m4.
‘ .«rnwvfi‘
4 .19.!

a $44.: . .
éuvaft...‘ r. 45‘

MW 437‘!

ungufiau
u...“ .. ,

1.1.5....
.11! 1. o It!

. . «.3... .v

. an?”

a .

’10:?“
:3. .
x“?

. {7,
3. «1" I‘“
l. .. iii:
nu.) y iitll.

.31Mxlw ‘ .x
J 5!.- ,

. (u. Z
21;? Ar. (.1.
. n. 1‘ D “I
.A) 5.34.. ~\«\.~-i
...- ~ 5"...
. . ix ..

 

 

 

 

130? LIBRARY
Michigan State
University

This is to certify that the
thesis entitled

AFRICAN IMMIGRANTS’ ATTITUDES TOWARD AFRICAN
AMERICAN LANGUAGE/ENGLISH (AAUAAE)

presented by

CHRISTOPHER KURIA GITHIORA

has been accepted towards fulfillment
of the requirements for the

PhD. degree in ENGLISH

 

 

1914:41v04/Xgra‘ 4:41“ Kim \ /1 VA’C‘v/y

Major Professor’s Signature

Must 8, 2008

 

Date

MSU is an affinnative-action, equal-opportunity employer

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K’IProiIAcc8-PresIClRC/DateDue indd

AFRICAN IMMIGRANTS’ ATTITUDES TOWARD AFRICAN AMERICAN
LANGUAGE/ENGLISH (AAL/AAE)

By

CHRISTOPHER KURIA GITHIORA

A DISSERTATION
Submitted to
Michigan State University
In partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
English

2008

ABSTRACT

AFRICAN IMMIGRANTS’ ATTITUDES TOWARD AFRICAN AMERICAN
LANGUAGE/ENGLISH (AAL/AAE)

By

CHISTOPHER KURIA GITHIORA

This study analyzed language attitudes of twenty-four African immigrants in the USA
from West (WA), East (EA) and South Africa (SA) toward African American Language
(AAL), a.k.a. African American English (AAE), African American Vernacular English
(AAVE), Black Language, and US Ebonics. The study adapted the use of twelve
stimulus voices to help elicit responses from the 24 respondents, using the matched-guise
technique; a procedure commonly used in language attitude studies and that was
originally developed by Lambert (1960) and Lambert & Tucker (1972). This quantitative
research methodology has not been widely used in attitude studies of languages of
African ancestry, neither in Africa nor in the African Diaspora.

A female and male African American read three selected passages in African American
Language (AAL) and then code-switched into Standard American English (SAE) or the
Language of Wider Communication (LWC). The raters evaluated the twelve stimulus
voices on twelve traits/characteristics using a seven-point Semantic Differential Scale.

Using the Statistical Package for the Social Sciences (SPSS), analysis involved Ratings
and Rankings of the Mean Scores, Correlation Coefficient and Factor Analysis for each
of the twenty-four evaluators on each of the twelve adjectives on the Semantic

Differential Scale. While no significant differences were found in perceptions of the

language varieties based on the African immigrants’ age, gender and years spent in the
USA, language attitudes based whether they come from WA, EA and SA were significant
among the 24 raters.

The 24 evaluators perceived the 12 stimulus voices as either negative or unfavorable
when the language varieties sounded “Egocentric,” “Patronizing” and “Proud.” By
contrast, the respondents used attributes such as “Charming,” “Dependable,” “Sociable,”
and “Humble” when they had positive or favorable perceptions toward a language
variety. Using Factor Analysis, raters from WA and SA were found to perceive AAL as
more “Egocentric” than SAE. EA raters, however, perceived SAE as more “Egocentric”
than AAL. All 24 raters identified AAL as more “Patronizing,” “Charming” “Proud,”
and “Dependable.” Both EA and SA raters perceived AAL to be more “Sociable” and
“Humble.” The EA raters viewed the SAEM variety as more “Egocentric” and thus less
favorably than AAL but also rated its SAEF counterpart more positively, followed by
AALM, AALF, and SAEM voices. Ultimately while WA raters described the African
American Language female (AALF) and the African American Language Male (AALM)
voices more favorably than either the Standard American English male (SAEM) or
female (SAEF) voices, both SA and EA raters, who have a shared history of large
European settlements in their regions, expressed preference for the four voice varieties in
the same order, namely, SAEF, AALM, AALF and SAEM.

Findings of this language attitude study suggest that perceptions of African immigrants
in the US toward AAL and African American Culture have larger implications for how

Africans View African Americans in particular and US culture in general.

Copyright
CHRISTOPHER KURIA GITHIORA
2008

DEDICATION
I dedicate this work to my loving family and especially my wonderful mother, the late
Mwalimu Alice Wambui Githiora (May 8th 1929 to July 22nd 2004): Uromama Kuraaga!
(May you RIP where it rains forever!) I also dedicate this work to Marjory Njeri, my
beautiful daughter. I could not have succeeded without their great love and prayers. I
further dedicate this work to all those warrior—scholars working tirelessly to preserve the
dignity and continuing relevance of endangered languages and their speakers all over the
world.

ACKNOWLEDGEMENTS

This dissertation marks the pinnacle of several years of persistence in my studies and
an enduring desire to achieve the highest academic honors possible. It also marks the
fulfillment of many days, months and years spent searching, reading, consulting and
attending classes and humble learning at the feet of my teachers. This journey first started
when I joined Mang’u Full Primary School in Central Kenya, many years ago, where the
teachers instilled in my young mind and tender heart the value of asking the right
questions in order to get the right answers. The completion of this work also symbolizes
the culmination of an incessant research curiosity about issues that involve both the
healing and renewal of faith in the course of bridge-building and re-establishing critical
linguistic discourses and related communication channels using African languages in
Africa and the Diaspora. This project hopes to contribute positively to preserving and
advancing the human pride and dignity embodied in these languages whose value has
historically been seriously undermined and compromised.

I would like to acknowledge the excellent assistance of Professor Geneva Smitherman
a.k.a Dr. ‘G.’ my Guidance Committee Chairperson, who has over the years been both a
wonderful teacher and mentor. She has always offered “round the clock” support by
keeping it “Real!” Thanks to her wisdom, selfless guidance, dedication to her students,
and scholarship, along with her generous offers of both time and academic
resourcefulness embodied in inspiring words, both written and spoken. Her mentoring
has been the Soul food that has helped nourish and sustain me, especially when times
were extremely rough for this brother. Because of you Dr. ‘G.’ I feel much better

prepared and more determined than ever to join the global African community in

vi

preserving otherwise silenced voices of the Word, Sound and Power - the ‘Spoken Soul’
of millions of people in Africa and the Diaspora.

Many thanks also to Dr. David Dwyer for his support and tutelage in the value of
meticulously learning the many linguistic and technical aspects of African languages and
on the need to preserve them through modern archiving. Thank you also for selecting me
to work with you on the Webbook of African Languages now available on the World
Wide Web (w) and in helping to coordinate the study of African languages under the
prestigious auspices of SCALI 2002 (Summer Cooperative African Language Institute),
the first time ever the institute was held at Michigan State University. Dr. Dwyer and his
wife Annabelle constantly offered great friendship to me and to many African students in
East Lansing. Thank you also for the many wonderful dinners and fellowships with
numerous peace activists who honored us with their presence and great mind-opening
conversations in your home. I would also like to thank the wonderfirl support by
Professors Jeff Wray and Pero Dagbovie, the other two members of my Guidance
Committee. Together with their families, they have also over the last several years
generously helped guide my academic and social-cultural growth enormously through
both an unconditional love and inspiration in both words and deeds here in East Lansing.

Many thanks also to Jackie Campbell, Graduate Secretary at the English Department
and to Professor Dennis Preston and his wife Carolyn for their friendship and patience
even when I unexpectedly dropped by their home or even stopped them on numerous
occasions to ask various questions in Sociolinguistics. I cannot also forget the many
other wonderful and incredibly supportive people in my life who have been an invaluable

part of my global village experience here in East Lansing.

vii

I acknowledge the love and support from my loving and wonderfully wise and patient
parents, Mr. Raphael Githiora and my late and wonderful Mother, Mwalimu Alice
Wambui, my brothers and sisters, as well as my graduate school colleagues. I would also
like to acknowledge my employers and colleagues at Goodrich’s Shoprite, where I
worked for several years and where I was constantly reminded of the need to always
humble myself even as I learnt and greatly appreciated perhaps more than ever, both the
value of hard work along with learning new skills. I also greatly enjoyed meeting many
customers, now lifelong friends, at Goodrich’s Shoprite, from the local East Lansing and
Lansing communities as well as many Michigan State University students, staff and
professors from around the world. Working at Goodrich’s Shoprite often provided me a
much-needed break and respite as I temporarily escaped, albeit briefly, from the rigors of
academic life.

This dissertation belongs to all of you for after all it “takes an entire village to raise a
child,” to echo an often quoted African proverb. Thanks to the many other silent heroines
and heroes who have been strong pillars that have helped support my life here at MSU
who will forever remain life-long friends and memorable colleagues at Michigan State
University. Let us all, through education and community activism, help to change the
world for the better, in as multiple ways as is possible. Mzirorimia gzikumia.’ (May you
labor where it yields abundantly!) To all, 1 also say a big Shukran! Asantem’ Sana!

Ngaatho Nyingi Mrhino cia Mdanya! (Many Special Thanks!)

viii

 

 

TABLE OF CONTENTS
LIST OF TABLES ................................................................................ x1
LIST OF FIGURES ............................................................................ XIII
CHAPTER 1: INTRODUCTION 1
PURPOSE OF THE STUDY .......................................................................................... 2
LANG UA GE USE AND IMMIGRANTS IN THE USA ............................................... 3
AFRICAN IMMIGRA TION TO THE USA IN THE LA TE 19” AND EARL Y 20”
CENTURIES ............................................................................................................... 9
SIGNIFICANCE OF THE STUDY .............................................................................. 12
ORGANIZATION OF CHAPTERS ............................................................................. 14
CHAPTER 2: REVIEW OF LITERATURE 16
INTRODUCTION ........................................................................................................ l6

THEORETICAL FRAMEWORK FOR LANGUAGE ATTITUDE STUDY OF
AFRICAN IMMGRANTS TOWARD AFRICAN AII/[ERICAN LANGUA GE (AAL). I 7

 

LANGUAGE ATTITUDE RESEARCH IN AFRICA .................................................. I9
SHIFTING A TTITUDES TOWARD EUROPEAN LANGUAGES IN AFRICA: THE
GLOBAL CONNECTION ......................................................................................... 24
CHAPTER 3: METHODOLOGY AND RESEARCH PROCEDURES ................... 34
DATA COLLECTION PROCEDURES ....................................................................... 37
DATA ELICITATION ................................................................................................. 40
THE SAII/fl’LE ........................................................................................................... 42
SAMPLING FRAIl/E AND PROCEDURES .............................................................. 43
DATA ANALYSIS ....................................................................................................... 50
CORRELATION COEFFICIENT ............................................................................. 51
FACTOR ANALYSIS ................................................................................................. 5]
SCOPE AND LIMITATIONS OF THE STUDY ......................................................... 52
RESEARCH QUESTIONS ........................................................................................... 53
RESEARCH HYPOTHESES ....................................................................................... 53
MAJOR ASSUMPTIONS IN THE STUDY ................................................................. 54
CHAPTER 4: ANALYSIS AND PRESENTATION OF DATA 55
MEAN SCORE RATING AND RANKING ANALYSIS ............................................ 55
CORRELATION COEFFICIENT ANALYSIS ........................................................... 75
FACTOR ANALYSIS .................................................................................................. 88
CHAPTER 5: SUMMARY, CONCLUSIONS AND STUDY IMPLICATIONS 112
SYNOPSIS OF THE STUDY ..................................................................................... 112
CONCLUSIONS ........................................................................................................ 124
SUGGESTIONS F OR FURTHER RESEARCH AND IMPLICATIONS OF STUDY
.................................................................................................................................... 128

ix

APPENDICES -- - -- - ...... 132
APPENDIX A: QUESTIONNAIRE USED IN THE ELICITATION OF LANGUAGE

 

ATTITUDES .............................................................................................................. 132
APPENDIX B: INFORMED CONSENT DOCUMENT ........................................................ 146
REFERENCES -- _- 147

LIST OF TABLES
TABLE 1A: PROFILES OF THE 24 AFRICAN RATERS BY LANGUAGE & REGION ........... 46

TABLE 1B: PROFILES OF THE 24 AFRICAN RATERS BY LANGUAGE, GENDER & REGION
........................................................................................................................... 48

TABLE 2: MEAN RATINGS AND RANK OF MEANS OF ATTITUDES TOWARD
AAL AND SAE BY ALL RATERS FROM WEST AFRICA, EAST AFRICA

AND SOUTH AFRICA .................................................................................... 5 7
TABLE 3: MEAN RATINGS AND RANK OF MEANS OF AAL AND SAE BY
EIGHT WEST AFRICAN RATERS ................................................................. 60
TABLE 4: MEAN RATINGS AND RANK OF MEANS OF AAL AND SAE BY
EIGHT EAST AFRICAN RATERS ................................................................. 63
TABLE 5: MEAN RATINGS AND RANK OF MEANS OF AAL AND SAE BY
EIGHT SOUTH AFRICAN RATERS .............................................................. 65
TABLE 6: MEAN RATINGS AND RANK OF MEANS OF AAL AND SAE BY
GENDER BY ALL RATERS ........................................................................... 69
TABLE 7: RANKINGS OF AAL & SAE SPEAKERS BY ALL RATERS ................ 72
TABLE 8: RANKINGS OF AAL SPEAKERS BY ALL RATERS ............................ 72
TABLE 9: RANKINGS OF SAE SPEAKERS BY ALL RATERS ............................ 73

TABLE 10: CORRELATION COEFFICIENTS OF 12 SEMANTIC DIFFERENTIAL SCALES FOR
WEST AFRICAN RATERS .................................................................................... 79

TABLE 11: CORRELATION COEFFICIENTS OF 12 SEMANTIC DIFFERENTIAL SCALES FOR
EAST AFRICAN RATERS ..................................................................................... 82

TABLE 12: CORRELATION COEFFICIENTS OF 12 SEMANTIC DIFFERENTIAL SCALES FOR
SOUTH AFRICAN RATERS ................................................................................... 85

TABLE 13: FACTOR LOADING FOR WEST AFRICAN RATERS ........................................ 90

TABLE 14: INTERPRETATION OF FACTOR LOADINGS FOR THE WEST AFRICAN RATERS
IN TABLE 13 ....................................................................................................... 91

TABLE 15: FACTOR ANALYSIS: WEST AFRICAN RATERS’ ATTITUDES TOWARD AAL &
SAE ................................................................................................................... 93

xi

TABLE 16: FACTOR LOADINGS FOR EAST AFRICAN RATERS ...................................... 97

TABLE 17: INTERPRETATION OF THE FACTOR LOADINGS FOR THE EAST AFRICAN

RATERS IN TABLE 16 .......................................................................................... 98
TABLE 18: EAST AFRICAN RATERS” ATTITUDES TOWARD AAL & SAE .................... 99
TABLE 19: FACTOR LOADING FOR SOUTH AFRICAN RATERS ................................... 103

TABLE 20: INTERPRETATION OF THE FACTOR LOADINGS FOR SOUTH AFRICAN RATERS
IN TABLE 19 ..................................................................................................... 104

TABLE 21: SOUTH AFRICAN RATERS’ ATTITUDES TOWARD AAL & SAE ............... 105

TABLE 22: RATERS’ IDENTIFICATION OF THE TWELVE ANONYMOUS STIMULUS VOICES
......................................................................................................................... 108

xii

LIST OF FIGURES

FIGURE 1: AN ILLUSTRATED EXAMPLE OF THE SEVEN POINT RICHTER SCALE (LE. THE

SEMANTIC DIFFERENTIAL SCALE) USED IN THE STUDY ...................................... 35
FIGURE 2A: RANKINGS OF AAL & SAE BY REGION .................................................. 73
FIGURE 2B: RANKINGS OF AAL & SAE BY VARIETY ................................................ 73
FIGURE 3A: EGOCENTRIC & CHARMING ..................................................................... 94
FIGURE 38: PATRONIZING& DEPENDABLE ................................................................. 94
FIGURE 3C: PROUD & EGOCENTRIC ........................................................................... 94
FIGURE 4A: EGOCENTRIC& CHARMING ................................................................... 100
FIGURE 4B: PATRONIZING & DEPENDABLE .............................................................. 100
FIGURE 4C: PROUD & SOCIABLE .............................................................................. 101
FIGURE 5A: EGOCENTRIC & CHARMING ................................................................... 106
FIGURE SB: PATRONIZING & DEPENDABLE .............................................................. 106
FIGURE 5C: PROUD & SOCIABLE .............................................................................. 107
FIGURE 5D: DEPENDABLE & HUMBLE ...................................................................... 107
FIGURE 6A: IDENTIFICATION OF AAL & SAE .......................................................... 109
FIGURE 6B: IDENTIFICATION OF AAL & SAE .......................................................... 109

xiii

CHAPTER 1: INTRODUCTION
Now, if this passion, this skill, this (to quote Toni Morrison) “sheer intelligence,” this
incredible music, the mighty achievement of having brought a people utterly unknown to,
or despised by “history”-- to have brought this people to their present, troubled, troubling,
and unassailable and unanswerable place--if this absolutely unprecedented journey does
not indicate that black English is a language, I am curious to know what definition of
language is tO be trusted. (If Black English Isn't a Language, Then Tell Me, What IS?
James Baldwin, The New York Times, Op-Ed, July 1979)

African American writer James Baldwin captures the persistent ambivalence that
African Americans continue to experience and the attitudes that continue to question the
legitimacy of Afiican American Language (AAL), a.k.a. African American Vernacular
English (AAVE), African American English (AAE), Black English, Black Vernacular,
Black English Vernacular (BEV) and Black Vernacular English (BVE), and Ebonics.
Several scholars, including Dillard (1972) and Rickford (1999), argue that AAVE shares
so many characteristics with Creole dialects spoken by black people in much of the world
that AAVE itself is a Creole. It has also been suggested that AAVE has grammatical
structures in common with West African languages. As with all linguistic forms, age,
status, topic and setting influence the usage of AAL.

Together with other colonized and enslaved African people, AAL speakers continue to
face the linguistic dilemma and constant challenges in their choice of whether to
communicate in their variety or Standard American English (SAE)/ Language of Wider
Communication (LWC), or both. Africans and Afiican Americans, according to Ngr'rgi
wa Thiong’o in Decolonising the Mind (1986), share the same bio-geographic roots. . .the
same past of humiliation and exploitation under slavery, and colonialism. . .the same

aspirations for the total liberation of all black people, in the world (98). This means that

for both Africans and African Americans European languages have historically

dominated most Of their social, cultural and political discourses. However while AAL
and many African languages, continue to “imitate” Standard American English (SAE),
and as they also continue to adapt themselves to what is considered standard European
language varieties, both on the Continent and in the Diaspora, they still retain many
distinctly African linguistic features. New African varieties that are similar to US
Ebonics in this regard, include, for example, “Sheng,” a Swahili-English slang (Githiora
2002), usually spoken in urban areas of Kenya and F laitaal, an argot (slang) often
“spoken in South African townships, a mixture of local African languages with an
Afiikaans syntax” (Thiba 2000, 19) which is mainly spoken by males but not exclusively.
In this regard, since both language and culture are dynamic processes, they also respond
to their users’ needs. As seen with AAL and with both Sheng and F laitaal, both the

language and culture they embody are continually evolving and regulating themselves.

PURPOSE OF THE STUDY

The purpose of this study is to investigate the nature of African immigrants’ attitudes
toward African American Language (AAL). Since the days of European colonialism in
Africa, the majority of language attitude studies have historically focused on European
languages. This has helped to influence much more favorable attitudes toward these
varieties at the expense of indigenous African languages (Adegjiba 1994; Mbaabu 1996;
Mazrui & Mazrui 1998). This study hopes to help address this research gap. Perceptions
of African immigrants in the US toward African American language and culture, as
manifested in the case of the present study on language attitudes, have larger implications

for how Africans View African Americans in particular and US culture in general.

The significance of the study can also be extended to various global realities that have
resulted in recent huge increases in the number of Africans migrating to the US in search
Of better social, economic and political opportunities and environment. The broader
assumptions in this study are centered on the notion that in spite of the many years of
social, cultural and linguistic divisions due to European slavery and colonialism, (Fanon
1967; Ngt'rgi 1986) both in Africa and in the Diaspora, both Africans and African
Americans share a common African ancestry. Additionally, given such historically
significant complexities, coupled with the presence of varieties of both African and
European languages in West, East and South Africa, this study raises the question of
whether language attitudes are complicated by the hierarchical statuses in Afiica and the
Diaspora that privilege RP--Received Pronunciation or the “Queen’s English”--and
European languages such as French or Portuguese over indigenous African languages and
the varieties of African English. While the study is not designed to respond to all of these

issues, they nonetheless are embodied in this study.

LANGUAGE USE AND IMMIGRANTS IN THE USA
The largest increase in African immigration to the US has been witnessed since
around 1990 with the number of immigrants from the Continent rising by as much as fifty
thousand per year (Roberts, Feb. 21, 2005). This is the largest number of Africans
comprised of some of the most highly educated of all immigrants to the US, Since the
Official outlawing Of slavery in 1807. Against this background the 1990 census
“established that the United States is a nation of some 248 million, of whom 2,015,143

are Native American and 205,501 Hawai’ian (Lippi-Green 1997). Within this population

a “total Of31,844,979 persons — many of these not foreign-born — reported that they
spoke a language other than English in the home” (220). Lippi-Green adds that, it is also
the inescapable truth that the majority of people residing in the US are immigrants, or the
descendants of immigrants, with the greatest portion coming to the US out of their own
free will, while others came in chains.

Having made the transition and established itself, the US has a strong urge to be
protective of what is here, Americans talk at great length about closing the door behind
them. Lippi-Green (1997) points out that European Americans have acted on this
impulse as the following shows: 1) In the 18405 during a depression, mobs hostile to Irish
Catholic immigrants burned down a convent in Boston; 2) Congress passed the Chinese
Exclusion Act in 1882, one of the nation’s first immigration laws, to exclude all people of
Chinese origin; 3) In 1942, 120,000 Americans of Japanese descent had their homes and
other property confiscated, and were interned in camps until the end of the Second World
War. At the same time, many Jews fleeing Nazi Germany during the War were excluded
under regulations enacted in the 19205. The 1990 US census figures, however,

do not specify a single language from the continent of Africa beyond the Arabic
languages of the north. It must be assumed that as immigration from the mid- and
southern African nations is limited, speakers of languages such as Swahili and

Zulu are subsumed under the category “Other” and unspecified languages.”
(Lippi-Green 1997, 220)

What is missing in such census data, then, is the great variety of languages spoken in
other parts of the world. Also missing, according to Lippi-Green, is the fact that the
majority of non-native English speakers claim a very good command of their second
language. What this raises is that millions of US residents are not native speakers of

English and communicate in languages other than English in their homes and personal

lives. These L2 English-language (i.e. speakers of English as second language) speakers
comprise a significant number in the “Other language” group in the US census.
Additionally there are preconceived notions about non-native speakers of English, which
have repercussions even in the way we count their numbers and talk about them. “The
US Census Bureau,” argues Lippi-Green “distinguishes between Spanish, Asian, and
other languages. It is from this departure point that one needs to take a careful look at the
way foreign-language groups and the language stereotypes associated with those groups
are used to classify - and often dismiss — individual needs and rights” (1997, 221).

Amidst the foregoing issues, attitudes of African immigrants in the USA, another
group of L2 speakers, toward African American Language (AAL) can be contextualized
within such global forces as enslavement, colonialism, the rise of modern nation-states,
the easier availability of goOds and services worldwide, and the global dispersion of
Black Popular Culture--all Of which have had complex effects on global language
attitudes toward AAL. These global forces echo Bakhtin’s suggestion (1986) that
language gains meaning within its social and dialogic or intertextual contexts. That is,
language and modern identities continue to be re-made, re-created, and re-negotiated
every time we speak or write. This also means we make use of language in different
situations and re-create ourselves differently for both our audiences and ourselves.
Additionally, these global forces coupled with current global migrations, necessitated by
various global dynamics, contribute to numerous language contact situations and changes
that affect culture, social class, gender, age, ethnicity, race, and identity. These forces
also continue to both impact and complicate — and also to redefine - the effects of

globalization on language attitudes. In this project, we seek to explore and define the

attitudes of African immigrants toward AAL, particularly as impacted by the various
language contact circumstances that have now been enhanced through globalization and
related global migrations of people, goods and services as well as associated socio-
cultural, political and economic discourses.

Language attitudes of African immigrants toward language varieties both in Africa
and the Diaspora reflect the myriad forms of marginalization, particularly during such
language contact (or lack of) situations both in Africa and in the African Diaspora, which
both Afiicans and African Americans continue to experience. According to Roberts
(New York Times, February 21, 2005) Afiicans have been migrating to the US at the rate
of over fifty thousand per year since 1990. The group is among the most highly educated
immigrant group to the US in recent years. These large numbers of African immigrants
represent some of the grim realities of today’s multi-polar globalization with its many
attendant effects. As these Africans immigrate, they bring with them various indigenous
African and European languages as important linguistic resources. For these immigrants
and those they encounter, language remains the tool of communication that best mirrors
the various global social encounters that continue to both enhance and strengthen contacts
between them and various US communities.

Language is a critical resource and mode of communication, as well as a form of
cultural expression and a carrier of memory and history. For African immigrants and
Afiican Americans as well as others, it also defines who they are by equipping them with
an identity and a sense of belonging in their new circumstances. According to Berger and
Luckmann (1966), language is the repository of knoWledge and cultural meanings, which

we use to socially construct the respective world realities that surround us, making it, “the

most important item in socialization” (59). The two sociologists assert that through
language “the relationship between man, the producer, and the social world, his product,
is and remains a dialectical one” (61).

While exploration of the nature of African immigrants’ attitudes toward AAL falls
under the area of language attitude research in sociolinguistics, the cross-cultural
perspective in this research also embraces multilingualism, critical in the contemporary
world of globalization, which requires that speakers have a much more profound
understanding, appreciation and proficiency in a variety of local, national, and
metropolitan world languages and cultures. There are serious drawbacks to privileging
Languages of Wider Communication (LWCs), which are most often metropolitan
European languages (i.e. English, French, Portuguese, etc.) over indigenous languages.

This study taps into scholarship that advances key principles embodied in
sociolinguistic theory and research in areas such as bi/multilingualism, language and
ideology, language and globalization, and language policy and planning as well as
language change and variation. From a political perspective this study speaks to F anon’s
challenge that “Each generation out of relative obscurity must discover their mission,
fulfill it or betray it” (1963, 207). It explores the View, also articulated by F anon in Black
Skin, White Masks ( 1967), that “Every colonized people. . .in whose soul an inferiority
complex has been created by the death and burial of its local cultural originality — finds
itself face to face with the language of the civilizing nation; that is, with the culture of the
mother country” (18). In this broad manner, the study hopes to contribute to language
attitude studies, both those focused on AAL and SAE (Standard American English) and

those with a more general focus. This is likely to help us understand how globalization

has both affected and contributed to the complex nature of language attitudes among
African immigrants in the US toward both AAL and SAE. Additionally, using such a
broad perspective, this study is likely to add to our knowledge of how attitudes of African
immigrants toward both AAL and SAE help them View African Americans (and US
culture in general), a community they also share a common African ancestry and
domination with, through slavery and colonialism by European languages and cultures.
AAL users have for example had a linguistic experience in the US, described by
Smitherman (1986) as a “linguistic pull-push.” Consequently, AAL speakers share a “co-
authorship” (Morgan 1993) of the American experience, accounting in part for the
ambivalence of being both Black and American.

Through a history of struggle amidst hostility by education officials among others in
the US (see the “Black English Case” aka. the “King Trial” and Oakland School District
“Ebonics” controversy in Smitherman 1986, 2000) to survive within an “English-only”
environment, how is AAL then viewed by African immigrants in the US, who hail from
Afiican nation-states, with “dense multilingualism” (Adegbija 1994a)? These African
immigrants come from nations where language attitudes have historically been
complicated, by the need to speak the “Queen’s English” in former British colonies
(Mbaabu 1996) and where, according to Mfitonya (1997), European languages have
historically been strongly recommended for use in both government and private
transactions.

With the global dissemination of AAL in US popular cultural products such as music
and films, the use of AAL seems to continue to be accepted both nationally in the US and

globally as part of the language and culture of African Americans. At the same time, the

variety also faces resistance in US mainstream education, business, and law enforcement
institutions and among some members of the African American middle and upper classes.
The availability of AAL in the global dissemination of US mass media products also
implies that most likely many Afiican immigrants to the US have previously been
exposed to AAL as embodied in US mass media products and music back home in
Africa. Since these African immigrants also most likely identify and associate certain
characteristics of AAL with its speakers, it is possible that, despite many years of
linguistic discrimination, AAL still enjoys “covert prestige” (Trudgill 1972, 179-95)
among both users and nonusers as it facilitates the global dissemination of US popular
culture.

BACKGROUND AND OVERVIEW OF THE STUDY

AFRICAN IMMIGRATION TO THE USA IN THE LATE 19TH AND EARLY 20TH
CENTURIES

Dr. Aggrey from Ghana (then known as the Gold Coast) studied and lived in the US
from 1898 until 1927. He both studied and then taught at Livingstone College, a
religious and historically Black College and University (HBCU) in Salisbury, South
Carolina where he arrived in 1898 and where he also married an African American
woman. Dr. Aggrey taught and lectured widely in North America, UK and in Africa
where he also presided over the establishment of Achimota College in Ghana and where
he was Kwame Nkrumah’s teacher. He died in 1927 in New York City whilst
completing his doctorate at Columbia University.

The other prominent African immigrant to the US around this time was Kwame
Nkrumah, Dr. Aggrey’s student in Ghana who would become Ghana’s first President.

Ghana was the first sub-Saharan African nation to gain independence. Nkrumah attended

Lincoln University in Pennsylvania, another HBCU, from 1935-45. Influenced by both
the Pan-Africanism of Marcus Garvey and that of Dr. W.E.B. Dubois, the prominent
African American scholar and activist, Nkrumah would later join hands with Dubois
among others, in the fight for African independence from European colonial rule. The
bond between Nkrumah, an African who once attended school in the US and W.E.B.
Dubois, an African American was further strengthened and saw the latter taking up
Ghanaian citizenship in 1958 until he died in his newly adopted country in 1963.

John Langalibalele Dube of South Africa is another African student whose journeys to
the US, 1887-1892 and 1896-1899, helped “establish the intellectual rationale for his
educational and political thought for the rest of his adult life” (Marable 1976, abstract) in
South Africa. Marable, an African American historian, suggests that Dube’s progressive
ideology based on his American education coupled with sponsorship by the American
Congregationalist missionaries, under the auspices of the American Board of
Commissions for Foreign Missions, helped turn him into a Black youth who believed in
the Calvinist work ethic and industrial training for Africans. This Marable adds, was
seen in Dube’s work among the Kholwa people of South Africa from a pre-industrial
state into a bourgeois society. Dube’s work in education was modeled along that of
another prominent African American educator, Booker T. Washington, the first teacher of
Tuskegee Institute in Alabama, US, another HBCU, when Dube established Ohlange
Institute in Victoria County, north of Durban. The college was the first Black operated
institution to emulate Tuskegee Institute.

Dube’s educational and religious efforts, like those of Dr. Aggrey before him

benefited from assistance by philanthropist Anson Phelps Stokes. Emaroy June Smith

10

wife of a Chicago millionaire also supported him. Sadly, according to Marable (1976),
Dube assumed many of his American benefactors’ political and economic thoughts which
coupled with the Americans’ unconscious racism and support for segregation, had a
critical impact upon his important decisions within colonial society. Dube was a creation
of the British colonial system and American thought, whose politics and economic
activities were rooted within western tradition and practices. Although Dube, to use
Chinua Achebe’s (1966) words, was “a man of the people,” his failure and that of
Kholwa society, to appreciate the corrupt nature of segregation and to oppose white
racism at all levels helped to bring about the system of South African race relations called
apartheid.

With the advent of independence in Africa coupled with the increasing “cold war’s”
social and economic competition for global dominance between the Soviet Union and the
US from the 19508 through to the mid—19805, large numbers of Africans were airlifted to
study in the US including Barrack Obama Sr. and father to US presidential candidate
Senator Barrack Obama Jr. of Illionois. They were expected to return and help in
building their new nations along the US model of democracy. In Kenya, for example,
trade unionist and later Kenyan government minister Thomas Joseph Mboya together
with Dr. Julius Gikonyo Kiano, the first Kenyan Ph.D. holder, and government minister,
educated at UCLA, would in the years between 1959 and 1963 work ceaselessly to help
airlift 230 East Afiican students to the US with assistance from then US Senator John F.
Kennedy. With the Cold War between the US and the Soviet Union accelerating,
Mboya’s political rival in Kenya on the other hand, Oginga Odinga, the country’s first

Vice President, a socialist and author of Not Yet Uhuru (1967) presided over scholarships

11

for Kenyan students to the former Eastern bloc countries. This scenario was replicated
throughout Africa and the developing world. The early 19605 continued to expose
African immigrants to African Americans with many students choosing to study and live
in the US along with many visits by African Americans to Afiica in support of African

independence.

SIGNIFICANCE OF THE STUDY

My quarrel with English language has been that the language reflected none of my

experience. But now I began to see the matter in quite another way. . .perhaps the

language was not my own because I had never attempted to use it, had only learned to
imitate it. If this was so, then it might be made to bear the burden of my experience if
I could find the stamina to challenge it, and me to such a test... [An] immense
experience had forged this language, it had been (and remains) one of the tools Of the
people’s survival, and it revealed expectations which no white American could easily
entertain. (African American writer James Baldwin (1924-1987) writing in the

London Observer in 1985, about his experience with the English language, as

reported in Transition, NO. 58 (1992): 90-113)

Baldwin’s statement encapsulates the need to investigate the nature of underlying
attitudes of African immigrants to the USA toward AAL using a social and psychological
approach (e.g., Giles, 1971), among others. The significance and purpose of this study is
to both enhance and utilize empirical methods in linguistic research among people of
African descent, especially in the African Diaspora. It aims to accomplish this by testing
methods and theories used in research related to issues of multilingualism in both
monolingual and multilingual situations.

Because of what can be often referred to as an uneasy relationship and numerous gaps
in knowledge about each other between continental Africans and those in the Diaspora,

the present study is important and relevant in helping re-open the well-worn ties that bind

the two groups of Africans. This study also reflects the numerous challenges posed by a

12

largely US and Western European driven contemporary globalization with English as the
dominant language, which has had complex and significant effects on many global
populations, particularly those in poor African American neighborhoods and African
countries. These diverse and complicated global dynamics also continue to influence and
redefine the relationships between Africans and African Americans. Faced with recent
increases in the number of African immigrants in the US, this study should help pioneer
the exploration and testing of linguistic methods and theories on language attitudes
among people of African ancestry in an increasingly multilingual world. For these
Afiican immigrants and those they encounter, language remains the tool of
communication that best mirrors the various global and social encounters including the
“shifts in linguistic identities in a global world” (Djite 2006).

This study has exploited the benefits of a “matched guise” using anonymous voice
stimuli, an approach which is measurable and quantifiable. In this manner the study has
tried to maximize the salient aspects of this kind of empirical linguistic research.

While the main focus in this study is on attitudes of African immigrants toward AAL
and SAE, it is important to briefly situate this project within a historical-cultural context.
This is because African immigrants’ responses to AAL form part of a broader historical
continuum that continues to help re/define the relationship between continental Africans
and African Americans within the United States. In other words, how Africans view
AAL is a larger part of how Africans View African American culture as a whole. Further,
AAL symbolizes both their African ancestry and the painful history of enslavement
within which they’ve maintained a unique system of language and thought to help

express their struggle, life and culture in the US. In Africa, on the other hand, the 1884

13

Berlin Conference divided Africa among European powers and marks what both Ali
Mazrui & Alamin Mazrui (1998) suggest was a strange division of Africa into a linguistic
anomaly and zones on the basis of imperial European languages--Anglophone Lusophone
and Francophone Africa. The Berlin Conference effectively marked the advent of
Africa’s linguistic dependence on Europe. Mazrui & Mazrui (1998) also argue that this
colonial legacy and dependency by Africans appear to have, over the course of time,
spilled into other areas of life in Africa, leading to social, economic and political
upheavals that have created conditions for people to migrate to the US and to the West in

general in search of a better life.

ORGANIZATION OF CHAPTERS
Chapter One presents the introduction to this study and includes brief details on socio-
historical background to the study. It discusses early interactions between leading
African students and the Diaspora. Some of these Africans would later return to become
post-independence African leaders. The chapter also discusses the significance of the
study. Chapter Two provides a review of literature on research on language attitudes.
Chapter Three outlines the methodological procedures, sample design and data collection
and analysis conducted in this study. It discusses the benefits of the matched-guise
technique using twelve stimulus voices specially prepared for evaluation and rating by
the twenty-four African immigrant raters, based on a seven-point Semantic Differential
Scale. Chapter Four presents a discussion of the analytical techniques in the study and
includes analysis and results of the investigation. Chapter five presents the summary and

implications for further research. The chapter makes several recommendations for the

14

kind of studies needed to advance our knowledge of attitudes toward both African

languages and languages of the Diaspora.

15

CHAPTER 2: REVIEW OF LITERATURE
Where shall I turn, divided to the vein?
I who have cursed
The drunken officer of the British rule, how choose
Between this Africa and the European tongue I love?
‘A Far Cry from Africa’ in Collected Poems, 1948-1984 (1986)
By Derek Walcott
INTRODUCTION

Derek Walcott’s words in the above poem, captures the ambivalence experienced by
both linguistically dispossessed Africans, especially in the Diaspora, and those
linguistically disempowered speakers of various African languages. As mentioned, both
Afiican immigrants and African American Language (AAL) users largely due to forces
of both slavery and colonialism have historically been forced to View their native
languages negatively and to also abandon them in favor of European languages.

The study investigated attitudes of African immigrants in the US toward AAL and is
contextualized within language attitude research that deals with the following themes:

1) Theoretical framework for language attitudes of Afiican immigrants toward AAL

2) Research on attitudes toward varieties of African and European languages in

Africa

3) Research in language attitUdes and the global perspective

4) Shifting nature of language attitudes toward European languages in Africa

The aim of this chapter is to provide a synopsis of language attitude issues and
research as they pertain to attitudes of African immigrants in the US toward a variety of

languages in Africa. The summary of language attitude research also makes global

connections by reviewing similar language attitude studies in Australia (Callan & Gallois

l6

1987) and North America (Agheyisi & F ishman 1970; Baker 1992; Fisher 1974; Lambert
& Wallace 1960; Lambert, Wallace, & Tucker 1972; Schmied 1985; Shuy & F asold
1973; Taylor 1973; Williams 1972, 1973 & 1974). The history of language attitude
research in sub-Saharan Africa is fairly recent (Adegbija 1994a, 2000; Akere 1982;
Angogo & Hancock 1982; Berry & Greenberg 1966; Bokamba 1982; Dirven 1990; Dyers
1999; GOrIach 1991; Kachru 1986; Putz 1995; Rubagumya 1986; Saab 1986; Schmied
1985, 1991; Sure 1991).

THEORETICAL FRAMEWORK FOR LANGUAGE ATTITUDE STUDY OF
AFRICAN IMMIGRANTS TOWARD AFRICAN AMERICAN LANGUAGE
(AAL)

Much of language attitude research includes the three main components of attitude:
the cognitive, the affective and readiness for action components (Baker 1992, 12;
Adegbija 1994, 49). The cognitive component according to the Baker relates to thought
and beliefs. The affective component concerns feelings toward the attitude object (i.e.
toward the language). The readiness for action (conative) component of attitude is
described as “behavioural intention of plan of action under defined contexts and
circumstances” (Baker 1992, I3; Adegbija 1994, 49). Baker emphasizes that the
relationship between attitudes and action is neither straightforward nor simple:
Attempting language shift by language planning, language policymaking and the
provision of human and material resources can all come to nothing if attitudes are
not favorable to change. Language engineering can flourish or fail according to
the attitudes of the community. Having a favourable attitude to the subject of
language attitudes becomes important in bilingual and policy and practice. (Baker,
1992, 21; Adegbija 1994, 49)

A general attitude of ambivalence is sometimes evident with regard to perceptions

toward European languages in contrast to indigenous African languages. Such

l7

ambivalence is reflective of “the conflict between loyalty to one language and the utility
of another: the choice between learning a language because it is useful (instrumentalism)
and learning another because it marks the individual’s cultural, ethnic or national identity
(integration)” (Sure 1991, 251). This seems to reveal another dimension of the
instrument/integrative dichotomy of motivation in the learning of languages in the
Afiican context (Adegbija 1994, 64).

While ethnolinguistic minorities are also sensitive to language issues and are often
closely attached to their languages and cultures, Baker (1993, 5) adds, “the attitudes of
individuals toward a particular language may affect language maintenance, language
restoration, language shift or language death in society.” Africa’s colonial past, along
with a dense multilingual and multicultural character, means there is a growing
acceptance of European languages in many parts of the continent because of the
perception that they serve unifying roles in largely multi-ethnic societies in Africa.
However the importance of native varieties of European languages and their increasing
acceptance means that many Africans perceive these languages as no longer the
properties of Europeans as such, but as international commodities with universal
ownership, which the entire world can therefore lay claim to.

Language attitude research seems to have had its start in the early 1960’s from the
field of social psychology (Lambert, et al, 1960; Lambert 1967). Linguists started getting
involved in language attitudes studies in the 1970’s (Agheyisi & F ishman 1970; Cooper
1974, 1975; Cooper & Fishman 1974; Shuy and Fasold 1973; Williams 1972, 1973 &
1974) especially as it pertained to teachers’ attitudes toward students’ speaking styles.

The majority of these studies, however, focused on Western countries and Western

18

languages. African countries and African immigrants in the Diaspora as well as African
Americans received scant attention in these earlier studies despite the fact that
bilingualism, multilingualism and multiculturalism are so prevalent both in Africa and the

Diaspora.

LANGUAGE ATTITUDE RESEARCH IN AFRICA

Starting in the 1980’s, language attitude studies began to be conducted in West Africa
(Akere 1982; Saab 1986) with the 1990’s seeing an increase in the number of language
attitude studies in Southern Africa partly due to the abolition of Apartheid in South
Africa and the beginning of Black majority rule in South Africa (1994) and previously,
independence in Namibia from South Africa (1990). The creation of these two free and
independent nations resulted in language policy legislation that for example saw South
Afiica in 1996 enact official language policy to make eleven languages, including
Afiikaans and English, the official languages of the country.

Much of the research that has been published in Africa on language attitudes (e.g.,
Adegbija 1994a, 2000; Akere 1982; Angogo & Hancock 1982; Berry & Greenberg 1966;
Bokamba 1982; Dirven 1990; Dyers 1999; GOrlach 1991; Kachru 1986; Putz 1995;
Rubagumya 1986; Saab 1986; Schmied 1985, 1991; Sure 1991) focuses on attitudes
toward former colonial languages, especially English, and not on indigenous African and
Afiicanized languages or African languages spoken in the Diaspora. The present study
seeks to address this research gap.

Adegbija (1994, 50) suggests the importance of language attitude studies in sub-
Saharan Africa does not appear to have been recognized and in spite of the densely

multilingual nature of the region and frequency of language-related problems, studies on

19

language attitudes relevant to it are very few at the moment. He also examines what he

suggests should be the aims of language attitude research in Africa. These aims are:

Generally aim at pinpointing the patterns and bottom-line determinants of
attitudes toward particular languages;

More finely-tuned research that pinpoints attitudes toward European languages
and African indigenous languages;

Be relevant for changing attitudes and for ensuring effective language policy
making, language planning and action;

Try out fairly Simple and flexible techniques which can easily be applied in
developing countries despite their problems and research limitations;

Show the results of two different language attitude concepts and to explain their
relationship in a complex sociolinguistic situation;

Demonstrate how a subtle interpretation of language attitudes can help to throw
light on problems of language policy, language use and language learning (taken

from Schmied, 19852237).

He also includes the major results of African language attitude research in sub-

Saharan Afiica along with discussions of the bottom-line language attitude determinants

in African multi-lingual and multicultural contexts and important areas for future African

language attitude research. He singles out two key language attitude studies, one by Sure

(1991) in Kenya and another by Schmied (1985) in Tanzania for critical review. He also

suggests:

Deficiency in methodology appears to be the principal weakness of most language
attitude studies hitherto carried out in Sun-Saharan Africa. Some of such studies
(e.g. Saab, 1986) appear to be largely impressionistic in approach and sampling
techniques and instruments, when indicated, are often weak. Sure (1991), though

20

a very interesting and commendable study because it is one of the pioneering
studies by Africans that directly focuses on language attitudes, partly illustrates
the use of Simple instruments and statistical techniques. Instruments included
twelve attitude statements, six favourable and six unfavourable for both Swahili
and English in the primary school test. He indicates that he has used a total of 405
primary pupils and 358 secondary pupils (763 on the whole) drawn by stratified
random sampling of pupils in seven secondary schools and seven primary schools
spread over four town centres and four rural districts For the secondary school
subjects, the same statements plus an additional eighteen were used. Respondents
had to simply tick whether they agreed or disagreed with each

statement. . .Attitudes, however, like many aspects of life, are far more complex
than merely agreeing or disagreeing with particular statements. (Adegbija 1994,
53-54)

Schmied (1985) carried out two attitude tests in Tanzania in which, according to
Adegbija, “he attempted greater scientific rigour in the presentation of his results, thus
making his study more scientifically detailed” (54). In his first test, Schmied investigated
stereotyped attitudes toward English and Kiswahili, which, he referred to as

‘H (igh)-variety rivals,’ as well as French, taught in some Tanzanian secondary
schools because it is perceived, as an important pan-African language, and
Arabic, which has historically been influential along the east African coast and on
the development of Swahili. Unfortunately, as he points out, the various
vemaculars were not taken into consideration. (We must again reiterate at the risk
of overemphasis, the need for attitude studies on African vernaculars.) Adjectives
such as beautiful, colourful, rich, pure, precise, logical, rhythmical, pleasing to the
ear, refined, superior, sophisticated and intimate were employed to assign inherent
attributes to the four-world language...

The second test, which sought to investigate ‘language beliefs concerning the
importance and use of English in Tanzania’, used statements similar to Sure’s
about the importance and use of English and Kiswahili in Tanzania (Schmied
1985, 238). Subjects for the study were ‘educated Tanzanians’ and ‘primary
school teachers’. The statements used to relate to language-inherent arguments,
national arguments, personal instrumental arguments, educational arguments, and
cognitive arguments. (54)

Based on the techniques used by Schmied (1985) and Sure (1991), Adegbija (1994)
concludes that measuring language attitudes is, however, difficult, like many aspects of
life because they are far more complex than merely agreeing or disagreeing with

particular statements. He also argues that, one

21

is very skeptical about the possibilities of such statements used by both Sure and
Schmied to actually point at deep-seated language attitudes, since respondents
Often answer in accordance with what they think the researcher wants to hear.
Moreover, responses are influenced by such wide-ranging variables as level of
education, sex differences, ethnicity, etc (55).

He notes that his study (Adegbija 1992f) of a sample of six hundred subjects randomly
chosen from Kwara State, Nigeria was mindful of such variables in its research design
and also relied on such qualitative techniques as interviews and observational data. He,
however, points out that his study of attitudes is quantitatively weak. Adegjiba further
critiques his study (1992f) by asserting that like Sure’s (1985), he uses only simple
percentages in analyzing his results. There is no search for multidimensionality, and
inferential statistics are absent even though they are demanded by the study. Baker’s
(1992) study however argues that, the research by Adegbija is particularly revealing in
that it attempts to control variables such as age, gender, and type of school, language
ability and provides some insights into how‘such variables can affect the measurement of
language attitudes.

While emphasizing that the methodologies used in most African studies on language
attitudes thus far need improvement, Adegbija also points out that, in their research, Sure
(1991), Schmied ( 1985) and Adegbija (19921) are hardly representative of Kenya,
Tanzania and Nigeria, respectively, nor are the research instruments they used and the
statistical techniques employed powerful enough to truly reveal fine nuances and
complexities of language attitudes. Such criticisms, he adds, are equally applicable to

much of language attitude research in Afiica (e. g. Saab 1986; and studies reported in

Webb 1992).

22

Adegbija finally suggests that future research on language attitudes in sub-Saharan
Africa would need to be less impressionistic, more sophisticated in research design and
more thorough in sampling and analytical techniques along with the use of more detailed
statistical procedures and methods. He also recommends that future language attitude
research begin by grounding itself in attitude theory and research while also placing high
premium on the need for utilizing attitude scales that meet the following three crucial
criteria: 1) Internal consistency in response and in the use of many items to ensure
internal reliability, 2) Validity and the concern that the attitude scale should actually
measure attitude through a thorough checking of items (content validity); the relation of
the scale to a variety of present variables (criterion-related validity); and future variables
(predictive validity); and to variables within an established theoretical formulation
(construct validity), 3) Dimensionality: testing of whether one or more entities has to be
measured and unidimensionality or multidimensionality can be measured if a large initial
pool of attitude items is subjected to an exploratory confirmatory factor analysis
(following recommendations by Baker 1992, 25).

The complicated set of attitudes also reflect various forms of marginalizing
historically experienced by both Africans and African Americans due to both colonialism
and enslavement and a lopsided globalization that often privileges European languages
and their dominance over both AAL users’ and Afiican linguistic worldviews. In Africa
itself, research on language attitudes still contains information gaps that needs filling with
new information to help define the nature of attitudes toward both AAL and African
languages. While Myers (1996) acknowledges the existence of some comprehensive

studies on the language situation in Africa, it also however notes a glaring fifteen-year

23

gap in research studies that have traditionally focused on language attitudes in Africa.
There are however relatively recent studies by Smitherman (1998) and Mr'rtonya’s (1997,
2000), along with other studies by Thiba (2000) and Makalela (2005) on attitudes toward
African languages among various African groups. These studies cover West, East and
South Africa by Mfrtonya and among South Africans (Thiba and Makalela) and toward
African languages among African Americans (Smitherman 1998). The four studies are
significant and relevant for this study while the quantitative components and designs in
three of them (Mr'rtonya 1997; Smitherman 1998; Makalela 2005) offer invaluable leads.
These studies also help fill the gap noted earlier by Adegbija’s (1994) on the dearth of
available and published studies that focus on language attitudes toward varieties of
African languages in Africa and in our view that of varieties of African languages in the
Diaspora:
Also worthy Of research interest. . .are attitudes toward the nativisation of
implanted varieties [European languages]. . .What are the attitudes of those who
use the nativised varieties of implanted languages toward the varieties they use?
much sociolinguistic work specifically relating to attitudes of nativisation still...
has to be done. This is particularly so when we take cognizance of the fact that

these new varieties play crucial roles in the educational systems of the countries
concerned. (Adegbija 1994a, 6)

SHIF TING ATTITUDES TOWARD EUROPEAN LANGUAGES IN AFRICA:
THE GLOBAL CONNECTION

While the prestige and status associated with English or French still remains in
Africa today, there is recognizable shift in attitudes toward these European languages as
Africans increasingly turn to their native languages to broadcast news and information.
This is enabled by the rapid increase in availability of broadcasting by F M radio stations

and continental-wide television media such as on M-NET from South Africa. Africans

24

are also actively recording various genres of music and producing films, while also
publishing, advertising and instructing their children in their mother tongue languages.

Emerging language studies are also increasingly recognizing the many varieties of
these European languages in Africa (e.g. “Sheng” in Kenya, “F laitaal” in South Afiica,
varieties of African Englishes (Mirtonya 1997; Bokamba 1981) such as West, East and
South African Englishes (WAfrE, EAfrE and SAfrE). These varieties are being
recognized as the resultant linguistic manifestations as non-native European languages go
through the process of Moag’s (1982) model and perspective of the lifecycle of non-
native languages varieties first studied in the South Pacific. This process involves five
phases in the life cycle of English in Afi'ica, namely: contact, institutionalization,
expansion, recognition or suppression, and finally, adoption or acceptance by native
populations.

In studying English in Afiica using Moag’s model, Schmied (1991) suggests that the
language was in its fourth or fifih stages toward the end of the 20‘h Century. This, he
suggests, are precarious phases for English because the language is facing growing
confidence by native Africans who are increasingly turning to their indigenous languages
and other socio-cultural expressions for proudly depicting their African identity in
contrast to being defined using previously preferred forms of European linguistic and
socio-culture norms. This revival of African Centered socio-cultural discourse is perhaps
akin to that witnessed in the late 19505 and early 19605 when African countries were first
winning independence from their European colonizers. Consequently, Schmied, among

others, is of the view that English in Afiica is faced with losing out to the gradual

25

institutionalizing of indigenous African languages, a process that also implies the
embodiment of new language attitudes toward English.

Evaluative judgments about languages and their speakers are the subject of studies by
Lambert & Wallace (1960) and Lambert, Wallace, and Tucker (1972). According to
Lambert, et.al:

A useful technique has been developed at McGill University to measure, in an
indirect fashion, the Views that members of one social group have of
representatives of some other contrasting group. Described briefly, a sample Of
‘judges’ is asked to listen to a series of tape recordings of different speakers
reading a standard passage, and to evaluate relevant personality characteristics of
each speaker, using only voice characteristics and speech styles as cues. The
technique appears to expose the listeners’ more private feelings and stereotyped
attitudes toward a contrasting group or groups whose language, accent, or dialect
is distinctive, and it appears to be reliable in that the same profile of reactions
emerges on repeated sampling from a particular social group. The procedure has
been used to compare the reactions of judges listening to the two guises of
bilingual speakers presenting (a) contrasting languages, (b) contrasting dialects, or
(c) contrasting accents. (1972, 175-176)

In their work, Delpit and Dowdy suggest that language is deeply intertwined with
identity: “Since language is one of the most intimate expressions of identity, indeed, ‘the
Skin that we speak,’ then to reject a person’s language can only feel as if we are rejecting

him” (2002, 47). Such rejection may be due to the person’s race, ethnicity, socio-

economic status, educational level, age, or gender.

Taylor’s (1973) survey of 422 teachers of various races in the US demonstrated that
“while 40 percent expressed negative opinions about the structure and usefulness of
AAVE and other vernacular varieties, 40 percent expressed positive opinions” (183).
This research assessed teachers’ attitudes on language differences between teachers and
students in schools with substantial Black and other minority group children. It involved

developing and administering a Language Attitude Scale (LAS), a Likert-type scaling

26

instrument specially designed to solicit data on what teachers think about non-standard
and Black English, and how (or if) this dialect should be used in the classroom. The
study included 117 test items, which, were selected as a function of their ability to help
differentiate between teachers with positive Black English attitudes from those with
negative Black English attitudes. Taylor notes previous work lacked any controlled
study, which discusses in-depth teachers’ attitudes as a function of such variables as race,
sex, geography, teaching experience, grade taught, etc. The test was administered to a
large cross section of teachers to obtain such data. The 422 respondents included
teachers from one rural and one large urban school in each of nine Federal Census
districts with at least 20 teachers (10 males and 10 females) selected in each of the
settings. As Rickford (1999) notes in his commentary on Taylor’s (1973) study, these
teachers’ attitudes “could not be characterized simply as positive or negative; they varied
depending on the aspect of dialect use under discussion, length of teaching experience

(those who had been teaching for 3-5 years were most positive), and other factors” (284).

Research by Williamson-Ige (1984) is a cultural critique on the controversy
surrounding black language, which, she argues remains unresolved (Brasch 1981;
Harrison and Trabasso 1976; R.L. Williams 1975). She suggests that careful analysis of
the prevailing theories on black communication reveals two developments: 1) Attitudes
and practices toward black language in the United States have been politically based, and
2) African Americans have the right to self-definition in determining the meaning and
implications of black language. These two developments are symmetrical — the first is
rooted in white culture and the second finds its source in the black community. This

study, using a cultural analytical approach to language attitudes, helps us to situate the

27

problem of attitudes toward AAL by Afiican immigrants in terms of the dominance
exercised by SAE and its users within the politics, economics and historical—cultural
realities that continue to both dominate and marginalize speakers of a what is considered
non-standard varieties such as AAL. Despite AAL being globally available in US mass
media products, we use this article to justify this study’s design and instrumentation in
that our sample includes African immi grants with attitudes largely informed outside of
US political, economic and social-cultural constraints. Despite sharing a common
heritage with AAL users, these immigrants’ View of AAL is nonetheless that of a non-

USA Black population.

Callan, V. & Gallois, C. (1987), two Australian researchers review experimental and
survey research methods often used in language attitude studies. They focus on Anglo-
Australians’ and immigrants’ attitudes toward language and accent. They argue that
research on the language attitudes of members of dominant and minority speech
communities has special importance in countries such as Australia and in situations
where governments are in the process of developing a national language policy. Research
in Australia suggests that as late as 1987 Anglo-Australians remained strongly
monolingual and Anglophile in their attitudes; they supported educational programs on
other languages mainly for their children’s own educational advantage. These Anglo-
Australians also showed preference in most situations for standard or prestige varieties of
English. The researchers observe that while second generation members of immigrant
groups are often under strong pressure to assimilate and to abandon their community
languages, opportunities to learn and use community languages are somewhat restricted.

In addition, young second generation Australians may in some cases have even more

28

negative attitudes toward non-standard accents in English than Anglo-Australians
although they may value their own ethnic language as a signal of solidarity with their
ethnic community. This study is relevant to our project in that it deals with an immigrant
population in a predominantly English-speaking country. It also deals with issues of
contemporary globalization and the various dynamics that inform this phenomenon.
Research in linguistic attitudes by Akere, Funso (1982) centers on language use and

language attitudes in a Yoruba suburban town as a sociological response to factors of

traditionalism and modernity. This study explored the attitudes toward dialect selection

along with other aspects of language use in a Yoruba community. The study also
highlights the problems of using Western social science survey techniques to help elicit
sociolinguistic data in traditional Afiican societies. One problem of using Western
science in Africa is the issue of standards, norms and models (terms often used
interchangeably in the literature), which, is also a frequent concern of many researchers
in language attitude studies. When it comes to attitudes, a split commitment was
discovered in that Yoruba, one of the study’s languages is seen as a symbol of larger
ethnic identity, whereas the Akoko languages are positively regarded as Vital links with
ancestors. The article underscores the need for collaborative efforts between
communities, linguists, and governments in order to salvage the languages from eventual
death. The issue is also tied to that of an appropriate language model which, also
influences language attitudes (English as a written language versus Yoruba which, is
largely based on mythology and orature) has assumed importance because of its
pedagogical implications. Also informing attitudes toward languages in both mother

tongue and second-language contexts is concern with the variety of English (RP versus

29

African Englishes) that Should be presented as a model of English. Also important is the
issue of how either mother-tongue dialectal forms or local second-language varieties
ought to be treated.

Baker’s (1992) study of attitudes and language highlights several key issues in
language attitude scholarship and bilingualism. The researcher shows advances in current
trends in language attitude research. This work is helpful as it helps tie previous
theoretical frameworks between research language attitudes and bilingual education,
which form an important part of our research. Baker establishes the concept of attitudes
as more central to the study of minority and majority languages and demonstrates that
language attitude theory and research is informed by the strong tradition of attitude
theory and research from social psychology. He also demonstrates the considerable
relevance of attitude change theory to language restoration and decay while also
illustrating how language attitude research can use recent developments in attitude
measurement and model building to increase understanding of language attitudes. Baker
shows that at a conceptual and operational level, the attitude toward bilingualism is
distinct from attitudes toward a language. By examining the origins of attitude to a
language and attitude to bilingualism in terms of individual and contextual variables, he
finally focuses on factors that create positive and negative language attitudes. Finally,
Baker’s work focuses on factors that influence attitudes over the crucial teenage years of
a child.

Berry J. & Greenberg J. (1966) reported on behalf of the Committee on
Sociolinguistic Research in Afiica (CSRA) to the African Studies Association (ASA).

They argue that much of the linguistic work done in the colonial era and even in more

30

recent years is inadequate, because of lack of reference to the relevant social context. In
surveying the field as a whole, the CSRA came to the conclusion that, for its immediate
purposes, three subgroups of studies could be distinguished, and research conducted on
the following topics: 1) Description of habitual language usage; 2) Behavior toward
language; 3) Dynamic study of social and psychological variables. As regards language
usage, the report suggests that, linguistic research in sub-Saharan Afi'ica has traditionally
concentrated on the description of the linguistic situation of Negro-African languages.
The majority of these descriptions have been characterized by anthropological “purism”
in that the selection of typical rural varieties of indigenous languages has been the rule.
The report recommended that future linguistic research in Africa would hopefully
broaden its scope to include other kinds of linguistic phenomena, which occupy a central
place in communication in modern Africa. They add that research into the nature of
purely African varieties of standard languages of foreign origin is long overdue and has
much practical relevance. This study is important for acknowledging the research and
knowledge gap in the value and utility of indigenous African languages and attendant
language attitudes by speakers of these varieties, both inside and outside Africa.
Language attitude research in Africa leads to the assumptions that there are some
distinct linguistic features that identify “African accentedness” (Mr’rtonya 1997; 2000) in
the English varieties spoken by Speakers from Africa. Although he studied attitudes of
educated Africans toward forms of Afiican English, this study bears some similarities
with his work in that it also departs from other research approaches and perspectives in

the adjectives used with European based languages.

31

African English has been defined by Schmied (1991) as “. . .forrns of English spoken
by African speakers; this does not imply that there is an acknowledged variety or that
there are several distinct varieties of the language, nor that these forms are already
standardized and codified in any way” (2). This regionally based definition also groups
together all forms of English, including Creole and Pidgin varieties mostly spoken in
West Afiica and the islands of Malagasy, Reunion, and Cape Verde, right up to the many
other English varieties spoken in Africa.

Focusing on syntactic and semantic properties of educated Africans, Bokamba (1981,
78) notes that African Englishes “share certain properties” identifiable as “Africanisms,
in that they reflect structural characteristics of African languages” and that “these
properties can be discovered at all linguistic levels: phonological, morphological,
semantic and syntactic.” The research supports previous findings that argue native
English speakers, based on simple African intonation in English, can clearly identify such
speakers. These African speakers are also likely to recognize similarly distinctive
linguistic properties in both AAL and SAE. Bokamba’s focus on syntactic and semantic
language properties to determine Africanisms in varieties of African English among
educated groups of Africans is also based on how these new forms of English varieties
deviate “from authentic English” (1965, 198) as a result of the transfer of linguistic
features and discursive strategies by educated Afiicans. The transfer is of the same
features and discursive strategies along with utterance of the educated Africans’
respective mother tongue languages.

This evidence helps to accentuate the differences between African English variety and

RP —— Received Pronunciation. GOrlach (1991) argues this is crucial as it is not only “the

32

most reliable test for localizing a speaker” (24) but according to Schmied (1991) it is also
“the most flexible element, which can be used (subconsciously) to express subtle
sociolinguistic messages of speaker identity and of distance from or solidarity with the
listener” (57). In Africa “pronunciation of English. . .is of particular importance because
(non-standard) pronunciation features seem to be the most persistent in Afiican varieties
i.e. they are retained even in the Speech of the most educated speakers” (Schmied 1991,
57). Consequently, how Afiican immigrants in the US use and perceive varieties of
English, including Standard American English (SAE) amongst other European languages
used in Africa (i.e. French and Portuguese), African languages and Afiican American
Language (AAL), is related to the foregoing language attitude research, particularly in

Africa.

33

CHAPTER 3: METHODOLOGY AND RESEARCH PROCEDURES

To elicit language attitudes from the respondents, this study utilized the matched
guise method, developed by Lambert and associates (1960, 1967) in which a bilingual
speaker says the same thing in two different languages or language varieties. The “guise”
is that the speaker is represented as being a different speaker when speaking a different
language, thus “fooling” respondents into thinking that they are rating a different speaker
when in actuality, only the language is different. The matched guise allows researchers
to Obtain feedback on perceptions of language per se since the “two speakers” being rated
are the same person. Thus all speaker variables are controlled, with only language being
manipulated.

The matched -guise research procedure has been modified on numerous occasions in
its various applications by researchers. Normally, the technique involves prior recording
of anonymous voices of respondents selected to help provide voice stimulus as they read
from a selected word-list or a reading passage. Subsequently, these voices are then
presented to other respondents selected for a study who are often drawn from both a
variety and broad spectrum of linguistic and social backgrounds in regards to such
variables as the level of education, socio-economic status, gender, age, linguistic identity,
regional and ethnic affiliation, etc.

The raters or evaluators who have been selected as the subjects for the study are then
requested to make subjective evaluations and judgments about each anonymous voice,
also referred to as the voice stimulus, along a seven-point Semantic Differential Scale,
the main tool utilized for attitude measurement in the field of language attitude studies.

The deployment of the semantic differential scale typically involves evaluating a

34

speaker’s voice by rating the latter on the scales comprised of adjectival opposites
(Williams 1983). In order to design a semantic differential scale which will be used to
test whether a language variety such as African American Language (AAL) or Standard
American English (SAE), evokes in its listeners a feeling or sense of being socially close
or being socially distant, a researcher would need to design a scale comprising such
adjectives as “friendly,” “gentle,” “trustworthy,” etc.

FIGURE 1: AN ILLUSTRATED EXAMPLE OF THE SEMANTIC DIFFERENTIAL SCALE USED

IN THE STUDY

Voice#1

1 2 3 4 5 6 7
Friendly not friendly

 

l 2 3 4 5 6 7
Gentle not gentle

 

1 2 3 4 5 6 7
Trustworthy -------------- not trustworthy

Respondents were asked to listen carefully to each voice and to write down any
impressions of the speaker that struck them when they listened to the anonymous voices-
(see Appendix A). Respondents were asked to rate each Of the “twelve” speakers on a
Semantic Differential Scale of twelve traits (Well-educated, Confident, Responsible,
Gentle, Friendly, Polite, Leader-like, Patient, Proud, High status, Trustworthy and Social)
using the questionnaire provided (see Appendix A). The order in which the twenty-four
respondents listened to the voices was reversed so that of the twenty-four respondents,
the first twelve respondents heard the AAL voice first, followed by the SAE voice and

the next twelve respondents heard the SAE voice first, followed by the AAL voice.

35

Respondents were asked to place a check on the scale next to each adjective to
indicate their feeling or impression of the voice using that language variety. Each point
on the scale was given a numerical value, ranging from l-7 as seen in Figure l. The
twelve adjectives comprising the scale used in this study were generated in a pilot study
in 2003 involving a purposively selected group of twelve African immigrants from West,
East and South Africa who were asked to provide open-ended descriptions of Afiican
American speech and Standard American English. The adjectives most frequently used
by respondents during the aforementioned pilot study were selected for the construction
of the Semantic Differential Scale in the present study. The full Survey Questionnaire is
given in Appendix A of this study.

As we conducted this study of attitudes of African immigrants in the US toward
African American Language we found only one related empirical study (Smitherman
1998). No other studies in the US on the same topic were found. Needed therefore are
similar language attitude research studies that are multifaceted and/or which also utilize
multidimensional research methodologies and generate equally sophisticated data. Most
language attitude research in Africa is largely non-empirical in the methods used to both
collect and analyze the data. While insisting the solution to this deficiency is the use of
“more sophisticated ...more thorough. . .sampling and analytical techniques (Adegbija
1994a, 55), other scholars such as Baker (1992, et al) recommend the use of protocols
that utilize a wider variety and combination of research methods and procedures.

Seeking to begin to fill this research gap, this study used empirical research techniques
that involved scientific sampling and data collection using the matched-guise method and

a seven-point Semantic Differential Scale. These empirical techniques have enabled us to

36

measure the language attitudes of African immigrants, currently residing in the USA,
toward African American Language (AAL).

In terms of its broader relevance to sociolinguistic studies, this project has tried to
combine research procedures commonly utilized in language attitude research. We
acknowledge that the sample size is limited, which somewhat also limits the number of
statistical analyses. Our study, nevertheless, gains strength in the purposive choice of the
sample, along with the use of research techniques and approaches that are social
psychological. Additionally, our statistical analyses, using the Statistical Package for

Social Science Research (SPSS) belongs to the social psychological research model.

DATA COLLECTION PROCEDURES

Data was collected from a purposively selected group of twenty-four African
immigrants in a medium-sized Mid-western city in the US. The city is home to many
new immigrants from Africa and around the world. Based on his knowledge of Afiica,
the researcher identified the respondents from West, East and South Africa. Their ages
ranged from 19 to 43. Respondents were carefully selected in order to accommodate
both their various age levels and language experiences with the two varieties of English
i.e. SAE (or LWC) and AAL. The respondents were pre-screened for their ability to
speak an indigenous African language as their primary language. All of the respondents
had learned English (as well as both French and Portuguese along with Afrikaans) as
their second and/or third language.

The sample of twenty-four respondents was divided into eight respondents from each

of the three African regions (West, East and South Africa) selected for this study, with

37

each region represented by four females and four males. The study used twelve audio-
recorded voices reading from three short passages based on writings by l) Chinua
Achebe a Nigerian writer, 2) Caribbean writer and poet, Derek Walcott from St. Lucia, in
the Caribbean, and 3) Alice Walker, African American writer and poet. The three writers
were selected because their works represent the global Afiicana and Pan-African spaces
represented in this study.

Two African Americans, a male and a female, who are bilingual that is, skillful in
code switching from AAL to LWC and vice versa, were, selected to read the three short
passages. Chinua Achebe’s passage is taken from Home and Exile (2000, 1-2); Derek
Walcott’s is from The Bounty (1997, 19); and African American writer Alice Walker’s is
from The Color Purple (1982, 11). Each of the two speakers read the passages in the
order listed below. Both were asked to read the three passages in African American
Language and in Standard American English (SAE)/Language of Wider Communication
(LWC), producing twelve voices as stimuli. The readings were tape-recorded and varied
to control for order effect. The following are the three audiotaped passages, which the
twenty-four African respondents were asked to listen to and rate:

a. Chinua Achebe (Home and Exile (2000, 1-2):
One of the earliest memories I can summon from the realm of childhood was a
homecoming that was extraordinary even for such recollections. I was returning to
my ancestral home for the first time. The paradox of returning for the first time
need not detain us now because there are more engaging things at hand. I was five
years old and riding in a motor vehicle for the first time. I had looked forward
very much to this experience, but it was not working out right. Sitting in the back
of the truck and facing what seemed the wrong way, I could not see where we
were going, only where we were coming from. The dust and the smell and the
speed and the roadside trees rushing forward as we rushed back finally overcame

me with fear and dizziness. I was glad when it finally came to a halt at my home
and my town.

38

b. Derek Walcott (The Bounty 2004, I9):

I cannot remember the name of that seacoast city,

but it trembled with summer crowds, flags, and the fair
with the terraces full and very French, determinedly witty,
as perhaps all Europe sat out in the open air

that was speckled and sun-stroked like Monet that summer
with its grey wide beach, ah yes! It is near Dinard,

a town with hyphens, I believe in Normandy

or Brittany, and the tide went far out and the barred

sand was immense. I was inhabiting a postcard.

The breeze was cold, but I did a good watercolour,

and it stands there on the wall. And though it is dated,
time races across its surface but nothing changes

its motion, the tidal flats not clouded, the tiny

figures in the distance, the man walking his dog. Any
stroke and tint have eluded time. Still, it estranges.

Now, so many deaths, nothing short of a massacre

from the wild scythe blindly flailing friends, flowers, and grass,
as the seaside city of graves expands its acre

and the only art left is the preparation of grace.

So, for my Hic Jacet, my own epitaph, “Here lies

D.W. This place is good to die in.” It really was.

6. Alice Walker (The Color Purple (1982, 11):

Dear God,

It took him the whole spring, from March
to June, to make up his mind to take me.
All I thought about was Nettie. How she
could come to me if I marry him and he be
so love struck with her I could figure out a
way for us to run away. Us both be hitting
Nettie’s schoolbooks pretty hard, cause us
know we got to be smart to git away. I
know I’m not as pretty or as smart as
Nettie, but she say I ain’t dumb.

The way you know who discover America,
Nettie say, is think about cucumbers. That
what Columbus sound like. I learned all
about Columbus in first grade, but look
like he the first thing I forgot. She say
Columbus come here in boats call the
Neater, the Peter, and the Santomareater.

39

Indians so nice to him he force a bunch of

‘em back home with him to wait on the

queen.

DATA ELICITATION
The three reading passages were preferred over a word-list, another choice in language

attitude research that uses the matched-guise technique. On a stylistic continuum, a
reading passage was preferred because it is a relatively less formal style of speech as
opposed to the more careful word-list reading. Additionally, the reading passage helps
elicit the various distinctive linguistic features often present in an individual’s speech and

makes such features much more distinct and discernible to raters.

PILOT STUDY

This study follows a pilot study conducted in 2003 on the same topic, in which
twelve African respondents were able to identify AAL and SAE in a matched-guise
exercise. They rated both varieties, which had been audio-recorded as stimulus voices to
help disguise them for evaluation on a semantic differential scale. As detailed in the
section that follows, “THE SAMPLE,” the probability sampling techniques included
stratified random, systematic and purposive random sampling.

The twelve respondents in the pilot study preferred using various adjectives such as
“Well Educated,” “Intelligent,” “Social,” etc. to help identify language varieties they
were familiar with. The final preference of the respondents in the pilot study resulted in
the following list of twelve adjectives: “Well Educated,” “Confident,” “Responsible,”
“Gentle,” “Friendly,” “Polite,” “Leader-like,” “Patient,” “Proud,” “High status,”

“Trustworthy” and “Social.” The choice of the adjectives was also selected following

40

previous work on attitudes Of educated Africans in the US toward varieties of African
Englishes (WAfrE, EAfrE and SAfrE) by ML’Itonya’s (1997).

The selection of appropriate adjectives in the data elicitation procedure from the
twelve selected African respondents was of major importance in helping to determine
how these African immigrants in the Diaspora would evaluate both AAL and SAE.
Mr’rtonya’s (1997) study of educated Africans’ attitudes toward African Englishes noted
that borrowing adjectives from previous language attitude research might not work with
African immigrants due to differential cultural experiences and worldviews.

Based on the findings of the pilot study in 2003, and following MUtonya’s work
(1997), this study used nine adjectives from the pilot study--“Confident,” “Responsible,”
“Gentle,” “Polite,” “Leader-like,” “Patient,” “Proud,” “High Status,” and “Social”). Only
three adjectives were used from Tucker and Lambert’s study (1972): “Friendly,”
“Educated,” and “Trustworthy.” The final twelve adjectives that respondents used to
evaluate the twelve anonymous stimulus voices in both AAL and SAE were: “Well -
Educated,” “Confident,” “Responsible,” “Gentle,” “Friendly,” “Polite,” “Leader-like,”
“Patient,” “Proud,” “High status,” “Trustworthy,” and “Social.”

The above list of adjectives differs from those used in earlier studies, such as the non-
African (Tucker and Lambert 1972, 178-84) and African-oriented (Wood 1994) attitude
studies. The adjectives that Tucker and Lambert (1972) used to determine the attitudes of
White and Black American students toward varieties of American English included,
“Upbringing,” “Intelligent,” “Disposition,” “Speech,” “Faith in God,” “Talented,”
“Character,” “Personality,” and “Considerate” among others. In his study of attitudes

toward French, national and mother tongue languages in the Republic of Congo, Wood

41

(1994) elicited adjectives such as “Beautiful,” “Intelligent,” “Trustworthy” and
“Friendly.”

This study, which is designed to elicit attitudes of African immigrants in the USA
toward AAL, differs from the method used by Wood in two ways. First, he used the
direct method to elicit data while we use the indirect method. Secondly, Wood asked his
respondents to evaluate the various languages, while we asked our respondents to
evaluate anonymous voices of speakers of both AAL and SAE. The method and data
collection of this investigation are therefore different from those previously used with

Black groups in North America as well as in Africa (Republic of Congo).

THE SAMPLE
The twenty-four respondents from West, East and South Africa, ages 19 to 43, who
all speak an African language as their first or primary language, had also lived in the
USA among English-speaking communities from one year to sixteen years. They
included both female and male raters in equal proportions (i.e. four females and four
males from each region).

The respondents came from English, French and Portuguese speaking sub-Saharan
African countries and had learned English as a second (for most raters from Anglophone
Africa) or third language (for many raters from Francophone and Lusophone Africa)
mostly, in the early years of their formal education. All were asked to evaluate each of
the twelve anonymous stimulus voices and to rate them on twelve adjectives using a

Semantic Differential Scale (See Appendix A).

The location Of the study in a medium-sized Mid-westem city in the USA

notwithstanding, and while we were somewhat limited in the size of the sample, we tried

42

to maximize the benefits through the use of stratified random, systematic random and
purposive probability sampling procedures. These procedures were needed in order to
Obtain a representative sample of Afiican immigrants in an academic community
surrounded by an equally small USA community. The sample size of twenty-four
African respondents is therefore a reflection of the universe from which it is drawn. As
generally accepted in applied social research, we consider the sample to be accurate and

rigorous. A detailed discussion of our sampling procedures follows.

Overall, most sampling methods are purposive in nature because researchers usually
approach the sampling problem with a specific plan in mind. In using purposive
sampling to help ensure gender balance for example, the study concurs with the logic that
sampling the population always has a purpose in mind. In the study the specific and
predefined group under focus was African immigrants in the USA who originate from
West, East and South Africa and whose primary language is an indigenous Afiican
language. Additionally, the respondents must also have lived in a USA English-speaking
community.

SAMPLING FRAME AND PROCEDURES
This study selected twenty-four African immi grants from a medium-sized Mid-

western city that has a population of approximately 47,000 Of which over 60% are
college students. There is a bigger city adjoining it with a population of approximately
119, 000, where some of the African immigrants in our study also live. At the beginning
of spring 2003 when we conducted the pilot study that followed similar sampling
procedures albeit with half of the current sample, there were twelve African respondents
and six stimulus voices. At the start of on the present study, approximately one year ago,

we sent out email and poster notices seeking research participants through both the

43

African Studies Center and the African Students Union as well as through local churches
with specific guidelines outlined in these notices. Two crucial requirements were that
research participants should first be from West, East or South Africa and must be
speakers of an indigenous African language as their primary language. They must also
have lived in a US English-speaking community. To enable the study to draw an
acceptable sampling frame, we made a list of fifty-five African immigrants who
responded to the notices.

To select the final list of twenty four raters that we felt represented the population
under investigation, the study used a stratified random sampling in which the fifty-five
potential research participants where classified into three regional groups: West, East and
South Africa, equally divided into 3 groups of 18 each (SS/3:18.33 or 18). This
sampling technique was needed in order to ensure the final sample equally represented
the three African regions, the focus of the study. A systematic random sampling
followed where every second respondent (55/24=2.29 or 2) was selected from each of the
three lists representing the three regions with eight names selected from each stratum.
This yielded a total of twenty-four respondents who were then selected for the study. To
attain gender balance, we chose a purposive sampling procedure. We also used the same
criteria in the pre-screening procedure that required the respondents to have the ability to
speak an indigenous African language as their primary language and membership as an
African immigrant from West, East and South Africa. They must also have lived in an
English-speaking community in the US. While ensuring the adequacy of our probability
sampling techniques, we also realize that with a purposive sample used to ensure gender

balance in the study, we were likely to get the opinions of our target population, but also

44

likely to overweight subgroups that are more readily accessible. We therefore can
perhaps only draw limited inferences from our findings. The three sampling procedures
used (stratified, systematic and purposive) were all deemed necessary in order to
adequately include African immigrants who were born in Africa and came to the US
later. Most important for the study, however, were the requirements that they also had to
be speakers of an indigenous African language and originated from West, East and South
Afiica.

These two pre-screening measures were useful in selecting only respondents who were
motivated and showed interest and cooperation in participating in the entire study. Final
communication to all pre-selected participants was made by telephone, and email
contacts were made in order to arrange for individual appointments during which the
respondents listened to the twelve stimulus voices, rated them and also completed the
questionnaires in Appendix A. In our pre-screening, we were required to drop two
potential respondents due to what we gauged as extreme bias verbally expressed toward
either AAL or SAE. This did not, however, affect our sampling frame and related
procedures.

The following Table 1a is the breakdown of both African and European languages

spoken by the twenty-four African raters in the study:

45

TABLE 1A: PROFILES OF THE 24 AFRICAN RATERS BY LANGUAGE & REGION

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Languages Region Respondents Gender
Afrikaans SA 1 1 Female
0 Afrikaans-English SA 1 1 Female
Amharic EA 2 2
(IF emale;
lMale)
Akan/Twi WA 2 2 (IF emale;
lMale)
Ashanti WA 1 1 Male
Bambara WA 3 3 (2Female;
1 Male)
Bamileke WA 1 1 Female
Bemba SA 3 3 (IF emale; 2
Males)
Chichewa SA 1 lMale
Dholuo EA 1 2 (lFemale; 1
Male)
Egusii EA 1 lMale
English WA, EA & SA 24 24 (12F emales;
12 Males)
French WA&EA 7 7 (4 Females; 3
Males
0 French-English WA&EA 7 7 (4 Females; 3
Males)
Kikuyu EA 2 2 Females
Herero SA 1 1 Female
Kimeeru EA 1 1 Female
Kinyarwanda EA 1 1 Male
Kiswahili EA, SA 9 9 (4 Females; 5
Males)
Lingala EA, SA 2 2 Males
Luganda EA 1 1 Male

 

 

 

 

46

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table la (Continued)

Ndebele SA ‘ l 1 Female

Portuguese SA 2 2(1 Female; 1
Male)

0 Portuguese-English SA 2 2 (1 Female; 1

Male)

Ronga SA 1 1 Male

Sena SA 1 1 Female

Shona SA 1 1 Female

Xitswua SA 1 1 Female

Yoruba WA 2 2 Males

No Response 00 00 24 (12 Female;
12 Males

African Languages Total 22

(SA=8, EA=9, WA=5)=22

European Languages Total 4 (Afrikaans, English,

(SA=3: Afrikaans, English and French and

Portuguese; EA=2: English and French; Portuguese)

WA=2: English and French)

Languages Total =22 African and 4 26

European =26

Note: European languages= (Afrikaans,

English, French and Portuguese)= 4

Total = 26 3 Regions 24 Respondents

(WA=8, EA=8
and SA=8)

 

Key (for Tables la and lb):

0 Respondents who officially speak more than one European language

1. WA=West Africa, EA=East Afiica, SA=South Africa

47

 

TABLE 13: PROFILES OF THE 24 AFRICAN RATERS BY LANGUAGE. GENDER & REGION

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Languages Gender African Region Respondents

Afrikaans Female SA 1

0 Afrikaans-English Female SA 1

Amharic 1 Female EA 2
1 Male

Akan/Twi 1 Female WA 2
1 Male

Ashanti 1 Male WA 1

Bambara 2 females WA 3
1 Male

Bamileke 1 Female WA 1

Bemba 1 Female SA 3
2 Males

Chichewa 1 Male SA 1

Dholuo 1 Female EA 2
1 Male

Egusii 1 Male EA 1

English 12 Females WA, EA and SA 24
12 Males

French 4 Females WA, EA 7
3 Males

0 French-English 4 Females WA, EA 7
3 Males

Gikr'ryu 2 Females EA 2

Herero 1 Female SA 1

Kimeeru 1 Female EA 1

Kinyarwanda 1 Male EA 1

Kiswahili 4 Females EA, SA 9
5 Males

 

 

 

 

48

 

Table 1b (Continued)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Lingala 2 Males EA, SA 2

Luganda 1 Male EA 1

Ndebele 1 Female SA 1

Portuguese 1 Female SA 2

1 Male
0 Portuguese-English 1 Female SA 2
1 Male

Ronga 1 Female SA 1

Sena 1 Male SA 1

Shona 1 Female SA 1

Xitswua 1 Female SA 1

Yoruba 2 Males WA 2

NO Response 00 00 00

24 Respondents: 24

50% females and 50%

males per region

African Languages Total 22

(SA=8, EA=9, WA=5)

European languages Total 4 (Afrikaans,
English,
Portuguese and
French)

Languages Total 26

 

 

 

 

 

 

Tables 1a and 1b presents a profiles of the twenty four African immigrants spoke a
total of twenty-two African and four European languages, including Afrikaans spoken in
the Southern African region. We will not consider Afrikaans an indigenous African
language in this study as the variety does not belong to the Niger-Congo family of

African languages (MI'Itonya 1997). The East African raters speak the most group of

49

diverse African languages (9) followed by South Africans (8) and West Africans (5).
South Africans speak more European languages (i.e. Afrikaans, English and Portuguese)
than both West and East Africans who speak two European languages, namely English
and French. More female respondents in the sample spoke a wider variety of African
languages (19) than the males (15) and also more (18) of them spoke a wider variety of
European languages than their male counterparts (15). More East African raters spoke a
wider variety of African languages than their WA or SA counterparts. More males from
both EA and SA spoke a wider variety of African languages (8) than males from WA (4).
While many more Afiican languages are spoken in both EA (9) and SA (8) than in WA
(5), this latter group has more speakers who are fluent in both English and French (7);

that includes all female raters (4) and all but one male respondent.

DATA ANALYSIS
MEANSCORE RATING

Ratings from the twenty-four respondents of both AAL and SAE on a Semantic
Differential Scale in matched-guise exercise are herein considered to be a reflection of
these respondents’ attitudes toward the AAL and SAE varieties that this study focused
on. Using an SPSS (Statistical Program for Social Sciences) computer program, we were
able to calculate the Mean Score Rating, based on African region, gender, age, and
number of years respondents have lived in the USA, for the twelve stimulus voices on
each of the twelve adjectives used in this study. We achieved this task by using the
technique used by Tucker and Lambert (1972) of ranking of each of the 12 traits or

characteristics in order to determine how the evaluators acting as the judges rated each

50

voice on the twelve traits. The mean score was then calculated by dividing the total value
by the total number.

The resulting mean score for all measured features of the twelve voices was calculated
and a Factor Analysis conducted as well. In order to obtain a correct estimate of the
raters’ attitudes toward the two language varieties and in particular toward AAL speedily,
we used two methods of analysis as recommended by Tucker and Lambert (1972). This
method of Rating and Ranking of the Mean Scores helped indicate the raters’ preferences
of the voices and also helped indicate how they associated the voices with the twelve

characteristics on the Semantic Differential Scale.

CORRELATION COEFFICIENT
Next, a Correlation Coefficient Matrix was designed for African region, gender, age,
and for the number Of years respondents have spent in the USA to help determine the
degree of correlation of the traits based on significance level. We recognized that a
correlation coefficient might not be as efficient as the Factor Analysis in achieving the
aforementioned task. We nonetheless took this into account even as we sought to exploit

the relevant techniques proposed elsewhere as seen in the pioneering work by Tucker and

Lambert work (1972).

FACTOR ANALYSIS
A Factor Analysis was used to analyze firrther the mean scores in this study and to
also help determine the range of underlying factor groups of the semantic differential
scale. Recommendations by Hatch and F arhady (1982) suggest that a Factor Analysis is

helpful in this kind of study as it is premised on the assumption that in every statistical

51

test, there probably are multiple characteristics that are being simultaneously explored
and assessed. A Factor Analysis also helps to condense and cluster variables tested in the
study and helps correlate both the factors and observed scores. These scores help to show
which items have a common variance and/or can be isolated and identified as the
underlying factor under a new identifying tag or label. Take for example a hypothetical
factor such as “Competence,” which could be comprised of such variables as
“Responsible,” “Well Educated,” and “Leader-like.” In an exploratory study such as the
present one, the relationship among the aforementioned traits or variables could therefore
be described using these same variables if the scores indicated that they correlate as an
underlying factor group.

Consequently, factors can then be rotated using the Varimax Rotation Technique
available in SPSS in order to help maximize the correlation between a variable and the
factor to which it correctly belongs, while at the same time, minimizing the correlation
between a variable and the factor to which the variable does not belong. Ultimately, a
variable is therefore considered to be positively loaded in any given factor group if its

loading is .30 or greater.

SCOPE AND LIMITATIONS OF THE STUDY
This study limits itself to investigating language attitudes of African immigrants in the
US toward AAL amidst a variety of social, cultural, political and economic global
discourses, historically dominated by European languages and which have historically
marginalized both African languages and AAL as well as their speakers. Consequently
language attitudes toward both European and African languages as well AAL have most

likely been impacted by these circumstances.

52

The study recognizes that controlled passages rendered as stimulus voices may be
inadequate in revealing the wide range of unique features in AAL. However the three
reading passages by Achebe, Walcott and Walker were selected in order to control for
both AAL and SAE/LWC varieties. The three passages are from works by Achebe, an
African writer and Walcott, a Caribbean poet, both of whom write in SAE/LWC in
contrast to the third passage by Alice Walker, which displays prosodic as well as
syntactical features of AAL. The Walker passage was also read in SAE.

The focus of this study is on African immigrants in the US and their attitudes toward
AAL and SAE, with the assumption that the global use of AAL and SAE by the stimulus

voices will have an impact on these attitudes.

RESEARCH QUESTIONS
Notwithstanding many other myriad issues that may have arisen in the course of this
project, this study posed the following principal research questions:
1) How do African immigrants who hail from countries with “dense multilingualism”
(Adegbija 1994a) View Afiican American Language (AAL), a.k.a. African American
English (AAE) or US Ebonics?
2) How do African immigrants View the Language of Wider Communication (LWC),
a.k.a. “Standard American English” i.e. SAE?
3) What is the nature of these attitudes toward AAL and SAE based on African region,
gender, age and number of years the African immigrants have lived in the USA?
RESEARCH HYPOTHESES
1) Respondents will respond differently to both AAL and SAE based on what they

will hear in listening to both varieties.

53

2) Respondents’ attitudes will be either favorable or unfavorable toward AAL and
SAE.

3) Respondents’ attitudes will be influenced by the voices reading the passages, in
AAL and SAE, selected for this study.

4) Respondents’ attitudes toward AAL and SAE will likely be influenced by age,
gender, years lived in the USA, level of education, and by where they come from in
Africa (i.e. West, East and South Africa).

5) This study’s main null hypothesis predicts that there would be variation in attitudes
of the three sets of African immigrants in the USA toward AAL and SAE. There is
also the likelihood that there might not be significant differences in attitudes toward

AAL and SAE among the three sets of respondents from West, East and South Africa.

MAJOR ASSUMPTIONS IN THE STUDY
1) New African immigrants to the US will in general have favorable attitudes toward
a language variety or varieties they are most familiar with.
2) African immigrants are more likely to be familiar with the “Queen’s English”
variety spoken in England, or French and Portuguese. They’ll also likely identify
with AAL based on its distinctive pronunciation and prosodic patterns.
3) These African immigrants will also identify with SAE because as immigrants they
are aware of the need to adapt to what is considered a standard variety, which because
it is a European language is similar to the “Queen’s English” or French and
Portuguese in status. This is based on the reality these immigrants face in accessing
power and privilege using a standard lingua franca in their new circumstances in the

USA.

54

CHAPTER 4: ANALYSIS AND PRESENTATION OF DATA
This chapter will focus on the analytical techniques applied in our study to determine the
nature of language attitudes of African immigrants to the USA toward African American
Language/English (AAL/AAE) and SAE. Data presentation uses the mean score rating
and ranking of means to provide initial results of the evaluators’ attitudes toward both
SAE and AAL based on a trait-by-trait analysis of what the study identified as twelve
distinctive voice characteristics. Both the Correlation Coefficient Matrix and Factor
Analysis supplement this analysis.

The Correlation Coefficient Matrix explored how the twenty-four African raters
unconsciously correlated the twelve adjectives they used to rate the stimulus voices of
AAL and SAE speakers on the Semantic Differential Scale. This technique is unique in
that it combines unconsciously correlated adjectives into a cluster that provides a more
clearly defined and identifiable measure for each member of the set of twelve adjectives.
In this manner we tried to obtain a much more balanced and organized reduction of the
various individual voice stimulus traits into more manageable clusters. To achieve this,
we conducted intensive analyses even as we tried to exploit the value of this technique to
help uncover what is often considered deep-seated and hidden attitudes within the
evaluators’ minds. In this case, our focus was on evaluators’ attitudes towards AAL while
we used SAE for contrastive purposes.

MEAN SCORE RATING AND RANKING ANALYSIS

To Obtain the Mean Score Ratings and Ranking of Means, a score of each Of the

twelve adjectives was calculated for the three African regions. We used this method,

which was also used by Tucker and Lambert (1972). The results are shown in Tables 2,

55

3, 4 and 5. The [brackets] in each column refer to the rankings of each variety in regards
to that particular language feature.

The Mean Score Rating of each of the twelve voices was calculated for each of the
twelve adjectives, based on African region (WA, EA and SA), age, gender and years
spent in the USA. The overall rank for each of the twelve stimulus voices is shown in the
“sum” column. This score is derived after adding all the rank scores for each of the
twelve voices on each of the twelve traits. The ranks are based on the high (or positive)
scores or values meaning the “sum” score is lowest for the most positively ranked voice,
since this score is based on rank and not on value. Comparison of attitudes across the
three African regions (WA, EA and SA), gender, age and years spent in the USA is
presented in the following Table 2 through 8. We first ascertained there were no
significant statistical differences in terms of gender, age, and years spent in the USA
among the twenty-four raters from West, East and South Africa in their attitudes toward
AAL and SAE in the matched-guise exercise. There were, however, important

differences in how the raters ranked AAL and SAE based on the three Afiican regions.

56

TABLE 2: MEAN RATINGS AND RANK OF MEANS OF ATTITUDES TOWARD

AAL AND SAE BY ALL RATERS FROM WEST AFRICA, EAST AFRICA

 

 

 

 

 

 

 

 

 

 

 

 

AND SOUTH AFRICA
Traits
EDI CONF RESP GEN FRIE POLI LEA PAT PRO STA TR SOC Sum
D U

VOI 4.0 3.0 3.2 3.1 3.0 3.2 4.0 3.0 3.0 4.0 3.2 3.0 54
CE 1 [3] [5] [5] [5] [5] [5] [3] [5] [5] [3] [5] [5]

AA

L

(M)

V01 2.3 2.1 3.0 2.1 2.5 2.3 3.2 2.4 3.2 3.0 3.0 2.4 64
CE 2 [6] [6] [5] [6] [5] [6] [5] [5] [5] [5] [5] [5]

SAE

LM)

VOI 3.0 30 3.0 3.3 4.0 4.0 3.4 4.0 3.0 3.1 3.1 3.3 54
CE 3 [5] [5] [6] [5] [3] [3] [4] [3] [5] [5] [5] [5]

AA

L

(F)

VOI 3.0 3.0 3.1 3.4 3.4 3.4 3.0 4.0 2.4 4.0 3.0 3.4 53
CE 4 [5] [6] [5] [4] [4] [4] [5] [3] [5] [3] [5] [4]

SAE

(F)

VOI 3.0 2.5 3.0 3.0 3.0 3.0 4.0 3.2 3.0 4.0 3.4 3.0 55
CE 5 [5] [5] [5] [5] [5] [5] [3] [5] [5] [3] [4] [5]

AA

L

(M)

VOI 2.1 2.0 2.1 2.1 2.3 3.0 3.1 2.4 3.0 3.2 3.0 3.0 67
CE 6 [6] [6] [6] [6] [6] [6] [5] [5] [6] [5] [5] [5]

SAE

(M)

VOI 3.0 2.0 2.5 3.2 3.0 3.1 3.1 3.5 3.0 3.0 3.0 3.0 60
CE 7 [5] [6] [5] [5] [5] [5] [5] [4] [5] [5] [5] [5]
AA

L
IF)

VOI 3.0 2.2 3.0 3.4 3.3 3.2 3.0 4.0 3.0 4.0 3.1 3.3 54
CE 8 [5] [6] [5] [4] [5] [5] [5] [3] [5] [3] [5] [3]

SAE

(F)

VOI 4.0 3.3 4.0 3.3 3.3 3.5 4.4 3.4 3.3 5.0 4.0 3.3 46
CE 9 [3] [5] [3] [5] [5] [4] [2] [4] [5] [2] [3] [5]

AA

L

M)

VOI 3.0 3.0 3.0 3.0 3.0 3.0 4.1 3.1 4.0 4.0 3.1 3.0 54
(1333 [5] [5] [5] [5] [5] [5] [3] [5] [3] [3] [5] [5]

SAE

(M)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

57

 

Table 2 (Continued)

 

VOI
CE

I 1
AA
L

(F)

3.0
[5]

30

4 0 .
[5]

. 4 0
[3]

I . 3 5
5] [3]

. 53
[3]

2.2
[6]

30

. 4 0
[5]

. 3.0 3.2 3.3
[3]

[5] [5] [5]

f—lw

 

 

VOI
CE
12
SAE

L131

 

4.0
[31

40

4 0 .
[3]

. 4 0
[3]

. 3 2
[3]

. 4 l
[5]

. 3 o
[3]

. 47
[5]

22

. 3 0
[6]

. 4 0
[5]

. 4 0
[31

. 3 2
[3]

[51

 

 

 

 

 

 

 

 

 

 

 

 

 

Key (applies to Tables 2-6):

1.

@9999?!”

ED= EDUCATED, CONF=CONFIDENT, RESP=RESPONSIBLE,
GENT=GENTLE, FRIE=FRIENDLY, POLI=POLITE, LEAD=LEADER-
LIKE, PAT=PATIENT, PRO=PROUD, STA= STATUS,
TRU=TRUSTWORTHY, SOC=SOCIAL

AAL= AFRICAN AMERICAN LANGUAGE

SAE: STANDARD AMERICAN ENGLISH

M=MALE

F =FEMALE

Mean ratings are rounded to one decimal place

Rank of mean ratings is set in brackets

The ranks are based on the high (or positive) scores or values meaning the
“sum” score is lowest for the most positively ranked voice, since this score is
based on rank and not on the value

Table 2 shows the ratings and rankings of the stimulus voices by all twenty—four raters

from the three African regions (West Africa, East Africa and South Africa). An overview

of the rankings indicates a generally more favorable ranking of both female and male

African American Language (AAL) varieties. This group of African raters included

those first twelve raters who listened to the AAL voices first followed by SAE voices as

well as the second set of twelve raters who listened to the SAE voices first followed by

the AAL voices. There were no significant differences based on order effect.

58

 

It is worth noting that while the raters perceive the AAL varieties more favorably, the
SAE female voice, at times, also earned a more favorable rating as well. The African
American Language male (AALM) variety gets the most favorable ranking along with
African American Language female variety (AALF) and both voices also share their most
favored ranking with the SAEF voice. The SAEM voice is perceived less favorably of all
twelve voices by all twenty-four African judges. There is a striking numerical break
between the sums of AALM (46) voice and SAEM (54) voice but less between AALM

(46) and SAEF (47).
' Overall the two AAL voices (i.e. AALM and AALF) can be described as contrasting
with both SAE voices (i.e. SAEF and SAEM), especially when one looks at the twelve
traits that the voices were being rated on. The AALM voice is ranked high (namely, 3 and
above) on all traits except for “Confident.” The AALF voice also gets high rankings
(namely, 3 and above) for all traits except “Confident” and “Responsible.” The SAEM
voice, on the other hand, is ranked less favorably on most of the traits except for
“Gentle,” “Leader-like,” “Proud” and “High Status.” The SAEF voice was ranked highly
on seven of the traits: “High Status,” “Patient,” Responsible,” “Well Educated,”
“Gentle,” “Friendly,” and “Polite.”

There appears to be a close correlation in the rankings of AALF, AAL and SAEF
voices. The high rankings of these three voices, in contrast to that of the SAEM voice,
whose overall summary rankings are lowest of all twelve voices, distorts what initially
appeared to be an easily predicted pattern of ranking for all the voices. It had been
assumed that raters would respond favorably to one variety in contrast to the other.

However, the data have proven to be more interesting and more complex and call for

59

additional analysis to unpack why the SAEM variety is ranked so unfavorably in contrast

to the other three voices (i.e., AALM, AALF and SAEF).

TABLE 3: MEAN RATINGS AND RANK OF MEANS OF AAL AND SAE BY
EIGHT WEST AFRICAN RATERS

 

 

 

 

 

 

 

 

 

 

 

 

 

Traits
EDI CON RES GEN FRI POL LEAD PAT PRO STA TRU SOC Sum
F P E I
VOIC 4.0 3.4 4.0 4.0 3.3 4.0 4.0 3.4 3.0 4.3 4.0 3.3 44
E 1 [3] [4] [3] [3] [5] [3] [3] [4] [5] [3] [3] [5]
AAL
(M)
VOIC 3.0 2.3 3.0 2.3 3.0 3.0 3.1 2.3 4.0 3.3 3.0 3.0 58
E 2 [5] [6] [5] [6] [5] [5] [5] [4] [3] [4] [5] [5]
SAE
Q4)
VOIC 5.0 3.3 3.4 4.0 4.0 4.0 5.0 4.1 3.0 4.4 4.0 4.0 38
E 3 [2] [5] [4] [3] [3] [3] [2] [3] [5] [2] [3] [3]
AAL
(F)
VOIC 2.3 3.1 3.1 3.0 3.0 3.0 3.0 3.0 3.0 3.4 3.0 3.0 60
E 4 [6] [5] [5] [5] [5] [5] [5] [5] [5] [4] [5] [5]
SAE
(F)
VOIC 3.0 2.3 3.0 3.2 3.0 3.0 4.0 3.0 3.0 4.0 3.3 3.0 55
E 5 [5] [6] [5] [5] [5] [5] [2] [5] [5] [3] [4] [5]
AAL
(M)
VOIC 2.0 2.0 2.0 2.0 2.3 2.3 2.3 2.0 3.2 3.0 2.2 2.3 70
E 6 [6] [6] [6] [6] [6] [6] [6] [6] [5] [5] [6] [6]
SAE
(M)
VOIC 4.0 2.2 3.2 4.0 4.0 4.0 4.0 4.3 2.2 4.0 4.0 4.0 44
E 7 [3] [6] [5] [3] [3] [3] [3] [3] [6] [3] [3] [3]
AAL
(F)
VOIC 2.0 2.0 2.0 2.3 3.0 2.2 2.0 2.0 4.0 3.0 2.2 3.0 66
E 3 [6] [6] [6] [6] [5] [6] [6] [6] [3] [5] [6] [5]
SAE
F)
VOIC 3.3 4.0 4.1 4.3 4.0 4.2 5.0 4.0 3.0 5.0 4.0 4.0 34
E 9 [5] [3] [3] [3] [3] [3] [2] [3] [5] [2] [3] [3
AAL
M)
VOIC 3.0 3.2 3.3 4.0 4.0 4.0 5.0 4.0 4.0 5.0 4.3 4.0 36
E 10 [5] [5] [5] [3] [3] [3] [2] [3] [3] [2] [3] [3]
SAE
(M)
VOIC 4.0 3.0 4.0 4.0 4.0 4.2 4.0 5.0 4.0 5.0 4.2 4.0 32
E 11 [3] [5] [3] [3] [3] [3] [3] [2] [3] [2] [3] [3]
AAL
F)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60

 

Table 3 (Continued)

 

VOIC
E 12
SAE

3.0
[5]

2.2
[61

3.0
[5]

3.0
[5]

3.0
[5]

3.0
[5]

3.0
[5]

3.0
[5]

3.3
[5]

4.0
[3]

2.3
[6]

3.0
[5]

60

 

(F)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 3 shows the ratings and rankings of the twelve stimulus voices by West African
raters. An overview of the rankings indicates, once again, generally more favorable
rankings of both female and male African American Language (AAL) varieties. This
group of WA raters, as also seen in Table 2, included those first twelve raters who
listened to the AAL voices first followed by SAE voices as well as the second set Of
twelve raters who listened to the SAE voices first followed by the AAL voices. As
mentioned, there was no significant difference for order effect. It is worth noting that the
AAL varieties occupy the first two rankings in terms of being the most favored varieties
by the WA raters. The African American Language female (AALF) variety takes the first
ranking followed by the African American Language male variety (AALM).

While the West African raters rank the AALM and AALF most favorably, the SAEM
variety is rated third and SAEF rated! less favorably and most often so by this group of
WA judges. In these rankings, there is a noticeable numerical break between the sums of
the AALF (32) voice and that of SAEF variety (60). There is less numerical break
between the AALM (34) voice and the SAEM (36) voice.

Again as we saw in Table 2 the AAL voices (both AALF and AALM) are perceived
much more favorably in comparison to both SAEF and SAEM voices by the West
African raters. While the WA evaluators also perceive the SAEM voice somewhat
favorably, they have less favorable perceptions and ratings of the SAEF variety. In this

case we can describe the favorable ratings for both AALF and AALM voices as being in

61

contrast to both SAEF and SAEM voices, especially in the case of the latter voice. This
becomes more evident especially when one looks at the twelve traits that the voices were
rated on.

The AALM voice is ranked high (namely, 3 and above) on all traits except for
“Confident.” The AALF voice also gets high ranking (namely, 3 and above) for all the
traits except for “Confident.” The SAEM voice on the other hand is ranked highly on
seven of the twelve traits except for “Well Educated,” “Confident,” “Responsible,”
“Gentle,” and “Patience.” The SAEF voice rating is noticeably poor among the WA
raters in their perception because it doesn’t sound “Well Educated,” “Confident,”
“Responsible,” “Gentle,” “Polite,” “Leader-like,” “Patience” and “Trustworthy.”

Overall among the WA raters, there appears to be a close correlation in the rankings of
AALM and AALF voices while that between SAEM and SAEF voices although evident
is harder to clearly delineate at this juncture. The WA raters’ low ranking of the SAEF
voice, whose overall summary rankings of this voice is lowest of all the 12 voices for
WA needs further analysis. This will be conducted in the Factor Analysis section that
follows later in this study, as the current ranking seems to distort what initially appeared
to be an easily predicted pattern of ranking for all the voices. We saw in Table 2 that in
combination the African raters from all three regions (WA, EA and SA) ranked the
SAEM voice as the lowest among the 12 voices. In Table 3, when asked to rate the voices
separately from the other two African regions (i.e. EA and SA) the WA raters seem to
rank the SAEF voice the lowest. The following is the order of preference for the four

voices among the WA raters: AALM, AALF SAEM and SAEF.

62

TABLE 4: MEAN RATINGS AND RANK OF MEANS OF AAL AND SAE BY
EIGHT EAST AFRICAN RATERS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Traits
EDI CON RES GEN FRI POL LEA PAT PRO STA TR SO Sum
F P E I D U C

VOICE 3.6 2.4 3.1 3.0 2.3 3.0 3.4 3.0 3.0 3.3 3.0 2.3 60
(11:):414 [3] [5] [5] [5] [6] [5] [5] [5] [5] [5] [5] [6]
VOICE 2.4 2.4 3.3 2.0 2.0 2.0 4.0 3.0 3.0 3.0 3.0 2.0 62
2 [5] [5] [5] [6] [6] [6] [3] [5] [5] [5] [5] [6]

SAE

(M)

VOICE 2.0 2.0 3.0 3.1 3.3 4.0 3.0 4.0 2.4 2.0 3.0 3.0 59
3 [6] [6] [5] [5] [5] [3] [5] [3] [5] [6] [5] [5]

AAL

(F)

VOICE 3.4 2.4 3.1 3.4 4.1 4.0 3.0 4.0 3.0 4.0 3.1 4.1 49
4 [4] [6] [5] [4] [3] [3] [5] [3] [5] [3] [5] [3]

SAE
(F)

VOICE 3.0 3.0 3.1 3.0 3.0 30 4.0 4.0' 3.3 4.0 4.0 3.0 52
5 [5] [5] [5] [5] [5] [5] [3] [3] [5] [3] [3] [5]

AAL

M)

VOICE 2.3 20 2.0 2.0 2.3 3.0 3.0 2.3 2.0 3.0 3.0 2.3 68
6 [6] [6] [6] [6] [6] [5] [5] [6] [6] [5] [5] [6]

SAE

(M)

VOICE 2.0 2.0 2.3 3.0 3.0 3.0 2.1 3.1 2.4 2.0 2.4 3.0 65
7 [6] [6] [6] [5] [5] [5] [6] [5] [5] [6] [5] [5]

AAL

(1")

VOICE 4.0 3.0 3.0 4.0 3.0 3.3 3.1 5.0 3.0 4.1 4.0 4.4 49
3 [3] [5] [5] [3] [5] [5] [5] [2] [5] [3] [3] [4]

SAE
(F)

VOICE 4.0 3.0 3.3 3.0 3.0 3.0 4.0 3.4 3.4 4.3 4.1 3.0 50
9 [3] [5] [5] [5] [5] [5] [3] [4] [4] [3] [3] [5]

AAL

(M)

VOICE 2.4 2.3 2.3 2.0 3.0 2.1 4.0 3.0 3.0 3.0 3.0 3.0 63
10 [6] [6] [6] [6] [5] [6] [3] [5] [5] [5] [5] [5]

SAE
(M)

VOICE 2.1 2.0 2.1 3.0 2.4 3.0 23 3.1 3.0 3.0 3.3 2.4 63
11 [6] [6] [6] [5] [5] [5] [6] [5] [5] [5] [5] [5]
AAL
(D

VOICE 4.0 2.4 3.1 4.3 4.3 4.0 3.0 4.4 3.4 4.0 4.0 4.3 43
12 [3] [5] [5] [3] [3] [3] [3] [2] [5] [3] [3] [5]
SAE
(F)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

63

 

According to Table 4, the East African respondents rated the SAEF voice (43) much
more favorably than any other voice followed by the AALM voice (60). There is a
striking numerical break in the rankings between the two voices. While the SAEF voice
was ranked by the EA raters in positions 1 through two, both the AALF (65) and SAEM
(68) voices were ranked lowest by this group. The SAEF voice was most positively
associated with being “Patient,” “Social,” but was rated less favorably on “Confident.”
The AALM voice was rated most favorably for sounding, “Well Educated,” “High
Status,” Trustworthy,” “Proud” and “Leader Like.” The order of preference among EA
raters for the four voices is as follows: SAEF, AALM, AALF and SAEM.

While the EA raters ranked the AALF voice less favorably on many traits, they
nonetheless rated the voice less harshly on “Confident,” “Polite,” and “Social.” While the
EA judges also rated the SAEM voice less favorably on most of the traits, they
nonetheless rated this voice less harshly for sounding “Leader-like.” Further analysis is
required to establish possible reasons for both the high and low rankings for AAL and
SAE) and why both variety and gender play such important roles in language attitudes
among the evaluators from all three Afiican regions (WA, EA and SA). Both a
Correlation Coefficient Matrix and a Factor Analysis will most likely be useful in this

regard.

64

TABLE 5: MEAN RATINGS AND RANK OF MEANS OF AAL AND SAE BY
EIGHT SOUTH AFRICAN RATERS

 

Traits
ED CON RESP GEN FRI POLI LEA PAT PR ST TRU SOC Sum
1 F E D O A

 

VOIC 3.3 2.0 3.3 3.3 2.3 3.0 4.0 2.0 33 43 40 3.3 47
El [5] [6] [5] [5] [6] [5] [3] [6] [5] [31 [3] [5]
AAL
(M)

 

VOIC 2.0 1.3 2.3 23 30 3.0 3.0 30 40 3.0 3.0 3.0 63
E2 [61 [7] [6] [6] [5] [5] [5] [5] [3] [5] [5] [5]
SAE
(M)

 

VOIC 2.0 2.3 2.3 33 33 3.0 3.0 4.0 3.0 3.0 3.0 3.3 61
E3 [6] [6] [6] [5] [5] [5] [5] [3] [51 [5] [5] [5]
AAL
(F)

 

VOIC 3.3 2.0 3.0 4.0 30 40 40 5.0 2.0 40 3.0 3.0 41
E4 [5] [6] [5] [3] [5] [3] [3] [2] [6] [3] [5] [5]
SAE
(F)

 

VOIC 2.3 2.0 2.0 3.0 3.0 30 33 3.0 3.0 30 3.0 3.0 63
ES [6] [6] [6] [5] [5] [5] [5] [5] [5] [5] [5] [5]
AAL
(M)

 

VOIC 3.0 21 3.0 23 23 33 50 3.3 4.0 50 4.0 2.3 56
E6 [5] [6] [5] [6] [6] [5] [2] [5] [3] [2] [3] [6]
SAE
(M)

 

VOIC 2.0 2.0 2.0 3.3 3.0 23 40 3.0 4.0 3.0 3.0 3.0 60
E7 [6] [6] [6] [5] [5] [6] [3] [5] [3] [5] [51 [5]
AAL
(F)

 

VOIC 30 2.0 3.0 4.0 3.0 33 40 43 3.0 4.0 3.0 3.0 65
ES [5] [6] [5] [3] [5] [5] [3] [3] [3] [5] [5]
SAE
(F)

 

VOIC 5.0 40 4.0 3.0 3.3 3.3 5.0 30 50 33 3.3 3.0 45
E9 [2] [3] [3] [51 [5] [5] [2] [5] [2] [5] [5] [5]
AAL
(M)

 

VOIC 3.0 3.3 3.0 20 20 2.0 3.3 30 5.0 4.0 2.0 2.0 60
510 [5] [5] [5] [6] [6] [6] [5] [5] [3] [3] [6] [6]
SAE
(M)

 

VOIC 3.0 2.0 3.0 5.0 3.0 3.0 4.0 3.0 3.0 3.0 3.0 3.0 56

E11 [5] [6] [5] [2] [5] [5] [3] [5] [5] [5] [5] [5]
AAL

 

 

VOIC 4.0 2.0 3.3 4.3 4.3 4.0 4.0 5.0 3.0 5.0 3.0 4.3 43

512 [3] [6] [5] [3] [3] [3] [3] [2] [5] [2] [5] [3]
SAE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

65

 

In Table 5 above, the South African raters’ view of the two US English voice varieties
was most favorable toward both SAEF and AALM voices and less favorable for both
AALF and SAEM varieties. Among the sample of 12 stimulus voices, the SAEF voice
was ranked highest (41) followed by AALM (45). The other two voices, namely, the
AALF (56) and SAEM (60) followed with large numerical breaks and ranked in third and
fourth positions respectively.

This group of SA raters perceived the SAEF variety as particularly strong on being
“Patient,” but also rated it less favorably especially on sounding “Confident,” and
“Proud.” They also rated the AALM voice, which they ranked as the second highest
voice among the 12-stimulus voices, relatively less favorably on “Confident,”
“Responsible,” “Friendly,” “Polite” and “Patient.”

When it came to their attitudes toward the AALF variety, the SA judges rated the
voice less favorably on most of the traits especially on “Well Educated,” “Confident,”
and “Responsible.” They nonetheless perceived the AALF voice more favorably on
“Gentle.” The SA judges rated the SAEM voice less favorably on most traits and rated it
lowest especially on “Confident,” “Well educated,” “Responsible,” “Friendly,”
“Trustworthy” and “Social.” They were however relatively more favorable toward the
SAEM voice on sounding “Proud” and “Leader-like.”

Like both the West African and East African raters in this study, this group of SA
raters included both the twelve respondents who listened to the AAL voices first followed
by SAE voices as well as the second set of twelve raters who listened to the SAE voices
first followed by the AAL voices. It seems that among the twenty-four raters from all

three African regions generally, the rankings for both AAL voices were generally higher

66

than for the SAE voices, albeit there were high rankings for SAEF by both EA and SA
raters.

Overall, WA raters identified AALF, AALM, SAEM and SAEF in order of preference
while EA raters identified them in the following order: SAEF, AALM, AALF and
SAEM. Likewise SA raters picked the four voices in the following order: SAEF, AALM,
AALF and SAEM.

A summary of the mean ratings and rank of means by all the 24 African judges shows
that West African raters ranked the AALM and AALF voices most favorably but ranked
both SAEF and SAEM less favorably. Both East and South African judges, on the other
hand, had much more favorable attitudes toward both the SAEF and AALM voices and
were less favorable toward both AALF and SAEM voices. The reasons for the different
language attitudes by WA raters as compared to both EA and SA raters are wide-ranging.
On reason can perhaps be attributed to the lack of significant European settlement in
West Africa in comparison to that in both EA and SA.

Another viable reason is perhaps familiarity by this group of West African raters with
varieties of African American English and Krio, commonly spoken in Sierra Leone, a
Creole language derived from English and various African languages. Similar Creoles
are also spoken in the Diaspora and generally in West Africa. Additionally, the more
positive attitudes toward AAL by WA raters, though complicated by both indigenous
African language and French, the other important European language spoken in WA, can
perhaps be attributed to the regional presence of speakers of varieties of African
languages such as Mende and Temne. These two African languages, spoken in Sierra

Leone are also the base languages for Krio and for many languages spoken by enslaved

67

Africans to the Americas (Beryl Bailey 1965; Smitherman 2000). In WA also Sierra
Leone, which was first established in 1787, is home to many freed slaves from English
colonies in the Caribbean, particularly Jamaica and North America. It also includes
Liberia established in 1821 by the American Colonization Society for African American
settlers.

Both EA and SA raters ranked both SAEF and AALM voices much more favorably
than they did the AALF and SAEM voices which were ranked lowest by this group of
raters. While there is a clear distinction in the rankings of the three groups of evaluators
from the three African regions (WA, EA and SA) when their ratings are analyzed as one
data group, their most favorable and highest rankings are for the AAL voices in which the
AALM voice earns the most favorable ranking followed by the AALF voice. While as a
group all twenty-four African raters ranked the SAEF voice in third place, they are
particularly less favorable in their ranking of the SAEM voice, which earns the least
favorable rankings of all four voices.

Tables 2 through Table 5 present data for the mean ratings and rank of mean rankings
for all the three African regions combined and then for the individual regions (WA, EA
and SA). In the data in Tables 2 through Table 5, we saw higher rankings for both of the
AAL voices, with no significance for order effect.

There were no significant differences based on age and the numbers of years spent in
the US in the way the twenty-four raters evaluated the twelve stimulus voices. However,
based on their African region origin, many of the twenty-four raters ranked the SAEM
voice as the least favorable. This was with the exception of the WA raters, who rank the

SAEM voice third while ranking the SAEF variety the least favorable voice in fourth

68

position. In the majority of the cases, however, the SAEM voice is perceived as the least

favorable of all four voices.

TABLE 6: MEAN RATINGS AND RANK OF MEANS OF AAL AND SAE BY

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

GENDER BY ALL RATERS
Traits
EDI CON RES GEN FRI POL LEA PAT PRO STA TRU SOC Sum
F P E I D

VOICE 13:4. = = = = F= 1:: = = = = = F
“ML 0 3.1 3.4 3.0 3.2 3.3 4.1 3.0 3.0 4.3 3.1 3.0 =53
(M) 131 151 141 151 151 151 131 151 151 131 151 151 M=6

M: 3 M= M= M= M= M= M= M= M= M= M= M= l

.3 2.2 3.0 3.2 3.0 3.0 3.3 3.0 3.0 3.3 3.3 3.0 =

151 161 151 151 151 151 151 151 151 151 151 151 114
vorcra =2. = = = = F= F= = = = = = F=6l
2 3 2.2 3.1 2.0 2.4 3.3 4.0 2.2 4.0 3.3 3.0 22 M=6
33:3 161 161 151 161 151 151 131 161 131= 151 151 161 3

M: 2 M: M: M: M: M: M: M= M= M= M= =

.3 2.0 2.4 2.4 2.2 3.0 3.0 3.0 3.0 2.4 3.0 3.0 124

161 161 151 15__1 161 151 151 151 15__1 151 151 151
VOICE F=3. F= F=3. F= F= F= F= F= F= F= F=S4
3 0 3.0 2 3.4 3.4 4.1 3.0 4.0 2.4 30 3.2 3.1 M=
3:)“ 151 151 151 141 141 131 151 131 151 151 151 151 6

M=3 M= M: M: M= M= M= M= M= M= M= M= =

.0 3.0 3.0 3.2 4.0 3.0 3.1 4.0 3.0 3.2 3.0 4.0 110

151 15_1 15:1 15__1 151 151 151 131 151 15:1 151 131
VOICE F=3. F: F: F= F= F= F= F= F=47
4 1 3.1 3.4 4.0 4.0 4.0 3.0 4.0 3.0 4.2 3.0 4.0 M=6
SAE 15] 151 141 131 131 131 151 131 151 131 151 131 2
(F) M: 3 M: M: M= M: M: M: M: M: M= M= M:

.0 2.0 3.0 3.1 3.0 3.1 3.1 3.3 2.2 3.0 3.0 3.0 =

151 16__1 15:1 15__1 15_1 151 151 15_1 16:1 15:1 51 [SL 109
VOICE F: 3 F: F: = = F=55
5 0 2.3 3.0 3.0 3.0 3.0 4.0 3.2 3.3 4.0 4.0 3.0 M=5
3,11} 151 161 151 151 151 151 131 151 151 131 131 151 9

M= 3 M= M= M= M= M= M= M= M= M= M= M=

.0 3.0 3.0 3.0 3.0 3.0 3.4 3.1 3.0 3.3 3.0 3.0 =

151 151 15__1 151 151 151 141 151 151 15_1 15:1 151 114
VOICE F=2. F=2. 0 F: F= F= F: F: F= M= F=68
6 3 [6] 20 2.3 2.3 2.1 3.0 2.0 3.0 3.1 3.0 2.0 M=6
SAE 16] M= 161 161 161 161 151 161 151 151 151 161 4
(M) M= 2 2. 0 M= M= M= M= M= M= M= M= M= M=

.0 [6] 2.0 2.0 3.0 3.1 3.2 3.0 3.0 3.2 3.0 3.0 =

161 16:1 1F6__1 15__1 151 151 151 15:1 15__1 15_1 15___1 132
VOICE F=3. = F= M: F: [3:52
7 0 2.0 3.0 40 4.0 4.0 3.4 4.1 3.0 3.3 3.1 4.0 M=6
3,)“ 151 161_ 151 131 131 131 141 151 151 151 151 131 4

M: 2 M: M: M: M: M: M: M: 3. M: M: M:

.0 2. 0 2.1 3. 0 2.1 3.4 3.0 3.0 0 3.0 3.0 21 =

161 161 161 151 161 141 151 151 151 151 151 161 116

 

69

 

Table 6 (Continued)

 

VOICE =2. = = = F= F: F: F= = = = = F=50
8 4 2.1 3 0 4.0 4.0 3.4 3.0 4.0 3.3 4.0 3.0 4.0 M=
SAE [5] [6] [5] [3] [3] [4] [5] [3] [5] [3] [5] [3] 8

(F) M= 3 M= M= M= M= M= M= M= M= M= M= M=

.3 2.4 3.0 3.1 3.0 3.0 3.0 4.0 3.0 3.1 3.3 3.0 =
[5] [51 15] [5:] [5] [5] [51 [3] [5:] [F53 15] [5:] 108

 

VOICE F=4. F: F: F= F= F= F= F= F=46
9 4 3. 2 3. 0 3. 3 3.0 3.0 5.0 4.0 3.1 5.2 4. 0 3. 4 M=4
ML 121 151 151 151 151 151 121 131 151 121 131 141 s
(M) M=4 M= M= M= M= M= M= M= M= M= M= M= T=94
.0 3. 0 4.0 2.4 3.0 3.2 4.0 3.3 4.0 4.0 4.0 3.1
131 151 131 151 151 151 131 151 131 131 131 15:1

 

v01c F=3. F= F= F: F: F= F= F= F= F= F= =53
1310 0 3. 0 3. 0 3. 0 3.0 3.0 5.0 3.2 4. 0 4. 2 3. 3 31 M=5
SAE [51 [5] [5] [5] [5] [5] [2] [5] [3] [3] [5] [5] 6

(M) M=3 M= M= M= M= M= M= M= M= M= M= M=
.0 3.0 3.0 3.0 3.1 3.0 4.0 3.0 3.4 3.4 3.0 3.0 =
[5] [5] 15:] [5] [5] [5] [3] [5] [4] [4] 15:] [5:] 109

 

vorc F=3. F= F= F= F= F= F= F= F= F=46
1311 1 24 3.1 4.1 4.0 4.0 3.3 4.2 3.2 4.2 4.1 4.0 M=6
AAL [5] [5] [5] [3] [3] [3] [5] [3] [5] [3] [3] [3] 1

(F) M= M= M= M= M=2 M= M= M= M= M= M= M=
3.0 2.0 3.0 3.3 .4 3.0 3.3 3.0 3.0 30 3.0 2.4 =
[5] [6] [5:1 [5] [5] [5] [5] [5] [5] [5] [5] 15_1 107

 

 

VOIC F=4. F= F= F= F= F= F: F: F= F=45
1312 0 2.2 30 40 4.1 4.0 3.0 4.1 3.4 44 3.3 4.1 M=5
SAE [31 [6] [5] [3] [3] [5] [3] [3] [4] [2] [5] [3] 2

(F) M= M= M= M= M= M= M= M= M= M= M= M=
3.4 2.3 3.0 3.4 4.0 3.1 3.4 4.0 3.0 4.0 3.0 34 T=97
[4] [6] [51 [4] [3] [5] [4] [3] [5] [3] [5] [5]

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 6 presents the mean ratings and rank of mean rankings by gender, which was not
statistically significant. However, the trend results are interesting in that, the female
raters first ranked the AALM voice more favorably (46) than did their male counterparts
(48). Females also tended to rank the SAEF higher (45) than did the male raters (52).

The AALF voice tended to be ranked third by both genders with the female evaluators
first ranking the voice significantly higher (46) than the males (61). The SAEM voice
was ranked as the least favorable with the female raters ranking the SAEM voice initially

at number five with lower scores (68) than the male raters (64). Female raters in

70

 

particular viewed the AALM most favorably on “Leader-like” and “High Status,” and the
male evaluators rated the voice least favorably on “Confident.”

When it came to the SAEF voice, which was ranked as the second most favorable
voice, the female raters viewed the voice least favorably on “Confident” and “Proud.”
The male raters also judged the SAEF unfavorably on one of the same traits (i.e.
“Confident”), along with “Well Educated” and “Responsible.” The male raters also
viewed AALF least favorably on “Confident,” along with “Well Educated,
“Responsible,” “Friendly,” and “Social.” The female raters, on the other hand, rated the
AALF voice least favorably on “Confidence” just as they did with the SAEF voice. The
SAEM voice, which was ranked lowest by all raters, was ranked relatively more
favorably by the female raters on “Patient” and “Leader-like” with the male raters
ranking the same voice favorably on “Patient,” “Proud,” “High Status” and “Leader-
like.” These are interesting trends, but the gender differences were not statistically
significant.

Both AAL voices seem to be rated much more favorably and much more often
throughout this analysis. However, the often interchanging ranking positions of SAEM
and SAEF and the need to continue to unravel what we consider to be deeply held
language attitudes by the African raters regarding both AAL and SAE, make it necessary

to conduct further analysis.

71

TABLE 7: RANKINGS OF AAL & SAE SPEAKERS BY ALL RATERS

 

Region

AAL (M)

AAL (F)

SAE (M)

SAE (F)

RANKING
ORDER

 

WA

2

1

3

4

1=AAL (F)
2=AAL (M)
3=SAE (M)
4=SAE (F)

 

EA

1=SAE (F)
2=AAL (M)
3=AAL (F)

4=SAE (M)

 

SA

1=SAE (F)
2=AAL (M)
3=AAL (F)
4=SAE (M)

 

 

Sum

 

 

 

ll

 

 

 

Key (for Tables 7-9):

AAL=AFRICAN AMERICAN LANGUAGE
SAE=STANDARD AMERICAN ENGLISH
M= MALE

F= FEMALE
WA=WEST AFRICA

EA-EAST AFRICA

SA=SOUTH AFRICA

TABLE 8: RANKINGS OF AAL SPEAKERS BY ALL RATERS

 

 

 

 

 

 

Reflm AAL (M) AAL (F) RANKING
WA 2 1 l= AAL (F)
2=AAL (M)
EA 2 3 2=AAL (M)
3=AAL (F)
SA 2 3 2=AAL (M)
3=AAL (F)
Sum 6 7

 

 

 

 

 

72

 

TABLE 9: RANKINGS OF SAE SPEAKERS BY ALL RATERS

 

 

 

 

 

 

 

 

 

Region SAE (M) SAE (F) RANKING
WA 3 4 3= SAE (M)
4: SAE (F)
EA 4 1 1= SAE (F)
4= SAE (M)
SA 4 l 1= SAE (F)
4: SAE (M)

Sum 1 1 6

 

 

Figure 2a: Rankings of AAL & SAE

 

 

 

73

 

 

IWA
I EA
c1 SA

 

 

 

@AALM
.AALF
D SAEM
D SAEF

 

 

 

In summary, the Mean Ratings and Rank of Means analysis presented in Tables 2
through 9 and Figures 23 and 2b demonstrate that, the raters from all the three African
regions (WA, EA and SA) are clearly able to differentiate and to make the necessary
distinctions between speakers of AAL and SAE upon hearing these two different varieties
of English spoken in the US. The AAL variety evidently gets much more favorable
rankings in general than the SAE variety despite individual preferences, as seen with high
ratings of SAE variety by some of the raters. In Table 3, for example, the WA raters
ranked both AALM and AALF varieties highest overall with the SAEF earning the
lowest ranking. However, both the EA and SA raters seem to change this ranking trend
somewhat, because while ranking the SAEF voice the highest followed closely by both
AALM and AALF voices, they rated the SAEM voice as the least favorable.

Overall, the AAL variety, regardless of whether it is male or female, is more often
associated with the more positive attributes and traits of the Semantic Differential Scale
while the SAE variety, regardless of whether it is male or female, is more often
associated with the less favorable and negative attributes and traits in this matched-guise
study of twelve anonymous voices rated by twenty-four African immigrants from West,
East and South Africa. Attitudes toward the two varieties of US English, i.e. AAL and
SAE seem to vary considerably among the raters even though the rating of means and
ranks of mean rankings helps us to resolve this variation somewhat. There is nonetheless
the need now to conduct further analysis to help us evaluate the various distinctions
revealed by the Means Rating and Ranking.

After we conducted an analysis of the raters based on their region of origin in Africa

(WA, EA and SA) as well as based on their gender, age and years spent in the USA, we

74

can perhaps conclude and confirm safely at least one of the hypotheses in this study that,
respondents will, based primarily on African region, respond differently to both AAL and
SAE, and their attitudes toward the two US English varieties will be both diverse and
varied.

There is nevertheless the need to evaluate collected data further in order to help
confirm the following other hypotheses in this study: (a) Respondents’ attitudes will be
influenced by readings of the passages selected for this study (b) Respondents’ attitudes
will either be un/favorable attitudes toward AAL and SAE depending on their familiarity
with the variety/ies; (c) Respondents’ attitudes toward AAL and SAE will be influenced
by both their length of time in the US and their association with speakers of the two
language varieties.

CORRELATION COEFFICIENT ANALYSIS

A Correlation Coefficient helps indicate both the strength and direction of the
relationship between a pair of variables. Bryman and Cramer (2005, 213-214) suggest
that, two types of measures can be distinguished: measures of linear correlation using
interval variables and measures of rank correlation using ordinal variables. In this study,
the correlation coefficient is used to measure the linear relationship between at least any
two variables. In this case these two variables will be two semantic differential scale
adjectives.

When variables are interval/ratio, by far the most common measure of correlation is
Pearson ’5 Product Moment Correlation Coefficient often referred to as Pearson’s r. “This
“measure of correlation presumes that interval variables are being used, so that even

ordinal variables are not supposed to be employed, although this is a matter of debate”

75

(Bryman and Cramer 2005, 214). In this manner a correlation coefficient measure help to
provide a yardstick whereby the intensity or strength of a relationship between two
adjectives on the Semantic Differential scale was gauged. The measure also provided us
with clearer assessments of the closeness of the relationship among pairs of variables. In
our case this was the relationship among our list of twelve traits or adjectives of the
semantic differential scale.

What all this means is that a high correlation coefficient score, such as .62 indicates a
relatively strong linear relationship between the column variable (i.e. language trait) and
the corresponding row variable (another language trait). On the other hand, a smaller
measure or number indicates a weak relationship between semantic differential
adjectives, which is usually closer to zero.

Tables 10- 12 help illustrate the Correlation Coefficient Matrices with each entry in
the tables representing the correlation coefficient (i.e. how the 12 language features
correlate and cluster together in any significant way between the row variables (language
traits) and the column variables (measures of the same language traits when correlated).
Given the symmetric nature of the matrices in the three tables, we only represent the
lower triangular part of the matrix. This is because it is always the case that in a matrix
of this kind the correlation between the same variables, such as seen with the column
variable against the row variable is always equal to 1.00. It therefore makes sense for us
to omit that diagonal representation in the matrix.

Consequently, the matrices’ representations in Tables 10-12 helped us to determine
which of the twelve traits of the semantic differential scale, which, we used in this study

can logically be clustered together for appropriate interpretation and analysis. The

76

guidelines we employed to determine the strength of the linear relationship implied by
the strength of the correlation of the variables in our study was in effect their level of
significance. The strength of the correlation coefficient in the three matrix tables is
shown with a single * to help indicate the level of significance of .05 and a double ** to
help indicate the correlation coefficient is significantly different from zero at a
significance level of .01. Thus by using the correlation coefficient analysis, we
determined the trends and relationships in the various categories based on trait-by-trait
analysis.

The factor analysis that follows this section and which will form a major part of our
analysis will also show these relationships from a broader viewpoint. The correlation
coefficient analysis in this section will occupy itself in identifying the strongest
correlation sets and trends in order to show the linear relationships among the twelve
language traits or adjectives used for rating both AAL and SAE in this study.

Tables 10-12 also help to portray the pattern of the broader perceptions on the
language attitudes toward AAL and SAE by the 24 African raters using the correlation
coefficient analysis. These patterns are:

1) WEST AFRICAN RATERS: In Table 10, the cluster sets among the WA raters
indicate that two language traits namely, “Proud” and “High status,” are often the only
two traits characterized by only one positively significant correlation each. “Proud” has a
positively significant correlation only with “Social,” (.59*). “High status” has one
positively significant correlation with “Trustworthy.” It also has a significant but

. negative correlation with “Proud.”

77

2) EAST AFRICAN RATERS: In Table 11, there are at least three emerging cluster
patterns among East African raters: a) a group of clusters that contain at least one or both
of the following two traits: “Trustworthy” and “Proud.” b) The second group contains at
least one or both of the following traits: “High status” and “Social.” c) The third group is
a complement of the first two and can be described as a group of clusters that do not
contain any of the four language traits.

3) SOUTH AFRICAN RATERS: In Table 12, the correlation matrix of the South
African raters indicates a large group of clusters that share often negatively correlated
membership. The commonality of the group seems to be derived from the many negative
correlations that all the 12 traits have in common with each other and most important for
us, so few positively significant correlations. The only five positive correlation clusters
among the SA raters are: “Well Educated” and “Confident”; “Well Educated” and
“Proud”; “Trustworthy” and “Proud”; “Social” and “Proud” as well as both “Social” and

“High status.”

78

TABLE 10: CORRELATION COEFFICIENTS OF 12 SEMANTIC DIFFERENTIAL SCALES FOR
WEST AFRICAN RATERS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ED CON RESP GEN FRIErraI:(t)SLI LEA PAT PRO STA TRU SOC
F D

ED
CONF .80*
RESP .14 .35*
GEN .20 .31* .45*
FRIE .22 .38* .43* .56*
POLl .40* .41* .57“ .67* .70*
LEAD .30* .35"' .45* .62M .68“ .81"
PAT .32* .31 .50* .63* .62* .75* .63*
PRO - -.41 -.40 -.10 .08 -.08 -.00 .00

.32*
STA -.27 .22 -.26 -.26 -.10 .03 .26 .06 -

.35*
TRU -.08 .13 .14 .27 .39* .53* .47* .67“ .27 .69*
SOC .22 .38" .43“ .56** .63** .76** .66* .82* .59* .10 .62*
:1-

 

 

 

 

 

 

 

 

 

 

 

 

 

* = Significance .05

**= significance .01

(2-tailed)

Table 10 above presents the results of the evaluation of the 12 language traits that

represent the attitudes of the eight West African raters on how they perceive the 12

stimulus voices (divided into 6 AAL and 6 SAE voices). The data results indicate a

pattern that reflects the following cluster combinations. Overall then, the language trait

clusters among the West African raters beginning with for example “Well Educated,”

Show that it correlates significantly with “Confident,” “Polite” “Leader-like” and

79

 

“Patient” (Cluster set 1) which is a relatively small set. It is also significantly negatively

correlated to “Proud.” The other cluster sets are as follows:

 

(Cluster set 2): “Confident”: correlates with being “Well Educated,” Responsible,”
“Gentle,” “Friendly,” “Polite,” “Leader-like,” “Patient,” and “Social.” This set of
correlations is not entirely similar to the first set though it Shares at least three common
traits (i.e. “Well Educated,” “Leader-like” and “Patient”). It also Shares a significant but

negative correlation with “Proud.”

 

(Cluster set 3): “Responsible”: correlates with “Confident,” “Gentle,” “Friendly,”
“Polite,” “Leader-like,” “Patient,” and “Social.” Like the first two sets it also has a
significant but negative correlation with “Proud.”

(Cluster set 4): “Gentle”: correlates with “Confident,” “Friendly,” “Polite,” “Leader-

 

like,” “Patient” and “Social.” It also shares a Significant but negative correlation with
“Proud.” (Cluster set 5): “Friendly”: correlates with “Confident,” “Responsible,”
“Gentle,” “Polite,” “Leader-like,” “Patient,” “Trustworthy,” and “Social.” It compares
favorably with the preceding sets of clusters by not sharing either a significantly positive
or negative significant correlation with “Proud.”

(Cluster set 6): “Polite”: correlates with “Well Educated,” “Confident,”
“Responsible,” “Gentle,” “Friendly,” “Patient,” “Leader-like,” “Patient” “Trustworthy,”
and “Social.” It also bears neither negative nor positive and Significant correlations with
“Proud.”

(Cluster set 7): “Leader-like”: correlates with “Well Educated,” “Confident,”

 

“Responsible,” “Gentle,” “”Friendly,” “”Polite,” Patient,” “Trustworthy,” and “Social.” It

also has no (i.e. -.OO) correlation whatsoever with “Proud.”

80

(Cluster set 8): “Patient”: correlates with “Well Educated,” “Confident,”

 

“Responsible,” “Gentle,” “Friendly,” “Polite,” “Leader-like,” “Trustworthy” and
“Social.” It bears no significant correlation (positive or negative) with “Proud”

(Cluster set 9): “Proud”: is the only feature that doesn’t correlate significantly and
positively with any other trait. It however correlates significantly, albeit negatively, with
both “Confident” and “Social” and has in particular no significant correlations which are

either positive or negative with “Leader-like” (i.e. -.00) and “Patient” (i.e. .00).

 

(Cluster set 10): “High status”: correlates significantly only with, “Trustworthy” and

negatively so and like all the preceding sets, has a significant but negative correlation

 

with “Proud.” Cluster set 1 l): “Trustworthy”: correlates with “Friendly,” “Polite,”
“Leader-like,” “Patient,” “High status” and “Social.” Like all the other sets identified so
far, it also has neither a significant and positive nor a negative correlation with “Proud.”

(Cluster set 12): “Social”: correlates with “Confident,” “Responsible,” “Gentle,”

 

“Friendly,” ‘Polite,” “Leader-like,” “Patient,” “Proud” and “Trustworthy.” It is
noteworthy that among all the cluster combinations that precede it, “Social” is the only
language trait that has a positively significant correlation with the trait “Proud” (*.59).
The cluster sets among the WA raters indicate that two language traits namely, “Proud”
and “High status,” are often the only two traits characterized by only one positively
significant correlation each. “Proud” has a positively significant correlation only with
“Social,” (.59*). “High status” has one positively significant correlation with
“Trustworthy” (62*). It also has a significant but negative correlation with “Proud” (-

35*).

81

TABLE 11: CORRELATION COEFFICIENTS OF 12 SEMANTIC DIFFERENTIAL SCALES FOR

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EAST AFRICAN RATERS
Traits
ED CONF RESP GEN FRIE POL LEA PAT PRO STA TRU SOC
ED I D
CONF .30*
RESP .42* -.031
GEN -.15 .06 -.37
FRIE -.50 -.18 -.42 -.04 .02
POLI -.40 -.20 -.42 -.04 .02
LEAD -.33 .19 - .15 .13 -.04
.66**
PAT -.18 .09 -.22 .15 .25 .18 .33*
PRO -.21 .50* -.24 .44* .66* .44* .52* .50*
STA .01 .14 - .23 .06 .15 .40* .33* .00
.46**
TRU .49* .26 .00 .44* .18 .47* 56* .40* 33* .00
I.
SOC -.46 -.14 -.44 -.02 .07 -.10 .21 -.08 .21 .53* -.50

 

 

 

 

 

 

 

 

 

 

 

 

 

* = Significance .05

**= Significance .01

(2-tailed)

Table 11 presents an overview of all the clusters sets among the East African raters and

helps to indicate at least three emerging patterns in terms of their language attitudes

toward both AAL and SAE: a group of clusters that contain at least one or both of the

following two traits: “Trustworthy” and “Proud.” The second group contains at least one

or both of the following traits: “High status” and “Social.” The third group is a

82

 

complement of the first two and can be described as a group of clusters that do not
contain any of the four language traits.

In the first cluster grog, which also belongs to the second cluster group, we have a

 

cluster such as “Well Educated” which correlates significantly with “Confident,”
“Responsible” and “Trustworthy.” This cluster also includes “Gentle,” “Friendly,”
“Polite,” “Leader-like,” and “Patient” as its significantly correlated members. The cluster
also includes “High status” which significantly correlates with “Leader-like” and
“Social”.” In addition “Friendly” correlates significantly with “Proud” while “Polite,”
which can also be included in both the first and second cluster of combination adds,
“Proud” and “Trustworthy” to its list of combinations.

The second cluster group, which also belongs to the first cluster, includes, “Leader-

 

like” and is also made up of “Proud,” “High status” and “Trustworthy.” This second
cluster also includes “Patient” which correlates significantly with “Proud” and
“Trustworthy.” The final set of combination of language traits in this second cluster also
includes, “High status” which correlates significantly with “Social.”

The third cluster group is noteworthy as it includes “Gentle” with significant
correlation with both “Proud,” “Trustworthy” but also with no significant correlations
with both “High status” (.23) and “Social” (-.02) as well as “Friendly” (-.00)

“High status”: correlates with “Leader-like” and “Social.” But the trait does not
correlate significantly with both “Proud” (.27) and “Trustworthy” (.00). It has a

particularly significant but negative correlation with “Responsible” (-.46*).

83

“Trustworthy”: correlates significantly with “Well Educated,” “Gentle,” “Polite,”
“Leader-like,” “Patient,” and “Proud.” It has a Si gnificant but negative correlation with

“Social” (-.50*) and has also none with “High status” (.00)

The following are the cluster sets and combinations for the East African raters:

 

(Cluster set 1): “Well Educated”: correlates significantly with “Confident”

 

“Responsible,” and “Trustworthy.”

(Cluster set 2): “Confident”: correlates significantly with “Well Educated” and “Proud.”

 

(Cluster set 3): “Responsible”: correlates significantly with “Well Educated,” and
“Confident.”

(Cluster set 4): “Gentle”: correlates significantly with “Proud” and “Trustworthy.”
(Cluster set 5): “Friendly”: correlates significantly with “Proud.”

(Cluster set 6): “Polite”: correlates significantly with “Proud” and “Trustworthy.”

 

(Cluster set 7): “Leader-like”: correlates significantly with “Patient,” “Proud,” “High

 

status” and “Trustworthy.”
(Cluster set 8): “Patient”: correlates significantly with “Proud” and “Trustworthy.”

(Cluster set 9): “Proud”: correlates significantly with “Confident,” “Gentle,” “Friendly,”

(Cluster set 10): “Polite,” “Leader-like,” “Patient,” and “Leader-like,” “Proud” and

 

Trustworthy.”

(Cluster set 1 1): “Trustworthy”: correlates significantly with “Well Educated,” “Gentle,”

 

“Polite,” “Leader-like,” “Patient,” and “Proud.” It is also significantly but negatively

correlated with “Social.”

84

(Cluster set 12): “Social”: correlates significantly with “High status” and is also

significantly correlated though negatively with “”Well Educated,” “Responsible,” “High

status,” and “Trustworthy.”

TABLE 12: CORRELATION COEFFICIENTS OF 12 SEMANTIC DIFFERENTIAL SCALES FOR

SOUTH AFRICAN RATERS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ED CON RESP GEN FRIE 1‘leng LEA PAT PRO STA TRU SOC
F D

ED

CONF .75*

RESP -.50 .19

GEN -.30 -.21 -.69

FRIE -.17 .00 -.54 -.67

POLI -.39 .26 .05 -.42 .21

LEAD -.33 .00 -.4O -.67 -.30 -.55

PAT -.02 .42* -.14 -.62 .14 -.70 -.48

PRO .41"‘ .51* .28 -.18 -.20 -.27 .02 -.14

STA .13 .29 -.17 -.64** -.29 -.63** -.45 -.37 .15

TRU -.48 .017 -.12 -.39 -.02 -.55 -.21 -.09 .52* -.12

SOC -.40 -.02 -.50 -.74** -.I7 -.77** - -.45 .61* .48* .02
.62*
4:

 

 

 

 

 

 

 

 

 

 

 

 

 

* = Significance .05

**= Significance .01

(2-tailed)

According to Table 12, among the South African raters of both AAL and SAE, the

following are the significant correlations in their perceptions of the two US English

varieties using the 12 stimulus voices in this study which, they evaluated on each of the

12 language traits of the semantic differential scale namely, Well Educated”: correlated

85

 

significantly with both “Confident” and “Proud” (Cluster set 1). The language trait,
“Well Educated” is however also significantly correlated but negatively so with
“Responsible,” “Gentle,” “Polite,” “Leader-like,” “Trustworthy” and “Social.”

(Cluster set 2): “Confident”: correlated significantly with “Well Educated,” “Proud,”
“Patient,” and “Proud.”

(Cluster set 3): “Responsible: correlated significantly but negatively with “Well
Educated,” Friendly,” “Leader-like” and “Social.”

(Cluster set 4): “Gentle”: correlated significantly but negatively with both “Well
Educated,” “Responsible,” “Friendly,” “Polite,” “Leader-like,” “Patient,” “High status”
and “Social.”

(Cluster set 5): “Polite”: correlated significantly but negatively with “Well Educated,”

 

“Leader-like,” “Patient,” “High status,” “Trustworthy” and “Social.”

(Cluster set 6): “Leader-like”: correlated significantly but negatively with “Well
Educated,” Patient,” “High status,” and “Social.”

(Cluster set 7): “Patient”: correlated significantly with “Confident,” but negatively so
with “Gentle,” “Polite,” and “Leader-like.”

(Cluster set 8): “Proud”: correlated significantly with both “Well Educated,” and
“Confident.”

(Cluster set 9): “High status”: correlated significantly but negatively with both “Gentle,”
“Polite,” “Leader-like,” and “Patient.”

(Cluster set 10): “Trustworthy”: correlated both positively and significantly m with

“Proud” but negatively so with “Well Educated,” “Gentle,” and “Polite.”

86‘

 

(Cluster set 11): “Social”: correlated significantly with both “Proud” and “High status,”
but negatively so with “Well Educated,” “Responsible,” “Gentle,” “Polite,” “Leader-like”
and “Patient.” The correlation matrix of the South African raters, indicate a large group
of clusters that share interesting and often negatively correlated membership. The
commonality of the group seems to be derived from the many negative correlations that
all the 12 traits have in common with each other and most important for us, so few
positively significant correlations.

The only five positive correlation clusters among the SA raters are: “Well Educated”
and “Confident”; “Well Educated” and “Proud”; “Trustworthy” and “Proud”; “Social”
and “Proud” as well as both “Social” and “High status.”

We acknowledge that Tables 2 —- 6 of the mean ratings and rank of means present only
a general overview of both the rank of the two US English varieties based on the 12
stimulus voices which were rated on each of the 12 adjectives or traits used in their
evaluation in relationship to one another. We however contend that the correlation
coefficient matrices in the Tables 10-12 on the other hand, help to portray the pattern of
the broader perceptions on the language attitudes toward both AAL and SAE evaluated
by all 24 African raters from West Africa, East Africa and South Africa.

The three correlation coefficient matrices have also helped indicate to us that the 24
evaluators/raters from the three Afiican regions have different ratings for evaluating the
two US English language varieties. The matrices also provide us with additionally useful
insights into how the evaluators/raters perceive both African American Language (AAL)
and standard American English (SAE). As mentioned hereto, different historical events

in WA, EA and SA, such as significant European settlements in both EA and SA, the

87

establishment of freed slave colonies of Sierra Leone and Liberia in WA, along with
pertinent social and linguistic realities with attendant social and cultural discourses, may
have accounted for the differences in perceptions and attitudes toward both AAL and
SAE.

FACTOR ANALYSIS

Factor analysis is a statistical data reduction technique used to explain variability
among observed random variables in terms of fewer unobserved random variables called
factors. The observed variables are modeled as linear combinations of the factors, plus
“error” terms. The factor analysis helps in dimension reduction as well. This means it
helps to ensure that each member of a factor group identified belongs fully to that
assigned group. As seen with the correlation coefficient matrices in the previous chapter,
the statistical phenomena that resulted in the clustering of variables around a common
theme was typified by both the overlapping and combination of those features which had
both weak and strong correlations. In factor analysis we used the Varimax rotation
technique available in SPSS to help us rectify this anomaly of overlapping and
combination of for example, a variable with two disparate correlations.

The Varimax rotation technique also helped us to ensure that there are no overlaps and
that the correlation between one variable and a factor that it belongs to is maximized fully
while at the same being minimized in the factor (5) in which it doesn’t belong to. As a
result, the factor group sets have been refined using the Varimax rotation technique.

According to Bryman and Cramer (2005),

The scores of people on three or more variables primarily concern factor analysis
with describing the variation or variance, which is Shared. This variance is

referred to as common variance and needs to be distinguished from other kinds of
variance. Specific variance describes the variation which is specific or unique to a

88

variable and which is not shared with any other variable. Error variance, on the
other hand, is variation due to the fluctuations, which inevitably result from
measuring something. If for example you weigh yourself a number of times in
quick succession, you will find that the readings will vary somewhat, despite the
fact that your weight could not have changed in so short a time. These
fluctuations in measurement are known as error variance. So the total variation
that we find in the scores of an instrument (such as an item or test), to assess a
particular variable can be divided or partitioned into common, specific and error
variance. Total variance = Common variance + specific variance + Error variance.
Since factor analysis cannot distinguish specific from error variance, they are
combined to form unique variance. In other words, the total variance of a test
consists of its common and unique variance. (327)

Using Factor Analysis, we have re-computed scores from all twenty-four raters from
each of the three African regions (WA, EA and SA) and also employed the technique
using the principal component extraction feature of the technique along with the Varimax
rotation technique. We were thus able to establish that the evaluators from the three
African regions belong to different factor groups, which in our view provide unique
insights in this study. The purpose of Factor Analysis is to discover simple patterns in the
nature of relationships among variables. In particular, it seeks to discover if the observed
variables can be explained largely or entirely in terms of a much smaller number of
variables called factors.

Unlike many statistical methods used to study the relationship between independent
and dependent variables, Factor Analysis studies the patterns of relationship among many
dependent variables, with the goal of discovering something about the nature of the
independent variables that affect them, even though those independent variables were not
measured directly. We tried to maximize the value of Factor Analysis in this study and

consider the following results a departure from previous language attitude studies given

the unique nature of the present investigation. In conducting Factor Analysis in this study

89

we considered a factor as belonging to its loading category if it was 0.80 or higher. These
loadings are marked with an *. Loading indicators in this study are higher than the .30
factor loading often considered the benchmark. The following tables present the factor
groups of each of the three African regions (WA, SA and SA).

FACTOR LOADING FOR WEST AFRICAN RATERS

TABLE 132 FACTOR LOADING FOR WEST AFRICAN RATERS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Trait Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 Factor 7
High Status *.95 .40 .51 .60 .69 -.48 .59
Leader-like *.94 .46 .40 *.78 .68 .47

Polite *.92 .64 .68 *.74 .42 .47

Proud *.91 .58 *.89 .51 .65 .55

Trustworthy *.89 *.84 .65 .37 .49 .47

Friendly *.88 -.68 .63 .57 .30 .47 .59
Social *.88 -.68 .63 .57 .30 .47 .59
Confident -.70 .72 .66 *.75 -.48 .32 .35
Gentle *.84 *.83 *.70 .55 .38 .35 .17
Patient *.82 .40 .49 -.50 .53 .18 -.45
Responsible -.49 .53 -.37 *.79 .33 .38 -.43
Well -.60 *.84 *.79 .61 *.71 .41
Educated

 

 

 

 

 

 

 

 

 

* Indicates a Significantly High Trait and Score

Table 13 helps to interpret the scores into features and an attempt has been made to
provide an overall identification label for the factor group. Efforts have also been made
to uncover the best representative label for each of the seven factor groups identified for

all the 24 raters in all the three African regions (WA, EA and SA). This means that the

90

highest scores in each set that have been indicated with an asterisk (*) are both viewed

and interpreted as of higher value and meaning in this exercise.

TABLE 14: INTERPRETATION OF FACTOR LOADINGS FOR THE WEST AFRICAN RATERS

 

 

 

 

 

 

 

 

 

 

 

 

IN TABLE 13
Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 Factor 7
*High Status *Trustworthy *Well *Responsible *Well- Proud Friendly
Educated educated
*Leader-like *Well- *Proud *Leader-like High Status Friendly Social
educated
*Polite *Patient *Trustworth *Confident Leader-like Social High
Y

Status

*Proud Leader-like Gentle *Polite Proud Polite Not
Leader-like

*Not Friendly Polite High Status Responsible Not High Not

Confident Status Responsibl
e.

Not Not Confident Confident Not Patient Not Confident

Responsible

Not Well Not Social Not

Educated Responsible

Friendly

Trustworthy

EGOCENT PATRONIZI CHARMIN PROUD DEPENDAB SOCIABL HUMBLE

RIC NG G LE E

 

 

 

 

 

 

 

* Indicates a Significantly High Trait and Score

As seen in Table 14 above, the mean scores of the features that have significant

loading on each factor are calculated using the initial mean scores as shown earlier in

Tables 2-4. In Table 13 for West African raters we obtained the mean score for Factor 1

by calculating the mean of the mean scores for nine traits, namely, “high status,” “leader-

like,” “polite,

99 6‘

proud,” “confident,

,9 6‘

responsible,” “educated,” “friendly,” and

“trustworthy,” as now shown in Table 14 for the West African raters.

91

 

As a result the new mean of the mean scores constitute the Factor 1 mean. Similar
calculations are done for the other six Factors identified in this Factor Analysis, while at
the same time keeping our overall focus on analyzing the two US English varieties (i.e.
AAL and SAE). We also paid special attention and focus on language attitudes toward
AAL by these African immigrants from West, East and South Africa.

In order to determine the feature, which we refer here to as “EGOCENTRIC,” for both
AAL and SAE varieties, we calculated the mean of the mean scores for each of the nine
traits that loaded more significantly into Factor I mentioned hereto, for AAL variety first
followed by SAE variety. We then followed this procedure by calculating the means for
the rating scores generated from the three African regions (WA, EA and SA). The results
from this latter computation are presented as Tables 14 and 15 (WA), Tables 16 and 17
(EA) and Tables 18 and 19 (SA) in which we have tried to explain language attitudes
toward AAL and SAE by the 24 raters in terms of Afiican regions.

Additionally, Figures 3a (Egocentric & Charming), 3b (Patronizing & Dependable) and
3c (Proud & Egocentric) for WA raters, Figures 4a (Egocentric & Charming), 4b
(Patronizing & Dependable) and 4c (Proud & Sociable) for EA raters and Figures Sa
(Egocentric & Charming), 5b (Patronizing & Dependable) and 5c (Dependable &
Humble) for SA raters, help to illustrate specific language features or factors that were
identified in this study as they were used by the raters to describe both AAL and SAE.
These language features or factors are important because they help to define each group’s

language preferences based on the raters’ African region.

92

TABLE 15: FACTOR ANALYSIS: WEST AFRICAN RATERS’ ATTITUDES TOWARD AAL &

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SAE
Traits Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6
HIGH STATUS AAL=4.4
SAE=3.4
LEADER-LIKE AAL=4.2 AAL=4.2
SAE=3.0 SAE=3.0
POLITE AAL=4.0 AAL=4.0
SAE=3.0 SAE=3.0
PROUD AAL=3.2 3.3
SAE=3.3 AAL=3.2
SAE=3.3
TRUSTWORTH AAL=4.0 AAL4.0 3.2
Y SAE=3.3 SAE=3.0 AAL=4.0
SAE=3.0
FRIENDLY AAL=4.0
SAE=3.0
SOCIAL AAL=4.0
SAE=3.0
CONFIDENT AAL=3 .0 AAL=3.1
SAE=2.4 SAE=2.4
GENTLE AAL=4.0
SAE=2.5
PATIENT AAL=4.0 AAL=4.0
SAE=2.5 SAE=3.0
RESPONSIBLE AAL=4.0
SAE=3.0
WELL EDUC. AAL=4.0 AAL=4.0 AAL=4.0
SAE=3.0 SAE=3.0 SAE=3.0
FACTOR EGOC EN T PATRONIZI CHARM] PROU DEPENDAB HUMBL
Rl-C N-G -NG D -LE 1E

 

93

 

Figure 3a: Egocentric & Charming

 

 

 

 

  
  
  

 

 

 

 

 

 

 

 

 

 

 

4
35
3
25
2 IAAL
1-5 ISAE
1
05
EGOCENTRw CHARMING
Figure 3b: Patronizing & Dependable
4
3.5 g
3
2.5 g .AAL
2 i .SAE
1.5 : DAAL
1 E DSAE
0.5 E
o ,. .
PATRONIZIN DEPENDABLE
Figure 3c: Proud & Egocentric
4
35
3
25
2g EAAL
1-5 ISAE
1 1
0.5 7;
0 35:

PROUD EGOCENTRm

9,

l. Egocentric (Factor 1): WA raters perceive AAL speakers to be more “Egocentric

than the SAE users.

94

2. Patronizing (Factor 2): WA raters perceive SAE speakers to be less “Patronizing”

than the AAL users.

3. Charming (Factor 3): WA raters perceive AAL speakers to be more “Charming”

than the SAE users.

4. Proud (Factor 4): SAE users are considered less “Proud” or even less “Arrogant”

than AAL speakers by WA raters.
5. Dependable (Factor 5): SAE variety is considered as less “Dependable” than the

AAL variety by WA raters.

AS Figures 3a, 3b and 30 show, five features (Egocentric, Patronizing, Charming,
Proud and Dependable) out of the seven features identified in this study (the other two
are Sociable and Humble) are important in defining this group of West African raters’
attitudes toward the two US English varieties, i.e. African American Language (AAL)
and standard American English (SAE). The single most important factor, namely
“Egocentric” is considered by the West African raters to be relatively stronger for AAL
speakers as compared to SAE users. Important for us in this study is that SAE speakers
are also perceived to be relatively less “Patronizing,” “Charming,” “Proud,” and
“Dependable.” While the perception of SAE speakers as being less “Egocentric” and
being less “Patronizing” to the West African raters, are positive attributes, these SAE
users are also perceived as less “Charming,” “Proud” and “Dependable”; features which
can also be considered less positive. The AAL speakers are positively perceived by these
WA raters as more “Charming,” more “Proud” and as being more “Dependable.”

This study considered seven socio-linguistic features as the key factors (i.e. “Egocentric,”

“Patronizing,” “Charming,” “Proud,” “Dependable,” “Sociable,” and “Humble”) to

95

measure in the two US English varieties (i.e. AAL and SAE) used in this study depending
on the African region. Overall, however, AAL speakers have the more positive
evaluation in at least three of the five features identified by this group of West African
raters (“Charming,” “Proud” and “Dependable”). They are however rated less favorably
on two of them, i.e. being “Egocentric” and “Patronizing” by this same group of West
African judges. In summary we note the attitude of the West African raters toward AAL
and SAE, the two US English language varieties (Figure 3a, Figure 3b and Figure 3c) in
this study, also indicate a clear distinction in the perceptions they hold toward the two
varieties by the eight raters. Please note that perhaps for many of the African raters in this
study, it is our View that being “Proud” and/or “Arrogant” and “Self Important,” may
have carried the same meaning. This socio-linguistic feature can perhaps also be treated

as a positive language feature amongst raters from the three African regions.

96

FACTOR LOADINGS FOR EAST AFRICAN RATERS

TABLE 16: FACTOR LOADINGS FOR EAST AFRICAN RATERS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Trait Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 Factor 7
Trustworthy *.94 *.89 *.82 .34 .47
Social *.92 .78 *.82 —.61 .53 .53 .67
Friendly *.92 .78 *.82 -.61 .33 .46 .67
High Status *.90 *.81 *-.76 .56 .59

Patient *.74 —.72 -.65 .69 .43 .58

Responsible *.88 -.51 *-.71 *.75 *.87 .55 .41
Polite *-.7O *.93 .57 *-.84 .39 .36 .35
Well *.85 *.82 .59 .65 .65

Educated.

Proud *.84 *.9O *.82 .40 .37 -.67 .42
Leader-like -.63 -.56 .69 .52 .52 .40 -.51
Confident -.48 .53 *.80 .52 *.77 *.77 -.53
Gentle -.63 .34 *-.70 *.71 .41

 

 

 

 

 

 

 

 

* Indicates a significantly high score

 

Among the East African raters (Table 16) the Factor 1 mean scores were computed for
nine traits: “trustworthy,” “social,” “friendly,” “high status,” “patient,” “responsible,”

99 66

“polite, well-educated” and “proud.” Table 17 helps to interpret these factor loadings.

97

TABLE 17: INTERPRETATION OF THE FACTOR LOADINGS FOR THE EAST AFRICAN
RATERS IN TABLE 16

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 Factor 7
*Trustwort *Polite *Social *Trustworthy *Responsible *Confident Social
by
*Social *Proud *Friendly *Responsible *Confident Well Friendly
Educated
*Friendly *Trustworthy *Well *Polite *Gentle High Status Not
Educated Confiden
1
*High *High Status *Proud *Not Patient Not Social Proud Trustwor
thy
Status
*Responsib Friendly *Confident *Not Gentle Not Proud Responsible Not
le Leader—
like
*Patient Social *Not H. Not Friendly Not Patient
Status
*Proud Not Patient *Not Not Well Not Patient
Responsible Educated
*Not Polite Not
Responsible
Well
Educated
Not
Leader-like
Not Gentle
EGOCEN PATRON IZIN CHARMIN PROUD DEPENDAB SOCIABLE HUMBL
TRIC G G LE E

 

 

 

* INDICATES A SIGNIFICANTLY HIGH TRAIT AND SCORE

98

 

TABLE 18: EAST AFRICAN RATERS’ ATTITUDES TOWARD AAL & SAE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TRAlT FACTOR FACTOR FACTOR FACTOR FACTOR FACTOR FACTOR
1 2 3 4 5 6 7
TRUST AAL=4.0 AAL=4.0
WORTH SAE=3.2 SAE=3.2
-Y
SOCIAL AAL=3.0 AAL=4.0 AAL=3.0
SAE=3.3 SAE=3.2 SAE=3.3
FRIEND AAL=3.0 AAL=3.4 AAL=3.0
-LY SAE=3.3 SAE=3.0 SAE=3.3
HIGH AAL=3.0 AAL=4.0 AAL=3.3
STATUS SAE=3.3 SAE=3.4 SAE=3.4
PATIEN AAL=3.5 AAL=3.5 AAL=4.0
-T SAE=3.4 SAE=3.4 SAE=3.0
RESPON AAL=3.0 AAL=2.4 AAL=3.0 AAL=3.4 AAL=3.0
-SIBLE SAE=3.0 SAE=3.0 SAE=3.0 SAE=3.0 SAE=3.0
POLITE AAL=3.0 AAL=3.0 AAL=4.0
SAE=3.2 SAE=3.2 SAE=3.2
WELL AAL=3.0
EDUCA- SAE=3.0
TED
PROUD AAL=3.4 AAL=3.1
SAE=3.0 SAE=3.0
LEADE-
RLIKE g
CONFID AAL=3.0 AAL=3.0 AAL=2.5
-ENT SAE=2.3 SAE=2.3 SAE=2.3
GENTL- AAL=3.0 AAL=3.0
E SAE=3.5 SAE=3.0
FACTO— EGOCE— PATRON CHARMI PROUD DEPEND- SOCIABL HUMBL-
R NTRIC -IZING -NG ABLE -E E

 

 

 

 

 

 

 

 

Following the results of attitudes by EA toward AAL and SAE in Table 18, we added

Figures 4a, 4b and 4c below to supplement the findings. Both also help to further

illustrate these attitudes using six factors (Egocentric, Charming, Patronizing,

Dependable, Proud and Sociable).

The figures Show overlapping in the language features used to rate the two varieties.

While some distinctions emerge in the EA raters’ perception toward the two varieties, the

99

 

overlapping seems to sometimes blur a much clearer distinction of the features used to

rate both AAL and SAE.

Figure 4a: Patronizing & Dependable

 

AAL
I SAE

 

 

 

 

 

AAL
I SAE

 

 

 

 

PATRONIZING DEPENDABLE

100

Figure 4c: Proud & Sociable

 

 

 

 

4
3.5
2.5
IAAL
1.5 ISAE

 

 

 

    

PROUD SOCIABLE

1) Egocentric (Factor 1): EA raters perceive SAE speakers to be slightly more
“Egocentric” than AAL.
2) Patronizing (Factor 2): EA raters perceive AAL users to be more “Patronizing” than
the SAE speakers.
3) Charming (Factor 3): EA raters perceive AAL speakers to be slightly more
“Charming” than SAE.
4) Proud (Factor 4): EA raters perceive AAL users to as more “Proud” and perhaps
“Arrogant’” in this manner than their SAE counterparts.
5) Dependable (Factor 5): EA raters perceive AAL speakers to be more “Dependable”
than SAE speakers.
6) Sociable (Factor 6): EA raters perceive the SAE variety to be less “Sociable” than the
AAL variety.

Both the results of Table 18 and Figures 4a, 4b and 4c further suggest that, one of the

few distinctions between the two varieties is most evident in the mean score differences

101

of factor group 2 (“Patronizing”) and factor group 4 (“Proud”). Speakers of African
American Language (AAL) variety are perceived by the East African raters to be more
“Patronizing” and “Proud” and perhaps “Arrogant” in this respect than speakers of
standard American English (SAE) variety. The mean scores differences for what we
consider to be more positive and favorable features such as “Charming,” “Dependable,”
and “Sociable,” are less pronounced compared to the differences in the two
aforementioned (“Patronizing” and “Proud”) negative and less favorable language
features.

The EA raters perceive speakers of the SAE variety. as “Egocentric,” “Patronizing and
“Proud.” They however also perceive these SAE speakers, as “Charming.” but not very
“Dependable” or “Sociable.”

Interestingly, the East African raters also perceive both AAL and SAE as
“Patronizing” and “Proud,” to some extent given the relatively small mean differences
between them. The mean differences between these two factors are relatively higher than
the mean differences between the less pronounced values for the positive but favorable
features such as, “Charming,” “Dependable,” and “Sociable.” This seems to suggest that,
the EA raters have more positive attitudes toward varieties that they perceive as
“Egocentric,” “Proud” and “Patronizing,” without sounding either too “arrogant,”

“Egocentric,” or overly “Patronizing.”

102

FACTOR LOADINGS FOR SOUTH AFRICAN RATERS

TABLE 19: FACTOR LOADING FOR SOUTH AFRICAN RATERS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Trait Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 Factor 7
Confident *.97 .56 *.75 *.94 *.82 -.63 *. -79
Leader-like *-.97 —.57 *.71 *.93 *-.83 .45 -.62
Well Educated *-.97 .59 *.81 *.94 *.79 .30 *-77
Responsible *.93 *.81 *.90 *-.79 *-.79

Polite .64 *.84 *-.76 *-.82 *.74

Patient *-.90 *.84 .68 *.82 .37 .34
Social *.88 *-.72 .57 *.75 .38
Friendly *.88 *.75 *.83 .38
High Status *-.87 *.86 *-.76 *.86 *-.93
Trustworthy *.86 .67 .45 .65 .56 *.93

Gentle *.83 *.90 *76 .30 .59 .34
Proud *.70 -.68 .52 *-.77 -.67 .30

 

 

 

 

 

 

 

 

* Indicates a Significantly High Trait and Score

103

 

TABLE 20: INTERPRETATION OF THE FACTOR LOADINGS FOR SOUTH AFRICAN RATERS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

IN TABLE 19
Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 Factor 7
*Confident *Responsibl *Confident *Confident *Confident *Not Polite *Not
Confident
e

*Not Leader- *Polite *Leader-like *Leader-like *Not Lea.- *Friendly *Not Well
like like Educated
*Not Patient *Patient *Well *Well *Well *Trustworth *Not High

Educated Educated Educated y Status
* Well *High Status *Responsibl *Gentle *Not Not Conf. Leader-
Educated e Responsible like
*Responsible *Not Social *Not High *Not Polite *Not Polite Leader-like Patient

Status
*Social *Not Gentle Trustworthy *Not *Patient Well Social

Responsible Educated
*Friendly *Not Proud Proud *Not Proud *Social Patient Friendly
*Trustworthy Confident High Status *Friendly Gentle Gentle
*Gentle Leader-like Patient Trustworthy Proud
*Proud Trustworthy Gentle
Not Proud

EGOCENTRI PATRON I- CHARMIN PROUD DEPENDAB SOCIABLE HUMBL-
-C ZING -G -LE E

 

 

 

 

 

 

 

* Indicates a Significantly High Trait and Score

The results in Table 20 for SA raters indicate there are at least two broad categories of

clusters that represent the more favorable and therefore more positive values that are

comprised of linguistic characteristics such as “Polite,” “Patient,” “Friendly,” and

“Social.” These are also the traits of the group that loads as Factor 5 for “Dependable.”

The other broad cluster of traits represents the negative values and only contains some of

(other than “Egocentric” which contains all four traits) of the afore-mentioned four

positive attributes. In conducting this factor analysis, we concluded that the other five

groups represent the broad combination of what the raters perceive as negative language

attributes.

104

 

TABLE 21: SOUTH AFRICAN RATERS’ ATTITUDES TOWARD AAL & SAE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TRAIT FACTOR FACTOR FACTOR FACTOR FACTOR FACTOR FACTOR
1 2 3 4 5 6 7
CONFIDENT 3.0 3.0 3.0 3.0 3.0
AAL=3.0 AAL=3.0 AAL=3.0 AAL=3.0 AAL=3.0
SAE=2.5 SAE=2.5 SAE=2.5 SAE=2.5 SAE=2.5
LEADER- 4.0 4.0 4.0 4.0
LIKE AAL=4.0 AAL=4.0 AAL=4.0 AAL=4.0
SAE=3.3 SAE=3.3 SAE=3.3 SAE=3.4
WELL- 3.1 3.1 3.1 3.1
EDUC AAL=3.4 AAL=3.3 AAL=3.3 AAL=3.3
SAE=3.0 SAE=3.0 SAE=3.0 SAE=3.0
RESPONSIB 3.0 3.0 3.0 3.0 3.0
LE AAL=3.1 AAL=3.l AAL=3.1 AAL=3.l AAL=3.1
SAE=3.0 SAE=4.0 SAE=3.0 SAE=3.0 SAE=3.0
POLITE 3.2 3.2 3.2 3.2
AAL=3.3 AAL=3.3 AAL=3.3 AAL=3.3
SAE=3.1 SAE=3.1 SAE=3.1 SAE=3.1
PATIENT 4.0
AAL=4.0
SAE=4.0
SOCIAL 3.1 3.1 3.1
AAL=3. l AAL=4.0 AAL=3.l
SAE=3.2 SAE=3.2 SAE=4.0
FRIENDLY 3.1 3.1 3.0 3.1
AAL=3. l AAL=3. l AAL=3. l AAL=3. l
SAE=3.2 SAE=3.2 SAE=3.2 SAE=3.2
HIGH 4.0 4.0 4.0 4.0
STATUS AAL=4.0 AAL=4.0 AAL=4.0 AAL=4.0
SAE=4.0 SAE=4.0 SAE=4.0 SAE=4.0
TRUSTWOR 3.1 3.1 3.2
THY AAL=3.3 AAL=4.0 AAL=4.0
SAE=3.0 SAE=3.2 SAE=3.0
GENTLE 3.2 3.2
AAL=3.4 AAL=3.4
SAE=3.0 SAE=3.0
PROUD 3.4 3.4
AAL=3.4 AALflD
SAE=3.4 SAE=3.3
FACTOR EGOCE- PATRON CHARMI- PROUD DEPEND— SOCIABL HUMBLE
NCENTR -IZING NING ABLE -E
IC

 

 

 

 

 

 

 

 

The results of Table 21 above are supplemented by Figures 5a, 5b, 5c and 5d that

follow below. Both suggest that, the South Afiican raters seem to regard speakers of

African American Language (AAL) to be “Egocentric,” yet more “Patronizing,” and

relatively more “Charming,” “Proud,” and more “Dependable” than SAE Speakers. They

105

 

(AAL speakers) are perceived to also be more “Sociable,” and less “Humble” than SAE
speakers. This suggests to us that AAL speakers are perhaps less reserved and more

outgoing than SAE users in their general speech demeanor.

Figure 5a: Egocentric & Charming

 

 

 

 

 

 

 

 

UAAL
I SAE
3.6
3.5
3.4
3.3 IAAL
3.2 .SAE

 

 

 

  

 

3.1 g]

PATRONIZIN DEPENDABLE

 

106

Figure 5c: Proud & Sociable

 

IAAL
ISAE

 

 

 

 

 

IAAL
ISAE

 

 

 

 

2.7
DEPENDABLE HUMBLE

1) Egocentric (Factor 1): SA raters perceive both AAL speakers and to be
more “Egocentric” than SAE users.
2) Patronizing (Factor 2): SA raters perceive AAL users to be more “Patronizing” than
SAE speakers.
3) Charming (Factor 3): SA raters perceive SAE speakers to be less charming than AAL

users.

107

4) Proud (Factor 4): SA raters perceive AAL Speaker to be more “Proud” than SAE users.
5) Dependable (Factor 5): SA raters perceive the SAE speakers to be less “Dependable”
than the AAL users.

6) Sociable (Factor 6): SA raters perceive AAL speakers as more “Sociable,” than the
SAE users.

7) Humble (Factor 7): SA raters perceive SAE speakers as less “Humble” than the AAL
users.

Figures 5a, 5b and So help to also indicate that Standard American English (SAE)
speakers are perceived by the South African raters to be “Egocentric,” yet less
“Patronizing,” “Proud,” and “Charming” than AAL speakers. They are however also
perceived to be less “Dependable,” “Sociable” and “Humble” than AAL speakers.

IDENTIFICATION OF AFRICAN AMERICAN LANGUAGE (AAL) AND
STANDARD AMERICAN ENGLISH (SAE) VOICES BY ALL RATERS

TABLE 22: RATERS’ IDENTIFICATION OF THE TWELVE ANONYMOUS STIMULUS VOICES

 

 

 

 

 

 

WA (8) EA (8) SA (8) Total (24)
AALF 6 6 7 19 (79%)
AALM 4 4 5 13 (54%)
SAEF 8 5 6 20 (83%)
SAEM 5 3 3 14 (58%)
23 (96%) 18 (75%) 19 (79%)

 

 

 

 

 

 

 

108

Figure 6a: Identification of AAL & SAE

 

IWA
I EA
USA

 

 

 

o-xrowhurmuoo

 

 

IAALF
IAALM
DSAEF
EISAEM

 

 

 

 

Table 22 and Figures 6a and 6b both help summarize how the 24 African raters from
South, East and South Afiica identified both AAL and SAE. All the evaluators also took

the voice and region matching test in part II of the questionnaire (Appendix A). The

109

results of this identification exercise show what we consider a generally balanced
recognition of the twelve stimulus voices that represented both AAL and SAE varieties in
the matched guise exercise used in this study.

The most recognized variety was the Standard American English Female (SAEF)
variety, identified by more than two thirds (83%) of all raters. It was followed by the
African American Language Female (AALF) variety, which, was identified by 79% of all
the evaluators from all three African regions (West, East and South Africa). The African
American Language Male (AALM) variety was recognized by 54% and closely followed
that of Standard American English Male (SAEM) variety identified by 58% of all the
respondents. It is clear that the 24 African raters could easily identify and differentiate
both AALF and SAEF voices much more clearly than either the AALM or SAEM voices.
It is also likely that for the raters as well, the high-pitched female voices were much
clearer and easier to distinguish when compared to the deeper and low-pitched male
voices, making the task of identifying the two females voices easier than of the two male
voices. From the findings of this study, when the raters successfully identified the AAL
voices, they also generated rated them relatively higher than the SAE voices. On the other
hand the findings also indicate that overall, the SAEF variety earned higher ratings than
the SAEM variety.

The raters with the highest percentage in the correct identification of the stimulus
voices are the West Africans (96%). The East Africans were correct 75% of the time
while the South African raters had a 79% identification rate.

One of the objectives of this study was to determine whether the twenty-four raters

would positively identify the two varieties of US English (i.e. AAL and SAE), especially

llO

the African American Language (AAL) variety. This is in view of prevailing language
attitudes toward Languages of Wider Communication (LWC) such as Standard American
English (SAE) and European languages in general both in Africa and in North America in
contrast to AAL and indigenous African varieties, which have historically been
marginalized in favor of these European languages.

Based on the data we have obtained, it is evident that raters from all three African
regions often recognize the two US English varieties (i.e. AAL and SAE), albeit at
different rates. Although this is an exploratory study, limited by the lack of a larger and
much more diverse sample, the study represents an important inroad into language
attitude studies on Black populations. There is a need for similar studies to continue to
be methodologically sophisticated even as they explore new dimensions for expanding
the nature of studies of marginalized languages and language groups, such as both
African languages and languages with an African heritage such as African American.

Language (AAL).

111

CHAPTER 5: SUMMARY, CONCLUSIONS AND STUDY IMPLICATIONS

“The more people you meet, the more human you become.”
A Bambara proverb (Djite 2006, l)

The quote by Djite, a linguist from the Bambara community in West Africa articulates
the current state of global encounters in the world today among diverse people and
languages. For the twenty four African immigrants from West, East and South Africa in
this study and those they encounter, language use remains the tool of communication that
best mirrors the various global and social encounters including the “shifts in linguistic
identities in a global world” (Djite 2006). The aim of this study was to investigate the
nature of language attitudes of African immigrants in the US from West, East and South
Africa toward African American Language (AAL). Attitudes toward this US English
variety by the twenty four African immigrants in this study, as measures of language
perceptions, were tested using a matched-guise technique by Tucker and Lambert (1960
and 1967) to rate twelve anonymous stimulus voices, using both AAL and SAE, on
twelve language traits along a seven—point Semantic Differential Scale.

The purpose of this chapter is to provide a synopsis of the study, conclusions from the
findings and implications for further work by language attitude researchers, as well as

implications for language planners, policy makers and bi/multilingual educators.

SYNOPSIS OF THE STUDY

The problem under investigation was presented as the complete lack or limited
research in language attitude studies of Africans toward African American
language/English (AAL/AAE). The problem is complicated by the focus on attitudes

toward major European languages commonly spoken in Africa such as English, French

112

and Portuguese by many language attitude studies in both Africa and the Diaspora. This
language attitude research scenario, which, helps to also continue to inform prevailing
language and education planning policy practices in Africa, is largely a legacy of
European colonialism (Adegjiba 1994;Mbaabu 1996; Mazrui & Mazrui 1998). Coupled
with uni-dimensional research approaches (Adegjiba 1994), the dearth and shortcomings
of language attitude study in these circumstances seem, to have helped contribute to
perceptions that favor European languages, often associated with high status and prestige
at the expense of indigenous Afiican languages.

Makalela (2005) and Thiba (2000) note that in South Africa negative attitudes and
misrepresentation of the status of African languages as distinct languages, when they
could in certain cases have been harmonized as one, have contributed to the history of the
unchanging low status of indigenous African languages. This situation has, Since the
1884 Berlin Conference that helped to divide Africa among European powers, marked
what Mazrui & Mazrui (1998) suggest was a strange division of Africa into a linguistic
anomaly and zones on the basis of the imperial European languages such as Anglophone
Afiica, Lusophone Africa and Francophone Africa. African immigrants in the US, the
focus of our study also bring with them many of these language attitudes toward both
European and African languages.

The rationale was provided for the need to study attitudes toward African American
Language (AAL) by African immigrants in the USA in order to help understand their
attitudes toward AAL in contrast to Standard American English (SAE), a variety that
enjoys high prestige and status in the US comparable to that of the ‘Queen’s English’

(Adegbija 1994; Mbaabu 1996) or even that of both French and Portuguese in Africa.

113

Understanding the nature of these attitudes through this language attitude study is

necessary to help bridge the many years of systematic and deliberate linguistic divisions

that were wrought in both the lives of Africans and African Americans through European

colonialism and enslavement. Chapter One outlined the purpose of the study and set out

the following objectives:

0 To investigate the attitudes of African immigrants in the US toward AAL

0 To determine the nature of these attitudes

0 To help contribute to scholarship on attitudes toward AAL and indigenous African
languages and to language attitude study both in Africa and the Diaspora in general.

It was established that Language attitudes of Afiican immigrants to the US are
complicated by those held toward varieties of African languages and European languages
both in Africa and beyond. These attitudes have also been complicated by a history of
European colonialism and languages as well as the existence of numerous African
languages. In both West and East Africa for example both English and French are
Spoken alongside various African languages while English, Portugueseand Afrikaans are
used in South Africa alongside a variety of indigenous African languages. Within this
complex web of historical events that impact attitudes toward languages, future language
attitude studies would perhaps need to investigate and also describe the evolving nature
of attitudes toward AAL by Africans both in the Continent and the Diaspora.

Further discussions focused on language and immigrants along with the history of
individual immigration by leading Afiican leaders such as African pioneer educator, Dr.
Aggrey and his future student, Kwame Nkrumah of Ghana. This was followed by

discussion of the immigration of John L Dube from South Africa, among other senior

114

African figures who visited and interacted with African Americans and other US citizens
many years before the advent of African independence from European colonialism in the
late 19503 and early 19605. This part of the discussion also linked the present study to _
recent increases in the number of Afiicans migrating to the US, which, is estimated at
over fifty thousand per year since 1990 (Roberts, New York Times, February 21, 2005).
This is the largest group of African immigrants to arrive in the US since the outlawing of
the transatlantic Slave trade in 1807 and includes some of the most highly educated recent
immigrants to the country. These African immigrants represent some of the grim realities
of today’s multi-polar globalization with its many attendant effects. When they
immigrate, these Africans bring with them various language attitudes embodied in their
use of both indigenous African and European languages.

The experiences of both Africans and African Americans are contextualized in the
study within the realm of cross-cultural research base on a shared history of European
enslavement and colonialism, of which Fanon articulated in Black Skin, White Masks
(1967): “Every colonized people. . .in whose soul an inferiority complex has been created
by the death and burial of its local cultural originality — finds itself face to face with the
language of the civilizing nation; that is, with the culture of the mother country” (18).
From a political perspective the study also echoes F anon’s challenge that “Each
generation out of relative obscurity must discover their mission, fulfill it or betray it”
(1963, 207).

The introduction to this study also pointed out how the global availability of AAL in
the dissemination of US mass media products is further impacting language attitudes

toward the variety. It is therefore most likely that many African immigrants to the US

115

have previously been exposed to AAL in Africa and are likely to identify and associate
certain characteristics of AAL with its speakers. It is also possible that, despite many
years of linguistic discrimination, AAL still enjoys “covert prestige” (Trudgill 1972, 179-
95) among both users and nonusers as it facilitates the global dissemination of US
popular culture.

Next was a review of literature on language attitudes and the theoretical context for
language attitude study in sub-Saharan Africa in general (see Mbaabu 1996; Adegbija
1994; Mazrui & Mazrui 1998 etc.). It indicated that language attitudes toward both
indigenous African and European languages remain an important component. This is
because problems of post-independence education and language planning policies in most
of sub-Saharan Africa are historically and closely associated with both language attitude
studies and related attitudes that have often privileged European languages over African
languages.

The first part of the review of literature revealed a research gap and hence the need to
study prevailing language attitudes toward both African languages in Africa and
languages spoken by speakers of African ancestry in the Diaspora. The review of
literature noted a general attitude of ambivalence is sometimes evident with regard to
perceptions toward European languages in contrast to indigenous African languages. The
review observed that such attitudes reflect “the conflict between loyalty to one language
and the utility of another: the choice between learning a language because it is useful
(instrumentalism) and learning another because it marks the individual’s cultural, ethnic

or national identity (integration)” (Sure 1991, 251). This seems to reveal another

116

dimension of the “instrument/integrative dichotomy of motivation in the learning of
languages in the African context” (Adegbija 1994, 64).

The review of literature then focused on several shortcomings also observed by
Adegbija (1994) who suggested that, future language attitude study in sub-Saharan Africa
would need to be less impressionistic, more sophisticated in research design and more
thorough in its sampling and analytical techniques. The scholar adds that, such research
would also need to use more detailed statistical procedures and methods and should begin
by grounding itself in attitude theory and research. It should also place high premium on
utilizing attitude scales that adhere to criteria for internal consistency in response and in
the use of many items to ensure internal reliability along with issues of both content and
criterion-related, predictive and construct validity. The review cited Baker (1992, 25)
who recommends that language attitude research must adhere to the dimensionality that
includes the testing of whether one or more entities need to be measured while also
considering various aspects of uni-dimensionality or multidimensionality along with
whether these two can be measured if a large initial pool of attitude items is subjected to
an exploratory confirmatory factor analysis.

The second part of the review of literature concerned itself with language attitude
research in Africa and beyond. It tried to make pertinent global connections with other
language attitude studies in for example Australia and with similar research toward AAL
in the US. It also linked these studies to challenges posed by others that detail attitude
shifts among Africans toward European languages in the Continent. By further tracing
language attitude research to its origins in the early 1960’s from the field of social

psychology (Lambert, et al, 1960; Lambert 1967), the review was then linked the work of

117

linguists involved in language attitudes studies in the 1970’s especially as it pertained to
teacher’s attitudes toward students’ speaking styles. Unfortunately, the majority of these
studies focused on western countries and western languages. African languages in the
Continent and their counterparts in Diaspora such as African American Language (AAL)
received scant or negative attention in these earlier language attitudes research despite the
fact that bilingualism, multilingualism and multiculturalism are prevalent among both
Africans and African Americans.

It was concluded that, based on the foregoing review of literature, the present study is
methodologically significant in helping to expand current body of knowledge on
language attitude study in both Africa and the Diaspora. The present study also aims to
make important contributions to the study of attitudes of Africans in the USA toward
AAL, the variety spoken by many African Americans. It departs from many and previous
approaches in language attitude research in socio-linguistics which are based on less
reliable methods such as observation or intuition by exploiting the benefits of a matched
guise technique by Lambert and associates ( 1960, 1967). Basing its work on pertinent
attitude research theories, it successfully achieved this aim by using 12 anonymous
stimulus voices audio-taped in both AAL and SAE to test both research hypotheses, and
approach, among 24 African evaluators. This technique was ascertained as adequately
measurable and quantifiable. The technique is detailed in the following part of this
synopsis.

Following the review of literature, details and rationale of the methodology used in the
study were provided. It was clarified that due to the complexities of accurately

measuring both hidden and “deep—seated” attitudes toward language, the most feasible

118

technique would need to offer both anonymity and adequately measurable attitude
ratings. The study utilized the matched guise method by Lambert and associates (1960
and 1967), in which a bilingual speaker says the same thing in two different languages.
The “guise” is that the speaker is represented as being a different speaker when speaking
a different language, thus “fooling” respondents into thinking that they are rating a
different speaker when in actuality, only the language is different. The matched guise
allows researchers to obtain feedback on perceptions of language per se since the “two
speakers” being rated are the same person. Thus all speaker variables are controlled, with
only language being manipulated.

We used twelve sample voices from a female and male AAL Speaker, who are skillful
in code switching between AAL and SAE. They read a passage by African writer Chinua
Achebe, Caribbean writer and poet Derek Walcott and African American writer and poet
Alice Walker in both varieties generating a total of twelve anonymous stimulus voices.
These voices were audio-recorded and replayed to the 24 raters who then evaluated them
on a seven-point scale based on the twelve adjectives or language traits of the semantic
differential scale

Presentation and analysis of data and results followed. This first part of the data
analysis using Ratings and Rankings of Mean Scores showed that perceptions and
attitudes of the 24 African immigrants from West Africa, East Africa and South Africa
toward both SAE and AAL, are both favorable and unfavorable, based on their region of
origin in Africa.

The Ratings and Rankings of the Mean Scores helped us to also ascertain there were

no significant statistical differences in terms of gender, age, and years spent in the US. A

119

Correlation Coefficient analysis helped to indicate both the strength and direction of the
relationship between a pair of language variables/traits in the study. It was used to
measure the linear relationship between at least any of two language variables. In this
case these two variables were two semantic differential scale language adjectives on the
semantic differential scale. The measure also provided us with clearer assessments of the
closeness of the relationship among pairs of variables. In our case this was the
relationship among our list of twelve traits or adjectives of the semantic differential scale.
In summary, cluster sets, identified in the Correlation Coefficient Analysis, among the
WA raters indicate that two language traits namely, “Proud” and “High status,” are often
the only two traits that are characterized by only one positively significant correlation
each. For example, “Proud” has a positively significant correlation only with “Social”
while “High status” also has one positively significant correlation with “Trustworthy.” It
also has a significant but negative correlation with “Proud.” Both EA and SA raters
shared two cluster sets that are patterned around “Trustworthy” and “Proud” as well as
around “High Status” and “Social.” They also both have the same order of voice
preferences (i.e. SAEF, AALM, AALF and SAEM). However the ratings for both
regions are patterned around different cluster groups, namely three clusters among East
African raters that contain at least one or both of the following two traits: “Trustworthy”
and “Proud.” The second group contains at least one or both of the following traits:
“High status” and “Social.” Finally, the third group is a complement of the first two and
can be described as a group of clusters that do not contain any of the four language traits.
Among the SA raters, the pattern of attitudes cluster around a large group of traits that

share often negatively correlated membership. Common to the SA group is the many

120

negative correlations that all the 12 traits have in common with each other in the 12
voices. These clusters have such few positively Significant correlations, namely, 1)
“Well Educated” and “Confident”; 2) “Well Educated” and “Proud”; 3) “Trustworthy”
and “Proud”; 4) “Social” and “Proud” as well as both 5) “Social” and “High status.”

Finally, using the Factor Analysis, which was the third part of the data analysis and
presentation, the study found that AAL is perceived as more “Egocentric” than SAE by
raters from WA and SA and compares favorably with the same description for SAE by
EA raters. Raters from all three regions also identified AAL as more “Patronizing,”
“Charming,” “Proud,” and “Dependable.” Both EA and SA raters also perceived AAL to
be more “Sociable” and “Humble.” The EA raters however perceived SAE as much more
“Egocentric” “Patronizing,” and “Proud,” than AAL.

99 ‘6

These seven factor groups, namely, “Egocentric,” “Patronizing, Charming,”
“Proud,” “Dependable,” “Sociable” and “Humble” helped us to identify and rate the 12
language traits and features used in this study on the basis of them being either favorable
(i.e. positive) or unfavorable (i.e. negative). While these attitudinal features and
perceptions by the 24 African raters help us to describe the two language varieties (i.e.
AAL and SAE), they also help provide information used to categorize the language
varieties using various combinations of traits that make up the seven factor loadings.
These findings helped to answer one of our research questions, which, asked whether
there would be any significant differences in attitudes by the 24 raters based on African
region. The findings further helped to reveal the nature of the 24 African respondents’

attitudes as either favorable or unfavorable toward AAL and SAE depending on African

region. However, the results did not help reveal significant findings to help answer the

121

research question that asked whether there would be differences in the raters’ attitudes
toward both AAL and SAE based on gender, age and years spent in the US. The findings
also did not answer the research question that asked Whether attitudes toward AAL' and
SAE would be based on familiarity with the varieties.

There are overlaps in terms of attitudes and related rankings of the 12 stimulus voices.
The measurement of these attitudes is by no means absolute and only helps indicate
general trends in the relationship among the twelve language traits used to rate both
varieties in the study. The findings also have important implications of such measures of
attitudes and attendant beliefs of African immigrants in the US toward AAL, which like
language attitudes in general, are often hard to measure.

Both AAL and SAE varieties were not described in sharply contrastive manner, as
expected. However, the findings suggest there perhaps are other influences in the 24
raters’ attitudes such as prior attitudes toward both indigenous African and European
language and cultural backgrounds. Other factors include gender, age, years lived in the
US and whether they were exposed to French and Portuguese as their second language
before learning English, etc. Other influences might also be the choice of reading
passages by Achebe, Walcott and Walker, which, are written by scholars from three
varied linguistic experiences. Both Achebe and Walcott write in the ‘Queen’s English’
(Adegjiba 1994; Mbaabu 1996) using What can be considered West African and
Caribbean linguistic styles respectively, while Walker’s writing is AAL in style
especially when read in AAL. We used the matched- guise technique (Lambert and
associates 1960 and 1967) to help control for some of these factors. It is possible

however that, the contents of the passages themselves and choice of the two AAL readers

122

who also code-switched to SAE in audio-recording the 12 stimulus voices had a bearing
on the attitudes and ratings of both varieties by the 24'African evaluators.

The results present a picture of the trends that the various combinations of the twelve
language traits by the 24 raters helped set for describing both language varieties, based on
rating the twelve anonymous AAL and SAE stimulus voices. The results nonetheless
help represent the perceptions held by all the 24 raters toward the two varieties. In terms
of the twelve language characteristics (a.k.a. traits, adjectives, features and attributes),
“Proud” is the one trait that reveals itself most prominently and which perhaps seems to
both infer both favorable and negative evaluation of the 12 stimulus voices, by the 24
African raters. This trait might also require firrther study and analysis. While the trait
“Proud” might be construed as positive, it is possible that among the 24 African
respondents, it is also viewed negatively. This observation is supported by the results of
both the Correlation Coefficient and Factor Analyses. This is less evident in the Mean
Score Rating and Ranking Analysis.

The final analysis of data compared the rates of identification of AAL and SAE in the
twelve anonymous stimulus voices. The results Show the most recognized variety is the
Standard American English Female (SAEF) variety (83%), followed by the African
American Language Female (AALF) variety (79%). The African American Language
Male (AALM) variety was correctly identified by 54% of all raters, while Standard
American English Male (SAEM) variety was correctly identified by 5 8% of the raters.
Where both major voice varieties were identifiable, the AAL variety was rated more
favorably and more often than the SAE voices by all raters and in particular when it came

to rating the SAEM voice. Overall, while both East and South Afiican raters rated the

123

SAEF variety most favorably, it nonetheless occupied the third ranking after both the
AALF and AALM voices.

It is possible that for the 24 African raters both the higher-pitched AALF and SAEF
voices were rendered in much more clearer and distinct styles. Consequently, they were
easier to distinguish from one another in comparison to AALM and SAEM voices which
were rendered at lower and harder to distinguish vocal levels. The raters with the highest
percentage in the correct identification of the stimulus voices are the West Africans
(96%). The East Africans were correct 75% of the time while the South Afiica raters had
a 79% identification rate.

CONCLUSIONS

The main conclusion of the study is that attitudes of African immigrants in the US
toward African American Language (AAL) in contrast to Standard American English
(SAE) are based largely on the region of Africa they originate from. The West African
raters consider AAL Speakers to be more “Egocentric” than SAE users. They also
perceive these SAE speakers to be less “Patronizing,” “Charming,” “Proud,” and
“Dependable,” features which can also be considered unfavorable and/or negative. The
East African raters perceive AAL to be more “Egocentric,” “Patronizing” and “Proud,” in
comparison to SAE variety, which, they perceive as more “Charming.” They however
also rated SAE users less favorably for being “Dependable” and “Sociable.” The South
African raters regard AAL users to be more “Egocentric,” “Patronizing,” yet relatively
more “Charming” “Proud,” and more “Dependable” than SAE speakers. They also

perceive AAL speakers as more “Sociable,” and “Humble.”

124

Overall, WA raters identified AALF, AALM, SAEM and SAEF in order of preference
while EA raters identified them in the following order: SAEF, AALM, AALF and
SAEM. SA raters picked the four voices in the following order: SAEF, AALM, AALF
' and SAEM. There were no significant differences found in these attitudes toward AAL
and/or SAE based on gender, age or the number of years the respondents had lived in the
US. The findings helped to also prove the prediction in our main null hypothesis, that,
based on African region of origin (i.e. West, East and South Afiica), there would be
variation in attitudes among the three sets of African immigrants in the USA toward
AAL.

While the findings indicated that attitudes toward AAL and SAE based on African
region was an important factor among the 24 African evaluators, it did not produce
further evidence to suggest that linguistic backgrounds of the African raters affected these
language attitude. It is however possible that their attitudes and ratings of both AAL and
SAE were influenced by corresponding attitudes toward varieties of both indigenous
African languages and European languages spoken among the 24 raters. This is possible
given the fact that, French is spoken by a number of East and West African raters while
Portuguese is spoken among selected South African raters, in our sample. All of the
raters speak English along the other European languages.

These speakers of both French and Portuguese might also have had different attitudes
toward both AAL and SAE, which are two English language varieties. Further research
might help isolate attitudes of specific raters toward AAL or SAE, in terms of whether

they speak these two European languages instead of English.

125

Based on the literature review in the study it was assumed however that, attitudes
toward European languages including both SAE and to some extent AAL, will also be
influenced by the fact that, historically these varieties have continued to enjoy high status
against the low status of indigenous African languages (Adegjiba 1994) and languages
spoken by people of African heritage that includes both indigenous African languages
and AAL. We therefore assumed that attitudes of the raters would be impacted by these
linguistic and language attitude experiences. We however, pre-screened the 24 African
respondents in order to ensure they at least spoke an indigenous Afiican language as their
first language. In this manner we assumed that a European language would be their
second or third language and attitudes toward both African languages and European will
likely be contrastive. The findings however suggest that attitudes of these African raters
toward both AAL and SAE are much more complex and calls for additional research.

Also, based on the variations in the ratings of both AAL and SAE by region of origin
in Africa, the findings follow the argument that language is often used to share a socially
and culturally shared communal worldview among Africans in both the Continent and in
the Diaspora (Mbiti 1969; Asante 1988). In this case regional identification largely based
on shared linguistic, socio-cultural and historical experiences as West Africans, East and
South Africans is a plausible explanation for the resultant variation in language attitude
by region. .

This study has generated important findings based on some of its assumptions,
research hypotheses and questions. However some of the assumptions in this study will
need further study. They include the following: 1) African immigrants and AAL users

will identify with one another based on language usage of AAL and/SAE depending on

126

existing attitudes they both hold toward one another; 2) Knowledgeable African
immigrants will voluntarily enter into dialogue with AAL in solidarity to help confront
negative global and historical circumstances, which have often been beyond their control;
3) African immigrants will relate to AAL and its speakers, in the attempt to learn useful
skills needed to help them adapt, transform and improve their welfare and that of their
families socially and economically in the USA.

It is important to continue to build on the results of this study in order to continue to
effectively respond to issues of language and culture, both dynamic processes that
continue to evolve and regulate and that often also respond to their users’ needs. This
especially important among marginalized people of African descent both in Africa and its
Diaspora. It is also crucial that future language attitude research continue to investigate
in what ways languages such as varieties of African languages and AAL as well as
Languages of Wider Communication (LWC), etc. help to create self and collective
efficacy. This is vital for social transformation by injecting faith and belief in one’s
ability individually and that of others (Njogu, 2006).

Further, among Africans, both in the Continent and in the Diaspora, languages also
form the “centrality of a worldview based upon Africa as the essence of Afrocentricty,
found in every compartment of postmodern history” and which as Diop (1978) shows in
his research, “finds its place in the origins of civilizations” (Molefi Asante 1979, 9).
Interpreted in this manner, the findings echo the notion that language helps communicate
the spiritual and communal nature of the African people’s worldview. This African
worldview is embodied in the various African Centered languages and the socio-cultural

discourses they embody, including African American Language (AAL). Mbiti

127

demonstrates in African Religions & Philosophy (1969) that the African worldview
including language is inseparable from the African daily religious beliefs and socio-
cultural practices. Consequently this future research might explore in what ways one’s
language and identity are intrinsically tied since language “is one of the most intimate
expressions of identity, indeed ‘the skin that we speak’” (Delpit and Dowdy 2002, 47).
Similarly, such research might help answer the question of how language is often bound
to power and privilege (F airclough 2001) including the worldview or ideology we
subscribe to.

The findings in the present study help sustain the notion that our attitudes and
worldviews are all embodied in our languages and cultures. According to Ngr'rgi (1986)
language and culture are inseparable:

[A] specific culture is not transmitted through language in its universality, but in
its particularity as the language of a specific community with a specific history.
Written literature and orature are the main means by which a particular language
transmits the images of the world contained in the culture it carries.

Language as communication and as culture are then products of each other . . .
Language carries culture, and culture carries, particularly through orature and
literature, the entire body of values by which we perceive ourselves and our place
in the world . . .. Language is thus inseparable from ourselves as a community of
human beings with a specific form and character, a specific history, a specific
relationship to the world. (15-16)

SUGGESTIONS FOR FURTHER RESEARCH AND IMPLICATIONS OF STUDY
1. Further research should be conducted using a larger sample to help determine
language attitudes toward AAL among African immigrants in the USA, based on
whether they reside in a large US metropolis compared to living in a small town

and whether there are significant differences among respondents who live in the

US East, the Mid-west, and the West Coast. This research should also investigate

128

the nature and significance if any, of the differences in attitudes toward AAL
based on whether respondents are college students or already in the workforce. It
is recommended similar research be conducted in Africa using the same research
approach and analytical models

. Research is also needed with a larger sample of African immigrants in the US in
order to exhaustively replicate the present study using similar data analyses such
as Ratings of Means and Rankings, Correlation Coefficient and Factor Analysis
available in the Statistical Package for Social Sciences (SPSS). This study should
also utilize both the matched-guise technique and the semantic differential scale
and should also test for any significant differences based on region, gender, age,

and years spent in the USA, level of education, etc.

. Future research Should help to determine the language attitudes of both teachers in

higher education and those of prospective employers in various US business
establishments. This should help alleviate the stereotype that speaking non-
standard varieties such as AAL and other minority languages is detrimental to
one’s career and that SAE is the only language needed to obtain gainful

employment and more equity as both global and US citizens.

. Additional research that includes surveys and evaluations should be incorporated

in programs that include immigrants from around the world along with both AAL
and SAE speakers that help train teachers across various school curricula in the

US, preferably from elementary school through to college.

129

. Research is also needed to help monitor language use among immigrants and
speakers of both SAE and AAL, to help measure allocation of resources in
advancing multilingualism in US schools.

Broaden language attitude studies to include writing, speaking and other forms of
communication in order to examine how AAL’s phonology, syntax, lexicon,
discursive strategies, etc. impact attitudes of both non—users and users toward the
variety.

. Maximize the benefits of multidimensional research approaches (i.e. diversify
methodology to include both qualitative (e. g. in-depth interviews) and
quantitative (e. g. structured questionnaires) methodologies.

. Research is needed to understand how language attitudes toward AAL among
other varieties are mediated and impacted by various forms of emerging/evolving
“new media” available on/off the www (i.e. Chat rooms, You Tube, My Space,

Cell Phone technology, iPods, etc.)

The aim of this study was to determine the nature of attitudes of African immigrants in

the US toward AAL. The findings of the study suggest there is immense potential for

replicating them in future research. The study also lends itself to more extensive Pan-

Africanist language attitude studies that are focused on various attitudes toward

languages spoken in Africa and its Diaspora. Such an approach will be following the Pan

African linguistic studies pioneered by Turner (1949), Beryl Bailey (1965) and

Smitherman (1977, 1986, 2000, et a1) among other scholars and activists. Because our

area of study is in the US among African immigrants, we note that many language

attitude studies have in the past mainly focused on European and very rarely and/or

130

positively on African languages. These studies have also been conducted separately
either among Africans on the Continent or among both AAL users and nonusers in the
USA. Major language attitude studies toward AAL among Africans on the Continent or
as immigrants in the US are largely nonexistent. This study hopes to begin to fill this
research gap.

Finally, the findings in this study are likely to help counter both the reproduction and
maintenance of negative attitudes by Africans and African Americans toward one
another. The results might also assist in helping to address negative academic discourses
that have traditionally occupied themselves in conceptualizing AAL and African
languages “within the framework of a linguistic deviant model.”

Constructive language attitude studies should continue to be more methodologically
sophisticated and to proactively re-assess research both in AAL and in African languages.
This should be combined with consistent advocacy and meaningful scholarly
contributions in the preservation of these languages. This type of scholarship and
activism will also most likely help encourage both users and non-users of marginalized
languages globally, to have favorable attitudes toward these languages as well as their
own. It can also positively impact and add value toward more meaningful forms of
language education and tolerance manifested in more peaceable citizenship and co-
existence. This is critical given the multiple challenges posed by the numerous
significant global socio-cultural, political and economic dynamics present within an

increasingly multilingual world.

131

APPENDICES

APPENDIX A: QUESTIONNAIRE USED IN THE ELICITATION OF
LANGUAGE ATTITUDES
Instructions
All your responses will be treated confidentially. Thank you very much for participating
in this survey.

Please note that this questionnaire is in three parts

Part 1: Please listen to the twelve speakers that you will hear in the recordings.

In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed. For example, consider the
characteristic of friendliness:

l 2 3 4 5 6 7
Friendly unfriendly

 

If you think the speaker sounds “extremely or very fiiendly,” you would rate him/her a
“1.” If you think the speaker sounds “extremely or very unfriendly,” you would rate the
speaker a “7.” If you think the speaker falls somewhere in between “extremely or very
friendly” and “extremely or very unfriendly,” you would use the numbers from 2 to 6 to
indicate the degree of friendliness or unfiiendliness that you think the speaker’s voice
represents.

After your rating of the voices, please try to identify each speaker’s language pattern
as African American Vernacular English or Standard English (Part 11).

Finally, please provide information about yourself (Part 111).

Voice # 1

Please listen to the speaker that you will hear in the recording.

In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed.

1 2 3 4 5 6 7
Well-educated not well educated

 

1 2 3 4 5 6 7
Confident not confident

 

l 2 3 4 5 6 7
Responsible not responsible

 

l 2 3 4 5 6 7
Gentle not gentle

 

132

l 2 3 4 5 6 7
Friendly not friendly

 

1 2 3 4 5 6 7
Polite not polite

 

 

Leader-like not leader-like

 

Patient not patient

 

Proud not proud

 

High status low status

 

Trustworthy not trustworthy

 

Social - not Social

Voice # 2

Please listen to the speaker that you will hear in the recording.

In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed.

1 2 3 4 5 6 7
Well-educated not well educated

 

l 2 3 4 5 6 7
Confident not confident

 

1 2 3 4 5 6 7
Responsible not responsible

 

l 2 3 4 5 6 7
Gentle not gentle

 

l 2 3 4 5 6 7
Friendly not friendly

 

133

l 2 3 4 5 6 7
Polite not polite

 

l 2 3 4 5 6 7
Leader-like - not leader-like

 

l 2 3 4 5 6 7
Patient not patient

 

l 2 3 4 5 6 7
Proud not proud

 

1 2 3 4 5 6 7
High status low status

 

1234567

 

Trustworthy not trustworthy

1 2 3 4 5 6 7
Social not Social

 

Voice # 3

Please listen to the speaker that you will hear in the recording.

In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed.

1 2 3 4 5 6 7
Well-educated not well educated

 

1 2 3 4 5 6 7
Confident not confident

 

1 2 3 4 5 6 7
Responsible not responsible

 

l 2 3 4 5 6 7
Gentle not gentle

 

l 2 3 4 5 6 7
Friendly not friendly

 

1 2 3 4 5 6 7
Polite not polite

 

134

l 2 3 4 5 6 7
Leader-like not leader-like

 

1 2 3 4 5 6 7
Patient not patient

 

l 2 3 4 5 6 7
Proud not proud

 

1 2 3 4 5 6 7
High status low status

 

l 2 3 4 5 6 7
Trustworthy not trustworthy

 

1 2 3 4 5 6 7
Social not Social.

 

Voice # 4

Please listen to the speaker that you will hear in the recording.

In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed.

1 2 3 4 5 6 7
Well-educated not well educated

 

l 2 3 4 5 6 7
Confident not confident

 

l 2 3 4 5 6 7
Responsible not responsible

 

1 2 3 4 5 6 7
Gentle not gentle

 

1 2 3 4 5 6 7
Friendly not friendly

 

1 2 3 4 5 6 7
Polite not polite

 

l 2 3 4 5 6 7
Leader-like not leader—like

 

1 2 3 4 5 6 7
Patient not patient

 

135

1 2 3 4 5 6 7
Proud not proud

 

l 2 3 4 5 6 7
High status low status

 

1234567

Trustworthy not trustworthy

 

l 2 3 4 5 6 7
Social not Social

 

Voice # 5

Please listen to the speaker that you will hear in the recording.

In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed.

1 2 3 4 5 6 7
Well-educated not well educated

 

l 2 3 4 5 6 7
Confident not confident

 

l 2 3 4 5 6 7
Responsible not responsible

 

Gentle not gentle

 

Friendly not friendly

 

Polite not polite

 

Leader-like not leader-like

 

Patient not patient

 

Proud not proud

 

136

 

High status low status

1 2 3 4 5 6 7
Trustworthy not trustworthy

 

1 2 3 4 5 6 7
Social not Social

 

Voice # 6

Please listen to the speaker that you will hear in the recording.

In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed.

1 2 3 4 5 6 7
Well-educated not well educated

 

l 2 3 4 5 6 7
Confident not confident

 

1 2 3 4 5 6 7
Responsible not responsible

 

Gentle not gentle

 

Friendly not friendly

 

Polite not polite

 

Leader-like not leader-like

 

Patient not patient

 

1 2 3 4 5 6 7
Proud not proud

 

l 2 3 4 5 6 7
High status low status

 

137

1 2 3 4 5 6 7
Trustworthy not trustworthy

 

1 2 3 4 5 6 7
Social not Social

 

Voice # 7

Please listen to the speaker that you will hear in the recording.

In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed.

1 2 3 4 5 6 7
Well-educated not well educated

 

l 2 3 4 5 6 7
Confident not confident

 

l 2 3 4 5 6 7
Responsible not responsible

 

l 2 3 4 5 6 7
Gentle not gentle

 

l 2 3 4 5 6 7
Friendly not friendly

 

l 2 3 4 5 6 7
Polite not polite

 

Leader-like not leader-like

 

Patient not patient

 

Proud not proud

 

High status low status

 

Trustworthy not trustworthy

 

Social not Social

 

138

Voice # 8

Please listen to the speaker that you will hear in the recording.

In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed.

1 2 3 4 5 6 7
Well-educated not well educated

 

l 2 3 4 5 6 7
Confident not confident

 

1 2 3 4 5 6 7
Responsible not responsible

 

l 2 3 4 5 6 7
Gentle not gentle

 

1 2 3 4 5 6 7
Friendly not friendly

 

1 2 3 4 5 6 7
Polite not polite

 

Leader-like not leader-like

 

 

Patient not patient

Proud not proud

 

High status low status

 

 

Trustworthy not trustworthy

 

Social not Social

139

Voice # 9

Please listen to the speaker that you will hear in the recording.
In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed.

1 2 3 4 5 6 7
Well-educated not well educated

 

1 2 3 4 5 6 7
Confident not confident

 

l 2 3 4 5 6 7
Responsible not responsible

 

1 2 3 4 5 6 7
Gentle not gentle

 

1 2 3 4 5 6 7
Friendly not friendly

 

l 2 3 4 5 6 7
Polite not polite

 

 

Leader-like not leader-like

 

Patient not patient

 

Proud not proud

 

High status low status

 

Trustworthy not trustworthy

 

Social not Social

140

Voice # 10

Please listen to the speaker that you will hear in the recording.
In the space provided, rate the speaker on the seven—point scale that represents your
evaluation of the speaker for the characteristic listed.

1 2 3 4 5 6 7
Well-educated not well educated

 

l 2 3 4 5 6 7
Confident not confident

 

1 2 3 4 5 6 7
Responsible not responsible

 

1 2 3 4 5 6 7
Gentle not gentle

 

1 2 3 4 5 6 7
Friendly not fiiendly

 

 

Polite not polite

Leader-like not leader-like

 

 

Patient not patient

 

Proud not proud

 

High status low status

Trustworthy not trustworthy

 

Social not Social

 

141

Voice # 11

Please listen to the Speaker that you will hear in the recording.
In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed.

Well-educated

234567

 

not well educated

234567

 

Confident

6

not confident

 

Responsible

Gentle

6

 

Friendly

Polite

Leader—like

Patient

Proud

High status

Trustworthy

Social

 

 

 

 

 

 

 

7
not responsible
7
not gentle
7
not friendly
7
not polite
7
not leader-like
7
not patient
7
not proud
7
low status
7
not trustworthy
7
not Social

 

142

Voice # 12

Please listen to the speaker that you will hear in the recording.
In the space provided, rate the speaker on the seven-point scale that represents your
evaluation of the speaker for the characteristic listed.

1 2 3 4 5 6 7
Well-educated not well educated

 

1 2 3 4 5 6 7
Confident not confident

 

1 2 3 4 5 6 7
Responsible not responsible

 

1 2 3 4 5 6 7
Gentle not gentle
1 2 3 4 5 6 7
Friendly not friendly

 

 

 

Polite not polite

Leader-like not leader-like

 

Patient not patient

 

Proud not proud

 

High status low status

 

Trustworthy not trustworthy

 

Social not Social

 

143

Part 11
Please match each voice with its owner. Indicate if you think the voice is that of a speaker
of African American Language (AAL) or speaker of Standard American English Speaker
(SAE).

Voice # 1
Voice # 2
Voice # 3
Voice # 4
Voice # 5
Voice # 6
Voice # 7
Voice # 8
Voice # 9
Voice # 10
Voice # 11
Voice # 12

 

 

 

 

 

 

 

 

 

 

 

 

Part III
Please provide the following information about yourself.
1. Gender (please circle)
a) Male
b) Female
2. Place of region (please circle)
a) East Africa
b) West Africa
c) Central Africa
(1) Southern Africa
3. Age (please circle)
a) 18-21
b) 22-25
c) 26—29
CI) 30-33
e) 34-39
f) 40 and over
Number of years you have lived in the US
Which languages do you commonly speak?
What is your occupation?
If you’re in college, what is your major?
If a student, what is your status in college?
a) Freshman
b) Sophomore
c) Junior
(1) Senior
e) Graduate

 

 

 

 

P°>'.°‘.U‘:P‘

Thank you very much for participating in this research.

144

Christopher Kuria Githiora

Department of English

201 Morrill Hall, Michigan State University
East Lansing, MI, 48824

Telephone: 517-355-7572

Personal Telephone: 517-353-0975

Email: gIthiorl@msu.edu

145

APPENDIX B: Informed consent document

The purpose of this research is to find out about African immigrants’ attitudes toward
African American Language (AAL) in Lansing and East Lansing, Michigan. The study
is being conducted by the researcher as part of the fulfillment of the requirements for a
Ph.D. degree in Language Studies and Sociolinguistics, in the Department of English at
Michigan State University.

There are no known risks associated with participation in this study. Your
participation is entirely voluntary. While you will not benefit from participating in this
study, your participation in this study may contribute to the understanding of
relationships between Africans and African Americans. You are being requested to listen
carefully to the recorded voices and then complete the accompanying questionnaire to the
best of your knowledge and ability. Please listen to the four speakers that you will hear in
the recordings. In the space provided, rate the speaker on the seven-point scale that
represents your evaluation of the speaker for the characteristic listed. You are welcome
to ask any questions to the researcher at any time during and after the course of the
survey. You can also withdraw from participating at any point during the survey without
being penalized. You are also free to complete only those parts of the questionnaire you
feel comfortable with. The time for participating in the study is thirty minutes or less.

All your responses will be treated confidentially and no personal names or addresses
will be required and/or used in this study. Your privacy will be protected to the
maximum extent allowable by law. If you have any questions about this study, please
contact the Responsible Project Investigator: Dr. Geneva Smitherman, University
Distinguished Professor, English Department, Michigan State University, 221 Morrill
Hall, East Lansing, MI 48824-1036, Telephone: 517-353-9252;Email
smither4@@su.edu.

In addition, if you have any questions or concerns regarding your rights as a study
participant, or are dissatisfied at any time with any aspect of this study, you may contact -
anonymously, if you wish - Peter Vasilenko, Ph.D., Director of the Human Subject
Protection Programs at Michigan State University, by phone: (517) 355-2180, fax: (517)
432-4503, email: irb@msu.edu, or regular mail: 202 Olds Hall, East Lansing, MI 48824.

 

Thank you very much for participating in this study

Christopher Kuria Githiora

1634H Spartan Village, East Lansing, MI. 48823
Telephone: 517-353—0975

Email: githiorl @msu.edu

 

I voluntarily agree to participate in the study.
Name

Signature
Today’s Date

 

 

 

146

REFERENCES
Achebe, Chinua. A Man of the People. London: Heinemann, 1966.
---. Home and exile. Oxford & New York: Oxford UP, 2000.
Adegbija, E. “Language Issues in Post-Literacy and Basic Level Education: Three.
Nigerian Experiences and Their Implications,” L.A.U.D., Series B: Applied and
Interdisciplinary Papers, No. 219: Duisburg, 1992.

---. Language Attitudes in Sub-Saharan Africa. Multilingual Matters Ltd.
Philadelphia: Clevedon, 1994.

---. “Language attitudes in West Africa. International Journal of the Sociology of
Language (141) 2000: 75—100.

Agheyisi, R. & F ishman, J. “Language attitude studies: A brief survey of methodological
approaches.”Anthropological Linguistics, Vol. 17(5), 1970: 137-157.

Akere, F. “Language use and language attitudes in a Yoruba suburban town: A
sociolinguistic response to the factors of traditionalism and modernity.”
Anthropological Linguistics, Vol. 24 (3) 1982: 344-362.

---. “Language Attitudes in West Africa.” International Journal of the Sociology of
Language, (141) 2001: 75-100.

Angongo, R. & Hancock, 1. “English in Africa: Emerging Standards or Diverging
Regionalisms?” In English World- Wide: A Journal of Varieties of English 1 (1)
1980: 67-96.

Asante, Molefi Kete. Afrocentricity, rev. ed., Trenton, N.J.: Africa World Press, 1988.

---. Kemet, Afrocentricity, and knowledge. Africa World Press, 1990.

Bailey, Beryl L. “Toward a New Perspective in Negro English Dialectology.” American
Speech, Vol. 40 (1965): 171-177.

Bailey, L. & GOrlach, M. (Eds.). English as a World Language. Ann Arbor: UM Press,
1982.

Baker, Collins. Attitudes and Language. Clevendon. Philadelphia: Multilingual Matters
Ltd., 1992.

Bakhtin, M.M. The dialogic imagination. Austin: U of TX Press, 1981.

147

Baldwin, James. If Black English Isn't a Language, Then Tell Me, What Is? The New
York Times, Op-Ed Section, July 29, 1979
<http://www.nytimes.comfbooks/98/03/29/specials/baldwin-english.html>

---. “My Quarrel with the English language.” In A Prophet Is Not Without Honor.
Talking at the Gates: A Life of James Baldwin by James Campbell. Review author[s]:
Ekwueme Michael Thelwell. Transition, No. 58 (1992): 90-113.
<http://www.jstor.org/view/0041 1 l91/dm994470/99p00592/0>

 

Bamgbose, Ayo, Ayo Banjo, and Andrew Thomas. Eds. New Englishes: A West African
Perspective. Ibadan, Nigeria: Mosuro, c1995.

Baugh, John. Black Street Speech. Austin: UTX P, 1983.

Berger, Peter & Thomas Luckman. The Social Construction of Reality: A Treatise in the
Sociology of Knowledge. Garden City, NY: Doubleday, 1966.

Berry, Jack & Greenberg, Joseph H. “Sociolinguistic Research in Africa.” African
Studies Bulletin, Vol. 9 (2) Sept. 1966:1-9.

Bokamba, Eyamba G. “The Africanization of English.” In Kachru, B.B. (Ed.). The
Other Tongue. English Across Cultures. Urban Champaign: U of Illinois Press, 1981:
77-98. ‘

Bourdieu, Pierre. Language and Symbolic Power. Trans. Gino Raymond and Matthew
Adamson. Ed. John B. Thompson. Cambridge, MA: Harvard UP, 1991.

Bryman, Alan & Cramer, Duncan. Quantitative Data Analysis with SPSS 12 and 13.
A Guide for social scientists. London and New York: Routledge Publishers, 2005.

Callan, Victor J. & Gallois, Cynthia. “Anglo-Australians’ and Immigrants’ Attitudes
toward Language and Accent: A review of Experimental and Survey Research.”
International Migration Review, Vol. 1 (21) spring 1987: 48-69.

Ciapopo, J .T. and RE. Petty. “Language variables, attitudes, and persuasion.” Attitudes
Toward Language variation: Social and Applied Contexts. Ed. Ryan, EB. and Howard
Giles. London: Edward Arnold, 1982. 189-207.

Delpit, L., & Dowdy, J .K. Eds. The skin that we speak: Thoughts on language and
culture in the classroom. New York: The New Press, 2002.

leite, Paulin G. “Shifts in linguistic identities in a global world.” Language Problems
and Language Planning, Vol. 30 (2006): 1—20.

148

Dillard, J .L. Black English. Random House, 1972.

Diop, Cheikh A. Cultural Unity of Black Africa: The Domains of Patriarchy and of
Matriarchy in Classical Antiquity. Chicago: Third World Press, 1978.

Du Bois, W.E.B. The souls of Black folk. Introduction by John Edgar Wideman. New
York: Vintage Books/The Library of Congress, 1990.

Dwyer, David J. An Introduction to West African English. African Studies Center,
Michigan State U., 1969.

Edwards, Walter F. “Sociolinguistics behavior in a Detroit inner-city black
neighborhood.” Language in Society, Vol. 21 (1992): 98-115.

F abrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. “Evaluating
the use of exploratory factor analysis in psychological research.” Psychological

Methods (4)1999: 272-299.

F arb, Peter. “Linguistic Chauvinism.” Word Play: What Happens When People Talk.
New York: Vintage Books, 1973. 138-164.

F airclough, Norman. Language and Power (2nd Ed.). Harlow, England: Pearson Ed. Ltd,
2001.

F anon, F rantz. The Wretched of the Earth; pref Jean-Paul Sartre. Trans. Constance
Farrington Harmondsworth, UK: Penguin, 1967.

---. “The Negro and Language.” In Black skin, white masks. Trans. Charles Lam
Markmann. London: Pluto, 1986, c1967. 17-40.

F ishman, J. Bilingualism in the barrio. Bloomington: IN UP, 1971.

Giles, H. “Patterns of Evaluation in reaction to RP, South Welsh and Somerset accented
speech.” British Journal of Social and Clinical Psychology, Vol. 10 (1971): 280-281.

Gilroy, Paul. The Black Atlantic: modernity and double consciousness. Cambridge, MA:
Harvard, 1993.

Githiora, Cege. “Sheng: peer language, Swahili dialect or emerging Creole?” Journal of
African studies, Vol. 15 (2), (2002): 159-181.

Graff, David, William Labov, and Wendell A. Harris. “Testing listener reactions to
Phonological markers of ethnic identity: A new method for sociolinguistic research.”
Diversity and diachrony. Ed. Sankoff, David. Amsterdam: Benjamins, 1986. 45-58.

149

Green, Lisa J. African American English: A Linguistic Introduction. Cambridge, UK:
Cambridge UP, 2002. 216-244.

Harper, Frederick D., Kisha Braithwaite, and Ricardo D. LaGrange. “Ebonics and
Academic Achievement: The Role of the Counselor.” Journal of Negro Education,
Vol. 67 (Winter 1998): 25-33.

Hinnebusch, Thomas J. “Bantu” in J. Bender-Samuel (Ed.). The Niger-Congo Languages.
Lanham, MD: UPenn Press, 1980.

Hymes, Dell H. Foundations in Sociolinguistics, Philadelphia, PA: U of PA P, 1974.

Ibrahim, A. “‘Whassup, homeboy?” Joining the African Diaspora: Black English as a
symbolic site of identification and language learning.’ Black Linguistics: Language,
Society, and Politics in Africa and the Americas. Ed. Sinfree Makoni, Geneva
Smitherman, Ametha F. Ball, Arthur K. Spears. London and New York: Routledge,
2003. 169-185.

Jackson, Jennifer V. & Mary E. Cothran. “BLACK VERSUS BLACK: The Relationships
Among African American and African Caribbean Persons.” Journal of Black Studies,
Vol. 5 (May 2003): 576-604.

Karenga, Maulana. Introduction to Black Studies, 2nd ed., Los Angeles: University of
Sankore Press, 1993.

Labov, William. The Social Stratification of English in New York City. Washington,
DC: CAL, 1966.

---. “The logic of non-standard English. Language in the inner city: Studies in the Black
English Vernacular. Philadelphia: UPenn. Press, 1972. 201-240.

Lambert, Wallace E. at al. “Evaluational reaction to spoken language.” Journal of
Abnormal Social Psychological Psychology. Vol. 60 (1960): 44-51.

---. and Otto Klineberg. Children's views of foreign peoples; a cross-
national study. New York: Appleton-Century-Crofts [1966, c1967].

---. & Tucker, G. R. The Bilingual Education of Children. Rowley, MA:
Newbury House, 1972.

LeMay, Michael C. US. immigration: a reference handbook. Santa Barbara, CA: ABC-
CLIO, 02004.

Linguistics Society of America (LSA). Resolution on the Oakland “Ebonics ” Issue.
Unanimously Adopted at the Annual Meeting of Chicago, Illinois, January 3, 1997
<http://www.-personal.umich.edu/~ilawler/ebonics.lsa.html>

150

Lippi-Green, R. English with an Accent: Language, Ideology and Discrimination in the
United States. London: Routledge, 1997.

Makalela, Leketi. “Language Harmonization in South Africa: Practices and Attitudes of
University Students.” Unpublished PhD. Dissertation, Michigan State University,
2005.

Marable, William M. “Afiican Nationalist: The Life of John Langalibalele Dube.” PhD.
Dissertation, University of Maryland, 1976.

Mazrui, Alamin M. “Linguistic dilemmas of Afrocentricty: The Diaspora experience.
Language and Ideology: Cognitive Theoretical Perspectives. In Current Issues in
Linguistic Theory Series. Ed. Rene Dirven, Bruce Hawkins and Esra Sandikclioglu.
Philadelphia: John, 2001. 141-164.

Mazrui, Ali A. & Alamin M. Mazrui. The Power of Babel: Language and governance in
the African experience. Oxford: James Currey, 1998.

Mbaabu, Ireri. Language Policy in East A fiica: A Dependency Theory Perspective.
Nairobi: Educational Research and Publications, 1996.

Mbiti, John S. Afiican Religions and Philosophy. New York: Doubleday, 1969.
Mboya, Tom. The Challenge of Nationhood: A Collection of speeches and writings by
Tom Mboya. Fwd. by Jomo Kenyatta, President of the Republic of Kenya. Postscript
by Mrs. Pamela Mboya. New York & Washington: Praeger Publishers, 1970.
Milroy, Lesley. Language and social networks. Oxford: Blackwell, 1980.

Moag, Rodney F ., Albert J. Schiitz, and Paul Geraghty. Ed. Albert J. Schiitz. Fijian
language studies: borrowing and pidginization. Suva, Fiji: Fiji Museum, 1982, c1978.

Morgan, Marcelina. “The Africanness of Counter- Language among Afro-Americans.”
Ed. Mufwene, S.S. Africanisms in Afro-Americans Language Varieties. Athens, GA:
UGA Press. 1993.

Mr’rtonya, Mungai. “Attitudes of Educated Africans Toward Regional Varieties of
English.” Unpublished MA Thesis, Michigan State University, 1997.

Mwangi, Evan. Guide to Encounters from Africa. New York: Longhorn, 2002.

Ngt'rgi wa Thiong’o. Decolonising the Mind: The Politics of Language in African
Literature. Portsmouth, NH: Heinemann, 1986.

---. Moving the Centre: The Struggle for Cultural Freedoms. London: James Curry, 1993.

151

Niedzielski, Nancy. “Attitudes Toward Midwestern American English.” In Niedzielski,
Nancy & Dennis R. Preston. Folk Linguistics. Berlin; New York: Mouton de Gruyter,
2000, c1999. 217.

Njogu, Kimani. “Language key tool in changing societies.”
East African Standard online Edition, Monday September 18, 2006:
<http://www.eastandard.net/archiveslcl/print/news.php?articleid=1 143958347>

NoruSis, MJ. SPSS 13.0 guide to data analysis. Upper Saddle River, NJ: Prentice Hall,
2005.

Obama, Barack. Dreams from my father: a story of race and inheritance. New York:
Times Books & Random House, c1995.

Odinga, Ajuma Oginga. Not yet Uhuru: the autobiography of Oginga Odinga. Fwd.
Kwame Nkrumah. London: Heinemann, 1967.

Ogawa, David. “Small Group Stereotypes of Black Americans.” Journal of Black
Studies, Vol. 1 (1971): 273-81.

Preston, Dennis R. “Talking black and talking white. A study in variety imitation.” Old
English and new: Papers in honor of Frederic G. Cassidy on the occasion of his 85"2
birthday. Ed. Joan H. Hall, Nick Doane, and Dick Ringler. New York: Garland, 1992.
327- 55.

Ruether, Katie. “A Study of the Effects of Awareness of Black English on Language
Attitudes.” February 28, 2006 <http://www2.truman.edu/~mshapiro/Reuther.html>

Rickford, J. R and Penelope Eckert Eds. Style and sociolinguistic variation.
Cambridge, UK: Cambridge UP, 2001.

---. Afiican American vernacular English: features, evolution, educational
implications. Malden, MA: Blackwell Publishers, 1999.

--- and Russell John Rickford. Spoken soul: the story of Black English. Fwd. by Geneva
Smitherman. New York: Wiley, 02000.

Roberts, Sam. “More Afiicans Enter US. Than in Days of Slavery.” New York Times
(February 21, 2005). May 4, 2005

<http://www.amren.com/mtnews/archives/2005/02/more_africans_e.php>

 

Reidlinger. Trans. From French by Wade Baskin. Newy York: Philosophical Library,
1959.

152

Schmied, J. “Language use, attitudes, performance and sociolinguistic background: A
Study of English in Kenya, Tanzania and Zambia.” In English World Wide. 11
(2) 1990: 217-238.

Smitherman, Geneva. T alkin and T estijfvin: The Language of Black America. Boston:
Houghton Mifflin, 1977.

---. “Black English ” and the Education of Black Children and Youth: Proceedings of
the National Invitational Symposium of the King Decision, 2-13 February. Ed. And
Introduction by Smitherman, Geneva. Detroit: Center for Black Studies, Wayne State
University, 1981.

Smitherman-Donaldson, Geneva. “Discriminatory discourse on Afro-American speech.”
Discourse and discrimination. Ed. Geneva-Smitherman-Donaldson and Teun A. van
Dijk. Detroit: Wayne State UP, 1988.144-175.

---. “Discourse, Ethnicity, Culture and Racism” (in collaboration with Teun A. van Dijk,
Stella Ting- Toomey and Denise Troutrnan). Discourse as Social Interaction. Ed. Teun
A. van Dijk. Thousand Oaks, CA: Sage, 1997.144-180.

---. “ ‘The Chain Remain the Same’: Communicative Practices in the Hip-Hop Nation,”
Journal of Black Studies, September 1997.

--. “From ‘Hujambo to Molo’: The Study of and interest in Afiican Languages among
African Americans.” In K.K. Prah (Ed.), Between Distinction and Extinction (243-253).
Johannesburg: Witwatersrand UP, 1998.

---. “African Americans, Ebonics and US. Language Planning-Policy.” Comparative
Perspectives on Language and Literacy. Ed. Leslie Limage. Dakar, Senegal: UNESCO,
1999. 51-64.

---. T alkin The Talk: Language, Culture and Education in African America.
London: Routledge Publishers, 2000.

---. “A Commentary on Ebonics: From a Ghetto Lady Turned Critical Linguist.” Ebonics
and Language Education. Ed. Clinton Crawford. New York and London: Sankofa
World Publishers, 2001. 214-234.

---. “Language and Democracy in the United States of America and South Africa.”

Language and Institutions in Africa. Ed. Sinfree B. Makoni and Nkonko M.
Kamwangamalu. Cape Town: Centre for Advanced Studies of African Society, 2001.

---. Word from the Mother. New York: Routledge, 2006.

153

Shuy, Roger W. & Ralph W. F asold (Eds.). Language Attitudes: Current Trends and
Prospects. Washington, DC: Georgetown UP, 1973. 174-201.

Singler, John. An introduction to Liberian English. With J. Gbehwalahyee Mason [et
al]. [East Lansing]: Peace Corps/ African Studies Center, Michigan State University,
1981.

Taylor, Hanni U. Standard English, Black English, and Biadelictalism. A Controversy.
New York: Peter Lang, 1989.

Taylor, Orlando L. “Teachers’ attitudes toward black and nonstandard English as
measured by the language attitude scale.” Language Attitudes: Current Trends and
Prospects. Roger W. Shuy and Ralph W. F asold (Eds.) 174-201. Washington, DC:
Georgetown UP, 1973. 174-201.

Thiba, Tumelontle, M. “The Language Attitudes and Practices of High School Learners,
Teachers and Parents in the Mafikeng and Vryburg Districts of the North West
Province, South Africa.” Unpublished PhD. Dissertation, Michigan State University,
2000.

Trudgill, Peter. “Sex, covert prestige and linguistic change of the urban British English of
Norwich.” Sociolinguistics: An Introduction to Language and Society.
Harmondsworth: Penguin (1983), 215-46.

Tucker, G. Richard and Wallace E. Lambert. “White and Negro Listeners’ Reactions
to Various American English dialects.” In F ishman, Joshua A. (Ed.), Advances in
Sociology of Language, Vol. II. The Hague and Paris: Mouton & Co. Publishers,

1972. 178-184.

Turner, L.D. Africanisms in the Gullah Dialect. Chicago: U of Chicago P., 1949.

Van Sertima, Ivan. “My Gullah Brother and 1: Exploration into a Community’s Language
and Myth through its Oral Tradition.” Black English: A Seminar. Ed. Harrison, D.S.
and Trabasso, T. Hillsdale, NJ: Erlbaum Associates, 1976.

Walcott, Derek. ‘A Far Cry from Africa’ in Collected Poems 1948-1984. New York: The
Noonday Press, 1986. 18.

---. The Bounty. New York: F arrar, Strauss and Giroux, 1997. 19.

Walker, Alice. The Color Purple: A Novel. New York and London: Harcourt Brace
Jovanovich, 1982.

Walker, Alice. The Color Purple. San Francisco: Aunt Lute Books, 1996, 59-60.

154

Williams, Robert L. Ed. Ebonics: The True Language of Black Folks. St. Louis, MO: The
Institute of Black Studies, 1975.

Williamson-Ige, Dorothy K. “Approaches To Black Language Studies: A Cultural
Critic.” Journal of Black Studies, Vol. 15 (1) Sept. 1984:17-29.

Wilson, Clint et a1. Racism, Sexism and the Media: The Rise of Class Communication in
Multicultural America. Thousand Oaks: Sage, 2003.

---. Media Research: An Introduction (Sixth Ed.) Belmont, CA: Wadsworth Publishing
Co., 2000. 390-395.

Wolfram, Walt and Natalie Schilling-Estes. American English: dialects and variation.
Malden, MA: Blackwell Pub. 2006.

Woods, David R. “Attitudes toward French, national languages, and mother tongues
across age and sex in Congo.” In Akinbiyi Akinlabi (Ed.). Theoretical Approaches to
African Linguistics. Trenton, NJ: African World Press, 1995. 403-15.

155

llllllllllfllfllllljjjlljlllljl|