Mind the gap: A tale of two curriculum fallacies

Throughout my professional life, I have been interested in the relationship between teachers and curriculum. As someone who has taught languages, educated teachers, and developed curriculum and materials, I have been puzzled by the separation of curriculum and teaching. In the US, this separation is encapsulated in the phrase ‘curriculum and instruction’, where they are seen as separate domains of research and responsibility (Doyle, 1992; Kaplan & Owings, 2015). Indeed, as a teacher educator, I would often hear the refrain from teachers, ‘I know how to plan a good lesson, but I'm not sure what the big picture is. How do the lessons fit together as a whole?’ I interpreted this to mean that they did not have a sense of the overall curricular structure and aims for their students’ learning. As a materials and curriculum developer, I saw my responsibility as providing a map for teachers that would show how the parts added up to a sensible whole.


Introduction
Throughout my professional life, I have been interested in the relationship between teachers and curriculum. As someone who has taught languages, educated teachers, and developed curriculum and materials, I have been puzzled by the separation of curriculum and teaching. In the US, this separation is encapsulated in the phrase 'curriculum and instruction', where they are seen as separate domains of research and responsibility (Doyle, 1992;Kaplan & Owings, 2015). Indeed, as a teacher educator, I would often hear the refrain from teachers, 'I know how to plan a good lesson, but I'm not sure what the big picture is. How do the lessons fit together as a whole?' I interpreted this to mean that they did not have a sense of the overall curricular structure and aims for their students' learning. As a materials and curriculum developer, I saw my responsibility as providing a map for teachers that would show how the parts added up to a sensible whole.
Determining a sensible whole for a language curriculum is not a straightforward undertaking. When I started out in materials development in the 1980s, I became fascinated by the possibilities of what could or should be included in a language course. I saw that authors made decisions that got fixed on a page, but at any point the decisions could have resulted in something different, depending on one's reasoning. For me, this meant that when textbooks were used in the classroom, they should be regarded as malleable not fixed. The key was to have good reasons for changing, moving, or dropping a topic or activity. The person who made those decisions was the teacher. The curriculum provided the map; the navigator was the teacher. Thus, there needed to be a balance between a good curriculum and teacher know-how and agency in using it. As a teacher educator, my responsibility was to support teachers in understanding how a curriculum is put together, how to evaluate one, and how to make decisions about how to adapt or design one to meet the needs of their students.
This belief in teacher decision-making was tested when I went on tour to promote my first textbook series, East West (Graves & Rein, 1988, 1989. The series was very much of its time in the hybrid structural-CLT (communicative language teaching) mold. It was built around a grammatical core, with pair work activities designed to foster oral communication. On the tour, I met a teacher who told me how much she liked the books, but that she always skipped the pair work. I remember thinking, 'But she doesn't get the point! Pair work is where they learn to use the language. How can she NOT do the pair work?' In other words, if her students did not learn, the problem was not the textbook, it was the teacher. The textbook was good, the teacher just was not using it correctly.
In retrospect, of course, I should have asked the teacher why she had decided to skip the pair work. What was her reasoning? What contextual factors came into play in her decision? There could have been any number of reasons, such as lack of time, the challenges of managing a large class, student reluctance to speak, or her own beliefs about how languages are learned. Indeed, research on how teachers understand and adopt a communicative approach shows that one or all of these factors may affect whether teachers choose to have students do pair work in the target language (Humphries, 2014;Sefarej, 2014;Le, 2018;Tram, 2021). Instead of probing the teachers' reasoning, I left the encounter feeling I should have convinced her to do the pair work.
At the root of my desire to convince her to change her practice was the assumption that the teacher needed to change, not the textbook. I have come to see, over the years, that this assumption permeates the field of education and can have dire consequences for educational reform and for teachers. The assumption is based on two intertwined fallacies about curriculum and its relationship to teachers. The first is that a good curriculum is the foundation for effective teaching and learning. The second is that when teachers faithfully follow a curriculum as designed, they will produce consistent learning results. I call them fallacies because the arguments are deceptive or misleading, but widely believed to be true. 1 These particular fallacies are too often the bedrock of curriculum planning and innovation, at both large and small scale, and too often taken to be true. The resulting focus on the curriculum and not on the teachers who use it, can lead to a 'lose-lose' situation. Curriculum aims are not achieved, and teachers feel undermined and unsupported. Yet these fallacies persist. Why? In this paper, I will discuss the fallacies and how they contribute to the gap between curriculum aims and classroom achievement. I will then propose a different way of looking at this relationship that can help to close the gap.
2. Versions of curriculum I want to first define what I mean by curriculum. Curriculum is concerned with WHAT students learn, broadly defined as 'the substance or content of schooling' (Doyle, 1992, p. 487). This broad definition raises questions about who decides what students learn, on what basis these decisions are made, and what the relationship is between what students should learn and what they actually learn. When one starts to explore these questions, it becomes apparent that there is not 'a curriculum', but different versions of curriculum with different people involved, different purposes and different discourses. Doyle (1992) and Deng (2017Deng ( , 2018 define these different versions in ways that I find helpful for understanding the fallacies. Doyle (1992, p. 487) notes that 'curriculum discourse operates at both institutional and experiential levels. There is a formal curriculum that defines the core substance of schooling and an experienced curriculum that is taught and learned in classrooms.' The two discourses are not equal. The institutional discourse of policy makers and curriculum developers is privileged over the discourse of teachers and learners. The gap between these two discourses and how they interact is at the heart of the fallacies. Deng (2017Deng ( , 2018 proposes that curriculum making occurs in three layers of contexts, each with its own version of curriculum. The first, the societal curriculum, comprises beliefs and expectations about the purpose of schooling that circulate in the society at large. The second, the institutional curriculum, comprises both policy and programmatic curricula. The policy curriculum translates the societal curriculum into policy guidelines and directives. The programmatic curriculum uses these guidelines to shape the programs, courses of study, and materials to be used in schools. The third is the classroom curriculum, also known as the enacted curriculum, in which 'The externally created curricular materials … are seen as tools for students and teacher to use as they construct the enacted experience of the classroom' (Snyder, Bolin, & Zumwalt, 1992, p. 418).
These versions of curriculum are often depicted as hierarchical and sequential. Societal expectations are translated into policy; policy is translated into programs and materials; programs are translated into classroom learning experiences (Deng, 2017). However, it is more accurate to say that these processes co-exist in relation to each other (Brown, 1995). Policies may change frequently, but it is not easy to change programs and materials. Thus, programs endure as policies change. When programs ARE being revised, classroom instruction goes on. When a new curriculum is introduced, it is introduced into an existing classroom culture (Cohen, 1990).
In general, among the three, when we refer to 'curriculum', we mean the institutional curriculum. The institutional and enacted curriculum are distinct from each other in terms of the people who are responsible for them, the processes of creating them, and the forms they take. This gap means that those who develop the curriculum are generally separate from those who use it. Developers are responsible for developing a sound curriculum and materials. Teachers (and learners) are responsible for realizing the aims of the curriculum by transforming it into educative experiences in the classroom (Deng, 2017).
It seems reasonable, then, to posit that a good curriculum is the foundation for effective teaching and learning. It also makes sense that when teachers follow a curriculum as it is designed and intended by the developers, the learning that results will consistently achieve the aims of the curriculum. However, upon closer examination we will find that these fallacies sustain the gap between the institutional and enacted curriculum.
3. Two fallacies about the relationship between curriculum and teachers 3.1 Fallacy #1: A good curriculum is the foundation for effective teaching and learning Here, I define a 'good' curriculum not by its content, because that varies according to subject and to educational trends, but by how it is conceived and 'packaged'. From the point of view of curriculum developers ' … a curriculum is a systematic program, often benefiting from substantial expertise and resources, to achieve valuable educational purposes' (Doyle & Rosemartin, 2012, pp. 137-138). In this paper, a 'good' curriculum is determined by its systematic coherence and its perceived value. A 'good' language curriculum is conceptually systematic when it is based on a coherent interpretation of what learning a language involves, that is, theories of language and how it is learned (Nation & Macalister, 2010;Graves & Garton, 2017). Language curriculum developers use these principles to develop theoretically sound curricular guidelines that describe 'the valuable educational purposes', that is, curricular aims.
A language curriculum is structurally systematic when it uses the guidelines to develop syllabi and materials that provide a clear pathway for learning so that the aims can be achieved. Materials developers use these syllabi to develop well-conceived materials, such as textbooks. Assessment instruments are developed that document the extent to which the aims have been achieved. These layers of curriculum products range across a spectrum from abstract goals to usable instruments, which should all share the same theoretical DNA, so that one should be able to see, for example, which principle a textbook activity is designed to operationalize.
According to the fallacy, a curriculum is the foundation for effective teaching and learning because of its theoretical coherence, structural cohesiveness and clear outcomes. It provides needed resources for teachers and learners. It provides a rationale and a syllabus for what is to be learned and the materials necessary for learning. It frees teachers to devote their energy and time to implementing it in the classroom with their students, rather than to developing their own. Without a curriculum, there would be no clarity as to goals, how to achieve them, or assessing the extent to which they had been achieved. This fallacy is the cornerstone for the second fallacy. That is, if the curriculum is based on a coherent understanding of what needs to be learned, is organized so as to provide a clear pathway for learning, and provides appropriate, well-conceived materials, then it follows that if teachers use it in the way it was intended, the results will be good.
3.2 Fallacy #2: When teachers faithfully follow a curriculum as designed, they will produce consistent learning results This fallacy applies at different levels of scale. When the curriculum is uniform for a nation or a program, consistency of results should be assured across levels and contexts. If everyone follows the same syllabus and uses the same materials, all students should learn the same thing and educational aims should be achieved. On a national level, 'Such uniformity, or "homogeneity" … is reasonable to imagine from a ministry of education, which society charges with conceiving and diffusing a coherent, equitable, and consistent education program for all a nation's people' (Muskin, 2015, p. 94). From a policy standpoint, a goal of the societal curriculum is homogeneity, which creates coherence and continuity. If different curricula were developed for different schools, school districts, or regions, there would be a danger of variation in rigor or amount of content, thus potentially leading to inequitable learning results. A standard curriculum provides teachers with needed guidance about what to teach and how to teach it. An underlying assumption is that curriculum and materials can be used in the same way by teachers and students in each context and produce similar results.
The curriculum is thus seen as an instrument to be used in specific ways. Teaching is viewed as a technical process that involves operationalizing the curriculum by doing exactly what it says to do. Teachers (and learners) are implementers. Their role is to faithfully follow the curriculum. If they do so, they will achieve the desired results. This is the normative or 'fidelity' view of implementation (Snyder et al., 1992). This view assumes that with excellent materials, teachers who have not previously used them can easily change their teaching by simply following what they prescribe.
To summarize the reasoning thus far: A good curriculum is aspirational. It shows the way forward by providing theoretical coherence and a clear learning sequence. It defines outcomes, which can be tested. It provides teachers and learners with needed resources. In other words, the curriculum is so good and well-conceived that when teachers and learners use the resources as intended, effective learning will result and the aims will be achieved.
This reasoning seems very sensible. So why are the fallacies deceptive? The problem is that they are based on a wrong understanding of the relationship between the institutional and the enacted curriculum.

Why the fallacies are misleading
The fallacies are misleading because they assume that the institutional curriculum is the most important determinant of what happens in the classroomthat its use leads to effective learning. However, curriculum frameworks and materials, OF THEMSELVES, do not cause or ensure effective learning. They may be theoretically coherent, provide clear pathways, define learning outcomes and so on, but they are inert and meaningless formulations until they are used in the classroom (Markee, 1997;Graves, 2008). How materials are used in the classroom depends on the users. Curriculum plans and materials are only 'good' if teachers and learners are able to use them in ways that make sense to them and contribute to their learning. In other words, 'goodness' emerges through enactment. Conversely, if teachers and learners are not able to use curriculum plans and materials in ways that make sense to them, can we say that the curriculum is 'good'?
Studies of how teachers and learners use new textbooks can help to shed light on this question. No matter how well-conceived the textbook, it will not have the intended impact in the classroom if it is not workable for teachers and learners. The reasons that textbooks do not 'work' for their users depend on a range of contextual factors. Consider one Argentinian teacher's experience of working with a new curriculum that emphasized integration of the four skills, but where physical factors such as lack of equipment and, occasionally, electricity, interfered with her ability to teach the listening activities (Soto, 2018). Her students' English level and learning preferences made it difficult for them to understand the reading and listening passages. The teacher explained that they 'don't want to listen to English. They want to listen to ME because I'm clear. … And because I speak slowly for them to understand. So, no listening, no readings because they have to read with ME working as a walking dictionary.' (Soto, 2018, p. 211). Likewise, students' background knowledge created challenges for a teacher in Albania whose new communication-oriented textbook included activities on topics that are not part of the children's life experience, such as describing an automobile accident observed on a holiday in another country (Sefarej, 2014). Her beliefs about how students learn led her to omit pair work. 'We … keep hearing 'pair work, pair work.' However, any experienced teacher knows that this idea, along with other initiatives that successive governments have tried to foist on secondary schools, simply does not work.' Students expect 'to be instructed by someone who has the skills and ability to teach, not by a pair who lack an understanding of English' (Sefarej, 2014, p. 99). Expectations that students should be prepared for high-stakes exams are a common obstacle to following a new communication-oriented curriculum, as voiced by one Chinese teacher, 'We have only one class every day, so we can only focus on what is to be tested, i.e. vocabulary, reading, writing, and grammar in our teaching, practice and assignment. We have no time for oral practice and assignment' (Yan, 2018, p. 73).
In the eyes of their developers, I am sure that the curriculum materials these teachers were using were 'good'; however, physical, social, and educational factors in these contexts caused the teachers to work around them. These factors highlight a second misconception: The institutional curriculum is not the most important determinant of learning; it is only ONE element of the enacted curriculum. It is one part of a system that includes the teacher, the learners, the physical context and the school and community context, mutually influencing and interacting with each other over time. Figure 1 is a schematic of classroom instruction that depicts the different elements at play in the enacted curriculum: learners, teacher, subject matter or content to be learned, and the contexts or environments in which learning takes place.
I would argue that the teacher's role is to ensure the connection of the learners to the content through their teaching so that they learn. (This does not mean that each learner makes the connection or that the connection is made because of the teacher.) Orchestrating effective learning depends on how learning is defined, which depends on how content is defined. The content in Figure 1 thus represents both what the teacher is expected to teach (the institutional curriculum) and what is actually taught in the classroom (the enacted curriculum).
A study by Menkabu and Harwood (2014) of seven teachers using the same textbook in a medical English for specific purposes (ESP) course illustrates the interplay of the elements in Figure 1. It shows how each teacher omitted or changed textbook activities in some way, for different reasons, including one teacher's concern that her lack of background knowledge about a topic would undermine her authority; the appropriateness of certain topics for the cultural context; or students' concerns about emphasizing skills that were not tested on the university exam. The institutional curriculum, represented by the textbook, is one of a constellation of factors that are at play in enactment. Each of these factorsthe teacher, the learners, the context, and the textbook itselfmutually influence how the textbook is used, and what actually gets done in the classroom. As a result, the curriculum enacted and experienced by the teacher and learners was different in each classroom.
In terms of a national curriculum, differences across contexts in which it is used, including school cultures, community expectations, and available resources mean that it will be interpreted differently in different contexts (Muskin, 2015). Even when teachers are well prepared to use a national curriculum, there is simply no way it can be implemented exactly as it is intended. When teachers try to do so, they inevitably fail to meet the needs of all, or even most, of their learners (Tram, 2021). The assumption that one curriculum can fit all contexts is patently false. Studies in language curriculum enactment over the last decades bear this out (e.g. Kirkgöz, 2008;Zhang & Hu, 2010;Humphries, 2014 (Ball & Forzani, 2009) A further misconception underlying the fallacies is that enactment is a technical process that involves operationalizing a curriculum by doing what it says to do (Apple & Jungck, 1990). This misconception ignores the relational nature of teaching and learning where teachers must constantly negotiate the elements depicted in Figure 1. They do not mechanically implement a curriculum, they mediate it for their context in relation to their own beliefs and experience, to their students, to societal expectations and to a myriad of other contextual factors (Muskin, 2015). This point is particularly important for curriculum innovation, which, by definition, is directed at new aims and ways for students to learn. Innovation assumes that the old aims and ways need changing. This means that teachers both have to unlearn old ways while also learning new ways of teaching (Cohen, 1990). This kind of (un)learning takes time and may be perceived as too risky if it asks teachers 'to abandon many of the classroom practices and behaviours they are familiar with for new approaches that are less familiar, and which pose a potential threat to their professional self-perceptions and their professional relationships with others' (Grassick & Wedell, 2018, p. 263). Curriculum change does not occur by telling teachers what to do (Wang, 2015).
The processes of curriculum making also affect the relationship between the institutional and the enacted curriculum. An institutional curriculum is a producta reification, over time, of a group of people's processes of thinking, debating and negotiating ideas, and deciding how to organize and express them (Graves, 2008). As a product, it seems set in stone. But at any point the product could have turned out differently, depending on who was making the decisions. Institutional curricula tend to be overly ambitious and impossible to cover within the time allotted because they address broad aspirations and because each contributor has an agenda about what should be included (Wiggins & McTighe, 2005). Teachers therefore have to make choices about what to teach that make sense for their context.
To summarize why the fallacies are misleading: A curriculum, of itself, does not cause learning. It does not matter how 'good' it is if teachers and learners cannot make sense of it or are unable to use it. A curriculum is not the most important determinant of what happens in classrooms; it is one of several elements at play in enactment. How a curriculum is used depends on the interplay of these elements; therefore, the same curriculum will be used differently in different contexts. Teachers do not mechanically implement a curriculum, they mediate it for their context. All of this means that it is IMPOSSIBLE to follow a curriculum with fidelity or to implement it consistently. This is true even when teachers design their own courses or their own syllabus or choose their own materials. The actual teaching of a course always involves changing the design in some way. The design is a reasoned conceptualization, while enactment unfolds in response to the layered realities of the context. Accepting these fallacies can have serious impacts on how teachers are positioned and supported, particularly when a new curriculum is introduced. Innovation is seen as depending on 'experts' developing a new curriculum. Teachers are given the new curriculum, usually in the form of textbooks, sometimes in the form of curriculum frameworks . They are expected to change their current practice and adopt a new practice by being 'briefed' about its purpose and rationale (Wang, 2015). They are expected to figure out how to do this in the classroom by using the materials as designed. If they are lucky, they may have some one-off training opportunities in how to use the materials (Tram, 2021). Teachers are thus positioned as receivers and consumers of new ideas and practices. Once teachers receive the new curriculum, administrators and policy makers are 'off the hook.' Since they have ensured quality in the institutional curriculum, they have done all they can do, so if the enacted curriculum does not produce the desired results, teachers are at fault (Doyle & Rosemartin, 2012). As a result, if there is a gap between the aims of the curriculum and its workability in the context, it should come as no surprise that the innovation does not take root. One only has to look at the uneven success of introducing communicative curricula in a variety of contexts to see the effects of the fallacies on language education (Graves & Garton, 2017).

Turning the fallacies on their heads
How can we close the gap, bringing closer alignment between the institutional and enacted curriculum? Such alignment can only happen by turning the fallacies on their heads, namely by focusing on the enacted curriculum and acknowledging that the success of curriculum innovation is determined not by how 'good' the institutional curriculum is, but by how well teachers are able to use it in their context. Therefore, the aim of curriculum innovation should be for TEACHERS to be successful in using the curriculum. Closer alignment thus comes down to three main points: the curriculum needs to be workable in the context, it needs to make sense to teachers, and teachers need to be able use it confidently with their students in their context. These points have implications for how innovation is introduced, for teacher education and for curriculum research.

Curriculum innovation
What would curriculum innovation look like if it focused on teachers' making sense of the curriculum and being able to use it confidently? Efforts to develop Content and language integrated learning (CLIL) in Europe and Latin America offer good examples of this approach at local and regional levels. These efforts have recognized the need for teacher involvement and training for CLIL to be successful (e.g. Lasagabaster & Ruiz de Zarobe, 2010; Banegas, 2019). However, I could find few examples of large-scale language curriculum reform that consistently involved teachers. One that stood out was a reform that introduced English as a foreign language (EFL) in primary grades 1 and 2 in Greece (Dendrinos, 2013(Dendrinos, , 2015Karavas, 2014;Karavas & Papadopoulou, 2014). It required new thinking and skills from teachers and involved them from the outset of the reform. Change came from how teachers were able to make sense of the curriculum, not from imposing it and expecting them to implement it.
To give some background: In 2010 the Greek Ministry of Education introduced EFL in grades 1 and 2 of primary public all-day schools (Karavas, 2014). The development and implementation of the curriculum was overseen by a team at the Research Centre for Language Teaching, Testing and Assessment at the University of Athens. The team were charged with developing a curricular framework for EFL in grades 1 and 2, as well as syllabi and materials for each grade (Dendrinos, 2013, p. 12).
Unlike most reforms that start by ignoring teachers or making assumptions about what they can or cannot do, this reform started with 'a national survey documenting the needs and characteristics of teachers, learners, and school units' (Karavas, 2014, p. 245). The survey found that most teachers in the project, although specialist language teachers, were not familiar with methodology for teaching English to young learners.
The team prioritized 'efficient communication networks and stakeholder ownership' (Karavas, 2014, p. 246) including teachers, parents and administrators. This approach stands in contrast to the top-down approach to communicating its aims so that teachers and parents would be presented with a fait accompli. Notably, information about the reform was disseminated through a project website that was set up at the outset. Teachers were invited to attend two two-day conferences in two different cities, which were also videotaped and uploaded to the website.
Rather than depending on 'experts' to develop curriculum materials that would then be handed off to teachers, teachers were recruited to join either a materials development team or a materials evaluation team. The teams also included 'EYL (English for young learners) specialists, applied linguists and young researchers' (Dendrinos, 2015, p. 32). The materials took the form of cycles of tasks and resources appropriate for very young learners that could be used in flexible ways (PEAP project website). In an effort to promote local authority and ownership, all curriculum information and guidelines for the teachers were in Greek. The materials were tried out and evaluated by teachers in their classrooms over a two-year period, after which the grade-level syllabi were developed. The syllabi thus reflected 'the implemented curriculum rather than the planned one' (Karavas, 2014, p. 248). The curriculum was workable for the context because it had been shaped by use in the context.
Teachers were supported in understanding the theoretical underpinnings of the reform and how they would be reflected in classroom practice through conferences, face-to-face and online courses. They were given time and support to learn the new practices. The reform did not expect teachers to change or adopt a new practice by being 'briefed' about the reform and given new materials they were expected to understand and follow. In particular, teaching young learners required a 'learning by doing' pedagogy that was different from what the teachers were used to (Dendrinos, 2015). Teacher training was paramount. The findings from the initial survey were the basis for developing an approach to teacher training that was 'capable of (a) addressing a very diverse group of teachers with a wealth of different training needs, (b) covering the needs of the less and more experienced teachers, and (c) effecting change in teacher beliefs without being overly didactic and theoretical, but grounded in examples of actual practice' (Karavas & Papadopoulou, 2014, p. 183).
Teachers were viewed as producers of curriculum knowledge based on the ways they enacted the curriculum. Their expertise was recognized and promoted, unlike reforms that treat teachers as receivers and consumers of new ideas and practices. A 'Teachers' Corner' was set up on the project website, which teachers used to share and exchange ideas about classroom experiences and activities, upload and access resources, and discuss problems and give advice (Karavas, 2014, p. 248).
On its face, the reform appeared to operate based on the two fallacies: The aim was to develop a 'good' curriculum to be the foundation for effective teaching and learning of English. In this case though, teachers felt ownership of the curriculum precisely because they understood it and were involved in its development in a substantive and ongoing way. This involvement meant that their use of the materials would likely result in student learning consistent with the aims of the curriculum. The reason this curriculum was 'good' and could be used effectively was in large part due to how teachers learned about the principles underlying it; the flexible nature of the materials, which gave them choices; how they designed, taught, and evaluated them; and opportunities to share this knowledge through the website.
There was thus a reciprocal relationship between institutional curriculum-making and curriculum enactment. The process of developing the institutional curriculum was shaped by the enacted curriculum, that is, teachers using and adapting the materials to determine what worked for their learners. The framework, syllabi and materials grew out of this relationship. As participants in both development and enactment, teachers were able to make use of these products in ways consistent with their aims. Thus, the aims of the institutional curriculum and how they were realized in the enacted curriculum were closely aligned.

Educating teachers to be curriculum thinkers
We know that teachers who are involved in developing a curriculum are better able to align what they do in the classroom with its aims, as the Greek example illustrates. I argue that this is because they have learned to be 'curriculum thinkers.' They understand how a curriculum is shaped and are able to 'act locally' on their understanding (Wang, 2015); they know how to use and adapt it and to create their own variations. Put another way, they learn to use knowledge about curriculum-making in the institutional context to inform curriculum enactment in the classroom context. Although involvement in curriculum development is not realistic or feasible for most teachers, they can still learn to analyze, evaluate, adapt and even create a curriculum and materials. These skills address two of the points I made earlier: being able to make sense of a curriculum and to use it confidently with their students in their context.
One way of making sense is joining global theories of curriculum with local practice. Wang (2015) underscores the importance of making these links when she describes a collaborative effort among practicing teachers, curriculum developers and teacher educators in China. Teachers mapped the new curriculum to their textbooks, which enabled them to identify 'how curriculum ideas were reflected in the textbook and could be translated into classroom practice' (p. 141). They designed lessons, taught them in their classrooms, and revised them based on post-lesson discussions and feedback from the team members who had observed the lessons. The teachers later taught them as 'public lessons' that were observed by other teachers. Wang (2015) comments that the initial trial and feedback experience required them to engage in a continual process in which they had to 'make sense of themselves, their beliefs about how learners learn, and their choices of methods and activities' (p. 143). As discussed earlier, for curriculum innovation to take hold, teachers have to unlearn old ways while also learning new ways, but this can only happen through practice and reflection. One teacher in Wang's study acknowledged this process: 'The changes in my teaching were significant. By repeated instructional design of a grammar lesson, I became increasingly aware of my misunderstandings about teaching. During this process, I refuted myself many times after which I moved to a better understanding of myself ' (p. 143).
Pre-service teachers can likewise learn to make sense of a curriculum conceptually and practically. Augusto-Navarro, de Oliveira, and de Abreu-e-Lima (2014) describe a teacher preparation course in Brazil in which pre-service teachers analyze the theories of language and learning underlying their textbooks and then prepare, peer teach and evaluate lessons based on those books. This analysis pushes them to look beneath the surface to examine the views and biases represented in the activities. They see that textbooks with similar aims may be organized in different ways, and that there may be mismatches between the underlying theories and the activities (Graves, 2000). Then rehearsing lessons with peers can provide valuable insights about what students need to be able to learn from and with materials in a classroom context (Lee, 2015). To round out the process, teachers learn through reflecting on and discussing these experiences to evaluate the viability of their lessons and the extent to which they reflect curriculum principles.
Adaptation is a natural and needed part of the process of classroom teaching (Bosompen, 2014). As I have argued, teachers are always mediating a curriculum in their context. They are adapting it according to who their students are, the resources they have, and the culture of their classrooms. That said, most teachers are not taught how to adapt materials (Tomlinson & Masuhara, 2018;Graves & Garton, 2019). It need not be so. New teachers can learn, for example, how to adapt activities or sequences in the same textbook for different kinds of learners or for different contextual constraints (Graves, 2000). Making and justifying these kinds of decisions helps them to see curriculum as a tool they use to support their learners.
Curriculum adaptation is part of the process of developing agency and autonomy. Teachers would gain greater confidence in their ability to adapt curriculum if they learned to plan lessons based on the underlying frameworks. Working with frameworks and syllabi gives them greater autonomy because they are not constrained by textbook content and sequences (Vilches, 2018). Hult (2018) writes about how pre-service teachers engaged with national syllabus documents through scenarios that required them to interpret these documents in light of classroom practice. They came to see their pros and cons: 'how vague the curriculum and syllabus on which I am to base my future work are', but at the same time 'that we can steer our future classrooms and students in many different ways and still be justified by the rules that are set up for how we should be teaching' (p. 254).
There is a certain irony here. We know that most practicing teachers never see or become familiar with the curricular frameworks on which their textbooks are based (Soto, 2018;Tetiurka, 2018;Tram, 2021). This is hardly surprising since many teacher educators do not themselves understand the connections between curriculum documents and classroom learning materials. Gulyamova, Irgasheva, and Bolitho (2014) describe how teacher educators themselves must learn to work from curriculum documents to produce relevant learning materials so that they can teach teachers to do so. As they learned to interpret the syllabus and select materials on a principled basis, their own attitudes changed. Where before they felt that if textbooks were prescribed there was no point in evaluating them, they came to see that teaching according to the syllabus and not the textbook gave them freedom to work towards course objectives in ways they deemed appropriate.
To summarize, I argue that educating teachers to be curriculum thinkers positions them to have knowledge and thus to exercise agency over the curriculum as a tool. Rather than being subordinate to the curriculum, curriculum thinking positions teachers as mediators which reverses the fallacies. They become partners with curriculum rather than being positioned as unquestioning implementers.

Curriculum and materials research
These changes of attitude start as much with research as they do with policy. Research on language curriculum needs to rethink how teachers are positioned with respect to curriculum. Too often, such research focuses on how closely teachers adhere to the curriculum or how they adapt it, rather than on why they make such adaptations (Tram, 2021). Similarly, in language materials research, the focus has largely been on how materials are developed, or on evaluation and critiques of their content, rather than on how teachers use materials in the classroom and their reasons for doing so (Tomlinson & Masuhara, 2018;Graves, 2019). The net result is that teachers are again positioned as unquestioning users, unable to exploit materials in critical ways, and unaware of their impact (Santos, 2013;Bori, 2018). As Doyle and Rosemartin (2012, p. 138) observe of such research in general education, 'The direction of much of this work is from the curriculum to the teacher.' There are notable challenges to this directional assumption. Promising research in language education explores teachers' experience of curriculum or their centrality in educational change (Tribble, 2012;Le et al., 2020;Tram, 2021), but more of this type of research is needed to challenge the fallacies. Otherwise, teachers will continue to be cast as 'reluctant' or 'resistant' to change. Resources will be put into developing 'better' curricula rather than into the kinds of systematic support teachers need to understand and mediate the curricula they are told to teach. If, as I have argued, the aim of these innovations is for teachers to be successful in using curriculum, then research that focuses on teachers and context is imperative. Research needs to focus on enactment not implementation so that it can inform curriculum development as well as teacher preparation and support.

Conclusion
We need to understand how teachers make sense of a curriculum, how teachers can be prepared to be curriculum thinkers, and what enables them to confidently use and adapt curriculum. More than three decades ago, Johnson (1989) described decision making in what he called a 'specialist' versus a 'coherent' approach to language curriculum development. In the specialist approach, different experts make decisions in different domains of policy and curriculum development, with teachers and their students at the bottom of the chain of decision making. The decisions are hierarchical, which result among other things in lack of connection and accountability between domains. In the absence of connection come the two fallacies I have described. Accountability is not operational; it is assumed. In the coherent approach, on the other hand, decisions are interdependent such that the success of one depends on the success of the others. There have been very few instances of this coherent approach since Johnson outlined it. The Greek reform introducing English into grades 1 and 2 that I described is a notable counterexample. Here, the focus was on involving teachers in the innovation, which turned the fallacies on their heads. Decision-making processes of policy, development, and training in the institutional curriculum were interdependent with the enacted curriculum throughout the two years.
This interdependence contributed to its success, but not forever. I wrote one of the Greek authors, whose work I consulted, to find out if the innovation was alive and well. I asked her if the project teachers were still using the online platform and Teachers' Corner to exchange materials and ideas. She told me that in 2016 the Minister who had supported the reform changed. The new Minister decided, without consulting the development team, to cut the weekly hours of English instruction from three to one. She continued: The programme was designed for three hours of instruction a week and so this drastic reduction created many problems especially for teachers who had to find ways to select and creatively synthesise tasks from the various thematic 'cycles'. The platform is still being used by teachers to download tasks and materials. Teachers also access the online teacher training modules especially designed for this programme. They do not, however, contribute to the teachers' corner (they're so busy trying to figure out how to condense the material to fit in the current work timetable). (Karavas, personal communication) Her comments vividly illustrate the need to focus on teachers and enactment. It is not the policy makers who feel the impact of changes like this reduction in instructional hours, it is the teachers. Ironically however, these teachers are in a much better position to cope with the gap between policy and enactment because of the skills they have developed as curriculum thinkers. But they should not have to.
I am not immune to these mistakes. In the early 2000s, I was involved in a curriculum reform project in a country in Northern Africa. We were tasked with interpreting policy guidelines for a new secondary school English language curriculum in order to develop a curriculum framework based on the holy grail: reaching a CEFR B1 level upon school-leaving. Although I met with Ministry officials, I did not once meet a teacher. Nor did I ask to do so. My team and I toiled away at developing the framework. Our work was guided by questions such as how skills in one level could build toward the next, how to develop culturally appropriate themes and how to thread the themes through the grade levels. I subsequently worked with a group of school inspectors whose role was to support teachers in their regions. They told me that both they and the teachers had been completely taken by surprise by the reform and were left scrambling to figure out how to implement it.
If I were to be involved in such a project again, my work would be guided by a very different set of questions. I would start by asking how teachers could participate in the development of the curriculum so that it would benefit from what they know and need. I would ask how the development process could engage teachers in making links between the theories and aims of the curriculum with examples from their practice. I would also ask what planning was in place for ongoing training of teachers that involved them in understanding how the curriculum could work in their classrooms and why. I would further want to know what provisions were in place for teachers to try out the new curriculum and learn from their experience as well as how they could share their knowledge about using the curriculum with each other.
I would like to think that had I been guided by these questions, things might have turned out differently for the teachers and inspectors.