Teacher politics bottom-up: theorising the impact of micro-politics on policy generation

ABSTRACT

Education policy is generally understood as a multi-layered process, consisting of diverse interconnected phases. Studies of these interconnections typically ask whether and how the generation of official policy from the top-down affects micro-politics, i.e. how teachers experience and execute their work. The assumption that policy is influenced in the reverse direction is widely held, but has seldom been studied empirically. Little is known, therefore, about how this dynamic operates. This study delineates mechanisms that link teachers’ micro-politics to the macro-politics of policy generation. Analytically, it combines concepts from the literature on teacher involvement in macro- and micro-politics in order to develop a framework bridging the two. Empirically, it harnesses the theoretical potential of Swiss language education policy, tracing the process of reforms through which teachers, though formally excluded from policy-making, were able to influence the choice of languages included in official curricula. The analysis identifies three mechanisms through which they exerted influence: voicing experience, subversive enactment, and open resistance. None of these are dependent upon higher levels of teacher unionisation or particular institutions of governance. These findings highlight the importance of engaging with the processual dimension of politics to advance our theories of educational policy.

KEYWORDS:

• Educational policy
• teachers
• politics of education
• language policy
• curriculum-making
• process tracing

Introduction

Schooling embodies different meanings for different stakeholders. Some see it primarily as an engine of economic growth, others as a means to secure social mobility. Still others believe its main focus should lie in shaping the next generation’s character and morals. For some of those involved, including teachers, education systems also constitute a workplace and a source of revenue. Accordingly, contentious political processes normally precede the design and enactment of new education policies, which Ball (2006) characterises as ‘cannibalised products of multiple (but circumscribed) influences and agendas’. In order to advance our theories about why formal education looks the way it does, it thus is crucial for us to understand how power is wielded in education politics. What means allow actors to shape policy? Systematic answers to this question, from either a comparative or historical research perspective, are largely lacking (Ball 2015; Moe and Wiborg 2017; Sivesind and Westbury 2016). This is, by and large, the case or one of the actors considered most critical in shaping formal education: teachers.

Studies of teachers’ influence on education policy can generally be classified into two strands according to the policy phase they focus on (see section 2). These strands show markedly different understandings of teachers as political actors. On the one hand, studies of the policy generation phase (macro-politics) are primarily concerned with teachers’ roles in the making of the authoritative policy texts that regulate formal education – what I call official policy. These studies usually understand teachers as complex actors – unions or pressure groups – driven by professional concerns or vested interests. On the other hand, research on micro-politics focusses on the policy enactment phase. This literature tends to conceive of teachers as individual actors whose political scope resides in their monopoly over the enactment of official policy in schools.

Both strands of research draw on predominant conceptualisations of education politics as a complex, multi-layered and multi-directional process (e.g. Ball 2006; Ricento and Hornberger 1996; Sutton and Levinson 2001). More specifically, both call for analysis that acknowledges macro- and micro-politics’ respective institutional constraints, constellations of actors, and dominant mechanisms. But while recognising these two layers as interconnected, scholarship tends to be somewhat unidirectional. Policy enactment studies often inquire into how the generation of official policy affects teachers’ practices and self-understanding. We know little, however, about the reverse process: the mechanisms by which teachers’ micro-politics might shape official policy.

Therefore, this study combines the above approaches to devise an analytical framework that allows for the identification of mechanisms connecting teachers’ micro-politics – their policy enactments and communication about them – with official policy generation. Section 2 presents this framework. To identify and specify such mechanisms empirically, I perform process tracing on selected reforms regarding the languages included in the official curricula of Swiss primary and lower secondary schools. The study concretely asks: what mechanisms link teachers’ micro-politics with the generation of official Swiss language education policy?

This type of decision is particularly suited to my investigation. Which languages the next generations are (and are not) encouraged to learn has major implications for crucial political issues, from civic equality to economic development, from international relations to the viability of democratic deliberation. These decisions therefore constitute a subset of education policy where politicians’ ideological and economic concerns are expected to predominate. This is especially true for an officially multilingual country such as Switzerland, where languages assume extra political meaning, sometimes creating a political consensus around the need to teach children additional languages (see section 3). The mechanisms through which teachers’ concerns and micro-political actions affect official policy here are thus likely to also operate in other education policy areas.

The Swiss case offers a second crucial advantage. After constituting itself as a modern federal state in 1848, Switzerland continued to delegate education policy to its 25 (since 1979, 26) semi-sovereign states, called cantons. Therefore, since the mid-nineteenth century, Switzerland has deliberated a large number of language curriculum reforms, providing a multitude of potential cases for analysis. Since my aim is to explore connections between teachers’ micro-politics and policy generation, I focus on reforms whose outcomes, according to previous studies (Giudici 2019), were affected by the teachers who were supposed to enact curricular provisions, despite their not being formally included in negotiations. Hence, the analysis excludes the more recent reforms undertaken since the 1980s, when most cantons adopted provisions institutionalising teachers’ representation in education politics (Höhener and Criblez 2018).

My analysis finds three different mechanisms connecting teachers’ micro-politics to policy generation. I call them voicing experience, subversive enactment, and open resistance (section 4). All three are triggered by teachers’ micro-political actions outside the panels formally tasked with generating official education policy. They are thus largely independent from teachers’ degree of unionisation or their veto-power (or lack thereof) in deliberations. It is teachers’ role as the professionals who deliver education, and their licence to communicate their experiences of doing so, that allow them to exert an impact via these mechanisms, not their organisation and position as a political pressure-group.

As argued in the conclusion (section 5), my findings support calls for investigations that bridge the divide between the generation of education policy and its enactment, and explore their interaction. To improve our understanding of how today’s formal education came to be conceptualised and regulated, we need to theorise not only the institutions and formal power distribution characterising the underlying processes, but also the processes themselves.

Extant approaches to teachers’ role in the politics of education

There is a common assumption within the politics of education that official policy determines what happens in classrooms (Hopmann 1998). This linear understanding, however, has been largely dismissed within the scholarly fields of education (Ball 2006; Schulte 2018), curriculum (Hopmann 1998; Sivesind and Westbury 2016), and language education policy (Ricento and Hornberger 1996), as well as by policy analysts more generally (Kingdon 1984). These fields have claimed the need to understand policy-making as a complex, multi-layered process, which engages actors not only at the level of government, but also within classrooms and all the bodies and communities of practice in between.

The dismissal of linear models of policy-making has triggered different analytical reactions. Arguing that actors ‘produce linguistic and scholastic norms anew in each relevant movement of political negotiation or practice’ (Hurdus and Lasagabaster 2018, 230), some scholars suggest abandoning any analytical distinction between policy generation and enactment. Most researchers in the aforementioned fields, however, contend that scholarship should retain and even reinforce these distinctions. They claim that it is only by engaging separately with every layer, that the specific actors, institutional constraints, and mechanisms shaping policy at each stage can be identified and theorised.

As a result, the literature on educational policy-making has progressed in two distinctive directions (Bowe, Ball, and Gold 1992; Schulte 2018). On the one hand, we find studies on what is usually referred to as policy generation or macro-politics. Following what the language education literature calls the historical-structural public sphere paradigm (Tollefson 2012), this strand mostly constitutes ‘macro-based theoretical analyses of policy documents and the activities and organisation of groups of policy makers’ (Bowe, Ball, and Gold 1992, 6). Its focus lies in relating official policy texts to institutional constraints, dominant ideologies, and power distribution; less attention is paid to political processes and actors’ formative roles therein (Ball 2015; Tollefson 2012). On the other hand, another set of studies engages with policy enactment (micro-politics). This scholarship is mainly interested in theorising how actors engage with policies on the ground. Aligning with what Tollefson (2012) dubs the creative public sphere paradigm, these studies often involve in-depth ethnographies in schools (Björk and Blase 2010), and are predominantly concerned with individuals’ formative roles in the policy-making process.

The two phases are often couched in opposite terms, drawing on dualisms such as ‘structure vs. agency, global/national policies vs. local actors’ (Schulte 2018, 624, see also Tollefson 2012). As the next sections show, this dualistic opposition also characterises research on teachers’ political involvement. Studies on policy generation (section 2.1) and on policy enactment (section 2.2) rely on different conceptions of teachers as political actors, as well as different theoretical frameworks to explain their impact.

Policy generation: teachers as a complex actors

Research on policy generation is interested in the development phase of official policy. This phase is typically understood to be dominated by organised interests and ideas whose impact is mediated by government-related institutional constraints (Bowe, Ball, and Gold 1992; Tollefson 2012). Therefore, studies on language and education policy generation tend to conceive of teachers along the lines of Scharpf’s (1997) definition of a complex actor: as a collective whose internal decision-making procedures enable it to act on behalf of all its participating individuals. Teachers’ status and working conditions are regulated via official education policy, and therefore hinge on their capacity to introduce their concerns and interests into policy generation. Teachers thus have strong incentives to overcome collective action problems and organise into complex actors such as unions (Boyd, Plank, and Skyes 2000; Moe and Wiborg 2017).

Comparative and historical studies mainly relate teachers’ strategies and impact on policy generation to two interconnected factors: their degree of organisation and the institutional framework in which they operate (Boyd, Plank, and Sykes 2000; Moe 2015; Ozga and Lawn 1981). In his analysis of U.S. teachers’ politics, for instance, Moe (2015) argues that it was the states’ adoption of collective bargaining laws that turned teachers from ‘an atomized and politically weak constituency’ into massive, extraordinarily powerful unions. Similarly, the language education policy literature finds that the organisational degree and effectiveness of different language-subject communities affects the status of their corresponding subjects in curricula (e.g. Dubois 2012; Extermann 2013). The institutional aspect considered most significant is whether teachers are represented and furnished with veto-power within the fora formally entrusted with deliberating official education policy. Teachers are found to have a stronger political influence in countries that rely on corporatist models of governance to integrate professional demands into education policy – either de jure, as in some Northern European countries, or de facto, as in France or the U.K. (Moe and Wiborg 2017). Indeed, institutional changes, such as the introduction of market-inspired governance models, have been identified as key explanations for the declining degrees of unionisation since the 1980s, and teachers’ consequent loss of political influence (Boyd, Plank, and Sykes 2000).

The mechanisms ensuring teachers’ influence on policy generation have also been shown to depend on the institutional setup of government. Where teachers are collectively represented in official negotiations, they can make an impact via mechanisms of persuasion or coalition-forging (Hopmann 1998; Sivesind and Westbury 2016). Where this is not the case, such as in the U.S., studies find teachers’ organisations rely on typical pressure-group strategies such as lobbying, forging alliances with political parties, supporting sympathetic candidates, or even running for office themselves (Boyd, Plank, and Skyes 2000). Where the possibility exists, they can also engage in referendum campaigns or choose more disruptive forms of political protest such as strikes or demonstrations (Wiborg and Larsen 2017).

Individual teachers’ micro-politics and enactment spaces

Scholars interested in policy enactment understand teachers’ politics in a markedly different way (Blase 1991; Ball, Maguire, and Braun 2012; Hornberger and Johnson 2007). Their focus is not on unions, but rather on teachers as individual actors or localised communities whose work and self-understanding are affected but not determined by official policy. Indeed, this work shows that what is practised in and around classrooms seldom corresponds to what is officially legislated, highlighting how policy-making by no means ends with the generation of new legislation.

The mismatch between text and practice is generally reduced to two clusters of factors. First, schools are no tabula rasa. How they deal with new policy texts always depends on contextual and institutional constraints that are specific to the enactment phase, such as school location, infrastructure, and budget, as well as prior policies and expectations (Ball 2006; Ball, Maguire, and Braun 2012). Second, teachers have agency. As theorised by Ball, Maguire, and Braun (2012), the enactment of policy texts requires a process of interpretation and translation; it thus implies repeated negotiations about policies’ meanings and ways they may be put into practice. Studies show this process can yield different results depending on teachers’ knowledge and practical experience (Choi 2017; Hornberger and Johnson 2007; Kirsch 2018), political convictions (Schulte 2018), or interests (Kelchtermans and Ballet 2002; Leonardi 2017). Furthermore, how teachers deal with official policy can be shaped either by their participation in communities of practice (Colburn and Stein 2006), or by personal experiences and interests (Schulte 2018; Leonardi 2017).

Consequently, these scholars argue, individual teachers’ actions on the ground, ‘consciously or unconsciously motivated, have political “significance”’ (Blase 1991, 11). Teachers’ singular role within the education system provides them with the power to resist policy change, alter it, or forge new policy on the ground (Hornberger and Johnson 2007; Ball, Maguire, and Braun 2012) – even under repressive governments (Schulte 2018; Choi 2017). This can have a substantial impact on schooling. For the case of language education, for instance, ethnographic studies find some teachers inadvertently or deliberately recognise and employ multiple languages in their classrooms, despite official policy requiring monolingualism (Kirsch 2018). Conversely, others ignore official policy’s multilingual orientation and allow the use of one language only (Hornberger and Johnson 2007).

Interactions between micro- and macro-politics

The importance of scrutinising the different policy layers’ interconnections is regularly reiterated in the literature. Still, one effect of the diversification between macro- and micro-political studies has been that policy generation has come to be considered as taking place remotely from, and prior to, policy enactment (Bowe, Ball, and Gold 1992; Schulte 2018). Thus, while several analytical tools have been developed to theorise the connections that stretch from the generation of official policy to teachers’ micro-politics (e.g. Björk and Blase 2010; Ball, Maguire, and Braun 2012; Schulte 2018), the reverse connection has seldom been the object of research. Studies of micro-politics sometimes even explicitly limit teachers’ power to their classrooms, schools, or local communities (Björk and Blase 2010; Kirsch 2018). Hence, while their re-interpretation of, or resistance to, official policy – e.g. by using different languages in the classrooms than those stipulated by official regulations – is shown to have profound effects on communities and pupils’ learning, research seldom considers whether and how these actions might inform subsequent policy generation, e.g. altering the languages included in the official curriculum.

This unidirectionality has been repeatedly problematised. Ball (2006, 47) criticises extant policy education scholarship for ‘assuming the adjustment of teachers and context to policy but not of policy to context’ (see also Bowe, Ball, and Gold 1992; Sutton and Levinson 2001). Similarly, language education scholars Ricento and Hornberger (1996, 418) call for researchers to not only investigate policy-making from a top-down perspective, but to acknowledge that teachers ‘can transform classrooms, thereby promoting institutional change that can lead to political and, ultimately, broader social change’ (see also Hornberger and Johnson 2007; Johnson 2018).

That such an impact is possible has recently been argued by studies on teachers’ roles in the contemporary resistance to high-stakes testing in the U.S. (Sundstrom 2018; Lingard and Hursh 2019; Lynch 2015). They find that, with the major teachers’ unions adopting ambiguous stances towards these reforms, it was teachers’ (and parents’) micro-politics – their submitting letters to the authorities, connecting with the community via local media and students, or incorporating discussions about high-stakes testing into the curriculum – that impelled state legislators to water-down these reforms, for instance allowing parents to opt out. This begs the question, what are the mechanisms that make such impact possible?

Analytical and methodological approach

This study combines insights from both approaches described above in order to uncover and specify the mechanisms through which teachers’ micro-politics can impact the generation of official policy. Methodologically and empirically, this is done by performing process tracing on selected language education policy reforms in Switzerland.

Process tracing as a method of analysis

Process tracing is a method explicitly designed to uncover causal mechanisms (Bennett and Checkel 2015). Relying on a mechanistic inferential logic, it posits that, to establish a relationship between determinate starting conditions and the outcome of interest, researchers must identify the process that leads from the former to the latter, and disclose its underlying mechanisms. Different conceptualisations of mechanisms have been put forward. This study draws on Mayntz’s (2004, 214) contribution to the process-tracing literature, which defines mechanisms as ‘recurrent processes that link specified initial conditions and a specific outcome’ (her italics).

This definition implies that mechanisms are neither abstract law-like statements nor idiosyncratic explanations that hold in a single case only. Instead, they should ‘appear repeatedly in reality if certain conditions are given’ (ibid., 253). Mechanisms of this type must specify the actors involved in the underlying process, their motivations and policy preferences, as well as how these preferences aggregate and are institutionally mediated to produce the outcome. They must provide the most likely explanation for all the steps delineated by the data documenting the process, as well as for its timing and pacing (Gryzmala-Busse 2011; Mayntz 2004). Hence, as a method that stays close to empirical data, but uses them to refine generalisable mid-range theories, process tracing offers an ideal instrument to relate ‘the ad hocery of the macro with the ad hocery of the micro without losing sight of the systematic bases and effects of ad hoc social actions’ (Ball 2006, 43).

Tracing Swiss language education reforms

Literatures in history, socio-linguistics, and education have developed different explanations for why languages are included in official curricula. Following Bennett and Checkel (2015), these theories can be grouped into three categories. First, there are cognitive explanations based on powerful actors’ ideas about either education, nationalism, or justice. Second, structural theories postulate the impact of economic or political constraints such as trade patterns or strategic alliances. A third category comprises theories relying on the views and interests of crucial stakeholders: political elites, parents, or teachers (see, Giudici 2019, for an overview).

Each of these theories has largely been studied in isolation. Studies rarely discuss whether a theory retains its explanatory power when alternative explanations are also considered. In line with the macro-structural orientation of policy-generation scholarship, nationalism and economic constraints dominate the aforementioned literatures. When teachers’ engagement is considered, the primary focus has been the lobbying efforts of subject-specific pressure-groups (e.g. Dubois 2012; Extermann 2013). Some historical reconstructions also hypothesise that teachers’ refusal to comply with official language policy requirements in certain contexts – for instance, the banning of regional languages in nineteenth-century-French classrooms (Judt and Lacorne 2004) – might have facilitated corresponding changes of official policy in the long term (see also, Harp 1998). They do not, however, specify the mechanisms that made this possible.

The tendency to couch language education reforms in nationalist terms is particularly pronounced in studies on Switzerland. The country became one of the first multilingual modern states when, in 1848, its Federal Constitution declared German, French, and Italian ‘national languages’ (art. 109).¹ In retrospect, this act has often been represented as a declarative statement in favour of a plurilingual Swiss society and national identity (e.g. Haas 2000). Empirical studies, however, do not support this interpretation (Widmer et al. 2004). Indeed, curriculum documents show that the great majority of Swiss cantons – most of which are monolingual – did not introduce the general population to more than one language earlier than any of Switzerland’s officially monolingual neighbours (Giudici 2019).

In a previous project, I used deductive process tracing (Bennett and Checkel 2015) to systematically assess which of the aforementioned theories provide the most (and least) suited explanation for Swiss language education reforms. In line with other scholarship (Widmer et al. 2004), the analysis finds that Switzerland’s early institutional multilingualism and economic relations with its different-speaking neighbours politicised language-related issues, often also turning language education policy into a conflictive topic. Still, in some instances, these circumstances also created political consensus around the need to introduce children to additional languages. Corresponding reform proposals did not always succeed, however. In a number of cases, the process leading to this outcome did not align with the implications of theories based on influential ideas, economic or political structures, nor elites’ and parents’ concerns. The most valid explanation was the opposition of the teachers tasked with actually delivering foreign language teaching, even though they had no formal representation or veto-power during deliberations (Giudici 2019). This analysis performs inductive process tracing on these cases.

Inductive process tracing is a theory-guided method to decompose a process and identify the links between its constituting events. The aim is to ‘identify different patterns of sequences and their related causes and consequences’ (Falleti 2016, 457), and thus to detect the mechanism linking starting conditions with the outcome. This requires paying attention to theoretically relevant aspects of the process such as the timing, sequencing, and pacing of decision-making; the actors involved and their stances; as well as how their actions aggregate to produce change. This type of analysis allowed me to specify the mechanisms by which teachers’ micro-politics aggregated to produce change, group reforms in which teachers’ influence relied on the same mechanism, identify scope conditions, and exclude reforms which resulted from idiosyncratic coincidences rather than generalisable mechanism.

Methodologically, Swiss language education reforms provide ideal presuppositions for this analysis. First, Swiss language education policy represents an ideologically-charged area, making it more difficult for teachers to make their concerns heard. Second, during the period of analysis, Switzerland followed a non-corporatist state-led tradition of curriculum making (Hopmann 1998; Sivesind and Westbury 2016). Until the 1980s, most cantons did not integrate teachers’ interests into official policy-making, and, while advanced schools enjoyed more autonomy, the Swiss primary and lower secondary schools examined here had no formal say in issues regarding the subjects included in the official curriculum; these lay within the competency area of parliaments, governments, and administrations. Moreover, historiography shows Swiss primary and lower secondary school teachers – unlike high school or university teachers – did not hold a particularly high status in official politics. Until the 1990s, teaching these grades required only basic vocational education, and women – who lacked political rights until the 1970s – often predominated (Grunder 2008). The mechanisms that find scope here, therefore, are also likely to function in other countries and education policy areas.

A last methodological advantage is provided by Swiss federalism. The fact that Swiss education policy lies with the country’s 25 (since 1979, 26) cantons – and in the early nineteenth century, even with cities – accordingly multiplies the number of potential reforms to trace. This study considers the language education policies of an economically and politically diverse sample of 13 cantons from Switzerland’s four language regions,² from the first half of the 19th century, when Swiss education policy was beginning to take its modern, state-led shape, until the participative turn of the 1980s (Höhener and Criblez 2018). Mechanisms operate under different temporal conditions (Gryzmala-Busse 2011). The consideration of such a broad period of time, therefore, not only multiplies the potential cases for analysis, but also allows the observation of mechanisms that develop more slowly.

Data and analysis

Following the methodological rationale of inductive process tracing, sources’ probative value is determined, first and foremost, by their usefulness for reconstructing theoretically relevant aspects of a process. This rationale has two implications (Bennett and Checkel 2015; Falleti 2016). First, it is not primarily the number of sources, but sources’ discriminating power and diversity that matters. This study thus draws on diverse types of data. Second, sources require a particularly careful assessment. This is because sources documenting political processes normally have been produced by the actors involved. Following methodologists’ advice, the analysis prioritises private communication or deliberations over statements made in public, where there are greater incentives to strategically misrepresent one’s position by justifying it with popular ideas.

Data gathered and analysed for this study fall into two categories (for detailed information on data selection, see the Online Appendix). A first type of source allowed me to systematically reconstruct which languages were included in the 13 cantons’ mandatory schooling during the analysed time-frame. These are these cantons’ official curriculum regulations – namely, laws syllabi, and decrees –, which have been systematically compiled for the inter-regional research project Construction and transformation of school knowledge since 1830.³

A second more heterogenous set of sources was used to trace the processes underlying relevant reforms. These include, first, reports and files documenting educational administrations’ activities, on the basis of which I reconstructed reform processes’ pacing and timing, as well as the actors involved. Second, minutes of deliberations of as well as correspondence to and from educational administrators and the various commissions deliberating curriculum policy were used to detect the positions of the actors represented in the policy generation phase, as well as uncover whose views these actors were exposed to, including those of teachers. To further document the views and strategies of the latter group, I relied on an extensive collection of newspaper articles, letters, and statements gathered by the educational ministries, as well as on teachers’ organisations proceedings and a systematic collection of curriculum-related articles appearing in nine canton-based and regional teacher reviews.

Three mechanisms linking micro- and macro-politics

The next three sections present the mechanisms which, according to the analysis, connect teachers’ micro-politics with change in the generation of official policy. They are called: voicing experience (4.1), subversive enactment (4.2), and open resistance (4.3). Each is exemplified by one or two reforms in German-, French-, and Italian- speaking cantons, selected because they provide the most exhaustive and illustrative evidence.

Voicing experience

The most intuitive (but perhaps analytically underestimated) way for teachers to influence policy with their concerns and practical experiences is to convince those in power of their validity. In mechanismic terms, teachers have to frame and deliver their position in ways that change key actors’ beliefs. Since teachers were not formally included in negotiations in the cases analysed here, they mostly achieved this by writing letters, either alone or in groups, or submitting petitions to parliaments or governments. Sometimes they were aided by an intermediary, such as a school inspector, who helped them introduce their positions to the authorities. The sheer quantity of communications stored in education departments’ archives shows that teachers made abundant use of these options from early on. In cases in which this strategy was successful, drawing on their experiences in the classroom to frame the teaching of multiple languages as educationally impracticable or unnecessary (or both), they managed to win over some participants formally involved in negotiations. The following illustrates how this mechanism operated in practice.

One of the earliest examples of voicing experience can be found in nineteenth-century Basel. In 1817, this wealthy German-speaking city decided to modernise its schooling system. As part of this reform, following the advice of an expert commission composed of humanists from the local university, the authorities introduced Latin to the boys’ compulsory primary school curriculum. According to the experts, Latin provided ‘an introduction to all language teaching’ (Universitäts-Commission 1817, 6), thus laying the groundwork for the systematic acquisition of German, as well as the modern and ancient languages taught in subsequent schooling.

Contemporary sources document that primary school teachers were not happy with this decision, and reached out to the authorities directly. As later reported by the school inspector, many teachers considered Latin as requiring ‘much effort and work for nothing’ (Hess 1889, 9–10), and some had given up teaching it. After repeatedly voicing their opinion to the authorities, in 1827 several teachers instructed the inspector to submit a petition to the education board, asking for the elimination of Latin from the curriculum. No written record of the board’s subsequent deliberations seems to have survived. However, shortly after receiving the petition, the board decided to make German the only language taught in primary schools. A ministry report issued years later to justify the decision before Parliament explicitly relies on teachers’ arguments contending that Latin lessons for primary schoolchildren are infeasible and ineffective (Erziehungsdirektion Basel-Stadt 1839).

Just over a decade later, a similar scenario unfolded in the Italian-speaking canton of Ticino. In 1841, Ticino’s Parliament passed legislation to create a new type of lower secondary schooling tailored to the needs of middle-class boys. During deliberations, it was agreed across the entire political spectrum that, given the dominance of French and German in local trade and politics, it was crucial to offer the future tradesmen and administrators the opportunity to learn both. Shortly after the first schools opened, however, the administration began receiving letters from teachers and inspectors, claiming that delivering all the subjects politicians had included in the curriculum was impossible. The senders’ opinions as to the subjects to prioritise differed. Almost everyone, however, wanted to eliminate either German or French.⁴ Their main arguments were the impracticality and the costs associated with offering both languages in schools, especially since most teachers were unable to teach three languages on their own.

No other actor intervened in the debate, meaning that these were the only arguments against additional language teaching present when, in 1846, the Ticino Parliament met to deliberate whether and how to pare back these schools’ curriculum. Still, in deliberations, German and French were the only subjects actually up for elimination. Some politicians did request further investments in order to keep them both in the curriculum. However, a majority finally voted to make offering these languages optional. All the members of Parliament who spoke in favour of this legislation reiterated teachers’ arguments, declaring that they would have preferred schools to continue offering both languages, but that this had proven impossible to implement.⁵

It must be noted that at these points in time neither the teachers in Basel nor those in Ticino had formed any organisation. Beginning in the late nineteenth century, as teachers began to organise, these uncoordinated forms of communication were sometimes complemented with more strategic and concerted actions. Still, even when deploying the standard pressure-group toolbox to re-frame an issue and convince those in power of their position, teachers continued to rely on their legitimacy as practitioners and on more localised forms of action.

This development can be illustrated by the case of French-speaking Geneva, where, in the late 1880s, Parliament introduced German as a compulsory subject in primary schooling. In the years that followed, numerous complaints and requests from teachers to revoke this decision reached the authorities in the form of letters and petitions. They called German-teaching ‘a nightmare for teachers and pupils’ (Pesson 1905, 230) and argued that the idea of conveying this language to pupils ‘even in a rudimentary way is utopian’ (Mercier 1904, 19). This pressure, however, did not suffice to change politicians’ consensus that German had to be kept in the curriculum for economic reasons.

Consequently, in 1922, a local primary teachers’ organisation launched an enquiry into the benefits of German, involving two hundred firms and teachers of advanced types of schooling. While it is impossible to determine the reliability of the enquiry’s findings, the organisation did use them (in a letter to the authorities) to refute the economic benefits of learning German, adding that the subject also caused education problems as it ‘does not conform to modern ideas regarding the reduction of pure intellectualism’ (quoted in Jordi 2003, 105). After hearing this combination of arguments, the minister of education, the inspectorate, and the administrative bureau of primary schooling decided to defy Parliament. Contradicting the law, as well as their previous statements, one year later, this group of actors adopted a new syllabus which postponed the start of German lessons from the third to the sixth year of schooling and limited their access to more academically successful students. The authorities waved the policy through despite its dubious legal status (Rosier 1923).

Subversive enactment

In the processes just outlined, teachers conveyed their message by reaching out to the authorities. Direct communication, however, is not the only mechanism allowing practitioners’ concerns to be heard. In some cases, policy change can be linked to teachers’ collective (though not necessarily coordinated) micro-political acts of non-compliance. This mechanism is best illustrated by a scenario that unfolded in the German-speaking canton of Basel-Stadt.

In the 1870s, the government of Basel-Stadt issued a legislative proposal suggesting the introduction of French in all the canton’s primary schools. Even before the consequent deliberations officially started, several teachers had issued letters and statements asking members of Parliament to reject this suggestion. They argued that the subject was too difficult for most primary school pupils (e.g. Koch 1877).

Legislators did not accede to these requests. For different reasons, both of the parties then in Parliament favoured compulsory French-language instruction. Conservatives believed the language to be important for the local, trade-based economy. Liberals, on the other hand, saw the inclusion of French in the schooling offered to working-class children as a means to enhance social equality.⁶ French thus became part of the primary school curriculum, but upon the request of the education board, Parliament adopted a provision that allowed teachers to grant pupils a dispensation from French-language instruction if they and the education board judged the pupils absolutely incapable of receiving it. Several politicians emphasised that this rule was only intended for highly exceptional cases (hence the need to ask for approval on a case-by-case basis). As the minister of education, Paul Speiser, put it, the creation of a French-less track for the ‘spiritually poor’ should be avoided at all costs (ibid.).

This, however, is exactly what subsequently developed. The official statistics show that teachers made extensive use of the dispensation provision, and that since they were the ones who actually knew the pupils, their requests to do so were usually granted. In the mid-1880s, the number of pupils receiving a dispensation in each school was so large that teachers began to group them together. The result was the establishment of a heterogeneous array of courses, which compensated for the lack of French with other content (Regierungsrat Basel-Stadt 1911). In 1912, the Department of Education tried to bring some order to the situation. Administrators grouped these classes under the label of B-classes and gave them a unitary, practically-oriented programme: girls’ B-classes compensated for the lack of French-instruction with lessons in house-keeping and needlework, while boys learnt additional German, gardening, and woodwork (Regierungsrat Basel-Stadt 1922, II 10).

This provision sat at first on legally shaky ground, but quickly became regularised. Indeed, the proposal for a new education law submitted by Basel-Stadt’s government to Parliament in 1922 declared the B-classes’ curriculum the new standard for regular primary school (Regierungsrat Basel-Stadt 1922). Unsurprisingly, members of Parliament were far from pleased. In its official statement, the parliamentary commission charged with pre-deliberations called this provision ‘a painful capitulation’ (Grossratskommission Basel-Stadt 1927, 16). Similar reactions dominated in Parliament itself, where representatives from all parties were outspoken in their discomfort with limiting the teaching of French to gifted students (Grosser Rat Basel-Stadt 1928). However, in the end, a majority voted against an amendment aimed at keeping mandatory French in the curriculum. As several representatives noted – not without frustration – the provision only legalised a situation that schools already practiced: fighting it was pointless (Grosser Rat Basel-Stadt 1928). This decision was not revised until the 1960s.

Open resistance

I label the third mechanism open resistance. This mechanism seems to be more specific to the policy-making models introduced in Switzerland and elsewhere, starting in the mid-twentieth century (Höhener and Criblez 2018). Developed by contemporary education- and curriculum-experts, these models stipulated that potential reforms should be tested prior to the legislative process. This practice was meant to furnish politicians with evidence on the effects of potential reforms.

The new policy-models did not give teachers formal representation or veto power in negotiations. However, the analysis shows that, inadvertently, these models furnished teachers with additional implementational spaces, not only after, but also before official policy was generated. In fact, with evidence-gathering and trials becoming a prerogative for policy change, authorities needed teachers’ collaboration to embark on reforms – collaboration teachers could choose to withhold. Hence, in the reforms grouped in this section, teachers not only voiced their experience to those in power, but actively withheld, or threatened to withhold, their collaboration in the preparation or enactment of official policy. Contrary to the subversive enactment mechanism, teachers thus openly resisted policy change, and policy-makers were aware of this resistance.

One internationally prominent reform to be achieved through this policy-model was the introduction of foreign language teaching in early primary education. In the 1960s, several countries engaged in this reform, promoted by international organisations such as Unesco and the Modern Language Association (Giudici 2019). The main Swiss teacher union also endorsed it (Konferenz der Schweizerischen Lehrerorganisationen 1975). To them, an expansion of the primary school curriculum represented an opportunity to strengthen some long-standing syndical claims, including reducing class-sizes and raising salaries. Some Swiss teachers also favoured the idea of early foreign language teaching in itself, and several actively participated in the international reform movement (Giudici 2019). These teacher-activists initiated the first trials, which took place on a voluntary basis. Their evaluations were overwhelmingly positive (e.g. Wymann 1954). According to the teachers involved, pupils were highly motivated and their achievements met all expectations. Thus, they called for the expansion of the reform.

As soon as these calls were picked up by politicians, however, many teachers came out against the reform. In places where the authorities were themselves unconvinced of the benefits of early foreign language teaching, the mere announcement of teachers’ opposition sufficed to halt the reform for decades. In rural, German-speaking Uri, the authorities started preparing the reform in 1975, pressured by an interregional treaty. In response, the local teachers’ organisation conducted a survey among Uri’s teachers, finding only 15% to be in favour of the reform. The organisation subsequently announced teachers would oppose any policy change on the ground, confidently declaring that ‘without the approval of this partner there will be no reform’ (KLVU 1975, 2). This firm stance probably only reinforced a position the authorities already held; the issue was not brought up again until the late 1990s.

A different configuration characterised Basel-Stadt. Here, experts and politicians formed a strong coalition aimed at placing the urban canton on the Franco-Swiss border at the forefront of what the government called ‘the revolution of foreign language teaching’ (Regierungsrat Basel-Stadt 1967, 3). To accelerate the reform, in 1969 Basel-Stadt’s ministry tasked an expert commission with the planning of early French-teaching trials. The commission’s report, however, did not read quite as the ministry had anticipated. The main problem was that the authorities, in an effort to coordinate the reform with surrounding cantons, had agreed to introduce French lessons from the fourth year of schooling while experts considered an earlier start to be ‘closer to the optimal age … according to the findings of learning psychology’ (Kommission Versuche mit Französischunterricht 1969, 5).

The disagreement provided teachers with a window of opportunity. In fact, just a couple of weeks after the publication of the report, the headmasters and teachers of Basel-Stadt’s primary schools announced their opposition to the ministry. Their letter calls the reform a misguided attempt to inflate the primary education curriculum ‘from outside’ (Rektoren Primarschule Basel-Stadt 1969), without considering pupils’ real needs, which teachers knew best. The education board and ministry agreed that, under these conditions, the reform had to be suspended.

Two years later, the ministry made a new attempt. Ignoring the experts, administrators planned a staggered introduction of early French-teaching, starting trials with fourth graders in 1972. Eight teachers answered their call to participate in these trials under the supervision of a scientific monitoring committee. According to the committee’s subsequent assessment report, the teachers enjoyed the experience, especially since academically weaker pupils seemed to thrive in French lessons. They thus favoured the institutionalisation of the reform (Begleitkommission Französischunterricht 1972). This, however, did not happen. In fact, despite repeated and insistent requests, in the following years the ministry was unable to recruit enough teachers to realise the training and trials it had planned. In an enquiry conducted to investigate the reasons behind their obstruction, many teachers cited feeling already overburdened with the current curriculum, and being unsure about the benefits of early French lessons (Versuchsleiterin Französisch 1973).

A last attempt to push the reform through failed in 1974. In that year, the education minister addressed a letter to the local teacher-trainers, asking what they thought might happen if the government simply declared the reform official policy. In their response, the teacher-trainers offered their support for this plan, but also warned the minister that, if the government were to make an official decision on the matter, teachers would probably make their opposition public. They might even try to force a ballot initiative which, given their public credibility as experts on children’s educational needs, was likely to succeed (Lehrerseminar Basel-Stadt 1974). The authorities shied away from such a public confrontation and subsequently halted the reform. Indeed, foreign languages were not taught in Basel-Stadt’s primary schools until the 2000s. Similar forms of open resistance, often combined with displays of public opposition, also allowed teachers in other cantons to override the authorities and hinder the reform effort for decades. In Schaffhausen, for instance, teachers not only declined to support the authorities’ reform plans, but also organised press conferences with titles such as ‘French? In primary school jamais!’ (Stadtschulrat Schaffhausen 1986).

Discussion and conclusion

The literature on education, curriculum, and language education policy tends to depict policy generation as a stage that is separate from, and prior to, policy enactment. When concerns related to educational practice enter the policy generation process, it is supposedly through teachers’ organised interests. However, this study finds three mechanisms that link policy generation to teachers’ micro-politics in and around classrooms. First, by voicing their experiences and concerns, teachers can change the beliefs of those in power. In the analysed cases, teachers’ key role in the enactment of official policy conferred their views a particular kind of legitimacy. Indeed, teachers usually backed up claims with their practical educational expertise, framing their arguments in terms of concrete feasibility and benefits. Second, through the mechanism I call subversive enactment, teachers’ micro-politics, deliberately or inadvertently, can produce outcomes that differ markedly from official policy’s intended effects. As a result, policy-makers find themselves forced to adapt legislation. Third, teachers’ specific roles can also allow them to openly resist policy-making, at least since evidence-based assessments of a reform’s feasibility and effectiveness have become a prerogative for political decision-making.

A growing multifaceted body of literature in (language) education policy shows how micro-political constraints and teachers’ agency produce, transform, and renew policy on the ground. Unlike such in-depth ethnographies, this historical analysis is not methodologically equipped to delve into the classrooms and untangle the complex motives shaping teachers’ enactment. What this study does show, however, is that enactment can have an impact on policy generation. It also reveals how teachers’ actions can aggregate to produce an impact.

Indeed, the mechanisms identified here, while affecting policy generation, are partly independent from the factors typically considered by studies of macro-politics. While this study does not deny that organised interests or specific governance institutions are crucial regarding teachers’ ability to make their interests and ideas heard, it shows that these actors have additional channels of influence. These channels do not rely on teachers’ efficacy as complex actors, but rather on their role as educators in the classroom; on the micro-political, enactment-related actions this role enables them to perform; and on the legitimacy they can draw upon to justify their stances. It is this role that allowed them to translate policy in unintended ways, obstruct policy implementation, or legitimate their stances to those in power. Combining these findings with the recent literature on resistance and teachers’ grassroots mobilisation in the U.S., certain conditions emerge that need to be met for these mechanisms to unfold.

First, none of the mechanisms described here could have been triggered by one teacher alone. While not requiring coordination, each analysed process implies a certain degree of actions resulting noticeable patterns of behaviour. As mentioned, there have been situations in which teachers stopped teaching a foreign language (e.g. Latin in Basel) but no one in power noticed. While this surely changed schooling in practice, it did not have an impact on official regulations. This observation resonates with Lingard and Hursh's (2019) suggestion that the appearance of there being large numbers of teachers and parents opposing high-stakes testing in the U.S., e.g. by them keeping a high profile in public, contributed to the amendment of such reforms in some states.

The need to be noticed also indicates sympathy or at least lack of opposition by certain intermediaries – school inspectors, administrators, the media – as a second condition for these mechanisms to unfold. However, the nature of this sympathy differs, as evidenced by the different relationship between the three mechanisms. It is in reforms characterised by voicing experience and open resistance that, in addition to writing letters and petitions, teachers also relied on intermediaries such as school inspectors or the media to voice their concerns or announce opposition. Indeed, in most cases where teachers openly resisted reforms, they also voiced their experiences to those in power. This not the case for the processes characterised by the subversive enactment mechanism. Here, the silent institutionalisation of teachers’ unorthodox enactment strategies hinged on administrators or inspectors not opposing them at the outset, but was not accompanied by an effort to publicise these strategies or connect them to specific claims. Hollander and Einwohners' (2004) discussion of types of resistances provides a helpful tool to elaborate on this difference. The authors distinguish between visible and generally recognised ‘overt’ resistance and ‘covert’ resistance, whose efficacy can also rely on it not being recognised as such by those in power. Indeed, the fact that Basel’s members of Parliament perceived the creation of French-less classes as a consequence of the impossibility of teaching this language to less gifted students rather than to teachers resisting a regulation many of them disliked, might be a requirement for the subversive enactment mechanism to unfold.

Third, all the cases included in the analysis share the lack of strong differences of opinion within the teaching professions. This manifest unity appears to be another scope condition, since it helps teachers’ views and actions to be perceived as the result of professional experience and knowledge, rather than political views or vested interests. Indeed, evidence from Swiss language education reforms not included in this study suggests that, when teachers were divided, and regarded as either partisan or protecting their interests, their views were more easily dismissed (Giudici 2019). Referring to more recent reforms, Lingard and Hursh (2019) also argue that the perception of teachers as fighting for vested interests in debates on high-stakes testing undermined their public credibility, making parents more effective opponents in these processes. Lynch (2015) draws a similar conclusion from her own activism in these reforms when arguing that ‘[t]he most important requirement is to be informed, to position ourselves as knowledgeable, reflective practitioners’.

The specific design and setting of the study leave some questions unanswered, suggesting avenues for future research. Indeed, by focussing only on cases in which teachers’ micro-politics were the most likely explanation for policy change, the analysis does not provide evidence regarding related mechanisms’ incidence or frequency. The fact that micro-politics were influential on an ideologically- and economically-laden issue such as Swiss language curricula, and that they allowed teachers with rather low political status and connections to have an impact, suggests that these mechanisms are portable across contexts. Still, further research is needed to assess how likely they are to unfold in other policy areas, polity structures, and time periods.

Regarding policy areas, reforms in the selection of taught languages are a prescriptive and incisive type of education policy. They imply new time-tables and schooling materials, and it is relatively easy for outsiders to recognise whether children have been taught an additional language or not. Whether the same mechanisms could also work for more subtle policy change where the mismatch between legislation and practice is less noticeable is open to question. As mentioned earlier, the findings also suggest the necessity of teachers to be perceived as professionals rather than defenders of vested interests for these mechanisms to unfold. Therefore, these mechanisms might also be more likely to emerge in debates concerning educational methods and contents rather than teachers’ salaries or working conditions. In addition, because of the specifics of the topic considered here, teachers’ preferences were always directed either at preserving or restoring the status quo. This invites the question as to whether mechanisms that link innovation-oriented micro-politics to policy generation are the same as those that allow teachers to resist reforms.

By relying on the methodological advantages offered by a federalist state, this study cannot assess how federalism might influence the emergence of these mechanisms. An argument often brought in favour of federalism is that, by bringing politics closer to the people it offers more access points to government, and incentivises active citizenship (Weinstock 2001). Lingard and Hursh (2019, 242) make a similar argument when linking the emergence of grassroots resistance to high-stakes testing to the ‘localism’ characterising U.S. democracy and schooling. One could imagine that the perception of authorities as distant might discourage teachers from engaging with them, as necessary for the overt obstruction and voicing experience mechanisms to unfold. However, empirical evidence, for instance based on comparisons between more centralised and federalised polities, has yet to be produced – both with regard to schooling as well as politics more generally.

Further investigations are also needed to assess whether the restructuring of education systems and governance in the last few decades has affected teachers’ political leverage. Indeed, new policy technologies such as market-like governance or managerialism have been shown to alter teachers’ understanding of their professional role, as well as the conditions in which they operate (Holloway and Brass 2018; Ball, Maguire, and Braun 2012). In turn, this might affect the ways by which mechanisms like those described here operate, eliminating some channels of influence or opening new ones.

While these questions need more empirical scrutiny, the findings presented here have implications beyond the study of teacher politics. On the one hand, they call into question the analytical primacy of macro-politics. If there are mechanisms translating teachers’ actions in the classroom into official policy, then the classroom, as a site where teachers, students, and other individuals deal with different contextual and policy requirements ‘in agentive ways’ (Johnson 2018, 462), becomes a potential factor of influence that has to be built into our theories of macro-political change. The study thus reinforces the appeals of those scholars who have argued that research on policy generation should take the multi-directional nature of the policy-process more seriously, and incorporate knowledge of micro-political dynamics into its methodological and theoretical considerations (Ball 2006; Ricento and Hornberger 1996; Schulte 2018; Sutton and Levinson 2001). Indeed, since these mechanisms do not take place in institutionalised arenas, nor rely on actors holding formal power or representation, they are difficult to detect by analysing policy texts and institutional rules. In fact, they might only reveal themselves if we trace the political processes and how power is wielded in practice. Political and historical scholarship can provide a precious complement to ethnographic and critical studies on real-time education politics, allowing the identification of mechanisms that operate in the long term (Gryzmala-Busse 2011) – such as the subversive enactment mechanism here.

On the other hand, the findings valorise individual agency. They suggest that, in theorising education policy, it might be important to consider not only organised interests, but also ways in which the behaviour of individual or minimally coordinated stakeholders can aggregate to produce change (or stability). Not reducing actors such as parents or teachers to their official organisations and representation might also prove important from another perspective. The more complex, professionalised, and internally structured an actor becomes, and the more integrated it is in institutional politics, the more distance grows between the leadership and its base (Scharpf 1997). Indeed, in one of the cases delineated here (section 4.3), the interregional Swiss teachers’ organisation advocated a different position from local teachers assemblies and, apparently, a large share of individual teachers. A similar situation can be observed today, as the major U.S. teacher unions and parts of their base seem to disagree on the issue of high-stakes testing (Lingard and Hursh 2019). The existence of these mechanisms, then, may be good news to teachers and other unorganised groups, especially in times characterised by declining unionisation rates and the apparent failure of participatory models of educational policy-making (Björk and Blase 2010).

Acknowledgements

I would like to thank the anonymous reviewers as well as the participants at the 2019 ESPANET Politics of Education stream in Stockholm for their helpful comments and advice. Research for this article is supported with funding from the European Research Council (ERC) with a Starting Grant for the project ‘The Transformation of Post-War Education: Causes and Effects (SCHOOLPOL)’. Grant Number 759188. Research is also supported by the Swiss National Science Foundation, Grant Number P2ZHP1_184086.

Disclosure statement

No potential conflict of interest was reported by the author.

Supplementary material

Supplemental data for this article can be accessed here

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This work was supported by the European Research Council [759188];and the Swiss National Science Foundation [P2ZHP1_184086]

Notes on contributors

Anja Giudici

Anja Giudici is a postdoctoral research fellow at the Department of Politics & International Relations at the University of Oxford. She holds a PhD in Education and is currently working on the ERC-funded project SCHOOLPOL, The Transformation of Post War Education: Causes & Effects. Her research focuses on how political ideologies, nationalism, and gender impact on the politics of education and language.

Notes

1. A fourth language, Romansch, was officially recognised in 1938.

2. Aargau, Basel-Stadt, Bern, Fribourg, Geneva, Graubünden, Lucerne, Schaffhausen, Schwyz, Ticino, Valais, Vaud, Zurich.

3. The project involved five Swiss universities and was sponsored by the Swiss National Science Foundation (2012–2017, nr. CRSII1-160810).

4. The correspondence is stored at the Archivio di Stato del Canton Ticino, Fondo ottocentesco (DPE), fascicolo VI.

5. See the minutes of parliamentary debates: Verbali Gran Consiglio, sessione ordinaria primaverile, 1846, 597–600, online: https://www.sbt.ti.ch/bcbweb/vgc/ricerca.

6. The minutes of the parliamentary debate are reported in the local newspaper Basler Nachrichten, 4 May 1880.