Using Comprehensible Input in the Latin Classroom to Enhance Language Proficiency

Abstract

This study originates from my curiosity to explore a different Latin teaching approach to address some students’ misconception that Latin is an unnecessarily complicated language that no one speaks anymore.

Introduction

This study originates from my curiosity to explore a different Latin teaching approach to address some students’ misconception that Latin is an unnecessarily complicated language that no one speaks anymore.

Having learned Latin and ancient Greek - both at school and university - through a strictly grammar-translation approach, I am almost tempted to admit that this misconception may have some reason. In stating this, I am not expressing disapproval of the grammar-translation approach because, on the contrary, I appreciated it so greatly that I chose a master's degree in Linguistics. I simply came to the conclusion that it does not always work for everyone. Even when it works, as in my case, it leaves some gaps to fill. Despite having an in-depth knowledge of how the language works, I have never felt confident enough to read an authentic passage in Latin with confidence and fluency. For the same reason, one cannot help but recognise that learning a modern language only through the grammar might expose one to the risk of being speechless in simple communicative situations, even when very complex linguistic structures have been mastered. However, as far as modern languages are concerned, linguistic studies over the past 50 years have brought significant innovations in the field of Second Language Acquisition (henceforth abbreviated to SLA). The aim of my study was to evaluate the impact of these theories, developed for modern languages, on Latin teaching. Encouraged by the practice of very active Latin teachers in this field (such as Keith Rogers in the UK and Robert Patrick in the US), I therefore created my Unit of Work (UoW) with the intent to convince novice Latin students that Latin ‘is not different’ (Patrick, 2011). To engage the students actively, I used resources based on the Teaching Proficiency through Reading and Storytelling (TPRS) approach that transformed my Latin lessons with opportunities for the students to discover a language that continues to live and speak today, by means of modern languages and authentic texts by ancient authors.

Latin in the National Context

In the past ten years the number of state-maintained schools in the UK offering Latin as a curricular subject has risen from 100 to 600, according to the data reported by Hunt (2016, p. 25). However, in the National Curriculum published by the Department of Education in 2014 the guidelines for the study of ancient languages are outlined only at Key Stage 2, with the clarification that ‘a linguistic foundation in ancient languages may support the study of modern languages at Key Stage 3’ (DfE, 2014, p. 253). At Key Stage 3 there is, in fact, no mention for Latin in the National Curriculum. Latin comes back at Key Stage 4, where Ofqual (2015) has outlined very precise assessment criteria for ancient languages. The absence of pre-established norms for teaching Latin in secondary school at Key Stage 3 gives teachers some freedom in creating their schemes of work.

Latin in the School Context

My placement was an independent, co-educational day school in Hampshire for students aged 11–18, well-known for its academic prominence, extensiveness of its extra-curricular opportunities, and its passion for sport and arts. Classics at this school comprises the languages of Latin and Greek, and the study of Classical Civilisation from Key Stage 3 to Key Stage 5. The teaching of Latin at Key Stage 3 until the first year of Key Stage 4 adheres to the principles of the reading approach offered by the Cambridge Latin Course. Latin is compulsory in the first year, whereas in the second year students can decide either to carry on with Latin or to opt for Classical Civilisation. With this configuration of the curriculum, Latin teachers have free will to select and adapt resources for the first years according to their needs.

Latin in the Class Context

I delivered my Unit of Work to a first-year class of 20 students, including 10 boys and 10 girls with an average age of 11.6 years. The scheme of work planned for this class involved the use of the Cambridge Latin Course Book 1 but, as mentioned above, this plan could be adapted according to the specific needs of the students. The Latin teacher had in fact decided to address the focus of his lessons on the historical and cultural background of Pompeii in the 1st century AD rather than on an in-depth study of the language. For this reason, it seemed to me very convenient to try out a communicative approach to Latin in this class, in order to allow the students to consider Latin as a modern language.

Literature Review

Rationale

Beginners in grammar are so overwhelmed by precepts, rules, exceptions to the rules, and exceptions to the exceptions, that for the most part they do not know what they are doing, and are quite stupefied before they begin to understand anything (Comenius, tr. Keatinge, 1907, p. 196).

Comenius is considered the most famous educational reformer of the 17th century. In his major work, entitled Didactica Magna, he discloses some contemporary educational ideas. His pansophic ideal is summarised in his motto omnia omnibus omnino, which indicates the need to extend education to all in order that ‘no man, in his journey through life, may encounter anything so unknown to him that he cannot pass sound judgment upon it and turn it to its proper use without serious error’ (Comenius, tr. Keatinge, 1907, p. 70).

Inspired by this pedagogical principle, introduced by Comenius in such an innovative fashion for his time, in my study I have reviewed modern language learning theories in search of an innovative approach to the teaching of Latin. As a result, I created a Unit of Work aimed to encourage first year students to reconsider Latin as an interesting and modern subject. As outlined in the introduction, I moved away from the traditional Latin class based on the translation-grammar approach and embraced current SLA theories. Terrell first in 1977, on the basis of his teaching experiences, introduced a ‘natural approach’ to modern language pedagogy, which gave importance to the content rather than to the form of communication and considered errors as a central part of the acquisition process. This approach was then scientifically theorised by Krashen and Terrell in 1983. Since then, Krashen has continued to influence language teaching with his own prolific publications. Considering that the areas investigated by Krashen are numerous and wide-ranging, I decided to narrow the focus of my analysis on his Comprehensible Input Hypothesis (1982) and, more specifically, on the Teaching Proficiency through Reading and Storytelling (TPRS) approach, introduced by Riley (1980) and promoted by Krashen (2015).

To appreciate how the teaching of classical languages can be re-evaluated in the light of modern language teaching studies, it is useful to retrace the different phases and theories that have contributed to the birth of the field known as SLA.

Early Second Language Acquisition Theories and Limits

In the second half of the 20th century, the first scientific studies on language learning developed from the debate between Skinner's theory of behaviourism (1957) and Chomsky's (1959) innatism.

Behaviourism

According to behaviourists, every form of learning - therefore also language learning - develops from a series of reactions to stimuli from the external environment, without involving any mental or internal processes. Pavlov's experiments at the beginning of the 20th century can help to exemplify this concept: a dog hears a bell every time he is fed. After a series of repetitions of this event the dog begins to associate the sound with food and thus begins to salivate as soon as he hears the bell. Even when the dog hears the sound but does not receive food, it still salivates. This type of behaviour is called classical conditioning and is the basis of Skinner's studies on operant conditioning. Van-Patten and Williams (2015) describe the concept as ‘a feedback system, in which reinforcement and punishment can induce an organism to engage in new behaviors: Chickens can learn to dance, pigeons to bowl, and people to speak new languages’ (VanPatten & Williams, 2015, p. 18).

Taking up the idea that learning consists of responses to stimuli from the external environment, positive reinforcement to a correct response would create repetition, while punishment would lead to abandoning the wrong behaviour. Likewise, children who learn their mother tongue reproduce sounds in response to stimuli they hear around them. If these behaviours are reinforced by a positive reaction from the mother, for example, they will certainly be repeated by the child, otherwise they will be abandoned. According to these theories, learning a second language (L2) would happen in the same way, that is, through an automatic repetition of stimuli received from outside, by analogy and without any mental effort or implication coming from inside. The limit of a theory that places language learning exclusively in the external environment seems clear enough. It should also be added that when it comes to learning a L2, learners must not only respond to new stimuli, but they also need to change the pre-existing habits applied to their L1. In fact, failures in learning a L2 are partly explained by behaviourists as negative transfer effects from L1; more specifically, developed habits in L1 may prevent learning new habits in L2.

Contrastive Analysis

The study of the differences between L1 and L2 contributed to the rise of Contrastive Analysis, which sees among its pioneers Uriel Weinreich and Robert Lado. Facing the theme of bilingualism, Weinreich (1953) analyses the contact between different languages at various levels. He states the obvious presence of the interference that:

…implies the rearrangement of patterns that result from the introduction of foreign elements into the more highly structured domains of language, such as the bulk of the phonemic system, a large part of the morphology and syntax, and some areas of the vocabulary (Weinreich, 1953, p. 1).

Analysing the question from a language learning perspective, Lado (1957) underlines the importance of a systematic comparison between L1 and L2 to obtain valuable clues in determining potential errors from the learner. Wrong habits in the L1 can negatively affect the learning of L2. Moreover, most of the errors would derive from the differences or overlaps between the two languages. Lado explains that:

Individuals tend to transfer the forms and meanings, and the distribution of forms and meanings of their native language and culture to the foreign language and culture - both productively when attempting to speak the language and to act in the culture, and receptively when attempting to grasp and understand the language and the culture as practiced by natives (Lado, 1957, p. 2).

However, Spada and Lightbown (2020) have reported that the Contrastive Analysis hypothesis showed all its limitations when it was tested. It was noted that in some cases it did not predict mistakes that learners made, in others it predicted errors that speakers did not make. Furthermore, L2 learners from different L1 backgrounds made the same mistakes. These results and the failures of behavioural theories led scholars between the 1970s and 1980s to believe that ‘there was, in fact, very little L1 influence in second language acquisition’ (Spada & Lightbown, 2020, p. 114). This conclusion has certainly been revisited in the light of subsequent research aimed at restoring the importance of L1 in language learning. However, such a rejection of both Behaviourism and Contrastive Analysis might have been strongly influenced by the theory of innatism introduced by Chomsky.

Chomsky's Universal Grammar

Chomsky is considered to be one of the most influential figures of modern linguistics. His ideas about language acquisition marked a profound change in the studies of applied linguistics. According to his innatist theory, there are universal principles underlying all human languages. In his review of Skinner's book Verbal behavior in 1959, Chomsky argues that experience and stimuli from the external environment have very little to do with children's acquisition of language. They would instead be biologically equipped with a system of abstract laws that govern every human language. He suggests that children produce sounds, words and utterances that go far beyond the stimuli they receive from outside and this happens because they were born with the special innate ability to discover the general abstract rules of the language by themselves thanks to Universal Grammar (UG). Chomsky stated ‘It is appropriate to regard universal grammar as the study of one of the essential faculties of mind. It is, therefore, extremely interesting to discover, as I believe we do, that the principles of universal grammar are rich, abstract, and restrictive, and can be used to construct principled explanations for a variety of phenomena’ (Chomsky, 1970, p. 387).

The theory of UG is widely accepted today to explain L1 acquisition (Spada & Lightbown, 2020). However, it is less clear what the implications are for L2 acquisition.

Krashen's ‘Second Language Acquisition Theory’ (1982)

Krashen's Second Language Acquisition Theory represents the most valid application of the principles of UG to second language acquisition.

In 1982, Krashen presented his theory by structuring it into five hypotheses.

1) The Acquisition / Learning Hypothesis

According to this hypothesis, defined by Krashen as the most important, adults develop competence in a second language in two ways: ‘The first way is language acquisition, a process similar, if not identical, to the way children develop ability in their first language’ (Krashen, 1982, p. 10). People acquire a language unconsciously and spontaneously, focused on communication rather than on form. ‘The second way to develop competence in a second language is by language learning’ (Krashen, 1982, p10). In this case learners are aware of the rules of the language and can speak about it. Learning is more knowing about the language rather than communicating.

2) The Natural Order Hypothesis:

Enlightened by research conducted during the 1970s on the order of acquisition of grammatical structures, Krashen believes that there is a predictable natural order in language learning such that some structures are learned before others. The limit of this hypothesis is that its examples are mainly based on research on the English language. Krashen also states that there would be no difference between the linguistic acquisition of an adult and a child because ‘that ability to pick up languages does not disappear at puberty’ (Krashen, 1982, p. 10). The issue is still controversial. On the one hand, scholars like Singleton and Ryan (2004) show that ‘with specific regard to L2 acquisition, it is no longer possible to accept the view that younger L2 learners are in all respects and at every stage of learning superior to older learners’ (Singleton & Ryan, 2004, p. 226). But on the other hand, the hypothesis of age-related differences in the acquisition of L2 is a very important and strongly debated issue: it seems that, in the long term, children immersed in a L2 natural context reach a higher level than adolescents and adults in the same conditions.

3) The Monitor Hypothesis

According to Krashen, the monitor is a specific tool of learning. It is a mechanism that controls or modifies the form of the message. This function is carried out if three conditions are met: (1) the learner must have enough time to develop the linguistic product; (2) the focus must be on the form of the statement; (3) they must know the rule involved.

4) The Input Hypothesis

Learning occurs when the learner, having reached a stage of knowledge, receives an input that is slightly above their competence and that they will be able to understand using the linguistic and extra-linguistic context.

5) The Affective Filter Hypothesis

For the input received by the learner to be taken in the learner must be open, without anxiety and fear, and their emotional filters must be very low. Only in this way will acquisition take place successfully. Krashen's study seems very convincing and every language teacher will have dealt with shy students who feel uncomfortable in the classroom. However, in the absence of scientific data, how is it possible to say how much the filter really affects acquisition? While Krashen claims that ‘research over the last decade has confirmed that a variety of affective variables relate to success in second language acquisition’ (Krashen, 1982, p. 31), Oxford states that ‘it is impossible to overstate the importance of the affective factors influencing language learning’ (Oxford, 1990, p. 140). However, recognising that the most capable language learners are probably those able to control their emotions, in his study Oxford lists ‘affective strategies’ that allow to decrease the affective filter (Oxford, 1990, p. 140).

It has been widely recognised that Krashen's ‘ongoing research has influenced language education for almost half a century’ (Ash, 2017, p. 71). However, it should not be forgotten that Krashen has also been criticised for the vagueness of the hypotheses that are sometimes difficult to verify through data (Spada & Lightbown, 2020).

The application of Krashen's Input Hypothesis to Classical Languages Teaching

Although many doubts about Krashen's hypotheses remain unresolved due to a lack of scientific data, the Input Hypothesis represents a valid way forward for all language teachers. Ellis suggests:

If learners do not receive exposure to the target language, they cannot acquire it. In general, the more exposure they receive, the more and the faster they will learn (Ellis, 2005, p. 217).

Krashen opens his introduction dedicated to the Input Hypothesis with clear and unequivocal words:

The Input Hypothesis claims that we acquire language in an amazingly simple way when we understand messages. We have tried everything else - learning grammar rules, memorizing vocabulary, using expensive machinery, forms of group therapy, etc… What has escaped us all these years, however, is the one essential ingredient: comprehensible input (Krashen, 1985, p. 7).

As for modern languages, the path is easily accessible because, even in a classroom context, it is possible to recreate for the learner conditions of total immersion in the target language. When it comes to classical languages the situation becomes more complicated as we do not have any feedback from native speakers, we essentially deal with a non-living language. Despite this, the question to ask if one decides to face this challenge will be, according to Macdonald (2011, p. 3), ‘Who will be the better reader of Virgil? The grammarian, or the student who can communicate in Latin and read it without pause?’

Some Latin teachers may object that integrating teaching techniques used for modern languages in the Latin class might be impracticable because the criteria established by Ofqual (2015) for the assessment of ancient languages at Key Stage 4 do not give enough liberty. As far as the language assessment is concerned, the requirements for the students are very clearly oriented to the translation of passages from Latin to English. However, Hunt has noted that ‘the Department for Education has not issued any guidelines for what a programme of study in ancient languages should comprise in Key Stage 3’ (Hunt, 2016, p. 10). Therefore, it seems worth trying to introduce at least at this stage – in schools where Latin is part of the curriculum – a teaching approach that adheres to the SLA results and is ‘aimed specifically at the acquisition of Latin with fluent reading as the goal’ (Carlon, 2013, p. 106).

There is another fundamental reason why introducing variation in Latin teaching is worth trying. As highlighted by Deagon (2006), a one-style-oriented approach – Deagon refers here to the traditional grammar-translation method but it seems appropriate to extend the reasoning to all the approaches that do not provide variation - runs the risk of discouraging students who do not find that particular style congenial because they learn differently. Deagon argues that, in order to gain interest in Latin from students with all kinds of cognitive styles, we must distance ourselves from the ideal of teaching based on ‘information transfer, and adopt an approach that teaches the processes by which language may be understood’ (Deagon, 2006, p. 45). Therefore, incorporating in the Latin classroom SLA theories, although researched for modern languages, could represent a step forward in this direction.

Unit of Work

Teaching Proficiency through Reading and Storytelling (TPRS)

Hunt (2016) notes that Latin teachers in most state-maintained secondary schools in the UK have chosen the reading approach – based on the Cambridge Latin Course – over the grammar-translation one that is more common in the independent sector. By contrast, in the USA ‘the grammar-translation and reading approaches have equally strong adherents, with the communicative approach in a strong third place’ (Hunt, 2016, p. 71). A form of communicative approach used in the USA is Teaching Proficiency through Reading and Storytelling (TPRS), which Rogers describes as an ‘approach which has enjoyed much success in the classroom’ (Rogers, 2019, p. 33). Some teachers, such as Patrick (2011) and his followers, promote the use in the classroom of stories told in Latin and question-answer sessions always in Latin. Patrick (2019) argues that students who are learning a language achieve better results when they receive comprehensible inputs, in an environment where they feel at ease, if the story is interesting and when the material presented is only slightly beyond what they have already acquired. As a matter of fact, this approach is beginning to spread to other countries including the UK, as demonstrated by a series of initiatives designed to bring a wind of change in the teaching of Latin.

Hunt (2016) claims that Clark's (2013) study has shown how TPRS is very successful in attracting students’ attention at the beginning but that, in the end, it risks losing its effect and causes teachers to return to traditional approaches. As suggested by Rogers during his lecture for Classics PGCE trainees at the University of Sussex in September 2019, it would be very advantageous to introduce the TPRS approach in small doses alongside other current methods to promote students’ interest in different aspects of the language. Encouraged by his enthusiasm and considering all his suggestions, I planned the following Unit of five lessons (see Figure 1).

Evaluation

Aims of the Unit

My interest in CI started after observing and teaching a few lessons in a Year 7 Latin class. Argetsinger (2006) describes a fairly common situation in the traditional first-year Latin class. She observes a sense of general boredom during the Latin lesson, due to the fact that some students are perfectly capable of mastering the language right away but must ‘sit idle’ and wait for other students who struggle to understand the language and translate (Argetsinger, 2006, p. 69). The sense of boredom of the more able students has the same negative impact on the lesson as the sense of bewilderment that affects the weaker ones. Argetsinger (2006) explains that this is due to a traditional (grammar-translation) approach to the teaching of Latin. Although a reading-approach was adopted in the first-year Latin class where I was teaching, I seemed to notice the same dichotomy. While two or three students were always attentive and interested, I realised that many others had many misconceptions about Latin. Some wondered if it was still possible to speak Latin today, others were aware that Latin is only a written language but, when tackling stories from the Cambridge Latin Course, they could not help thinking that they would never be able to understand the meaning of the whole text. Therefore, I came to the resolution that using a communicative approach in this class and making use of Comprehensible Input in Latin could be very effective for at least four reasons pointed out by different studies:

• 1. Enhancing competence and understanding (Ellis, 2005; Krashen, 1985a).

• 2. Promoting students’ curiosity with engaging material (Argetsinger, 2006; Patrick, 2011).

• 3. Developing fluency in reading (Carlon, 2013; MacDonald, 2011).

• 4. Offering variety for students of different learning styles (Deagon, 2006).

At the end of the five lessons, given the enthusiastic response of the students to the lessons and the good results they achieved in the end-of-unit test, the impact that the unit has had on the class is positive. However, it would have been necessary to devote much more time to a communicative approach whose effectiveness – according to Krashen's Input Hypothesis (Krashen, 1982) – depends on a gradual release of new linguistic inputs that are understood and therefore acquired only if just beyond the current level of competence.

Addressing Misconceptions

When I first introduced the unity of work to the class, I did not immediately tell the students that we were about to undertake an alternative path into the study of Latin. Inspired by some food for thought offered by Carlon (2013), I started asking them a simple question: ‘Why should you study Latin?’ The answers were, as I expected, quite confused. One girl said ‘Because it's mandatory’, so I replied, ‘Yes, it's true. In this school the study of Latin is compulsory in the first year. And why do you think it is that?’. Another girl said ‘Because we have to study two languages’. One boy continued ‘Yes, but why do we have to study two languages, one of which is dead?’ I then asked him: ‘Why do you think that Latin is a dead language?’ He replied ‘Because my mum says so’. After a few more minutes of discussion one student finally replied, ‘We study Latin because English and other languages derive from Latin’. He studied Latin in primary school, so he might have already come across this metalinguistic reflection. Therefore, building up this well-reasoned answer, I explained that not only is the study of Latin essential to increase knowledge of their own language and promote competence in other modern languages, but also that it offers the possibility to read ancient authors who continue to speak to us today in a very modern way despite their Latin words.

Compelling Material to Forget the Language

After the introductory discussion, I revealed to the students that we were about to face an important mission together: to reconsider Latin, in Carlon's words, as ‘a fully-functional and communicative language, one that can indeed be acquired just as readily as any modern language’ (Carlon, 2013, p. 106). To achieve this, throughout my entire unit of work, I have been considering one of the requirements of Teachers’ Standard 4, as outlined by the Department for Education: ‘To promote a love of learning and children's intellectual curiosity’ (DfE, 2013, p. 11). Therefore, I presented to the students the world of Legonium, where Lego characters speak Latin and live adventures in a small imaginary city.

Legonium is by Anthony Gibbins who, inspired by the works of Latin fabulae creators, decided to create his own Latin stories (first published in 2017) with the hope that they were not only fun to read but could also easily be included and adapted in the Latin classroom. The stories of Legonium, and other stories created by eager Latin teachers, are built around the TPRS approach discussed earlier. Krashen, in his theory of Second Language Acquisition, outlined the basics through the Comprehensible Input Hypothesis: ‘The best input is so interesting and relevant that the acquirer may even forget that the message is encoded in a foreign language’ (Krashen, 1982, p. 66).

When the pupils saw the first pictures of the PowerPoint I had prepared for the lesson, with Lego characters and Latin speech bubbles, they had a reaction of surprise and curiosity. The first question was: ‘Why are they dressed in modern clothes if they speak Latin?’ I thought that this question was a good starting point because it indicated first some awareness of the spatiotemporal position of Latin, and also an interest in the story rather than a focus on the language.

The target I set for the first lesson was to identify and use the structures to greet and introduce oneself in Latin. For this purpose, the introduction of Lego characters never seen before worked well because in the pictures they were presenting themselves to the students. The choice of using imaginary characters who speak Latin in PowerPoint presentations had already proved very effective in some previous Latin lessons with other students. According to Krashen (1982), compelling material elicits interest, allows the students to immerse themselves in the language without even realising it, and the distance between students and teacher decreases because the latter turns into an intermediary between them and the story displayed on the board.

The first word the students learned to say in Latin was salvē, a greeting addressed to one person only. In particular, with the help of magistra Livia (one of the main characters of our comic strips) they learned to say salvē sodālis – ‘Hello, friend’. When I showed them this comic and explained that magistra Livia was talking to only one student, the question asked by more than one student was ‘How do you say hello, friends?’ Therefore, I showed them the new structure without dwelling on any grammar point.

The Natural Order of Acquisition: is it applicable to Latin?

The decision not to overload the students with grammatical information during the different phases of my unit of work responds to the Natural Order Hypothesis included by Krashen (1982) in his Second Language Acquisition Theory. Patrick summarises Krashen's hypothesis by arguing that it ‘functions like an operating system on a computer - always there and running in the background, perhaps gathering data for later use, but never quite obvious to the user’ (Patrick, 2019, p. 40). Based on this hypothesis, people acquire languages in a specific order. We do not know what this order is for most languages. Even if we knew, it would be useless to create a syllabus that follows this order, because in that case we would be working on learning rather than acquisition – that is, we would be working on the form of the language rather than on communication. It is therefore essential for the teacher to keep in mind that the learner will acquire a grammatical structure only when he/she is ready. In the meantime, the teacher must continue to provide understandable input. Following these suggestions, the criterion that guided the planning of my five lessons was thematic and not grammatical. I started the first lesson by introducing the topic of greetings in Latin, then I continued with the linguistic structures at the basis of any beginner course syllabus that adopt a communicative approach: to ask and say the name, to express feelings to answer the question How are you?, and to talk about one's favourite animal.

By applying Krashen's hypothesis to the teaching of classical languages, it would seem that a traditional syllabus which gradually presents to students more complicated grammatical structures does not follow in any way the natural order of language acquisition. However, as argued by Macdonald (2011, p. 3) ‘The natural order of acquisition for classical languages is, due to paucity of speakers, difficult to deduce’. What I have observed in my short experience is that, although it is possible to develop a syllabus that follows a thematic rather than grammatical order in Latin – in fact the absence of native speakers does not represent a limit as long as the stories presented, although not realistic, are interesting for students – the real limit is students’ curiosity. In particular, for English-speaking students, it is very difficult not to wonder why the endings of verbs and nouns change. My position when questions of this kind were raised in the classroom was rather uncomfortable because I realised that as soon as I introduced some grammatical terms their expression changed and they became confused simply as they were not ready for that sort of explanation.

The affective filter

Oxford states that ‘the affective side of the learner is probably one of the very biggest influences on language learning success or failure’ (Oxford, 1990, p. 140). The scholar continues by arguing that negative emotions can inhibit even the performances of the learner who completely overcomes the technical difficulties of the target language. On the contrary, a confident attitude and positive emotions can make the difference and be conducive to the acquisition process. The role of the teacher, according to Krashen, is ‘to provide input and help make it comprehensible in a low anxiety situation’ (Krashen, 1982, p. 32). But how can we be sure that, only by producing understandable input, the learner gains more confidence? And how can we do it in Latin, a language that, as much as we can strive to make it alive, it is not spoken by anyone else but ourselves in the classroom context? Patrick (2019), elaborating Krashen, advises: ‘When we plan our lessons, they should focus on lots of understandable input, stories, and communicative tasks in which students forget that we are working in Latin and become lost in the flow of the language even in its simplest forms’ (Patrick, 2019, p. 42). Oxford (1990) also provides advice to help teachers have a positive impact on the ‘emotional atmosphere’ of the class. In particular, he suggests three techniques: increasing the use of natural communication, giving students more responsibility, teaching students ‘emotional strategies’ (Oxford, 1990, p. 140). The latter are indicated by Oxford with the acronym LET and consist of ‘Lowering Your Anxiety, Encouraging Yourself, and Taking Your Emotional Temperature’. Although Oxford is referring here to the context of modern language classes, these suggestions are clearly transferable to the Latin class. During my lessons I tried to follow both Patrick's (2019) and Oxford's (1990) ideas. I tried to create the ideal conditions for students to not feel any pressure on the language. In Lesson 3. I introduced the topic of animals, being aware that the focus would be on the pictures of Lego animals and on the discussion about their favourite animals. Hence, provided the resources available on Legonium website, I created PowerPoint slides to catch students’ attention. In this way, they did not only acquire new words, but they did also unintentionally use a grammatical structure that they already had come across in Lesson 2 to say their name. In this case the structure was slightly more complex due to the addition of the Latin word for favourite to indicate preferences. This technique refers once again to the Input Hypothesis theorised by Krashen (1982), according to which if the current level of acquisition is ‘i’, it will be possible to acquire further understandable input at level ‘i + 1’. Anything beyond this level becomes comprehensible through extra-linguistic information such as the context, the images, the activities.

Conclusions

When I planned my Unit of Work based on CI in Latin, I did not take into account a fundamental aspect that relates very well to Krashen's hypotheses - in particular the affective filter, but it applies to me as a teacher rather than to the students. For a person like me, who studied Latin with the grammar-translation approach, it is not so obvious to be familiar with Latin vocabulary and structures related to aspects of everyday life. In fact, I never happened to say phrases like ‘My favourite animal is …’ in Latin. During these lessons I was helped by the fact that the unit only included a few very simple linguistic structures. I keep wondering whether, when dealing with more complex structures embedded in a broader communicative syllabus, I would be comfortable in speaking a language that I have never really treated as a spoken one. According to Rogers (2019), the answer is being patient and accepting that at first our affective filter as teachers could be very high but, with a lot of practice, incredible results are achievable.

To conclude, the most evident limitation of my Unit of Work was the time available. Undoubtedly, in order for a syllabus based on the TPRS approach to be successful, teachers and students need more than five lessons. The starting point must be the foundation of a common vocabulary that could facilitate the reading of understandable and interesting stories. Starting from this experience I would like, in the future, to expand the practice to a whole year plan with the aim of, together with the students, rethinking Latin as a language that can be learned in a similar fashion to the modern ones.