PORTLAND DIALECT STUDY: THE FRONTING OF /OW, U, UW/ IN PORTLAND, OREGON by MICHAEL WARD A thesis submitted in partial fulfillment of the requirements for the degree of MASTER OF ARTS in TEACHING ENGLISH TO SPEAKERS OF OTHER LANGUAGES Portland State University 2003 ABSTRACT An abstract of the thesis of Michael Ward for the Masters of Arts in Teaching English to Speakers of Other Languages presented December 10, 2002. Title: Portland Dialect Study: The Fronting of /ow, u, uw/ in Portland, Oregon The fronting of the mid and high back vowels /ow, u, uw/ differentiates the major dialects of English in North America. In particular, the fronting of these vowels has been described as a feature of two regional vowel movements: the California Movement and the Canadian Shift. The purpose of the present study is to document the production of these vowels in the speech of residents of Portland, Oregon, and see if this Pacific Northwest speech illuminates dialect research by seeing how patterns here conform or do not conform to patterns others elsewhere.
Data were collected from a sample of eighteen Portland speakers. Nine males and nine females from three different age groupings and two different social classes were interviewed in order to elicit naturalistic speech. Word-sized files containing relevant vowel tokens were extracted from each interview. Each vowel token was then analyzed in a three-step process: 1) Vowel formants were calculated and measured with spectrographic analysis.
2) The data from each speaker were normalized to account for physiological differences. 3) Vowel format measurements were then correlated with social factors, including age, gender and class. The study found that a pattern of fronting of /ow, u, uw/ was present across age groups, with “Young Adult” speakers showing the greatest degree, followed by “Teen” speakers. Patterns of fronting relative to gender and social class were not as salient, although “Female” and “Working Class” speakers generally showed a greater degree for fronting compared to “Male” and “Middle Class” speakers.
Fronting among ‘Young Adult” speakers provides evidence for sound change in Portland, similar in nature to preliminary descriptions of both the California and Canadian. TABLE OF CONTENTS LIST OF TABLES. vi LIST OF FIGURES.vii CHAPTER ONE: INTRODUCTION. GOAL FOR THE PORTLAND DIALECT SURVEY.
PRESENTATION OF FINDINGS: WEBSITE. 6 CHAPTER TWO: LITERATURE REVIEW. Significance of F1 and F2. Vowel measurement conventions.3 SOCIALLY DEFINED VARIATION.
Age and the apparent time construct. CHAIN SHIFTING AND LANGUAGE CHANGE. Historical vowel classes. North American English chain shifting pattern.
32 REGIONAL DIALECT STUDIES. Lexical studies of the Pacific Northwest. Phonology-based regional studies. The California movement.
The Canadian vowel shift. Where does Portland fit in?. 43 CHAPTER THREE: METHODOLOGY AND DESIGN. Defining social class.
Instruments and procedures. Selection of vowel tokens. 61 CHAPTER FOUR: RESULTS. Vowel class expansion.
Grand speaker means for post-vocalic environments. Combined social categories. 76 CHAPTER FIVE: CONCLUSION. REVIEW OF RESEARCH QUESTIONS OR HYPOTHESIS.
Are /ow, uw, u/ fronting in Portland?. Variability across age groups. Variability across gender. Variability across gender.
Is Portland participating in the Western dialect area?. LIMITATIONS OF THIS STUDY. DIRECTIONS FOR FUTURE RESEARCH. ADDITIONAL DATA TABLES.
VOWEL CHART OF IPA VOWELS. INFORMED CONSENT LETTER. INTERVIEW WORD LOG FILE. 106 iv LIST OF TABLES Table 1.
Labov’s vowel notational system and IPA equivalents. Data sample of speaker formant values before and after normalization for “Marcia,” Middle Class Female, age 17. PDS Subjects for present study. Socio-economic index for Portland.
Selection of vowel tokens. Vowel classes under variation analysis. Young Adult, Teen, Older Adult and Grand means. Age comparison of speaker means: OA vs.
Gender comparison: Female and Male means. Social class comparison of speaker means: Working Class vs. Young Adult gender and social class comparison of speaker means (combined table): Female vs. Male; Middle Class vs.
Age comparison of F2 means: YA vs. Teen and YA vs. Age comparison: Statistical analysis of differences. Gender comparison: Statistical analysis of differences.
Social class comparison: Statistical analysis of differences. Female speaker means. Male speaker means. 102 v LIST OF FIGURES Figure 1.1 Hierarchical display of North American English dialects.
Schematic of vowel acoustic space and phonological processes. Relationship of F1 and F2 to vowel position. Formant chart of selected vowels. Northern Cities Vowel Shift.
Southern Vowel Shift. Comparison of American English dialectical divisions. The California vowel movement. Canadian Vowel Shift.
Spectrographic analysis of the words “those” and “totally”. Grand means (all speakers) relative to post-vocalic environment. Age comparison: OA vs. Curvilinear distribution of frontedness by age.
Gender comparison: Female vs. Social class comparison: Working Class vs. Young Adult gender comparison: Male vs. Young Adult social class comparison: Working Class vs.
Vowel chart of IPA. 105 vi CHAPTER ONE: INTRODUCTION The word on the street is that Portlanders don't have an accent—or at least nothing as easily caricatured as the speech of Texans, New Yorkers or California surfer dudes, all of whom native Portlanders immediately recognize as outsiders. Linguists, on the other hand, recognize that everybody speaks a distinct variety of English. What distinguishes Portland speech is not known—we can only know if there is a unique Portland speech variety, or dialect, by identifying the features that would make it so, and by comparing those features to other speech varieties.
As of yet, there is no strong linguistic evidence to suggest that the Portland dialect is unique per se, mostly because the data collected are incomplete. Because of this, we can’t say very much at all about Portland speech until it has been properly and systematically studied. GOAL FOR THE PORTLAND DIALECT SURVEY The goal of the Portland Dialect Survey (PDS) is to develop a comprehensive database that documents the character of Portland-area speech through the systematic description of its most salient features. The PDS could ultimately determine whether or not Portland speech is distinctive, and if so, in what ways.
Provided with this database, the PDS will explore three main research strands: comparing Portland speech data to that generated by other regional dialect studies; describing language change taking place in Portland; and investigating the underlying social structure and its relationship to language as revealed by the former two strands. The PDS, in turn, will complement the efforts of others describing the many varieties of English. Dialect studies of the past have mostly focused on differences in word choice, forming lexically-defined regional isoglosses that roughly manage to divide the United States into three main dialectical areas: The North, the Midlands and the South (Carver 1987). Yet Labov (2002) claims these lexical isoglosses fail to show structural relationships that would allow for comparison among different regions and this has contributed to reluctance among traditional dialectologists to define discreet dialectical boundaries.
Labov has instead focused almost exclusively on the acoustic analysis of speakers’ vowels. Instead of purely regional features, he extends the study to social features such as class and gender that may influence dialectical differences among speakers within regions. By attempting to make a systematic study of all the phonological relations in the vowel system across the United States and by placing it within a social context, Labov seeks to more closely define the dialectical boundaries of American English than have previous researchers. The following figure by Labov shows the relationships among the North American English dialects that he has defined to date: 2 Figure 1.
Hierarchical display of North American English dialects (Labov 2002) Each main area he defines is characterized by fairly rapid language change involving the chain shifting of different vowel systems (to be explained later in full detail). As he states, “It is radical rotations of vowel systems, and not differences of inventory, that account for the greatest differences between vowel systems. In these rotations, whole sets of vowels reverse their relative positions to each other; phones that represent one phoneme in one dialect represent an entirely different phoneme in 3 another” (Labov 1991: 3). above illustrates these chain shifts often result in one salient feature, either a merger or a split, separating different dialects.
The West, where Portland, Oregon is physically located, has not experienced the degree of change undergone by dialects in the North, Midlands and South. It is most easily identified by the merger of /oh/ (as in caught) and /o/ (as in cot). This reduces the inventory of back vowels, and subsequently there is no raising of the vowel /ow/ (as in coat). To some extent, the relative stability of /æ/ (as in cat) also prevents any fronting of back vowels.
However, Labov recognizes that the stability of the West has also been overestimated (Labov 1991: 33) and there is indeed a fronting of the back vowels described in a few Western areas. Some work has been done describing varieties of Utah speech (Di Paolo 1990); California speech (Luthin 1987), (Hinton, Bremmer, Corcoran, Learner, Luthin & Moonwoman 1987); and similarities between California and Canadian speech (Clarke, Elims & Youssef 1995). Yet for the most part, no comprehensive effort has been made to systematically describe the West. In addition, Labov’s TELSUR project, a telephone survey of hundreds of speakers from throughout the United States, is still pending completion.
The PDS may help to ultimately determine if the West is as cohesive as Labov seems to suggest by showing how Portland does or does not differ from other previously studied Western dialects. 4 So far two other Portland State University Master's students have contributed to the PDS by examining Portland speech. The first is Wolff (2000) Portland Dialect Survey: High Rising Contours in Portland Speech which deals with pitch contours, what have been called “upsweep” or “High Rising (Non-final) Terminals.” Pitch is a suprasegmental feature that gives even declarative statements a question-like intonation. Because her study was limited to just this feature, her findings do not generalize to Portland speech behavior as a whole.
Conn (2000) The Story of /æ/ in Portland, seems to uncover some uniqueness in Portland speech, although ironically it is in the direction of /æ/ lowering rather than raising as he had anticipated. These two studies represent only limited snapshots of Portland speech. RESEARCH QUESTIONS Does Portland has a distinct dialect and if so, what are its distinguishing features? This larger question must be addressed piecemeal by investigating candidate features of a Portland dialect. The focus of this study is three vowels: /ow/ (as in coat), /u/ (as in could) and /uw/ (as in boot).
The PDS–the purpose of which is to create a detailed and accurate description of Portland speech—will amalgamate the separate findings into a coherent picture. The specific questions of this study are: 1. Is the fronting of the mid back vowel /ow/ and high back vowels /uw/ and /u/ taking place in Portland, Oregon? a. Does the general phenomenon show variability across age groups, suggesting language change? 5 b.
Does it show patterned variability between genders and among social groups, providing evidence that particular groups can be identified as leading the change, while other groups demonstrate a more conservative speech style? 2. If so, does this provide evidence for Portland speaker’s participating in the Western dialect area, or does it constitute a separate trend forming a dialect isolate? What kind of social factors can help explain this change or lack of change? 1. PRESENTATION OF FINDINGS: WEBSITE In addition to the written thesis itself, the thesis will be made available on the PDS website. The PDS website will detail the ongoing results of the Portland Dialect Survey which will include the findings of PSU students and researchers as they relate to the PDS.
NOTATIONAL SYSTEM For describing vowels and vowel classes, this thesis employs a notational system developed by William Labov (1994), which will be explained in greater detail in Section 2. of the next chapter. However, a table is provided below to assist the reader in understanding the system as it relates to the International Phonetic Alphabet (IPA).