Can creative writing in Latin support students' confidence in and enjoyment of A Level prose composition?

Abstract

This research aims to explore the ways in which creative writing may be used as a pedagogical tool in the Latin language classroom, in particular how creative writing may benefit students in Latin prose composition. The lesson sequence delivered as part of this research was undertaken in an academically-selective, independent coeducational school in an affluent, inner-metropolitan area. The sequence of four 60-minute lessons formed part of the language (as opposed to literature) portion of timetabled Latin lessons for a group of nine Year 12 students (aged 16–17). As part of their language lessons, the students had been following a course of study in prose composition based upon Andrew Leigh's (2019) Latin Prose Composition: A Guide from GCSE to A Level and Beyond¹. The lesson sequence was intended to build on this work by making use of, and thus consolidating, grammatical constructions and vocabulary which the students had already encountered in the context of prose composition. The sequence was designed in such a way that students were required to apply their linguistic knowledge in new and creative ways. Students' responses to the various activities were positive and they expressed enjoyment in the methodologies.

Introduction

My research question originated from observations I carried out before taking on the teaching of the class myself. I noted that the generally positive and highly engaged cohort would seem to utter a collective groan whenever the lesson turned to prose composition. I heard individual students frequently speak negatively about both prose composition itself and their own ability with regards to it. My discussions with the class teacher after lessons revealed that she too had a sense that the class as a whole did not enjoy prose composition, but she also said that this was to be expected as, in her experience, students tended not to enjoy it.

This made me think back to a personal experience I had while studying Classics at university. My tutor set a task to write a piece of original prose in Latin. I remember finding the task incredibly freeing: rather than having to grapple with the often-awkward act of translating from English into Latin, I was able to write directly in Latin and produce a piece of prose that I would never have written in any language except Latin. The act of writing in Latin gave me a sense of ownership over the language. I could express what I wanted to express and I was much more motivated to do that, than I had been to simply translate something predetermined. Given my own positive experience, I was motivated to explore whether I could bring similar experiences to the students in my class.

Fundamentally this research is predicated on three key ideas:

1. 1) Current approaches to Latin prose composition are not serving students' as they should (Dugdale, 2011), hence the students' dislike of it and the class teacher's seeming resignation to their dislike.

2. 2) Creative writing is a powerful tool for aiding second language acquisition (Smith, 2013), and so it likely follows that it can also be a powerful tool in developing Latin literacy.

3. 3) Engaging students' creativity and setting tasks which allow them to explore and experiment with Latin in a more free-form way is likely to give students a greater sense of ownership over the language (Dugdale, 2011), and thus build their confidence in, and relationship to, Latin.

This research will examine to what extent these ideas are valid, and to what extent creative writing should be incorporated as a pedagogical tool in the Latin classroom.

Prose composition in the Latin classroom

The current literature examining the value of prose composition in the Latin classroom tends to imply that there are two diametrically opposed camps when it comes to prose composition, those fervently for and those vehemently against. Saunders describes it as a debate with ‘few fence-sitters among teachers of classical languages’ (1993, 385) and Davisson notes that ‘Latin prose composition has provoked considerable argument’ (2000, 75). In fact, my reading has found that most research on the subject of prose composition is largely in favour of it, but there are widely recognised issues with it in its current form. When I describe prose composition in its current form, I refer to the reality for most students of Latin today; that is prose composition as simply the translation of pre-existing English sentences into Latin, sentences that are generally ‘not part of any connected narrative..; that have no audience but one, the teacher; and have as their primary purpose to learn morphology, syntax, and vocabulary’ (Gruber-Miller, 2006, 190). It is important to note that prose composition can and should refer to any manner in which students compose prose in Latin and there are many ways that this can be done in a manner which allows for expression and creativity, yet the current approach in many British schools is generally narrow, highly prescriptive and lacking creativity on the students' part.

A key paper which lit the fire of the prose composition debate was Ball and Ellsworth's (1989) inflammatory article, Against Teaching Composition in Classical Languages, in which they tie prose composition's legacy very closely to an elitist intellectual approach to language learning which emphasised the ‘thorough memorisation of complicated grammatical rules’ (Ball and Ellsworth, 1989, 55) as ends in themselves, and the composition of sentences ‘which teachers selected more for intellectual rigor than pedagogical value’ (Ball and Ellsworth, 1989, 55). They consider prose composition as a tool which elitist teachers use to humiliate those they consider less knowledgeable and which has ‘no substantial pedagogical value’ (Ball and Ellsworth, 1989, 60–61). In this way they dismiss prose composition as both culturally problematic and pedagogically bereft.

Responses to these claims have been produced ever since, with proponents arguing that in fact the benefits of prose composition in language acquisition are evident, even essential (Davisson, 2000; Gruber-Miller, 2006; Saunders, 1993). However, even amongst its proponents, there is often sympathy with the concerns raised, and prose composition is often acknowledged as problematic, something that when delivered poorly lacks enjoyment and even usefulness (Fogel, 2002, 79), and in its current form deprives students of ‘the joy of expressing ideas’ (Gruber-Miller, 2006, 190). This is perhaps why so many of those extolling the virtues of prose composition have done so, not by advocating for the merits of current practices in prose composition, but by proposing alternative approaches to it (Davisson, 2000; Fogel, 2002; Matz, 1986).

My research follows a trend therefore when it comes to advocating for the value of prose composition, by presenting an alternative model in which students are encouraged to explore and play with the language, and ultimately to develop a sense of ownership and personal connection to it.

Creative writing in the Latin classroom

While there is only a small body of research which looks specifically at the use of creative writing in the Latin classroom, what does exist is overwhelmingly positive about its use. Research studies carried out in a school context have found that students respond positively to the chance to write in a way which allows them a means of individual and often quite personal expression (Barrett, 2020). A notable study conducted by Dugdale finds that for students beginning Latin there is a great benefit to using a variety of creative tasks for ‘fostering a greater sense of investment in the language’ (Dugdale, 2011, 3). Through actively constructing the language, rather than passively receiving it, and by maintaining authorial independence, Dugdale makes the case that students gain ‘a greater sense of personal engagement both with the language and its cultural contexts’ (Dugdale, 2011, 18). This finding is backed up by Barrett (2020) who suggests that students have significantly higher levels of enjoyment when carrying out creative storytelling tasks as compared to more traditional English-to-Latin prose composition.

Research carried out at a university level finds similarly positive outcomes, suggesting neither age nor level of proficiency in Latin diminishes the benefits of creative writing as a pedagogical tool. Davisson, whose research is conducted with university students, finds that there is a great benefit to students being able to grapple with the language they are learning in ‘the context of topics… of interest to the student’ (Davisson, 2000, 80). That is to say, by allowing students creative freedom in choosing the content of their written work, they are more likely to choose content which will keep them personally interested and invested in its production. The research also points towards clear benefits when it comes to students' progress in developing their grammatical and lexical knowledge of the language. Dugdale finds that for beginning Latin learners ‘creative writing exercises can provide needed grammatical reinforcement’ (Dugdale, 2011, 3). Davisson posits that ‘students’ sensitivity to Latin vocabulary and syntax as well as idiom and rhetoric increases with active use’ (Davisson, 2000, 75).

Despite these findings, I would argue there has been little, if any, change in most teachers' approaches to prose composition. However, there is some hint that change may be on its way. The new International Baccalaureate Diploma in Classical Languages (2023) assessment now includes a ‘creative and innovative’ coursework task in the form of ‘free composition in Latin… prose’ accompanied by a commentary in which the student ‘outlines their intentions, discusses their rationale… and explains their choices and processes’ (Trafford, 2022, 3). This is quite distinct from any other pre-university qualification in the UK. It is not yet clear whether the Classical Languages IB is an outlier or a pioneer, but it certainly suggests that there is some momentum towards seeing more creative, or ‘free’ as the IB has termed it, approaches to prose composition.

The case for creative writing in the process of second-language acquisition

While research into the effects of creative writing in Latin is fairly limited, there is much that has been written about the benefits of creative writing as an instrument in second language acquisition in general.

The positive findings regarding creative writing in the Latin classroom are mirrored by similar findings in research conducted in other language learning contexts. There is a general consensus that creative writing supports the development of more sophisticated use of morphology, syntax and vocabulary, in an engaging and intrinsically motivating way. Smith (2013) highlights the connection between the creativity of language tasks and students' enjoyment of them which in turn leads to greater intrinsic motivation. He also argues that in language learning there is often a divide between a focus on form, that is syntax and morphology, and a focus on meaning. In particular, he argues, the learning of grammar can be problematic, ‘inauthentic, controlling and non-communicative’ (Smith, 2013, 13); this is an argument which has much in common with those who raise concerns about the disjointed nature of prose composition activities which require the translation of sentences disconnected from any narrative and lacking any real meaning (Gruber-Miller, 2006). Smith argues that, by contrast, creative writing ‘provides a means of combining meaning-focused and form-focused tasks’ (Smith, 2013, 13). I would argue that in Latin we deprive students of the joys of meaning-making, a crucial element in the process of teaching and learning (Ignelzi, 2000; Nash and Murray, 2010), when we do not allow them opportunities to construct meaning in the language we are teaching them.

The significance of meaning-making is also key to research carried out by Tin who argues that ‘in a creative language task which requires learners to communicate about new meanings… [they] must somehow innovate and complexify their language, by re-analysing and combining known utterances and structures to create new ideas and forms’ (Tin, 2012, 179). There are studies which show specifically that creative language play can motivate learners to make use of more complex and ambitious grammatical constructions (Kim and Kellogg, 2007) in a way which may be lacking from non-creative tasks (Joyce, 2009). This reflects the findings explored above from the studies of creative writing in the Latin classroom, which found that students' output tended to be more sophisticated and grammatically complex in creative tasks.

There is also an argument amongst the second-language acquisition literature, which is yet to appear in the Latin-specific literature, but which must surely be of particular interest to those Latinists concerned with accusations of elitism (Ball and Ellsworth, 1989), and that is the argument that creative writing helps to level the playing-field when it comes to learners of differing abilities (Roberts, 2013; Ross, 2007; Thorpe, 2021). Because students are focused on personal expression and have freedom to choose which words and grammatical forms they play with and which they choose not to make use of, they can develop their linguistic skills at a rate suitable to their own individual learning.

Assessment for learning: assessing creative work and its associated learning outcomes

In conducting this research, it has been necessary to consider the role of assessment in gauging the success of the activities I introduced into the classroom. The literature on assessment is quite clear on the value of formative assessment. When assessment informs the teacher's practice, it can be said to be formative assessment, as opposed to summative assessment which simply tells students how well they have done (Dixson and Worrell, 2016). Formative assessment is ‘at the heart of effective teaching’ (Black and Wiliam, 1998, 140) and so it was crucial that my lesson sequence allowed for, indeed revolved around, the formative use of assessment.

My research was interested in the production of written work as a means of consolidating pre-existing knowledge of grammar and vocabulary, which meant that only certain forms of assessment were appropriate. For example, while questioning is a key tool I typically use to assess my students' learning in the classroom, in this context it would not be appropriate. The only way to assess the success of my students' work ultimately was to read it. I therefore needed to ensure that I used the information gathered from my reading of their work in a formative way. It is generally agreed in the literature on assessment that teacher-to-student feedback, whether written or verbal, is ‘an essential part of formative assessment’ (Black et al., 2003, 41). This is because not only does feedback provide students with a means to improve their work but it provides information based upon which a teacher can ‘modify the teaching and learning’ (Black et al., 2003, 41) happening in their classroom.

While there is a good deal of literature on the topic of assessment for learning, I have found its intersection with assessing creative writing tasks to be limited. Given the subjective nature of creativity and the difficulties in its assessment (Bulman, 2008) I knew my research would require experimentation regarding the modes of feedback and assessment which I carried out. I have focused on discussion of this experimentation in the evaluation of each lesson, and how my findings led me to approach the next lesson differently.

Methodology and planning the sequence

Key to my research was the use of a qualitative method of data collection to measure student opinions, as my primary research interest concerned not the students' work per se but their opinions towards prose composition following the completion of work.

I needed to establish student opinions both before the lesson sequence and after to establish any changes between the two. I decided to survey the students using an ordinal Likert scale. To address one of the limitations of such a scale, I also included a section which invited students to offer any other comments.

I knew that for each lesson I would require the students to produce a piece of creative written Latin. To facilitate this, in every lesson I would provide a creative prompt and a set of grammatical constructions to be included in students' work. It has been found that ‘too much freedom… can limit rather than enable the exploration and transformation of learner language’ (Tin, 2012, 178) and can in fact have a ‘paralyzing’ effect (Joyce, 2009). By providing prompts I was providing a certain level of scaffolding for students. Beyond this, given the importance of formative assessment and reflection in shaping my lesson plans, I essentially planned each lesson only once I had taught the one before, implementing the adaptations I deemed necessary based on my assessment of the students' work.

To ensure the anonymity of all participants, all the names referred to below are pseudonyms. While the class activities carried out were part of my normal professional duties and so did not require the consent of students, I wished to conduct questionnaires at the start and end of the sequence of lessons to gather data on student opinions.

Lesson sequence

Below I have provided an overview of each lesson's content, and an accompanying evaluation. In the interest of brevity, I have chosen to focus the evaluations on the assessment of student output. Full lesson plans and evaluations can be found in the supplementary appendices 1–4 for each lesson.

Lesson 1 (See Supplementary Appendix 1)

Overview

The first lesson in the sequence required that students read the story of Pygmalion as found in Cullen and Taylor's Latin Stories: A GCSE Reader (2011). They then wrote a creative response to the story (see Supplementary Appendix 1.A for lesson slides and Pygmalion handout). Students were required in their creative response to include one purpose clause, one indirect statement or indirect command, and one ablative absolute.

Evaluation

I took in the students' work to read and provided brief written feedback. I found it quite difficult to know exactly how to feed back as I had not anticipated how different the experience of reading their very varied responses would be to reading traditional prose composition answers. In my concern to not stifle or dampen their first attempts at creative writing in Latin, I focused my feedback too much on general praise with a little area for improvement flagged, and I did not highlight each individual error a student had made (Supplementary Appendix 1B). This was a regrettable decision on my part because, looking back over the course of the lesson sequence, I see now that by not highlighting and targeting every linguistic error a student made, I was privileging the act of creativity itself at the expense of developing students' accuracy and linguistic skill, and was therefore not maximising the formative potential of the feedback I was providing. Therefore, in evaluating this lesson I found that while the lesson itself went as planned and students worked in a focused way, students struggled to shoehorn in the grammar constructions required of them, and I failed to implement an adequately robust system for providing meaningful feedback to the students. Therefore, for lesson 2 I decided to implement a mark scheme which would make up for my error in lesson 1.

Lesson 2 (Supplementary Appendix 2)

Overview

The first part of this lesson was spent recapping the use and formation of ablative absolutes as I identified these as an area requiring particular focus while reading students' work from lesson 1. This had not been in my initial lesson sequence but is an example of where I used my assessment of students' work formatively (Black et al., 2003), and adapted my teaching accordingly. Students were next required to imagine they had received a letter from home and write a passage explaining what was in the letter. The grammatical focus of this task was indirect statement and ablative absolute (Supplementary Appendix 2.A).

Evaluation

While assessing the work created in this lesson, the limitations and shortcomings of the mark scheme I had devised became clear to me. For example, one student, Brandon, produced the most accurate passage insofar as his work contained no linguistic errors, and so his mark was the highest of the whole class. However, Brandon only produced a three-sentence long passage and did not attempt to take any creative risks with his language or aim for more ambitious iterations of the grammatical constructions required. Compare the first third of Brandon's passage to the first third of another student Bobby's passage (note that any student errors have been retained).

1. 1. Brandon: ‘respondet, urbe deleta, populum perterritum Romae tugere.’

2. 2. Bobby: ‘urbe deleta, veritus sum ut Synnodus non advenisset; dice mihi, amice, quid faciam? Synnodus dixit se iter ad urbem facuturum esse; non solum hostes castra Synnodi appropinquant, sed etiam vos proximi sunt.’

Once my mark scheme was implemented Brandon received higher marks than Bobby, yet these extracts from their work clearly show that of the two, Bobby has extended herself further and attempted the use of much more sophisticated language. My limited mark scheme had come up against exactly the problem which concerned Trafford (2022) in his article on the new requirement for creative composition in the International Baccalaureate. Trafford notes that ‘there is no apparent consideration or allowance for how the difficulty level of each composition will be assessed’ and asks ‘does that mean that a student can produce an excellent piece of Latin comprising simple vocabulary and grammar? One would hope not!’ (Trafford, 2022, 3). I therefore decided to focus on written feedback, in green pen for corrections and pink for suggestions to help the students acquire a more Latinate style, and I put the results of the mark scheme on the back to diminish focus on it. I explained as much to the students when returning their work, and suggested next time I shouldn't use a mark scheme. The students, however, were unanimous that they found the inclusion of marks somehow comforting. One student, Fiona, volunteered the following explanation as to why: ‘when I see lots of comments, it makes me feel like I've done a bad job but the marks reassure me that actually it's ok, I have done well.’ I therefore decide to keep the mark scheme and adapt it for the next lesson.

Lesson 3 (Supplementary Appendix 3)

Overview

In this lesson, students chose one of three visual prompts (Supplementary Appendix 3.A) and produced a creative response to their chosen prompt. Students were required to incorporate into their passage one purpose clause, one result clause, one indirect command and one indirect question. Students were also required to provide an English translation of their work, as I had found that the marking process after lesson 2 would have been helped by me having a greater understanding of what the students were aiming to say (Supplementary Appendix 3.B).

Evaluation

In assessing the work created in this lesson, I found the new mark scheme I had devised to result in scores which were somewhat more reflective; however I still found it unsatisfactory in being truly reflective of the extent to which a student had successfully extended themselves to create a piece of Latin which was at once ambitious in its complexity, strong in its accuracy and compelling as a coherent, original work. It has since occurred to me that in running these activities in the future, I should apply a rubric style system of assessment which would categorise students' work into bands, rather than a numerical marking system as I had been attempting. The value of a rubric would be on the one hand, a clearer more in-depth guide for students as to the expectations surrounding their work, and for the teacher, a way to grade students' work which was more reflective of their output as a whole and allowed for more qualitative considerations such as the strength of their authorial voice (Morton et al., 2021). While the development of such a rubric would be initially time-consuming, once established I believe it would go a great way to alleviating the problems I encountered in marking the students' work.

Lesson 4 (Supplementary Appendix 4)

Overview

The final lesson in the sequence was designed to consolidate the grammatical structures the students had been working on, while providing a different approach to creative writing (Supplementary Appendix 4.A). Whereas the previous lessons all centred around individual tasks and required completely independent work from the students, this lesson required students to work in pairs to create a story. To support this collaborative process, I created a gamified starter, intending to set a fun and playful tone, and I encouraged this tone of playfulness throughout the lesson. Student work was this time not marked by me but read and commented on by peers at the end of the lesson.

Evaluation

This lesson achieved something quite different to the others, insofar as it focused more explicitly on the fun and playfulness of using language to create meaning. I think it worked well as a final lesson in the sequence as it required students to have enough confidence in their ability to construct meaning in Latin to then play with that meaning without recourse to fear that they might make an error. Students appeared to have a lot of fun during the task, and lots of laughter was shared. The final part of the lesson ended up very rushed so while students didn't have time to give the level of detailed feedback that I had hoped, the comments left on their peers' work reveal that they responded to the passages as entertaining stories first and foremost, rather than as examples of grammatical constructions to be deemed correct or incorrect (Appendix 4.B). Comments such as ‘oh dear…’, ‘had me worried for a sec’ and ‘he should've died!’ show the spirit of fun with which the students responded to their peers' work. Prose composition should ultimately be about communication and the joy of expression (Fogel, 2002), not about making students feel inadequate for every little mistake they make, and so I consider this lesson a huge success for tapping into that joy.

Research findings

Findings from Assessment of Student Output and Lesson Evaluations.

I have drawn conclusions from across all nine students' work in the process of teaching my lesson sequence and some key findings are as follows.

1. 1. Student output varied in such a way that by lesson 3 I was able to sense a unique authorial voice becoming apparent in each student's work. This was particularly apparent in Emma's work – which contained lots of poetic and mournful imagery; ‘segestes aureas arescere’, ‘vir caerulee, tuum fletum retinere’ and ‘tam acriter lacrimas ut odium meum vinceris [sic]’ – and in Bobby's work which contained vivid emotional appeal with the frequent interjections of direct speech and jussive subjunctives – ‘dice mihi amice’, ‘cives gaudeamus!’, ‘quid faciamus?’ and ‘eam vidistine?’ These authorial voices I would argue are evidence of the opportunity these tasks allow for self-expression (Gruber-Miller, 2006).

2. 2. The majority of students tended to attempt quite complex grammatical forms. So much so that the time I had planned for students to read each other's work in class ended up being insufficient for students to be able to comprehend the passages in front of them. For example, Fiona produced the following sentence: ‘rogabat utrum egrediantur ne [sic] manerent quia tametsi senex dicet [sic] turrim periculosum esse, timuit cedere.’ This is a much more complex sentence than a student at A Level would be required to produce. This chimes with the literature which suggests that creative language tasks lead students to be more willing to grapple with complex linguistic constructions and ‘complexify’ their language to create new ideas (Tin, 2012, 179).

3. 3. Students were so motivated to complete their work that they would frequently ask to have extra time to complete it as homework. This is why after lesson 1, I switched to having them all write on laptops as it made it easier to submit their work after the lesson had ended. This is in contrast to previous prose composition lessons, where not a single student asked to be allowed to finish traditional practice sentences for homework; all were content to stop at whatever sentence they had reached as soon as the lesson ended. I would argue this is a clear indicator of a change in the students' intrinsic motivation levels. These findings chime with the literature which suggests that creative writing tasks are more intrinsically motivating for students and lead to a greater quality of written work. The ‘innate human drive to expression’ takes over and students write ‘with more care, dedication and… far greater output’ (Smith, 2013, 17). I have certainly been impressed, at times even moved by the creative output of the students, and can see the care and dedication both in their attitude during lesson tasks and the quality of the work they have handed in.

Findings from survey of student opinion

Enjoyment of Latin prose composition

Given my research aims to explore the impact of creative writing on students' enjoyment of Latin prose composition, I asked students to respond to the statement ‘you enjoy Latin prose composition’ at the start of the sequence and the end of the sequence to see if their feelings had changed.

Those who had, on the initial questionnaires, agreed or strongly agreed with the statements ‘you consider yourself a creative person’ and ‘you enjoy creative writing in your native language’ showed the biggest positive change in regard to their perception of prose composition overall. This category of self-identified ‘creative’ students comprised two thirds of the class – Emma, Sienna, Fiona, Bobby, Oriana and Evie – and it is interesting to note that it is only in this category that drastic changes occur regarding the statement ‘you enjoy prose composition’. Notably Fiona and Oriana moved two points on the Likert scale from strongly disagree to agree, as did Emma and Sienna from neutral to strongly agree.

In contrast, two participants in the group, Brandon and Jess, did not identify as creative people and disagreed with the statement ‘you enjoy creative writing in your native language’. Interestingly both their responses regarding their enjoyment of prose composition remained neutral both at the beginning and at the end of the sequence.

This suggests that creative writing as a means for improving a students' relationship with prose composition is most effective when the student already has a positive relationship with creative writing in their native language, and that for students who do not enjoy creative writing, other approaches to constructing engaging prose composition tasks should be explored.

Regarding the creative writing tasks themselves, seven students either agreed or strongly agreed with the statements ‘You have found creative writing in Latin enjoyable’ and ‘You have found creative writing in Latin more enjoyable than traditional prose composition’, while Brandon and Jess responded to both statements that they neither agreed nor disagreed. Students' enjoyment is also indicated in the optional comments box at the end of the questionnaire. Examples of students' comments include:

1. 1. Fiona – ‘This approach to prose composition made me see it in a whole new light! Really enjoyed!’

2. 2. Sienna – ‘You have more freedom – v fun!’

3. 3. Bobby – ‘I really enjoyed doing prose composition creatively.’

The results of the questionnaires are therefore overwhelmingly positive about students' enjoyment of the creative writing tasks and, more importantly, about their improved perception of prose composition as a whole.

Perception of Latin prose composition's difficulty level

It is interesting to note that the smallest change in participants' responses was regarding the perceived difficulty level of prose composition, that is to say the lesson sequence had the least impact on this area of their thinking about prose composition. In response to the statement ‘Latin prose composition is difficult’, seven of the nine participants reported the same response at the end of the sequence as at the beginning (all seven responses either agreed or strongly agreed with the statement), while the other two reported only a slight improvement insofar as their response moved one step along the Likert scale indicating they perceived prose composition as slightly less difficult than before.

The fact that the majority of students' perception of the difficulty level of prose composition did not change but their enjoyment of it did, is very interesting. Furthermore, eight of the nine students reported a more positive response to the statement ‘you are happy with your current level of skill in Latin’, with seven moving from disagree or strongly disagree to neither agree or disagree, and one moving from strongly disagree to disagree. Therefore, despite the fact that students perceived prose composition to be just as difficult as before, their sense of their own skill level had changed for the better, moving towards greater confidence. I would argue these findings suggest therefore that difficulty is not the primary cause for students' lack of enjoyment and confidence when it comes to the traditional approach to prose composition, and therefore lowering the level of difficulty need not be a leading concern when developing new approaches to it.

Some student comments in the optional comment box at the end of the questionnaire were revealing on the topic of difficulty. For example, Emma said: ‘Its still difficult… Although I still struggle with some vocab and constructions, the creative prose comp made it more fun so it didn't seem as much as an issue. And I wanted to do it.’

This comment reflects the findings already extant in the literature regarding creativity as an intrinsically motivating force (Smith, 2013); the student ‘wanted to do’ the creative writing tasks intrinsically. It is also interesting to note that, as with most of the class, Emma's perception of the level of difficulty did not change; however the negative effects of her ‘struggle with some vocab and constructions’ was apparently diminished due to the element of fun introduced into the activity. I would posit that by making the composition of Latin prose about more than just vocabulary and grammatical constructions, by making it instead about the communication and expression of original ideas, a far more positive experience of composing prose in Latin has been created.

Limitations

The reliability of my research findings is limited by several factors. I worked with a small group of students, in an academically-selective context, and so did not have the opportunity to examine the impacts of creative writing for mixed-ability sets, or lower-attaining students. Furthermore, none of my students have a recognised Special Educational Need or Disability (SEND), which means that their experience of creative writing in Latin may be very different from a student who, for example, has dyslexia. Furthermore, there are limitations with the use of ordinal scales such as I have used to gather data, especially as a level of subjectivity is required to interpret the results, and participants, for various reasons, may not always select options which are most truly representative of their opinions (Cohen et al., 2018).

A limitation regarding the application of my research findings is a practical one which I feel should not be ignored. The process of marking and providing meaningful feedback on students' creative work has been much more time-consuming than marking traditional English-to-Latin translation. The subjective nature of marking creative work coupled with the huge variability in student output makes it a far more involved process, which may not be feasible given a teacher's time constraints. There are possible ways that may work to mitigate this such as putting in place a system in which students mark each other's work, for example.

Conclusion

I set out, in my research, to examine whether creative writing in Latin could benefit students' enjoyment of and confidence in Latin prose composition. The findings of my research suggest strongly that it can bring significant benefits, particularly to students who already enjoy creative writing in their native language, but also, to a more moderate extent to those who do not.

A key and surprising finding of this research is that students a) chose to grapple with grammatical constructions more complex than those they would be expected to produce at A Level, and b) self-reported higher levels of enjoyment in prose composition and greater happiness with their current level of skill in it, while c) maintaining the belief that prose composition is not only difficult but more difficult than other aspects of their Latin A Level. This finding suggests that difficulty was not at the root of students' negative feelings towards prose composition, or at least was not its primary cause. Based on my findings I strongly suggest that lack of intrinsic motivation, rather than difficulty, is at the root cause of students' negativity towards traditional English-to-Latin prose composition, and that by tapping into the intrinsically motivating nature of creativity the students were able to meet the challenge presented by prose composition head on.

Moving forward, I believe more research into the potential benefits of creativity in the Latin classroom will yield rich results. The limitations of my research project are such that replication of my lesson sequence in other contexts will not necessarily be possible without significant adaptation. However, given the potential for creative writing tasks to be endlessly adapted and tailored to a class's particular needs, the requirement for adaptation should not be a barrier to replication of the core elements of my project, namely the introduction of creativity into the process of writing in Latin. Further research into the benefits of this approach for younger students, mixed-ability classrooms, and those with SEND, is desirable to understand in greater depth the full impact and potential of creativity as a transformational pedagogical tool in all Latin classrooms.