k.

y
.1. . . . ... A. .
H21... . .352...

by...

. 5.... .

£33 a.
.119... ..

.l. 3)}...

.312...ng

.iisvutv. 3
2 \9 41.31.;
i... 1A (t

I v!»
“:3va
xxx}..-

11. I

‘1
.1

 

niifhmt. 1 , , , ‘ ,, 1.9M»......=..J-..u.flugxsxaaafifiiymmimgac

 

 

 

 

This is to certify that the
thesis entitled

I’M NO HOOSIER!
EVIDENCE OF THE NORTHERN CITIES SHIFT IN
ST. LOUIS, MISSOURI

presented by

JILL CHRISTINE GOODHEART

has been accepted towards fulfillment
of the requirements for the

Master of degree in Linguistics
Arts

 

 

 

 

Major Professor’s Signature

5'//3'/o‘/

Date

MSU is an Affirmative Action/Equal Opportunity Institution

—a-u-n-n-o-n--Lugo-o-o-o-n—- —

 

 

 

LIBRARY

Michigan State

University

 

 

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

 

DATE put; a

5

DATE DUE

DATE DUE

 

0 C 2 U6

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClFIC/DateDuepssop. 1 5

 

I’M NO HOOSIER!
EVIDENCE OF THE NORTHERN CITIES SHIFT IN
ST. LOUIS, MISSOURI

By

Jill Christine Goodhean

A THESIS

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

MASTER OF ARTS

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

2004

 

 

IN I

 

is null:
()I only
Illh sm

'1
“’18 CHI
cmpli if
Irtmlir
Sllnltu
Class II
,E nu

ABSTRACT

I’M NO HOOSIER!
EVIDENCE OF THE NORTHERN CITIES SHIFT IN ST. LOUIS, MISSOURI

By

Jill Christine Goodheart

According to the Atlas of North American English, the Northern Cities Shift (NCS)
is making its way to St. Louis, Missouri. This claim is, however, based on a very limited set
of only four speakers. The following thesis, therefore, more fully examines the existence of
this speech phenomenon in St. Louis among young speakers there.

To determine the presence (or absence) of the N CS in St. Louis, acoustic analysis

was conducted on the speech of 29 young St. Louis residents. Statistical tests were

employed to determine the extent of two early stages in the shift: /28/ raising and /(l/

fronting. In addition, two social variables were examined: sex and social class. Results
showed that the NCS is likely in the beginning stages in St. Louis, and that upper middle

class men and boys are resisting the NCS feature which was exhibited by most informants:

/ae/ raising. Surprisingly, however, it is the upper working class men and boys who are
leading :mother early stage in the shift: /a/ fronting.

Why the NCS would be emerging in St. Louis is somewhat puzzling, since it has
traditionally been a unique dialect region. However, attitudinal data from these young St.
Lousians shows that they reject their local dialect, as well as associations with the state of
Missouri. They also reject the nearby Southcm dialect region, which they associated with
being a hoosier, a major insult in the Gateway City. Thus, it appears that these young

residents of St. Louis must look elsewhere to find their linguistic identity.

ACKNOWLEDGEMENTS

This study would have been impossible without the assistance of many. I would
first like to offer sincere thanks to my advisor Dr. Dennis R. Preston for his guidance,
expertise, and patience throughout the course of this project. Thanks also to the other
members of my committee, Drs. Barbara Abbott and Grover Hudson, for their advice and
support. In addition, I would like to thank Judy and Paul Heaton for their generosity,
hospitality, and incredible networking skills during my data collection. And, obviously, this
thesis would never have been completed without the cooperation of many people in St.
Louis who opened their homes, their places of business, and their mouths. Finally, for his
help on some of the more tedious aspects of this project and for his emotional support, I

thank Dave Heaton, who is neither ghetto nor hoosier.

iii

 

 

listol
Listol

Chapt
1.1

1.2
1.3 ,
1.3.1 ;_
1.3.2 |
1.3.2.1
1.3.2.."_
1.3.3

1.3.1

1.1

 

Chaptt
t) 1

$2
2.3
2.1
2.5
€113th
3.1

3.2

3.2.1
3.2.2
3.2.3

CO Q: m
<2 o v - cc to
I ."

nus

co co co 9° go
1 :4 I Q; . ‘-
w w h

h—

‘ Co
.

. -

i ‘- 4

CO C u
C C‘”
,_ . _ .
-0 rg. '--

w
it

 

 

TABLE OF CONTENTS

list of Tables ....................................................................................................... vi

list of Figures. ..................................................................................................... vii
Chapter 1.0 Introduction ................................................................................... 1

1.1 Research Questions and Aims ...................................................... l

1.2 The Northern Cities Shift. .............................................................. 3

1.3 St. Louis ........................................................................................... 7

1.3.1 Claims about St. Louis and the N CS .............................................. 7

1.3.2 St. Louis: linguistically unique from the rest of Missouri historically ..... 8

1 .3.2. l Phonology ......................................................................................... 9

1.3.2.2 “Hoosier” .......................................................................................... 10
1.3.3 Linguistic influences on St. Louis .................................................... 13
1.3.4 St. Louis becoming “Northern”? ...................................................... 16
1.4 Summary ........................................................................................... 17
Chapter 2.0 Methodology .................................................................................... 19
2. 1 Respondents ..................................................................................... 20
2.2 Data Collection ................................................................................. 20
2.3 Social Class ....................................................................................... 22
2.4 Data Analysis .................................................................................... 23
2.5 Summary ........................................................................................... 27
Chapter 3.0 Results .............................................................................................. 28
3.1 /2B/ raising .......................................................................................... 28
3.2 Social factors and /2e/ raising ............................................................ 31
3.2.1 Sex and /2B/ raising ........................................................................... 32
3.2.2 Social class and /ae/ raising ............................................................... 33
3.2.3 Sex & class interaction and /aa/ raising ............................................. 37
3.2.4 Age and /a:/ raising ........................................................................... 39
3.3 Linguistic factors and /28/ raising ...................................................... 39
3.4 /a/ fronting ......................................................................................... 43
3.5 Social factors and /(1/ fronting ........................................................... 47
3.5.1 Sex and /a,/ fronting .......................................................................... 48
3.5.2 Social class /(1/ fronting ..................................................................... 49
3.5.3 Sex & social class interactions and /(l/ fronting ................................. 50
3.6 Discussion .......................................................................................... 5 1
3.6.1 Inconsistencies with regard to social factors ..................................... 51
3.6.2 Inconsistencies with regard to linguistic factors ................................ 52
3.6.3 Possible explanation for inconsistencies ........................................... 53
3.7 Summary ............................................................................................ 54

iv

Chapter 4.0 Language Attitudes ......................................................................... 56

4.1 The local dialect ................................................................................. 56
4.2 St. Louis vs. Missouri ......................................................................... 59
4.3 Hoosiers .............................................................................................. 65
4.4 Hoosier speak ..................................................................................... 69
4.5 Discussion ........................................................................................... 71
4.6 Concluding Remarks .......................................................................... 72
Appendices ............................................................................................................... 74
Appendix A: Index of Social Class ......................................................................... 75
Appendix B: Interview Questions ........................................................................... 77
Appendix C: Word List. .......................................................................................... 78
Appendix D: Reading Passage ................................................................................. 80
Appendix E: Individual Vowel Plots ...................................................................... 81
Appendix F: Supplemental Attitudinal Data. ......................................................... 110
Appendix G: Transcription conventions ................................................................. 115
References ................................................................................................................. 116

LIST OF TABLES

Chapter 2.0

2.1 F1 index of the raising of/ze/ ....................................................................... 26
2.2 F2 index for the fronting of /a/ .................................................................... 27
Chapter 3.0

3.1 /2B/ index scores for all subjects ................................................................... 29
3.2 /33/ F1 index score by sex ............................................................................ 33
3.3 Normalized F1 values of /m/ by social class ................................................ 34
3.4 F1 index scores of /a%/ by class .................................................................... 35
3.5 /a3/ raising index scores for sex and class .................................................... 38
3.6 Influence of following manner of articulation on /tB/ raising across studies... 40

3.7 Influence of following place of articulation on /a=:/ raising across studies... 41
3.8 Influence of preceding place of articulation on /ae/ raising across studies... 42

3.9 Overall results for /a/ fronting ..................................................................... 44
3.10 /a/ fronting individuals via the Labovian measure ....................................... 45
3.11 Normalized F2 values of /(1/ by social class ................................................. 49

LIST OF FIGURES

Chapter 1.0

1.1 Pre-NCS vowel system based on Peterson and Barney (1952) .................... 4
1.2 The Northern Cities Chain Shift. .................................................................. 5
1.3 Boundaries of the St. Louis Speech Island ................................................... 9
Chapter 2.0

2.1 Overview of Sample Design .......................................................................... 19
2.2 Map of the St. Louis, Missouri area .............................................................. 21
2.3 Example of Spectrogram ............................................................................... 24
2.4 Vowels of Laura (individual tokens) ............................................................. 24
2.5 Vowels of Laura (means) .............................................................................. 25
Chapter 3.0

3.1 Vowels of Maria (means) ............................................................................. 30
3.2 Vowels ofJackie (means) ............................................................................. 31
3.3 Percentage of /tB/ raising by class ................................................................. 35
3.4 Normalized Fl of /22/ values for sex and class ............................................. 38
3.5 Cassie’s vowel system ..................................................................................... 46
3.6 Nate’s vowel system ....................................................................................... 47
3.7 F2 values of/O/ in hertz by sex and social class ............................................ 50

vii

1.0 Introduction

1.1 Research Questions and Aims

The Northern Cities Shift (hereafter, NCS) has been called the most robust sound
change currently going on in the US. While there has been and continues to be extensive
work done on the NCS throughout Michigan and Illinois (Callary 1975, Eckert 1989a,
Labov 1994, Habick 1993, Hemdobler 1993, Ito 1999, Gordon 2000 and 2001, Evans
2003, Jones, 2003), the work done examining its presence in St. Louis, Missouri is quite
limited. In fact, St. Louis as a region of linguistic study has been ignored by many, despite
claims that it is a unique dialect area in the middle of robust change. There has been
nothing published examining the N CS in any great depth in St. Louis, despite the evidence
presented in the Atlas of North American English that the N CS is gaining a foothold in the
St. Louis area, although not in the rest of the state. This is, however, based on a very
limited study of St. Louisians, mostly over the telephone (Labov 2003). As there are a
great many sociolinguists who find the N CS to be one of the most significant changes in
American dialects in a century, it is of interest to many in the field that we examine just how
quickly it is spreading and where it is headed, and perhaps gain some insight into why it is
moving to particular regions. In addition, the northeast and southern parts of the US have
received much more attention from linguists than most of the "middle" states, so this study
will contribute to filling in a gap in our knowledge of Midwestern dialects.

For those unfamiliar with Missouri and its politics, attitudes, and diversity, it may
seem as though the state could have linguistic uniformity. In fact, one may even believe
something like "Missouri speech" could possibly exist. Missouri, however, has both

attitudinal and linguistic diversity. Urban and rural areas cannot agree on concealed

weapons; many Missouri residents eschew the University of Missouri and look outside the
state (Iowa, Kansas) for college sports teams to root for, and the two major cities of the
state, Kansas City and St. Louis, cannot even agree on the infamous pop/soda controversy.

St. Louis has the largest metro area in the state and is situated on the
Missouri/Illinois border. And while suburbs continue to sprawl to the north and west, the
city itself tends to be the geographical identifier of its residents (not their particular suburb)
when traveling outside the area.

These linguistic, social, and cultural facts bring up many questions regarding the
current linguistic practices of the people of St. Louis. Are they content to sound like a St.
Louisian? Or do people in the Gateway City instead look outside the state, to cities such as
Chicago, for their linguistic identity? Or do they look towards the rest of Missouri or
surrounding southern states? Ultimately, is there evidence of the Northern Cities Shift in
St. Louis, Missouri which is replacing the receding, but historically unique speech of St.
Louis? If so, how advanced is such a shift? Are there any attitudinal factors that would
increase the likelihood of such a shift in St. Louis? The following study will focus on
evidence of the NCS in young residents of the Gateway City and what regional attitudes
may play a role in its adoption. Additionally, the social factors of sex and class will also be

examined to determine what role they have in the presence (or absence) of the NCS in St.

Louis.

1.2 The Northern Cities Shift
As is well known among sociolinguists, the Northern Cities Shift is an urban sound

change observed primarily in the northeastern part of the United States, noted especially in

Rochester, Syracuse, Detroit, Buffalo, Cleveland and Chicago (Labov 2001: 7). Credit as
the original "discoverer" of this shift has been given to Fasold, from an unpublished 1969
work. He first described the raising of /ae/ and the fronting of /o/ and /o/ in Detroit (Labov

1994: 178).

These shifted vowels (fronted, raised, etc.) are given relative to an older American
English vowel system, such as that described by Peterson and Barney (1952), which has
become a baseline for American English vowel studies. Such a system can be seen in
Figure 1.1. The NCS has been discussed in comparison to this model. The Peterson and
Barney study does not, of course, give us definitive results on exactly how Americans
sounded in 1952, as the research was based on a less-than-ideal set of 76 speakers from
various pans of the US, and a total number oftokens of only 1520. However, since the use
of Peterson and Barney's vowel system is so widespread, it will be used here as the
benchmark of pre-shifted vowel systems. Unfortunately, there is a lack of data on older St.
Louisians' speech. When comparing language changes in progress, researchers in the past
have compared current language use to recordings of much older speakers, such as from
the Dictionary of Regional English (DARE). Unfortunately, the only data collected for
DARE from St. Louis were from African American speakers. And, as the current study
focuses Oil the speech of European Americans, a comparison would be inconclusive. III
addition, Missouri appears to have been neglected in linguistic atlas projects, so there is no
data available from such sources On St. Louis speakers. Therefore, it is an unfortunate gap

in this study that St. Louis speakers from generations gone by are not examined.'

 

' Although there are no European American DARE recordings for St. Louis, it is likely that other recordings of older
generations are available for future researchers of the speech of St. Louis. It is possible, for example, that recordings have
been made for the purpose of documenting oral histories, story telling, or family histories.

2400 2200 2000 1300 1600 1400 1200 1000 800
200 I I I I I I I I I

 

beet

soou boot

400.

500 -

bought
600.

 

700 -

 

hot

 

Figure 1.1: a pie-NCS vowel system based on Peterson and Barney (19.52); these
researchers did not include /o/ or /e/ in their study.

Sociolinguist William Labov (1994: 178, 194) claims the N CS progresses in several

continuous and connected steps. He claims the steps are as follows: the raising of /a:/,
followed by the fronting of /a/, and the lowering and fronting of /o/. Later steps include /I/
lowering, /e/ backing or lowering, and /A/ backing. The claim of Labov is that the NCS is a

chain shift: the space in the vowel system created by the raising of /23/ "pulls" /a/ to a more
fronted position to [ill the void. Subsequently, /:)/ is "pulled" to a lower and more fronted

position to fill the space left by /(1/. Figure 1.2 shows a vowel system with arrows indicating

the movements of the vowels from pre-shifted to shifted positions:

8—-—-—>A———>3

$4— (1

 

Figure 1.2: The Northern Cities Chain Shili; from Gordon (2001: 197)
alter Labov 1991.

There have been some hypotheses that the Northern Cities Shift is not, in fact, a
chain shift (Gordon 1997, 2001). Matthew Gordon, who makes some of the major
arguments against the NCS as a chain shift, does concede that when one of these changes is
occurring, it is accompanied by the other phonetic changes as well, despite the fact that he
says the "relatedness of the NCS vowels can only be partially confirmed“ (Gordon 2001:

196).

Despite this claim, there have been several studies to confirm that, indeed, the
raising of/w/ is a very early step in the N CS (Labov, Yeager, and Steiner 1972, Labov
1994, Eckert 1988, Gordon 2001). There are some claims, however, that the fronting of
/(1/ actually occurs before the raising and fronting of/tB/; thus, /a=:/ is "pushed" by /a/ to its
new position (Gordon 2001: 207). Both /(1/ fronting and and /ae/ raising are examined in

the present study. It is important to note that the raising of/ze/ primarily refers to its onset.

III the most advanced speakers, /£B/ usually raises to close to /1/ position with an inglide.

The resulting vowel is something such as /I°/. Therefore, only the onset will be considered

for the purposes of this study.

Interestingly, this speech phenomenon is (and historically has been) on the move.
Labov calls it "one of the most vigorous sound changes now in progress in the United
States" (1994: 178). His claim is that the N CS began in New York state and made its way

westward to urban areas (2001: 7). Callary's research also supports this sound change as an

urban one; he found that in Illinois, the larger the community, the more raised /a3/ was in

that community (1975). However, there has also been some work done which shows that
in the regions where the N CS is most advanced, it is also making its way to less populous
areas (Ito 1999, Gordon 2001). l

The NCS is a language change in progress that is below the level of consciousness
of its speakers. There are several predictions made about such "change from below.” One
such prediction is that women should lead this change. Much evidence done on N CS has
supported this claim, as females tend to lead males in the use of the innovative NCS forms
(Fasold 1969, Eckert 1988, 1989a). III fact, Labov claims that "women are very much in
the lead" in this particular speech phenomenon (2001: 285).

It is predicted that lower middle class and upper working class speakers are more
likely to embrace changes from below. And while class appears to play a role in the NCS
phenomenon, evidence is not all OII the same page (Fasold 1969, Hemdobler 1993, Eckert
1988, 1989a). It appears, however, that middle class and lower middle class individuals,
especially women in these groups, tend to lead this change. It is also the case that young
people, such as young adults and youth in late adolescence, tend to lead changes from

below, and this is borne out in NCS research as well (Labov 2001, Ito 1999).

The vast majority of research‘done on the NCS has included primarily European
Americans (with the exception of J ones 2003 and Gordon 2000). N CS is considered
primarily (or, at least, initially) a European American phenomenon. This is home out by
Gordon (2000), who did not find African Americans or Mexican Americans participating
in the NCS in a community where the sound change was at the beginning stages. J ones
found some evidence of African Americans using N CS features, but her investigation was
in a region where the shift is well advanced.

According to Labov (1994: 100), there are certain phonetic environments which
promote or favor the raising of /x/ more than others; for example, before word-final apical
nasals (e.g. man). He gives the following hierarchy in regards to the manner of articulation

of the following phone, starting with the most favorable to /a:/ raising: nasals, voiceless

fricatives, voiced stops, voiced fricatives, voiceless stops. He also gives a hierarchy for the
following place of articulation: palatal, apical, labial, velar. Certain proceeding consonant
clusters, however, such as obstruent + liquid (as in the word black), have been shown to

retard the process (Labov 1994, Ito 1999, Jones 2003).

1.3 St. Louis
1.3.1 Claims about St. Louis and the NCS

According to work done on the Atlas of North American English, the Northern
Cities Shift has begun to make its way to St. Louis (Labov 2003). Of the four St. Louis
individuals interviewed by project researchers, three are said to have characteristics of N CS
in their speech. It is important to note that the Atlas of North American English is a

telephone survey, which is not the optimal environment for collecting speech data. Despite

this small data set, Labov makes the claim that "St. Louis is undergoing a massive shift
towards the pattern of the Inland North, including the Northern Cities Shift" (2003).
Another claim is that St. Louis is "transitional" in the spreading of the NCS to more

southern states and the southern expansion of the Northem Cities Shift (Labov 2003).

1.3.2 St. Louis: linguistically unique from the rest of Missouri historically

Traditionally, dialectologists have viewed most of Missouri as a Midlands dialect
area, though the Midlands is often defined by the lack of defining dialect features of those
regions which surround them - a kind of transitional area between the north and the south
(Labov 2003, Murray 2002, Lance 1974, 2003). Most of the time, as in Donald Lance's
dialect map, Missouri is simply divided into North and South Midland. Indeed, Lance
called this particular region 'ambiguously midland" in 1974 (9-10). And while the dialects
of east coast cities such as Philadelphia, New York, and Boston have been studied
extensively, many Midlands cities, such as St. Louis, have been largely neglected. From
what has been published, it appears that all of Missouri does not fit into the same dialect
category. In addition to claims that St. Louis comprises its own unique dialect region and
claims of Missouri being a Midland region, there are also pockets of southern speech in
Missouri (Labov 2003).

Based primarily OII lexicon, and confirmed by an analysis of "pronunciation," Frazer
(1979) argued that the western part of Illinois opposite St. Louis, as well as several Missouri
counties surrounding the city, comprise a "speech island." In his most recent. research,
much of which is over two decades old, this pocket of unique speech was aligned in both

pronunciation and lexicon with the North and North Midlands, "contrasting sharply with

the region speech of the surrounding area" (Frazer 1979: 186). A rendition of this speech

island is given in Figure 1.3:

 

 

  

ILLINOIS

 

 

 

MISSOURI

 

 

 

 

 

Figure I .3. Boundaries of the Northem/North Midland speech island
surrounding St. Louis. (Murray 1.993, based on Frazer 1979: [92111)

It is also notable, however, that Labov and others assert that, although larger regions of the
US are "becoming increasingly differentiated from each other," this does not "apply within
the major regions" (2003). Thus, although there is increased diversity from one region to
another (such as between the North and the South), speech within those larger regions is

becoming more homogenized.

1.3.2.1 Phonology

As anyone would know who has spent much time there, St. Louis has a dialect
characteristic which is quite recognizable, as it is often the butt of jokes about the region.
III fact, Labov calls this St. Louis' "most distinctive traditional feature" (2003). It is the
merger of words such as cord/ card (to card) and for/far (to far) while four remains distinct

(Murray 2002). Since two major highways in St. Louis are interstate 40 and state highway

44, a common way of poking fun at this dialect feature is to exaggerate its presence in the
word "forty" and (incorrectly) in "four.” Due to the fact that this speech phenomenon is
highly stigmatized, however, it appears to be receding (Murray 2002, Labov 2003).
Regardless, this merger is still recognized by most inhabitants of St. Louis and is "strong
enough to act as a defining feature of St. Louis dialect" (Labov 2003). Indeed, every
individual interviewed for the purposes of this study was aware of this speech characteristic,
even though none exhibited it and few said they knew many young people who spoke this
way.

Despite the fact that the cord/card merger appears to be dying out, St. Louis' dialect
still remains distinct, at least from the rest of the state of Missouri; for example, there is
evidence that the widespread caught/cot merger is occurring in nearly every part of the
state, except St. Louis, which keeps this distinction (Majors 2003, Labov 2003). This is

also noteworthy because a resistance to the caught/cot merger is a feature of the Northern

Cities Shift, as the fronting of /(1/ and the fronting and lowering of/o/ are some of the first

stages in the shift.

1.3.2.2 'Hoosier'

Another distinct speech characteristic of St. Louis is not a feature of phonology, but
is a lexical item: the word hoosicr. For most Americans familiar with this term, it is usually
associated with inhabitants of Indiana, as it is Indiana's nickname: The Hoosier State (Graf
2000). For the most Americans, it canies no derogatory overtone (Seely 2003). For some
reason, however, St. Louisians have, in their collective lexicon, a unique definition of

hoosier which aligns itself with terms such as hick or hillbilly. The use of this term in St.

10

Louis has been of interest to researchers for at least a few decades. Crinklaw said in a 1976
article that inhabitants of St. Louis associate hoosier with a "displaced country man who
moves into a city neighborhood and tears it up” (60). Murray found that his informaan
associated this term with characteristics such as "lazy, slow moving, derelict, and
irresponsible" (1987: 3). Historically, the origin of this term is unclear, though there are
dozens of hypotheses. Indiana historian Jeffrey Graf, who has done extensive research on
the temI, explains some of these accounts:

Like bamacles, a thick crust of speculation has gathered over the word ”Hoosier" to
explain the origin of Indiana's nickname. The popular theories, diligently and often
sincerely advanced, form a rich, often amusing body of folklore. Those theories
include: "Who's here P” as a question to unknown visitors or to the inhabitants of a
country cabin; Hussar, from the fiery European mounted troops; 'Huzzah!"
proclaimed after victory in a fight; Husher, a brawny man, capable of stilling his
opponents; Hoosa, an Indian word for corn; Hoose, an English term for a disease
of cattle which gives the animals a wild sort of look; and the evergreen "Who's ear?"
asked while toeing a tom-off ear lying on the bar room floor the morning after a

brawl (2000: 1).

Graf also points out that when lroosier originally came into use, it was probably "a term of
contempt and opprobrium common in the upland South and used to denote a rustic, a
burnpkin, a countryman, a roughneck, a hick or an awkward, uncouth or unskilled fellow"
(2000: 1). And despite the fact that this meaning has largely fallen off, he points out that St.
Louis is an "important pocket of linguistic resistance" to embracing the current use of the
term in Indiana and elsewhere.

Graf points out that lroosier in its original sense may be heard infrequently by those

familiar with the term. However, it appears to be quite widespread in the Gateway City. In

11

fact, Murray points out in his 1987 study of lroosier that "few epithets in St. Louis carry the
pejorative social connotations or the potential for eliciting negative responses that hoosicr
does" (3). In the current study, and all but one of the respondents in the entire corpus of
47 were very familiar with the term, and all associated it negatively. The most common
synonyms given were: hick, hillbilly, white trash, redneck, and country person. In
Murray's study, hoosicr was the insult of choice for white males, though its use for women
and African Americans was also robust. He defined St. Louisians' use of hoosicr with the
following analogies:

If the driver of another car swerves in front of a St. Louisian who is also driving, the
person who swerved is a Hoosier. Similarly, if someone attends a social event or

even merely appears in public and is inappropriately underdressed, that person is a
Hoosier. In short, any person whose behavior is perceived as nonstandard by a St.

Louisian in any way is a prime candidate for Hoosier status. (1987: 3).

Murray also found that when he played recordings of speakers from various social classes
for his infonnants, they could accurately distinguish upper and lower class speakers (his
labels). Lower class speakers were most often assigned the label hoosicr. When asked
where the lroosiers were likely from, respondents claimed they were likely from southern
Missouri (Murray 1987). And interestingly, according to the Atlas of North American
English, southern Missouri, most notably southwest Missouri and the so—called boot heel, is
the only section of the state included in the southern speech reg'on (Labov 2003).

Murray predicted that, despite being ”alive and well” in 1987, hoosicr would soon
"die the ignoble death at the hands of Father Time" (1987: 7). However, evidence from
popular culture and the current study prove use of the derogatory hoosicr is still thriving.

In addition to the data from the present study which supports this fact, evidence from a less

12

academic realm also shows lroosier is not going to die an "ignoble death" any time soon.
For example, MU330, a St. Louis ska band popular in the late 19905, has a song entitled
"Hoosier Love" on their 1993 album, which contained lyrics such as: We don't need no
high school/No high school! I don't want to go/We'll have kids at 17/ Gettin' laid at Dairy
Queen/Hoosier love, Hoosier love, southside city Hoosier love." Such lyrics indicate that
the image of a hoosicr as a working class, less educated person with low morals is still a
salient one for listeners of the song. In addition, Mike Seely, a columnist for St. Louis'
weekly paper The Riverfront Times, has written about the term as recently as 2003.
Moreover, a short film by two St. Louisians entitled Hoosiers Are From Mars addresses
the St. Louis hoosicr. In this film, hoosiers drive beat-up pickup trucks, date ugly women
(or are ugly women themselves), and wear mullets (the short on top, long in the back
hairstyle often the object of much ridicule). In the opening clip of the film, a hoosicr is
defined: "St. Louisians agree that hoosicr is a noun but use the term negatively to describe

individuals a step above white trash" (Henroid and Shah).

1.3.3 Linguistic Influences on St. Louis’

Historically, St. Louis has had diverse linguistic input from the immigrants and
migrants who settled in and near the city in the mid and late 19th century. During this

time, St. Louis saw a large growth in population from individuals moving to the Gateway

 

' It should be noted that this examination into immigration and migration to St. Louis focuses primarily on Europeans
and European Americans. This is not to say there was a lack of African American presence in St. Louis. However,
despite the city being aligned with the union cause against southern secession and slavery, there was not a large population
of African Americans just before the Cisil War, at which time African Americans comprised just two percent of the
population (Kamphoefner 2003: 89). And after the Civil War, the number of African Americans in the state as a whole
dropped even further, as many former slaves left Missouri to find work elsewhere (Holland 2003: 68). Much like the rest
of the country, there was enforced and de facto segregation in St. Louis between whites and blacks for decades. Indeed,
the largest number of African Americans who live in the St. Louis metropolitan area today live in the nearly all-black
community of East St. Louis, Illinois which lies across the Mississippi River from St. Louis, Missouri. Thus, any linguistic
influences African Americans have historically had on the speech of St. Louis is unclear.

l3

City, primarily from other countries, including England, Ireland, Switzerland, Alsace, the
Austro-Hungarian Empire, Germany, various Slavic countries, and Canada. In fact, by the
18505, the majority of St. Louisians had been born abroad (Kamphoefner 2003).

The earliest European settlers in St. Louis were largely French and Irish
immigrants, initially bringing with them linguistic influences, including the introduction of
numerous French creoles. According to some dialectologists, however, these groups did
not leave a lasting impact on the speech of St. Louis (Frazer 1979). The most sigrificant
group of immigrants was from Germany, who began to arrive around 1830; their
immigration to St. Louis increased at a rapid rate throughout the mid-19th century.
Germans, who were concentrated primarily in the north and south sides of St. Louis, had
considerable linguistic and cultural influence on St. Louis (Marckwardt 1980, Lance 1993).
By the time fighting during the Civil War broke out in'the United States, there were ten
languages spoken in St. Louis, the predominant one being German, which was taught in
schools and was nearly part of the cultural mainstream (Kamphoefner 2003). Most likely,
direct influence of GemIan linguistic heritage in St. Louis is evident most strongly in certain
lexical items, although there has been some evidence that some phonological features from
German were not so quick to die out (Marckwardt 1980, Lance 1993). However,
according to Frazer (1979), the most dramatic influence of the German immigrants was not
their use of the German language features, but their embracing of a more "Yankee" style of
, speech, not the speech of any Southerners they encountered. Frazer's claim is that North
and North Midland speech was most likely the preferred speech variety due to the fact that
"Yankee” settlers to the St. Louis region "regarded themselves as representatives of a

superior culture, and their prejudice extended to language as well" who viewed individuals

14

from the South Midlands as "their linguistic inferiors" (1979: 188-189). And indeed, the
majority of school teachers in St. Louis were not of Southem heritage, and most likely
spoke a dialect that Frazer calls the "prestige model" of more northern speech (1979: 190-
192). He even claims that Germans living in the rural areas surrounding St. Louis “did not
feel comfortable with [their Southeml neighbors as cultural or linguistic models," and thus
adopted a more "Yankee" style of speech (Frazer 1979: 190).

During and after the Civil War, immigration to St. Louis steadily declined, much to
the dismay of its inhabitants there. In fact, losing population growth to Chicago by the turn
of the 20th century gave St. Louis what one historian called an ”inferiority complex" (Meyer
1963: 501). Indeed, St. Louis has never seen immigration like it saw in the 19th century;
even when other large immigration waves came to the US, they "largely bypassed St. Louis"
(Kamphoefner 2003: 97).

III addition to immigration from foreign lands, St. Louisians saw a migration from
their fellow countrymen during the mid-19th century. At this time, large numbers of both
northern and southem Americans migrated to St. Louis, impacting the speech of the
Gateway City. The largest number of migrants coming to the state of Missouri and western
Illinois came from southern states; the majority of them came from Kentucky, Tennessee,
and Virginia (Kamphoefner 2003, Frazer 1979). However, by 1860, most migrants coming
to St. Louis were from "Yankee" states: New York, Pennsylvania, Ohio, and Illinois
(Kamphoefner 2003: 89). In fact, as one Missouri historian explains, "in temIs of its
population, St. Louis remained a Northern city in a Southern state” (Kamphoefner 2003:
89). This was indeed the case for most residents during the Civil War, as most St.

Lousians aligned themselves with the union cause, including a vast majority of Gennan

l5

immigrants and German Americans (Kamphoefner 2003: 85).

According to Frazer, there was also an interesting class split based on these regional
differences in the city of St. Louis and surrounding areas: N ortherners tended to be more
educated and live in the urban areas, while Southerners tended to have less education and
live in the more rural surrounding communities (1993, 1979: 188-192). Migrants from the
North also tended to have more political and economic power than those from southern
states (Frazer 1993: 63-64).

As has been noted, southern speech is generally a stigmatized dialect in the United
States (Preston 1993). This general tendency is the case in St. Louis as well, as discussed in
Frazer 1979 and Murray 1993. Murray claims that St. Louisians are aware of "more
correct" or "more standard" language use, especially with younger informants (1993: 129).
Although this data was collected in 1986, the tendency appears for the trend to continue.

III addition, there also seems to be a desire in St. Louis not to sound like a hoosicr.

1.3.4 St. Louis Becoming 'Northern'?

III addition to evidence regarding the uniqueness of St. Louis' dialect region, and
influences of the early preference for "Yankee" speech, it has also been suggested for
decades that St. Louis continues to align itself dialectically with northern dialect areas.
Thomas Murray, who has done some extensive data collection on St. Louis speech, though
not the NCS specifically, has also made a claim that the speech of St. Louis is becoming
more 'northem" with each generation. And while Murray initially worries that ”one could
easily surmise that the language of St. Louis is a hopeless amalgam of Northern, Southern,

and North and South Midlands speech traits," he also claims that patterns towards

l6

'northernness" are evident (Murray 1993: 129). Based on his collection of data from 1986,
which included an examination of lexicon, syntax, and phonology, Murray concludes that
St. Innis aligns itself with Northern and North Midland dialect regions, more so than
Southern or South Midlands. In a more recent study, where he compares his older data
with another collection from the 213i century, St. Louisians are moving to an even more
Northern dialect. He says that both "culturally and psychosocially, the Northern/North
Midland standard seems to be growing more robust, further displacing that of the
South/South Midlands" in St. Louis (Murray 2002: 349). In the youngest generation of
speakers from his newest. data collection, all of the linguistic factors he examined have
become more "northern," most robustly in ”pronunciation" (Murray 2002).

According to Labov, Ash, and Boberg, St. Innis "has long been recognized as a
center of Northern linguistic influence" (1997). III fact, based on a very small number of
informants, Labov claims St. Innis is undergoing a "massive shift towards the pattern of the
Inland North, including the Northem Cities Shift" (2003). Here Labov also claims that the
embracing of the NCS in St. Innis differentiates it from the surrounding Midland area.

This continues to keep St. Innis distinct from other areas of Missouri.

1.4 Summary

In addition to giving the background on the speech phenomenon examined in the
present study, this chapter has shown how St. Innis is historically a unique dialect region,
based on both phonology and lexicon. It has also provided a historical reference for the

speech of the region as well as given an update on the allegedly increasing "northem'

speech exhibited by the inhabitants of St. Innis.

l7

The following chapters will explain how data was collected, analyzed, and
interpreted in an attempt to address the question of whether the NCS is making its way

further south to St. Innis.

l8

2.0 Methodology
2.1 Respondents

Since it has only recently been documented that the N CS may be making its way to
St. Innis, it is likely that the speakers of this new dialect will be young individuals.
Therefore, this study focused on young St. Inuisians, ages 15-26. Two social factors were
used to further examine the social embedding of the Northem Cities Shift if evidence of its
presence is found: sex and social status. Attitudes about St. Innis, American English
dialects, and Missouri were also examined. The data examined in this study are taken from
a larger set of collected data. A total of 47 individuals were interviewed during November
of 2003 and January of 2004. From this data set, 29 will be examined in the current study.
Inforrnants were excluded from acoustic analysis because they have lived for several years
outside of St. Louis, are currently attending college in another city, live in Illinois, or
because their social class was diffith to determine. Those respondents who were chosen
for acoustic analysis are distributed nearly evenly with respect to three social classes (upper
working class, lower middle class, and upper middle class) and both sexes. The
distribution of informants is summarized in Figure 2.1. As is clear from the figure, five

informants were sampled for each cell except for upper working class males.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Male (4)
Upper Working Class
(9) Female (5)
Male (5)
Respondents Inwer Middle Class
(10) Female (5)
/ Male (5)
Upper Middle Class

(10) Female (5)

 

 

 

 

 

 

 

Figure 2. I : Overview of Sample Design (Total N -29)

_19

III addition to the variable demographic information of social class and sex, and the static
demographic of age, informants were selected based on their status as St. Innis natives and
their ethnicity. All respondents in this study are natives of the St. Innis area or moved
there as young children.’ The vast majority of respondents' parents were also born in St.
Innis. And as the NCS is historically a European American phenomenon, all respondents

grew up in European American families and communities?

2.2 Data Collection

Young speakers from various neighborhoods in the St. Innis, Missouri
metropolitan area were interviewed in order to determine the existence of the Northern
Cities Shift. By its inhabitants, St. Innis is divided into two major areas: the city and the
county. Oddly, St. Innis city proper withdrew from St. Innis county in 1875 (Meyer 1963:
425). Therefore, residents of St. Innis are quite aware of the boundaries between the city
of St. Innis and the county of St. Innis. These two regions, of course, have many smaller
divisions. For example, "north county" and "south county” are working and lower middle
class areas, while "west county" is much more affluent. East St. Innis lies across the
Mississippi in Illinois; it is its own incorporated city and is primarily an African American
working class city. In addition, there are some suburbs of St. Innis, Missouri which lie

across the river in Illinois, but are not part of East St. Innis.

 

’ Two of the 29 respondents were not born in St. Innis: one was adopted, one moved to St. Innis during elementary
school.

i All respondents but one were European American; the one exception is a Korean-American respondent who was born
in Korea, but adopted by European American parents as an infant. He grew up in a European American family,
neighborhood, and church.

20

The quantitative data for this study was gathered from inhabitants of St. Innis city,
north county, south county, west county, and Fenton (which lies in both "south county" and
Jefferson county, just south of St. Innis county); attitudinal data was gathered from
inhabitants of these areas as well as some of the Illinois suburbs. All such areas are
considered part of the St. Innis metro area. The basic regions and neighborhoods of St.

Innis can be seen in the following map (Figure 2.2):

  
  

  

b Hilly?" stalling 1'”
GI‘Anile-

 

,
I. T I

Cnllinsville

  
 

 

    

 

     
 

 

 

t.
. , , rat .............
= ‘13. 1363;": . lurk
, BEDW (19“ 0:2

  
 
   

3‘1
I

'I
H- '\".\l I.
,. Tillman Swén "walk:

.‘v, 3' ..~\
Mehfiilille Du "pt- 1. Imhs.

' ‘ 3!
TR '“n°°_.\ . ,_.,-:A Birth °/ 1 . ’
I I s- , .3... ------ 1}"
Murphy. ‘Ltioit ille N's» \ ‘I , ,3-
Q 1'93:

   
 
    

. ' F 6*: Kiopp'Fld Millsudt
\ T ' 1 {cliff Caves- Columbia 1

Figure 2.2: Map of the St. Louis, Missouri area

Inforrnants were contacted in a variety of ways, primarily through the "friend of a
friend" method described in Milroy (1980) and Milroy and Gordon (2003). In addition,
some respondents were approached, usually by the researcher, and simply asked if they
would like to take part in a linguistic study. Interviews were conducted at individuals'
homes or the homes of friends, a high school, a bowling alley, and respondents' workplaces

(including fire stations, a construction site, an elementary school, and a university).

21

workplaces (including fire stations, a construction site, an elementary school, and a
university).

Respondents were first interviewed in order to obtain demographic information
which was used to determine social class (e .g. occupation and education of individual or
parents). Questions were also asked in order to elicit the informant's attitudes about the
city of St. Innis (in comparison to other large cites, to the rest of the state, etc.), hoosr'ers
and hoosicr speech, and their future plans. In addition, informants read a word list and a
reading passage. The word list, which consisted of 105 words presented to individuals on
flashcards, predominantly contained words with the vowels of the first three steps of the

Northern Cities Shift: /a=:/, /a/, and /o/. Other vowels and diphthongs were also elicited

through the word list and reading passage. Data on vowels not involved in the shift were
collected so that the targeted N CS vowels could be examined with regard to their relative

position in the respondent's entire vowel space.

2.3 Social Class

One of the two demographic variables examined for the purposes of this study was
social status. Three social class groups were examined: upper working, lower middle, and
upper middle. Class was determined based on Warner's Index of Social Characteristics
(1960). In his research, Warner determined which economic and social factors are most
indicative of an individual's social class in the United States. This model has been used by
various linguistic researchers in determining social status (Ito 1999, Evans 2003, Jones
2003). Therefore, the individuals in this study were assigned a social class index based on

their occupation, type of dwelling, education, and reputation of neighborhood. If the

respondent was a high school or undergraduate student or a non-working spouse, the
characteristics of the main wage earner in their household were used. Each of these
categories was given a scale of 1-7, then the categories were weighted, and finally they were
added together to determine social status. See Appendix A for a full explanation and

breakdown of social class features and categories.

2.4 Data Analysis

According to Labov, chain shifts, such as the N CS, require acoustic analysis to be
adequately detected (2003). Therefore, the first stage in data analysis for this study
involved extracting the first (F l) and second (F2) formant frequencies of the respondents'
vowels. This was done through computerized linear predictive coding (LPC) analysis on
the word list data. For the most part, the word list provided enough data for a robust
analysis, though a few tokens were taken from the reading passage to supplement data for
some individuals. It has been shown that there is no stylistic variation for the N CS in
controlled speech such as word lists, as it is below the level of consciousness and not
stigmatized (Ash 1999).

The LPC analysis was perfomred using the computer program Praat (version
4.1.23), designed by Paul Boersma and David Weenink of the Institute of Phonetic
Sciences at the University of Amsterdam for the purposes of speech analysis. The
spectrogram of each vowel was examined in order to extract the vowel frequency scores.
Such a spectrogram can be seen in Figure 2.3. The vowels' fomrant frequencies were

extracted at the onset of the steady state of each vowel.

23

l‘t

 

Figure 2.3: An example ot‘a spectrogram: this is the word pat as said by Laura. The lower ofthe two
dark bands is the lirst formant tirquency (F I) and the higher is the second formant frequency (F2).

The formant frequencies were recorded and next inputted into PLOTNIK (version
7.0). This vowel system analysis program was designed by William Labov at the University
of Pennsylvania and allows the researcher to visualize informants' vowel spaces by plotting

individual vowel tokens. Samples of vowel plots are given in Figure 2.4 and 2.5:

 

mnzs 00 26.00 2100 22:00 2000 1a00 16100 1100 12:00 1000 890 600
"0 bad
400 1 "R70 m... d' a boot
0" pool
. I horse
500‘ o ' OflSh o ood K<>°"
”'9 Qlflis’t 9 Kb
"0"“ K0 0. 0p: 0 '°9i° from Ofoot 3 "°'°
600 . Q) 0 h" ‘7' hope
0 pull
0 ”'"d v lost
7'30 ' & nap 0 mesh . I
I tab A dust V f
06) bet A bun V ”903.050
800 - A duck A sub A I I watch
I bade. I “W1; mewful
<> fl 5 I no A I u
900 - I I u ""9"
I cabin
hon! I mo
( P
1000 - T I P“
I 1 00 -

 

 

 

 

Figure 2.4. Laura's vowels (individual tokens) as plotted in PLOTNIK.

24

 

[E‘Zl

2:000 2:;00 2e00 2100 2:300 2000 1e00 16300 1100 17.00 1000 800 600

 

.0. ® ®

600 ~ (I) (9:)

700.

900 1 ©
1000 ..

®
@ @@
®

1100-.

 

 

 

 

 

 

 

Figure 2.5. Lamas vowels (mean) as plotted in PLOTNIK

In the figures above, F1 and F2 frequencies of Laura's vowels are superimposed on a
traditional vowel chart. The F1 frequencies are associated with the height of the tongue in
vowel production; the lower the frequency, the higher the tongue is raised when the vowel
is produced. The F2 frequencies are associated with the front-back dimension. The
higher the frequency, the more front the vowel is produced in the mouth of the speaker.
The mean scores of the speakers' vowel systems were then plotted, creating a map
for each respondent's vowel system. (See Appendix E for vowel plots of all respondents).
As detecting whether an individual is a speaker of the NCS cannot be done simply by
eyeballing the vowel system of the speaker, an index system was used in the current study

to determine any divergence from a non-shifted vowel system. Such an index system

"allows for the quantification of the position of each vowel relative to other vowels in the
individual's system" (Evans 2003). Therefore, discrete differences are observable, which
are not necessarily seen when simply examining vowel formant data or even individual

vowel plots. This kind of assignment of index scores requires comparing a stable vowel to

a vowel which may have shifted. The vowels chosen for this study were /8/ as in bet, and
/A/ as in but; the movement of these vowels occurs in a later stage of the NCS. And, as this
speech phenomenon is presumably a recent arrival in St. Innis, it is likely that /e/ and /A/

remain stable. Thus, the data from acoustic analysis of respondents' /a/ and /a/ vowels
were examined statistically relative to other vowels in the system using the objective
statistical measure of a t-test. The raising of /a/ was determined via a comparison of the F1
of /ae/ to the F1 of / 8/ . The fronting of /(1/ was determined by comparing the F2 of /(1/ to
the F2 of /A/. Indices of both raising of /EB/ and fronting of /a/ were determined by the
following means, as employed by Ito (1999), Evans (2003), and Jones (2003):

Table 2.1: Fl index of the raising of /m/

 

/w/ is significantly lower than /6/

 

/ze/ is not significantly different from /8/

 

 

/ae/ is significantly higher than /e/, and closer to /1/ than /8/

 

 

l
2
3 /tB/ is significantly higher than /E/, but closer to /8/ than /t/
4
5

22/ is not significantly different from /I/

 

 

 

26

Table 2.2: F2 index of the fronting of /a/

 

/0,/ is significantly back of /A/

 

0
170/ is not significantly different from /A/

2/(1/ is significantly different from /A/, but closer to /A/ than /8/
3

4

 

 

/a/ is significantly front of /A/ but closer to /8/ than /A/
/a/ is not significantly different from /t-:/

 

 

 

 

 

After respondents' vowels were plotted and t-tests run on the /EB/ and /a/ vowels relative to

/e/ and /A/, F1 and F2 values were nonnalized, also via PLOTNIK, which includes a

system for such a task, developed by Neruy (1977). Nomralization allows researchers to
compare the vowel formant data across speakers, regardless of vocal tract size (Evans and
Preston 2001). Social and linguistic factors were analyzed statistically via t-tests, analysis of
variance (ANOVAs), and/or regressions. Such tests were run on SYST AT for Macintosh.

Data from these statistical tests were also compared to the index scores on prenormalized

data from PLOTNIK in terms of social factors. In addition, normalized data for /a,/

fronting was also subjected to a measurement proposed by Labov (see Section 3.4).

2.5 Summary

This chapter explained the methods used in the present study, which includes how
respondents were approached to be a part of the study, how their social class was
determined, how data was elicited, and how the collected data was analyzed to detemrine
participation in the NCS. The following chapters will discuss both the qualitative and

quantitative results of the research described.

3.0 Results
Chapter 3 examines the extent to which St. Innis respondents are exhibiting

features of the Northern Cities Shift. Although there are as many as six vowels involved in

this shift, only the raising of /z£/ and fronting of /(1/ will be treated in the present study, as

they are considered to be the earliest st. res in the shift (Labov 1994, Gordon 2001).

These NCS features were examined statistically in terms of social factors; the linguistic

factors of /&:/ raising of the present study were also compared to findings from previous

researchers.

3.1 lae/ raising

When examining the indices resulting from t-tests comparing the F l of /2B/ versus
the F1 of / 8/ of individuals' pre-nomralized vowel plots, one sees there is a general
tendency for the young people of St. Innis to raise /as/. The majority of respondents in the
study, 22 out of 29 (75.86%), exhibit /aa/ raising in relation to /e/. And of those
respondents with a raised /ae/, two respondents raised /a:/ to the index level of 3, and two

raised /ae/ to the index level of 4. Indices for all respondents are given in Table 3.1:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 3.1: /m/ index scores for all subjects
Respondent /m/ index
score sex Class age
Cassie 4 female LMC 21
Daisy 2 female LMC 24
Jackie 2 female LMC 15
Jennifer 2 female LMC 15
Kathy 2 female LMC 16
Anne 3 female UMC 18
Carrie 2 female UMC 25
Laura 1 female U MC 21
Sally 2 female UMC 23
Sara 2 female UMC 17
Amanda 3 female UWC 19
Billie 2 female UWC 25
Karen 2 female UWC 19
Maria 4 female UVV C 21
Terri 2 female UWC 17
Bob 2 male LMC 24
Craig 2 male LMC 26
J arnes 2 male LMC 16
Michael 2 male LMC 25
Peter 1 male LMC 22
Aaron 1 male UMC 22
Jacob 1 male UMC 25
Luke 2 male UMC 18
Rex 2 male UMC 16
Timothy 1 male UMC 1 7
Dan 2 male UVV C 26
Nate 1 male UWC 23
Rob 1 male UWC 21
William 2 male UWC 26

 

 

 

 

 

 

Social class abbreviations are as follows: UM C - upper middle class; LM C - lower
middle class; UW C = upper working class.

One of the respondents with the most raised /ze/ is Maria, who received a value of 4 for her

Fl index score. Her (mean) vowel system is shown below in Figure 3.1:

 

[171

1000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600
l L l l l l l l L l L l

 

500%

6004 © ©
® ©

900-‘ (::>
21

B

1000‘ O

I

1100

 

 

 

 

 

 

 

Figure 3.1 The vowel system of Maria, who has an F 1 index score of 4.

As you can see, the position of her mean value for /ze/ is over 100 hertz higher than /e/.
Since the majority (18 of 22; 81.82%) of individuals who raised /$/ in St. Innis,

raised to an index level of 2 in relation to /8/, a much more typical /'ce/ raiser of the St.

Inuis respondents is Jackie, who received an index score of 2. In her mean vowel system,
there is no significant difference between /&/ and /8/, as they are practically on top of one

another. Jackie's vowel system is shown in Figure 3.2:

30

 

F

El

303000 2am 2§00 2100 22:00 2000 12.100 115.00 1100 Igoo 1900 800 60G

'°°“ G)

5004

 

Q“ ® @
...

1100-1

 

 

 

 

 

 

 

Figure 3. 2: Jackie's vowel system; she received a 2 index score for /ce/ raising.

3.2 Social Factors and /ae/ raising

In order to shed some light on the social embedding of /83/ raising in St. Innis,

various social factors were examined statistically. Analysis of variance (AN OVA) tests and

t-tests were run on the normalized vowels of informants to determine the effect of these

social categories on /2£/ raising. In addition, chi-square calculations were done on the

/m/ raising index scores of respondents.

31

3.2.1 Sex and /m/ raising

A t-test run on the nonnalizcd F1 values of /8e/ for sex indicated that there was no
significant difference between males turd females (p<0.6530). In fact, the difference
between the Flof /a:/ between males and females was less than five hertz in the normalized

data. This result is surprising, given the fact that historically women tend to lead linguistic
change, including change in the NCS (Labov 1994, Hemdobler 1993, Eckert 1989a, Ito

1999, Evans 2003, Gordon 2001, Jones 2003). However, according to Evans and Preston

(2001), discrete differences of individualsf'vowel systems can be lost in normalized data. In

addition, these normalized data only show F1 values of /ae/ in isolation; the data on /20/

raising is not compared relative to other vowels of the system. Therefore, a t-test was run

on each individual's pre-nomralized vowel system (as described in Section 2.4), comparing
F 1 values of /EB/ and /8/. These results show that Fl index scores for /$/ raising indicate
that young women in St. Innis do have a tendency to raise /a=:/ in relation to /t-:/ more so
than young men in the Gateway City. In fact, only females scored higher than an index of
2 for /m/ raising (4 out of 15 total or 26.67% of females received an index score above 2).
And only one female received an index score of 1 (only 6.67%), indicating no signs of /EB/
raising in relation to /8/. In contrast, six males received an index score of l (42.86% of total

males) and no oun men scored above an index of 2, which would indicate more
,

advanced /a:/ raising. This data is summarized in Table 3.2 below:

Table 3.2 /w/ Fl index
score by sex

 

 

F
- kB/ 1 females males
index score
1 6
10

2
2

1503501—

 

To determine the significance of these results, a chi-square (non-parametric test) was
employed. In the comparison of males and females who raised /m/ to those who did not
via this statistical test, the result is significant (chi-square: 5.179, DF=1, p<0.025). This
result is therefore in keeping with the other work on language change and the NCS that

indicates females would be more likely to raise /a=:/. Importantly, this phenomenon of

women leading change is not unique to the N CS; Labov states that "in most vowel shifts

...women are considerably more advanced than men" (1994: 156). A striking example of

this is Cassie and Peter. Cassie has an /$/ raising index score of 4, and Peter only 1. This

is noteworthy because they are brother and sister and only a year apart in age.

3.2.2 Social class and /ae/ raising

In contrast with the data on sex, an ANOVA statistical test did indicate a significant
difference between the three social status groups of upper working class, lower middle
class, and upper middle class (p<0.00005) for /m/ raising when examining normalized F1
values. Because there are three groups examined here, a Tukey post hoc analysis was also

run to see if the three social class groups were statistically different from one another

individually. And, in fact, all three groups are significantly different from one another

33

(p<0.04l7 for UWC versus LMC, p<0.00005 for UWC versus UMC, p<0.0209 for LMC
versus UMC). This is not surprising, given that the F1 hertz values for social class are
nearly equidistant apart, going from lowest to highest social class. Mean F1 values in hertz

are given for the normalized data in Table 3.3:

Table 3.3: Normalized
F1 values of /a:/ by
social class

 

UWC 681.11 hz

 

LMC 700.14 hz

 

 

 

 

UMC 720.61 hz

 

From this data, a conclusion could be made that lower social status indicates a more raised

/$/ for individuals. And though not as evenly distributed as the AN OVA findings, /28/
raising index scores appear to correlate with the AN OVA findings. From Table 3.4
below, it appears that fewer upper middle class respondents raise /a%/. In fact, only 60% of

upper middle class individuals raise /2e/, compared to the lower middle class (90% of

whom are /£e/ raisers) and the upper working class (77.78% of whom are /E/ raisers). In

addition, of those individuals who raise beyond the index level of 2, half of these are upper
working class individuals, while the lower middle class and upper middle class have only
one respondent each who raised above this level. Raw values for index scores can be seen

in Table 3.4. The percent. res of these values are represented graphically in Figure 3.4.

34

Table 3.4: Fl index scores of /1e/ by class

 

 

 

 

 

F 1 Index
Score UWC LMC U MC
1 2 1 4
2 5 8 5
3 1 0 l
4 l 1 0

 

 

 

 

 

 

 

Figure 3.3: Percentage of /tel raising by class

 

IUVVCTi
ILMC .
DUMC

 

 

 

 

 

 

 

   

l 2

lie! index score

 

 

 

As the NCS has been dubbed a change from below (Labov 1994), it makes sense

that working class and lower middle class individuals would exhibit more evidence of /a3/

raising than the upper middle class. Although previous N CS data has not come to a
unified conclusion on leaders of change and social status, what has been shown is that there
is a tendency for upper middle class individuals to be lagging behind those individuals from

lower classes (Fasold 1969, Hemdobler 1993, Eckert 1988, 1989a). Therefore, if St. Innis

/a:/ raising is indeed caused by the NCS, most likely lower status speakers and women

would lead the change there.

35

It is also worth noting that there may be some explanation as to why at least one of

the upper working class individuals is not raising /2e/ like the rest of his social status group.

Both of these respondents are males, who would be more likely to resist such a shift, but

there may be a more nuanced explanation for Rob's lack of /ae/ raising. Like the rest of the

sample, Rob is European American; however, through the interview, it became clear that

he aligns himself much more with the African American cormnunity in St. Innis. This was

evident not only in the topics he discussed during the interview, but also through his

speech, which included features of African American Vernacular English (such as habitual

be, copula deletion, and various AAVE lexical items). An excerpt from his interview

appears below in which he is discussing his brother and his brother's manner of speaking

(see transcription conventions in Appendix G):

F."

:e‘rugefie

poi-.1

J:
R:

yeah, he come in here. He's, uh, he's (pause) I'm not gonna say he's a - you
ever heard of a wiger?

yeah

OK. He's kinda - he's ghetto, but lroosier

yeah

He's (unintelligible). I’m the same way, but I'm more - I'm not gonna say
I'm higher class, but I'm more proper.

yeah, you're less hoosier (laughs)

You know, I talk a little more proper. I mean, you wouldn't could tell by
the way I dress, but I'm more - kinda - 'cause I graduated from high school
and he didn't, you know? And I went with a lot of black people. I learned
the slang and the country talk, you know. That's all slang is.

yeah

counU‘y talk.

And since the N CS is a speech phenomenon more likely to be exhibited by European

Americans, it is understandable that an individual who aligns himself outside the European

36

American corrununity would not exhibit NCS features, especially since this speech

phenomenon is most likely new to the area.

3.2.3 Sex & class interaction and /2e/ raising

Statistically, there are possible interactions between social factors that may help
shed light on the speech phenomenon at hand. Therefore, an ANOVA statistical test and
Tukey post hoc analysis were run on normalized F1 /33/ values to detemrine if there was
any interaction between sex and social class. And it turns out that there is a significant
interaction between social status and sex (p<0.0059), even though sex was not significant in
the normalized t-tests on F1 of /&/. When sex and class were examined together, the only
group that was significantly different from any other social group was upper middle class

males, who were significantly different from every other social category'. The comparative

results in hertz are graphed in Figure 3.5 below. Upper middle class males exhibit far less

/2e/ raising than any other social group, regardless of sex or class:

 

5 A Tukey post-hoe analysis determined that UMC males were significantly different from other groups as per the
following probabilities: versus UWC females p<0.000l, versus LMC females p<0.0395, versus UMC females p<0.0428,
versus UWC males p<0.00005, versus LMC males p<0.0021.

37

 

Figure 3.5: Normalized F1 of /a:/ values for sex 8: class

Upper \Vorking Class Inwer Middle Class Upper Middle Class

610
650
660
670
680
5 690
E 700
7 10
720
730
740
750

 

7"“ _, .

 

rile-females

 

 

 

 

This is also apparent from the t-tests run on individuals' pre-normalized vowel spaces.

Notice in Table 3.5 below that it is the upper middle class males who have the most

instances of individuals with an /ze/ raising score of 1, indicating no evidence of raising.

Table 3.5: /m/ raising index scores for sex and class

 

 

 

 

 

 

 

 

 

 

 

UWC UWC LMC LMC UMC UMC
males females males females males females
1 2 0 1 0 3 1
2 2 3 4 4 2 3
3 0 1 o o O 1
4 o 1 0 1 0 0

 

 

 

Again, given what we know about the N CS and language changes in general, especially

changes from below, it stands to reason that upper middle class males would be the least

likely to raise /a/.

38

3.2.4 Age and /ae/ raising

Although age was not the primzn'y focus of the present study, there is still a range of
11 years (15 year-olds - 26 year-olds) among the infonnants. Therefore, a regression

analysis was run on the nomralized vowel system in order to determine if there were any

age effects on the data. Age, however, is not significant. in this population in terms of /33/

raising.

3.3 Linguistic factors and /3/ raising

The following and preceding segments were examined in this study to determine
which environments promoted and inhibited the raising of /a:/ in these young St. Inuisians.
Adjacent phones were coded in PLOTNIK for following manner, following place,
following voicedness, and preceding segment. The coding was done for the normalized
vowels of all 29 respondents. Following place and manner were recoded in SYSTAT to
include i voice. Analysis on adjacent segments has been done extensively by sociolinguists
investigating the NCS (Labov 1994, 11.0 1999, Evans 2003, Jones 2003). A comparison was
made of the results of the present study and those of previous research. Although all

researchers' findings are not identical, they illustrate a tendency for certain features to

promote and inhibit /&/ raising. And the results for adjacent phones in this study do not

stray too far from the previous research. The comparative data on the influence of

following manner of articulation is illustrated in Table 3.6 below:

39

Table 3.6: Influence of following manner of articulation

on IE / raising across studies
most promoting are the farthest left, most inhibiting are farthest right

 

 

 

 

 

 

 

 

 

 

 

Detroit
Labov (1994) nasal VI. fricative V stop V fricative VL stop
Rural Mid-Michigan V fricative VL fricative
Ito (1999) nasal V stop VL fricative VL stop
V affricate
V stop
Ypsilanti Appalachian V fricative VL stop
Exams (2003) nasal lateral VL fricative
African Americans
in Lansing, MI
ones (2003) nasal V stop V fricative VI. fricative VL stop
V affricate
V stop VL stop
St. Innis nasal V fricative VI. fricative lateral

 

It can be seen that all researchers found following nasals to be the most promoting of /ae/
raising. This is the most striking example of agreement of the influence of adjacent phones
on /&/ raising. In addition, the overall results for the promotion of /EB/ raising in the
present study are roughly in keeping with Ito (1999), Evans (2003), and Jones (2003), with

the exception of laterals, though only Evans (2003) included laterals in her investigation.

And, with the exception of Labov (1994), all researchers found voiced fricatives, affricates

(if included), and stops promote /w/ raising, which is in keeping with the findings of the

current study.

The picture for both following place and preceding phone is a bit rnurkier. The

comparative results of following place are given in Table 3.7 below:

 

Table 3.7: Influence of following place of articulation

on /2/ raising across studies
most promoting are the farthest left, most inhibiting are farthest right

 

 

 

 

 

 

 

 

 

 

 

 

DCU'OII labial
1.me ( 1994’ palatal apical velar
Rural Mid-Michigan
Ito (1999) no significant results
V velar
V apical
V labiodental
VL interdental
V palatal
VL labial
Ypsilanti Appalachian VL palatal VL apical
Evans (2003) V labial VI. labiodental VL velar
African Americans
tin Lansing, MI
ones (2003) no significant results
V labio-dental
V apical VL interdental VL apical
V velar VL palatal VL velar
St. Innis V labial V palatal VL labial VL labiodental

 

 

Clearly following place of articulation is not nearly as salient a feature in influencing the
raising of /a:/ as manner of articulation, as two researchers presented here found no
statistical difference with regard to following place, and for those studies which found
significant results, there is often significance only between large groups. But while the
evidence presented here is at first somewhat inconsistent, tendencies across studies can be
observed. First, it is likely that the reason the results of the Labov (1994) study are

inconsistent with the Ypsilanti Appalachian and St. Innis speakers is that Labov did not

include voicing in his investigation. This is important because a primary tendency for both

the St. Innis and Ypsilanti speakers is that they are more likely to raise /ee/ when the

following phone is voiced. The most salient finding between the St. Louis and

Appalachian speakers is that they both show voiced labials to promote /m/ raising

41

significantly more than other feature pairs. This distinction is lost in the Labov data since
voicing was not included.
Results for preceding segment are also somewhat inconsistent between studies, but

again, there appear to be some general tendencies.

Table 3.8: Influence of preceding segment

on /2e / raising across studies
most promoting are the farthest fell, most inhibitingare farthest right

 

 

 

 

 

 

 

 

 

 

 

Detroit
Labov (1994) notes obst + liquid inhibits raising
Rural Mid-Michigrur apical
Ito (1999) velar obst + liquid liquid labial
velar
apical
palatal
labial liquid
Ypsilanti Appalachian nasal obst. + liquid
Evans (2003) velar apical nasal labial
African Americans
in Lansing, MI
Jones (2003) No significant results, but notes obst + liquid inhibits raising
/W/
liquid labial
nasal apical
St. Inuis palatal velar obst + liquid

 

 

 

 

As can be seen in Table 3.8, Ito (1999) and Evans (2003), are the only two studies, besides

the present study, to have found significzurt results; they both showed that preceding velars

have a promoting effect on /&/ raising. This supports the assertions of Stevens (1998) that

velar phones either in following or preceding position tend to lower the Flof/m/ (573).

However, this was not. the result for the St. Innis speakers, who show a tendency to raise
when preceded by a liquid, nasal, and palatal more than velars. A possible explanation for
this is that there is a small number of tokens for both velars and nasals in the St. Innis

study (29 and 27 respectively, which is roughly one token per speaker). A more consistent

42

. 1a E1 .l . . ..

finding is that the preceding segment obsmient + liquid inhibits raising. With the exception
of the rural mid-Michigan group, there is a consensus on this cluster among these studies,
even those that did not find much statistical significance with regard to preceding phone.

Though seemingly muddled at first glance, this compiled evidence shows that, for
the most part, there are tendencies as to 110w adjacent segments will affect /83/ raising.

Clearly in terms of acoustic properties, however, manner of articulation is the most
influential. Such effects, therefore, look as if they are based on physiological acoustic facts

and universal phonetic laws, not variation from one speech cornmnnity to another.

3.4 /a/ fronting

The fronting of /0/ within the speech of young St. Inuisians was examined for a
vmiety of reasons. The first of these is that, despite the fact that there is disagreement as to
whether /(1/ fronting happens before or after /a‘:/ raising, /(1/ fronting is, by all accounts, one

of the early st. res in the shift (Labov 1994, Gordon 2001). In addition, a region-specific
baseline for the vowel system of St. Innis speakers was unable to be established. And,

unlike many other recent studies on the NCS (Ito 1999, Evans 2003, Gordon 2001, Jones

2003), the present study does not explore accommodation to a local norm, as St. Innis

historically has a unique dialect. Thus, an examination of /CI,/ fronting is done in order to
detemiine if the /a‘:/ raising in St. Innis is indeed due to the N CS. /c1/ fronting was

examined by the same mezms as /m/ raising; that is, through an index of pre-normalized

vowel systems (as described in section 2.4) and through statistical tests such as ANOVAs

43

and t-tests on the nomralized vowels of all respondents. The overall results of the /a,/
fronting index (juxtaposed next to the /28/ raising index) can be seen in Table 3.9:

Table 3.9: /u/ fronting results for all subjects

 

Respondent

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

/3/ index /0/ index sex class age
Cassie 4 0 female LMC 21
Daisy 2 1 female LMC 24
Jackie 2 0 female LMC 15
J ermifer 2 1 female LMC 15
Kathy 2 1 female LMC 16
Anne 3 1 female UMC 18
Carrie 2 0 female U MC 25
Laura 1 1 female UMC 21
Sally 2 1 female U MC 23
Sara 2 1 female U MC 17
Amanda 3 1 female UWC 19
Billie 2 1 female UWC 25
Karen 2 1 female UWC 19
Maria 4 1 female UWC 21
Terri 2 1 female UVV C 17
Bob 2 1 male LMC 24
Craig 2 1 male LMC 26
James 2 1 male LMC 16
Michael 2 1 male LMC 25
Peter 1 1 male LMC 22
Aaron 1 0 male UMC 22
Jacob l 1 male UMC 25
Luke 2 1 male UMC 18
Rex 2 0 male UMC l6
Timothy 1 1 male I lMC 17
Dan 2 1 male UWC 26
Nate 1 1 male UWC 23
Rob 1 1 male UWC 21
William 2 1 male UW C 26

 

 

It appears from this table that /(1/ fronting is not occurring to a large extent within

the vernacular of young St. Inuisians, since none of the informants fronted /(1/ beyond the

level of l, and five respondents actually have /a,/ in back of /A/ . This result is quite curious
given the fact that /a/ fronting is supposed to occur so early in the NCS. This, combined
with the fact that four respondents have an /2e/ raising score over 2, suggests that perhaps if
these respondents are indeed participating in the NCS, they may actually have a backed /A/,

making the index values somewhat inconclusive, although /A/ backing is a very late move in
the shift, one we would not expect to encounter in St. Innis. Nevertheless, an additional
criterion for /0/ fronting that Labov (2003) lays out was employed. His claim is that in a
normalized vowel system, /(1/ is fronted if the F2 of /(1/ is less than 375 hertz back of /£/ .

Therefore, such measurements were made on all respondents. By this gauge, seven

respondents are shown to have a fronted /(1/. The data from this test is summarized in

Table 3.10:

Table 3.10: /a/ fronting individuals

(via the Labovian measure for normalized vowel systems)

 

 

 

 

 

 

 

 

 

 

 

hertz /(l/ is
Respondent /a=:/ index /0/ index back of /8/ Sex class age
Anne 3 l 359 Female UMC 18
Cassie 4 l 28 1 female LMC 21
Dan 2 1 312 male UWC 26
Luke 2 l 332 male UMC 18
Maria 4 1 343 female UWC 21
Nate 1 1 29.5 male UWC 23
Peter 1 l 341 male LMC 22

 

 

 

 

 

 

Of the four respondents with an /m/ index score over 2, three have evidence of /a/

fronting via the Labovian strmdard. And two respondents with an /2e/ index score of 2 also

showed evidence of /a/ fronting. Interestingly, however, two of the individuals who showed
evidence of /0/ fronting show no signs of /a'-:/ raising, including Nate, who has one of the
smallest distances between /a/ and /€/. Figures 3.5 and 3.6 show the mean vowel plots of
the two respondents with the most fronted /a/, Cassie and Nate respectively. It is
noteworthy that Cassie has a very raised /m/, with an index score of 4, while Nate does not

raise /w/, as his index score is only 1.

 

[r2]

, 3000 2800 2600 2400 2200 2000 1800 1600 1100 1200 1000 800
‘500 l l l l I l l l 1

""°' (D
@

"M" (9 ® @
?00~

LIEU @

soot

900 . Q

I000!

 

 

1|00u

 

 

 

 

 

 

 

Figure 3.5: Cassie's pre-norrnalizcd vowel system; she has both a raised /w/ and a fronted /a/.

In Cassie's vowel system, /w/ is significantly more raised than /€/. It is also clear that /a/ is

approaching /e/. Nate's vowel system, on the other hand, shows that his /ae/ is lower than

46

his /e/. Despite this, he has a fronted /u/ in relation to /€/. Such inconsistencies will be

discussed in detail in section 3.6. Nate's vowel system can be seen in Figure 3.6 below:

 

g [F2]
300mm 2§00 211.00 2100 2&00 2:300 tqoo 1§00 1100 1.300 1000 800 600

 

400.

o o @
.... @ ®

7004 ‘:iEI)
8009
900-

1000~

 

1100~

 

 

 

 

 

Figure 3. 6: Nate's pie-normalized vowel system; he has a fronted /a/, but a non-raised /ce/.

3.5 Social factors and /a/ fronting

In order to get a more complete picture of who is fronting /a/ in St. Inuis, social

class and sex were examined statistically with regards to this feature. Once again, AN OVAs
and t-tests were run on all the normalized vowels of respondents. Chi-square calculations

were also employed where relevant.

47

3.5.1 Sex and /a/ fronting
The distribution of /0/ for sex, using the Labovian determination, is quite even.

Four males and three females fronted /(1/ in this population. Therefore, it is not surprising
that, according to a chi-square calculation on this variable, sex was found to be insignificant
(p<l, chi-square: 0.2905, DF=1) in /a/ fronting. In addition to this test, however, another

means of determining the significance of sex was also employed. AN OVA tests were run

on the F2 of the normalized vowel systems of all respondents. Results of this test show that
the /(1/s of male respondents are significantly more fronted than those of females
(p<0.00005). In addition, four out of five of the respondents who received an index score
of 0 for /(1/ fronting were female. This result is surprising for two reasons: 1- it. is contrary

to the fmdings of the chi-square on the Labovian standard, and 2- women tend to lead

linguistic change, including the NCS. Comparing the siblings Cassie and Peter is again
interesting. Remember that Cassie had an /aa/ raising index value of 4, while her brother
only received a 1, indicating he showed no evidence of /$/ raising. Yet, according to the

Labovian standard, both siblings show evidence of /(1/ fronting, though Cassie's /a/ is more

fronted than Peter's. Because of this curious fmding on sex, social status and sex were
examined together to see if there are any interactions between these two social categories.

This will be discussed in section 3.5.3.

48

3.5.2 Social class /a/ fronting

As is evident from Table 3.11, the higher the social status, the less likely an
individual will have a fronted /o/. Of the seven respondents who fronted /a/, three of these
were upper working class individuals (21.43% of total U W C), while two were lower middle

class (13.33% of total LMC) and two upper middle class (13.33% of total UMC). While

this finding is not robust, it does show a tendency for upper working class individuals to

front /a/. This was confirmed by ANOVAs on the normalized vowels, which found class

to be significant (p<0.0013).

Table 3.11:
Normalized F2 values of

/a/ by social class

 

 

 

 

 

 

 

social class F2 of /o/
UWC 1440 hz
LMC 1388 hz
UMC 1374 hz

 

In keeping with the data from the Labovian measure of /a/ fronting, the upper working
class is significantly different from both the lower middle class (p<0.0172) and the upper
middle class (p<0.0013), yet the two middle class groups are not significantly different from
one another (p<0.7035). Thus, as was the case with /a:/ raising, it appears the working class
individuals are leading the change with /0/ fronting. Interactions between social class and

sex are examined in the next section.

49

3.5.3 Sex & social class interactions and /u/ fronting

Because more than one variable was considered in this study, possible statistical
interactions were explored. An AN OVA statistical test was run on both sex and social
class. The result was significant (p<0.0424). Strikingly, the only group that was statistically
different from any other group was upper working class males.'S Indeed, two of the three

working class individuals who fronted /0/ according to the Labovian measure of /0/ relative

to /e/ are male. This difference, based on the normalized data, is graphed in hertz in

Figure 3.7 below:

 

Figure 3.7 F2 values of /a/ in hertz
by sex and social class

l§S()0 '- T‘-—— !- \ ‘—"' A ‘¥ 7 —"'#—_’ fi‘f “—f *w —"

 

 

 

 

 

 

 

 

1450 \
a \\\ Lima...
3 ”'00 ~-—I—— males '
J
1 350 - —— .
1 300
1 250

 

 

 

Upper Working Class Inwcr Middle Class Upper Middle Class

 

 

 

As the NCS is a change from below, and as upper working class individuals were leading

/ae/ raising, it is not surprising that it is the upper working class respondents who show

evidence of /(1/ fronting. This also helps explain the discrepancy between the chi-square

results on the Labovian measure and the ANOVA results on the normalized data. It is,

 

6 Results of the AN OVA for UWC males are as follows: versus U\VC females p<0.0001, versus LMC females
p<0.00005, versus UMC females p<0.00005, versus LMC males p<0.0029, versus I'MC males p<0.0003).

however, unexpected that the male respondents would be fronting /a/ so much more than

the respondents in other social categories, as men tend to lag behind in the NCS.

3.6 Discussion
Examination of the results of the speech of young people of St. Innis, shows
support for Labov's claim that the N CS is making its way to the Gateway City. There is

evidence of both early stages in the shift that were examined for the purposes of this study:

/ae/ raising and /(1/ fronting. The results, however, are not always consistent with previous

research done on the N CS. This sub-section explains some of these inconsistencies and

attempts to account for them.

3.6. l Inconsistencies with regard to social factors

Many of the findings in this study are consistent with previous research done on the
NCS. There is, however, some evidence gleaned from these St. Innis informants that is
not consistent with the previous NCS research.

While previous research on social status groups and the NCS has not entirely
reached consensus, there does appear to be a tendency for lower middle class and/or
upper working class individuals to exhibit NCS features first (Hemdobler 1993, Labov
2001, Ito 1999). This is usually the case with a change from below such as the NCS. The

current study is consistent with these previous findings, as individuals in the upper working

class lead /a/ fronting and upper middle class respondents are resisting /a=:/ raising.

The vast majority of sociolinguistic research shows that women have the tendency

to lead linguistic change, and work on the NCS has shown this tendency as well (Labov

51

1994, Hemdobler 1993, Eckert 1989a, Ito 1999, Evans 2003, Gordon 2001,Jones 2003).
Thus, it comes to no surprise that females in St. Innis are leading /33/ raising, possibly the
earliest st. re of the shift, and that it is a male group (upper middle class) that is the most
resistant to the change. There are many hypotheses as to why women lead such linguistic
change. Trudgill (1972) makes claims that women seek overt prestige; that is, align
themselves with the standard, in linguistic practice. Eckert (1989b) points out that women
are often powerless socially, and thus must use "linguistic capital" in order to assert their
identities. And Hemdobler (1993) suggests that women are the cultural bearers in society,
and this is played out linguistically. Whatever the reason, this phenomenon of women's

use of innovative forms has been shown time and time again with regard to the NCS.

It is for this reason, then, that it is so surprising that more men exhibit /a/ fronting
than women in the present study. Evidence that females are behind males in /a/ fronting is

seen through both the AN OVAs and by Labovian stzmdard (where 20% of women front /0/

versus 29% of men).

3.6.2 Inconsistencies with regard to linguistic factors
Results regarding linguistic factors are almost more puzzling than the social factor
results. The young speakers in St. Innis who exhibit N CS features are not exhibiting them

in the same systematic way Labov (1994) and others have suggested. For example, two of

the seven respondents who show evidence of /a,/ fronting received an index value of only 1
for /m/ raising, indicating a lack of /2e/ raising. This initially seems to support Gordon's

(2001) proposal that /a/ shifts before /£e/. However, all but one of the five informants who

52

received an index score of 0 for /a/ fronting (mezming /a/ is back of /A/) also received an
index score for /m/ raising of 2 or 3, indicating some degree of raising. Thus, for these
individuals, /a/ raising occurs before /a/ fronting. So, can we say /2e/ is the first. stage when
/(1/ fronting precedes /a:/ raising for some individuals? Can we say that /a/ fronting occurs
first when several /a=:/ raisers do not front /a/? Indeed, the man with the frontest /a/, Nate,
shows no signs of raising /a:/. Thus, it is difficult. to conclude whether /33/ drags /a/ in the

chain shift, or that /a/ pushes /ze/.

3.6.3 Possible explanation for inconsistencies

While there may not be a completely clear-cut reason as to why such social and
linguistic inconsistencies exist for the St. Innis population, an explanation may lie in the
fact that there are several differences between the St. Innis population and those examined
in the most recent NCS research. Probably the most significant difference is that, unlike
European Americans in rural mid-Michigan (Gordon 2001 , Ito 1999), Ypsilanti
Appalachians (Evans 2003), and African Americans in Lansing (Jones 2003), the N CS is
making its way to the mouths of St. Innis for reasons other than accommodation to a local
norm. These other studies examined the extent to which individuals outside of the NCS
speech community (that is, outside the urban, European American community) altered
their speech to accommodate the speech nomrs of the larger community surrounding
them. In the case of Ypsilanti Appalachians and African Americans in Lansing, the
populations studied were at least geographically embedded in an area where the NCS is

very advanced. In the case of rural mid-Michigan, the respondents were in close proximity

53

to a more urban community where the NCS has a strong foothold. Thus, the NCS is
simply spreading from these more urban communities to smaller ones nearby. On the
other hand, St. Innis is an urban community which is historically a distinct dialect region.
And the infomrants studied here, as they are European Americans who are native to St.
Innis, would not be at the fringe of the NCS community if the sound change were well
established. St. Innis lacks close proximity to major N CS territory, and therefore has
limited daily contact with a community with advanced NCS speakers. For these reasons, it
may be the case that this new speech phenomenon in St. Innis is not embedding itself the

same way socially as in those communities which are in closer proximity to NCS territory.

3. 7 Summary
For the population studied here, two early stages of the NCS were examined, /a/
raising and /a/ fronting, to detennine the presence of this language change in St. Innis.

From the various statistical tests, it appears that to some extent young residents of the

Gateway City are embracing this new language change. For the most part, the means in
which this is occurring is consistent with changes from below: females lead /£B/ raising,
upper middle class males are resisting /ze/ raising, and the upper working class is leading /a/
fronting. In addition, the results on adjacent segments which either promote or inhibit /'£/

raising are quite consistent with previous research on the N CS. Because speakers in St.

Innis are not simply accommodating to a community norm, however, there are some
inconsistencies with other findings on the NCS. Men lead /(1/ fronting, and the linguistic

factors do not appear to be playing out the same systematically across speakers. Despite

these inconsistencies, the general tendency for these speakers to exhibit features of the
NCS allows us to draw the conclusion that, in keeping with Labov's claim, St. Innis maybe
losing some of its distinctive dialect features as the NCS makes its way down Highway 55

from Chicago to St. Innis.

55

4.0 Language Attitudes

Although there is quite a bit of evidence that St. Innis has traditionally been a
unique dialect region, the evidence presented here shows that a widespread language
change from outside the cormnunity, the Northern Cities Shift, is making its way to the
Gateway City. Why would a city with distinguishable linguistic features essentially give up
its dialect for a less distinct one? Why would St. Inuisians align with more distant urban
areas than cities equally close or closer to their hometown? Evidence has shown that even
with changes from below, identity and attitude are reflected in language use. For example,
in her NCS studies in suburban Detroit, Eckert (1988, 1989a) showed that the social
identities and attitudes of the Jocks and Bumouts she interviewed affected their linguistic
practice. Ito (1999) and Gordon (2001) also showed that negative attitudes about people
from N CS territory inhibited its adoption in certain rural communities. Therefore, data
collected on the language attitudes of young St. Innis residents should give some insight
into the intrusion of the NCS into St. Innis. These attitudes will be explored in this
chapter. In addition to the 29 respondents whose speech was subjected to acoustic

analysis, 14 other individuals' attitudes were included in this portion of the study.7

4.1 The local dialect
As stated in section 1.3.2.1, none of the respondents in this study exhibited the

cord/card merger, which is the most recognizable phonological feature of the St. Innis

 

7 Of the original corpus of 47 respondents, four were not included in the present study because they were not European
American, were over the age of 26, or had moved to St. Innis as young adults.

area“. The vast majority recognized this as a stigmatized feature, and several said they knew
older people, like their parents, grandparents, or teachers, who exhibited the feature. Only
a handful of respondents, however, said they knew any young people who had the

cord/card merger. And in all of these cases the respondent claimed to only know perhaps
one or two people who said the word forty as /farti/. Of all the individuals who recognized

this dialect feature, none had positive things to say about it. When I mentioned this feature
to Jackie, a lower middle class lS-ycar-old, her immediate response was "0000, I Irate that."
She noted that some of her friends' parents speak this way. When asked if she and her

peers spoke much differently from their parents, she responded as follows:

Ja: yeah (whispers). (laughs) Especially my friends' parents. They'll say
/fartifcrr/ and /warj/, and unugh, it drives me crazy. My mother says
/warj/, uungh. (laughs), so annoying.

J: And, you guys, none of you say those things?

J a: Well, some of my friends do, but, I don't know. I'm the one who corrects
everyone (laughs). I'm a big loser (laughs).

Jackie clearly shows contempt for the historical dialect of St. Innis, to the point that when
she does hear one of her friends using its features, she "corrects" them by pointing it out.
A couple of respondents said they thought this dialect feature was "hoosicr," others
said it was a 'mispronunciation," was indicative of less education, or simply "doesn't sound
as nice." Kammi, an upper working class 17-year-old, even said it was 'dorkish." Many
individuals said they thought this feature was on the decline because of improvements in

education, traveling and being exposed to non-local people, or that they were taught to "say

 

' It should be noted that a few individuals mentioned a supposed local feature of African American Vernacular English
which has been popularized by the rap star Nelly, where here is pronounced /h(1r/, there is pronounced /60r/ and so on.

57

it correctly" when they were in school. When I asked Carrie, an upper working class 24~
year-old, if she thought the young people of St. Innis spoke differently from their parents,

she said she thought her dialect was different because of who her friends are:

It's probably, too, who I hang around with, like most of the people I hang around
with are, like educated, um, people with, like, white collar jobs, or whatever, so I
don't know if it's just like from college or being exposed to people from different
areas? or, you know, perhaps people my age who are in more, like, blue collar jobs
perhaps still have the same speech patterns.
Carrie also pointed out that her mother uses these traditional dialect features, and Carrie
feels no qualms about telling her mother to ”correct" her speech. In addition, Carrie
herself said "every once in a while I'll slip and say, say it, like O-R words like A-R. And so,
I have to sto-, so I mean it's not like it's totally, like, disappeared." She did not, however,
exhibit any such features during the course of our interview.

Like Carrie, Sally, another upper middle class 24-year—old, thought that younger

people do not exhibit this feature because they know it is stigmatized and consciously try to

avoid it:
8: I know I don't say /fartifar/, but um I me- I've heard a couple older people
say that, but I don't hear as many younger people saying it.
J : Right
S: Probably because they're so aware of it now.
J: Right, right. They don't want to be made fun of.

l l

S: They CHOOSE not to say it that way, 'cause they think about it when
they're gonna say it, like, oh, I can't say it like /fartifar/ or everybody's
gonna think I'm stupid, so-

J : think I'm a hoosier

S: Exactly (laughs)

When asked about the cord/card merger, many young people, like Jackie, also
discussed the intrusive /r/ sometimes heard in wash in the local dialect (though it should be
noted this is a feature of other dialects as well, particularly nearby North and South
Midland areas). Of those interviewed, only Kammi exhibited this feature, though she
recognized that this pronunciation was "wrong." Others, like William, a 26-year-old upper
working class man, said they "used" to include the intrusive /r/ wash, but eventually realized

it was something of a mispronnnciation:

J: But you don't say /farti/, you sazzzy?
W: /forti/ yeah, and don't, you know, uh, I did use to say /warj/, but I fixed
it myself, I started-

J: I know one of my friends from Indiana, he quickly-
W: Well, someone pointed it out that it's not co-, pronounced, there's no 'r,"

you know, and you look at in spelling, you're like, yeah, it's /waj/ (laughs).
You know?

It lras been asserted (Labov 2003, Murray 2002) that the unique phonological features of
the St. Innis dialect are regressing. This must in part be due to the fact that so many young
people in St. Innis have such negative associations with it, as nearly all of the young people

interviewed for this study who were familiar with these features showed disdain for the local

dialect.

4.2 St. Louis vs. Missouri
There was a wide range of opinions on the city° of St. Innis itself. Many of the St.

Inuisians who were under 21 found the area "boring" or thought that going to downtown

 

' For the purposes here, “city" refers to metropolitan area, not the city of St. Innis proper. This clarification is
made because St. Innis residents have a very clear distinction of the 'city' versus the 'county.' See Section

2.2 for full discussion.

was undesirable, though some thought it was a "great." city for sports and other activities. Of
the respondents in their 20's, opinions on the city were even more mixed. Some said St.
Innis was lacking in arts and culture, was ”not that happening,” or lacked diversity. On the
other hand, others said they were "proud" to be from St. Innis, that good schools were
available, it was an "important" city, or there was a lot to do. Despite these disagreements,
there was more of a consensus on the difference between the St. Innis area and the rest of
the state. Obviously, the majority of Missouri, in terms of land mass, is not urban; this
contrast between urban and rural appeared to be a very salient one to many of the
respondents, not just in terms of lifestyle, but also language and values, as can be seen from
the views of Laura, a 21 year-old upper middle class woman:

L: I think there's a pretty major difference. I mean, you can, I kind of group
St. Innis and Kansas City together, um, it's just the fact that it's, uh, any
kind of metropolis. I mean, and then you go down to the Springfieazlds
and cities like that that are almost just like giant groups of suburbs.

J. yeah.

L: And then after that it's just completely rural, and once you get down sontlr,

kind of backwards.

What to you mean by backwards?

Like, um, I don't kno-, real-, um, I mean, that's where a lot of

fundamentalists live, and things like that, that's kind of back in the woods,

and um, I, I took a bunch of sociology classes, and any time you talk about
major Irate groups, they always almost have a base in the Ozarks, which is

kind of scary. So, I mean, you get farther down into that kind of
fundamentalist, backwards thinking.

z-‘fi‘

Laura was not alone in her thinking that ”rural" had a negative connotation, though other
respondents were a bit less specific in their reasons why. I asked individuals what they
would say if they were on vacation far from home and someone asked them where they
were from: St. Innis or Missouri. All but one respondent said St. Innis, though a handful

included both city and state. The one individual who said he says "Missouri" also said he

60

used to say "St. Innis" when he went to Florida on vacation, but was surprised when people
did not know where his hometown was located.

When the other respondents were asked why they said the name of their city, the
majority of responses were something to the effect that they did not think "Missouri" gave
the same impression as ”St. Innis" (though some people said "St. Innis" was just nrore

specific), such as Rex, a 16-year—old upper working class boy:

R: Like some people might think of Missouri as being hoosier-ish, maybe, I
mean, I don't really know, so, 'cause some part of Missouri are really
hoosier-ish

OK. Which parts? Do you know?

Like small cities, like Lesterviazlle

Lesterville? I've never even heard of Lesterville.

yeah, see.

Do you think the people are different? Do you think the people are
hoosier-ish?

Like in Lesterville?

yeah. Or just wherever.

yeah, well, yeah, they are (pause) different

You think the people in the city, or in the metro area, are more-

more (pause) I don't know what the word is. Mozzre, less hoosier-ish
maybe? I don't know, more cultured? I don't really know what the word is.

wfiawfiw rawi-‘wrr‘

Upper middle class Sam, 21, said he thought St. Innis is "too cultured" for other
parts of the Show-me state, and the city is ”more refined” and ”sophisticated“ than the rest of
Missouri. This negative opinion was not specific to the upper middle class, however.
Cassie, a lower middle class 2l-year—old, thought the rest of Missouri sounded like
”hoosierville” and conjured up images of farming which she did not want to associate with.
A man from the upper working class, 2l-year-old Rob mentioned in Chapter 3, said he
liked to be associated specifically with St. Innis because "you gotta represent" and the rest

of the state was too "hoosicr.” Amanda, a 19-year-old from the upper working class,

61

thought the rest of Missouri sounded "too counU‘y." Maria, an upper working class 21-year-
old, said she thought that telling someone she is from Missouri would be “totally different”
from saying she is from St. Innis because the rest of the state is "just like farmland."
Fifteen-year-old Kathy from the lower middle class simply thought Missouri gave the

"wrong impression" of the way she lived:

F

I don't know, they always get the wrong impression when you say Missouri
(laughs)

What impression do you think they get?

Well, like a lot of my friends have said they like, they ask them if they like
wear sho:::::es, I don't-

(laughs)

(laughs) like we live on farms or something? I don't know.

7:“:

William, the upper working class 26-year-old mentioned earlier, thought the rest of
Missouri is too rural and Republican for his tastes. He has relatives in the rural areas of
Missouri, and he said he thongfrt “people out in the rural areas are different" because ”they
go more about just gut reaction than actually thinking things through, but maybe that's just
my bias towards a lot of their views, 'cause I don't feel the sanre way they do.” Political
beliefs and values were brought up by several other respondents as well. Three individuals
mentioned a proposition on concealed weapons, which was voted on by the state a few
years prior (and despite the fact that it was narrowly voted down, the legislature had
recently passed a bill making the carrying of concealed weapons in the state legal). For the
most part, according to these St. Inuisians, the rural areas all voted for the concealed
weapons proposition, while residents of urban areas voted it down. All three respondents

said they were against the bill.

Although they were never asked specifically about language differences within the

state, a few respondents said one of the major differences between St Innis and the rest of

Missouri was that. people in other parts of Missouri spoke differently than residents of the

city. Craig, a 26-year-old lower middle class man, gave the following account of St. Innis

and the speech surrounding it:

(3%:

Well, it’s like, if you leave, I think if you leave pretty much (pause) maybe
like a 90-mile radius, urnrn, say if you’re starting with, uh, directly west and
go half a circle all the way to the east, anywhere down south you can
DEFinitely tell, like there’s, uh, HUGE accents that come in, you can hear
‘em. North, I don’t really hear anything until like Chicago, like an accent,
but. (pause) I think that, uh, like in Cincinnati, or in Kansas, or anywhere in
the middle of the country, I think people kind of talk normal. If you go out
to the east, you got like a thi-, you know, New York accent kinda, you got,
or, New England kinda accent. South of the southwest had their California
kinda talk, I guess. I think-

valley girl?

Right, and the north there’s more like the:::: Fargo or, you know, North
Dakota, kinda Canada-sounding stuff. I think people here are pretty much
just normal.

Despite the fact that Craig thinks people from ”the middle of the country" talk "normal," he

still said he thought residents in southern Missouri had "southern accents." Interestingly,

Sam, mentioned above, also gave a distance in miles to where these southern accents are

first heard; his estimation (100 miles) was only ten miles different from Craig's (90 miles).

He was interviewed here with his mother Inrraine:

You can see that, you can hear the change in, in I guess it'd be dialect, as
you go further south. 100 miles south of St. Innis people talk completely
different than they do here. And it's just a different way of thinking down
there too. It's, I don't know, more farming, more agriculture, things like
that. It's not so much hustle and bustle of the city.

63

When you say people's dialect change, you think, are you saying, like, you
say you go 100 miles further south, you think it's like the southern features?
Is that what you're saying? Like-

Yeah, they'll pick up a lot more of a drawlll, and I don't know, we noticed
that a lot when we're- My sister went to Old Miss and on the way to drop
her off you would notice-

I'd go to Sikeston, Missouri, which is about halfway between Memphis and
St. Innis, and you'd get that, that (imitating southern dialect features) twang
would come out, that, that draw]. And it wasn't that nice southern, in
Memphis it's a very southern, soft accent, but when you get to the boot
heel it becomes a very nasal, (imitating southern features) kinda they don't
sound real intelligent, you know.

Again, this opinion is not unique to the middle class, as Nate, an upper working

class 23-year-old, also commented on the speech of other areas of the Show-me State. I

asked Nate if he thought the people in the rest of Missouri were similar or different from

residents of St. Innis.

2%:

Hrnmrn, uh, probably, both. Like, there's a lot of, you know, people are
similar, people are pretty similar, I guess, but yeah, pretty much. It seems,
except, more populated here.

Right, right. So you think people are pretty much just the same.

Uh, except when you get further south, and people start talking with that
hick, you know, twang.

(laughs) yeah

I mean, not even that far south, either, so you know, like maybe when you
get down to like Farmington and stuff. So, yeah, you don't gotta go too far
and people talk totally different.

Nate could not manage to peg down any difference between St. Innis residents and people

from the rest of the state except for their linguistic practice. In fact, it is the only difference

he could come up with. However, it is clear from the use of the term "hick" that he does

not feel as though the way these Southern Missourians speak is desirable, and we can

64

gather from his description of the "hick" dialect as "twang" that it likely has southern
features.

The consensus from these residents is that they do not necessarily want to be
associated with the rest of Missouri. They equate "Missouri" with ruralness and farmland,
which means a lack of sophistication and culture; it is therefore something they do not want
to be a pan of. From the few individuals who discuss the dialect features of "other"
Missourians, we can gather that simply traveling a few dozen miles south of St. Innis will

put them in an undesirable dialect region.

4.3 Hoosiers

As mentioned in section 1.3.2.2, one of the nrost widely used derogatory terms in
St. Innis is lroosier. From the data collected here, it seems that this insult, which is
completely separate from the term for residents of Indiana, is usually a noun, but can also

be an adjective (as in, "that's so lroosier"). It may also take the adjectival form hoosicr-i511 or

lroosier-y, or be truncated in noun fonrr to 11002 (pronounced /hu3/). All but one of the

respondents in this study knew very specifically what a St. Innis lroosier was when asked to
describe the term. The one person, 1.5-ycar-old Tamara, who did not have a mental image
of hoosicr still said she knew it was sorrretlring you called someone you do not like.

A few infomrants mentioned that they knew hoosicr was a term used in Indiana,
but they were clear this was not. the definition they meant when they used the word. In fact,
three of the respondents had recently gone together to see a concert in Indianapolis, and it
was the first time any of these young people (who are all in their 20's) had ever heard

hoosicr used outside of the derogatory St. Innis meaning (and all three mentioned this

65

experience independently of one another during their interviews). They all said they were
shocked by the fact that hoosicr was so embraced in Indiana. After this encounter Sam still

indicated that that the Indiana definition is not the most common one:

What's interesting is when I was in Indianapolis to see the Flaming Lips, we were
waiting outside to go in, and people were, you know, we brought up the fact that
they were hoosiers, and they were proud of that fact. And I don't think they
realized it had a negative connotation everywhere else BUT Indiana.
Rex, who was mentioned above, was not aware this was a region-specific term; he was quite
surprised that I had not learned this meaning of hoosicr until I met a group of St. Inuisians
in college. A handful of infomrants, including Rex, used hoosicr during the interview to
describe certain areas of the city or other parts of the state before I ever asked them about
the term.

So, what is a lroosier? For some it was hard to describe, and it seemed the only
words to describe hoosicr were "just hoosier!" or "you know, hoosier-y.” Others
enthusiastically gave detailed descriptions of the lroosier lifestyle, hoosicr haircuts, and
lroosier speech. It seems there has not been rnnclr evolution of the term in the last few
decades. Many of the informants in this study gave definitions of lroosier that roughly
matched Crinklaw's 1976 definition of a ”displaced country man who moves into a city
neighborhood and tears it up" (60). In addition, many respondents gave definitions that
were similar to those of Murray's 1987 informants who said that hoosiers were ”lazy, slow
moving, derelict, and irresponsible" (3). It is impossible to say, however, that either one of

these definitions fully describes the St. Innis hoosicr. The young St. Innis respondents in

this study put hoosiers into roughly two camps: "rednecks" (or "hicks”) and "white trash,"

66

though most people were not satisfied with any of these temrs. This difficulty can be seen

in Carrie's description below:

It's, like, I think, like, uh, a no-, or, uh, a non-local term would be probably like a

(pause) not a hick, but like a, like, you don't have to be like a country bumpkin to

be a hoosier, but you have to have a certain, like, um, it's easier if you have like

someone in mind that you can point out as an example (laughs), but it's sort of, you
know, anyone who-, I think like a, maybe like a redneck, kind ozf, it's not exactly
the same thing as a redneck, but that would be a more, like, less local term, I guess
for it.
Despite this difficulty in describing what hoosicr means exactly, Carrie showed great
affection for the tenn; she said it made her "so happy" to be able to use it because ”it's such
a good term too, it's so descriptive of, like, what you mean."

The two camps of thought — "redneck" and "white trash" - basically boil down to
whether hoosiers reside (or originated from) ”the country" and can therefore fall into the
”redneck" camp, or if they are lower class city dwellers, and therefore fall into the "white
trash" canrp.

Jackie said she and her friends considered hoosiers to be "people who are, like,
from the country, wear overalls and talk funny." She said she could not imagine hoosiers in
the city. Rex had a similar reaction; he also said hoosiers wear overalls, and that they live in
a trailer in small towns. When asked if such a thing as an "urban hoosicr” could possibly
exist, he said he could not "see that happening." Even within these two categories, however,
the term is a bit slippery, as many respondents who initially said hoosiers were "rednecks,"
”hicks," or "country bumpkins" also said hoosiers could reside in the city. Elly, a 24-year-old

lower middle class woman, initially said a lroosier is a 'slobby, country, redneck," but when

asked if a hoosicr necessarily lives in the country, she said:

67

N o, no, definitely, like, I would call someone who's just, like, and this is a bad term
to use, but white trazzsh. I would call them a hoosier, like, and they could be
from anywhere. They don't have to be a bumpkin to be a hoosier.

Some of the St. Innis infomrants who said they felt hoosiers could be urban

dwellers painted very specific pictures of what hoosiers are like (these individuals also

usually dubbed hoosiers as "trashy" or "white trash," though some were hesitant to use this

somewhat racially charged phrase). Such descriptions usually included trailer parks, cars

up on blocks in the front yard, drinking lowbrow beer, and sporting the mullet hairstyle.

For exaruple, when I asked Nate if he knew what a hoosier was, he responded as follows:

2%:

Can you tell me what a hoosier is?

Oh, hell, yeah. Hoosier is somebody who, uh, basically it just means,
uh, somebody who, uh, is just kind of a slob kind of person, you know,
drinks cheap, cheapest beer, smokes the cheapest cigarettes, goes home

with the fattest girl from the bar and brags about it, don't even care, you
know.

Billie, a 25-year-old upper working class woman, enthusiastically and quite descriptively

spoke about hoosiers:

as“:

This is a word that I didn’t, like it's sort of localized to St. Innis, that I

didn't know until I met St. Innis people, and that's "hoosicr"

Oh my-

Can you tell me what a hoosier is?

-GO::D. There are hoosiers everywhere (laughs). Um, like, people

that, you drive by a house and you see four broken down cazzrs in their front
yard, um, that's a lroosier. Um, they just don't care what they LOOK like
or- (pause) mullets. You know what a mullet is?

yeah

[Bob's] big on mullets, talkin' about mullets, uh, that's a hoosier.

68

J: That's a hoosicr thing.

B: They just don't care. And, it's like, there's nothing wrong with not.
carin', but they just leave stuff everywhere, you know, I mean, you
could drive down the street and see somebody's house.

Billie, like rrrany other St. Inusizurs who said they thought hoosiers could be city dwellers,
also gave specific locations in the area of St. Innis with high hoosicr populations, such as
south city, "north county," and various areas of "south county": LeMay, Affton, Fenton, and
Maplewood. The conclusion several infonnants reached, however, was similar to Maria's,
who said hoosiers are "everywhere," as she distinguished between urban hoosiers and
"country" hoosiers.

Regardless of location, however, many informants agreed that hoosiers are "messy,"
"dirty," "nasty," or "nu-kept" (sic), or that they just ”don't care" about their property or
possessions. Several people mentioned that hoosiers are often less educated than non-
hoosiers, though just as many informzurts said that having a poor education was not a
prerequisite to hoosier-ness. For exarrrple, Rob said that "it's just how you act. You could
be the smartest man in the world and be dirty and hoosier." For more examples of similar

attitudes, see Appendix F.

4.4 Hoosier speak

The word hoosicr has a fairly specific, derogatory meaning in St. Innis, and it is not
simply another term in the mental dictionary of residents of the Gateway City. Its use is
widespread, as can be seen from the popular culture examples in section 1.3.2.2. In
addition, several informants for this study, as mentioned above, used hoosicr before they

were asked to discuss it. People, such as Carrie, have great affection for the term, yet

69

maintain contempt for what it stands for. Due to the widespread use of the tenn, I chose
to examine what attitudes in St. Innis are among young people as to how a hoosicr speaks.
Only a small number of respondents said there was essentially no difference
between lroosier speech and non-hoosicr speech, except for hoosiers perhaps being more
"loud and obnoxious" than the average person. A slightly larger group of people claimed
that hoosiers simply have "poor grammar" and use words such as ”ain't," which are basically
signs of their lack of education. Two individuals said that hoosiers might use features of
African American Vernacular English in their speech. However, well over half (25
respondents or 59%) of those interviewed described the way hoosiers speak as having
southern features. Besides simply saying hoosiers had a 'southem accent" (which many
did), descriptions such as ”twang," "country," and "drawlll," which allude to southern speech,
were also used. Some of these individuals also claimed hoosicr speech did not include
"proper English," was "slower," or sounded "dumber." Several individuals even imitated
hoosicr speech; all of these individuals included features of American English spoken in

sorrthem states such as /(1j/ monophthongization. Elly described lroosier speech this way:

They have a bit of a souzzthem thing goirr', and I guess when you think of southern,
you think of slower, dumber? speech. I guess, 'cause they usually
portray southern accents as dumber or whatever, but, yeah.

Linda, a 21-year-old lower middle class woman, said initially that lroosier speech was
different from her own only because it sounded less educated. However, later in the same
explanation, she also claimed hoosiers have southern features in their speech, perhaps

equating a "southern accent" with lack of education:

70

L: I think they would speak differently, but that would be based on education,
not necessarily accent versus non-accent.

J: Right

L: So, you know, they would use double negatives, a:::nd they would have
ALMOST more of like a SOUTHERN accent.

J: uh-huh.

L: Whereas, you know, like, almost- not necessarily like an Atlanta

accent, but you know what I'm saying? Like, there's a different twang to it
or somethin'.
It has been shown, as mentioned in section 1.3.3, that southern speech is generally

a stigmatized dialect, yet there still is some evidence that so-called southern accents have
been rated "pleasant" by both northerrrcrs and southemers. (Preston 1999). St. Inuisians,
however, have nothing nice to say about the lroosier version of southern speech. It seems
that some residents of St. Innis associate this lroosier southern dialect with low class
individuals. For exarrrple, Mark was reluctant to even call hoosicr speech southern even
though this is apparently what he had in mind: "I don't want to say southern sounding, but
I guess that would be it. (pause) I guess white trash southern, you know (laughs)." This
sarrre disdain for hoosiers :urd their southern speech was repeated throughout my

interviews with the young people in St. Innis. For more examples, refer to Appendix F.

4.5 Discussion

'The young people of St. Innis apparently have no desire to embrace the traditional
linguistic features of their hometown. They find that it sounds uneducated or just plain
ugly. They therefore do not look locally for their linguistic identity, and are forced to look
elsewhere. From the majority of the respondents' views on the rest of the state, it appears

they do not have any desire to look to other parts of Missouri for their linguistic identity.

71

 

 

111).! iyf..llll r

 

The rest of Missouri is too rural, too "country," too politically backwards for this urban and
more sophisticated population. And, indeed, this is reflected in the fact that some areas of
Missouri exhibit southern dialect features. It seems that in the minds of these St.
Inuisians, backwardness and ruralness may also be reflected in southern speech. This
disdain of sonthem speech is compounded for St. Inuisians, as one of their most
commonly used insults, hoosicr, is also indicative of southern speech for many.

Such attitudes may help to shed some light on why the Northern Cities Shift is
making its way to St. Innis. The N CS is an urban dialect which cannot be confused with
southern speech, a dialect to be avoided. And adopting NCS features distinguishes St.
Innis from the rest. of the state, which is not exhibiting such features, as discussed in

section 1.3.4.

4.6 Concluding Remarks

Despite the limited number of respondents in his telephone survey, William
Labov's assertion that the NCS is likely making its way down Highway 55 from Chicago
into the mouths of young residents of St. Innis appears to be well-founded, based on the

data collected and analyzed here. Since there is some debate over what the first stage in the

shift is, both /2B/ raising and /a/ fronting were examined in the present study. Some degree
of /a2/ raising was exhibited in 22 of the 29 respondents, according to an index score
comparing the relative position of /&/ to /e/. /a/ fronting was exhibited by fewer

respondents - none showed evidence via the /a/ fronting index score, though seven

72

individuals showed evidence of /(1/ fronting via the Labovizm measure of /(1/ as <375 hertz
back of / 8/ .

In terms of social factors, the presence of NCS features was consistent with changes
from below for the rrrost part, with the exception of upper working class males exhibiting

the most /(1/ fronting. There were also several linguistic inconsistencies, such as /a=:/ raising

happening to an advanced degree in sorrre respondents who showed no /(1/ fronting, and

some individuals who showed high degrees of /a,/ fronting and no signs of /ae/ raising. As

this is such a new speech phenomenon in m area lacking close proximity to a region with
advanced NCS speakers, it may be the case that the NCS in St. Innis is not embedding
itselfin the same way socially as in other NCS regions. It will be interesting to see how the
social and linguistic factors play out when the shift has a more substantial foothold on the
region in a generation or two.

Clearly from the attitudinal data collected here, the local dialect of St. Innis is being
rejected, and the southern dialect close by is also heavily stigmatized. Thus, it appears that
these young residents of the Gateway City are looking north to find their linguistic identity

in an attempt to not sound like a lroosier.

73

APPENDICES

74

Occupation:

Housing:

1

Appendix A: Index of Social Class
based on Warner (1960)

lawyers, doctors, engineers, judges, architects, and managers of
large businesses

high school teachers, trained nurses, librarians, small business
owners, accountants, large farm owners

social workers, elementary school teachers, optometrists, minor
officials of business, bank clerks, auto sales, contractors

small business managers, typists, mail clerks, most store clerks,
factory foremen, private repainnen (e.g. plumbers)

beauticians, carpenters, plumbers (employed by others), barbers,
frremen, bartenders, restaurant cooks, tenant farmers

semi-skilled worker, skilled worker assistants, watchmen, truck
drivers, waitpersons (in small restaurants)

heavy laborers, janitors, newspaper delivery, odd job persons,
migrant workers

grand, ostentatious

very good, attractive, roomy, landscaped

good, only slightly larger than utilitarian demands, more
conventional and less showy than the first two categories

average, private, one and a half stories, nice lawn, some extra
room, small, well-cared—for lawn

fair, just enough room for needs, well kept, but no extras

poor, run-down, often too small for needs, not in shambles or
beyond repair

very poor, perhaps not even designed as housing, beyond repair,
crowded

ReDuLaL'on of
Neighborhood:

Education:

Computation:

Ratings:

1

3

in

vain-BOON;—

very high - the best place to live in the area; known as "well to do"

high - an area with an excellent reputation, low crime, good schools,
large houses and yards

above average - not pretentious but a nice, clean, and tidy

neighborhood

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

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

low, area regarded as slum

I , I
tenement areas; shacks, lean-tos, squatters areas

Graduate or professional school
college

high school

some high school

junior high school

elementary school

little or no schooling

Occupation x4 + Education x3 + Housing x3 + Neighborhood x2

12-17
18-22
23-24
25-33
34-37
38-50
51-53
54-62
63-66
67-69
70-84

upper

upper-upper middle
upper-middle-npper
upper middle

upper rniddle-lower-middle
lower middle

lower middle-upper lower
upper lower

upper lower-lower lower
lower-lower-upper-lower
lower lower

High school students and non-working spouses were assigned the smne scores as the
principal wage earner in their family.

76

 

Appendix B: Interview Questions

How long have you lived in St. Innis? How long have your parents, grandparents lived
here?

Compared to the rest of the country, what kind of city do you think St. Innis is? Where
do you think it fits in the scope of the US?

When you go out of town to somewhere far away like California or Florida, and someone
asks you where you're from, do you say St. Innis or Missouri? Why?

When you think about St. Innis corrrpared to the rest of the state, do you think it's very
similar to the rest of Missouri? How is it similar/different?

Do you think there's enough for young people to do in St. Innis?
When I went to college I learned this word from my St. Innis friends, and I was wondering
if you could explain it to me. Can you tell me what a 'hoosier" is? Or, what does the term

"hoosier" mean to you?

Where do ”hoosiers” live? Can there be such a thing as an ”urban hoosier"? Does being a
"lroosier" have anything to do with education level? How does a 'hoosier" speak?

Do you know this stereotype about how people in St. Innis speak, that they say /farti/

instead of /forti/? Do you know anyone who talks this way? Do you think there's any real
difference in the way you speak and the way your parents or other older people speak?

77

Appendix C: Word List

Vowels in the early stages of the NCS:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

/m/ labial/labio-dental/ apical
N-30 interdental (alveolar) palatal velar
rack, black,
voiceless stops nap, apple, zap at, mattress n/a brag
voiced stops tab, cabin dad badge rag
voiceless
fricatives bath, laugh, Affton ask, past cash, mash n/a
voiced
lfricatives have has, jazz n/a n/a
gang,

nasals gmnble, Sam lant, thank n/a banker
liquids n/a a] n/a n/a
/0/ abial/labio-dental/ apical
N- l 7 interdental (alveolar) palatal velar
voiceless stops mop, stop pot Watch rock, block
voiced stops Bob body logic n/a
voiceless
fricatives profit possible gosh n/a
voiced

ricatives father rr/a n/a n/a
nasals Tom John n/a n/a

iquids N/a car, (1011 n/a n/a

 

78

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

/9/ labial/labio-dental/ apical

N- l 3 interdental (alveolar) palatal velar
voiceless stops n/a caught n/a chalk
voiced stops n/a rr/a n/a dog, fog
voiceless

fricatives awful, moth lost n/a n/a
voiced

fricatives n/a pause, closet n/a n/a
nasals n/a gone n/a song
liquids n/a horse, tall n/a n/a

 

Vowels in the late stages of the NCS:

/8/: pen, mesh, bet, fed, step, neck, bend (N=7)
/A/: bun, puff, cup, sub, duck, dust (N-=6)
/I/: tin, hit, kid, tip, pig, fist, fish, pill (N=-8)

()tlrer vowels (N =24):

boot, food, pool, good, foot, pull, hope, hole, road, sleep, peel, meat, head, the, state,
make, bite, night, ride, house, loud, mouse, toy, oil

Total: N-105

79

 

Appendix D: Reading Passage

A Bad Day for Ducks
Tom and Bob were supposed to meet at Tom's house. They planned to go to a nearby
pond and watch the ducks. While waiting for Bob to get there, Tom picked up around the

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

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

better.

He grabbed a dusty hat out of the closet and shook it, but he couldn't get all the dirt off.
He got a cap instead and put a scarf around his neck and put on his socks and boots.
There was a big hole in his sock, and Bob was really late. It was already past 2:00.

Nothing was working out.

Just then Bob phoned and said he wanted to talk. He told ' ‘om that the flock of ducks

had left the pond. A pack of dogs had chased them off. Tom was sad; he had really

warrted to see the ducks, but Bob said they could go shoot some pool instead. Tonr

thought that was a good idea and forgot all about the ducks and his burned hand.

80

Appendix Ezlndividual Vowel Plots

L723

3000 22100 26300 2100 22100 2000 13300 1,6300 1100 race 1000 300
. . vl

 

1500
"°°‘ C9

.500. © ®

ISOO 1

*900 a

11300-

‘

IDO-

 

 

Vowels of Aaron, an upper middle class man, age 22
/ze/ raising index score: 1 /a/ fronting index score: 1

81

 

 

 

1C

11?

 

00

EIOO

2g00

2§00 2100 2200 2000 11100 1§00 1100 1300 1000 800

6C

 

 

@ @

 

 

 

 

Vowels of Amanda, an upper middle class woman, age 19
/$/ raising index score: 3 /a,/ fronting index score: 0

82

 

 

 

 

[E]

3003000 2830 2100 2100 22,00 2000 1.2500 1100 1100 1100 1000 800

 

400

A

ED @

500

6004 ©@

...1 ~09 @

900 - 8

1000-

 

1100-1

 

 

 

 

Vowels of Anne, an upper middle class woman, age 18
/EB/ raising index score: 3 /(1/ fronting index score: 1

83

 

 

 

Lrg]

 

L1)

 

500

400

500

500

700

2800 2§00 2400 2200 2000 1800 1600 1400 1200 1000 800
I l 1 l l l 1 J J '

 

 

F1

 

 

300

900

000

100

 

<9 <9 @
©

©© @
O© @

@
a

(1L).

 

 

 

Vowels of Billie, an upper working class woman, age 25

L

/a:/ raising index score: 2 /a/ fronting index score: 1

 

 

 

 

1 IE! 7

 

3003000 21100 2100 .2100 2200 2000 11100 1100 1100 1200 .1000 300

.... (0 ®

®
ia @

10004

1100‘

 

 

 

Vowels of Bob, a lower middle class man, age 24
/£/ raising index score: 2 /a/ fronting index score: 1

 

 

 

IE1

500200 3000 21.100 2100 2100 22:00 2000 11.100 1100 1100 1200 1000 300

t
2
s
i
i
1
1
1
r
1

 

400-1
(9 ®
600 -
700 -

800-

@
©
0
©
1000 -1 ®

1100-

 

 

 

 

 

 

 

Vowels of Carrie, an upper middle class woman, age 25
/£8/ raising index score: 2 /a/ fronting index score: 1

86

 

gnaw“

IE

:00 21100 2100 2100 2200 2900 1§00 1100 1100 12100 1000 800

 

$00

4004

500—

600-

700+

 

800-

900d

11001

 

©

 

 

 

 

Vowels of Cassie, a lower middle class woman, age 21
/m/ raising index score: 4 /(1/ fronting index score: 1

87

 

 

 

 

 

3C
300

100 2100

CW]

2§00 2100 2200 2000 1000 1§00 1100 1200 1000 800

6

 

400~
$004
600%
'i'OO-

IFII

800-

 

 

900~

1000-

1|00~

 

 

 

 

Vowels of Craig, a lower middle class male, age 26
/2B/ raising index score: 2 /a/ fronting index score: 1

88

 

 

 

 

 

 

. 00

2100 2100 2100 2100 2000 1100 1100 1100 1100 1000 .800

 

 

 

 

 

Vowels of Daisy, a lower rrriddle class woman, age 24
/$/ raising index score: 2 /a/ fronting index score: 0

89

 

 

 

 

 

5 .. , .l
1 IE]

300311100 2100 2100 2100 2100 2900 1100 1100 1100 1200 .1100 800

 

.00. Q)
500. ©

600d

7°:- @713

900 -
1000 -1

1100-1

 

 

 

 

 

 

Vowels of Dan, an upper working class man, age 26
/ae/ raising index score: 2 /a/ fronting index score: 1

90

 

 

 

{El
3003000 2100 2100 2100 2100 2100 .1100 1100 1100 1200 1100 800

 

‘°°' CD

500-

600- © 6) ® @

[iii— @ @
3°“ @@ ©
900 -«

1000-

1100-1

 

 

 

 

Vowels of Jackie, a lower middle class girl, age 15
/2e/ raising index score: 2 /a/ fronting index score: 0

91

 

 

 

 

I
i

 

E2] ,

 

30035300 28.002100, 2100 22:00 2900 1100 1100 1100 1100 1900 390 6
400 ~ (9 Cu)
50... © © @
®
600 . ®
<0
E @ ©
@
800 .
900 .
10001
1100 1

 

 

 

 

 

Vowels of Jacob, an upper middle class man, age 25
/a:/ raising index score: 1 /a/ fronting index score: 1

92

 

 

 

1
3000
300

2100

21011 2100

2200

 

IE1

2100 1100

1 100

1400
L

1 300

1900890

6

 

400-

600-

7001

800~

900-

1000

L

1100-

 

1

®

®

@

@

 

 

 

 

 

/2e/ raising index score: 2

93

Vowels of J arnes, a lower middle class boy, age 16
/a/ fronting index score: 1

 

 

 

[E

3001120 2100 2100 2100 2100 2100 1100 1100 1100 1100 1100 000 100

 

 

400-

5°°‘® ®
600 - © 6) @

900a

@ (9 "GD

 

1000

1

11004

 

 

 

 

 

 

Vowels of J essica, a lower middle class girl, age 15
/m/ raising index score: 2 /(1/ fronting index score: 1

94

 

 

E23

3200 3000 2000 2§00 2100 2200 2000 1000 1600 1400 1200 1000 800

 

400«

CD @

1100-1

 

 

 

 

 

 

Vowels of Karen, an upper working class woman, age 19
/a3/ raising index score: 2 /a/ fronting index score: 1

95

 

 

3065000 28LOO . 2§00 . 2100 2200 2000 1000 1§00 1100 1&00

11.00890

600

 

400 -1

5°“: 6) ®

700. ©® @”

@ <23 0 @

1000-1

1100-1

 

 

 

 

 

Vowels of Kathy, a lower middle class girl, age 16
/2e/ raising index score: 2 /a/ fronting index score: 1

96

 

 

 

if]

300-. r zqoo 2600 2100 _ 2goo 2000 1000 16:00 1100 12100 7 1900 800 600'

1:)
ID

 

 

.00. CD (9

500 ~ (9 @1

7004
[Ft

80d~ ©
900~ ®
@ @ @
@

 

 

@

 

1100-1

 

 

 

 

 

 

Vowels of Laura, an upper middle class woman, age 21
/a:/ raising index score: 1 /(1/ fronting index score: 0

97

2 1
2

3000

2100

2600 2400 2200 2000
L l l J

[E

1100 1100

1 400 1 2w 1 000
l I 1

800
l

30

 

500

400-1

500-

500-

7130-1

 

 

 

000-1

100-1

 

G)

6)

G)

09
©@

@
®

%@

 

 

 

 

 

Vowels of Luke, an upper middle class man, age 18

/$/ raising index score: 2

/a/ fronting index score: 1

98

 

 

 

[E2]
30 100 2100 2100 2100 2:100 2100 1100 11m 1100 1100 1100 810 60c

“ooonuaum .

 

400

500- © @
600 - ® @

700

©
900 .1 ©

1000~

®
©

 

 

1100

l

 

 

 

 

 

 

Vowels of Maria, an upper working class woman, age 21
/a':/ raising index score: 4 /a/ fronting index score: 1

99

40C
50C
60C
70C
E
90C
100(

110C

_

[1‘2]

3
30915000 2100 2100 [“2100 2100 2000 1100 1100 1100 12100

 

I-1

I-

I-1

1.1

[—1

)-

 

L1)

©
@@

 

 

 

Vowels of Michael, a lower middle class man, age 25
/ae/ raising index score: 2 /a/ fronting index score: 0

100

 

 

 

 

 

1
30031000 2100 12100, 2100 21:00 2100, 1100 1100 1100112100 1100 810 100

 

400~

<9 © @

E?” 0111 @

900-

 

1000-

1100~

 

 

 

 

 

Vowels of Nate, an upper working class man, age 23
/ze/ raising index score: 1 /(r/ fronting index score: 1

101

 

E2]

1 3000 2800 2600 2400 2200 2000 1 800 1 600 1 400 1 200 1 000 800
1 500 l L L l I l l l I

 

1100 -

1500 -1 ©
1500 -1 @ © ®

700 -
0

GB
800 -1 G

‘300 -

 

 

1000J

1I00~

 

 

 

 

 

 

Vowels of Peter, a lower nriddle class man, age 22
/a:/ raising index score: 1 /a/ fronting index score: 1

102

 

 

 

 

. 5 :0 2800 2600 2400 2200 2000 1000 1600 1400 1200 1000
1506 L l L l l L l 1 l

 

 

@
@

 

900a

1000-

 

‘

l0?-

 

 

 

 

Vowels of Rex, an upper middle class boy, age 16
/m/ raising index score: 2 /a/ fronting index score: 1

103

 

 

 

1
1 E2]
3003000 2100 .2100 2100 2200 2100 1100 1100 1100 12:00 1100 800 100

.00. (D

 

(D @@®

36 ©
..1 0.0 ©

700-

8004

1000a

1100-

 

 

 

 

 

 

Vowels of Rob, an upper working class man, age 21
/'ce/ raising index score: 1 /a/ fronting index score: 1

104

 

 

f

i

3000 2100

2100 2100 22100 2000
. 1

[El

1100 1100 _1100 1100

1000
1 .

39°

 

300

400-

500-

600-

700-

800«

900-

1000‘

1100-

 

6)

© 0)

@@

@j

@
@

'@

 

 

 

Vowels of Sally, an upper working class woman, age 23

/w/ raising index score: 2

/a/ fronting index score: 1

105

 

 

 

 

 

I2
305mm 3900 2900 2100 2100 2200 2900 1900 1100 110:1 1900 1m0 800

 

 

9001 o

 

1100-1

 

 

 

 

 

 

 

f 1

. Vowels of Sara, an upper middle class girl, age 17
/$/ raising index score: 2 /(r/ fronting index score: 1

106

 

 

3000

[El

2900 2100 2100 2900 2900 1900 1100 1100 1900

1900 890

00

 

' 300

400-

500-

600-

700-

80011

900-

10004

1100~

 

@ @

 

 

 

 

 

Vowels of Terri, an upper working class girl, age 17

/m/ raising index score: 2 /(r,/ fronting index score: 1

107

 

 

 

 

5000

L2]

2900 2100 2100 2900 2900 1900 1100 1100 1900 1900 890

600

 

600-

700-1

900-1

1000-1

1100-1

 

@ @
o
e Q§@

g§®®

.n‘.." .‘17 '1'

 

 

 

 

 

 

 

 

Vowels of Timothy, an upper middle class boy, age 17
/a/ raising index score: 1 /a/ fronting index score: 1

108

 

 

 

 

:
5
g E
:

mo31300 2900 2100 2100 2900 2900 1900 1100 1100 1900 1900 890 601

 

 

1000 .1

1100-1

 

 

 

 

 

 

 

 

Vowels of William, an upper working class man, age 26
/2e/ raising index score: 2 /a/ fronting index score: 1

109

 

 

 

Appendix F: Supplemental Attitndinal Data
Some responses to interview questions about hoosiers:
From Alan, LMC man, age 22:

Now, hoosier means you're like a, (pause) like, kind of the same thing as redneck,
kind of a country bumpkin kinda guy. Ummm, if you did something very hoosicr,
it would he say you were at a, (in a mock sophisticated voice) say you were at a fine
restaurant eating caviar, and you said (exaggerated stereotype of a working class
person) FISH EGGS? It's like that guy in the commercial where he says 'pass the
jelly'? (laughs) EXACle like that. That's the best analogy there is for a hoosier.

From Manny, LMC man, age 23:
Somebody that dresses scrubby, their hair's all over the place, um, like unshaven,
um, just downright srrrellin' awful.

From Linda, LMC woman, age 21:
They don't really care what they look like. Their house is trashed, and, you know,
they don't drive nice cars, and they don't wear nice clothes.

From Karl, UWC boy, age 17:
A hoosier "wears scrubby clothes and doesn't give a crap about anything."

From Anne, UWC woman, age 18:
Oh, they can live in the city, where they're just not taking care of their hon:::se.

You walk in, and you feel dirty when you walk outta there.

From Peter, LMC man, age 22:
Someone who's (pause) dirty, kind of, but not, like, a bum. See, there's a difference
between a bum and a hoosier. A hoosier is someone who can help, help it, but

they kinda act like a burn, and look like a bum

Also from Peter, LMC man, age 22:
They wear dirty clothes (laughs) don't have any teeth

From William, UWC, age 26:
(laughs) That's somebody that don't cut their gra:::ss, they have cars in their
yazzrd, they don't bathe very often, uh, they're just un-kept.

From Dan, U WC, age 26:
Around here, a hoosier would be someone who doesn't take care of themselves or
their fazzmily, or their property, for the most part

Also from Dan, UWC, age 26:
A lot of people, you know, they think, Metallica T-shirts, you know, heavy, hard

110

 

 

rock 'n' roll, five cars in the lawn

From Jamie, LMC woman, age 24

Ja:

Ja:

A hoosier, yeah, would just be, uh, (laughs), can I draw a picture? um,
somebody who is (pause), I wanna say redneck, I wanna say, um, trailer
park trash, and things like that, but. (pause) it's just somebody who doesn't
present themselves in a very, like, professional way? or, um, just kind of
rough around the edges, umm. Hoosier. J nst, they do things that aren't
with the norms of society, I guess, like, you know, have eight million cats, or
have mullets, or (laughs) I don't know, umm.

So, can a hoosier be an urban dweller? or is it like a country person? or
can it be either one?

I think of it more as a country person, but even out in St. Peters, there's
many a hoosicr out there, so, I mean, I don't know, that's not really the city,
but it's not really the country either? Um, I definitely don't associate it with
people who live in the city. I mean, it could be, I guess, but most of the
time I think it's people who would have to have a lot of land to put
refrigerators out on it and stuff like that, you know (laughs) and couches and
stuff.

From Sally, UMC woman, age 23
Ok, hoosier, um, you know you're talking your wife-beater shirts, driving your
Camero cars, or F irebird I should say. Firebird, or even more hoosicr is a truck, a
pick up truck that has no back end to it, it's white, it's rusted, um, hoosicr I think,
you know, you drink Busch beer, uh, wife-heaters, I said that already, but kind of
don't-give-a-crap kind of attitude, you know, real white trashy.

Luke, UMC man, age 18

L:

Wul, I've always thought a hoosier was like a hick, somebody that lives, like,
out in the middle of nowhere, just like typical redneck, like makes
everything themselves, does everything, self-reliant, can do whatever they
want, but really like country born, strict morals, a real country, country kind
of attitude.

So, you don't think there can be like an urban hoosicr.

Well, yeah, there can be, but I don't think I've seen one.

From Jessica, LMC girl, age 1.5:
Um, well, I don't know, going along with the statement of white trash, I think that it
would be, like, one of those peo:::ple who don't exactly have a stable jo:::b type of
thing, so I wouldn't think, just stereotypically, I don't think that they would be very
well-educated because it seems like well educated people don't have THAT nruch

111

problem finding a job? Um, anyway, it just seems like they would be the kinda
person who lives in a run down mobile hozzme off on the side of, I don't know,
some polluted area or something (laughs), I'm not really sure, that's just the
stereotypical hoosicr environment.

From Laura, UMC woman, age 21
You can, you can be like white trash with nrorrey, and with like Canreros and stuff
like that, it's just that kind o::f flasfry cheap

From Betsy, LMC woman, age 22
Hoosiers are just like tasteless people, do you know what I mean? just, they're,
they're hoosiers. It's kind of like you know it when you see it almost. You know, I
wouldn't say if you had like a mullet, you're a hoosicr, but it's like, I don't know, you
just think of a hoosier as somebody that wouldn't really couldn't take anywhere, do
you know what I mean? You kinda have to watch how they act and stuff like that.
But I use the term really loosely, do you know what I mean?

Also from Betsy (about her neighbor she claims is a hoosier):
He revs his motorcycle for no known reason, does not take care of his house, does
not take care of his yazzrd, his, um, truck is completely up on four blocks, you
know, and just, the engine of his car just sat out in his driveway for like six months.
Do you know what I nrean? yeah, he's a hoosier.

112

 

Some responses to the question: 'Does a boosi er speak differently than a
non-hoosicr?"

From Alan, LMC man, age 22:
A: I'm thinking, ummm (pause), in a country accent you would suppose, but
not one of those smart country accents where they' re kinda en: :.te It would
sound, uh, are you asking me to perform a hoosier accent?

J: No, no, not unless you want to, but, no I’m not asking you to perform

A: (deeper voice) /ma for wil brok dorm for tamz las wik a eigk 3
turn 01 glt A nu WAn/ (exhibits features of Southern US speech, including
four instances of /aj/ monophthongization)

From Betsy, LWC woman, age 22:
You know, like the the southern Illinois, like, drawl kind of? you know, kind of,
um, I keep wanting to say ”like lroosier." (laughs) you know what I mean, like with a
Southern acce:::nt and, you know, just tasteless, just kind of like classless kind of

thing.
From Aaron, UMC rrran, age 22:

I mean, a hoosicr, uh, I would think is generally the, uh, kind of southern twazzng
type, definitely. I MEAN, I asso- I mean, I think people generally associate
hoosiers with the south, and, kinda, I think it's southern, a southern accent, yeah.

From Jessica, LWC girl, age 15:

I think maybe when you're depicting a:: hoosier, you might have a little southern
accent or somethizzng, just not exactly proper Englizzzsh? maybe like slur your
wo:::rds type of thing, that's just my perception

From Luke, UMC man, age 18:
Well, a hoosier's just got, like, the southern acce::nt because they haven't been
influence by, like, like, I figure they do have a TV, maybe? but, we have TV's, we
have radios, and we learned, the way we learned to speak, our way, from our
parents, probably they're influenced by TV (unintelligible) how everybody else
speaks, and hoosiers just speak how they feel is comfortable, let everything go, don't
care, like how it sounds or anything, they just say it, or how bad it grammatically is.

From Dan, UWC man, age 26:
J: Does a hoosicr have a different way of speaking, than like a norr-hoosier
that you know or that you think about right away?
D: Um, can you tell, uh, by the way they talk? Is that what you're asking?
J: yeah

113

 

D: Yeah, yeah, I think there's a hoosicr twang, like we call it- that's what I call it,

hoosicr twang, or, I don't know, it's just weird how some people will look
and say different words that you can tell hoosicr month, I don't know.

From Rob, UWC man, age 21:

J: What about the way hoosiers speak, do you think-
R: country

J: yeah? What do you mean by country?

R:

like down south country, like, you know, Inuisiarra kinda country.

 

114

Appendix G: Transcription conventions

CAPITAL LETFERS

(comment)

spoken with errrphasis, higher volume
lengthened speech segrrrent

rising intonation

speaker overlap

researcher comment or explanation

115

REFERENCES

116

Ash, Sharon. 1999. Word list data and the measurement of sound change. New Ways of
Analyzing Variation 28 conference. University of Toronto/York University,
Toronto, Canada.

Callary, Rovert E. 1975. Phonological change and the development of an nrbarr dialect in
Illinois. Language and Society. 4: 155-70.

Cricklaw, Don 1976. Ladue Inckjaw, or How I Learned to Inve the St. Innis Language.
St. Inuisian (Oct): 59-61.

Eckert, Penelope. 1988. Adolescent social structure and the spread of linguistic change.
Language and Society. 17: 183-208.

—— 1989a. Jocks and Bumouts: Social categories and identity in the high school. New
York: Teachers College Press.

— 1989b. The whole woman: Sex and gender differences in variation. Language
Variation and Change. 1: 245-68.

Evans, Betsy. 2003. Dialect Contact and the Northern Cities Shift in Ypsilanti, Michigan.
East Lansing, MI: Michigan State University dissertation.

Evans, Betsy and Dennis Preston. 2001. When it's not nice to be normal: What's

missing from normalized data? University of Pennsylvania Working Papers in
Linguistics, ed. by Tara Sanchez and Daniel Ezra Johnson. 7.3: 59-65.

Fasold, Ralph. 1969. A sociolinguistic study of the pronunciation of three vowels in
Detroit speech. Georgetown University mirneograph.

Frazer, Timothy. 1979. The Speech Island of the American Bottoms: A Problem in
Social History. American Speech. 54: 185-193.

-— 1993. The Language of Yankee Cnltrrral Imperialism: Pioneer Ideology and
"General American." Heartland English: Variation and Transition in the
American Midwest, ed. by Timothy Frazer. Tnscalaoosa: University of Alabarrra

Press. 59—68

Gordon, Matthew. 1997. Urban sound change beyond the city limits: the spread of the
Northern Cities Shift in Michigan. Ann Arbor, MI: University of Michigan
dissertation.

—-—- 2000. Phonological correlates of ethnic identity: evidence of divergence?
American Speech. 75: 115-135.

-—-— 2001. Small Town Values and Big City Vowels. Publication of the Anrerican
Dialect Society. 84. Duke University Press.

117

Graf, Jeffrey. 2000. The Word Hoosier. Indiana Notes and Queries. Online:
http://indianaedu/ ~ libcsd/intemet/extra/hoosier.html

Habick, Timothy. 1993. Farmer City, Illinois: Sound Systems Shifting South. Heartland
English: Variation and Transition in the American Midwest, ed. by Timothy
Frazer. Tnscalaoosa: University of Alabama Press. 97-124.

Henroid, Paul and Ali Shah. 2002. Hoosiers Are From Mars. EnFocus Productions.

Hemdobler, Robin. 1993. Sound change and gender in a working class community.

Heartland English: Variation and Transition in the American Midwest, ed. by
Timothy Frazer. Tnscalaoosa: University of Alabama Press. 137-156

Holland, Antonio F. 2003. African Americans in Henry Shaw's St Innis. St. Innis in
the Century of Henry Shaw, ed. by Eric Sandweiss. Columbia, MO: University of
Missouri Press: 51-78.

Ito, Rika, 1999. Diffusion of urban sound change in rural Michigan: A case of the
Northern Cities Shift. East Lansing, MI: Michigan State University dissertation.

J ones, J amilia. 2003. African Americans in Lansing and the Northern Cities Vowel Shift:
Language Contact and Accommodation. East Lansing, MI: Michigan State
University dissertation.

Kamphoefrrer, Walter D. 2003. Learning from the "Majority-Minority" City. St. Innis in
the Century of Henry Shaw, ed. by Eric Sandweiss. Columbia, MO: University of
Missouri Press: 79-102.

Labov, William 2003. Ch 11 of the Atlas of North American English. Online:
http://www.ling.upenn.edu/phono_atlas/Atlas_chapters/Ch1 l/Chl l.html

-—-. 1994. Principles of Linguistic Change: Internal Factors. Oxford: Blackwell.

—. 2001. Principles of Linguistic Change: Social Factors. Oxford: Blackwell.

Labov, William, Sharon Ash, and Charles Boberg. 1997. A national map of regional
dialects of American English. Online:
http://www.ling.upenn.edu/phono_atlas/NationalMap/NationalMap.html.

Labov, William, Malcah Yaeger, and Richard Steiner. 1972. A quantitative study of
sound change in progress. Philadelphia: US Regional Survey.

Lance, Donald. 1974. Dialect Divisions in Missouri. Paper presented to the Midwestern
Regional Meeting of the American Dialect Society.

118

--— 1993. Some Dialect Features in the Speech of Missouri Germans. Heartland
English: Variation and Transition in the American Midwest, ed. by Timothy
Frazer. Tnscalaoosa: University of Alabarrra Press. 187-198

-—- 2003. The Pronunciation of Missouri: Variation and Change in American English.
American Speech. 78: 255-284.

Majors, Tivoli. 2003. Acoustic examination of the low back merger in Missouri speech.
New Ways of Analyzing Variation 32 conference: University of Pennsylvania,

 

Philadelphia, Pennsylvania.
Marckwardt, Albert H. 1980. American English: Second Edition, revised byJL Dillard. F
New York: Oxford University Press. E
Meyer, Duane. 1963. The Heritage of Missouri: A History. St. Innis, MO: State ,
Publishing Company. ':
Milroy, Lesley. 1980. Language and Social Networks. Oxford: Blackwell. E

Milroy, Lesley and Matthew Gordon. Sociolinguistics : method and interpretation.
Malden, MA: Blackwell.

M11330. 1997. Hoosier Inve. Press. Asian Man Records.

Murray, Thomas E. 1986. The language of St. Innis, Missouri: Variation in the Gateway
City. Frankfurt am Main: Peter Lang.

— 1987. ' You @#$%’\&’ hoosierl': Derogatory rrarrres and the derogatory name in St.
Innis, Missouri. Names. 35:1-7.

——- 1993. The Language of St. Innis, Missouri: Dialect Mixture in the Urban
Midwest. Heartland English: Variation and Transition in the American Midwest,
ed. by Timothy Frazer. Tnscalaoosa: University of Alabama Press. 125-136

—— 2002. Language variation and change in the urban Midwest: The case of St. Innis,
Missouri. Language Variation and Change. 14: 347-361.

Neary, TM. 1977. Phonetic feature systems for vowels. University of Connecticut
dissertation.

Peterson, GE and H. Barney. 1958. Control methods used in a study of the vowels.
Journal of the Acoustical Society of America. 22: 75-84.

Preston, Dennis. 1993. Two Heartland Perceptions of Language Variety. Heartland

English: Variation and Transition in the American Midwest, ed. by Timothy
Frazer. Tnscalaoosa: University of Alabama Press. 23-48

119

— 1999. A language attitude analysis of regional US speech. Cuadcmos dc Filologia
Ingiesa 8: 129-46

Seely, Mike. 2003 (March 26). Hoosiers. The Riverfront Times. Online:
http://www.riverfronttimes.com/ issues/ 2003-03-26/1eature.html/ 1 /index.htrnl

Stevens, Kenneth. 1998. Acoustic phonetics. Cambridge, MA: MIT Press.

Warner, W. Lloyd. 1960. Social class in America: a manual of procedure for the
measurement of social status. New York: Harper.

"Arm‘s...

 

Ell

 

120