Shulman (2005a, 2005b) describes the surface structure of a signature pedagogy as the recognisable acts of teaching and learning how to perform as a professional. Each of the models of teacher education outlined in Table 1 takes up an implicit position in relation to both the recognisable acts of teaching (that is, what one expects the teacher educator to do) and the actions that are then necessary to learn how to perform as a professional. For example, implicit within stage theory is an assumption that the new teacher will move through a series of stages as they progress in their competence. For the teacher educator, it is then assumed that they will adapt their teaching to reflect the different stage of development and, one would anticipate, elicit different behaviours from the new teacher as they learn how to perform like a teacher. There is then an assumed, and perhaps presumed, sequence to learning to teach that would be reflected in the teacher education programme.

There are two distinct approaches to the surface structure of the models in Table 1, each drawing up a different privileging of teacher education pedagogy; what we are referring to here as knowledge first, or person first. We use these terms as a way of categorising how the various approaches to initial teacher education represent learning to teach in different ways.

In the category of knowledge first, the knowledge that teachers need is foregrounded, but situated externally to them. There are different ways of conceptualising that knowledge, perhaps the most famous being Shulman’s (1986, 1987) own taxonomy of the knowledge bases for teaching. Other more recent accounts offer similar but nuanced categorisations or conceptualisations of teacher knowledge (see, for example, Grossman, 1990; Bransford et al., 2005; Rowland et al., 2005; and Ball et al., 2008). While these accounts do not necessarily outline the sequential stages of knowledge that teachers need, they do recount the depth and breadth of knowledge areas that teacher education programmes need to cover. Teacher education programmes based on, for example, the stage-theory model or the theory-into-practice model might organise these knowledges to ensure their coverage during the programme. Within the apprenticeship model, one might expect the apprentice to demonstrate deepening knowledge in these areas.

Pedagogical content knowledge is often treated a little differently, since, particularly in the elementary or primary school phase, it is seen as an organising principle through which other knowledges are situated (see Turner-Bisset, 2001). In Shulman’s (1986: 9) own definition, pedagogical content knowledge is described as ‘the most useful forms of representation … the most powerful analogies, illustrations, examples, explanations, and demonstrations – in a word, the ways of representing and formulating the subject that make it comprehensible to others’.

Here, knowledge is placed first. The description reflects the presentation or laying out of subject matter; the teacher then has the task of adapting that knowledge in a manner which makes it accessible for students – what Dewey may have called psychologising subject matter. This is an approach that can be seen in stage theory, theory into practice, core practice or indeed the apprenticeship model, as each implies that the knowledge needed to be a teacher somehow sits outside of the teacher themselves and needs to be absorbed or accumulated by them in order for them to progress. However, where the knowledge sits varies in each of these models. In the theory-into-practice model, the knowledge sits inside the academic canon, articulated as ‘theory’ to be understood and then applied by the new teacher. The teacher educator selects and curates this knowledge, and prepares it for the new teacher to learn and then apply to their practice. For the apprenticeship model, the knowledge sits within the expert teacher, who acts as the ‘master’ from whom the apprentice learns. This is also the case with approaches based on core practices, where the teacher educator rehearses, appropriates or represents certain practices for the new teacher to copy and develop: the knowledge sits with the teacher educator to be imparted and demonstrated. In each of these approaches, however, the knowledge is seen as sitting external to the new teacher, and as being something that needs to be acquired by them. The pedagogy of the teacher education programme in such cases is designed to make knowledge accessible to them. Ellis (2010: 106, 107) argues that general approaches to course design, particularly in light of policy reform in England and the focus on school-based initial teacher education, have ‘relied on an acquisition view of learning and a view of knowledge as a thing that is transferred’, leading to ‘tensions and contradictions’ which ‘arise out of contradictory conceptualisations of knowledge and learning in relation to experience in schools’. These approaches contrast to what we have described as person-first approaches. In these approaches, the new teacher is positioned as a competent but as yet uninitiated member of the teaching community. The approach assumes that the knowledge and skills required to be a teacher are developmentally within reach of the new teacher, and the pedagogy is designed to enable it to come to the fore. Two good examples here are reflective practice and situated learning, or communities of practice, which are often referred to as approaches to professional learning evident in any of the overarching models outlined above. Donald Schön (1983, 1987) is most widely regarded for explaining how and why reflective practice is effective, although the idea stems as far back as Dewey. Distinguishing between reflection on action, and in action, Schön (1983, 1987) notes how it is through considering what happens in the ‘white heat’ of a professional’s repertoire, and how they consider novel problems that their expertise grows. This may be supported by critique which occurs through peer or mentor feedback, alternative viewpoints or theories, or the use of appropriate critical questions, but in each case, it is the practitioner at the centre of the community who is responsible for this development through their active engagement in reflecting on their practice.

While critiques of reflective practice have noted its individualised nature, Lave and Wenger (1991) introduce further social elements through their description of situated learning and communities of practice. Here, again, the learning takes place through social and reflective encounters, and such an approach is well suited to apprenticeship models of learning, as professional skills are acquired through involvement within the community of practice. This is described as ‘legitimate peripheral participation’ (Lave and Wenger, 1991: 29), within an apprenticeship model, which can lead to professional learning (for both beginning and experienced teachers).

In these person-first approaches, learning how to perform as a professional requires the new teacher to question and challenge themselves and their actions, as well as to build on the work of the community. The ‘answer’ lies within the teachers and their social network, rather than being external to them, and this is why such an approach operates within models of teacher learning where the teacher works with a supportive community. This is also the case with clinical practice, where the teacher has to question and review their own practice and experience in order to develop their expertise. In clinical practice, this may be more systematised, through a research-aligned pedagogical process, with due attention to data and evidence to substantiate learning, but the presumption is still that the answer is obtainable to the knower through their engagement with the learning process.

Shulman (2005a: 55) refers to a signature pedagogy’s deep structure as being ‘a set of assumptions about how best to impart a certain body of knowledge and know-how’. As with the above discussion of surface structure, notions of knowledge first or person first are similarly pertinent, reflecting assumptions, at the most fundamental level, as to whether the teacher education curriculum is best delivered through direct instruction or through a process that places the beginning teacher at the heart of the learning process.

Returning to the example of stage theory which is informed predominantly by an analysis of beginning teacher concerns (Fuller and Bown, 1975), the deep structure does not focus on these concerns at the individual level, but rather on the way in which they can be seen to reflect stages of development for the majority. Stage theory identifies the way in which beginning teachers move from an early stage, at which their concerns are about self and survival in the classroom, to a stage where these concerns focus on tasks and situations, to a final stage where concerns are about the impact of their teaching on students and individual student learning. Conway and Clark (2003: 467) argue that the model has endured because ‘the normative expectation that teachers’ ultimate concern ought to be about students and student learning rather than self or situations seems to blend description of what development is like with a prescriptive template for teacher development’. The model therefore becomes one that leads to programmes being structured around these anticipated stages of development and, crucially, places an emphasis on supporting beginning teachers through these stages, each of which presents its own challenges. Development only occurs as the concerns are worked through in the context of teaching itself, so the role of school-based mentoring becomes paramount, with mentors being guided as to how to respond to beginning teachers’ developing concerns at different stages of the programme (Maynard and Furlong, 1995).

At the heart of the models described above is an inherent stance in relation to the roles of both theory and practice within teacher education programmes. While traditional theory-into-practice models, such as those prevalent in England and elsewhere around the 1970s, where the focus was on teaching the foundation disciplines of education within teacher preparation programmes, are no longer in evidence, some aspects of a theory-first approach can still be seen, where evidence of what works is used to direct beginning teachers as to what needs to be applied to their own practice. Elsewhere, the increasingly significant influence of the core practices movement can be seen in policy reforms internationally, where emphasis is placed on the practice of teaching, and the way in which beginning teachers are taught how to enact specific, carefully defined practices (Ball and Forzani, 2009; Grossman et al., 2009). Here, the deep structure is premised on a notion that professional competence comes primarily from the teacher’s ability to replicate a series of prescribed practices, acquired through what could be seen as a reductive, technicist approach to teaching (Ellis et al., 2019), which also includes ‘the risk of peripheralizing equity and justice’ (Philip et al., 2019: 251).

The idea that practice should precede theory, or that practice is of itself sufficient, can also be seen to underpin ‘craft’ approaches to teaching, and a broader range of apprenticeship models. Such models reflect the conception of the teacher as ‘craft worker’, or ‘executive technician’ (Winch et al., 2015: 202). In these models, the underpinning structure is one in which theory and practice are in constant tension, held between the drive for efficiency in ‘acquiring and using well-learned schemas and routines’ (Hammerness et al., 2005: 374) and effectiveness in the ability to ‘rethink key ideas, practices and even values in order to respond to novel situations’ (Hammerness et al., 2005: 358–9). The deep structure of the approaches that attempt to achieve the latter have, at their heart, the belief that the only model of teacher learning that will develop effective professional judgement by teachers is one that offers sufficient opportunity for ‘critical reflection’ (Winch et al., 2015: 204).

Kriewaldt and Turnidge (2013: 106), in discussing clinical practice models of teacher education, refer to the need for teachers to exercise such critical reflection through a process of ‘clinical reasoning’, which they define as ‘the analytical and intuitive cognitive processes that professionals use to arrive at a best judged ethical response in a specific practice-based context’. While the concept of clinical reasoning can be seen as being applicable to a range of professional learning contexts, the specific context of teacher education differs in that beginning teachers are often required to practise (and demonstrate competence) at an early stage in their professional preparation. The evidence on which any ‘clinical reasoning’ is based may therefore be inevitably limited, and clinical practice models must provide opportunities for practice and theory to be fully integrated in a coherent way. Underpinning this is the need for beginning teachers, working with more experienced teachers within a community of practice, to engage in a process of enquiry. Such a process requires beginning teachers to draw on multiple sources of knowledge, and to subject all ideas for practice to critical examination (Hagger and McIntyre, 2006) by engaging in what McIntyre (1993) refers to as practical theorising. McIntyre (1993) sees this as a dialectic process, through which beginning teachers are equipped to interrogate different kinds of knowledge, both theoretical and practical, in order to use the results of such reflection to inform their own developing practice (Burn and Mutton, 2015). The above approaches inevitably emphasise the centrality of the beginning teacher in the process of learning to teach, and acknowledge the complex role that individual teacher beliefs play in the process of professional learning (Pajares, 1992; Tatto, 1999, 2019; Fives and Buehl, 2008, 2016), as well as the extent to which individual dispositions to learning influence the way in which teachers engage with the process of learning to teach (Oosterheert and Vermunt, 2001; Hagger et al., 2008; Pedder and Opfer, 2013).

For Shulman (2005a: 55), the third fundamental dimension within a signature pedagogy comprises the moral dimension: the ‘implicit structure’ which signals the set of beliefs about ‘professional attitudes, values and dispositions’. This is the dimension that showcases how future professionals within the field, in this case beginning teachers, learn how to ‘act with integrity’ (Shulman, 2005a: 52). Our contention is that foregrounding different theoretical underpinnings showcases how different models of teacher education shape future practice, and convey the values and hopes of that profession: the implicit structure.

Approaches to teacher education which are predicated on evidence from empirical observations of what works in teaching, such as the core practice model above, convey that teaching techniques that foreground surface acts of teaching have value because they lead to immediately visible results in terms of control of student behaviours or test scores. This and other practice-based knowledge-first models foreground that the role of the novice professional teacher is to focus on acts of schooling which will enable learners to function appropriately within a specific school environment, and to learn the appropriate knowledge as prescribed in a specific curriculum. To act with integrity within the framing of the apprenticeship models and the core practice models is, for the beginning teacher, to learn the practices which deliver results and outcomes. To do this, the beginning teacher needs to absorb ‘ready to use integrated knowledge that has been learned implicitly in the same context’ (Philpott, 2014: 29). By practising techniques within that context, the beginning teacher performs the acts of teaching which align with what that context values, thereby reproducing recognisable surface acts of what it is to be a teacher in that specific context. What is valued in these models is that teachers produce learning outputs from their students which can be measured and compared. In addition, the practices that are promoted are framed as of benefit to all children, regardless of their individual circumstances. This shifts the moral dimension of how to act with integrity as a teacher from that of teaching the whole child, through an individualised holistic approach, to delivering a generalised set of standardised practices, activities and normative outcomes. The focus is on the ‘collective craft knowledge of teachers’ as to what works best for most students (Alexander, 2000: 275); the beginning teacher’s personal experiences and beliefs about what might be best for individual learners are rendered less important. In this way, the implicit structure highlights that the professional values of teaching are grounded in an economic framing which measures student and teacher performance, and rewards those who achieve the highest measurable outcomes. In knowledge-first models of teacher education, to act with integrity as a professional teacher is to deliver outcomes within a highly regulated and meritocratic system of education.

We can contrast this to people-first models of teacher education which foreground the social and relational aspects of learning to act as a teacher. In these models, attention is paid to the ways in which the novice learns within a professional community of practice. Through reflective dialogue with peers, with more expert colleagues within the field, and with scaffolded interactions with the knowledge base of teacher education and teaching (reading and debating what has been written about within the field), novice professionals learn how to value relationships with individual children, with groups of learners and with colleagues. Through understanding how these pedagogic relationships are influenced by specific contexts, beginning teachers learn how to make judgements about pedagogical acts which are ethically grounded. To act with integrity as a professional teacher in this framing is to understand that teaching is a social act, that relationships and context matter, and that the profession requires teachers to be prepared to make the best decisions for the learners in any given situation. And, as such, these decisions are based on ‘value-laden choices, which can and will be questioned by others’ (Kelchtermans, 2021: 1506). In person-first models of teacher education, ‘acting with integrity’ is not without risk, but it is predicated on moral and ethical ‘wise judgement’ (Kelchtermans, 2021: 1507).