For eighteenth and nineteenth century Pennsylvania Dutch speakers, a variety of German was the language of their books, their newspapers, and their schools. Being far from the European homeland created a hegemonic shift in the linguistic lives of these early German Americans; they were adopting an American regional identity. Along with their shift in identities and in linguistic hegemony, structural aspects of the languages they used also changed: their written German was in contact with English and with their spoken Pennsylvania Dutch. In addition, the limitations of formal education in German at rural schools meant that the emphasis among most Pennsylvania Dutch was on the receptive knowledge of German and not on productive control of the language. In time, a variety of German called Pennsylvania High German emerged in the publications, writings, and schools of Pennsylvania. This article shares recent findings of a large corpus of written attempts at Pennsylvania High German by Civil War soldiers. It discusses both the structural aspects of their written language as well as the negotiation of their American regional identities through a multilingual lens at the first major moment of increased contact with outsiders.
When do we expect rural farmers to write? The question, posed today, seems odd in postindustrial America. In the nineteenth century, however, the question takes on new meaning. Farmers constituted the semi-literate.  It is a familiar drive among those with dense, isolated networks in preindustrial America to engage daily in face-to-face “kin work” – a term rooted in anthropology and defined as “the conception, maintenance, and ritual celebration of cross-household kin ties” (di Leonardo 1987: 442).  The American Civil War, 1861–1865, which pulled millions of Americans from their homes (often for the first time) and sent them to distant fighting posts throughout the nation, created the ideal situation for relying on kin work that goes beyond the usual visits and rituals of family life that had been a mainstay of their socialization since birth. For young people, caught in the prime of their own identification and withdrawn from their everyday kin work, writing comes to fore as the outlet of continued connection with their homes. Indeed, we know that many soldiers kept diaries and wrote letters home during the Civil War. Zimm (2012) estimates that 4500 letters were sent per day at the start of the War in 1861. His work draws on the 11,000 letters housed solely in the Wisconsin Historical Society’s Civil War collection. Many other organizations have substantial Civil War letter collections, including a digital (nationally focused) corpus of Civil War soldier’s letters, the Corpus of American Civil War Letters (Ellis and Montgomery 2014).
The entirety of these corpora and of the vast majority of scholarly analyses of Civil War writings has focused on Anglo-Americans. There are, however, a handful of examples of German-American writings during the Civil War: Wolf’s (2005) study of four diaries, Kamphoefner and Helbich’s (2006) letters written by German-Americans, and Keller’s (2007) letters from Chancellorsville, Virginia. These contributions address only the German-American experience of the then recent, i.e. mid-nineteenth century immigrants to the United States. Such examples are important for study – while we know much about the literary endeavors of German in the nineteenth century – we know little about the German of everyday people in the United States before it was specifically recorded for preservation (Elspaß 1998). On the other hand, the Pennsylvania Dutch had already been in the United States for about a century by the time of the Civil War, having emigrated from Central Europe before the American Revolutionary War (1775–1783). They spoke Pennsylvania Dutch. Louden (2016: 64, 70) provides strong evidence that immigrant dialects “coalesced” to form Pennsylvania Dutch in the second half of the eighteenth century, particularly for the children of German immigrants who reached adulthood between 1750 and 1780 and who were in close contact with each other in rural southeastern Pennsylvania. Unfortunately, their experiences in the Civil War have been largely mediated by out-group primary sources. Did they write diaries and letters during the Civil War and, if they remain, what language did they write in and what topics did they cover? This manuscript analyzes their writings during the American Civil War and highlights their regional sociolinguistic identities through their writings, instead of relying on out-group mediating sources.
The manuscript is organized as follows: Section 2 reviews the necessity of locating sociolinguistic identities historically, before introducing the corpora and contextualizing them (Section 3). Section 4 looks at the corpora from two angles: (1) what varieties are present in the ego-documents as a reflection of the historic sociolinguistic situation and (2) how sociolinguistic identities are constructed through the ego-documents. The final section (Section 5) summarizes the research.
2 Ego-documents in historical sociolinguistics
Ego-materials are written by everyday speakers of the language and are often unedited for publication. The benefits of incorporating them into the historical view of language use provides a global and holistic approach to the history of language use (Elspaß 2007, van der Wal and Rutten 2013). Structurally-motivated studies note that innovation in language use “from below” – i.e. by everyday speakers – is imperative for incorporation into linguistics as these writers are those “involved in ongoing socio-communicative verbal interaction” (Auer et al. 2015b: 283). As a result, historical sociolinguistics moves from a perspective based on a standard language and explores new text types and the heterogeneity in the history of languages (Hernández-Campoy and Conde-Silvestre 2015: 16). By looking at ego documents, we are looking at varieties that are close to speech and represent different registers from all of society (van der Wal and Rutten 2013: 1; Watts 2015: 5). Other, and perhaps more closely speech-related, sources of linguistic data on historical speech beyond the scope of this manuscript are those that reference actual speech in the form of official and court transcriptions, witness depositions, prose fiction, etc. (see Kytö and Walker 2003; Culpeper and Kytö 2010 for additional sources).
The use of ego-documents – in part due to their scarcity – is fairly new; only recently, for example, have ego-documents for English-writing members of Heritage German communities become a focus of research (Bagwell et al. Forthcoming). More often, however, researchers downplay the incorporation of ego-documents in linguistic inquiry simply because of “bad data” – the fact that historical written forms of semi-literate people do not necessarily show actual language use and, combined with their scarcity, represent fairly idiosyncratic formulations of attempts at the standard language. Sometimes these supposed shortcomings are presented as an apologia to the readership. To view language production in such a way, we ignore the heterogeneity of language users and the heterogeneity of the history of language use. The history of language does not always revolve around the formation of a standard language – in fact, this manuscript will show the exact opposite of that. Speakers use language to develop historical sociolinguistic identities that work for them and demonstrate who they are as they negotiate their identities in the world in a historical snapshot of time. Terming language data as “bad” is further misleading in that all linguistic data are limited, whether studying languages diachronically or synchronically – not every speech community has a linguist at its disposal, not every utterance a recording device to capture it. To combat these arguments, historical sociolinguistics must aim for “informational maximalism,” i.e. “the utilization of all reasonable means to extend our knowledge of what might have been going on in the part, even though it is not directly observable” (Janda and Joseph 2003: 37). To that end, historical sociolinguists must keep mining archives and other repositories for the establishment of ego-document corpora. This manuscript serves that endeavor; namely, its data are drawn entirely from the ever-growing digital archive: Heritage Language Corpora (http://www.heritagelanguagecorpora.org/).
Ego-documents, whether historical or contemporary, are products of social practice. As artifacts of an individual’s thoughts, they are spaces where individuals construct their “knowledge of the world” (DeHaan 2010: 108). Indeed, historical sociolinguistics is not only the study of language within a historical social context, but also encompasses the study of historical language use as “conscious or unconscious acts of identity and social distinction” (Auer et al. 2015a: 9). This manuscript shows not only what language variants emerged in the writing of the Pennsylvania Dutch, but also how they used every language in their verbal repertoires to negotiate sociolinguistic identities. The heritage languages they used to write about their heritage language identities provide a nuanced view of ego-documents during the American Civil War.
3 Socio-historic and sociolinguistic aspects of the present research
This section follows Elspaß’s (1998) model of foregrounding ego-documents with necessary socio-historic and sociolinguistic aspects.  In so doing, the materials are better contextualized for analysis.
Heritage Language Corpora (http://www.heritagelanguagecorpora.org/) is a growing website for ego-documents written by everyday members of heritage language communities in the United States. This manuscript focuses only on the Pennsylvania Dutch writings contained there. Three sets of ego-documents have already appeared in print elsewhere and are included in the corpora by permission. The Shuler Family Correspondence consists of nine letters written largely in English during the California Gold Rush from 1851–1855 (Parsons and Heimburger 1986), ten letters in Pennsylvania High German written before and during the American Civil War, and two letters written in English in 1863 (Parsons and Heimburger 1980). The letters are sets of correspondence from brothers Asa and Amandus Shuler whose family arrived in colonial Pennsylvania in 1712. The Stahl Letter is a single manuscript written in Pennsylvania High German in 1863 (Sauer 2006). Peter Stahl, a Mennonite from Snyder County, Pennsylvania, had the letter written back to his wife while he served in the Union Army during the Civil War. The Weidle Letters are four documents written in Pennsylvania High German from 1863–1864 (Quinter 1993). Private Benjamin Franklin Weidle wrote them back to his family in Myerstown, Pennsylvania while he served in the Union Army during the Civil War. For this article, I focus largely (though not solely) on the two sets of ego-documents not previously published: (1) the Haupt Diary and (2) the Shoemaker Letters.
Private Joel Haupt was recruited at Shamokin, Pennsylvania and enlisted in the 46th Regiment of Pennsylvania Volunteers, Company K in February 1864. He mustered out with the rest of the company in July 1865. He kept a daily diary of his actions during Sherman’s capture of Atlanta and March to the Sea campaign. It begins with his enlistment and ends with his muster out of service more than a year later.
Two brothers, William and Benjamin Shoemaker, were mustered into service of the 128th Regiment on August 15, 1862, which was recruited in Berks County, Pennsylvania. Privates Shoemaker then joined the Regiment to Washington, Arlington Heights and Fairfax Seminary by the end of August. Both William and Benjamin would later re-enlist for the duration of the war. The 128th Regiment is noteworthy for its participation at Antietam in September 1862. Both brothers wrote a total of fifteen letters home to family between 1862 and 1865. I am indebted to the owners of the documents (both in private collections) for allowing me to work with them.
The Pennsylvania Dutch are an American ethnicity, which traces its origin to the 60,000–100,000 immigrants to “Greater Pennsylvania” from the Rhineland and Palatinate regions of Central Europe in the eighteenth century (Bronner and Brown 2017: 1). Religious wars, political unrest, and widespread famine made Central Europe in the late seventeenth and early eighteenth centuries an unwelcome place. It is no surprise that religious freedom-seeker William Penn specifically targeted this area to recruit farmers and the religiously oppressed for his newly acquired “Holy Experiment” in the American colonies; Penn succeeded in luring farmers. The larger group consisted of members of Lutheran and Reformed churches. Smaller groups were made up of the Amish, Mennonites, and other sectarians, who often faced religious persecution in the European homeland. When they settled southeastern Pennsylvania between the Blue Mountains and Philadelphia, they adapted customs from the old country to the new colonial environment. They became the Pennsylvania Dutch – ‘Pennsylvania’ from their destination and ‘Dutch’ from the then-contemporary English word applied to anyone from Central Europe (Louden 2016: 2).
Historians note the many contributions of the Pennsylvania Dutch to the Civil War, and also note the often severe prejudice against them (Valuska 2010). Thirty-one regiments in the Pennsylvania Volunteer Infantry had a strong Pennsylvania Dutch majority, though they were often integrated into larger units with the “English.” The Pennsylvania Dutch were primarily Democrats, thinking of the Lincoln-helmed war as a Republican effort. As rural farmers, they felt more inclined to protect their own property and way of life over a unified notion of a “Union.” In fact, although many outsiders considered the Pennsylvania Dutch to be wholly apolitical, they were not uninterested in national issues (Louden 2016: 85). They were bilingual, using both English (as evidenced by loanwords into their Pennsylvania Dutch and Pennsylvania High German). Although the sectarian Pennsylvania Dutch, citing religious tenets of pacifism, usually paid for exemptions in military service, they were not completely isolated from the war, as letters and oral tradition recount their interactions with Union and Confederate soldiers (Yoder 1962). The non-sectarian Pennsylvania Dutch were a visible contingent of soldiers fighting for the Union effort. Some viewed the Pennsylvania Dutch faction as exemplary. Others were not as favorable to the Pennsylvania Dutch, calling them a “queer lot,” whose women had “hard features” with “dirty-looking” children (Keller 2010: 59).
As a means of reconciling this paradoxical stance – on the one had valuing the personal and the local and on the other being devoted to the patriotic cause of national preservation – historian Steven Nolt (2002) has characterized the Pennsylvania Dutch as embodying peasant republicanism, an ideology derived from elements of German Pietism, which valued freedom of an individual’s right, unaffected by forced change from the outside. Peasant republicanism is characterized as “endors[ing] a collective self-interest derived from a strong local base. It could produce seemingly positive subjects who compliantly yielded to hierarchies of merit, but its advocates actually based their actions on political principles that could also evoke stiff opposition and vigorous protest” (Nolt 2002: 31). The Pennsylvania Dutch showed their patriotism through the expression of their distinctive identity, contained rurally and regionally. They maintained their language and cultural identity not in opposition to ideals of the nation-state, but rather in complete consonance with the right to individual freedoms advocated by the United States.
3.3 Language contact and variation
The arrival of many tens of thousands of immigrants from Central Europe brought many German dialects to Greater Pennsylvania before the Revolutionary War. The children of those immigrants to rural southeastern Pennsylvania, having reached adulthood between 1750 and 1780, created a language that was similar to the dialects of the Palatinate region, but not exactly the same as any of them (Louden 2016: 64). The resulting language was Pennsylvania Dutch. Pennsylvania Dutch was primarily an oral medium, though it did appear in print as early as 1794 in the Unpartheyische Readinger Zeitung (Louden 2016: 101). The languages available to the Pennsylvania Dutch – Pennsylvania Dutch, English, and European German – were used in socially constructed domains, such that English was used with outsiders and was the language of most written communication in the late nineteenth century (Donner 2008: 200). Although most of the Pennsylvania Dutch could read and write German in the eighteenth century, very few possessed the “ability, need, or inclination” to use German in the nineteenth century (Louden 2016: 134). 
Although on the surface, the parochial schools operated by the Pennsylvania Dutch were teaching German, there were several limitations on the schooling that their children received (Louden 2016: 155). Weakened ties to the European homeland, contact with English and (even more so) Pennsylvania Dutch, and schooling that did not amount to much face-to-face interaction resulted in a shift from German to a new variety: Pennsylvania High German. Pennsylvania High German is characterized by loans from English, nonstandard German forms as a result of influence from Pennsylvania Dutch, and – when compared to contemporary standard German – is reflective of an archaic, historical variety (Louden 2016: 130).
Early ABC books and grammars published locally frequently used “German-American” terms (Wood 1945). Muhlenberg’s Dictionary (1812) includes an appendix of English loans “[b]ecause of the constant contact with English we use words in conversation, without knowing whether they are English or German.” The examples include those spelled exactly as in English (Patchwork, Security), those with adapted spelling to German pronunciation (Dschäl ‘jail’ and Tscherrybaunz ‘cherry bounce [a cordial]’), and English words with German morphology (whippen ‘to whip’ and schwampig ‘swampy’). In the only scholarly work solely on Pennsylvania High German, Wood (1945) focused primarily on the maintenance of archaic forms in Pennsylvania High German and measured its phonetics and morpohosyntax against the norm of the standard language.
Some nineteenth-century newspapers from the Pennsylvania Dutch area include examples of Pennsylvania High German, notably the Reading Adler (Stark 2017). Figure 1 printed in the Alt Berks newspaper on March 26, 1844 shows the influence of English and Pennsylvania Dutch resulting in Pennsylvania High German. The advertisement is an offer of $5 for the capture of anyone who damaged Fensen (‘fences’ from English) or stole produce (Welschkorn ‘corn’ from Pennsylvania Dutch and Grundberen ‘potatoes’ from Pennsylvania Dutch) from the named farmers’ Lotten (‘lots’ a loanshift from Palatine German).
Figure 2 again shows similar borrowings from English resulting in German that the Americans could understand. It is a humorous advertisement for a Schutingmätsch (‘shooting match’ from English).
Importantly, Pennsylvania High German should not be viewed solely as a deviant from the standard and limit the analysis of it solely to comparisons with European German. Pennsylvania High German was a language contextually important for Pennsylvania and the regional identities of the Pennsylvania Dutch. For as, Mattheier (2010: 353) warns: the standard language should not be the “genuine teleological goal of any historical language development.” Moreover, by the second half of the nineteenth century, source material from Europe in German had diminished considerably.
Pennsylvania lacked a comprehensive public education system and was the last of the northern states to adopt one with the Public School Act of 1834. The creation of comprehensive public schools meant, to the Pennsylvania Dutch, that their parochial German schools were under threat and they largely refused to adopt public curriculum. Aware of the situation, the state threatened the parochial schools by withholding financial aid in 1854. Still, in 1866, just after the end of the American Civil War, there were 23 school districts in the core Pennsylvania Dutch area of Central and Southeastern Pennsylvania, which refused to accept a common school curriculum (Stine 1938). As a result, many of the Pennsylvania Dutch in the nineteenth century would have had a curriculum centered on learning penmanship and becoming literate in German.
As noted earlier, those writing in Pennsylvania High German were not the educated upper classes among the Pennsylvania Dutch. They had obviously had some schooling and education in German as evidenced from their penmanship. However, the content of the writing shows in some instances that the writers were more comfortable with Pennsylvania Dutch and thus wrote mostly Pennsylvania Dutch in the Kurrent script with some standard German words. Each of the writers under study here were raised in farming families, e.g. Life-Mode 1 (cf. footnote 1). After their enlistments in the military, each entered a trade, e.g. Life-Mode 2. As such, the ego-documents are definitely “from below,” i.e. how the majority of Pennsylvania Dutch would write, if they were taught how to in their parochial schools. In fact, Louden (2016: 156–158) includes examples of writing in “Dutchified German,” essentially “dressed up” German that is actually Pennsylvania Dutch written in perfect Kurrent script.
Throughout this paper, I stress Milroy’s (2001: 532) dictum: “Uniformity … is a property of the language system, not of the speakers”. Not every borrowing into Pennsylvania High German from either Pennsylvania Dutch or English was done by every writer. Not every nuance of word order was done with uniform precision. When looking at the writing of the semi-literate Pennsylvania Dutch who wrote in their attempts at Pennsylvania High German, there is considerable variation between speakers and within individual speakers. Those aspects, however, do not deter from finding commonalities and unearthing the sociolinguistic identities being expressed. Especially in a monolingual bias of literate writers, one expects uniformity and the “errors” will lead to linguistic analyses. In looking at ego-documents and the ego-documents of multilinguals writing in a language of non-dominant society, the uniformity – the (socio)linguistic realities that persist in their writing – will be the most telling for historical sociolinguistic analyses. Moreover, this situation represents one of dismantling the European standard for their own purposes in new contexts. They are using Pennsylvania High German (with influence from Pennsylvania Dutch and English) for kin work at a very stressful time in their lives – not only in terms of identity and language contact, but also in terms of psychological and physical stress. In many historical sociolinguistic analyses, the formation of the standard receives the greatest attention, in this case, however, the Pennsylvania Dutch mold the languages in their repertoires to communicate with kin and, as a result, the language used is one which seeks to fulfill kin work and not necessarily one which strives toward standard-like norms.
4 Written “Dutch” from below
This section examines the Pennsylvania Dutch contents of the Heritage Language Corpora and highlights two previously held assumptions that now need additional qualification. The first, namely that the Pennsylvania Dutch wrote in English, is true to a large extent of the known materials. In recent years, many more have come to light that force researchers to question the existence of additional writings in attempts at Pennsylvania High German. The second, namely that writing in the language of non-dominant society as a way of fending off assimilation is not supported by the writings of the Pennsylvania Dutch, nor by historical evidence. As more ego-documents emerge, the writings of everyday people in the only language they were literate in will demonstrate the sociolinguistic identities they negotiated.
4.1 Muddling through?
Keller and Valuska (2010: xvi–xvii) in their influential work on the Pennsylvania Dutch in the Civil War note that “many Dutch soldiers and civilians, more familiar with written English than with written German, wrote their letters and diaries in English – which they could muddle through better than the High German they heard only in church or read only in newspapers.” However, this statement glosses over the rich variation that exists in the writings of the nineteenth century Pennsylvania Dutch. To suppose a strict delineation between “High German” on the one side and English on the other, is to ignore the interwoven qualities of the languages available to the Pennsylvania Dutch. Allen (2015) posits the existence of a continuum for non-standard writing and Macha (1997) develops a typology specifically for German immigrant writing based on the linguistic competence displayed in their writing in both English and German. He delineates three levels of competence: rudimentary (level 1, indicated by signatures on documents), restricted or marked (level 2, writing that shows heavy dialect interference or “phonographic” spelling), and elaborated or unmarked (level 3, writing that is close to the standard). The German and English writings in the Pennsylvania Dutch content of the Heritage Language Corpora show considerable variation in competency depending on the writer. Most, however, would fall into level 2 in both their English and Pennsylvania High German writings. Of course, there is even considerable variation within that single level ranging from near rudimentary competency in the written language to near full competency. The attempts at Pennsylvania High German may in fact be largely Pennsylvania Dutch with German morphology or Pennsylvania Dutch with English lexemes. Especially here, I provide the caveat over terminology. These writers most likely were not “attempting” to write in German at all. Louden (2016: 156–158) aptly shows that some writing of the nineteenth century Pennsylvania Dutch is “dressed up” Pennsylvania Dutch – what he refers to as “Dutchified German.” In the end, I feel that the writers were seeking a communicative goal. The script they used was the one taught in their parochial schools. The language they used was closer to Pennsylvania Dutch with some formulaic expressions from German letter-writing manuals. Of course, there still is a wide variety in the resulting written ego-documents.
The English writing can be heavily influenced from either Pennsylvania Dutch or Pennsylvania High German. Having three languages in their repertoire, makes this historical contact situation complicated and interesting for understanding the range of non-standardness that can exist in the writings of rural farmers. There are many examples of written English, usually by the more educated Pennsylvania Dutch. Their English, however, is not completely void of language contact, i.e. through “imposition” of linguistic elements in the resulting writing.  William J. Reichard, also in the 128th Regiment with the Shoemaker brothers, has several examples of lexical borrowings from Pennsylvania Dutch in his English letters:
The English letters of the Shuler Family Correspondence also displays several lexical items from Pennsylvania Dutch:
Additionally, their English writings show contact with Pennsylvania Dutch on the resulting English writing. In word order, Reichard uses more Pennsylvania Dutch structure than English:
Depending on the English literacy of the individual, the resulting English shows heavy influence from Pennsylvania Dutch. Another Shuler Family letter written in English hints at the lower proficiency of the writer in English due to the prevalence of contact-induced elements from Pennsylvania Dutch, namely word final devoicing :
Some of the writing shows complete code-switches from English to Pennsylvania High German or Pennsylvania Dutch or a mixture of both:
In looking at these attempts at German, the result was what has become known as Pennsylvania High German. This term, however, is largely defined by its distance from the standard, rather than its existence as a language in and of itself. The most telling feature of these writings is the precise handwriting in Kurrentschrift, an older German script. This visual cue would force some to believe that the writing is German. Parsons and Heimburger (1980: 98) categorize the Schuler Family Correspondence as “the real Pennsylvfaanisch Deitsch [Pennsylvania Dutch] which can still be heard in the Pulverdahl [Powder Valley] any warm Summer day.” I disagree with their assertion. Closer inspection of the Shuler Family writing, while showing some influence from Pennsylvania Dutch, is closer to German:
Another letter within the same collection, but written by a different person shows inflection of verbal and adjectival morphology, but it also shows Pennsylvania Dutch word order and borrowings from English like Morketsch ‘mortgage’ and gesut ‘sued’:
The newest additions to the Heritage Language Corpora show just as much variation in the writing. The Haupt diary might be best classified as a solid level 2 in Macha’s (1997) typology with occasional dips to level 1. Although written in Kurrentschrift with an obvious nod to German that he was exposed to in his early years at school, there is lexical borrowing from English:
Haupt clearly has knowledge of Pennsylvania High German/European German with forms like morgen ‘tomorrow’ instead of Pennsylvania Dutch mariye in example (8d), but Pennsylvania Dutch appears in his plural auxiliary hen instead of German haben in example (8e) (cf. Frey 1985: 53).  Pennsylvania Dutch phonology is applied to the phonographic orthography inschpektion ‘inspection’ in example (8d), krasrot ‘crossroad’ in example (8f), and Sauß ‘south’ in example (8h).  Examples of Pennsylvania High German morphology on English loans are found in (8f) gemet ‘met’, (8g) gedietelt ‘detailed’, and (8h) gestert ‘started’ (cf. Frey 1985: 58).
There is not only lexical borrowing from English, but also imposition from Pennsylvania Dutch:
These examples show lexical imposition from Pennsylvania Dutch mir ‘we’ in (9b) and (9c), strehl ‘combs’ in (9a), and ob ‘before’ in (9b), as well as lexical calques from English in Pennsylvania Dutch like rigelweg ‘railroad’ in example (9c). Examples (9a) and (9c) show delabialisation that exists in Pennsylvania Dutch: leit (German Leute) ‘people’, zu rick (German zurück) ‘back’, brick (German Brücke) ‘bridge’.
The Shoemaker letters are much different than the Haupt diary. The letter collection was written by a number of people within the Shoemaker relation and friends, all living in the Berks County, Pennsylvania area. The diary entries are rarely more than a sentence or two, but the Shoemaker letters are longer and as such provide numerous instances of linguistic points. I would place the Shoemaker letters (as a whole collection) to be in the higher end of level 2, depending on the writer. They are all written in Kurrentschrift, unless they overtly switch to English in the letter. The letters have more indication of knowledge of German, but show influence from both Pennsylvania Dutch and English. The following section examines the lexical, phonological, and syntactic influences from Pennsylvania Dutch and English that shaped the writers’ attempts at German and created their own type of Pennsylvania High German.
A common orthographical anomaly schücken ‘to send’ (10a) instead of schicken is commonplace in Pennsylvania High German. Pennsylvania Dutch lexical items like sel ‘that’ (10b) and eb ‘before’ (10c) are found in the letters. Additionally, English borrowings, similar to the examples from the Haupt Diary occur. krasen ‘to cross’ (10d) represents a lexical borrowing from English with Pennsylvania High German morphology, similar to Haupt Diary examples (8f), (8g), and (8h). gedruweld (8e) ‘troubled’ is a loanshift from Palatine German. Labilization of front rounded vowels is present in Pennsylvania Dutch forms like eich ‘you’ (10f). Additional Pennsylvania Dutch forms like mir ‘we’ and kat ‘had’ (10g) appear alongside Pennsylvania High German standards like haben, something that the diary does not do.
In addition to the lexical contact from Pennsylvania Dutch and English, the attempts at Pennsylvania High German in both the letters and the diary exhibit many morphosyntactic influences from Pennsylvania Dutch and English. Examples (11a) and (11b) contain purpose clauses that are common in Pennsylvania Dutch. Examples (11c) and (11d) have progressive forms. The frequency of which in Pennsylvania Dutch is shown to come from contact with English, but also have taken on a life of their own (cf. Brown and Putnam 2015). English-like word order is found in example (11e) with a topicalized prepositional phrase followed not by V2 order of Pennsylvania, but rather a structure that is more like English (cf. Page and Brown 2006). The use of the subjunctive for det in example (11f) is also unique to Pennsylvania Dutch (cf. Frey 1985: 76). Example 11f contains a typical three-verb cluster with IPP (Infintivus Pro Participio). This 312 (auxiliary verb + modal infinitive + lexical infinitive) is a feature found in both historical and modern Pennsylvania Dutch (Louden 2011).
The letters occasionally switch between Pennsylvania High German and English. The English exhibits influence from Pennsylvania Dutch and gives us a glimpse at what their English may have sounded like. For example, the following from a letter dated February 14, 1865 at the end of the war contains several examples of word final devoicing, as well as confusion over capitalization:
Such imposition from Pennsylvania Dutch (or maybe too Pennsylvania High German) on the written English of the Pennsylvania Dutch is something neglected in the historical research.  Some Pennsylvania Dutch did write in English, as noted earlier, but these individuals would be even more educated than those who learned the basics of Pennsylvania High German.
In sum, the diary and the letters show the range of linguistic creativity from multilingual individuals. I echo the findings of Lattey and Tracy (2001). In their study of a German immigrant’s letters in New Jersey, they found that her German contained influence from English if there was no German equivalent. However, in some situations the writer did not need to borrow as the German form was known to her in other instances. They conclude that this is not due to “language loss or confusion” but “making full use of the two co-activated language systems in the writer’s overall linguistic repertoire” (Lattey and Tracy 2001: 429). Not only do the letters, in particular, show us the individual’s linguistic repertoire, but also hint at the linguistic repertoire of many of the Pennsylvania Dutch, since the recipients needed all codes (English, Pennsylvania Dutch, and Pennsylvania High German) at their disposal to understand the meaning. The idea of “muddling” through one of the languages in their repertoire is clearly not supported by the evidence in the preceding section.
Sebeok’s (1943) analysis of diaries and narratives of German in the nineteenth century found that “[t]he German language served as a fence to keep out American culture and as a weapon to ward off assimilation” (280). Indeed, then-contemporary historical accounts have shown the reticence of the Pennsylvania Dutch from participating in the American Civil War and being apolitical or uninterested in national affairs in general (Louden 2016: 85). Here especially is the value of historical sociolinguistics in uncovering the changing sociolinguistic identities of rural farmers through their ego-documents. They are writing at a time of national conflict and are confronted with people who are outside of their usual dense social networks in rural Pennsylvania. By using their ego-documents, we are able to re-examine the history of the Pennsylvania Dutch and cast them in a very different historical light. For this analysis, I rely on Pavlenko and Blackledge’s (2004) theory of the negotiation of multilingual identities. For them and my purposes here, the negotiation of identity is dynamic, multiple, and discursive. Sociolinguistic identities are contextually-based, competing identities are differently valued, and, those identities may be negotiable, non-negotiable, or not negotiated. Importantly, their work has been employed historically as well (Pavlenko 2004). This manuscript draws on just such contested multilingual spaces – the rural Pennsylvania Dutch speakers entered an American war with the English-speaking majority from a variety of ethnic backgrounds.
Haupt makes extra reference to his review from Sherman himself on December 30, 1930 1864 over a week after he participated in the siege of Savannah:
In the following year, upon their return to Washington, Haupt recounts presenting an officer with a gift in this entry from May 26, 2026 1865:
Earlier entries reveal that it was Private Haupt himself, who wrote for the sword for (presumably) Thomas Alderson, who was made first lieutenant just days before.
The Shoemaker brothers make even more overt pro-Union statements. William wrote in an undated letter that he has no wants from home and invokes Uncle Sam as the patriotic symbol of support for his “boys”:
In the same letter, he repeats those sentiments by switching to English:
Importantly, his switches between English and Pennsylvania High German show the changing nature of these languages in his sociolinguistic identification. When writing of his mother and home, he invokes Uncle Sam; when writing only about Uncle Sam “in Washington” – far removed from his rural Pennsylvania home, he switches to English. These sentiments show that the Pennsylvania Dutch were not unaware of, unaffected by, or uninterested in national issues. They were devoted to national causes, including fighting for the preservation of their individual freedoms (cf. the discussion on “peasant republicanism” above).
As is true of many enlisted and faced with the very real threat of death, they wish many times in the letters that they could go home. At the start of their enlistment while stationed at Maryland Heights, they wrote home:
Beleaguered by the psychological and physical strain – they wish they could return home. Just before their regiment’s battle at Chancellorsville in May 1863, the brothers wrote home in a letter dated April 7, 1863 about the trials of the war on their regiment and on a neighboring friend, Private Charles Geist, who (in their eyes) died more justly as a result of disease than on the battlefield:
William Shoemaker writes of illness in an undated letter about Felde (Valentine) Geist. The uncertainty in times of war among the Pennsylvania Dutch are by no means unique, many felt unprepared and perhaps unwilling to fight in the war:
More overtly, the brothers write about their pride in Lincoln, the Republican president with politics differing from many of the Pennsylvania Dutch. They negotiate how being removed from home presents situations in which they are “gentlemen” and privy to valuable goods befitting a president.
Another letter from Summit Point on April 16, 1864, following the assassination of Lincoln, shows their sadness for the event.
His sentiments earlier in the year in February 4, 1865 from Camp Fairview were notably pro-Union in his English-written letter home concerning the capture of Lincoln’s assassin John Wilkes Booth. Again, the negotiation of identities is often conflicting and can change throughout one’s narrative, across time in a war (for example), and across the lifespan.
Again, when writing of home and individuals known back home, the writing is in Pennsylvania High German (example 19). As the letter writers invoke Lincoln and the Union, they switch to English (examples 21–22). Another letter contains similar sentiments about Booth. Here the switch between languages is even more pronounced:
Writing about Lincoln is in English, invoking home and greeting everyone there is in Pennsylvania High German, and cheering for the union necessitates a further switch back to English. A similar sentiment is found years earlier on December 12, 1863 letter from Benjamin:
Notably the most favorable sentiments toward the Union are written in English and not in an attempt at Pennsylvania High German. They were already patriotic and invested in the ideals of the nation, so the language switch here is likely due to increased talk about the nation and the war that they would have encountered with the English-speaking majority. Unfortunately, the effort behind kin work seemingly ended when they returned home from the war. No additional writings of these individuals are known to exist. When they returned to their rural homes, kin work could resume as it had done before the war. Although their identities were in a constant state of negotiation, the kin work that they sought out through their writing was no longer needed as a venue of the negotiation of their identities. The data in this section show that the Pennsylvania Dutch were committed to national issues. They did not use their writings in a language other than English to ward off assimilation, for they were already fully American. Their expressions of patriotism emerge in all the languages available to them as a reflection of their commitment to a peasant republican ideology.
This analysis shows that although the Pennsylvania Dutch may have often been portrayed as apolitical and isolated, these ego-documents reveal great pride in serving not only President Lincoln, but also in the pride toward the Union. The Civil War was a complex time in the lives of these men. By the nineteenth century, the Pennsylvania Dutch would have had increased contact with the English-speaking majority. Their identities as Americans would have been firmly entrenched. The nation was growing, railroad travel was increasing, and the Civil War challenged the role of the nation in lives of Americans. The resulting reversal back to regional identities immediately after the Civil War, notably through American literary regionalism movement, characterized the second half of the nineteenth and early twentieth centuries. Together, both collections show the experiences of Pennsylvania Dutch soldiers as intimately linked to the formative processes at work in nationhood – formative processes that have deep resonations even today. America was built by Americans, yes, but Americans bound by strong and tenacious ethnic ties – and, moreover, sometimes arguing for the preservation of that nation in a language other than English.
This analysis has also shown that the language used in those negotiations of identities manifested as their entire verbal repertoire. They created their own Pennsylvania High German – a language that deviated from a standard to match and facilitate the changing linguistic capabilities of the Pennsylvania Dutch removed from their German-speaking homeland. The borrowings from English and impositions from Pennsylvania Dutch do not represent lapses or a confusion of tongues, rather a balance of using their multilingual abilities to relate the realities of their experiences to kin back home.
Not only does this manuscript represent the tip of research into the possibility of exploring these writings within their historical context, but it allows the stepping stone for more work in social and linguistic areas of historical research. Are the ego-documents in this manuscript and the other five examples previously published the tip of the proverbial iceberg? The ego-documents in this analysis were both relegated to an attic or chest for safekeeping. For their descendants who came across these ego-artifacts, the writings presented first the obstacle of the writing system used, and then the peculiarities of the language, incorporating Pennsylvania High German, Pennsylvania Dutch, and regional American English. The result, once deciphered, reveals an amazingly neglected aspect of a shared American history. Perhaps – invoking the idea of “kin work” again – more Pennsylvania Dutch wrote home in their heritage language than we are currently aware.
I am grateful for the feedback from two anonymous reviewers and participants of the 8th Workshop on Immigrant Languages in the Americas, especially Joe Salmons and Aneta Pavlenko. The usual disclaimers apply. The writing of this article was supported by the University of Wisconsin-Eau Claire Faculty Sabbatical Leave Program and the Kreider Fellowship through the Young Center for Anabaptist and Pietist Studies at Elizabethtown College.
Allen, Barbara. 2015. A non-standard standard? Exploring evidence from nineteenth-century vernacular letters and diaries. In Anita Auer, Daniel Schreier & Richard J. Watts (eds.), Letter writing and language change, 202–222. Cambridge: Cambridge University Press.10.1017/CBO9781139088275.012Search in Google Scholar
Auer, Anita, Catharina Peersman, Simon Pickl, Gijsbert Rutten & Rik Vosters. 2015a. Historical sociolinguistics: The field and its future. Journal of Historical Sociolinguistics 1(1). 1–12.10.1515/jhsl-2015-0001Search in Google Scholar
Auer, Anita, Daniel Schreier & Richard J. Watts. 2015b. Epilogue: Where next? In Anita Auer, Daniel Schreier & Richard J. Watts (eds.), Letter writing and language change, 277–287. Cambridge: Cambridge University Press.10.1017/CBO9781139088275.016Search in Google Scholar
Bagwell, Angela, Samantha Litty & Mike Olson. Forthcoming. Wisconsin immigrant letters: German imposition on Wisconsin English. In Raymond Hickey (ed.), Keeping in touch. Familiar letters across the English-speaking world. Amsterdam: John Benjamins.Search in Google Scholar
Bronner, Simon J. & Joshua R. Brown (eds.). 2017. Pennsylvania Germans: An interpretive encyclopedia. Baltimore: Johns Hopkins University Press.Search in Google Scholar
Brown, Joshua R. & Michael Putnam. 2015. Functional convergence and extension in contact: Syntactic and semantic attributes of the progressive aspect in Pennsylvania Dutch. In Janne Bondi Johannessen & Joseph Salmons (eds.), Germanic heritage languages in North America, 135–160. Amsterdam & Philadelphia: John Benjamins.10.1075/silv.18.06broSearch in Google Scholar
Culpeper, Jonathan & Merja Kytö. 2010. Early modern English dialogues: Spoken interaction as writing. Cambridge: Cambridge University Press.Search in Google Scholar
DeHaan, Kathleen A. 2010. Negotiating the transnational moment: Immigrant letters as performance of a diasporic identity. National Identities 12(2). 107–131.10.1080/14608940902929789Search in Google Scholar
Di Leonardo, Micaela. 1987. The female world of cards and holidays: Women, families, and the work of kinship. Signs: Journal of Women in Culture and Society 12(3). 440–453.10.1086/494338Search in Google Scholar
Donner, William W. 2008. “Neither Germans nor Englishmen, but Americans”: Education, assimilation, and ethnicity among nineteenth-century Pennsylvania Germans. Pennsylvania History 75(2). 197–226.10.2307/27778830Search in Google Scholar
Elspaß, Stephan. 1998. Bridging the gap: Fixed expressions in nineteenth-century letters of German immigrants. Paper presented at Defining tensions: A fresh look at Germans in Wisconsin, University of Wisconsin-Madison, 15–17 October.Search in Google Scholar
Elspaß, Stephan. 2007. Sprachgeschichte von unten: Untersuchungen zum geschriebenen Alltagsdeutsch im 19. Jahrhundert. Tübingen: Max Niemeyer.Search in Google Scholar
Frey, J. William. 1985. A simple grammar of Pennsylvania Dutch. Lancaster, PA: Brookshire Publications.Search in Google Scholar
Hernández-Campoy, Juan Manuel & Juan Camilo Conde-Silvestre. 2015. Assessing variability and change in early English letters. In Anita Auer, Daniel Schreier & Richard J. Watts (eds.), Letter writing and language change, 14–34. Cambridge: Cambridge University Press.10.1017/CBO9781139088275.003Search in Google Scholar
Højrup, Thomas. 2003. State, culture, and life-modes: The foundations of life-mode analysis. Hants, England & Burlington, VT: Ashgate.Search in Google Scholar
Janda, Richard D. & Brian D. Joseph. 2003. On language, change, and language change - or, of history, linguistics, and historical linguistics. In Brian D. Joseph & Richard D. Janda (eds.), The handbook of historical linguistics, 3–180. Oxford: Blackwell.10.1002/9780470756393Search in Google Scholar
Kamphoefner, Walter D. & Wolfgang Helbich (eds.). 2006. Germans in the Civil War: The letters they wrote home. Chapel Hill: The University of North Carolina Press.Search in Google Scholar
Keller, Christian B. 2010. The Pennsylvania Dutch and “the hard hand of war”. In David L. Valuska & Christian B. Keller (eds.), Damn Dutch: Pennsylvania Germans at Gettysburg, 56–73. Mechanicsburg, PA: Stackpole Books.Search in Google Scholar
Keller, Christian B. & David L. Valuska. 2010. Introduction. In David L. Valuska & Christian B. Keller (eds.), Damn Dutch: Pennsylvania Germans at Gettsyburg, xiii–xix. Mechanicsburg, PA: Stackpole Books.Search in Google Scholar
Kytö, Merja & Terry Walker. 2003. The linguistic study of early modern English speech-related texts: How ‘bad’ can ‘bad’ data be? Journal of English Linguistics 31(3). 221–248.10.1177/0075424203257260Search in Google Scholar
Lattey, Elsa & Rosemarie Tracy. 2001. Language contact in the individual: A case study based on letters from a German immigrant in New Jersey. In P. Sture Ureland (ed.), Global eurolinguistics: European languages in North America - migration, maintenance and death, 413–433. Tubingen: Max Niemeyer.Search in Google Scholar
Louden, Mark L. 2011. Synchrony and diachrony of verb clusters in Pennsylvania Dutch. In Michael Putnam (ed.), Studies on German-language islands, 165–186. Amsterdam & Philadelphia: John Benjamins.10.1075/slcs.123.08louSearch in Google Scholar
Louden, Mark L. 2016. Pennsylvania Dutch: The story of an American language. Baltimore: Johns Hopkins.Search in Google Scholar
Macha, Jürgen. 1997. Rückbindung und Neuanfang: Zur Schreibsprache deutscher Amerika-Auswanderer im 19. Jahrhundert. Rheinisch-westfällische Zeitschrift für Volkskunde 42. 203–221.Search in Google Scholar
Muhlenberg, Henry. 1812. English-German and German-English dictionary. Lancaster, PA: William Hamilton.Search in Google Scholar
Nolt, Steven M. 2002. Foreigners in their own land: Pennsylvania Germans in the early Republic. University Park, PA: The Pennsylvania State University Press.Search in Google Scholar
Page, B. Richard & Joshua R. Brown. 2006. From V2 to SVO? A quantitative analysis of word order in Pennsylvania German. Paper presented at the Germanic Linguistics Annual Conference 12, University of Illinois at Urbana-Champaign, 28 April.Search in Google Scholar
Parsons, William T. & Mary Shuler Heimburger. 1980. Shuler family correspondence. Pennsylvania Folklife 29(3). 99–113.Search in Google Scholar
Parsons, William T. & Mary Shuler Heimburger. 1986. Letters and reports from Pennsylvania Germans in the American West. Pennsylvania Folklife 35(2). 50–65.Search in Google Scholar
Pavlenko, Aneta. 2004. “The making of an American”: Negotiation of identities at the turn of the twentieth century. In Aneta Pavlenko & Adrian Blackledge (eds.), Negotiation of identities in multilingual contexts, 34–67. Clevedon: Multilingual matters.10.21832/9781853596483-004Search in Google Scholar
Pavlenko, Aneta & Adrian Blackledge. 2004. New theoretical approaches to the study of negotiation in identities in multilingual contexts. In Aneta Pavlenko & Adrian Blackledge (eds.), Negotiation of identities in multilingual contexts, 1–33. Clevedon: Multilingual Matters.10.21832/9781853596483Search in Google Scholar
Quinter, Edward E. 1993. Letters home to Myerstaun (1863–1864). Der Reggebogge: Journal of the Pennsylvania German Society 27. 1–5.Search in Google Scholar
Reichard, William J. 1958. Civil War letters of William J. Reichard 1862–1863. Proceedings of the Lehigh County Historical Society 22. 137–281.Search in Google Scholar
Sauer, Walter. 2006. “Ein grus an dig meine liebe frau”: A civil war letter from a Pennsylvania German soldier to his wife. In Joshua R. Brown & Leroy T. Hopkins (eds.), Preserving heritage: A festschrift for C. Richard Beam, 67–76. Yearbook for German-American Studies. [supplemental issue 2].Search in Google Scholar
Stark, Joel Reuben. 2017. Sociolinguistic and structural aspects of Pennsylvania Dutch in the nineteenth century: An analysis of the Reading Adler and the dialect letters of Dr Frank R. Brunner. Madison, WI: University of Wisconsin-Madison dissertation.Search in Google Scholar
Stine, Clyde S. 1938. Problems of education among the Pennsylvania Germans. Ithaca, NY: Cornell University dissertation.Search in Google Scholar
Valuska, David L. 2010. The Pennsylvania Dutch as first defenders. In David L. Valuska & Christian B. Keller (eds.), Damn Dutch: Pennsylvania Germans at Gettysburg, 44–55. Mechanicsburg, PA: Stackpole Books.Search in Google Scholar
van der Wal, Marijke J. & Gijsbert Rutten. 2013. Ego-documents in a historical-sociolinguistic perspective. In Marijke J. van der Wal & Gijsbert Rutten (eds.), Touching the past: Studies in the historical sociolinguistics of ego-documents, 1–17. Amsterdam & Philadelphia: John Benjamins.10.1075/ahs.1Search in Google Scholar
Watts, Richard J. 2015. Setting the scene: letters, standards and historical sociolinguistics. In Anita Auer, Daniel Schreier & Richard J. Watts (eds.), Letter writing and language change, 1–13. Cambridge: Cambridge University Press.10.1017/CBO9781139088275Search in Google Scholar
Wolf, Andrea. 2005. Kriegstagebücher des 19. Jahrhunderts: Enstehung, Sprache, Edition. Frankfurt: Peter Lang.Search in Google Scholar
Yoder, Mary Elizabeth. 1962. Amish settlers and the Civil War. Christian Living 9(1). 28–29, 39–40.Search in Google Scholar
Zimm, John. 2012. This wicked rebellion: Wisconsin Civil War soldiers write home. Madison, WI: Wisconsin Historical Society Press.Search in Google Scholar
© 2019 Walter de Gruyter GmbH, Berlin/Boston