Minding the gaps: the politics of differentiation in Swedish education from 1842 to the 1960s

ABSTRACT The concept of differentiation holds immense significance in education, touching upon aspects like access, inclusion, justice, and equality. However, it is also a complex and elusive notion, which acquires different meanings across historical and cultural contexts. This article explores the shifting reasoning about differentiation in the Swedish educational context. Inspired by Foucault’s account of disciplinary power, it conceptualizes differentiation as a technique for marking and addressing gaps between individuals. Drawing on an analysis of governmental and scholarly reports from 1842 to the late 1960s, the article identifies three shifts in the reasoning on differentiation: 1) from differentiation by socioeconomic class as a given factor to the search for scientific rationales for differentiation based on measurement of intellectual ability, 2) from viewing differences in intelligence as biologically conditioned and stable to viewing them as amenable to training and correction through education, and 3) from a focus on inputs to a focus on outputs. Overall, the article argues that even if the term ‘differentiation’ itself has been discursively replaced by others, the ideas underlying it—the search for gaps—continue to shape education in Sweden and beyond.

In Sweden, the ways of posing and answering this question have varied considerably over time, competing and coexisting with each other.Most often, however, the search for rationalized criteria for differentiation has focused on deficiencies and problems identified on the basis of examinations, quantifications, and statistical calculations.What or who is to be differentiated is, thus, defined by gaps that separate the 'desirable' from the 'problematic', the 'normal' from the 'abnormal', the intended and the achieved in the dominant political and scientific discourse.Such gaps are crucial to educational governance and the fabrication of 'kinds of people' (Hacking, 1986), whether the intention is to maintain or bridge them.
Inspired by Foucault's (1986Foucault's ( , 1991) ) account of disciplinary power, in this paper we approach differentiation as a specific technique for marking and addressing gaps between individuals, groups or institutions.This technique is based on measurement and comparative reasoning and renders some qualities 'useful' and hierarchizes them in terms of nature, capacity, aptitude, values, and so on.It includes and excludes according to ranks, grades, levels, and other qualities.In so doing, it also individualizes.However, as Foucault (p. 193) noted, such individualization is 'descending' as it relies on 'gaps' rather than deeds and aims at normalization and homogenization rather than diversification: In a system of discipline, the child is more individualized than the adult, the patient more than the healthy man, the madman and the delinquent more than the normal and the non-delinquent.In each case, it is towards the first of these pairs that all the individualizing mechanisms are turned in our civilization; and when one wishes to individualize the healthy, normal and law-abiding adult, it is always by asking him how much of the child he has in him, what secret madness lies within him, what fundamental crime he has dreamt of committing.(Foucault, 1991, p. 193) In other words, 'descending' individualization, according to Foucault, does not distinguish individuals from the masses, but is accomplished through comparative measurements which have the 'norm' as a reference point.It replaces the individuality of a 'memorable' man with the individuality of a 'calculable' man (Foucault, 1991, p. 193).

Exploring differentiation in the Swedish context
In this article we explore the shifting 'modes of thought' or 'styles of reasoning' (Elden, 2006;Hacking, 1991;Porter, 1995) about differentiation using examples from the Swedish context.Through an analysis of governmental and scholarly reports from 1842 to the late 1960s, a large portion of which was identified through previous studies related to differentiation in and of education in Sweden (e.g.Axelsson, 2007;Dahllöf, 1967;Husén & Härnqvist, 2000;Marklund, 1985;Nordström, 1968), the article unpacks the knowledges, assumptions, and beliefs that underlie various practices of differentiation in education.In so doing, it argues that the politics of differentiation are predicated on the detection, visualization, measurement, and control of various kinds of gaps.
The selected period, from 1842 to the 1960s, can be described as a time of intense debates about how, when and on what grounds the school system should be differentiated.Consequently, the account presented below does not claim to be exhaustive.For instance, it does not touch upon differentiation by age, race, or biological sex, which were among the major issues in the early 20 th century.Nor does it provide a detailed description of differentiation practices within a comprehensive school.Instead, the article attempts to provide an overview of what we consider to be the major shifts in the reasoning about what or who should be differentiated, when this should take place, and why.In particular, we focus on shifts: 1) from differentiation by socioeconomic class as a given factor to the search for scientific rationales for differentiation based on measurement of intellectual ability, 2) from viewing differences in intelligence as biologically conditioned and stable to viewing them as amenable to training and correction through education, and 3) from a focus on inputs to a focus on outputs, both in terms of achievement gaps and gaps between the intended and the achieved effects of school activities.These shifts developed in leaps and bounds; for example, the basic ideas behind the creation of a unified, undifferentiated school were already in place in the late 19 th century.For the sake of clarity and focus, however, we present them in chronological order.
The paper is organized as follows.The main part of the paper consists of two larger sections.First, we focus on the period from the mid-19 th century to the late 1930s, when the main concern was to maintain social order by placing the 'right' child into the 'right' school.During this period, a whole range of differentiating techniques were developed to identify and deal with 'developmentally disabled, psychopathic and so-called problem children' (SOU 1943:19, p. 33).Second, we outline developments from the 1940s to the late 1960s.This period is characterized by growing political ambitions to create a single school for all children.In this context, differentiation became a central concern.Several investigations were conducted to provide a scientific perspective on the most appropriate methods for differentiating instruction within a heterogeneous school.Thus, the search for gaps refocused from children's abilities to internal workings of the school.This way of thinking was further emphasized after the introduction of one school for all, with the emphasis shifting from measuring individual abilities to evaluating outcomes in relation to the intended purposes.The concluding section summarizes the key points and discusses how the historical trajectories of differentiation and gaps continue to shape educational policy, research, and practice in Sweden and beyond.

Sorting the poor out: political tensions and early differentiating techniques
The problem of individual differences, which according to Husén (1968, p. 325) is a fundamental problem of education, emerged as a specific political concern in connection with the rise of schooling in the early nineteenth century.In Sweden, this problem was brought to the fore after 1842, when the first public schools were established (SFS, 1842, p. 19).The formative years of public education saw a gradual transition from the Bell-Lancaster monitorial system to classroom instruction that laid the conditions for students to stand out from the masses and be recognized as individuals (Andersson, 1986).It was at this time that methodological guidelines for teachers began to appear, discussing the need to adapt teaching to the individual characteristics of students in order to make the school content relevant and meaningful for all (see e.g.Mikhaylova et al., in press).However, this approach to teaching was not discussed in terms of educational differentiation and was to be applied in homogeneous classes in terms of students' social backgrounds.
In fact, there existed different types of schools for children from the upper social classes (läroverk) and from less affluent families (folkskola).It was widely assumed that people from different social classes are by nature destined for different roles and should therefore be educated differently.By the end of the 19 th century, this assumption was increasingly reinforced, not least by the spread of eugenic ideas.The term eugenics, coined by the British psychologist Francis Galton (1869), refers to the belief that certain groups of people are inherently superior to others, and that a careful selection and reproduction of 'desirable' traits would create a 'better' society.In education, this meant that individuals from certain racial, ethnic, or socioeconomic backgrounds were deemed 'unfit' and therefore excluded from better educational opportunities, regardless of their actual abilities (see for instance Pearson, 1905Pearson, , 1909;;Key, 1900; see also Vikström, 2021).This view of school as an institution reflecting (rather than, for example, challenging) society was prevailing in late 19 th century Sweden (cf. Marklund, 1985).
The early critique of this parallel school system and the division of people by socioeconomic background came already in the 1840s, when Thorsten Rudenschöld (1845) proposed the notion of social circulation.He claimed that every individual should be given the opportunity to pursue a profession or achieve a status in life that corresponds to their aptitudes and abilities.In this way, some people could rise in their social position while others would fall down the status scale.In the 1880s, the idea that a child's abilities are more significant for their success in school than the social status of their parents was expressed by Fridtjuv Berg.The soon-to-be Minister of Education pointed out the fallacy of dividing children into 'superior' and 'inferior' based on the socio-economic status of their parents (Berg, 1883, pp. 59-60).He advocated a single basic school for all, noting that differentiation and gradual individualization should be the goal of education rather than its starting point: A distinct individuality is not the beginning of development.It is its goal.For development means nothing else but the transition from an incoherent, indeterminate, and more similar state to a coherent, definite, divided, and apparently different state, i.e., it is the same as articulation, differentiation, gradual individualization.(Berg, 1883, pp. 25-26;italics in original) In other words, Berg saw differentiation as the way of adapting the school to the abilities of individual students, not the selection of students to the 'right' types of schools.However, his ideas did not find much support.For a long time, the dominant view was that differentiation by social background, 'while not ideal, is better than no differentiation at all' (SOU 1943:19, p. 9), as John Elmgren, professor of psychology and education, put it decades later.

The birth of scientific selection: intelligence tests and ability gaps
As society modernized towards the end of the 19 th century, children's health, bodies, morals, and other 'inner' characteristics became a matter of increasing scientific interest (Axelsson, 2007).At the same time, the question of children's ability to benefit from education gained great social and political importance.Remarkably, this ability came to be linked to the notion of intelligence.In connection to this, it became important to develop tools that could be used to measure intelligence and assign students to different educational streams and 'treatments' according to their intellectual abilities (cf.Jaederholm, 1914a).Galton (1883) was one of the first to design such a tool in the form of a test that measured visual and auditory acuity, speed of reaction to stimuli, etc.Although his test was soon considered a failure, the idea was taken up and elaborated by his followers, mainly psychologists from Germany, France, England, and the USA (see e.g.Binet & Simon, 1905;Ebbinghaus, 1885;1897;Pearson, 1909;Spearman, 1904;Stern, 1920Stern, [2018]]; Terman, 1919;Thorndike, 1904).Overall, it can be said that in contrast to Galton the tests they developed refocused attention from the senses to mental capacities and the ability to learn.The legitimacy of intelligence tests was bolstered by statistical methods, including the correlation principle developed by Galton's disciple Karl Pearson (1909) and Charles Spearman's (1904) factor analysis, which provided scientific criteria for calculating norms and deviations.
The view of individual differences as something measurable and comparable had profound political and social consequences in many countries.As Gustaf Jaederholm (1914a, p. 11), an early adopter of intelligence testing in Sweden, enthusiastically noted, standardized tests allowed for the 'accurate treatment' of variations between individuals with respect to their intelligence.In line with Binet and Simon (1905), who are considered the fathers of ability testing, Jaederholm (1914a, p. 11) argued that intelligence tests were a valuable tool for singling out 'mentally defective' children who could not profit from the ordinary school.Furthermore, Jaederholm (p.15) pointed out that there is a strong positive correlation between various kinds of 'desirable qualities' of individuals; that is, high scores on intelligence tests correlated with high moral qualities and higher school grades (p.379-380).
The 'discovery' of intelligence and the development for methods for its measurement, which was actively pursued in Sweden (e.g.Jaederholm, 1914aJaederholm, , 1914b;;Siegvald, 1928), soon permeated political discourse.A governmental report for example, argued that while it was important to put the right man in the right place, in a democratic society individuals must not be selected by social class but by ability (SOU 1924:24).This, it was stressed, was of paramount importance to both the society and the individual: For society, it is important to find the most suitable people for its many tasks, both public and private, who can make the best contribution to the work.But this also promotes the best interests of the individual, for only those who find a job suited to their abilities can achieve the satisfaction and inner equilibrium that everyone seeks, and which enables them to fully develop and make use of their abilities.(SOU, 1924:24, p. 13) Certainly, this idea was made possible not only by the developments in pedagogy and experimental psychology but also by the liberalization of politics in the early 1920s.
Around the same time, however, the scientific debate about intelligence testing as a means of differentiation took a somewhat new direction.First, numerous studies conducted in Sweden and abroad questioned the validity of these tests.Second, it was found that differences in intellectual abilities depended not only on innate qualities, but also on social environment and education (see e.g.Gadelius, 1924;Siegvald, 1944).Thus, it was suggested that although genetics and socioeconomic status had an impact on children's development, there may be other factors that influenced the ability to learn and academic achievement (e.g.Siegvald, 1944).Hence, intelligence, or at least some aspects of it, can be trained (Anderberg, 1931;Jaederholm, 1914a).Third, it was recognized that while statistical evidence suggested a relationship between socioeconomic status and intelligence, it was impossible to draw firm conclusions about an individual, because what is true for a group is not necessarily true for everyone in that group (Anderberg, 1932).
Despite growing uncertainty regarding the validity and applicability of intelligence tests, they were still seen as the best available means for differentiation.As the seminary teacher Herman Siegvald (1928) explained, it was easier for teachers to refer to the results of an intelligence test to overcome parents' mistrust when their children were transferred to remedial classes.According to Siegvald, this was a 'consequence of the respect for numbers that characterizes most people' (p.110).
Furthermore, it was suggested that intelligence tests could be used not only as a diagnostic aid for the examination of mental retardation, but also as potential tools for the detection of gifted and talented children (Anderberg, 1932;Jaederholm, 1914a).Rudolf Anderberg (1932, p. 295) pointed out, for instance, that this category of children 'was left to fend for themselves' because schools, welfare institutions, and scientific research were 'more concerned with those who deviated downward from the norm'.There was, however, no consensus on the use of intelligence tests for this purpose.The psychiatrist Bror Gadelius (1924, p. 42) argued that only teachers can detect giftedness because they had a close and good relationship with their students.For Gadelius, differentiation was more a matter of the teacher's talents and abilities.Consequently, he considered careful selection of teachers through teacher training programmes more important than selection of students through examinations (p.44).
Overall, by the 1930s, differentiation came to be increasingly seen as an educational rather than a psychological concern and was more often discussed in terms of individualization.In this context, intelligence tests came to be seen as a diagnostic tool for the teacher to select teaching methods (see e.g.Siegvald, 1928, pp. 97-98).As Anderberg (1932, p. 296) noted, 'differentiation by ability is a key condition for rationally implemented individualization'.This line of reasoning was further reinforced over the next two decades, to which we will turn next.

The path to "one school for all": reframing differentiation in post-war-Sweden
The 1940s can be considered a turning period in the history of Swedish education.With the continuous democratization of society, Fridtjuv Berg's idea of a single bottom school eventually gained general recognition.It was suggested that the school system should be organized in such a way that each type and each level of education would be possible for each young citizen, irrespective of sex, class, living-and economic conditions (SOU 1944:23, p. 41).For that, schools must be able to adapt to the different educational preconditions of students (SOU 1948:27, p. 9).
Certainly, these ambitions were informed by post-war optimism and a strong belief that education was a factor in peacebuilding (cf.Marklund & Söderberg, 1969).However, there was no clarity on how to translate these ambitions into practice.On the one hand, it was argued that a democratic school should be differentiated in a variety of ways to be able to accommodate different types of giftedness in a more flexible and nuanced way than it was possible with the parallel school system (SOU 1948:27, p. 204).It was also pointed out that an early differentiation carries the risk of splitting the population into 'gifted' and 'others' (p.9).According to the 1946 School Commission's final report, 'the correct form of differentiation in the early grades [. ..] is individualization of teaching, so that due account can be taken of the condition and abilities of each student' (SOU 1948:27, p. 9, italics in original).On the other hand, initial differentiation by means of school readiness tests was still considered necessary to achieve the desired intellectual homogeneity (p.10).The Commission also emphasized the importance of achieving balance between 'manual' and 'intellectual' labour and encourage students to choose manual occupations (p.68).This suggestion was consistent with the conclusions of John Elmgren (1952), who was commissioned to determine different types of talents and aptitudes and to propose a suitable differentiation by them.However, according to Alva Myrdal (1946), a member of the Commission, the division between 'theoretical' and 'practical' talents would merely replace the old economic-based class society with a talent-based one, with a 'talented' upper class and an 'untalented' lower class.In her view, all children should attend the same school for as long as possible and be given access to both theoretical and practical knowledge (pp. 208-209).
The intensified debates on how to organize the school led to the decision in 1950 to create a nineyear comprehensive school for all children (Proposition, 1950, p. 70) and to abolish the elementary school (folkskola).The question of whether the new school would also substitute for other types of schools (realskola, läroverk, etc.) had to be carefully evaluated before the school reform could be implemented more widely.As a result of this 'precautionary principle', the 1950s was a time when numerous evaluations and investigations were carried out to determine whether the political aspirations for a single school were achievable.
The school reform was also accompanied by a lively discussion about access to higher education for children from economically disadvantaged groups.The discussion started in the mid-1940s (Husén, 1946(Husén, , 1947) ) and overlapped with questions about the number of years of compulsory schooling, its content and structure, and of course the differentiation of students into different study paths.The questions were examined in several investigations (Agrell, 1950;Hallgren, 1943;Härnqvist, {{year}} 1960;Johannesson, 1960;Sjöstrand, 1954;SOU 1949:48;SOU 1958:11), which all indicated that there were many more people qualified for higher education than had been previously thought.It was hoped that this problem would disappear with the unification of the school 'if only the problem of differentiation could be solved in a rational way' (Sjöstrand, 1954, p. 158).Other studies showed that in the higher levels of compulsory education there were large differences between students in terms of aptitude and academic achievement (Härnqvist, 1960).Against this background, Kjell Härnqvist argued that a single school with a common goal for all was unrealistic (p.112).Therefore, he continued, there must be differentiation in the broad sense of the term with courses of varying difficulty at the same age level and opportunities for variation in the range of subjects (p.112).
Thus, of all the problems faced by the experimental school of the 1950s, the question of differentiation was the most pressing.The key point was to find a balance between so-called internal -educational -differentiation and external -organizational -differentiation.According to the 1946 School Commission's recommendations, the former was to be practiced as much as possible while the latter as little as possible.To get a scientific perspective on this dilemma, the government turned to the country's professors of psychology and education, who were asked to comment on the use of aptitude and school readiness tests, as well as other appropriate methods of selecting students for different educational pathways.Four professors, one in psychology-John Elmgren-and three in education-Torsten Husén, Wilhelm Sjöstrand and Arne Trankell-responded to the call (Elmgren et al., 1959).The answers provided by them covered both children's psychological development, the question of giftedness, and the problem of how to deal with individual differences through subject choice, ability-grouping or other educational means.It goes beyond the scope of this article to summarize the answers.However, some points deserve attention as they signal a paradigmatic shift in the treatment of the question of differentiation in Sweden.
First, in defining the term 'differentiation', Husén (1959, p. 74) states that it refers to 'all measures taken to provide students with an education that meets their individual needs'.Thus, differentiation is discussed as a matter of adapting the school to the child, not the other way around.Similarly, when discussing school readiness tests that were used to predict children's ability to follow instruction in the first grade of elementary school, Trankell (1959, p. 141) noted that the very concept of 'school readiness' is artificial because it is 'constructed from a ready-made organization whose demands on children and teachers are shaped by its structure and goals'.He argued that children's ability to cope with the many different demands placed on them by school can hardly be expected to result from a uniform biological maturation process.Here again, the criticism concerns the inadequate notion that every child, regardless of his or her needs, interests, talents, and abilities, must be equally prepared for the standardized and inflexible organization called school.
Second, there was a growing distrust of measurement of any kind as a solution to social and educational problems.As Trankell (1959, p. 145)

noted,
No school problem, whether it is one of intelligence development, reading deficiency, or emotional adjustment, is inherently solvable by perfect measurement alone, because each child lives in a complex and multifaceted world of external and internal influences that must be assessed as a whole if the interpretation of individual test results is to be meaningful.
In connection to this, it was stated that it was impossible to make general recommendations regarding the best age differentiation and to predict children's success in school on the basis of intelligence or school readiness tests.Instead, as Sjöstrand (1959, p. 110) put it, 'each child must be taken for what it is at the time' and 'strive for the highest possible goal for each individual'.
Third, it was asserted that school differentiation does not merely reflect societal differentiation but also produces it (see e.g.Husén, 1959, p. 98).This recognition necessitated a discussion of the purposes of public education and the values on which it was based.As Husén claimed, 'the question of differentiation is ultimately a question of values' (p.76) and, thus, cannot be answered by psychological and educational research.
In 1962, the experimental period led to the establishment of a 9-year school for all (Proposition, 1962, p. 54), upon which some of the previously major issues disappeared from the political agenda or were relegated to the background.Instead, the responsibility for dealing with differences among students was placed on teachers.As stated in the first curriculum for the unified school, 'any form of individualization of work in a normal-sized class makes great demands on the teacher, on his interest in the pupils, his ability to adapt quickly, his knowledge and organizational skills' (Skolöverstyrelsen, 1962, p. 51).This is not to say that the issue of differentiation has lost its relevance; rather, it has taken on a new shape, which will be discussed in the next section.

Beyond ability and performance: evaluation and the gap between outcomes and purposes
This shift from organizational differentiation to individualization is clearly reflected in two editions of Husén's publication Pedagogical Psychology.Whereas in the 1958 edition Husén spoke of 'educational measurement in relation to the problem of differentiation' (Husén, 1958, p. 11), the 1968 edition discusses the same issues under the heading 'educational measurement in relation the problem of individualization' (Husén, 1968, p. 11).Husén notes, however, that both differentiation and individualization have become so 'heavily politically loaded that their meanings become emotionally charged and indefinite' (p.326).
According to Husén, the shift to individualization also meant a reorientation of the use of educational measurement from students' abilities to the intellectual, emotional, and social effects of all school activities 'in the service of evaluation' (Husén, 1968, p. 12).Moreover, it was suggested that differences in educational achievement were less dependent on differences in student's ability than on differences in the received instruction (Husén, 1968, p. 325).Hence, when discussing educational measurements, in the 1958 edition Husén argued that they could be used to estimate "the extent to which students' reactions and performance are given by his innate disposition" and how much these dispositions could be modified by education (Husén, 1958, p. 12).In contrast, the 1968 edition does not even mention innate dispositions; instead, it suggests that educational measurements enable the teacher to determine 'to what extent student reactions and performance can be modified by education and teaching?' (Husén, 1968, p. 12).Put differently, in one school for all, students' performances became more important than abilities.Furthermore, while in the earlier edition Husén argued that various psychological and pedagogical methods could be used to identify 'practical' and 'theoretical' talents and aptitudes (Husén, 1958, p. 13), in the later edition this discussion is absent.
Comparing the two editions of the same book reveals two interrelated shifts: (1) A shift from viewing differences among students as differences in their innate abilities and aptitudes to viewing them as outcomes of educational processes.This transformed the psychological and organizational problem of differentiation into an educational problem of individualization.
(2) A shift from ability tests to evaluation, broadly defined as 'the measures taken to determine the results of the school's activities' and their intellectual, emotional, and social effects (Husén, 1968, p. 12).
Whereas the first shift was already evident in the political and academic debates of the 1940s, the second one became more pronounced after the advent of comprehensive education.In essence, it refocused the question of differentiation from input to output, from the search for socially and intellectually 'defective' children to the search for ineffective or inadequate teaching practices.
Together, these shifts brought to the fore the discussion of the purposes of education.For example, referring to the American educator Ralph Tyler, who worked in the field of curriculum and evaluation, Husén (1968) states that evaluation is an inherent part of the educational process (p. 6) and starts with the formulation of the objectives to be achieved (p.12).It is in relation to these objectives, he states, that the effectiveness of differentiating and individualizing measures should be assessed and compared.Echoing Tyler,Husén (p. 175) points out that the goal of evaluation is not to get numerical results but to interpret them in relation to the objectives of school and teaching.In this description, evaluation appears as a technique for detecting gaps between intended and achieved results.At the same time, according to Husén (p. 6), evaluation involves measuring not only the more tangible matters, such as students' academic performances, but also many other things, including values, attitudes, emotions, empathy, and aesthetic taste.Moreover, since the primary goal of evaluation is to identify areas of improvement (cf.Tyler, 1942), it also becomes a new tool for calculating gaps and identifying deficiencies on both individual, group, and institutional levels.

Summary and conclusion
From the early debates about eugenics and intelligence testing to the more recent focus on individualization and evaluation, differentiation has been central to Swedish educational policy, research, and practice.For a long time, the question was not whether differentiation was necessary but when it should be introduced and on what grounds (SOU 1948:27).While in the mid-19 th century differentiation by social background was considered nearly self-evident, in the early 20 th century the problem of selection became the object of scientific expertise.By that time, a variety of techniques-ranging from gathering genealogical information of each student's family to standardized examinations and intelligence tests at the group and individual levels (Axelsson, 2007)-were developed to collect knowledge of the individuals for comparing and hierarchizing them (cf.Foucault, 1991).
Deeply embedded in biological and psychological styles of reasoning, the politics of differentiation during this period relied on a seemingly boundless faith in the ability of numbers and scientific judgement to provide objective criteria for the selection of individuals.Eventually, the ubiquitous measurements and anonymous 'bureaucracy of statistics' (Hacking, 1991) allowed for the calculation of gaps between individuals in terms of intelligence, personality, developmental stages, and aptitudes for education.In this way, experimental psychology created a dominant account of normalcies and deficiencies which came to play a crucial role in establishing classifications and regulating access to education.In the spirit of eugenics, this account focused on social, genetic, and intellectual gaps between 'superior' and 'inferior' individuals and sought correlations between these gaps.Moreover, until the late 1920s there was little effort to overcome the gaps.On the contrary, the quantified knowledge of individuals legitimized 'negative' or 'positive' selection of students, which was considered necessary for homogenizing schools and facilitating teaching.Thus, for nearly a century, public education in Sweden was concerned with how to 'fit' the child to the school, not the school to the child.
With the democratization of society, accompanied by the 'epistemological "thaw" of the sciences of the individual' (Foucault, 1991, p. 191), the notion of intelligence gradually gave way to aptitudes and talents, signalling a new way of reasoning about differentiation.Whereas intelligence was seen as something stable, aptitudes and talents were seen as something that could be developed under the right conditions.Consequently, while low scores on intelligence tests could lead directly to exclusion, the idea that different talents could be of value for society sparked a discussion of how to embrace them in school.Although democratic in its ambition, this shift created a gap between theoretically and practically oriented students, which, according to some critics, would merely replace economic-based social hierarchies with talent-based ones.
Overall, by the 1940s and especially in the post-war period, differentiation came to be seen not as a technique for solving other problems, but as an ethical concern that should be taken care of within the school.After the creation of a single comprehensive school in 1962, the very notion of differentiation acquired negative connotations and was eventually replaced by individualization.It was recognized that variation in student behaviour and performance could be attributed to different norms, demands and expectations in the environment.Accordingly, individualization, as opposed to organizational differentiation, implies accommodating differences by adapting instruction and environments to individual abilities and needs, not vice versa.These arguments fully reflected the progressive individualist ideas of the time and provided the ideological basis for individually differentiated education in one school for all.Paradoxically, these ideas were realized in the same decade that Sweden introduced its first standardized curriculum (Skolöverstyrelsen, 1962).
As a form of differentiation, however, individualization involves similar styles of reasoning to that of organizational differentiation, since it presupposes the identification of gaps between students through comparison, measurement, and examination.The 'uniqueness' of each student is thus established along a scale of differences within a normalized and standardized framework (cf.Foucault, 1991).Even though these differences and gaps are no longer attributed solely to individuals' innate characteristics, individualized instruction means identifying and focusing on those who deviate from the norm in some way.As such, individualization does not really concern the average student, whose social background, abilities, and academic performance are considered 'normal'.
The rise of evaluation practices in the 1960s refocused the question of differentiation from diagnostic measurements of individuals to retrospective measurements of school activities, from input to output.Nonetheless, like other differentiation techniques, evaluation is essentially 'defectoriented' in that it seeks to detect gaps at the individual, group, and institutional levels and to locate the failures and problems to be addressed.Thus, throughout the history of public education in Sweden, the question of differentiation has undergone significant changes, reflecting shifts in societal values, scientific paradigms, and educational goals.These changes are also reflected at the discursive level, i.e. in specific concepts (e.g.general ability, intelligence, giftedness, talent) and binaries (e.g.differentiation/individualization, inclusion/exclusion, normal/'retarded' or normal/gifted) that have accompanied political and scientific debates on the matter at different times.Rooted in institutionalized hierarchies of values, such concepts and binaries do not simply describe and label individuals as 'gifted', 'normal', or 'weak', but discursively establish and perpetuate gaps by hierarchizing differences and assigning individuals to particular social positions.
Certainly, the issues described in this paper are not confined to Sweden; neither are they limited to a specific historical period.As a specific technique for detecting and addressing gaps, differentiation has been a crucial element of educational reasoning.Moreover, with the advent of international large-scale assessments, its application has been extended from the level of individuals, groups, and institutions, to entire national school systems, allowing for the identification of racial, ethnic, and gender gaps in student achievement across countries.As with other differentiation practices, international large-scale assessments tend to generate interest for deviations and outliers, while the 'averages' receive little attention (see e.g.Steiner-Khamsi, 2003).In other words, even if the term 'differentiation' itself has been discursively replaced by others, the ideas underlying it-the search for gaps and the distribution of differences along a normalized scale-continue to shape education in Sweden and beyond.