Intentional, tacit, contingent: knowledge recontextualization in the official History curriculum - a Critical Discourse Analysis

ABSTRACT

This paper reports a Critical Discourse Analysis (CDA) exploring the nature, causes and epistemic effects of knowledge recontextualization in the ‘official’ Key Stage Five History curriculum in England. “Recontextualization” refers to inevitable changes that occur to knowledge as it is ‘pedagogized’, due to the value-laden practices and contexts which enact and shape curriculum-making. Five accredited examination specifications and three ‘generative’ policy documents were analysed using Fairclough’s methods and interpreted through a Bernsteinian theoretic lens. Five problematic forms of knowledge recontextualization will be discussed: canonization; commodification; de-diversification; knowledge made static; and epistemic inconsistency. The application of CDA to everyday ‘official’ curriculum artefacts illuminates the role of micro-level ‘language-in use’ in bringing particular constructions of subject knowledge into being. It is suggested that not all these recontextualizations were intentional, nor fully explainable through macro and meso structural factors. Some had been enacted ‘by accident’ in the contingencies of fragmented and negotiated local practice. Epistemic and discursive ‘literacies’ are suggested as key to engendering agentic practice within the official curriculum-making community, as well as enabling teachers to select and pedagogically mediate specifications in line with local epistemic aims.

KEYWORDS:

• Curriculum
• history instruction
• policy
• epistemology
• discourse analysis
• evaluation

Introduction

The ‘official’ curriculum—the curriculum that those with political authority expect pupils to follow (Kelly, 2009)—now generates significant ‘washback’ across the whole education system due to the associated regimes of accountability and ‘high-stakes’ assessments of pupils and schools (Baird et al., 2017; Stobart, 2008) that have become a preferred policy lever (Ball, 2015). Despite teacher mediation of official curricula (Priestley & Philippou, 2018), this washback not only influences forms of schooling (Westbury et al., 2016), it has deeper epistemic effects. Processes of selecting which knowledge to include/exclude within official curricula, sequencing, connecting and assessing that knowledge, and putting it to work for social, cultural and political purposes, as well as pedagogic ones, construct and represent ‘subject knowledge’ in particular ways (Bernstein, 1990, 2000). Official curricula, particularly manifested in ‘curricula-as-documents’ (Westbury & Sivesind, 2016, p. 760), construct conceptualizations not just of what knowledge ‘counts’, but also what knowledge ‘is’ (Stearns et al., 2000), and how we come to ‘know’ (Elwood & Murphy, 2015). Questions of what episteme official curricula construct, and how those episteme came to be constructed in that way, therefore demand sustained attention from curriculum researchers (Lambert, 2018; M. Young, 2015).

This paper reports a Critical Discourse Analysis (CDA) (Fairclough, 2003) exploring ‘official’ constructions of historical knowledge within the Key Stage 5 (KS5) History curriculum for 16–18 year olds in England. The ‘specification’ and policy documents analysed form part, in England, of what Sivesind and Westbury (2016) have described as an architecture of documentation in which the official curriculum is instantiated. Drawing on critical realist theories of discourse (Fairclough, 2001, 2003) they can be understood as not only constructive agents in their own right, but as also bearing the imprint of what are known to be complex, negotiated and contested contexts of production (Baird & Lee‐Kelley, 2009; Cadwallader & Tremain, 2013). They thus offer valuable evidence for curriculum researchers looking to address calls to, ‘surface[e] … processes, assumptions and influences that shape curriculum-making,’ (Priestley & Philippou, 2018, p. 156), something which CDA allows us to accomplish. This study therefore not only examines what historical knowledge is understood to be in the official KS5 History curriculum in England, it sheds light on the real social practices which lie behind the typically anonymous and institutional character of official curriculum documents (Gerrard & Farrell, 2013).

The conceptual focus of the research is recontextualization (Bernstein, 1990, 2000), the transformations which occur to knowledge when it is ‘relocated’ (or, perhaps preferably, reconstructed) into pedagogic communication. Five problematic forms of knowledge recontextualization identified will be presented: (i) canonization; (ii) commodification; (iii) de-diversification; (iv) knowledge made static; and (v) epistemological inconsistency. The findings of the CDA suggest that to explain these recontextualizations we must turn our gaze beyond only macro structural factors (politics, ideology, ‘market’, culture), and the ‘meso’ level objectives and professional norms of assessment. The specific recontextualizations arose, at least in part, contingently, at the ‘micro’ level, out of the interaction of unconscious day-to-day practices and ‘language-in-use’ (Gee, 2011) within a curriculum-making process that was simultaneously complex, resource-constrained and highly delegated (Baird & Lee‐Kelley, 2009). These explanations suggest new directions for research-practice dialogues focused on engendering reflexive linguistic and epistemic ‘literacies’ to support more agentic recontextualizing amongst assessment professionals, as well as to render the ‘invisible’ semiotic properties (Hasan, 2012) of official curriculum texts more visible to teachers.

Knowledge recontextualization

Recontextualization (Bernstein, 1990, 2000) refers to the transformations that occur to knowledge when it is ‘relocated’ (or perhaps preferably, reconstructed) into pedagogic communication as a consequence of the value-laden social practices through which this is enacted. There is a reservoir of ‘meaning potential’, residing in fields of epistemic and discursive ‘production’, such as disciplinary research communities (Bernstein, 2000), on which curriculum-makers can draw when developing pedagogic discourse. As this is drawn on selectively, and often indirectly (Lombard & Weiss, 2018), and the knowledge that is selected is subsequently re-represented, re-combined, and ‘put to work’ in different ways, for different explicit and tacit purposes, it is changed in the process. Essentially, knowledge in the curriculum bears the imprint of the curriculum-maker(s) and their (micro to macro) contexts of practice. For sociologists of knowledge this is a fruitful process to study not just in terms of the structure and properties of the resultant pedagogic discourse, but also because the process reveals something of the ‘distributive rules’ (social structures and power relations), ‘recontextualizing rules’ (pedagogic norms) and ‘evaluative rules’ (social systems of value) (Bernstein, 1990, 2000) in operation in a given educational context at a given time.

Recontextualization has been identified as a conceptual tool with significant potential for establishing greater dialogicity between theorized ‘ideal types’ of curriculum and their expression in real sites of pedagogic practice (Deng, 2015b; Gericke et al., 2018; Lambert, 2018). This may be particularly pertinent in relation to curriculum theory that has emerged out of the ‘knowledge turn’ (Hoadley et al., 2019), given the emphasis in ‘Social Realist’ approaches on a normative epistemic relationship between the structured conceptual and procedural knowledge of (evolving) ‘academic disciplines’ and ‘school subjects’ (Muller & Young, 2019), although as Deng (2015a) points out, problematization of the origins, positioning and nature of ‘subject knowledge’ within the curriculum is by no means confined to this tradition. Recontextualization foregrounds rather than backgrounds the multiple transformations that knowledge must inevitably and necessarily go through when it is ‘relocated’ and reminds us that this process is contextually-specific and non-linear. It is therefore, a powerful reflexive concept. The question for curriculum-makers becomes not ‘how can I reduce/eliminate recontextualization?’ but, ‘how can I make the ways I am transforming knowledge more visible, so I can evaluate their pedagogic effects, and recontextualize with greater agency?’

This study addresses increasing calls for more research into the nature, mechanisms and outcomes of knowledge recontextualization, particularly at the ‘micro’ empirical level of classrooms and curricula (Deng, 2015b; Gericke et al., 2018; Lambert, 2018; Lim, 2017b; M. Young, 2015). It makes a distinct contribution to the need for better understandings of recontextualization within ‘official’ curriculum-making, and the contribution of assessment as both a site of and mechanism for recontexualization (Lambert, 2018; M. Young, 2015). The research complements and extends the work of other scholars who have made recontextualization in different epistemic domains and jurisdictions their object of analysis (e.g. Lim, 2014, 2017a, 2017b [Critical Thinking]; G. McPhail, 2013, 2016 [Music]; Ormond, 2017b [History, New Zealand]; Pandraud, 2011 [Science]; Sharma and Anderson, 2009 [Science]; Wright & Froehlich, 2012 [Music]). The work is offered in the spirit of, ‘build[ing] little by little a strategic knowledge,’ (Ball, 2010, p. 69) of a phenomenon which is ultimately crucial to understanding and theorizing interrelationships between epistemic and discursive ‘production’ and ‘reproduction’/reconstruction (Bernstein, 2000), and between the multi-layered and dispersed ‘official’ and pedagogic artefacts, discourses and social practices that collectively encompass curricula and curriculum-making (drawing on Priestley & Philippou, 2018).

Official curriculum-making for ‘high stakes’ assessments

In England, awarding bodies author and manage ‘high stakes’ summative assessments and publish supporting resources, including specifications which outline what will be assessed and how. Given recognition of the, ‘disproportionate,’ role that awarding bodies can have, ‘in shaping the curriculum,’ (Young, 2011, p. 271) there is great potential in detailed empirical examination of the ways in which they represent and construct ‘subject knowledge’, and the factors which influence that process (Lambert, 2018; M. Young, 2015). There is now substantial empirical evidence that ‘high stakes’ qualifications and their attendant textual artefacts and practices shape pedagogic practice in schools globally, discursively as well as practically (Torrance, 2017, p. 83)—so-called ‘backwash’ (Baird et al., 2017), which ‘trickles down’ throughout the system in anticipation of assessment requirements (Madaus & Russell, 2010). Qualification reform has thus become a preferred policy instrument for effecting system change in many countries (Madaus & Russell, 2010). In England, the initial purpose of specifications to improve transparency in the assessment system (Chapman, 2011) has now evolved into a much deeper endeavour by awarding bodies to actively influence the curriculum in schools—‘intentional backwash’ (Baird et al., 2017). Indeed, in the specifications analysed in this research the terms ‘specification’ and ‘qualification’ were used interchangeably with ‘curriculum’ and ‘course’, confirming that, in this context, initially separate meanings have indeed been synonymized. Therefore, whilst recognizing the growing body of work demonstrating that policy enactment in schools is a heavily negotiated (Ball et al., 2012; Priestley et al., 2012; Priestley & Philippou, 2018), dynamic, and highly contextually-specific process (Helsby & McCulloch, 1997), examination specifications are a good empirical proxy for the intended official curriculum, and not just a site in which recontextualization and its causes can be made visible, but also a significant causal mechanism in their own right. They might best be understood as, ‘working interim recontextualization[s]’ (Westbury & Sivesind, 2016, p. 760), which generate boundaries, or ‘zones of enactment’ (Spillane, 1999), for the schools which use them, as well as influencing (drawing on Priestley & Philippou, 2018) other sites in the constellation of curriculum-making practices, and hence merit sustained critical attention from curriculum researchers.

In many ways English¹ qualification specifications exemplify what an ‘official’ process of curriculum-making looks like in an era of what has been described as ‘network governance’ (Ball & Junemann, 2012), where officially mandated curriculum-making has been ‘translocated’ from the state to quasi-independent organizations (Deng, 2018). In England, assessment specifications are produced by independent awarding bodies, following subject-level content generated by the Department for Education (DfE) in dialogue with ministers and ‘Criteria’ and ‘Guidance’ authored by the exams regulator Ofqual, an agency accountable to Parliament (Baird & Lee‐Kelley, 2009). In one direction, therefore, they are accountable to the state, requiring official accreditation (Ofqual, n.d.), a ‘ritual of verification’ (Ball, 2015) conferring authority on awarding bodies to regulate the assessment work, and related curricula and pedagogic practice, of schools. Facing the other way, they are accountable to schools/colleges through a qualifications ‘market’, where choice of awarding body is exercised. This might be viewed as a ‘Commercial Recontextualizing Field’ (Robertson, 2012), where competition for market share leads to ‘commodification’ of knowledge (Ball, 2007), with ‘aestheticized’ (Fairclough, 2003) specifications used to promote the qualification to prospective ‘consumers’. For example, visually attractive front-covers and layouts and rhetorical, informal and/or literary language are often incorporated. Nonetheless, it would be too simple to understand assessment-related curriculum-making solely in terms of an intersection of politics/policy and ‘market’. Assessment itself is a specialized domain, with its own norms and imperatives (Elwood & Murphy, 2015; Higham & Yeomans, 2011), and there is also significant interchange with the ‘pedagogic recontextualizing field’; for example, in England most assessment specification authors are practising or former teachers (Baird & Lee‐Kelley, 2009). Official, assessment-related curriculum-making in England might best be understood as a hybridized, ‘entangled’ (Baird & Lee‐Kelley, 2009), recontextualizing field where constant negotiation is taking place between the ‘official’, the ‘pedagogic’, the ‘commercial’ and the imperatives of assessment itself. A study of recontextualization in the documents that result from this complex, negotiated process, as presented here, is therefore highly relevant in understanding the nature and outcomes of official, assessment-related curriculum-making in the contemporary networked state.

Theoretic relevance of KS5 history education in England

Engagement with the concept of recontextualization begs the fundamental question of what knowledge is being recontextualized/reconstructed. Bernstein spoke in terms of recontextualization opening up a gap between university-based disciplines which were sites of epistemic and discursive ‘production’ and school subjects that ostensibly shared the same name. To investigate recontextualization in these terms clearly requires an empirical setting where a normative epistemic relationship between the ‘school subject’ and the ‘academic discipline’ is endogenously relevant; imposing such an analytic lens in a setting where it was not an aim of the curriculum-making in the first place would certainly find ‘gaps’ but these would not merit the label ‘recontextualizations’. History education in England, and the KS5 official curriculum, offer such a theoretically relevant site.

Pupils studying History at 16+ (KS5) in England—where it is an optional subject—follow two year courses of study that lead to ‘A-Level’ (or equivalent) qualifications, which facilitate the transition out of schooling into higher education or work. The A-Level assessment specifications and generative policy documents analysed for this study have their origins in government reforms which sought to significantly increase the role of higher education (HE) in determining the KS5 curriculum (Gove, 2010)—a project in which History received particular attention (Haydn, 2012). Ofqual Conditions outline that the first purpose of KS5 History qualifications in England is to support, ‘progress[ion] to undergraduate study at a … [HE] establishment, particularly … in the same subject area,’ (2014a, p. 3) suggesting a degree of epistemic continuity between these contexts is desired. This Discourse of progression and epistemic connectivity with the university ‘discipline’ of History is visible across all specifications. For example, Pearson Edexcel describe practices of, ‘drawing on independent subject specific expertise to challenge and validate our qualifications,’ (2017, frontmatter); OCR describes consultation with ‘Higher Education’ to design a course which is, ‘a suitable foundation for the study of history … in [HE]’, ‘designed to emphasise knowledge,’ (2018, p. 2); AQA describes students as, ‘historians’ (2018, p. 5). Even if such comments are interpreted as rhetorical legitimating or positioning devices (drawing on Fairclough, 2003), the presence of a consistent Discourse emphasizing links between the KS5 curriculum and HE/‘academic history’ suggests that the relationship between them is broadly understood to be a relevant and desirable one by stakeholders.

Prevailing pedagogic norms in the history education community in England overlap to an extent with those of policy makers and awarding bodies outlined above in viewing epistemic connections between ‘school history’ and ‘disciplinary history’ as highly desirable (Counsell, 2011; Fordham, 2017), albeit underpinned by a differing rationale, emphasizing ‘disciplinary knowledge for all,’ (Counsell, 2011, p. 201), rather than progression into HE for some. There are distinct ‘recontextualizing rules’ in operation in the English history education community that emphasize organizing curricula and pedagogic practice around ‘substantive’/‘first order’ and, particularly, ‘second order’ analytic and procedural historical concepts, for giving students access to ‘disciplinary’ modes of historical thinking (Counsell, 2011). ‘Substantive’ concepts categorize, organize and connect features of the past—such as ‘revolution’ or ‘monarchy’ (Lee, 2011), whereas ‘second order’ concepts emphasize modes of historical thinking and practice—particularly cause/consequence, similarity/difference, change/continuity, significance, evidence and interpretation (Counsell, 2011). Muller and Young (2019) have identified History as perhaps the paradigm case of a pedagogic community seeking to connect the ‘academic discipline’ and school subject for this reason. To analyse English KS5 History specification documents in terms of an ‘academic discipline’/‘school subject’ process of knowledge recontextualization is therefore not only an appropriate theoretic device to apply to make constructions of subject knowledge in the specifications, and their origins, analytically visible, it also reflects aims and reflexive lenses which are internally relevant to both the ‘official’ and ‘pedagogic’ fields within this curriculum-making space.

Research design

The research design synthesized the Critical Realist models of Corson (1991) and Fairclough (2001), where empirical identification and description of a phenomenon (in this case, the recontextualization of ‘disciplinary’ historical knowledge), moves into deeper analysis and explanation of how and why it has come into being in that way, in a given context; a mode of analysis which is ultimately oriented to contributing to a theory of change.

Data was analysed using Fairclough’s CDA, a Critical Realist form of discourse analysis (2001). CDA is a particularly powerful method for studying the official curriculum because it provides methodological tools which connect detailed linguistic evidence to a much wider network of Discourses and social practices (Fairclough, 2003), moving beyond content analysis into what have been described as, ‘difficult but necessary,’ (Livingston et al., 2015, p. 515) explorations of the processes and structures which may have shaped the texts (Fairclough, 2001) as well as their potential effects as constructive artefacts in their own right (Fairclough, 2003). CDA is commonly associated with enquiries into power, ideology (Blommaert & Bulcaen, 2000) and, increasingly, neoliberalisms (Fairclough, 2013). Whilst these themes are of course, far from absent from enquiries about curriculum, this study contributes to the growing interest in applying CDA to studies of knowledge itself and the discourses that express and regulate it (van Dijk, 2011).

The five accredited KS5 History qualifications with market share in England at the time of conducting the empirical research (early 2019) formed the corpus analysed: three ‘A-Levels’ (AQA, 2018; OCR, 2018; Pearson, 2017) and the minority qualifications (Gill, 2016) the International Baccalaureate Diploma (IBO, 2015), and the ‘Pre-U’ (CIE, 2017). The minority qualifications are not substantively discussed in this paper due to different contexts of production and circulation, but provided a comparative analytic lens which contributed to the findings and discussion. Three subject-specific policy documents which serve as ‘generative texts’ (Prior, 2009) for the specifications were also analysed (DfE, 2014; Ofqual, 2014a, 2014b).

Schools select from a wide range of optional units within specifications, which must, at the time of writing, comprise a ‘substantial’ proportion of British History, the history of at least one other country or state, and cover a chronological range of at least 200 years (Ofqual, 2014a). Specifications outline intended content for each optional unit as well as making more general statements about the purpose and nature of the qualification, advising how students will be assessed and providing administrative information. Each document was analysed in full. Protocols for quality and validity based on Steinke (2004), Antaki et al. (2002), and Fairclough (2003) were followed. The research was ethically approved via institutional procedures and British Educational Research Association (BERA) ethical guidelines (2018) were followed throughout.

The research reported here should be understood as a form of ‘close-to-practice’ research (Wyse et al., 2018, 2021) due to my ongoing professional relationship with an awarding body, through which I had, six years prior, been involved in authoring parts of one of the specifications analysed, as well as a previous career in history teaching when I had been a ‘user’ of the specifications. The Critical Realist research design, and the use of CDA, enabled the ‘familiar’ to be ‘made strange’ (drawing on Delamont et al., 2010). Personal experiences and perceptions were challenged through systematic data analysis and iteration with theory and empirical literature, thus allowing (in Critical Realist terms) the ‘actual’ phenomenon and its ‘real’ underlying mechanisms to come into view (O’Mahoney & Vincent, 2014). There is a small but growing interest in CDA as a method for reflexive practitioner research amongst curriculum-makers (for example, Williamson, 2012), and it is hoped this paper will open up wider conversations about the affordances of this methodology in ‘close-to-practice’ curriculum research.

Findings

This paper will focus on five problematic forms of knowledge recontextualization identified through the analysis:

(i) canonization;

(ii) commodification;

(iii) de-diversification;

(iv) knowledge made static;

(v) epistemic inconsistency.

Whilst there was also significant discursive evidence of forms of recontextualization that generated connectivity between the KS5 curriculum and ‘disciplinary history’ these problematic forms merit particular attention. They act contrary to stakeholders’ and curriculum-makers’ aims to generate epistemic connectivity and progression between KS5 History and HE, and also have likely negative epistemic effects on pupils’ and teachers’ understandings of what historical knowledge is, as well as problematic pedagogic consequences, which will be outlined below.

Each form will be briefly described, with indicative evidence presented, before moving into consideration of how they can be explained, with a view to identifying implications for research and practice.

(i) Canonization

The first form of recontextualization identified was canonization. History is a multi-perspectival discipline where multiple acceptable interpretations and approaches to understanding the past coexist (Grever & Stuurman, 2007; Rublack, 2012). Despite this, in many places across the corpus single perspectives of historical events and processes had been selected, at the exclusion of others, and these were then linguistically presented as if they were the only, or the ‘most valid’, perspective, canonizing them. This may inhibit students’ development of ‘disciplinary’ understandings of the nature of historical knowledge, particularly given the ‘high stakes’ context which attaches examination success to viewing the past through these given lenses.

Canonization was particularly enacted through three linguistic features, illustrated here using AQA (2018).² Firstly, the use of a strong epistemic modality when describing interpretive or conceptual elements of the subject content—‘issues and perspectives which are central to the period of study,’ (p. 16)—and the use of the definite article, which implies there is only one perspective on the issue—‘the nature of change over time,’ (p. 5, italics added). Secondly, the privileging of particular conceptual lenses in the titles and framing of certain topics; for example, ‘revolution’ in relation to Russia’s experience of political change in the twentieth century, compared to ‘transformation’ in the case of China (p. 8). Finally, by pre-interpreting the ‘significance’ of events, rather than leaving this open for teachers and students to problematize—a ‘Discourse of significance’ further explained in relation to commodification below.

To some extent this appears to be intentional, and possibly a consequence of the way that different awarding bodies had interpreted regulatory expectations. For example, in describing a conscious attempt to transmit, ‘similar perspectives,’ about complex historical issues, AQA refers directly to ‘coherence’ (p. 15), a regulatory requirement (DfE, 2014, p. 2). In the case of AQA, canonization also appears to have arisen out of a commitment to ensuring students experience the conceptually-oriented ‘disciplinary’ pedagogy viewed as an ideal in English history education—the canonized perspectives, embodied in, ‘key questions,’ are explicitly intended to, ‘inform and guide how the content should be studied,’ (p. 17, italics added). Nonetheless, in the same specification, canonization co-existed with language which legitimizes more ‘open’ epistemic practices, including ‘explor[ation]’ (p. 38) and ‘reflect[ion]’ (p. 43). Room was left for pedagogic encounters with multiple conceptual tools through linguistic constructions including, ‘such as,’ (23 uses across the unit descriptions), and occasional suggestion was made that substantive disciplinary concepts ought to be problematized, by placing terms in inverted commas (for example, ‘reflect on … what makes a “state”,’) (p. 34). This suggests competing epistemic beliefs underpin the curriculum-making, where attempts to assimilate arguably incompatible commitments to both multi-perspectivity and conceptual ‘coherence’ can result in overall epistemic inconsistency.

(ii) Commodification

A commodifying Discourse was visible to greater and lesser extents across the whole corpus, and is likely to be contributing to, as well as reflective of, an increasingly marketized qualifications system in England where competition for market share between awarding bodies is mirrored at school level in competition between optional subjects (Harris & Haydn, 2012). Historical knowledge is selected and recombined into what Ball (1998, 2007) has termed ‘bite-size chunks’ (here, optional units of study) which are ‘aestheticized’ (Fairclough, 2003) with snappy titles (a form of ‘branding’) which act to pre-interpret the knowledge included, contributing to canonization (above). For example,

‘In search of the American Dream: the USA, c1917–96ʹ;

‘South Africa, 1948–94: from apartheid state to ‘rainbow nation’;

‘Industrialisation and social change in Britain, 1759–1928: forging a new society’;

‘The changing nature of warfare, 1859–1991: perception and reality’

(all Pearson, 2017, frontmatter, italics added).

This is problematic because it risks fragmenting historical knowledge into an episodic ‘list of topics’ rather than an interconnected body of knowledge (Harris & Reynolds, 2018; Lee & Howson, 2009; Ormond, 2014).

Commodification is also visible in discursive hybridization within unit descriptions, where lexical choices more commonly associated with literary or rhetorical genres than academic or policy texts are used to present the material favourably to potential ‘consumers’ (teachers/schools) This particularly emphasizes the affective (as opposed to epistemic) properties of the historical knowledge selected, as well as enacting a ‘sensationalization’ of the past, including consistent insistence on the (indisputable) ‘significance’ of the topics selected:

Affective properties of historical knowledge

‘The engaging topics available to them throughout the course … ’

(AQA, 2018, p. 5);

‘a variety of exciting historical topics.’

(OCR, 2018, p. 2);

Sensationalizing lexical choices

‘The aftershocks of these changes’;

‘at their starkest’;

‘underwent the turbulence of’;

‘bitter divisions’;

‘a shaky start’;

‘an imperial catastrophe’;

‘passion, tensions and commitment’;

(all Pearson, 2017, pp. 49, 52, 68, 76, 82, 92, 98).

A Discourse of ‘significance’

‘powerful influence’, ‘legacy’;

‘shape the course of its history’;

‘repercussions for the centuries that followed’;

‘era of decisive change’;

‘decisive shift’;

‘momentous effect’;

(all Pearson, 2017, pp. 16, 16, 18, 23, 24, 32, 36).

Whilst seeking student engagement in learning is not problematic in its own right, it is historical knowledge itself that is represented as the source of that engagement, rather than pedagogic strategies, arguably devaluing the role of the teacher and constructing an evaluative rule which suggests historical knowledge should be primarily appraised on its affective, rather than epistemic qualities. Knowledge is evaluated at least partially, as Lyotard (1989) argued, on the basis of its ‘saleability’ above its ‘truth’. The emphasis on including only the most geographically and temporally ‘significant’ events is also likely to be having an effect on awarding bodies’ topic selection, contributing to de-diversification of the curriculum (below).

(iii) De-diversification

Knowledge has been selected in a partial and unrepresentative way for inclusion with the KS5 curriculum, resulting in History being ‘de-diversified’; some subjects, topics and approaches are accorded particular prominence and others are rendered significantly less visible. In the context of a ‘high stakes’ assessment, a system of value is being constructed as to which aspects of the past are ‘worth knowing’. As Table 1–4 demonstrate,³ there are significant imbalances in chronological and geographical coverage, with pre-colonial, non-western histories particularly invisible at this level. There is significant over-representation of political history, and concomitantly a high visibility of white, male elites, restricting the visibility of non-elite actors to those who came into contact with these political processes (Mandler, 2013). Social and economic history, and, particularly, cultural history are significantly under-represented when their prominence as modes of historical research are considered (Rublack, 2012; Tosh, 2006). This is a fundamental misrepresentation of the direction of travel in historical scholarship in recent years, and, in the light of recent research on inclusion and progression in History (for example, in England, Alexander & Weekes-Bernard, 2017; Atkinson et al., 2018; Harris & Reynolds, 2014) raises significant questions about the suitability of the curriculum for promoting equitable attainment and progression across all social groups.

Table 1. Number of available units across A-Level specifications by chronological period (AB = Awarding Body)

	AB1	AB2	AB3	Total
Pre-Medieval Up to ca. 1000	5	0	0	5
Medieval ca. 1000 to ca. 1500	7	2	4	13
Early Modern ca. 1450 to ca. 1750	21	9	7	37
Modern ca. 1700-c.1950	14	8	10	32
Majority/all Twentieth Century	11	9	19	39

Table 2. Number of available units across A-Level specifications by geography

	AB1		AB2		AB3		Total
	M	S	M	S	M	S	M	S
Great Britain and Ireland	20	2	10	0	15	0	45	2
Europe	25	5	13	0	17	2	55	7
Middle East	3	2	0	2	0	3	3	7
Africa	2	1	0	1	1	1	3	3
North America	3	5	4	0	4	2	11	7
Latin and Central America and the Caribbean	0	3	0	0	0	0	0	3
Asia, the Pacific and Australasia	5	3	1	2	3	14	9	19

M = main focus; S = some content

Table 3. Number of available units across A-Level specifications by method/perspective

	AB1		AB2		AB3		Total
	M	S	M	S	M	S	M	S
Political	50	4	28	1	28	9	106	14
Military	2	27	0	6	2	6	4	39
Economic	0	38	0	19	0	29	0	86
Social or Cultural	6	35	1	21	10	26	17	82

M = main focus; S = some content

Social and cultural history differ methodologically, but intended emphasis could not always be reliably inferred, so the categories were combined.

Table 4. Cross-tabulation of geographic and chronological coverage

	Pre-Medieval	Medieval	Early Modern	Modern	Twentieth Century
British	2	8	16	14	5
European	2	4	18	11	20
Middle East	1	0	1	0	1
Africa	0	0	1	0	2
North America	0	0	0	5	6
Latin/Central America	0	0	0	0	0
Asia-Pacific	0	1	1	2	5

The consistency and prevalence of de-diversification suggests that an historic content bias in the English History curriculum (Cannadine et al., 2011) towards what has been termed the ‘knowledge of the powerful’ (Young, 1971) has interacted with a ‘meso’ level normative narrative in the assessment field that minimal change is preferred by teachers (see below)—who in a marketized qualifications context are the target ‘consumers’—to preserve a curriculum structure which is disciplinarily ‘de-diversified’, and socially unrepresentative.

‘we have kept our most popular topics so you can re-use or adapt your departmental resources … ’. (AQA, 2018, p. 5)

Minimal change is represented as an asset based on assumptions about teachers’ preferred way of preparing lessons.

‘topics are included to encourage centres to look beyond the most commonly taught areas of history … ’. (OCR, 2018, p. 2)

This suggests an assumption that teachers tend to gravitate towards longstanding, ‘commonly taught’ areas.

There is certainly evidence that schools are reluctant to take up the limited number of newly introduced ‘diversifying’ options (Dunn et al., 2016), and can be reluctant to change the examination units or topics they teach, tending to cluster around a narrow range (Child et al., 2014, 2015; Dunn et al., 2016; Harris & Reynolds, 2018). Nonetheless, de-diversification is better understood as an epistemic and discursive phenomenon that exists in the dialogue between pedagogic practice in schools, ‘high stakes’ assessments and accountability regimes, and socio-political contexts, than as the ‘fault’ of one particular group. As the analysis above suggests, ‘distributive rules’—social hierarchies and power structures which act to distribute access to knowledge and confer worth on knowledges differentially—have interacted with field-specific ‘norms’ (‘recontextualising rules’) to de-diversify the curriculum, and de-value historical knowledge of and about particular social groups (responding to, and in turn constructing, ‘evaluative rules’). More research is merited into how official curriculum-making intersects with pedagogic norms in schools, in contextually-specific ways, to generate this de-diversified ‘canon’, and how this might be more effectively challenged.

(iv) Knowledge as static

The specification discourse predominately represents historical knowledge as ‘given’ and static, rather than dynamic, challengeable and evolving. This is enacted through three main discursive features. Firstly, in some, but not all, cases, continuity of pedagogy is represented as normative; for example, as outlined above, teachers’ ability to, ‘re-use … resources,’ was represented as a selling point of one specification (AQA, 2018, p. 5), suggesting that ‘good teaching’ is quite possible without ‘updating’ historical knowledge content. Secondly, in all cases, specification continuity was positioned as a selling point:

‘Some new topics,’ (AQA, 2018, p. 5 italics added)

‘Some’ suggests a deliberate moderation of change.

‘A large range of traditional and new topics,’ (OCR, 2018, p. 2)

This suggests content changes have occurred at the ‘topic’ level, rather than at the level of content within topics. ‘Traditional’, in particular, directly appeals to the curricular past.

Both these features suggest an underlying assumption that historical knowledge itself is not subject to substantial or regular development. Finally, there are moments of unproblematic privileging of historical perspectives that can be considered somewhat ‘out of date’. For example, characterizing events of 1625–1660 as an ‘English Revolution’ (AQA, 2018, p. 40) draws heavily on interpretations that originated over half a century ago, and would generally no longer be used uncritically in historical scholarship (Braddick, 2015).

This static construction of historical knowledge has potential implications for pupils’ epistemic development—what they understand the nature of historical knowledge and knowledge generation to be—and may be eliminating their access to the ‘syntactic’ (Schwab, 1964) aspects of disciplinary knowledge that give it its potential ‘power’, based on Muller and Young’s recent theoretic reflections (2019). The normalization of ‘static’ knowledge and pedagogies also has potential implications for teachers’ professional learning and resultant classroom practice. For example, Foster (2013) has demonstrated that problematic definitions of the Holocaust were being unwittingly included in classroom practice where teachers had not been supported in ‘updating’ their knowledge of this issue.

(v) Epistemic inconsistency

Overall, representations of what historical knowledge, and the nature of historical ‘knowing’, are understood to be are inconsistent, not only across the corpus of documents, but within individual specifications themselves. This has parallels in findings relating to other curriculum contexts, including Brady (2015), Ormond (2014), Smith (2016, 2019) and Priestley and Humes (2010), suggesting that this may be a common outcome of curriculum-making that merits further attention. In the case of KS5 History, such inconsistency is partly the product of a regulatory context which requires students to demonstrate understanding of the historical practices through which knowledge is generated in some parts of the curriculum, through critical analysis of ‘primary sources’ (historical evidence) and historians’ ‘interpretations’ of specified issues, whilst simultaneously devaluing such ‘syntactic knowledge’ by according it significantly lower weighting in the assessment objectives (Ofqual, 2014a), prioritizing recall and manipulation of historical knowledge divorced from knowledge of the circumstances of its production. Nonetheless, an epistemic ‘fuzziness’ is also represented in, and constructed by, awarding bodies’ own discourse. This is evident, for example, in the way historical texts are represented, where some texts are accorded the status of ‘interpretations’ which require critical analysis and other similar texts are, at the same time, to be used for uncritically accessing information:

‘students are expected to consult a range of resources, which may include textbooks, course books and work of academic historians … there must be explicit analysis and evaluation of two differing interpretations by academic historians.’. (AQA, 2018, p. 68)

Here, texts are divided into two categories – ‘resources’ and ‘interpretations’, which should be approached differently.

‘The interpretations selected will always be deliberate constructs by historians … ’. (OCR, 2018, p. 68)

This implies that not all texts, or indeed ‘interpretations’, are ‘deliberate constructs’

Even where pupils are expected to work with ‘primary sources’ or engage with competing ‘interpretations’ of the past, the conceptualization of these concepts is arguably thin. In relation to sources, what Jordanova (2006) termed the ‘cult of the archive’ is very evident, with documentary, primarily written sources dominating, and very limited acknowledgement of methodological developments in the discipline such as growing use of oral history or material artefacts. Indeed, the term ‘methods’, whilst used in the International Baccalaureate Diploma (IBO, 2015), is not used in the A-Level curriculum, discursively separating what students do with historical texts and artefacts from the practices of historians. ‘Interpretations’ are largely synonymized with ‘works of [named] academic historians’, as opposed to all historical narratives. This not only occludes the ‘memory turn’ that has been such a focus of the discipline in recent years (Jordanova, 2006), it simultaneously devalues and removes from critical scrutiny the narratives of non-academic communities.

Recontextualizations: intentional, tacit and contingent

The consistency of certain forms of recontextualization outlined above—such as de-diversification—both within and across specifications and policy texts, suggests that certain macro and meso level structural factors and professional norms are certainly relevant in explaining the forms of recontextualization seen. Some of these were clearly intentional; for example, appeals to ‘consumers’ about the attractive qualities of the specifications (resulting in ‘commodification’ of knowledge) were consistent and self-conscious. Others might be better understood as tacit. For example, OCR represented newly introduced, ‘pre-colonial, non-western,’ (2018, p. 2) units as a selling point of the specification, suggesting it believed it was offering a more diverse view of history than its competitors, yet nonetheless remained de-diversified overall.

Not all of the forms of recontextualization outlined above appeared consistently, however, not only across different specifications, but also within individual texts. For example, AQA was highlighted as an example where the canonization of particular historical interpretations directly competed with a multi-perspectival view of the past. Such internal inconsistencies are only explainable if we view recontextualization, like all aspects of curriculum-making, as Priestley and Philippou (2018) have persuasively argued, as highly contingent and contextually-specific. Discursive fragmentation and inconsistency make visible a system in which multiple, often weakly networked, authors make curricula in a highly devolved, sequenced, and therefore ultra-localized, way (Baird & Lee‐Kelley, 2009), ‘trading off’ competing stakeholder priorities. ‘In the moment’ discursive decisions solidify transient and fragmented ‘moments’ of practice into discursive and epistemic artefacts which, drawing on Prior (2009), become agents in their own right. Tronsmo (2020) has recently suggested that more research into micro-interactions in the small-scale environments of practice where ‘real time’ epistemic decision-making and negotiation occur is needed to fully understand teachers’ ‘knowledge work’; this study suggests the same is true of official curriculum-makers. This is an important direction for future research because the texts that result from these chains of contingent ‘moments’, ‘become’ the (intended) curriculum (Westbury & Sivesind, 2016), and the competing recontextualizations therein ‘become’ official constructions of historical knowledge—ones which have been revealed to be ‘fuzzy’ at best, and problematic in parts.

Discussion and conclusions

This research makes a contribution to our theoretic understanding of knowledge recontextualization, and the mechanisms which generate it, in official, assessment-related curriculum-making. It also proposes a working typology of forms of recontextualization within the KS5 History curriculum in England, offered as a starting point for further conceptual and empirical dialogues within and across related contexts. The focus on a specific empirical ‘case’ in depth speaks to Deng’s (2015a) call for an orientation to ‘the practical’. The application of CDA responds to the call for methods which engage systematically with the, ‘surfacing of processes, assumptions and influences that shape curriculum-making,’ in the interests of not only greater theoretic understanding, but also more sensitive practice (Priestley & Philippou, 2018, pp. 156). It complements existing work which emphasizes the importance of macro and meso level factors in influencing the recontextualization of knowledge, such as political and ideological context (e.g. Lim, 2014, 2017a), differences between popular and elite culture (e.g. G. J. McPhail, 2016, 2013), or assessment structures (Ormond, 2017b), by suggesting we also need to attend to ‘micro’ level, contingent factors related to ‘language-in-use’ and ‘in the moment’ curriculum-making practices and decisions. This surfacing of what is often a tacit or backgrounded element of curriculum-making (Deng, 2015b; Gericke et al., 2018; M. Young, 2015) is a necessary precondition for reflexivity and agency amongst practitioners and it is hoped that the work will open up wider conversations about the affordances of CDA as a methodology not just for curriculum researchers but, as suggested by Rogers et al. (2016), also for practitioner-researchers as a way of making knowledge recontextualizations, and the factors and contexts generating them, more visible.

The study builds on emerging research into teachers’ curriculum decision-making (e.g. Harris & Reynolds, 2018; Livingston et al., 2015; Ormond, 2017a; Smith, 2019), agency (e.g. Alvunger, 2018; Campbell, 2012; Harris & Graham, 2019; Priestley et al., 2012) and reflexivity (Hizli Alkan & Priestley, 2019), extending it into the field of ‘official’ assessment-related curriculum-making. Its ‘micro’ level linguistic analysis contributes to these dialogues by raising further questions about how far the episteme and outcomes which official curriculum-makers believe are being enacted are actually realized, once contingent, ‘in the moment’ practices, and the complexities and constructive properties of language-in-use, act on and within the curriculum-making process. Without such ‘control’ over the conceptualizations of subject knowledge the specifications they author express and construct (drawing on Campbell, 2012), official curriculum-makers cannot be regarded as agentic.

The research suggests that agentic recontextualization is unlikely to be realized within official curriculum-making without reflexive consideration of the role that language-in-use plays in instantiating subject knowledge in the written artefacts that embody the ‘official’ curriculum; what might be termed a ‘discursive’ or ‘textual’ literacy (Scott, 2000). Nonetheless, significant questions are raised as to the degree to which such discursive agency could currently be engendered in complex, resource-constrained environments, where risk is high and responsibilities are diffuse and distributed (Baird & Lee‐Kelley, 2009; Cadwallader & Tremain, 2013), and therefore frequently lacking in the collaboration and relationality that has been identified as necessary to set and achieve shared curricular goals (Bladh et al., 2018).

Recontextualization is offered here not just as a theoretic lens but as a powerful reflexive concept in its own right. Whilst recontextualization might be an inevitable, and indeed necessary, outcome of curriculum-making, the specific transformations it enacts on knowledge are not. These emerge from highly specific contexts of practice, language- and episteme-in-use. Deng (2015a) encourages curriculum researchers and practitioners to view recontextualization not as a mere mechanism, but rather a moment for intentionality and deliberation. To become more agentic recontextualizers, this research suggests that assessment professionals must be supported in what might be termed ‘epistemic literacy’; a form of professional knowledge which is not only fluent in what Scheffler (2007) described as ‘philosophies of’ a subject, but where a dialogic professional architecture is in place to externalize, negotiate, ‘notice’, challenge and reflexively instantiate ‘epistemes-in-use’ at all stages of what are known to be long, complex, multi-authored production processes (Baird & Lee‐Kelley, 2009). This is likely to require collective professional examination of what are often ‘routinized’ (Sivesind & Westbury, 2016) and unconscious curriculum-making practices.

This research has endeavoured to illuminate one site amongst a complex constellation of curriculum artefacts and practices. It demonstrates how a particular set of contextually-specific curriculum-making practices have enacted particular recontextualizations of historical knowledge. According this recontextualized knowledge the status of ‘official knowledge’ generates material and semiotic boundaries shaping the enactment of these curricula at school level (Campbell, 2012; Kontovourki et al., 2018; Spillane, 1999). Whilst the focus of this discussion has been the agency of official curriculum-makers involved in the specification authoring process, the research has repercussions for History teachers too as they, in turn, undertaken their own local curriculum-making within the ‘zones of enactment’ (Spillane, 1999) constructed by these specifications. How far these epistemic ‘zones of enactment’ are mirrored in international ‘official’ History curricula would be a fruitful line of further enquiry. For teachers of KS5 History in England the conceptual language of canonization, commodification, de-diversification, knowledge made static, and epistemic inconsistency may provide a sense of direction for agentic recontextualizations of their own as they enact and mediate the official curriculum in line with local epistemic and pedagogic aims. Discursive and epistemic literacy is as relevant at the level of schools as it is for official curriculum-making, so that the tacit episteme that result from official recontextualizations can be rendered visible and reflexively negotiated in a way which puts pupils’ disciplinary (and holistic) epistemic development at its heart.

Ethical standards

This research was ethically approved by the University of Cambridge and was conducted according to British Educational Research Association (BERA) ethical guidelines.

Data availability

There is no data set associated with this research. The documents analysed are available in the public domain, with weblinks provided in the bibliography.

Copyright

The very short extracts from the OCR, AQA and Edexcel A-Level specifications that appear in the manuscript are included for the purposes of criticism and review. Copyright for the full documents resides with the authors of these documents, for which details can be found in the bibliography.

Acknowledgments

The author would like to thank the anonymous peer reviewers for their constructive and detailed comments; Dr Mary Earl for her critical observations on the development of the work reported, and for reviewing an early version of this article; and Dr Riikka Hofmann for her wider support.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

This research was supported by the Economic and Social Research Council under grant ES/P000738/1.

Notes on contributors

Siobhan Dickens

Siobhan Dickens is a doctoral researcher at the Faculty of Education, University of Cambridge, UK. She was formerly a secondary History teacher for ten years and remains a senior assessor with a large awarding body.

Notes

1. Education and assessment policy is devolved in the United Kingdom. Qualifications produced by Welsh, Scottish and Northern Irish awarding bodies can be submitted to the English regulator Ofqual for accreditation for use in England, but at the time of conducting the research (2019) these had a market share very close to zero in England (Ofqual, 2018) so were not included in the corpus.

2. In some cases, linguistic examples are all drawn from the same specification to demonstrate the existence of a Discourse (as opposed to isolated examples). This is by no means to argue that this Discourse was not present in other specifications.

3. Geographical and chronological analysis was undertaken using content analysis. Each specification was coded twice, with a third round undertaken if a discrepancy resulted. Where coverage was broadly equal (50/50) across two categories, it was coded in multiple columns, hence overall totals do not correspond directly to the numbers of units available. Methodological/perspectival emphasis was determined using CDA.