Reconstructing multilingualism in the Habsburg state: lessons learnt and implications for historical sociolinguistics

1 Introduction

Language contact occurs when speakers of two or more languages or varieties interact. The analyses of resulting contact phenomena are usually carried out in line with four main bundles of criteria, namely language internal, language external, sociolinguistic and metalinguistic criteria (see Kim and Prochazka 2019: 6–7). In addition, historical language contact research introduces the dimension of real time to the study of language, its use and its social context. In this context, we are almost immediately confronted with one of the major challenges of historical linguistics, i.e. the ‘bad’ data problem. For instance, many contact phenomena are associated with nonstandard registers and have not survived in present-day varieties; reliable historical evidence is often rare or simply non-existent. Therefore, one of the primary tasks of historically oriented contact linguists as well as sociolinguists is to develop a theoretically motivated and data driven methodology for reconstructing historical languages and varieties in contact situations frequently characterised by fragmentary (linguistic) documentation and the lack of language data.

The same applies to the historical reconstruction of languages in contact and conflict in Austria, which is a relatively small, but linguistically extremely diverse country in Central Europe. Its linguistic diversity can be regarded as a prime example of “internal” as well as “external multilingualism” (see Wandruszka 1979). The external multilingualism – especially in the urban centres – reflects, on the one hand, the multilingual tradition in the Austro-Hungarian Empire and, on the other, the linguistic consequences of the migration movements of the late 20^th and early 21^st centuries. The co-existence of different languages and the resulting language contact concurs with strong “internal” multilingualism of Austria’s majority language – German – that is still present today. This “internal” multilingualism is shaped by Alemannic and Bavarian dialects that have existed since the early Middle Ages and, in some regions, have remained virtually unchanged over the centuries. The standardisation processes and the ensuing inter-varietal linguistic contacts of the last two and a half centuries have led to the emergence of new intermediary varieties between dialect and standard, and have also had an effect on the linguistic structures of the traditional dialects. In the light of this linguistic diversity, Austria offers an ideal research laboratory and testing ground for studies on historical language contact and multilingualism. In this contribution, I will first briefly outline this special linguistic situation in the Habsburg state, before presenting the sociolinguistic research that my colleagues and I have conducted so far and the methodological approaches we have used to this end. I will conclude with the lessons learnt and implications for further sociolinguistic research in historical contexts.

2 Historical foundations of Austria’s societal multilingualism

As mentioned above, German is by far the strongest language in Austria in terms of speakers. However, it displays a high degree of internal variation to this day. More important is the fact that we are also confronted with a high degree of external multilingualism. On the one hand, the reasons for this are historical. Influences from other languages in Austria have existed for a long period of time. From around the 6th and 7th centuries, peoples of Slavic origin have settled in Central Europe, including much of present-day Austria. The subsequent spread of the Hungarians, or Magyars, as well as the Bavarianisation of the area under research separated northern and southern Slavs. For the former Habsburg state, we have to reckon with eleven main languages in addition to numerous smaller ones. Moreover, these most spoken languages already represented a number of widely divergent languages: German; two Romance tongues, Italian and Romanian; a range of Slavic languages from all the three branches of that family — western, eastern, and southern; and Hungarian, or Magyar, from the Finno-Ugric group. In addition, in the majority of the crown lands two, three and more languages were officially in use at the same time.^[1]

The multilingual situation has remained relatively stable ever since. In present-day Austria, there are seven officially recognised minority languages, the so-called languages of the six autochthonous ethnic groups protected by the Ethnic Groups Act (Burgenland Croatian, Slovene, Czech, Slovak, Hungarian and Romani), plus the Austrian Sign Language. However, the 20th century brought about significant changes regarding the societal conditions for these minority groups and thus a shift in the importance of certain groups can be observed. Whereas, for example, the influence of Czech and Slovak speakers declined, others – such as speakers of Turkish, Polish (especially after 1978) or South-Slavic languages (especially Serbian, Croatian and Bosnian) – gained in importance. Other groups, for example speakers of Slovenian or Burgenland Croatian, have maintained their position even though the sociolinguistic circumstances under which the minority languages are used have changed considerably. By far the largest minority groups to date are migrants from Germany, Turkey, the former Yugoslavia, followed by migrants from Romania, Poland and Hungary (see Newerkla 2013).

What also needs to be mentioned in this context is the fact that the largest share of migrants can be found in Vienna as well as in the urban agglomerations of Vorarlberg, Tyrol and Salzburg, and not directly in the border regions. Moreover, the metropolitan area of Vienna represents a principal contact area in Central Europe, still influenced by the languages spoken in the Habsburg Empire. Due to substantial migration from what is now the Czech Republic, a micro-area emerged during the 19th century that was particularly affected by Czech-German language contact (see Newerkla 2007). Throughout this time, extra-linguistic processes fostered contact phenomena and also influenced the varieties of German in Austria (see Lenz et al. 2020). But Vienna and its surroundings remained multilingual not only due to migration movements. A varying degree of German-Slavic bi- and multilingualism had been common in the rural areas east and north-east of Vienna for centuries (see Kim 2019). As I have shown, one can conceive the whole region as a focal point of linguistic convergence and language contact within a larger Central European area (see Newerkla 2020).

3 Research on Austria’s societal multilingualism – methodical approaches and results

One of the ground-breaking research efforts with regard to this topic was initiated by Rosita Rindler Schjerve at Vienna University during the 1990s and published in the anthology Diglossia & Power: Language Policies and Practice in the 19th Century Habsburg Empire (Rindler Schjerve 2003; see also Rindler Schjerve and Nelde 2003). Taking the historical reconstruction of societal multilingualism and language policies in the conflict-driven 19th-century Habsburg state as a point of departure, she and her colleagues explored how the struggle for power was reflected in attempts to control language use and how language became a prominent site for inter-ethnic controversies and conflicts.^[2] Moreover, Rindler Schjerve tried to integrate the conclusions drawn from the historical reconstruction of languages in contact and conflict into a more general debate on historical sociolinguistics as a discipline in its own right. There were several topics already back then that attracted her attention, such as the relationship between historical sociolinguistics and neighbouring disciplines, i.e. historiography proper, socio-historical linguistics and, most prominently, synchronic sociolinguistics, with which historical sociolinguistics shares common theoretical grounds (see Rindler Schjerve and Vetter 2007). Those major theoretical and methodological issues relating to historical sociolinguistics in general and to the historical study of Austria’s languages in contact and conflict in particular were taken up again within the framework of the Special Research Programme FWF 60-G23 “German in Austria. Variation – Contact – Perception” (see Budin et al. 2018, 2019; Lenz 2018), especially in its task cluster^[3] on language contact.^[4]

In our joint research, we started from the assumption that the social as well as individual multilingualism of the Habsburg Empire has left traces in several respects in the (constructed) social as well as individual monolingualism of the Second Republic of Austria. On the one hand, it has shaped the handling of and attitudes towards multilingualism, but also towards German in Austria itself; on the other hand, it has led to linguistic contact phenomena in varieties and registers (from dialects to standard) on all linguistic levels, which distinguish the varieties of German in Austria from other varieties and registers of German (in other countries). One main aim of our research was to shed light on how German in Austria was used as an instrument of social interaction and as a reference point for cultural construction. Historical multilingualism undoubtedly had an impact on the linguistic structure of German varieties in Austria, but little was known about the characteristics of the multilingual setting in which German in Austria was embedded and which most probably continue to affect the language behaviour of Austrian speakers of German today. In this context, one central goal was to provide a historically founded and multilingualism-based understanding of the functional and metalinguistic dimensions of German varieties in the multilingual Habsburg state in order to relate them to the situation in present-day Austria. Of special interest was the political discourse concerning linguistic diversity in Austria and Austrian speakers’ linguistic self-awareness and perception.

Methodologically, we used a combination of two well-established approaches as our main analytical framework to address the well-known challenges of historical sociolinguistics, such as its cross-disciplinary nature at the intersection of language history and social history or the so-called “bad” data problem (see Auer et al. 2015; Nevalainen 1999; Säily et al. 2017):

1. the scenario approach – According to Muysken (2010: 267), a language contact scenario is “the organised fashion in which multilingual speakers, in certain social settings, deal with the various languages in their repertoire”. This approach allows for predictions regarding the kind of contact phenomena that are most likely to occur in the framework of individual multilingualism. It thus takes the multilingual individual as a starting point from which to make inferences about the speech community and possible changes to the language system (see Muysken 2013 for further details).

2. the domain-specific approach – We blended the scenario approach with a domain-specific approach (see Rindler Schjerve 1996), which is an adequate analytical instrument for the description of habitual language choice patterns in multilingual societies. It is particularly useful for describing historical contact scenarios as it allows for the drawing of conclusions about who might have used which language to whom, when and where on the basis of the data available. This approach presupposes that societal multilingualism is diglossic or rather multiglossic, i.e. that language choice is not arbitrary, but functionally organised. Therefore, it abstracts so-called domains of language use from the specific social settings, such as school, office or family. These domains derive from and reflect the societal super-structure and thus not only enable us to describe specific language use in multilingual societies, but also to generalise and predict it to a certain degree (see Rindler Schjerve 1996: 797).

Specifically, we focused on the reconstruction of language contact scenarios by combining data from quantitative and qualitative sources to gather the sociolinguistic information required for the plausibility assessment of contact explanations of linguistic features. Additionally, we aimed at comprehensively describing complete regions longitudinally, for example from 1867, when the Austro-Hungarian Compromise was established, until the dissolution of the monarchy in 1918 and beyond. To achieve this goal, we used quantitative analyses to identify places of interest. These were then investigated in depth and supplemented with qualitative information and analyses. To this end, three main data sources were used: census data, Wenker’s questionnaires (a survey conducted for his atlas of the German dialects Deutscher Sprachatlas; see Fleischer 2017) and school annual reports (Schulprogramme). While census results offer quantitative statistical data, Wenker’s questionnaires and school programmes provide both quantitative and qualitative data concerning multilingualism, for example data on the prevalence of languages other than German in the village, or data on students’ mother tongues. At the same time, these data represent a mix of objective and attitudinal data. To compensate for the potential unreliability of census data, we also integrated other written historical sources (e.g., textbooks, parish reports, etc.) into our multilevel research corpus, employed re-analyses of existing data, discourse analyses of official documents, contemporary journals and newspapers, but also made use of new mathematical models to emulate the transformation of language use in minority constellations.

Our research focused mostly on three different geographic areas and language contact situations in the territory of former Austria-Hungary: (a) Czech-German in Southern Moravia, Lower Austria (then including Vienna) and Brno, (b) Slovenian-German in Carinthia (the part now belonging to Austria) and (c) German-Hungarian in Southern Hungary.

1. The choice of Southern Moravia was motivated by several factors. In contrast to neighbouring Bohemia, Moravia is considerably under-researched. This region is also of central importance to the (perceived) influence of Slavic languages on the “Viennese” variety of German. A significant fraction of people born in Southern Moravia later resided in Vienna (roughly 10 % in 1910). Moreover, people of Moravian origin were more likely to assimilate than those from Bohemia. Thus, studying the linguistic conditions in Southern Moravia was paramount for understanding the language situation in Vienna by providing the sociolinguistic background for language contact in Vienna. In addition, we considered the full stretch between Vienna and Brno, including both Lower Austria and Moravia’s capital.

2. The second research area was Carinthia with its language contact between Slovene and the German varieties spoken in Carinthia. We chose this crown land to be able to compare local processes in different parts of the Habsburg state and identify polycentric developments.^[5] It also provided a counterpoint to Czech in that it featured a different Slavic language.

3. The third focus area, Southern Hungary with its language contact between the local German varieties and Hungarian, was situated in the Hungarian part of Austria-Hungary – in contrast to Carinthia and Moravia. It enabled comparative studies to see how the two different language policies used by the Austrian and the Hungarian parts of the Empire were reflected in language shift scenarios.

For the chosen areas, we performed analyses on the language policies and their effects on the educational system, compared different sources on the languages used in the area and awareness of multilingualism among the population, analysed cultural institutions as sources of nation-building and their impact on data concerning language, reviewed sources on language use and looked at factors influencing language shift. While several findings have already been published (e.g. Kim 2019; Kim and Newerkla 2018; Prochazka 2019; Schinko 2019), detailed results will ultimately be freely accessible on the ‘German in Austria’ research platform, which is an integral part of the digital research infrastructure of the above-mentioned special research programme.^[6]

The results for Southern Moravia support the hypotheses from historiographical publications which characterise Moravia in general, and Southern Moravia in particular, as highly multilingual up until the end of World War II. Southern Moravia displayed prolonged stable bilingualism in various shades, with local domain-specific practices. However, the studies also show that, by the end of the 19th century, the region had become a battleground over language. Unexpectedly, our results also suggest that the northern part of Lower Austria was shaped by bilingualism in various, locally determined forms (see Kim 2019).

The analysis of school annual reports from all four gymnasia in Brno (two German and two Czech-medium, see Schinko 2019) during the period 1868–1918 showed that the percentage of multilingual students (in both crown land languages) increased – from 60 % at the beginning to around 80 % in the school year 1913–14. The increase in multilingualism is mostly due to an increase in the number of students attending classes in the second language. At the Czech gymnasia, 100 % of students attended these classes throughout the period, while at the German gymnasia the percentage rose from 20–30 % (1870–1880) to 30–40 % (1880–1900) and eventually to almost 60 % (1914). However, students’ mother tongues aligned over time with the language of instruction. At the end of the period, the Czech gymnasia were attended exclusively by Czech students, and only a few Czech students attended the German gymnasia. These longitudinal observations are consistent with the findings of Kim and Newerkla (2018), who detected a similar increase in importance of Czech as a second language^[7] during the first decade of the 20th century in gymnasia in Southern Moravia where German was the language of education. In contrast, the case studies on the importance of the two languages in South Moravian elementary schools (Volksschulen) indicate that early bilingual education was not supported at that time. In the domain of education, various actors clearly differentiated between common and elite multilingualism. Whereas for the common people (early) monolingualism was considered desirable, the elites required multilingual competencies in order to pursue their careers and/or strengthen their nationality’s power position.

For Carinthia and Southern Hungary, the research focused on a specific language contact scenario, namely language shift. To this end, a new method was developed. In line with the goal of describing a longer time period and identifying places of further interest, mathematical modelling based on methods from physics was employed to describe the shift from Slovenian to German (see Prochazka and Vogl 2017, 2018; Prochazka 2019).^[8] Census data as well as other data about parish and school language were used for testing the model. This made it possible to identify specific factors influencing language shift and to quantify their influence, and in turn reconstruct the language shift in Carinthia for two periods (1880–1910 and 1971–2001). The results underscored the need for a critical interpretation of census data. The broader historical sociolinguistic context is essential for a meaningful interpretation of the mathematical results. In Carinthia, interaction with people who spoke the same language was identified as the most important factor for language maintenance globally. This includes interaction with people in both the same and surrounding villages.

The basic model was extended to measure the impact of additional local “habitat” factors, which occur only in some villages. Three factors were tested: bilingual schools, parish language and urbanity. Bilingual schools and parish language did not seem to have a noticeable impact. In fact, bilingual, or so-called utraquist schools^[9] had a slightly negative impact on the maintenance of Slovenian. Such a result can be seen as an indication that utraquist schools actually served to “Germanise” the population, something that is suggested in the literature (e.g., see Kurz 1990). A difference in rural vs. urban areas was found. In the capital, Klagenfurt, the number of Slovenian speakers declined faster in the first period (1880–1910) than elsewhere in Carinthia. This means that moving to a larger urban centre increased the chance of giving up the minority language in favour of the majority language. An interesting development occurred during the second time period (1971–2001), where the number of Slovenian speakers in Klagenfurt decreased slower than elsewhere, and even increased both in percentage and absolute number.

The newly developed model was applied to a second language contact situation, namely the shift from German to Hungarian in Southern Hungary. Again, census data was used for testing, in this case for the time period 1880–1930. Here, the Carinthian model did not prove as suitable, for a number of reasons. First, the Carinthian villages were smaller and closer to each other, while in Hungary, they were larger and situated further apart, which led to different interaction patterns. Second, migration is visible in the Hungarian data to a larger extent than in the Carinthian data, so a more complex model is needed for changes in the population. Thirdly, differences in the data material (the Austrian census asked for everyday language, the Hungarian census for mother tongue) might play a role as everyday language and mother tongue are affected differently by language shift and thus need to be described by different processes.

4 Lessons learnt and implications for (historical) sociolinguistics: an outlook

Multilingualism in the Habsburg state encompassed individual as well as societal multilingualism. Depending on local hegemonic and linguistic constellations (e.g., with German as a majority language, a minority language, or just a dynastic language), it resulted in different forms of diglossia or even polyglossia specific to each crown land. The status, functionality and structural heterogeneity of German in the Habsburg state were thus shaped by the demographic, legal and historical conditions prevailing in the diverse centres of the crown lands. In combining macro- and micro-sociolinguistic approaches we tried to gain valuable insight into the interplay of officially imposed language regulations and unofficial multilingual practices in the domains of administration, the judiciary and education as well as to uncover the impact of historical multilingualism on the linguistic structure of German in Austria.

As our research highlights, some well-attested methodological approaches (see Auer et al. 2015; Säily et al. 2017) are useful. Some novel approaches (see Prochazka 2019) help us expand our horizons, but simultaneously are limited in their applicability conditions, which is why we are always on the lookout for new considerations and more efficient approaches to the analysis of historical multilingualism and historical language contact. So, what do I think are the most important lessons we have learnt in this regard from our research so far?

(1) A central motivation of the studies in Rindler Schjerve (2003) was to gain insights for the multilingual Europe of the 21st century by analysing in particular how state actors deal with multilingualism in historical situations. Current historical-cultural approaches, which belong to the paradigm of similarity (e.g., see Bhatti and Kimmich 2015), also pursue similar goals, but shift the focus to the level of everyday practices of cultural encounter. This brings several individual aspects of multilingualism to the fore, which are oriented towards everyday communication. Given the basic assumption of contact linguistics with the multilingual individual as the locus of language contact (see Weinreich 1979 [1953]: 1), this is an indispensable prerequisite for researching the traces left by this kind of multilingualism in Austria and (Central) Europe in the 20th and 21st centuries.

However, due to the limited accessibility of data on individual language use, historical language contact scenarios pose a particular methodological challenge, which must be taken into account through extensive consideration and the linking of diverse quantitative and qualitative data sources. In the historical context, such a mixed-methods approach and the combination of multiple sources is necessary because only this particular procedure provides a way to cross-validate findings and bridging data gaps which are unavoidable in longitudinal studies. New methodological and interdisciplinary approaches, such as physics-inspired diffusion modelling of language shift (see Prochazka and Vogl 2017, 2018; Prochazka 2019), can be of particular help in this context. As we have seen, such approaches are quite capable of bridging data gaps. An absolute prerequisite for this, however, is that sufficient data are available and that the factors that have a significant influence on the mathematical modelling can be identified to a manageable extent. As soon as this is no longer the case, this form of mathematical analysis very quickly reaches its limits, because the modelling either cannot be carried out (if crucial initial data are missing) or becomes too complex (e.g., if the influencing factors are too many or change too quickly).

(2) From a macro-linguistic perspective, historical sociolinguistics also comprises themes that touch upon how language(s) and varieties are embedded in complex societies, such as multilingualism, language contact, the institutionalisation of language as expressed in language policy and planning (see Auer et al. 2015). In the Habsburg monarchy, German was embedded in a state that was characterised by social and individual multilingualism on a communicative and discursive level and this was reflected in its language policy. It stands to reason that in the 19th century, when language was understood as an “objective characteristic” indicating nationality or nationhood, language policies became the arena and venue for conflicts over hegemonic claims between the non-national Habsburg state and the increasingly important ethnic groups, especially in the traditional power domains of education, administration and jurisdiction. However, throughout this century, no fixed fronts can be drawn between the conflicting parties. Rather, we see a wide variety of fragile and only temporarily upright models of equilibrium in the execution of political power (see Wallnig 2003). However, from such a double function of language(s) as a politically charged label and at the same time a means of communication, we can only indirectly infer the actual language competence of certain population groups or individuals. No doubt, in the Habsburg state German had the function of a lingua franca in public contexts whereas other vernacular languages, i.e. geographically neighbouring languages, were also acquired and learned. Against the background of the strong regional differentiation of the diglossic and often polyglossic situations, I think that it remains a desideratum to support these findings with studies on further crown lands and domains and to attempt to classify them in a scenario-specific typology. In our opinion, the acquisition contexts of the first, second and foreign languages as well as the intra-linguistic variation should also be considered in order to make statements about the individual characteristics under certain conditions of social multilingualism.

(3) In the ideological framework of “one language equals one culture equals one nation” that was dominant in Central and Eastern Europe from the mid-19th century onwards, language use is directly related to cultural attachment, usually understood as belonging to a certain nation. Thus, multilingualism signifies multiple attachments or a non-national conception of the self. This connection between language and culture was especially important in an explicitly multicultural and multilingual entity like the Habsburg state, where German was a lingua franca, but other languages played significant roles at a local level and were used as a means to create group identities by several ethnic groups. In contrast, present-day Austria is largely considered a monolingually German-speaking state, although (individual) multilingualism exists to a great extent and the linguistic make-up of the state keeps changing, for example through immigration which brings in non-indigenous languages. This area is above all also directly linked to spoken language, which again increases the demands on historically oriented sociolinguistics due to the meagre supply of data relatively unaffected by conventions of writing in order to reconstruct multilingual orality and historical language attitudes concerning spoken language.

In this context, personal correspondence, trial proceedings, drama texts, sermons, proclamations, etc. have proven to be useful for the reconstruction of oral registers from written sources. The use of so-called ego-documents such as private letters, diaries or travelogues has recently gained importance (e.g., see van der Wal & Rutten 2013). One of our case studies in this methodological framework was based on the examination of a corpus of personal letters written by aristocratic women from the 17th century. The data comprised not only a great amount of conversational language material, but also very vividly showed in which context and in which way multilingualism was perceived as a phenomenon at that time and became the object of discursive communication. Embedded in statements on topics such as religion, faith and religiosity, marriage and matrimony, pregnancy, birth, children, family, deaths, social and interpersonal relationships as well as the social stratification of society, functional attributions of multilingualism, social positioning and partly also ideological projections of language were made (see Newerkla forthcoming). Based on former research results and especially on the findings of our joint research, we have thereby also been able to unveil a continuity of views and attitudes towards multilingualism in the Habsburg state and its successor states. In other words, we see that the former situation has had an impact on (a) the language users’ self-awareness and perception, (b) the political discourse concerning linguistic diversity, and subsequently (c) the language policies implemented in the domains of administration, the judiciary and especially education. Such results clearly support the demand for longitudinal in-depth comparison.

But, of course, this methodological approach is also subject to its restrictions and limitations. Written sources from times when literacy was not common have a strong bias toward the educated classes, which makes it difficult to form a complete picture of the social distribution of any linguistic feature or language attitude at a given time in the past. The literacy bias entails a social bias, which is why women or the lower classes, for example, have been under-represented in conventional language histories (see Auer et al. 2015). Efforts to make increased use of sources from parts of society that have remained largely unheard in traditional historical linguistics, such as private letters by women in our case, are in principle to be encouraged, but often fail because such sources usually have not been preserved in archives and handed down to the present day. In our opinion, it would therefore be all the more important to make the existing sources as low-threshold and publicly accessible as possible.

(4) A worthwhile attempt to arrive at a better foundation for the analysis of data on students’ mother tongues or the languages of instruction would be the digitisation of school annual reports, especially from bilingual and multilingual regions. Moreover, for German in Austria-(Hungary) during the 19th (and early 20th) century, no specific linguistic corpus is available so far. There are, however, some highly specialised text collections and corpora, such as a searchable online edition of the newspaper Die Fackel (The Torch, edited by Karl Kraus 1899‒1936) published by the Austrian Academy Corpus (meanwhile merged into the ACE – Austrian Corpora and Editions)^[10] and the ABaC:us – Austrian Baroque Corpus,^[11] including a corpus of texts by the Baroque preacher Abraham a Santa Clara (1644‒1709).

Additionally, it is possible to draw from the ANNO-database^[12] of the Austrian National Library, which digitises its historical newspapers and publishes them online as a collection of OCR-scans. ANNO is the largest searchable collection of texts of historical German varieties in Austria. However, what is lacking so far, is a joint effort to prepare – at least selected – texts from ANNO for corpus linguistic analyses and thereby create a yet unprecedented corpus of Austrian German in historical newspapers. Such texts could cover several time periods throughout the 19th century to facilitate diachronic analyses. Regionally, such an annotated corpus could also comprise texts from capitals of various Austrian crown lands, including at least Vienna (Lower Austria), Innsbruck (Tyrol and Vorarlberg), Prague (Bohemia) and Lemberg (Galicia) to allow for investigations on aspects of polycentricity. Such a corpus would be comparable genre-wise to the more recent data from the Second Austrian Republic available in the AMC – Austrian Media Corpus.^[13] Such annotated diachronic corpora would enhance the feasibility of diachronic studies on the linguistic material expressing attitudes and the perception of particular linguistic concepts in non-scientific press publications. By using a frame-semantic approach (e.g., see Ziem 2014) for the reconstruction of historical concepts, these annotated texts could then allow for new insights, for example into the attitudinal distinction between an increasingly standardised variety of Austrian German and non-codified varieties of German in Austria (e.g., dialects) by language users of past time periods.

Again, all of these resources for historical language varieties only include written texts. Additionally, most of them represent a standardised variety of the respective language. The resulting and expected lack of data for non-standard varieties and authentic learners’ or bilinguals’ speech is crucial for the whole research process, since the vast majority of contact phenomena has been and is ascribed to those varieties and registers. To some extent, this problem could be approached by considering imitations of bilingual speech in literary and satirical works with a methodology adapted from Lea Schäfer (2017), applied by her to Yiddish texts.

5 Conclusion

In this contribution, I have tried to show that, in my opinion, it is precisely the comparison of historical sociolinguistic research with synchronic sociolinguistic research which represents an added value that has not been fully exploited, despite its potential to provide a deeper understanding of ongoing linguistic processes in complex multilingual societies. By analysing contemporary language use against its historical background, we can for example shed light on how, in the context of the other languages, a certain variety was and is used and valorised as an instrument of social interaction and as a reference point for cultural construction. Additionally, we can unveil the manner in which different ethnic groups experienced the use of a certain variety – mediated through the multiple lingua-cultural practices – in their everyday life and how they experience it today. Last but not least, we can thus reach a better understanding of the impact this historical multilingualism has had on domain-specific communication and language attitudes not only in the past, but even today.

I think that international research on historical multilingualism will profit from such findings and results, since research tasks like these not only comprise the analysis of the characteristics of a specific historical multilingual setting. They also deal with their lasting consequences by distilling the implications of a certain historical context on the language-specific attitudes of speakers of such varieties involved to this day. Ultimately, such a research design – when applied in different areas – will allow for the identification of comparable, distinct and universally applicable aspects of language contact in the areas under investigation. In this respect, it will contribute to general theoretical questions posed by historical sociolinguistics in its relationship to neighbouring disciplines, i.e. historiography proper, socio-historical linguistics and, most interestingly, synchronic sociolinguistics, with which historical sociolinguistics shares common theoretical grounds. Together, these approaches will provide a holistic picture of a multilingual setting by looking at its conditions, effects and perception both diachronically and synchronically.