Talking about rights without talking about rights: on the absence of knowledge in classroom discussions

This article reports on research in three secondary schools in England where students were engaged in deliberative discussion of controversial issues. The teaching resources used illustrated rights-based dilemmas and the data analysis focused on the nature of the talk and the types of knowledge the students drew upon to inform their discussions. The article shares four insights: (i) there is a need to be more explicit about what constitutes human rights knowledge; (ii) human rights education requires the development of political understanding, which moves beyond individual empathy; (iii) educators need to value the process of deliberative discussions and avoid a push for conclusive answers; (iv) students need support to draw on knowledge from a range of disciplines. If these issues are not addressed, some students are able to engage in rights-based discussions with little knowledge and understanding of rights.


Introduction
The starting point for this article is Parker's challenge to human rights educators (in the first edition of this journal) that we need to be more explicit about what knowledge might sit at the heart of a curriculum for human rights education (HRE) (Parker, 2018). He argued that HRE frameworks often leave this issue relatively open and unexplored. Whilst learning 'about' human rights is one of the three defining elements of HRE according to the United Nations (UN) Declaration on Human Rights Education and Training (the others being learning 'through' and 'for' rights), it is not clear what we should expect to be sufficient knowledge, nor indeed what an informed student should understand about human rights.
Parker's discussion focuses on a number of issues but crucially observes there has been a kind of 'knowledge blindness' within HRE, where advocates tend to avoid engaging with questions about human rights knowledge. He builds on Vygotsky's distinction between the everyday knowledge that arises from reflection on experience, and more conceptual scientific knowledge, which has to be more consciously taught (Karpov, 2003). These scientific concepts unlock new ways to understand the world and therefore have significant transformative potential. Despite curriculum theorists' relative neglect of such issues, the importance of knowledge and understanding has been described by writers as different as Freire (1996) (in his dictum that we should learn to read the word and the world) and Young and Muller (2016) (in their recent controversial work on 'powerful knowledge'). Whilst these authors adopt very different perspectives on the curriculum, schools and pedagogy, they draw attention to the role of knowledge for unlocking new conceptual connections and new understandings about how the world works.
In this article, we are concerned with what is meant by human rights knowledge and what constitutes a higher level of understanding of human rights. As Parker argues, without answers to these basic questions, it is of little surprise that we continue to struggle to articulate what HRE might offer, and to frame HRE meaningfully when we do have the opportunity to embed it in curricula (Jerome, 2018). Rather than engage with this debate as an abstract theoretical challenge, we use it to frame our reflections on empirical data we have generated in classrooms. By reflecting on how secondary school students (aged 12-18) talked about human rights in the classroom, we seek to illuminate what constitutes an informed understanding, and reflect on some of the obstacles.

The Deliberative Classroom project as HRE
The data were generated during a project called the Deliberative Classroom (Jerome, Liddle &Young, 2020). This was framed in the policy context of the English education system, where schools must 'promote the fundamental British values' (FBVs); these are defined in statutory guidance (Home Office, 2019) as democracy, the rule of law, equality and tolerance. It has already been documented that calling these concepts 'fundamentally British' and 'values' may well have distorting effects, as it tends to deemphasise the extent to which they could be taught as contested political concepts and encourages teachers to focus on a vague notion of 'Britishness' (Busher & Jerome, 2020). By contrast, the Deliberative Classroom aimed to teach the FBVs as complex concepts requiring discussion, evaluation and application in different contexts. One of the resource packs (Association for Citizenship Teaching [ACT], 2018) focused on freedom and toleration through exploring the nature of the right to freedom of religion, and the circumstances under which the right could be limited. Another pack focused on the rule of law and democracy, and considered how one might seek to justify breaking the law for political ends. Human rights provided the foundation for the project and students were encouraged to think about how rights are defined, balanced, restricted and enacted in various circumstances.
Pedagogically, the project built on two main ideas. First, it aimed to engage the students in discussions incorporating some of the principles of deliberative democracy (Dryzek, 2002;Habermas, 1996). This includes a commitment that coming together for open, honest and respectful conversations across our differences can lead to better, more inclusive outcomes. Whether or not it leads to a consensus, the process of deliberation is transformational in that it changes the way individuals perceive their own position, and that of others, and may well encourage people to review their beliefs to some extent. Deliberative fora typically involve learning phases, where participants deepen their understanding of an issue, often in dialogue with experts. But, as with HRE, some commentators have criticised deliberative democratic theorists for failing to develop an adequate account of the role and nature of knowledge (Walter, 2008;H. Young, 2017). Our research complements the approach adopted by Al-Daraweesh and Snauwaert (2018), who have begun to explore how deliberative pedagogy might contribute to HRE knowledge. Whilst they provide an illustrative sequence of classroom activities where students deliberate human rights principles, our work adapts a case study approach to consider human rights in specific contexts, as described below.
The second pedagogic influence was Hess's (2009) work on teaching controversial issues. Again, this was influenced by the policy context, where schools are both required to promote the FBVs and induct students into controversial discussions, including those about terrorism and extremism (Department for Education, 2015). Hess argues that teachers have to think about controversial issues on at least two levels. At the most immediate level, there is the specific case which has been identified as controversial, such as gay marriage, outlawing religious sects, abortion, online hate speech, etc. But lurking beneath the surface is a 'perennial issue', which generally resembles a Vygotskyan scientific concept, i.e. an abstract idea which requires interpretation and application (e.g. freedom, equality or minority rights). The Deliberative Classroom aimed to set up discussions about specific case studies in order to engage students with the perennial issues that underpinned them. The resources were written to engage with the FBVs as contested perennial issues and to address the human rights dimension to these issues, such as freedom of religion, freedom of speech, and freedom to organise. Some HRE advocates have long recognised the value of case studies as a means to explore multiple perspectives and counter-narratives, and to develop empathy as the basis for action (Adami, 2004;Osler, 2016: Ch.4;Osler & Zhu, 2011).
These two ideas of deliberation and teaching controversial issues informed the design of the project, but it was a genuinely open question as to what the students would do with the resources, and what kinds of conversations would be generated. One source of uncertainty stemmed from the fact that the resources encouraged the small group discussion which Pace (2019), in her small-scale study, observed as typical in the approach adopted by English educators, an approach that left students free to veer off task. The resources reflected McAvoy and Hess' (2013) recommendations that resources should use genuinely controversial issues (open questions), distinguish between the case and the underlying issue, embrace ideological diversity and avoid prescribing a correct opinion. This meant students had considerable freedom to lead the conversation in different directions. Mercer (1995) describes talk between teachers and students as the 'guided construction of knowledge' and argues that, through conversations in class, 'knowledge is neither accumulated nor discovered by learners, it is shaped by people's communicative actions' (Mercer, 1995, p.19). Mercer contends that a key role for teachers is to induct students into new discourses (in the same way that Parker [2018] argues students need to be inducted into a human rights discourse). This reflects Vygotsky's conception of scientific concepts which have to be consciously taught and acquired outside of normal everyday experience but, once acquired, can be used to transform the way one understands the everyday world (Karpov, 2003). Whilst students will gain some level of understanding of what constitutes rights from everyday life and conversations, a deep understanding will draw on political theory and law, and will include understanding what it means to claim and recognise a right (Jerome, 2018). This involves learning some specialist political knowledge, such as how rights are enacted and how those processes can be influenced (Barton, 2020). It may also involve deliberately teaching about different contexts, from the local to the national and international, in order to help students think consistently about the enactment of human rights principles (Kim, 2019). This process may be made more compelling if learners encounter detailed authentic accounts of other people's lives (Barton & Ho, 2020). These forms of knowledge tend to focus on additional factual information (e.g. how political processes work, how other people live, how political systems differ), but there is also a deeper Vygotskian sense in which students should also acquire concepts which form a collective cultural tool.
The curriculum does not consist simply of subject knowledge of a factual kind, but embodies ways of using language-discourses-which students need to be enabled to understand and use if they are to become educated. (Mercer, 1995, p.85) Teaching about human rights should attend therefore to the acquisition of factual and conceptual knowledge, the development of competence in using that knowledge to interpret different situations, an awareness of one's own political agency in the broader development of rights and opportunities to tackle misconceptions (Jerome, Emerson & Lundy 2015).
In reflecting on how dialogue is supposed to facilitate learning, Reznitskaya and Gregory argue there are three processes: (i) Discussion gives students experience of rational thinking, which they can then internalise. Here they speculate that individual reasoning is a kind of internal argument with a generalised other, and that our thought is shaped through struggle with others' thoughts (Reznitskaya & Gregory, 2013, p.118).
(ii) The cumulative experiences of such discussion help students build an 'argument schema' for a general understanding of how arguments are constructed and questioned, which enables students to judge between different interpretations. (iii) Students develop an increasingly complex understanding of disciplinary knowledge, i.e. what it means to understand and think with concepts like rights.
These general observations about the nature of learning through argument also resonate with Al-Daraweesh and Snauwaert (2018), who suggest that deliberation can both develop knowledge about human rights and a deeper grasp 'of the forms of thought characteristic of the discipline', including the structure of moral and political discourse that underpins human rights (p. 86). Reznitskaya and Gregory (2013) also argue that teachers may unconsciously promote pseudo-enquiries by simply accepting all student contributions as valid, rather than testing and evaluating them appropriately. This alerts us to the observation that just because valuable learning can derive from classroom talk, we should not assume that all classroom talk leads to valuable learning. One way in which learning might be limited results from teachers not being clear about the purposes of such talk (Mercer, 1995). Another problem might be seen where students 'do not advance their understanding ... they merely reiterate halfunderstandings which they already possess' (Barnes & Todd, 1978, p.73 cited in Mercer, 1995. This may happen when students never fully articulate and explain their positions and therefore fail to make themselves accountable to the group for what they say, which can lead to a 'set of unexamined platitudes'. A third problem arises when students simply perform the role the teacher wants to see and rehearse the kinds of lines the teacher wants to hear as a form of classroom ventriloquism (Pace, 2015).
In addition to being alert to how classroom talk might go awry, we are also conscious that there are additional problems that may prevent students making progress in their learning about human rights. As Parker (2018) suggests, students may have already learned to be dismissive of rights, or to think in simplistic and absolutist ways about them. They may have been taught that rights are directly dependent on responsibilities, and therefore may be conditional and easily forfeited (Howe & Covell, 2010). They may simply fail to see rights as relevant to the problems being discussed, and solve those problems using other reference points (Jerome & Lalor, 2015). Or they may see rights as largely relating to other people in other parts of the world and reframe rights problems as cases for charity or volunteerism . They may have a restricted view of rights as moral values (Kim, 2019) and/or a set of international declarations and documents (Keet, 2015), and fail to understand the ways in which rights feature within political struggle (Barton, 2020;Jerome & Lalor, 2015).

Research approach
As noted above, the research project generated data from classroom discussions that resulted from students engaging with the Deliberative Classroom resources. Schools were approached in a variety of ways, including a general email through the Association for Citizenship Teaching invitations to teachers with whom the researchers had worked on similar projects, and personal approaches to other schools. The intention was to record the classrooms where discussions based on the themes in the resource packs (ACT, 2018) were taking place. This paper draws on data generated in three of the schools.
All three schools focused on the topic of religious freedom, hence our analysis discusses this concept in greater depth. The resources in this pack started with an historical account of the development of religious freedom in the UK. This clarified how religious freedom and toleration have developed over time, and demonstrated how contemporary assumptions about religious equality have only relatively recently become enshrined in law. The second activity included several case studies where religious freedom had to be interpreted and restricted in practice. These case studies included the following issues: • Gay marriage and the Church of England as an employer (and the Church's exemption from aspects of UK equalities legislation).

•
The UK welfare system's rules on family and marriage in relation to polygamous relationships (and under what circumstances the state should acknowledge such relationships).

•
The UK's new £5 banknotes (which include small quantities of beef fat), and whether the state has a responsibility to respect the beliefs of vegans and Hindus.

•
How and when cults should be accepted, and when they should be viewed with suspicion and restricted.
Each case study included specific questions to focus discussion on the issue and then built up to a common question: 'How does this case study help you think about the limits on religious freedom?' The final activities included a variety of suggested debates and discussions, either to debate a single issue or case (e.g. everyone has a right to wear religious symbols or dress to work) or to draft a school resolution, setting out principles for the school to follow and encouraging students to debate, amend and vote on each element to maximise consensus. It should be noted that one aim of this research was to explore how the resources were received in schools by students and teachers, and to evaluate to what extent they scaffolded the discussion. By recording the sessions, generating transcripts, and identifying a variety of concepts to analyse the data, we are provided with what Mason describes as 'flashes of insight' (2017, p. 44) from the student talk.

Data collection
As mentioned above, the data was collected in three schools. In two of these, the lessons were with year 8 pupils (aged 12-13) in Religious Education (RE) classes. In the third, the session was run with the school's debating society. Although there is a national curriculum in England, schools have great flexibility in what they teach in history, citizenship and RE, which makes the prediction of prior knowledge particularly difficult. RE is influenced by local frameworks and whether the school is designated as a faith school. In addition, academies (two of our schools) are exempt from following the national curriculum, which has particularly adversely affected citizenship education (Hahn, 2020, p. 252). We are aware that as this is a small-scale study we are not able to generalise widely. However, with the classroom materials freely available, the study is replicable. The three schools are as follows:

Arun Grammar
This academically selective boys' academy has very few children with special educational needs (SEN), and very few receive free school meals (FSM) (FSM is widely used as a proxy indicator of disadvantage/poverty). A third of the students speak English as an additional language (EAL). The material on religious freedom was taught by an RE specialist teacher (with substantial experience of citizenship teaching), to year 8 (aged 12-13) boys. The researchers attended two lessons but only recorded discussions during the second visit, when the teacher included extended discussions.

Stour Grammar
This academically selective girls' school has lower than average numbers of children with SEN, FSM and EAL. The material on religious freedom was taught by an RE specialist teacher (who also had extensive experience of citizenship teaching) with year 8 girls. The researchers attended two lessons and generated data in both. The students sat around 15 small tables. This made it impossible to have a recorder on every table, so five small groups were recorded, plus the whole class discussions.

Avon School
This comprehensive Church of England academy has average numbers of children with SEN and FSM but lower than average EAL. It has roughly equal numbers of boys and girls and for this project a mixed-gender group of 14 student volunteers (12-18 years old) from the debating society agreed to participate in one deliberative discussion on religious freedom in school (the Christian nature of the school does not mean only Christians attend, many students are Muslims). This was facilitated by two of the researchers, and small group and plenary discussions were recorded. The session was supported by the teacher who ran the society.
In each school, recorders were placed on tables in the classroom if the pupils consented. Recordings of small group discussions and whole class plenaries were transcribed, but in cases where there was a lot of off-task talk (in some of the small groups), transcriptions focused only on the sections of the recordings which were relevant to the tasks.

Analysis of the data and selection of themes
Two of the researchers applied an initial coding system on one transcript each in Nvivo, then met to review each other's work and discuss issues arising. This led to a revised set of codes, which were then applied to the whole data set and further moderated. The coding focused initially on classifying individual utterances and then on longer sections of conversations, where utterances are related to one another in relation to a topic or argument (Mercer, 1995). Full details on the coding system used and how these methods build on traditions in the field of classroom talk are in the project report (Jerome, Liddle & Young, 2020). For this article, we have focused on those themes that seem to be related in some way with our concern for clarifying the role of knowledge in HRE discussions

Discussion of data
A number of key themes around HRE knowledge emerged from our analysis. First, we consider whether starting with students' own emotional responses might inadvertently restrict their capacity for engaging with other perspectives. Second, we provide some examples that illustrate a tendency to cut short discussions and prevent students from deepening their understanding. Third, we argue that the consensus-seeking behaviour inherent in deliberative discussions may lead to a form of cumulative talk, which discourages students from adopting a more critical perspective. It is this criticality that might be expected to lead to deeper engagement with underlying concepts. Fourth, we noted very few examples of students drawing on knowledge gained elsewhere in the curriculum.

Opinions and restricted empathy
In this section we consider whether the tendency to start with personal responses, which seems to make pedagogical sense, might ultimately restrict students' ability to frame the issue from different perspectives. We suggest this may both limit their capacity for genuine engagement with the other (Barton & Ho, 2020;Osler, 2016) because they essentially project their own emotional response onto others. This in turn could limit their political understanding of the case(s) being discussed because conflicting points of view are sidestepped. This presents a challenge for inducting students into the human rights discourses discussed above.
Deliberative democratic theorists have noted several relevant issues. First, in situations where the deliberating group is not diverse, this can lead to 'limited argument pools' where participants cluster around a shared perspective (Sunstein, 2000, p. 75). Second, there is a debate about the balance between the rational and affective dimensions to deliberation. Habermas (1996) tends to focus on the need to articulate and exchange reasons for one's view, but there have been many critiques of this emphasis (I. Young, 2002). Barton and Ho (2020) point out that teachers should acknowledge that emotions are not simply problems to be tamed or managed in the classroom, but are actually integrally bound up with the positions we take and the reasons which we find compelling. Whilst they suggest emotional responses, such as compassion, can be fundamental to developing regard for the other, we note here that the combination of limited argument pools and emotional empathy may also lead to problems.
In a small group discussion in Stour Grammar, focusing on polygamy, one student reported: Speaker 1: I watched a TV programme about that once, and he had like three wives, and they didn't like each other because they were jealous of the other wife, because he went out to dinner with one of the wives, and she was like, but that's my husband. [Speaker 2: but that's weird]. Yes, that is weird, and he had three kids with one of them, and two with the other one, and then the other one was pregnant.
The students went on to discuss how the children would feel in this situation, before the conversation turned back to the case study on the card, and then another participant introduced a second example from the media: Speaker 3: I read this thing the other day, it was 'I went to my best friend's wedding and I fell in love with her and now we're in a three-way marriage'.
During this discussion, one of the researchers joined the group and tried to introduce an argument that one might consider polygamy from a different perspective (imagining how it might be seen in a different culture), but the students sidestepped the point, and returned to the media coverage as evidence of how women actually experience polygamous marriages.
Speaker 1: I feel if I was in that position having a husband with two other wives, I'd just think... Speaker 2: I think even if I was brought up in that religion, I don't think I would… Speaker 1: I watched a documentary on it, as I've said like five times, but the women in it, they said they didn't really… it wasn't that, it was like they didn't really want to say anything about it. Speaker 2: It was like all the women were jealous of the other women [Speaker 1: yeah] when they went out on dates and it was like they were trapped in it.
In this example the documentary fills an experience gap. Whilst the researcher is providing a general, relatively speculative argument about how the construct of marriage might be perceived, the students are keen to return to the concrete example provided by the programme they have watched. However, the main purpose of the programme seemed to be to bolster their individual emotional responses. Here the participants couch the discussion in terms of how they would feel in a situation and focus on the emotional responses of the people in the programme. Whilst this seems like a reasonable way to engage in the discussion, it is one-dimensional. Notably, this framing entirely ignores the context of the case study, which had been carefully written to frame the issue in terms of religious diversity, migration to the UK, and the fact that the state's welfare rules have acknowledged polygamy but its marriage rules have not. Rowe's (2005) work on young people's political thinking similarly involved students discussing case studies in class. He argued that schema theory helps to describe the development of political literacy, as students learn to adopt a variety of different schemas (or ways of framing the problem) in order to understand the complexity of political phenomena and decision making. Here, for example, the question can be viewed from an individual emotional perspective, which may include notions of romantic love, and may also include other emotional aspects, such as jealousy, bullying, etc. However, additional schemas might include a religious perspective, exploring which religions, and under what conditions, acknowledge polygamy. One might also adopt a feminist perspective, inviting questions about the gender inequalities within those relationships and in the societies that acknowledge polygamous marriage. One could also adopt an historical perspective, to think about the roots of such practices and how changing social norms affect one's judgement. This case study was also framed from a bureaucratic perspective, to consider how the state inevitably has to take a position on the issue, because people engage with it through tax and welfare systems where marriage is important. In this discussion, the group's schema is limited to the emotional framing, and even here the participants appear to generalise from their own emotional response to the emotional response of all others, thus erasing possible cultural or religious differences. As such, this represents a restricted sense of empathy, in which the participants seem to project their emotional responses onto others, in order to feel sorry for them. The more rigorous form of cognitive empathy required to appreciate diversity seems largely absent (Endacott & Brooks, 2013). This effectively means the case study is depoliticised; students pitch their discussion at an emotional and individual level, and reflect a level of thinking which might be described as 'pre-political' (Jerome & Lalor, 2015). This restricts the extent to which students apply a human rights framework; they paint all women in polygamous relationship as victims who need to be helped, rather than engaging with the legal challenge represented by different perceptions of what constitutes the right to a family life.

Task completion and 'certainty'
The distinction between deep and surface learning is a common one, although it has been criticised for a lack of clarity (Howie & Bagnall, 2013). For our purposes, it reflects the distinction made in Hess's (2009) work that students move beyond the specific case towards the deeper perennial issue. This also resonates with Shulman's (1986) work on the acquisition of underlying concepts and processes that define an area of knowledge. In this section we consider a tendency in some classes that limits the students' pursuit of greater depth, leading them to settle for a relatively superficial level of engagement.
Students commonly announced, in a rather arbitrary manner, that they had completed a task and were ready to move to the next. Obviously, these are all openended controversial issues that can never be fully 'completed' in the sense of unpicking all the possible principles. None of this was surprising to us, but in this context we became very aware of the relationship between task completion and students' engagement with 'deeper' knowledge.
Culturally, schools in England are very much structured around task completion, reflecting pressures associated with high-stakes testing (Ball, 2017) and the school day being divided into very structured time sections (Lingard & Thompson, 2017). From a simple behaviourist point of view, task completion is the way to gain teacher praise. This has implications for open-ended deliberation, as is illustrated in the following extract from a small group discussion in Stour Grammar, relating to case studies on cults and on gay marriage. After an extended conversation, which began to identify different perspectives and possible tensions, the conversation was brought to a halt: Completing the task in this arbitrary fashion narrows explorations and obscures possibilities for ambiguity and uncertainty. Dryzek (2002) says deliberation can 'multiply dimensions and options' (p. 41), but this can only occur when deliberation is extensive and unconstrained by a need to complete tasks. The student exchanges demonstrated their ability to open up the issue as they engaged with the fact that there may be rights in tension, that there is a distinction to be drawn between freedoms to (such as joining a cult) and freedom from (harm), and that we should think cautiously about whether to advise free citizens or actually restrict their freedoms. The discussion appears to be moving in the direction of Mill's dictum 'that the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of action of any of their number, is self-protection' (Mill, 1859, pp. 21-22). This reflects precisely the intention behind the resources, but the students did not sustain their enquiry in order to pursue these initial thoughts. It seems as though the task is judged as 'done' when the students feel they have expended sufficient energy for a short task. Earlier we noted the possibility of 'pseudo-enquiries' and this exchange seems to exemplify this. At the least, it indicates a very limited enquiry.
Another discussion illustrates the same point. In the following small group discussion in the same lesson, students unpacked the issue of polygamy much more thoroughly than the example cited above (in 'Opinions and restricted empathy'). Having drawn a distinction between polygamy and polyamory, discussed the nature of secular government in diverse societies, considered attitudinal change over time and touched on the sexual politics of marriage, the conversation is suddenly closed down: Speaker 2: Decades ago you could have said the same thing about gay people, you know like when people might have said you can't explain gay people to children, that was their perception of society. Speaker 1: Yes, that is very true. Speaker 2: It will be interesting to see what the views on this will be in 30 years when… Speaker 3: When we're older, because it's definitely more common now.
Speaker 2: Exactly, and less frowned upon I think. Speaker 1: Shall we go to case study 3, what's this, Hindu complaints about tallow in the new £5 notes [reads information]. Reznitskaya and Gregory (2013) argue that deliberation will help to shift students from simplistic single narratives about socio-political issues towards an understanding that there are multiple perspectives and explanations which can be evaluated against the evidence and the standards of rational argument. In other words, deliberative discussion should enable students to develop their epistemological understanding of social issues such as human rights. In this section we consider whether the search for a consensus might restrict the extent to which students engage in sustained critical argument.

Consensus
Deliberative democracy often emphasises consensus, and in our research we were struck by how the assumption of group discussions was that they should come to a consensus to report to the teacher or to us, rather than that they should reflect on the complexity of the issue. This was closely related to the issue of task completion and an unwillingness to stay with the uncertainty (explored above). An emerging consensus can create strong social pressure for others to agree and so closes down possibilities for further exploration of different perspectives. As a consequence, some deliberative democrats, such as I. Young (2002), challenge the focus on consensus and emphasise the importance of acknowledging and living with conflict and disagreement: Too strong a commitment to consensus as a common good can incline some or all to advocate removing difficult issues from discussion for the sake of agreement and preservation of the common good. (I. Young, 2002, p. 44) In the classroom context, Mercer (1995) notes that talk quite often tends towards an uncritical consensus, as group members are interested in demonstrating recognition of others and maintaining positive social relations. Critiquing one another, testing arguments, asking for evidence or seeking clarification can be perceived as an aggressive or disruptive challenge, especially when working with friends.
This was reflected in our data, in conversational chains coded as 'cumulative' talk. The following extract from a small group discussion in Stour Grammar illustrates this. The group began with a case study of gay marriage and the Church, and then moved into a more general discussion of attitudes towards LGBT+ relationships.
Speaker 2: Basically, in this day and age, love is love. … Speaker 1: Our grandparents for example were born in a time when being gay was seen as being something bad [Speaker 2: a crime]. My grandma, I went to her house and it was this time last year and like there were loads of shows on Pride, and she was like, turn it off I don't want to watch it, because that's how they've been brought up. It's nothing, it's not, they don't mean to be discriminatory. Speaker 3: It's just how they think.
Speaker 1: That's how they were brought up, they have not been shown a different world. Speaker 3: Also now our grandparents they're like 70, 80 [Speaker 2: They're fixed in their ways] yes, they're not going to change are they? They're not going to change.
Here the students simply build on previous statements to construct a shared worldview. They agree with one another, they extend earlier points, they interject to confirm their agreement, and they finish off one another's thoughts. This may all genuinely reflect their understanding, but ultimately it does nothing to extend their understanding of the political process of liberation, and it seems to reflect a rather uncritical perception that rights simply come with the progressive passage of time. This marginalises the role of LGBT+ activists who struggled over generations to achieve these changes in the UK, and who continue to do so here and elsewhere. The easy consensus the students reached also precluded them from exploring the tensions that were illustrated in the case study material, where the Church of England is exempted from the Equalities Act so that religious beliefs are not compromised by LGBT+ rights.

Sources of knowledge
Students tended to draw primarily on their own experiences and on the vicarious individual experiences of others (drawn from friends and family or the media). Such experiences seemed to serve as additional case studies or examples to help reinforce a point. For example, in one discussion a student went to some length to describe her mother's choice to be a more traditional 'stay at home mum'. Even though this seemed at first to be irrelevant to the conversation, which was actually about polygamy, she used the anecdote to make a point about how easily outsiders can misinterpret aspects of a relationship. By contrast, it was relatively rare for students to draw on knowledge from the school subjects they were studying.
We noted just two exceptions where students did draw on knowledge from elsewhere in the curriculum. At Stour Gramma a student brought in previous knowledge from history: Well when it was Tudor times there was conflict between Protestants and Catholics and like the Church of England and there was more of a gap between those before other religions were brought into the equation. So they seemed very different until compared with something else.
Here her background knowledge helped to generate some insight into how one might account for the pace of change in the legal framework for religious freedom. In Avon School, the debating society students came from different year groups and therefore had different types of curriculum knowledge to share. Some of these students were taking A-level subjects (the advanced courses generally studied by 16-18 year olds), and were able to share relevant insights from sociology and law to discuss different ways to think about how to define and deal with offensive speech in the context of respecting religious diversity. In these examples the students drew on relevant prior learning to expand their discussions. However, these few examples represent all the instances in our data and so the most significant finding here is how rarely students drew on what they had already learned in school.

Concluding remarks on knowledge and HRE
We have shared these findings and our discussion of them in the spirit of open and honest reflection on our HRE practice. Our data includes other examples of conversations which seemed to be more successful in some regards, but by focusing on instances where problems arose, we have attempted to sharpen up our critique of the project and its theoretical foundations. From such a small data set we cannot seek to make generalised statements about HRE; rather we offer four insights that we have identified to inform future work as researchers, teachers, and resource writers.

Insight 1: We need to be more explicit about teaching human rights as a discourse that can usefully frame our consideration of socio-political issues
We believe it has been fruitful to follow Parker (2018) in exploring what it means to have learned 'about' human rights. We suggest that taking Vygotsky's distinction between everyday and scientific knowledge together with Mercer's (1995) understanding of learning to perceive and think with specialist discourses is a useful way to think about our aims in HRE. Whilst we thought we had a clear sense of what this meant within the Deliberative Classroom project, we now appreciate that it might be useful to be more explicit about this human rights discourse. The argument that one can teach through case studies to perennial issues is compellingly made by Hess (2009) and we saw some examples of this in our data, but the reality of busy classrooms means these conceptual connections are not always thoroughly explored or understood.
It has particularly struck us that a human rights discourse holds tensions and confusions (where does one draw the line between different rights?) and offers a process for working through those tensions. It does not so much provide students with ready answers, as operate as one powerful way to frame a problem and reason through the complexities. As they stand, the resources we have used (and many others with which we are familiar) do not address this. As a consequence, it is possible for students to simply note the tensions and move on, without learning how to work within a human rights discourse. This is compounded by the possibility that students can draw on everyday understandings to 'solve' these problems without using a human rights discourse. Many of the conversations resolved the scenarios drawing on personal experience and morality, without the need to invoke specialist discourses, so there is a role here for being more explicit about what a human rights discourse does and why that is useful.

Insight 2: We need to attend to the connections between understanding human rights and the broader development of political understanding
Our reflections have clarified that an understanding of human rights discourse is a part of the broader project of political education. Barton's (2020) work makes the point that many students who have studied human rights still lack an understanding of how human rights are enacted through international and local institutions, and thus their understanding of political action is limited. However, we feel that this does not quite go far enough, as it merely identifies another block of facts that could be taught. Whilst Keet (2015) argued that we should move beyond a 'declarationist' approach and avoid simply teaching about human rights agreements, so we believe Barton's invitation to teach more content knowledge is similarly restrictive.
Following Rowe (2005), we want to assert that political understanding is not just a facility with political concepts, nor grasping what constitutes the political discourse. We rather perceive it as an ability to understand that the same issues can be understood through multiple framings, and from multiple perspectives. In turn, this helps us to understand any particular case study as a site of conflicting interests, priorities, and perceptions. Importantly, this approach to political understanding also enables one to engage with the 'other' in more authentic ways-acknowledging their perceptions through acts of cognitive empathy, rather than through universalizing emotional empathy, which can erase difference, and thus obscure politics. The students' discussions here have clarified the rather obvious point that human rights thinking is inevitably also political thinking, but that some students avoid this through excessively individualised and emotional responses to problems.

Insight 3: Deliberation and consensus-seeking may restrict critical engagement with others' views and thus encourage a superficial engagement with the topic if processes are not clearly explained and modelled.
Both McAvoy and Hess (2013) and Barton and Ho (2020) make a powerful case that classroom deliberation can foster critical enquiry and constructive argument. Similarly, Al-Daraweeh and Snauwaert (2018) assume that deliberative activities will lead to a deeper understanding of human rights discourse and how to build arguments about human rights. Our experience here demonstrates that, even within highly structured deliberative classroom activities, students may engage in talk which undermines these aims. In those discussions where talk was more 'exploratory', students pushed one another to consider alternative views and occasionally tested one position against another, but the examples of 'cumulative' talk in the extracts demonstrate that some talk narrows the focus and replicates some aspects of the pseudo-enquiries and enclave deliberation discussed above.
This suggests the resources might usefully include a more explicit and structured focus on giving reasons and evidence, and allow them to be tested and analysed by the group. This requires teachers to focus on the deliberative process rather than a conclusive outcome. Not only does this engage more with underlying perennial issues, but it seems to model a form of political reasoning which reveals the distinctive nature of political understanding, and also demonstrates how it is practically useful in the light of conflict and difference. Deliberative discussion may well develop an argument schema, and thus contribute to an understanding of the process of reasoning through a human rights discourse, but this is not inevitable.

Insight 4: Interdisciplinary perspectives are helpful but may need to be facilitated more deliberately as students do not spontaneously make connections between HRE and other relevant school knowledge
The students' limited range of knowledge in these discussions reinforces how the application of prior learning might benefit from more deliberate practice. Here we were undoubtedly hampered by the fact that the sessions we observed and recorded were one-off, short-lived projects that were not embedded in longer-term schemes of work. As Mercer (1995) points out, it is more likely that students will learn to think with discourses as they become more familiar with them and the kinds of conversations they enable. Nevertheless, this data has made us reflect on the challenge of connecting to relevant specialist knowledge and the problems with revisiting relevant curriculum content. The fact that only one student related their lesson to the history of the Reformation indicated that although the other year 8 students in this school had covered the same history, they had not perceived it to be useful or relevant. The specialist knowledge that constitutes the human rights discourse is interdisciplinary, drawing on history, politics, sociology, economics and ethics. Students may encounter this through history, geography, religious education, citizenship and other school subjects, but we cannot assume this knowledge will be drawn on by students without support.
In this article, we have explored the ways that students responded to the Deliberative Classroom resources. Those resources sought to build on Keets' (2015) warnings about a potentially conservative over-emphasis on declarationist knowledge and so focused on critical engagement with complex issues, rather than focusing primarily on teaching about human rights agreements and organisations. Our use of diverse case studies reflected a commitment to engaging with others' experiences and perspectives (Adami, 2014;Osler & Zhu, 2011) and through these case studies we sought to engage with the concepts that underpin HRE (Parker, 2018). In reflecting on our experience we have argued, first, that there is a need to be more explicit about what human rights can do (and the power of a human rights discourse), with a stronger recognition of both the value of scientific knowledge as a powerful complement to everyday knowledge and the tensions within a rights discourse. Second, HRE must enable students to recognise the broader political context beyond just focusing on empathising with individual cases or acquiring information about political institutions (Barton, 2020). Third, we suggest educators need to be explicit about the process of deliberative discussions over the outcome and avoid a push for conclusive answers. Fourth, educators need to support students to draw on knowledge from a range of disciplines. If these issues are not addressed, some students are able to engage in rights-based discussions with little knowledge and understanding of rights.