Variation and change in pronominal address in 19th and early 20th-century German private letters

1 Introduction

The pronominal system of address in use today in German originates in an innovation of the 18th century. Whereas addressing a single person with the pronoun Du (2nd person singular) had been the default form in private symmetrical communication (‘language of immediacy’, cf. Koch and Oesterreicher 2012) almost throughout the history of German, the Sie form of address (grammatically: 3rd person plural) was established as the dominant form of address for the polite ‘language of distance’ (cf. Simon 2003a: 93, 110–114). According to von Polenz (1999: 383), the binary Du versus Sie system quickly gained ground in the 19th century.

The present paper focuses on the pronominal form of address in private communication. Deplored by grammarians and philologists at the time, the use of the familiar form Du by children addressing their parents is reported to have spread rapidly in the first decades of the 19th century (Besch 1996: 109, 2003: 2617). So far, however, there is only anecdotal evidence for this and other findings. The aim of our paper is to provide a sound corpus-based investigation of the variation and change of pronominal address in 19th and early 20th century private letters, especially in letters to parents, but also to other family members and acquaintances.

After a brief overview of the variation and change of the pronominal address system in the history of German and a critical review of the relevant research literature (Section 2), we will present our data in Section 3. Our study draws on two historical corpora: a corpus of historical patient letters and a corpus of historical emigrant letters, with 2,748 private letters in total. Each letter was annotated with metadata about the individual writers, such as age, gender and writing experience, as well as information about the letters, such as date, addressee(s) and address pronouns used (see Section 4 for our methodology). Section 5 presents the results of our study. It shows and discusses the diachronic developments as well as social and individual factors influencing the variation of address pronouns in the patient letters (see Section 5.1) and the emigrant letters (see Section 5.2). The conclusion in Section 6 evaluates the findings against the background of the more general question of how the overall change can be explained. The results permit a differentiated picture of the development of the present-day pronominal address system in German.

2 Pronominal address in the history of German

Like many other European languages, present-day German has a dual pronominal address system, with the older du form (2nd person singular for one addressee) and the ihr form (2nd person plural for more than one addressee) as the default forms for symmetrical communication in private contexts (T) and the Sie-form (grammatically: 3rd person plural; for one or more addressees) as the polite variant (V).^[1] The 19th century is of particular interest for the study of variation and change of the pronominal address system in German, as – according to the research literature – it saw the change from a complex system with four (or five) pronouns to the present-day dual system. Figure 1 gives an overview of the development of the pronominal address system in reference to a single addressee throughout the history of German according to the current state of research.^[2]

Figure 1:

Pronominal paradigms used in reference to a single addressee in the history of German (from Simon 2003b: 86; cf. Simon 2003a: 93).

Starting from a single form (du, stage I), German has developed a complex pronominal address system, which reached its peak in the 18th and early 19th centuries. The only constant throughout the history is the du address as the default form for the informal ‘language of proximity’ (or ‘language of immediacy’, in the terms of Koch and Oesterreicher 2012), which already materialises in the earliest sources. In the Middle Ages, the ihr-form (originally spelt ir; grammatically: 2nd person plural, cf. French vous) was added as a form of politeness in communicative formal situations (or ‘language of distance’, cf. ibid.); this two-part system survived until the Early New High German period (stage II). In the 17th century, the system was extended by Er (masc.)/Sie (fem.; grammatically: 3rd person singular) as polite forms of address (stage III) and finally, in the 18th and early 19th century, by the – even more polite – Sie form of address (stages IV and V).^[3] Like in other languages, the emergence of 2nd and 3rd person plural and 3rd person singular forms of address reflects the application of widespread politeness strategies: the use of plural forms to metaphorically elevate the addressee or the use of indirect address as a face-saving act (Dammel 2017: 207–212; Listen 1999; Simon 2003a: 128–133).

Against this background, it seems particularly worthwhile to conduct an in-depth corpus-based analysis of the developments in the 19th century. The most obvious motivation for such an investigation is that, according to Figure 1, it is the century in which the abrupt change from a four-tier system, which had been built up over more than three centuries, to a two-tier system, which is still valid in the standard language, took place. Dammel (2017: 211) describes the fairly sudden reduction to the dual system with Sie as the formal and du as the informal address pronoun in the course of the 19th century (stage VI) as a collapse of the ‘overloaded five-tier system’ (“Dieses überladene fünfstufige System”), thus attributing the collapse primarily to language-internal factors, which may be associated with societal changes following the French Revolution (cf. von Polenz 1999: 383). However, the question would then arise as to why this overloaded system was able to function for about a century and a half rather than collapsing earlier.

A further motivation is that we now finally have sufficient data to conduct an in-depth analysis for the 19th century. Such an investigation has only become possible after many new textual sources from this century have been made accessible in the past two decades. In particular, private correspondence from all parts of the population have been made available – for the first time in the history of German – to an extent that allows comprehensive historical sociolinguistic studies (cf. Elspaß 2012a, 2015; Schiegg 2022).

It must be emphasised that the six-stage model outlined above and its sudden reduction accounts for single addressees and the written tradition of German only; the conditions in dialects and other regional varieties of German may even today deviate from this model (cf. Simon 2003b, 2004 for Bavaria). For example, while the use of Ihr as formal address had become uncommon in the written language of the early 19th century (cf. Besch 2003: 2616) – and therefore is missing in stage VI in Figure 1 –, it was still used in some regional varieties until the recent past (cf. Besch 2003: 2601; von Polenz 1999: 384). Grober-Glück (1994: 92–93) demonstrates that in the 1930s the use of Ihr was the dominant variant for farmhands addressing their employers throughout most parts of Germany, particularly in the west and south-west. Further information on non-standard forms of address comes from dialect grammars and dialect atlases. The dialect atlas of Bavarian Swabia (SBS 2: 173–182), for example, documents the language use in the 1980s and provides evidence for plenty of variation in the address of parents among the older, rural population, born in the first third of the 20th century. While the informants characterise the use of Sie as a rare or old variant for addressing parents (one informant used it until around 1945), Ihrzen was still far more common. In any case, the loss of the Ihr address to single addressees in Bavarian Swabia often only took place in the 20th century.^[4] Overall, however, cross-dialect studies and work on diatopic variation in the use of forms of address are lacking so far (cf. Simon 2003a: 125).

Moreover, most previous studies on the system of address in German are based on edited texts from the Middle Ages and the early modern period, and for New High German even only on printed sources. Apart from sporadic observations in the correspondence of the Mozart and Goethe families (cf. Reiffenstein 2009: 69; von Polenz 1999: 384), studies on variation and change in the use of pronominal address forms within family networks are lacking. Thus, hypotheses that are put forward in the research literature and in textbooks on the history of German are often not backed up by empirical corpus-based data of language usage. For example, Besch (2003: 2617), who claims that a relatively rapid change in the pronouns of address used by children towards their parents took place in the first decades of the 19th century, bases his assumptions solely on accounts in various editions of a foreign language grammar of German. Von Polenz (1999: 384) states – without specifying a source – that spouses from the higher social classes mostly used Sie (instead of the old Ihr) to address each other in the 19th century, as the use of Du was considered ‘plebeian’ and ‘obscene’. Findings in patient letters, however, hint at more individuality and variation in actual usage than previously assumed (cf. Schiegg 2022: 227).

To examine variation and change of the German pronominal address system in the 19th and early 20th century more closely, we will use a corpus of patient letters and emigrant letters (see Section 3). The following research questions will guide our investigation:

In the concluding Section 6, we will discuss and try to find explanations for the results on the five research questions.

3 Corpora: patient and emigrant letters

Our data come from two historical corpora of patient and emigrant letters from the 19th and early 20th centuries, comprising more than 5,000 letters overall. Extensive metadata is available for most of the writers, so we usually know their hometown, their year of birth, their occupation and their confession, among other things.

The patient letters were taken from the Corpus of Patient Documents (CoPaDocs) that has been compiled by Markus Schiegg’s research group at the Friedrich-Alexander-Universität Erlangen-Nürnberg. It consists of more than 4,000 texts written by patients and their relatives and acquaintances from different German psychiatric hospitals, mainly between 1850 and 1940.^[5] The letters have survived because they were censored and preserved in the patient files and case books where they can still be found today (Schiegg 2022). The data mostly encompass letters, but also texts of other types, such as autobiographies, notebooks, and poetic texts. Most of the material was retrieved from the southern German psychiatric hospital Kaufbeuren-Irsee, but a few hundred texts also stem from other places in the north (Hamburg) and north-west of Germany (Westphalia). The letters can be differentiated into private letters to family members or friends and official letters to doctors or official institutions. As the present study focuses on pronominal address in private communication, only the 1,767 private letters, written by 432 different people, were selected. 1,424 of these letters by 343 writers are from Kaufbeuren-Irsee, the rest – 343 letters by 89 writers – is of northern German origin.^[6]

The emigrant letters come from a corpus of 648 letters compiled by Stephan Elspaß in the context of his research on a ‘language history from below’ project (Elspaß 2005, 2015, 2023). About 600 letters were taken from the German Emigrant Letter Collection (‘Deutsche Auswandererbriefsammlung (DABS)’), which comprises a total of 12,000 letters archived in the manuscript department of the Gotha Research Library.^[7] The other letters come from various published and unpublished sources (cf. Elspaß 2005: 539–557 for details). Elspaß’s original corpus of 648 documents was expanded over the years by approximately 350 further texts. A large number of these additional letters were newly transcribed by Schiegg’s research group so that the emigrant letter corpus on which the present study is based consists of around 1,000 texts. The letters were written between 1830 and 1938 by emigrants, mostly to North America, to their families at home in the German-speaking areas of the time. There are also a few letters from relatives and friends to the emigrants. As only few texts had to be excluded from the present analysis (one diary, a few official letters, letters written in Dutch etc.), 981 private letters by 356 different writers were considered for this study. Table 1 gives an overview of the composition of the two corpora.

The two corpora show a good diachronic distribution of patient and emigrant letters over the 19th and early 20th centuries, both, however, containing fewer material from the earliest and latest parts of this period. On average, our writers from the psychiatric institutions were born 17 years later than the emigrants.^[8]

As usual in historical data, certain asymmetries are unavoidable, which leads to limitations in the scope and the methods applied for the analysis. In the following, we discuss the diatopic, diaphasic and diastratic distribution of the letter material.

The clearest asymmetry in the patient letters is their regional focus on southern Germany, as 80.6 % of the private letters stem from a southern German institution in Kaufbeuren-Irsee. This results from both the good accessibility of the archive where the material is stored today (‘Archiv des Bezirkskrankenhauses Kaufbeuren’) and the research interests of the Erlangen project, which focuses particularly on intra-individual variation. The emigrant letters, in contrast, were collected with the intention to analyse inter-individual regional variation (Elspaß 2005: 68), which leads to a broad regional distribution of the data in the German-speaking areas.

The Erlangen project’s focus on intra-writer variation and in particular addressee design led to a selection of such writers with letters to a number of different addresses (Schiegg 2022: 95). Although the official letters were excluded in the present paper, there is still a high variety of addressees among the private letters, which allows for comparisons of address pronouns in these groups. Emigrant letters, in contrast, were often directed at a number of people at home (Elspaß 2005: 63), which reduces the possibilities of a differentiated analysis of diastratic variation of pronouns used with regard to the addressee.

The two corpora contain texts from writers of diverse social backgrounds. As both corpora were compiled with an interest in the language use of ‘ordinary people’, the majority of the writers are from the lower social classes. This is particularly the case for the emigrant letters, where 317 of the 356 writers (89.0 %) have a low degree of schooling (Elspaß 2005: 46). In our corpus of private patient letters, 292 of the 432 writers (67.6 %) have a manual profession with no necessity to write longer texts on a regular basis so that they can be classified as less-experienced writers (Schiegg 2022: 328). A considerable number of writers in both corpora are women, which is a notable exception among historical corpora as they usually represent predominantly or solely male writing (Nevalainen and Raumolin-Brunberg 2017: 26). The emigrant letter corpus consists of texts by 95 female and 252 male writers,^[9] i.e. 27.4 % of the writers were female. The corpus of patient letters has 185 female and 245 male writers^[10] (43.0 % female writers). Thus, the patient letter corpus has a better balance of writers’ social backgrounds (their profession and gender) than the emigrant letters, which facilitates comparisons with regard to these variables in the patient letters.

4 Methodology

The 2,748 private letters selected for the analysis were annotated manually with regard to metadata about (a) the writers and (b) the letters as well as (c) the address pronouns used in each letter.

1. For each writer, we annotated their year and decade of birth, regional provenance according to the main dialect areas in the German-speaking countries (cf. Elspaß 2005: 70), gender and profession/education. For the writers of patient letters, we inferred their writing experience from their professions that either involve regular writing practice or consist of manual activities (Schiegg 2022: 328). For acquaintances with an unclear profession (metadata for acquaintances is scarce), we used the same category as for the patient. In the emigrant letters, we referred to the categories basic versus higher educated and thus less experienced versus experienced writers (Elspaß 2005: 46).

2. Each letter was annotated with the following metadata: the year and decade of writing^[11] and the addressee(s) of the letter. The addressees were placed into six categories: parents, spouses/lovers, siblings, children, relatives, acquaintances/friends. Often we found more than one addressee group in a letter, which led to mixed groups. Due to our research interests in the contextual functions of address and the limited amount of data, we do not restrict our analysis to single addressees, as has been done in previous work (e.g. Simon 2003a).

3. We are interested both in the morphology of the pronouns and the contextual functions of address (T or V). Therefore, the letters were also annotated with regard to the address pronouns used and their contextual functions. This is reflected in the glosses to our examples that first explain the morphology and then the function of the pronoun. The following three functional categories are relevant: ‘Duzen’ (T: informal address; morphology: 2nd person singular/plural), ‘Siezen’ (V: formal address; morphology: 3rd person plural), ‘Ihrzen’ (V: formal address; morphology: 2nd person plural). Thereby, we differentiate between one addressee (1) and more than one addressee (>1) (cf. Table 2). When variation appears within the same letter, more than one category of address pronouns was annotated. For example, when different people are addressed in the course of the letter, the address pronouns can change.

One difficulty in the annotation of the address pronouns is the overlapping of two categories, since 2nd person plural pronouns can be used both as informal address to several persons (T) and as ‘Ihrzen’ (V), i.e. a formal address to one person. Only if it is clear from the context that it is just one person who is being addressed with a pronoun of the 2nd person plural, we can safely assign this case to the category ‘Ihrzen’. A close reading of the entire letter is always necessary, as the writers sometimes write a letter to one person but then direct wishes or questions to other people as well. For example, our corpus contains two letters by the maidservant Babette B. (kfb-52)^[12] (born 1891), both addressed to her mother. In the letter of 9 April 1911, she uses the formal address of the 3rd person plural (example 1.a), but also 2nd person plural pronouns. In example (1.b), the writer repeats the salutation “Liebe Mutter” (‘Dear Mother’) so that the directly following “Euch” (‘you’-ACC) may first be interpreted as ‘Ihrzen’. However, later in the sentence she writes “Euch alle” (‘you all’-ACC), which suggests that Babette B. sends her Easter wishes to several relatives and/or acquaintances at home, who she addresses together informally.

The high diversity of the two corpora both with regard to the different text types and the asymmetries in the distribution of the material (see Section 3) led to the decision to analyse the corpora separately and compare the results afterwards. From the imbalances, it also follows that not all possible factors of variation can be analysed in the same way in both corpora. For example, the high variety of addressees in the patient letters allow for more detailed diaphasic analyses than in the emigrant letters.

The imbalances in some dimensions of variation are not necessarily disadvantages but may even facilitate analyses as they reduce the possible factors of variation. For example, the clear dominance of southern German material in the patient letters on the one hand obviates comprehensive diatopic analyses but on the other hand allows a clearer focus on the southern German letters without greater regional imbalances in the use of address pronouns. As a methodological consequence, quantitative analyses of patient letters are conducted only with the southern German material.

The good diachronic distribution of the material makes it possible to trace changes in the use of pronominal address over time using quantitative methods. Consequently, the most important independent variable for the quantitative analyses is time. For a better overview, the diachronic results in Section 5 are therefore presented in 20-year periods, so that the use of different ‘writer generations’ can be compared. At the same time, we need to consider the high diversity of data of the individuals, ranging from one to more than thirty letters per writer (on average: 4.2 letters per patient vs. 2.8 letters per emigrant). Qualitative analyses of intra-individual variation will be of particular interest with authors of multiple letters, as they can provide more detailed insights into the individual language use of the writers to different addressees and over time. The data thus makes it possible to analyse both variation and change in pronominal address.

5 Results of the corpus analyses

5.1 Pronominal address in patient letters

In the following, we present the results of our analyses of pronominal address in the patient letters. As described in the previous section, we focus on the letters from Kaufbeuren-Irsee, drawing on the material from the other institutions for comparison.

5.1.1 Letters to parents: overview on language change

123 of the 343 people from Kaufbeuren-Irsee wrote at least one of their letters to their parents. Four writers do not use any address pronouns. 54 use formal address pronouns (‘Siezen’), while 71 use informal ones (‘Duzen’). Some of the writers use both formal and informal pronouns, so they were placed in both categories. Formal address with the 2nd person plural (‘Ihrzen’) is quite rare in the letters, as only twelve writers use this variant – five write only Ihr, seven use both Ihr and Sie or Du in the same letter.

In most cases, the 2nd person plural pronoun could be connected with wishes or requests directed to relatives at home (cf. Section 4, example (1.b)). Clear cases of ‘Ihrzen’ are, for example, references to contents that can only be attributed to one addressee, such as the mother’s name day in the letter by Theodor H. (kfb-882) (born 1850) (example 2), or letters with clearly separate addressees in different parts, such as a letter by Karl W. (born 1839) (kfb-1217) to his mother and siblings, in which he directs a request for trousers and waistcoats directly to his mother (example 3):

(2)

Ich wünsche Euch zum neuen Jahre u. zu Eurem Namenstage Alles erdenkliche u. mögliche Gute

‘I wish you-DAT-PL-(1-V-Ihrzen) all the best for the new year and for your-DAT-PL-(1-V-Ihrzen) name day.’

(Theodor H., kfb-882, postal assistant, letter to mother, 5 January 1886)

(3)

Libe Mutter Seid ihr so gut u last mir ein par Sontaghosen u westen machen

‘Dear Mother, would you-NOM-PL-(1-V-Ihrzen) be so kind as to have a pair of Sunday trousers and waistcoats made for me?’

(Karl W., kfb-1217, shoemaker, letter to mother and siblings, 26 September 1864)

As ‘Ihrzen’ was still a common phenomenon in Bavarian Swabia in the 20th century (see Section 2), it comes as no surprise that there is no connection with diachrony in the corpus (birth years 1819 [Katharina D., kfb-667] to 1882 [Josefine S., kfb-2287]). Even in the most recent letters from the 1930s, we found instances of Ihrzen.^[13] Social class or gender do not seem to play a role either.

Since the number of occurrences for ‘Ihrzen’ are comparatively small, they were not included in the quantitative analyses. Also, we have merged the results for one and more than one addressee, as our focus in the following is on the question whether the different groups of addressees are addressed formally (V) or informally (T).

Figure 2 presents the data for the letter writers’ use of formal Sie towards their parents. The black graph shows the proportion of writers using the formal address pronoun over time. The writers were grouped according to their birth date, with the earliest writer addressing his parents born in 1807, the latest in 1907. Around 55 % of the writers born in the first half of the 19th century used formal Sie when addressing their parents, while there is a slight decrease to 44 % for the years 1850–69 and a sharper decrease to around 20 % by the end of the century.

Figure 2:

Proportion of Kaufbeuren-Irsee writers and letters with formal Sie address to parents over time.

The psychiatric hospital in Irsee opened in 1849 and the earliest letters in our corpus are from 1850. The other two graphs consider language use in the letters addressed to parents. As there are only two letters addressed to parents from the 1930s in the corpus (both du), this period was not included in the figure. The solid red line shows the development of the Sie address in all the 291 letters from Kaufbeuren-Irsee in letters to parents. In 47 cases, the letters were also addressed to other relatives such as siblings. These letters were excluded to plot the dotted red line. Again, there is a clear decrease in the use of the Sie pronoun over these 80 years. When writers address only their parents, the proportion of the formal address pronoun is higher, particularly in the first (62.0 % vs. 53.3 %) and last period (26.3 % vs. 13.5 %). Contrary to what has been stated in previous research literature (cf. Section 2), the proportion of the formal Sie pronoun is still rather high in the second half of the 19th century, and it is, furthermore, commonly used also in the first decades of the 20th century.

5.1.2 Letters to parents: social factors for the use of address pronouns

Of the 123 people writing to their parents, 80 (65.0 %) came from a background of manual professions such as farmers, craftsmen and maids and were therefore classified as less-experienced writers. 43 (35.0 %) were considered experienced writers as they had worked in professions that involved regular writing such as bookkeepers, merchants, and students. Figure 3 shows the diachronic developments of the Sie address in letters to parents divided into these two groups.^[14] The rather low number of experienced writers leads to larger fluctuations than in the data of the less-experienced writers. Both groups use the Sie address regularly and follow its general decrease over time. However, there seems to be the tendency of a slower decrease in the group of less-experienced writers. Of such writers born in the first half of the 19th century, 51.2 % (22/43) use the formal address (compared to 60.9 % [14/23] of the experienced writers) and of the less experienced writers born in the second half of the 19th century and the beginning of the 20th century (until 1909), 32.5 % (13/40) employ the Sie address (compared to 22.2 % [4/18] of the experienced writers). Thus, experienced writers seem to lead this change.

Figure 3:

Proportion of Kaufbeuren-Irsee writers using the formal Sie address to parents with respect to writing experience (birth years).

The analysis of use of the Sie address with respect to gender shows that both female and male writers regularly use the different forms of address, and that the Sie pronoun decreases over time. As Figure 4 illustrates, in any of the five writer generations examined, the proportion of the Sie address is higher for females than for males. While again the numbers are rather small, the consistency of these proportions is striking and may either result from women being – or being expected to be – more polite, or it may correlate with the generally lower writing experience of women and the more conservative behaviour of this group with regard to address pronouns. This would conform to our results on writing experience (see Figure 3).

Figure 4:

Proportion of Kaufbeuren-Irsee writers using the formal Sie address to parents with respect to gender (birth years).

5.1.3 Letters to parents: factors of intra-individual variation

About two thirds of the writers using the Sie pronoun (36 out of 55) did this consistently. Interestingly, their average birth year (1837) is 13 years earlier than that of those writers who varied in their language use (1850), which may be an indicator of change. Sometimes, the variation cannot be explained and seems to be random. For example, the corpus contains 12 private letters by the factory worker Viktorie B. (kfb-2853) (born 1864), most of them to her father and her sister. When she addresses her father, she fluctuates between Sie and Du, sometimes even in the same sentence:

(4)

muß ich an Dich, lieber Vater schreiben, daß Sie mich oder meine Lina, sobald wie möglich heimholen!

‘I must write to you-ACC-SG-(1-T-Duzen), dear father, that you-NOM-PL-(1-V-Siezen) bring me or my Lina home as soon as possible!

(Viktorie B., kfb-2853, letter to father and sister, 7 January 1890)

When both parents are addressed, we sometimes find variation between formal and informal plural forms. This can, for example, be observed in a letter by the butcher’s daughter Aloisia S. (kfb-38) (born 1827).

(5)

Der Allmächige, Wird und soll Euch Beite Hier, Und Jensets belohnen Mit Sieben Kronen; Seht Ihr in Welch Schöner Lage ich mich befinde. […] um dieses herzli Sie Bittend […] Es that mir sehr Leid, mich von Ihnen getrennt sehen zu müßen –, ohne Abschied, Und Abbidte getan zu Habn. […] Kön̄ens wen̄ Ich nuntr kom̄, leicht vurstelln.

‘The Almighty will and shall reward you-ACC-PL-(>1-T-Duzen) here and hereafter with seven crowns; you-NOM-PL-(>1-T-Duzen) see in what a beautiful situation I find myself. […] I sincerely ask you-ACC-PL-(>1-V-Siezen) for this […] I was very sorry to see myself separated from you-DAT-PL-(>1-V-Siezen) – without having said goodbye, and having apologised. […] You-NOM-PL-(>1-V-Siezen) can easily imagine when I come down [i.e. home].

(Aloisia S., kfb-38, letter to parents, around 1850/51)

She starts her letter with the informal pronouns Euch (acc.) and Ihr (nom.), but later also uses the formal pronouns Sie (acc.) and Ihnen (dat.). The syntactic contexts for the formal pronouns are rather formal with a present participle verb (“Bittend”) and a complex sentence structure with an infinitive construction and with a tripartite verbal complex (“mich… getrennt sehen zu müßen”). This may have triggered the formal address pronouns. Interestingly, Aloisia S. uses a third form, the enclitic pronoun s (for Sie; for example, in Kön̄ens [can-you-PL-(V)] and Schrebs.: abbreviated for Schre[i]b[en]s [write-you-PL-(V)]). This form can be characterised as a southern German regional variant (Simon 2004: 358; Werth 2020) of the formal address. However, due to the affinity of regional variants with informal linguistic registers (cf. Schiegg 2022: 335–339) this variant can be classified as less formal than Standard German Sie but at the same time more formal than the informal Ihr (you-PL-(T)) address. Here, the co-occurring forms are informal dialect variants,^[15] which again probably triggered the clitic form.

Particularly in formulaic contexts at the beginning of letters, writers sometimes use the formal pronouns. For example, the nailsmith Johann H. (kfb-789) (born 1835) usually addresses his mother with the informal pronoun Du, but in a routine formula at the top of his letter from 22 June 1870 he uses Sie: “Ich füle mich verpflichted an Sie zu Schreiben” (‘I feel obliged to write to you-ACC-PL-(1-V-Siezen).’). Similarly, the factory worker Josef R. (kfb-1342) only uses the formal pronoun Ihnen in a routine formula at the beginning of most of his letters:

(6)

Liebe Eltern! Ich teile ihne mit, daß soll Vater u. Tante u. Viktor zu uns kom̄e. Und Sei so gut, u. bringt mir gleich mein Werktaganzug

‘Dear parents! I would like to inform you-DAT-PL-(>1-V-Siezen) that father, aunt, and Viktor are to come to us. And [you-NOM-SG (1-T-Duzen)] be so good as to bring me my workday suit right away.’

(Josef R., kfb-1342, letter to parents, 24 September 1907)

His verbal forms indicate informal address, but he fluctuates between singular (“Sei”) and plural (“bringt”). His letter in general indicates a lack of competence in written standard German that is, among other phenomena, expressed in such switches. Individual uses of the pronominal system can also be observed in a letter by Jakob H. (kfb-1909) (born 1842), a farmer’s son who wrote one letter to his father on 28 September 1883 and consistently uses Sie in the nominative, Ihnen in the dative and Euch (instead of Sie) in the accusative. Thus, there is linguistically conditioned variation between the formal Sie and formal Ihr address in this text.

Two more experienced writers also show individual uses of the address pronouns. The bookkeeper and salesman Georg B. (kfb-966) (born 1851) consistently addresses his mother and father as individuals with Du, but in letters to both of them with Sie. Another bookkeeper, Josef B. (kfb-1269) (born 1844), wrote eleven letters to his parents in the years 1876–1878. The nine letters from 1876 are all addressed to both parents and he fluctuates between the formal and informal plural address. In his last letter of this year, he provides a metalinguistic comment on the use of his address pronouns:

(7)

Sei es mir gestattet den Vater mit “Du” anzureden. Ich will einstweilen noch Umgang daran nehmen. Die Mutter würde es beleidigen.

‘May I be allowed to address father as “you”-NOM-SG-(1-T-Duzen). For the time being, I will avoid it. It would offend mother [if I used it with her].’

(Josef B., kfb-1269, letter to parents, 29 November 1876)

In this quotation, he ponders whether he may be permitted to address his father with Du, but wants to avoid addressing his mother with Du, as this would offend her. Later in this letter, he addresses his father directly and uses the formal pronoun: “Sei es von Ihnen oder von der Mutter, das ist mir gleichbedeutend, lieber Vater.” (‘Be it from you-DAT-PL-(1-V-Siezen) or from mother, it is all the same to me, dear father.’) In his two later letters, dated 24 December 1877 to his mother and 22 January 1878 to his father, he uses the informal pronouns.

These examples show that there are other factors besides time (or more precisely, date of birth and time of writing) that influence the variation in address pronouns to parents. The following two sections will examine the variation and change of address pronouns to the other addressee groups.

5.1.4 Letters to siblings, children, spouses and lovers

The writers almost consistently use the informal address pronouns (‘Duzen’) when addressing their siblings (179 writers with 504 letters from all psychiatric hospitals) and their own children (58 writers, 187 letters).^[16] The corpus contains only one formal address in a letter to children, namely when the shoemaker’s wife Maria V. (kfb-2226) (born 1838) uses a formulaic expression at the beginning of a letter to her children on 24 April 1898: “wie geths Jhnen” (‘how are you-DAT-PL-(>1-V-Siezen)’).

Similarly, in letters to siblings, formal address pronouns appear only sporadically and in specific contexts. This can again occur when writers use formulae, such as in a letter by the maidservant Therese W. (born 1851) to her sister, when she closes with a rhyming couplet:

(8)

Die Drei Feiertage sind verflosen die wir haben genosen Es wünschet Ihnen Prosit Neujahr Ihre liebe Schwester Theres.

‘The three holidays that we have enjoyed are gone now. Your-NOM-PL-(1-V-Siezen) dear sister Therese wishes you-DAT-PL-(1-V-Siezen) a Happy New Year.’

(Therese W., kfb-2091, letter to sister, 27 December between 1882 and 1895)

Another context in which the use of formal and informal pronouns of address varies are letters to different addressee groups, including one that the writers usually address formally. For example, the seamstress Josefine S. (kfb-2287) (born 1882) changes between formal and informal address when she writes to her siblings and father on 17 February 1909 but addresses her father alone (22 July 1906) with the formal address. A few cases are difficult to explain and may result from general politeness, such as in a letter by the son of a factory owner Karl R. (kfb-659) (born 1831) to his sister (2 July 1856) or an intended broader group of recipients in the factory worker Albert F.’s (kfb-2032) (born 1882) postcard to his brother (undated, around 1905–14). Mental illnesses are also a factor that must always be considered in patient letters and that leads to difficulties in writing, for example when the former infantry soldier Karl F.’s (kfb-2882) (born 1849) mixes informal and formal address pronouns in an undated letter to his brother. A strange contrast between different degrees of formality can also be observed in the clerk Karl D.’s (kfb-1614) letters to his sister. While he writes a cordial letter to her on 10 April 1903 (e.g. the closing formula: “schließe ich mit bestmöglichen Grüßen von Deinem Dich liebenden Bruder Carl” ‘I end [my letter] with the best possible greetings from your-DAT-SG-(1-T-Duzen) you-AKK-SG-(1-T-Duzen) loving brother Carl’), his letter three years later on 25 February 1906 is excessively formal (“Es grüßt Sie mit freundlicher Hochachtung Karl D.” ‘I greet you-ACC-PL-(1-V-Siezen) with kind regards, Karl D.’ [surname abbreviated by us]). The writer died in November 1907 and the inconsistencies in his texts could have resulted from his deteriorating health.

Another group of people that is usually addressed with informal address pronouns are spouses or lovers (131 writers with 310 letters from all psychiatric hospitals). Occasional examples of formal address pronouns in letters to spouses may again be explained with multiple addressees, such as in the bookkeeper Georg B.’s (kfb-966) (born 1851) letter to his wife and mother-in-law from January 1894, where he uses the Sie pronoun. There is yet another case in which a writer changes his language use over time: The shoemaker Cosmas R. (kfb-2108) (born 1825/26) uses the informal address in two letters to his wife in 1872. Ten years later, in April 1882, he wrote a last, overly formal letter to her, before he died from his paralytic illness in July of the same year. His writing ability is clearly connected to his illness, as shown in Schiegg (2022: 367–371) and may have influenced the degree of formality used in his letters. Another context for using the formal address to her husband are the poetic passages that Maria W. (kfb-1745) (born 1848), a tailor’s wife, employs in her letter on 22 June 1882. She metalinguistically reflects on the use of her address pronouns and explains the formal Sie with the estrangement from her husband. She varies with the informal address in her letter and also reflects about the “Du Wort” (‘Du word’) that she characterises as faithful, pure, slender and fine:

(9)

Ach lieber werther Herr Gemahl, wie ich Sie titelier allchier, o Sie liebe, edle Man̄eszier. Den so wie das Schicksall geschieden hat, so hat auch Das Herz geschieden in stumer Gnad […] thuet leide mir, weil ich Sie gnädiger Herr titelier Und gar Euer Gnaden noch größere Zier, komts Du Wort mitten drunter hinein, so treu, so rein, so schlank so fein, und gar so treu wie die Altbeyerlein

‘Oh dear esteemed husband, as I address you-ACC-PL-(1-V-Siezen) here, oh you-ACC-PL-(1-V-Siezen) dear, noble man. Because as fate has parted us, so has the heart parted in silent grace […] (I) am sorry, for although I address you-ACC-PL-(1-V-Siezen) “gracious Sir” and even “your Grace” or (use) even greater adornments, the “you-NOM-SG-(1-T-Duzen)” word comes in sometimes, so faithful, so pure, so slender so fine, and even as faithful as the little people from Old Bavaria [Altbayern].’

(Maria W., kfb-1745, letter to husband, 22 June 1882)

Due to the low number of 38 texts (by 16 writers), letters to lovers and fiancées were annotated together with the spouses. Again, we find nearly a consistent use of informal address pronouns but, again, with some exceptions. Karl R. (kfb-659) (born 1831), for instance, the son of a factory owner, who has addressed his sister formally (see above), also uses the formal pronouns in a love letter from 2 November 1856 to his girlfriend. Similarly, the pastry cook Friedrich G. (lip-420) (born 1846), treated in the north-west German psychiatric hospital Lippstadt-Eickelborn, addresses his lover in two letters from 22 March 1869 formally. The maidservant Therese W.’s (kfb-2091) four letters to her ‘beloved’ Johann show variation in the use of address pronouns. In an emotional letter from 27 December 1891, she varies between the Sie pronoun, then uses informal verbal forms (imperatives singular: “Sei” ‘be’/“schreibe” ‘write’) and repeats and thus intensifies her request, using the southern German enclitic pronoun s (“Schreibens” ‘write’; “habens” ‘have’):

(10)

Haben Sie mich nah als ihre Gelibte oder als Freundin! Sei so gut und schreibe mir warum Schreibens mir den doch einmal, habens nicht mer’s Gorasch Geliebter aber ich bitte Sie Darum.

‘You-NOM-PL-(1-V-Siezen) have me close as your-ACC-PL-(1-V-Siezen) lover or as girlfriend! Be so kind and write me, why do[n’t] you-NOM-PL-(1-V-Siezen) write me once, don’t you-NOM-PL-(1-V-Siezen) have the courage for it anymore, beloved, but I ask you-ACC-PL-(1-V-Siezen) for it.

A short-term diachronic change can be observed in the merchant Otto D. (ham-18626)^[17] (born 1903) from Hamburg in his four letters to his lady friend Gertrud in June 1930. He proposed an engagement with her on the 24th, using informal address pronouns. As he did not receive a reply within the following days (his letters were never sent out), he wrote a short card on the 29th, in which he excuses his directness and withdraws his offer, using formal address pronouns.

5.1.5 Letters to relatives and friends

In letters to relatives and acquaintances, we finally observe more variation in the use of address forms than in those to siblings, children and spouses/lovers, which again allows for a quantitative perspective. Since we do not have sufficient data for a regional/diatopic comparison, we again restrict our analysis to letters from Kaufbeuren-Irsee. The group of relatives contains uncles/aunts, cousins, nieces/nephews, godparents, sisters- and brothers-in-law. Despite the heterogeneity of this group, a diachronic development in the use of formal address pronouns towards relatives is noticeable. 42.6 % (29/68) of the writers born in the first half of the 19th century use formal address pronouns with relatives, which declines to 37.0 % (20/54) for writers born in the second half of that century (see Figure 5, light green bars). In the group of relatives, we find plenty of addressee-dependent variation in the use of address pronouns. Relatives of the same and younger generations are usually addressed informally, while in letters to older relatives, writers sometimes use the formal address. For example, the daughter of a professor, Johanna R. (kfb-343) (born 1823), uses formal Sie address for her aunt (9 January 1861) and informal Du for her cousin (14 November 1861). Similarly, the farmer’s daughter Maria C. G. (kfb-2827) (born 1845) addresses her uncle formally (31 December 1889), but her nephew informally (3 November 1904). Often, the individual closeness or otherwise to a relative influences the form of address. For example, the day labourer Martin B. (kfb-1621) (born 1832) writes rather informal letters to his “base” – Base may refer to a first cousin or other third- or fourth-degree female relative – (15 May 1899; 12 June 1903), using informal address pronouns. In contrast, the letter to his brother-in-law (24 September 1901), who he has not met in person, is highly formal and also contains the formal address pronouns (cf. Schiegg 2022: 303).

Figure 5:

Proportion of Kaufbeuren-Irsee writers using formal Sie address with relatives and friends over time.

The group of friends not only encompasses close friends, but also sometimes neighbours, fellow patients and other unspecified persons. Despite the heterogeneity of this group, there is again a clear diachronic development, from 50.0 % of the Kaufbeuren-Irsee writers born in the first half of the 19th century using the formal address towards acquaintances, to only 27.5 % in the second half (see Figure 5, dark green bars). Again, individual closeness to the addressee can best explain variation in this group. For example, the maidservant Maria E. (kfb-2817) (born 1848) wrote a letter (20 May, year unclear) to “Em̄a Und Mathias”, Emma being an acquaintance from her hometown and Mathias her new husband. When the writer addresses Emma, she uses informal pronouns. As she presumably does not know Emma’s husband personally, she switches to formal pronouns when she directly asks him to pick her up from the psychiatric hospital. It is striking that still in the 1920s and 1930s, some writers differentiate between their address pronouns for friends. For example, the innkeeper and farmer Ludwig F. (kfb-2087) (born 1875) addresses two friends formally (1928/29: Siezen and Ihrzen) and four other friends and acquaintances informally (1927/28: Duzen).

The proportions and general diachronic trends in the use of address pronouns are similar in the groups of relatives and acquaintances and also in comparison to the letters to parents (cf. Figure 2). This confirms the validity of the observations made in the corpus of patient letters.

5.2 Pronominal address in emigrant letters

Like the patient letters, emigrant letters show a considerable amount of variation and change in the use of pronominal forms of address. The patterns of intra-individual variation are often similar to those described in the previous sections. For example, the bookkeeper Robert Hager^[18] (born 1857, emigrated 1880) varies between informal address (Duzen) in a letter to his brother Hans in Germany (8 June 1881) and formal address (Siezen) in a letter to his mother (23 February 1884). His younger brother Andreas (born 1865, emigrated 1883), who worked as an ordinary sailor, appears to be more progressive, as he uses informal pronouns of address when writing to his mother (3 letters in 1886). As in the patient letters, formal pronouns of address are sometimes restricted to formulaic structures. This comes as no great surprise, as such formulae have already been shown elsewhere to be preservation areas of conservative linguistic – including grammatical – structures (cf. Elspaß 2012b). For example, in a letter to her nephew (18 April 1882), the maidservant Allegonda Look (born 1824) uses the formal address only in the request formula “sind sie so gut” (‘be you-NOM-PL-(1-V-Siezen) so kind’). Like in the patient letters, cases for Ihrzen as formal address are rather uncommon, with only 15 of the writers using this form of address. See example (11), where the writer shifts between Ihrzen and Siezen:

(11)

Liebe Onkel. Den Brief, den Ihr uns geschrieben habt die haben wir Richtig erhalten und daraus gesehen das ihr noch alle Gesund waren. Nun Lieber Frund den ich anspreche als Onnkel jetz will ich euch Schreiben wie es uns hier geth den ich habe aus den Brief gesehen sie wolten gerne wißen wie es mich mit meine Frau geth.

‘Dear Uncle. We have correctly received the letter you-NOM-PL-(1-V-Ihrzen) wrote to us and saw from it that you-NOM-PL-(>1-T-Duzen) were all still healthy. Now dear friend, who I address as uncle, I will now write to you-DAT-PL-(1-V-Ihrzen) how we are doing here because I have seen from the letter that you-NOM-PL-(1-V-Siezen) would like to know how I am doing with my wife.’

(Heinrich Elderinck, farmer from Westphalia, letter to uncle, 10 January 1870)

The regional distribution, however, is striking, as only writers from the western part of the German-speaking countries use the Ihrzen variant, which confirms Grober-Glück’s (1994: 92f.) observations (cf. Section 2).

Two aspects limit the validity of the analysis of emigrant letters. First, their number in our corpus is far smaller than the number of patient letters – emigrant letters account for only a third of the entire corpus, which makes quantitative analyses difficult. In addition, a larger proportion of emigrant letters has more than one addressee (358 of the 953 that have an identifiable addressee: 37.6 %) than the patient letters (195 of 1743: 11.2 %); this makes it difficult to compare the forms of address used by different groups of addressees. As a consequence, we limit our quantitative analysis to letters in which writers address their parents exclusively, compared to letters in which writers address their parents and other relatives.

Figure 6 confirms the general trend that the pronominal address with the formal Sie variant decreases during the 19th century. Like Figure 2, the graph arranges both the number of writers using the formal address according to their year of birth (black line) and the number of letters with formal address (red lines) on the timeline. Only 24 of the 155 emigrants (15.5 %) use the formal address,^[19] and even the numbers for those born in the early 19th century are always below 20 %. The dotted red line shows the development of the Sie address in letters to parents only, while the solid red line includes these letters and also letters to parents and other relatives. The differences in these two lines are strikingly large in the first two periods analysed (1830–49 and 1850–69). The letters in which writers do not solely address parents – as the prototypical ‘persons of respect’ in the family – show already for these two early periods a largely completed change to the Du-form of address. It can therefore be concluded that letters addressed not only to parents but also to addressees for whom the Du-form of address had already been established by the beginning of the 19th century (such as siblings, cousins) acted as a gateway for a general Duzen towards relatives – including parents.

Figure 6:

Proportion of emigrant writers and letters with formal Sie address to parents over time.

In comparison to the patient letters, the occurrences of formal address are far fewer in the emigrant letters throughout the data. While only 15.5 % of the emigrants use the formal address (see above), 54 of the 123 writers of patient letters from Kaufbeuren-Irsee (43.9 %) do so, with numbers above 50 % in the first half of the 19th century. Concerning the date of the letters, already in the first period with overlapping data (1850–69), the proportion of the Sie address in letters to parents only is far lower in the emigrant letters (41.3 %) than in the patient letters (62.0 %). In addition, the decrease in the use of formal Sie address in emigrant letters is very rapid, with only very few instances of formal address from the 1870s onwards, while in 1870–89, 45.3  % of writers of the patient letters still address parents formally.

At this point, the question may arise as to whether the contact of US-emigrants with English and the writers’ increasing familiarity with the simplified system of pronominal address in English could have had an influence on their use of Du and Sie and, in particular, on the comparatively abrupt change to the Du-form of address in letters to their parents. However, this question cannot be answered satisfactorily on the basis of the available data. The emigrant letter corpus consists largely of letters written during or in the first years after emigration, so that the actual period of language contact was relatively short for most of the writers (Elspaß 2005: 71–72). But even in the few cases where series of letters from individual writers over a longer period are available, there is no indication of an influence of language contact on the use of pronominal forms of address. To give just one example, the corpus contains 22 letters written by the four Wesslau siblings (born between 1824 and 1834; see also Helbich and Kamphoefner 2002: 123–136) who emigrated from Jüterbog (North-Eastern Germany) to New York in the 1840s and 1850s. In all these letters, written between 1860 and 1867, they use the formal address when writing to their parents at home. The corpus even provides evidence for the fact that the use of the formal address was taught to the emigrants’ descendants. It contains a letter by the eleven-years old Phyllis Neumayr (born 1895). She was a daughter of the clerk Benno Neumayr, who had emigrated to America in 1881, and his American wife. Phyllis wrote a (fragmentarily preserved) letter to her grandmother in Germany, describing her life in America and using formal address pronouns. For example, she closes her letter as follows:

(12)

Ich schike Ihnen tausend herzliche Grüße von Vater und Mutter besonders aber von Georg und mir. Ihre Sie herzlich liebende Enkelin Phyllis Neumayr

‘I send you-DAT-PL-(1-V-Siezen) a thousand warm greetings from father and mother, but especially from Georg [her brother] and me. Your-NOM-PL-(1-V-Siezen) you-ACC-PL-(1-V-Siezen) loving granddaughter Phyllis Neumayr’

(Phyllis Neumayr, letter to grandmother, 1905)

Benno, in contrast, consistently uses the informal pronouns when writing to his parents, be it to both parents, only to his mother or to parents and siblings (38 letters between 1881–1907).^[20]

These observations show that the emigrants on the one hand were rather progressive with the comparatively low numbers of formal address pronouns to their parents. On the other hand, some individuals exhibited conservative behaviour by retaining the linguistic norms acquired in Germany that they – or the German schools – also taught to their children.

6 Conclusions

Based on the results presented in Section 5, we can answer our five research questions as follows:

Figure 7:

Adapted model of pronominal address in the history of German.