Diagrams & maps by Wikimedia user Mahagaja
The Northern Cities Vowel Shift
The Northern Cities Vowel Shift is a chain shift that defines the Northern Inland dialect region of Northern United States English. It is a sound change resembling a chain shift which effects the vowels in TRAP, LOT/PALM, THOUGHT, STRUT, DRESS, and KIT.
Stages of the NCVS
Stage 1 sees the vowel in TRAP /æ/ tensed into a diphthong, in the range of [ɛ͡ə ~ e͡ə ~ ɪ͡ə]
Stage 2 sees the vowel in LOT and PALM /ɑ/ front towards the space left by TRAP, into the range of [ᴀ ~ a]. The vowel in START will usually go along with this change.
Stage 3 sees the vowel in THOUGHT /ɔ/ lower to fill the space left by LOT, becoming [ɑ]
Stage 4 sees the vowel in DRESS /ɛ/ back and lower towards the range of [ɐ] unless followed by /ɹ/
Stage 5 sees the vowel in STRUT /ɜ/ back and sometimes round into [ʌ ~ ɔ], filling the space left by THOUGHT
Stage 6 sees the vowel in KIT /ɪ/ back and/or lower into the range of [e ~ ə̝]
Examples of Speakers with the NCVS
Jeff Kujawa, from Buffalo, has a complete NCVS. His clip starts at 5:17 in the linked video.
Steve Adamko, from Kalamazoo, has an inconsistent NCVS.
Garrett Ryan, from Chicago, has strong Stages 1 and 2, but his THOUGHT vowel is almost unusually high and diphthongal, thus preventing any further stages.
Vince Incandela, from Kenosha, is the absolute paragon of the Northern Cities Vowel Shift! Listen closely to how backed his STRUT vowel is! His clip starts at 1:14 in the linked video.
Geography & History of the NCVS
In this diagram, the brown zone represents areas where the LOT vowel is fronter than the STRUT vowel; the zones enclosed in blue represent areas where the DRESS vowel is retracted, as in Stage 4 of the NCVS; the zones enclosed in red represent areas where the TRAP vowel is universally tensed, as in Stage 1 of the NCVS.
The NCVS originated in the middle of the 20th Century, in cities around the Great Lakes like Chicago, Detroit, and Syracuse, and has since been slowly developing and spreading out into rural areas. The areas effected by the NCVS are largely restricted to the northern half of the Rust Belt, with the notable exception of the St. Louis Corridor, which follows the Interstate 55 from Chicago to St. Louis. In this Corridor, old people are effected by the NCVS, presumably because of migrations southward from Chicago in the 20th century, but I'm not totally sure about that.
The Origins of the NCVS
Many linguists believe that the Northern Cities Vowel Shift originated from some kind of linguistic anomaly around the Erie Canal in the 1860’s. The construction of the Erie Canal brought a lot of people into the inland northern region of the United States, who spoke many different dialects of English. The theory goes that, somehow, the different features of these accents clashed and merged together to create some kind of phonological situation which facilitated the Northern Cities Vowel Shift. The details beyond that are contested.
The theory that I personally believe is Bill Labov's theory: Labov posits that in the 1860’s, the TRAP vowel was treated very differently by people in different parts of the northeastern United States; as, indeed, it is today. Today, New York has something called a Short-A Split System, where TRAP is split into two different phonemes, one tense diphthong [ɛ͡ə] in bad and bath, and one lax monophthong [æ], like in bat. Eastern New England also had a split in its TRAP vowel, but this one wasn’t phonemic: the TRAP vowel was simply tensed into a diphthong before nasal consonants, like in pan and animal. Western New England, on the other hand, has what’s called a continuous short-a system. The TRAP vowel does allophonically raise in certain contexts, but in a more subtle way; there isn’t any overt tense-lax split like in the other two dialects. Labov hypothesises that people from the northeast moved over to the Erie Canal, and their dialects began mixing together and homogenising. However, homogenisation was complicated by these dialects’ contradictory treatments of the TRAP vowel. In the mess of tense and lax pronunciations that followed, phonetic compromise was reached in the end simply by universally tensing the TRAP vowel into a diphthong. Thus, the TRAP vowel was raised in the vowel space, and Stage 1 of the NCVS was born.
There are, however, problems with this theory. A 2019 study by David Durian and Richard Cameron showed that the first people in Chicago to express the beginnings of the NCVS were women born in the 1870’s and 1880’s, who showed very slight fronting of TRAP and LOT/PALM. In this study, they conclude that the fronting of LOT/PALM was the earliest stage of the NCVS within Chicago. An earlier 2016 study by Matthew Gordon and Christopher Strelluf found that people born around the turn of the 20th century in Buffalo and Grand Rapids showed almost no signs of the modern NCVS, with the notable exception of a usually central, or even front-of-central, LOT/PALM vowel. This would seem to imply that the earliest manifestation of what would become the NCVS was in fact the fronting of LOT/PALM, and the tensing of TRAP happened independently later on.
The Rise & Fall of the NCVS
In the early years, the Northern Cities Vowel Shift seemed to hold an amount of prestige in the minds of those living around the Great Lakes. Between 1998 and 2001, Rika Ito and Dennis Preston found that in small-town Michigan, a universally tense TRAP vowel was most commonly exhibited by those “with positive attitudes toward city life and aspirations of moving out of their rural town”. In 1977, Robin Herndobler attributed Chicago women’s tendency to tense their TRAP vowels “to their internalization of this variable as ‘citified and sophisticated.’” In 1989, Penny Eckert found that, in suburban Detroit, white teenagers were adopting the tense TRAP vowel “due to its association with tough, street smart, autonomous white kids of inner-city Detroit.”
However, linguists noticed in 2016 (a little late to the party, in my opinion) that people around upstate New York had begun find the tensed TRAP of the NCVS to be annoying. Lately, more and more young people who we would expect to have the NCVS have instead had much more mainstream vowel systems, and have lost that regional speech indicative of their homeland. Most people would probably blame this decline on social media and increased interconnectedness, but it's actually much more complex than that.
A 2020 study by Anja Thiel and Aaron Dinkin saw the pair go over to Ogdensburg, New York, where the NCVS has been seen to be retreating, and test peoples' reactions to the tense TRAP vowel. They discovered that, at least in that area, the feature had come to be associated with working-class people. Recently, Monica Nesbitt did a study on Michiganders and found that they held the exact same implicit bias.
Say what you will about classism, but in the vast majority of cases, if a speech feature holds working-class connotations, people will shy away from using it. This is usually subconscious: people usually don't think of what they're doing as eliminating the working-class features of their speech, but instead they think of it as talking in a more upper-class way. This often includes working-class people themselves! Think about it: if you’re a lower-class person and you choose to speak in a more middle-class or upper-class manner, that gives you a certain edge when it comes to upward social mobility, as members of those classes will more readily accept you as one of their own if you talk like them. Meanwhile, members of the middle and upper classes will want to lose a feature that is marked as working-class. And so, when a feature becomes socially marked like the tense TRAP vowel has, it usually results in a rapid decline.
In 2021, Monica Nesbitt had a look at the social distribution of Stage 1 of the Northern Cities Shift in Lansing, Michigan. She found that the Silent Generation already had a fronted and diphthongal TRAP vowel, and that the Baby Boomers saw a raising of the vowel. However, comparing Generation X to Baby Boomers, she found that their TRAP vowels were comparatively lower, more backed, and less diphthongal. Millennials had even less tense TRAP vowels.
Nesbitt used this observation, in connection with her aforementioned observation of the tense vowel’s new marking as a working-class feature in Michigan, to conclude that the decline of the NCVS in Lansing coincided with the decline of the Rust Belt.
Nearing the end of the 20th century, the industrial area around the Great Lakes previously known as the Steel Belt was beginning to see massive economic change and decline. Eventually, it became the Rust Belt we know it as today, and the whole region seemed to switch from a manufacturing economy to a service-based economy. Generation X was born into this new and changing Rust Belt, and came to associate the NCVS they heard around them with the economic decline they saw around them. They came to associate it with their unfortunate, industrial, working-class seniors: and so they began to abandon it. This isn’t to mention the fact that their new service-based economy, where they would be working face-to-face with customers much more than their forebearers, sorta demanded they abandon their local accents in favour of more supraregional ones.
Home
