A Cacophony of Classifications: Education and Identification in a Prenational Empire

Abstract

The article seeks to fill a gap in existing scholarship on explicit and implicit linguistic, ethnic, and national classifications in Habsburg schools and their effects. It attempts to reconstruct the classifications that appeared in textbooks and other teaching materials as well as in daily practice, pupils’ exposure to them, and their engagement with these categories. Temporally, the study begins with the establishment of compulsory education (1774) and ends in the revolutionary period of 1848–49 and focuses on the Slavophone population of the Habsburg crownlands Carniola, Carinthia, Styria, and the Austrian Littoral. Our research suggests that systematic and uniform classification schemes were not yet in place in the school environment in the period under review. As a result, the influence of classificatory systems on identifications was limited. If anything, the schools inadvertently reproduced existing local and provincial identifications while the students’ limited internalization of emerging transregional identifications only happened through their personal relationships with a few teachers and peers. The transition from familiarization with emerging transregional ethnolinguistic identifications through personal networks to the systemic (and often completely unintentional) reproduction of nationalist ideology happened only after 1848.

1 Introduction

In his memoirs, Josip Sernec, an important Slovene politician during the turn of the 20th century, acknowledged that he came from a bilingual Slovene- and German-speaking family. As a child, he did not identify nationally. He only “became aware of his Slovene nationality” in school, and this largely happened after one of his teachers classified him as a Slovene and put him in a Slovene class in 1857 (Sernec 1927, 3, 4). A few years later, in 1861, the principal of a school attended by Anton Šantel, a future teacher of mathematics and physics, entered the classroom and started asking pupils about their first language. For Šantel and his friends, the question was “a new one, we were not used to such questions and not ready to answer them.” In response, Šantel began “to stammer” that he could speak German and Slovene equally well but eventually told the principal that his first language was Slovene. From then on, the school categorized him as a Slovene. In forms and reports, his family name was changed from Schantel to Šantel, and the bilingual boy started to identify as a Slovene (Šantel 2006, 229, 230).

Stories like these underlie the claim that in the second half of the 19th century Habsburg schools—at least in Cisleithania—contributed to the emergence and reproduction of ethnolinguistic nations as relevant categories of identification. This is rather surprising: the Habsburg Empire was not a nation-state seeking to nationalize its population. Still, the Habsburg schools created the literate elites that formed the backbone of national movements (Cohen 2007). As Pieter Judson put it, they “seemed to offer nationalists unlimited opportunities for the indoctrination of new generations into a permanent commitment to a particular national identification” (Judson 2016, 302). ¹ This was especially important because national identifications were not integral to one’s primary socialization within the family until the turn of the 20th century. For most parents, “nationness” only became relevant at that time. For others, such categories were never relevant (Judson 2016, 269–332). ² Research has already suggested that all of this was enabled by the ethnolinguistic classification of pupils. Schools grouped their pupils according to language, and in so doing they established boundaries and familiarized children with their supposed belonging to a nation, its language, culture, geography, and history. They were, in short, establishing and promoting ethnolinguistic categories—ethnic boxes—that schools and nationalists could then fill with appropriate content (Stergar and Scheer 2018).

However, if we want to fully understand the role Habsburg schools played in shaping the identifications of the empire’s heterogeneous population, we need more research. One thing we know very little about, for instance, is the interplay of classifications and identifications in the first half of the nineteenth century. The second half of the century is well researched, for good reason: it was only through the transformations of the 1848 Revolutions that nationality became one of the basic principles of political action and group-building in the Austrian Empire. The state acknowledged nations in the 1849 Constitution, making them into a legal category (Stourzh 1985, 17–41; Judson 2016, 155–217). This was also the moment when the state started publishing its official gazette in nine languages, establishing a modern state-sanctioned system of ethnolinguistic classification for the first time (Stergar and Scheer 2018, 579). ³ Yet, Habsburg ethnolinguistic nationalisms started to emerge earlier—in some cases at the turn of the 19th century—whereas nonnational modes of group-building obviously also existed before 1848.

Our article seeks to fill this gap in existing scholarship. We examine school classifications during the period from the establishment of compulsory education in the Habsburg Empire (1774) to the revolutionary period of 1848–49. We attempt to reconstruct the explicit and implicit linguistic, ethnic, and national classifications—in textbooks, other teaching materials, and daily practice—without ever trying to impose modern taxonomies on them or passing judgment on their supposed correctness. We also observe how much the pupils were exposed to these classifications and whether they engaged with these categories.

To be as detailed in our analysis as possible and to keep the number of sources manageable, we approach this topic through a case study. We focus on the Slavophone population in the crownlands of Carniola, Carinthia, Styria, and the Austrian Littoral. Because the administrative and legal frameworks were significantly different in the Hungarian part of the Habsburg Empire (see Csáky 1978), we exclude Croatia and Slavonia and present-day Prekmurje, even though ideas and textbooks certainly circulated among Slavophones on both sides of the divide separating the so-called Austrian lands from Hungary.

Our main sources were textbooks, but we also analyzed additional texts used in classrooms even though they were not necessarily intended for scholastic use. ⁴ The corpus includes primers, readers, textbooks for specific subjects, catechisms, various teacher handbooks, etc. Their use was rather diverse. A considerable part of the texts that make up the corpus were primarily used by children in class, whereas some of the material was originally intended for teachers. The border between use for one case or the other was quite porous in practice, however, and that is why it made sense for us not to exclude the second type from our analysis. Our corpus included slightly more than one hundred texts published in several Slavic varieties and in German. We also examined published and unpublished memoirs written by teachers and former pupils. The number of relevant texts is limited, slightly more than 20 volumes. Despite the introduction of compulsory education in 1774, elementary education only really became universal at the end of 1860s. Until then, a considerable proportion of the population remained illiterate.

Their characteristics presented us with a major problem from the outset. Autobiographical writings of the period are unbalanced in terms of gender and class. Only two autobiographical texts known to us were written by a woman, both by Luiza Pesjak, and even she was not educated in public schools. As the daughter of a well-off bourgeois family, she was educated in a private school for girls and by tutors (Pesjak 1886). ⁵ In addition, most of the authors came from affluent peasant families. Unlike most of their peers, they were also educated in urban schools and later continued their studies at various institutions of higher learning.

Although this article should certainly improve our knowledge of the prenational ethnic and linguistic classification of the Slavophone population in former Inner Austria, our goal is wider. ⁶ Not ignoring the obvious differences between the crownlands—despite attempts at centralization, the Habsburg Empire had many features of a composite state throughout the 1774–1848 period—we are fairly certain that our findings will have a relevance that goes beyond this region. With our case study, we strive to shed light on the relationship between the schools in the Austrian half of the Habsburg Empire and group-building and to verify whether they already functioned as “ethnic boxes” in this era. Finally, the article is also a contribution to the wider debate on the role that prenational ethnic identifications had in nation-building.

2 Linguistic and Ethnic Classifications in Everyday School Activities

First, a few words about Habsburg compulsory education before the Revolutions of 1848. As a response to the requirements of the modernizing state set out in Theresian and Josephinian periods as well as the product of economic necessity and ideological currents of the time, the General School Regulation (Allgemeine Schulordnung) of 1774 introduced compulsory education for all children, boys, and girls between the ages of six and 12—extended to the age of 15 in 1816. At the time, several types of primary schools were established. The more demanding four-year Normalschulen and Hauptschulen enabled continuing education at the secondary level. The less demanding Trivialschulen, found in villages, focused on reading, writing, and basic arithmetic—the so called trivium—and did not allow a direct transition to the secondary level (Schmidt 1988b, 178–83; Engelbrecht 1984, 89–145, 223–43; Cvrček 2020, 25, 26). ⁷ Whether a child went to a Normalschule, a Hauptschule, or a Trivialschule did not just depend on geography but—as already hinted in the previous section—on the status and the ambitions of their parents too. Gender also mattered. Although women were included in elementary education from the very beginning, secondary education only became partially accessible from the late 1860s (Cohen 1996, 73–75, 171–211).

Even though these newly established schools were called German, the lessons in the lower forms of elementary education were held in many other languages as well. The reasons were largely pragmatic: many children simply did not understand—let alone speak—German. But some enlightened educators also began to question the established hierarchy of languages and advocate for instruction in the language of the children. They were convinced that it made more sense and was also the right thing to do. Indeed, the use of one’s mother tongue in elementary education was explicitly suggested even if the authorities also hoped that all the children would eventually learn German, which was the administrative language but also the language of most secondary and tertiary education (Engelbrecht 1984, 130). Therefore, German was present alongside local languages in the classroom. German textbooks were used, and many were bilingual; in Haupt- and Normalschulen, German was often dominant even where most of the pupils did not come from German-speaking families. ⁸

Existing research claims and our autobiographical sources confirm that the Slavophone pupils in Carniola, Carinthia, Styria, and the Littoral were mostly educated in bilingual classrooms before 1848. In village schools, especially in Carniola, classes were often held almost exclusively in the local Slavic vernaculars (Schmidt 1964, 192–235; 1988b, 232–43). ⁹ Consequently, pupils were exposed to everyday on-the-fly classifications at least to a degree. As the teachers decided who needed to use German and who could employ a different language, which pupils they would only talk to in German, and which in another vernacular, they were not only solving a practical problem but also categorizing the children according to a scheme.

Let us now look at which categories and classificatory schemes were employed in these situations and in schools more generally. Texts from Carniola show how the pupils and their Slavic vernaculars were routinely categorized as Carniolans and Carniolan. Janez Lesjak, a Catholic priest, recounts an illustrative example. In 1824, he entered a new school in Novo mesto/Rudolfswerth. One of the teachers asked him a question in German, and he responded that he could only speak Carniolan (“po kranjsko”) and not German. From then on, he was allowed to speak Carniolan, at least until he mastered German well enough (Marinko 1891, 116). Sources also indicate that the Carniolan category often excluded the speakers of both Gottscheer—a Germanic vernacular spoken by the (mainly) rural population in the southeastern part of Carniola around Gottschee/Kočevje—and German, as it was only applied to the Slavophone pupils from the province. This goes to show how important languages—and not just provincial borders—were in these classifications. It reminds us simultaneously that they were largely based on prenational and nonnational categories.

In fact, the case of Gottscheer is a good illustration of the discrepancy between prenational and national categories. Pupils from Gottschee were referred to as “Gotho-Suevus” or “Gotho-Svevus” in the catalogs of pupils in the Ljubljana/Laibach and Celje/Cilli gymnasia, implying their Germanic, yet not German, classification (Juventus Caesareo-Regii Gymnasii Celejensis e Moribus et Progressu in Literis Censa Exeunte Anno Scholastico M. D. CCC. XX. 1820; Črnivec 1999, 371, 384, 388, 390, 393, 396, 416). Andrej Marušič’s description of the situation at the Trivialschule in Gorizia/Gorica/Görz around 1836 is also quite revealing. According to Marušič, the teacher grouped his pupils into Friulians and Slavs in some situations, telling them quite explicitly: “You are Friulians … and he is a Slav” (Marušič 1898a, 21). ¹⁰ He did not categorize them as Italians and Slovenes, something that was the norm a few decades later, when—as Jan Fellerer succinctly summarized—“unitary glossonyms … became key descriptive as well as normative terms” (Fellerer 2023, 189).

This is not surprising, at least as far as the Slovene category is concerned. Research has clearly shown that local intellectuals had only started to formulate the very idea of Sloveneness after the turn of the 19th century—that is, the idea of the existence of a distinct ethnolinguistic group living on the territories between the Bay of Trieste to the west and the Pannonian plain to the east, the rivers Kolpa and Sotla to the south, and the Klagenfurt basin to the north. Until it established itself more broadly in the second half of the 19th century, the Slavophones in former Inner Austria largely identified and were categorized differently. Whereas they were as a rule Carniolans (“Kranjci”) in the eponymous crownland—occasionally also in some of the neighboring regions of other provinces—they were Slovenes (“Slovenci”) in Styria and Carinthia. Clearly, the term “Slovene” meant something else than it did later, when Slovene nationalism appropriated it and transformed its semantic meaning (Hösler 2006; Kosi 2010; 2013; Stergar and Kosi 2016). ¹¹

At least until 1848, the noun “Slovene(s)” (Slovenec and Slovenci), the adjective “Slovene” (slovenski), and the phrase “Slovene language” (slovenščina, slovenski jezik) were primarily used to describe a category that included either all the Slavs and/or their language or a particular group of Slav speakers—for example, Slavs from Styria and Carinthia or those from western Hungary, present-day Prekmurje, and/or their language (Müller 2011; Dular 2022). Sources confirm that classifications based on the latter understanding of the term were regularly used in the schools. We already mentioned the use of the Carniolan name for the people and their language, and the case of School Commissioner Blasius (Blaž) Kumerdej nicely confirms the habitual distinction between Carniolans and Slovenes. When he advocated for the use of the Slavic vernacular in Carniolan schools, he called it “the Carniolan language” (kranjski jezik); in Styria, however, he called the local Slavic variety “Slovene” (Schmidt 1988b, 202–238 passim; see also Kidrič 1929, 228). On the other hand, the sources give no hint that the pupils were ever explicitly classified as Slovenes in the modern sense in everyday school activities.

Finally, let us return to classifications in school yearbooks, the alphabetized lists of pupils, classified according to their supposed ethnic or provincial background, which the schools published yearly. These classifications were largely administrative; their influence on everyday activities was not large, but they were not entirely unknown to the schoolchildren: yearbooks were sometimes read out loud and given to some pupils at the end of the school year (Torkar 1895, 548; Marušič 1898b, 37). ¹² In any case, they are a good illustration of the arbitrary application of various categories and the instability of taxonomies. The yearbooks of the gymnasium in Marburg/Maribor are just one highly informative example. Throughout the last decades of the 18th and the first half of the 19th centuries, the classification of pupils continuously changed. Besides the categories, based on provincial and other administrative borders—the children appeared as Carniolans, Styrians, Carinthians, Gorizians, or Triestines—the Illyrian category also featured prominently. At first it was applied to pupils from Croatia and Slavonia, which probably means that it was understood in the traditional ethnolinguistic sense, equating some South Slavs with Illyrians. But after the French Illyrian Provinces were established, pupils from those territories were often categorized as Illyrians as well. However, this category was not applied in a systematic way, and some pupils from the Illyrian provinces were still categorized according to the old provincial borders as Carniolans, Carinthians, etc. Moreover, some pupils from the Littoral were categorized according to their language use as Friulians or—if they were Slavophone—as Carniolans, even though they were not from Carniola. ¹³ With the potential exception of a few pupils that were categorized as Italians, modern national categories—either German or Slovene—were never used. ¹⁴

3 Classifications in Textbooks

What about textbooks and other printed materials? What kind of categories did the Slavophone children in Carniola, Styria, Carinthia, and the Littoral encounter in their compulsory education before 1848? Even a cursory glance at the textbooks suggests that linguistic, ethnic, and national classifications were not widely used in textbooks. It also becomes clear that the terms and their semantic meanings changed throughout the late 18th and early 19th centuries.

Several translations of Thomas à Kempis’s popular religious book The Imitation of Christ (first published in 1418), for instance, were published in the 18th and 19th centuries and went through an interesting terminological evolution. The first translation from 1719 used the phrase “Slovene Carniolan language” (Slovénski Crainski Jesik) to describe its language, whereas all the subsequent editions called its language “Carniolan” (Krajnʃka ʃhpraha). The intended readers were sometimes Carniolans but also “Slovenes and Carniolans” (Slovenzi inu Crainzi) and, in 1745, “those Slovenes” (tim Slovenzam). ¹⁵ The terminology in Marko (Markus) Pohlin’s language textbook, which went through at least five editions between 1765 and 1813, was more stable: its language was called Carniolan in every edition. ¹⁶ His basic mathematics textbook from 1781 was also published for “boys and girls of the common Carniolan people” (ʃa Kraynze) so that they could learn how to calculate in Carniolan (Pohlin 1781). Johann Nepomuk Edling’s 1779 catechism also defined its language as Carniolan, but only on the German title page (Krainerische Uibersezung); the Slavic title page omitted the language (Edling 1779). The 1777 catechism, which was probably printed for the Slavophone schoolchildren of the Gorizia/Gorica/Görz Catholic diocese, defined its language as ʃlovenski jesik, which literally means “Slovene language.”

As we have explained in the previous section, the term Slovene did not have a fixed semantic meaning. It could mean either all Slavs and/or their language, a particular group of Slavic speakers and/or their language, or Slovenes in the modern sense—that is, Slavic speakers in Carniola, Lower Styria, southern Carinthia, and the Austrian Littoral, and/or their language. Consequently, the noun Slovenci could mean all Slavs or a particular group of Slavs, and the term slovenski jezik could mean “Slavic language”—an umbrella term for all the Slavic varieties—or a particular Slavic vernacular, used in a particular region. ¹⁷ In the textbooks, we have come across all these variations, but most often the terms Slovene or Slovenes (adj. slovenski, n. Slovenci, and their variations but also the German adjective windisch and even the noun Slowenen) were used to describe Slavophone Styrians or Carinthians. A table in the 1833 alphabet book from Styria, for instance, listed Carniolans as a separate category, as a separate people among the Croats, Hungarians, Germans, Slavonians, Dalmatians, Serbs, Poles, and Czechs but called the language of the book Slovene (slovenski and windisch in German; Machner 1833, 6, 7). ¹⁸ The translator, Peter Dajnko, a prolific writer from Radkersburg/Radgona, called the language Slovene—and the people using it Slovenes—in other textbooks as well (Dajnko 1824a; 1833). Although he did not define the terms there, he did so in his German textbook for learners of Slovene. He called the book’s language Slovene (windisch) and its potential readers Slovenes (Slowenen) but quite explicitly defined it as “the Styrian dialect (die Steyrische Mundart)” in the introduction. He also listed Carniolans and Carinthians as well as Croats among “our Slavic neighbors,” clearly showing that ‘Slovenes’ meant Slavophone Styrians (Dajnko 1824b, iii, iv).

This common distinction between the Carniolans and the Slovenes was also reproduced implicitly, as identical texts were published in a separate edition for each crownland. A catechism, for instance, was published in Ljubljana/Laibach, the capital of Carniola, several times but also in the Carinthian capital Klagenfurt/Celovec. In some places the spelling in the Carinthian translation differed slightly from the almost simultaneous Carniolan translation (Japelj 1789; Gutsman 1790). In several Styrian textbooks, especially those written or translated by Dajnko, these differences were even more noticeable as the books used the Styrian Slavic variety and sometimes also Dajnko’s own novel orthography (Dajnko 1816, 1824a, 1833; Machner 1833). ¹⁹

Soon after the turn of the 19th century, however, an important innovation appeared in a German-language history book. Authored by Valentin Vodnik, a priest, teacher, and important member of educated Carniolan circles, the 1809 book presented the history of Carniola and the Austrian Littoral and was intended for pupils of secondary schools, the elitist Austrian gymnasia. Because of that, the ethnicity of the people from the region was presented in a much more detailed manner. Vodnik explained how the Slavophone population of the region identified as Slovenes (Winden, Slovenzi) in Styria and Carinthia but as Carniolans (Krainer) in Carniola and added that they in fact all spoke a single Slavic dialect, Slovene (Slowensko; Vodnik 1809). ²⁰ Almost at the same time as the Vienna-based linguist Bartholomäus (Jernej) Kopitar proposed this novel idea in his groundbreaking 1808 grammar, Vodnik had already introduced it to schoolchildren, which confirms that it had already been circulating among the educated (Kosi 2013, 317). More importantly, this was the first documented appearance of the classification on which Slovene nationalism was later built in Habsburg schools.

In the following decades, this notion began to sporadically appear in textbooks and other teaching materials. Janez Nepomuk Primic’s 1814 bilingual alphabet book, intended primarily for teachers but occasionally used in classrooms as an unofficial textbook, not only defined the Slovene language transregionally but even explicitly demarcated a Slovenia (Slovenien in German, Slovénſko in Slovene). Primic explained that Slovenia comprised Carniola and the Slovene parts of Styria and Carinthia and that potentially parts of western Hungary and the so-called Provincial Croatia could also be seen as parts of this imagined land (Primic 1814). ²¹ Slovenia—and Slovenes—were also mentioned in an 1847 grammar book published in Graz. Neither was explicitly defined, yet the geographic focus of the book, the examples given in the text, made it rather clear that the novel transregional understanding of both terms was implied. The Slovenes in this book were not just the Slavophones of Styria, and they were not all Slavs either: they were the Slavic-speaking inhabitants of the Austrian Littoral, Carniola, Syria, and Carinthia (Muršec 1847).

Two books by Matija Vrtovec on viticulture and agricultural chemistry, both published a few years before the 1848 revolution and subsequently used in schools, are also interesting and worth examining even though it is not entirely clear how much they were used before 1848 (Schmidt 1988a, 92, 336). ²² In both introductions, their author, a Catholic priest, subscribed to the novel definition of what a Slovene was—he explicitly distinguished them and their language from the Czechs, the Poles, the Slovaks, and the Croats—and he also called them a nation and spoke of their motherland, Slovenia (Slovenja; Vertovc 1844, 1847). ²³ Several textbooks that came out in Carniola in the early 1830s were less explicit but nevertheless important. All of them used the adjective Slovene, not Carniolan, in their titles which was a departure from the prevailing practice in the province (Metelko 1830; Burgar, Jerin, and Jelovšek 1831; Machner 1832). Although textbooks for the Slovene language, mathematics, and religion neglected to define the term in any way, we know enough about their authors, especially the linguist Franc Serafin Metelko, holder of the newly established chair for the Slovene language at the Ljubljana/Laibach Lyceum, to be able to say that their authors were employing the term according to Kopitar’s emerging definition. Metelko, also an author of a new orthography, defined the boundaries of this language quite explicitly in his German textbook of Slovene, published in 1825. Citing Josef Dobrovský, he argued that Slovene was one of the Slavic dialects (Mundarten) and that it was spoken in Carniola, Styria, and Carinthia (Metelko 1825, xviii; see also Kosi 2013, 171).

Yet traditional classifications continued to be used. A language textbook published in Ljubljana/Laibach in 1831 and using the term Carniolan in its title is just one of many examples (Krajnski abezédnik sa uzhenze na kmétih 1831). The novel taxonomy did not establish itself as hegemonic, and an unambiguous semantic meaning of terms was only achieved in the second half of the 19th century—largely because of state-imposed standardization (Almasy 2018b). Until then, the semantic meaning of the word “Slovene” (Slovenec) was not fixed. Vodnik, for instance, used the adjective “Slovene” (slovenski) for Slavic and the noun “Slovenes” (Slovenzi) for all the Slavs in his 1811 language textbook, published in the short period of French rule over the region (Vodnik 1811). In his 1812 primer, the adjective “Slovene” (slovenski) was also used in its widest—Slavic—sense, and Carniolans (Kranjzi) were defined as a part of the larger Slavic whole (Vodnik 1812, 4 [original not paginated]). The best example of inconsistent taxonomies, however, is Primic’s 1813 reader—he started it as a textbook but later expanded it into a book for a general readership. Although Primic, a professor of Slovene at the Lyceum in Graz, embraced the novel understanding of the Slovene language, the eponymous people, and their land in his 1814 alphabet book, his 1813 book employed conflicting definitions across the text. In a part of the book, he called the Slavs (Slaven in German) “Şlovani” or “Şlovénzi” and further explained that Carinthian and Styrian Slovenes (Şlovénzi) and Carniolans (Krajnzi)—along with Russians, Poles, Czechs, Serbs, etc.—were all parts of the Slavic nation. Later in the book, however, he described Slovenia (Slovenien in German), the land of Slovenes, as a land comprising Carniola, Carinthia, Lower Styria, western Hungary, and Provincial Croatia. In a single book, Primic used the same term (Slovenes) in the traditional way—for all the Slavs and for Slavophones in Carinthia and Styria, respectively—and, indirectly, through the demarcation of their land, in the novel way, based on Kopitar’s new classification of Slavic varieties (Primic 1813). ²⁴

In short, the classifications that pupils encountered in the classroom and in textbooks and other printed materials were hardly uniform before 1848. They reflected the often imprecise, unsystematic, and occasionally even contradictory classifications of their era.

4 Linguistic and Ethnic Classifications and Categories of Practice

We have seen that trying to reconstruct the linguistic and ethnic classifications used in schools is complicated. But how might we measure their effects? Research has shown that classifications influence group-building. Although this may be true in principle, trying to ascertain whether it happened in concrete historical situations is much harder because the existence of categories does not tell us anything about their resonance (Brubaker 2006, 53, 54).

Autobiographies can help us tease out how individuals identified. However, a range of factors shape identification. The role of the parents cannot be overlooked. Our sources confirm contemporary psychological research, which suggests that parents initiate their children into group identifications in early childhood (Barrett 2012). Informal forms of education, where a considerable part of learning took place, also mattered. Autobiographies show that masons, village seamstresses, but mostly parish priests took on the role of informal teacher (Godina 1879, 13–17; Alešovec 1884, 38; Marinko 1891, 115; Kopitar 1973, 70, 71). ²⁵ And when it came to schools, classifications were not necessarily the most important factor. Sometimes they might not even have mattered. Thus, it is hard to establish whether a causal relationship between school classifications and pupils’ identifications existed in each case.

We can, however, provide some provisional answers, especially in the case of Slavophone Carniolans. Even though our sources suggest that linguistic and ethnic self-categorizations were anything but stable and clearly defined, in Carniola the Slavic speakers regularly called their vernacular “Carniolan” and an identification with Carniolanness was very noticeable. Even those who described the events before 1848 after they had fully internalized the basic postulates of Slovene nationalism and had themselves become its important creators and propagators usually stated quite explicitly that their language was perceived (by both themselves and their teachers) as Carniolan in their childhood. Whether Bartholomäus (Jernej) Kopitar, who received his primary education in the late 1780s and early 1790s, Janez Lesjak, who entered school in the mid-1820s, or Matija Torkar in the late 1830s and early 1840s, they all called the Slavic vernacular of their youth “Carniolan” in their autobiographical texts (Marinko 1891, 116; Torkar 1895, 547; Kopitar 1973, 69). In the case of Janez Trdina, the identification with Carniolanness went beyond the mere designation of linguistic practice: “I thought that the Carniolan was the best in everything, followed by the Frenchman and the German. I picked up ‘Carniolan’ patriotism at home, our peasants always spoke with a kind of respect about Carniolans and the Carniolan land” (Trdina 1946, 40, 41).

Whether this Carniolan patriotism—which, according to Trdina, was already present before children entered school—was further bolstered during elementary education and whether this was a consequence of the lectures on the history of Carniola or the classifications they encountered in textbooks is, once again, difficult to say. However, given that the idea of Carniolanness manifested itself across the board, it seems reasonable to assume that these factors worked in a mutually reinforcing way, that schools were one of the institutions that helped maintain the imagined Carniolan community. Moreover, it does not seem entirely implausible that the preschool understanding of Carniolanness was not only strengthened but also slightly transformed in schools. As Trdina’s memoir indicates, this had certainly been happening in early secondary schools. Through recitations of Valentin Vodnik’s “Revived Illyria” (Ilirija oživljena) and lectures on Carniolan history, their teacher, Franz Heinrich (1797–1844), wanted to strengthen the students’ interest in the Carniolan past. Although Trdina, his student between 1843 and 1844, states that these lectures eventually contributed to his internalization of Slovene nationalism, it seems that at this point they contributed mainly to the strengthening and transformation of his sense of Carniolanness, which was subsequently based on a clearly defined historical narrative and high literature (Trdina 1946, 93, 94, 103, 104).

Gauging the effects of ethnic and linguistic classifications on the identifications of the Slavic speakers in Lower Styria, southern Carinthia, and the Littoral is even more challenging. There are very few autobiographies, and they were written comparatively late, which suggests that the influence of later classifications and identifications on what was written might have been more noticeable. Another problem is the already mentioned semantic indeterminacy of the term “Slovene,” which was especially common outside of Carniola. When the authors described their language as Slovene or called themselves Slovenes, it is therefore possible that this was an after-the-fact nationalization or that they adopted a new understanding—one that was only starting to spread—quite early. But it is also equally conceivable that this reflected the prenational convention so common in their youth. Consequently, it is difficult if not impossible to ascertain whether in their youth they identified as speakers of Kopitar’s transregional Slovene and even as members of the eponymous ethnic group or rather as speakers of a regional Slavic vernacular or simply as Slavs in the widest sense. The autobiography of the writer from the Littoral, Josip Godina, is a good illustration of these problems. In his 1879 memoirs, he consistently used the adjective “Slovene” for the language of his youth—Godina was born in 1808—and described several people as Slovenes. But it is hard to believe that this was anything but an after-the-fact nationalization. He and his biographers each acknowledge that he only “became aware of his Sloveneness” in the early 1850s (Godina 1879; Jevnikar 2013; Pirjevec 2013). Yet trying to guess how he really identified in his youth and which terms he used for his in-group would be futile.

Despite these challenges, the prevalence of local and provincial identifications is hard to deny. Whether in Carniola or in Styria, Carinthia, and the Littoral, pupils mostly identified in this way all the way to 1848. Beside personal narratives, the frequent resistance to the use of Slavic textbooks from neighboring provinces that—as we have mentioned in the previous section—used a slightly different Slavic variety and sometimes a different orthography certainly suggests that large parts of the Slavophone population did not identify as members of a community speaking Kopitar’s transregional Slovene. Thus, for instance, the Diocese of Ljubljana/Laibach, which supervised elementary education in Carniola, rejected textbooks published by the state publishing house from Vienna (the Schulbücherverlag) and written by the Lavantal bishop Anton Martin Slomšek, a Styrian, because they used a new orthography and were written in a variety that—so the Carniolan clergy claimed—“differed significantly from the Carniolan dialect.” They argued that parents would keep their kids out of school in case these textbooks were introduced; parents would assume their children were having to learn “Croat.” (Schmidt 1988a, 116).

Sources from our sample, however, tell us that before 1848 a few students in all these lands did identify with transregional Slavic groups. Two competing—or at least not entirely identical—conceptualizations were in circulation: Sloveneness, based on Kopitar’s novel classification of Slavic varieties, and Illyrianism, the idea that South Slavs are together a distinct national group. However, the existence of these modern identifications is only documented for secondary education, or—to be even more exact—among the students at elite high schools, the gymnasium and the lyceum. Also, there are no indications that these identifications spread in a systematic way or that classifications had anything to do with them. This holds even though the new understanding of Sloveneness had already made an appearance in a few texts used in the classrooms and even if the pupils were sometimes categorized as Illyrians. At the time, these were exceptions, not the rule.

The new idea of Sloveneness as a collective identifier—unlike religiosity, estate consciousness, or provincial patriotism—did not reproduce itself within peasant communities during this period. Nor did it spread among those who had attended school or had received the most basic education through informal channels. It simply could not because it had only been formulated recently and only a few educated individuals subscribed to it (Hösler 2006; Kosi 2010, 2013; Stergar and Kosi 2016). The nationalization of the masses, however, only happened—as much as it did—toward the end of the 19th century. Illyrianism, on the other hand, had a longer history. The idea that the South Slavs spoke a single Slavic variety and belonged to a single community already started to spread in the 15th century. Through its various transformations, Illyrianism established itself and local intellectuals often identified as Illyrians. Toward the end of the 18th and at the beginning of 19th centuries, this identification gained some purchase among the rest of the population but never—despite the French establishment of the Illyrian Provinces in 1809 and the subsequent existence of the Habsburg Illyrian Kingdom—became a mass phenomenon. That did not change, even as neighboring Provincial Croatia became the center of the modern Illyrian movement in the 1830s (Petrè 1939; Blažević 2008; Greenberg 2011; Jež 2016; Stergar 2017). Essentially, both Illyrianness and Sloveneness remained limited to smaller circles of lay intellectuals, parts of the clergy, and an occasional educated peasant in the first part of the 19th century.

How and why did some high school students nevertheless identify as Illyrians or Slovenes (in the modern sense of the word) in the 1840s? For once, the sources provide a clear answer! At the beginning of the 1840s, some students at the lyceum in Ljubljana/Laibach internalized either Sloveneness or Illyrianism under the influence of their classmates educated at the gymnasium in Karlovac in Croatia (Globočnik pl. Sorodolski 1898, 1; Bohinjec 1902, 123, 124). Similarly, contacts with the students from Ljubljana/Laibach facilitated the spread of these ideas among students in Marburg/Maribor gymnasium and Gorizia/Gorica/Görz seminary (Macun 1883, 136; Kociančič 1887, 1–2 [original not paginated]; Ilešič 1907). Several teachers also contributed significantly to the penetration of transregional identifications among the students (Levec 1895, 170, 172, 176; Rupel 1941, 128–47; Trdina 1946, 132, 171, 188; Šedivy 1970, 76–79). ²⁶

By the late 1840s, these transregional identifications—while entirely unnoticeable among elementary school children—were already a differentiating principle among high school students and, in some cases, possessed considerable mobilizing potential. Of course, these were fluctuating phenomena. Periods of fervent activity, which were usually triggered by certain external stimuli, alternated with periods of relative calm. Josip Vošnjak, the future prominent Slovene politician, recalled the marked oscillations between periods of his pronounced nationalist activism and those when he had almost completely forgotten about nationalism (Vošnjak 1982, 30–32; see also Kidrič 1929, 626–29). Janez Trdina also reported that in the 1840s to him “freedom mattered much more than nationality” (1946, 133).

Additionally, sources show that in the first decades of the second half of the 19th century, peasants in Lower Carniola still understood and used the Carniolan label the same way Janez Trdina did when he wrote about the period before 1848 (Trdina 1884, 118). The fact that Carniolanness continued to be a relevant category of identification until well into the second half of the 19th century suggests that identifications changed much more slowly in everyday practice than the classifications in textbooks and other educational texts—or in the minds of nationalist activists.

5 Conclusions

The linguistic and ethnic classifications that appeared in textbooks and in schools’ everyday activities varied throughout the period, and no uniform state-imposed taxonomy existed. For pragmatic reasons, state and provincial authorities as well as the Catholic church made efforts to reduce the number of linguistic varieties in administrative and scholastic use. In an attempt to shape and control emerging nationalisms, the Metternich regime supported local activists in their efforts to standardize and promote transregional linguistic varieties (Domej 1999; Kappus 2002; Decker 2022). The modern Slovene national category that emerged out of this process did appear in a few textbooks, and the Illyrian category was also occasionally used to classify the pupils. Yet these transregional national categories remained marginal. Traditional prenational classifications continued to dominate. Additionally, the classifications and the meaning of certain categories were often only implicit, and it is quite possible that the pupils did not even notice them or understood them in unintended ways. Although Peter Dajnko, for instance, surely knew who the Slovenes in his textbooks were, he often neglected to define this category, leaving room for the pupils to understand it differently.

Most importantly, classifications and identifications were not really something the authorities cared a lot about. The priority of primary education, especially in the countryside, was to teach the illiterate children of mostly illiterate parents the basics of reading and writing (Cvrček 2020, 26–28, 36, 37, 200–232). The goal of the state was to turn children into pious, loyal, and productive subjects, whereas their identifications—beyond dynastic loyalty and Christianity—were seen as irrelevant. Emperor Francis I summarized what the state expected of schools and teachers in his 1821 speech at the Ljubljana/Laibach lyceum: “I do not need learned men, but honest, well-behaved subjects” (cited in Cvrček 2020, 47).

Not surprisingly, primary education—whether formal or informal—had at best a very limited influence on the identifications of school-age youth during the period under consideration. The main (admittedly quite prosaic) reason for that was probably the young age of the children involved. Additionally, low school attendance rates surely played a role. Although the 1774 regulations made elementary school compulsory in principle, in practice many kids only attended sporadically or not at all. In 1820, only 13% of Carniolan children of school age attended elementary school; in Carinthia, the share was 58%, and in the Cilli/Celje district of Styria it was 19%. In Carniola it was still only 20% in 1847 (Kosi 2013, 119). ²⁷

The effects were limited even in the case of those children who attended school somewhat regularly. Schools were simply not focused on generating identifications but on making children literate—and they were not necessarily successful in that either. A sophisticated three-tiered system of teaching identifications was developed only in the second half of the century. Simultaneously, it was supposed to make children into loyal and patriotic Habsburg subjects, members of an ethnolinguistic nation, and proud inhabitants of their crownland (Almasy 2018a; Horejs 2020; Moore 2020). Before 1848, this simply did not exist. During the period under consideration, dynastic loyalty was certainly emphasized, provincial patriotism was not entirely absent, and elements of ethnolinguistic nationalism made an appearance. But none of them were taught systematically. In short, there was a distinct lack of what Siniša Malešević termed “organizational grounding,” an essential precondition for any “imagined community” to reproduce itself (Malešević 2019). Changing and even conflicting taxonomies and poorly defined categories, however, also did not help.

Although the state had no intent to influence pupils’ identifications, the effects of linguistic and ethnic classifications were not entirely absent. It certainly seems that they contributed to the strengthening of some existing identifications. The Carniolanness that pupils often learned in the family, for instance, seems to have been further reinforced by the employment of the Carniolan category in textbooks and in classroom activities. Similarly, the fact that the identifications of Styrian or Carinthian pupils remained local, that they continued to be limited by provincial boundaries, seems to have been at least partially related to the use of textbooks that were largely produced locally for a group limited by provincial and diocesan borders and that used differing written standards, sometimes even different orthographies.

On the other hand, textbooks did talk about bounded human groups. The Biblical tradition, particularly the tower of Babel parable, played a significant role in this case. Language was often mentioned as a boundary: several texts mentioned the “German lands (na Nemškem),” which were implicitly defined by spoken language. This only confirms that languages were used to define groups well before the advent of ethnolinguistic nationalism, and Johann Gottlieb Fichte was not the first to use them in such a way (Kostantaras 2016). Occasionally, pupils were presented with a set of other characteristics that separated “us” from “them.” A few textbooks also provided a description of the way in which humankind was divided into groups. Kumerdej’s 1796 bilingual reading textbook even employed a novel Slavic terminology and spoke of a “flock” (krdelo) for the German Volk, and a “genus” (rodovina) for the German Nation. ²⁸ In short, the basic vocabulary of ethnolinguistic nationalism was already present in textbooks before the middle of the 19th century. But this was not yet a fully elaborated ideology; it was more of a set of disparate concepts. Nevertheless, it is possible to surmise that schools did play a role in the subsequent nationalization of the masses even when textbooks did not yet reproduce nationalist worldviews. They familiarized the children with certain concepts—even if only in a very rudimentary form—and thus made the later spread of nationalism somewhat easier.

Classifications based on modern national categories, however, only made sporadic appearance in schools before 1848. Their effects were minimal at best. Existing research clearly shows how the nationalization of the masses gradually became an inseparable part of the Habsburg educational system after 1848, especially after the introduction of new legislation in 1869 (Almasy 2014, 2018b; Govekar-Okoliš 2017; Moore 2020). To a large degree, this was a consequence of the state-sanctioned classification that developed after 1848 and was, of course, complemented by the influx of increasingly nationalist-minded teachers. One development reinforced the other, and both were reinforced by other modes of dissemination of nationalist ideology. The new national taxonomy made, as Rogers Brubaker has argued, “certain categories readily and legitimately available for the representation of social reality, the framing of political claims, and the organization of political action” (2006, 54). In short, after 1848 classification provided the framework that could—and indeed was—instrumentalized by the nationalists.

During the first half of the 19th century, on the other hand, a new state-sanctioned taxonomy was only emerging and lacked any institutional grounding. If anything, the schools inadvertently reproduced existing local and provincial identifications while the students’ limited internalization of emerging transregional identifications only happened through their personal relationships with a few teachers and peers. The transition from familiarization with emerging transregional ethnolinguistic identifications through personal networks to the systemic (and often completely unintentional) reproduction of nationalist ideology happened only after 1848.

Finally, there is one more conclusion that we can draw from our research. We certainly could not avoid noticing that the idea of a Slovene nation comprising Slavic speakers in Carniola, Carinthia, Styria, and the Littoral only made an appearance in the schools after the turn of the 19th century. Still, it remained limited in its reach—as a category and as a self-identification. Other categories and other identifications were much more present and relevant throughout the period under observation. This adds further weight to the assertions of those scholars who have argued that the Slovene nation is a modern phenomenon that did not evolve from an existing ethnic community because the latter simply did not exist (Hösler 2006; Kosi 2010, 2013; Stergar and Kosi 2016).

This then begs the question of what happened during the next decades to turn Sloveneness from a marginal to a dominant mode of group-building among Slavic-speakers in the region? Which circumstances and which mechanisms were at play? Obviously, so-called ethnic roots could not have played a role because they simply did not exist. On the other hand, nationalists and their efforts were surely instrumental in this development. As elsewhere across the Habsburg Empire, Slovene nationalists worked tirelessly to make identification with the Slovene nation ubiquitous. They founded papers and political parties, established myriad associations of which teachers’ associations were certainly not the least important. Yet before 1848, the Slavic activists were not single-minded, and their support for Sloveneness was far from guaranteed. Some championed the Illyrian movement, and some still identified locally and worked to promote their provincial Slavic variety and a culture based on it. Others still, as the Slavic activists in Trieste/Trst for example, identified simply as “Slavs” (Slavjani; Ličen 2022). In short, although Slovene nationalism had potential, there were alternatives that were just as feasible if not more.

That is why we want to suggest that our research confirms the claim that the role of the Habsburg state in the establishment of ethnolinguistic nations—and the Slovene nation in particular—was in many respects crucial but only in the second half of the 19th century (Almasy 2018b, forthcoming; Stergar and Scheer 2018). When the state introduced its official linguistic taxonomy in 1849, it not only made existing attempts at statewide or provincial civic nationalisms futile (Decker 2023). It also decisively intervened in the debate about classifications, and this intervention was the start of the mostly inadvertent organizational grounding of some ethnolinguistic nationalisms. In other words, Alexander Maxwell was right when he argued that nationalism scholars would do well to study classifications precisely because they play such a large role in the history of nationalisms (Maxwell 2018b).

Yet, nation-building was complicated. Although later developments certainly suggest that the Habsburg linguistic taxonomy from 1849 and its long-winded and hesitant institutionalization had a noticeable influence, other contingencies and people’s agency played an important role as well. This was certainly not an algorithmic process, and identifications that did not conform to the state-sanctioned taxonomy retained some of their relevance all the way to 1918—sometimes longer. That, however, is a topic for another article.