LEARNERS’ AND TEACHERS’ BELIEFS ABOUT LEARNING TONES AND PINYIN

This paper reports a study of the perceptions of English-speaking learners and teachers about the challenges and difficulties of Chinese as a Second Language (CSL) learning in England. The study involved a Likert-scale questionnaire and follow-up interviews with 37 university student learners, 443 school students and the 42 teachers of both groups. The questionnaires and interviews explored beliefs about language learning, about Chinese language learning and about language learning strategies. This paper focuses on the findings concerning the perceived challenges of speaking Chinese and of tones in learning Chinese. The findings of this study present a picture of teachers who are keen for their students to learn to speak and communicate in Chinese, and of students who are keen to take risks in speaking. However, in contrast to earlier findings about learners’ views about learning Chinese, the learners in this study claimed to be very tone aware and reported that they found listening and understanding Chinese more difficult than production. This is explored in relation to the pupils’ views about learning tones and pinyin and raises questions about the ways they address tones and pinyin learning in the context of their expressed aim of communicating and taking risks in speaking. The discussion raises issues about the possible effects of communicative teaching of languages in English schools. We ask whether an emphasis on communicative approaches may affect how learners address difficulties of the Chinese pronunciation system and the use of pinyin.


Introduction
A British Council strategic analysis of the UK's long-term language needs (Tinsley and Board, 2013) placed Chinese as the fourth of ten languages which will be of crucial importance for the UK's prosperity, security and influence in the world in the years ahead. However, a recent survey conducted for the British Council highlighted that 75 per cent of the adult UK population are unable to hold a conversation in any of the ten languages identified as important and, indeed, that less than 1% of the UK population could do this in Chinese (YouGov, 2013).
Against this background of recognised need, but a poor track record, of languages learning, England is currently experiencing a wave of popular and political good will towards the learning of Chinese, evidenced by statements by politicians, the press and popular publications (e.g. Watt et al, 2013). Entries for Chinese A-level (an examination taken aged 18) have grown rapidly, making Mandarin Chinese the fourth most popular language in schools after French, German and Spanish. Mandarin Chinese is available at degree level in around 30 universities in England, Scotland and Wales (JCQ, 2013;SQA, 2013).The UK has 13 Confucius Institutes as well as a network of school-based Confucius Classrooms. This is despite the perception that Chinese is difficult for English speakers to learn (Stevens, 2006).
In England, language learning remains optional for both children and adults and motivation is known to be one of the most significant predicators and determinants of success in second and foreign language learning. (Dörnyei 1994;Oxford and Shearin 1994;Coleman, Galaczi et al. 2007). A large-scale survey of UK secondary students' motivation to learn foreign languages conducted by Coleman et al. (2007) reported declining motivation towards foreign language learning, the causes of which have been debated at length in a number of substantial articles (Coleman, Galaczi et al. 2007;Coleman 2009). In this context it is particularly important that learners' beliefs about their learning and language are understood by teachers and policymakers. Pajares (1992) examined the "messy construct" of teacher beliefs in a substantial review of the literature and discussed the way that poor definitions of what is meant by beliefs has caused difficulties in this area of study. Pajares (1992) argues argued that knowledge is based on objective fact, whereas belief is based on evaluation and judgment. Woods (1996, p.195) suggests suggested that while knowledge "refers to things we 'know'conventionally accepted facts, beliefs refer to an acceptance of a proposition for which there is no conventional knowledge, one that is not demonstrable, and for which there is accepted disagreement". Most authors agree that beliefs are created through a process of enculturation and social construction (Fleet 1979;Lasley 1980;Pajares 1992;Poulson et al. 2001) but while a number of studies describe how teacher beliefs appear resistant to change (Brousseau and Freeman 1988;Golombek 1998), more recent research, most usually in the teaching of languages to adults using a particular survey instrument devised by Horwitz (1988) (Beliefs about Language learning Inventory-BALLI) suggests that beliefs about language may be susceptible to change over time.

The importance of beliefs about language learning
Learners bring to language learning a complex set of beliefs about language learning and a number of authors have emphasised that failure to address unrealistic language learning expectations can lead to reluctance and loss of motivation in pupils (Richards and Lockhart 1994); a breakdown in learning (Ellis 1996); and language learning anxiety for students (Young, 1991). Riley (2006). This echoes the arguments about beliefs outside the ELT world ) that if learner beliefs are consistent with good learning practices, the effect of beliefs is likely to be positive, but that inconsistent beliefs and practices may have negative learning consequences. Studies such as that by Kern (1995), using the BALLI referred to above, investigated the stability of pupil beliefs in relation to teacher beliefs and challenged the accepted wisdom that language learners' beliefs remained stable, whilst suggesting that the beliefs of individual pupils were influenced by those of their teachers. In this context it is interesting to see whether adult students and child pupils share the same beliefs about learning Chinese, and whether these reflect the beliefs of their teachers.
Beliefs about language learning, and in particular Chinese learning, may affect teacher and student practices at all levels. One area in which this is important is in risk taking, That is, speaking up in class learning situations and taking the risk of being wrong. This risk-taking is important in successful classroom learning of a language (Thoma, 2011;Zafar and Meenakshi, 2012). High risk-taking does not always result in positive outcomes in FL learning (Ely, 1986). Good learners are perceived more likely to take the medium-risk tasks rather than make wild decisions, on the basis of their judgment of success or failure under the actual circumstances (Beebe, 1983). However, learners need to believe the risk worth taking, if they are to use their language in class and learn from feedback. Beliefs about risk-taking may affect the degree to which learners are able to practice language use. In addition, beliefs about the challenges presented by a language may affect both motivation to learn and strategy use (Dornyei, 2003).

The challenge of tones and pinyin in learning Chinese
One of the significant challenges for foreign learners is the tonal nature of the Chinese sound system, which poses a particular challenge for a beginning learner in speaking and reading Chinese (Wang, Perfetti, & Liu, 2003). Chinese is based on single syllables, each of which is normally formed by an initial consonant and a final vowel or vowel-like element and (with the exception of -er), individual syllables are generally pronounced independently from each other. Unlike intonations in English, Chinese tone is a supra-segmental marker used to help foreigners pronounce the characters, including 21 consonants and 36 vowels or semivowels, which combine to form the syllables in Chinese.
For speakers of alphabetic, non-tonal languages, the challenge in pronouncing Chinese therefore lies both in pronouncing the difficult sounds from Pinyin representations, involving positioning the tongue accurately, and changing the relative pitch of one's voice when sounding out not only the isolated but also large number of homophones renders tone discrimination problematic. Studies suggest that the slight differences of the five tones are difficult for foreigners to discriminate, especially for beginning learners (Lin, 1985) and that difficulty with pronunciation is related to difficulty in discrimination (Hao, 2012). Because the syllable (the rime or vowel ending) is a carrier of the tone, the processing of syllables involves segments and tones simultaneously. This means that the spoken syllable is temporally integrated and can be difficult to decompose by non-tonal language speakers (Lee & Nusbaum, 1993). In short, it is likely to make listening and understanding difficult for English speaking learners. Wang et al. (2003) found that English-speaking beginner learners encountered great difficulty in acquiring tone skills. Wang et al. used a matching task to test beginning Chinese learners' phonological processing skills. When matching was based on perceiving the same tone in two syllables, the Chinese learners showed poorer performance than when the matching was based on syllable onset and rime. This is probably because onset and rime are phonological components of syllables in both English and Chinese, whereas tone is a feature of Chinese but not English.
The high degree of homophony in Chinese is also a discrimination issue for English-speaking learners. Taylor and Taylor (1995) identify around 1300 tonal syllables in spoken Chinese and that number represents almost all of the characters (morphemes). This means that a large number of homophones exist in the language creating difficulty especially in listening, when learners are likely to be exposed to many similar sounds which in fact have different meanings. A spoken word can be of one, two or, in some cases, three syllables and the combination of the syllables compensates for the lack of sufficient phoneme compounds. However, Hu (2010) noted that this can result in particular difficulty for the CFL learners, who have to try to detect the semantic borders among the flowing single syllables which bear more or less equal stress.
Pinyin as a Romanisation system represents the syllable elements (onset and rime) and each syllable is marked by one of the four pronounced tones or the neutral tone which vary in terms of pitch, amplitude and duration level (Liu et al., 2011;Zhang, 2006). This allows learners of Chinese to mimic the pronunciation of Chinese, which is not always evident in characters. Chinese is usually taught through both pinyin and simplified characters in England, but there is little literature about how early characters and pinyin should be used in Chinese learning, although plenty of general debate among teachers. However, the area which has not been explored is the beliefs of pupils and teachers about this issue. Horwitz's (1988) Beliefs about Language Learning Inventory has been used extensively to examine beliefs about language learning and to align these with pedagogical "orientations", especially with adults. Horwitz's (1988) survey with participants learning German, French and Spanish suggested that most believed that it was easier to read and write their target language than to speak and understand it. This inventory has been used to investigate the beliefs of small numbers of adult learners of Chinese.

Investigating beliefs about Chinese Language learning
A small study by Samimy and Lee (1997) collected the beliefs of 10 language instructors and 34 first-year university students studying Chinese in the US, using a modified BALLI, and reported that students and instructors perceived speaking Chinese as more difficult than understanding the language, while reading/ writing skills (taken together) were more difficult than the oral skills of speaking and listening. Indeed, Samimy and Lee also found that the majority of their participants, both students and teachers, believed that, of reading, writing, speaking, listening and cultural behaviours, writing was the hardest aspect of learning Chinese, followed by speaking which was the second most difficult for both teachers and pupils. Listening was rated the third most difficult by the students (although cultural behaviours was third for the teachers) and reading (ranked least difficult by both groups).
In a study investigating the major difficulties encountered in learning Chinese, Huang (2000) asked fifteen intermediate and advanced university students (including nine from a Cantonese background) to rank eight areas of potential difficulty: pronunciation; tones; grammar; writing Chinese characters; vocabulary building; oral communication; delivering a prepared oral presentation; and written compositions. There was a clear distinction between the rankings of those with English as their primary language, who mostly selected written compositions as most difficult, and Cantonese participants for whom pronunciation and tones were the main concern. L1 background has a large part to play in what learners find difficult.
In contrast to these two studies Chiang's (2002) research did not find that writing was considered the greatest difficulty. In her study, which asked more open questions, 20 second-year Chinese students in the US were asked to write down which aspects of learning Chinese they found easy or difficult, from which the researcher identified nine difficulties in the following descending order: memorisation; tones; speaking; listening; characters and writing; semantic distinction; reading; integrating; and thinking deeply.
To identify the kinds of learning difficulties encountered by anglophone CFL learners at UK universities, Hu (2010) used an adapted form of the BALLI with 164 19-25 year old university students. A factor analysis was used to extract six factors which were named as Grammar, Aural Reception, Words, Oral Production, Pronunciation and Recall. Hu's (2010) sample included university students taking Chinese as a major, some taking it as additional work and some school pupils. As students' level of proficiency rose, so, in general, Hu found them to perceive the factors identified above as less 'difficult'. However, when the six factors were examined individually, a link between proficiency level and perceived difficulty was only found for Aural Reception and Oral Production at significance level (p<0.5). In other words, tone production was perceived as difficult by both novices and experts, although tone perception was perceived as less difficult by more expert students.
These studies have provided some interesting initial findings concerning the learning difficulties of Chinese for University students, but University students are, by definition, those who have been academically successful and this restricts the conclusions which can be drawn about the wide range of younger learners on the basis of these studies. Moreover, as Song (2002) pointed out, in the UK, it is likely to be non-major students of Chinese who find language learning most useful and the British Council report discussed above, called for the greatest growth of languages learning among school children, who might reasonably be seen to have that as a life focus and for whom a language has future personal potential gains.
The school learners in this study were studying the national Curriculum-a languages curriculum which Block, (2005), has described as a particular version of communicative language teaching. Although often non-statutory, this curriculum is used in primary, independent and academy schools and includes speaking, listening, reading and writing. The public examinations at 16 and 18 years of age also include speaking and listening as well as reading and writing. This paper seeks to explore school students' and teachers' beliefs about the relative difficulty of spoken aspects of Chinese learning.

Method
This study used a 4-point attitude scale to examine learners' beliefs about language learning, including items about the relative difficulties of different aspects of Chinese as a Foreign Language (CFL) learning, risk taking strategies, and use of pinyin. A number of statements about these issues in different situations were used to measure the strength of the respondents' agreement and disagreement. This Likert-scale instrument was first applied to beliefs research by Horwitz (1988) and the Beliefs about Language Learning Inventory (BALLI) is now a well-established instrument that has been used in many studies of EFL (Diab, 2006;Hong, 2006;Kern, 1995;Peacock, 1999Peacock, , 2001Truitt, 1995;Wu, 2010;Wang, 2006;Yang, 1992) and other foreign languages, such as Spanish, German and French (Horwitz, 1988;Kern, 1995;Kuntz, 1996;Mantle-Bromley, 1995). Horwitz (1999) pointed out that, although the original aim of BALLI is to identify "the individuality in beliefs about language learning" (p.558), in the general sense, the inventory has also been used to investigate whether common beliefs were shared among different language learner groups, as well as learner groups and their teacher groups. Nearly all BALLI studies have focused on alphabetic languages so the questionnaire was adapted to the CFL context, to provide new insights into beliefs, by making comparisons with previous findings on ESL and FL learner groups, as well as teacher groups. Question items specifically for Chinese features were added to the BALLI inventory based on studies with selfdesigned questionnaires about perceptions of linguistic difficulties of Chinese learning, and learning strategies for Chinese tones and characters (Hu, 2010;Hu and Tian, 2012;Shen, 2005Shen, , 2011. A Chinese beliefs questionnaire with 25 items in total was generated for this study (See Appendix 1) Both student and teacher surveys included questions about the nature of CFL learning; the difficulties of CFL learning and the strategies respondents used, but the questions in the background section and the wording was slightly different for each group. This paper considers the findings relevant to tones and pinyin in these areas. The questionnaire responses were analysed using SPSS software.
This questionnaire was supported with interviews with school and university students and teachers. The interview questions in this study were based on the focused aspects discussed previously in the review and participants' answers to the questionnaire statements.
The focus topics generated from the questionnaire items were: a) Difficulties of Chinese linguistic items, i.e. tones, Pinyin, words, grammar and characters; b) Comparison of the four skills in Chinese learning: speaking, listening, reading and writing; c) What is good to start for beginner Chinese learners, Pinyin, characters or oral words? d) How to learn pronunciation, including tones? e) How to learn Chinese characters and words? These interviews were semi structured and were recorded and transcribed. Analysis used NVivo to seek themes and links between them.

Learners-school and university students
The difficulties of locating CFL students and teachers in England have been documented in other studies (CILT, 2007). This survey sought access to teachers and students nationwide through electronic fora, Confucius Institutes and CFL teaching associations. A convenience sample of school children in 10 secondary schools across England, teachers in 15 schools and CFL learners in a University in the West Midlands of England was selected, so that the students in the classes and their teachers were included in the study. This sample cannot claim to be stratified, but it did include whole classes and their teachers in state, independent and academy schools at all relevant age phases.
In total, 443 school students, 37 University students, and 42 teachers responded to the questionnaire (a return rate of 95%). One adult learner, 67 school students and 13 teachers were interviewed, including teachers and students from all age phases and types of school. 92% of students (N=408) were first language English speakers, and only 8% of students (N=35) reported that they came from other language backgrounds. 54.6% of students (N=242) were boys and 45.1% of them (N=200) were girls. This is slightly surprising because there is a longstanding trend of more girls than boys choosing to study languages at secondary school and entering for examinations in England. In 2013, 64% of A level languages examinations were girls (Board and Tinsley, 2014). 39.5% of students (N=175) were aged 7-11 and so in primary school, 58.7% of students (N=260) were aged 12-15, and 1.8% of students (N=8) were aged 15-18 years, all in secondary schools. The pattern of few children choosing to take language qualifications is an established pattern in England. Studying a language qualification at 16 or 18 is seen by pupils as difficult, compared with other options, of limited practical value and risky (Stables and Wikeley, 1999;Coleman et al, 2007).
In contrast with the studies of University students reported by Hu (2010) and Huang (2000), the sample of school students in this study is heavily weighted to beginner learners. 369 school pupils (83.3%) reported that they had been learning Chinese for less than a year, 53 students (12%) had been learning for 1-2 years, and 21 students (4.7%) were in their third to fifth year of learning. The pupils' own estimations of their expertise reflected their experience. Nearly a third of students classified themselves as total beginners, 20% as experienced beginners with less than 5% of students reporting that they had reached intermediate level. Less than 1% of students thought they were at the advanced level. The small number of intermediate and expert learners, might be expected, as so many children give up languages before age 14 and languages are optional before age 11.

CFL Teachers
42 teachers participated in the survey. 83.3% of them (N=35) were L1 Chinese speakers, only 16.7% (N=7) were L2 speakers of Chinese (six L1 English speakers and one L1 French). 88.1% of the teachers were female (N=37), with only five male. Seven teachers taught both primary and secondary pupils. Eight were teachers of Chinese in secondary schools but not of exam classes, nineteen were GCSE Chinese teachers for all children, two were GCSE teachers for Chinese or heritage Chinese students. There were also university teachers of Chinese, two of them for non-Chinese major students, and only one for students who was majoring in Chinese. One teacher was a trainee teacher, and one was the tutor of an online Mandarin course.
Most of teachers in this sample had been teaching Chinese for about 3-5 years in England (N=11) and somewhere else in total (N=15). Some teachers had taught Chinese, to L1 Chinese students for more than 10 years in China, and had started to teach Chinese in England in the last year through Hanban or Taiwan government programmes.

Findings
The questionnaire used a number of Likert-type statements about the learning of Chinese against which respondents chose scores ranging from 1 to 4, signifying levels of disagreement and agreement. To prevent bias, both positive and negative statements were included and, as part of data analysis, the scores were reversed to follow the regular cognitive pattern (Coolican, 2004). Therefore, the higher scores of 3 and 4 represent positive views about the items. In contrast, lower scores of 1 or 2 are negative views. With 2.5 operationalized as a baseline for degree of difficulty, scores above 2.5 signify that an item is was regarded as difficult by respondents and a score below 2.5 signifies that respondents found the item relatively easy. The overall descriptive analyses of beliefs about learning Chinese related to tones and Pinyin is set out in Table 1. The interview results are presented in relation to the questionnaire data.

Is learning Chinese difficult?
Although both the adult students and school students agreed quite strongly (3.35 and 3.10) on statement 12 that "Everyone can learn to speak a foreign language", they are were slightly less strong in their agreement with the proposition that "Everyone can learn to speak Chinese" (2.84 and 3.03), suggesting both groups of students recognized that it is a difficult language for English speakers to learn. School students agreed with the statement that "learning Chinese will be more difficult than learning European languages" (2.89). However, the teachers were neutral about whether Chinese is more difficult than European languages for English speaking learners (2.50), possibly because most of them were L1 Chinese speakers. Teachers estimated the difficulty students face in learning Chinese as lower than students. The interviews confirmed this pattern, and both school pupils and adult learners discussed why learning Chinese is likely to be harder than learning European languages, due to the nonalphabetic features of Chinese language. One school student stated, "I am learning German as well. I find that is easy because it's the same alphabet as English. … With Mandarin you have to learn the characters with the Pinyin, and learn to say it." The degree to which students believed that "I will ultimately learn to speak Chinese very well" differed between the groups. Adult students were more confident (2.95) whereas the school students were ambivalent (2.50) about their ultimate success. This may reflect the positive choice to learn Chinese made by adult learners, whilst primary and secondary pupils below the age of 14 had no choice.

What is important in learning Chinese?
The results suggest some interesting differences between teachers' and pupils' beliefs about what is important in learning Chinese. The responses of the teachers were rather communication-oriented, showing a belief that learning how to converse in Chinese is more important than learning to read and write (M=2.67). In contrast, students held a slightly negative view about this point (M=2.34). In terms of the importance of grammar, half of students believed that grammar rules are important in learning Chinese (M=2.48), whilst the teachers were rather less positive (M=2.21). This suggests that, while the teachers were communication focused, students focused more on the linguistic forms and written scripts of Chinese, rather than on spoken Chinese, possibly because they believed that the linguistic forms and written scripts were more problematic for them.
The teachers and the adult learners in this study were very positive (3.69 and 3.16) about the need to have a go at speaking, even when they made mistakes, and the school pupils agreed, although to a much lesser degree (2.95) than their teachers. However, all three groups agreed strongly that "It is important to speak Chinese with correct pronunciation and intonation" (3.03, 3.35 and 3.00) and this apparent contradiction could suggest that, whilst excellent pronunciation is a goal, they all recognised the need to take risks and make mistakes. In the interviews this was a view shared by pupils students and teachers. As one school pupil pointed out: "Because if you don't make mistakes at start, and you will make mistakes when you are older in China or speaking to people, then the meanings are different and they won't understand you." The adult students and teachers agreed with the item "If you are allowed to make mistakes in the beginning, it will not be hard to get rid of them later on" (2.62 and 2.69), but the pupils were unsure or ambivalent about this (2.51).

Is speaking more difficult than listening and understanding?
Overall, students and teachers had different opinions about the relative difficulty of speaking and listening for understanding. Teachers tended to rate speaking (production) as more difficult than understanding (adjusted M=2.64), whereas both school and adult students thought that listening for understanding is actually harder than speaking (M=2.21>2.50). However, the interviews revealed that most pupils thought that whether speaking is easy or not depends on what they are saying. For instance: "Speaking is probably the easiest, because you just say it and you don't have to know how to write the characters. It doesn't involve the characters at all. It just the Pinyin." An adult learner expressed the beliefs that speaking is easier than writing, but was still clearly concerned about tones. As she pointed out, "speaking is all about pronunciation which makes a bit more difficult, if you want to get a correct intonation and the accuracy of your speaking you have to add the tone, that's what perhaps makes it a slightly difficult, I suppose." The interviews also revealed multiple interpretations of what speaking and listening meant.
"If it is pretty long sentence it is quite hard, but if it is just something that we keep going over and over, like I am late, I remember, yeah, it's easy".
So speaking can mean repetition of a phrase or sentence, or composition and pronunciation of a meaningful utterance.

What is difficult in learning Chinese?
When asked how difficult they found key elements of Chinese learning (tones, pinyin, vocabulary, grammar rules and characters), both groups of students identified the most difficult aspect as homophones, descending in difficulty through grammar, matching the sound with the forms, remembering words, remembering characters, with tones and pinyin as the easiest aspects of Chinese learning. However, this was not the order of difficulty for teachers, who identified tones as the most difficult, followed by homophones, descending in difficulty through characters, matching sound with characters, words, grammar and pinyin. Tones were rated as the most difficult aspect of learning Chinese by the teachers (who might be expected to understand students actual performance) and least difficult by both adult students and school pupils.
The interviews included a great deal of discussion of tones by school pupils and adult students. The degree of discussion about tones and analysis of the discussion suggested that school pupils' and adult students' understandings of tone, and their difficulties with them, were not straightforward. Typically, adult learners and school pupils made it clear they understood the importance of good pronunciation, and pupils in the interview overwhelmingly agree it was because of the tones. They pointed out, "there are four tones, they could be the same words with different tones, with meanings are completely different. So you have to make sure you say them properly, otherwise you could say something completely different." However, some responses about tones cast doubt on the understanding of them by learners. One adult learner said, "There is nothing that sounds anything like an English word. So Chinese is much harder….The tones for me are much harder than the pronunciation." This suggests that this learner (and the others who made this point) saw pronunciation of the onset and rime of the syllable as separate from the tone.
Moreover, many of the school-aged pupil interview responses gave the impression that they did not attend very closely to tones in their speaking, despite the claim to the contrary in the questionnaires. For instance: "I know tones mean different things, but I don't have the feel for tones, so I probably just say it wrong." and "We don't really focus on tones. Probably when you say it, sounds like the accents." or "You can just say it, because sometimes when you say the tones, it sounds different. Anyway, it's not so important every time." And, very clearly "if you get the pronunciation right. Well, you are supposed to get the tone right but it's not really possible, is it?" These comments are typical of a school-aged pupil view that tones were not so important that they needed constant attention.
Whilst there was a consensus about pinyin being the easiest aspect of Chinese, the interview results suggested that this may be based on misconceptions about pinyin by students. Many school-aged pupils did not seem to know the term 'Pinyin' well and merely considered it as a tool for pronunciation. One pupil stated, "Pinyin could be more complicated than the rest of the Chinese, because not like some English vowels make the same sounds. For instance, 'i' in Pinyin is like 'yi', but we don't have that similar things, so it's quite different." All the teachers claimed to teach pinyin and most thought it was a good way to start Chinese, but a large proportion of the teachers expressed concern that pinyin and characters were "a lot to take on at the start" and some of the teachers also said they did "a very little" pinyin. One Year 7 and 8 teacher stated, "I just do a little pinyin to give them confidence. They need to hear the words most of all".
Another suggested "I do Pinyin very lightly -with just the main points and not too much emphasis".
It was noticeable that many of the school-aged pupils had a limited understanding of pinyin and there was doubt about how much it helped in learning Chinese.
"I am ok when I speak tones, but when I write them (in Pinyin) it's a bit confusing".
Listening and understanding Chinese on the basis of tone was perceived as a major obstacle for English speaking learners in the interviews. As one school pupil pointed out, "we have been trying to recognize which tone it was and know them. It was pretty difficult to get them just from someone speaking in Chinese. I'm not sure how we would necessarily get to recognize which tone it actually was." Other pupils mentioned the discrepancies of tones spoken by themselves and Chinese L1 speakers, saying that "I think listening probably the hardest of all, because you know your own voice well, but when you listen to other people saying Chinese, that's another thing." Some pupils believed listening was comfortable for them. However, their comments suggest that "listening" was not always listening on their own, but listening with some additional help from their teacher. They mentioned their Chinese teacher provides English translations or uses simple words when speaking to them.
"listening is OK when I know what words are coming. Or she (the teacher) points out words as they come up on the tape".
For this reason, they did not think that listening to Chinese was a problem. Some pupils pointed out the fast speed did affect listening comprehension. Sometimes the speech was 'so fast that I cannot get my head around it quickly enough.'

What is a good starting point for learning Chinese?
When asked what to start with when learning Chinese, teachers most strongly supported starting with oral words (M=2.93), as might be expected based on their global goal of communicating in Chinese. An early emphasis on oral words was repeated a number of times in interviews, for example "I do not even want to write down the Pinyin for them. The character sounds like this. You just have to listen to what the teacher says." However, the school students believed that starting with pinyin and characters was preferable, with the adult students, and teachers, showing a preference for pinyin. As one school student said "pinyin gives me a way to know how something sounds when the teacher isn't there. I mean, you don't know otherwise, see?" This may reflect the experiences of pinyin, discussed above.

Discussion
The beliefs of the students and teachers about learning spoken Chinese in this study are likely to affect their experiences of learning and teaching and so are important. The emphasis on learning and using spoken Chinese is reflected in the communicative orientation of the teachers, but the learners clearly found the linguistic elements and written forms to be an important focus.
Both the school pupils and the adult students in the present study believed that speaking was easier than listening to or understanding Chinese. This finding does not reflect the pattern of belief in older studies of University students in the USA (Samimy and Lee, 1997), which reported that students and instructors perceived speaking Chinese as more difficult than understanding the language. However, Gabbianelli et al. (2015) (this volume) in a study of 85 beginner learners of Chinese in Italy, also found that students perceived aural reception as more difficult than speaking, and they note that aural reception is generally perceived as the most difficult ability to achieve in language learning (Krashen 1982). The findings of the present study could be related to the beginner students and pupils' need to focus on the new, salient elements of speaking in Chinese and the theoretical argument above, that listening and discriminating Chinese homophones on the basis of tone is particularly difficult for English speaking learners. To explore this further, it is interesting to consider the teacher and student beliefs about difficulty for sound related elements of Chinese.
It is somewhat surprising that the questionnaire findings suggested that tones, a very complex item which English L1 students are unlikely to have experienced before, were not seen as the most problematic element by pupils, but were believed to be a major difficulty by teachers. As the teachers were likely to have evaluated the students' performance and understood the significance of tones better than the pupils, it is possible that the learners were underestimating the challenge that tones pose or the importance of the tones for comprehensibility.
This finding might be related to the expectations of learners and teachers in a language learning setting that emphasized spoken language, a communicative approach and risk taking in speaking. An alternative explanation could be that learners simply do not attend to tones in the beginning stages of learning Chinese. Although the students agreed that "I am aware of my tones when speaking Chinese" in the survey, the interview showed that students said they did not think this was a priority. A typical comment from a school student was "I try, but it is not so easy. I just can't get them all right or I wouldn't say anything".
And "I think tones are OK but I sometimes just pronounce it anyway and don't worry about the tone".
A great many of the school students interviewed knew about the importance of tones but also discussed them as an addition to a syllable, rather than an integral part. It is reasonable to assume that Englishspeaking learners might find the phonological aspects of Pinyin familiar from the onset-rime structures of English syllables, whereas the tones are likely to be unfamiliar. This may also explain the school students' approach.
Some further information can be derived from what the pupils, students and teachers thought should be introduced at the very start of Chinese learning. Whereas teachers favoured beginning with oral words, pupils were much more strongly in agreement with starting with characters and pinyin. This may reflect beliefs about what is important in language learning, because pupils did not agree that carrying on a conversation was more important than reading and writing, whereas the teachers did. The teachers seemed to have a greater focus on oral performance than the pupils, at least at this beginner level.
Finally, it was interesting that the interviews suggested a "pinyin-lite" approach for beginners in this sample, which may be the result of teachers' wish not to over burden beginner learners. Learners may have a very limited understanding of pinyin and be over-optimistic about the challenges it presents.
This study is based on students' and teachers' reports about their beliefs, which may not be a direct basis for their actions. More research in this area is needed to establish a robust understanding of the beliefs of students, especially those of school age.

Conclusion
This study presents the beliefs about Chinese language, language learning and, particularly, aspects of spoken language learning of school and university students, and their teachers. It is a picture of teachers who believe their students need to use language and take the risks inherent in speaking in a new language, and necessary to benefit from feedback. In this respect, the student's own beliefs reflect those of their teachers. They aim to use their limited Chinese in class, but find the unfamiliar linguistic elements are important and seem to be ambivalent about how much attention to allocate to them. Moreover, there seems to be a tension between the teacher's beliefs about to focus on spoken language, at least initially, and the student's wish to use pinyin and concentrate on the familiar, phonological aspects of pronunciation. It would be useful to explore further how students can engage with tones from the outset of their studies.

Beliefs about Chinese Language Learning (Student Survey)
The purpose of this survey is to help us find out more effective ways of teaching and learning Chinese. Your feedback is important to us. Please answer the questions below. This survey will take about 20 minutes to complete. Your participation in this survey is completely voluntary, and your responses will be kept confidential and anonymous.

Part 1 About You
18. It is ok to guess the sound of the character if you only know part of it. 19. I pay attention to my grammar when speaking Chinese. 20. It is important to practice in language laboratory with audio-visual and e-learning materials. 21. If you have other comments on learning strategies, please share them here. ____________________________________________________________

Part 6 Why do you learn a language?
Please indicate to what extent you agree or disagree with the following statements. 1. If I learn to speak Chinese very well, I will have many opportunities to use it. 2. I believe English people think that it is important to speak Chinese. 3. I would like to learn Chinese so that I can get to know Chinese people better. 4. If I learn Chinese, I will learn more about how other people think. 5. I would like to learn Chinese characters so that I can understand Chinese materials. 6. If I learn Chinese I will learn more about my own language. 7. I don't want to learn how to write Chinese characters because it is boring. 8. I believe Chinese people think that it is important to learn characters. 9. If I learn to speak Chinese very well it will help me get a good job. 10. If you have other comments on the purpose of learning a language, please share them here.

____________________________________________________________
Thank you for your time and cooperation. If you are interested in this study and would like to get further involved in the follow-up interview, please tell me your email address.