What are they talking about? Pupil talk while translating Latin stories: a case study of a Year 7 class, using the Cambridge Latin Course

Introduction

The objective of this case study is to consider the role of pupil talk in the translation of Latin. I consider that ‘talk’ within the Latin classroom is an important pedagogical tool to support the development of pupil knowledge and understanding and to promote independent learning. Firstly, I am particularly interested in pupils’ early acquisition of the Latin language through ‘talk’. In this study, I wanted to explore how they use talk to support each other and the issues they encounter, the use of available and additional support resources and the extent of grammatical terminology within their talk. With the latter, I was curious to investigate how much terminology infiltrated pupils’ speech, or indeed if it was absent, and the subsequent effect both of these might have. The oral translation of Latin is an accepted and traditional part of learning for pupils. Throughout my PGCE teacher training course, I have observed that the majority of pupils frequently and independently consult or choose to work with another pupil when translating. I considered it was both a valid and valuable learning strategy to explore and analyse in some depth. Such research would provide a constructive insight into this accepted but sometimes overlooked learning tool, and therefore could realistically inform and improve my future teaching practice. Lastly, research into pupil talk has its own intrinsic worth, as transcripts provide both thought-provoking and informative evidence of their own. Thus, I was keen, within my placement school, to discover and consider how the pupils responded and talked about the language together in pairs.

For my second placement, I taught in a co-educational, independent day school in Norfolk. The school (hereafter ‘School X’) is divided into two sections, the Lower School and the Senior School. I conducted my research in the Senior School, which caters for the 11-18 age range (Years 7-13). The Senior School is comprised of 863 pupils, with 546 boys and 317 girls in total. There are 81 pupils listed with Individual Learning Plans within the Learning Support Department and 3 pupils with English as a Foreign Language. School X attracts pupils from a wide catchment area, predominantly from Norfolk and north Suffolk, and consequently many pupils travel significant distances to attend. School X is academically selective and thus a high-achieving school, with exceptional results: in 2014, 98.2 per cent of GCSE results were graded A*- C and 88.9 per cent of A Level pupils achieved A*- B. In 2010, the Independent Schools Inspectorate inspected School X, and the report states, “the quality of the pupils’ achievement throughout the school is outstanding” (ISI, 2010, p.2). However, it is important to note for this research that as pupils are not set by ability until Year 9 the classes in Year 7 and Year 8 can often consist of pupils across the ability range for this type of school. For example, a Year 8 Latin group had a class average of 80 per cent for marked homework from September 2014 to March 2015, but the individual pupil averages ranged from 63 per cent to 98 per cent. It is also important to note, although girls were first admitted in 1994 and intake has steadily increased since, School X still maintains a higher ratio of boys to girls in all year groups. The class, on which this study is based, is a Year 7 mixed-ability class, comprised of 24 pupils, 16 boys and 8 girls. Within the class, there are two pupils listed with specific learning difficulties: one girl has developmental co-ordination disorder (DCD) and one boy (Pupil G) has a statement of special educational need (SEN) of Autism syndrome. The Classics department is well established, and has four full-time teachers. At KS3, Latin is a compulsory subject in Year 7 and Year 8, but in Year 9, pupils who either prefer the background material or struggle with the language, can change to studying Classical Civilisation. School X uses the Cambridge Latin Course, a reading course, as its primary textbook throughout the KS3 Latin course, but classwork is supplemented significantly with other resources. However, for the most part, the Cambridge Latin Course (hereafter CLC) is employed in some form in every lesson. The Classics department aims for Year 7 pupils to read to the end of Book One of the CLC by the end of the Summer term.

Literature Review

Within this literature review, I will explore the concepts and theory surrounding my area of interest. I will first discuss the issues and debates surrounding pupil talk. I will also consider the use of dialogic teaching and Vgotsky's theory of the ‘zone of proximal development’.

Alexander demonstrates the utility of pupil talk as an effective pedagogical tool: ‘… it is through language, especially spoken language, that teachers teach and children learn’ (Alexander, 2004, p.2). He outlines that dialogic teaching is ‘a distinctive pedagogical approach’ (ibid, p.5). Dialogic teaching is the employment of varying forms of ‘talk’ within the classroom to support and develop pupil attainment. The term itself, dialogic, defined by the Oxford Dictionary, is the adjectival form of a dialogue (Oxford Dictionary, 2015), thus something relating to or in the form of a discourse. Alexander states: ‘Dialogic teaching harnesses the power of talk to engage children, stimulate and extend their thinking, and advance their learning and understanding’ (ibid, p.1). He suggests effective classroom talk should not be confined to the basic teaching catalogue, labelling this, ‘the familiar and traditional bedrock’ (ibid, pp. 27-28), of rote, recitation and instruction. Alexander is not in favour of discarding such forms of interaction but believes they should supplement ‘talk’ in the classroom: ‘The two groups are not mutually exclusive’ (ibid, p.28). Thus he argues for greater inclusion of discussion and scaffolded dialogue alongside traditional and accepted classroom interactions. He emphasises that these forms of talk are not limited to teacher-pupil settings, but also include teacher-group or pupil-pupil talk. It is pupil-pupil talk that I have chosen to form the basis of my own research as the data collection seeks to support both Alexander's aforementioned argument and provide evidence to answer my research questions.

Alexander was prompted to write about dialogic teaching after carrying out previous research: in 2000, he helped to conduct an educational research project involving several countries, in which they systematically observed and videotaped children and teachers at work. Alexander considered that data provided by the videotaping was the ‘most exciting’ as the classroom talk provoked the most interest (ibid). However, although Alexander states succinctly ‘talking is easy’ (ibid, p.5), he then acknowledges its downfalls. It is fleeting: ‘unless our talk is taped or transcribed … it soon fades’ (ibid, p.4). Furthermore, it appears the transitory nature of dialogue is also responsible for the lack of reflection on its use in classrooms today: Alexander notes the tendency to ‘be less reflective about what is said in our classrooms than about what is written’ (ibid, p.5). Mercer similarly argues that our language, although ‘rarely held up for special consideration in research’, is an effective ‘system for thinking collectively’ (Mercer, 2000, p.15). These conclusions prompted my own research methodology: a permanent record of qualitative data to analyse its value as a pedagogical strategy, and to inform my professional practice.

Barnes’ academic paper (2008) considers the role and importance of experimental pupil talk in today's classrooms. He outlines his ‘constructivist’ view concerning the nature of learning (Barnes, 2008, p.1). He first states ‘Learning in schools is a social activity’ (ibid, p.1). The author is interested in the varying roles of communication in the classroom, between the teacher and pupil and between the pupils themselves. He divides and labels such interactions, based upon their intention, into ‘presentational’ and ‘exploratory’ talk (ibid). He argues that the phrase ‘active learning’ is often misconstrued to denote some form of physical movement, but is, in fact, learning which is ‘attempting to interrelate, to reinterpret, to understand new experiences and ideas’ (ibid, p. 2). Alexander equally noted that dialogic teaching is ‘distinct from the question-answer-tell routines of so-called “interactive” teaching’ (Alexander, 2004, p.1). Thus, through talk, Barnes contends, ‘the learner actively constructs the new way of understanding’ (Barnes, 2008, p.2). The pupil is seen to build their own understanding through communication with their peers and the teacher. Pupil talk is, therefore, viewed as key to the construction of new knowledge and understanding. It is the teacher's role to guide and frame the subject matter for discussion and equally to provide resources, which enable productive and progressive talk. In the case of this study, the Latin stories - the ‘narrative and dramatic passage’ of the CLC (CLC, 1999, p.9) - provided rich material for pupil talk: opportunity for a challenging but achievable co-constructed translation and engaging content.

Barnes then refers to the importance of talk as a support and scaffold to pupil understanding: ‘at the centre of working on understanding is the idea of “trying out” new ways of thinking’ (Barnes, 2008, p.4). He appears to imply that the experimental nature of talk is equally decisive in developing pupil confidence: he states whilst pupil talk is an important method for understanding, learners are less likely to interact unless they feel at ease, or free from being contradicted or mocked (ibid). Thus, talk forms an instrumental part of a social as well as cognitive process. Alexander equally noted in his paper that dialogic teaching must allow pupils to ‘articulate their ideas freely … and … help each other to reach common understandings’ (Alexander, 2004, p.22). Furthermore, Barnes notes the flexibility in speech as an effective tool to permit pupil experimentation. He outlines his expectations of the nature of joint enquiry:

Exploratory talk is hesitant and incomplete because it enables the speaker to try out new ideas, to hear how they sound, to see what others make of them, to arrange information and ideas into different patterns. (Barnes, 2008, p.4)

Talk is viewed as a non-intrusive and experimental way of providing pupils with the means to develop their own understanding. Alexander and Barnes both refer to the research of Vygotsky, who theorised issues concerning human and social development, to support their findings and conclusions. Thus, it is important to consider the role of Vygotsky's theory in relation to pupil talk. Alexander states that it was ‘the Vygotskian view that the child's cognitive development also requires it to engage, through the medium of spoken language, with … other children’ (Alexander, 2004, p.7). Therefore, the other children with whom a pupil interacts assume ‘a critical role’ in scaffolding (ibid p.8). Vygotsky termed this the ‘zone of proximal development’, which Schutz explains ‘is the difference between the child's capacity to solve problems on his own, and his capacity to solve them with assistance’ (Schutz, 2014). Therefore, other pupils’ assistance forms a key part of the scaffolding process, ‘providing non-intrusive intervention’ (ibid). Significantly, such assistance is more productive if the peer ‘has already mastered that particular function’ (ibid). Alexander similarly states that the peer must be of higher attainment to achieve an effective dialogue (Alexander, 2004, pp.7-8). In addition, the task itself must be constructed ‘in advance of the child's current level’ to challenge the learner sufficiently and so ensure progress (Tudge & Hogan, 1999, p.62).

Seranis’ paper (2003) explores how pupil discussion about classical literature can encourage a divergent and valuable construction of knowledge. Although the paper does not directly relate to language study, it considers the role of ‘talk’ within Classics and its benefits. The paper evaluates the evidence of an educational study, which investigated A Level pupil responses to the teaching of Homer in translation. As the study progressed, the researchers noted that pupil dialogue was crucial to developing their understanding of classical literature. Seranis states:

What is more significant in the classroom setting is not a recollection of likely interpretations of the intended audience, but a dialogic investigation into how students perceived the text. (Seranis, 2003, p.2)

This demonstrates that pupil discussion is a useful teaching tool within Classics to enable pupils to share and build upon others’ viewpoints: the ‘opportunity … to exchange views and ideas with their peers and modify their responses if needed’ (ibid, p.3). Seranis further highlights the importance of active learning by both pupil and teacher ‘to facilitate and encourage personal engagement’ (ibid, p.3). It could be concluded that the employment of classroom talk creates a learning environment, which promotes debate and the shared construction of knowledge.

Critical reading of the literature has highlighted the importance of pupil talk in the classroom and consequently significantly guided the direction of my research. Classroom dialogue plays a key role in the cognitive and social development of children and the lack of studies surrounding pupil talk in Classics, in part, prompted my research. In this case study, I will examine closely pupil talk within the Latin classroom, focusing on the nature of their dialogue and the influence of the composition of the pupils themselves, with regards to achievement, gender and special educational need. In order to analyse their talk, I will define and analyse the extent of their collaboration, their use of resources and, significantly, the language they use to refer to the Latin.

Teaching sequence

After two observation lessons, I began to teach the Year 7 mixed- ability class in late January 2015. Thus, before I conducted my research, I had a thorough understanding of the class dynamic as well as the pupils’ abilities and personalities. After the half-term break, in consultation with the class teacher and Pupil G's key worker, I rearranged the seating plan in anticipation of the desired pairings. This meant the pupils were accustomed to the new seating plan and surrounding pupils by the time the research began. Alexander suggests in his paper that, ‘dialogic teaching is facilitated and supported when teachers are prepared to change the classroom layout’ (Alexander, 2004, p.29). The school timetable consists of nine periods per day, each lasting 30 or 35 minutes. The Year 7 research class was allocated three Latin lessons a week: a double period on Tuesday afternoon and single 30-minute lesson on Friday morning. The research was carried out over four continuous lessons in a two-week period in March 2015. The research was carried out as the pupils began a new stage in the CLC. In Stage 10, the new language feature is the 1st and 2nd person plural verb endings of the present tense. The lessons were planned with the intention that the case study would not disrupt the flow of the lessons: the recordings were built in and not a separate and detached exercise, thus ‘embedded in an extended sequence of work’ (Barnes, 2008, p.6). I aimed to record the pupils translating different sections as they progressed through the stage. I varied the end product each translation task required to avoid repetition and monotony, but also to see how the expected outcome affected how they translated.

In the first research lesson, pupils were introduced to the model sentences of Stage 10. I explained that the model sentences would follow the form of a debate between the Romans and the Greeks, and we would discover what each side said about themselves and their opponents. I projected images of the first four model sentences onto the whiteboard and modelled the new grammar to the pupils, using open and directed questioning. I asked the pupils to spend five minutes orally translating in pairs the final four sentences (here, the Pupils A and B, and Pupils C and D were first recorded) with unfamiliar vocabulary displayed on the board. At the end of the lesson, the pupils revealed what they had discovered: some quickly mentioned the new verb endings, however, others noted new vocabulary, and others picked out the claims each side had made. The second lesson began looking at the cultural background material of Stage 10, which provided an outline of the Roman educational system. The pupils then translated the first seven lines of the story of controversia in pairs (Pupils A and B, and C and D were recorded for the second time) and wrote their translation in their exercise books. However, in this lesson, I instructed pupils to think about how they translated the section together: either to look carefully at the endings of words, or to give themselves roles, one person in charge of nouns, the other verbs, or to divide the section up. In the following lesson, the pupils continued to translate the story of controversia in pairs (Pupils E and F, and G and H were recorded for the first time). The pupils completed worksheets, which required them to unravel the information they translated. The task reinforced their understanding of the new grammar point, as, each time, they untangled the meaning between nos and vos. In the final research lesson, the pupils first developed their own grammar note of all the persons of the present tense, now having met them all. The pupils then moved onto the story of statuae: a teacher-led translation of the first seven lines and then a written translation in pairs of lines 8-14 (Pupils E and F, and G and H were recorded for the second time).

Methodology

My investigation into pupil talk meant that I primarily wanted to know how the Year 7 pupils talked about the Latin. Taber defines the term case study as an ‘in-depth exploration of a particular case’ (Taber, 2007, p.73). The author Yin similarly notes, in his discourse, a case study is the preferred method when ‘the main research questions are “how” or “why”’ (Yin, 2013, p.2) and require ‘an extensive and “in-depth” description of some social phenomenon’ (Yin, 2013, p.4). In researching pupils’ interactions, I was investigating a pre-existing, ‘normal’ occurrence, as Yin outlines: a ‘social phenomenon’. Pupil talk is a given of any classroom: ‘Talk has always been one of the essential tools of teaching (Alexander, 2004, p.5). My aim was to explore and analyse pupil dialogue, as it existed, ‘within its real-life context’, not to change it (Sarantakos, 1998, p.191). Thus, I wished to avoid artificially creating a scenario that would not accurately reflect this through any form of active research. Furthermore, case study research fundamentally seeks to reflect and study ‘reality’: ‘case studies are grounded in “lived reality”’ (Hodkinson & Hodkinson, 2001, p.3). Hodkinson and Hodkinson argue that case studies retain ‘more of the “noise” of real life than many other types of research’ (ibid, p. 3). Thereby they argue the research maintains a truer representation of a ‘social phenomenon’. Thus, case study research seemed the most applicable methodology to provide a detailed and candid account of pupil talk. Case study research primarily analyses a social process. My research considered the role and advantages of pupil-pupil talk in Latin translation. Therefore I was predominantly focused on how the pupils engaged with each other and the Latin. Case study research was equally ‘…concerned with how people experience and engage with the topic’ (Hamilton & Corbett-Whittier, 2012, p.24). A case study was also best suited to producing qualitative data. The empirical data derived from pupil talk would describe and characterise the interaction, rather than measure or define it. Additionally, as discussed in the literature review, pupil talk is expected to be hesitant, unpredictable and, importantly, unmanageable. Taber notes that case study research is ‘a useful approach when the boundary between case and context may not be distinct’ (Taber, 2007, p.74). Yin equally states it is the preferred methodology when ‘a researcher has little or no control over behavioural events’ (Yin, 2013, p.2). Consequently, it is a valuable research method when many variables are at play, and control of these is not possible. It is impossible to claim the results are typical as the individualism of case study research prevents any potential definitive conclusion. The data gathered in a case study will only ever be representative of a small cross-section of individuals. The issue of small samples arises in any form of educational research and idiosyncratic results are expected from the study of any social process. It is important to highlight that the qualitative nature of the results means that their interpretation is subject to the researcher's subjective analysis of the evidence. There will always be ‘doubts about their objectivity’ on account of this, as ‘case study researchers are constantly making judgments about the significance of data’ (Hodkinson & Hodkinson, 2001, p.10). However, if the researcher supports their findings with clear evidence, these doubts can be somewhat lessened. Having taken these limitations into account, I feel that a case study is the most appropriate methodology to answer my research question.

Research methods

My research aimed to record paired pupil talk - the joint process of translating a passage of Latin. Therefore, to achieve the most reliable representation of pupil talk, I digitally recorded pupil conversations for later transcription. Recording the pupils’ voices would provide a more authentic representation of their talk and abundant qualitative data. Additionally, I chose only to record aurally the interaction, as my primary interest was simply how the pupils talked. I considered asking the observing teacher to transcribe pupils’ conversation, but thought this could be both subject to human error (the teacher may not hear or miss parts of the interaction) and their presence may have worried or distracted the pupils or influenced their interaction and the words used. I contemplated videotaping the pupils working in pairs, but decided this was intrusive and unnecessary. The presence of a camera would undermine my intention to record ‘usual’ pupil talk, as it would easily divert the pupils’ attention, and the video data would be superfluous to my investigation. Thus, I considered a Dictaphone would be the most discreet method and allow me, as the teacher, to move away from the pupils, so as to not influence their conversation. I perceived that despite initial interest in the Dictaphones, the pupils appeared to become less aware of the recording, and speak naturally and at ease. The transcription evidence provided qualitative data, which was chronological and accurate. In my opinion, such evidence is incomparable: it delivers an accurate and unaltered reflection of what was actually said.

Undoubtedly, there are limitations as to the extent that transcription can provide accurate representations of pupil talk. Inevitably, transcription ‘loses information as the concrete event or emotional response is translated into written language’ (Markle et al., 2011). The researcher loses some of the naturalism from the initial conversation, yet I contend that this is unavoidable in transcription evidence. The transference of the intangible spoken language into cold, concrete written data will certainly mean it loses some of the original inflections. However, for the purposes of this study, written data was essential, therefore I have included, where possible, to infer such inflections (see Transcripts Key): indications of emotion, pauses, and verbal stress.

A sample size of four sets of pairs was chosen to represent how the pupils conversed whilst translating Latin. I considered that four pairings was an adequate and achievable sample for effective comparison and it would provide a sufficient amount of qualitative data. Moreover, a sample of four pairs meant that eight out of the 24 pupils in class would be recorded. An evidence base of a third of the class could claim to be representative of the class as a whole. In addition, in practical terms, the resources available meant I was unable to record more than two pairs within a single lesson. I recorded each pair twice, both to check for consistency and to allow comparison between the two. I was aware of the many external factors, which could affect how the pupils talked on a given day, and so considered a ‘one-off’ recording would be neither trustworthy nor reliable. Alexander similarly remarks that the quality and regularity of classroom talk ‘depends on many factors’ (Alexander, 2004, p.29). Therefore, I repeated the recordings to limit the effect of this. The pairings were selected to gather a wide representation of interactions in terms of gender, achievement and special educational need. These included two pairs of pupils of the same gender and similar achievement (Pupils C and D, E and F). I decided to record these pairs in particular, as they were the ‘norm’ of the Latin classroom: I have detected, throughout my PGCE course, whilst observing lessons, pupils, unless instructed otherwise, appear instinctively to work with others of the same gender and similar capability. I was intrigued about the talk that these ‘standard’ pairings would produce and how this would inform my own teaching. I also chose to pair a pupil of low attainment with a high-achieving pupil, of different genders (Pupils A and B). I was interested in applying Vygotsky's theory concerning the ‘zone of proximal development’ to Latin translation, and to study the possible affects this would generate. Finally, I paired two pupils of high attainment, one with a statement of SEN, of the same gender, (Pupils G and H), to observe and examine this unusual pupil dynamic.

Data and Findings

Within this section, I will collate and discuss varying themes within the research findings. These themes will offer insight into how the pupils work together and build upon each other's knowledge. I will first consider the extent of collaboration within each pair - how much of the translation is a joint process. I will then examine the pupils’ language and their use of resources, considering the role of the dictionary, visual prompts and a more capable peer. It is important to acknowledge that the analysis of these findings is my interpretation of the qualitative data.

Collaboration

It could be argued that the extent of the pupils’ collaboration is defined by how much they contribute and respond to the dialogue. The indications of teamwork may be portrayed through pupils’ receptiveness to what was previously said, sharing ideas, asking questions or praise. Alexander noted that an effective dialogue between pupils required them to ‘acquire the capacity to: … listen, be receptive to alternative viewpoints, think about what they hear, give others time to think’ (Alexander, 2004, pp.28-29). Tudge and Hogan noted that the extent to which pupils treat collaborative tasks as ‘a joint endeavour … the more likely it seems to be that children will learn’ (Tudge & Hogan, 1999, p.61). In many respects, the recorded pupils demonstrated many elements of such collaboration and the transcripts display evident signs of productivity and progression: the pupils repeatedly build upon each other's knowledge and ideas. For example, Pupils A and B co-construct their translation and act and build upon the previous response:

A: The teacher … iuvenibus … it's young man but in a different case… Dative.

B: I'll look it up too…. Plural.

A: To the young men.

B: So, the teacher announced the debate to the young men.

A: nuntiabat. Was announcing.

(Appendix 1, Lines 75-79)

There is a deliberative building of ideas, with both clearly aware of what is and should be translated and discussed. Likewise, Pupils C and D, both of similar attainment, demonstrate aspects of collaboration: they are alert to their partner's replies and willing to help:

D: This teacher was Theodorus and prope, exercise area, was something.

C: prope is near. Right, lives near the exercise area.

D: Was living.

(Appendix 2, Lines 54-56)

As Alexander noted, in his discourse on dialogic teaching, the ability to listen is key to collaborative learning (Alexander, 2004). Pupil C identifies Pupil D's vocabulary issue with prope as he scans the sentence, articulating aloud what he understands. Pupil C is quick to supply the vocabulary and develop Pupil D's translation. There is further evidence of such collaboration later in their transcript:

C: Where Theodorus docebat.

D: docet is teaches. So, it's was teaching.

C: iuvenes is young men. It's plural.

D: Where Theodorus was teaching young men.

(Appendix 2, Lines 64-67)

Here, the pupils’ interaction is still collaborative, but each appears to contribute to the overall translation through concentrating independently on particular words. The method is effective, precise and swift, and consequently there is little deliberation over the finished translation. Conversely, the findings equally contain evidence of disjointed verbal interactions. This is most evident in the transcripts of Pupils E and F, which recorded the conversations of two girls of similar attainment:

E: Right, I think optimi is very well. So like the best architects… nos vias et pontes…

F: Wait, what line are we on?

E: Line 18. You vias… No, let's do line 19, urbs Roma? So maior quam means greater than or bigger than.

F: Er… urbs Roma…

(Appendix 3, Lines 22-27)

E: Is it after Alexander bought the statues, he hurries back to the villa where Quintus is?

F: Yeah.

E: Right so duo means two, fratres means brothers. Two brothers… in horto? Can you find out what horto means?

(Appendix 3, Lines 72-76)

Pupil E is evidently controlling the direction of the translations. In the first extract, Pupil F appears to struggle with the frequent deviations and consequently there is no development of ideas. The pupils do not appear to gather, act and build upon their partner's responses, but ‘jump around’ the translation in an effort to piece it together. As such, the interaction seems stilted and unproductive. It is significant to acknowledge here that the overall task design may have affected how pupils approached the translation. As opposed to a written, line-by-line translation, the task required the pupils to group their translation into themes. Therefore, the desired product of the translation afforded pupils some freedom as to how they tackled the Latin. It is difficult, however, to know how much the task design would have influenced the process. In the second extract, although the translation follows a chronological pattern, Pupil F still has little input and Pupil E's audible thought process dominates the dialogue. Nonetheless, it could be contended that there are features of collaboration in their transcripts. The pupils are proactively aware of the joint process, and the necessity to complete the task together. I detect this shared understanding through their use of the shared pronoun, ‘we’, and the joint sense of accomplishment:

E: So we need to find out what numquam means.

F: We've done quite a lot

E: … If I find out what picturam is, will you find out what pingebat is?

(Appendix 3, Lines 40-41, 79)

However, it could be argued that it is a fundamental lack of understanding, which subsequently prevents these pupils from building upon each other's knowledge. Thus, in this case, it could be concluded that pupils of similar abilities appear not to be the most effective pairings, despite being the most common in the classroom. Yet, such a conclusion is unlikely, as Pupils C and D, who were also of comparable attainment, but both boys, portrayed more features of collaboration and progression during their recordings (Appendix 2). This leads me to question if gender or social conditioning plays a prominent role in the pairings, for the transcripts of Pupils E and F exemplify many features of misunderstanding and unconfident questioning. Tudge and Hogan suggest in their research that girls are less likely to show signs of collaboration: girls are ‘more likely to regress, following collaboration with other girls, in part because the girls seemed more interested in preserving good relations with their partners than in arguing with one another’ (Tudge & Hogan, 1999, p.48). However, I am wary of overstating such a claim without further evidence to support this, and am mindful that such analysis is my interpretation of their transcript.

It is important to state that in discussing the extent of pupil collaboration within my findings, Pupil G's input is expectedly less. I had no expectation of a translation, achieved by collaborative work, but was intrigued by how their interaction might develop. Their first transcript almost reflects independent work; Pupil G progresses at his own quicker pace and makes few verbal comments to Pupil H. The transcript is characterised by its long pauses and Pupil G's frequent streams of consciousness. For example:

G: Very brave. Barbarians. Peace. Superamus?

H: The Romans are very brave, fortissimi.

(Appendix 4, Lines 1-2)

In contrast, their second transcript shows greater interaction between the two pupils. Pupil G's key worker commented after the second lesson that Pupil G had ‘felt more at ease and therefore less guarded in his partnership with (Pupil H)’ (Informal conversation, 2015).

G: Find out what legebat is.

H: Reads, so was reading.

G: Yes, it's in the imperfect.

(Appendix 4, Lines 60-62)

Language

C: … pugnatis looks like a verb.

(Appendix 2, Line 18)

To further examine how the pupils talk about the Latin, it is necessary to consider the language they employ. In this section, I will consider how the pupils refer to the Latin and their use of grammatical terminology. Throughout my own schooling, an understanding of the terminology was seen as the prerequisite of comprehending Latin. However, during my research, it became clear that this was not always the case. The CLC Teacher's Guide states that the course ‘draws a distinction between knowledge about the language and skill in using the language’ (CLC, 1999, p.8). Thus, it contends that naming a particular tense or case is not fundamental to being able to read Latin. The pupils are able to elicit and comprehend the meaning and function of a Latin word, before it is formalised as a grammatical rule or term. The CLC is a reading course - ‘a whole language approach’ (Lindzey, 2003, p.8). Whilst a grammar textbook relies on mastery of grammar features before a pupil is able to progress, the CLC immerses the pupils in a continuous and progressive motivational text. In addition, I was interested to see if pupils referred to terminology in their speech and the impact this had on their translation. Would specifying the grammatical terminology be of any real benefit, if it were already understood?

Throughout the transcripts, all the pupils show some awareness of the emphasis placed upon word endings and how these affect the translation. The imperfect tense appears to be most commonly identified piece of grammar. The pupils frequently refer to the bat tense, presumably on account of its distinctive ba signpost. The Key Stage 3 Pupils at School X had often commented that the imperfect tense was easily recognisable and so they felt confident in accurately translating it (Informal conversation, 2015). The pupils’ references to the bat tense, or their correction of their partner's translation (transferring the verb to ‘was doing something’) demonstrate a real understanding of the tense without naming the terminology. This is evident in this extract from Pupil E and F's dialogue:.

E: …. If I find out what picturam is, will you find out what pingebat is?

F: pingebat?

E: Yeah, the bat means was doing something, so look for it without that.

(Appendix 3, Lines 79-81)

Whilst advising Pupil F about how to locate the verb pingebat in the dictionary, Pupil E inadvertently reveals a clear understanding of tense endings. She demonstrates awareness of the interchangeable nature of verbs and how the verb will look different in the dictionary. It is possible that Pupil E chooses not to label the tense: having given the English meaning, such information would be redundant. Furthermore, it seems that the labelling of grammar may simply add a layer of confusion to the dialogue: it may not be fully understood by the other pupil and it is unnecessary if the pupils have already extracted the meaning of the word.

Throughout the findings, I noted that pupils frequently appeared to identify grammar by its characteristics. In place of the terminology, the pupils highlighted the grammatical content of the word by referring to its distinctive features. It is essentially a memorising technique, a learning strategy, personalised to the pupils - their own method to identify and remember the grammar. For example, Pupil B recognises the object of the sentence, the accusative, simply by verbally isolating the last letter, and Pupil F picks out a superlative by noticing the addition to the stem of the adjective:

B: palaestram is the ‘m’ one.

(Appendix 1, Line 46)

And:

F: It's an issi one. Very.

(Appendix 3, Line 7)

In contrast, Pupil A, a pupil of high attainment, uses the terminology to distinguish quickly the meaning of porticum:

A: Young men. Afterwards, what's porticum?

B: Won't it be the same as porticus?

A: Oh, it's the accusative of porticus.

(Appendix 1, Lines 64-66)

Moreover, throughout the transcripts, I noticed that pupils were unlikely to specify grammatical terminology if they were unsure of being correct. Additionally, sometimes the terminology seemed to only be mentioned once the pupils had decided the translation (for example, Appendix 2, Line 58). Thus, it seems pupils are, to an extent, careful and apprehensive about stating any terminology. However, there are limitations to this argument; the task itself required pupils to translate the passage, not to identify grammatical terminology. It is, therefore, important to note that the purpose of translation exercises may affect how pupils talk about the Latin and subsequently the focus of their dialogue. A translation exercise may be used to achieve a variety of learning objectives: to comprehend the story, to translate the passage, to model new grammar, or to identify and state particular syntax. It is the role of the teacher to instruct and guide the pupils clearly as to what they should be able to do. Therefore, as the pupils were instructed simply to translate the passage, it is perhaps unreasonable to criticise the lack of spoken terminology. Consequently, the purpose and aim of translation exercises could be argued to have significant impact on how pupils refer to the Latin.

It is also important to acknowledge the role of guesswork and intuition in the pupils’ joint translation. The CLC encourages the use of intuition to support pupils to ‘elicit, correct, concrete answers’ (CLC, 1999, p.12). In the following extract, Pupil E queries the number of the noun statuas, by picking out the case ending for deciphering. Pupil F's reply, although correct, is not verbally supported by any explanation:

E: Alexander buys the statues for the brothers. Is it statue or statues? It's got as on the end.

F: Statues, I think.

(Appendix 3, Lines 48-50)

Finally, the language used by Pupil G, who has been diagnosed with Autism syndrome, is predictably different but noteworthy. His language and verbal translation of Latin are almost ruled by the use of grammatical terminology. The terminology for Pupil G provides a methodical framework to follow and engage with:

G: Graecas must be first declension, because statuas is first declension. Plural accusative.

(Appendix 4, Lines 24-25)

Thus, overall, it seems whilst the pupils do not need to explicitly refer to grammatical terminology when translating, it is simply useful for pupils to comprehend the terms when discussing linguistic features.

The learning of any language is supported through sounding and reading words aloud and although Latin is not spoken today, this strategy is still a useful exercise. Lindzey states that ‘the importance of learning noun endings is reinforced when we read’ (Lindzey, 2003, p.9). The CLC Teacher's Guide states that when pupils ‘hear the words organised into phrases or clauses … they glean clues to the meaning’ (CLC, 1999, p.13). It seems that the reading aloud of Latin is able to act as an aural stimulus for pupils to locate known vocabulary or phrases. In these extracts, pupils read the Latin aloud, and select unfamiliar words for further analysis:

A: hic rhetor erat Theodorus et prope … What does hic mean?

(Appendix 1, Line 50)

And:

G: What's ad? … Something villam cum Quinto condendit… something to Quintus. Quinto is dative. What's ad villam?

(Appendix 4, Lines 43-44)

Resources

In this section, I will examine how pupils employ the resources available to them. I will consider how the pupils use the dictionary in pairs, but also the difficulties it brings. I will then consider the pupils’ use of visual prompts in the CLC. Finally, having discussed Vgotsky's theory in the literature review, I will review the role of a higher-ability peer as a resource within paired translation.

Every transcript includes examples of pupils looking up unknown words in the dictionary. It is expected that as the CLC builds up the vocabulary stage-by-stage, new words are frequently introduced, which are appropriate for the linguistic demands being met at the time. Therefore, it is not unusual for pupils to check the dictionary repeatedly. Throughout the transcripts, I observed that dictionary use was primarily a shared process. Most commonly, the pupils either declared to their partner they were looking up a particular word (Appendix 1, Line 13), perhaps to prevent the other needlessly doing so, or asked their partner to find a word's meaning (Appendix 2, Line 18). In my findings, there appeared to be a distinct balance between these two methods. However, I often found that looking up unknown words in the dictionary was a problematic exercise in itself: the fundamental unfamiliarity of the word often meant pupils found or chose the wrong English meaning. Pupils then became perplexed as they attempted to fit the word into the context of their translation and to manipulate the rest of the sentence to support this. For example, Pupils E and F's vocabulary check of contentiones, meaning arguments, mistakenly leads to an English meaning of satisfied (contentus), and there is much confusion before they decide to recheck contentiones (Appendix 3, Lines 11–21).

Pupils generally looked up vocabulary rather than grammar. But the following extract contains a rare example of the pupils purposefully looking up the grammar. It could be suggested that this word is looked up chiefly on account of its ‘more unusual’ ending; here, both Pupil A and B collaboratively look up the case ending for iuvenibus:

A: The teacher … iuvenibus … it's young man but in a different case… Dative.

B: I'll look it up too…. Plural.

A: To the young men.

(Appendix 1, Lines 75-77)

The CLC employs pictures to provide an additional support resource. The pictures can be used to elicit and infer the meaning of the Latin. It is a process of semiotics, by which the meaning is conveyed through the relationship between the Latin and the pictures. The model sentences are accompanied by line drawings to facilitate and prompt pupil understanding. The pictures are similarly a support to the new vocabulary, and so help the learner to grasp, in some way, the sense of the model sentence. For example, in this first extract, Pupil A attempts to deduce the meaning of res utiles facimus from the actions taking place in the accompanying picture:

Figure 1 | CLC, Book 1, Model Sentences, p. 133 (by permission of CLC).

B: Is it we use tools?

(Appendix 1, Line 21)

Although Pupil A's ‘educated guess’ is incorrect, importantly, it is a close interpretation and effectively shows his engagement with the task. However, Taylor argues that the CLC encourages an ‘over-reliance on instinct and guesswork’ (Taylor in Lister, 2005, p.14). Such dependence is noticeable in Pupils C and D's translation, which appears to be entirely guided by their interpretation of the picture, in relation to the word spectatis:

C: We actors are always.

D: No, you actors. Always sitting down?

C: You actors always sit down. That doesn't make sense.

(Appendix 2, Lines 9-11)

The pictures are designed to act as a visual aid, if required, to the translation of the introductory sentences. However, in this second extract, they have an adverse effect, as it appears Pupils C and D's translation becomes reliant upon the picture, rather than the Latin. Yet, there are limitations to this argument; significantly, the pupils’ independent translation of the model sentences is not the course's intended approach. The CLC Teacher Guide's recommended handling of these sentences stresses the importance of guiding pupils through them (CLC, 1999). The pupils should be directed through all the introductory sentences and pictures orally (CLC, 1999): the teacher should carefully control and model the vocabulary and grammar, and verbally reinforce the way pupils talk about the Latin. Therefore, the pictures are foremost a support to the new vocabulary, and the Latin is the first alert to the new grammar feature. The translation of the model sentences should be guided by reason, rather than instinct or guesswork. Thus, without such teacher guidance, Pupil C and D's reliance on the supporting pictures is not unexpected. Furthermore, it is important to remember that the pupils have never met the grammatical feature before and therefore, the task design, although appropriate to the case study research, was not the most effective pedagogical approach.

The pupils also appeared to employ the pictures to support their understanding of the story's context, the cultural background material. In this extract, Pupil A utilises the accompanying line drawing to the story of controversia to illustrate what a colonnade is:

Figure 2 | CLC, Book 1, Title picture for ‘controversia’, p. 134 (by permission of CSCP)

A: What's a porticus?

B: I'll look it up… a colonnade. What does colonnade mean?

A: I'm just writing down the words we've looked up… I think it's that area with all the columns and pillars, like the picture there.

(Appendix 1, 57-60)

Lastly, I will consider the use of a higher-ability pupil as a learning resource. In my literature review, I discussed how Vygotsky theorised the importance of cooperative learning between peers. His theory asserted, ‘children gain mastery … in the course of interacting with others … often more competent’ (Tudge & Hogan, 1999, p.39). Therefore, pupil talk between those with differing abilities is considered to affect positively the learning and understanding of both pupils. Pupils A and B were employed as an example of a mixed-attainment pairing and their interaction is key to exploring the applicability of this theory to Latin translation. During the first transcript, it could be argued that Pupil A, the high-achieving pupil, dominated and guided the translation, and Pupil B's input consisted mostly of dictionary assistance and summarising the agreed translation (Appendix 1, Lines 11, 23, 31). However, in contrast, in the second transcript, Pupil B had an increased input into deciphering the Latin and collaboratively building the translation (Appendix 1, Lines 65, 71). I consider that Pupil B's progress is motivated, in part, by the challenge of working alongside Pupil A and so appears to be fully engaged with the process. Thus, it is possible to conclude that mixed-attainment pairings, once established, can have a constructive effect on joint translation exercises.

Conclusion

Overall, my experience of undertaking this research project has been a positive one. The qualitative data collected has provided me with a rich source of evidence to analyse and reflect upon. The aim of this case study was to examine how Year 7 pupils talked about Latin whilst translating in pairs. External and internal factors, for example, the pupils’ gender, achievement and special educational needs, all appeared to impact on how the pupils utilised ‘talk’ whilst translating. This is evident in the transcripts and shown through the extent of the pupils’ collaboration, the frequency and use of grammatical terminology and the employment of available resources. It is possible to conclude that the use of grammatical terminology is not a necessary aspect of translation, but an understanding of the roles of words is important. Technical vocabulary should not be seen as a barrier. Pupils should be supported to develop their own strategies for translation, especially in the early stages of pupils’ engagement and motivation to learn Latin. I consider that this is an important concept if we wish to give all pupils the opportunity to learn and enjoy Latin.

On reflection, this research project has been a valuable professional development experience. However, further research is needed into the value and use of pupil talk in Latin and how this might differ from other curriculum areas. In future research, I would be interested to see how pupils’ interactions differ within the non-linguistic fields of Classics. Mercer notes that research into the language of pupils would help us to ‘understand why joint activity is sometimes more or less effective’ (Mercer, 2000, p.15). I have undoubtedly learnt that giving pupils the opportunity to talk in pairs about Latin is a positive tool to empower learning. It is one however that requires careful lesson planning to ensure pupils are paired effectively, possibly by attainment, and the purpose of the translation is clear. Thus, this case study has shown that paired translation in Latin can effectively allow pupils to support and build upon each other's knowledge and understanding.