-down, often too small for needs, not in Shambles or
beyond repair

Very poor, perhaps not even designed as housing, beyond repair

crowded

Very high -— the best place to live in this area; known as the
area of the "well to do"

High — an area with an excellent reputation, low crime, good
schools, large houses and yards

Above average — not pretentious but a nice, clean, and tidy
neighborhood

Average, solid working class area; neat, not fancy but a nice
place to live

Below average, some run-down housing, close to industrial
or other undesirable residence areas

Low, areas regarded as slums

Tenement areas; shacks, lean-tos, "squatters" areas

135

Education

1 Graduate or professional school 2 College

3 High school 4 Some high school
5 Junior high school 6 Elementary

7 Little or no schooling

Computation Occupation X 4 + Education X 3 + Housing X 3 + Neighborhood X 2
Ratings: 12-17 Upper

18-22 Upper-Upper Middle

23-24 Upper-Middle-Upper

25-33 Upper Middle

34-37 Upper Middle-Lower-Middle

38-50 Lower Middle

51-53 Lower Middle-Upper Lower

54-62 Upper Lower

63-66 Upper Lower-Lower Lower

67-69 Lower-Lower-Upper-Lower

70-84 Lower Lower
High school students and non-working spouses have the same scores as the principal

working member of the family.

136

APPENDIX B: The Questionnaire

Urban sound change among African Americans

Give consent form.

1. Where were you born? Are you originally from this area?

2. How long have you lived in this area? Have you or your family moved to many
places? ( If yes) Where? How long did you stay there?

3. What are the names of the schools that you attended?

5. (Workers and professionals) What do you do for a living? Do you like working
there?
What is your title and position?

6. What do you plan to do when you finish school (student)? What do you plan to
do when you retire (employee)?

7. Network relations:
A. Do your best or closest friends live in your neighborhood?

B. Do you have any relatives who live in your neighborhood?

C. Do you know people who also work at your workplace from your
neighborhood?
Do you have coworker of the same sex as you who also live in your
neighborhood?

137

D. Do you spend time with your co-workers after work? How often?

8. What do you usually do when you have spare time?

9. Is this a good place to grow up? Why?

10. Are you planning to stay here afier graduation?

11. Have you ever wanted to live somewhere else?

12. What is the best and worst thing about living in this area?

13. Give the word list and reading passage.

14. Gating experiment (respondents will listen to a tape and write the words that they

hear). Say; “I am going to play a tape for you and I want you to write down what
you hear on this sheet of paper.”

15. Do you know people who say pin when they mean pen?
16. Do you know people who don’t pronounce their rs and say doh for door?

17. Read the debriefing script.

18. Do you have any questions that you would like to ask concerning the interview?

138

APPENDIX C: The linguistic environments used in this research.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

/a=3/ Step 1 Labial/labiodental/ Apical Palatal Velar
N = 31 interdental (alveolar)
Voiceless nap, apple, zap pat, mattress N/A rack, black,
stops
Voiced tab, cabin Dad badge saginaw, brag,
stgps rag
Voiceless bath, laugh ask, past cash, N/A
fricatives mash
Voiced have has, jazz N/A N/A
fricatives
Nasals gamble, Sam Lansing, plant, N/A gang, banker
thank
Liquids N/A pal N/A N/A
/0/ Step 2 Labial/labio- apical palatal velar
N = 17 dental/interdental
Voiceless mop, stop pot watch rock,
Stops block
Voiced bob body logic N/A
Stops
Voiceless profit possible gosh N/A
Fricatives
Voiced father N/A N/A N/A
F ricatives
Nasals Tom John N/A N/A
Liguids N/A car, doll N/A N/A

 

 

 

 

 

139

 

 

 

 

 

 

 

 

 

 

/o/ step 3 Labial/labiodent apical palatal velar
N = 13 al/interdental

Voiceless Stops N/A caught N/A chalk
Voiced Stops N/A N/A N/A dog, fog
Voiceless awful, moth lost N/A N/A
Fricatives

Voiced N/A pause, closet N/A N/A
Fricatives

Nasals N/A gone N/A N/A
Liquids N/A horse, tall N/A N/A

 

 

 

 

 

Vowels of the last three steps of the NCS:

Step 4 (/£/): pen. mesh. bet, fed, step, neck, bend (7)

Step 5 (/A/: bun, puff, cup, sub, duck, dust (6)

Step 6 (/I/): tin, hit, kid, tip, pig, fist, fish, pill (8)

8 other vowels:

(/u/): boot, food, pool (3)

(AM): good, foot, pull (3)

(/o/): hope, hole, road (3)

(/i/): sleep, peel, meat, bead (4)

140

 

(/e/): hate, state, make (3)

(/dI/): bite, night, ride (3)
(/ou/): house, loud, mouse (3)

(/or/): toy, oil (2)

A BAD DAY FOR DUCKS
Tom and Bob were supposed to meet at Tom’s house. They planned to go to a nearby
pond and watch the ducks. While waiting for Bob to get there, Tom picked up around

the house. He put the electric fan away for the winter and did the dishes.

He wanted a snack before he left, so he peeled an apple and cut it into slices. He bit
into one, but it was awful, probably rotten. He spit it out and tried to rinse his mouth
out with hot, black coffee. He poured it into a tin cup, but when he put it to his lips, he
spilled it on his hand. His hand puffed up and hurt a lot, so he stuck it under the faucet
to make it feel better.

He grabbed a dusty hat out of the closet and shook it, but he couldn’t get the dirt off.
He got a cap instead and put a scarf around his neck and put on his socks and boots.
There was a big hole in his sock, and Bob was really late. It was already past 2:00.
Nothing was working out.

Just then Bob phoned and said he wanted to talk. He told Tom that the flock of ducks
had left the pond. A pack of dogs had chased them off. Tom was sad; he had really
wanted to see the ducks, but Bob said that they should go shoot some pool instead.
Tom thought that was a good idea and forgot all about the ducks and his burned band.

141

APPENDIX D: Debriefing script
The Northern Cities Chain Shifi in the Greater Lansing area.

Certain Northern cities in the United States such as Buffalo, Philadelphia, New York,

Chicago and Detroit are undergoing a vowel shift or change. /ae/ as in cat and /a/ as in

pot are two of the vowels which are part of this shift. This means that these vowels
exist in variation with other sounds. That is, the short (a) sound usually found in
words such as cat, man and bag is sometimes pronounced as (e) as in bet or even (i) as
in bit.

Research indicates that certain minority groups and rural groups are not participating
in this vowel shifi. You have been asked to read certain words to determine the extent
of participation by people living in the Lansing area, and to determine whether social

identification is a greater predictor of phonological variation than ethnic identification.

142

Alana, Lansing, 24, Working

APPENDIX E: F1 and F2 /2e/ scores

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
695 1655 apple
693 1719 apple
728 1822 ask
640 1781 badge
306 21 88 banker
667 1 844 bath
749 1675 black
703 1 81 1 brag
732 1744 cabin
640 l 801 cash
662 1 793 dad
691 l 742 gamble
683 1801 has

732 1 700 have
621 1 824 'azz
762 1645 laugh
638 l 854 Lansing
653 1682 mattress
858 1891 mash
595 1660 pal

721 1888 Kit

725 1874 past
679 1 81 7 plant
749 1792 rack

71 0 1 847 rag

749 1750 Saginaw
369 1865 Sam
684 171 1 snack
698 1814 tab
701 1980 zap

 

 

 

143

 

Alton, Lansing, 54, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
992 1932 apple
760 1741 ask

819 1 867 badge
474 l 507 banker
780 2145 bath

871 1 859 black
792 2006 brag
787 1 836 cabin
888 21 76 cash

5 1 7 1464 dad

722 1 897 gamble
601 2646 gang
898 1924 has

81 8 2055 have
743 2095 jazz

781 2944 laugh
673 l 787 Lansing
859 2191 mattress
646 1 507 mash
685 2327 nap

743 1447 pal

860 1873 past

716 2330 plant
689 2072 rack
755 1939 rag

823 2203 Saginaw
664 2036 Sam
655 2645 thank

91 9 2299 zap

 

I44

 

Ann, East Lansing, 42, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
828 2298 apple
628 2358 badge
798 243 1 bad

762 2363 bath
961 l 894 black
784 1939 brag

71 7 2404 cabin
754 2329 cash
663 2563 dad

784 2450 gamble
523 2624 gang
654 2469 has

676 2462 jazz
874 2261 laugh
640 2767 Lansing
945 2305 mattress
833 2374 mash
801 2495 nap

742 2507 pal

764 2487 pat

91 5 21 5 1 plant
715 2216 rack
728 2073 rag

75 1 21 99 Saginaw
658 2566 tab

836 2227 thank
833 201 3 zap

 

145

 

Bertie. Lansing, 42, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
962 2095 apple
960 2273 ask

801 2045 badge
693 2452 banker
908 2261 bath
852 2087 black
785 2 1 1 7 brag
873 2008 cabin
836 2052 cash
665 1998 dad

883 2139 gamble
687 2290 gang

8 l 5 21 80 hand

81 8 23 16 has

890 21 13 have
775 2258 jazz

83 1 2304 laugh
790 2449 Lansing
904 2 1 90 mattress
81 5 2055 mash

83 8 2 178 pal

896 2247 pat

757 2349 plant
854 201 5 rack
794 2328 rag

849 2145 Saginaw
822 2296 tab

762 2344 thank
866 2168 zap

 

146

 

Cassy, Lansing, 21, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
964 1626 apple
81 2 2074 ask
758 2021 badge
833 1922 bath
589 1998 black
802 l 864 brag
503 1 896 cabin
757 1 953 cash
804 1987 dad
580 1648 gamble
537 2194 gang
41 6 1961 has
599 l 881 have
75 1 l 85 1 jazz
589 1998 Lansing
404 2097 mash
406 1995 nap

3 83 1466 pal
842 201 3 pat

63 5 1670 past
561 2239 plant
853 l 875 rack
768 1995 rag
635 1670 Sam
784 1932 sad
477 1 827 tab
446 221 8 thank
745 l 889 zap

 

 

 

147

 

Chuck, Lansing, 28, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
770 1 599 apple
623 1609 ask
630 1630 badge
444 1639 banker
73 1 1702 bath
723 1378 black
670 l 560 brag
714 1693 cabin
469 1725 cash
646 1638 dad
397 1723 gamble
53 8 1 799 gang
579 1662 has
443 l 788 have
723 1688 jazz
719 1538 laugh
400 1626 Lansing
700 1675 mattress
65 8 1702 mash
755 1601 pal

667 1761 pat

648 1 554 past

3 70 l 51 1 plant
609 1 596 rack
622 1650 rag

619 1524 Saginaw
407 l 848 Sam
656 1636 tab

405 1767 thank
750 1605 zap

 

148

 

Crawford, Lansing, 45, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
605 1678 ask

5 1 7 1482 badge
456 1582 banker
548 143 1 bath
624 1443 black
539 1458 brag
608 1619 cabin

5 l 9 1680 cash
528 1569 gamble
550 1603 gang
591 1662 has

561 l 566 have

5 1 7 1637 jazz

601 l 555 laugh
489 1614 Lansing
560 1 507 mattress
640 1701 nap

570 1470 pal

595 1 563 ‘ pat

569 1632 past
623 1 5 16 rack
546 1450 rag

571 1432 Saginaw
499 1659 thank
574 1481 zap

 

149

 

Curtis, East Lansing, 28, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
706 1 5 1 8 apple
577 1452 ask

623 1589 badge
609 2249 black
53 8 1 539 cash
591 l 810 dad

530 1 780 gamble
799 1604 have
451 1 548 jazz
476 1678 Lansing
662 1546 laugh
654 1490 mattress
661 1 63 1 mash
653 1446 ack
599 1555 pal

621 1618 gt

544 1568 past

693 l 863 plant
649 1386 rack

61 8 1 526 rag

562 1 523 Saginaw
592 1788 Sam
521 1748 sad

557 1599 tab

394 1635 thank
571 1503 zap

 

 

 

150

 

Dave, Lansing, 21. Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
744 1573 apple
683 1739 ask

666 l 866 badge
605 1871 banker
733 1723 black
647 161 7 brag
683 1621 cabin
619 1801 cash
571 1778 dad

595 175 l gamble
571 1843 gang
633 1713 has

651 1779 have
619 1698 jazz
687 1641 laugh
581 1787 Lansing
681 1781 mattress
652 1765 mash
644 1658 pal

668 181 5 pat

656 1765 past
612 1772 plant
662 1603 rack
645 1652 rag

603 1675 Saginaw
584 1914 Sam
646 1730 tab

594 1793 thank
674 1656 zap

 

151

 

Debbie, East Lansing, 24, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
919 1 803 ask
646 1852 badge
743 21 74 banker
775 l 809 bath
990 1575 black
775 1766 brag
881 1841 cab
560 2283 cabin
732 1895 dad
689 2024 gamble
603 2196 gang
855 1689 have
602 1 879 jazz
909 1762 laugh
726 2053 Lansing
81 5 l 865 mattress
657 2033 mash
775 1809 nap
949 1709 pal

913 1850 pat

646 1809 past
456 1698 past2
560 1 723 plant
775 1 776 rack
775 1680 rag
689 l 723 sad
819 1826 Saginaw
884 1947 Sam
603 1 809 thank
564 1901 2gp

 

152

 

Dolly, Lansing, 39, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word

8 10 201 6 apple
842 2144 bad

8 1 5 2561 banker
876 2099 black
716 2203 mg
567 2557 dad

674 2483 grabbed
773 2347 has

730 21 l 3 jazz
820 2337 laugh
812 2429 Lansing
796 21 58 map
773 2096 mash
795 2226 nap

795 2249 pat

82 l 21 93 pat2
900 2193 past

895 241 8 planned
788 2004 rack
827 2248 Saginaw
933 21 79 snack
785 2546 tab

497 2129 thank

7 75 2342 zap

 

 

 

153

 

Gregory, Lansing, 66, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
646 1543 ask

625 1595 badge
689 1651 banker
693 1559 bath
738 1239 black
560 1430 brag
633 1659 cabin
561 1751 cash
599 1582 dad

548 1616 fan

680 1807 gamble
659 1720 gang
619 1610 grabbed
615 1630 has

703 1553 hat

554 181 1 jazz

691 1401 laugh
703 1516 Lansing
737 1 5 54 mattress
672 1600 mash
667 1632 nap

694 1561 pat

628 1451 planned
661 1478 plant
662 1523 rack
618 1540 rag

728 1613 snack
639 1556 tab

656 1586 thank

 

154

 

Henry, Lansing, 56, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
653 1 880 ask

522 1959 badge
664 2090 banker
625 l 894 bath

741 l 71 9 black
521 2058 brag
581 1901 cab

634 1 896 cabin
539 1947 cash

5 l 1 1908 dad

695 2074 has

539 1976 have
524 1942 jazz

653 1955 Lansing
73 1 l 842 mattress
670 1959 mash
622 1944 pal

5 10 l 575 planned
729 1942 plant
679 1 845 rack

623 l 806 Saginaw
505 2000 Sam
745 1699 zap

 

155

 

Ibrahim, Lansing, 52, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
728 1630 apple
732 l 881 ask

548 1652 bad

5 1 7 1 804 badge
576 l 814 banker
712 1474 black
547 l 503 brag
588 l 862 cabin
556 l 892 cash
525 1 825 dad

576 1 886 gamble
269 21 59 gang
592 2006 has

619 1 892 have

43 8 1787 jazz
626 1755 Lansing
678 1736 laugh
676 1667 mattress
71 5 1 762 mash
703 1714 pal

601 1 786 pat

692 1 766 planned
668 1 8 1 5 plant
672 1 582 rack
567 1739 rag

666 l 509 Saginaw
673 1672 Sam
524 1752 tab

669 l 727 thank
636 1606 zap

 

 

 

156

 

Jane, Lansing, 25, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
753 2 1 81 apple
563 2286 ask
649 2272 badge
706 2604 banker
738 2335 bath
825 2138 black
780 2274 cabin
433 21 38 cash

61 l 2346 dad
452 2649 gamble
823 2358 has
767 2199 have
662 2248 jazz
903 21 74 laugh
891 23 10 mattress
883 2326 nap
808 1 812 pal

743 2302 pat

739 2273 past
388 2369 plant
753 2160 rack
709 2282 rag

71 1 2282 rag2
691 2817 Saginaw
444 2456 Sam
649 2270 tab

91 8 21 72 zap

 

 

 

157

 

Lorna, Lansing, 28, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
769 1921 apple
733 21 77 ask

803 221 8 badge
494 2756 banker
792 2156 bath
808 21 57 black
705 1870 brag
783 21 79 cabin
834 2190 cash
796 21 83 dad
604 2309 gamble
671 2295 gang
753 2308 has
840 2245 have
716 2103 jazz
890 21 60 laugh
616 223 l Lansing
777 2107 mattress
799 21 35 mash
774 2209 pal

792 2267 pat

848 2089 ast
520 2638 plant
847 2074 rack
794 2059 rag
812 2153 sad
605 2097 Sgginaw
534 2412 Sam
776 2055 tab

61 6 241 0 thank
830 2019 zap

 

158

 

Mali, Lansing, 28, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
91 7 1 559 apple
759 1 757 ask

693 1 841 badge
5 1 7 2 149 banker
686 2038 bath
857 l 588 black
71 8 1 896 brag
744 l 604 cabin
748 1958 cash
788 l 71 1 dad
701 l 71 3 gamble
683 2104 gang
746 1977 has
806 1761 have
760 1926 'azz
788 1957 laugh
605 1 841 Lansing
740 1 837 mattress
678 21 53 mash
814 1474 pal

737 1 507 pat

736 2093 past
723 21 57 plant
784 1 737 rack
709 1 859 rag

740 1 720 Saginaw
660 2208 Sam
780 1 778 tab

685 2236 thank
807 l 776 mg

 

159

 

Mandy, Lansing, 39, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
941 1808 apple
907 1985 ask
708 2052 bad

81 1 2012 badge
796 2148 banker
842 1901 bath
727 1826 brag
830 2080 cabin
778 2097 cap
749 1999 cash
673 2168 dad
774 2256 gang
842 1959 has
814 2016 hat
917 1924 have
794 21 1 0 jazz
867 l 871 laugh
871 1928 mattress
865 21 10 mash
866 21 87 nap
805 1829 pack
702 1879 past
853 2096 gast2
747 1 896 planned
853 2041 plant
802 2230 rack
825 1946 rag
787 221 5 sad
772 1839 Saginaw
854 2064 Sam
833 2194 thank

 

160

 

May, Lansing, 19, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
646 2024 apple
756 2341 ask

6] 6 2163 badge
823 1982 bath
842 1 845 black
745 21 32 brag
744 1 879 cabin
571 21 33 cash
670 1584 dad

664 2285 gamble
768 1956 grabbed
720 2105 has

8 1 l 191 6 hat

721 223 1 jazz
859 2101 laugh
553 2258 Lansing
744 1958 mattress
75 1 1962 mash
780 1 852 nap

565 2294 pal

732 2196 pat

734 1626 past
550 23 1 7 plant
923 2100 rack
673 l 735 rag

748 1 873 sad

764 1994 Saginaw
43 8 1743 Sam
837 1995 snack
707 1937 tab

964 21 3 1 thank
918 2161 zap

 

I61

 

Nancy. Lansing, 27, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
965 1709 apple
934 2058 ask
753 21 1 5 badge
562 2508 banker
854 2098 bath
844 2238 black
764 2030 brag
844 2132 cabin
804 2198 cash
797 2142 dad
793 2136 gang
799 21 87 has
860 2123 have
781 2093 jazz
780 2103 laugh
765 21 54 mattress
881 213 1 mash
800 1958 pal

821 21 84 pat
840 2104 past
524 2349 plant
856 1979 rack
740 2047 Saginaw
585 2364 Sam
758 2200 tab
467 245 1 thank
849 191 7 zap

 

 

 

162

 

Norma, Lansing, 42, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
825 1983 ask
794 2089 badge
799 21 1 8 bath
880 1 883 black
796 1952 brag
816 2052 cabin
819 21 10 cash
766 2129 dad
786 1 51 9 gamble
421 2413 gang
81 l 2025 has

801 l 881 have
840 21 1 8 jazz
744 2134 laugh
766 2533 Lansing
862 2008 mattress
769 1979 mash
85 1 2051 nap
758 2027 pal

856 21 70 pat

81 5 201 5 past
792 l 881 rack
749 1993 rag
723 23 19 Sam
748 2161 tab

476 2420 thank
843 21 14 zap

 

163

 

Nubia, Lansing, 27, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
872 21 14 ask

832 2077 badge
909 1 803 black
801 1 879 brag
860 l 800 cash
804 2279 dad

815 2437 gamble
807 21 59 has

807 1964 have
737 2020 jazz
862 1 820 laugh
841 2487 Lansing
921 1943 mattress
627 2125 mash
939 1621 pal

777 2647 pat

834 2048 past
956 2410 rack
817 2007 rag

771 1975 sad

732 l 852 Saginaw
847 2034 Sam
888 2223 tab

429 2461 thank
821 1676 zap

 

 

 

164

 

Olive, East Lansing, 48, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
982 2402 apple
543 191 8 ask

63 8 1 739 badge
626 2357 banker
829 21 23 bath
884 1663 black
673 1648 brag
765 1 825 cabin
896 2104 cash
699 2 l 21 dad

641 l 837 gamble
772 2300 gang
671 1927 grabbed
782 1 762 jazz
864 l 720 laugh
625 1 963 Lansing
799 1 786 mattress
802 2246 mash
772 1750 nap

794 1981 pack
827 2083 pal

901 2200 pat

592 1 743 planned
356 2219 plant
782 1689 rack
723 2043 rag

746 1968 sad

768 1916 Saginaw
41 5 2597 Sam
798 21 56 tab

330 l 710 thank
804 1673 zap

 

 

 

165

 

Paul, Lansing, 41, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
539 1663 ask

5 l 4 1660 badge

3 14 23 12 banker
496 l 573 bath
633 1411 cap
497 1685 cash
475 1658 gamble
471 1730 gang
507 1503 has

501 163 1 jazz
445 21 73 Lansing
520 l 563 mattress
307 2274 nap

573 l 539 pal

528 1712 pat

502 1452 past
528 1532 pack

5 63 1 530 rack
575 1493 Saginaw
505 1 546 snack
467 1794 thank
502 2326 zap

 

 

 

166

 

Qasim, Lansing, 53, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
778 1661 apple
635 2017 ask

644 1909 badge
626 2014 banker
632 1897 bath
764 1933 black
704 1828 cabin
538 1991 cash
537 1989 dad

73 8 1 593 gamble
523 2082 gang
550 1979 has

638 1826 have
576 1982 jazz
639 1689 laugh
704 l 828 Lansing
789 l 812 mattress
656 1 840 mash
503 1980 pal

708 1836 pat

701 1914 planned.
466 1441 plant
712 1855 rgg

755 1567 Saginaw
379 1950 Sam
579 1851 tab

774 1888 thank

 

167

 

Rachel, Lansing, 48, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
881 1749 apple
784 1 885 ask

73 1 2469 banker
735 1899 bath
819 1 807 black
730 1924 brag
791 l 809 cabin
717 1982 cap

73 1 1946 cash
682 2619 dad

807 1933 gamble
737 2334 gang
770 2269 hand
754 1937 have

73 1 1936 jazz

741 2002 laugh

71 5 2468 Lansing
850 l 749 mattress
755 1998 mash
762 l 746 pal

741 1930 pat

782 1914 path
819 2224 plant
748 2004 rag

720 2055 sad

772 1795 Saginaw
791 2296 Sam

73 1 1750 tab

 

 

 

168

 

Rana, Lansing, 50, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
970 1 847 ask
566 2090 badge
720 2563 banker
533 2163 bath
880 2069 black
655 221 7 brag
604 1 863 cabin
603 221 8 cash
658 2599 dad
577 2078 gamble
775 1697 gang
798 2048 has
912 1 910 have
607 2063 jazz
638 l 854 laugh
732 l 852 mattress
403 2007 mash
824 1633 nap
978 1 576 pal

636 1798 Lat

999 l 805 past
632 2537 plant
560 1971 rack

81 5 1 636 rag
667 1935 Saginaw
545 2484 Sam
691 2301 tab

534 2606 thank
906 2366 zap

 

169

 

Rhonda, Lansing, 38, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
73 1 1602 apple
532 1989 ask

535 1999 badge
791 1573 banker
521 1961 bath
719 l 724 black
700 1746 brag
655 1 890 cabin
477 1996 cash
481 2108 dad
637 1613 gamble
467 2138 gang
529 1948 has

501 2036 have
498 1934 jazz
621 1 81 7 laugh
504 2355 Lansirg
603 1740 mattress
594 2044 mash
702 1776 nap
603 l 853 pal

5 1 8 1988 past
748 1939 pat

483 2201 plant
712 1782 rack
500 1 874 rag

667 1 797 Saginaw
476 1904 Sam
652 1703 tab

715 1 870 zap

 

170

 

Thomas, Lansing, 27, Middle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F 1 F2 word
863 1 825 apple
737 l 806 ask

658 l 71 3 bad

647 l 845 badge
583 2029 banker
694 171 8 bath
792 1464 black
832 1653 black2
685 1476 brag
722 1692 cabin
658 1753 cash

61 7 1786 dad

563 1951 gamble
63 l 1865 gang
653 1 562 grabbed
524 2328 hand
669 1 867 has

745 1 582 hat

7 1 5 1 714 have
658 1773 jazz
776 1782 laugh
686 1 693 mattress
649 1 770 mash
798 1799 fill

714 1778 past
639 1641 planned
599 1838 plant
788 1785 rack
671 1608 rag

696 1636 Saginaw
541 1758 Sam
734 1645 snack
709 1807 tab

533 2066 thank
737 1724 zap

 

 

 

l7l

 

Veronica, Lansing, 42, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
837 1904 ask

606 2123 badge
329 241 5 banker
752 2061 bath
745 1975 black
661 1998 cash
610 2248 dad

761 2123 gamble
465 2417 gang

63 8 1648 has

746 l 888 have
574 2054 jazz
710 1914 laugh
429 2237 Lansing
941 1980 mattress
744 2033 mash
827 1975 pat

694 2212 past
809 1991 plant
754 21 3 1 rack
659 1956 rag

687 1812 Saginaw
404 2365 Sam
764 21 34 tab

350 1888 thank

 

 

Winston, Lansing, 74, Working

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 F2 word
754 l 500 apple
605 1 838 ask

600 1739 badge
591 1 807 banker
653 1 703 bath
660 1460 black
635 l 852 brag
662 1747 cap
576 1 843 cash
500 1927 dad

589 1646 fan

566 1 865 gamble
557 1932 gang

5 10 l 567 grabbed
5 1 5 1 846 hand
646 1 75 l hat

528 l 862 have
645 1623 laugh
621 1 705 Lansing
660 1927 mattress
645 1 766 mash
656 1708 nap

737 1490 pack
677 1 8 1 7 at

576 1 728 past

601 1753 past2
589 1619 planned
686 1476 plant
725 1645 rack
699 l 532 Saginaw
592 1 71 5 Sam
643 1 503 snacks
529 1 871 tab

695 l 587 thank
649 1684 zap

 

173

 

APPENDIX F
ANOVA Summary

Analysis of Variance (Significant Results)
Dependent Variable: F1

Independent Variables: HS = Preceding segments

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

, source DF Sum of ; Mean F Value Pr>F
‘ ' Squares " Squares 1
[Ann 0H8 5 . 117415.0225 23483.0045 7 2.69 1 0.0497
Dependent Variable: F 1
Independent Variables: = Following manner of articulation
Name Source DF Sum of Squares Mean Squares F Value Pr>F
Alana manner 4 168700.9473 42175.2368 4.36 0.0101
Alton manner 4 297931 .2023 74482.8006 6.08 .0002
Bertie manner 4 66491 .91534 16622.97884 4.78 .0063
Crawford manner 4 26869.25217 6717.31304 5.72 .0038
Jane manner 4 340441.1099 851 10.2775 6.96 .0010
Mali manner 4 83033.17619 20758.29405 6.82 .0009
Nancy manner 4 24891 1.9400 62227.9850 l 1.51 .0001
Norma manner 4 126174.6409 31543.6602 4.10 .0166
Olive manner 4 387749.4167 96937.3542 7.86 .0003
Thomas manner 4 149318.2902 37329.5725 18.68 .0001
Veronica manner 4 297102.2714 742755679 4.81 .0061

 

 

174

 

APPENDIX G

INDIVIDUAL VOWEL PLOTS

175

2500 2300

Alana'a vowel plot

F2

2100 1900 1700 1500 1300

 

 

O—n-

 

 

Name: Alana

Sex: Female
Age: 28
Class: Working

F 1 /2e/ index score: 2

Net E Score: 4
Net S Score: 0
Gating Score: N/A

176

500

100
200 ‘
300
400
500
600
700
800
900

1000 ‘

Alton'a vowel plot

F2
2500 2300 2100 1900 1700 1500 1300 1100 900 700 500

 

 

 

 

‘ 100
i
. . 300
e e u
, o
I
° .0 500
a: e
o
O o A o 700
(11 °
0 0 all
0
a
900
Name: Alton Fl /ae/ index score: 2
Sex: Male Net E score: 4
Age: 64 Net S score: 3

Class: Working Gating: 10

177

Ann's Vowel Plot

 

 

 

 

3000 2500 2000 1500 1000 500
200
300
i
9 u
. 400
o
I O O 500
0 0y
. O
600
8
as e 700
O
A 0° 800
O
900
1
.q' du ,0
. 1000
1100
Name: Ann Fl /ae/ index score: 2
Sex: Female Net E score: 2
Age: 42 Net S score: 1

Class: Middle

Gating: 1 5

178

2600 2100

Vowel plot for Bertie

F2
1600 1100 600

100

 

 

100

200

300

l: 400

500

600

700

- 800

 

 

Name: Bertie

Sex: Female
Age: 42
Class: Working

Fl /ae/ score: 2

Net E score: 4
Net S score: 3
Gating: 13

179

900

Casey's vowel plot

 

r:
2500 2300 2100 1900 1700 1500 1300 noo goo 700 500
. U
l
. e o
O
I e A °
0 . .
.8
0
e

 

 

 

Name: Cassy
Sex: Female
Age: 21
Class: Middle

F l /m/ index score: 2

Net E score: 3
Net S score: 1

Gating: 14

180

200

300

400

500

600

700

800

F2

Chuck‘s vowel plot

 

 

 

 

2300 2100 1900 1700 1300 1100 900 700 500
' . L 200
. u
1 . 300
O
or
0)’
e I u
o . e 0 ° 400
e
O A .0
O
» 500
a: ’3
0 cu 600
O
o
e 700
800
Name: Chuck F1 /a=: / index score: 1
Sex: Male Net E score: 3
Age: 28 Net S score: 3

Class: Working

Gating: 16

181

Crawford's vowel plot

F2

 

1900 1700 1500 1300 1100 900 700 500
, j , L - - 1 200
300
i
O
O U
U 400
e 0 I 0 0
0 ° . or
0y
0
e 0 cu 500
O
O
O a .A O
600
01 Cl
. .
700

 

 

 

Name: Crawford F1 /a2/ index score: 2

Sex: Male NetEscore: 4
Age: 45 NetSscore: 3
Class: Middle Gating: 14

182

Curtis' vowel plot

F2
2100 1900 1700 1500 1300 1100

500

 

ll 01'
O
O
I
O .
l .U 0y
0 A O .
O 0
e 8
8 0
GI
$ 0
. Cu O
O O
.0

 

200

300

400

500

600

 

700

 

Name: Curtis F 1 /2e / index score: 1

Sex: Male NetEscore: 3
Age: 28 NetSscore: 2.
Class: Middle Gating: 12

183

Dave's vowel plot

F2

 

2200 2000 1800 1600 1400 1200 1000 800 600 400 200
i
o
u
e
e I
0
e .0 0y e
co
8 31'
O A
e
2e 0
° 0
(11
e
C
. all
e

 

 

 

Name: Dave

Sex: Male
Age: 21
Class: Middle

Fl /2c/ index score: 2

Net E score: 4
Net S score: 1
Gating: 17

184

200

300

400

500

600

700

800

2500

Debbie's vowel plot

 

 

F2
2000 1500 1000 500
u
C
e
i e
e U ° 0y
0
e
e
E
e
a . (II 0
au
0 ° .
0A
0
e

 

 

Name: Debbie
Sex: Female
Age: 24
Class: Middle

F /m/ index score: 2

NetE: 2
NetS: 3
Gating Score: 10

185

300

400

500

600

700

800

900

1000

Dolly's vowel plot

F2

2500 2300 2100 1900 1700 1500 1300 1100 900 700 500
* ' * ‘ 200

 

300
u 400

e 500
° 0 600

31' 700

O
O A O
. 800

90‘)

900

 

 

1000

 

Name: Dolly F 1 /m/ index score: 2

Sex: Female NetE score: 3
Age: 39 NetSscore: 0
Class: Middle Gating: 12

186

Gregory's vowel plot

 

F2
1900 1700 1500 1300 1100 900 700 500
. 200
- i 300
’ u
0
400
u
e e .
e . 0 or 500
. 0
O
o
A o 0 y 600
E O 0
° ou
(11 e 700
o ’0
aoo

 

 

 

Name: Gregory F1 /ee/ index score: 1
Sex: Male NetEscore: 4

Age: 66 NetSscore: 1

Class: Middle Gating: 9

187

Henry‘s vowel plot

 

 

 

 

F2
2500 2300 2100 1900 1700 1500 1300 1100 900 700 500
A ‘ ‘ ‘ 200
i 300
0 u
e o
. 400
OI U 0.
° 500
0 0y
8
O
.33 600
A
° 700
.01 0 cu
a
o 800
900
Name: Henry Fl /ae/ index score: 2
Sex: Male NetEscore: 4
Age: 56 NetSscore: 3

Class: Middle

Gating: 13

188

Ibrahlm'o vowel plot

 

 

 

 

F2
2100 1900 1700 1500 1300 1100 500
‘ 200
i
O
u 300
0
e 400
s I U
0
e o o 0
y 500
or ’
O
A o
3‘3. , 600
au 0
a
or .
. 700
800
Name: Ibrahim F1 /ae/ index score: 1
Sex: Male Net E score: 4
Age: 52 NetSscore: 2

Class: Middle

Gating: 14

189

Jane's vowel plot

 

 

 

 

F2
3000 2500 2000 1500 1000 500
l ' 300
400
L1
’° 0 ° or
I U: o
o 0), 500
e . A 600
a O
O 0'
. 700
O
0“ 800
900
Name: Jane F1 /ae/ index score: 2

Sex: Female
Age: 25
Class: Middle

Net E score: 4
Net S score: 4

Gating: 16

190

Lorna‘s vowel plot

 

F2
3000 2500 2000 1500 1000 g 500
U
0
C
O
U C
l . . o
8
° or
O
A 9 cu
. 0
28
C11 0 . 3
O

 

 

 

Name: Lorna
Sex: Female
Age: 28
Class: Working

F /a:/ index score: 1
Net E: 4

Net S: 5

Gating: 12

191

300

400

500

600

700

800

900'

 

 

 

 

Mali's vowel plot
: F2
; 2400 2200 2000 1300 1600 1400 1200 1000 800 600
' A A ‘ 200
- 300
u
| O 400
O
r 6 0y 0 soc
° 0
. O
. 0 U
0
600
.or
.08 01 A V 700
E . 3.
am 800
00
900
Name: Mali F 1 /ae/ index score: 2

Sex: Female
Age: 28
Class: Middle

Net E score: 3
Net S score: 4
Gating: 12

192

Mandy's vowels

 

 

 

 

F2
2500 2000 1500 1000 500
200
300
.i u
° 400
r u
e
° ° ° » 500
.0
600
7
.8 2 00
A
f o 9 800
or au
0 . 900
1000
Name: Mandy F 1 /2e/ index score: 1
Sex: Female Net E score: 2
Age: 39 NetSscore: 2

Class: Middle

Gating: 16

193

May's vowel chart

 

 

 

 

F2
2300 2100 1900 1700 1500 1300 1100 900 700 500
A ‘ - , 200
i 300
O
u
o 400
e
1 U o
. ° ° 0 y 500
.0
e 600
° A o
O
a: o 700
O
or
° 0 cu 800
a
. 900
1000
Name: May F1 lae/ index score: 1

Sex: Female
Age: 19
Class: Working

Net E score: 4
Net S score: 0
Gating: 15

194

3000

Nancy's vowel plot

F2

2500 2000

1500

1000

500

 

 

200

300

400

500

600

700

800

900

 

 

Name: Nancy

Sex: Female
Age: 27
Class: Working

F1 /a=:/ index score: 2

Net E score: 4
Net S score: 5
Gating: 14

195

1000

3000

Norma‘s vowel plot

F2

 

 

2500 2000 1500 1000 500
‘ ' 100
r300
u 01'
' O 0
e l
o o O 0 U 500
I ’0y
9
A
8 700
E o .
0
0U
01
O a .
O
r 900
1100

 

 

Name: Norma

Sex: Female
Age: 42
Class: Working

F /22/ index Score: 2
Net E: 3

Net S: 1
Gating: 13

196

2800 2300

Nuble'e vowel plot

 

 

F2
1800 1300
8
O
a: A . 011
° 0 e 9
CI
0 .0

 

 

Name: Nubia

Sex: Female
Age: 27
Class: Working

F /2e/ index score: 2

Net E Score: 4
Net S: 1
Gating Score: 13

197

300

200

300

400

500

600

700

800

900

1000

 

2300 2100

Ollve'e vowel plot

F2

1900 1700 1500 1300 1100 900 700

500
200

 

 

300

400

500

600

700

800

900

 

1000

 

Name: Olive
Sex: Female
Age: 48
Class: Middle

F1 /ze/ index score: 2

Net E score: 4
Nets score: 4
Gating: 14

198

Paul's vowel plot

 

 

 

 

F2
2100 1900 1700 1500 1300 1100 900 700 500 300
- t 300
350
i “.
° 400
e
o I. u
e 450
° 0
O
or 3
O
a: A'ou o 500
a
O 550
600
Name: Paul F lae/ index score: 2
Sex: Male Net E score: 3
Age: 4 Net S: 2

Class: Working

Gating score: 14

199

Oaslm'e vowel plot

 

F2
2300 2100 1900 1700 1500 1300 1100 900 700 500
: . 1 ‘ : 200
. O
L l u
. 300
I
O U
9 . 400
O
E
500
O
A
’0 or
O O
‘ e
(l 0 (111
O

00 700
800

 

 

 

Name: Qasim
Sex: Male
Age: 53

Class: Working

F1 /m/ index score: 1

NetE Score: 4
Net S Score: 0
Gating: 11

200

Rachel's vowel plot

 

 

 

 

F2
2500 2300 2100 1900 1700 1500 1300 1100 900 700 500
A - ' * 200
300
1 , u 400
e O
O
1 . 500
U 0 CV or
O
e 600
° 0
e. A . on 700 .
$° e
,3 800
or. a
e 900
1000
Name: Rachel F /ae/ index score: 2
Sex: Female NetE Score: 4
Age: 48 NetS Score: 2

Class: Working Gating Score: 15

201

2600 2400

2200 2000 1800 1600 1400 1200

Rana's vowel plot

F2

1

600

 

i
e

 

.I 0y °
e 0 e
8 Q A
. 01'
O
a. .0“ O

' 200

300

400

500

600

700

800

900

 

1000

 

Name: Rana
Sex: Female
Age: 50

Class: Middle

F /&/ index score: 2
Net E: 3
NetS: 2

Gating score: 16

202

2500 2300

2100

Rhonda's vowel plot

F2

1900 1700 1500 1300

A

1100 900

700

500

 

 

200

300

400

500

600

700

 

800

 

Name: Rhonda

Sex: Female
Age: 38
Class: Working

F1 /&/ index score: 1

Net E score: 5
Net S score: 1
Gating: 16

203

 

2300 2100

1900

Thomas' vowel plot

F2
1500 1300 1100 900 700

500
200

 

QC

 

to

Go

300

400

500

600

700

800

 

900

 

Name: Thomas

Sex: Male
Age: 27
Class: Middle

F 1 /2e/ index score: 2

Net E score: 3
Net S score: 3
Gating: 13

204

2300 2100

Veronica's vowel plot

1900 1700 1500

1300

1100 900 700

500
200

 

’98

 

01

.>

0!:

03'. e or

300

400

500

600

700

800

900

 

1000

 

Name: Veronica

Sex: Female
Age: 43
Class: Working

Fl /&/ index score: 2

Net E score: 2
Net S score: 2
Gating: 15

205

2000

Winston's vowel plot

 

 

F2
1800 1600 1400 1200 1000 800 600 400 200
A i L 2 , 200
i
O
u
r ’ 400
u
0 e
e . ‘2
e 0)’
O O
A 600
a , ° 3
0
cu
or e
O
O a
800

 

 

Name: Winston F1 /2e/ index score: 1

Sex: Male
Age: 74

Net E score: 4
Net S score: 4

Class: Working Gating: 12

206

REFERENCES

Abu Lughud, Lila. 1993. Writing Women's Worlds: Bedouin Stories.
Berkeley: University of California Press.

Appiah, Kwame Anthony. 1998. The uncompleted argument: Du Bois
and the illusion of race. In Race, Class, Gender, and Sexuality:
The Big Questions. eds. Naomi Zack, Laurie Shrage, and Crispin
Sartell. Oxford: Blackwell.

Ash, Sharon and John Myhill. 1986. Linguistic correlates of inter-
ethnic contact. In Diversity and Diachrony. (ed.) David
Sankoff. Amsterdam: John Benjamins Publishing Company.

Bailey, Guy and Erik Thomas. 1998. Some aspects of African-
American vernacular English phonology. In African American
English. eds. Salikoko S. Mufwene, John Rickford, Guy Bailey,
and John Baugh.

Bailey, Guy and Natalie Maynor. 1989. The divergence controversy.
American Speech. 64: 12-39.

Baugh, John. 1983. Black Street Speech. Austin: University of Texas
Press.

1999. Out of the Mouths of Slaves: African American
Language and Educational Mal-Practice. Austin: University of
Texas Press.

. 2000. Beyond Ebonics: Linguistic Pride and Racial Prejudice.
Oxford: Oxford University Press.

Benjamin, Lois. 1991. The Black Elite: Facing the Color Line in the
Twilight of the Twentieth Century. Chicago: Nelson-Hall
Publishers.

Butterfield, Fox. 2002. Study finds big increase in Black men as
inmates since 1980. The New York Times. (www.nytimes.com).

Callary, Robert E. 1975. Phonological change and the phonological
development of an urban dialect in Illinois. Language in Society.
4: 155-170.

Childs, Becky and Christine Mallinson. 2003. The Regional alignment

207

of African American English in the Smoky Mountains. Paper
presented at ADS Annual Meeting. Atlanta, Georgia.

Davis, Angela Y. 1983. 2nd. ed. 1998. Women, Race and Class. New
York. Vintage Books.

Denning, Keith. 1989. Convergence with divergence: A sound change
in vernacular Black English. Language Variation and Change.
Cambridge: Cambridge University Press.

Deser, Toni. 1991. Dialect Transmission and Variation. Indiana
Linguistic Club.

Dorill, George. 1986. A comparison of stressed vowels of Black and
White speakers in the South. In Language Variety in the South.
eds. Michael B. Montgomery and Guy Bailey. University:
University of Alabama Press.

Eckert, Penelope. 1989. The Whole woman: Sex and gender
differences in variation. Language Variation and Change.
Cambridge University Press. 245-266.

1988. Adolescent social structure and the spread of linguistic
change. Language in Society. Cambridge: Cambridge University
Press.

. 2000. Linguistic Variation as Social Practice. Oxford:
Blackwell Publishers.

. 1997. Age as a sociolinguistic variable. In The Handbook of
Sociolinguistics. ed. Florian Coulmas. Malden: Blackwell
Publishers.

Edwards, Walter F. 1992. Sociolinguistic behavior in a Detroit inner-
city black neighborhood. Language in Society. 21: 98-115.
Cambridge: Cambridge University Press.

Evans, Betsy. 2001. Dialect contact and the Northern Cities Shift in
Ypsilanti, Michigan. Ph.D. dissertation. Michigan State
University.

Feagin, Crawford. 1996. Peaks and glides in Southern States short-a
Towards a Social Science of Language: Vol. 1. Variation and
Change in Language and Society. eds. Gregory Guy, Crawford
Feagin, Deborah Schriffin, and John Baugh. Amsterdam:

John Benjamin.

208

Feagin, Joe R. and Melvin P. Sikes. 1994. Living with Racism: The
Black Middle-Class Experience. Boston: Beacon Press.

Fridland, Valerie Michelle. I998. The Southern Shift: Linguistic and
Social Factors. Ph. D. dissertation. Michigan State University.

Flege, James Emil. 1995. Second language speech learning theory,
findings, and problems. In Speech Perception and Linguistic
Experience: Issues in Cross-Language Research. ed. Winnifred
Strange. Baltimore: New York Press.

Gal, Susan. 1995. Language, gender and power: An anthropological
review. In Gender Articulated: Language and the Socially
Constructed Self. eds. Ira Hall and Mary Bucholtz. New York:

Routledge.

Giles, Howard. 1973. Accent mobility: A model and some data.
Anthropological Linguistics. 15: 87-105.

1991. Accommodation theory: Communication, context and
consequence. In Contexts ofAccommodation: Development in
Applied Sociolinguistics. eds. Howard Giles, Nikolas Coupland
and Justine Coupland. Cambridge: Cambridge University Press.

Goldstein, Ursula Gisela. 1972. An articulatory model for the vocal
tract of growing children. Ph.D. dissertation. Massachusetts
Institute of Technology.

Gordon, Mathew James. 1997. Urban sound change beyond city
limits: The spread of the Northern Cities Chain Shift in
Michigan. Ph.D. dissertation. Ann Arbor: University of
Michigan.

Hagiwara, Robert. 1997. Dialect variation and formant frequency: The
American English vowels revisited. Journal of the Acoustical
Society ofAmerica. 102 (1). 655-658.

Hawkins, Homer. 11979. The destruction of a Black community: A
Case study of Lansing, Michigan. In Blacks and Chicanos in
Urban Michigan. eds. Homer C. Hawkins and Richard W.
Thomas. Michigan Department of State.

209

Jones, Jamila. 1997. lae/ raising among biracial children in the
Greater Lansing area. unpublished paper. Michigan State
University.

Kernan, Claudia Mitchell. 1971. Language Behavior in a Black Urban
Community. Monographs of the Language-Behavior Research
Laboratory. Number Two.

Kohl, Herbert and James Hinton. 1972. Names, graffiti, and cultures.
In Rappin and Stylin Out: Communication in Urban Black
America. ed. Thomas Kochman. Urbana: University of
Illinois Press.

Kuhl, Patricia, and Paul Iverson. 1995. Linguistic experience and the
perceptual magnet effect. In Speech Perception and Linguistic
Experience: Issues in cross-language research. ed. Winifred
Strange. Baltimore: York Press.

Labov, William. 1972. Sociolinguistic Patterns. University of
Pennsylvania Press.

______ 1972. Language in the Inner City. University of
Pennsylvania Press.

1980. The social origin of sound change. In Locating
Language in Time and Space. ed. William Labov. New York:
Academic Press.

1994. Principles of Linguistic Change: Internal Factors.
Oxford: Blackwell.

. 2001. Principles of Linguistic Change: Social Factors.
Oxford: Blackwell.

1996. The organization of dialect diversity in North America.
Telsur project. Linguistic laboratory. University of
Pennsylvania. (www.ling.upenn.edu/phono_atlas/ICSLP4.htm1).

_____ and Malcah Yaeger and Richard Steiner. 1972. A Quantitative
Study of Sound Change in Progress. Philadelphia: The US.
Regional Study.

and Wendell Harris. 1986. Defacto segregation of Black and
White vernaculars. In Diversity and Diachrony. ed. David
Sankoff. Amsterdam: John Benjamins Publishing Company.

210

Ladefoged, Peter. 1993. A Course in Phonetics. Third edition.
Harcourt Brace College Publishers.

LaFerriere, Martha. 1979. Ethnicity in Phonological Variation and
Change. Language. Vol. 55. No. 3. Baltimore. Linguistic
Society of America.

Le Page RB. and Andree’ Tabouret-Keller. 1985. Acts ofidentity:
Creole-Based Approach to Language and Ethnicity. Cambridge:
CambridgeUniversity Press.

Lippi-Green, Rosina. 1997. English with an Accent. London and New
York: Routledge.

Martinet, Andre. 1952. Function, structure and sound change. Word.
8:1-32

Meyer, Douglass K. 1970. The changing Negro Residential Patterns
in Lansing, Michigan, 1850-1969. A Ph.D. Thesis. Department
of geography. Michigan State University.

Milroy, Leslie. 1980. Language and Social Networks. Oxford: Basil
Blackwell.

_____ and James Milroy. 1992. Social network and social class:
Toward an integrated sociolinguistic model. Language in
Society. 21:1-26. Cambridge: Cambridge University Press.

Morgan, Marcyliena. 1994. Theories and politics in African American
English. Annual Review ofAnthropology. 23:325-45.

_____ 1998. More than a mood or an attitude: discourse and verbal
genres in African-American culture. In African-American
English: Structure, History and Use. eds. Salikoko S.
Mufwene,John R. Rickford, Guy Bailey and John Baugh.
London: Routledge.

1994. The African-American speech community: Reality and
sociolinguistics. In the Social Construction of identity in
Creole Situations. ed. Marcyliena Morgan. 121-148.

Nieddzielski, Nancy and Dennis R. Preston. 2000. Folklinguistics.
Berlin: Mouton de Gruyter.

Payne, Arvila. 1980. Factors controlling the acquisition of the
Philadelphia dialect by out-of-state children. In Locating

211

Language in Time and Space. ed. William Labov. New
York: Academic Press.

Peterson, Gordon E. and Harold Barney. 1952. Control methods
used in a study of the vowels. Journal of the Acoustical
Society ofAmerica. Vol. 24. No. 2.

Peterson, Gordon B. and Ilse Lehiste. 1960. Duration of syllable
Nuclei in English. The Journal of the Acoustical Society
OfAmerica. Vol. 32. Number 6. 693-703.

Pickett, J.M. 1999. The Acoustics ofSpeech Communication:
Fundamentals, Speech Perception Theory, and Technology.
Boston: Allyn and Bacon.

Preston, Dennis R. 1986. Five visions of America. Language in
Society. 15: 221-40.

1993. Two heartland perceptions of language variety. In
"Heartland" English: Variation and Transition in the American
Midwest. ed. Timothy C. Frazer.

_____ , Evans, Betsy, Rika Ito and Jamila Jones. 2000. Change on top
of change: social and regional accommodation to the Northern
Cities Chain Shift. In Jan Stroop (ed), Festschrift for Jo Daan.
Amsterdam: Meertons Institute.

Rosner, BS. and J.B. Pickering. 1994. Vowel Perception and
Production. Oxford: Oxford University Press.

Sachs, Jacqueline, Philip Lieberman and Donna Erickson. 1973.
Anatomical and cultural determinants of male and female speech.
In Language Attitudes: Current Trends and Prospects. eds.
Roger W. Shuy and Ralph W. Fasold.

Smedley, Audrey. 2001. Social origins of the idea of race. In Race in
21st. Century America. eds. Curtis Stokes, Theresa Meléndez,
and Grace Rhodes Reed. East Lansing: Michigan State
University Press.

Smitherman, Geneva. 2000. Talkin That Talk: Language Culture and
Education in African America. London: Routledge.

1977. Talkin and Testifyin. Boston: Houghton Mifflin.

Spears, Arthur K. 1999. Race and Ideology: Language, Symbolism,

212

and Popular Culture. Detroit: Wayne State University Press.

_____ .2001. Ebonics and African American English. In Ebonics and
Language Education. ed. Clinton Crawford. London: Sankofa
World Publishers.

Stanback, Marsha Houston. 1985. Language and a Black woman's
place: Evidence from the middle-class. In For Alma Mater.
eds. Paula A. Treichler, Cheris Kramarae and Beth Stafford.
Urbana: University of Illinois Press.

Steele, Shelby. 1998. A Dream Deferred: The Second Betrayal of
Black Freedom in America. HarperCollins Publishers.

Stevens, Kenneth N. 1998. Acoustic Phonetics. Cambridge: The
MIT Press.

Stockman Ida and Newkirk, Brandi L. 2000. Regional differences and
selected forms of African American English in 3 year old
African American Head-Start students. Unpublished paper
Department of Audiology. Michigan State University.

Strange, Winifred, James J. Jenkins and Thomas L. Johnson. 1983.
Dynamic specifications of coarticulated vowels. Journal of
the Acoustical Society ofAmerica. 74 (3), September 1983.

Trudgill, Peter. 1986. Dialects in contact. Oxford: Basil Blackwell.

Thomas, Eric. 1997. A Compendium of Vowel Plots. Raleigh, North
Carolina: North Carolina Language and Life Project.
Department of English, North Carolina State University.

. 2001. An Acoustic Analysis of Vowel Variation in New
World English. Publication of the American Dialect Society.
No. 85.

_____ , and Guy Bailey. 1998. Parallels between vowel subsystems
of African American Vernacular English and Caribbean
Anglophone Creoles. Journal of Pidgins and Creole Languages.
Amsterdam: John Benjamins. 13:2, 267-296.

Warner, W. Lloyd, Marchia Meeker and Kenneth Eells. 1960.

Social Class in America: The Evaluation of Status. Harper
Torch Books.

213

Waters, Mary C. 1998. American identity and culture. In Race, Class,
and Gender. eds. Margaret L. Anderson and Patricia Hill
Collins. Belmont, California: Wadsworth Publishing Company.

Weinreich, Uriel, William Labov, and Marvin Herzog. 1968.
Empirical Foundations for a theory of language change. In
Directions for Historical Linguistics. eds. W.P. Lehmann and
Yakov Malkiel Austin: University of Texas Press, 97-195.

Weerker, Janet. 1995. Age-related changes in cross-language speech
perceptions standing at the crossroads. In Speech Perception and
Linguistic Experience: Issues in cross-language research. ed.
Winifred Strange. Baltimore: York Press.

West, Cornel. 2001. 2nd ed. Race Matters New York:
Vintage Books.

Williamson, Juanita. 1968. A Phonological and Morphological Study
of the Speech of the Negro of Memphis, Tennessee. University
Alabama: University of Alabama Press.

Wilson, William Julius. 2001. Socioeconomic inequality: Race and/or
class. In Race in ZIst. Century America. eds. Curtis Stokes,
Theresa Melendez, and Grace Rhodes Reed. East Lansing:
Michigan State University Press. 435-454.

Wolfram, Walter A. 1969. A Sociolinguistic Description of Detroit
Negro Speech. Washington, DC. Center for Applied
Linguistics.

1973. Sociolinguistic Aspects ofAssimilation: Puerto Rican
English in New York City. Washington, DC: Center for
Applied Linguistics.

. and Natalie Schilling-Estes. 1998. American English. Malden:
Blackwell Publishers.

_____ , Kirk Hazen and Jennifer Ruff Tamburro. 1997. Isolation within

isolation: A solitary century of African American English.
Journal of Sociolinguistics. 1:7-38.

Van Keulen, James E., Gloria Toliver Weddington, and Charles E.
Debose. Speech, Language learning and the African American
Child. Boston: Allyn and Bacon. 137-159.

Zack, Naomi. 2001. Different forms of mixed race: Microdiversity and

214

destabilization. In Race in 21st. Century America. eds. Curtis
Stokes, Theresa Melendez and Grace Rhodes Reed. East Lansing:
Michigan State University Press. 49-59.

1998. Mixed Black and White race and public policy. In
Race. Gender, and Sexuality: The Big Questions. eds. Naomi

Zack, Laurie Shrage, and Crispin Sartwell. Malden: Blackwell.
73-83.

215