This article first presents an overview of the social and demographic phenomena specific to the language shift situation in Sanpete County, Utah, focusing on the biggest non-English-speaking group, the Danes. This overview includes the assimilation norms that were present in the community (including from the dominant religion, The Church of Jesus Christ of Latter-day Saints), social and geographical isolation, and related issues of identity and language maintenance. Using interdisciplinary methods under the rubric of sociocultural linguistic research, our analysis presents an overview of the state of Danish in today’s Sanpete County, then further divides the Danish linguistic elements into two main categories: overt and covert. The analysis of these items makes use of the notion of postvernacular language use, as well as highlighting the female and domestic-related networks of transmission. This study of the Danish-language situation in Sanpete County offers a glimpse of the final stages of complete language shift, revealing information about a rare and under-examined linguistic community within the American context.
The first large-scale, permanent settlements by Europeans and European Americans in the present-day state of Utah1 were a result of the relocation of members of The Church of Jesus Christ of Latter-day Saints (lds, commonly known as the Mormons) from the Midwest. The first settlers to arrive, setting up communities in the Great Salt Lake Valley in 1847, were predominantly English speaking. Soon after settling the Great Salt Lake Valley, groups of settlers were sent by Mormon Church leaders to various locations around the territory, with a goal of establishing a Mormon stronghold in the western United States. Among these settlements was a location now known as Sanpete County, which lies in a high valley in the geographical center of the current state of Utah. The first settlement in Sanpete County was established in 1849 by English speakers from other parts of the United States (Antrei and Roberts, 1999; Carter, 2015). Thus, English was established as a “founder language” (Mufwene, 1996) both in Utah territory as well as Sanpete County.
Shortly after settlement, however, starting in 1850, thousands of immigrants of Scandinavian background joined the English-speaking settlers in the new territory. This was due to intense proselytizing in the Scandinavian countries—especially in Denmark, where newly relaxed religious laws, coupled with political and economic upheaval, made the situation ripe for religious conversion and also migration from Denmark. By 1850, no fewer than five Mormon missionaries were working in Denmark, which at the time had a population of about 1.4 million (Grøngaard Jeppesen, 2010: 17ff). In total, nearly 25,000 Scandinavians migrated to Utah from about 1850 to 1920. The majority of these, 54 percent, were Danes, with most of these coming from the Jutland area, along with 33 percent Swedes and 13 percent Norwegians (Henrichsen and Bailey, 2010). Salt Lake City, as the capital/most urban area of Utah, attracted the highest number of Scandinavian settlers, but the highest density per capita of Scandinavians settled in Sanpete County, where, in 1890, 22.54 percent of the total population had been born in Scandinavia (Henrichsen and Bailey, 2010).
Utah, then, and especially Sanpete County, serves as a relatively early (and relatively unexamined) example of a language contact situation between Scandinavian languages and English—in addition to American Indian Languages—in the American West. It is also unusual in that settlers of Danish descent constituted the majority of the immigrant population (Grøngaard Jeppesen, 2010: 19). These aspects of the social and cultural situation in Utah and Sanpete County are explored further in Section 2 of this article, with additional information related to demographics and the cultural domination of the Mormon Church.
The situation involving Scandinavian immigrants in Utah and Sanpete County has been studied from a number of research perspectives, including the architecture (Carter, 2015), ethnicity (Simmons, 2007), material culture (Abbot, 2013), humor (Geary, 1999; Jenson, 1999) and folklore (Eliason and Mould, 2013). The language contact situation in Utah has also been explored, including references to Sanpete County. However, Sanpete County has not been singled out as a focal point of investigation in work on language shift. In addition, the research that has been conducted thus far on Utah’s language contact situation (with regard to Scandinavian languages) is probably most accurately characterized as focusing on the language policy of the Mormon Church and Utah Territory, and from a top-down perspective using written records and archive material (for example, see Henrichsen and Bailey, 2010; Henrichsen, Bailey et al., 2009; Simmons, 2007). There is relatively robust research on linguistic features of Utah English, including work on phonological and syntactic features (including, e.g., Bowie, 2003, 2008; Di Paolo, 1993; Di Paolo and Faber, 1990), but this work has focused on the English speakers in Utah, rather than language shift among non-English speakers.
The present study differs from previous accounts in that, in addition to relying on Mormon Church-held records and previously published accounts, we have entered the Sanpete County community at the grassroots level, conducting face-to-face interviews with descendants of Scandinavian heritage. The aim of this ongoing investigation is to gain an understanding of the language shift situation in this rare (if not unique) social setting, with its predominance of Danish-speaking immigrants in a relatively isolated Western u.s. religious community. The two key questions driving this investigation are:
What elements of Danish as a heritage language are still present in Sanpete County? What linguistic domains of use are represented?
How were the remaining elements transmitted from the immigrant generation to the present day?
The main hypotheses are that women have been the transmitters of Danish, and the linguistic elements that have been retained relate to the domestic sphere. The rationale for these hypotheses is presented in Section 3.
The article is laid out as follows: Section 2 offers background information on Scandinavian/Danish Mormon immigration to Utah, the language policy of the lds Church, and the early social structure of Sanpete County contrasted with the Scandinavian profile there today. Section 3 details the nature of the data used in the study, as well as how it was obtained, with special attention paid to the participants in the study. Section 4 gives a composite overview of the lexical items and other linguistic features still evident in Sanpete County, followed by an account of these items according to the divisions of reported speech, covert vs. overt transmission and networks of transmission. Section 5 concludes the article.
This section lays the background for the transmission of Danish in Utah and Sanpete County by first describing the movement of Danish and other Scandinavian groups to Utah, the language situation in Utah, and demographics in Sanpete County.
2.1 The Danish Mormon Exodus
The first mass migration of Danes to the United States occurred only with the Mormon influx; there were 3,749 Danes who migrated to the u.s. in the 1850s, and more than three quarters of these were Mormons, the majority of whom ended up in Utah Territory (Nielsen, 1981: 89). The main movement of Danes to Utah began on September 30, 1853 (Antrei and Roberts, 1999; Carter, 2015). By 1860, Utah had the highest number of Danish-born population in the United States (Nielsen, 1981: 59): 1,824 immigrants, constituting 4.53 percent of the territory’s total white population (Henrichsen, Bailey et al., 2006). Utah’s Danish-born population peaked at 9,132 in 1900 (Henrichsen, Bailey et al., 2006: 65). Utah’s Scandinavian immigrant majority was short-lived, however. Although Mormon Scandinavians continued to come to Utah, subsequent Scandinavian immigrants tended to settle in the more temperate, familiar, and livable habitat of the American Midwest. By 1870, Wisconsin had the highest number of Danish immigrants, and in 1890, the state of Iowa had the highest number (Nielsen, 1981).
There are a few key features that separate the Danish/Scandinavian Mormon exodus from later immigrant groups to the United States. These features are key in explaining the language shift process in Utah. Importantly, it was typical of Mormon immigrants to travel as entire families, even including multiple generations. This is different from the general trend in the later Danish migration when single young men outnumbered family migration (Hvidt, 1975). In fact, female Scandinavian immigrants to Utah slightly outnumbered the men, 53 to 47 percent (Mulder, 1957). This demographic feature is crucial; it is well-established that in traditional Western societies women contribute to the persistence of a heritage language, as the option to marry within the ethnic/linguistic group makes it possible—or at least more likely—to carry on heritage customs and language (see, for example, Boas, 2009 on exogamy). In Utah, it was possible for Danish men to marry Danish women, whereas in the later settlements in the Midwest, for example, there were relatively few Danish women, and, accordingly, little chance to uphold Danish as a family language.
Another important factor regarding the Scandinavian Mormon immigrants was the overall social status of the group. In general, European immigrants to the United States have not been among the highest socioeconomic classes (with a few exceptions, such as the middle class religious dissenters who settled New England) (Tamasi and Antieau, 2015; Wolfram and Schilling-Estes, 1998). The Danish Mormons were no exception, with the majority of the converts and eventual settlers to Utah stemming from “the abused peasantry” (Antrei and Scow, 1982). Many of them were shunned from their families and home communities when they converted to what was considered a new, untrustworthy, and even blasphemous religion. In Utah, although among people who shared similar religious views, they were still often socially shunned for their “un-American” customs, languages, traditions, and due to their sheer number (Mulder, 1957). While some accounts claim that ethnic differences between Mormon settlers were diminished in the face of overall religious cohesiveness, we propose that the social disenfranchisement of the Scandinavian settlers to Utah is another key factor in the language shift process, particularly in Sanpete County. This theme and others will be expanded in Section 2.3, which deals with the demographics and social situation in Sanpete County.
The missionaries tempted with great possibilities, free crossing and the company of guides the whole way from Denmark to Utah. The migrants just had to sign that they owed the Mormon Church for the crossing and the journey. Suddenly, the migration to the u.s.a. became possible also for the property-less domestic servants and farmhands. Until now, it had only been small farmers and others with a certain amount of money who could afford the long journey and establishment in the new land.grøngaard Jeppesen, 2010, citing Hvidt 1971; authors’ translation
2.2 The Contemporary Language Policy of the Mormon Church
Mormon settlers to Utah established their first settlement in the Great Salt Lake Valley in 1847. This settlement was the outcome of a long period of conflicts with other communities in the Midwest and elsewhere. The leaders of the Mormon Church, including its second leader, Brigham Young, had an explicit desire to create an independent nation state of “Zion” in the American West. For this reason, as well as for external reasons that were inherent to the location—such as the extreme climate and the American Indians who already inhabited the region—early life in Utah was under tight control, for the survival of the Church and also for its people and their settlements.
With this backdrop in mind, it can be assumed that immigrants to Utah were expected to assimilate to their new norms at an even higher rate than immigrants elsewhere in the United States—which, as has been noted extensively in the literature, was already quite high. Within the broader context of the United States, Scandinavian immigrants have been perceived as among the most “successful” to assimilate to American life, due to the similarities of the Northern European lifestyle and values with English-speaking communities (Nielsen, 1981; Mulder, 1957; Grøngaard Jeppesen, 2010). In Utah as well, the expectations were that the non-English-speaking settlers would integrate both culturally and linguistically, although in reality the integration was not without complications. A well-known quotation from then-lds president Brigham Young was that [immigrants] should learn English, “the language of God, the language of the Book of Mormon, the language of these Latter Days” (Mulder, 1957: 181). This “duty” (ibid.) was placed second only after the duty of learning how to live in the harsh landscape; thus, language skills were placed only after physical needs such as food, water, and shelter. To this end, English language skills were taught by lds Church missionaries to immigrants in Denmark as well as en route to Utah.
At the same time, there appears to have been an awareness that adult learners of English were not likely to be able or willing to abandon their native language; while efforts were made to teach English to the Scandinavian immigrants, there were also allowances made that enabled them to participate on at least some level in community life in a Scandinavian language in the form of church services, newspapers, cultural events, and so on (the early language situation in Utah is documented in Henrichsen, Bailey and Huckaby, 2006).
Thus, there was overt pressure to learn English and thereby participate fully in civic and church life, but with some concessions for immigrants. This “mixed message” regarding language seems to have been interpreted, based on our research, at the level of the individual, depending on factors such as family ties, the community, and the date of migration.
2.3 The Demographics and Social Structure of Sanpete County
When the Scandinavian immigrants began to arrive in Sanpete County, English-speaking settlers had already established a fledgling community at the command of Mormon Church leaders from Salt Lake City. The first group of Scandinavian settlers was sent to Sanpete County in 1853, along with other groups to Sevier County, Cache County and Box Elder County (Henrichsen and Bailey, 2010). Explanations for this move vary; some accounts claim that the higher elevation and colder weather was familiar to the Scandinavians and would thus improve their chances of survival there (Antrei and Roberts, 1999; Geary, 2013). Others claim that it was important for the image of the Church to remove foreign elements from the capital area of Salt Lake City, thus decreasing the image of Utah and the Mormon Church as being something alien and strange (Antrei and Scow, 1982: 75; Simmons, 2007).
There are conflicting reports concerning any social tensions that would have existed in Utah and Sanpete County due to differences in ethnic and language background. Some accounts gloss over the issue (Carter, 2015;4 Nielsen, 1981; Simmons 2007), while others attempt to at least minimally address the cultural clash Scandinavian immigrants met in Utah (Abbott, 2013; Antrei and Roberts, 1999; Henrichsen and Bailey, 2010; Mulder, 1957). Regardless of the explanation (some sources claim it was a simple a matter of lack of other available land, see Carter, 2015), one indisputable fact is that most Sanpete County towns were segregated, with the Danes and other Scandinavians living in a quadrant with poorer farm land, and the English speakers (of American and British descent) living in a more socially prominent and more fertile area.
…one may hear the various changes in Danish from North Jutland to Copenhagen, and listen to Norwegian as spoken in Christiania, Trondhjem and in the mixed-German Bergen, also to the worst Skane or southern Swedish, and to the best as spoken in Goteborg and Stockholm, or hear the different varieties of Swedish from Upsala to Ystad, and yet not hear anything quite like the mixture which is called Danish, Norwegian and Swedish in Utah.cited in Mulder, 1957: 249
Going hand-in-hand with geographical segregation, multiple social divisions are documented in the literature as well as in our own fieldwork. Both written and oral accounts of the early decades of settlement report not only social tensions between English-speakers and Scandinavian speakers, but also between the different Scandinavian groups. For example, several participants in our study discussed tensions between Swedish speakers and Danish speakers over which language should be used for church services at the local Scandinavian congregations. While we cannot generalize after the fact, our research indicates that for many Scandinavian settlers and their families, there was a distinct language and cultural community embedded within the matrix English-speaking society. It was already noted that the majority of Sanpete County towns exhibited social and geographical segregation. This division is still outwardly apparent, for example, in the town of Manti, where even today the former Danish quadrant is made up of smaller, poorer residences that are distinct from most of the rest of the town (see Carter, 2015). With regard to social segregation, it is important to note that Danish people lived in the same neighborhoods with each other. Due to their number and the presence of women, they had the possibility to marry within the ethnic group, which, according to some sources among our interviews, was a practice for at least the first generations after settlement (see also Abbott, 2013). They were able to conduct their shopping in Danish, they were able to go to church in Danish (until the 1940s), they sang in Danish choirs, they read Danish-language newspapers (Bikuben was published in Utah from 1875 until 1935), they danced to Danish music, and they socialized with other Danes. Schooling and civic life in Danish was not possible. Participation within the matrix social order demanded unidirectional bilingualism and even a diglossic situation on the part of the early Scandinavian settlers.
As is typical for many immigrant situations in the United States and elsewhere, the first Utah-born generation, at least in many families, was bilingual. There are no known audio recordings from this period, but written evidence and contemporary accounts (for example those chronicled in Mulder, 1957) point toward an Anglicized variety of Danish. In addition, much of the humor and folklore accounts that have been reported from Sanpete County (the town of Ephraim, apparently, was referred to as “the funniest town in America” see Geary, 2013) are based on the L2 use of English among Scandinavian immigrants, or what was locally referred to as “brogue.” Many of the second generation Danish Americans in Sanpete County could speak Danish, but they did not appear to know how to write in Danish. Our linguistic investigations in Sanpete County so far indicate that transmission of Danish after the first Utah-born generation is best characterized as unpredictable; there were a few families who carried on speaking Danish as a home language to some extent, although these seem to constitute a minority. In most cases, it seems that after the second generation, Danish elements were restricted to a few core vocabulary items related to the home front, in addition to nursery rhymes, songs, and other cultural features that were carried on as ethnic relics (cf. Fishman, 1991; Haugen, 1953). These linguistic items and others are listed in Section 3 of this article.
2.4 Sanpete County Today
Today, the most overt manifestations of the Scandinavian input to Sanpete County lies with physical features such as architecture, or on signs on the landscape, including surnames used as names for businesses, in phone directories, or the names on local graves. Modern day immigration and resettlement from other communities has altered the demographics of Sanpete County. However, acknowledgement of the Scandinavian background of the area is brought to the forefront during public celebrations and in public spaces, notably in the form of the local Scandinavian Festival, which has been hosted annually since 1975 in the town of Ephraim (the population of which, during its inception, was more than 90 percent Scandinavian-born) This two-day event includes an aebleskiver breakfast, speeches and Scandinavian-themed stories, folk dancing, a half-marathon, a bike race, and a parade down the town’s Main Street. These types of events are reminiscent of what Pico Larsen (2006) refers to as the “Disney-fication” of Danish culture in her investigation of Solvang, California. While interesting in their own right, these overt attempts at the public level to connect to a Scandinavian past are not the focus of the present research. Rather, our work attempts to connect at the individual level with members of the Sanpete County community to determine if/which elements of Danish language and culture that have taken on post-vernacular status.
3 Materials, Methodology, Analysis
The data gathered for this investigation has mostly been informed by sociolinguistic/ anthropological methods under the rubric of sociocultural linguistic research (Bucholtz and Hall, 2008), using a network method (Milroy, 1980) to establish relationships within the community. The findings presented in this article were gained primarily through interviews and participant observation that were carried out in Sanpete County over three summers, 2012, 2013, and 2015. However, the overall data for our larger investigation of the language situation in Sanpete County includes printed and written material gained from archives, libraries, as well as from personally held collections in Utah. These materials include letters, journals, photographs, pamphlets, books, newspapers, and other printed matter (for example, wedding invitations and holiday cards). In addition, we have photographed Scandinavian artifacts and objects in museums, homes, and public places.
Finally, the maintenance of field notes during research visits has rounded out the overall robustness of the data. For example, it is possible that in the day-to-day interactions in Sanpete County, both in public and private, notions of Danish and Danishness enter spontaneously into a conversation—for example at the cash register of a local shop, a scenario that has been encountered on several occasions. Such events have been written down afterwards in a field notebook.
3.1 Interviews, Participants and Recordings
In this article, we report mostly on oral accounts of language that were gained through interviews and discussions. So far, there have been 16 recorded interviews with Sanpete County residents, involving 16 different speakers. A few members of the community were already familiar to the researchers at the time fieldwork was initiated. These pre-existing relationships opened up new acquaintances within the Sanpete County community among those who have Danish/Scandinavian5 background. In a community of this composition and background—that is, a relatively closed, tight-knit community with a strong religious presence—it was seen as necessary (or at least as an advantage) to have some sort of personal connection to the research site prior to entering with a goal of conducting research. The background of the researchers was probably conducive to achieving credibility with participants and to ultimately meet the goals of the study; one of the researchers is originally from Utah (and thus has an insider’s understanding of the social and cultural structure), and the other is from Denmark and researches Scandinavian languages worldwide (and is therefore qualified to make judgments concerning Danish elements found in the data).
In total, there are 18 hours and 12 minutes of recorded interviews, involving different constellations of speakers. The interviews each began with a verbal informed consent, and were recorded using a handheld recording device. The interviews lasted from 30 minutes to two and a half hours in length, and included between one and four speakers in addition to the interviewer. Eleven of the recorded interviews have been with one speaker and one interviewer, although there have been interview settings where there have been up to four speakers in addition to the interviewer. Some of the participants have been interviewed more than once, as well as appearing in group interviews. The oldest participant was 94 years old at the time he was interviewed. The youngest participant was 63 years old at the time he was interviewed. The average age of the participants has been approximately 77 years old. The majority of the participants are female: 10 out of 16.
Subsequent to the initial visit to Sanpete County in 2012, there was a concerted effort to seek out interviews with women. This is because it became evident, based on early networks in the community, that local women tended to be more likely than men to have concrete recollections of Danish sourced linguistic elements (and other cultural knowledge) based on personal experience in the home front, particularly if, as children, they had interacted extensively with female relatives who were Danish. Based on our research and interviews, it appears that men tended to carry out a more public role in the community, with the result that interviews with men have tended to focus more on church and civic life—which for the most part excluded the use of Danish.6 Research on the same region supports this finding: in an overview of the architecture of Sanpete County, Carter (2015) notes that in reading more than fifty years’ worth of church congregation (called a “ward” in the lds Church) minutes, he encountered “no female voice.” When he did find a woman’s voice, it was in the form of a personal journal or diary, and discussed daily routines (ebook location 2970). An investigation of the material culture of Utah’s Scandinavian immigrants discusses the importance of the lds women’s organization called “Relief Society,” which offered the primary means of community involvement for women, as well as a means for affirming and perpetuating Scandinavian culture (Abbott, 2013: 293).
The interviews that have been conducted so far have necessarily focused on older members of the community. These are those who have any recollection of speakers of Danish and who may themselves speak a few words; younger community members seem to have little or no connection to the heritage language community. In this sense, the Sanpete County community is at a critical crossroads: with the oldest generation, we are still dealing with a heritage identity that was passed on in the home setting. Within the same community, at the same time, the connection of the younger generation to Denmark and Scandinavia is something remote, foreign, and “other,” embodied through such public means as official historical accounts, monuments, public events, and second language learning.
Yet even among the older generation, the relationship with Danish language and culture is not always straightforward. Not everyone we approached was willing or able to discuss their family heritage. It appears that for some families, there was a concerted effort to hide or shed the Scandinavian past in favor of a new Utah and American one. Some people responded to our queries by stating that that their family had become American when they arrived in Utah, and there was nothing more to say on the matter.
To date, there have been three fieldwork visits to Sanpete County. The first two are best characterized as exploratory, with the aim of gaining an understanding of the community, meeting local residents, and attempting to ascertain if and what kinds of linguistic and discourse elements are connected to Danish and Denmark. During the first two field visits, entailing a total of three weeks in Sanpete County over the course of two consecutive summers (2012–2013), semi-structured interviews were conducted with 14 people.
The third research visit to Sanpete County, which lasted one week, took place during the summer of 2015. During this visit, interviews with three women and two men were recorded, with an additional two interviews that were not recorded due to inferior sound quality and other external factors. These interviews were conducted with targeted individuals who had been fruitful participants during previous research visits. However, additional networks opened up on this visit, resulting in two interviews with new participants.
The interview format of the third research visit differed slightly from the previous two in that questions were more targeted toward certain themes and outcomes, making use of knowledge gained from previous visits. With this newer set of interviews, there was a more focused effort to elicit Danish names for household items and other linguistic and cultural elements. Part of the interview format asked participants to visualize and describe the contents of a Danish-ancestry family member’s home, and also to describe any shared activities with those family members. This elicitation technique was considered appropriate due to the fact that the Danish language in Sanpete County is nearly gone; we are dealing with historical factors that are no longer in everyday discourse or consciousness. The language situation in Sanpete County is far into the process of complete shift into English, making traditional dialectology formats such as questionnaires and word recognition tasks non-applicable. Any remaining elements of Danish and Danishness appear, for the most part, to be tied to the private realm, which mandates a personal and individually tailored interview format. This is also a realm relating more to the traditional role of women in the home (a theme addressed further in Section 4 of this article). Furthermore, there is an intricate relationship of Danish-sourced elements; what remains often appears to be outside the level of awareness. If the researchers were to ask (as we naively did on previous visits) “Do you know any Danish?” the answer would be “No,” but if we ask, “What was in your grandmother’s living room? … Was there a special name for anything in there?” the answer might be, “Yes, there was something she called a skammel (‘footstool’).” In other words, attempting to evoke a specific memory of a (female) family member’s home has been useful in our data collection.
3.1.1 Transcribing the Interviews
At this stage in our research, the interview data has been only partially transcribed. The recordings from the more focused 2015 fieldwork visit have been reviewed in their entirety by both authors. For this article, the portions of the interviews that contained Danish sourced lexical items and quotations about Danish-speaking community members have been transcribed. Excerpts from the 2012 and 2013 recordings containing relevant linguistic items have also been reviewed by both authors. In this article, the identity of the participants of the study have been anonymized. The speakers are presented in the article according to their year of birth and their gender; for example, the identity 46F refers to a woman who was born in 1946. In cases where more than one speaker was born in the same year, they are distinguished with a numeral after the gender designation (e.g., 28F vs 28F1).
4 What is Left of Danish in Sanpete County?
In the analysis of our data, we have followed two key questions: (1) What is left of Danish in Sanpete County? and (2) How was Danish transmitted? A quantitative approach to our data is meaningless as it would only confirm what is obvious, namely that there is no however hesitantly spoken Danish left. What is left are a limited collection of Danish lexical items, almost all of them denoting items from the household, food and family, as well as nursery rhymes and short songs. Section 4.1 provides examples of this. These lexical items seem to exist in the mind of the speakers on different levels of active speech production and consciousness: on the one hand, there are lexical items that the participants themselves use or have used during childhood (we include both productive and receptive competence in the notion of ‘use’). The speakers perceive most of these items clearly as Danish. Other items, however, are (or were) not recognized by the speakers as being Danish but seem to have been attributed to a local or even personal way of speaking. This difference is discussed in Section 4.3, along with examples. On the other hand, the participants rendered language use or speech routines that were produced originally by older generations, typically their mothers or grandparents, not by the participants themselves. This is detailed in Section 4.2.
4.1 Maintenance of Lexical Items
Most of the items and lexical chunks that the participants of the study use, or remember using in childhood, are connected tightly to the immediate home surroundings and close family members. The domain that is represented most strongly is food, followed by household items and nursery rhymes. The food domain includes what seems a quite complete list of traditional Danish food and dishes with no American equivalent, for example æbleskiver ‘traditional Danish fried pancake balls,’ frikadeller ‘meatballs made of meat, crumbs, egg and milk,’ grønkål ‘kale’ and surkål ‘sauerkraut, choucroute,’ klatkager ‘cold rice pudding fried as a small pancake,’ finker ‘chitterlings,’ surmælk ‘curdled milk,’ smør ‘butter,’ knapost ‘cheese made of buttermilk,’ pølser ‘sausages’ and snaps ‘schnapps.’ The interviews contain descriptions of other Danish food items and customs, but these are rendered in English, e.g., ‘red mash’ (cooked red cabbage, Danish ‘rødkål’), sweet soup and the traditional Danish Christmas dessert made of rice pudding, whipped cream and one whole almond (risalamande). In addition, other Danish food items have taken the form of direct translations, e.g., sistercake (Danish søsterkage).
Household items that are remembered by our participants are bitte hus ‘little house, outhouse,’ stol ‘chair,’ skammel ‘(foot) stool,’ æbleskivepande ‘a pan for making æbleskiver,’ stue ‘living room,’ klædeskab ‘clothes cupboard,’ blomster ‘flowers’ and hakkeboard, a bilingual combination of Danish hakkebræt and English ‘cutting board.’ Also skruk ‘skrook’, denoting a ‘broody hen’ or ‘a nesting hen’ is remembered by our participants, sometimes in the sense of a grumpy woman, or a woman who wishes to become pregnant.
The third domain where Danish lexical items are maintained by the speakers are nursery rhymes and short songs. For example, the Danish finger-naming rhyme Tomme(l)tot slikkepot langemand guldbrand og lille bitte spillemand is rendered by several speakers in slighty different versions (e.g. speaker 28F, 43F, 24F). One explanation for this variation is that the rhyme has slightly different Danish versions, as well as similar rhymes in Norwegian and Swedish. In addition, the speakers differ in their level of English influence. Speaker 39F offers an Anglicized version of dikke dik min far, sikke lår, sikke ben, sikke skridt, han ta’r ‘my daddy what legs he has, what steps he takes’. Further, speaker 28F reports that she and her siblings would sing kør kør kør ‘turn turn turn’ when they helped churn the butter. It should be noted that the speakers attribute these elements of Danish to speakers other than themselves or recalling events from their childhood (see Section 4.2); as such, this is not firsthand language production but rather reported use. There is no evidence that any of the speakers have transmitted these linguistic items to their own children or that they are in active use today.
In the rendering of the lexical items, chunks and songs, there is ample evidence of grammatical and semantic knowledge that is either reduced due to linguistic attrition across the life-span of the speakers or because it never was passed on from the previous generations. The speakers differ greatly with regard to the degree that they know the semantic content of the Danish words. Some participants were able to discuss the making of æbleskiver ‘traditional Danish fried pancake balls’ in great detail (conversation among the speakers 39M, 39F, 28F and 49M) and were even able to produce and comment on different pronunciations of the word æbleskiver and the sociolinguistic notions tied to it (i.e., Low and Standard Danish, speaker 28F1). However, other speakers remembered lexical items as chunks but did not have any knowledge about the concept. For example, speakers 43F and 35F talk about the Danish dish grønkål ‘kale’ but to them it is some mysterious, unknown plant related to parsley. In this conversation grønkål, which would be [ɡ̊ʁɶ̝nkʰɔ̝̈ːl] in Standard Danish pronunciation, is pronounced [ɡɻŋkʰl]. This realization cannot be attributed neither dialectal nor historical variation in Danish. This is a clear example of how the lexical items have been acquired by our participants as chunks they are not able to analyze, as they have no productive structural knowledge of Danish (no lexical and/or grammatical knowledge, no orthographic knowledge).
A common morphosyntactic feature throughout the group of participants is the rendering of lexical items with the English plural marker –s, despite the item already including the enclitic Danish plural marker –er or -r, e.g., æbleskive-r-s or frikadelle-r-s. This doubling of lexical-grammatical information occurs with definiteness markers, as well, e.g. “a nissen” (‘elf, gnome’), where the Danish enclitic definite article –(e)n is doubled by the preponed indefinite English article. This implies that the participants have no lexical-grammatical knowledge enabling them to differentiate between a lexical stem and inflectional morphology. What they remember of Danish are lexical chunks that are not productive in any derivational sense.
It becomes apparent that many of these lexical items do not have an English language equivalent within the immediate context. As such, they fall within the general definition of what has been termed a “gap” in the repertoire of bilingual or even semi-bilingual speakers (Matras, 2009). In choosing the resources from one language as opposed to another, speakers “attempt to avail themselves of their full inventory of linguistic resources, at all times and in all contexts of interaction” (Matras, 2009: 150). Unlike core forms, these cultural loans that do not have an equivalent in the recipient language do not presuppose bilingualism of the speakers in order to be maintained. Their ability to fill the “gap” in the community’s lexicon can lead to long-term incorporation in the recipient language—sometimes resulting in language borrowing (Matras, 2009: 110).
4.2 Reported Speech Routines
Apart from the lexical inventory that is or has been part of the participants’ everyday speech practices, they also offered accounts during the interviews of the Danish language use by former generations. Such reports are, of course, non-attested speech which must be treated with caution as a data source. At the same time, however, this kind of reproduction opens up a window on language use by former generations during the childhood of the participants. This grants an insight into speech practices that is hard to achieve by other means, due to the absence of recordings and the controlled and gender-biased historiography that is available.
The speaker in (1) had Danish-speaking parents; she also married into a family of Danish heritage. Other speakers remember older family members using fixed phrases such as skål ‘cheers’ (still used to some extent today) and exclamations such as oh my herre gud ‘oh my Lord’ (speaker 28F1’s Scottish-Danish grandmother). Danish kinship terms and terms of endearment seem to have persevered to some extent, too, e.g. bedstemor and mamse, both denoting ‘grandmother’, and bitta [sic] Laura ‘little Laura’ and little skidt ‘poor little thing’. These speech routines seem to have been used to express local belonging and Danishness as well as to fill gaps in their American lexicon with cultural loans. The functionality of the codeswitches thus lies on the continuum between a stylistic choice and a default expression (Matras, 2009: 111).
(1) 24F: and from time to time a customer would come in and and- and there were a few xxx and they would say ‘How are De?’ meaning ‘How are you?’ and then the answer would be ‘Temmelig godt.’ And then they would ah ‘Tak skal du ha’ ‘Thank you.’ And then there would be a reply and I don’t remember what that was. And that would be the extent of the Danish I would hear.
int: Ja (‘yeah’). And the rest would be in English?
interview with 24F, recorded in 2012
4.3 Covert and Overt Code-Switching
Our analysis of the Danish items in our data brought two concurrent phenomena to light, which we have labeled “covert” and “overt,” referring to how the Danish linguistic items were acquired and how they are categorized by the speakers themselves.
Overt Danish items include those that the participants are consciously aware derive from Danish. These items are used by the speakers to fill gaps in their American English lexicon, mainly within the domains of household and food, but they may also fulfill a function in adding “local flavor” as acts of ethnic identity and performance and maintenance of that identity, i.e., as a stylistic choice. Such a use does not presuppose precise awareness of the semantic field of the lexical item in question. Examples of lexical items that the speakers recognize as Danish are offered in Section 4.1. It is important to note that while some speakers may be aware of the linguistic origin of certain lexical items, others may not.
Covert Danish items are by definition those that are challenging both to elicit and to identify. These include items that the speakers themselves do not recognize as being Danish, as they were embedded in the everyday linguistic repertoire or home milieu. For example, in a joint interview, two participants discussed at length how they had been puzzled to not find the word skruk ‘a brooding hen’ in an English dictionary, discovering that it was a Danish lexical item only after one of them thought to look it up in a Danish dictionary (speakers 43F and 35F, recorded in 2013). In the covert category, we include lexical items that were part of the home domain, usually the grandparents’ home. At this point, such items hang in the balance between covert and overt status, because they have been subsequently identified as Danish lexical items—sometimes during the interview process, with participants contributing metalinguistic commentary such as, “We used to sit on something called a skammel when we talked to my grandmother. Could that be a Danish word?” (speaker 28F). Another speaker (43F) reports that she thought that her grandmother’s name was bedstemor and that it took some time before she found out that it meant ‘grandmother.’ Another interview (speakers 45F and 46F) rendered the use of kikkers to describe a toy viewing device from the grandparents’ house; this term was the only the speakers had for this particular item.
Another woman (speaker 28F) who, in her English translation of the finger-naming rhyme, said, “Do you know why you call it slikkepot? Because this is the one that you take the peanut butter out of the jar and slik it.” In this case, the speaker has assigned a very specified meaning to the Danish verb, probably supported by the similarity of the Danish verb slikke and the English verb ‘to lick.’
In such a situation of receptive bilingual language use, the Danish items would merely have been a local or personal way of speaking, not the remains of a migrant language. Such an implicit integration into the language use of the community might look like the beginning of a language shift variety, meaning that these linguistic traces could become part of a local way of speaking English, thus losing their affiliation to Danish. This has happened, e.g., in Wisconsin English where the Wisconsin German varieties have left some traces (Wilkerson, Livengood and Salmons, 2014), but also in European contexts. Braunmüller (2009) provides a synchronic view on what he calls “covert lexical codeswitching” (ibid: 61) between German, the Danish dialect South Jutish and Danish, respectively, and connects these instances of “covert lexical codeswitching” to the emergence of new languages. However, such a process seems to be very unlikely to happen in Sanpete County. Knowledge of overt Danish items is mainly restricted to the now older generation of speakers, and even among these, the knowledge is already scarce.
(2) I remember once when we were gathered around, all the sisters, and she [the great-grandmother] went to the shelf took the jar down, came and handed it to me and told me in Danish I mean she was speaking in Danish she told me to pass it- to pass one around to each one of the sisters but I just- all of a sudden I knew what she said and even though she said it in Danish.speaker 28F
There is, however, one feature that deserves further investigation with regard to Danish or Scandinavian influence on Utah English. The agreement marker ja [ʝɑ]’yes, yeah,’ is evident in everyday conversations. Further investigation will reveal if this feature can be traced to Scandinavian diaspora language use (Wilkerson, Livengood and Salmons, 2014). It could be the case that a feature such as ja ‘yes, yeah’ does not fill a gap, is not performative, but rather belongs to a class of “gesture-like” features that are not cognitively distinguished among the languages of the bilingual, leading to a possible merger of forms in the recipient language (Matras, 2009).
4.4 The Networks of Transmission
The communities that make up Sanpete County historically have been dense, multiplexical, and relatively self-contained. The participants who have been interviewed so far know each other, and many of them are related either genetically or through marriage. While a network analysis has not been the overarching impetus of the study, it is evident that elements of Danish language and culture have been transmitted among people who have direct family ties to Danish and Scandinavian immigrants, making this a promising avenue of investigation in the future. So far, there have been two major family networks involved in the study. One family network has yielded six participants; the other has yielded five. Furthermore, the two networks are connected through marriage. The families have been connected, in fact, for generations; even before the current generation, the grandparents and great-grandparents lived in the same neighborhood (the Danish neighborhood) and were part of the same dense, multiplexical networks. Several of the participants interviewed for the study live in homes that were built and inhabited by now-deceased Danish family members.
These two family networks were fruitful sources of Danish cultural and lexical information, especially the women who are part of these networks. As discussed in Section 3, it became apparent during our fieldwork visits to Sanpete County that women, and more specifically women who had spent time with their grandmothers, were those who had the most in-depth knowledge of Danish lexical and cultural items. The Danish lexical inventory that is still present with the speakers in Sanpete County depends, of course, “on their experience in different situational domains, with frequently used vocabulary in common domains the most persistent as the language dies” (Wolfram, 2002: 776). This implies that we, on the basis of the remnants of the Danish language in Sanpete County, can determine the domains where Danish was maintained the longest. These were obviously the traditional domains of women, i.e., food, household and child-caretaker-interaction. In these domains, Danish seems to have been a natural part of everyday life, and it is here that Danish language has been passed on from mothers and grandmothers to daughters and granddaughters within the dense community of Sanpete County and, more specifically, the two family networks that have been involved in our study. The important role of grandparents has been discussed by several researchers. For example, Boas (2009: 255) mentions that 95% of his Texas German respondents report having spoken Texas German with their grandparents, as opposed to 60% to their parents. He attributes this to changes in the socioeconomic factors (e.g., parents working outside the home) and fears of stigmatization, as well as increased exogamy. The presence of a non-English-speaking grandmother has a stronger effect on the grandchild’s non-English language use than the presence of a grandfather, as grandmothers are more likely to be caregivers and to live longer than grandfathers (Ishizawa, 2014).
Supporting data must be obtained in order to confirm, but the results so far indicate that relatively early settlement in the community may also have contributed to the perpetuation of Danish linguistic items. Interviews with a participant whose family members migrated relatively late, around the World War I era, demonstrate a lack of transmission to younger generations of the family, despite multiple family members who spoke Danish. This halt to the transmission process likely coincides with the overall lack of heritage language tranmission due to external factors such as gradual integration into the matrix community.7
At present, the language situation in Sanpete County with regard to Danish is probably best classified as “dormant,” according to the classifications offered on the Language Vitality Scale (Lewis, Paul and Simons, 2010). In 2010, the notion of a “dormant language” was added to the scale to account for heritage languages which still carry symbolic status within a community and serve as a means of self-identification for an ethnic group, even though there are no remaining speakers of the heritage language.
For many community members of Sanpete County, there is still an ethnic association with Danish and Denmark, and there is some heritage knowledge of Danish customs and traditions. However, the current oldest members of the community are most likely the last who will have any inherited (as opposed to acquired) linguistic knowledge of heritage Danish, and their knowledge is already scant. After this point in time, any Danish language elements most likely will have to be acquired formally.
Within the well-established frameworks of language shift and language death, the situation in Sanpete County can best be described as a rather traditional case of gradual language shift to interrupted intergenerational language transfer where the language proficiency correlates with different generations of speakers (cf. Wolfram, 2002: 766). The gradual intergenerational language shift by the Danish immigrants and their descendants has been supported by the official assimilation ideology of the Mormon Church, as discussed in Section 2.4 of this article. Due to isolation and population density, the Danes/Scandinavians in Sanpete County probably maintained both Danish language and Danish customs for a longer time than, for example, the Danes in Salt Lake City, but the assimilation pressure was still strong.
Given the external factors present in Sanpete County, the relatively low level of lexical items from Danish is hardly surprising: the literature on language contact and shift notes that in cases of intergenerational shift, a speaker’s goal is to learn the vocabulary of the target community, with any remaining words from the heritage language belonging to the domains of food and other cultural items (Thomason and Kaufman, 1988: 39); this is certainly the case in our investigation of Sanpete County. Rather than lexical items manifesting in the target language, it is expected that grammar and pronunciation are likely to be affected. These issues remain to be investigated in Sanpete County.
In the case of Sanpete County, “generation” is not the only factor relevant for the state of language shift. Gender seems to play a role, too, pushing the generational shift rate. Women kept Danish alive longer than their husbands and sons. This complies with the society of Utah, where men’s voices would be heard, while women were confined to interactions with other women, in their homes and immediate surroundings, where Danish/Scandinavian language use was an option. However, it should be noted that while our investigation does not show a correlation between male networks and the transmission of Danish linguistic items, it cannot be ruled out that such networks may have existed.
Although in many ways this is a traditional language shift scenario, our investigation illuminates an area and a population that has been overlooked and also over-simplified. Previous accounts of this population have viewed the situation of settlers in Sanpete County through a wide angle lens, without addressing the intricate and mixed relations between the various subgroups to the greater community, such as Scandinavians and women. It has been noted that such a treatment of women is common in Western u.s. history, considering the admonition: “We need to approach western women’s history, not through the filters of prescriptive literature or concepts of frontier liberation and oppression, but through the experience of the people who lived the history” (Armitage and Jameson, 1987, cited in Abbot, 2013: 31). The inclusion and active seeking out of women in our study has yielded an account of the transmission of Danish that otherwise would have been lost.
We postulate that the combination of covert and overt Danish linguistic elements, transmitted primarily through the female networks we have explored, are indicators of the final stages of the language shift process, after which any elements of Danish influence are likely to manifest only below the level of awareness within the local variety of English. As such, given its in-depth view of a small, tight-knit, and established community, this investigation adds to our overall knowledge about the last stages of language shift. In addition, the investigation of a community with these extra-linguistic parameters—including Danes, the settling of the American West, and the role of a dominant religion—helps to reveal historical and sociolinguistic information about a society and immigrant group that remains relatively unexamined within the broader context of American immigrant groups.
The authors express their sincere gratitude to the many people with ties to Sanpete County who have aided them with their research, especially those who have participated in interviews. We also appreciate the careful comments of the external reviewers for this issue.
Henrichsen Lynn Bailey George and Huckaby Jacob . 2006. English-Language Learning and Native-Language Maintenance among Danish Immigrants to Utah, 1850–1930. In Larsen Birgit F. and Chimenti Linda M. (eds.) Danish Culture Past and Present: The Last Two Hundred Years62–91. Decorah, Iowa: Anundsen Publishing.
Larsen Hanne Pico . 2006. Culture for Sale in Solvang, California: A Little Bit of Denmark, Disney, or Something Else? Danish Culture Past and Present: The Last Two Hundred Years/Proceedings of an international conference sponsored by the Danish American Heritage Society . Ames, Iowa: Danish American Heritage Society130–139.
Petersen Peter L. and Nielsen John M. . 1992. Peter Sørensen Vig – Danish-American Historian. In Larsen Birgit F. and Bender Henrik (eds.) Danish emigration to the u.s.a. Aalborg: Danes Worldwide Archives in collaboration with the Danish Society for Emigration History (Udvandrerhistoriske studier, 4): 124–141.
Utah would not become a state until 1896, but the present-day settlement by Europeans/Americans began in 1847.
The details of the journey from Scandinavia to Utah are fascinating in their own right, but are not relevant to the aims of this article. The most comprehensive account to date on the Mormon exodus from Scandinavia to Utah is by the historian William Mulder, Homeward to Zion: The Mormon Migration from Scandinavia (University of Minnesota Press 1957).
The American Indian languages most likely would have been at a border area between Southern Paiute and Ute varieties, which are collectively referred to as Ute (Marianna Di Paolo, personal communication, 24 March 2015).
Carter’s (2015) architectural investigation of Sanpete County, for example, concludes that the settlers “melded rather seamlessly” and that “church members existed as a multicultural society bound together by a common commitment to the building of the Kingdom of God. One wasn’t so much a New Yorker or a Dane as a Mormon” (ebook location 1736). However, he admits that there was variability in assimilation in the community, and notes that there was a great deal of individualism and holding onto their previous identity (location 2168).
While there were no active efforts during the fieldwork to elicit Swedish or Norwegian based linguistic items, it was sometimes the case that speakers discussed family members from these immigrant groups, as well. In this article and in our work in general, the focus is on Danish immigrants and their language and culture.
There are a few exceptions to this generalization about the role of men and women with regard to the use of Danish, however. For example, two of the men interviewed recalled Danish men in Manti who would use Danish expletives, including at the workplace. Another man recalled that male factory workers at a pea cannery used Danish amongst themselves.
Another explanation for the apparent lack of Danish language transmission during this era is the shift in patriotism and associated behaviors, including linguistic behavior, in the United States in reaction to the wars in Europe (for the Danish immigrants see Petersen and Nielsen, 1992: 134).