NOTES ON PERSONAL LANGUAGE LEARNING EXPERIENCE

John Whelpton

The reminiscences and reflections collected here cover almost six decades of language learning, from childhood in Nottingham, where I was born in 1950, through study at Oxford (1968-72), teaching English in Nepal (1972-74), working as a civil servant in London (1975-81), graduate studies and teacher training in London, Nepal, India and Manchester (1981-87) to the last twenty-two years when I have been teaching English in Hong Kong but paying regular return visits to the UK and to Nepal. I began the compilation early in 1997, when I was teaching only part-time and occupied mainly with an intensive course in Cantonese and with work for an M.A. in Applied Linguistics. At the suggestion of my course director, Professor David Nunan, I had decided that my M.A.dissertation would be a diary study of my efforts with Cantonese and I needed a summary of my previous language learning experience as part of the exercise as well as for incorporation, in condensed form, in the eventual dissertation (completed in September 1998). I included any language which I had been formally taught for any length of time and also any others which I had worked at on my own over long periods, but not those which I occasionally looked at just out of linguistic interest or to learn a few phrases for short holiday trips. Earlier drafts were circulated to friends and colleagues for comments in the light of their own experience as language learners. I slightly updated it slightly in April 2001 and again in April 2009, after I had again become involved with modern European languages, and, as I am still interested in exchanging experiences of language learning, will continue using it in the same way as before. Because it was intended for readers brought up both in Europe and in Hong Kong Chinese society, I included explanations of points that are common knowledge to one group but not to the other . I also tried to link my reminiscences with some of the academic controversies over language acquisition. I therefore have to apologise for sometimes telling people what may seem either obvious or pedantic.

1.English: According to my mother's account, I began talking earlier than average. From my own memories of childhood, I suspect that compared with many children, my L1 learning was probably more from adults (parents and teachers) and less from peers. Later on, I became increasingly dependent on reading and radio, the latter becoming important as I had a lot of time off school because of my annual bouts of bronchitis. At primary school, my linguistic development was normally `ahead' of most of my schoolmates and I had some problems with peer relationships as I concentrated more on pleasing the teachers than following student norms. I had an early interest in story-telling - both co-operatively with my younger brother and at school (sometimes being requested to tell stories to the class), but this `creative' bent seemed to give way later to more analytical work. In contrast to the quick development of my verbal skills, my mother had a lot of trouble teaching me to tie my own shoe laces. I can remember at the start of Primary 1 still finding this difficult and my embarrassment at having to ask a female classmate to help me tie them quickly if they came undone. I continued to have slight problems with motor-coordination and spatial skills, never developing any real proficiency at ball games. I also tended not to notice easily incidental details of my surroundings, for example names of different shops in the town centre, perhaps as a result of concentrating on verbal intake and on my own thoughts. While I had a reasonable singing voice until it `broke' at 13, since then my natural vocal range has not been wide enough to sing comfortably. My normal speaking voice tends towards a monotone and, while undergoing teacher training in 1985/6, I was warned by a tutor that this would give me difficulties in maintaining students' attention and students have in fact occasionally commented on this. Though I have been much more geographically mobile than he was, my accent remains very similar to my father's - i.e. the (British) dictionary standard (`Received Pronunciation') modified by some northern features, which are not always detected by those who do not themselves have a full RP accent.[1] At my two primary schools, my classmates had a variety of accents and so presumably my inclination to follow parental and teacher norms was not offset by the consistent peer pressure which Labov (1992: 138, 304) sees as the normal determinant of a child's pronunciation (see also Kazakis (1970)). I thus had no reason to try to alter my pronunciation before adolescence and since then have never automatically assimilated to the pronunciation used around me, nor am I good at deliberate imitation of different accents: I cannot mimic properly the `broad' Nottingham accent spoken by some of my primary school class mates (on dialect acquisition v. Chambers 1992). However, I can be affected by syntax and vocabulary around me and this can alter my own speech even if definitely do not want it to do so. As a Hong Kong secondary school teacher, I have frequently told my students not to say `open the air-conditioner'when they mean switch on', but on at least one occasion in class found myself using the non-standard phrase - I realised the mistake as soon as I made it but had not been able to anticipate it. I also often use `bye-bye' on the telephone (especially with non-Europeans?), although this is Hong Kong English (in British English the word is typically used only by small children).[2] The pattern is consistent with Fathman's study of immigrant children in suggesting younger ones had an advantage in pronunciation but older ones in syntax and morphology.[3] On the other hand, if imitation of a foreign accent is regarded as a key component of L2 acquisition, my own case is perhaps counter-evidence to the suggestion that early acquisition of L1 is a predictor of success in L2 acquisition (Ioup et al).

My mother told me that she had made me basically literate by the time I started school but I have no conscious memories of learning from her and not many memories of literacy instruction in the early years at school: just as I have no clear memories of being unable to talk, I have no real memory of being unable to read. I do remember that at primary school I enjoyed reading and that a lot of time was spent practising handwriting. I recall I had trouble initially distinguishing `b' and `d' and I was told I initially produced right-to-left writing which had to be held up to the mirror for decipherment - probably this was connected with my left-handedness. I had a problem with spelling until some time into secondary school - my mother believed I was cured by the regular `punishment' of having to copy out misspelled words though I myself think implicit learning from extended reading was probably more important! I do still have some trouble remembering whether some expressions should be written as a single word hyphenated or written as two words and where I am unsure over the actual letters I often find that I have been unknowingly hesitating between a British or American version. I wrote quite fluently at primary and secondary school but at university found writing academic essays rather stressful. Possibly because I was aware of too many alternative ways of arranging the material, I wrote rather slowly and also had problems pacing myself in examinations; I frequently had to finish off my last answer in skeleton, note-form. This has remained a problem ever since, whether in the civil service or in my academic writing, though in the latter case interest in the subject matter normally outweighs the pain if I am not under too much time pressure.

Although, like L1 learners in general, I must have relied on implicit methods for acquiring both pronunciation and vocabulary, I do have some memories of conscious reflection on language. I remember (probably at the age of 7 or 8?) being fascinated by the fact that I could predict from the way the radio announcer pronounced the name of the second team whether the result of the football match was a home win, away win or draw. The ability to interpret intonation in this way is normally acquired considerably later than command of segmental phonology (Cruttenden (1974) cited in Crystal (1997: 243)),[4] so I was presumably reacting to a recent discovery, even though I have no conscious memory of a time when I was unable to predict the result. Just after starting secondary school, I remember making a conscious effort to remember one or two rarer words which I think I had come across in a Reader's Digest `It Pays to Increase Your Word Power' section. I was anxious to remember, and keep separate, the meanings of `prophylactic' and `exigencies', words which later must have been very easy to retain because of the high frequency of the Greek and Latin words from which they derive[5]

2.Latin: As a child in a Catholic family, I was exposed to Latin in church but perceived the language as pure gibberish. One of my earliest childhood memories is of my brother and myself arranging our stuffed toys in a circle so that I could pretend to be the priest, holding up a fake communion host and speaking nonsense syllables. My brother (two years younger than myself) and I used to tell collaborative stories in the bedroom about a make-believe world centred on a country called `Sillyvillage' with an incomprehensible language and Latin might possibly have been the model for this. I began formal study of Latin at 11 and enjoyed it immensely from the start. One of my Latin teachers, who was also interested in educational psychology, once (in my second year in the school?) asked the class which of us enjoyed Maths and preferred translating from English to Latin rather than from Latin to English (I was myself one of these). He then told us that this combination of preferences showed an analytical learning style.[6] I remember once in F1 our teacher dictated sentences in English which I mistakenly thought we were supposed to translate instantly and write down in Latin. I was unable to do this and consequently was always asking for repetition; I was concentrating so much on the task that I did not realise that if I couldn't keep up the others certainly wouldn't be able to do so. In fact, we were only supposed to write down the English to translate later and it was only because translation then came so naturally to me that I imagined we had to do the task at once. Just before starting on Latin at school, I had begun training as an altar boy, which then involved learning the Latin responses for the Mass: I have now forgotten most of these but still remember from that rote learning just the priest's opening words and our response:

Introibo ad altare Dei. (I will go to the altar of God) Ad Deum qui laetificat iuventutem meam. (To God who gives joy to my youth)

The switch to the vernacular made in the wake of the Second Vatican Council happened just at the point that I was beginning to be able to understand the Latin. In Form 2 I was one of three students whom the Latin teacher invited to study on our own, going through the book at our own pace and just meeting him every ten days or so for tutorials. Making rapid progress in this way gave me a great sense of achievement. I remember, for example, completing a Certificate Paper in the summer holidays at the end of F2. When I took the real Certificate exam in F4
(under a school policy which allowed selected students to take some subjects early) I finished one of the papers with about half-an-hour to spare and spent the rest of the time copying out the question paper whilst watching Fifth formers still working on their answers. Another feeling of perceptible progress came from being able to read whole books; in F1, when we could only handle short, simplified passages, I had been amazed by a sentence in my history textbook mentioning Julius Caesar's De Bello Gallico as something `you may one day read in the original Latin'. An unfortunate side-effect of working almost entirely on my own was the lack of any spoken input and also of the need to say things out loud myself. Although Latin was learnt as a written language and the standard school pronunciation for reading out aloud used purely English sounds, you were supposed to observe differences in vowel length etc. I never realised that I pronounced very badly, so it came as a shock when I was told about the problem by my tutor when I was a first year undergraduate, majoring in Classics (viz. Latin, Greek and ancient Roman and Greek culture). At university I also began to find English-Latin translation (`prose composition') a chore and my tutor became worried because I didn't seem to have any real sense of idiom or to remember stock phrases from Latin authors - I'd in fact earlier relied on a very mechanical, engineering approach to prose composition which did not work so well at a higher level. In general, though still having an interest in the subject, I did not get so much satisfaction at university level because, while still keeping ahead of college-mates in overall results, I no longer felt so completely in command of the subject. My Latin reading ability decayed after graduation in 1972 but was always a long way short of English reading speed. In addition, a kind of `translation' was always involved because once the `echo' of the actual words in short-term memory was lost, there was only a `propositional' memory in English with perhaps a few particularly striking words or phrases from the original (cf. MacDonough 1986).[7]. I do, however, still remember very clearly the Latin grace used at college dinners, which I often had to recite and which, apart from the responses for the mass, was the only Latin I had to use for non-academic purposes:

Miserere nostri te quaesumus domine, tuisque donis quae de tua benignitate percepturi sumus benedicito, per Jesum Christum Dominum nostrum, Amen. (Take pity on us we beg you, Lord, and bless your gifts which we are about to receive from your kindness, through Jesus Christ, Our Lord, Amen)

I began to give more attention to the language from around 1999 as I became aware of the Latin resources on the Internet, brought over to HK classics books earlier left in the UK and began to acquire some new ones. In summer terms from 2005 onwards till I retired from full-time teaching in 2010 I taught an elementary Latin course to interested students at my secondary school, relying mostly on the first part of the Cambridge Latin Course and the accompanying website.[8] In summer 2009 I was one of the instructors for the first two weeks of an intensive introductory Latin course at the Chinese university of Hong Kong,[9] The same year I began regular use of the weekly Latin news bulletins from Finnish Radio, making a point of listening to them before reading the printed text on their website and from 2007 uploading hyperlinked English glosses to the words in each bulletin on the Wordchamp language learning site.[10] I also began using the Latin locutorium (chatroom) to `converse’ with other enthusiasts and in 2008 joined the Grex Latine Loquentium, a group of around 200 members who email each other in Latin; I mostly just read others’ letters but occasionally contribute myself.[11] The same year, I joined SCHOLA., a Latin-language social networking site which works roughly like a simplified version of Facebook. After retirement from full-time teaching in 2010, I began writing to new members of this group informing them of other key resources on the web as well as publicizing my own classics page (http://linguae.weebly.com/latin--greek.html) All of these sites form part of a wider `Living Latin Movement’, which, despite the near collapse of Latin as a school subject in many countries in the `1960s, is attempting to reinstate the use of the language, in both written and spoken form, for actual communication. I first became aware of in the 1990s when I came across Wilkes et al. (1993); this teaches in Latin exactly the same content as found in the authors’ parallel volumes on French and other modern European languages.[12] I am now (May 2011) teaching Latin to several private students and normally add some practice of spoken Latin to more conventional instruction.

In summer 2010 I attended a `Seminarium’ or Latin-immersion conference run by Sigrides Albert in Trier. As well as formal sessions in which expert speakers delivered papers in Latin on various topics, all participants had to prepare a short presentation on a Latin word they had been given by the organizer. [13] We were also supposed to continue the use of Latin in informal conversation over meals, during sight-seeing excursions etc. and, although this required a lot of effort from the novices, most people kept to this rule. This was an enjoyable experience though neither this, nor a number of Skype conversations which I have had from time to time with other Latinists,[14] were enough for me to develop real fluency. In Trier, I found that I could generally understand most speakers but had difficulty with one or two, either because of their pronunciation or their use of particularly complicated language. In my own speech, I was very much aware of still having to mechanically assemble words and select endings, except for a small number of phrases that functioned for me like lexical items. However, since all but a small minority of practitioners of spoken Latin have a similar problem, I don’t have the same feeling of inadequacy I can feel when struggling with a modern language and so the psychological pay-off is quite high, even though I would still like to be fully fluent.

I have attempted to make Latin conversation practice available in Hong Kong through the Circulus Latinus Honcongensis, established at the end of 2010 and normally meeting once a month over dinner in a restaurant.[15] Circuli Latini in most country are attended by people who have a good knowledge of the written language and just need to get used to listening to Latin and to speed up their own deployment of phrases they already know. In Hong Kong, some attendees are near-beginners in the language so some use of English has to be allowed and sessions can resemble elementary Latin lessons rather than conventional oral practice.

All this activity, particularly listening every week to Nuntii Latini has considerably improved my ability to listen to the language but I still do not read it as fluently as I do English. I can read the very straightforward Helsinki news bulletins and also some of the Grex emails about as easily as I do French but with the emails written for literary effect or on philosophical topics and with classical authors using a more complete style or with poetry I still have trouble. My reading technique in such cases is probably in some ways reminiscent of that of HK students grappling with English textbooks, but I normally get the main structure of each sentence fully understood, whilst probably missing nuances.

Both in teaching and in private use of the language I try to pay attention to the pronunciation but still have difficulty if a text does not have the long vowels marked. My knowledge of vowel quantity has, however, improved a little in recent years, since modern published teaching material does normally have macrons (length marks) and I myself, when preparing materials of my own, spend a great deal of time consulting dictionaries to add them

3.French: I studied French at secondary school for five years. This was also taught by grammar translation but pronunciation was stressed more than with the classical languages and dictation exercises were important. I did well with translation and composition exercises but I had a poor accent (even by British standards!) and difficulties with sound discrimination, resulting in problems with dictation. Because of the close similarity between the two languages, reading was not difficult, but initially I got confused between the acute (/) and grave (\) accents - paralleling the confusion of b/d I think I initially had with written English.[16] Throughout F1 and in the opening months of Form 2, we had a French teacher who was completely unable to maintain discipline. Amidst chaotic classroom conditions, I got used to studying on my own. Later I read about two-thirds of Les Misérables and also enjoyed looking at anthologies of earlier French literature, including material on the sound changes from Latin to French. I got my first opportunity to use the language `for real' during a hiking holiday in 1968, two years after finishing formal study of the language. I managed to communicate and kept up a short correspondence in French with a girl we had met on the trip. She once asked me to write in English so that she could practise the language but, when I did so, she wrote back explaining she couldn't understand and asking me to revert to French. I think the resultant feeling of having the key to someone else's room when they haven't got the key to yours may be one of the important motivators in language learning for me and for a lot of other people.[17] In later years, I occasionally needed to read French texts as an undergraduate and I continued to do so when researching on Nepalese history as well as occasionally reading French for pleasure. I became more involved with the language again from 2003 when I asked a junior colleague who had recently studied the language as a subsidiary element for her degree to teach elementary extra-curricular classes at my school. She was unavailable to teach in subsequent years and when unable to find a native speaker or proficient second-language speaker I started giving the class myself, using my colleague’s materials plus my own supplementary materials and relying as much as possible on recordings of authentic speech.[18] Although I could read and (at an elementary) level write the language, problems with listening and speaking remained. I visited Paris in 1981 to consult newspaper archives in the Bibliothèque Nationale and was able to do this without much difficulty but found conversation with the library staff difficult. The pattern was similar when I was there in 2008 for a conference (conducted in English) on Nepal and again on holiday and attending another conference in 2010, even though in winter 2008-9 I began occasionally attending meetings of a French conversation group in Hong Kong. None of this has been sufficiently frequent to make any significant improvement. . My younger brother specialised in modern languages and studied French formally until the end of his first year at university. In contrast to me, he acquired a reasonable accent and, after a year working in France between school and college, was able to function efficiently in the language. A wish to emulate this achievement may have been one of the reasons behind my eagerness to become part of the local language community whenever I have myself worked abroad.

4. Greek: I started studying Ancient Greek in F3, and, as with Latin, was soon mostly working on my own with occasional `tutorials'. Also as with Latin, my pronunciation was poor but I did not realise this until I got to university. Before beginning the subject I was worried about having to use a new alphabet and imagined I would need to transliterate words into Roman script to aid memorisation. In fact, because the Greek alphabet is so similar to the Roman one, which developed from the Greek, there was no need for this. As was usual in secondary schools at that time, we were not required to memorise the accents used in printed Greek texts. I made an attempt to learn the system at university but was not very successful - the written accents do not affect the way Greek is pronounced in western Europe and so they must be memorised simply as arbitrary spelling conventions. If you have neglected such conventions in initial study, it is very difficult to make up for it later. School and university Classics courses do not normally include any coverage of Modern Greek, which bears a similar relation to ancient Greek as does Italian to Latin. I spent a little time studying this on my own, before a visit to Greece in 1970, but never developed more than a phrasebook knowledge. After graduating in 1972, I did not often look at a Greek text, though since 1999 I have had more of my Greek books with me and browse in them from time to time. When reading the (Romanised) examples of modern Greek in Smith and Tsimpli's 1991 and 1995 studies of `Christopher' I enjoyed reconstructing mentally the Greek orthography and the Ancient Greek pronunciation. I have sometimes read the news reports in classical Greek posted regularly on the Akropolis world News site (http://www.akwn.net/ ). As with the better-known Latin bulletins from Finnish radio, these are considerably easier to read than most classical authors. Most recently, as a spinoff from my interest in conversational Latin, I have begun collecting links to web resources for spoken classical Greek and also for introductory courses in the language but have put much less time into this than I have done with Latin

5.German: Probably in F4, I took extra-curricular lessons in German from one of the Greek teachers. He commented on my poor pronunciation and said that, though I had no ear for languages, I definitely was a linguist and therefore well-suited to specialise in Classics (i.e.Latin and Greek). By this time I had already decided myself to study Classics in F6/7, abandoning ideas of becoming a scientist which I had entertained when first starting at secondary school. I only really played with German and never got a solid grasp of the vocabulary (unlike French, the longer words are normally not recognisable from English loan-words or cognates). During my undergraduate years, German, while not essential, would have been useful as some important work in classical studies was available only in German. However, despite some attempts I never got on top of the language. In summer 1969 I spent a few days hiking in the Rhineland and my brother and I had enough stock phrases to order meals or ask for permission to sleep in farmers' barns but not to hold real conversations. While we were in Germany, the Northern Ireland `Troubles' erupted and, with no English newspapers available in the small village where we were staying, I remember standing outside the closed office of the local paper, reading the front page coverage on display there. In 1976, when employed in the Ministry of Defence in a job which involved attending occasional conferences with German representatives, I spent some time listening to elementary tapes but not long enough to produce any real improvement. Whilst working on Nepalese history I once or twice struggled thorough an entire book in German (most recently Karl-Heinz Krämer, Nepal – der lange Weg zur Democratie ), but otherwise had little contact with it until 2005-06 when, under my school’s Europe Club programme, I got an exchange student to give a few lessons and, as with French and Spanish, produced some elementary materials myself. In 2008-09 I did this more elaborately, for delivery by another exchange student, drawing mainly on material in Wilkes and Shackell’s German for Beginner’s and on dialogues on the Internet. However working at this elementary level has not been enough even to properly consolidate my knowledge of German’s complex morphology, let alone develop any spoken competence..

6. Nepali: Before learning in 1972 that Voluntary Service Overseas would send me to Nepal rather than India or Pakistan, I did not even realise Nepali was an Indo-European language. On discovering that the language was in fact closely related to Hindi and written, like Hindi and Sanskrit, in the Devanagari script, I started learning the characters - 42 basic ones, with modifications when they are used in combination.[19] On arrival in Kathmandu, I was given an audio-lingual course over about seven weeks. Class sessions, with one tutor to about four students, were conducted mostly in the target language but we could study the grammar and learn vocabulary from our bi-lingual textbook. Before any written material was issued, we had a brief introductory session seated cross-legged on the floor at the American Peace Corps training centre. We were taught the structure Tapainko nam X ho ( Your - name - X - is, viz. `Your name is X') and, partly under the influence of my training in Classics, I at first interpreted the ho as a nominative case termination belonging to the noun! The teacher's use of `direct method' could not prevent me constantly analysing the input and, of course, doing so in L1. Because we were all used to a more deductive method of presentation, several of us collectivelyprotested against the ban on the teacher giving us grammatical explanations during class. We cited an American research project we had recently been told about by our British Council TEFL instructors as evidence that the deductive method could in fact be more effective than the one then in vogue.[20]
In later years, however, I looked back on this course as successful and as a teacher become (excessively?) eager to minimise the use of LI during language lessons. While under formal instruction we also got some practice outside; for example, a monolingual Nepali cook was working for myself and two classmates. However, I was myself told by one of the language instructors that while I was good in the classes I did not make enough effort to initiate use of the language outside; I remember telling her (or accepting her suggestion?) that this was just a matter of character. Although I absorbed the vocabulary, my pronunciation was poor; one of my classmates, who was herself an accomplished linguist, thought that I was deliberately mispronouncing the Nepali as a kind of joke! Another classmate remarked some years later that there was a very competitive atmosphere on the course. I don't myself remember this being excessive during those initial two months but it was certainly true that volunteers' status among their peers rested in part on their ability to acquire good Nepali and also to `fit in' with local society. I was immediately able to put the language to use during a two-weeks trek and then in the town on the Indian border where I was sent as a college lecturer. I think I actually learned new vocabulary and structures through use of supplementary textbooks (e.g. Clark's Introduction to Nepali ) and then gained some fluency by interaction with native speakers rather than acquiring new language directly from what was said to me.[21] A large proportion of my early practice was with the young son of my landlord, whose English was very limited. There were some communication difficulties, and I remember once, when he had erased part of one of my music tapes, I had to use the dictionary before I could tell him Timile sangita nasta gareko chau (`You have destroyed the music'). In my work as a teacher, I was frustrated at not being able to communicate with my students in their own language well enough to explain particularly difficult points. For example, when it seemed students in an `English for Science' class did not understand the concept of a controlled experiment, I could not be sure whether the problem really was conceptual or purely the result of the language barrier. I was also frustrated because I felt I did not get the same respect from the students as did teachers fluent in Hindi or Nepali and felt less powerful than my colleagues because, although my English was a little bit better than theirs, I was immeasurably less proficient in the local language. But there was also some gratifying success: I recall one class, knowing I could read Nepali, asking me to read out a notice sent round by the principal rather than handing it to a student to make the announcement. After transfer to a Kathmandu college in spring 1973, I worked through a short novel and then had to use a lot of Nepali dealing with another monolingual cook (a bright 18-year-old from the northern hills who had had virtually no formal education). My flatmate had himself been working in the hills where there was greater need for Nepali than in the town where I had first taught and he was a little better than me at communicating with the cook but, because of my interest in the written language, I was the one who supervised the writing down of household expenses in a Nepali accounts book. I also still remember my feeling of satisfaction the first time I left a simple note for the cook explaining in Nepali when we wanted dinner. I also continued trying to improve my reading by working through Brahmand ra Saurmandal (The Universe and the Solar System), a book on cosmology written by one of the lecturers at my college. The strong contrast between my attitude and the more instrumental one of locals learning English was highlighted by a student who came upon me writing out Nepali vocabulary and asked me, `Sir, why you do all this labour? You already know Nepali.' He knew I could already give simple instructions and ask for things at a shop and could not understand why I wanted to do more than that. My flatmate and another British volunteer who regularly ate with us also showed no interest in extending their basic, functional command of Nepali. This was partly because they were already planning to leave Nepal at the end of their first year, but also perhaps because they were Modern Language graduates who had already achieved full command of a second language and so it was no longer so important for them to `prove themselves' by mastering a new language. After their departure in summer 1973 until I left for the UK in December 1974, the cook and myself were the only occupants of the flat and I got some satisfaction from the feeling that, only a year after beginning study of the language, I was in effect the head of a Nepali speaking household. I also needed to use Nepali with other monolingual or near-monolingual Nepalis, including a youngster who I had given some lessons to in Birgunj and who later stayed with me for some time in Kathmandu. It was after building up a basic fluency with people like this that I began using more Nepali with some of my English-speaking Nepali friends. In February 1974 the Birgunj youngster and I had inadvertently wandered into the inner court of the Hindus-only temple of Pashupatinath just outside Kathmandu and were put under `citizens' arrest' by an indignant crowd. I subsequently gave a lengthy account of this in Nepali to a lecturer friend, with whom I now normally speak in English but often correspond with in Nepali.[22]In the autumn of the same year I made several journeys through the hills investigating possible job opportunities for future VSOs and normally spoke Nepali with all but the most senior local civil servants. On one occasion, the Chief District Officer summoned a meeting of the town panchayat (council) for my benefit. My Nepali performance was rather strained, but the one or two members speaking to me carried on without any attempt to switch. They may have felt even less confident about their own English, although one of the members watching silently, ex-prime minister Matrika Prasad Koirala, was certainly fluent in it and probably rather amused by my performance. Despite managing one-to-one communication in Nepali reasonably well, I was disappointed that I could not properly follow native speakers conversing with one another. This was particularly galling when there were student demonstrations in summer 1974 protesting the `integration’ of Sikkim into the Indian Union. I remember my frustration in a Kathmandu restaurant when some excited students came in but I was unable to eavesdrop on their conversation. In summer 1974, a few months before returning to the UK I began keeping a journal in Nepali. I discontinued this some time after coming back to the U.K. and beginning a job in the civil service but resumed when I became a graduate student at SOAS in 1981 and, with a brief interruption when highly disorientated after beginning work in Hong Kong, have maintained the habit ever since. Writing in Nepali in this way was easier when I was also speaking it in daily life and has seemed rather artificial in recent years, but I am reluctant to discontinue the practice. After return to the UK I also kept up correspondence in Nepali with some Nepali friends and did some reading. In summer 1979, on the suggestion of a Nepali friend then doing postgraduate work in Edinburgh, I began a translation of a 19th. century Nepali text (Whelpton 1983) and this led on to graduate work and subsequent research projects up to the present, involving both extensive use of written Nepali sources and also interviews in Nepali. In the academic year 1982-3 I was again in a Nepali-language environment, spending about 8 months in Kathmandu in the home of the Nepali friend who had suggested and helped with the translation project. There was also a shorter period in New Delhi with a (third!) Nepali-speaking cook who had no English. In Kathmandu, although my friend was a university lecturer in English and his wife was also well educated, I adopted a strict Nepali-only rule, waived only if we had to discuss something highly technical (e.g. his plans to lecture on the development of comparative philology in the 19th. century) or if we were with non-Nepali speakers. I have stayed regularly with the same friends on shorter visits since then but the Nepali-language environment has been attenuated somewhat because I began speaking English to his elder son to help the latter's acquisition of the language. The son is now completely fluent in English (speaking it arguably more naturally than his father since he grew up with English TV programmes and visits from English-speaking friends) and I use a mixture of English and Nepali with him and also with a younger brother. We also naturally use mostly English when my wife, who does not know Nepali, is with me on visits. My fluency in Nepali was achieved within a limited vocabulary and range of topics and my pronunciation has remained very poor. Many foreigners in similar circumstances to mine achieved a better command of the spoken language and Narendra Mani Dixit, an elderly Nepali scholar renowned for his own command of English, once playfully told me that my Nepali was `execrable'![23] Rather than compare overall performance, however, it is perhaps more useful to note the difference in the balance of skills between my VSO contemporaries and myself. My reading ability and knowledge of the formal vocabulary drawn from Sanskrit was greater, whilst my pronunciation and comprehension of everyday conversation was poorer. I recall one of my 1973 flatmates understanding better when our cook, rather than speaking as usual about household matters which directly concerned us, related an incident in his life outside. Twenty-four years later, when acting as bestman at the wedding of another VSO contemporary to a Nepali, I noticed that the ex-volunteer understood his bride's rapid Nepali much better than I did but that, when we both had to use some Nepali for sections of our speeches at the reception, I spoke much more fluently. The groom's agricultural extension work in the hills in 1972-4 had involved constant use of Nepali and he had clearly retained a very good receptive knowledge whilst his active command of the language had deteriorated. In contrast, my reading and writing and more frequent return visits to Nepal allowed me to marshal words and phrases more easily, even if the pronunciation may have been less accurate. Despite my lack of all-round competence, I obtained occasional work in 1984-85 as a teacher of Nepali in London, and the then principal lecturer in Nepali at London University accepted me as an assistant because I could give grammmatical explanations in English following the system of analysis in the textbook he had himself written (Matthews 1984). When giving one lesson to a British Council officer about to leave for Kathmandu, I remember being very nervous because a Nepalese acquaintance was going to be present in the room. I was confident that my general, declarative knowledge of Nepali was adequate and, having already been frank with the student about my limitations, I was prepared to be corrected on particular linguistic points, but I was worried that I might fail to understand the native-speaker's Nepali when he was pointing out my mistakes! In fact, I did write one character wrongly on the board and the Nepalese commented on this; luckily, although not fully understanding what he had said, I guessed correctly what the problem was and made the necessary amendment. I think I was able to get as far with spoken Nepali as I did partly because it is reasonably close to the written form, so my stronger visual skills could support my weaker audio-aural ones. Another important factor was people's willingness to listen and speak to me despite my inaccuracy. This attitude was in turn made possible by several factors. First, the phonological structure of Nepali is sufficiently close to English for the language to be comprehensible (though rather grating on the ear!) if spoken more or less as if it was English. Second, Nepali's status as a lingua franca in a multi-ethnic country has made native speakers more tolerant of divergence from the standard variety. Third, English is not a medium of instruction in government schools nor a language of administration so it is quite common for officials below senior ranks to have a poor command of it and to prefer speaking in Nepali if a foreigner knows it. Fourth, a foreigner's social and economic status in a very poor country also makes it likely that mono-lingual Nepalis will be motivated to associate with them and to accept any difficulties in communication (this factor explains why British Gurkha officers can go on making the same mistakes throughout their military careers without protest from those they speak to.[24]) Fifth (though this may just be a subjective impression), many Nepalis exhibit an open and friendly attitude, contrasting with the wariness that is often encountered in India. In my own case, it was particularly important that, for whatever reason, people would quite often initiate a conversation with me thus compensating for my own tendency not to `risk' interaction unless the situation appeared to invite it in some way. Sixth and finally, there was the tolerance of English-speaking Nepali friends who have been prepared to provide me with a kind of `language-shelter' on a long-term basis. In reading Nepali, I use much the same technique as formerly with Latin and Greek - i.e. reading much more slowly than with English and frequently needing several `runs' over difficult sentences and also sometimes consultation of dictionary before understanding fully. My slowness is partly due to the script being considerably different from Roman, but lack of familiarity with the less common words and problems identifying immediate constituents in a complex sentence may be more important factors (cf McLaughlin et al. 1983 - section on reading). My spoken ability has perhaps deteriorated slightly over the last few years although I have continued to make annual return visits. The reduction in time spent in genuine interaction in Nepali, combined probably with frustration over my failure to make as much progress as I had hoped with Cantonese, seems to have put me under increasing tension. In particular, I often feel frustrated when speaking English with Nepalese, including the children of my old Kathmandu friends. Although this use of English is generally the most efficient means of communication, I am both unhappy at feeling less closely integrated into Nepali society and also irrationally resentful of Nepalese who manage to get a really good command of English in contrast to my own mediocre Nepali. As with Latin and Greek, I made good progress when setting off to climb the mountain but then found my path blocked well short of the peak.

7.Hindi: When posted to Birgunj on the Indian border immediately after finishing my Nepali-language training in 1972 I found that Hindi rather than Nepali served as the lingua franca in the staffroom as well as in the town generally. I spent a lot of time in the company of an Indian lecturer in the English Dept., accompanying him on a round of visits to Hindi-speaking friends and hoping that I would somehow just `pick it up'. In fact, I learnt almost nothing this way: I merely noticed the prevalence of hoga (may/will be), ekdum (very, exceedingly) and chelenge (will go). It was only after getting to Kathmandu and spending some time with Hindi grammars (particularly Macgregor's) that I managed to develop a reading knowledge and basic conversational ability, helped by the close relationship between Hindi and Nepali.[25] I also used the language sometimes to write to Hindi-speaking friends and in 1979 attended evening classes in London (I think an hour and a half weekly for one term) to revitalise it. However, I have never developed the kind of fluency in Hindi that I have in Nepali. Even had I spent more time in India, the resistance to the use of Hindi to Westerners by many educated Indians would probably have been a problem (cf. Gumperez quoted in Gibbons 1987).

8. Sanskrit: Whilst still in Kathmandu I purchased a Sanskrit grammar and continued rather casual study of the language on my own in the U.K. before finally taking the first year Sanskrit course at London University when I began postgraduate work there in 1981. I was able to make early progress because the grammar is quite similar to that of Greek and Latin whilst the script is the same as for Nepali and Hindi, which evolved from colloquial Sanskrit and still borrow `learned’ vocabulary directly from it.. However, I never became fluent in reading original texts and I have made little use of the language in recent years.

9. Italian: The Ministry of Defence posting which involved contact with German speakers (section 5 above) also took me to Italy and motivated some self-study of Italian. I also enrolled for two months or so in a weekly evening class. The similarities with Latin and French made it extremely easy to read simple texts but was not in an Italian-language environment long enough to make progress in conversation. My reading knowledge has now also decayed and, if presented with an Italian text, I am now dependent largely on my general familiarity with the Romance language group rather than my memory of specific features of Italian.[26] This was still the case when I produced a number of lesson units in elementary Italian for delivery by a native-speaker exchange student at my school in 2004. As with other European languages, I drew material from Wilkes and Shackell’s series and from material on the Internet.

10. Portuguese: Because a teacher training course in 1985/6 included six weeks teaching practice in Lisbon, I leant some basic grammar and vocabulary on my own over six weeks or so before leaving Manchester, using Hugo's Portuguese in Three Months. As with Italian, familiarity with Romance languages in general made it easy to obtain an elementary command of the new language. Pronunciation problems and the short period of time in Portuguese-speaking society made it impossible to develop any real fluency but I managed basic communication with hotel-staff, shopkeepers etc. I enjoyed occasionally acting as interpreter' with hotel staff for classmates who had made no attempt to learn the language, but I was frustrated at having to speak English all the time with teachers and students at the school where I was attached. On the other hand, developing a reading knowledge of the language was very easy and while in Lisbon I read one or two newspaper articles each day - the same learning strategy I first applied with Nepali and which I afterwards used for Chinese. I was also able to read Portuguese materials for some of my university assignments

11.Cantonese: Throughout over twenty years in Hong Kong I have normally been in a Cantonese working environment, working from 1987 to 1996 and again since autumn 1997 in secondary schools, where I was either the only member of staff not fluent in Cantonese or one of only two such teachers. In 1987-89 and again in 1996-97 I had one British colleague. In 1988-91 I also had a Malaysian Chinese colleague who knew little Cantonese and preferred to speak English with anyone (Chinese or non-Chinese) who was fluent in it. Based on my Nepalese experience, I had expected to acquire basic competence in Cantonese quite quickly but found that progress was very slow, despite attending Cantonese courses from time to time, including over a hundred hours (class and language laboratory time) at the Hong Kong Bible Language Centre in 1988-90 and self-study from various textbooks. Not only could I not follow properly discussions between those around me, but phonological problems (both in reception and production) and lack of fluency and limited vocabulary made one-to-one conversation difficult. Factors holding me back seem to have been:

A. The fact that I rarely needed to say anything at all elaborate to anyone whose English was not better than my Cantonese (in the government quarters where I lived for my first four years, my room was cleaned by a Chinese amah but, in contrast to my Nepalese cooks in earlier years, her duties were laid down by the management of the building and I did not need to give detailed instructions, discuss pay or household expenses etc.)

B. Reinforcing the practical difficulty of communicating in Cantonese were the prestige factors implicit in a diglossic situation with a colonial background (Pennington 1994, Gibbons 1987 and Whelpton 1998). In addition to wishing to assert their own status as competent users of English, many educated local people also seem to find it difficult to understand why foreigners would want to learn to speak Cantonese other than for strictly transactional purposes (e.g. shopping in street markets); a small-scale survey (limited to tertiary institutions and a few businessmen) conducted by David Li and Jack Richards did indeed suggest that most expatriates (other than special categories such as missionaries and the police) reporting a wish to learn Cantonese have transactional rather than interactional reasons (Li & Richards 1995: 4), whilst the motivation of Cantonese learning English is itself primarily economic.

C. I have always pronounced foreign languages poorly, but the problem became particularly acute in Cantonese with so many difficult distinctions (tonal, vowel length) being phonemic. Imposing the phonetic structure of English on French or Nepali does not prevent relatively efficient communication taking place (especially once someone is used to our pronunciation) but it is a more severe handicap with a language as different from English as is Cantonese. This difficulty, of course, reinforced the problem of lack of real necessity to communicate in the language (para A above). The latter factor made it more likely that, even if conversation began in Cantonese, either my conversation partner or myself would initiate a switch to English.

D. Since I rely heavily on the written code when learning a language, the non-phonetic nature of the Chinese writing system (and the fact that most writing in Hong Kong is in standard Chinese rather than Cantonese) meant that written Chinese, which I was studying at the same time (see below), was of relatively little help. With hindsight, I should probably have spent more time reading as much Romanised Cantonese as possible, since there would be more reinforcement for my deficient audio-oral skills (cf Bell, 1994); however, this would have gone directly against what was for me the most natural way of tackling a new language, viz. moving as quickly as possible to receptive mastery of the writing system which the speakers of that language themselves use.

E. The thousands of hours of exposure to Cantonese spoken by native speakers were of limited use, since (1) such input is not generally comprehensible to the learner until he reaches an advanced stage (cf Schmidt Frota 1985 - Schmidt is still often uncomfortable listening to one Arabic native speaker conversing with another) (2) even when I guess the general meaning from key words and the context, the `unknown' stretches are generally too long and/or (especially for someone without a good `ear' for languages) not sufficiently clear for me to learn from. The `eavesdropping' strategy preferred by John Schumann (F.E. & J.H. Schumann 1977) is of limited value to me for increasing my communicative ability, nor does it improve my pronunciation, since I automatically impose English phonological structure on whatever I hear. The key problem is, of course, that the learner cannot control the speed of communication with listening in the same way that he can with reading or with his own output (c.f.F.R.Jones (1994: 450-1) on his own experience with Hungarian: `when I had the opportunity to interact with native speakers..understanding them proved difficult;...in contrast to speaking, this hardly seemed to improve over time....With listening...I had little direct influence on message speed: negotiation strategies, if used more than occasionally, quickly become tiring for both parties, as well as threatening my face as a conversational partner.') Burling, in his Learning a Field Language (p.44) suggests that the problem is self-limiting: The time will come when you can understand the natural conversations that surround you easily enough to make them the most useful means for further learning. Of course, you will often miss things, but you will reach the point where your misunderstandings are sufficiently infrequent to allow you to stop people and ask them to explain on the spot. This will be a point of great emancipation I am still waiting for `emancipation' after more than twenty years! [27]

F. Unlike Schmidt, who is `used to chatting all day', I am quite happy just `silently talking' to myself in English for long stretches of time, which reduces my motivation to overcome the hurdle of initiating and maintaining non-essential conversation. L1 `silent speech', which I have enjoyed for as long as I can remember, is regarded by Vygotsky and others as playing an important role in the development of verbal intelligence (John-Steiner et al. 19??) but I believe it has become so overwhelmingly important to me that it now acts as a barrier making my own mental world less permeable to L2 utterances and the alternative thought-world they might eventually create. Whenever I have, through accident or design, been cut off from aural English input, I am not really immersed in the L2 environment since a much greater amount of time is spent heeding my own English thoughts rather than listening to messages from the outside. My hesitancy in initiating or sustaining communication is, of course, increased by the tendency of some Hong Kong people to react with ridicule or even anger when addressed in poor Cantonese. In Nepal, I benefited from people around me taking the initiative in communication, but in Hong Kong it's usually the English-speakers who will do this!

G. Personality factors also probably make it more difficult to repair breakdowns in conversation e.g. I prefer just to stay silent when I don't understand, hoping that things will become clearer later or that a particular point is not vital. This trait is reinforced by fear that if I revealed how much difficulty I am in, a non-speaker of English may break of the interaction and an English speaker initiate a switch to English.

This basic situation remained unchanged even though since 1992 I have been married to a Cantonese and have spent a lot of time attending family dinners where the conversation would normally always be in Cantonese. My monolingual parents-in-law are not particularly good at adjusting their Cantonese to my level, though my mother-in-law can do this to some extent when I am the only person available for her to talk to. With my wife I do practise Cantonese to some extent but the temptation to revert to English for anything at all complex has always been too strong. Despite all these difficulties, during my 1987-96 secondary-school career I still managed better than the average expatriate teacher and colleagues at school often assumed I understood more than I really did. This was partly because on some occasions I could guess well, and partly because by smiling at the right time or just not crying out for help you naturally lead others to think that you are catching the gist. Precisely this point is made in Moore's 1977 account of his experiences in a Danish university, written just over two years after he took up a professorial appointment there:

I watch the speaker's faces, and probably my own expression to some extent mirrors theirs and gives the impression that I am following the meaning. Certainly when everyone laughs I invariably smile; it would feel stupid to sit poker-faced; but this must suggest that I understand more than I do. People usually seem surprised when they realize that I have not grasped the gist.... The sense of exclusion is greatest when everyone roars at a joke I have not taken in. After trying in vain for some time to follow, one's attention inevitably wanders. The effort of trying to understand is so tiring and yields so little return that one opts out, simply to conserve energy. But this is an insidious and damaging course.(ib.). (Italics supplied)

The tendency to overestimate a foreigner's listening comprehension powers is perhaps increased in Chinese culture if the foreigner makes any attempt to speak the language. Many Chinese people don't risk speaking until their aural comprehension reaches a high level and they may assume that foreigners follow the same rule. The failure to make as much progress as I had originally hoped for nevertheless became a source of psychological pressure for me, compounding other difficulties with the secondary school working environment. Not knowing the language spoken around you obviously contributes to a general feeling of alienation; another teacher on the Expatriate English Language Teacher scheme, who had decided to leave at the end of the first contract, explained to me in May 1989 that he was unhappy in Hong Kong because he missed the feeling of being part of local society which he enjoyed in France; there he spent all the time speaking and listening to French, reading French newspapers, following French football etc. The kind of frustration that the very different situation in Hong Kong can generate is again well expressed by Moore (ib.).

I had not realized before how importantly a satisfactory working life depends on being able to take part in casual discussion. Without it, one loses touch with events and never becomes aware of the currents of opinions among one's colleagues. One cannot contribute to the ongoing dialogue, let alone make any initiative, and increasingly feels isolated and debarred from playing one's proper part in the corporate life of one's working group. One may feel incompetent, resentful and depressed by turns, or may simply select to withdraw from the fray and pursue one's own interests. The latter course may free time for productive work, but at the cost of social and professional isolation.(110 - Campbell 217).

In 1989, my own dissatisfaction was within bounds because I still had confidence that things would come right in the end but this changed in subsequent years.

Even more important than this sense of exclusion was that of failing to succeed in an area - language-learning - where success had always been psychologically important for me. Very inconveniently for an English teacher, I began almost to resent the success of many of those around me in learning English, since their achievement threw my own failure into higher relief as well as depriving me of the opportunity to improve.[28] This meant, for example, that when I had to deal with the parents of my students I was unhappy when they themselves could speak English or when I had a fluently bilingual colleague interpreting for me, but felt better when, by accident or design, I found myself struggling to communicate in Cantonese with a monolingual parent. The actual efficiency of communication was drastically reduced, but a major source of aggravation had been removed! I still, of course, could enjoy conversations in English with local friends, but only because either the interest of the topic or my rapport with the individual concerned outbalanced the linguistic frustration. After resignation from my teaching position in summer 1996, two terms of intensive Cantonese (11 or 12 hours per week) in classes of 6-8 students at the Chinese University of Hong Kong in autumn and spring of 1996/97 did not make the radical difference to my predicament that I had hoped for. However, particularly during the period immediately after the end of formal study, it did at least boost my vocabulary (more so in `High Cantonese' which is much closer to Mandarin and thus maps more closely onto written Chinese text) and TV and radio news broadcasts became a little easier to understand. I do, though, remain heavily dependent on `top-down processing' and, with TV, on the Chinese subtitles. From time to time, especially when a word or phrase is repeated during a broadcast, I am able to isolate it and ask a native speaker for the meaning - this technique is, of course, most effective when the person concerned has been listening to the broadcast with me since they are then likely to recognise the word even if I badly mispronounce it. Since joining a third secondary school in September 1997, I have found many members of staff and also students willing to talk to me in Cantonese but, as before, my still limited listening comprehension makes this highly inefficient and generally forces me back into English when things get at all complicated. Frustration at being surrounded by a language I am not in command of remains a major source of stress. Even though they did not make a radical difference to my proficiency level, there were significant changes in the linguistic environment at home with the birth of our daughter in May 2000. In the first four years, my mother-in-law spent a lot of time helping to look after her and I had more occasion to speak to her as well as occasionally interpreting for her and our Philipino or Indonesian helper. By agreement, my wife also used Cantonese when speaking directly to the baby whilst I used only English. Thus, although not usually practising my spoken Cantonese with my daughter myself, I naturally picked up words repeated often by the others – mum (a `baby word’ for rice or food) and lau gai (display bad temper) were two of the most frequent. As before, I `acquired’ interlanguage forms without the tone; I had thought that lau gai was two level tones before seeing from Yip and Matthews’ Intermediate Cantonese that both syllables are high-rising! When my daughter began speaking herself at around 12 months she initially spoke to me, as to everyone else around her, in Cantonese, even calling me baba before I noticed her using mama or mummy to my wife, but usually appeared to understand when I spoke to her in simple English. At 18 months she began speaking to me in English but inserting Cantonese vocabulary when she did not know, or could not immediately recall, the English word. As she has normally been in a Chinese-language environment, her Chinese is still (2009) a little ahead of her English, this pattern has continued. Though still normally speaking English to her myself, I sometimes use Cantonese with her as a joke and she enjoys making fun of my poor pronunciation![29]

12.Putonghua (PTH): Whilst in Nepal in summer 1987, prior to taking up teaching duties in H.K. in the autumn, I spent a few hours listening to tapes and reading Tung and Pollard's Colloquial Chinese, but switched my attention to Cantonese once I arrived in H.K. and found it was overwhelmingly more important as a means of daily communication. I picked up some more information from the Pinyin in the textbook I used for written Chinese (see next section). Whilst at Kiangsu-Chekiang College, where particular emphasis was put upon PTH, I attended a few lessons provided by the school for staff but, as these were designed for those who were already fully literate in Chinese, this was of limited value. Subsequently exposure has been largely limited to listening to Mainland spokesmen appearing on TV bulletins. I can in fact often understand them quite well but this is because of being able to rely on the subtitles and on the fact that (as with government spokesmen generally) it is often very easy to predict what they are going to say! On a solo trip to Beijing in 1990, I relied upon a phrasebook and occasionally writing down (badly!) a Chinese character. My original intention in 1987 had been to concentrate on Cantonese for a couple of years and then switch to PTH but failure to make satisfactory progress with the former has stymied this plan.

13.Written Chinese: In October 1987 I began self-study of written Chinese using John De Francis's series (Beginning ChineseReader, Intermediate Chinese Reader and Advanced Chinese Reader). I did not practise writing characters but just concentrated on getting a reading knowledge. I continued using the series for about three years, reaching about half-way through the Advanced volume, and after that read newspaper articles instead. In the early 90s I often pasted articles into a scrap book and tried to look up and copy some of the new characters or phrases but this was very time-consuming given the difficulty of locating characters by the radical/stroke-count method and my lack of command of the writing system. I now largely just read (with occasional use of the dictionary) and hope words will `stick' without more formal study methods. In some cases I associate a character with an English word without knowing the spoken Chinese, in others I link it with an (inaccurate!) Cantonese pronunciation. I can probably recognise around 1500 characters and I can usually get the gist of a short report or editorial on a subject I am familiar with but I am very slow and the process is often more like decipherment than reading in the proper sense. The great advantage of reading is, of course, precisely the fact that you can go at your own pace! This concentration on reading rather than writing contrasts strongly with the way Chinese themselves become literate, since this involves painstaking practice of individual characters (see Bell(1995) for an account by a non-Chinese of her impatience when, as a postgraduate student, she was taught to write by a Chinese teacher using the traditional method.) When needing to count up the number of strokes in a character to consult a dictionary, Hong Kong Chinese can therefore rely on a strong, motor memory, tracing out the character with a finger.[30] My own method is to shut my eyes and simply try to visualise the printed character. Apart from the newspaper and occasionally attempting academic articles, I now make regular, though limited, use of my Chinese reading proficiency dealing with simple notices and instructions at school and also whilst watching T.V: many programmes on the Chinese-language channels are sub-titled, presumably mainly as assistance to Chinese who can read the standard literary language but do not follow spoken Cantonese easily. However, with TV drama (in contrast to news bulletins) my processing abilities are much too slow to cope with the presented text, even if I think all the characters look familiar. Though I have never learned to input Chinese characters into a computer, I do `write’ a few words and phrases with the help of translation programmes (e.g. www.dictionary.com) and copy-and-paste from various websites both in e-mails and (since late 2008) on Facebook. As Hong Kong students communicating on Facebook, sometimes use written Cantonese (as well as standard written Chinese and English) this enables me to copy and paste colloquial expressions which would not be available in a standard English-Chinese translation programme. I was initially able to cope with most of the Chinese my daugherter read or wrote for school but this is becoming more difficult now (2009) when she is studying in Primary Grade 3.

14.Spanish I had spent a very short time looking at basic Spanish grammar in 1971 before accompanying my brother traveling around Spain for a fortnight but it was my brother who did most of the communicating as he had been learning on his own more intensively so that he would be able to study it as well as French when he started university the following autumn. I only gave attention to the language again thirty years later when helping establish a club at my school to encourage interest in mainland European languages and cultures. In the academic year 2003-04, after another teacher who had studied the language at school in Canada had to leave HK, I myself taught a few lessons on `Reading Spanish’, making it clear that I was not a speaker of the language myself but teaching some basic vocabulary and grammar from a textbook. I also began using and encouraging others to use the introductory Spanish lessons in the BBC’s on-line `Spanish Steps’ course. I was particularly keen to promote Spanish in my own school both because of the language’s very large number of native speakers and because, as the most grammatically regular of the Romance languages, it probably provides the easiest point of entry to those languages generally (Hawkins.1981) In subsequent years I have got native-speaker exchange students to do the actual teaching but designed some supplementary materials myself. These include a summary of the grammar, condensed mainly from Butt’s Oxford Spanish Grammar, and intended perhaps more to help systematize my own knowledge than to assist students directly.[31] Although this strengthened my reading knowledge of the language, my command is nowhere near my level in French which I studied formally for five years. Very recently (January 2009) I have begun attending a Spanish conversation group in HK but have the problem of weaker aural comprehension than most of the others taking part and, in the meetings which take place over dinner in Spanish or Mexican restaurants, am either frequently lost or succumb to the temptation to switch back into English. I would probably need to study intensively and in a more structured manner for some months to make any major improvement.

Conclusion A VSO contemporary, now working as a TEFL teacher trainer, to whom I showed an earlier draft of this paper commented that my basic approach to language learning had remained the same despite many changes of circumstances. My strategy seems to have been dictated both by problems with phonetic decoding and by personality factors that often make me feel happier with a book than with a conversation in which I cannot perform competently. In Krashen's terms, I have been dependent on acquisition as well as on explicit learning, since reading has been so important, but, since my inter-language is probably built up through my own (generally inaccurate) internal phonetic representation of the printed word, listening to informal speech is naturally a problem. The anxiety provoked by feeling out of my depth has also probably made me over eager to speak myself even though my own consciously-held belief about language learning is that listening, whilst not the sole key to progress, is nevertheless more important than production. Despite its limitations, my method does give a basic communicative ability if enough relevant vocabulary has been learnt explicitly or absorbed through reading but there has always seemed to be a ceiling on this. A possible explanation is that I rely heavily on building associations between single words or short phrases in L2 and items in my L1 vocabulary and do not develop the large stock of longer phrases and sentence-stems which Pawley and Syder see as an important factor in building native-like command of a language. Whether this might change if I was placed for long periods of time in a situation where communicating in L2 was a necessity rather than a luxury is a still unanswered question. My continuing motivation is probably best described as `performative' rather than integrational or instrumental, though integrational factors are also of some importance: what matters most is my interest in a language for its own sake and enjoyment of a sense of (albeit limited) control over it. I feel a sense of control either because my own command of the language is perceptibly increasing or, in a bilingual situation, because other people are to some extent dependent on my own bilingualism. With Latin and Greek at school, the feeling of satisfaction was provided by a rapidly growing reading ability and also by a fascination for the actual mechanics of the languages. The grammar translation method thus suited me well though it was a source of frustration for many of classmates.[32] Going to Nepal enabled me to add the excitement of successful communication in real life to the academic pleasures of processing in text. Being faced in later years with apparent fossilization of my reading abilities and a lack of opportunity to use my latest languages in conversation inevitably brought degree of frustration and I felt uncertain whether to get off the mountain and rest or to redouble the effort to climb higher. At the moment (May 2011), whilst still perhaps clinging to a faint hope that I might eventually `arrive’ in one particular language, I seem to have mostly settled for maintaining and very slightly extending the competence I already have in a number of them, and particularly in Latin, where my level is reasonably good as judged against the rather low average standard among learners.. It is, of course, unsafe to try to draw any conclusions about learners generally from one individual's account of his own experiences, particularly since I have so often been struck by the contrast between my own approach and that of those round me. However, the very fact of individual variation is in itself a key factor to be considered by any learner, teacher or course designer. Resource constraints may seem to dictate the provision of similar treatment for every student, but the difference in motivation and learning style between one student and the next is a powerful argument for a learner-centred curriculum (Nunan 1988) and for greater emphasis on self-access methods.

REFERENCES

Arbeitsstelle für Neuelatein. 2010. Material presented at, or accounts of, the 2010 Seminarium Trevericum, Vox Latina, 46 (182): 192-218.
Aitchison, J. 1989. The Articulate Mammal (3rd.ed.) London: Routledge.
Anttila, Raimo. 1989. Historical and Comparative Linguistics (2nd. ed.) Amsterdam: John Benjamin's.
Bauer, R.S. & P.K. Benedict. 1997. Modern Cantonese Phonology. The Hague: Mouton de Gruyter.
Burling, R. 1984. Learning a Field Language. Ann Arbor: University of Michigan Press.
Bell, J.S. 1995 `The Relationship Between First and Second Language Literacy: some Complicating Factors'. TESOL Quarterly. 29:4. p.687-704.
Butt, John, 1997. Spanish Grammar New York: Oxford University Press
Cambridge School Classics Project. 1998. Cambridge Latin Course1 (Student’s Book), 4th ed. Cambridge: Cambridge University Press
Chambers, J.K. 1992. `Dialect Acquisition' Language. 68:673-705.
Clark, T.W. 1963. Introduction to Nepali - A First Year Language course. Cambridge: Heffer.
Corson, D. 1997. `The Learning and Use of Academic English Words'. Language Learning, 47(4):671- 718
Cross, J.P. 1984. `The Gift of Tongues: Natural or Induced?' Incorporated Linguist. 23:3.
Cruttenden, A. 1974. `An experiment involving comprehension of intonation in children from 7 to 10.' Journal of Child Language, 1: 221-31.
Crystal, D. The Cambridge Encyclopedia of Language (2nd.ed.) Cambridge: CUP.
De Francis, J. 1977. Beginning Chinese Reader. London: Yale University Press.
__________ 1967. Intermediate Chinese Reader. London: Yale University Press.
__________ 1968. Advanced Chinese Reader. London: Yale University Press.
Des Chene, M.K. 1991. `Relics of Empire: a Cultural History of the Gurkhas 1815-1987'. PhD thesis, Stanford University.
Fathman, A. 1975. `The relationship between age and second language productive ability' Language Learning, 25:2, 245-54.
Freed, B.F. 1995a. `Language Learning and Study Abroad'. In: Barbara F. Freed (ed.), Second Language Acquisition in a Study Abroad Context. Amsterdam: John Benjamins.
Freed, B.F. 1995b. `What makes us think that students who study abroad become fluent?' In: Barbara F. Freed (ed.), Second Language Acquisition in a Study Abroad Context. Amsterdam: John Benjamins.
Gibbons, J. 1987. Code-mixing and Code choice: a Hong Kong case study. Clevedon: Multilingual Matters.
Gumperz, J.J. 1966. `Linguistic Repertoires, Grammars and Second Language Instruction'. In Languages in Social Groups: essays by John J. Gumperz. Selected and edited by A.A.Dil. Stanford: Stanford University Press.
Hawkins, Eric. 1981. Modern Languages in the Curriculum. Cambridge: Cambridge University Press.
Honey, J. 1997. Language is Power: the Story of Standard English and its Enemies. London: Faber & Faber.
Ioup, G., E. Boustagui, M.E.Tigi & M.Moselle. 1994. `Re-examining the critical period hypothesis: a case study of successful adult SLA in a naturalistic environment.' Studies in Second Language Acquisition 16.73-98.
John-Steiner, V., C.P.Panofsky & L.W.Smith (eds.), Sociocultural Approaches to Language and Literacy: an Interactionist Perspective, Cambridge, NY: Cambridge University Press.
Jones, F.R. 1994. `The Lone Language Learner: a Diary Study'. System, 22,4: 441-54.
Kazazis, Kostas. 1970. `The relative importance of parents and peers in first-language acquisition.' General Linguistics, 10: 111-20.
Krashen, S.D. 1989. `We acquire vocabulary and spelling by reading: Additional evidence for the input hypothesis.' Modern Language Journal, 73(4): 440-464.
Krashen, S.D. 1994. `The Input Hypothesis and its Rivals.' In: N. Ellis (ed.), Implicit and Explicit Learning of Languages. London: Academic Press
Lafford, B.A. 1995. `Getting into, through and out of a survival situation: a comparison of communicative strategies used by students studying Spanish abroad and `at home''. In: Barbara F. Freed (ed.), Second Language Acquisition in a Study Abroad Context. Amsterdam: John Benjamin's.
Li, D.C.S. & Richards, J.C. 1995. Cantonese as a second language?: a study of learner needs and Cantonese course books. Hong Kong: City University.
Macgregor, R.S. 1986. Outline of Hindi Grammar. Oxford: OUP. (2nd. ed.?)
Matthews, D. 1984. A Course in Nepali. London: S.O.A.S.
Matthews, Stephen & Virginia Yip. 2007 The Bilingual Child: Early Development and Language Contact (Cambridge Approaches to Language Contact) Cambridge: CUP
McDonough, S.H. 1986. Psychology in Foreign Language Teaching. 2nd. ed. London: Allen & Unwin.
McLaughlin, B., T.Rossman & B.McLeod. 1983. `Second Language Learning: an information processing perspective.' Language Learning, 33: 135-58.
Moore, T. 1977. `An experimental language language handicap (personal account).' Bulletin of the British Psychological Society, 30: 107-110.
Nunan, D. 1992. Research Methods in Language Learning. Cambridge: Cambridge University Press.
Nunan, D. 1989. The Learner-Centre Curriculum. Camberidge: Cambridge University Press.
Orberg, Hans. 2005. Lingua Latina Per se illustrata: Pars 1 Familia Romana Newport, MA: Focus Publishing/R Pullins.
Pawley, A. & F.H. Syder. 1983. `Two puzzles for linguistic theory: nativelike selection and nativelike fluency.' In: J.C. Richards & R.W. Schmidt (eds.), Language and communication, London: Longman.
Pennington, M. Forces shaping a dual code society: an interpretative review of the literature on language use and language attitudes in Hong Kong. Hong Kong: City University of Hong Kong. (Research Report No.36).
Schmidt, R. 1990. `The Role of Consciousness in Second Language Learning'. Applied Linguistics, 11: 129-158.
Schmidt, R. and S.Frota. 1985. `Developing basic conversational ability in a second language: a case study of an adult learner of Portuguese' In R.Day (ed.), Talking to Learn. Rowley, MA: Newbury House.
Schumann, F.E. & J.H.. 1977. `Diary of a language learner: an introspective study of second language learning'. In H.D.Brown, R.H.Crymes, and C.A.Yorio (eds.), On TESOL '77: Teaching and Learning English as a Second Language: Trends in Research and Practice. Washington, DC:TESOL.Lafford 1995.
Smith, N. & Tsimpli, I. 1991. `Linguistic Modularity? A case study of a `Savant' linguist.' Lingua, 84: 315-351.
Smith, N. & Tsimpli, I.M. 1995. The Mind of a Savant - Language Learning and Modularity. Oxford: Blackwell.
So, L.K.H. & B.J. Dodd. 1995. `The acquisition of phonology by Cantonese-speaking children'. Journal of Child Language Acquisition, 22: 473-495.
Sthapit, Motikaji. 1967/8 [2024 V.S.] Brahmand ra Saurmandal Kathmndu: Askal Prakashan
Traupman, John C. 2004. Conversational Latin for Oral Proficiency. Wauconda, Illinois: Bolchazy-Carducci.
Tung, P.C. & D.E. Pollard, 1982. Colloquial Chinese London: Routledge.
Varley, Rosemary & Lydia K.H. So. 1995. `Age effects in tonal comprehension of Cantonese'. Journal of Chinese Linguistics, 23(2): 76-97.
Volterra, V. & T. Taeschner. 1978. `The acquisition and development of language by bilingual children.' Journal of Child Language, 5: 311-26.
Whelpton, J. 1983. Jang Bahadur in Europe: the First Nepalese Mission to the West. Kathmandu: Sahayogi.
Whelpton, J. 1999. `English, Cantonese and Putonghua in a Hong Kong Secondary School: Language Use and Language Attitudes.' Hong Kong Journal of Applied Linguistics 4(2): 79-92.
Weir, R.H. 1966. `Some questions on the child's learning of phonology.' In: Smith, F. & Miller, G.A. (eds.), The Genesis of Language, Cambridge, Mass.: MIT Press.
Wilkes, Angela, J. Shackell and John Shackell. 1993. Latin for Beginners London: Usborne.
Wilkes, Angela and John Shackell. 1986. German for Beginners London: Usborne
Yip, Virginia and Stephen Matthews. 2001. Intermediate Cantonese: a Grammar and Workbook London: Routledge.
Zhao, Yong. 1997. `The Effect of Listeners' Control of Speech on Second Language Comprehension', Applied Linguistics, 18,1: 49-68.

-----------------------
[1] Honey (1997: 121) points out that, whether or not the oft-cited figure of 3% for British speakers of full RP is correct, `much more significance attaches to the power that RP has exercised...in causing speakers to adapt their accents in the direction of RP, even though most do not make the adaptation complete, but retain in their accent at least some trace of their regional origins.'
[2] Both the failure to `pick up' the pronunciation and the involuntary adoption of some other features are explicable in terms of Schmidt's `notice the difference' model (Schmidt & Frota 1985; Schmidt 1990) since I certainly did become strongly aware of the vocabulary, syntactical features mentioned but do not accurately perceive the nuances of pronunciation ones in normal speech. This evidence also bears out Schmidt's stress in his 1990 article on the importance of separating the notion of `noticing' (as opposed to unawareness) from that of deliberate intention (as against an automatic process) or possessing explicit (declarative) knowledge (in contrast to implicit knowledge where one knows what is right in a particular case but cannot give a general rule. The use of the terms `conscious'/`subconscious' can be misleading because normal English usage conflates the three distinctions. In the main text I had myself originally written `subconsciously assimilate' and `conscious imitation' where I now have `automatically assimilate' and `deliberate imitation'. We also need to distinguish the issue of deliberateness in learning from that of deliberateness in later production - a word could be originally learned from a vocabulary test but later `pop out' in production without any deliberate choice to employ it.
[3] Certainly, acquisition of L1 resists any attempt to `correct' by learning later on - c.f. at Thakur Ram Campus in Nepal I taught the use of subjunctive `were' for singular in second conditional but then an instant later unwittingly used the singular myself. The importance of acquisition for second language learners generally, and its resistance to the influence of explicit knowledge, was also highlighted for me by the mistake of a teacher I observed at a Lisbon school in 1986: she reverted to universal use of the `isn't it?' tag just after she had carefully drilled her students in several of the alternative, idiomatic forms (`hasn't he?' etc.)
[4] In contrast, speakers of tonal languages acquire a command of lexical tone very early on. Cantonese children have normal mastered all tonal contrasts by the age of two (So & Dodd (1995: 484, 490) cited by Bauer & Benedict (1997: 109); Varley & So (1995:79-80)). It has also been claimed that Chinese babies at the babbling stage seem to be trying out tonal variation on individual syllables rather than imitating intonation contours over longer stretches of sound (Weir 1966, cited in Aichison 1989:78).
[5] The difficulties for both native speakers ad L2 learners of the Graeco-Latin element in English vocabulary and the assistance provided in this by a knowledge of the Classical languages or a modern Romance language is discussed in Corson (1997).
[6] What I enjoyed about maths at this time was its presentation as a system or rules and principles which could be applied deductively. I did not (and do not now) feel particularly comfortable with calculation as such. In F6 (1996/7) the school participated in the trial of a newly `Academic Aptitude Test' which it was hoped would provide a better correlation with eventual university results was done by A-Level grades. I remember finishing the verbal section within the time allowed but not the numerical one.

[7] I recall once make a conscious effort to get longer stretches of Latin into my brain by trying to memorise a passage on the ancient Scythians' way of life from Virgil's Georgics (a didactic poem on agriculture) memorized verbatim but found this very difficult. Similarly, I was not very good at remembering long quotes from Shakespeare when doing O-level English Lit. I have assumed in the text that my non-verbatim ( `propositional') memory was always in English but another possibility is that it is in long-term memory in non-verbal form (the Chomskyan `mentalese') but converted into English simultaneously with recall
[8] http://www.cambridgescp.com/page.php?p=clc^oa_book1^intro
[9] See the notes at http://humanum.arts.cuhk.edu.hk/~lha/latin_intensive/download/WHELPTON_NOTES2.pdf
[10] The bulletins are available at www.yleradio1.fi/nuntii and the glossing, intended to make the texts accessible to beginning learners of Latin, is explained in detail at [pic]KÚÝ¢
À
È
Ø
:
O
†
‡
Â
CNQÊÍZ\¦¬¼ or (for a more up-to-date account) at http://linguae.weebly.com/using-nuntii-latini.html
[11] See http://cirlapa.org/locutorium/index.php and http://www.alcuinus.net/GLL/index.htm
[12] The attractive illustrations in the Usborne internet-linked introductory courses make them especially suitable for young learners just needing a `taste’ of the language. For `Living Latin’ see http://www.latinitatis.com/latinitas/menu_gb.htm and http://en.wikipedia.org/wiki/Recent_Latin (both consulted 12/4/09) and sites tagged under `livelatin’ on http://delicious.com/Velptonius For serious students, a good supplementary textbook embodying this approach is Traupman (2004), with graded dialogues both on classical and contemporary topics. For those wanting to dispense with translation altogether, there is Orberg (2005)’s direct method course with accompanying audio material. Professional Latin teachers are themselves deeply divided over communicative Latin, with some seeing it as a pointless diversion from the serious study of written texts.
[13] Papers delivered at the conference were published in Vox Latina tomus 46 (fasc.182), 2010, pp.519-618
[14] It is possible to consult a lit of those wishing to take part in such exchanges, and to add one’s own name, on the website of the Circulus Latinus Interretialis (http://www.circuluslatinusinterretialis.co.uk/)
[15] See the Circulus webpage at http://linguae.weebly.com/circulus-latinus-honcongensis.html This includes demonstration dialogues with english translations that I prepare before each meeting.
[16] My parents often reminded me that the acute/grave problem actually reduced me to tears the first week of term but, interestingly, I seem to have `censored out' any direct memory of this!
[17] Crystal (1997: 365, reporting Volterra & Taeshner, 1978), in discussing the linguistic development of bilingual children, cites the case of a four-year-old who always used German with her mother and Italian with her father becoming distressed (`No, tu non puoi!') when the father one day spoke to her in German. Was the reaction purely because a seeming regularity in her world had suddenly disappeared or partly because she felt she had lost a hitherto perceived linguistic superiority over the parent?
[18] Some of the supplementary materials, including Powerpoint presentations and songs with English translations and links to Youtube videos of performances, are available for download from http://french.meetup.com/835/files or http://linguae.weebly.com/french.html
[19] Sanskrit, the classical language of India, stands in roughly the same relation to Nepali, Hindi, Bengali and the other North Indian languages as does Latin to the Romance languages.
[20] I think this may have been the Pennsylvania Project, which showed a slight superiority of the traditional method in the 58 schools surveyed. The findings of the Project may, however, have been unreliable because teachers involved probably did not teach exclusively according to one method (Nunan 1992:92)
[21] Freed (1995a & b) and Lafford (1995) report research suggesting that it is `fluency' rather than gains in other areas of competence which gives students studying abroad in an L2 environment their edge over those exposed only to classroom learning at home. They also discuss `fluency' as a construct. Repeated use of the language by the learner would be expected to contribute to the `automaticisation' of production of items learnt formally, which is part of McLaughlin's model of 2LA. In addition, by using (or even just mentally rehearsing) linguistic items) the learner is providing him or herself with `auto-input' (Schmidt 1990)) and thus such forms will come to `sound' increasingly correct and even become impervious to explicit correction. Initial `internalising' of items through reading is not, of course, necessarily inconsistent with Krashen's insistence on true command of a language being achievable only through `acquisition', since Krashen himself has made it clear that he regards `visual acquisition' as important for vocabulary building both in L1 and L2 (Krashen 1989 & 1994: 46)
[22] In recalling this incident now, I am struck by the way in which the power implications of language were highlighted by it. When we first inadvertently entered the central courtyard, I attracted odd looks but no one actually approached me - perhaps because they were unsure of their own linguistic competence. Then one young man, dressed in a leather jacket and appearing slightly drunk, came up to me and started off, `What is this? ' A crowd then rapidly gathered. Code-switching constantly between English and Nepali I attempted unsuccessfully to talk our way out of the situation. When brought by the crowd into the presence of the senior police officer at the local station, my nervousness led me into taking a rather aggressive stance and I began, `Can you speak English? Well?' The retort was, `I can teach you English!' Fifteen years later I was on the hill overlooking the temple when two young Nepali boys began calling out to me in English in a rather cheeky manner. I switched to Nepali and explained that I had been on that same hillside before they were born. This transformed the situation and I now had two rather shy boys asking me in Nepali what differences I had noticed between Nepal then and now.
[23] In this, of course, I resemble many of the British officers in the Brigade of Gurkhas, whose poor pronunciation is sometimes mimicked (not, of course, in their presence!) by their men (Des Chene, 1990).
[24] One example that sticks in my mind is that of a Lt.-Colonel who had been in command of a Gurkha regiment but had never realised that pa:ni (water – with a long vowel) was different from pani (also). There are also, however, British officers who have attained an extremely high proficiency in Nepali. Lt.-Colonel John Cross, for example, who now lives in Nepal with an adopted local family, is indistinguishable from a native-speaker when heard over the radio. For his own highly interesting account of his experiences learning Nepali and many other languages, see Cross 1984.
[25] The relationship is roughly similar to that between Italian and Spanish.
[26] The high degree of commonality in the vocabularies of these languages and (through loan-words) of English provides the lexical basis for artificial `link-languages' such as Esperanto, which has been aptly described as `a kind of pidgin Romance' (Anttila 1989:175). Crystal (1997: 357) illustrates this neatly with a quotation in the more recently created `Interlingua': `Tote le membros del communitate de linguas occidental son in un certe senso dialectos individual que devia plus o minus de un patrono commun.... Le termine interlingua... representa un lingua que es international proque su elementos existe de facto o potentialmente in un gruppo de linguas national.'
[27] Possible way of getting round the problem of `authentic' input being too fast to process. First (as recommended in Burling (1984)) recorded monologues by informants (or authentic dialogues?) and then analyse at leisure. Process could be aided by the kind of digital, computerised playback system discussed by Zhao (1997), in which the listener is able to vary the speed of playback himself.
[28] In theory, I should also have become jealous of the success of foreigners who have had more success than me in learning Cantonese. In fact, probably because I am not very frequently in contact with such people, I tend on the contrary to feel pleased when I do meet or read about them. They do not seem so much rivals as role-models, giving me hope that it is not, after all, hopelessly unrealistic for gweilo to aspire to move back and forth across the language barrier in the same way that educated Cantonese can do! However, I can feel jealous when with a westerner who seems to be communicating more effectively in Nepali than I can myself; presumably I accept that I am `on the way' in Cantonese but consider I ought definitely to have `arrived' in Nepali!
[29] Recordings of my daughter’s English and Cantonese speech between ages two and four formed part of the data base for a research project on bilingual language acquisition by two Hong Kong linguists (Matthews and Yip 2007).
[30] Whether or not Krashen (1994) is right in rejecting an important role for routinisation of explicitly taught output skills in becoming a proficient writer of languages using an alphabetic script, his theory clearly does not apply to mastering Chinese characters.

[31] The grammar notes (`Reading Spanish’) and some of my other materials can be downloaded from http://linguae.weebly.com/spanish.html
[32] Seeking to highlight the sterility of teaching rules of grammar before concentration on sentences in use, Crystal (1997:89) compares the procedure to teaching biology by copying down labelled diagrams of plants before collecting and examining real plants. In fact, I enjoyed my F2 introduction to biology when we did indeed just copy diagrams and learn names and descriptions of organisms. I did not enjoy it so much from F3 onwards when we were also required to handle real specimens in the laboratory!