Were there national school systems in the nineteenth century? The construction of a regionalised primary school system in Sweden

ABSTRACT The nation-state remains fundamental to our understanding of nineteenth-century schooling, which is commonly referred to in terms of national school systems or national education systems. While nineteenth-century school systems were often national in scope and promoted with the purpose of creating nationally minded citizens, this article examines whether such systems were national in the sense of being designed to impose national uniformity and standardisation on schools, teachers and pupils. Based on an investigation of public regulations of primary schooling in Sweden 1842–1920, this article shows that the Swedish school system promoted regional variation by officially sanctioning a wide range of school types, including ambulatory schools, junior schools and minor primary schools. As a result, this case study encourages considerate use of the term primary schooling and raises questions as to whether nineteenth-century school systems are, in this respect, more aptly described as regionalised rather than national.


Introduction
The rise of mass schooling during the nineteenth century is commonly referred to in terms of the rise of 'national education systems', 'national school systems' or 'nationwide systems of elementary education'. 1 In these contexts, the term 'national' has been used to emphasise at least three different aspects of these education or school systems. First, the term has denoted educational systems governed by national (central or federal) government. In this sense, the USA did not establish a national education system in the a stricter national curriculum intended to create a more uniform school system. As a result, this article raises questions about how other school systems managed regional variations and the extent to which they promoted national uniformity. In other words, in what respects were nineteenth-century school systems national?

Theoretical and methodological framework
The general theoretical inspiration for this article is the historical nominalism of Michel Foucault, which implies a disbelief in universals and inspires further investigations into the pluralities hidden under the entities we take for granted, including truth, power, reason, sexuality, and in this case national school systems. 9 Addressing this concept, this study is more specifically inspired by the expanding research on regions and regionalism. 10 In relation to the concept of a national school system, such studies are vital because they address the diversity within a nation, and the divisions of government introduced for administrative and fiscal reasons. 11 Studies of the administrative regions that nation-states created or reinforced have engaged with a wide range of questions. These include the societal construction of regions, their internal structure and their relationships with other regions and the central government, using terms such as regionalisation, regionalism, regional systems, regional cooperation and regional integration. 12 The historical dimension of regions has also been studied in terms of regionality and regional identification; that is, in terms of how regions have been experienced and discursively constructed, in relation to perceptions of the nation, for example, or in the unequal relations between centre and periphery. 13 Of particular importance for this article are studies on regional variations in schooling, which have, among other things, generated remarkable insights into the significant regional variations of educational provision in countries such as France, Italy and Prussia. 14 By examining education at the sub-national level, such studies of a 'Europe of regions' surely pose important questions regarding the sense in which there really were national school systems in the nineteenth century. 15 In this article, the term region mainly refers to the Swedish counties (län) (see Figure 1). Consequently, regional differences denote the accumulation of local differences that are evident at the level of regions. Although Sweden was not marked by the substantial ethnic, religious and linguistic conflicts that divided many other countries in the nineteenth century, there were certainly a wide range of social, economic, demographic and cultural differences across Sweden's counties, including those between sparsely and densely populated regions, and between those regions that benefited from the economic development of the nineteenth century and those that did not. These variations -indicating striking differences in population density, industrialisation level or relative regional GDP -were also understood at the time in terms of regional temperaments and dispositions: the population of Halland could be perceived as particularly frugal, while the inhabitants of Västerbotten were described as hardy, law-abiding and honest. 16 For the purposes of this article, I have used the distinction between regionalisation, which is the central government's handling of regional issues, and regionalism, which highlights societal and non-state processes. 17 As I argue in this article, the Swedish nineteenth-century school system is most aptly described as regionalised, as it was intended to manage (rather than eradicate) regional differences. This investigation of how the Swedish central government managed regional variations in schooling focuses on primary schooling -that is, the basic schooling provided for the general population during the nineteenth and early twentieth centuries. 18 In contrast to the main bulk of research into the local or regional features of schooling, which addresses specific villages, parishes, towns or counties, this investigation is based on national regulations issued by the central government concerning school types. Vital documents include the School Act of 1842 and the revised School Acts of 1882 and 1897, the school standards (normalplanerna) of 1878, 1889 and 1900, and the national curriculum (undervisningsplanen) of 1919. Starting with these, the study considers a range of documents published by the central government. These include parliamentary debate on school types and school standards, state school inspector reports (published from 1865 onwards), minutes from state school inspector meetings (1870-1881), official educational statistics when available (1868,, and the debate concerning school standards in a selection of prominent educational journals. 19 Of particular interest are educational statistics, because they indicate the school types created in educational policy and also provide evidence of the extent to which these types were used at local and regional levels. Sweden is a suitable starting point to critically examine assumptions of nineteenthcentury national school systems as one might expect its national school policies to promote a uniform and homogeneous school system. During the wars of the seventeenth and eighteenth centuries, Sweden developed a strong fiscal-military state to sustain its armed forces, and has often been referred to as a case (alongside Denmark and Prussia) in which the state constructed an education system. 20 In addition, schooling in Sweden was national in the sense that most children attended school in the late nineteenth century and that schooling was used to promote a national identity. 21 How these features were translated into policies addressing regional variations, however, remains to be seen.

School types and national conceptualisations of schooling
By examining the regionalised features of the Swedish nineteenth-century school system, this article builds on existing research in three ways. The first contribution concerns the conceptualisation of nineteenth-century school systems as national. Despite challenges from studies emphasising the international and transnational dimensions of education, the historiography of the nineteenth and early twentieth centuries has been firmly marked by notions of the nation-state. While the use of the term 'national' to denote schooling in the eighteenth century has been questioned by James Albisetti among others, primary schooling of the nineteenth and twentieth centuries has often appeared as a direct expression of the nation-state. 22 As Elsie Rockwell and Eugenia Roldán Vera have described this kind of methodological nationalism, landmark studies have tended to treat the state as a unitary agent of power and regulation, and have used the nation-state as a self-evident unit of analysis. 23 In paradigmatic publications, state ambitions to improve national school systems have been described as a 'bruising' adjustment between the powers of central government and local government, or as a tension among the state's attempts to impose uniformity, limited financial resources and the will of local communities. 24 An excellent example of this national conceptualisation of education is John Boli's landmark analysis of the Swedish national school system. According to Boli, state involvement was decisive in the creation of a school system intended to mould new citizens -productive and responsible economic and political citizens -from the ignorant children of nineteenth-century Sweden. Here, the state was able to provide and gradually tighten state supervision and control over an increasingly 'standardised, universal system of schools' that was basically in place by 1880, by which point 'a standard school year, school day and school curriculum had been established'. 25 Other authors have also emphasised the increasingly uniform and standardised nature of the nineteenthcentury school system. 26 By contrast, the present study indicates that the Swedish government did not ascribe to a simple policy of uniformity or standardisation, but instead attempted to incorporate regional variations as a formal part of the structure of the school system.
Second, this article contributes to the literature that has examined actual regional variations in nineteenth-century schooling. Acknowledging that schooling was highly decentralised in organisational and financial terms, previous studies have measured variations in literacy, enrolment rates, teacher density and educational expenditure, and have explored how these differences were affected by determinants such as wealth, land ownership and political voice. These studies have uncovered some remarkable patterns, including regional variations in the number of primary schools in Portugal (1878), literacy ratios in Spain ranging from 18.7% in rural districts to 50.6% in Madrid (1860), and differences in literacy that have traditionally been said to divide France according to a line drawn from St Malo in the west to Geneva in the east. 27 Such regional differences certainly raise questions as to whether they existed despite the best efforts of the central governments, or whether they were actually the intentional result of national educational polices. Third, this article adds to the somewhat overlooked history of primary school types. While rarely in the focus of historical investigations, evidence suggests that the main, often officially sanctioned, categories of nineteenth-century primary schools included those based on organiser (private/public), location (permanent/ambulatory; rural/urban), and gender (for boys/for girls). Apart from the simple divisions between two types of schools, such as the distinction between permanent schools (faste Skoler) and ambulatory schools (Omgangsskoler) in the Norwegian School Act of 1827, some primary school systems provided a wider range of schools. 28 In Belgium, distinctions were made between municipal schools, so-called adopted schools (private schools supported by municipal government) and private schools. 29 In the Netherlands, public statistics distinguished between public and private schools, and also between day and evening schools. 30 There were also more complex systems of school types. In Russia, primary schools included urban parish schools (school years 1-3), municipal schools (years 1-6, divided into three classes), primary schools (years 1-4), two-class primary schools (years 1-5) and literacy schools (years 1-2). 31 In Austria, the main primary school types were the major schools (Hauptschulen), intended for cities, and the minor schools (Trivialschulen), for market towns and rural communities. In order to cater to local demand, the Austrian primary school system also included parish schools (Pfarrschule) with an enlarged minor school curriculum; branch schools (Excurrendo-Schule), which were a kind of ambulatory school; and provisional schools (Notschule), which were schools with inadequate school premises. 32 In Australia, nineteenth-century primary schools included travelling schools, so-called half-time schools, house-to-house schools, provisional schools, public schools and superior public schools. The central government policies towards the wide range of primary school experiences that marked the late eighteenth and nineteenth centuries varied. In some instances, the government opposed certain kinds of schools. In Imperial Austria, the School Act of 1805 banned unauthorised so-called Winkellehrern, and in the Netherlands, the government opposed illegal schools established by Orthodox Calvinist parents. 34 In other instances, governments simply distinguished between officially sanctioned and unofficial schools. In Spain, the Montesino regulation of 1838 distinguished schools with a complete curriculum from those with an incomplete curriculum. A similar distinction was made in Italy between regular and irregular schools (scuole irregolari), where teachers in the latter type of school consequently remained at a rather insignificant 6% of the total staff of primary school teachers in the late 1880s. 35 The present article adds to this evidence, emphasising that nineteenth-century school systems did not consist of just one kind of permanent school. Instead, the nineteenth century featured a myriad of ambulatory schools, minor schools, parish schools, provisional schools and even irregular schools of various kinds. Thus, this article indicates not only the complexities that must be taken into account when using the concept of national school system in nineteenth-century settings, but also the need for care when applying the concept of primary school in an era marked by a diversity of school types.

Permanent and ambulatory schools
From the earliest legislation, the Swedish primary school system was designed to allow for local and regional differences. The first primary School Act of 1842 created the basic framework for schooling during the investigated period. It required Sweden's 2300-plus parishes to create school districts that were responsible for hiring teachers and establishing at least one coeducational school for both boys and girls. Like nineteenth-century School Acts in general, the School Act of 1842 consequently established a rather decentralised school system where the main responsibility for funding and organising schooling was placed on the local level. 36 In this School Act of 1842, local and regional variations in schooling were enabled by the introduction of a fundamental distinction between two kinds of schools: permanent and ambulatory. The term used for a permanent school was fast skola, indicating a school settled or fixed in one place. The term used for an ambulatory school was flyttbar skola; that is, a school (in the sense of an institution, not a building) that would move between two or more locations. According to the act, the ambulatory schools were intended for poor parishes that could not afford a permanent school, or for sparsely populated parishes where children would not be able to travel daily to a permanent school. 37 Ambulatory schools required less spending from school districts and offered a different experience for both teachers and children. These schools did not demand investments in new school buildings. Instead, by travelling between villages, the teacher could teach in homes or outbuildings provided by the children's parents. By shifting from village to village, one teacher could also teach more children by catering to, for example, 30 children in Village A for three weeks, 20 children in Village B for the next three weeks, and 15 children in Village C over the following couple of weeks. Despite this, the teachers' qualifications and their teaching were not supposed to differ from those of permanent schools. In both school types, teachers were required to hold a degree from a teacher training institute and they enjoyed the same minimum salary. In both permanent and ambulatory schools, children from the age of nine (at the oldest) would be provided with at least a minimum level of knowledge that included catechism, biblical history, reading, writing, arithmetic and church singing. 38 The differences between permanent and ambulatory schools had been debated prior to the School Act of 1842. There were those who mistrusted ambulatory schools. Wilhelm Gumaelius, a member of parliament, argued that the problem with ambulatory schools was their movable nature. Moving between places, these schools would be located in houses not fit for purpose, with children forced to huddle in cramped rooms where they would be disturbed by the house owners. Consequently, Gumaelius opposed a formulation of the School Act that allowed both permanent and ambulatory schools, since he believed the end result would be a proliferation of the cheaper ambulatory schools. 39 However, the members of parliament who favoured ambulatory schools won the debate, partly because of their perceived benefits in certain regions. For example, the minister Erik Agrell argued that permanent schools were not required in all parishes, but that they would be useful in sparsely populated areas. Instead, Agrell criticised the terms used: he questioned the 'long and non-Swedish word' of ambulatorisk skola, preferring the term flyttskola (moving school) or flyttbar skola (movable school). 40 Agrell was consequently in favour of a national school system, in the sense that even the terms used to describe it should be proper Swedish words. Nevertheless, he preferred a school system that was not nationally uniform but instead allowed local school districts to choose between ambulatory and permanent schools.
In the years that followed the School Act of 1842, the need for additional revisions of the school system became apparent to central government. While the cheap and flexible ambulatory schools proved a success -in 1847, 48% of enrolled schoolchildren attended these -schooling expanded at a slow pace. 41 Available statistics indicate that in 1847 only 57% of school-age children (aged 7-13) were enrolled in primary schools, and the low standards of schooling were apparent to contemporary commentators. 42 Various measures were suggested to promote enrolment, attendance and the quality of teaching. These included the introduction of additional school types, a solution that I will examine 38 SFS 1842:19, § 5, 7. 39  in the following section. In the context of this article, such proposals are notable, since they implied an attempt to strengthen schooling by effectively promoting further regional variations among school types.

Primary schools, junior schools and minor primary schools
The creation of additional schools apart from permanent and ambulatory primary schools was not a new idea. In the debate preceding the School Act of 1842, members of the Peasant Estate of the Parliament argued that parishes in remote and povertystricken areas should be allowed to establish village schools (byskolor), serving a function that seems to have been comparable to that of the Austrian provisional schools. 43 These village schools were supposed to be more loosely regulated in terms of the salaries and education of the teachers, and would provide the children with just a minimum level of knowledge. 44 At this point in time, however, this school type remained disputed and was not mentioned in the act of 1842. The main argument against such village schools was that they were perceived as an inferior school type. If sanctioned by the central government, parishes would settle for village schools instead of the more expensive permanent primary schools, which required actual school buildings and trained teachers. 45 After the School Act of 1842, this debate continued along similar lines. As in the case of ambulatory schools, the main purpose of introducing new school types was not to provide uniform school provision, but to promote enrolment by adapting schooling to local and regional conditions. Gumaelius, who had been critical of ambulatory schools, supported the introduction of so-called hjälpskolor (help or auxiliary schools) in the parliament of 1844-1845. He claimed that such schools, managed by suitable teachers without formal training, were necessary. The local conditions that rendered proper primary schools impossible could not be changed -only addressed in the best possible manner. 46 Similar suggestions were voiced elsewhere. Among the most influential were those of the educationalist Torsten Rudenschöld. He promoted the use of so-called district schools (roteskolor). These were intended to lower the cost of popular education by employing young district schoolmasters (aged 15-20) who lacked formal teacher training, would accept a low salary and would teach in the homes of schoolchildren. By investing in many of these district schools, instead of the expensive permanent primary schools, Rudenschöld believed that a beneficial reduction in the length of school routes and class sizes would be accomplished. 47 As a result of this debate, and with the intention of promoting schooling in rural areas, so-called minor schools (mindre skolor, literally 'smaller schools') were launched in 1853 as a distinct school type. These schools, either permanent or ambulatory, were originally intended for remote areas, from which children could not travel 'without difficulties' to 43 Cvrcek, Schooling under Control, 54. 44  regular primary schools. As the vague phrasing 'without difficulties' indicates, these minor schools were a less regulated, and certainly cheaper school type. Unlike teachers at primary schools, teachers at minor schools had no minimum salary and were not required to have a degree from a teacher training institute. Instead, their qualifications included merely being able to read and write and having lived a Christian life. 48 In parliamentary debate, these schools were described as an emergency measure that should be managed by women, so-called school-mothers (skolmödrar), who, with their tender disposition, were more suitable for this task than men. 49 The purpose of these minor schools was revised in 1858. Instead of targeting children living at a considerable distance from primary schools, minor schools were intended to provide the youngest school-age children with their first limited amount of knowledge before they entered the ordinary primary schools. In that respect, the minor schools were redefined as a kind of lower primary school. Their purpose was partly to counter the resistance in some regions towards sending younger children to school, especially over long distances, and partly to strengthen the regular primary schools by providing them with preparatory schools. A specific state subsidy for minor schools was distributed from 1858 onwards in order to promote this school type, which probably added to the increasing popularity of this school type (see Figure 2). 50 Despite this new definition of the minor schools, their position remained vague. There were no regulations stating the age of the children expected to attend them, or the number of years they were supposed to remain in minor schools. 51 Therefore, further decisions were made by the central government. In 1863, småskola (junior school) was established as the new official term for this school type and attempts were made to clarify the relationship between junior schools and primary schools. In school districts with junior schools, children were only allowed to attend primary schools if they were already able to read to an acceptable standard. 52 The junior schools -which through these efforts of the 1850s and early 1860s became 'recognised as a state institution' according to a member of parliament -were defined as a preparatory stage for the ordinary primary schools; a kind of lower primary school. 53 As a result, the 1860s also saw the first teacher training institutes for junior school teachers, even though knowledge of the Bible and the three Rs remained the criteria for being eligible for a teaching position at junior schools. 54 However, these efforts to define junior schools remained difficult to implement. In 1863, a state school inspector in Uppsala diocese noted that it was difficult to distinguish junior schools from primary schools, while another inspector argued that the junior school was a 'masked villain' (kascherad skurk) that included all kinds of unwanted schools. 55 In 1868, another school inspector provided a more nuanced account, arguing 48 SFS 1853:65, § 1:3, 4:4, 6:2. 49  that primary schools and junior schools often had varying and problematic roles. Primary schools could be responsible for the preparatory education that junior schools were supposed to have catered for, and junior schools were either focused on preparatory teaching (mainly catechism and biblical history) or took on the role of minimum-schools (minimi-skolor), providing a basic level of schooling to any child, regardless of age. 56 The proliferation of informal terms denoting such schools in between junior schools and primary schools, including replacement schools (ersättningsskolor), village schools (byskolor) and auxiliary schools (hjälpskolor), added to the confusion. The unclear and overlapping definitions of these schools were known to school inspectors. One school inspector noted that so-called replacement schools in previous years had been 'sailing under a false flag' as junior schools. Another inspector argued that there was even another school type, which could be described as an intermediate between junior schools and replacement schools. 57 Although these school types remained unofficial, in the sense that they were not mentioned in public regulations, they were used and discussed by school inspectors. As this discussion indicates, nineteenth-century school inspectors were not only (or even primarily) a symbol of a 'uniform socialization process' as John Boli argued, but could also promote school types that added to the regional variations of the Swedish school system. 58  school inspector meeting of 1870, replacement schools were defined as an intermediate or a mixture between junior and primary schools. 59 There were also attempts to distinguish between two kinds of replacement schools: (1) proper replacement schools that, as their denotation indicated, were true replacement schools that provided an education of almost similar quality, and (2) cheaper replacement schools that could not replace regular primary schools, but only provide schooling of significantly lower quality. 60 Apart from these attempts to define unofficial school types, school inspectors discussed the officially sanctioned school types. Some preferred solutions that reduced the number of school types. One suggestion involved distinguishing junior schools from primary schools by limiting the former to basic teaching in reading, writing, arithmetic and religion. There were also those who promoted a school system that would further adapt schools to local conditions. At the school inspector meeting of 1875, some inspectors argued that schools of a type somewhere in between primary schools and junior schools should not be banned, but should remain somewhere in between official and unofficial; they should be tolerated, but not sanctioned in official documents. 61 Some inspectors even suggested that certain unofficial school types should be formally recognised, as they represented a genuine effort by poor parishes to strengthen their school systems. 62 The replacement schools were an example of such a school type. Although not acknowledged by the School Acts, replacement schools were recognised by some school inspectors, and in some areas they were an important part of the school system. For example, in 13 school districts in Örebro County, the school inspector reported that replacement schools were the most common school type after regular primary schools. As many as 31% of the total number of schools were listed as either permanent or ambulatory replacement schools. 63 The solution adopted was the third suggestion mentioned above, which embraced regional differences in schooling. In December 1875, a Royal statement recognised socalled minor schools (mindre skolor) as an official school type in addition to junior and primary schools. 64 In the new School Act of 1882, they were renamed minor primary schools (mindre folkskolor). While junior schools were to provide school-age children with their initial knowledge, these minor primary schools were intended for villages where children could not attend the ordinary primary schools 'without difficulty'. This phrasing -which implied that these school could be applied in all kinds of situationsmeant that the minor primary schools became an officially recognised, albeit rather vaguely defined, category that covered at least some of the schools that had previously been known as replacement schools, village schools and auxiliary schools. According to the School Act of 1882, the teaching at these minor primary schools was supposed to be equivalent to that of primary schools, but could be reduced in accordance with local circumstances. 65 Compared with regular primary schools, the regulations regarding the teaching staff at minor primary schools were more flexible. As in junior schools, teachers at minor primary schools were not required to hold a degree from a teacher training institute. In addition, neither the employment nor the dismissal of junior and minor primary school teachers was regulated by the School Act of 1882, which strengthened the position of school boards in relation to these teachers. 66 Because the salaries were lower -the minimum level for schools to be eligible for state subsidies was 200 SEK for these two categories of teachers in 1882, compared with the 500 SEK minimum wage of primary school teachers -these teacher categories were also rapidly feminised. By 1900, 88% of minor primary school teachers and 97% of the junior schoolteachers were female, while 28% of primary school teachers were female. 67 By adding the minor primary school, the school system was, by the early 1880s, structured according to three main school types: (regular) primary schools, minor primary schools and junior schools, either permanent or ambulatory. Their relative importance varied across regions. As Figure 3 shows, the Swedish school system featured significant regional differences that transcended indicators such as enrolment or number of teachers. Particularly striking examples were the taste for ambulatory primary schools in the school districts of Jönköping County, the emphasis on permanent junior schools in Kopparberg, the prevalence of ambulatory junior schools in Jämtland, and the number of ambulatory minor primary schools in Norrbotten County. As these regional differences indicate, the different school types provided a national framework for a regionalised school system.

Recognising all schools or promoting certain school types?
This system of three permanent or ambulatory school types, established during the late 1870s and formalised in the School Act of 1882, was further elaborated by the school standard (normalplan) issued in 1878 by the Ministry of Education and Ecclesiastical Affairs (Ecklesiastikdepartementet). This plan did not provide a single standard for primary schools; instead, it provided recommendations for a total of 11 school types: five types of junior schools and six types of primary schools. These standards included guidelines not only on the content of teaching, but also the organisation of school days, school weeks and school years. In total, the school standard featured 20 schedules that described five-and six-day weeks and included schools where children were taught both each school day and every other school day. 68 Although minor primary schools had been acknowledged as an official school type in 1875, and thereafter by the School Act of 1882, the minor primary schools were not defined as a 'normal school type' in the standard of 1878. Therefore, they were not included in the standard, even though its authors admitted that they fulfilled an important role in some regions. 69 This distinction between normal and sub-normal school types satisfied politicians and schoolmen who saw junior and primary schools as the basis for the proper development of the Swedish school system. Several statements in school inspector reports expressed satisfaction with the guidance that the standard provided, describing it as a step forward in the development of the Swedish school system. 70 Nevertheless, the school standard of 1878 received mixed reviews. Although it included 11 school types, critics argued that it promoted an excessively homogeneous school system. Some teachers and state school inspectors argued that the standard was too strict, describing it as 'the primary school forced into a uniform' and questioning what was perceived as the ambitions of the Ministry of Education and Ecclesiastical Affairs to regulate schooling 'into the smallest detail'. 71 One school inspector noted that the school standard did not include the three-year junior schools, which he felt represented the best option for rural school districts. 72 Attempting to find the middle ground, a teacher stated that although the standard contained useful information, it was too restrictive and should therefore be treated merely as providing indications or advice. 73 Among the opponents of the school standard was the member of parliament S. A. Hedlund. His critique dealt with the core question examined in this article: should school systems be nationally uniform and standardised? Hedlund's position was that even the 11 school types set out in the standard could not meet the demands of the diverse local and regional conditions in Sweden. Consequently, he questioned the ambition of the standard to reduce this diversity 'with a stroke of the pen', without the necessary knowledge of the varied local conditions of Sweden. Thus, Hedlund argued, the school standard was an example of how a standard should not be formulated. 74 The problems of an excessively homogeneous national school system were also discussed from the standpoint of existing schools. In the keynote to the seventh school inspector meeting, the Minister of Education and Ecclesiastical Affairs Carl Hammarskjöld stated that regulations had to adapt to an infinite diversity of traditions and geographical conditions a solution fitting for the southern parts of Sweden might not work at all in the north. 75 The school inspector J. H. Bergendal supported this argument by discussing the problems that the standard created in the southern county of Scania (Skåne). Although the school system of Scania was comparatively strong, the school standard's detailed prescriptions did not match the traditional organisation of primary schools in the county. These included a kind of primary school with a shorter school year, and a school type in which children from junior schools and regular primary schools were taught simultaneously. Bergendal also criticised the school standard for not including guidelines for ambulatory schools moving among three villages. Although he admitted this school type was problematic, he claimed that it remained popular and was therefore in need of guidelines. 76 Responding to this criticism, and in an attempt to strike a better balance between uniformity and regional heterogeneity, a new school standard was published in 1889. This revision was described as the result of experiences since 1878, particularly the opinions of the state school inspectors. Instead of a limited selection of accepted school types, the standard of 1889 included all types, even those that, according to the authors, 'barely meet the requirements'. 77 This choice was motivated by the need to adjust the school system to local contexts. As a result, the new school standard included a total of 15 school types: five types of junior schools, eight types of primary schools and two types of minor primary schools. Among these were the much-criticised ambulatory primary schools that moved between three locations -and as a result clearly reduced the children's number of school weeks per year in each location -and minor primary schools (mindre folkskolor). The standard of 1889 also included an extended junior school that, by covering three school years, could be described as a new intermediate school type in between junior and minor schools. 78 Like its predecessor, the school standard of 1889 provided several recommendations for standard curriculums, school days and school years: it provided standards, not a single standard. In total, the school standard entailed 33 schedules, some with as many as six variations. These included junior schools where the second year was taught on Mondays, Wednesdays and Fridays, and the first year on Tuesdays, Thursdays and Saturdays, and school types in which teachers shifted between different groups of children across the semester. The school weeks remained five or six days long. While the School Act of 1882 noted that a school year should be at least eight months, this practice of half-time schooling (every other week or every other day) created school years of varying length. 79 Since the school standard of 1889 included additional school types, it is not surprising that it was criticised for being too permissive. One of its strongest opponents, the teacher and school politician Alfred Dalin, was given ample room in the teacher journal Svensk Läraretidning to criticise both the standard and its defenders. His arguments indicated that there were certainly educationalists and politicians who wanted to establish a more uniform national school system. Under the heading 'False Label', Dalin noted that the Swedish government had not set a single standard for its schools in the school standard of 1889. Therefore, those expecting a single curriculum from a public document named 'school standard' (normalplan) were bound to be disappointed. Instead, Dalin stated with sarcastic hyperbole, the purpose of the school standard was seemingly to include 'all possible or at least currently existing school types', and that it did not distinguish good school types from bad. As a result, Dalin argued, the standard of 1889 effectively became 'a textbook of how school districts, in the most convenient and cheapest way' could manage their obligation to run schools. 80 However, some perceived the standard of 1889 differently. Some argued that it was not very different from its predecessor -in some respects, it even placed higher demands on schools. 81 There were also active supporters of the new school standard's inclusive vision of the Swedish school system that allowed for a plurality of school types. The influential educationalist Carl Kastman claimed that the new school standard had many advantages. Kastman argued that including junior schools with three school years had many benefits, not least that such a popular school type had finally been provided with an official curriculum. Kastman made similar claims regarding the minor primary schools, which were, in many locations, believed to be necessary for the expansion of school provision. If minor primary schools had not been allowed, regular primary schools would not have been established in their place. Under such circumstances, Kastman preferred to include such schools in the school standard. 82 Regardless of this debate, the school standard of 1889 enabled both the development and the mapping of a school system that featured significant regional variations (see Figure 4). Some regions featured primarily a combination of permanent primary schools and permanent junior schools (for example, Malmöhus, Södermanland and Uppsala), while ambulatory schools were more popular in other regions (for example, Värmland, Jönköping and Kronoberg). In yet other regions, the so-called minor primary schools were common. Notable in that respect was Västerbotten County, where minor primary schools had largely replaced junior schools during the years 1882-1900 to become the basis of its school system. As Figure 4 indicates, nineteenth-century school policy did not imply uniformity, but instead promoted striking regional variations.

Creation of a more uniform national school system
As is evident from above the nineteenth-century primary school system in Sweden was marked by significant regional differences. These were the result of local demand but also due to central government policies that created a wide range of school types adapted to local conditions. In this respect, it was a regionalised rather than a national school system.
However, from the turn of the century these policies did change, and attempts were made to create a more homogeneous national school system. The social, cultural and political contexts of this change require further explorations. Within the scope of this article, it may suffice to note that the decades around 1900 were marked by improved communications, converging regional differences in, for example, salaries, and a changing understanding of the scope and target of central government intervention. 83 In 1900, the controversial school standard of 1889 was replaced by a new set of regulations, which might be described as a compromise that limited the officially sanctioned school types to 13. This restriction targeted what were described as the least effective and popular school types, removing the ambulatory schools that shifted between three locations and schools in which six classes of children were taught simultaneously. 84  The restriction was also positively received by the teaching press, which welcomed the fact that the 'worst school types of the 1889 school standard were removed', and that school types that involved a reduced number of school hours had been clearly defined as exceptions. 85 Although the homogenisation efforts implied by this reduction of school types were limited, they marked the start of a series of reforms that aimed to create an increasingly uniform national school system. The slightly more restrictive stance of this school standard was given additional weight by new regulations for state school inspectors (1904), which stressed their responsibility to determine whether children were being given the correct amount of teaching in relation to local conditions. 86 These efforts to improve the organisation of the Swedish school system also included the introduction of a national board of primary schooling (Folkskoleöverstyrelsen) in 1913, increases in central government grants through the grant reform of the same year, and the creation of a government committee to investigate primary school matters in 1906. 87 In addition to these efforts, a critical debate regarding individual school types continued to highlight the problems of the Swedish school system. In 1903, an article entitled 'The Decline' (Kräftgången) described how the share of the problematic minor primary schools had actually grown during the period 1896-1900 -a development termed as 'abnormal'. 88 This was also a debate that, based on available statistical compilations and individual examples, concerned regional variations in the provision of schooling. In the aforementioned article, Västerbotten County was described as most abnormal in terms of minor primary schools. These regional variations were sometimes described with reference to a formulation made famous by the explorer Henry Morton Stanley in the title of his book In Darkest Africa, published in 1890. In Sweden, this formulation was used to describe culturally 'backward' regions in Sweden but was also applied to discussions of unacceptable regional differences in schooling. For example, the phrase 'darkest Småland' was used to describe unacceptable examples of schooling in that area. 89 Initiatives were also taken to reform the school standard of 1900. The Minister of Education and Ecclesiastical Affairs Hugo Hammarskjöld proposed an inquiry for a new school standard in 1906, but, because of various practical and administrative circumstances, the inquiry was still unable to deliver a proposal in 1914. In 1917, however, the finalisation of the proposal was promoted by a letter from parliament concerning a reorganisation of primary schooling with three main purposes: systematic restriction of the use of inferior school forms, prevention of poor or no school attendance, and training of sufficiently educated primary school teachers. 90 As a result, the curriculum of 1919 (Undervisningsplan för rikets folkskolor) was enacted. While still covering various school types, it clearly reduced the amount of variation possible by distinguishing between the main school types (Huvudformer) and the exceptions (Undantagsformer). The four main types (A, B1, B2 and B3), which were termed primary schools (folkskolor), provided full-time teaching. All of these primary schools gave children six or seven years of schooling, taught by trained primary school teachers, albeit with some variation in the organisation of the classes. The exceptionswhich were only permitted in rare circumstances -consisted of three kinds of parttime primary schools (C1-C3), and three kinds of minor primary schools (D1-D3). Part-time primary schools were allowed only when the distance between home and school was an obstacle to full-time schooling, and minor primary schools (defined here as primary schools run by junior schoolteachers) were allowed only in small and remote villages. 91 By making this distinction between normal and exceptional school types, the curriculum of 1919 demanded a level of uniformity far beyond that of previous school standards. It required school districts, under all but the most exceptional circumstances, to establish primary schools where all children attended school full-time, with teachers who were trained primary school teachers. Starting from a primary school system that consisted of a wide range of school types, the 1919 curriculum stated that all schools should be regular primary schools. To promote uniformity, this curriculum also included timetables for the number of hours spent on each school subject per week, as well as additional comments on how the teaching was to be organised. These included lesson length, breaks and morning prayers, and comments on the method and content of the teaching of each school subject. 92 Although the curriculum of 1919 had a mixed reception -ranging from controversy concerning the reduction in the hours and content of religious education to more minor discussions regarding the methodological assumptions behind the teaching of music -it had a definite impact on the Swedish school system. 93 According to a government inquiry of 1940, the curriculum brought primary schooling into a new era in which parttime schooling and school types without properly trained teachers were marginalised. Although this development was slow to take effect in certain regions, the share of children enrolled in school types deemed problematic was significantly reduced within five years. 94 Beyond its short-term impact, the ambitious curriculum of 1919 certainly inaugurated a new epoch in the history of Swedish school policy. From this point on, all Swedish children were not only supposed to attend school but also to attend similarly organised primary schools.

Conclusion
While the Swedish nineteenth-century school system was certainly national in the sense of being overseen by central government with the purpose of creating national citizens, this article has shown that it was not the intention of central government to create a uniform national system of primary schooling. Instead, the Swedish government sought to promote schooling by sanctioning a wide range of school types. Starting from a simple distinction between permanent and ambulatory schools, defined in the School Act of 1842, new school types were created to promote school enrolment and to adjust schooling to local demand. While some school types remained proposals (such as village schools, auxiliary schools and district schools), and others remained unofficial, recognised only by school districts and school inspectors (including the so-called replacement schools), the number of officially sanctioned school types increased to include various kinds of junior schools (småskolor), minor schools (mindre skolor) and minor primary schools (mindre folkskolor).
Instead of creating a uniform school system based on one kind of primary school, this school system featured a wide range of officially sanctioned school types. The school standard of 1878 included 11 types of primary and junior schools, but was nevertheless criticised for imposing an overly restrictive uniformity on schools. As a result, this school standard was replaced by the standard of 1889, which featured 15 school types. Criticised for being too inclusive, the next school standard of 1900 included 13 school types. The first attempt to create a more uniform system of schooling was the national curriculum of 1919, which consisted of four main types of regular primary schools. Barring exceptional circumstances, all children were, for the first time, expected to attend full-time regular primary schools with properly trained teachers.
Intended to adapt schooling to local conditions, this wide range of school types promoted a publicly sanctioned regional diversity in schooling during the nineteenth century. As is evident from Figure 3 and Figure 4, the cheaper junior schools, minor schools and ambulatory schools were preferred in some regions, while school districts in other regions chose to spend more on permanent primary schools. Despite being national in scope, the Swedish school system was certainly regionalised in this respect. Instead of imposing a national standard, School Acts and school standards in Sweden promoted regional diversity by offering local school districts a menu of school types to choose from.
This regionalised policy should not be perceived as merely a pragmatic adjustment that could be contrasted to a more idealistic promotion of a uniform national school system. The latter was not an alternative in a debate that focused on what school types should be included and the range of regional differences in schooling that should be promoted. As is evident from the discussion in this article, even a school system featuring 11 school types could be perceived as overly restrictive, also by school inspectors.
As a result, this article encourages us to reflect on the concepts used when describing nineteenth-century schooling. First and foremost, the article promotes considerate use of the notion of national school systems. In this respect, I have raised the question of whether it is generally more apt to describe nineteenth-century school systems as regionalised systems that promoted or sanctioned regional differences, rather than as national systems that promoted a single national standard. While James Albisetti has questioned whether it makes sense to denote schooling as national prior to 1815, I reflect on the extent to which it makes sense to describe schooling prior to the twentieth century as national. 95 Second, this article promotes careful use of the concept of primary schools. Although the twentieth century saw the creation of a national school system in Sweden consisting solely of primary schools, this was not the case in the nineteenth century, when a wide range of school types flourished offering a wide range of school experiences. Although Sweden might be an extreme case, varying kinds of school types were also established in other countries during the nineteenth century, as is evident from the introductory sections of this article.
These suggestions are vital because the concepts we use determine our interpretations of the past. 'Primary school' tends to stress a uniformity of schools that overlooks the amazing richness of local and regional traditions in nineteenth-century schooling, including the history of nineteenth-century school types. In the Swedish case, almost half of the schoolchildren attended an ambulatory school in the 1850s (see Figure 2) and not the permanent primary schools that dominated the twentieth century. The term 'national' encourages us to focus on national conditions, standardisation processes and the agency of national (central or federal) government that passes and implements laws and regulations with the assistance of state subsidies and school inspectors, setting national governments against regional and local levels and emphasising tensions between regional diversity and national hegemonic uniformity.
In this context, this article promotes other concepts that can serve several functions in addressing this kind of methodological nationalism. Concepts such as 'school types', 'region' and 'regionalisation' offer an alternative to studies that stress national conditions and central governments' efforts to implement national policies on local levels. Instead, such terms enable us to explore the hybrid nature of school systems where national policies are not only resisted but used and refashioned at a local level, and where national policies may also stem from local or regional experiences. The case of school types certainly highlights such multifaceted interactions among the national, the regional and the local. Instead of uniformity, such studies furthermore stress local and regional diversity and enable us to examine the complex roles of regulations, state subsidies and school inspectors in nineteenth-century school systems. As this article has shown, even school inspectors could promote regional diversity in schooling.
As a result, this article encourages further research that extends beyond a national framework to examine processes of regionalisation rather than the nationalisation or internationalisation of education. Unlike such investigations this article is an argument to continue the exploration of nineteenth-century entanglements among local, regional and national levels. This includes further investigations into the local and regional features of national policy, and the heterogeneous experiences of nineteenth-century schooling. These questions include the social, economic, demographic or cultural basis for the shifting distribution of school types indicated earlier, and their impact on the expansion of schooling. These questions also include the school types studied in this article: Did such policies promote or hamper the development towards schooling for all? 95 Albisetti, 'National Systems: Europe', 149.