The University Next Door: Sound Change in an Urban Rochester Neighborhood

Abstract

Recent studies have documented the retreat of the Northern Cities Shift (NCS) in the Inland North, as well as the advance of a split between pre-nasal and pre-oral /æ/ (Bat/Ban Split) and a merger of /ɑ/ and /ɔ/ (cot and caught, Low-Back Merger) in dialects throughout North America. The present study investigates the status of these three changes-in-progress for fifteen speakers in an urban neighborhood of Rochester, New York, located in close proximity to a university. Our analysis draws on a series of oral history interviews collected in collaboration with a neighborhood organization, alongside participant observation in the community. We describe the clear progression in apparent time of the three dialect features within one Rochester family, then zoom out to the sample as a whole. In the larger sample we find evidence for the presence of both the “new” and “old” dialect features but find that it is not as simple as advance and retreat in apparent time. Rather, we conclude that the social meaning of the NCS is undergoing incipient change and displays a great deal of variability as a marker of localness in the community. Our findings also highlight an underexplored social factor in the study of language change: interaction with higher education, particularly as it relates to sense of place. Rather than analyzing education as a proxy for social class, we propose that higher education impacts sound change in a more nuanced way: through its impact on local culture and social networks, as well as its influence on speakers’ access to and awareness of local and supralocal features and evaluation of what it means to be local.

Keywords

language variation and change Inland North dialect region Northern Cities Shift American English higher education

1. Introduction

The Northern Cities Shift (NCS), first described by Labov, Yeager, and Steiner (1972), is the defining feature of the Inland North dialect of North American English (Labov, Ash & Boberg 2006), affecting primarily five lax vowels, especially trap and lot. Its more recent reversal, documented throughout the Inland North from New York to Wisconsin (Niedzielski 1999; Labov 2010; McCarthy 2011; Driscoll & Lape 2015; Wagner, Mason, Nesbitt, Pevan & Savage 2016; Milholland 2018; Nesbitt 2018, 2021; Durian & Cameron 2020; Thiel & Dinkin 2020), has been attributed to a growing awareness of the NCS and increasing levels of stigma associated with it (Bigham 2010; Driscoll & Lape 2015; Savage, Mason, Nesbitt, Pevan & Wagner 2015; King 2018; Milholland 2018; Nesbitt 2018; Nesbitt, Wagner & Mason 2019).

Concurrent with the retreat of the NCS and affecting a subset of the same vowels, two sound changes are currently advancing in many varieties of North American English: the divergence of the pre-oral /æ/ of bat and the pre-nasal /æ/ of ban (the Bat/Ban Split; also known as the nasal system [Labov, Ash & Boberg 2006] or Nasal Short-a System [Dinkin 2011]), and the merger of /ɔ/ and /ɑ/ (the Cot/Caught or Low-Back Merger). Both of these have been observed to be advancing in the Inland North: King (2018) finds that the Bat/Ban Split correlates with the rejection of the NCS, while multiple studies (Bigham 2010; McCarthy 2011; Nesbitt, Wagner & Mason 2019) have found connections between the Low-Back Merger and the retreat of the NCS.

In the present study, we investigate the status of the NCS, the Bat/Ban Split, and the Low-Back Merger in Rochester, New York. In the context of an urban neighborhood with close and complicated connections to the University of Rochester, we explore especially the role of higher education in this reorientation and in the progression of these sound changes.

Full-time enrollment in institutions of higher education in the United States almost doubled from 1970 (~5 million) to 2019 (~10 million),¹ and yet the significance of higher education as a factor in language change remains under-investigated. Studies examining the role of higher education in language change have generally used education level as a proxy for social class (e.g., Wagner 2008; Prichard 2016; Duncan 2018), including in the Inland North (Thiel & Dinkin 2020). As the recent NCS reversal has been associated with higher levels of education, this is typically interpreted as one indication that NCS-reversal is a change from above, with speakers of higher socioeconomic status leading in an ongoing change. Beyond serving as a proxy for class, however, access to higher education changes speakers’ access to linguistic features and may change their orientation toward localness. Prichard (2016) has found that higher education, especially at a nationally oriented university, can predict fewer local speech features and more supralocal ones, while Bigham (2010), Sayahi (2011), and McCarthy (2011:183) have pointed out that universities may be important sites of language contact. Moreover, both Eckert (1988) and King (2021) have demonstrated the value of understanding multiple individual motivations for the adoption or rejection of ongoing changes in the early adoption and now reversal of the NCS. Taken together, these point toward the need for a more complex analysis of the role of higher education in sound change, particularly in communities where universities have a large social and economic presence.

In the rest of this paper, we present additional background information about the sound changes in question, about higher education and language change, and about Rochester and the neighborhood we worked in. We briefly discuss some of our quantitative results with respect to the status of the NCS, the Bat/Ban Split, and the Low-Back Merger in relation to several social factors in the community, before turning to a more qualitative analysis of individuals’ connections to higher education and the social meaning of the three dialect features in the community. Our initial hypotheses were as follows: (1) that the NCS in Rochester’s 19th Ward is demonstrably rising above the level of consciousness as it is in other Inland North cities, and retreating in apparent time; (2) that speakers in the 19th Ward are at the same time adopting supralocal norms like the Bat/Ban Split and the Low-Back Merger; and (3) that association with higher education impacts local speech patterns in a more nuanced way than previously described, particularly when we focus on the variable behavior of individuals with respect to the adoption of supralocal speech forms and the retention of local forms. We provide some quantitative evidence for change in apparent time but mainly focus on qualitative evidence for the status of the NCS as a stereotyped marker of local speech by using a case study of an individual family. We argue that the vowel spaces of individual speakers reflect their varied relationships with higher education and their different ways of constructing identity with respect to Place.

2. Background

First described in depth in 1972 (Labov, Yaeger & Steiner), the NCS (Figure 1) is conjectured to have originated in the nineteenth century during the period of westward migration from New England (Labov, Ash & Boberg 2006; Dinkin 2009). Early scholarship documented the NCS’s advance throughout the Inland North (Labov, Yaeger & Steiner 1972; Eckert 1988; Niedzielski 1999; Evans, Ito, Jones & Preston 2000; Clopper, Pisoni & de Jong 2005; Roeder 2010; Durian & Cameron 2020). It was argued that because the NCS had “not yet risen to social consciousness and [was] not subject to social correction” (Labov, Yaeger & Steiner 1972:13), Inland North speakers “did not hear the shift in their own speech” (Niedzielski 1999:81).

Figure 1.

The Northern Cities Shift

More recently, the NCS has been widely documented as retreating in the same regions where it was previously found to be advancing (Niedzielski 1999; Labov 2010; McCarthy 2011; Driscoll & Lape 2015; Wagner, Mason, Nesbitt, Pevan & Savage 2016; Milholland 2018; Nesbitt 2018, 2021; D’Onofrio & Benheim 2020; Durian & Cameron 2020; Thiel & Dinkin 2020). In Upstate New York, findings from Ogdensburg (Thiel & Dinkin 2020), Cooperstown (Dinkin 2022), Syracuse (Driscoll & Lape 2015), Buffalo (Milholland 2018), and Rochester (King 2018, 2021) suggest a retreat of the NCS concurrent with a rising awareness of its features and an association with a blue-collar Rust Belt identity from which younger speakers in particular are trying to distance themselves. In perceptual dialectology maps and other contexts, Syracuse speakers characterized the vowels of their region as “horrible,” “flat,” and “harsh”; one described consciously avoiding using the fronted and raised bat (Driscoll 2016). Moreover, King’s (2018, 2021) work has demonstrated that NCS in Rochester functions in the region as a local linguistic resource that is closely associated with urban and working-class populations (and not only white speakers, as had previously been described)—in particular King identifies a retreat from the shift as a way for a persona she describes as the Mobile Black Professional to orient away from an industrial identity and “toward supralocal economic interests” (King 2018:36). This is consistent with findings from elsewhere in the Inland North (e.g., Nesbitt 2018; Durian & Cameron 2020).

Another ongoing change in progress in several varieties of North American English (including in the California Vowel Shift; Eckert 2008) is the divergence of the pre-oral /æ/ of bat and the pre-nasal /æ/ of ban. The Bat/Ban Split and the NCS both impact the bat vowel, but in contradictory ways. In contrast with the NCS, which pulls /æ/ forward and upward in all environments, the Bat/Ban Split, which is also known as Ban-raising or the short-a split, moves /æ/ forward and upward only in the pre-nasal position (ban). Thus, while the Bat/Ban Split and the NCS both promote the raising and fronting of ban, they move bat in opposite directions. This predicts the following: (1) that the two features, being contradictory to each other, would resist occurring together and (2) that the advance of one would correlate with the retreat of the other. While Dinkin (2011) has observed the simultaneous presence of the NCS and the Bat/Ban Split in some Upstate New York speakers, it seems to be an exception rather than a rule. Dinkin interprets the coexistence of the two systems as a sign of a not fully present NCS; a speaker who has truly adopted the NCS is unlikely to have the Bat/Ban Split. Dinkin argues that the non-split bat/ban is actually (although not exclusively) a feature of the Inland North dialect, of which the NCS is the defining feature. Dinkin (2009:198) has also argued that the Bat/Ban Split can act as an inhibitive force on the raising of bat by the NCS. In Rochester, King (2018) finds evidence for the adoption of the Bat/Ban Split in Mobile Black Professionals, led by women, younger speakers, and those with plans to leave Rochester for economic reasons. For King’s (2018) speakers, the advance of the Bat/Ban Split correlates with the rejection of the NCS.

Another advancing feature in American English is the merger of /ɔ/ and /ɑ/ (the Cot/Caught or Low-Back Merger). This is well attested in dialects throughout North America, including the Elsewhere/California/Western dialect (Labov, Ash & Boberg 2006; Irons 2007; Fridland 2015; Podesva, D’Onofrio, Van Hofwegen & Kim 2015), Canadian English (Clarke, Elms & Youssef 1995; Labov, Ash & Boberg 2006), and in West Virginia (Hazen 2005; Labov, Ash & Boberg 2006), Kentucky (Irons 2007), Eastern New England (Labov, Ash & Boberg 2006), Western Pennsylvania (Labov, Ash & Boberg 2006; Fridland 2015), Southern Illinois (Bigham 2010) and the South (Labov, Ash & Boberg 2006; Fridland 2015). There is also evidence for its spread into the Inland North region (Dinkin 2009; Bigham 2010; Nesbitt, Wagner & Mason 2019; Dinkin 2022), but no recent examination of these vowels in Rochester.

Both of the vowels affected by the Low-Back Merger are also involved in the NCS. While the lowering and fronting of caught is compatible with both phenomena, only the NCS motivates the lowering and fronting of cot; the Low-Back Merger encourages the movement of cot in the opposite direction. This interaction has been explored in depth in Lansing, Michigan by Nesbitt, Wagner, and Mason (2019), in Southern Illinois by Bigham (2010), and in Cooperstown, New York by Dinkin (2022). They have found the advance of the merger to be happening concurrently with, and perhaps even as part of, the retreat of the NCS. A possible link between the two phenomena has also been found in Chicago (McCarthy 2011).

In studies of sound change, including many of those cited above, education level has, because of its role as a vehicle for social mobility, often been used as a proxy for social mobility, social class, and associated ideologies (e.g., Wagner 2008; Prichard 2016; Duncan 2018). More importantly, orientation toward higher education—either through enrollment at a university or simply through ambition to enroll—can be a marker of upward social mobility and interest in moving beyond the local community (De Decker 2006; Wagner 2012) and can promote the rejection of local linguistic variants as well as the adoption of supra-local standards (Wagner 2012). These findings suggest that as the number of college-educated (and college-aspiring) Americans increases, we may find more adoption of supra-local norms, whether this is associated with changes in socioeconomic status or not.

In a study that specifically examined the role of higher education in local sound changes, Prichard (2016) found that Philadelphia speakers attending national universities were leading in the adoption of the nasal bat system and in the retreat from local, more stigmatized vowel patterns. Higher education, especially at a nationally oriented university, was found to predict fewer local speech features and more supralocal ones.

Universities are also settings of language and dialect contact (Bigham 2010; Sayahi 2011). McCarthy (2011:183) has suggested that because universities attract “people from all across North America, the university-educated simply have more opportunity to come into contact with speakers with a range of American dialects.” Bigham (2010), for example, found that university students in southern Illinois were converging upon a vowel space in between the vowel spaces of the contact varieties (including the Inland North dialect). As universities are sites of dialect contact, university students and others associated with this university are thus also in a position to develop a more heightened awareness of their own speech in comparison to others and to be introduced to a wider variety of linguistic features for potential adoption (Bigham 2010; McCarthy 2011).

With respect to the Inland North, in Buffalo (Milholland 2018), Syracuse (Driscoll 2016), Ogdensburg (Thiel & Dinkin 2020), and Chicago (McCarthy 2011), the decline of the NCS is associated with higher levels of education, and more highly educated speakers show less NCS-like vowels, especially for bat. These findings initially suggest that the retreat from the NCS is a change from above, with speakers of higher socioeconomic status leading in an ongoing change. But if we view these findings alongside previous findings with respect to higher education and language change, we may also want to consider the role that exposure to higher education plays in promoting a shift toward supralocal norms beyond its use as a proxy for socioeconomic status. As King (2021) has shown that the social meaning of the NCS and the retreat from NCS in Rochester goes beyond a simplified understanding of race and draws on a more nuanced analysis of local-ness, we consider here that in thinking about the NCS in Rochester we also need to move beyond a simple characterization of higher education as a marker of social class and instead consider connections to institutions of higher learning as part of a more complex ideology of what it means to be local in Rochester.

Rochester, located in the western part of New York State, is the fourth largest city in the state, with approximately 206,000 residents at the time of this study (U.S. Census Bureau 2018). Much like the Rust Belt region as a whole Rochester has experienced a long period of economic decline, leading to demographic shifts as people—especially those younger and more highly educated—leave Rochester for economic opportunities. This changing economy has resulted in a rising importance of higher education in the local economy, which is now dominated by the presence of several colleges and universities. Currently the largest employer in the area is the University of Rochester.

For the present study, we have chosen to focus on Rochester’s 19th Ward (also known by residents as “the Ward”), which is located directly across the Genesee River from the University. The Ward is the largest and most racially and economically diverse neighborhood in Rochester with a population of about 22,000 people (19wca.org). The proximity of the neighborhood to the University, as well as the complicated relationship between the city and the University, which we describe later, make the Ward an ideal location to explore the nuances of access to higher education and the role of “town and gown” relations in language change.

The complex relationship between the University of Rochester and the city and neighborhoods around it is neither new nor unique. The concept of town and gown relations—the interactions between universities and the communities in which they are located—is an old one, and recent work demonstrates the economic, social, and cultural impact that universities and their cities have on one another (Pickford 1989; Bender 1991; Mayfield 2001; Massey, Field & Chan 2014). Increasingly, cities and universities, including the University of Rochester and the City of Rochester, have been trying to foster more mutually beneficial relationships through programs such as the University of Rochester’s Community Ambassador program, although much progress remains to be made. For individuals, the relationship between “town” and “gown” is also changing, as more and more people in the United States are now getting a college education.

Beyond higher education, Rochester is also an ideal location to explore the interactions of the NCS, Bat/Ban Split, and Low-Back Merger. Labov, Yaeger, and Steiner’s seminal (1972) data documenting the NCS came from Rochester and Buffalo. Meanwhile, the retreat of the NCS and the advance of the Bat/Ban Split and Low-Back Merger are demonstrably progressing in Upstate New York in general (e.g., Dinkin 2009, 2022; Driscoll 2016; Milholland 2018; Thiel & Dinkin 2020), but the study of individual motivations behind the retreat is still early, especially in Rochester. With the exception of King’s (2018, 2021) work cited above, individual motivations for the retreat of the NCS in Rochester remain under-explored, and their connection to town-and-gown relations has never been investigated.

3. Methodology

Much of the foundation for our consideration of town-and-gown in this study comes from insider knowledge of the University of Rochester. The first author was a University of Rochester undergraduate student from 2015 to 2019, and the second author has worked at the University of Rochester since 2015. In this capacity we have had access to many informal conversations with students, faculty, and staff about the 19th Ward and the relationship between the University and the neighborhood.

In addition to our experience with the University, this study is also informed by three years of participant observation in the 19th Ward by the first author from 2016 to 2019. Initially, her time in the Ward was infrequent and consisted mainly of participation in small community service events such as tree-planting; as time progressed, she became more involved and completed the majority of her participant observation in the summer of 2017 and from June to October 2018. During this time, Author Kapner worked with members of the 19th Ward Community Association (19WCA) to catalog and curate the 19WCA’s archival collections chronicling local history. She attended and helped at neighborhood events including outdoor concerts and festivals. Kapner lived in the 19th Ward in Summer 2018 and volunteered weekly at the Westside Farmers’ Market, a popular gathering place which featured, in addition to produce for sale, free entertainment and educational events, and represented a pillar of community identity. Both authors also began a collaboration with the 19WCA on a joint oral history/sociolinguistic project, from which came the recordings used in the present study.

Participants in the project were recruited via word-of-mouth convenience sampling between 2017 and 2018 in a joint effort by both History and Archives (H&A) Committee members and Author Kapner. The participants all identified as residents of Rochester’s 19th Ward (although two—SR and JC—did not at the time actually live there). They ranged in age from twenty to eighty-five, with birth years clustering around the middle of the twentieth century (median birth year = 1956). Nine participants self-identified as men (including one who also self-identified as transgender) and six self-identified as women. A chart of demographic information for the participants is shown in Table 1.

Table 1.

Demographic Information: Age, Race, Gender, Education Levels

Participant	Year of birth (age as of 2018)	Gender	Race	Education level
BL	1943 (age 75)	M	White	Some college
DL	1943 (age 75)	F	White	Graduate
DW	1967 (age 51)	F	Black	Graduate
EM	1961 (age 57)	F	White	Graduate
HR	1962 (age 56)	M	White	Undergraduate
JB	1949 (age 69)	M	White	Undergraduate
JBI	1948 (age 70)	F	White	Graduate
JC	1948 (age 70)	M	White	Graduate
LC	1984 (age 34)	F	White	Graduate
LR	1933 (age 85)	M	White	Undergraduate
MER	1942 (age 76)	F	White	Graduate
MR	1970 (age 48)	M	White	Undergraduate
SB	1956 (age 62)	M	White	Undergraduate
SG	1969 (age 56)	M	White	Some high school
SR	1998 (age 20)	M	White	Undergraduate

Because of the recruitment methods, our group is not well balanced demographically: we included only one Black speaker (the rest were white), and ages skew toward the older end of the range. The makeup of the sample represents the reality of this social network, although it is not a faithful representation of the makeup of Rochester as a whole or even of the 19th Ward. While the neighborhood as a whole is quite diverse by race and age, this does not necessarily translate to diversity within individual social networks: the participants in this study were all members of a social network which consists mainly of older white residents. Our methods also resulted in confounding of certain variables, which ultimately made quantitative analyses difficult. For instance, the group overall was very highly educated, making it difficult to quantitatively assess effects attributable to level of education, and the youngest speakers were also those with higher terminal degrees. This too represents the reality of not only this social network, but of rising education levels in the community we studied, as well as nationally. At the same time, by limiting our sample to a single social network, we are able to understand in greater detail the factors impacting sound change on an individual level. Our application of qualitative analysis methods allowed us to maintain such a focus and also provide insights that set out a starting point for further study. Finally, our sample of fifteen speakers also included three speakers who are members of one family, which further narrows the representative diversity of the sample in some ways. It does, however, allow us to focus more narrowly on the progression of change, which we do in section 4.1.

Participants were informed before participation that this was a joint oral history/sociolinguistic project and that the recordings would be eventually made available to the public. Recording sessions took place from June to November 2018. Sessions began with an oral history interview conducted by the first author and members of the H&A Committee and also included a semantic differential task and a word list. The study materials are available from the authors by request. Following the completion of the word list, we often had an informal metalinguistic conversation with the participant(s) about the Rochester accent. We obtained just over one hour of speech from each speaker, and all but one interviewee completed the word list. The oral history recordings were publicly archived with the Rochester City Library as part of the Rochester Voices project (RochesterVoices.org).

Recordings were transcribed manually in Praat (Boersma & Weenink 2023). The recordings were force-aligned and the first and second formants measured using DARLA² (Kendall & Thomas 2010; Reddy & Stanford 2015). The measurements were averaged and normalized with the Lobanov (1971) method using NORM (Thomas & Kendall 2007), using first and second formants only. We completed additional post-measurement processing and sorting using custom programs in Python. A total of 18,900 tokens were included in the analysis. To quantify speakers’ degree of Bat/Ban Split and Low-Back Merger, we used a custom Python program to find the Euclidean distances between speakers’ mean normalized bat and ban vowels and between their mean normalized cot and caught vowels. To avoid the confound of the Bat/Ban Split, pre-nasal /æ/ was removed from all analyses except for the Euclidean distance measurements. All Euclidean distance and NCS scores were calculated separately for each speaker’s word list and interview.

As a way to measure speakers’ level of adoption or rejection of the NCS beyond measurements of individual vowels, we used Dinkin’s (2013) version of Labov’s (2007) criteria for the NCS to calculate a score for each speaker (Table 2). A speaker’s NCS score was the number of criteria they exhibited, with each of the two components of the EQ criterion counting as 0.5 point toward their score. The Labov criteria are calculated with benchmarks from the Atlas of North American English (ANAE; Labov, Ash & Boberg 2006), which are in turn based on the Labov ANAE normalization method rather than the Lobanov method. We therefore used Dinkin’s (2021) Lobanov-transformed versions of these benchmarks in calculating the scores. We produced all vowel space plots in R (R Core Team 2023).

Table 2.

NCS Criteria

Criterion	Description	Measurement method
AE1	Raised bat (pre-oral only)	Mean bat F1 < 0.329
O2	Fronted cot	Mean cot F2 > −0.219
EQ	Relative height and fronting of dress and bat are reversed	Mean dress F1 > mean trap F1 and mean dress F2 < mean trap F2
ED	Difference between mean dress F2 and mean cot F2 is less than 0.862	Mean dress F2 − mean cot F2 < 0.862
UD	cot is more fronted than strut	Mean cot F2 > mean strut F2

We conducted all statistical tests using mixed-effects linear regression models with the lme4 package (Bates, Mächler, Bolker & Walker 2015) in R (R Core Team 2023). In our analysis we considered the social factors that we hypothesized to be meaningful, including year of birth, education level, and a measure of degree of connection to higher education (which we termed HEI, Higher Education Index). We also included speech style (word list or oral history interview) and type of university attended as nominal variables, as well as the random effects of individual speaker and word. Below, we report the results of our analysis from participant observation, vowel measurements for individual speakers, and the content of the interviews, along with some of our quantitative findings.

4. Results

We begin this section by presenting a case study of sound change in a single Rochester family, which we use to set the stage for and motivate the rest of our analysis. We then zoom out to the broader group, describing our observations and results with respect to language attitudes, style-shifting and attention to speech, and finally the relationship between the University and the 19th Ward and its effects on local speech patterns. For reference, Table 3 provides each participant’s values for three of the variables we discuss here: NCS score, and Bat/Ban and Cot/Caught Euclidean distance, for the word list and oral history interview.

Table 3.

Participants’ Individual Metrics for the Relevant Variables

Participant	Speech style	NCS score	Bat/Ban Euclidean distance	Cot/Caught Euclidean distance
BL	Word list	0.5	0.41	2.24
BL	Oral history	1.5	0.62	2.00
DL	Word list	1.5	0.60	1.70
DL	Oral history	0.5	1.28	1.42
DW	Word list	1.5	0.67	1.57
DW	Oral history	1.5	1.09	1.73
EM	Word list	0.5	1.38	1.55
EM	Oral history	2.5	0.99	1.46
HR	Word list	1.5	1.17	1.45
HR	Oral history	2.5	0.88	1.50
JB	Word list	0	0.71	1.74
JB	Oral history	0.5	1.14	1.27
JBI	Word list	2.5	1.21	1.92
JBI	Oral history	1.5	0.60	1.97
JC	Word list	2.5	0.71	1.91
JC	Oral history	2.5	0.55	1.68
LC	Word list	0	3.03	1.43
LC	Oral history	0.5	0.89	1.35
LR	Word list	2.5	0.75	1.82
LR	Oral history	3.5	0.85	1.81
MER	Word list	Not recorded
MER	Oral history	4	0.41	1.72
MR	Word list	0	0.86	1.17
MR	Oral history	1.5	0.55	1.20
SB	Word list	3	0.21	1.94
SB	Oral history	3.5	0.36	1.76
SG	Word list	2	0.96	1.48
SG	Oral history	0.5	0.91	1.21
SR	Word list	0	1.56	0.98
SR	Oral history	0	1.61	0.82

4.1. Case Study: Change Within One Family

In this section we explore sound change in apparent time by focusing on one family, three members of which were interviewed for this study. At eighty-five years old, LR was the oldest participant in this study. We interviewed him first with his son, HR, who was fifty-six years old. In a second interview, we interviewed LR, HR, and HR’s twenty-year-old son SR. All three grew up in the 19th Ward and the family is quite close; as such, they represent an interesting case study from which to examine language change over time. And indeed, within this family we find support for the reversal of the NCS and the adoption of supralocal Bat/Ban Split and Low-Back Merger in apparent time. Vowel plots for the three speakers are presented in Figure 2 (LR), Figure 3 (HR), and Figure 4 (SR).

Figure 2.

LR’s Vowel Space (Oral History Interview): (a) Word List and (b) Oral History Interview

Figure 3.

HR’s Vowel Space—Oral History Interview: (a) Word List and (b) Oral History Interview

Figure 4.

SR’s Vowel Space: (a) Word List and (b) Oral History Interview

When we consider features of the NCS, bat and cot are most fronted for LR—the grandfather—and least fronted for SR—the grandson. Bat is also highest in the oldest generation and lowest in the youngest generation. For LR and HR, the lowest vowel in the system is cot, while for SR, the youngest, the lowest vowel is bat. Thus, when we consider these two vowels, the NCS appears to be retreating in apparent time in this family. Moreover, looking more holistically, the NCS scores decline through the generations: LR scores a 3.5, HR scores a 2.5, and SR scores a 0.

For all three speakers, the bat and ban vowels are quite separate, with ban approximately as far front as face and bat farther back, but their Euclidean distance increases with each generation. Similarly, the cot and caught vowels remain distinct for all three speakers, but the distance closes with each generation. For LR and HR, strut falls between cot and caught, which is not the case for SR, the youngest. Thus, in this family, the Bat/Ban Split and Low-Back Merger are advancing with each generation.

These three generations of a 19th Ward family illustrate something in line with our initial hypothesis and with previous studies. Over three generations we see an advance of the supralocal Bat/Ban Split and Low-Back Merger alongside a retreat of the NCS; these results are summarized in Table 4.

Table 4.

Three Generations in One Family

Feature	LR (father)	HR (son)	SR (grandson)	Conclusion
Mean bat F1 (Lobanov Normalized)	0.14	0.31	1.58	Retreat of the NCS
Mean bat F2 (Lobanov Normalized)	1.12	0.99	0.07
Mean cot F2 (Lobanov Normalized)	−0.17	−0.49	−0.99
Mean dress F2 (Lobanov Normalized)	0.20	0.32	0.44
NCS score	4.5	3.5	0
Bat/Ban Euclidean distance (Lobanov Normalized)	0.67	0.96	1.6	Advance of the Bat/Ban split
cot/caught Euclidean distance (Lobanov Normalized)	1.93	1.56	0.82	Advance of the low-back merger

Contra our expectations, however, this family is not fully representative of our sample as a whole. They represent an ongoing reversal of the NCS in apparent time, consistent with previous recent studies in the Inland North, and in this they are representative of larger community norms across the dialect region. However, their pattern is an oversimplification of the more nuanced relationship between the NCS, localness, and the university in Rochester, for which we provide more evidence below.

4.2. The Broader Social Network: Sound Change and Higher Education

4.2.1. Attitudes Toward the NCS

Early in our work in the 19th Ward we found that the bat and cot vowels are highly salient in this social network, as they are anecdotally in other parts of Rochester as well. The raised /æ/ and fronted /ɑ/ are often stereotyped, and Rochesterians frequently characterize the “Rochester accent” with an exaggeratedly raised and fronted /æ/ in “accent” and fronted /ɑ/ in “Rochester.” In our interviews, one participant (BL) described the Rochester accent as “when you have a pair of sacks,” and during the semantic differential task, another participant (JC) guessed, correctly and without prompting from the interviewer, that some words in the task were selected to elicit tokens of /æ/ and /ɑ/ but made no such comments about the words containing the other vowels.

Overall, participants express mixed attitudes toward the NCS, although none of the views expressed in the interviews were fully positive. SG says in his oral history interview that “you either have that Rochester accent or . . . you’ve been blessed,” even though in a later discussion with the interviewers he said that he thinks “it’s kind of a cool accent,” “kind of unique,” and that he does not find it “a negative thing.” MR says he finds the Rochester accent “obnoxious and annoying,” and BL, when asked whether he likes the sound of the Rochester accent, says “no” because “it could grate on you.” At least some participants considered their own awareness unusual. MR mentions the “Great Lakes Vowel Shift” but also commented that “ninety percent of people in Rochester don’t think they have an accent.” In general, our observation was that the participants in this study had relatively high awareness of the existence of a Rochester dialect and of some of its features (namely the bat and cot vowels), and that this awareness is not unique to the 19th Ward. Heightened awareness has not been accompanied by a demonstrable shift away from the NCS for this sample as a whole, however, as we discuss in the next section.

4.2.2. Apparent Time and Attention to Speech

In our quantitative analysis of the sample as a whole the NCS does not appear to be reversing in apparent time. We ran a series of linear mixed-effects regression models with the critical value Bonferroni corrected to p = .05/5 = .01. We found no significant correlation (p > .01) between age and the NCS features we explored: bat F1 or F2, cot F1 or F2, or NCS Score. Similarly, with respect to style-shifting according to task our quantitative results are inconclusive. Using speech task as a proxy for attention to speech (word list versus interview, following Labov 1972), we ran a series of linear mixed-effects regression models, again with a critical p-value of .01. The word list style predicted a very slightly lower bat F1 (estimate = −0.31; p = .00008) and cot F1 (estimate = −0.21; p = .0009) but failed to yield any significant effects for NCS score or bat or cot F2. Looking at individual speakers, we find that some individuals showed little difference between the tasks, with SR, SB, and BL showing little difference in their NCS score between tasks, while others showed a difference, but not always in the same direction: JB and DL show a greater distance between bat and ban in the oral history interview, while LC and JBI show a greater distance between bat and ban in the word list. Moreover, whether or not speakers exhibited differences between the two tasks with respect to these features was not cleanly associated with their expressed views about the NCS. But exploring this sample in the context of complex town and gown relations yields insights to better understand the patterning in this community.³

4.2.3. The University and the Neighborhood

As participant observers and from the content of the oral history interviews, we were able to glean substantial knowledge of the relationship between the University and the 19th Ward in general and of specific participants’ relationships with the University of Rochester and with higher education more generally. Two themes that emerged throughout were the complexity of the relationships and the ways that many 19th Warders place local identity in opposition to the University.

Because of its proximity and economic prominence, the University of Rochester has a significant influence in the 19th Ward, but it is a complicated relationship. The Genesee River is a physical and symbolic separator between “town and gown,” with the University campus on the east side and the Ward on the west. For many years there was a “taboo” (JB) on student housing on the West side of the river, and students rarely ventured into the Ward. The construction in 1991 of a foot bridge between the campus and the neighborhood reflected and supported a growing interest—at least by the University—in breaking down the divide. In the words of JB, it “changed everything.” Since that time, the University has constructed two student dormitory areas on the Ward side of the river, a decision met with mixed reactions by Ward residents. Many students also “buy or . . . rent houses and apartments in the neighborhood” (SR). At the time of this study, there were multiple new businesses opening in the Ward targeting student customers, but SR, at the time a University of Rochester undergraduate, observed that “people don’t go [to the Ward] so much just to hang out.”

On one hand, residents acknowledge that “the University of Rochester has surpassed Kodak and Xerox as . . . [Rochester’s] biggest employer, so [that is] a benefit for . . . the southwest quadrant” (JC), that is, the section of the city containing the 19th Ward. JB credits the “huge explosion of the U of R student population” with helping to raise property values and improve maintenance of buildings in the Ward. Residents appreciate the economic benefits of students for Ward businesses and the availability of volunteers for community service projects.

On the other hand, many residents express concern that the growing student presence in the Ward “is changing the character of the neighborhood” (JB) as students “don’t get a chance to develop roots” (JC) and the houses students rent are often “held by absentee landlords” (EM). In the view of some residents, students are not invested in the community, but bring noise, gentrification, and over-policing (EM).

For their part, students have mixed views of the Ward. The first author recalls being told upon her arrival as a first-year undergraduate—by many people including alumni and members of the University police force—that the 19th Ward was “dangerous” and that she should avoid spending time there. LC, a Ward resident and University employee, says that “hearing the way that students and faculty and staff talk about or assume” things about the neighborhood is “frustrating.” Students live in the Ward because it is close to Campus, and non-University-owned housing in the Ward is substantially cheaper than University-owned housing. They do not, in the assessment of both the students and Ward residents, live in the Ward out of particular interest in neighborhood culture and community; most would agree that there is a “lack of connection” (EM). In our observations, even students who live in the Ward rarely venture more than two or three blocks into the large neighborhood and rarely engage with Ward businesses.

Strong relationships between people “from the University” and Ward residents are perceived as rare, at least by those we spoke with. Only four of the fifteen speakers we recorded for this project had current or official ties to the University, and as a student Author Kapner felt herself to be an anomaly in her strong relationships with the Ward. EM describes an “us versus them” attitude between students and Ward residents. Efforts to build a better relationship between the University and the Ward are ongoing and supported by individuals on both sides but are finding limited success. At the time of the interviews, JB was directing the University’s Neighborhood Ambassadors program, which tried to connect students with the Ward to promote better relations; the impact of the program was unclear. In the words of JC, “the challenge for us is to welcome the U of R presence here, engage them as partners and work together.” “Movement ahead is slow” (JC).

Despite their relative lack of formal affiliation with the University of Rochester, the people in the social network interviewed for this study are as a whole quite highly educated: seven of the fifteen participants had graduate degrees, and all participants except one (SG) earned at least a high school diploma and attended college for a time. One participant (SR) qualified by the metric of education level as having achieved only some undergraduate education, but he was at the time of the study a current undergraduate student with aspirations to earn a PhD in the future. Another participant (MR) had earned an undergraduate degree only, but his father and his closest childhood friends’ parents were University of Rochester professors and his wife earned her PhD from the University; he cites the University as a force which repeatedly “drew [his] family” to Rochester. On the other hand, some participants with graduate degrees have had comparatively little connection with universities. Two participants (MER and JBI), for example, earned graduate degrees in nursing in the 1960s but have not had any interaction with higher education since then. One participant (SB) expressed an orientation away from higher education, saying that he went to college only out of fear that his girlfriend would not stay with him unless he did: “I went to college because . . . I realized if I don’t get a college degree she’s gonna end up dumping me.” In other words, terminal degree, attitude toward the University, and connection with the higher education in general and with the University of Rochester in particular are not necessarily related.

While the University of Rochester is geographically adjacent to—and even moving into—the Ward, there is a strong sense that the University is not part of the neighborhood. Even with formal and informal connections between the two, the University and the neighborhood remain seen in opposition to one another. In the following section we illustrate how this complicated relationship may influence the variable use of both local and supralocal dialect features for individual speakers.

4.2.4. Higher Education and Sound Change

LR, HR, and SR, described above, represent sound change over time in a 19th Ward family, but they also represent the varied and evolving roles of higher education in the community. LR’s parents were blue-collar workers in the garment industry. LR (the grandfather, born 1933) got a job at his uncle’s pharmacy and then went to a state university to become certified as a pharmacist; he spent his career as a pharmacist in Rochester. LR’s wife (HR’s mother) did not, as far as we are aware, go to college. HR (LR’s son, born 1962) went to the same state university as his father to become a civil engineer and met his wife (SR’s mother) there. SR (LR’s grandson, HR’s son, born 1997) was, at the time of interviewing, a student at the University of Rochester, with plans to obtain a doctoral degree. LR was the first generation to be college-educated, HR had one college-educated parent, and SR was raised by two college-educated parents. LR went to college to become certified to work at the family business, HR to become an engineer, and SR as a stepping-stone to further education. The family has therefore become more and more closely connected with higher education over time, and the differences in the vowel systems of the three speakers reflect this connection between higher education and change in apparent time.

As one of the oldest speakers in the sample, we might expect BL (born 1943) to demonstrate more traditional local patterns like LR or even HR. Yet BL displays only a few features of the Inland North dialect. He has an NCS score of only 0.5 on the word list and 1.5 on the oral history interview (although he does have some of the most fronted bat vowels in the sample [Figure 5]). BL is the speaker described above who was not aware of the NCS until he attended college in out of state, and he now espouses a very negative view toward the NCS. At the same time, despite his presumed exposure to other dialects in college, he exhibits no adoption of the supralocal Bat/Ban Split or the Low-Back Merger.⁴

Figure 5.

BL’s Vowel Space: (a) Word List and (b) Oral History Interview

It seems that BL’s exposure in higher education has led to a rejection of local speech patterns, but not necessarily to an adoption of supralocal ones. We see a similar pattern with MR, who has however relatively more contact with higher education as his parents and spouse were all University professors. As noted above, MR demonstrates high awareness of the NCS. He also has a low NCS score of 1.5, and his bat vowel is relatively un-fronted. At the same time, he is not an adopter of either of the supralocal features: he lacks both the Bat/Ban Split and the Low-Back Merger (Figure 6), and his lowest vowel is cot.

Figure 6.

MR’s Vowel Space: (a) Word List and (b) Oral History Interview

Both BL and MR serve as examples of how these two motivations—the avoidance of the local and the adoption of the supralocal—may not always operate in tandem, and we can illustrate this further with the contrast between two of our youngest speakers, SR and LC (Figure 7), who are the two speakers in this sample with the closest ties to higher education, but with different orientations with respect to the University and the Ward.

Figure 7.

LC’s Vowel Space: (a) Word List and (b) Oral History Interview

As a full-time student at the time of interviewing with college-educated parents and aspirations for attaining a doctorate, SR represents one of the speakers with the greatest level of connection to higher education. He is a part of the 19th Ward community, but he spends the vast majority of his time in the University setting. He grew up in the 19th Ward, but his aspirational ties are not to Rochester: he plans to move away from his hometown to pursue a PhD. SR describes having friends from many different places as a defining part of his social experience at the University, and he shows strong adoption of the supralocal features. SR’s speech thus reflects his supralocal aspirations more than his local reality. In this way, he represents a white persona similar to King’s (2021) Mobile Black Professionals, or to the students described by Wagner (2012) who aspired to attend non-local colleges. For SR, higher education represents a means to achieving those aspirations: he needs a research-oriented undergraduate degree in order to reach his next goal of a PhD program outside of Rochester. This orientation is reflected in SR’s relatively split bat and ban and relatively merged cot and caught.

LC is the second-youngest speaker in the sample; she was thirty-four at the time of interviewing. She grew up in the 19th Ward but left Rochester for college, again reflecting the broader diachronic patterns of greater orientation toward higher education in this community. LC was a full-time employee of the University of Rochester, in a role involving frequent interaction with students and faculty and mediation between the University and the Ward.

Like SR, LC’s bat is quite low relative to the other vowels but unlike SR, it shares the lowest part of the vowel space with cot. LC has a very low NCS score (0.5) and the greatest bat/ban distance in the sample is in her word list. On the other hand, unlike SR, LC does not seem to have adopted the Low-Back Merger, although those two vowels are slightly closer together on the word list than in the oral history interview. LC adopts the bat/ban split to a far greater extent than SR. Notably, while SR style-shifts slightly toward local 19th Ward patterns on the word list, LC style-shifts away from them. In other words, with greater attention to speech LC shifts even further away from local patterns.

For both speakers, supralocal speech patterns reflect their aspirational orientations. We suspect that for SR, this is largely subconscious, while LC’s style-shifting represents a more conscious awareness of the social meaning of local and supralocal patterns, reflecting an ideology that places higher education and Ward identity in opposition to each other such that she must mediate these through her speech patterns and the social capital they generate. LC has spent more of her life outside the Ward. She has more experience in higher education, and, as a woman, occupies a slightly less hegemonic place in society than SR. In her oral history interview, LC embraces her 19th Ward identity, but in the formal word list context, she leans on the supralocal pattern to (re)orient toward the University. LC and SR both demonstrate a retreat of the NCS in apparent time, driven in part by changing relationships to higher education, but the extent to which they draw on the social meaning of the NCS is demonstrably different. We can further illustrate this with the speaker JC.⁵

Like SR and LC, JC has ties with both the University and the Ward. As of 2018, he held leadership roles in multiple service organizations in the Ward and often served as a point person for volunteer groups from the University of Rochester and nearby Rochester Institute of Technology (RIT); this how Author Kapner first met him. JC’s involvement with the University is less immersive than LC’s and SR’s and takes place almost entirely in the 19th Ward. His master’s degree has little influence on his connection to higher education. JC expresses mixed attitudes toward the University, as described above, and he positions himself as an advocate for the Ward.

Despite JC’s regular connection with the University, he does not exhibit the Bat/Ban Split or the Low-Back Merger and the bottom vowel in his vowel space (Figure 8) is cot. We suspect that his strong identification with the Ward motivates his adoption of local features, and his lack of immersion in University life restricts his access to and motivation to adopt supralocal features. JC is aware of the NCS and its stigma: he mentions the “Rochester nasal ‘a.’” In spite—or perhaps because—of this, he embraces the local dialect features as part of his persona as a voice of the 19th Ward.

Figure 8.

JC’s Vowel Space: (a) Word List and (b) Oral History Interview

Finally, SB is the speaker described above who went to college only because of concerns that his girlfriend would break up with him if he did not. As of 2018, he was raising his children in the suburbs outside the 19th Ward and generally expresses perhaps the most neutral (as opposed to uniformly positive) view toward the Ward of anyone in the sample. Yet perhaps surprisingly, SB exhibits one of the most quintessentially Rochesterian vowel spaces in this sample (Figure 9): he has the relatively high NCS scores of 3.5 on the oral history interview and 3 on the word list. He is a non-adopter of the Bat/Ban Split and the Low-Back Merger, and the lowest vowel in his system is cot. SB exhibits little sign of style-shifting, indicating a lack of awareness of these features. In other words, for SB, like BL, the supralocal features are not available for him to adopt. However, unlike BL, he also lacks any particular reason to adopt or reject the local features because he is less aware of their stigma.

Figure 9.

SB’s Vowel Space: (a) Word List and (b) Oral History Interview

The varied and at times conflicting patterns in this group relative to these sound changes and the role of higher education reflect a great deal of variation and complexity within the community with respect to the NCS, supralocal speech patterns, and relationships with higher education. To make sense of the heterogeneity in this sample, we need to consider rejection of local and adoption of supralocal patterns separately, considering access to those forms as well as motivation to use or reject them, in light of other social factors about the speakers.

5. Conclusion and Discussion

Within a single family in Rochester’s 19th Ward (section 4.1), we find an illustration of several ongoing changes in progress, in line with our expectations: an apparent time retreat of the NCS and adoption of the Bat/Ban Split and Low-Back Merger.

Yet statistical analyses fail to reveal strong patterns for the sample as a whole in the ways that we expected (section 4.2.2). For the most part we feel that this is likely due to the small and relatively socially homogeneous sample. Yet it is also possible that the great deal of variability we are finding in the use of the erstwhile Inland North dialect in Rochester’s 19th Ward is a reflection of the fact that the NCS has only recently began to regress, and that the social meaning of the NCS as a marker of local-ness has not been fully established in the local community. It may be that we should expect to find this level of inter- and intra-speaker variability in the linguistic means for constructing identities with respect to place, particularly in the context of a newly stigmatized local dialect in a community that has undergone quite recent socio-economic upheaval. In other words, we are not on the surface able to demonstrate the type of orderly heterogeneity (Weinreich et al. 1968; Labov 1972) that we might expect, because of ongoing change in the meaning of the NCS locally. With respect to the Bat/Ban Split and the Low-Back Merger, it seems likely that these incoming supralocal changes have not yet been fully established in the community. Following King (2021), we argue that looking at the Rochester community as a whole is less useful than looking at individual motivations to adopt or reject changes.

Our interviews and participant observation reveal a complex situation of town-gown relations in Rochester (section 4.2.3). The participants in this study are quite proud of being 19th Warders and of their broader Rochester identities. This particular group of speakers, recruited for an oral history project about local culture, likely represents a social network with a particularly high level of local pride. Meanwhile, even as Ward residents find themselves in increasing contact with the University community of practice, and the University expands into the neighborhood, there is a strong sense that the University will never become part of the neighborhood. At least for some in this study, the 19th Ward identity is defined in opposition to the University. We also find that for these speakers, terminal degree, attitude toward the University, and connection with the higher education in general and with the University of Rochester in particular are not necessarily related: the interactions of higher education and identity are far more complex.

We observe that speakers in this community are generally aware of a “Rochester accent” and some of its features (section 4.2.1), particularly raised and fronted bat and cot. This gives us some reason to believe that the NCS is consequently available to speakers for manipulation; speakers may be able to use more or fewer features of the NCS in their speech. And although the NCS is stereotyped by many speakers, it is not necessarily rejected by them. With this group of speakers so consistently proud of their 19th Ward identity, it is possible that these speakers may be calling on the NCS to index their status as Warders, and that increased awareness of the NCS has made this shifting possible. Moreover, as these were interviews conducted in the context of an oral history interview project conducted jointly by members of the University and the 19th Ward communities, this context may have primed speakers to embrace their use of local features to index their status as Warders in opposition to the University. For the speakers like SR and LC who do exhibit fewer local features, we interpret this less as a rejection of the local and more as an adoption of the supralocal, driven by an aspirational orientation toward higher education. Within the one family, the apparent time changes we observe in apparent time also reflect changing relationships to higher education, as each generation orients more toward the supralocal aspirations the University represents. Looking at individual speakers rather than the sample as a whole reveals that higher education impacts sound change in a nuanced way that is different for each speaker (section 4.2.4).

In his review of literature on place, identity, and language, Reed (2020) points out that it is not uncommon for speakers to use stigmatized features of local language varieties as a way to index pride in being from that place. Summarizing Lane’s (1998) study of speakers of a variety of Danish local to the village of Thyborøn, Reed (2020:7) says, “a speaker’s identity with Thyborøn seemed to be more important than the stigma from using marked features.” The adoption of the NCS in the present study seems to follow this same pattern.

Race may have a role to play here as well. D’Onofrio and Benheim (2020) and Van Herk (2008) argue that the NCS represents a kind of “symbolic ‘White flight’” (D’Onofrio & Benheim 2020:485), and that its widespread reversal in apparent time reflects a rejection of racist ideologies. While the participants in this particular study all condemn racism and speak with pride about the diverse racial makeup of the Ward, the fact remains—reflected in the racial makeup of this sample—that this social network is primarily white, as were the interviewers. It is possible that for the white speakers in this sample, the NCS may, as described by King (2021:174), “indirectly index Whiteness” even in the context of stated opposition to racism. DW, the only Black speaker in this sample, seems to fit with King’s (2021) persona of the Mobile Black Professional, for whom the reversal of the NCS indexes a career “not defined by manual labor,” and as well as competence or professionalism (King 2021:174). Her lack of NCS reflects the patterns King describes for this persona.

Overall, our findings contribute to the understanding of the changing social meaning of the NCS in the Inland North, and the roles of place and identity. Drawing on a combination of participant observation and phonetic analysis, particularly in one family of speakers, we find that we can draw some conclusions about the regression of the NCS, and the progression of the Bat/Ban Split and the Low-Back Merger in Rochester’s 19th Ward. The complicated relationship between this social network and the university next door is reflected in these 19th Warders’ varied adoption of local and supralocal linguistic features. For some, growing connection to the university means greater availability and adoption of supralocal features, while for others, local pride and the university’s perceived oppositionality to this provide impetus to embrace local speech patterns. The impact of universities on local speech patterns remains, therefore, a promising area for further study.

Footnotes

Acknowledgements

The authors wish to acknowledge the detailed feedback of four anonymous reviewers, and to thank the fifteen Rochester speakers for their enthusiastic participation. We also received valuable assistance from Sandra Talarico, Jackie Farrell, MaryDan Cooper, Christine Ridarsky, Michelle Finn, Justin Davidson, Carolyn Caza, and many others at the 19th Ward Community Association, the University of Rochester, and UC Berkeley.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was partially funded by a University of Rochester Barth-Crapsey Undergraduate Research Award, awarded to the first author.

ORCID iDs

Julianne Kapner

Maya Ravindranath Abtahian

Notes

Author Biographies

Julianne Kapner is a Ph.D. candidate in the Department of Linguistics at the University of California, Berkeley. Her research focuses on sociolinguistics, phonetics, and language revitalization, particularly the dynamics of language variation and identity in urban settings.

Maya Ravindranath Abtahian is a sociolinguist in the Department of Linguistics at the University of Rochester. Her research focuses on language variation and change in the context of language shift and other social changes.

References

19th Ward Community Association. 2023. https://19wca.org (February 18, 2024).

Bates

Douglas

Mächler

Martin

Bolker

Ben

Walker

Steve

. 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67(1). 1-48.

Bender

Thomas

(ed.). 1991. The university and the city. Oxford: Oxford University Press.

Bigham

Douglas S.

2010. Mechanisms of accommodation among emerging adults in a university setting. Journal of English Linguistics 38(3). 193-210.

Boersma

Paul

Weenink

David

. 2023. Praat: Doing phonetics by computer [computer program]. Version 6.4.01, retrieved November 2023 from http://www.praat.org.

Clarke

Sandra

Elms

Ford

Youssef

Amani

. 1995. The third dialect of English: Some Canadian Evidence. Language Variation and Change 7. 209-228.

Clopper

Cynthia G.

Pisoni

David

de Jong

Kenneth

. 2005. Acoustic characteristics of the vowel systems of six regional varieties of American English. The Journal of the Acoustical Society of America 118(3). 1661-1676.

De Decker

Paul

. 2006. A real-time investigation of social and phonetic changes in post-adolescence. University of Pennsylvania Working Papers in Linguistics 12(2). Article 7.

Dinkin

Aaron J.

2009. Dialect boundaries and phonological change in Upstate New York. Philadelphia, PA: University of Pennsylvania PhD dissertation.

10.

Dinkin

Aaron J.

2011. Nasal short-a systems vs. the Northern Cities Shift. University of Pennsylvania Working Papers in Linguistics 17(2). Article 9.

11.

Dinkin

Aaron J.

2013. Settlement patterns and the eastern boundary of the Northern Cities Shift. Journal of Linguistic Geography 1(1). 4-30.

12.

Dinkin

Aaron J.

2021. Adjusting to the new normal(ization): Adapting Atlas of North American English benchmarks to Lobanov-normalized data. Paper presented at New Ways of Analyzing Variation 49, Austin, TX.

13.

Dinkin

Aaron J.

2022. Generational phases: Toward the low-back merger in Cooperstown, New York. Journal of English Linguistics 50(3). 219-246.

14.

D’Onofrio

Annette

Benheim

Jaime

. 2020. Contextualizing reversal: Local dynamics of the Northern Cities Shift in a Chicago community. Journal of Sociolinguistics 24(4). 469-491.

15.

Driscoll

Anna

. 2016. Cold winters, flat A’s: Linguistics, geography, and the Northern Cities Shift in Syracuse. Hanover, NH: Dartmouth College undergraduate honors thesis.

16.

Driscoll

Anna

Lape

Emma

. 2015. Reversal of the Northern Cities Shift in Syracuse, New York. University of Pennsylvania Working Papers in Linguistics 21(2). 41-47.

17.

Duncan

Daniel A.

2018. Language variation and change in the geographies of suburbs. New York, NY: New York University PhD dissertation.

18.

Durian

David

Cameron

Richard

. 2020. A new perspective on the development of the Northern Cities Shift in Chicago. In Ömer Eren, Asimina Giannoula, Sam Gray, Chi-Day Lam (Daniel), and Aurora Martinez Del Rio (eds.), Proceedings of the fifty-fifth annual meeting of the Chicago Linguistic Society, 115-128. Chicago, IL: Chicago Linguistic Society.

19.

Eckert

Penelope

. 1988. Adolescent social structure and the spread of linguistic change. Language in Society 17(2). 183-207.

20.

Eckert

Penelope

. 2008. Where do ethnolects stop? International Journal of Bilingualism 12(1-2). 25-42.

21.

Evanini

Keelan

. 2009. The permeability of dialect boundaries: A case study of the region surrounding Erie, Pennsylvania. Philadelphia, PA: University of Pennsylvania PhD dissertation.

22.

Evans

Betsy E.

Ito

Rika

Jones

Jamila

Preston

Dennis R.

2000. Change on top of change: Social and regional accommodation to the Northern Cities Chain Shift. Taal en Tongval 52(1). 61-86.

23.

Fridland

Valerie

. 2015. The spread of the cot/caught merger in the speech of Memphians. In Picone

Michael D.

Davies

Catherine Evans

(eds.), New perspectives on language variety in the south: Historical and contemporary approaches, 551-564. Tuscaloosa, AL: University of Alabama Press.

24.

Hazen

Kirk

. 2005. Mergers in the mountains: West Virginia division and unification. English World-Wide 26(2). 199-221.

25.

Irons

Terry Lynn

. 2007. On the status of low back vowels in Kentucky English: More evidence of merger. Language Variation and Change 19(2). 137-180.

26.

Kendall

Tyler

Thomas

Erik R.

2010. Vowels: Vowel manipulation, normalization, and plotting in R (R package). cran.r-project.org/web/packages/vowels/index.html (July 18, 2025).

27.

King

Sharese

. 2018. Exploring social and linguistic diversity across African Americans from Rochester, New York. Palo Alto, CA: Stanford University PhD dissertation.

28.

King

Sharese

. 2021. Rethinking race and place: The role of persona in sound change reversal. Journal of Sociolinguistics 25(2). 159-178.

29.

Labov

William

. 1972. Sociolinguistic patterns. Philadelphia, PA: University of Pennsylvania Press.

30.

Labov

William

. 2007. Transmission and diffusion. Language 83(2). 344-387.

31.

Labov

William

. 2010. Principles of linguistic change. 2: Social factors. Language in Society 29. Chichester: Wiley-Blackwell.

32.

Labov

William

Ash

Sharon

Boberg

Charles

. 2006. The Atlas of North American English: Phonetics, phonology and sound change a multimedia reference tool. Berlin: Mouton de Gruyter.

33.

Labov

William

Yaeger

Malcah

Steiner

Richard

. 1972. A quantitative study of sound change in progress. Philadelphia, PA: U.S. Regional Survey.

34.

Lane

Lisa Ann

. 1998. Emergence and transformation of a dialect: ThyborØnsk (Danish). Chicago, IL: University of Chicago PhD dissertation.

35.

Lobanov

Boris

. 1971. Classification of Russian vowels spoken by different speakers. The Journal of the Acoustical Society of America 49(28). 606-608.

36.

Massey

Jennifer

Field

Sean

Chan

Yolande

. 2014. Partnering for economic development: How town-gown relations impact local economic development in small and medium cities. Canadian Journal of Higher Education 44(2). 152-169.

37.

Mayfield

Loomis

. 2001. Town and gown in America: Some historical and institutional issues of the engaged university. Education for Health 14(2). 231-240.

38.

McAuliffe

Michael

Socolof

Michaela

Mihuc

Sarah

Wagner

Michael

Sonderegger

Morgan

. 2017. Montreal Forced Aligner: Trainable text-speech alignment using Kaldi. Proceedings of the 18th conference of the International Speech Communication Association, Stockholm, Sweden.

39.

McCarthy

Corinne

. 2011. The Northern Cities Shift in Chicago. Journal of English Linguistics 39(2). 166-187.

40.

Milholland

Agatha.

2018. Reversal of the Northern Cities Shift in Buffalo, New York. Paper presented at New Ways of Analyzing Variation 47, New York, NY.

41.

Nesbitt

Monica

. 2018. Economic change and the decline of raised trap in Lansing, MI. University of Pennsylvania Working Papers in Linguistics 74(2). Article 9.

42.

Nesbitt

Monica

. 2021. The rise and fall of the Northern Cities Shift: Social and linguistic reorganization of TRAP in twentieth-century Lansing, Michigan. American Speech 96(3). 332-370.

43.

Nesbitt

Monica

Wagner

Suzanne Evans

Mason

Alexander

. 2019. A tale of two shifts: Movement toward the Low-Back-Merger Shift in Lansing, Michigan. Publication of the American Dialect Society 104(1). 144-165.

44.

Niedzielski

Nancy

. 1999. The effect of social information on the perception of sociolinguistic variables. Journal of Language and Social Psychology 18(1). 62-85.

45.

Pickford

Giles

. 1989. Town and gown relations. Journal of Tertiary Educational Administration 2. 201-210.

46.

Podesva

Robert J.

D’Onofrio

Annette

Van Hofwegen

Janneke

Kim

Seung Kyung

. 2015. Country ideology and the California Vowel Shift. Language Variation and Change 27(2). 157-186.

47.

Prichard

Hilary

. 2016. The role of higher education in sound change. Philadelphia, PA: University of Pennsylvania PhD dissertation.

48.

R Core Team. 2023. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. https://www.R-project.org (December 26, 2025).

49.

Reddy

Sravana

Stanford

James N.

2015. A web application for automated dialect analysis. Proceedings of NAACL-HLT 2015, Denver, CO.

50.

Reed

Paul E.

2020. Place and language: Links between speech, region, and connection to place. Wiley Interdisciplinary Reviews: Cognitive Science 11(3). Article e1524.

51.

Roeder

Rebecca V.

2010. Northern cities Mexican American English: Vowel production and perception. American Speech 85(2). 163-184.

52.

Rosenfelder

Ingrid

Fruehwald

Josef

Evanini

Keelan

Seyfarth

Scott

Gorman

Kyle

Prichard

Hilary

Yuan

Jiahong

. 2014. FAVE (Forced Alignment and Vowel Extraction) Program Suite v1.2.2. doi:10.5281/zenodo.22281.

53.

Savage

Matthew

Mason

Alexander

Nesbitt

Monica

Pevan

Eric

Wagner

Suzanne Evans

. 2015. Ignorant and annoying: Inland Northerners’ attitudes towards the NCS short-o. Poster presented at New Ways of Analyzing Variation 44, Toronto, ON.

54.

Sayahi

Lotfi

. 2011. Code-switching and language change in Tunisia. International Journal of the Sociology of Language 211. 113-133.

55.

Thiel

Anja

Dinkin

Aaron J.

2020. Escaping the trap: Losing the Northern Cities Shift in real time. Language Variation and Change 32(3). 373-398.

56.

Thomas

Erik R.

Kendall

Tyler

. 2007. NORM: The vowel normalization and plotting suite. https://lingtools.uoregon.edu/norm/norm1.php

57.

U.S. Census Bureau. 2018. Demographic and Housing Estimates. American Community Survey, ACS 1-Year Estimates Data Profiles, Table DP05. https://data.census.gov/table/ACSDP1&2018.DP05 (February 24, 2020).

58.

Van Herk

Gerard

. 2008. Fear of a black phonology: The Northern Cities Shift as linguistic white flight. University of Pennsylvania Working Papers in Linguistics 14(2). Article 19.

59.

Wagner

Suzanne Evans

. 2008. Linguistic change and stabilization in the transition from adolescence to adulthood. Philadelphia, PA: University of Pennsylvania PhD dissertation.

60.

Wagner

Suzanne Evans

. 2012. Real-time evidence for age grad(ing) in late adolescence. Language Variation and Change 24(2). 179-202.

61.

Wagner

Suzanne Evans

Mason

Alexander

Nesbitt

Monica

Pevan

Eric

Savage

Matt

. 2016. Reversal and re-organization of the Northern Cities Shift in Michigan. University of Pennsylvania Working Papers in Linguistics 22(2). Article 19.

62.

Weinreich

Uriel

Labov

William

Herzog

Marvin

. 1968. Empirical Foundations for a Theory of Langague Change. Austin: University of Texas Press.