Changing our (Dis)Course: A Distinctive Social Justice Aligned Definition of Open Education

This paper investigates the degree to which recent digital Open Education literature is aligned to social justice principles, starting with the first UNESCO definition of Open Educational Resources (OER). A critical analysis of 19 texts was undertaken to track dominant and alternative ideas shaping the development of Open Education since 2002 as it broadened and developed from OER to Open Educational Practices (OEP). The paper begins by outlining the method of texts selection, including defining the three principles of social justice (redistributive, recognitive and representational justice) used as an analytical lens. Next the paper sets out findings which show where and how the principles of social justice became lost within the details of texts, or in other digital agendas and technological determinist debates. Finally, a new social justice aligned definition for Open Education is offered. The aim of the new definition is to provide new language and a strong theoretical framework for equitable education, as well as to clearly distinguish the field of Open Education from mainstream constructivist eLearning.


Introduction
Ten years have passed since the Cape Town Open Education Declaration (The Cape Town Open Education Declaration, 2007), so it is timely to review the progress towards the objectives laid out for Open Education.However, this is made difficult by the fact that the Declaration describes multiple possibilities, and Open Education continues to mean many different things to many different people.
The Cape Town Declaration avoided setting out a definition for Open Education, suggesting that this would leave what it termed an "emerging" movement free to develop and take advantage of technologies and innovations as yet unknown.In fact, Open Education has a long history of providing education to non-privileged learners via Open Universities.However, the rise of digital and Internet-based learning made Open Education seem new -it brought many new people, ideas and institutions to the field with different backgrounds.Scholars of the modern Open Education movement worked almost without reference to the early corpus of Open Education literature (Weller, Jordan, DeVries, & Rolfe, 2018).Therefore, this paper starts from the position that the literature from the 2002 UNESCO declaration (UNESCO, 2002) can be considered a new chapter of Open Education literature for the digital era.
Since 2002, Open Education research and practice has expanded outwards from Open Educational Resources (OER) to an interest in a broader set of Open Educational Practices (OEP).Researchers have debated the definition of "openness" common to both areas of practice (Peter & Deimann, 2013).
Definitions of openness are variable, but tend to highlight a sharing of effort and/or resources with all teachers and all learners, often positioned against closed practices as negative or lacking innovation (Wiley, Bliss, & McEwan, 2014).Yet, while conference keynotes, panels and debates have discussed Open Education initiatives with regard to their alignment with social justice aspirations, it was surprisingly difficult to find and therefore cite published Open Education literature focused on social justice and the enablement of excluded or disadvantaged learners.The 2002 UNESCO declaration certainly is clear about its intended benefit for excluded learners in developing countries (UNESCO, 2002).However, the recently published statement on the 10 th Anniversary of the Cape Town Open Education Declaration talks about collaboration, innovation and quality more than ideas of redistributing educational resources and opportunities to those who need them the most (Cape Town Open Education Declaration 10th Anniversary: Ten directions to move Open Education forward, 2017).Meanwhile, a consensus on a definition or purpose for Open Education remains elusive.This brings us to critical questions which lie at the heart of current unresolved definitional debates and which are the motivation for this research: Where is social justice in the contemporary Open Education literature?And similarly, is Open Education an innovation for everybody, or is it primarily about removing barriers to the marginalised and excluded?
The problem that this paper addresses is two-fold.Firstly, a lack of definitional clarity is a problem for those that consider Open Education as a valid field of endeavour.When there are easily shifting goalposts, we never really know if we have contributed or made substantive progress.As influential practitioner and commentator Wiley notes, "without clarity about our foundational commitments, it can be easy to wander" (Wiley, 2017b).Secondly, for those Open Education practitioners and researchers who consider social justice important aims, particularly in times of increasing not decreasing inequality, the problem is more than definitional.It means that the field's efforts towards reducing educational and societal inequalities are dispersed, inefficient and ineffective.Without a critical mass of social justice orientated views of what Open Education is capable of, we are unlikely to direct our energies where it is most needed, and, as a result, achieve less equitable educational outcomes.This purpose of this research, therefore, is to investigate the degree to which the discourse of contemporary Open Education literature is concerned with social justice principles and the ideal of fairness or equality of educational provision.The method of investigation is a critical analysis of a sample of contemporary Open Education texts.This paper begins by outlining the method of choosing and analysing the key texts including the definition of social justice used as a critical analytical lens.Next, findings are outlined showing where and how the principles of social justice became lost within the details of texts, or in other digital agendas and technological determinist debates.The paper concludes by offering a new definition for Open Education which is aligned to the principles of social justice, and some observations about how it may be applied.

Gaps in the Existing Literature
Open Education papers discussing social justice issues are hard to find.A search of the IRRODL journal found only two that had "social justice" in either the title or abstract, and the Australian Journal of Educational Technology (AJET) had none.While there is some work focussed on social inclusion and the digital divides (DiMaggio & Hargittai, 2001;Lane, 2013;Warschauer, 2003;Willems & Bossu, 2012) these are similar but somewhat narrower constructs than social justice.Similarly, related works about widening participation of formal education continues to emerge from the Open and Distance Education institutions, including some scoped to bridge informal and formal learning (Farrow, Arcos, Pitt, & Weller, 2015;Lane, 2016).However, as recent citation analysis has shown, the connection between the field of widening participation in formal education and the OER and OEP literature is extremely tenuous (Weller et al., 2018).While Open Education researchers have used textual analysis methods to critically appraise the social-justice alignment of Open Education policy documents in different global contexts (Cox & Trotter, 2016;Mukama, 2018) or institutional mission statements (Tait, 2013), however, to date no research exists which interrogates texts and definitions in an attempt to shed light on how assumptions might have constrained progress towards social justice.
Therefore, this research contributes a new understanding or explanation for a perceived lack of Open Education progress, an alternative account beyond the dominant discourse that if we could only improve awareness and uptake of Open Education policy or practice, we would be able to make it both "more accessible and more effective" (Cape Town Open Education Declaration 10th Anniversary: Ten directions to move Open Education forward, 2017)."Instead it argues that social justice outcomes for Open Education do not flow from the affordances of our technologies, nor any view of our "openness", but flow from our commitment to design explicitly for it via the application of one, two or three of the principles of social justice.

Theoretical Lens for the Research: Social Justice Definition
Making a judgement on the degree to which Open Education key texts align with social justice relies on a good working definition of social justice.For this research, the following definition of social justice was developed from the work of Keddie (2012), Fraser (1995), andYoung (1997) as: A process and also a goal to achieve a fairer society which involves actions guided by the principles of redistributive justice, recognitive justice or representational justice.
Redistributive justice is the most long-standing principle of social justice and involves allocation of material or human resources towards those who by circumstance have less (Rawls, 1971).Recognitive justice involves recognition and respect for cultural and gender difference, and representational justice involves equitable representation and political voice (Fraser, 1995;Keddie, 2012;Young, 1997).
The example of an open textbook can be used to show how these principles can be applied to Open Education.Providing a free textbook to learners of colour in the American two-year college system, is redistributive justice in action.It reduces the costs and increases the chances of success for learners who "by circumstance have less" -they are marginalised in education, workplaces and more broadly in society.But how "open" is the textbook for marginalised learners if indigenous, Hispanic and learners of colour are invisible inside the textbook and perhaps invisible in the whole curriculum?The editing of such a textbook to include images and cases featuring more diverse communities, businesses and people will be an act of recognitive justice.But what if the textbook features people of colour, but does not value their perspectives, knowledges or histories?What if the textbook takes a white colonial view of black lives, if black stories are told solely by white voices?The development or selection of a new version of a textbook (or perhaps a new resource altogether) written by people of colour where they are free to represent their own views, histories and knowledges would be an act of representational justice, to give voice to those who are often not heard.Table 1 summarises the three principles and provides some examples as applied to Open Education.From these examples, we can see that an open course or textbook might meet the principles of social justice, in three quite different ways.For the most socially just outcome, it would ideally meet all three principles.Most importantly, the example also shows that providing an open textbook to all learners, particularly if they are predominantly already educationally privileged, may not be social justice at all.Depending on the cohort and their needs, it may enable a range of outcomes for a range of learners.For elite cohorts, it may in fact give a further leg-up to those whom by circumstance typically have more.

Selection of Texts
Nineteen key texts were selected for analysis on the basis of the author's focussed searching and reading to locate social justice influences during a concurrent process of PhD research in Open Education.The aim was to locate a sample of influential texts which shaped the modern Open Education period, and could then be analysed for their major interests, social justice or otherwise.
Selection of texts started with the highly cited "Declarations" and authors such as Wiley and Weller who are associated with social-justice aligned organisations or projects in the Open University or American College sector.Selection then shifted to texts discussing what Open Education was, or was not -that influenced definitional debates in the literature and at international conferences.Couros was selected for completeness with regard to tracking the development and values of "the Open Movement".Additional texts were selected relating to definitions and representational/graphic models of OEP, which were often used in lieu of definitions.Another set were chosen as responses to MOOCs, where the definitional debates about what Open Education was and wasn't intensified again.
These texts, when arranged chronologically, roughly fit into three somewhat overlapping time periods with common concerns: Table 2 lists the sample texts in each phase.While any sample of texts cannot be exhaustive, and invariably omits more than it includes, nevertheless they have been chosen because of their influence on the definitional debates about what is distinctive or valuable about the field of Open Education.
It is important to acknowledge that none of the texts set out to write specifically about social justice as currently defined, nor may the authors have awareness of the three principles used in the social justice definition of the term.While each of the authors had their own purposes for their work, nevertheless these texts were picked up and discussed by others, and in the absence of a definition, often used as de facto definitions or discourse to shape, explain and justify the work and the purposes for the work that followed.Therefore, the extent to which these texts align with social justice ideas is indicative of the extent to which recent, digitally enabled Open Education aligns with social justice ideals.

"Openness Determinism" and the Iterative Analysis and Development of Themes
The analysis involved multiple readings and note-taking alongside excerpts from each text.The early re-readings identified relevant segment/s of the longer texts where social-justice ideas were addressed, partially addressed or implied.Where social justice ideas were found to be absent, notes were made on the alternative interests or themes.Early emerging themes included benefits to IT workers, potential/barriers, educational innovation and quality improvement.The "openness as good" theme emerged as a major alternative (non-social justice) theme in which the idea of what "openness" could achieve was fetishized as if it had some kind of inherent power, reminiscent of the technical determinist literature.Technological determinism is a problematic and ultimately ineffective approach to technology implementations, which assumes that the particular capabilities of new technologies will always improve the situations into which they are brought.Such over optimism tends to understate the influence of people and the social context for the success -and also the failure -of particular technologies.The technological determinist literature was consulted for a definition, and the "affordance account" was a match for the theme emerging in the analysis.The "affordance" account of technological determinism is the overemphasising or attributing of the power for improvements in education (or society) to the general decontextualized properties or "affordance" of technology, particularly to promote the uptake by others (Oliver, 2011;Selwyn, 2011).Therefore, within the subsequent analysis, the theme was labelled "openness determinism" and a final reading was made consolidating and finalising themes.
The findings of the analysis are outlined then discussed in the sections following Table 2.

Foundational Digital Phase Findings
The 2002 and 2012 UNESCO OER Declarations stand out for their clear social justice alignment, whereas social justice principles are absent or only weakly implied in the other key texts of this period.The 2002 UNESCO proposal to pursue OER is framed as an action of redistributive justice from those with greater resources to those with fewer, i.e., developing countries.The addition of the phrase with the "full participation" of those countries -is an action of representational justice.Other sections imply a degree of recognitive justice.The 2012 version acknowledges the limits of progress and is a renewed call to ten OER-based social justice actions.However, similar to the problem of the 2002 Declaration and also Wiley's "4/5Rs" specification text, the social justice intentions are lost if only the headings are cited as a de facto summary of the whole text.We are left with the purposeless and deterministic "foster awareness and use of OER" -as if it is an end to itself -which assumes that justice or equality will naturally follow from access to OER.The shorter versions or summaries as well as the more technical key texts from this phase, tend to read as technologically deterministic "affordance" accounts.Within this deterministic account, access to resources for all is paramount -OER adoption potential is promoted in terms of free resources online, highlighting new technology as a key enabler.
As Table 3 below summarises, the key texts from the Foundational digital period 2002-2012 either do not address social justice principles as currently defined, imply them weakly or are present in the context or purposes for the work but absent from the often-cited shorter texts and definitions.
The dominant discourses in these texts are not social justice ideas but rather that openness and OER is good, access is good, and that re-using OER have potential to change education for the better.A later variation on this dominant discourse continues to assert the potential value of increasing access and re-using OER, if only barriers to adoption (i.e., challenges related to people) could be overcome.The sample texts often frame openness, OER, open-software community and/or the open movement as good because they are based on collaboration, sharing, and democratisation of knowledge.However, none of these are necessarily good in terms of social justice if the sharing and collaboration is primarily between relatively highly privileged Global North IT workers.Because of this, the term democratisation of knowledge (or of education) raised in these texts (and in many other subsequent papers) cannot be conflated with social justice even though the term may sound like a synonym for social justice.Similarly, while for some readers the term "access" may imply ideas of social justice, however, as currently defined, access is not a synonym for social justice.
The Wiley et all literature review (2014) reports there is more OER traction for speeding up mainstream course development: benefits of OER are primarily institutional, relating to reputation and efficiency -prescient views that would be echoed and amplified in the MOOC or appropriation phase.Alternative themes -openness determinism (affordance); access to free materials online for everybody: The text discusses the potential of Open Education in terms of innovation and quality improvement in education for all; asserts that "we are on the cusp of a revolution", i.e., change afforded by improved access to materials and various forms of openness.This text is representative of many others like it, and by this stage it represents a dominant account around potential, re-use, access and education for all.

Paris OER Declaration (2012)
Social Justice present but hidden in context: Taken in its entirety, this Declaration is not a definition, but a call to action, and in particular a call to social justice actions.The text shows influence of the widening participation field in higher education (itself a social justice/inclusion process).The full texts of the first, second and seventh actions taken together call for OER as distributive, recognitive and representational justice actions.These details are easily lost when citing only the ten actions heading texts.
Social justice principles absent: OER as social justice actions for students are not discussed, and where the three types of social justice are mentioned or implied, they are framed as unrealised potential of OER.
Alternative themes -openness determinism (affordance); institutional benefits: The potential of Open Education dominant discourse is re-stated even while noting the reported reality that 12 years after OER was launched, using free digital materials for Global North education is still the majority practice.Re-use via technical editing and open licencing has not gained traction.OER noted as providing institutional reputational benefit.More equitable forms of education are framed as a "potential" or unrealised benefit of re-purposing OER -if barriers to adoption could be overcome.

Broadening Phase Findings
In these texts, Open Education is re-framed by some authors as broader than the dominant OER or "4/5Rs" account from the previous phase.For example, the Hodgkin-Williams et al first "continuum" of Open Education indicates that more than legal and technical solutions are required, re-asserting the complex people and social side of both learning and academic development around OER.An alternative view was developed of OEP as a set of innovative collaborative teaching and learning practices available to all and for all.However, in this account the notion of "all" still does not distinguish between those who by circumstance have more, or less.The benefit of a broadened OEP view is that researcher/practitioners in the field excluded by a narrow "4/5Rs" definition of openness were invited "into the tent".With more participants shaping the field, a dominant conversation developed around what kind of openness (resources/OER vs practices/OEP) would expedite a path towards a common good -without defining how such a "good" could be measured or whom the resources/practices actually enable.What is not questioned in these broadening phase key texts is the idea that openness can effect change for the better.This variation of determinism is known in the technological literature as the "normative" account, which is problematic as it constrains other alternative views (Oliver, 2011;Selwyn, 2011).Dominant discourse leaves alternative ideas -such as those relating to social justice -at the periphery of a field.A common alternative idea left marginalised was the utility of free materials (but not openly licenced so not "open") for direct benefit of learners -marginalised or otherwise.Despite their redistributive justice potential free resources were criticised for not being open to modification by other educators, seen as more efficient and hence costeffective.Further, the analysis identifies the problem of "reification" where normative accounts are further promoted by the use of shared visualisations, which tend to calcify a dominant understanding during a process of simplification to aid sharing and uptake by others (Wenger, 1998).Such reification may be an unfortunate unintended consequence of some of the "continuums" visual diagrams published during this period.
The OEP literature also introduce another term, "empowerment", that may have been read as a synonym or idea related to social justice.The use of the term may have created an impression that social justice was still being discussed and pursued.However, the term "empowerment" in the sample texts align with only half of the concept of recognitive justice where they are described as an innovation for everyone, because they leave open the very real possibility of further empowering the already privileged (who continue to dominate most universities) so they may take their place as future leaders, thus further propping up current societal inequality.

Social justice limited and hidden:
There is a very limited version of social justice via the principle of redistributive justice in the report.Also, social justice principles are absent from the simpler and easierto-reference definition and infographic which reads like a checklist for any eLearning innovation.
Alternative theme -eLearning innovation for everyone.Part of a broadening agenda that blurs boundaries with mainstream education.

Social justice is absent or hidden. Alternative theme -openness determinism (normative view):
Wiley adopts a new term as a response to the way OEP had broadened the field towards constructivist online learning."OEP enabled pedagogy" reifies the technologically deterministic account of technical re-use as central to change.An alternate view would be that organisational culture and investment in people wrapped around OER technical systems are the cause of the change.Such "people power" seems to be present in the actual collaborative work Wiley and his projects engage in but is absent from the cited and discussed blog posting texts.

Social justice partially implied (empowerment synonym).
Alternative theme -eLearning innovation for everyone: OEP definition: starts to imply the possibility of some form of recognitive justice through the use of the term empowerment -but like definitions of OEP that came before -does not discuss the importance of doing so to remediate differential opportunity between learner groups.The benefits of collaborative co-construction are framed as educational innovations for everyone.
Standing somewhat apart from the technological determinist discourses discussed, the Butcher (2015) and the dos Santos et al ( 2016) texts go beyond notions of access (dominant theme from foundational texts) to discuss the supported progress and success of learners.Access, progress, and success are three key terms from the field of widening participation in higher education.The use of these terms denotes an attempt to change the demographics of higher education learners and graduates to resemble the multi-cultural, gender, dis/ability, indigenous, and socio-economic mix that occur in wider society.Such definitions suggest different approaches to future research -beyond measuring self-reported learner satisfaction, to investigating the progress and graduation rates of more advantaged as compared to less advantaged learners in any particular context.
However, the most common themes in this set of literature is not social justice but OEP becoming an eLearning innovation for everyone.This broadening agenda blurred boundaries with mainstream education and its concerns of digitising curriculum, technology innovation and improving the quality of educational provision.These ideas were already being discussed in the earlier phase, and the Cape Town Declaration spoke in terms of the innovation and quality of education agenda (The Cape Town Open Education Declaration, 2007).However, in this broadening phase, the notion of OER/OEP as an eLearning innovation applicable to everyone becomes the dominant discourse.
Recently, the genesis and assumptions of OEP are tracked back to recent trends in the broader educational literature, namely social constructivist, student-centred learning (Cronin & MacLaren, 2018).I would suggest that OEP can alternatively be considered a contemporary online iteration of social constructivist learning, positioned against OER as a more positivist resource and teacherfocussed paradigm.The risk then, at this point in time, is Open Education broadening so far as to lose its distinctiveness and point of differentiation between mainstream eLearning or higher education.In other words, Open Education could wither as a separate field and become subsumed into the field of eLearning and/or social constructivist learning.

Appropriation Phase Texts
The analysis of the Appropriation phase texts (see Table 5 below) revealed a crisis point in the field where the term "openness" was overlaid with commercial meanings such that any sense of "open as common good" was lost and authors moved between wanting to abandon the term, to redefining and re-claiming what was "truly open."However, as previously discussed, the fetishization of openness is a problematic form of determinism, which reduces the effectiveness of Open Education by not attending to the complex socio-cultural context of learning and technology use.Table 5 summarises the findings and themes.The "Battle", in the context of commercialism of openness, is the battle for control of higher education and the fight between the idea of education as a greater good vs education as a saleable commodity.It's not the battle for reducing inequality in education and society, however, there may be an assumed view that student benefit flows from staff control over learning.

Watters's blog post (2014)
From "Open" to Justice Clear Social justice aim: Watters returns us to a social justice aspiration for Open Education and uses the term social justice explicitly.

Rohs and Ganz (2015) MOOCs and the claim of education for all: A disillusion by empirical data
Redistributive justice aims: This paper is one of the many emerging empirical and demographic studies showing the failure of online "access" to provide for fairer education.
Blog posts and comments from this period demonstrate anger towards commercial appropriation of Open Education as practitioners are seen to be "systematically forced out of the movement they started by their new slick corporate overlords."There is also recognition of "the damage…done" to the field by limiting the work of the field to only that which is profitable (Lamb, 2013).The kinds of justice called for in these texts relate to practitioners/educators and their ability to make ethical choices.There is a possible implication or assumption that such ethical choices on behalf of educators would benefit all learners but this connection is not made explicit in the texts analysed.
The 2014-15 texts from Watters and Rohs and Ganz signal a major shift in discourse as more empirical studies came to light identifying a lack of improvement in educational inequality from Open Education initiatives.Both texts signal a growing discomfort with dominant "access and openness as good" discourses, and identify a widening digital divide as likely outcomes should similar approaches continue.
Watters' text provides a powerful rejoinder to the determinist discourse that had been building over previous phases and which assumes "openness" can and will do the work of social justice.While Rohs and Ganz's texts frame their argument within the digital divide debate, Watters uses the term social justice explicitly.She draws attention to the relationship between normative openness determinist views of Open Education -"All the right nods from all the right powerful players within 'open'" and the failure of Open Education to enact or provide for more equitable education.What Watters describes is the outcome of openness determinism:  (Watters, 2014).

Summary of Themes Across the Three Phases
Social justice principles were present in 2002 at the start of the digital foundation phase but eroded over time as numerous alternative ideas and discourses developed.While the principles were sometimes implied or hidden in the detail of the digital foundation phase, in the broadening phase they were mostly absent.Throughout the 2002-2017 period multiple major alternative discourses were present and in flux, and some rose to dominate the literature and conversations, notably openness determinism, eLearning innovation and commercial threats to institutions and practitioners.Synonyms for social justice such as "access", "democratisation of education" and "empowerment" seemed to appear as red-herrings, potentially providing an impression that social justice ideas were being pursued.
As a theme in the literature, social justice faded, particularly as the field broadened and came to more closely resemble mainstream eLearning.However, it was subsequently re-asserted by influential authors -UNESCO and Watters, the latter in the light of the failure of dominant discourses to provide for more equitable learning.Towards the end of the period of analysis it was too late to be still claiming "potential" of OER, a number of more critical views had coalesced in the light of published evidence of learner outcomes, notably demographic inequality of MOOC access.

Comparison with Recent Published Views
Recently, prominent researchers working with a long-term viewpoint from within regional distance or Open Universities have also begun to note the way the advances in digital openness coincided with a move away from a more inclusive and widening participation stance (Tait, 2018;Weller et al., 2018).While Weller and colleague's 2018 citation analysis does not discuss the technical deterministic bent of current literature, it does, however, note the way that recent research occurred without recourse to the rich body of earlier research that emerged from open and distance learning.For example, they note with regard to emerging shortcomings of MOOCs, that the literature on "supporting students at a distance (e.g., Tait, 2004), e-learning costs (e.g., Bates, 1995;Weller, 2004), or student retention (e.g., Tinto, 1975) may well have provided useful contributions to this development, but was largely ignored."Rolfe's work also showed the tendency for positive, uncritical bias in recent work and very limited drawing on foundational theorising from the Open Education research of the 1970s (Rolfe, 2016).The present data and analysis backs-up and extends these observations to suggest that the dominant themes of contemporary literature not only missed out on the earlier insights but also took the discourse down a technological determinist pathway that requires a concerted changing of course to avoid a recurring lack of impact for diverse learners.
Of the 20 papers used to seed Weller et al's citation analysis, there are eight papers that fall within the "broadening" and "appropriation" phases of the current analysis, i.e., 2009-2017(Weller et al., 2018)).
A brief review of the themes of these papers was conducted as a comparison with the current findings and found to align to the dominant/alternative types of literature identified here.In the Weller et al sample, two of the three thematic clusters within those eight papers covered similar terrain (broadening towards eLearning, and focus on educator rather than student freedom or justice) while the third theme presents a somewhat different but sympathetic type of alternative narrative.
For example, three of the eight papers that cover similar ground argue for a broadening to reconsider the definition of openness within the context of "Web 2.0" (Friesen & Murray, 2013), the rise of informal learning in "a connected world" (McAndrew, 2010) alternatively described as learning "beyond the course" (Dalsgaard & Thestrup, 2015).Two texts also look at the freedoms of educators and the value of higher education (Weller, 2014) alternatively expressed as the impact of openness on "positive liberty in the enactment of academic practices" (Oliver, 2015).While the present analysis has made the point that such broadening accounts tend to unintentionally downplay social justice for diverse learners while focussing on other important commercial and institutional moves in the field, there were also four alternative historical overviews which sought to bring learner-centred approaches to open and distance learning back to the fore.These can be read as offering longer view historical accounts of Open Education as reduction of barriers for disadvantaged learners (Friesen & Murray, 2013;Longstaff, 2014;Peter & Deimann, 2013;Weller, 2014).
Longstaff's (2014) account has an interesting "take" on the fluxes and flows noted in this paper between dominant and alternative narratives.Through investigating the development of universities over time, including throughout the first few years of MOOCs, she finds, "a cyclical model of change, one in which waves of inclusivity alternate with bouts of exclusivity" in line with complex influences within education and society.
Perhaps the 2002-2017 more technically influenced Open Education literature will be viewed with the further passing of time as a more commercially-focussed period in between more learner and socialjustice focussed educational phases.The social-justice focussed new definition for Open Education offered in the next section aims to provide practitioner-researchers with a clearer research pathway towards a more social-justice-oriented future.Certainly the field of Open and Distance Education continues to develop and currently research is pursuing both humanist and post-humanist approaches to the integration of "bots" and innovative semantic technologies to increase support for students without sacrificing quality of experience (Bozkurt, Kilgore, & Crosslin, 2017;Knox, 2015;Santamaría Lancho, Hernández, Sánchez-Elvira Paniagua, Luzón Encabo, & de Jorge-Botana, 2018).
Interestingly, in the last few years issues of recognitive justice and representational justice have also been debated publically with regard to who has a right to be included and to speak at Open Education conferences.There has been criticism and rejection of "manels" (male only panels) and the underrepresentation of experts of colour as keynote speakers, particularly those from the Global South who are highly active Open Education participants.It seems timely, then, to also apply these social justice principles to the experience of our students and their learning environments.

Recentering a Social-justice Purpose and Definition of Open Education
To enable interested practitioner/researchers of Open Education to work more effectively towards more equitable forms of education, adoption of a definition of Open Education that is centred on social-justice principles is proposed.A more narrowly focussed and distinctive definition would also guard against further broadening, to ensure Open Education remains distinctive from eLearning.
A social-justice oriented definition would be useful then to shift the debate from what openness might look like, to whom we want our openness to ultimately serve and how our openness might achieve greater educational and societal equality.As Edwards notes, "An important question therefore becomes not simply whether education is more or less open, but what forms of openness are worthwhile and for whom; openness alone is not an educational virtue" (Edwards, 2015, p. 253).
Following this, there needs to be intention for reducing inequality -both in program design and research.Successful designs are more likely to be founded on an understanding of which communities and cohorts in our contexts are more and less privileged, and of ensuring that access, support and services are provided to them and that their progress as compared to their more privileged peers is always tracked.Such work could take an affirmative action approach, where organisations put in place additional resources to help minority groups overcome historical injustice and reach their full potential as learners.Considering the success of affirmative action policies in both the labour market and for university admissions (the effect on the latter being more for the more elite schools,) (Holzer & Neumark, 2006) this seems fruitful areas for future research and practice.

Proposed Definition
The following definition of Open Education is proposed as primarily about social justice, while still allowing space for secondary benefits by other learners: Open Education is the development of free digitally enabled learning materials and experiences primarily by and for the benefit and empowerment of non-privileged learners who may be under-represented in education systems or marginalised in their global context.Success of social justice aligned programs can be measured not by any particular technical feature or format, but instead by the extent to which they enact redistributive justice, recognitive justice and/or representational justice.
The inclusion of the phrase "by and for… non-privileged learners" maintains the original intention of the 2002 OER definition regarding active participation by developing countries and the marginalised -rather than neo-colonial practices of the Global North doing things to and for those they consider disadvantaged.
With such a definition in place, other related definitions such as OER may remain unchanged to denote different foci within the field.For example, the definition of OEP could remain more broadly about the processes of collaboration and sharing in a wide range of educational practices to improve pedagogy for all learners.The term "critical open pedagogy" could continue to be used to identify the set of intentionally empowering OEP which seek to shift the power balance between learner and teacher as a particular strategy to reduce inequality (DeRosa & Robinson, 2017).

Conclusion
This paper has tracked a sample of texts which shaped the definitions and practice of Open Education since 2002 through a series of broadening moves from OER to OEP and through various "continuums" and "dimensions." It has shown the points at which claims for the importance of social justice purposes became hidden in the larger documents but were absent from the most regularly used definitions.It has shown how a broadening of scope overlapped with more mainstream educational, eLearning/and Distance Education debates about quality and pedagogy, at the expense of discourse on social justice purpose.
A major discourse in the debates about the power of openness has been labelled "openness determinism" for the way it has inadvertently reinforced technological determinist ideas -that somehow openness will democratise education, as technology itself was expected to do previously.
This paper has argued for a viable alternative account of how Open Education can begin to shift educational inequality by focussing on one or more of the three principles of social justiceredistributive, recognitive and representational -which, via the 2002 UNESCO OER Declaration, launched the modern, digital OER movement in the first place.
The social-justice aligned definition of Open Education proposed here offers new opportunities for designs to be shaped as explicit social justice actions aligned to one or more of the three principles.It also offers the opportunity for new empirical research to measure the social justice impact of initiatives in terms of the way that learners who, by circumstance, have less are able to be provided with more resources, recognition or representation.It also suggests empirical research approaches attuned to demographics of privilege, so that access, progress and success rates can be investigated in both the more and less privileged cohorts in our educational systems.
sample texts in each phase suggest a reasonably similar sample size to adequately represent the periods.The first published literature review of OER -while published in 2014 -is included in the foundational digital texts phase as it covers literature from 2002-2013 and acts as a consolidation of the definitional issues of that period.

Social Justice hidden in context. Alternative theme -access to free materials online for everybody:
While the whole text is strongly aligned with redistributive and representational justice, with a lesser emphasis on recognitive justice, the often-cited shorter definition sentence contains none of the context and social justice principles for improving educational access in developing countries, but are about putting free materials online for everybody.At this level of granularity regarding legal and technical features of a digital resource, the "4/5Rs framework" or definition lacks any of the keywords or principles of social justice.Social justice is not present as either a goal or a process/action in the texts, even though redistributive justice regarding text-book costs for marginalised learners was a significant part of Wiley's application of the work in the American College sector.
Alternative theme -openness determinism (affordance); access to free materials online for everybody: the text discusses overcoming technical and legal hurdles; three of the "4Rs" (rework, remix, redistribute) are relevant only if you have technical skills and infrastructure; the text can be read as a conversation between IT workers promoting re-use affordance.Increasing educator re-use of openly licenced materials is assumed to lead to improved/changed educational access.4.The Cape Town Open EducationDeclaration. (2007)Social Justice principles weakly implied: The principles of redistributive, recognitive and representational justice are only implied, are extremely watered down.

Table 4 : Data Summary of Findings from Eight Broadening Phase Texts 2009 -2017
Social Justice is absent: Social justice principles are not mentioned in this paper.Alternative theme -eLearning innovation for everyone; staff and institutional benefits: Definition shows the broadening of the field towards mainstream higher education, benefits are noted as towards institutions first, higher education staff second, with student benefits noted third.10.Siemens' chapter contribution in bookMOOCs and Open Education around the world (2015)Social Justice is absent: Social justice principles are not mentioned in this paper.Alternative theme -eLearning innovation for everyone: Reflects the ideas in OEP regarding digital literacy and lifelong learning are the alpha trend in digital learning and the more important phenomena than MOOCs.Can be read as blurring the boundaries between openeducation and eLearning or online learning.11.Butcher's report for the Commonwealth of Learning (2015) A Basic Guide to Open Educational Resources (OER).Limited, partial redistributive justice approach.Provides a new definition for Open Education, which is about removing barriers to learning and addressing assessment and accreditation -these can be viewed as parts of redistributive justice.Alternative theme -Widening participation: access, supported progress, and success of disadvantaged learners.Works against the discourse that "access" alone will produce fairer education.12. 10 dimensions of Open Education in EU Science project report from dos Santos, A. I., Punie, Y., & Muñoz, J. C. (2016).

justice is absent or weakly implied:
The use of OER for social justice is not discussed -perhaps not considered, or is it just assumed that OER are social justice actions and commercial resources are not?

Social justice principles absent. Alternative theme -commercial threat to practitioners and field:
What happens when something is "open" in all the ways that Open Education and open source and open data advocates would approve.All the right open licenses...All the right nods from all the right powerful players within "open."And yet, the project is still not equitable.What if, in fact, it's making it worse?What are we to do when we recognize that "open" is not enough?I hope, that we recognize that what we need is social justice.We need politics, not simply a license.We need politics, not simply technology solutions.We need an ethics of care, of justice, not simply assume that "open" does the work of those for us