What makes a dialect a ‘dialect’? The roots of Upper Peninsula English

This post is based on the TEDxNMU Talk that I gave on March 24, 2018 at Northern Michigan University.

The history of English in the Upper Peninsula is relatively recent compared to other regional varieties found throughout the United States. It wasn’t until the mid-nineteenth century that English speakers regularly visited and then inhabited the region. The dialect’s recent history combined with its isolation, the people who have immigrated to the area, where they have settled, and the languages that they speak are all factors that maintain the dialect and its distinct features.

Four major factors have helped UP English to develop into the variety it is: geography—particularly the isolated location of the Upper Peninsula, historical events such as immigration and settlement patterns, economics—including mining and tourism, and language attitudes, what we think is “good” and “bad” English.

say ya to da up, eh! copyYet it wasn’t until the mid-1970s that the term Yooper was popularly used in the UP. In the timeline below we see attitudes at work and their effects on what it means to be a Yooper, some of which are tied to what it means to sound like a Yooper. For examples: 1975 “Yooper”; 1982 Michael Loukinen’s film and connections with perceptions about what it means to be a Yooper; and 1983 Jack Bower’s bumper sticker and 1986 Da Yoopers. These examples demonstrate the connections among dialect, people, and place. This connection most evident in the name Yooper, which is used today for both the residents and the dialect.

  • 1975  “Yooper” in an Editor’s Note, Pick & Axe, Bessemer, 1 March (p. 3).
  • 1979 Escanaba Daily Press holds contest to name the people of the Upper Michigan. “Yooper” is declared the winner.
  • 1982 Sociologist Michael Loukinen attempts to make a documentary about the cultural phenomenon of “Yoopers” but receives widespread criticism from older members of the community and instead releases “Good Man in the Woods”.
  • 1983 “Say Ya To Da U.P. eh?” Bumper sticker printed by Jack Bowers of Marquette in response to the “Say Yes to Michigan” tourism campaign.
  • 1986 Da Yoopers band is formed bringing wider exposure to the term. Originally called the Yoopers.
  • 2014 Yooper is added to the Merriam Webster Collegiate Dictionary as the result of a Scrabble game where Yooper was contested as a ‘word’ because it wasn’t in “the dictionary“. The entry is between yore and yoo hoo.

It’s important to realize that although Yooper dates in print to 1975, it existed well before that in  spoken language. Words must exist in spoken language, everyday conversation, before they appear in the popular press.

So, how did the dialect emerge?
The historical factors of immigration and settlement have had the most significant effect on the dialect. The people who have immigrated to the UP, where they have settled, how they have mixed and mingled, and the languages and dialect of English spoken, all contribute to shaping Upper Peninsula English. Evidence of the resulting language contact is easily found in place names: Presque Isle from Canadian French; Maas Street and Makinen’s Road, located outside of Negaunee are from Finnish. From Anishinaabemowin there are many place names, including Munising, from Gitchi Minissing ‘at the big island’, in reference to Grand Island; Negaunee, loosely translated as ‘pioneer’ because the first iron mine and furnace in the area; and Keweenaw, meaning ‘portage’ or ’passage way’. English place names in the Upper Peninsula are often direct translations from Anishinaabemowin. For example, Portage Lake and Laughing Whitefish River.

old and new kowsit lats

Kowsit Lats, a street in Houghton County, reflects contact between Finnish and English. Kowsit Lats is a result of the sociolinguistic history of the area, but the symbolism is typically lost on local residents as well as outsiders. The two signs, one hand-made and the other an official township street sign, are located on Quincy Hill, near the Quincy mine hoist, which today is a popular tourist attraction. The Quincy mine was in full operation from 1846 to 1945, and many of the miners lived in company owned housing that skirted the mine. Many families had their own cow and some houses included a cowshed or a single-stall barn. Each cow wore a numbered tag, the same number as the miner’s work identification number. Miners and their family members would take their cows to graze at the communal pasture provided by the mining company. This pasture was jokingly called “Kowsit Lats” (‘cowshit flats’) and was located on the right side of the road. In the 1980s, Wilber Salmi, who lived in the area, made a street name sign (in the foreground of the photo) that mirrors official street signs with its green background and white lettering. He placed the sign at the corner of Kowsit Lats Road and US 41 and approached the Quincy Township Board to request that the road be officially named “Kowsit Lats.” (At the time, streets did not have official names and instead were known by geographical features, homesteads, mining locations, or town names.) It was not until 2001 when the emergency telephone number 9-1-1 was established that the township officially named the road and placed a street sign on the east side of Kowsit Lats Road, while keeping the original handmade sign on the west side of the road. In addition to the history behind the name, understanding the meaning behind the street name relies on recognition of local linguistic features that have transferred from Finnish to English: /s/ for (sh), /l/ for /fl/, as well as knowing that the Finnish alphabet does not include the letter C and instead uses K for /k/ sounds. The pun is a result of language contact between English and Finnish: Finnish doesn’t have the (sh) sound, so people whose first language is Finnish, often substitute /s/ for the (sh) sound when speaking English. Similarly, the consonant cluster /fl/ is not a part of Finnish phonology, so the /f/ might be omitted as a result of language contact and transfer.

Kowsit lats is also an example of language change and local dialect features (especially in the northwestern UP) that have changed since the 1980s. Most speakers today do not substitute /s/ for (sh). This change can give the impression that the dialect is dying. But that’s not the case. However, it is changing. Regional dialects, such as local ways of speaking English in the UP, are not dying out, and in fact some features might become stronger over time. But dialects, like language in general, do change—they must change as society changes and as contact with heritage languages fade. These changes are what people notice when they think a dialect is dying.

So what makes a dialect a ‘dialect’?
Here’s an analogy: language is like pie, any old pie. Dialects, or varieties of that language, are different kinds of pie, even tarts, variations on a pie theme. Dialects have similar sounds, words, grammatical structures that make them recognizable as pies, but they differ enough, vary, to make them varieties of the language, or different kinds of pies. They also have enough similar linguistic features that they can be grouped together as part of a language, or as a pie. Even what we perceive as “standard English” is a kind of pie or variety of the English language.

In linguistics, “accent” refers to only to pronunciation, for example, people might speak English with a Japanese accent, a German accent, or Hindi accent. Whereas “dialect” and “variety” refer to the identifiable forms of language that vary on all levels, from pronunciation to grammar, regional and social varieties, such as Wisconsin English or Hispanic English.

sauna insurance signThese identifiable and named dialects vary on all levels of language: phonology or sounds, lexicon or vocabulary, word structure, and grammatical structures.For example, in the UP, many people pronounce sauna as “sow-na” rather than the pronunciation “saw-na,” which is more common in American English. These pronunciations are often a result of language contact: as settlers mixed and mingled and learned English, their languages came into contact, not only with each other, but with English. These sounds are evidence of the area’s history as well as how people learned English: typically from other people who spoke English as a second language, so the English that emerged was “accented” English.

Likewise, everyday language use provides evidence of language contact and a window into the area’s sociolinguistic past, and reflects the various languages that have come into contact in the area. Below are some examples of lexical features, or vocabulary, and word structure, or morphology, for example, yous for the plural ‘you’. No doubt readers from the UP are familiar with some of these examples and can add to this list. What is significant in terms of the idea of a dialect is that while some of these linguistic features are recognizable as local ways of speaking, others like bakery are not, and others like catarrh have disappeared from use.

  • Chook or chuke from touque: A knit winter cap (Canadian French, toque)
  • Catarrh, or “Lake Superior cattarh”: A cold, congestion (Canadian French)
  • Make wood: To cut firewood (Canadian French)
  • Bush: Woods, forest (Cornish English)
  • Yous: plural “you” (Irish English)
  • Place Names: Keweenaw, Ahmeek, Assinins (Anishimabemowin)
  • Choppers: Long-sleeved mittens, usually made with deer skin, removable finger flaps (from Anishinabemowin minjikaawan)
  • Ja, yah: yes (German, Swedish, Danish, compare with Finnish, jo)
  • Bakery: baked goods (German)
  • Camp: cabin, cottage (German)
  • Sauna (Finnish)
  • Sisu: perseverance in the face of adversity (Finnish)
  • Foods, from Finnish: nisu (pulla; sweet cardamom bread); korpu (cinnamon and sugar toast), pannukakku (oven-baked pancake), sauna makkara (ring bologna), juustoa (a kind of cheese) (Finnish), and from Slovenian, povatica (a nut and cinnamon rolled pastry)

bakeryHere we see bakery, meaning ‘baked goods,’ not the place where the baked goods are made, at a food booth at at the Covington Music Festival. In addition to pronunciation, vocabulary, and word structure, dialects also vary semantically, or by word meanings. Bakery is a particularly good example of semantic variation: meaning both ‘baked goods’ and ‘the building where baked goods are made’. Bakery is also an example of language contact between English and German in the UP, as well as in areas of Wisconsin, especially Milwaukee, where Germans settled in the late 1800s and early 1900s.

fullsizeoutput_154Pank is another particularly interesting example because many UP residents from mining areas recognize the word, use it, and understand that it is part of the local dialect. And even more significant is that many people in the UP who use pank think that it is unique to the UP. However, we can see in the entry from the Dictionary of American Regional English that pank is used chiefly in Michigan, Pennsylvania, and Upstate New York. What these three distinct regions have in common is a history of mining, and dialectologists think that the word must have come from miners and mining terminology, especially given the similar word, banka, in Swedish, Danish, and Norwegian. These languages were spoken by some of the same folks who immigrated to the UP and who worked in the mines. And, often the consonant /b/ in these languages is pronounced more like a /p/. But, there’s no solid evidence to support to support its origins; that’s why the beginning of the entry says “perhaps a blend…”

Language contact also affects grammatical structures, and often these are not recognized as dialect features or simply a variation of American English, and instead, people mistakenly label them “bad English.” One particularly recognizable example of UP English is what’s called the “illative phrase”, or a phrase that shows movement to or toward a place, for example, “Let’s go to the mall.” In English, this is usually signaled with the preposition “to”. Because of prolonged language contact with Finnish, which doesn’t use a preposition in this phrase and doesn’t include articles (a, an, the), you might here sentences like “Let’s go mall” or “I went post office.” This grammatical structure is a result of language transfer: grammatical rules of Finnish have been transferred to English. Another example of language transfer results from contact with three languages and the tag questions eh and hey, as in “Have a nice day, eh.” “That’s a pretty dress, hey.” This feature came from Cornish English (eh), Anishinaabemowin (en), and/or Canadian French (hien). In the early to mid-1800s, Canadian French and Anishinaabemowin speakers were in close contact and Cornish English speakers were often held up as models for new arrivals not only because they spoke English but also because they had some of the more desirable jobs in the mines. Because of the value placed on Cornish miners, the significant interaction between Canadian French and Anishinaabe, and the similar grammatical structures among the three languages, it might well be that all three  contributed to its current use.

As with any dialect, the sounds, words, phrases, and sentence structures combine in unique ways to create UP English. By knowing the history of UP English, we come to understand how people, place, and language are tightly bound together. We also become aware that, although the rules that make up regional dialects may vary from mainstream American English, they are different rather than “wrong,” “incorrect,” or “bad English.”

yooperpridemagnetHistory and social interaction resulting in language contact are the key ingredients of what makes a dialect a dialect. But our attitudes toward dialects and perceptions of dialect speakers are also significant factors. While there are negative stereotypes associated with what it means to be a Yooper and and the dialect, at the same time there is great pride in being a Yooper and sounding like a Yooper. For example, in the UP, there comes a fierce pride about being local, as well as a do-it-yourself, get by with what you have, independence that’s rooted a history of hard labor in an isolated place that has a climate that can be difficult to thrive in. This magnet reflects the connections between identity, place, and language with “proud, tough, independent,” and by mapping “Yooper” onto the UP. Language is our badge of identity—how else to we show where we’re from? Who are community and family are? Our genders, ages, even the values we hold dear?

So, what makes a dialect a ‘dialect’?
We do: as we shape history, as we share our attitudes about dialects in social media and conversations with friends, coworkers, and family, and as we interact with others and our dialects and languages come into contact. My hope is that by understanding how and why dialects develop that we can realize that variation isn’t “bad” or “good,” it just is. To paraphrase the blogger, Mr Verb: Language varies. Deal with it. Revel in it.







Yoopers, Trolls, and Detroiters: Perceptions of Michiganders’ Talk

Where in Michigan do you think people speak best? Why? Where in Michigan do you think people speak worst? Why? These are questions that Angela Tramontelli are investigating. We’re curious as to where Michganders think the best and worse English is spoken in our state and who they think speaks the best and worse English. 

Perceptual dialectology studies the beliefs that non-linguists have about language variation, the ways that language differs from place to place and/or group to group. While sociolinguists (those who study language variation and change) typically categorize regional dialects according to bundles of linguistic features (pronunciations, vocabulary, grammatical structures), perceptual dialect studies have found that non-linguists determine dialect boundaries not only by linguistic features (e.g., Mase 1964; Lance 1999; Benson 2003; Evans 2012), but also along political and civil boundaries (e.g., Sibata 1959; Preston 1986; Inoue 1996; Lance 1999), and according to cultural differences (Preston 2002). Most importantly, previous research (e.g. Lambert, Hodgson, Gardner & Fillenbaum 1960; Giles 1970; Giles & Ryan 1982; Niedzielski & Preston 2003), has demonstrated that speaker attitudes of dialects directly correspond to attitudes about groups of speakers. These beliefs play a critical role in understanding the effects of linguistic and social prejudices on language change: speakers of stigmatized dialects are often pressured to change the way they talk. Therefore, these perceptions and attitudes are important to an understanding of the factors that affect language change, the relationship between language variation and groups of speakers, as well as the relationship between linguistic prejudice and social prejudice. 
Our study contributes to a growing body of research in this area. We are conducting surveys with Michigan residents to investigate these beliefs and are analyzing the data for patterns in how respondents divide the state into dialect regions and the social characteristics they attach to the speakers who live there. Perceptions such as these within one state can reveal local social categories, including urban/rural distinction (Evans 2012) or the belief in the absence of dialect in a particular state or region (Preston 2002).
Our study further contributes to research on Michigan dialects (e.g. Remlinger 2006; 2007a, b; 2009; Remlinger, Salmons, & von Schniedemesser 2009; Simon 2005) and would add to existing research that examines language attitudes within an individual state (Benson 2003; Bucholtz et al. 2007; Evans 2012), rather than the majority of perceptual dialect studies which focus on regional boundaries within an entire country. A focus on Michigan is particularly significant because unlike most states, it has two main regional dialects so distinct that residents separate the two areas linguistically and socially, as reflected in the labels Yooper and Troll and web pages such as “Michigan Accent: Pronunciations Unique to Us” (www.michigannative.com) and “Yooper Dialect” (en.wikipedia.org/wiki/Yooper_dialect).
Our study also builds on research that demonstrates that the attitudes people hold towards regional dialects are often representative of attitudes about the speakers of those dialects (Preston 1989, 2003; Benson 2003; Bucholtz et al. 2007). The project relies on a language ideology framework (e.g. Lippi-Green 1997; Irvine & Gal 2000; Silverstein 2003), which explains how these attitudes and resulting prejudices reside within the association of particular linguistic features with certain social characteristics. Our goals include creating awareness of this association, examining the relationship between perceptions of and attitudes towards language use and local social categories, furthering the understanding of dialect boundaries in Michigan, examining factors that affect language change, and investigating the common belief that English in the Lower Peninsula is standard and English in the UP is non-standard. 
Benson, E. (2003). Folk linguistic perceptions and the mapping of dialect boundaries. American Speech, 78 (3): 307-330.
Bucholtz, M., Bermudez, N., Fung,V., Edwards, L., & Vargas, R. (2007). Hella Nor Cal or Totally So Cal?: The Perceptual Dialectology of California. Journal of English Linguistics, 35: 325-352. DOI: 10.1177/0075424207307780
Evans, B. (2012). “Seattletonian” to “Faux Hick”: Perceptions of English in Washington state. American Speech 86, 4: 383-414. 
Giles, H. (1970). Evaluative reactions to accents. Educational Review 22: 211-227.
Inoue, F. (1996). Subjective dialect division in Great Britain.” American Speech
71: 142–61.
Irvine, J. & Gal, S. (2000). Language ideology and linguistic differentiation. In P. Kroskrity (ed.), Regimes of language: Ideologies, politics, and identities, pp. 35-84. Santa Fe, NM: School of American Research Press.
Lambert, W., Hodgson, R. Gardner, R. & Fillenbaum, S. (1960). Evaluations reactions to spoken languages. Journal of Abnormal and Social Psychology, 60: 44-51.
Lance, D. (1999). Regional variation in subjective dialect divisions in the United States. In Preston 1999, pp. 283–314.
Lippi-Green, R. (1997). English with and accent: Language, ideology, and discrimination in the United States. New York: Routledge.
Mase, Y. (1964). Hôgen Ishiki to Hôgen Kukaku. Gengo Kenkyû, 36: 1–30. Repr. in English as Dialect Consciousness and Dialect Divisions: Examples in the Nagano-Gifu Boundary Region. In Preston, D. ed., Handbook of perceptual dialectology, pp. 71-99Amsterdam: Benjamins.
Niedzielski, N. & Preston, D. (2003). Folk linguistics. Berlin: Mouton de Gruyter.
Preston, D. (1986). Perceptual dialectology. Dordrecht: Foris.
———. (1993a). Folk dialectology. In D. Preston, ed., American dialect research, pp. 333–77. Amsterdam: Benjamins.
———. 1993b. Two Heartland perceptions of language variety. In Frazer, T., ed.,“Heartland” English: Variation and transition in the American Midwest, pp. 23-47. Tuscaloosa: University of Alabama Press.
———. 1996a. “Whaddayaknow?” The modes of folk linguistic awareness. Language Awareness, 5: 40–74.
–––. (1996b). Where the worst English is spoken. In Schneider, E., ed., Focus on the USA, pp. 297–360. Varieties of English around the World 16. Amsterdam: Benjamins.
—–, ed. 1999. Handbook of perceptual dialectology. Amsterdam: Benjamins.
–––. (2002). Language with an attitude. In Chambers, J. et al, eds., Handbook of language variation and change, pp. 40-66. Malden, MA: Blackwell.
—–. (2002). Perceptual dialectology: Aims, methods, findings. In J. Beerns and J. Van Marle,  eds., Present-day dialectology, pp. 57–104. Berlin: de Gruyter.
Remlinger, K. (2009). “Everyone up here”: Enregisterment and identity in Michigan’s Keweenaw Peninsula. American Speech, 84 (2): 118-137.
–––. with Salmons, Joseph and von Schneidemesser, Luanne. (2009). Revised perceptions: Changing dialect perceptions in Wisconsin and Michigan’s Upper Peninsula. American Speech, 84 (2):176-192.
–––. (2007a). The intertwined histories of identity and dialect in Michigan’s Copper Country. In A. Hoagland, E. Nordberg, & T. Reynolds, eds., New perspectives on Michigan’s Copper Country, pp. 62-84. Houghton, MI: Quincy Mine Hoist Association.
–––. (2007b). Newfies, Cajuns, Hillbillies, and Yoopers: Gendered media representations of regional dialects. Linguistica Atlantica, Journal of the Atlantic Provinces Linguistic Association, 26-27: 96-100.
–––. (2006). What it means to be a Yooper: Identity, language attitudes, and variation in Michigan’s Keweenaw Peninsula. In M. Filppula, M. Palander, J. Klemola and E. Penttilä, eds., Topics in dialectal variation, pp. 125-144. Joensuu, Finland: University of Joensuu Press.
Sibata, T. (1959). Hôgen kyôkai no ishiki. Gengo Kenkyû 36: 1–30. Repr. In English as “Consciousness of Dialect Boundaries” in Preston, D., ed., Handbook of perceptual dialectology, pp. 39-62. Amsterdam: Benjamins.
Simon, B. (2005). Dago, Finlander, Cousin Jack: Ethnicity and identity on Michigan’s Upper Peninsula. In D.Preston and B. Joseph, eds., Language profiles: Michigan and Ohio. Ann Arbor, MI: Caravan Books.
Silverstein, M. (2003). Indexical order and the dialectics of sociolinguistic life. Language and Communication, 23: 193-229.

Enregisterment and Semiotic Mediation: The Role of Recognition in Dialect-making

 Adam Sandler’s character, Cajun Man, a feature of Saturday Night Live skits in 2003, is defined as ‘Cajun Man’ by his mock Cajun French accent, particularly its stress on the nasalization of the final syllable –tion: “onION,” “vacaTION” “hallucinaTION,” “deficaTION”. But Cajun Man is also ‘Cajun’ and ‘man’ because his talk centers a stereotype of a Cajun man: spicy food, drunkenness, and scatological humor. The accent and content of his talk, in conjunction with his appearance—straw hat, suspenders, a plaid flannel shirt with cut off sleeves over a dingy white longjohn shirt—interdiscursively combine to index ‘Cajun man.’ Meaning is communicated semiotically—through recognizable linguistic features and cultural values attached to the visual image. Similarly, we recognize the authenticity of Boston Harbor Tea in the label’s index of a Boston pronunciation: “’Bawstonaba’ Registered” (‘Boston Harbor Registered’). Its authenticity is further reinforced with the intertextuality of the claim that the tea is “blended and packed for…the firm which supplied tea 1772-1774 for the historic Boston Tea Parties”. The enregistered pronunciations of aw in Bawston and the (r)-less and (h)-less pronunciations of aba (harbor) along with the claim to historic events semiotically mediate meanings linking language, place, and authenticity. Likewise, no one doubts the authenticity of Pauly D as a real north shore New Jerseyan on MTV’s reality show Jersey Shore. As a prototypical Italian-American East Coast male, his character’s validity depends on what he says, how he says it (in particular his use of –in for –ing and the lack of (r) after vowels), as well as within a particular context (sitting on a tricked-out motorcycle in a working class neighborhood): “I was bon (‘born’) and raised a guido. It’s just a lifestyle, it’s bein Italian, it’s representin, family, friends, tannin, heya (hair) gel, everything” (Jersey Shore, Episode 1). The combined linguistic features and images index ‘Italian American,’ ‘white,’ ‘working class,’ and ‘heterosexual masculinity.’ The semiotic mediation of regional dialect and speaker are recognizable in these examples through the use of specific linguistic features in given contexts, features and contexts that combine to evoke certain and specific meanings that yoke language, place, and people. Because these contextualized features are considered exemplar, their use effects the perception that best speakers are the most authentic locals. Similarly, the most authentic products, such as Boston Harbor Tea, are the best representations of localness through the use of enregistered features and intertextual links with a historic event. Moreover, used within specific contexts and juxtaposed with stereotypical images that evoke authenticity, dialect is often discursively constructed as gender-, sexuality-, race-, and class-based. Social meanings are as much a part of the semiotic mediation of ‘dialect’ as are pronunciation and word use and cultural categories: speaker authenticity relies on recognizable discursive practices, where language use is the determining factor, and indexicality, where the ‘best’ speakers of a regional variety are recognized as the most authentic locals. In these ways, the practice of masculinity—what it means to be heterosexual male, specifically, is tied to the practice of dialect—dialect-making.