Teaching and family: either or both? Work and family among women primary school teachers in northern Sweden, c. 1860–1937

ABSTRACT

This study examines the life and career trajectories of teachers recruited to primary schools in coastal northern Sweden. A variety of historical sources are combined to construct collective biographies which include approximately 500 primary school teachers divided in four birth cohorts. Main findings show that women teachers with a teaching certificate came to constitute a majority of the teachers from the 1870s and onwards in the rural region studied. These women teachers were increasingly inclined to form a family of their own and the share that never became mothers decreased continuously during the studied period. Furthermore, among those who formed a family the practice to return to work when the children were older increased over time. Towards the end of the 19^th century having children became an increasingly available option among women teachers and a few decades later it had become common. In sum, through the combination of multiple sources, this study suggests that women primary school teachers in Sweden were able to combine family and teaching well before these rights were protected by law in 1939.

KEYWORDS:

• women teachers
• work and family
• life history
• collective biography
• source pluralism

Introduction

In 1937, elementary school teacher Rudolf Lundmark compiled a teacher register of all primary school teachers that had been teaching in Skellefteå, a parish in coastal northern Sweden. Lundmark’s register was intended to be a part of the upcoming 100^th anniversary of the first Swedish school act (1842), to be celebrated five years later.¹ By collecting and combining several historical sources, Lundmark was able to obtain comprehensive information about the teachers. The information on each teacher included name, year of birth, teacher degree, years of service per school, and the reason for leaving the profession. When completed, the register included almost 500 individuals that had been engaged in teaching.² Through the teacher register, and by addressing other sources, this study will explore the link between teacher’s professional and private/family life and how this link changed in an era of school system establishment and societal development.³ There is a special focus on the link between parenthood and teaching and how this link functioned in practice and varied depending on the level of teacher training and gender, particularly with an interest in the lives of women teachers.

Inga Elgqvist-Saltzman has studied women elementary school teachers in Sweden that graduated from the same teacher training college. By investigating an alumnae survey from 1932, Elgqvist-Saltzman found that approximately one third (n = 503) of the 1,462 women teachers, who graduated between 1878 and the end of the 1920s, had married. Among those that did, 40% had left their position upon marriage, although many of them returned to teaching later in life. Additionally, Elgqvist-Saltzman found that the share of women who left their teacher position upon marriage was considerably lower (23%) if they married a male teacher.⁴

In contrast to Elgqvist-Saltzman this study focuses on both groups of primary school teachers in Sweden: junior school teachers (småskollärare) and elementary school teachers (folkskollärare). Throughout the studied period children generally started primary school at the age of seven, although it was allowed to postpone school start by up to two years.⁵ Junior school teachers were primarily intended to teach the first two grades (age seven to ten) while the elementary school teachers conducted teaching in grades three to six (age nine to fourteen). From the late 1870s and onwards, most junior school teachers undertook teacher training for two years while the teacher education for elementary school teachers was increased from three to four years in 1877.⁶ In the rural area studied, junior school teachers frequently taught in all six grades until the 1930s, commonly as the sole teacher in so called smaller elementary schools (mindre folkskolor), i.e. a one-room rural school.⁷ Both groups of teachers persistently worked to improve their working conditions and salaries. Previous research has shown that Swedish primary school teachers in the late 19^th century often had secondary occupations to ensure their livelihood.⁸ There was also great differences between the two groups of teachers. In 1900 a junior school teacher earned less than half of an elementary school teachers. Twenty years later the difference had been reduced to two-thirds.⁹ Compared to other occupations which required secondary or post-secondary education salaries among primary school teachers were low, especially for junior school teachers prior to 1920.¹⁰ I refer to previous research for further information on the two groups of primary school teachers.¹¹

Issues of how to combine work and family have been recurrently discussed in both historical and contemporary settings.¹² As for this study, relevant research focuses on how women, especially teachers, combined paid work and family in the decades before and after the turn of the 20^th century. In pioneering work from the 1970s, Louise Tilly and Joan Scott found that women teachers in France and England were prevented or discouraged from marrying. In France, it was forbidden until 1900, while officials in England up to the 1940s could stop women teachers from teaching if they married. Due to a shortage of teachers, a majority of the women teachers in France were married in 1922, but whether they bore any children is not examined by Tilly and Scott. After World War II, Tilly and Scott found a changing pattern according to which women increasingly combined family with work.¹³

France was far from the only country that banned women teachers from marriage during the late 19^th and early 20^th century. National or regional bans on married women teachers were applied at different points in time in the United States, Ireland, Australia and New Zealand, but never in Sweden.¹⁴ Geraldine J. Clifford criticizes the emphasis on the impact of marriage bars in studies such as Goldin’s (cf. infra). According to Clifford, married women teachers in service continued to be common even in areas where marriage bars were active and consequently some of these marital bans could be considered ‘dead letters’. Clifford argues that life histories and comparative approaches adds complexity and perspectives where policy studies fall short.¹⁵ As mentioned above marriage bars were never enacted in Sweden, but groups that wanted to ban married women from teaching existed from late 19^th century until the late 1930’s. In 1893, the national church council agreed to work for the introduction of a marriage bar for women teachers.¹⁶ Women elementary school teachers experienced a higher level of employment security than their junior school counterparts, who could not obtain permanent positions until 1918. In spite of the absence of a marriage bar in Sweden, women teachers that continued to teach after marriage recurrently received criticism in public debate during the study period. Critics objected that families where both spouses worked would earn a double salary which could have the consequence that another, sole family provider (i.e. a male teacher) would be left unemployed. In the late 1930’s these discussions came to an end, partly because the teacher surplus was turning into a teacher shortage, but also because employment security for women that got engaged, married or pregnant was legislated in 1939.¹⁷

Alison Oram pointed out that the marriage bar did not keep London women teachers from marrying in the 1930s. The number of women teachers that resigned upon marriage was more than twice as high when a marriage bar was active. By drawing on life-histories, Oram recounts how two women continued to teach after marriage in the mid-1930s as their marriages coincided with the abolishment of the marriage bar. However, both of them left their position when they had children a few years later.¹⁸ In another study of women teachers in London, Dina M. Copelman found that they had more working years, married later than other women did, and were more likely to keep their position upon marriage. Copelman’s results suggest that those women teachers could continue to be active in the two spheres of private and professional life although not without difficulties.¹⁹

This study will provide new results and perspectives on the link between teaching and having children by placing parenthood rather than marriage in the centre of attention. Life trajectories of about 500 teachers are analysed both on a group and individual level. The study’s approach covers approximately eighty years, from the 1860’s to the 1940’s, and includes all qualified primary school teachers in one geographical area.²⁰ The motivation behind the decision to start a family or not on an individual level remains unknown in most studies of historical populations. Researchers should therefore refrain from assuming that all individuals had a preference to find a partner and have children. Remaining childless could have various explanations, such as never wanting a family, never meeting a suitable spouse or issues of infertility. However, previous studies of historical populations in both Sweden and elsewhere have found that women with post-primary education experienced an ‘educational reduction’ by showing a lower probability of having children and getting married than other women.²¹

Influential studies from the 1960’s and 1970’s by Sten Carlsson and Gunnar Qvist discuss the increasing rights and labour market attainment among women. A combination of increased legal rights and changing norms meant that married women with a registered occupation emerged during the time examined. Unmarried women constituted a large group throughout the study period, in both 1860 and 1920 nearly half (48%) of the women aged 20 to 40 years had not married. Finding a spouse seems to have been particularly difficult for daughters in the middle and upper classes, a social background that was more common among women elementary school teachers than junior school teachers.²² As mentioned above, a lower fertility among middle-class women including teachers prevailed until the mid-20^th century baby boom. One previous study, which examined the same part of Sweden as this study, concluded that a considerably lower fertility among middle-class women (nurses, teachers and office clerks) was only true for women born prior to 1920. Furthermore, it was shown that the proportion of women teachers that remained childless decreased slowly over the decades around year 1900, moving from 45% (b. 1880–1900) to 39% (b. 1901–1920). In the following group, women teachers born 1921 to 1940, the share of childless women teachers plunged to 17%, almost reaching the same share as women with no registered occupation (15%).²³ Thus, it can be expected that the women teachers included in this study, all born prior to 1915, demonstrate a level of childlessness around 40%. Other studies within historical demography confirm this pattern and conclude that educational differences in fertility related to class were large during the late 19^th century and beginning of the 20^th century but started to decrease towards the end of the period investigated here.²⁴

The decades examined in this study correspond to the most intense decades of the fertility decline in Sweden when the number of children per woman decreased from approximately 4.5 in 1870 to 1.8 in 1930. However, in Västerbotten region, where the parish of Skellefteå is situated, the decline progressed slower than the national average. Additionally, the social and economic environment in Skellefteå was characterized by social stability, with village life with forestry and farming being the most important economic sectors.²⁵ Regarding primary education the area under study stands out through its high proportion of ambulatory schools which exceeded 50% in year 1900, especially ambulatory minor primary schools were more frequent here than in other parts of Sweden.²⁶

As opposed to previous research, this study provides an in-depth study of one specific group of women, namely primary school teachers, rather than focusing on a national or aggregated level. By focusing on a geographical region that experienced a later decrease in fertility and was regarded as social stable we can examine how women teachers responded to these circumstances. The aim of this study is to examine how the link between teacher work/career and parenthood developed over time (1860–1937) and to obtain results of what typical teacher life trajectories looked like and how these shifted across time. The following two research questions are addressed in the study:

1. To what extent did teachers enter parenthood and how did this vary with regard to gender, level of teacher training and birth year?

2. After becoming a parent, how did the possibility of returning to teaching change over time for women teachers and depending on the level of teacher training?

The practice of returning to paid work after becoming a mother has rarely been addressed beyond the individual level among women in historical populations, most likely because this information is difficult to attain. This paper draws on different techniques, source pluralism in combination with life course approach and life history examples, to study this question. The purpose of the research design is to increase our understanding of the historical processes and norms which affected decision making about family and work among women teachers.

Sources and methods

The research questions of this study call for a methodological approach addressing multiple sources, several methods and a multilevel analysis. I will first introduce the different sources compiled and examined in this study. This is followed by a presentation of source pluralism and the meso-level approach through collective biographies and life history, all of which have been employed throughout the study. Finally, I provide a brief explanation as to why one fifth of teachers, mostly unqualified teachers, are excluded from the main study.

Accounting for hundreds of teachers over a long period of time without neglecting the collective or individual experience imposes high requirements on the sources and the researcher. One challenge arises by the fact that sources that provide aggregate information, such as censuses and parish records, often provide less information about each individual, while sources such as letter offers more information, but for a smaller number of individuals. Consequently, the prerequisite of ‘knowing more about the many’ without reducing the level of detail regarding life experiences creates a need to combine a variety of sources that provide complementary information about the same individuals. Meeting these requirements, I have adapted the concept of source pluralism and collective biography when designing this study.²⁷ The 1937-register by Lundmark marks the end year for including new teachers in the study. However, for teachers that had an employment record in 1937 or earlier I have added information to their life trajectories from later sources. To write the teacher register, later labelled Teachers in Skellefteå since 1842, Lundmark mainly used data from three primary sources: parish records, school board meetings and staff service registers from specific schools.²⁸ Compiling three sources enabled Lundmark to construct an overview of both primary school teachers in service that year (1937) and teachers that had previously left their position and why.

In this study, 17 additional sources have supplemented Lundmark’s 1937-register (see Figure A in Appendix): six censuses, nine national teacher registers, and two teacher biographies; all of which were not available to Lundmark at the time he constructed his registers.²⁹ These additional sources serve two important purposes. First, in some cases these sources provided data that remained absent in the primary sources utilized by Lundmark. Second, the independent primary sources have worked to confirm, validate, and support information in the teacher register. When combining the 1937-register with supplementary sources, some cases of conflicting information were encountered, although minor errors such as spelling errors were most frequent. In a few cases, more substantial information differed, such as years of service or teacher degree, and triangulation was used to identify the source with the incorrect information. In cases where only two sources provided conflicting information, the (primary) source closest to the information in terms of time and place was used.³⁰ My intention with this in-depth empirical study of teachers’ lives in a local region is to provide findings that illuminate the social history of teachers in Sweden, which in its turn can help to understand similar processes elsewhere.³¹ In addition to the sources listed above, reports by regional state school inspectors (folkskoleinspektörernas berättelser) have been useful to understand the overall regional development of primary schooling. Lastly, secondary sources, such as village history books and literature published prior to 1950, have contributed life-history examples from the lives of individual teachers.

The return to teaching after parenthood is central in this article; teachers included in this group all had documented teaching through a teacher notation in one or more of the included sources after becoming a parent. In addition to gender and birth year, the level of teacher training has been central to dividing the primary school teachers into groups which help to differentiate their collective biographies.

Drawing on data from multiple sources, collective biographies constitute the main method used in this study, from which individual life histories are selected and highlighted to show results on the link between work and family among teachers.³² By using a mixed-methods approach on a structured dataset, this study has benefited from both quantitative and qualitative analysis in its examination of the collective biographies of teachers. This approach is flexible, as it enables a study of the teachers’ life histories ‘from below’, i.e. individual level of analysis, while also studying their collective experiences at a group or meso-level of analysis.³³ The individual level with life-histories has been a more common approach than collective biographies to study the history of teachers.³⁴ In Sweden, collective biographies have been used in combination with gender history, although primarily by drawing on qualitative sources and addressing elite women in more urban contexts.³⁵ In his thesis on return migration among Swedish Engineers in the period 1880–1940, Per-Olof Grönberg made use of collective biographies. He also included basic statistics of almost 6,000 engineers and complete careers for about 1,000 of these.³⁶

Based on the sources compiled, Figure 1 presents an overview of all teachers by gender, birth year and teacher training level. A total of 595 individuals had a notation indicating that they were engaged in primary school teaching at some point in life.³⁷ To enable time comparison, all teachers are divided into four cohorts by year of birth (1791–1859, 1860–1879, 1880–1899, and 1900–1914).³⁸

Figure 1. Individuals with teacher notation divided by gender, birth cohort and among qualified teachers - their level of teacher training.

Sources: Beskow 1885, Borgh 1889, Censuses 1870–1930, Dahlberg 1871, Folkskolans lärarekår i Luleå stift 1937, Francke 1905, Francke 1915, Gustafsson 1904, Göransson 1922, Lidell 1896, Linder 1945, Lundmark 1937, Parish records for Skellefteå 1791–1950. See further information in Appendix - Figure A and Table 1.

Table 1. Overview of all individuals with a teaching notation in the compiled data presented by birth cohort, teacher group, and notation of children. Qualified primary school teachers born 1860–1914 included in the collective biographies are displayed in bold. Excluded teachers, i.e. unqualified teachers, teachers born prior 1860 and one male junior school teacher (b. 1880), are displayed in italic.

Teacher group	Link between parenthood and teaching	−1859	1860–1879	1880–1899	1900–1914
Women junior schoolteachers (JST) N= 355	No documented child/children	9	50	74	34
	Teaching record after parenthood	2	14	41	15
	No record of teaching upon parenthood	11	41	56	8
	Total – Women JST in each cohort:	22	105	171	57
Women elementary schoolteachers (EST) N= 88	No documented child/children	4	24	11	13
	Teaching record after parenthood	0	1	6	12
	No record of teaching upon parenthood	1	6	5	5
	Total – Women EST in each cohort:	5	31	22	30
Male elementary schoolteachers (EST) N= 70	No documented child/children	0	2	7	6
	Teaching record after parenthood	8	3	26	16
	No record of teaching upon parenthood	1	0	1	0
	Total – Male EST in each cohort:	9	5	34	22
Unqualified teachers (N=79): Total (women of total):		56 (35)	20 (20)	3 (3)	N/A
Male junior school teachers (N=3):		2	1	N/A	N/A
Total number of teachers by birth cohort (N=595):		94	162	230	109
Total number of qualified teachers (N=516)		38	142	227	109
Qualified teachers included in the collective biographies (N=477):		N/A	141	227	109

Sources: Beskow 1885, Borgh 1889, Censuses 1870–1930, Dahlberg 1871, Folkskolans lärarekår i Luleå stift 1937, Francke 1905, Francke 1915, Gustafsson 1904, Göransson 1922, Lidell 1896, Linder 1945, Lundmark 1937, Parish records for Skellefteå 1791–1950.

Through replacement, retirement or supplementary teacher training, the group of unqualified primary school teachers diminished and eventually vanished during the last decades of the 19^th century. The unqualified teachers they played a key role during the mid-19^th century school expansion in rural areas. However, this study targets the teachers with documented teacher training. Four reasons motivate this selection-criterion. First, unqualified teachers had not invested the same amount of time to becoming a teacher. Second, unqualified teachers are likely to have found their ways to the occupation more coincidentally than those who were admitted to official teacher education. Third, unqualified teachers as a group disappear during the study period, making comparisons with trained teachers impossible over time. Fourth, teaching was a provisional duty for most of the unqualified teachers teaching, hence their average number of teaching years confirmed in the data set was less than half compared to the two groups of trained primary school teachers.³⁹

My decision to exclude the unqualified teachers had the consequence that the group of teachers born prior to 1860 were reduced by almost two-thirds, from 94 to 38 teachers. The later birth cohorts (1860–1879, 1880–1899, 1900–1914) comprised a minimum of one hundred teachers each. My next step in the selection process was therefore to exclude qualified teachers born prior to 1860 since these were too few in relation to the later cohorts. ⁴⁰ Below I will present a summary of the excluded qualified teachers before continuing to the collective biographies. However, for detailed overview of all primary school teachers encountered in the sources, both included and excluded, I refer to Table 1 in Appendix.

The vast majority of qualified teachers born prior to the 1860s were either born in the 1840s (n = 10) or the 1850s (n = 22), leaving six male elementary school teachers as the only qualified teachers born prior to 1840. These six and another three male elementary school teachers (born 1840–1859) all formed a family, while this was the case for one of the five women elementary school teachers born prior to 1860. The 22 women junior school teachers born before 1860 either left teaching upon marriage (n = 13) or remained childless (n = 9).⁴¹ After this selection process 80% or close to 500 teachers (n = 477) remain. Given the attention to work and parenthood in this study, it is important to point out the following. Firstly, in this study expressions such as ‘parenthood’, ‘having children’ and ‘becoming a mother/father’ refers to teachers that at some point had children living in the same household. Most of these cases refers to teachers that had biological children within marriage (94%, 240 cases). Among the remaining sixteen cases eight teachers either had a foster child/ren or a stepchild/ren while eight women teachers gave birth to one or more illegitimate child/ren outside wedlock. Seven of the women with illegitimate children married later in life, however from the sources available it is not possible to tell whether they married the father of the illegitimate child/ren or not.⁴² Secondly, the teachers that married but had no notation of children (including biological, foster and stepchildren) 13% (n = 63) were merged with teachers who neither married nor had any children.⁴³ This choice is motivated by the observation that the occupational track of married teachers without children was much more similar to the non-married and childless teachers than those that married and had children.

Results – work and family: either or both?

Figure 2 presents the three main groups of primary school teachers constructed for the collective biographies divided by birth cohort. The three bar-colours show the share of teachers that remained childless (grey), became a parent and returned to teaching after this (black), and had children but had no information of further teaching (white). The few qualified teachers born prior to 1860 are included only as a reference.

Figure 2. Parenthood and teaching among three groups of qualified school teachers (ST), presented as percent per birth cohort (N= 513). Actual numbers are presented in appendix, see Table 1.

Sources: Beskow 1885, Borgh 1889, Censuses 1870–1930, Dahlberg 1871, Folkskolans lärarekår i Luleå stift 1937, Francke 1905, Francke 1915, Gustafsson 1904, Göransson 1922, Lidell 1896, Linder 1945, Lundmark 1937, Parish records for Skellefteå 1791–1950. For further information, see Figure A and Table 1, both in Appendix.

Among women junior school teachers, it is evident that over time it became increasingly common to return to teaching after parenthood (black bar) while the share who had no known teaching record after forming a family decreased over time (white bar). The high proportion (60%) that had no notation of children (grey bar) in the youngest cohort of women junior school teachers should be interpreted with caution, as most women in this youngest group were still in their thirties in the late 1930’s. Some of the main sources of this study end around this time, and since the 1940’s baby boom took off just a few years later some of these women junior school teachers likely became mothers in their late thirties or early forties. Few women elementary school teachers in the first two birth cohorts formed a family as four out of five (28 of 36) in these groups remained childless. In contrast, a majority (28 of 52) of the women elementary school teachers in the two younger birth cohorts, born between 1880 and 1914, became mothers. In the two latter groups, it became more common to have documented teaching after parenthood, which was the case for four out of ten women elementary school teachers in the youngest group. This proportion is likely to be even higher since these women remained in working age throughout the 1950s and, in some cases, into the 1960s. The male elementary school teachers present an expected pattern in that they primarily either continued their teaching after establishing a family or never formed a family.⁴⁴ The two cases of male teachers with no record of teaching after becoming fathers (white bar) left teaching for another occupation prior to fatherhood.⁴⁵

In the following section, Figure 2 has been used to construct collective biographies of the three main groups: women junior teachers and women- and male elementary school teachers. Extracts from individual life-histories will be used to exemplify experiences that were shared by other teachers within the same birth cohort, gender, and teacher category.

Teachers born 1860–1879: A burst of women junior school teachers in the late 19^th century

The teachers presented in this section were born between 1860 and 1879. They were teenagers or young adults during an expansive phase of public schooling in the studied area, which included both more primary schools and a regionally accessible teacher training. Furthermore, the second school act, enacted in 1882, was less indulgent towards home education than the school act of 1842. For instance, paragraph 38 stated that home schooling would require a special permit by the local school board.⁴⁶ However, in the region under study and other rural parts of Sweden, the registration of school aged children and their low attendance rates continued to be an issue for decades. In 1887, the state school inspector reported that the proportion of children enrolled in school had reached 82%, although the low level of attendance persisted. The pupils had missed 85% of the school days in 1886 leaving a yearly average of 32 days in school, most of this absence was without any valid reason.⁴⁷

Irrespective of low pupil attendance, the process towards a more qualified teacher staff and suitable facilities for education, preferably permanent schools, continued. The establishment of teacher training within the area was crucial to the increase in certified teachers. Training for future women elementary school teachers (1878) and junior school teachers (1882) was initiated within a few years.⁴⁸ The regionally accessible teacher training had a tremendous impact on the increase in qualified teachers, exemplified by the fact that 94% (n = 317) of the junior school teachers in this study acquired their teacher training in Skellefteå. Some of these teachers came to teach in newly established schools while others replaced unqualified teachers.⁴⁹ The rapid development towards the end of the 19^th century can be seen by comparing the state school inspectors’ reports from 1886 and 1898. In 1886, the Skellefteå parish employed 42 primary school teachers in 39 schools (29 ambulatory, 10 permanent). Twelve years later, these figures were 65 teachers spread out over 61 schools (26 ambulatory and 35 permanent). The process of making ambulatory schools permanent is apparent in the continuation of new schools and additional teaching positions, both of which averaged to nearly two per year. However, given that the number of teachers and schools increased at the same pace, this meant that teaching remained largely solo work in a rural one-room school. More than 90% of the teachers remained the sole teacher at their school at the turn of the 20^th century.⁵⁰

The number of qualified teachers born 1860–1879 reached 142 – more than three times as many as all certified primary school teachers born before 1860. This was the case even if the number of male elementary teachers decreased from nine born prior to 1860 to only five in this cohort. Two of the latter five men held only short-term positions before moving elsewhere, while the remaining three worked at least ten years and fathered between three and eight children. Consequently, the increase was an explosion of women teachers among both elementary and primarily junior school teachers. As shown in the bar chart (Figure 2), most of them either remained childless, especially women elementary school teachers, or left teaching when forming a family, as was the case for 40% of the junior school teachers.

The different pattern between the two groups of teachers is not surprising since the elementary school teachers had invested twice the time in teacher training, i.e. four years. The decision to leave teaching when having a child seems to have been a more definite end to the teaching career among women elementary teachers. Only one of seven who had children also had a teaching record later in life. As a reference, one quarter of the women junior school teachers in this birth cohort taught after becoming a mother. One probable explanation is that women elementary teachers married men from higher social classes. Thereby the women elementary school teachers had less economic incentives to return to teaching. While junior school teachers commonly married farmers or workers, the women elementary school teachers had spouses from the middle or upper stratum such as bailiffs, higher clerks (kammarskrivare), or the county veterinarians. The marital pattern of primary school teachers is interesting, not least for exploring how the status of the teacher occupation varied across time, but it goes beyond the scope of this study.

The life-history of Hanna Paulina Lidén exemplifies how women elementary school teachers were more inclined to teach in various places around Sweden and were less likely than their junior school counterparts to settle down and form a family. Born in 1879, Lidén moved from inland northern Sweden to start elementary school teacher training in Umeå at the age of 16. After graduation in 1900 she worked a few years at various places in northern Sweden, including one and a half years in the area under study. She then returned to Umeå where she came to work until her sixties, never marrying or having any children.⁵¹ This high mobility among late 19^th and early 20^th century women elementary school teachers has previously been found in another Swedish parish.⁵²

Women junior school teachers were, in contrast, deeply rooted in the local community and seldom left the parish.⁵³ Nearly half of the junior school teachers born 1860–1879 (n = 50) had no record of any children. This group taught twenty years on average and their trajectories typically indicate teaching for several decades not leaving their position until age of retirement. Among women junior school teachers who formed a family, three quarters (n = 41) would not teach again. In the records this was commonly reported as ‘resigned upon marriage’.⁵⁴ Eugenia Öster embodies both the average teaching years and average number of children born by the women teachers in this group. After earning a junior school teacher certificate in 1894, Öster taught in a village for seven years before leaving teaching due to marriage. Her spouse, Olle Marklund, was a farmer in the same village that Eugenia had taught in, they had five children together between 1902 and 1914 and she never returned to teaching.⁵⁵

Fourteen of the women junior school teachers show a record of teaching after becoming mothers. Hulda Lundström and Lovisa Häggmark were the first two women in the parish to return to full time teaching after having children. They taught another twelve and nine years respectively after motherhood and were both granted pensions upon retirement in the 1920s.⁵⁶ However, returning to teaching was still mostly a temporary engagement, such as being a vacancy substitute for a limited period. This was the case for the remaining twelve women, among them Ida Kristina Hortelius. After graduation in 1883, she taught for seven years as a junior school teacher before ending her position due to her marriage to a farmer. They had seven children between 1892 and 1910. Between the sixth and the seventh child, Hortelius filled in as a vacancy substitute for three months in early 1910.⁵⁷

Teachers born 1880–1899: primary school teaching as a women’s world in the early 20^th century

Male elementary school teachers increased from five teachers (b. 1860–1879) to 34 teachers in this birth cohort. Among those, thirteen had done their training in the town of Luleå, where an elementary school teacher training had opened in 1907, reducing the distance to the closest teacher training for men by almost 200 kilometres.⁵⁸ Compared to the group born 1860–1879, women elementary school teachers were slightly fewer in this birth cohort. The number of women junior school teachers (Figure 3), boosted by additional positions and places for training, continued to increase from 101 in the previous birth cohort to reach its peak of 171 women, equivalent to a third of all qualified primary school teachers that worked in the area from 1842 to 1937.⁵⁹

Figure 3. Class of 1907, junior school teacher training college in skellefteå. Augusta saedén (front row, fifth from right) was the first headmistress when the teacher college became permanent in 1892 and continued as such retirement until retirement in 1917. Saedén was succeeded by ida Nordsvan (front row, sixth from right) whose term as headmistress ended when the college closed in 1929. A majority of the future teachers were born in either 1887 or 1888. The only man in the picture is one of the few men that entered junior school teacher training.

Source: Skellefteå museum, ID: C 12,015

Nine out of ten teachers in this birth cohort had earned their teacher certificate and entered teaching during the first two decades of the 20^th century. In the meantime, the slow but steady process of reducing half-time reading and ambulatory schools continued in the region. In 1920, when a vast majority of teachers born from 1880 to 1899 had started working, half-time reading was still practiced in every fifth school and, out of 106 schools, ten remained ambulatory, moving between two villages every spring and fall.⁶⁰ Primary school teachers with a long period of service from late 19^th century and into the 1920s experienced substantial improvement regarding employment security, salaries, and pensions. This is plausibly due to a new decree in 1918 that made it more difficult to dismiss an ordinary junior school teacher when she married. In effect, local school boards now needed to acquire a permit from the national school board to discharge a teacher.⁶¹

An interesting development between this and the previous birth cohort is found among women elementary school teachers. The share that remained childless fell from close to four out of five (24 of 31) to only every second (11 of 22). The share of teachers that combined teaching while having children also increased. One of them Estrid Smith, was born in Stockholm in 1882. She undertook her elementary school teacher training in Umeå, where she graduated in 1904 with good marks especially in ‘singing and music’ and ‘physical education’.⁶² Beside a one-year temporary position in the more southern parts of Sweden in 1912, she came to have three permanent positions at various places in Norrbotten, the northernmost region of Sweden, between 1904 and 1922. During these years, Smith met her husband, the elementary school teacher Knut Sjöberg, whom she married in 1913 and in the following years, they had four children. The family moved to the area under study after the birth of their fourth child in 1922. Estrid continued to teach part time until retirement, in the 1930 census she was listed as ‘wife/mother’ although with ‘teaching’ as an ancillary occupation. A few months after her 70^th birthday she passed away. In the obituary, she was described as ‘well prepared and educated, knowledge she made us of both within the sphere of home and school’.⁶³

Half of the women elementary school teachers had no recorded child, a share considerably lower compared to those born in 1860 to 1879 where almost eight out of ten remained childless. Among women teachers that had children in this group, it was almost as common to return to teaching later in life as it was to leave teaching permanently. Furthermore, in contrast to the previous birth cohort (b. 1860–1879), it was no longer only a question of being a substitute for a short period. The life-histories in this cohort (b. 1880–1899) were more diverse overall. For example, the first four divorce cases are seen in this group. In all four cases, it concerned women junior school teachers who taught several years after graduation before leaving their teaching position due to marriage followed by one or several children. ⁶⁴ Upon their divorce, they all returned to teaching suggesting the need to earn their own livelihood once again.

As Figure 2 showed above, approximately every fourth junior school teacher (b. 1880–1899) returned to teaching after motherhood. By scrutinizing this tendency within this birth cohort it is evident that the number that returned, at least occasionally, increased rapidly. Less than a third of the women junior school teachers born in the 1880s returned to teaching after becoming a mother while this was true for a majority of those born in the 1890s.⁶⁵ In the previous birth cohort, Hulda Lundström and Lovisa Häggmark were the only two mothers that returned to a permanent teaching position. In this birth cohort, more mother teachers returned to a permanent position; this was the case for those born in the 1880s and even more so in the 1890s. However, the distribution of years at home and teaching in school after the birth of the first child showed a great variation. This was partly, although not exclusively, related to the number of children reared. Typically, women that had a higher number of children usually had more years away from teaching.

From the mid-1920s, when more women junior school teachers kept their permanent position also after becoming a mother, it seems to have become more difficult for younger women teachers to be employed on a permanent position. These difficulties were intensified by the development towards more teaching positions for elementary school teachers and a subsequent decrease in the number of pupils. The life-histories of Gerda Enmark (b. 1896, m. Viklund) and Göta Hedlund (b. 1905) exemplify these difficulties. After junior school teacher graduation in 1915, Gerda Enmark worked as a long term (> one semester) substitute at three different schools for three years before she got a permanent position in 1918. Another ten years later, now in her early thirties, she married a timber merchant and between 1929 and 1938 they had four children. Throughout the years of forming her family, Gerda kept her permanent position that she had attained in 1918 and remained at the same position and school until 1945. The trajectory for the nine-year younger Göta Hedlund after she graduated in 1926 was rather different. During the first four years, Hedlund worked as a short-term substitute teacher (< one semester) at various schools in the parishes of Skellefteå and Jörn. The work as a substitute continued during the 1930s for Hedlund although she served a semester or more at the same school rather than weeks or months. During these longer substitute positions, Göta filled in for Gerda who was home with small children. This illustrates the harsher labour market conditions experienced by younger teachers that entered the labour market in the mid- to late-1920s. Göta never had any children, but in 1938, twelve years after graduation and having worked in almost as many schools, Göta got her first permanent position and was still working as a junior school teacher in 1949.⁶⁶

Teachers born 1900–1914: elementary school teachers on the rise – development in the 1930s and 1940s

Due to 1937 being the end year for including teachers in the study, the youngest birth cohort includes teachers born in 1900–1914. This excludes some individuals born during the first decades of the 20^th century that became teachers after 1937 or did not teach in the region prior to this year. However, the decreased number of teachers is mostly explained by the late-1920s reduction in junior school teacher training opportunities. The reduction in teaching positions was partly due to a political intention to train more elementary school teachers and partly due to falling birth rates during the national baby bust in Sweden during the 1930s. These shrinking opportunities for teacher training in the late-1920s for junior school teachers can be observed in this birth cohort. The shift from a majority of junior school teachers to elementary school teachers is distinguished by studying the number of teachers born each individual year. This examination shows future junior school teachers were more numerous every single year up until 1903; this then changed into a majority of elementary school teachers born between 1904 and 1914.⁶⁷ Within the group of elementary school teachers, if compared to the previous birth group, the number of women teachers increased by eight while male teachers had decreased by a dozen.⁶⁸ Before turning to the practice of combining work and family within this group, it should be noted that less is known about the later work and life-histories of the teachers in this cohort since most of the main sources end during the 1940s, when these teachers were still in their thirties or forties.⁶⁹ Nevertheless, Figure 2 shows that the share of women teachers who kept their position and taught after having children continued to increase in this group.

Notable among teachers born in 1900–1914 is the high share of primary school teachers that married and had children with another teacher. A few teacher couples were similarly encountered among teachers born in the 19^th century although it was far from the same extent as in this group. Among the 56 teachers in this youngest birth cohort that had at least one child, no less than 24 formed a family with another primary school teacher.⁷⁰ Furthermore, among the fourteen women teachers who formed a family with another teacher, all but one returned to teaching after becoming a mother. Among women teachers who had married a non-teacher only slightly more than half (14 out 26) taught after motherhood. Given that elementary school teacher training became increasingly more accessible for women and men in the same towns and the fact that they had a more equal sex composition, this increased the possibility of finding a spouse during teacher education or while working at the same school.⁷¹

Concluding remarks

This study investigated the link between work and family among primary school teachers, and how this link shifted across time. By way of introduction, two research questions were addressed, the first of which concerned how parenthood varied depending on birth year of the teachers, their gender, and level of teacher training. The second question considered the possibilities of returning to teaching after forming a family and changes over the study period. The results were obtained through collective biographies constructed from my various sources, in turn enabling a longitudinal examination of three birth cohorts (1860–1879, 1880–1899, and 1900–1914) and three teacher groups: women junior school teachers, and male and women elementary school teachers, respectively. Within each sub-group, the teachers were divided depending on whether they remained childless, had children and then left teaching permanently, or returned to teaching later in life.

The male elementary teachers that fathered children continued to work over their lifetime and parenthood did not affect their teaching careers. Women teachers follow a similar pattern as found in previous research whereas the level of childlessness oscillates between 40 to 50% with an exception for women elementary school teachers born prior to 1880 where nearly four out of five remained childless. This result confirms previous research by Carlsson and Qvist, suggesting that daughters coming from middle- or upper-class experienced difficulties to find a suitable spouse. That nearly half of the women teachers throughout the study remained childless is expected, as the included teachers were all born prior to 1915. One of the central findings in this study is that an increasing share of women returned to teaching after becoming mothers. This result is in line with what Elgqvist-Saltzman found among women elementary school teachers trained in southern Sweden. The combined longitudinal and qualitative approach further enabled this study to describe how the teaching carried out by these (mother) teachers changed over time. Moving from more temporary engagement in the 19^th century to the practice/norm of withholding their permanent positions upon motherhood a few decades into the 20^th century. Both Copelman and Oram found that some of the 1930s London women teachers continued to teach upon marriage, the empirical findings here add that married women teachers in northern Sweden did not only teach upon marriage, many of them also returned to full time teaching after having children.

Provided the societal changes and improved women civil rights in the late 19^th century and first half of the 20^th century, it is not surprising that women teachers took advantage of their extended opportunities to combine work and family. My findings show that women teachers, and especially women elementary school teachers who reached adulthood before 1900, had more limited possibilities to keep their teaching position upon motherhood than those who reached adulthood after 1900. However, junior school teachers life-histories of Hulda Lundström and Lovisa Häggmark, born in the 1860s and 1870s, demonstrate that even though it was unusual to have children and then continue a teaching career among women teachers born prior to 1880, it was not impossible. In the 1910s, they became the first two who returned to a permanent teaching position after having children. Both Hulda and Lovisa were junior school teachers; in comparison, among the women elementary school teachers, the increase in the share that became mothers and the proportion that returned to teaching occurred later, but went on to progress more rapidly. This suggests that the prevailing norm that women could either teach or be a mother had started to change or at least was possible to challenge.This was also the case in the socially stable area studied here.

By examining the link between work and having children, rather than work and marriage, this study has contributed new perspectives on the lives of women teachers. Studying the history of teachers by using both life history and a comparative perspective confirm Clifford’s argument that these approaches deliver a more complex picture than a single focus on policies. The results of this study provide a more complex account of women teachers in having identified their life histories, in turn indicating a variety of possibilities over the course of life within the group of women teachers. Indeed, there were many differences between women junior and elementary school teachers and the conditions they experienced over time in the rural context examined. However, time aspects and which teacher group they belonged to are far from the only features that shaped the link between family and work among women teachers. Further studies of how this link functioned and varied in relation to income level, occupation of spouse, number of children, school setting (urban/rural), etc. are needed to understand even more of the changing connections between work and family.

Another contribution of this study is its complete survey of how the entire teacher composition in one parish changed over time in terms of employment numbers, sex distribution, and type of primary school teachers. In the rural area under study, unqualified teachers were most numerous among the teachers born prior to 1860, followed by a long period where the vast majority of teachers were women junior school teachers until the 1930s when elementary school teachers increased their share. The combination of quantitative and qualitative methods proved to be a fruitful approach, which enabled me to obtain comparatively nuanced results. While the descriptive statistics show how the link between family and work changed over time, the life-histories highlighted throughout the study help to gain a deeper understanding of this link. When examining the changing link between work and family among the younger women teachers, the advantages of applying collective (meso-level) analysis on one group of teachers and their individual life histories constructed from multiple sources becomes evident. On a group level, the results reveal that an increasing share of women teachers kept their teaching position upon motherhood from the 1920s and onward. By drawing on information from Lundmark’s register and the life-trajectories of Gerda and Göta, the following argument could be proposed. When more women teachers kept their position after getting married and having children, as Gerda did, this decreased the need for new (permanent) primary school teachers. Consequently, this likely contributed to the 1930’s surplus of teachers that have been previously mentioned by Frangeur and Ursing. These conditions meant that young women teachers, such as Göta, had to accept a number of spatially scattered temporary positions for years before being hired on a permanent position. While this argument deserves further discussion in relation to structural demographic developments, such as lower birth rates, it certainly has contributed new knowledge and insights regarding a complex historical process, namely the changing link between family and work for women. Most likely, these insights would not have come to the fore without the multiple sources and methods used in this study.

In sum, focusing on the collective level and life histories of primary school teachers, this study suggests a way to examine the link between work and family in a historical setting. The results go beyond the focus on women either in their occupational or reproductive role by recognizing women teachers’ experience and positioning them within professional life as well as family life.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Notes on contributors

Emil Marklund

Emil Marklund is an associate professor at Umeå University, Sweden. In addition to history of education his research interests include social history, microhistory, spatial history but also contemporary issues regarding teacher recruitment and education in rural settings.

Notes

1. The three teachers’ unions in Sweden (during the 1940’s) jointly funded a book to commemorate the anniversary of the first Swedish School Act on a national level, Eljas, ed., Svenska folkskolan 100 år; Shorter books on the local level were also released. See Johansson, Umeå sockens skolväsen 1842–1942; and Essegård, Vilhelmina skoldistrikt 1842–1942.

2. Lundmark, “Lärare i Skellefteå”.

3. An overview of the political and economic development in Sweden during the latter part of the 19^th century and first decades of the 20^th century is provided in: Stråth, Sveriges historia, 1830–1920, see part II and part III; and Magnusson, Economic History of Sweden, chapter five and onwards; An overview of the Swedish history of education is found in, Richardson, Svensk utbildningshistoria.

4. Elgqvist-Saltzman, Lärarinna, kvinna, människa, 141–2.

5. For a compilation of the different school acts see: Hall, Sveriges Allmänna Folkskolestadgar 1842–1921. Paragraphs concerning the age of pupils can be found on page 8, 28, 57 and 101–2.

6. Sörensen, Det svenska folkundervisningsväsendet 1860–1900, 264.

7. For a recent discussion on regional forms of schooling in 19^th century Sweden, see: Westberg, “National School Systems”.

8. Westberg, “Teachers Make a Living?”.

9. Florin, Kampen om katedern, 43, 155; and Bruce, Det svenska folkundervisningsväsendet 1900–1920, 305.

10. Bengtsson and Prado, “Rise of the Middle Class”, see especially p. 7. Note that Bengtsson and Prado does not include figures for junior school teachers. These are available in, Florin, Kampen om katedern, 43.

11. Florin, Kampen om katedern; Persson, Läraryrkets uppkomst och förändring; Marklund, “To Become a Teacher”, 27–49; and Åberg, Sveriges småskollärare.

12. Scholarly interest in various connections between family life and paid work has been frequently addressed in recent decades both in contemporary and historical communities, not least within gender history. See for example: Stanfors, Mellan arbete och familj; and Tentler, Wage-Earning Women; Thistle, Marriage to the Market; in recent years also examining fathers, see: Kaufman, Superdads.

13. Tilly and Scott, Women, Work and Family, 184–5, 224.

14. Goldin, “Marriage Bars”; Redmond and Harford, “One man one job,” 639–54; Proctor and Driscoll, “Bureaucratic governance,” 157–70; Whitehead, “Fashioning the country teacher,” 1–14; and Aitken, “Wives and Mothers First,” 83–98.

15. Clifford, Those Good Gertrudes, 131–2.

16. Florin, Kampen om katedern, 148.

17. Florin, Kampen om katedern, 42–5; Frangeur, Yrkeskvinna eller makens tjänarinna?, 149–54; and Ursing, Fantastiska fröknar, 137–9.

18. Oram, Women teachers, 67–72.

19. Copelman, London’s women teachers, 194–5.

20. Of course, the word ‘all’ should always be used with caution, not least when the examination includes teachers over almost a hundred years. However, when compiling the different sources, all the included (qualified) teachers were present in more than one independent primary source. Furthermore, the compiled sources cover more than a century (1840’s to the 1940’s) however since the qualified teachers were few between 1840 and 1860 these were dropped from the main study, see also the extended discussion in the ‘Sources and method’ section.

21. Van Bavel, “Mid-twentieth century Baby Boom,” 925–62; and Sandström and Marklund, “The dual provider family,” 149–73; and Rowland, “Historical Trends in Childlessness,” 1311–37.

22. Carlsson, Yrken och samhällsgrupper, 270–80; and Carlsson, Fröknar, mamseller, jungfrur och pigor, 109–14; and Qvist, Blifva en god flicka, 163–84, especially 177–80. An overview of the different socio-economic background among primary school teachers is provided in Marklund, “To Become a Teacher?,” 27–49.

23. Sandström and Marklund, “The dual provider family,” 149–73, the numbers are extracted from Figure 4 on 162.

24. Dribe and Scalone, “Social Class Net Fertility”; Sandström, “Educational Gradient of Fertility”.

25. Junkka, Shared practices, 1–2, 46–7. See also, Hofsten and Lundström, Swedish population history.

26. Westberg, “National School System,” 201. Skellefteå is situated in the region of Västerbotten.

27. Myrdal, “Source pluralism,” 157–8.

28. Today these primary sources are found either in the municipal (Skellefteå) or the regional archive (Härnösand, c. 300 kilometres south of Skellefteå). However, in 1937 they were all accessible in the township of Skellefteå. During my visits to these archives, random samples were taken from the primary sources to cross-check the teacher register; all these samples agreed with the information provided by Lundmark.

29. After adding the information from these sources, it was evident that Lundmark did not know of/did not use these sources since he could have used some of the data to improve his own record. For example, this could be data on the year of first employment of a specific teacher, available in the national registers but absent in the record by Lundmark.

30. Regarding triangulation, see Vikström, “Identifying dissonant data,” 211–21.

31. My approach is similar to the idea of a ‘new social history’ as described by Westberg, “Global and Local Narratives,” 206–14.

32. I have used ‘collective biography’ rather than ‘prosopography’ although the two are often used as interchangeable concepts, see Stone, “Prosopography”.

33. This approach draws on the work by Cohen and colleagues, see: Cohen, Flinn, and Morgan, “Mixed Method Social History,” 211–29, especially 226.

34. Goodson, ed., Studying Teachers’ Lives; Goodson, “Story of Life History,” 129–42; and in addition to the work by Goodson, women teachers have been studied in: Middleton, “Educating Feminists,” 53–67; and Munro, Subject to Fiction.

35. Broady et al., “Formering för offentlighet”; Norrbin, Från isolering till integrering.

36. Grönberg also included a life-history perspective by examining experiences of individual engineers, see Grönberg, Learning and Returning, 40–1 and chapter five that addresses life-history examples.

37. This number is higher than the 492 teachers included in Lundmarks register due to two main reasons. First, Lundmark included fewer although some unqualified teachers. Second, and more importantly, Lundmark examined Skellefteå rural parish and therefore only included teachers in Skellefteå town parish until 1913 – when the town became a separate school district. Consequently, most of the primary school teachers included in this study but not present in the register are the teachers that worked in Skellefteå town parish at some point between 1913 and 1937.

38. Here one could argue that it would be better to separate the teachers by their first year as a teacher to obtain results regarding individuals that started their teaching about the same time. At least three objections could be raised to this approach: 1) 90% of the teachers started their teaching track when they were between 17 and 30 and a majority between 20 to 25; 2) Many of the teachers that had their first notation of teaching in the region after the age of 30 had previously taught elsewhere; and 3) By using year of birth, teachers that had somewhat similar possibilities during their formative years will be in the same group, including important aspects such as access to teacher training.

39. The unqualified teachers had on average six years of confirmed teaching, while elementary and junior school teachers had 14 and 16 years, respectively. That junior school teachers have more confirmed years of teaching is primarily an effect of the geographical delimitations. Elementary school teachers were more mobile and would have more years of service than their junior school colleagues if this were to be examined on a national level.

40. Including the one male junior school teacher born after 1860 that did not continue to elementary school teacher training, see Table 1 in Appendix.

41. Both women that returned to teaching after having children did so only as short-term substitutes.

42. Regarding children born outside wedlock birth/parish records rarely include any information on the father.

43. Cohabitation, with mutual children while not being married, was not encountered in any of the sources.

44. Although the Swedish male breadwinning system has been identified as ‘weak’ in comparison to the U.S., this was rather an effect of more mothers in the labour force. Swedish men definitely continued to work after becoming fathers, see Sommestad, “Welfare State Attitudes,” 153–74; and Mósesdóttir, “Dual Breadwinner Model,” 189–205; and for an example of how this male-privilege was practiced in a school setting, see Whitehead, “The Teaching Family”.

45. Johan Erik Löfgren (b. 1826) became an organist and Erik Anton Lindgren (b. 1885) an assistant vicar (komminister).

46. Hall, Sveriges Allmänna Folkskolestadgar, 29. In addition to the permit, a member of the school council would also hold early examination with home educated children.

47. Unæus, “Södra Lappmarkens kontrakt,” 82–96. The calculation of average school days combines the table on p. 90 with information on teaching on p. 87.

48. Shorter and irregular junior school teacher training courses had been arranged regionally during the 1870s, see Fellström, Folkundervisningen i Skelleftebygden, 69. The rapid increase in permanent teacher training colleges for junior school teachers were largely an effect following a law in 1878. This law declared that state subsidies for salaries were only to be paid if a junior school teacher had seven months of training or more, see Florin, Kampen om katedern, 124–5.

49. Among unqualified teachers, with information of last known teaching position, two-thirds (33 out of 51), had their last teaching record between 1870 and 1900. This coincides with the decades when approximately 130 trained junior school teachers, born 1850–1879, were in their twenties and entered the occupation.

50. This included both Skellefteå rural and town parish, however close to all the primary school teachers worked in the rural parish during the late 19^th century. Unæus, “Södra Lappmarkens kontrakt”; and Lundkvist, “Vesterbottens andra kontrakt”.

51. ”Umeå seminarium”, Lidén graduated in June 1900; and Lundmark, “Lärare i Skellefteå,” 72; Lärarekår i Luleå stift, 312.

52. Florin, Kampen om katedern, 66–7.

53. Less than 15% (n = 14) migrated at some point during the course of their life and among those that did, most common move was to another parish in the same county.

54. Lundmark, “Lärare i Skellefteå,” 51.

55. Lundmark, “Lärare i Skellefteå,” 137; “1900 års folkräkning”, Skellefteå stadsförsamling, 18, row 31. Accessed February 23, 2022. All the references to censuses were retrieved from the same source on the same day, henceforth the date will be excluded.

56. Lundmark, “Lärare i Skellefteå,” 63 (Lovisa Häggmark, born Ljuslinder), 99 (Hulda Lundström, born Dahlberg). In her study of a coastal Swedish parish approximately 300 kilometres to the south, Florin encountered six, out of 72, junior women school teachers that continued to work after they married, however it is not known if these women ever had any children, see Florin, Kampen om katedern, 46.

57. Lundmark, “Lärare i Skellefteå,” 61; “1910 års folkräkning”, Skellefteå landsförsamling, (Ida Kristina Hortelius), 418, row 39. Same comment from note 55.

58. Luleå is situated approximately 140 kilometres north of Skellefteå. A geographical closer teacher training should not be viewed as the sole explanation for the increase. For example, the number of teachers that graduated in the previously available facilities in Härnösand increased from four to eleven between the two age groups. The increased number of teaching positions for elementary school teachers as well as improved means of transportation probably contributed to this development.

59. For an overview of the development of junior school teacher training in the region, see Nylund, Småskollärarutbildningen i Sverige, 104–7.

60. The number of schools is from the report by the state school inspector. See Ahlman, “Berättelse över skolväsendet,” 4; For the different ways of arranging primary schooling, see the curriculum from 1919: Folkskolöverstyrelsen, Undervisningsplan för rikets folkskolor, 4–6.

61. Bruce, Det svenska undervisningsväsendet, 300–40, on the development of pensions and legal improvements regarding employment. See 337–9 for the mentioned improvements of junior school teachers.

62. ”Umeå seminarium”, (Estrid Maria Elvira Smith, graduated in 1904).

63. Göransson, Svensk Folkskolematrikel, 232; “1930 års folkräkning”, Skellefteå stadsförsamling, 17, row 44 “Estrid Maria Elvira Smith”; and Lärarekår i Luleå stift, 266–7; Norra Västerbotten, “Familjenyheter – Dödsfall – Estrid Smith,” 5.

64. A strong connection has been found between social class and divorce whereas higher social classes were also more likely to divorce, male teachers were more inclined to divorce than farmers and industrial workers for example. See Sandström, Socio-economic determinants of divorce, 292–307.

65. This example excludes those who remained childless. Among (mother) junior school teachers born in the 1880s, 15 of 47 had a record of teaching after becoming a mother, among those belonging to the same group but born the following decade, this number had increased to 28 of 50.

66. The two teachers are both present in multiple sources. See especially: Lärarekår i Luleå stift, 281 (Göta) respectively 285 (Gerda); Linder, Sveriges Småskollärarinnor, 1081 (Göta) respectively 1086 (Gerda); Hebbersfors, 175 (Gerda); Långviken förr och nu, 78 (Göta).

67. This meant a dramatic drop of junior school teachers in the examined sources going from a yearly average of nine teachers (b. 1891–1903, n = 115) to only two/year (b. 1904–1914, n = 22). By comparison, elementary school teachers showed a modest increase in yearly average moving from just below three teachers each year (1891–1903, n = 36) to just above three/year (1904–1914, n = 34).

68. Despite the decrease in actual numbers, male teachers increased their share due to the dramatic drop among junior school teachers. See Table 1 in Appendix.

69. The reason for leaving teaching is only known in approximately 50% of the cases in this group which can be compared with more than 90% among the primary school teachers born in the 19th century (i.e. the three first birth cohorts).

70. Including nine “teacher couples” (n = 18, where both partners belong to this birth cohort and another six teachers that either formed a family with a teacher belonging to the previous birth group (b. 1880–1899) or a teacher that worked in a neighbouring parish.

71. The first elementary school teacher training in Sweden to accept both men and women (1919) was the teacher education in Luleå. See Riksarkivet, “Folkskoleseminariet i Luleå arkiv”. Accessed February 23, 2022. In 1926 Greta Josefsson (b. Söderström) became the first woman in this study to graduate from the, previously all-male, elementary school teacher training in Luleå. The training for elementary school teachers in Umeå remained open exclusively for women until 1938 when it opened to male applicants. See Lärarekår i Luleå stift, 274; and Raninge, Folkskoleseminariet i Umeå, 64.

Appendix

Figure A. Chronological overview of the sources compiled in this study divided by longitudinal sources and synchronic sources. Bibliography and comments are provided below the figure. The first arrow/column is the register compiled and written by Rudolf Lundmark covering teachers active in primary schooling at some point between 1842–1937 (Skellefteå town parish until 1913, the larger Skellefteå rural parish until 1937). In some cases information was added in the 1940s and 1950s, but no new teachers were added after 1937. The second arrow/column records represent parish records, and: include information from a number of different records with information on birth, migration, occupation, marriage/spouse, children and death. The starting year (1791) is due to the birth year of the oldest teacher included. The third column presents the years of the six censuses that are included in the corpus of sources, which contain information on household and occupational information from that specific year. The fourth and final column represents teacher registers (R) which on a national level provide descriptive statistics on either all elementary school teachers (EST) or all junior school teachers (JST) or both groups. In these registers, the name of the teacher was accompanied by primarily numeric information such as which year the teacher graduated, years of service, number of pupils, type of school, salary level, etc. Finally, the two teacher biographies (B) commonly present the teacher with a small picture and a short text on primarily professional information although sometimes including spouse (name, occupation) and children (name, birth year).

Sources included Figure A – from left to right in the figure:

Rudolf Lundmark, “Lärare i Skellefteå sedan 1842,” unprinted teacher register, Skellefteå, 1937. Skolstyrelsen, Skellefteå landsförsamling. Skellefteå kommunarkiv, Sweden. The register is available in the municipality archive in Skellefteå, Lundmark provides references to the parish record or other sources used, thus enabling the possibility of cross-checking inconsistencies.

The parish records, 1791 to 1950, are in most cases digitized by the National Archives (Riksarkivet) and searchable here: https://sok.riksarkivet.se/nad (Retrieved: 13–10–2022), 1010174 – Skellefteå landsförsamlings kyrkoarkiv 1791–1950 and 1010175 – Skellefteå stadsförsamlings kyrkoarkiv 1913–1950. Information from the following series has been used: A – Catechetical examinations and parish records, B – Records on migration into and out of the parish, C – Birth and baptismal records, E – Marriage records, F – Death and burial records.

For digitized censuses – 1870, 1880, 1890, 1900, 1910, and 1930 – see the National Archives (Riksarkivet), Digitala forskarsalen, Specialsök, Folkräkningar https://sok.riksarkivet.se/folkrakningar (Retrieved: 13–10–2022),

Chronological bibliography – nine national registers and two teacher biographies

Dahlberg, N.J. Matrikel öfver rikets folkskolor och folkskollelärare jemte uppgift öfver rikets seminarier och vid dem anställde lärare. Helsingborg, 1871.

Beskow, Wilhelm. 1885 års Matrikel öfver folkskoleinspektörer samt lärare vid folkskolelärareseminarier, småskolelärarseminarier, folkhögskolor, högre folkskolor och ordinarie läraretjenster i egentliga folkskolor, innehållande jemväl vissa uppgifter rörande folkskoleväsendet i öfrigt. Stockholm: Ivar Hæggströms boktryckeri, 1885.

Borgh, Pehr. Utförlig och fullständig matrikel öfver Sveriges folkskollärare, organister, kantorer och klockare samt seminarii-lärare, folkskoleinspektörer och lärare vid högre folkskolor, jemte nogranna uppgifter rörande dessa tjänstemäns alla löneförmåner, folkskolornas beskaffenhet, boställenas storlek och afkastningsvärde, kyrkoorglarnes resurser m.m. Norrköping: M. W. Wallbergs förlag, 1889.

Lidell, Per. Sveriges folkskolelärarematrikel 1895. Stockholm: K. I. Beckmans boktryckeri, 1896.

Gustafsson, Clas. Svensk folkskole-matrikel omfattande uppgifter t. o. m. den 1 Januari 1904. Stockholm: Centraltryckeriet, 1904.

Francke, Sven P. Matrikel över lärarinnor och lärare vid Sveriges småskolor och mindre folkskolor även som biträdande i folkskolan vårterminen 1905. Malmö, 1905.

Gustafsson, Clas. Svensk folkskole-matrikel omfattande uppgifter t. o. m. den 1 July 1910. Stockholm: Centraltryckeriet, 1910.

Francke, Sven P. Matrikel över lärarinnor och lärare vid Sveriges småskolor och mindre folkskolor även som biträdande i folkskolan vårterminen 1915. Malmö, 1915.

Göransson, Edvard. Svensk folkskolematrikel År 1922, del I–III. Stockholm: P. A. Norstedt & Söner, 1922.

Folkskolans lärarekår i Luleå stift 1937. Malmö: Skånetr., 1937.

Linder, Ruth. Sveriges småskollärarinnor i ord och bild, vol. III. Uppsala: Biografiskt galleri, 1945.